text
stringlengths 649
4.42k
| synonym_substitution
stringlengths 759
4.5k
| butter_fingers
stringlengths 649
4.42k
| random_deletion
stringlengths 453
2.31k
| change_char_case
stringlengths 649
4.42k
| whitespace_perturbation
stringlengths 764
5.02k
| underscore_trick
stringlengths 649
4.42k
|
|---|---|---|---|---|---|---|
)$. *Panel IV.* Projection on $(q,E)$ plane of $C_T(q,E)$ with frequency normalization to account for the bending related to the edge of first pseudo-Brillouin zone.[]{data-label="MODEL_Tdisp2ord2"}](Fig5_short_2_prova_2.pdf){width="1.01\linewidth"}
Our results shows that the mixing of polarization can be generated by disorder. Beyond Rayleigh anomalies an approximate solution of the Dyson equation can quantitatively account also for this latter phenomenon.
Appendix A {#appendix-a.unnumbered}
==========
Expressions of the longitudinal and transverse self-energies in the Generalized Born Approximation {#expressions-of-the-longitudinal-and-transverse-self-energies-in-the-generalized-born-approximation.unnumbered}
--------------------------------------------------------------------------------------------------
We derive the expressions of the longitudinal and transverse dynamic structure factors in the Generalized Born Approximation, by exploiting Eq. 4 in the main text. We re- write it here for sake of clarity, $$\begin{aligned}
& \Sigma_j(\textbf{q},\omega) \simeq lim_{\eta \rightarrow 0^+}\hat{L}_{1jk}\big\{\frac{1}{(\tilde{q}_{0k}+i \tilde{\eta})^2-q^2}+\frac{1}{((\tilde{q}_{0k}+i \tilde{\eta})^2-q^2)^2} \cdot \nonumber \\& \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \cdot \frac{\epsilon^2}{\tilde{c}_k^2}q^2\Delta \tilde{\Sigma}_k^{1*}(0,\omega + i \eta)\big\},
\label{Sigma_K_ap1}\end{aligned}$$ where $\eta$ and $\tilde{\eta}$ are related to each other by the relationship $\tilde{\eta}=\frac{\eta}{\tilde{c}_k}$. We recall that $\hat{\textbf{L}}_1=\epsilon^2q^2\hat{\textbf{L}}_1'$, $\Delta \tilde{\Sigma}_k^1(\textbf{q},\omega)=\tilde{\Sigma}_k^1(\textbf{q},\ | ) $. * Panel IV. * Projection on $ (q, E)$ plane of $ C_T(q, E)$ with frequency normalization to account for the deflection relate to the edge of first imposter - Brillouin zone.[]{data - label="MODEL_Tdisp2ord2"}](Fig5_short_2_prova_2.pdf){width="1.01\linewidth " }
Our results show that the mixing of polarization can be generated by disorderliness. Beyond Rayleigh anomaly an approximate solution of the Dyson equation can quantitatively account also for this latter phenomenon.
Appendix A { # appendix-a.unnumbered }
= = = = = = = = = =
Expressions of the longitudinal and transverse self - energies in the Generalized Born Approximation { # expression - of - the - longitudinal - and - transverse - self - energies - in - the - generalized - born - approximation.unnumbered }
--------------------------------------------------------------------------------------------------
We derive the expressions of the longitudinal and transverse active structure factors in the Generalized Born Approximation, by exploiting Eq. 4 in the independent text. We re- write it here for sake of clearness, $ $ \begin{aligned }
& \Sigma_j(\textbf{q},\omega) \simeq lim_{\eta \rightarrow 0^+}\hat{L}_{1jk}\big\{\frac{1}{(\tilde{q}_{0k}+i \tilde{\eta})^2 - q^2}+\frac{1}{((\tilde{q}_{0k}+i \tilde{\eta})^2 - q^2)^2 } \cdot \nonumber \\ & \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \cdot \frac{\epsilon^2}{\tilde{c}_k^2}q^2\Delta \tilde{\Sigma}_k^{1*}(0,\omega + i \eta)\big\ },
\label{Sigma_K_ap1}\end{aligned}$$ where $ \eta$ and $ \tilde{\eta}$ are related to each early by the relationship $ \tilde{\eta}=\frac{\eta}{\tilde{c}_k}$. We recall that $ \hat{\textbf{L}}_1=\epsilon^2q^2\hat{\textbf{L}}_1'$, $ \Delta \tilde{\Sigma}_k^1(\textbf{q},\omega)=\tilde{\Sigma}_k^1(\textbf{q},\ | )$. *Pajel IV.* Projection on $(q,E)$ klane of $C_T(q,E)$ witk frequxncy nodmalizatkon to account for the bendiig rwlatee to the edge of first pseudo-Brpllouin zine.[]{deta-label="MODEL_Tdisp2ord2"}](Fig5_short_2_pdlva_2.pbf){xidth="1.01\linewidth"}
Oor results svows that the kibiug of polarization can be generated fy disotdfr. Beyond Raylgigh sgomampew an approximate solution of fhe Dysmn equation csn quantitatively account wlso for this latter pjenomenon.
Apksndyz A {#appendix-x.unnumbered}
==========
Expressions of the longitudinal and transvdrse xelf-energigs ij the Generaoized Born Approxlkation {#expresxions-of-the-longltudiial-abd-transverse-self-energmes-in-the-generalized-bjrn-approxhmction.unnumbered}
--------------------------------------------------------------------------------------------------
We deruvw the expsesskins of tie mongitkdiial and trahsverse dynqmic structure factprf in the Generzlized Bjrn Approximation, by exploiting Eq. 4 in uhe mzin text. We re- write it here for sake of clatity, $$\begin{wligned}
& \Sigma_j(\textbf{q},\omega) \simeq lim_{\eta \rightarrof 0^+}\hat{M}_{1gk}\bnn\{\nrac{1}{(\gulfe{q}_{0k}+i \tilde{\eta})^2-q^2}+\frac{1}{((\tilde{q}_{0k}+i \tilde{\eta})^2-q^2)^2} \cdot \noghmneg \\& \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \cdot \wrac{\epsilkn^2}{\tilde{c}_k^2}q^2\Delta \tipde{\Sigmw}_k^{1*}(0,\ometa + i \eta)\fig\},
\lsbel{Sigma_K_ap1}\end{aligned}$$ wherw $\eta$ and $\tijee{\eta}$ are related co each othex by tne rekationship $\tilde{\eta}=\frac{\zta}{\tilse{c}_k}$. We recwll that $\gxt{\textbf{L}}_1=\epsilon^2d^2\han{\texdbf{L}}_1'$, $\Delux \tilde{\Sigma}_k^1(\textff{q},\omega)=\tmlde{\Sngma}_k^1(\texgbf{q},\ | )$. *Panel IV.* Projection on $(q,E)$ plane with normalization to for the bending first zone.[]{data-label="MODEL_Tdisp2ord2"}](Fig5_short_2_prova_2.pdf){width="1.01\linewidth"} Our results that the mixing polarization can be generated by disorder. Rayleigh anomalies an approximate solution of the Dyson equation can quantitatively account also this latter phenomenon. Appendix A {#appendix-a.unnumbered} ========== Expressions of the longitudinal and transverse in Generalized Approximation -------------------------------------------------------------------------------------------------- We derive the expressions of the longitudinal and transverse dynamic structure factors in the Generalized Approximation, by exploiting Eq. 4 in the main We re- write it for sake of clarity, $$\begin{aligned} \Sigma_j(\textbf{q},\omega) lim_{\eta \rightarrow \tilde{\eta})^2-q^2}+\frac{1}{((\tilde{q}_{0k}+i \cdot \\& \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \tilde{\Sigma}_k^{1*}(0,\omega + i \eta)\big\}, \label{Sigma_K_ap1}\end{aligned}$$ where are related to each other by the relationship We recall $\hat{\textbf{L}}_1=\epsilon^2q^2\hat{\textbf{L}}_1'$, $\Delta \tilde{\Sigma}_k^1(\textbf{q},\omega)=\tilde{\Sigma}_k^1(\textbf{q},\ | )$. *Panel IV.* Projection on $(q,E)$ planE of $C_T(q,E)$ witH freqUenCy nOrMaliZatiOn to account for THe beNding related to the edge oF firsT pSEudo-bRiLlouiN zone.[]{daTA-lABEl="MoDeL_tdiSp2ORd2"}](fig5_shOrt_2_Prova_2.pdF){width="1.01\lineWidTh"}
our results shOWs That the mixIng Of polarizatiOn cAn be geNeRatED by diSorDer. BeYond RaYLeigh aNomalies aN aPProximATe solutION oF the dyson equation can qUAnTItatively accouNt also FoR ThIS LatTer Phenomenon.
apPendiX a {#appendIX-a.UNNUmbERed}
==========
ExpressionS of the longiTUdiNal and TrAnsVErse seLf-eneRgIEs iN the GeneralIzed born ApproXimatiON {#expresSIons-of-tHe-longItuDinAl-anD-TrAnSveRsE-SelF-EnErgIEs-iN-the-geneRaLiZed-boRn-apPROXImatIon.UnnuMbereD}
--------------------------------------------------------------------------------------------------
We derive the eXprEssiONs oF the lOngitUdinAl And trAnsverSe dynAmIc structure factOrs iN the GenerAliZeD BoRn approXImatioN, by ExpLoiting eq. 4 in the MAin TeXT. wE rE- write it here for sakE oF CLaRity, $$\begiN{alignED}
& \SIgMA_j(\textbf{Q},\oMegA) \simEQ Lim_{\etA \rigHTaRrow 0^+}\hat{L}_{1Jk}\big\{\fRAc{1}{(\TiLde{q}_{0k}+i \tIlDe{\eta})^2-q^2}+\FrAc{1}{((\tIldE{q}_{0k}+i \tILde{\eTa})^2-q^2)^2} \cdoT \nonumbeR \\& \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \cdot \FRac{\epsilon^2}{\tildE{C}_k^2}q^2\Delta \tilde{\sIgMA}_K^{1*}(0,\oMEga + i \Eta)\Big\},
\label{SigMa_K_aP1}\End{aLignED}$$ wHerE $\Eta$ anD $\tildE{\eTA}$ aRE related to each other By The relAtionShip $\tilde{\eta}=\fRac{\eta}{\tildE{C}_K}$. we recall That $\HAt{\TExtbf{L}}_1=\epsilon^2q^2\Hat{\teXtbf{L}}_1'$, $\Delta \TIlde{\SigmA}_k^1(\texTbf{q},\omegA)=\tilde{\SigMA}_K^1(\textbf{q},\ | )$. *Panel IV.* Projection on $(q,E) $ pla neof$C _T(q ,E)$ with frequenc y nor malization to accountfor t he bend i ng rela ted tot he e dge o ffir st ps eudo- Bri llouinzone.[]{da ta- la bel="MODEL_T d is p2ord2"}]( Fig 5_short_2_pr ova _2.pdf ){ wid t h="1. 01\ linew idth"}
Our r esults sh ow s thatt he mixi n g o f po larization can be ge n erated by diso rder.Be y on d Ray lei gh anomali es an a p proxima t es o l uti o n of the Dyso n equationc anquanti ta tiv e ly acc ountal s o f or this lat terphenomeno n.
Ap p endix A {#appen dix-a. unn umb ered }
= == === == = =
E xp res s ion s of the l on gitud inal a n d tra nsv erse self -energies inthe Gen e ral izedBornAppr ox imati on {#e xpres si ons-of-the-long itud inal-and- tra ns ver se -self - energi es- in- the-gen eralize d -bo rn - a p pr oximation.unnumber ed } -- -------- ------ - -- -- - -------- -- --- ---- - - ----- ---- - -- -------- ------ - -- -- ------- -- ------ -- --- --- ---
W e de rive t he expre ssion s of the longit u dinal and tra n sv e r se dyna mic structurefact o rs i n th e G ene r alize d Bor nA pp r oximation, by explo it ing Eq . 4 i n the main te xt. We re- w r ite it h eref or sake of clarit y, $$ \begin{ali g ned}
& \ Sigma _j(\text bf{q},\om e g a) \sime q l im_ {\e ta\ r ig htarrow 0^+}\ h a t{L} _{ 1jk}\bi g\{ \frac{1 }{( \ti lde {q} _{ 0k}+i \ti lde{\eta }) ^2 -q ^2 }+\ frac{ 1 }{((\til de {q} _{ 0k} +i \t i lde{\e ta})^ 2-q^ 2) ^2 } \c dot \no n um b e r \\ & \ \\ \ \ \ \\ \\ \\ \ \ \ \ \ \ \\ \ \ \\\\ \ \ \ \ \ \ \ \ \ \ cdot \frac {\ eps ilon^2 } { \tilde{c }_k^2}q^2\Delta \tilde{ \ Sigma}_ k^{ 1*}(0 ,\om ega + i \ eta )\big\ },\ label{ Sigma_ K_ap1 }\ end { a ligne d } $$ wh er e $\eta$ a n d $\ tilde {\ eta} $ are r elated to each oth e r b y the relatio nsh ip $ \ t il de{ \ et a }=\ fr a c{\ e t a}{\tilde{c}_k} $. We reca ll th at $\hat{\ t ext bf {L}}_1= \epsilo n^2q^ 2 \hat{\t extbf{L}} _1'$, $\D el ta \ t i lde {\Sigma}_k ^1(\text bf{q},\om e ga)=\ t il de{\S igm a}_k^1 (\ tex tbf{q },\ | )$. *Panel_IV.* Projection_on $(q,E)$ plane of_$C_T(q,E)$ with_frequency_normalization to_account_for the bending_related to the_edge of first pseudo-Brillouin_zone.[]{data-label="MODEL_Tdisp2ord2"}](Fig5_short_2_prova_2.pdf){width="1.01\linewidth"}
Our results shows_that_the mixing of polarization can be generated by disorder. Beyond Rayleigh anomalies an approximate_solution_of the_Dyson_equation_can quantitatively account also for_this latter phenomenon.
Appendix A {#appendix-a.unnumbered}
==========
Expressions_of the_longitudinal and transverse self-energies in the Generalized Born_Approximation_{#expressions-of-the-longitudinal-and-transverse-self-energies-in-the-generalized-born-approximation.unnumbered}
--------------------------------------------------------------------------------------------------
We derive the_expressions of the longitudinal and transverse dynamic structure factors_in the Generalized Born Approximation, by_exploiting Eq. 4_in_the_main text. We re-_write it here for sake of_clarity, $$\begin{aligned}
& \Sigma_j(\textbf{q},\omega) \simeq lim_{\eta \rightarrow_0^+}\hat{L}_{1jk}\big\{\frac{1}{(\tilde{q}_{0k}+i \tilde{\eta})^2-q^2}+\frac{1}{((\tilde{q}_{0k}+i \tilde{\eta})^2-q^2)^2} \cdot \nonumber \\& _ \ \ \ \ \_\ \ \ \ \_\ \_\ \ \ \ \_\ \ \_\ \_\ \ \_\ \ \ \ \ \_\ \_\ \ \cdot \frac{\epsilon^2}{\tilde{c}_k^2}q^2\Delta \tilde{\Sigma}_k^{1*}(0,\omega +_i_\eta)\big\},
\label{Sigma_K_ap1}\end{aligned}$$ where $\eta$_and_$\tilde{\eta}$_are related_to each other_by_the relationship_$\tilde{\eta}=\frac{\eta}{\tilde{c}_k}$._We recall that $\hat{\textbf{L}}_1=\epsilon^2q^2\hat{\textbf{L}}_1'$, $\Delta \tilde{\Sigma}_k^1(\textbf{q},\omega)=\tilde{\Sigma}_k^1(\textbf{q},\ |
mc] a scale, $r_{med}$, defined as the distance at which the value of the $T=0$ potential reaches the value $F_\infty(T)$, [*i.e.*]{} $$\begin{aligned}
V(r_{med})&\equiv&F_\infty(T)\;.\label{rmed}\end{aligned}$$ As $F_\infty(T)$ is a gauge invariant observable this relation provides a non-perturbative, gauge invariant definition of the scale $r_{med}$. While in pure gauge theory the color singlet free energies signal permanent confinement at temperatures below $T_c$ leading to a proper definition of this scale only above deconfinement, in full QCD it can be deduced in the whole temperature range. On the other hand, the change in the coupling $\alpha_{qq}(r,T)$ as function of distance at fixed temperature mimics the qualitative change in the interaction when going from small to large distances and the coupling exhibits a maximum at some intermediate distance. The location of this maximum, $r_{max}$, can also be used to identify a scale that characterizes separation between the short distance vacuumlike and the large distance medium modified interaction between the static quarks [@Kaczmarek:2004gv]. Due to the rapid crossover from short to large distance behavior (see Fig. \[fes\](a)) it should be obvious that $r_{med}$ and $r_{max}$ define similar scales, however, by construction $r_{max}\;\lsim\;r_{med}$.
To gain important information about the flavor and quark mass dependence of our analysis of the scales in QCD, we also took data for $F_\infty(T)$ from Ref. [@Petreczky:2004pz] at smaller quark mass, $m_\pi/m_\rho\simeq0.4$ [@Peterpriv], and calculated $r_{med}$ in $3$-flavor QCD with respect to the parameterization of $V(r)$ given in [@Petreczky:2004pz]. It is interesting to note here that a study of the flavor and quark mass dependence of $r_{med}$ and $r_{max}$ is independent of any undetermined and maybe flavor and/or quark mass dependent overall normalization of the corresponding $V(r)$ at zero temperature. Our results for $r_{max}$ ($N_f$=0,2) and $r_{med}$ ($N_f$=0,2, | mc ] a scale, $ r_{med}$, defined as the distance at which the value of the $ T=0 $ potential reaches the value $ F_\infty(T)$, [ * i.e. * ] { } $ $ \begin{aligned }
V(r_{med})&\equiv&F_\infty(T)\;.\label{rmed}\end{aligned}$$ As $ F_\infty(T)$ is a bore changeless discernible this relation provides a non - perturbative, gauge changeless definition of the scale $ r_{med}$. While in pure gauge hypothesis the coloring material singlet free energy signal permanent confinement at temperature below $ T_c$ leading to a proper definition of this scale entirely above deconfinement, in wide QCD it can be deduced in the whole temperature range. On the other hand, the variety in the coupling $ \alpha_{qq}(r, T)$ as function of distance at fixed temperature mimic the qualitative change in the interaction when going from small to large distances and the yoke exhibits a utmost at some intermediate distance. The location of this maximum, $ r_{max}$, can also be use to identify a scale that characterizes separation between the short distance vacuumlike and the big distance medium modified interaction between the static quarks [ @Kaczmarek:2004gv ]. Due to the rapid crossover from short to large distance behavior (see Fig. \[fes\](a) ) it should be obvious that $ r_{med}$ and $ r_{max}$ define alike scales, however, by construction $ r_{max}\;\lsim\;r_{med}$.
To profit authoritative information about the flavor and quark mass dependence of our analysis of the scales in QCD, we also accept data for $ F_\infty(T)$ from Ref. [ @Petreczky:2004pz ] at smaller quark mass, $ m_\pi / m_\rho\simeq0.4 $ [ @Peterpriv ], and calculated $ r_{med}$ in $ 3$-flavor QCD with respect to the parameterization of $ V(r)$ given in [ @Petreczky:2004pz ]. It is interesting to note here that a study of the flavor and quark mass dependence of $ r_{med}$ and $ r_{max}$ is autonomous of any undetermined and possibly flavor and/or quark mass dependent overall normalization of the corresponding $ V(r)$ at zero temperature. Our results for $ r_{max}$ ($ N_f$=0,2) and $ r_{med}$ ($ N_f$=0,2, | mc] w scale, $r_{med}$, defined as uhe distance at wkuch thx value of the $G=0$ potential reaches the valux $F_\ibfty(T)$, [*i.e.*]{} $$\begin{aligned}
V(r_{med})&\eduiv&F_\inftj(T)\;.\label{rmwd}\env{aligned}$$ As $F_\infvg(T)$ is a gauge lnvarnait observable tmis relatiot provides a nmn-oexturbative, gauge invariant definitiog of thr dcale $r_{med}$. Whije im purs gauge theory the color singlet fdee eneggies signal permsnent confinement at tempegatuges below $T_c$ leadijg to a proksr qwfinition of this scalt mnly above deconfinement, in full QCD it cxn be deduced ib rhe fhole tempecature range. On thc other hand, tne change in tme conplibg $\alpha_{qq}(r,T)$ as functmon of distance at fyxed temparcture mimics the qualutqtive chatge kb tfe jnverzction whxn going frkm small to large distances anc eye coupling eshibitf w maximum at some intermediate distance. Ths location of this maxinum, $r_{max}$, can also be osed to idqntify a scale that characterizes separation betwaen tie shirb dirrajce vacuumlike and the large distance medium ikdofped interaction bctween the static auwrlf [@Kaczmarek:2004gv]. Due to tge rapid crossover from srort ro large qistsnce behavior (see Fig. \[fes\](a)) ir should be jvvious that $r_{med}$ aud $r_{max}$ defiue simolar xcales, however, by constxuctioh $r_{max}\;\lsim\;r_{led}$.
To gaih important inforoatpon dbout tht flavor and quark mass depxndenee of ouf anslysis of the scwles lt QCD, we also took data fmr $F_\infty(T)$ from Ref. [@Petreczky:2004pz] at smaller quark mass, $m_\po/m_\sho\vimeq0.4$ [@Pecerpriy], and calculateq $r_{med}$ in $3$-flavpr QCD cith rdspect to nhe paramxterization jf $V(r)$ given it [@Petreczky:2004pz]. It is igterwstibg to nuge here that a study of the flavoe and quark mass dcpendgnde of $r_{med}$ and $x_{oqx}$ is independemt uf wnj uidetewkined and mafbe wlaxpr ana/or quark mcsr delendent overall normdlizztion of the correxpjnding $V(t)$ at zero temperature. Pur results for $r_{mwx}$ ($N_f$=0,2) anv $r_{med}$ ($N_f$=0,2, | mc] a scale, $r_{med}$, defined as the which value of $T=0$ potential reaches V(r_{med})&\equiv&F_\infty(T)\;.\label{rmed}\end{aligned}$$ $F_\infty(T)$ is a invariant observable this provides a non-perturbative, gauge invariant definition the scale $r_{med}$. While in pure gauge theory the color singlet free energies permanent confinement at temperatures below $T_c$ leading to a proper definition of this only deconfinement, full it can be deduced in the whole temperature range. On the other hand, the change in coupling $\alpha_{qq}(r,T)$ as function of distance at fixed mimics the qualitative change the interaction when going from to distances and coupling a at some intermediate The location of this maximum, $r_{max}$, can also be used to identify a scale that characterizes separation the short and the distance modified between the static Due to the rapid crossover from distance behavior (see Fig. \[fes\](a)) it should be that $r_{med}$ $r_{max}$ define similar scales, however, by $r_{max}\;\lsim\;r_{med}$. To gain important information about the flavor quark mass dependence of our analysis of the scales in QCD, we also took data from Ref. [@Petreczky:2004pz] at quark mass, $m_\pi/m_\rho\simeq0.4$ and $r_{med}$ $3$-flavor with respect the parameterization of $V(r)$ given in [@Petreczky:2004pz]. It is interesting to here that a study of the flavor and quark mass $r_{med}$ $r_{max}$ is independent any undetermined and maybe and/or mass dependent overall normalization corresponding at results $r_{max}$ and $r_{med}$ ($N_f$=0,2, | mc] a scale, $r_{med}$, defined as the dIstance at wHich tHe vAluE oF the $t=0$ potEntial reaches tHE valUe $F_\infty(T)$, [*i.e.*]{} $$\begin{aligneD}
V(r_{meD})&\eQUiv&F_\INfTy(T)\;.\laBel{rmed}\ENd{ALIgnEd}$$ as $f_\inFtY(t)$ iS a gauGe iNvarianT observablE thIs Relation provIDeS a non-pertuRbaTive, gauge invAriAnt defInItiON of thE scAle $r_{mEd}$. WhilE In pure Gauge theoRy THe coloR Singlet FREe EnerGies signal permaneNT cONfinement at temPeratuReS BeLOW $T_c$ LeaDing to a proPeR defiNItion of THiS SCAle ONly above deconFinement, in fULl QcD it caN bE deDUced in The whOlE TemPerature ranGe. On The other hAnd, the CHange in THe couplIng $\alpHa_{qQ}(r,T)$ As fuNCtIoN of DiSTanCE aT fiXEd tEmperatuRe MiMics tHe quALITAtivE chAnge In the Interaction whEn gOing FRom Small To larGe diStAnces And the CouplInG exhibits a maximUm at Some interMedIaTe dIsTance. tHe locaTioN of This maxImum, $r_{maX}$, Can AlSO BE uSed to identify a scalE tHAT cHaracterIzes sePArAtIOn betweeN tHe sHort DIStancE vacUUmLike and tHe largE DiStAnce medIuM modifIeD inTerActioN BetwEen the Static quArks [@KACzmarek:2004gv]. Due to THe rapid crossoVEr FROm SHort To lArge distancE behAVior (See FIG. \[fEs\](a)) IT shouLd be oBvIOuS That $r_{med}$ and $r_{max}$ defiNe SimilaR scalEs, however, by coNstruction $R_{MAX}\;\lsim\;r_{meD}$.
To gAIn IMportant informAtion About the flAVor and quArk maSs dependEnce of our ANAlysis of The ScaLes In Qcd, We Also took data fOR $f_\infTy(t)$ from ReF. [@PeTreczky:2004Pz] aT smAllEr qUaRk mass, $m_\pi/M_\rho\simeQ0.4$ [@PEtErPrIv], aNd calCUlated $r_{mEd}$ In $3$-fLaVor qCD wiTH respeCt to tHe paRaMeTEriZation oF $v(r)$ GIVen iN [@PEtReczKy:2004pZ]. IT is inTereSTinG to note Here that a StuDY of tHe FlAvor and Quark mass depeNdEnce of $r_{med}$ AnD $r_{mAx}$ is inDEPendent oF any undetermined and maybE Flavor aNd/oR quarK masS dependenT ovErall nOrmALizatiOn of thE corrEsPonDINg $V(r)$ aT ZErO teMpErature. Our RESulTs for $R_{mAx}$ ($N_f$=0,2) And $r_{med}$ ($n_f$=0,2, | mc] a scale, $r_{med}$, de fined as t he di sta nce a t wh ichthe value of t h e $T =0$ potential reachesthe v al u e $F _ \i nfty( T)$, [* i .e . * ]{} $ $\ beg in { al igned }
V (r_{med })&\equiv& F_\ in fty(T)\;.\la b el {rmed}\end {al igned}$$ As$F_ \infty (T )$i s a g aug e inv ariant observ able this r e lation provide s anon- perturbative, gau g ei nvariant defin itionof th e sca le$r_{med}$. W hilei n pureg au g e the o ry the colorsinglet fre e en ergies s ign a l perm anent c o nfi nement at t empe ratures b elow $ T _c$ lea d ing toa prop erdef init i on o f t hi s sc a le on l y a bove dec on fi nemen t, i n f u ll Q CDit c an be deduced in t hewhol e te mpera turerang e. On t he oth er ha nd , the change in the coupling $\ al pha _{ qq}(r , T)$ as fu nct ion ofdistanc e at f i x e dtemperature mimics t h e q ualitati ve cha n ge i n the int er act ionw h en go ingf ro m smallto lar g edi stances a nd the c oup lin g exh i bits a max imum atsomei ntermediate di s tance. The lo c at i o no f th ismaximum, $r _{ma x }$,cana ls o b e used to i de n ti f y a scale that char ac terize s sep aration betwe en the sho r t distance vac u um l ike and the la rge d istance me d ium modi fiedinteract ion betwe e n the sta tic qu ark s [ @ K ac zmarek:2004gv ] . Due t o the r api d cross ove r f rom sh or t to larg e distan ce b eh av ior (see Fig. \[f es \]( a) ) i t sho u ld beobvio us t ha t$ r_{ med}$ a n d$ r _{ma x} $defi nesi milar sca l es, howeve r, by con str u ctio n$r _{max}\ ;\lsim\;r_{me d} $.
To gai nimp ortant i nformati on about the flavor and quark m ass depe nden ce of our an alysis of the sc ales i n QCD ,wea l so to o k d ata f or $F_\inf t y (T) $ fro mRef. [@Petr eczky:2004pz] at s m all er quark mass , $ m_\p i / m_ \rh o \s i meq 0. 4 $ [ @ P eterpriv], andcalculated $ r _{ med}$ in $ 3 $-f la vor QCD with r espec t to the paramete rizationof $V( r ) $ g iven in [@ Petreczk y:2004pz] . It i s i ntere sti ng tono teherethat a stu dy of the f la vor an d qua rk mass de pendence of $r_{med}$ a nd $r_ {max} $ i s indepen den t of any unde term ined and m ayb e f lavor an d /or q uark ma ssd epend ento verall no r ma liz a t io n of the co r r e spo nding $V ( r)$ at zer o temperature. Ou r results for $ r_{m a x }$($N _ f$=0 ,2 ) and $r_{med} $ ( $N _ f $=0,2, | mc] a_scale, $r_{med}$,_defined as the distance_at which_the_value of_the_$T=0$ potential reaches_the value $F_\infty(T)$,_[*i.e.*]{} $$\begin{aligned}
V(r_{med})&\equiv&F_\infty(T)\;.\label{rmed}\end{aligned}$$ As $F_\infty(T)$_is a gauge_invariant_observable this relation provides a non-perturbative, gauge invariant definition of the scale $r_{med}$. While_in_pure gauge_theory_the_color singlet free energies signal_permanent confinement at temperatures below_$T_c$ leading_to a proper definition of this scale only_above_deconfinement, in full_QCD it can be deduced in the whole temperature_range. On the other hand, the_change in the_coupling_$\alpha_{qq}(r,T)$_as function of distance_at fixed temperature mimics the qualitative_change in the interaction when going_from small to large distances and the_coupling exhibits a maximum at some_intermediate distance. The location of_this maximum,_$r_{max}$, can also be used_to identify a_scale that_characterizes separation between_the short distance vacuumlike and the_large distance medium_modified interaction between the static quarks_[@Kaczmarek:2004gv]._Due to the_rapid_crossover_from short_to large distance_behavior_(see Fig. \[fes\](a))_it_should be obvious that $r_{med}$ and_$r_{max}$_define similar scales, however, by construction $r_{max}\;\lsim\;r_{med}$.
To_gain important information about_the_flavor and quark mass_dependence of our analysis of_the scales in QCD, we also_took data_for $F_\infty(T)$_from Ref. [@Petreczky:2004pz] at smaller quark mass, $m_\pi/m_\rho\simeq0.4$ [@Peterpriv], and calculated $r_{med}$_in $3$-flavor QCD with respect to_the parameterization of $V(r)$_given in_[@Petreczky:2004pz]._It is interesting_to_note here_that a study of the flavor and_quark mass_dependence of $r_{med}$ and $r_{max}$ is_independent of any undetermined_and_maybe flavor and/or quark mass dependent_overall normalization of the corresponding $V(r)$_at zero temperature. Our results_for_$r_{max}$_($N_f$=0,2) and $r_{med}$ ($N_f$=0,2, |
same as that of $\hat{\beta}_n$. These arguments can be justified similarly as those in @John:Stra:indu:2009.
Variance Estimation {#sect:vari}
===================
The asymptotic variance of $\tilde\beta_n$ is even harder to estimate for case-cohort data than for full cohort data because of the extra complexity caused by the data structure. The terms in the summation in $\tilde{U}_n(\beta)$ are not independent since the sub-cohort is drawn from the full cohort without replacement. We propose four variance estimators; one is fully resampling based while the other three use resampling to a component of the sandwich variance estimator.
Multiplier Bootstrap
--------------------
The multiplier bootstrap estimator of @Jin:Lin:Wei:Ying:rank:2003 is adapted to case-cohort data by inserting proper case-cohort weights, $h_i$’s, in the multiplier bootstrap estimating equations. Let $\eta_i$, $i = 1, \dots, n$, be independent and identically distributed positive random variables with ${\mathrm{E}}(\eta_i) = {\mathrm{Var}}(\eta_i) = 1$. Define $$\tilde{U}_n^{c*}(\beta) = \sum_{i=1}^n \sum_{j=1}^n \eta_i\eta_jh_j\Delta_i (X_i- X_j) \Phi \left[ \frac{e_j(\beta)-e_i(\beta)}{r_{ij}^2} \right].
\label{eq:UnSmoothC:Res}$$ For a realization of $(\eta_1, \ldots, \eta_n)$, the solution to provides one draw of $\tilde\beta_n$ from its asymptotic distribution. By repeating this process a large number $B$ times, the variance matrix of $\tilde\beta_n$ can be estimated directly by the sampling variance matrix of the bootstrap sample of $\tilde\beta_n$.
Since the asymptotic variance of $\hat{\beta}_n$ is the same as that of $\tilde{\beta}_n$, the covariance matrix of $\tilde{\beta}_n$ can also be estimated by through multiplier bootstrap. This is, however, not recommended because it would need to solve a large number $B$ nonsmooth estimating equations. As will be seen in our simulation study | same as that of $ \hat{\beta}_n$. These arguments can be justified similarly as those in @John: Stra: indu:2009.
Variance Estimation { # sect: vari }
= = = = = = = = = = = = = = = = = = =
The asymptotic discrepancy of $ \tilde\beta_n$ is even hard to estimate for case - cohort datum than for full cohort data because of the excess complexity induce by the data social organization. The term in the summation in $ \tilde{U}_n(\beta)$ are not independent since the sub - cohort is drawn from the full age group without replacement. We propose four variance calculator; one is fully resampling based while the other three use resampling to a component of the sandwich discrepancy estimator.
Multiplier Bootstrap
--------------------
The multiplier bootstrap estimator of @Jin: Lin: Wei: Ying: rank:2003 is adapted to case - cohort datum by inserting proper case - cohort weights, $ h_i$ ’s, in the multiplier bootstrap estimating equations. lease $ \eta_i$, $ i = 1, \dots, n$, be independent and identically distributed positive random variables with $ { \mathrm{E}}(\eta_i) = { \mathrm{Var}}(\eta_i) = 1$. Define $ $ \tilde{U}_n^{c*}(\beta) = \sum_{i=1}^n \sum_{j=1}^n \eta_i\eta_jh_j\Delta_i (X_i- X_j) \Phi \left [ \frac{e_j(\beta)-e_i(\beta)}{r_{ij}^2 } \right ].
\label{eq: UnSmoothC: Res}$$ For a realization of $ (\eta_1, \ldots, \eta_n)$, the solution to provides one draw of $ \tilde\beta_n$ from its asymptotic distribution. By repeating this process a large act $ B$ times, the variance matrix of $ \tilde\beta_n$ can be calculate immediately by the sample variance matrix of the bootstrap sample of $ \tilde\beta_n$.
Since the asymptotic variance of $ \hat{\beta}_n$ is the same as that of $ \tilde{\beta}_n$, the covariance matrix of $ \tilde{\beta}_n$ can also be estimated by through multiplier bootstrap. This is, however, not recommended because it would need to resolve a large number $ B$ nonsmooth estimating equations. As will be seen in our simulation discipline | sale as that of $\hat{\beta}_n$. Tmese arguments ccb be jnstifies similafly as those in @John:Stra:indu:2009.
Tariqnce Tftimation {#sect:vari}
===================
Ghe asympnotic varuanct of $\tilde\beta_n$ is even harder fl escinate for case-cphort data than for full cuhlrt data because of the extra complqxity csuded by the datw stgusturs. The terms in the summation in $\timde{U}_n(\beua)$ are not indepencent since the sub-cohort id drwwn from the full fohort withiut wwplacement. Wd propose four variancg estimators; one is fully resampuing yased while tye ljher three uwe refampling to a componant of yhe sandwich vsrience estimator.
Multiplier Uootstrap
--------------------
The multiplyer bootsdrcp estimator of @Jin:Lib:Wwi:Yinc:rann:2003 is qdaotes vo dase-cojorv data by ihserting priper case-cohort weibhew, $h_i$’s, in the jultipjiqr bootstrap estimating equations. Let $\eua_i$, $i = 1, \dots, n$, be independenr and identically disjributed pjsitive random variables with ${\mathrm{E}}(\eta_i) = {\mathrm{Ear}}(\ete_i) = 1$. Bcnine $$\ripde{U}_n^{c*}(\beta) = \sum_{i=1}^n \sum_{j=1}^n \eta_i\eta_jh_j\Delta_i (X_i- X_t) \Pni \left[ \frac{e_j(\beba)-e_i(\beta)}{r_{ij}^2} \right].
\lwbrj{eq:UnSmoothC:Rgs}$$ For c rsalization of $(\eta_1, \pdots, \eja_n)$, thw solutiog to provides one draw of $\tilde\veta_n$ from ins awymptotic distribucion. By repecting jhis ptocess a large number $B$ timss, the variwnce matrjb of $\tilde\beta_n$ zan ba estimaudd directly by thq samplinj varnance magrix of thq bootstrao samine of $\tilde\beta_n$.
Sijce tke asfmptotic vwriance of $\hat{\beta}_n$ is the same as that of $\tikda{\bena}_n$, the cjvarisnce matrix os $\tilde{\beta}_n$ ccn also ye estkmated by nhrough mnltiplier bojtstrap. This hd, however, nov recommegded becquse it dould need to xolve a lcxge numbee $B$ nonsmooth estikatkhg equations. As wiol be seen in oir rimtlwtmon segdy | same as that of $\hat{\beta}_n$. These arguments justified as those @John:Stra:indu:2009. Variance Estimation of is even harder estimate for case-cohort than for full cohort data because the extra complexity caused by the data structure. The terms in the summation $\tilde{U}_n(\beta)$ are not independent since the sub-cohort is drawn from the full cohort replacement. propose variance one is fully resampling based while the other three use resampling to a component of the variance estimator. Multiplier Bootstrap -------------------- The multiplier bootstrap of @Jin:Lin:Wei:Ying:rank:2003 is adapted case-cohort data by inserting proper weights, in the bootstrap equations. $\eta_i$, $i = \dots, n$, be independent and identically distributed positive random variables with ${\mathrm{E}}(\eta_i) = {\mathrm{Var}}(\eta_i) = 1$. Define = \sum_{i=1}^n (X_i- X_j) \left[ \right]. For a realization \ldots, \eta_n)$, the solution to provides $\tilde\beta_n$ from its asymptotic distribution. By repeating this a large $B$ times, the variance matrix of can be estimated directly by the sampling variance of the bootstrap sample of $\tilde\beta_n$. Since the asymptotic variance of $\hat{\beta}_n$ is the same of $\tilde{\beta}_n$, the covariance of $\tilde{\beta}_n$ can be by multiplier This is, not recommended because it would need to solve a large number nonsmooth estimating equations. As will be seen in our simulation | same as that of $\hat{\beta}_n$. These Arguments cAn be jUstIfiEd SimiLarlY as those in @John:sTra:iNdu:2009.
Variance Estimation {#sEct:vaRi}
===================
tHe asYMpTotic VariancE Of $\TILde\BeTa_N$ is EvEN hArder To eStimate For case-cohOrt DaTa than for fulL CoHort data beCauSe of the extra ComPlexitY cAusED by thE daTa strUcture. tHe termS in the sumMaTIon in $\tILde{U}_n(\beTA)$ ArE not Independent since tHE sUB-cohort is drawn From thE fULl COHorT wiThout replaCeMent. WE Propose FOuR VARiaNCe estimators; oNe is fully reSAmpLing baSeD whILe the oTher tHrEE usE resampling To a cOmponent oF the saNDwich vaRIance esTimatoR.
MuLtiPlieR boOtStrAp
--------------------
tHe mULtIplIEr bOotstrap EsTiMator Of @JiN:lIN:wei:YIng:Rank:2003 Is adaPted to case-cohOrt Data BY inSertiNg proPer cAsE-cohoRt weigHts, $h_i$’S, iN the multiplier bOotsTrap estimAtiNg EquAtIons. LET $\eta_i$, $i = 1, \DotS, n$, bE indepeNdent anD IdeNtICALlY distributed positiVe RANdOm variabLes witH ${\MaThRM{E}}(\eta_i) = {\maThRm{VAr}}(\etA_I) = 1$. definE $$\tilDE{U}_N^{c*}(\beta) = \suM_{i=1}^n \sum_{J=1}^N \eTa_I\eta_jh_j\deLta_i (X_i- x_j) \phi \LefT[ \frac{E_J(\betA)-e_i(\betA)}{r_{ij}^2} \righT].
\labeL{Eq:UnSmoothC:Res}$$ fOr a realizatioN Of $(\ETA_1, \lDOts, \eTa_n)$, The solution To prOVideS one DRaW of $\TIlde\bEta_n$ fRoM ItS Asymptotic distributIoN. By repEatinG this process a Large numbeR $b$ TImes, the vAriaNCe MAtrix of $\tilde\beTa_n$ caN be estimatED directlY by thE samplinG variance MATrix of thE boOtsTraP saMPLe Of $\tilde\beta_n$.
SINCe thE aSymptotIc vAriance Of $\hAt{\bEta}_N$ is ThE same as thAt of $\tildE{\bEtA}_n$, ThE coVariaNCe matrix Of $\TilDe{\BetA}_n$ can ALso be eStimaTed bY tHrOUgh MultiplIEr BOOtstRaP. THis iS, hoWeVer, noT recOMmeNded becAuse it wouLd nEEd to SoLvE a large Number $B$ nonsmoOtH estimatinG eQuaTions. AS WIll be seeN in our simulation study | same as that of $\hat{\be ta}_n$. Th ese a rgu men ts can bejustified simi l arly as those in @John:Str a:ind u: 2 009.
V arian ce Esti m at i o n { #s ec t:v ar i }===== === ======= ====
Theasy mp totic varian c eof $\tilde \be ta_n$ is eve n h arderto es t imate fo r cas e-coho r t data than for f u ll coh o rt data b ec ause of the extra com p le x ity caused bythe da ta st r u ctu re. The terms i n the summati o ni n $\t i lde{U}_n(\bet a)$ are not ind epende nt si n ce the sub- co h ort is drawn f romthe fullcohort without replace ment.Wepro pose fo ur va ri a nce es tim a tor s; one i sfu lly r esam p l i n g ba sed whi le th e other three us e re s amp lingto acomp on ent o f thesandw ic h variance esti mato r.
Multi pli er Bo ot strap ------ --- --- ------- -
Them ult ip l i e rbootstrap estimato ro f @ Jin:Lin: Wei:Yi n g: ra n k:2003 i sada pted t o cas e-co h or t data b y inse r ti ng proper c ase-co ho rtwei ghts, $h_i $’s, i n the mu ltipl i er bootstrap e s timating equa t io n s .L et $ \et a_i$, $i =1, \ d ots, n$, be in d epend ent a nd id e ntically distribute dpositi ve ra ndom variable s with ${\ m a t hrm{E}}( \eta _ i) = {\mathrm{Var }}(\e ta_i) = 1$ . Define$$\ti lde{U}_n ^{c*}(\be t a ) = \sum _{i =1} ^n\su m _ {j =1}^n \eta_i\ e t a_jh _j \Delta_ i ( X_i- X_ j)\Ph i \ lef t[ \frac{e_ j(\beta) -e _i (\ be ta) }{r_{ i j}^2} \r ig ht] . \ label { eq:UnS mooth C:Re s} $$ For a real i za t i on o f$( \eta _1, \ ldots , \e t a_n )$, the solution to pro vi de s one d raw of $\tild e\ beta_n$ fr om it s asym p t otic dis tribution. By repeating this pr oce ss alarg e number$B$ times , t h e vari ance m atrix o f $ \ t ilde\ b e ta _n$ c an be esti m a ted dire ct ly b y the s ampling variance m a tri x of the boot str ap s a m pl e o f $ \ til de \ bet a _ n$.
Since theasymptotic v a ri ance of $\ h at{ \b eta}_n$ is the same as that of $\til de{\beta} _n $, t h e co variance m atrix of $\tilde{ \ beta} _ n$ canals o be e st ima ted b y thr o ugh mult iplier b ootstr ap. T hi s is, ho wever, not recommendedbecaus e itwou ld need t o s o lve a largenumb er $B$ non smo oth esti mat i ng eq uati o ns . A s will bes een in ou r s imu l a ti on study | same_as that_of $\hat{\beta}_n$. These arguments_can be_justified_similarly as_those_in @John:Stra:indu:2009.
Variance Estimation_{#sect:vari}
===================
The asymptotic variance_of $\tilde\beta_n$ is even_harder to estimate_for_case-cohort data than for full cohort data because of the extra complexity caused by_the_data structure._The_terms_in the summation in $\tilde{U}_n(\beta)$_are not independent since the_sub-cohort is_drawn from the full cohort without replacement. We_propose_four variance estimators;_one is fully resampling based while the other three_use resampling to a component of_the sandwich variance_estimator.
Multiplier_Bootstrap
--------------------
The_multiplier bootstrap estimator of_@Jin:Lin:Wei:Ying:rank:2003 is adapted to case-cohort data_by inserting proper case-cohort weights, $h_i$’s,_in the multiplier bootstrap estimating equations. Let_$\eta_i$, $i = 1, \dots, n$,_be independent and identically distributed_positive random_variables with ${\mathrm{E}}(\eta_i) = {\mathrm{Var}}(\eta_i)_= 1$. Define_$$\tilde{U}_n^{c*}(\beta) =_\sum_{i=1}^n \sum_{j=1}^n \eta_i\eta_jh_j\Delta_i_(X_i- X_j) \Phi \left[ \frac{e_j(\beta)-e_i(\beta)}{r_{ij}^2} \right].
_ \label{eq:UnSmoothC:Res}$$ For_a realization of $(\eta_1, \ldots, \eta_n)$,_the_solution to provides_one_draw_of $\tilde\beta_n$_from its asymptotic_distribution._By repeating_this_process a large number $B$ times,_the_variance matrix of $\tilde\beta_n$ can be estimated_directly by the sampling_variance_matrix of the bootstrap_sample of $\tilde\beta_n$.
Since the asymptotic_variance of $\hat{\beta}_n$ is the same_as that_of $\tilde{\beta}_n$,_the covariance matrix of $\tilde{\beta}_n$ can also be estimated by through_multiplier bootstrap. This is, however, not_recommended because it would_need to_solve_a large number_$B$_nonsmooth estimating_equations. As will be seen in our_simulation study |
}$ and $J$. In the ‘large’ polaron limit this parameter takes a small value, while with increasing electron-phonon coupling it tends to unity, showing a distinct crossover from ‘large’ to ‘small’ polaron behavior. The measure of delocalization of the electron for various ranges of $g_{-}$ as well as $J$ will be evident from the kinetic energy. So we have also calculated the kinetic energy, given by, $$t_{eff}^{KE}=-E_{Kin}=<\psi_G|
t \frac{S_0 + \frac{1}{2}}{2 S + 1}
~ [c_{1}^{\dag} c_{2}~ \exp
(2 \lambda (d^{\dag}-d))
+ c_{2}^{\dag} c_{1}~\exp(-2 \lambda (d^{\dag}-d)) ~]
|\psi_G>,$$ where $\psi_{G}$ is the ground state wave-function, is evaluated upto the fifth order in the perturbation. The numerical evaluation of $E_{Kin}$ will be presented below in Sec. IV.
Phase diagrams and Specific Heat
================================
Recently, there have been many experimental reports on manganites at low doping and low temperatures with and without an external magnetic field [@cv; @cv1; @cv2]. Okuda et al have estimated the electronic specific heat for $La_{1-x}Sr_xMnO_3$ in the ferromagnetic regime and concluded that the carrier mass-renormalization near the metal-insulator transition at $x=0.16$ is minimal. They have also observed a decrease in the low temperature specific heat in the presence of a magnetic field. Motivated by these observations, we have carried out a calculation of the specific heat, based on the partition function of the system which, from a cumulant expansion upto the 2nd order, is given by [@sd],
$$Z(\beta) = Z_0(\beta) exp{(-\int^{\beta}_{0} d\beta^\prime
\int^{\beta^{\prime}}_{0} d\beta^{\prime \prime}\langle \tilde H_1(\beta^\prime)\tilde H_1(\beta^{\prime \prime})\rangle)},$$
where $Z_0(\beta) = Tr (e^{-\beta H_0})$ ; $\tilde H_{1} (\beta) = e^{\beta H_0 | } $ and $ J$. In the ‘ large ’ polaron limit this parameter takes a humble value, while with increase electron - phonon coupling it tends to unity, prove a distinct crossover from ‘ big ’ to ‘ little ’ polaron behavior. The bill of delocalization of the electron for various ranges of $ g_{-}$ equally well as $ J$ will be evident from the kinetic energy. So we have besides calculated the kinetic department of energy, given by, $ $ t_{eff}^{KE}=-E_{Kin}=<\psi_G|
deoxythymidine monophosphate \frac{S_0 + \frac{1}{2}}{2 S + 1 }
~ [ c_{1}^{\dag } c_{2}~ \exp
(2 \lambda (d^{\dag}-d) )
+ c_{2}^{\dag } c_{1}~\exp(-2 \lambda (d^{\dag}-d) ) ~ ]
|\psi_G>,$$ where $ \psi_{G}$ is the ground state wave - function, is evaluated upto the fifth club in the perturbation. The numerical evaluation of $ E_{Kin}$ will be presented below in Sec. IV.
Phase diagram and Specific Heat
= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
Recently, there have been many experimental reports on manganite at low doping and low temperatures with and without an external magnetic discipline [ @cv; @cv1; @cv2 ]. Okuda et al have estimated the electronic specific heat for $ La_{1 - x}Sr_xMnO_3 $ in the ferromagnetic regime and concluded that the carrier bulk - renormalization near the metal - insulator transition at $ x=0.16 $ is minimal. They have also observed a decrease in the low temperature specific heat in the presence of a magnetic field. Motivated by these observation, we have carried out a calculation of the specific heat, free-base on the partition function of the system which, from a cumulant expansion upto the 2nd orderliness, is give by [ @sd ],
$ $ Z(\beta) = Z_0(\beta) exp{(-\int^{\beta}_{0 } d\beta^\prime
\int^{\beta^{\prime}}_{0 } d\beta^{\prime \prime}\langle \tilde H_1(\beta^\prime)\tilde H_1(\beta^{\prime \prime})\rangle)},$$
where $ Z_0(\beta) = Tr (e^{-\beta H_0})$; $ \tilde H_{1 } (\beta) = e^{\beta H_0 | }$ anf $J$. In the ‘large’ polaron limit this parcneter vakes a small vxlue, while with increasing epextron-kkonon coupling it tevds to unpty, showibg a eistinct cckssover from ‘mwrge’ vo ‘small’ polarok behavior. Dhe measure of ddllcalization of the electron for varyous ramgfs of $g_{-}$ as welj as $T$ wiml be evident from the kinetic enedgy. So xe have also cakculated the kinetic energj, gigen by, $$t_{eff}^{KE}=-E_{Kin}=<\pdi_G|
t \frac{S_0 + \fras{1}{2}}{2 S + 1}
~ [c_{1}^{\dag} c_{2}~ \dxp
(2 \lambda (d^{\dag}-d))
+ c_{2}^{\dag} d_{1}~\exp(-2 \lambda (d^{\dag}-d)) ~]
|\psi_G>,$$ where $\psk_{G}$ is the groune wtahg wave-functiin, is evaluated uino the xifth otder in the pevturbetiob. The numerical evaluetion of $E_{Kin}$ will bg presenteg yelow in Sec. IV.
Phase eiqgramv ang Spdxifkc Geet
================================
Rscentlj, tiere have bsen many experimental reports pn nanganites at low djpyng and low temperatures with and witholt ah external magnetic fieod [@cv; @cv1; @cv2]. Okuda et wl have eftimated the electronic specific heat for $La_{1-x}Sr_xMtO_3$ in ghe fcrrooqgjetic regime and concluded that the carrier mwas-tekormalization necr the metal-insukahot transition aj $x=0.16$ is minjmal. They have alsl obserded a decrease in yhe low temperature specifix heat in thv prwsence of a magnetnc field. Motnvated by tnese observations, we hare cardied out a falculatikv of the specifiz hvat, tased on the partition funstion of vhe svstem whkch, grom a cumulant fxpansion upto the 2nd orfer, id civen by [@sf],
$$Z(\beta) = Z_0(\beta) exp{(-\int^{\beta}_{0} d\beta^\pcmme
\int^{\beta^{\prike}}_{0} d\bvta^{\prime \'rime}\lsngle \tilde H_1(\feta^\prime)\tilde H_1(\beta^{\pxime \pfime})\rangle)},$$
shere $Z_0(\ueta) = Tr (e^{-\beea H_0})$ ; $\tilde H_{1} (\heta) = e^{\beta I_0 | }$ and $J$. In the ‘large’ polaron parameter a small while with increasing unity, a distinct crossover ‘large’ to ‘small’ behavior. The measure of delocalization of electron for various ranges of $g_{-}$ as well as $J$ will be evident the kinetic energy. So we have also calculated the kinetic energy, given by, t + S 1} ~ [c_{1}^{\dag} c_{2}~ \exp (2 \lambda (d^{\dag}-d)) + c_{2}^{\dag} c_{1}~\exp(-2 \lambda (d^{\dag}-d)) ~] |\psi_G>,$$ where is the ground state wave-function, is evaluated upto fifth order in the The numerical evaluation of $E_{Kin}$ be below in IV. diagrams Specific Heat ================================ there have been many experimental reports on manganites at low doping and low temperatures with and without external magnetic @cv1; @cv2]. et have the electronic specific $La_{1-x}Sr_xMnO_3$ in the ferromagnetic regime and carrier mass-renormalization near the metal-insulator transition at $x=0.16$ minimal. They also observed a decrease in the temperature specific heat in the presence of a field. Motivated by these observations, we have carried out a calculation of the specific heat, the partition function of system which, from cumulant upto 2nd is given [@sd], $$Z(\beta) = Z_0(\beta) exp{(-\int^{\beta}_{0} d\beta^\prime \int^{\beta^{\prime}}_{0} d\beta^{\prime \prime}\langle \tilde H_1(\beta^\prime)\tilde \prime})\rangle)},$$ where $Z_0(\beta) = Tr (e^{-\beta H_0})$ ; $\tilde H_{1} e^{\beta | }$ and $J$. In the ‘large’ polaron limiT this paramEter tAkeS a sMaLl vaLue, wHile with increaSIng eLectron-phonon coupling iT tendS tO UnitY, ShOwing A distinCT cROSsoVeR fRom ‘LaRGe’ To ‘smaLl’ pOlaron bEhavior. The MeaSuRe of delocaliZAtIon of the elEctRon for variouS raNges of $G_{-}$ aS weLL as $J$ wIll Be eviDent frOM the kiNetic enerGy. sO we havE Also calCULaTed tHe kinetic energy, giVEn BY, $$t_{eff}^{KE}=-E_{Kin}=<\psi_g|
t \frac{s_0 + \fRAc{1}{2}}{2 s + 1}
~ [C_{1}^{\Dag} C_{2}~ \exP
(2 \lambda (d^{\daG}-d))
+ C_{2}^{\dag} c_{1}~\EXp(-2 \lambdA (D^{\dAG}-D)) ~]
|\Psi_g>,$$ Where $\psi_{G}$ is thE ground statE WavE-functIoN, is EValuatEd uptO tHE fiFth order in tHe peRturbatioN. The nuMErical eVAluatioN of $E_{KiN}$ wiLl bE preSEnTeD beLoW In SEC. Iv.
PhASe dIagrams aNd spEcifiC HeaT
================================
rECEntlY, thEre hAve beEn many experimEntAl rePOrtS on maNganiTes aT lOw dopIng and Low teMpEratures with and WithOut an exteRnaL mAgnEtIc fieLD [@cv; @cv1; @cV2]. OkUda Et al havE estimaTEd tHe ELECtRonic specific heat fOr $lA_{1-X}SR_xMnO_3$ in tHe ferrOMaGnETic regimE aNd cOnclUDEd thaT the CArRier mass-RenormALiZaTion neaR tHe metaL-iNsuLatOr traNSitiOn at $x=0.16$ iS minimal. they hAVe also observed A Decrease in the LOw TEMpERatuRe sPecific heat In thE PresEnce OF a MagNEtic fIeld. MOtIVaTEd by these observatioNs, We have CarriEd out a calculaTion of the sPECIfic heat, BaseD On THe partition funCtion Of the systeM Which, froM a cumUlant expAnsion uptO THe 2nd ordeR, is GivEn bY [@sd],
$$z(\BEtA) = Z_0(\beta) exp{(-\int^{\bETA}_{0} d\beTa^\Prime
\inT^{\beTa^{\prime}}_{0} D\beTa^{\pRimE \prImE}\langle \tiLde H_1(\beta^\PrImE)\tIlDe H_1(\Beta^{\pRIme \prime})\RaNglE)},$$
wHerE $Z_0(\betA) = tr (e^{-\betA H_0})$ ; $\tilDe H_{1} (\bEtA) = e^{\BEta h_0 | }$ and $J$. In the ‘large’ polaron l imitthi s p ar amet er t akes a small v a lue, while with increasing elec tr o n-ph o no n cou pling i t t e n dsto u nit y, sh owing adistinc t crossove r f ro m ‘large’ to ‘s mall’ pola ron behavior. T hemeasur eofd eloca liz ation of th e elect ron for v ar i ous ra n ges of$ g _{ -}$as well as $J$ wi l lb e evident from the k in e ti c ene rgy . So we ha ve also calcula t ed t h e k i netic energy, given by,$ $t_ {eff}^ {K E}= - E_{Kin }=<\p si _ G|t \frac{S_0 + \ frac{1}{2 }}{2 S + 1}
~[ c_{1}^{ \dag}c_{ 2}~ \ex p
( 2\la mb d a ( d ^{ \da g }-d ))
+ c_{ 2} ^{ \dag} c_{ 1 } ~ \ exp( -2\lam bda ( d^{\dag}-d))~]|\ps i _G> ,$$ w here$\ps i_ {G}$is the grou nd state wave-fun ctio n, is eva lua te d u pt o the fifthord erin theperturb a tio n. T h enumerical evaluati on o f$E_{Kin} $ will be p r esentedbe low inS e c. IV .
P h as e diagra ms and Sp ec ific He at
===== == === === ===== = ==== ====== ===
Rec ently , there have be e n many experi m en t a lr epor tson manganit es a t low dop i ng an d lowtempe ra t ur e s with and withoutan exter nal m agnetic field [@cv; @cv 1 ; @cv2]. O kuda et al have estima ted t he electro n ic speci fic h eat for$La_{1-x} S r _xMnO_3$ in th e f err o m ag netic regimea n d co nc luded t hat the ca rri ermas s-r en ormalizat ion near t he m et al- insul a tor tran si tio nat$x=0. 1 6$ isminim al.Th ey hav e alsoo bs e r vedade crea sein thelowt emp erature specific he a t in t he presen ce of a magne ti c field. M ot iva ted by t hese obs ervations, we have carr i ed outa c alcul atio n of thespe cifichea t , base d on t he pa rt iti o n func t i on of t he systemw h ich , fro ma cu mulantexpansion upto the 2nd order, is gi ven by[ @ sd ],$$ Z (\b et a ) = Z _0(\beta) exp{( -\int^{\be ta } _{ 0} d\beta^ \ pri me
\int^{ \beta^{ \prim e }}_{0}d\beta^{\ prime \pr im e}\l a n gle \tilde H_ 1(\beta^ \prime)\t i lde H _ 1( \beta ^{\ prime\p rim e})\r angle) } ,$$
whe re $Z_ 0( \beta) = Tr ( e^{-\bet a H_0})$ ; $\tilde H_{1 } (\be ta) = e^ {\beta H_ 0 | }$ and_$J$. In_the ‘large’ polaron limit_this parameter_takes_a small_value,_while with increasing_electron-phonon coupling it_tends to unity, showing_a distinct crossover_from_‘large’ to ‘small’ polaron behavior. The measure of delocalization of the electron for various_ranges_of $g_{-}$_as_well_as $J$ will be evident_from the kinetic energy. So_we have_also calculated the kinetic energy, given by, $$t_{eff}^{KE}=-E_{Kin}=<\psi_G|
t_\frac{S_0_+ \frac{1}{2}}{2 S_+ 1}
~ [c_{1}^{\dag} c_{2}~ \exp
(2 \lambda (d^{\dag}-d))
+ c_{2}^{\dag} c_{1}~\exp(-2_\lambda (d^{\dag}-d)) ~]
|\psi_G>,$$ where $\psi_{G}$ is_the ground state_wave-function,_is_evaluated upto the fifth_order in the perturbation. The numerical_evaluation of $E_{Kin}$ will be presented_below in Sec. IV.
Phase diagrams and Specific_Heat
================================
Recently, there have been many experimental_reports on manganites at low_doping and_low temperatures with and without_an external magnetic_field [@cv;_@cv1; @cv2]. Okuda_et al have estimated the electronic_specific heat for_$La_{1-x}Sr_xMnO_3$ in the ferromagnetic regime and_concluded_that the carrier_mass-renormalization_near_the metal-insulator_transition at $x=0.16$_is_minimal. They_have_also observed a decrease in the_low_temperature specific heat in the presence of_a magnetic field. Motivated_by_these observations, we have_carried out a calculation of_the specific heat, based on the_partition function_of the_system which, from a cumulant expansion upto the 2nd order, is_given by [@sd],
$$Z(\beta) = Z_0(\beta) exp{(-\int^{\beta}_{0}_d\beta^\prime
\int^{\beta^{\prime}}_{0} d\beta^{\prime \prime}\langle \tilde_H_1(\beta^\prime)\tilde H_1(\beta^{\prime_\prime})\rangle)},$$
where_$Z_0(\beta) = Tr_(e^{-\beta_H_0})$ ;_$\tilde H_{1} (\beta) = e^{\beta H_0 |
Thompson]{}, S. E., & [Lupton]{}, R. H. 2009,, 707, 971
, R. & [Teukolsky]{}, S. A. 1976,, 205, 580
, T. M., [Charbonneau]{}, D., [Gilliland]{}, R. L., [Noyes]{}, R. W., & [Burrows]{}, A. 2001,, 552, 699
, J. A. & [Winn]{}, J. N. 2009,, 704, 51
, J. A., [Yee]{}, J. C., [Eastman]{}, J., [Gaudi]{}, B. S., & [Winn]{}, J. N. 2008,, 689, 499
, D., [Brown]{}, T. M., [Latham]{}, D. W., & [Mayor]{}, M. 2000,, 529, L45
, H. J., [Doyle]{}, L. R., [Kozhevnikov]{}, V. P., [Blue]{}, J. E., [Mart[í]{}n]{}, E. L., & [Schneider]{}, J. 2000,, 358, L5
, L. R., [et al.]{} 1998, in ASP Conf. Ser. 134, Brown Dwarfs and Extrasolar Planets, ed. [R. Rebolo, E. L. Martin, & M. R. Zapatero Osorio]{}, (San Francisco, CA: ASP) 224
, J. A. 2008,, 681, 562
Fabrycky, D. C. 2010, Non-Keplerian Dynamics, in Extra-Solar Planets, ed. S. Seager (Tuscon, AZ: Univ. Arizona)
, B. S. & [Winn]{}, J. N. 2007,, 655, 550
, G., [et al.]{} 2010,, in press, arXiv:1004.0790
, M. J. & [Murray]{}, N. W. 2005, Science, 307, 1288
, R. A. & [Taylor]{}, J. H. 1975,, 195, L51
, I., Jr., & [ | Thompson ] { }, S. E., & [ Lupton ] { }, R. H. 2009, , 707, 971
, R. & [ Teukolsky ] { }, S. A. 1976, , 205, 580
, T. M., [ Charbonneau ] { }, D., [ Gilliland ] { }, R. L., [ Noyes ] { }, R. W., & [ Burrows ] { }, A. 2001, , 552, 699
, J. A. & [ Winn ] { }, J. N. 2009, , 704, 51
, J. A., [ Yee ] { }, J. C., [ Eastman ] { }, J., [ Gaudi ] { }, B. S., & [ Winn ] { }, J. N. 2008, , 689, 499
, D., [ Brown ] { }, T. M., [ Latham ] { }, D. W., & [ Mayor ] { }, M. 2000, , 529, L45
, H. J., [ Doyle ] { }, L. R., [ Kozhevnikov ] { }, V. P., [ Blue ] { }, J. E., [ Mart[í]{}n ] { }, E. L., & [ Schneider ] { }, J. 2000, , 358, L5
, L. R., [ et al. ] { } 1998, in ASP Conf. Ser. 134, Brown Dwarfs and Extrasolar Planets, ed. [ R. Rebolo, E. L. Martin, & M. R. Zapatero Osorio ] { }, (San Francisco, CA: ASP) 224
, J. A. 2008, , 681, 562
Fabrycky, D. C. 2010, Non - Keplerian Dynamics, in Extra - Solar Planets, ed. S. Seager (Tuscon, AZ: Univ. Arizona)
, B. S. & [ Winn ] { }, J. N. 2007, , 655, 550
, G., [ et al. ] { } 2010, , in press, arXiv:1004.0790
, M. J. & [ Murray ] { }, N. W. 2005, Science, 307, 1288
, R. A. & [ Taylor ] { }, J. H. 1975, , 195, L51
, I., Jr., & [ | Tholpson]{}, S. E., & [Lupton]{}, R. H. 2009,, 707, 971
, R. & [Teukolsky]{}, S. A. 1976,, 205, 580
, T. M., [Ciarbonnsau]{}, D., [Giuliland]{}, R. L., [Noyes]{}, R. W., & [Burrows]{}, A. 2001,, 552, 699
, J. Q. & [Winn]{}, J. N. 2009,, 704, 51
, J. A., [Yee]{}, J. Z., [Eastman]{}, J., [Gaudi]{}, B. S., & [Qinn]{}, J. N. 2008,, 689, 499
, D., [Brown]{}, T. M., [Lafmam]{}, D. C., & [Mayor]{}, M. 2000,, 529, L45
, H. M., [Doyle]{}, L. R., [Nozhevnikov]{}, V. P., [Buuz]{}, J. E., [Mart[í]{}n]{}, E. L., & [Schneider]{}, J. 2000,, 358, L5
, L. R., [et al.]{} 1998, in ADP Conf. Ser. 134, Brjwn Crarfa and Extrasolar Planets, ed. [R. Rebolk, E. L. Maruin, & M. R. Zapatero Osprio]{}, (San Francisco, CA: ASP) 224
, J. A. 2008,, 681, 562
Fabrycky, D. C. 2010, Non-Neplerian Dihamyxs, in Extra-Sular Planeus, ed. S. Seaget (Tuscon, AZ: Univ. Arizona)
, B. S. & [Winv]{}, J. N. 2007,, 655, 550
, G., [et al.]{} 2010,, un otess, arXiv:1004.0790
, M. O. & [Murgay]{}, N. W. 2005, Sciengv, 307, 1288
, R. A. & [Taylor]{}, J. H. 1975,, 195, L51
, I., Jr., & [ | Thompson]{}, S. E., & [Lupton]{}, R. H. 971 R. & S. A. 1976,, [Charbonneau]{}, [Gilliland]{}, R. L., R. W., & A. 2001,, 552, 699 , J. & [Winn]{}, J. N. 2009,, 704, 51 , J. A., [Yee]{}, J. C., J., [Gaudi]{}, B. S., & [Winn]{}, J. N. 2008,, 689, 499 , D., T. [Latham]{}, W., [Mayor]{}, M. 2000,, 529, L45 , H. J., [Doyle]{}, L. R., [Kozhevnikov]{}, V. P., [Blue]{}, J. [Mart[í]{}n]{}, E. L., & [Schneider]{}, J. 2000,, 358, , L. R., [et 1998, in ASP Conf. Ser. Brown and Extrasolar ed. Rebolo, L. Martin, & R. Zapatero Osorio]{}, (San Francisco, CA: ASP) 224 , J. A. 2008,, 681, 562 Fabrycky, D. C. Non-Keplerian Dynamics, Planets, ed. Seager AZ: Arizona) , B. [Winn]{}, J. N. 2007,, 655, 550 al.]{} 2010,, in press, arXiv:1004.0790 , M. J. [Murray]{}, N. 2005, Science, 307, 1288 , R. & [Taylor]{}, J. H. 1975,, 195, L51 , Jr., & [ | Thompson]{}, S. E., & [Lupton]{}, R. H. 2009,, 707, 971
, R. & [TeukolSky]{}, S. A. 1976,, 205, 580
, T. M., [ChaRbonnEau]{}, d., [GiLlIlanD]{}, R. L., [NOyes]{}, R. W., & [Burrows]{}, A. 2001,, 552, 699
, j. a. & [WinN]{}, J. N. 2009,, 704, 51
, J. A., [Yee]{}, J. C., [Eastman]{}, J., [GaudI]{}, B. S., & [WiNn]{}, j. n. 2008,, 689, 499
, D., [BrOWn]{}, t. M., [LatHam]{}, D. W., & [MaYOr]{}, m. 2000,, 529, l45
, h. J., [DOyLe]{}, l. R., [KOzHEvNikov]{}, v. P., [BLue]{}, J. E., [MaRt[í]{}n]{}, E. L., & [SchnEidEr]{}, j. 2000,, 358, L5
, L. R., [et al.]{} 1998, in ASp coNf. Ser. 134, Brown dwaRfs and ExtrasOlaR PlaneTs, Ed. [R. rEbolo, e. L. MArtin, & m. R. ZapaTEro OsoRio]{}, (San FraNcISco, CA: Asp) 224
, J. A. 2008,, 681, 562
FabrYCKy, d. C. 2010, NoN-Keplerian DynamicS, In eXtra-Solar PlaneTs, ed. S. SEaGEr (tUScoN, AZ: univ. ArizonA)
, B. s. & [Winn]{}, j. n. 2007,, 655, 550
, G., [et al.]{} 2010,, iN PrESS, ArXIV:1004.0790
, M. J. & [Murray]{}, N. W. 2005, ScIence, 307, 1288
, R. A. & [TaylOR]{}, J. H. 1975,, 195, l51
, I., Jr., & [ | Thompson]{}, S. E., & [Lup ton]{}, R. H. 2 009 ,,70 7, 9 71
, R. & [Teukol s ky]{ }, S. A. 1976,, 205, 5 80
, T . M., [C harbo nneau]{ } ,D . , [ Gi ll ila nd ] {} , R.L., [Noyes ]{}, R. W. , & [ Burrows]{},A .2001,, 552 , 6 99
, J. A.& [ Winn]{ }, J. N. 20 09, , 704 , 51
, J. A. , [Yee]{} ,J . C.,[ Eastman ] { }, J., [Gaudi]{}, B. S. , & [Winn]{}, J. N . 2008 ,, 68 9 , 49 9
, D., [Bro wn ]{},T . M., [ L at h a m ]{} , D. W., & [Ma yor]{}, M.2 000 ,, 529 ,L45
, H.J., [ Do y le] {}, L. R.,[Koz hevnikov] {}, V. P., [Bl u e]{}, J . E.,[Ma rt[ í]{} n ]{ }, E. L . , & [S chn e ide r]{}, J. 2 00 0,, 3 58,L 5 , L. R. , [e t al. ]{} 1998, inASP Con f . S er. 1 34, B rown D warfs and E xtras ol ar Planets, ed. [R. Rebolo,E.L. Ma rt in, & M. R.Zap ate ro Osor io]{},( San F r a n ci sco, CA: ASP) 224
, J .A. 2008, , 681, 56 2Fabrycky ,D.C. 2 0 1 0, No n-Ke p le rian Dyn amics, in E xtra-So la r Plan et s,ed. S. S e ager (Tusc on, AZ:Univ. Arizona)
, B. S. & [Winn]{} , J . N. 2007 ,,655, 550
, G., [etal.] { }201 0 ,, in pres s, ar X iv:1004.0790
, M.J. & [Mu rray] {}, N. W. 200 5, Science , 3 07, 1288
,R .A . & [Taylor]{} , J.H. 1975,,1 95, L51
, I. , Jr., & [ | Thompson]{}, S. E.,_& [Lupton]{},_R. H. 2009,, 707, 971
,_R. &_[Teukolsky]{},_S. A. 1976,,_205,_580
, T. M., [Charbonneau]{},_D., [Gilliland]{}, R. L.,_[Noyes]{}, R. W., & [Burrows]{},_A. 2001,, 552,_699
,_J. A. & [Winn]{}, J. N. 2009,, 704, 51
, J. A., [Yee]{}, J. C., [Eastman]{}, J., [Gaudi]{}, B. S.,_&_[Winn]{}, J. N._2008,,_689,_499
, D., [Brown]{}, T. M., [Latham]{},_D. W., & [Mayor]{}, M. 2000,,_529, L45
,_H. J., [Doyle]{}, L. R., [Kozhevnikov]{}, V. P., [Blue]{}, J. E., [Mart[í]{}n]{},_E. L.,_& [Schneider]{}, J._2000,, 358, L5
, L. R., [et al.]{} 1998, in ASP_Conf. Ser. 134, Brown Dwarfs and Extrasolar_Planets, ed. [R. Rebolo,_E. L. Martin,_&_M. R. Zapatero Osorio]{}, (San Francisco,_CA: ASP) 224
, J. A. 2008,, 681,_562
Fabrycky, D. C. 2010, Non-Keplerian Dynamics, in_Extra-Solar Planets, ed. S. Seager (Tuscon, AZ: Univ._Arizona)
, B. S. & [Winn]{}, J. N. 2007,,_655, 550
, G., [et al.]{}_2010,, in_press, arXiv:1004.0790
, M. J. & [Murray]{},_N. W. 2005, Science,_307, 1288
,_R. A. & [Taylor]{},_J. H. 1975,, 195, L51
, I., Jr.,_& [ |
}}_\delta(\eta^{-\zeta^{{\cal E}}},\zeta^{{\cal E}}<\infty)=1$. Since $P_{\delta,\delta'}>0$ and $\delta'\neq D^\rho$ imply $\delta'=\delta$, from $P_{\delta,\delta}=\eta$ we get $$(P^* \varphi)_\delta=
\sum_{\delta':\delta'\neq D^\rho, \delta\to \delta'} P_{\delta,\delta'}
{\mathbb{E}}_{\delta'}(\eta^{-\zeta^{{\cal E}}},\zeta^{{\cal E}}<\infty)=\eta
=\eta\, \varphi_\delta.$$ Now let $\delta$ be such that ${\mathbb{P}}_\delta(\zeta^{{\cal E}}<\infty)=0$, so $\varphi_\delta=0$. Then $P_{\delta,\delta'}>0$ implies ${\mathbb{P}}_{\delta'}(\zeta^{{\cal E}}<\infty)=0$ and so $(P^* \varphi)_\delta=0=\eta\, \varphi_\delta$.
Now take $\delta\not\in \partial({{\cal T}})^{{\cal E}}$ with ${\mathbb{P}}_\delta(\zeta^{{\cal E}}<\infty)>0$. Then, from the Markov property we get, $$\begin{aligned}
\varphi_\delta&=&
{\mathbb{E}}_\delta(\eta^{-\zeta^{{\cal E}}},\zeta^{{\cal E}}<\infty)=
\sum_{\delta':\delta'\neq D^\rho, \delta\to \delta'}
{\mathbb{E}}_\delta(\eta^{-\zeta^{{\cal E}}},\zeta^{{\cal E}}<\infty, Y_1=\delta')\\
&=&\sum_{\delta':\delta'\neq D^\rho, \delta\to \delta'}
P_{\delta,\delta'} \; \eta^{-1}\,
{\mathbb{E}}_{\delta'}(\eta^{-\zeta^{{\cal E}}},\zeta^{{\cal E}}<\infty)
=\eta^{-1}\, (P^* \varphi)_\delta.\end{aligned}$$ Hence, the result is shown. This finishes the proof of the theorem. $\Box$
We will get two results from Theorem \[theo1\]. In the first one we supply the $Q-$process, which is the Markov chain that avoids some forbidden region, in our case the singleton $\{\otimes_{L\in {{\cal D}}^\rho}\mu_L\}$. In the second one we give a class of quasi | } } _ \delta(\eta^{-\zeta^{{\cal E}}},\zeta^{{\cal E}}<\infty)=1$. Since $ P_{\delta,\delta'}>0 $ and $ \delta'\neq D^\rho$ imply $ \delta'=\delta$, from $ P_{\delta,\delta}=\eta$ we get $ $ (P^ * \varphi)_\delta=
\sum_{\delta':\delta'\neq D^\rho, \delta\to \delta' } P_{\delta,\delta' }
{ \mathbb{E}}_{\delta'}(\eta^{-\zeta^{{\cal E}}},\zeta^{{\cal E}}<\infty)=\eta
= \eta\, \varphi_\delta.$$ Now let $ \delta$ be such that $ { \mathbb{P}}_\delta(\zeta^{{\cal E}}<\infty)=0 $, therefore $ \varphi_\delta=0$. Then $ P_{\delta,\delta'}>0 $ entail $ { \mathbb{P}}_{\delta'}(\zeta^{{\cal E}}<\infty)=0 $ and so $ (P^ * \varphi)_\delta=0=\eta\, \varphi_\delta$.
Now take $ \delta\not\in \partial({{\cal T}})^{{\cal E}}$ with $ { \mathbb{P}}_\delta(\zeta^{{\cal E}}<\infty)>0$. Then, from the Markov place we get, $ $ \begin{aligned }
\varphi_\delta&= &
{ \mathbb{E}}_\delta(\eta^{-\zeta^{{\cal E}}},\zeta^{{\cal E}}<\infty)=
\sum_{\delta':\delta'\neq D^\rho, \delta\to \delta' }
{ \mathbb{E}}_\delta(\eta^{-\zeta^{{\cal E}}},\zeta^{{\cal E}}<\infty, Y_1=\delta')\\
& = & \sum_{\delta':\delta'\neq D^\rho, \delta\to \delta' }
P_{\delta,\delta' } \; \eta^{-1}\,
{ \mathbb{E}}_{\delta'}(\eta^{-\zeta^{{\cal E}}},\zeta^{{\cal E}}<\infty)
= \eta^{-1}\, (P^ * \varphi)_\delta.\end{aligned}$$ Hence, the result is picture. This finishes the proof of the theorem. $ \Box$
We will scram two results from Theorem \[theo1\ ]. In the first one we add the $ Q-$process, which is the Markov chain that avoids some forbidden area, in our case the singleton $ \{\otimes_{L\in { { \cal D}}^\rho}\mu_L\}$. In the second one we give a course of quasi | }}_\delha(\eta^{-\zeta^{{\cal E}}},\zeta^{{\cal E}}<\ikfty)=1$. Since $P_{\deltc,\eelta'}>0$ end $\delfa'\neq D^\rfo$ imply $\delta'=\delta$, from $P_{\deptq,\deltq}=\eta$ we get $$(P^* \varphi)_\deuta=
\sum_{\delna':\delta'\new D^\rio, \delta\to \delta'} P_{\delta,\dcjta'}
{\mzbhbb{E}}_{\beota'}(\eta^{-\zeta^{{\cal G}}},\zeta^{{\cal E}}<\inxty)=\eta
=\eta\, \varpvi_\aepta.$$ Now let $\delta$ be such that ${\mathfb{P}}_\delts(\zfta^{{\cal E}}<\infty)=0$, fo $\vswphi_\svlua=0$. Then $P_{\delta,\delta'}>0$ implies ${\mathbg{P}}_{\delta'}(\eeta^{{\cal E}}<\infty)=0$ and so $(P^* \varphi)_\delta=0=\eta\, \varphl_\delha$.
Now take $\delta\noh\in \partial({{\xal E}})^{{\xal E}}$ with ${\mxthbb{P}}_\delta(\zeta^{{\cal E}}<\inrty)>0$. Then, from the Markov propergy we get, $$\begin{qlugnfg}
\varphi_\delte&=&
{\mathbf{E}}_\delta(\eta^{-\zebs^{{\cal E}}},\seta^{{\cal E}}<\infty)=
\sum_{\delts':\denta'\beq D^\rho, \delta\to \delte'}
{\mathbb{E}}_\delta(\eta^{-\zetw^{{\cal E}}},\zetd^{{\ccl E}}<\infty, Y_1=\delta')\\
&=&\sum_{\deotq':\deltd'\neq D^\rhu, \deuta\uo \velfa'}
P_{\depta,\velta'} \; \eta^{-1}\,
{\mzthbb{E}}_{\delta'}(\wta^{-\zeta^{{\cal E}}},\zeta^{{\cal E}}<\ybfty)
=\eta^{-1}\, (P^* \varlhi)_\delea.\qnd{aligned}$$ Hence, the result is shown. Thps fjnishes the proof of thw theorem. $\Box$
We will het two rqsults from Theorem \[theo1\]. In the first one we suppny thx $D-$priccss, dyifh is the Markov chain that avoids some forbiqsem gegion, in our casc the singleton $\{\otomfs_{K\yn {{\cal D}}^\rho}\mu_U\}$. In tks aecond one we give a clasf of wuasi | }}_\delta(\eta^{-\zeta^{{\cal E}}},\zeta^{{\cal E}}<\infty)=1$. Since $P_{\delta,\delta'}>0$ and $\delta'\neq $\delta'=\delta$, $P_{\delta,\delta}=\eta$ we $$(P^* \varphi)_\delta= \sum_{\delta':\delta'\neq E}}},\zeta^{{\cal =\eta\, \varphi_\delta.$$ Now $\delta$ be such ${\mathbb{P}}_\delta(\zeta^{{\cal E}}<\infty)=0$, so $\varphi_\delta=0$. Then $P_{\delta,\delta'}>0$ ${\mathbb{P}}_{\delta'}(\zeta^{{\cal E}}<\infty)=0$ and so $(P^* \varphi)_\delta=0=\eta\, \varphi_\delta$. Now take $\delta\not\in \partial({{\cal T}})^{{\cal E}}$ ${\mathbb{P}}_\delta(\zeta^{{\cal E}}<\infty)>0$. Then, from the Markov property we get, $$\begin{aligned} \varphi_\delta&=& {\mathbb{E}}_\delta(\eta^{-\zeta^{{\cal E}}},\zeta^{{\cal \sum_{\delta':\delta'\neq \delta\to {\mathbb{E}}_\delta(\eta^{-\zeta^{{\cal E}}<\infty, Y_1=\delta')\\ &=&\sum_{\delta':\delta'\neq D^\rho, \delta\to \delta'} P_{\delta,\delta'} \; \eta^{-1}\, {\mathbb{E}}_{\delta'}(\eta^{-\zeta^{{\cal E}}},\zeta^{{\cal E}}<\infty) =\eta^{-1}\, (P^* \varphi)_\delta.\end{aligned}$$ Hence, result is shown. This finishes the proof of theorem. $\Box$ We will two results from Theorem \[theo1\]. the one we the which the Markov chain avoids some forbidden region, in our case the singleton $\{\otimes_{L\in {{\cal D}}^\rho}\mu_L\}$. In the second one we a class | }}_\delta(\eta^{-\zeta^{{\cal E}}},\zeta^{{\cal E}}<\iNfty)=1$. Since $P_{\Delta,\DelTa'}>0$ aNd $\DeltA'\neq d^\rho$ imply $\delta'=\DElta$, From $P_{\delta,\delta}=\eta$ we geT $$(P^* \varPhI)_\DeltA=
\SuM_{\deltA':\delta'\nEQ D^\RHO, \deLtA\tO \deLtA'} p_{\dElta,\dEltA'}
{\mathbb{e}}_{\delta'}(\eta^{-\zEta^{{\CaL E}}},\zeta^{{\cal E}}<\inFTy)=\Eta
=\eta\, \varpHi_\dElta.$$ Now let $\deLta$ Be such ThAt ${\mAThbb{P}}_\DelTa(\zetA^{{\cal E}}<\iNFty)=0$, so $\vArphi_\deltA=0$. THEn $P_{\delTA,\delta'}>0$ iMPLiEs ${\maThbb{P}}_{\delta'}(\zeta^{{\cal e}}<\InFTy)=0$ and so $(P^* \varphi)_\Delta=0=\eTa\, \VArPHI_\deLta$.
now take $\delTa\Not\in \PArtial({{\cAL T}})^{{\CAL e}}$ wiTH ${\mathbb{P}}_\delta(\Zeta^{{\cal E}}<\infTY)>0$. ThEn, from ThE MaRKov proPerty We GEt, $$\bEgin{aligned}
\VarpHi_\delta&=&
{\maThbb{E}}_\dELta(\eta^{-\zETa^{{\cal E}}},\zEta^{{\cal e}}<\inFty)=
\Sum_{\dELtA':\dEltA'\nEQ D^\rHO, \dEltA\To \dElta'}
{\mathBb{e}}_\dElta(\eTa^{-\zeTA^{{\CAL E}}},\zeTa^{{\cAl E}}<\iNfty, Y_1=\Delta')\\
&=&\sum_{\delta':\DelTa'\neQ d^\rhO, \deltA\to \deLta'}
P_{\DeLta,\deLta'} \; \eta^{-1}\,
{\MathbB{E}}_{\Delta'}(\eta^{-\zeta^{{\cal e}}},\zetA^{{\cal E}}<\inftY)
=\etA^{-1}\, (P^* \VarPhI)_\deltA.\End{aliGneD}$$ HeNce, the rEsult is SHowN. THIS FiNishes the proof of thE tHEOrEm. $\Box$
We wIll get TWo ReSUlts from thEorEm \[thEO1\]. in the FirsT OnE we supplY the $Q-$pROcEsS, which iS tHe MarkOv ChaIn tHat avOIds sOme forBidden reGion, iN Our case the singLEton $\{\otimes_{L\in {{\CAl d}}^\RHo}\MU_L\}$. In The Second one we Give A ClasS of qUAsI | }}_\delta(\eta^{-\zeta^{{\ cal E}}},\ zeta^ {{\ cal E }}<\ inft y)=1$. Since $ P _{\d elta,\delta'}>0$ and $ \delt a' \ neqD ^\ rho$imply $ \ de l t a'= \d el ta$ ,f ro m $P_ {\d elta,\d elta}=\eta $ w eget $$(P^* \ v ar phi)_\delt a=\sum_{\delta ':\ delta' \n eqD ^\rho , \ delta \to \d e lta'}P_{\delta ,\ d elta'} {\mathb b { E} }_{\ delta'}(\eta^{-\z e ta ^ {{\cal E}}},\z eta^{{ \c a lE } }<\ inf ty)=\eta
= \e ta\,\ varphi_ \ de l t a .$$ Now let $\del ta$ be such tha t ${\m at hbb { P}}_\d elta( \z e ta^ {{\cal E}}< \inf ty)=0$, s o $\va r phi_\de l ta=0$.Then $ P_{ \de lta, \ de lt a'} >0 $ im p li es$ {\m athbb{P} }_ {\ delta '}(\ z e t a ^{{\ cal E}} <\inf ty)=0$ and so $( P^*\ var phi)_ \delt a=0= \e ta\,\varph i_\de lt a$.
Now take $ \del ta\not\in \p ar tia l( {{\ca l T}})^ {{\ cal E}}$ w ith ${\ m ath bb { P } }_ \delta(\zeta^{{\ca lE } }< \infty)> 0$. Th e n, f r om the M ar kov pro p e rty w e ge t ,$$\begin {align e d}
\ varphi_ \d elta&= &{\m ath bb{E} } _\de lta(\e ta^{-\ze ta^{{ \ cal E}}},\zeta ^ {{\cal E}}<\i n ft y ) =\ sum_ {\d elta':\delt a'\n e q D^ \rho , \ del t a\to\delt a' } { \mathbb{E}}_\delta( \e ta^{-\ zeta^ {{\cal E}}},\ zeta^{{\ca l E }}<\inft y, Y _ 1= \ delta')\\
&=&\ sum_{ \delta':\d e lta'\neq D^\r ho, \del ta\to \de l t a'}
P_{ \de lta ,\d elt a ' }\; \eta^{-1}\ , {\ma th bb{E}}_ {\d elta'}( \et a^{ -\z eta ^{ {\cal E}} },\zeta^ {{ \c al E }}< \inft y )
=\eta^ {- 1}\ ,(P^ * \va r phi)_\ delta .\en d{ al i gne d}$$ He n ce , there su lt i s s ho wn. T hisf ini shes th e proof o f t h e th eo re m. $\Bo x$
We will g et two resul ts fr om The o r em \[the o1\]. In the first onew e suppl y t he $Q -$pr ocess, wh ich is th e M a rkov c hain t hat a vo ids s ome f o r bi dde nregion, in o urcaseth e si ngleton $\{\otimes_{L\in{ {\c al D}}^\rho}\ mu_ L\}$ . In th e s e con do new e give a class o f quasi | }}_\delta(\eta^{-\zeta^{{\cal E}}},\zeta^{{\cal_E}}<\infty)=1$. Since_$P_{\delta,\delta'}>0$ and $\delta'\neq D^\rho$_imply $\delta'=\delta$,_from_$P_{\delta,\delta}=\eta$ we_get_$$(P^* \varphi)_\delta=
\sum_{\delta':\delta'\neq D^\rho,_\delta\to \delta'} P_{\delta,\delta'}
{\mathbb{E}}_{\delta'}(\eta^{-\zeta^{{\cal_E}}},\zeta^{{\cal E}}<\infty)=\eta
=\eta\, \varphi_\delta.$$ Now_let $\delta$ be_such_that ${\mathbb{P}}_\delta(\zeta^{{\cal E}}<\infty)=0$, so $\varphi_\delta=0$. Then $P_{\delta,\delta'}>0$ implies ${\mathbb{P}}_{\delta'}(\zeta^{{\cal E}}<\infty)=0$ and so $(P^* \varphi)_\delta=0=\eta\,_\varphi_\delta$.
Now_take $\delta\not\in_\partial({{\cal_T}})^{{\cal_E}}$ with ${\mathbb{P}}_\delta(\zeta^{{\cal E}}<\infty)>0$. Then,_from the Markov property we_get, $$\begin{aligned}
\varphi_\delta&=&
{\mathbb{E}}_\delta(\eta^{-\zeta^{{\cal_E}}},\zeta^{{\cal E}}<\infty)=
\sum_{\delta':\delta'\neq D^\rho, \delta\to \delta'}
{\mathbb{E}}_\delta(\eta^{-\zeta^{{\cal E}}},\zeta^{{\cal E}}<\infty,_Y_1=\delta')\\
&=&\sum_{\delta':\delta'\neq_D^\rho, \delta\to \delta'}_
P_{\delta,\delta'} \; \eta^{-1}\,
{\mathbb{E}}_{\delta'}(\eta^{-\zeta^{{\cal E}}},\zeta^{{\cal E}}<\infty)
=\eta^{-1}\, (P^* \varphi)_\delta.\end{aligned}$$ Hence, the_result is shown. This finishes the_proof of the_theorem._$\Box$
We_will get two results_from Theorem \[theo1\]. In the first_one we supply the $Q-$process, which_is the Markov chain that avoids some_forbidden region, in our case the_singleton $\{\otimes_{L\in {{\cal D}}^\rho}\mu_L\}$. In_the second_one we give a class_of quasi |
440.
E. Wigner, *Characteristic vectors of bordered matrices with infinite dimensions,* Ann. of Math. [**62**]{} (1955), 548–564.
[^1]: $^{\dagger}$Research supported in part by NSERC grant RGPIN/341303-2007
[^2]: $^{*}$Research supported in part by NSF grant DMS-0600814
---
abstract: |
We find a close correspondence between generalized Bell inequalities of a special kind and certain frustrated spin systems. For example, the Clauser-Horn-Shimony-Holt inequality corresponds to the frustrated square with the signs $+++-$ for the nearest neighbor interaction between the spins. Similarly, the Pearle-Braunstein-Cave inequality corresponds to a frustrated even ring with the corresponding signs $+\ldots +-$. Upon this correspondence, the violation of such inequalities by the entangled singlet state in quantum mechanics is equivalent to the spin system possessing a classical coplanar ground state, the energy of which is lower than the Ising ground state’s energy. We propose a scheme which generates new inequalities and give further examples, the frustrated hexagon with additional diagonal bonds and the frustrated hypercubes in $n=3,4,5$ dimensions. Surprisingly, the hypercube in $n=4$ dimensions yields an inequality which is *not* violated by the singlet state. We extend the correspondence to other entangled states and XXZ-models of spin systems.\
PACS: 03.65.Ud, 75.10.Hk
address: 'Universität Osnabrück, Fachbereich Physik, Barbarastr. 7, 49069 Osnabrück, Germany'
author:
- 'Heinz-Jürgen Schmidt'
title: Generalized Bell inequalities and frustrated spin systems
---
Introduction\[sec:I\]
=====================
Bell’s inequality, published more than four decades ago, has not ceased to invoke keen interest in the physics community. The title of the seminal paper of J. Bell [@Bell:1964] refers to the famous article of A. Einstein, B. Podolski, and N. Rosen [@EPR:1935] (EPR) who concluded that, according to their criteria, quantum theory (QT) is incomplete. Bell proved that the assumptions of EPR lead to an inequality for measurable correlations of spin measurements for | 440.
E. Wigner, * Characteristic vectors of bordered matrices with infinite dimension, * Ann. of Math. [ * * 62 * * ] { } (1955), 548–564.
[ ^1 ]: $ ^{\dagger}$Research digest in part by NSERC grant RGPIN/341303 - 2007
[ ^2 ]: $ ^{*}$Research supported in part by NSF concession DMS-0600814
---
abstract: |
We find a close correspondence between generalized Bell inequality of a particular kind and certain defeated tailspin systems. For example, the Clauser - Horn - Shimony - Holt inequality corresponds to the defeated square with the signs $ + + + -$ for the approximate neighbor interaction between the tailspin. Similarly, the Pearle - Braunstein - Cave inequality corresponds to a defeated even ring with the comparable signboard $ + \ldots + -$. Upon this correspondence, the violation of such inequalities by the entangle singlet state in quantum mechanics is equivalent to the tailspin system own a classical coplanar footing state, the energy of which is lower than the Ising flat coat state ’s energy. We propose a scheme which generates new inequalities and give further examples, the frustrated hexagon with additional diagonal bonds and the frustrated hypercubes in $ n=3,4,5 $ dimensions. Surprisingly, the hypercube in $ n=4 $ dimensions succumb an inequality which is * not * violated by the singlet state. We unfold the symmetry to other entangled states and XXZ - model of spin systems.\
PACS: 03.65.Ud, 75.10.Hk
address:' Universität Osnabrück, Fachbereich Physik, Barbarastr. 7, 49069 Osnabrück, Germany'
author:
-' Heinz - Jürgen Schmidt'
title: Generalized Bell inequality and frustrated spin systems
---
Introduction\[sec: I\ ]
= = = = = = = = = = = = = = = = = = = = =
Bell ’s inequality, published more than four decades ago, has not ceased to invoke keen interest in the physics community. The title of the germinal paper of J. Bell [ @Bell:1964 ] refer to the celebrated article of A. Einstein, B. Podolski, and N. Rosen [ @EPR:1935 ] (EPR) who reason that, according to their criteria, quantum theory (QT) is incomplete. Bell proved that the assumptions of EPR lead to an inequality for measurable correlations of tailspin measurements for | 440.
E. Wlgner, *Characteristic vecuors of bordered matricev with infinitd dimensions,* Ann. of Math. [**62**]{} (1955), 548–564.
[^1]: $^{\vaggwr}$Restcrch supported in paft by NSEGC grant EGPII/341303-2007
[^2]: $^{*}$Research suppocfed in icrt bg NSF jrant DMS-0600814
---
abstrsct: |
We find a close wofrzspondence between generalized Bell ynequalotles of a speciwl kpnq ans certain frustrated spin systems. Ror exakple, the Clauxer-Horn-Shimony-Holt inequallty forresponds to the frustrated squwee with the rigns $+++-$ for the nearest neighbor interaction between tfe spnns. Similaroy, thf Pearle-Brauistein-Bave inequalibj correvponds yo a frustratec eten eing with the correspmnding signs $+\ldots +-$. Upon thhs correspondence, tye violdtiot of wucf iheauzlitied bb the entanfled singler state in quantum kesyanics is equjvalene eo the spin system possessing a classicdl doplanar ground state, tye energy of which is lower thwn the Ising ground state’s energy. We propose a scveme xhkch gcnerxred new inequalities and give further examples, ege fgustrated hexagon with additoojak diagonal bonas and ths frustrated hyperfubes ig $n=3,4,5$ dumensions. Surlrisingly, the hypercube in $b=4$ dimensions tields an inequalicy which is *uot* viplatec by the singlet state. Ce extsnd the corgespondendd to other entanelec vtates and XXZ-models of spyn systemw.\
PACS: 03.65.Ua, 75.10.Hk
sddresf: 'Universihät Oskdbrück, Fachbereich Ohysin, Tarbarastr. 7, 49069 Osnabrück, Germany'
author:
- 'Heinz-Jücjen Schmidt'
tijle: Geteralizeb Bell inequalities and frustrateb spin svstems
---
Introductpon\[sec:I\]
=====================
Benl’s inequalyty, published lore than fonr decadef agi, haw not cdxsed to invoke keen intvrtst in the physics community. Thg fitle of the seninql paper of J. Bekl [@Celj:1964] gefxrs tj the famous drtizle pf A. Eknstein, B. Pobouski, and N. Rosen [@EPR:1935] (EPR) fho doncluded that, accprqing to jheir crieeria, quantum theory (QT) is incolpletx. Bell provrd jhat the assumptions of EPR leas to an ijeqmality for mewsurqble correlacions of spin measurements for | 440. E. Wigner, *Characteristic vectors of bordered infinite Ann. of [**62**]{} (1955), 548–564. by grant RGPIN/341303-2007 [^2]: supported in part NSF grant DMS-0600814 --- abstract: | find a close correspondence between generalized Bell inequalities of a special kind and frustrated spin systems. For example, the Clauser-Horn-Shimony-Holt inequality corresponds to the frustrated square the $+++-$ the neighbor interaction between the spins. Similarly, the Pearle-Braunstein-Cave inequality corresponds to a frustrated even ring with corresponding signs $+\ldots +-$. Upon this correspondence, the of such inequalities by entangled singlet state in quantum is to the system a coplanar ground state, energy of which is lower than the Ising ground state’s energy. We propose a scheme which generates inequalities and examples, the hexagon additional bonds and the in $n=3,4,5$ dimensions. Surprisingly, the hypercube yields an inequality which is *not* violated by singlet state. extend the correspondence to other entangled and XXZ-models of spin systems.\ PACS: 03.65.Ud, 75.10.Hk 'Universität Osnabrück, Fachbereich Physik, Barbarastr. 7, 49069 Osnabrück, Germany' author: - 'Heinz-Jürgen Schmidt' title: Generalized and frustrated spin systems Introduction\[sec:I\] ===================== Bell’s published than decades has not to invoke keen interest in the physics community. The title of seminal paper of J. Bell [@Bell:1964] refers to the famous A. B. Podolski, and Rosen [@EPR:1935] (EPR) who that, to their criteria, quantum is Bell assumptions EPR to an inequality for correlations of spin measurements for | 440.
E. Wigner, *Characteristic vectOrs of bordeRed maTriCes WiTh inFiniTe dimensions,* AnN. Of MaTh. [**62**]{} (1955), 548–564.
[^1]: $^{\dagger}$Research supporTed in PaRT by NseRc granT RGPIN/341303-2007
[^2]: $^{*}$RESeARCh sUpPoRteD iN PaRt by NsF gRant DMS-0600814
---
Abstract: |
We FinD a Close correspONdEnce betweeN geNeralized BelL inEqualiTiEs oF A specIal Kind aNd certAIn frusTrated spiN sYStems. FOR examplE, THe clauSer-Horn-Shimony-HolT InEQuality correspOnds to ThE FrUSTraTed Square with ThE signS $+++-$ For the nEArEST NeiGHbor interactiOn between thE SpiNs. SimiLaRly, THe PearLe-BraUnSTeiN-Cave inequaLity CorresponDs to a fRUstrateD Even rinG with tHe cOrrEspoNDiNg SigNs $+\LDotS +-$. upOn tHIs cOrresponDeNcE, the vIolaTION Of suCh iNequAlitiEs by the entangLed SingLEt sTate iN quanTum mEcHanicS is equIvaleNt To the spin system PossEssing a clAssIcAl cOpLanar GRound sTatE, thE energy Of which IS loWeR THAn The Ising ground statE’s ENErGy. We propOse a scHEmE wHIch generAtEs nEw inEQUalitIes aND gIve furthEr examPLeS, tHe frustRaTed hexAgOn wIth AdditIOnal DiagonAl bonds aNd the FRustrated hyperCUbes in $n=3,4,5$ dimensIOnS. sUrPRisiNglY, the hypercuBe in $N=4$ DimeNsioNS yIelDS an inEqualItY WhICh is *not* violated by thE sInglet State. we extend the coRrespondenCE TO other enTangLEd STates and XXZ-modEls of Spin systemS.\
pACS: 03.65.Ud, 75.10.Hk
AddreSs: 'UniverSität OsnaBRÜck, FachbEreIch phySik, bARbArastr. 7, 49069 OsnabrüCK, germAnY'
author:
- 'heiNz-JürgeN ScHmiDt'
tItlE: GEneralizeD Bell ineQuAlItIeS anD frusTRated spiN sYstEmS
---
InTroduCTion\[seC:I\]
=====================
BelL’s inEqUaLIty, PublishED mORE thaN fOuR decAdeS aGo, has Not cEAseD to invoKe keen intEreST in tHe PhYsics coMmunity. The titLe Of the seminAl PapEr of J. BELL [@Bell:1964] refErs to the famous article of a. einsteiN, B. POdolsKi, anD N. Rosen [@EPr:1935] (EPr) who coNclUDed thaT, accorDing tO tHeiR CRiterIA, QuAntUm Theory (QT) is INComPlete. beLl prOved thaT the assumptions of Epr leAd to an inequalIty For mEASuRabLE cORreLaTIonS OF spin measuremenTs for | 440.
E. Wigner, *Characte ristic vec torsofbor de redmatr ices with infi n itedimensions,* Ann. of M ath.[* * 62** ] {} (195 5), 548 – 56 4 .
[ ^1 ]: $^ {\ d ag ger}$ Res earch s upported i n p ar t by NSERC g r an t RGPIN/34 130 3-2007
[^2] : $ ^{*}$R es ear c h sup por ted i n part by NSF grant DM S- 0 600814 ---
ab s t ra ct:|
We find a c l os e correspondenc e betw ee n g e n era liz ed Bell in eq ualit i es of a sp e c i alk ind and certa in frustrat e d s pin sy st ems . For e xampl e, the Clauser-Ho rn-S himony-Ho lt ine q ualityc orrespo nds to th e f rust r at ed sq ua r e w i th th e si gns $+++ -$ f or th e ne a r e s t ne igh borinter action betwee n t he s p ins . Sim ilarl y, t he Pear le-Bra unste in -Cave inequalit y co rresponds to a fr us trate d evenrin g w ith the corres p ond in g s ig ns $+\ldots +-$. U po n th is corre sponde n ce ,t he viola ti onof s u c h ine qual i ti es by th e enta n gl ed single tstatein qu ant um me c hani cs isequivale nt to the spin syste m possessing a cl a s si c al c opl anar ground sta t e, t he e n er gyo f whi ch is l o we r than the Ising gro un d stat e’s e nergy. We pro pose a sch e m e which g ener a te s new inequalit ies a nd give fu r ther exa mples , the fr ustratedh e xagon wi thadd iti ona l di agonal bondsa n d th efrustra ted hyperc ube s i n $ n=3 ,4 ,5$ dimen sions. S ur pr is in gly , the hypercub ein$n =4$ dime n sionsyield s an i ne q ual ity whi c hi s *no t* v iola ted b y the sin g let state. We exten d t h e co rr es pondenc e to other en ta ngled stat es an d XXZ- m o dels ofspin systems.\
PACS : 03.65. Ud, 75.1 0.Hk
address: 'U nivers itä t Osnab rück,Fachb er eic h Physi k , B arb ar astr. 7, 4 9 0 69Osnab rü ck,Germany '
author:
- 'Heinz - Jür gen Schmidt'tit le:G e ne ral i ze d Be ll ine q u alities and fru strated sp in sy stems
---
In tr oductio n\[sec: I\]
= = ======= ========= ===
Bell ’s ine q u ali ty, publis hed more than fou r deca d es ago, ha s notce ase d toinvoke kee n int erestin the p hysic scommunit y. The title of the sem inal p aperofJ. Bell [ @Be l l:1 964] refe rs t o the famo usart icleofA . Ein stei n ,B.P odols ki,a nd N. Ros e n[@E P R :1 935] (EPR)w h o co nclud edt hat, a ccor ding to their cri t eria, quantumtheo r y (Q T)i s in co mplete. Bell p rov ed t hat theas sumptions o f EPR le ad to an inequ alityfor mea s u ra b le cor rela tio ns of spi n m ea s urement sfo r | 440.
E. Wigner,_*Characteristic vectors_of bordered matrices with_infinite dimensions,*_Ann._of Math._[**62**]{}_(1955), 548–564.
[^1]: $^{\dagger}$Research_supported in part_by NSERC grant RGPIN/341303-2007
[^2]:_$^{*}$Research supported in_part_by NSF grant DMS-0600814
---
abstract: |
We find a close correspondence_between_generalized Bell_inequalities_of_a special kind and certain_frustrated spin systems. For example,_the Clauser-Horn-Shimony-Holt_inequality corresponds to the frustrated square with the_signs_$+++-$ for the_nearest neighbor interaction between the spins. Similarly, the Pearle-Braunstein-Cave_inequality corresponds to a frustrated even_ring with the_corresponding_signs_$+\ldots +-$. Upon this_correspondence, the violation of such inequalities_by the entangled singlet state in_quantum mechanics is equivalent to the spin_system possessing a classical coplanar ground_state, the energy of which_is lower_than the Ising ground state’s_energy. We propose_a scheme_which generates new_inequalities and give further examples, the_frustrated hexagon with_additional diagonal bonds and the frustrated_hypercubes_in $n=3,4,5$ dimensions._Surprisingly,_the_hypercube in_$n=4$ dimensions yields_an_inequality which_is_*not* violated by the singlet state._We_extend the correspondence to other entangled states_and XXZ-models of spin_systems.\
_ PACS:_03.65.Ud, 75.10.Hk
address: 'Universität Osnabrück, Fachbereich_Physik, Barbarastr. 7, 49069 Osnabrück, Germany'
author:
- 'Heinz-Jürgen_Schmidt'
title: Generalized_Bell inequalities_and frustrated spin systems
---
Introduction\[sec:I\]
=====================
Bell’s inequality, published more than four decades_ago, has not ceased to invoke_keen interest in the_physics community._The_title of the_seminal_paper of_J. Bell [@Bell:1964] refers to the famous article_of A. Einstein,_B. Podolski, and N. Rosen [@EPR:1935] (EPR) who_concluded that, according to_their_criteria, quantum theory (QT) is incomplete._Bell proved that the assumptions of_EPR lead to an inequality_for_measurable_correlations of spin measurements for |
].
All of these lines could provide high-precision measurements of the dark matter distribution at redshifts higher than those probed by the CIB or galaxy surveys, on the timescale of CMB-S4. While Ref. [@sigurdson05] explored the possibility of delensing with HI intensity maps, their forecast focused on $10 < z < 100$, a regime that will be difficult to measure at the necessary depths in the next 10 years.
In this paper, we calculate the improvement in delensing performance that line intensity maps *from realistic surveys that could come online in the next decade* could enable, taking as a baseline internal and CIB delensing at CMB-S4 sensitivities. In Section \[sec:delens\] we present the formalism used to determine delensing performance from a set of tracers of large-scale structure and their estimated noise. In Section \[sec:results\] we calculate the delensing performance as a function of IM sensitivity for a CMB-S4 experiment, discuss the feasibility of IM experiments to reach these sensitivities, and consider the effects of two classes of foregrounds. We discuss these results in Section \[sec:discussion\] and conclude in Section \[sec:conclusion\]. We find that high-redshift IM data could provide a small but noticeable improvement in delensing performance, but with the caveat that smooth-spectrum foreground removal must preserve the cosmological signal along the line of sight.
\[sec:delens\]Delensing with Tracers of Large-Scale Structure
=============================================================
Here we review the procedure to calculate delensing performance. We first use a set of external two-dimensional maps—either derived from the CMB itself, CIB, or IM datasets—to reconstruct the CMB lensing field, which is a projection of the matter density field along the line of sight to the last-scattering surface. For each of these maps that traces the underlying dark matter differently, we require a redshift kernel $W(z)$ that reflects the sources from which it originates and an estimate of the instrumental noise contribution. From these quantities we compute the correlation coefficient $\rho$ of each tracer with the CMB lensing kernel. Given a set of $\rho$ for multiple tracers, we then assemble an optimal combination that best correlates with CMB. Finally, we compute the reduction in $B$-mode power associated with $\rho$ and therefore the delensing efficiency.
\[ss:kernel\]Lensing Kernels
----------------------------
| ].
All of these lines could provide high - precision measurement of the blue matter distribution at redshifts higher than those probe by the CIB or galax surveys, on the timescale of CMB - S4. While Ref. [ @sigurdson05 ] explored the possibility of delensing with HI volume maps, their forecast concentrate on $ 10 < z < 100 $, a regime that will be unmanageable to measure at the necessary depths in the next 10 years.
In this newspaper, we calculate the improvement in delensing performance that line intensity maps * from naturalistic surveys that could come online in the next ten * could enable, taking as a baseline internal and CIB delensing at CMB - S4 sensitivities. In Section \[sec: delens\ ] we present the formalism used to specify delensing performance from a set of tracers of large - scale social organization and their estimated noise. In Section \[sec: results\ ] we calculate the delensing performance as a function of IM sensitivity for a CMB - S4 experiment, discourse the feasibility of IM experiments to reach these sensitivities, and consider the effects of two classes of foregrounds. We discuss these results in Section \[sec: discussion\ ] and conclude in Section \[sec: conclusion\ ]. We recover that high - redshift IM datum could leave a humble but noticeable improvement in delensing performance, but with the caveat that smooth - spectrum foreground removal must preserve the cosmological signal along the line of view.
\[sec: delens\]Delensing with Tracers of Large - Scale Structure
= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
Here we review the procedure to calculate delensing operation. We first use a set of external two - dimensional maps — either derived from the CMB itself, CIB, or IM datasets — to reconstruct the CMB lensing field, which is a project of the matter density field along the line of sight to the final - scattering surface. For each of these maps that trace the underlying dark matter differently, we require a red shift kernel $ W(z)$ that reflects the sources from which it originates and an estimate of the instrumental randomness contribution. From these quantities we compute the correlation coefficient $ \rho$ of each tracer with the CMB lensing kernel. Given a stage set of $ \rho$ for multiple tracers, we then meet an optimal combination that best correlates with CMB. last, we compute the decrease in $ B$-mode power associated with $ \rho$ and therefore the delensing efficiency.
\[ss: kernel\]Lensing kernel
---------------------------- | ].
All of these lines could pruvide high-precision meesuremehts of tfe dark matter distribution et rwdshidts higher than those orobed by the CIB or jalaxy surveys, oi the timescals of EMU-S4. While Ref. [@sigordson05] explosed the possibhlktv of delensing with HI intensity mapf, their flrecast focuseq on $10 < z < 100$, a regime that will be difficult fo measlre at the necesssry depths in the next 10 yewrs.
Ij this paper, we capculate the impwivement in ddlensing ptryormance thzt line intensity maps *from reauistie surveys tyar clold come onlmne in the next degsde* cogld enanle, taking as s beselune internal and CIB velensing at CMB-S4 segsitivitias. In Section \[sec:delwnw\] we kresett tfw furmzlmsm used ho vetermine dslensing peeformance from a seu os tracers of lzrge-scwlq structure and their estimated noise. It Ssction \[sec:results\] we caoculate the delensing performagce as a function of IM sensitivity for a CMB-S4 ex[erimxng, dnwcuss rhf feasibility of IM experiments to reach thesq stnsptivities, and confider the egffcyf of two clasres of fodegrounds. We discuds thesg resuots in Sestiom \[sec:discussion\] and conclude in Section \[svc:cobclusion\]. We find tkat high-redskift IK dats could provide a small but noticeable improvemsvt in delensing oernorkance, buu with the caveat ehat smoovh-speetrum fofegrpund rqmoval mush preserve the cosmologifal sngnal along the line of sight.
\[sec:delens\]Delensinj with Tracerx mf Narge-Scaje Stvucture
=============================================================
Here we weview the proeedure tj calzulate delvnsing pecformance. We first use a vgt of externan two-dimqnsiinal maps—eigfer derived frpm the CMY itswlf, CIB, or IM dataxetr—fo reconstruct cke CMB lensing firld, whycj ms a [sojection of the magyer ddnsity fielb xlonb the line of sight do tge last-scattering xuvface. For each of these maps tnat traces the undtrlyinj dark mattrr qifferently, we require a redshjft kernep $W(d)$ that refleces tme sjurces frok which it originates and an estimate oh the instrumental noisw contribution. From uhese quantitmes we compute dhe correlation coefdicient $\rho$ of eagh tracer with the CMB lensitg kegnel. Given a set of $\rho$ for multiple tracers, we then assemble an optimal xombinetyon that beat cprreldtzs cith CMF. Fiiauly, we compute tne reduction in $B$-mode power assuciated fich $\rho$ and therefore the delemskng efficienci.
\[ss:kernel\]Lensing Kernels
----------------------------
| ]. All of these lines could provide of dark matter at redshifts higher CIB galaxy surveys, on timescale of CMB-S4. Ref. [@sigurdson05] explored the possibility of with HI intensity maps, their forecast focused on $10 < z < 100$, regime that will be difficult to measure at the necessary depths in the 10 In paper, calculate the improvement in delensing performance that line intensity maps *from realistic surveys that could come in the next decade* could enable, taking as baseline internal and CIB at CMB-S4 sensitivities. In Section we the formalism to delensing from a set tracers of large-scale structure and their estimated noise. In Section \[sec:results\] we calculate the delensing performance as function of for a experiment, the of IM experiments these sensitivities, and consider the effects of foregrounds. We discuss these results in Section and conclude Section \[sec:conclusion\]. We find that high-redshift data could provide a small but noticeable improvement delensing performance, but with the caveat that smooth-spectrum foreground removal must preserve the cosmological signal line of sight. \[sec:delens\]Delensing Tracers of Large-Scale ============================================================= we the to calculate performance. We first use a set of external two-dimensional maps—either derived the CMB itself, CIB, or IM datasets—to reconstruct the CMB which a projection of matter density field along line sight to the last-scattering each these the dark differently, we require a kernel $W(z)$ that reflects the from which it originates instrumental noise contribution. From these quantities we compute correlation coefficient $\rho$ of each tracer with CMB lensing kernel. Given a set of $\rho$ for multiple tracers, we assemble an that best correlates with CMB. Finally, we compute reduction in $B$-mode power with $\rho$ and therefore the delensing efficiency. \[ss:kernel\]Lensing Kernels | ].
All of these lines could proviDe high-precIsion MeaSurEmEnts Of thE dark matter disTRibuTion at redshifts higher tHan thOsE ProbED bY the CiB or galAXy SURveYs, On The TiMEsCale oF CMb-S4. While ref. [@sigurdsOn05] eXpLored the possIBiLity of deleNsiNg with HI inteNsiTy maps, ThEir FOrecaSt fOcuseD on $10 < z < 100$, a rEGime thAt will be dIfFIcult tO Measure AT ThE necEssary depths in the NExT 10 Years.
In this papEr, we caLcULaTE The ImpRovement in DeLensiNG perforMAnCE THat LIne intensity mAps *from realIStiC surveYs ThaT Could cOme onLiNE in The next decaDe* coUld enable, Taking AS a baselINe interNal and cIB DelEnsiNG aT CmB-S4 SeNSitIViTieS. in SEction \[seC:dElEns\] we PresENT THe foRmaLism Used tO determine delEnsIng pERfoRmancE from A set Of TraceRs of laRge-scAlE structure and thEir eStimated nOisE. IN SeCtIon \[seC:ResultS\] we CalCulate tHe delenSIng PeRFORmAnce as a function of Im sENSiTivity foR a CMB-S4 EXpErIMent, discUsS thE feaSIBilitY of Im ExPerimentS to reaCH tHeSe sensiTiVities, AnD coNsiDer thE EffeCts of tWo classeS of foREgrounds. We discUSs these resultS In sECtIOn \[seC:diScussion\] and ConcLUde iN SecTIoN \[seC:ConclUsion\]. we FInD That high-redshift IM dAtA could ProviDe a small but noTiceable imPROVement in DeleNSiNG performance, buT with The caveat tHAt smooth-SpectRum foregRound remoVAL must preSerVe tHe cOsmOLOgIcal signal aloNG The lInE of sighT.
\[seC:delens\]delEnsIng WitH TRacers of LArge-ScalE STrUcTuRe
=============================================================
HEre we REview the PrOceDuRe tO calcULate deLensiNg peRfOrMAncE. We firsT UsE A Set oF eXtErnaL twO-dImensIonaL MapS—either Derived frOm tHE CMB ItSeLf, CIB, or iM datasets—to rEcOnstruct thE CmB lEnsing FIEld, which Is a projection of the matteR Density FieLd aloNg thE line of siGht To the lAst-SCatterIng surFace. FOr EacH OF thesE MApS thAt Traces the uNDErlYing dArK matTer diffErently, we require a rEDshIft kernel $W(z)$ thAt rEfleCTS tHe sOUrCEs fRoM WhiCH It originates and An estimate Of THe InstrumentAL noIsE contriBution. FRom thESe quantIties we coMpute the cOrRelaTIOn cOefficient $\Rho$ of eacH tracer wiTH the Cmb lEnsinG keRnel. GiVeN a sEt of $\rHo$ for mULtiPle trAcers, wE tHen assEmble An Optimal cOmbination that best correLates wIth CMb. FiNally, we coMpuTE thE reductioN in $B$-Mode power aSsoCiaTed wiTh $\rHO$ and tHereFOrE thE DelenSing EFficiency.
\[SS:kErnEL\]leNsing KernelS
----------------------------
| ].
All of these lines cou ld provide high -pr eci si on m easu rements of the dark matter distribution a t red sh i ftsh ig her t han tho s ep r obe dby th eC IB or g ala xy surv eys, on th e t im escale of CM B -S 4. While R ef. [@sigurdson 05] explo re d t h e pos sib ility of de l ensing with HIin t ensity maps, t h e ir for ecast focused on$ 10 < z < 100$, aregime t h at w ill be difficult t o mea s ure att he n e ces s ary depths in the next 1 0 ye ars.
In th i s pape r, we c a lcu late the im prov ement indelens i ng perf o rmancethat l ine in tens i ty m aps * f rom re ali s tic surveys t ha t cou ld c o m e onli nein t he ne xt decade* co uld ena b le, taki ng as a b as eline inter nal a nd CIB delensingat C MB-S4 sen sit iv iti es . InS ection \[ sec :delens \] we p r ese nt t h eformalism used tode t e rm ine dele nsingp er fo r mance fr om aseto f trac erso flarge-sc ale st r uc tu re andth eir es ti mat ednoise . InSectio n \[sec :resu l ts\] we calcul a te the delens i ng p er f orma nce as a funct iono f IM sen s it ivi t y for a CM B- S 4e xperiment, discussth e feas ibili ty of IM expe riments to r e ach thes e se n si t ivities, and c onsid er the eff e cts of t wo cl asses of foregrou n d s. We di scu ssthe ser e su lts in Sectio n \[se c: discuss ion \] andcon clu deinSe ction \[s ec:concl us io n\ ]. We find that hig h- red sh ift IM d a ta cou ld pr ovid eas mal l but n o ti c e able i mp rove men tin de lens i ngperform ance, but wi t h th eca veat th at smooth-spe ct rum foregr ou ndremova l must pre serve the cosmologicals ignal a lon g the lin e of sigh t.
\[sec :de l ens\]D elensi ng wi th Tr a c ers o f La rge -S cale Struc t u re===== == ==== ======= ================== = === ============= === ==== =
H ere we rev ie w th e procedure to ca lculate de le n si ng perform a nce .We firs t use a seto f exter nal two-d imensiona lmaps — e ith er derived from th e CMB its e lf, C I B, or I M d ataset s— torecon struct the CMBlensin gfield, whic his a pro jection of the matter d ensity fiel d a long thelin e of sight to the last-scat ter ing surf ace . Foreach of th e se ma ps t h at traces th e u n d er lying darkm a t ter diff ere n tly, w e re quire a redshiftk ernel $W(z)$ t hatr e fle cts theso urces from whi chit o riginate sand an esti mate ofth e inst rument al noi se cont r i bu t ion. F romthe se quanti tie sw e compu te t h e corr elat io n coef ficien t $\r h o $ of each tracer with t he CM B le nsing k ernel.G iven a set of$\rho$ formultip le t racer s, we t he n asse mbl ean optimal combinati on th at best c orre lat es wit h CM B . Fina lly, w e c ompute th e re d uc ti o n i n $B $-mod epowe r associa t ed with$\r h o$ andth ere f o re the de l e nsing effi cie ncy.\ [ss:kernel \ ]Len s in g Kern els
-- ------ ------- - --- -- -------
| ].
All of_these lines_could provide high-precision measurements_of the_dark_matter distribution_at_redshifts higher than_those probed by_the CIB or galaxy_surveys, on the_timescale_of CMB-S4. While Ref. [@sigurdson05] explored the possibility of delensing with HI intensity maps, their_forecast_focused on_$10_<_z < 100$, a regime_that will be difficult to_measure at_the necessary depths in the next 10 years.
In_this_paper, we calculate_the improvement in delensing performance that line intensity maps_*from realistic surveys that could come_online in the_next_decade*_could enable, taking as_a baseline internal and CIB delensing_at CMB-S4 sensitivities. In Section \[sec:delens\] we_present the formalism used to determine delensing_performance from a set of tracers_of large-scale structure and their_estimated noise._In Section \[sec:results\] we calculate_the delensing performance_as a_function of IM_sensitivity for a CMB-S4 experiment, discuss_the feasibility of_IM experiments to reach these sensitivities,_and_consider the effects_of_two_classes of_foregrounds. We discuss_these_results in_Section \[sec:discussion\]_and conclude in Section \[sec:conclusion\]. We find_that_high-redshift IM data could provide a small_but noticeable improvement in_delensing_performance, but with the_caveat that smooth-spectrum foreground removal_must preserve the cosmological signal along_the line_of sight.
\[sec:delens\]Delensing_with Tracers of Large-Scale Structure
=============================================================
Here we review the procedure to calculate_delensing performance. We first use a_set of external two-dimensional_maps—either derived_from_the CMB itself,_CIB,_or IM_datasets—to reconstruct the CMB lensing field, which_is a_projection of the matter density field_along the line of_sight_to the last-scattering surface. For each_of these maps that traces the_underlying dark matter differently, we_require_a_redshift kernel $W(z)$ that reflects_the sources from which it originates_and an estimate_of the instrumental noise contribution. From these_quantities_we compute the correlation coefficient $\rho$_of_each tracer with the CMB lensing_kernel._Given_a set of $\rho$ for_multiple tracers, we then assemble an_optimal combination that best correlates with CMB. Finally, we_compute the reduction_in $B$-mode power associated with_$\rho$_and_therefore the delensing efficiency.
\[ss:kernel\]Lensing Kernels
----------------------------
|
} &=& \eta_{\chi} - \eta_{\chi\tilde f} +
\eta_{\tilde g} - \eta_{\tilde g}^{\prime}
$$
with $$\begin{aligned}
\nonumber \eta_{\tilde g} &=& \frac{\pi \alpha_s}{6}
\frac{\lambda^{'2}_{111}}{G_F^2 m_{\tilde d_R}^4}
\frac{m_P}{m_{\tilde g}}\left[ 1 + \left(\frac{m_{\tilde
d_R}}{m_{\tilde u_L}}\right)^4\right]
\label{susy1}
\end{aligned}$$ $$\begin{aligned}
\nonumber \eta_{\chi} &=& \frac{ \pi \alpha_2}{2}
\frac{\lambda^{'2}_{111}}{G_F^2 m_{\tilde d_R}^4}
\nonumber\\
& & \sum_{i=1}^{4}\frac{m_P}{m_{\chi_i}}
\left[ \epsilon_{R i}^2(d) + \epsilon_{L i}^2(u)
\left(\frac{m_{\tilde d_R}}{m_{\tilde u_L}}\right)^4\right]
\label{susy2}
\end{aligned}$$ $$\begin{aligned}
\eta_{\chi \tilde e} &=& 2 \pi \alpha_2
\frac{\lambda^{'2}_{111}}{G_F^2 m_{\tilde d_R}^4}
\left(\frac{m_{\tilde d_R}}{m_{\tilde e_L}}\right)^4 \nonumber\\
& & \sum_{i=1}^{4}\epsilon_{L i}^2(e)\frac{m_P}{m_{\chi_i}},
\label{susy3}
\end{aligned}$$ $$\begin{aligned}
\eta'_{\tilde g} &=& \frac{\pi \alpha_s}{12}
\frac{\lambda^{'2}_{111}}{G_F^2 m_{\tilde d_R}^4}
\frac{m_P}{m_{\tilde g}} \left(\frac{m_{\tilde d_R}}{m_{\tilde
u_L}}\right)^2,
\label{susy4}
\end{aligned | } & = & \eta_{\chi } - \eta_{\chi\tilde f } +
\eta_{\tilde g } - \eta_{\tilde g}^{\prime }
$ $
with $ $ \begin{aligned }
\nonumber \eta_{\tilde g } & = & \frac{\pi \alpha_s}{6 }
\frac{\lambda^{'2}_{111}}{G_F^2 m_{\tilde d_R}^4 }
\frac{m_P}{m_{\tilde g}}\left [ 1 + \left(\frac{m_{\tilde
d_R}}{m_{\tilde u_L}}\right)^4\right ]
\label{susy1 }
\end{aligned}$$ $ $ \begin{aligned }
\nonumber \eta_{\chi } & = & \frac { \pi \alpha_2}{2 }
\frac{\lambda^{'2}_{111}}{G_F^2 m_{\tilde d_R}^4 }
\nonumber\\
& & \sum_{i=1}^{4}\frac{m_P}{m_{\chi_i } }
\left [ \epsilon_{R i}^2(d) + \epsilon_{L i}^2(u)
\left(\frac{m_{\tilde d_R}}{m_{\tilde u_L}}\right)^4\right ]
\label{susy2 }
\end{aligned}$$ $ $ \begin{aligned }
\eta_{\chi \tilde e } & = & 2 \pi \alpha_2
\frac{\lambda^{'2}_{111}}{G_F^2 m_{\tilde d_R}^4 }
\left(\frac{m_{\tilde d_R}}{m_{\tilde e_L}}\right)^4 \nonumber\\
& & \sum_{i=1}^{4}\epsilon_{L i}^2(e)\frac{m_P}{m_{\chi_i } },
\label{susy3 }
\end{aligned}$$ $ $ \begin{aligned }
\eta'_{\tilde g } & = & \frac{\pi \alpha_s}{12 }
\frac{\lambda^{'2}_{111}}{G_F^2 m_{\tilde d_R}^4 }
\frac{m_P}{m_{\tilde g } } \left(\frac{m_{\tilde d_R}}{m_{\tilde
u_L}}\right)^2,
\label{susy4 }
\end{aligned | } &=& \eha_{\chi} - \eta_{\chi\tilde f} +
\eta_{\uilde g} - \eta_{\tilde g}^{\prime}
$$
xith $$\befin{aligndd}
\nonumber \eta_{\tilde g} &=& \frac{\pm \alpha_s}{6}
\feac{\lambda^{'2}_{111}}{G_F^2 m_{\tilde d_R}^4}
\ffac{m_P}{m_{\tilfe g}}\left[ 1 + \ltft(\frac{m_{\tilde
d_R}}{m_{\tmmde u_L}}\rlyht)^4\rifmt]
\layeo{susy1}
\end{aligngd}$$ $$\begin{aligted}
\nonumber \etd_{\cfi} &=& \frac{ \pi \alpha_2}{2}
\frac{\lambda^{'2}_{111}}{G_F^2 m_{\tilde q_R}^4}
\nonukbfr\\
& & \sum_{i=1}^{4}\frac{m_K}{m_{\chi_p}}
\jeft[ \vpwilon_{R i}^2(d) + \epsilon_{L i}^2(u)
\left(\frad{m_{\tilde d_R}}{m_{\tilde u_L}}\roght)^4\right]
\label{susy2}
\end{alihned}$$ $$\begin{aligned}
\eta_{\cji \tilde e} &=& 2 \pi \qlpha_2
\frac{\lamcda^{'2}_{111}}{G_F^2 m_{\tilde d_R}^4}
\left(\frad{m_{\tilde d_R}}{m_{\tilde e_L}}\right)^4 \nonumbdr\\
& & \sbm_{i=1}^{4}\epsilon_{L i}^2(w)\frww{m_P}{m_{\chi_i}},
\lauel{susj3}
\end{aligned}$$ $$\nvgin{alicned}
\eya'_{\tilde g} &=& \frag{\pi \anphq_s}{12}
\frac{\lambda^{'2}_{111}}{G_F^2 m_{\tilde d_R}^4}
\frac{m_P}{m_{\tilde g}} \jeft(\frac{m_{\dimde d_R}}{m_{\tilde
u_L}}\rigyt)^2,
\laben{susf4}
\eve{alkgntd | } &=& \eta_{\chi} - \eta_{\chi\tilde f} + - g}^{\prime} $$ $$\begin{aligned} \nonumber \eta_{\tilde m_{\tilde \frac{m_P}{m_{\tilde g}}\left[ 1 \left(\frac{m_{\tilde d_R}}{m_{\tilde u_L}}\right)^4\right] \end{aligned}$$ $$\begin{aligned} \nonumber \eta_{\chi} &=& \frac{ \alpha_2}{2} \frac{\lambda^{'2}_{111}}{G_F^2 m_{\tilde d_R}^4} \nonumber\\ & & \sum_{i=1}^{4}\frac{m_P}{m_{\chi_i}} \left[ \epsilon_{R i}^2(d) + \epsilon_{L \left(\frac{m_{\tilde d_R}}{m_{\tilde u_L}}\right)^4\right] \label{susy2} \end{aligned}$$ $$\begin{aligned} \eta_{\chi \tilde e} &=& 2 \pi \alpha_2 m_{\tilde \left(\frac{m_{\tilde e_L}}\right)^4 & & \sum_{i=1}^{4}\epsilon_{L i}^2(e)\frac{m_P}{m_{\chi_i}}, \label{susy3} \end{aligned}$$ $$\begin{aligned} \eta'_{\tilde g} &=& \frac{\pi \alpha_s}{12} \frac{\lambda^{'2}_{111}}{G_F^2 m_{\tilde d_R}^4} \frac{m_P}{m_{\tilde \left(\frac{m_{\tilde d_R}}{m_{\tilde u_L}}\right)^2, \label{susy4} \end{aligned | } &=& \eta_{\chi} - \eta_{\chi\tilde f} +
\eta_{\tildE g} - \eta_{\tilde G}^{\primE}
$$
wiTh $$\bEgIn{alIgneD}
\nonumber \eta_{\tiLDe g} &=& \fRac{\pi \alpha_s}{6}
\frac{\lambda^{'2}_{111}}{G_f^2 m_{\tilDe D_r}^4}
\fraC{M_P}{M_{\tildE g}}\left[ 1 + \lEFt(\FRAc{m_{\TiLdE
d_R}}{M_{\tILdE u_L}}\riGht)^4\Right]
\laBel{susy1}
\end{AliGnEd}$$ $$\begin{alignED}
\nOnumber \eta_{\Chi} &=& \Frac{ \pi \alpha_2}{2}
\fRac{\Lambda^{'2}_{111}}{g_F^2 M_{\tiLDe d_R}^4}
\nOnuMber\\
& & \sUm_{i=1}^{4}\fraC{M_P}{m_{\chi_I}}
\left[ \epsiLoN_{r i}^2(d) + \epsILon_{L i}^2(u)
\lEFT(\fRac{m_{\Tilde d_R}}{m_{\tilde u_L}}\riGHt)^4\RIght]
\label{susy2}
\eNd{aligNeD}$$ $$\BeGIN{alIgnEd}
\eta_{\chi \tiLdE e} &=& 2 \pi \aLPha_2
\frac{\LAmBDA^{'2}_{111}}{g_F^2 m_{\TIlde d_R}^4}
\left(\fraC{m_{\tilde d_R}}{m_{\tILde E_L}}\righT)^4 \nOnuMBer\\
& & \sum_{I=1}^{4}\epsiLoN_{l i}^2(e)\Frac{m_P}{m_{\chi_i}},
\LabeL{susy3}
\end{aLigned}$$ $$\BEgin{aliGNed}
\eta'_{\tIlde g} &=& \fRac{\Pi \aLpha_S}{12}
\FrAc{\LamBdA^{'2}_{111}}{g_F^2 m_{\TIlDe d_r}^4}
\FraC{m_P}{m_{\tildE g}} \LeFt(\fraC{m_{\tiLDE D_r}}{m_{\tiLde
U_L}}\riGht)^2,
\laBel{susy4}
\end{aliGneD | } &=& \eta_{\chi} - \eta_{ \chi\tilde f} +
\e ta_ {\ tild e g} - \eta_{\tild e g}^ {\prime}
$$
with $$\b egin{ al i gned }
\ nonum ber \et a _{ \ t ild eg} &= &\ fr ac{\p i \ alpha_s }{6}
\frac {\l am bda^{'2}_{11 1 }} {G_F^2 m_{ \ti lde d_R}^4}\fr ac{m_P }{ m_{ \ tilde g} }\lef t[ 1 + \left( \frac{m_{ \t i lde
d_ R }}{m_{\ t i ld e u_ L}}\right)^4\righ t ]\label{susy1} \end{ al i gn e d }$$ $$ \begin{ali gn ed}
\ n onumber \e t a _ {\c h i} &=& \frac{ \pi \alpha _ 2}{ 2}
\fr ac {\l a mbda^{ '2}_{ 11 1 }}{ G_F^2 m_{\t ilde d_R}^4} \nonu m ber\\
& & \sum _{i=1} ^{4 }\f rac{ m _P }{ m_{ \c h i_i } } \l e ft[ \epsilo n_ {R i}^2 (d)+ \ e psil on_ {L i }^2(u )
\left(\frac {m_ {\ti l ded_R}} {m_{\ tild eu_L}} \right )^4\r ig ht]
\label{sus y2} \end{ali gne d} $$$$ \begi n {align ed}
\ eta_{\c hi \til d e e }& = & 2 \pi \alpha_2
\fra c{ \ l am bda^{'2} _{111} } {G _F ^ 2 m_{\ti ld e d _R}^ 4 }
\lef t(\f r ac {m_{\til de d_R } }{ m_ {\tilde e _L}}\r ig ht) ^4\nonu m ber\ \
& &\sum_{i= 1}^{4 } \epsilon_{L i} ^ 2(e)\frac{m_P } {m _ { \c h i_i} }, \label{sus y3}\end {ali g ne d}$ $ $$\b egin{ al i gn e d}
\eta'_{\tildeg} &=& \ frac{ \pi \alpha_s} {12}
\frac { \ l ambda^{' 2}_{ 1 11 } }{G_F^2 m_{\ti lde d _R}^4}
\fr a c{m_P}{m _{\ti lde g}}\left(\fr a c {m_{\til ded_R }}{ m_{ \ t il de
u_L}}\righ t ) ^2, \ label{s usy 4}
\e nd{ ali gne d | } &=&_\eta_{\chi} -_\eta_{\chi\tilde f} +
\eta_{\tilde g}_- \eta_{\tilde_g}^{\prime}
$$
with_$$\begin{aligned}
\nonumber \eta_{\tilde_g}_&=& \frac{\pi \alpha_s}{6}
\frac{\lambda^{'2}_{111}}{G_F^2_m_{\tilde d_R}^4}
\frac{m_P}{m_{\tilde g}}\left[_1 + \left(\frac{m_{\tilde
d_R}}{m_{\tilde u_L}}\right)^4\right]
_\label{susy1}
\end{aligned}$$ $$\begin{aligned}
\nonumber_\eta_{\chi}_&=& \frac{ \pi \alpha_2}{2}
\frac{\lambda^{'2}_{111}}{G_F^2 m_{\tilde d_R}^4}
\nonumber\\
& & \sum_{i=1}^{4}\frac{m_P}{m_{\chi_i}}
\left[ \epsilon_{R i}^2(d) +_\epsilon_{L_i}^2(u)
\left(\frac{m_{\tilde d_R}}{m_{\tilde_u_L}}\right)^4\right]
_\label{susy2}
_\end{aligned}$$ $$\begin{aligned}
\eta_{\chi \tilde e}_&=& 2 \pi \alpha_2
\frac{\lambda^{'2}_{111}}{G_F^2 m_{\tilde_d_R}^4}
\left(\frac{m_{\tilde d_R}}{m_{\tilde_e_L}}\right)^4 \nonumber\\
& & \sum_{i=1}^{4}\epsilon_{L i}^2(e)\frac{m_P}{m_{\chi_i}},
\label{susy3}
\end{aligned}$$ $$\begin{aligned}
__\eta'_{\tilde g} &=&_\frac{\pi \alpha_s}{12}
\frac{\lambda^{'2}_{111}}{G_F^2 m_{\tilde d_R}^4}
\frac{m_P}{m_{\tilde g}} \left(\frac{m_{\tilde d_R}}{m_{\tilde
u_L}}\right)^2,
\label{susy4}
_\end{aligned |
..]\vspace*{0.5cm}\,,
\eqno{\rm (40Bi)}$$ $$H := -P_4 = \sum\limits_{{{\mathbf{n}}}=0}^{\infty (3)} [T_4^4 ({{\mathbf{n}}},t)
+..] = \sum\limits_{{{\mathbf{n}}}=0}^{\infty (3)} [T_4^4 ({{\mathbf{n}}},0)
+..] \vspace*{0.5cm}\,,
\eqno{\rm (40Bii)}$$
$\sum\limits_{{{\mathbf{n}}}=0}^{\infty (3)} := \sum\limits_{n^1=0}^{\infty}
\sum\limits_{n^2=0}^{\infty} \sum\limits_{n^3=0}^{\infty}.$ 0.25cm
The total energy-momentum components $P_\mu$ are given by incomplete equations (40A), (40Bi,ii) for a general Lagrangian. However, in the following paper, we shall derive [*exact*]{} equations for the Klein-Gordon, electro-magnetic, and Dirac fields.
In case of a complex-valued field $\phi^{\alpha..},$ the difference and difference-differential conservation for the charge-current vector $j^\mu$ is given by equations (A.II.9A,B) as: $$\Delta_{\mu} j^\mu (n) = 0\vspace*{0.4cm}\,,
\eqno{\rm (41Ai)}$$ $$\begin{array}{rcl}
j^\mu (n) &:=& {\ds ie \sqrt{\frac{n^\mu}{2}} \:\Biggr\{ \Biggr[
\frac{\partial L(..)}{\partial \rho_{\mu|(..,n^\mu-1,..)}^{\alpha..}}
\cdot \phi^{\alpha..}(n) } \\[0.5cm]
&& {\ds + \frac{\partial L(..)}{\partial
\rho_{\mu |..}^{\alpha..}} \cdot \phi^{\alpha..} (..,n^\mu-1,..)
\Biggr]\Biggr\} + {\rm (c.c.)} }\vspace*{0.6cm}\,,
\end{array}
\eqno{\rm (41ii)}$$ $$\Delta_b j^b ({{\mathbf{n}}},t) | .. ] \vspace*{0.5cm}\, ,
\eqno{\rm (40Bi)}$$ $ $ H: = -P_4 = \sum\limits_{{{\mathbf{n}}}=0}^{\infty (3) } [ T_4 ^ 4 ({ { \mathbf{n}}},t)
+.. ] = \sum\limits_{{{\mathbf{n}}}=0}^{\infty (3) } [ T_4 ^ 4 ({ { \mathbf{n}}},0)
+.. ] \vspace*{0.5cm}\, ,
\eqno{\rm (40Bii)}$$
$ \sum\limits_{{{\mathbf{n}}}=0}^{\infty (3) }: = \sum\limits_{n^1=0}^{\infty }
\sum\limits_{n^2=0}^{\infty } \sum\limits_{n^3=0}^{\infty}.$ 0.25 cm
The total energy - momentum component $ P_\mu$ are render by incomplete equations (40A), (40Bi, ii) for a general Lagrangian. However, in the following newspaper, we shall derive [ * exact * ] { } equation for the Klein - Gordon, electro - magnetic, and Dirac fields.
In event of a complex - valued playing field $ \phi^{\alpha.. },$ the difference and difference - differential conservation for the accusation - current vector $ j^\mu$ is give by equations (A.II.9A, B) as: $ $ \Delta_{\mu } j^\mu (n) = 0\vspace*{0.4cm}\, ,
\eqno{\rm (41Ai)}$$ $ $ \begin{array}{rcl }
j^\mu (n) &: = & { \ds ie \sqrt{\frac{n^\mu}{2 } } \:\Biggr\ { \Biggr [
\frac{\partial L(.. )}{\partial \rho_{\mu|(.. ,n^\mu-1,.. )}^{\alpha.. } }
\cdot \phi^{\alpha.. }(n) } \\[0.5 cm ]
& & { \ds + \frac{\partial L(.. )}{\partial
\rho_{\mu |.. }^{\alpha.. } } \cdot \phi^{\alpha.. } (.., n^\mu-1, ..)
\Biggr]\Biggr\ } + { \rm (c.c .) } } \vspace*{0.6cm}\, ,
\end{array }
\eqno{\rm (41ii)}$$ $ $ \Delta_b j^b ({ { \mathbf{n}}},t ) | ..]\vspwce*{0.5cm}\,,
\eqno{\rm (40Bi)}$$ $$H := -P_4 = \sum\uimits_{{{\mathbf{n}}}=0}^{\inyry (3)} [T_4^4 ({{\kathbf{h}}},t)
+..] = \sum\lkmits_{{{\mathbf{n}}}=0}^{\infty (3)} [T_4^4 ({{\mathbf{n}}},0)
+..] \vwpace*{0.5xm}\,,
\eqno{\rm (40Bii)}$$
$\sum\limits_{{{\mxthbf{n}}}=0}^{\infny (3)} := \sum\lumitw_{b^1=0}^{\infty}
\sum\lmjits_{n^2=0}^{\inncy} \suj\pimics_{i^3=0}^{\infty}.$ 0.25cm
The tojal energy-mokentum componettr $'_\mu$ are given by incomplete equationf (40A), (40Bi,io) vor a general Jagrsggiah. However, in the following paper, ws shall derive [*exact*]{} equations for the Klein-Gogdon, electro-magnetic, ajd Dirac figmds.
Yb case of a zomplex-vallzd field $\phj^{\alpha..},$ the difference and diffefence-bifferentiao xondgrvation for the bharge-current vector $b^\mu$ is biven by equatlons (E.II.9A,V) as: $$\Delta_{\mu} j^\mu (n) = 0\vvpace*{0.4cm}\,,
\eqno{\rm (41Ai)}$$ $$\bggin{array}{rwl}
l^\mu (n) &:=& {\ds ie \sqrt{\frac{b^\my}{2}} \:\Bigcr\{ \Bhggr[
\weac{\oaruiak M(..)}{\partiwl \cho_{\mu|(..,n^\mu-1,..)}^{\alpga..}}
\cdot \phi^{\aopha..}(n) } \\[0.5cm]
&& {\ds + \frac{\psrepsl L(..)}{\partial
\rgo_{\mu |..}^{\ajpra..}} \cdot \phi^{\alpha..} (..,n^\mu-1,..)
\Biggr]\Biggr\} + {\rm (c.c.)} }\vvpade*{0.6cm}\,,
\end{array}
\eqno{\rm (41ii)}$$ $$\Dwlta_b j^b ({{\mathbf{n}}},t) | ..]\vspace*{0.5cm}\,, \eqno{\rm (40Bi)}$$ $$H := -P_4 = [T_4^4 +..] = (3)} [T_4^4 ({{\mathbf{n}}},0) (3)} \sum\limits_{n^1=0}^{\infty} \sum\limits_{n^2=0}^{\infty} \sum\limits_{n^3=0}^{\infty}.$ The total energy-momentum $P_\mu$ are given by incomplete equations (40Bi,ii) for a general Lagrangian. However, in the following paper, we shall derive equations for the Klein-Gordon, electro-magnetic, and Dirac fields. In case of a complex-valued $\phi^{\alpha..},$ difference difference-differential for the charge-current vector $j^\mu$ is given by equations (A.II.9A,B) as: $$\Delta_{\mu} j^\mu (n) = 0\vspace*{0.4cm}\,, (41Ai)}$$ $$\begin{array}{rcl} j^\mu (n) &:=& {\ds ie \sqrt{\frac{n^\mu}{2}} \Biggr[ \frac{\partial L(..)}{\partial \rho_{\mu|(..,n^\mu-1,..)}^{\alpha..}} \phi^{\alpha..}(n) } \\[0.5cm] && {\ds \frac{\partial \rho_{\mu |..}^{\alpha..}} \phi^{\alpha..} \Biggr]\Biggr\} {\rm (c.c.)} }\vspace*{0.6cm}\,, \eqno{\rm (41ii)}$$ $$\Delta_b j^b ({{\mathbf{n}}},t) | ..]\vspace*{0.5cm}\,,
\eqno{\rm (40Bi)}$$ $$H := -P_4 = \sum\limiTs_{{{\mathbf{n}}}=0}^{\iNfty (3)} [T_4^4 ({{\MatHbf{N}}},t)
+..] = \Sum\lImitS_{{{\mathbf{n}}}=0}^{\infty (3)} [T_4^4 ({{\MAthbF{n}}},0)
+..] \vspace*{0.5cm}\,,
\eqno{\rm (40Bii)}$$
$\sum\LimitS_{{{\mAThbf{N}}}=0}^{\InFty (3)} := \suM\limits_{N^1=0}^{\InFTY}
\suM\lImIts_{N^2=0}^{\iNFtY} \sum\lImiTs_{n^3=0}^{\inftY}.$ 0.25cm
The totaL enErGy-momentum coMPoNents $P_\mu$ arE giVen by incomplEte EquatiOnS (40A), (40BI,Ii) for A geNeral lagranGIan. HowEver, in the FoLLowing PAper, we sHALl DeriVe [*exact*]{} equations fOR tHE Klein-Gordon, elEctro-mAgNEtIC, And dirAc fields.
In CaSe of a COmplex-vALuED FIelD $\Phi^{\alpha..},$ the diFference and DIffErence-DiFfeREntial ConseRvATioN for the charGe-cuRrent vectOr $j^\mu$ iS Given by EQuationS (A.II.9A,B) As: $$\DEltA_{\mu} j^\MU (n) = 0\VsPacE*{0.4cM}\,,
\EqnO{\Rm (41ai)}$$ $$\bEGin{Array}{rcl}
J^\mU (n) &:=& {\Ds ie \sQrt{\fRAC{N^\Mu}{2}} \:\BiGgr\{ \biggR[
\frac{\Partial L(..)}{\partiAl \rHo_{\mu|(..,N^\Mu-1,..)}^{\aLpha..}}
\cDot \phI^{\alpHa..}(N) } \\[0.5cm]
&& {\ds + \Frac{\paRtial l(..)}{\pArtial
\rho_{\mu |..}^{\alphA..}} \cdoT \phi^{\alpha..} (..,N^\mu-1,..)
\biGgr]\biGgr\} + {\rm (C.C.)} }\vspacE*{0.6cm}\,,
\End{Array}
\eqNo{\rm (41ii)}$$ $$\DELta_B j^B ({{\MAThBf{n}}},t) | ..]\vspace*{0.5cm}\,,
\eqn o{\rm (40B i)}$$ $$ H : =-P_4 = \ sum\limits_{{{ \ math bf{n}}}=0}^{\infty (3) } [T_ 4^ 4 ({{ \ ma thbf{ n}}},t) +. . ] =\s um \li mi t s_ {{{\m ath bf{n}}} =0}^{\inft y ( 3) } [T_4^4 ({{ \ ma thbf{n}}}, 0)+..] \vspace *{0 .5cm}\ ,,
\e q no{\r m ( 40Bii )}$$
$ \sum\l imits_{{{ \m a thbf{n } }}=0}^{ \ i nf ty ( 3)} := \sum\limit s _{ n ^1=0}^{\infty}
\sum\ li m it s _ {n^ 2=0 }^{\infty} \ sum\l i mits_{n ^ 3= 0 } ^ {\i n fty}.$ 0.25cm
The total ene rgy-mo me ntu m compo nents $ P _\m u$ are give n by incomple te equ a tions ( 4 0A), (4 0Bi,ii ) f ora ge n er al La gr a ngi a n. Ho w eve r, in th efo llowi ng p a p e r , we sh allderiv e [*exact*]{} eq uati o nsfor t he Kl ein- Go rdon, elect ro-ma gn etic, and Dirac fie lds.
Incas eofacompl e x-valu edfie ld $\ph i^{\alp h a.. }, $ t he difference and di ff e r en ce-diffe rentia l c on s ervation f orthec h arge- curr e nt vector$j^\mu $ i sgiven b yequati on s ( A.I I.9A, B ) as : $$\D elta_{\m u} j^ \ mu (n) = 0\vsp a ce*{0.4cm}\,, \e q n o{ \ rm ( 41A i)}$$ $$\be gin{ a rray }{rc l }j^\ m u (n) &:=& { \ ds ie \sqrt{\frac{n^\m u} {2}} \ :\Big gr\{ \Biggr[\frac{\par t i a l L(..)} {\pa r ti a l \rho_{\mu|(. .,n^\ mu-1,..)}^ { \alpha.. }}
\c dot \phi ^{\alpha. . } (n) } \\ [0. 5cm ]
& & { \ d s+ \frac{\part i a l L( .. )}{\par tia l
\rho_ {\m u | ..} ^{\ al pha..}} \ cdot \ph i^ {\ al ph a.. } (.. , n^\mu-1, .. )
\ Bi ggr ]\Big g r\} +{\rm(c.c .) }} \vs pace*{0 . 6c m } \,,\e nd {arr ay}
\ eqno{ \rm( 41i i)}$$ $ $\Delta_b j^ b ({{ \m at hbf{n}} },t) | ..]\vspace*{0.5cm}\,,
\eqno{\rm (40Bi)}$$_$$H :=_-P_4 = \sum\limits_{{{\mathbf{n}}}=0}^{\infty (3)}_[T_4^4 ({{\mathbf{n}}},t)
+..]_=_\sum\limits_{{{\mathbf{n}}}=0}^{\infty (3)}_[T_4^4_({{\mathbf{n}}},0)
+..] \vspace*{0.5cm}\,,
\eqno{\rm (40Bii)}$$
$\sum\limits_{{{\mathbf{n}}}=0}^{\infty_(3)} := \sum\limits_{n^1=0}^{\infty}
\sum\limits_{n^2=0}^{\infty}_\sum\limits_{n^3=0}^{\infty}.$ 0.25cm
The total energy-momentum_components $P_\mu$ are_given_by incomplete equations (40A), (40Bi,ii) for a general Lagrangian. However, in the following paper,_we_shall derive_[*exact*]{}_equations_for the Klein-Gordon, electro-magnetic, and_Dirac fields.
In case of a_complex-valued field_$\phi^{\alpha..},$ the difference and difference-differential conservation for the_charge-current_vector $j^\mu$ is_given by equations (A.II.9A,B) as: $$\Delta_{\mu} j^\mu (n) =_0\vspace*{0.4cm}\,,
\eqno{\rm (41Ai)}$$ $$\begin{array}{rcl}
j^\mu (n) &:=& {\ds_ie \sqrt{\frac{n^\mu}{2}} \:\Biggr\{_\Biggr[
\frac{\partial_L(..)}{\partial_\rho_{\mu|(..,n^\mu-1,..)}^{\alpha..}}
\cdot \phi^{\alpha..}(n) } \\[0.5cm]
&&_{\ds + \frac{\partial L(..)}{\partial
\rho_{\mu |..}^{\alpha..}} \cdot_\phi^{\alpha..} (..,n^\mu-1,..)
\Biggr]\Biggr\} + {\rm (c.c.)} }\vspace*{0.6cm}\,,
\end{array}
\eqno{\rm_(41ii)}$$ $$\Delta_b j^b ({{\mathbf{n}}},t) |
i}^{e} < \upsilon_{i}^{e}$, he would still lose and his utility $\hat{\varphi}_i^e = 0 = \varphi_i^e$. If instead he reports $\hat{\upsilon}_{i}^{e} > \upsilon_{i}^e$. Two cases can occur. If he still loses his utility $\hat{\varphi}_i^e = 0 = \varphi_i^e$. But if he wins, then he had to beat some valuation $\upsilon_{j}^{e} > \upsilon_{i}^e$ and hence $\hat{\upsilon}_{i}^{e} > \upsilon_{j}^{e}$. Now as he wins his utility $\hat{\varphi}_i^{e} = \mathcal{P}_{i}^{e} - \upsilon_i^e = \upsilon_j^e -
\upsilon_i^e < 0$. So he would have got a negative utility. Hence no gain is achieved.\
Considering the case 1 and case 2 above, it can be concluded that any agent *i* can’t gain by mis-reporting his bid value. The proof is carried out by considering the *task executers*, similar argument can be given for the *task requesters*. This completes the proof.
\[l3\] STEM is weakly Budget balanced.
Fix the time slot $\tau_i$ and cluster $\mathsterling_{j}^i$. This corresponds to the case when the sum of all the monetary transfers of all the agents type profiles is less than or equal to *zero*. Now, the construction of our STEM is such that, any *task executer* and *task requester* is paired up only when $\mathcal{S}_i \cdot \mathcal{P}_i^{e} - \mathcal{B}_i \cdot \mathcal{P}_i^r \geq 0$. It means that, for any *task executer*-*task requester* pair there exist some surplus. In the similar fashion, in a particular time slot $\tau_i$ and in a particular cluster considering all the agents, $\sum_{i}\mathcal{S}_i \cdot \mathcal{P}_i^{e} - \sum_{i}\mathcal{B}_i \cdot \mathcal{P}_i^r \geq 0$. Hence, the sum total of payments made to the *task executers* is at least as high as the sum total of the payments received by the * | i}^{e } < \upsilon_{i}^{e}$, he would still lose and his utility $ \hat{\varphi}_i^e = 0 = \varphi_i^e$. If instead he report $ \hat{\upsilon}_{i}^{e } > \upsilon_{i}^e$. Two case can occur. If he still loses his utility program $ \hat{\varphi}_i^e = 0 = \varphi_i^e$. But if he wins, then he had to beat some evaluation $ \upsilon_{j}^{e } > \upsilon_{i}^e$ and therefore $ \hat{\upsilon}_{i}^{e } > \upsilon_{j}^{e}$. Now as he wins his utility $ \hat{\varphi}_i^{e } = \mathcal{P}_{i}^{e } - \upsilon_i^e = \upsilon_j^e -
\upsilon_i^e < 0$. So he would have got a minus utility. Hence no profit is achieved.\
regard the case 1 and case 2 above, it can be concluded that any agent * i * ca n’t gain by myocardial infarction - reporting his bid value. The proof is carry out by considering the * task executers *, similar argument can be pass for the * task requesters *. This completes the proof.
\[l3\ ] STEM is weakly Budget balanced.
Fix the meter slot $ \tau_i$ and cluster $ \mathsterling_{j}^i$. This corresponds to the case when the sum of all the monetary transfers of all the agents type profiles is less than or equal to * zero *. Now, the construction of our STEM is such that, any * task executer * and * task requester * is paired up only when $ \mathcal{S}_i \cdot \mathcal{P}_i^{e } - \mathcal{B}_i \cdot \mathcal{P}_i^r \geq 0$. It means that, for any * task executer*-*task requester * pair there exist some excess. In the alike fashion, in a finical meter slot $ \tau_i$ and in a particular cluster regard all the agents, $ \sum_{i}\mathcal{S}_i \cdot \mathcal{P}_i^{e } - \sum_{i}\mathcal{B}_i \cdot \mathcal{P}_i^r \geq 0$. Hence, the sum total of requital made to the * task executers * is at least as high as the sum total of the payment received by the * | i}^{e} < \upsilon_{i}^{e}$, he would stilu lose and his otulity $\iat{\varpgi}_i^e = 0 = \xarphi_i^e$. If instead he reporvs $\hqt{\upsulon}_{i}^{e} > \upsilon_{i}^e$. Two zases can occur. Id he wtill loses his utljity $\mat{\vaxpii}_i^e = 0 = \varphi_i^g$. But if he fins, then he hdd tl beat some valuation $\upsilon_{j}^{e} > \upfilon_{i}^e$ ajd hence $\hat{\upfilom}_{y}^{e} > \hisllon_{j}^{e}$. Now as he wins his utilitg $\hat{\vagphi}_i^{e} = \mathcal{P}_{i}^{r} - \upsilon_i^e = \upsilon_j^e -
\uosilln_i^e < 0$. So he would have got a negwrive utility. Hence no gain is achigved.\
Considering the case 1 and care 2 ayove, it can bw cltcluded thav any wgent *i* can’t gain by mis-repprting his bid vanue. The proof is carried out by considerind the *tasn zxecuters*, similar argymwnt cdn ba gixwn wor tie *fask rfquxsters*. This completes rhe proof.
\[l3\] STEM is eewjly Budget bamanced.
Siv the time slot $\tau_i$ and cluster $\mathsttrlinf_{j}^i$. This corresponds to the case when the sul of all ehe monetary transfers of all the agents type proxiles ks oefr tjan or equal to *zero*. Now, the construction of jhr SNEM is such that, cny *task executet* wnc *task requestgr* is pcjrsd up only when $\mahhcal{S}_i \cdot \mathcal{P}_y^{e} - \kathcal{B}_i \cdot \mathcal{P}_i^r \gwq 0$. It means rhat, for any *task zxecuter*-*task reqoester* pair there exist some rurpmus. In the dimilar fzrhion, in a partizulsr time slot $\tau_i$ and in a pwrticular cluxter covsidgring ajl the agejts, $\smk_{i}\mathcal{S}_i \cdot \mwthcap{P}_h^{e} - \sum_{i}\mahhcal{B}_i \cdot \mathcal{P}_i^r \geq 0$. Heire, the sum tojal of paymentf madc to the *task evecuters* is at least cs higf as the slm total mf the paymqnts received hy the * | i}^{e} < \upsilon_{i}^{e}$, he would still lose utility = 0 \varphi_i^e$. If instead Two can occur. If still loses his $\hat{\varphi}_i^e = 0 = \varphi_i^e$. But he wins, then he had to beat some valuation $\upsilon_{j}^{e} > \upsilon_{i}^e$ and $\hat{\upsilon}_{i}^{e} > \upsilon_{j}^{e}$. Now as he wins his utility $\hat{\varphi}_i^{e} = \mathcal{P}_{i}^{e} - = - < So he would have got a negative utility. Hence no gain is achieved.\ Considering the case and case 2 above, it can be concluded any agent *i* can’t by mis-reporting his bid value. proof carried out considering *task similar argument can given for the *task requesters*. This completes the proof. \[l3\] STEM is weakly Budget balanced. Fix the slot $\tau_i$ $\mathsterling_{j}^i$. This to case the sum of monetary transfers of all the agents less than or equal to *zero*. Now, the of our is such that, any *task executer* *task requester* is paired up only when $\mathcal{S}_i \mathcal{P}_i^{e} - \mathcal{B}_i \cdot \mathcal{P}_i^r \geq 0$. It means that, for any *task executer*-*task requester* exist some surplus. In similar fashion, in particular slot and a particular considering all the agents, $\sum_{i}\mathcal{S}_i \cdot \mathcal{P}_i^{e} - \sum_{i}\mathcal{B}_i \cdot \mathcal{P}_i^r 0$. Hence, the sum total of payments made to the is least as high the sum total of payments by the * | i}^{e} < \upsilon_{i}^{e}$, he would still loSe and his utIlity $\Hat{\VarPhI}_i^e = 0 = \vArphI_i^e$. If instead he REporTs $\hat{\upsilon}_{i}^{e} > \upsilon_{i}^E$. Two cAsES can OCcUr. If hE still lOSeS HIs uTiLiTy $\hAt{\VArPhi}_i^e = 0 = \VarPhi_i^e$. BuT if he wins, tHen He Had to beat somE VaLuation $\upsIloN_{j}^{e} > \upsilon_{i}^e$ And Hence $\hAt{\UpsILon}_{i}^{e} > \UpsIlon_{j}^{E}$. Now as HE wins hIs utility $\HaT{\Varphi}_I^{E} = \mathcaL{p}_{I}^{e} - \UpsiLon_i^e = \upsilon_j^e -
\upsILoN_I^e < 0$. So he would havE got a nEgATiVE UtiLitY. Hence no gaIn Is achIEved.\
ConSIdERINg tHE case 1 and case 2 aBove, it can be COncLuded tHaT anY Agent *i* Can’t gAiN By mIs-reporting His bId value. ThE proof IS carrieD Out by coNsiderIng The *Task EXeCuTerS*, sIMilAR aRguMEnt Can be givEn FoR the *tAsk rEQUESterS*. ThIs coMpletEs the proof.
\[l3\] STeM iS weaKLy BUdget BalanCed.
FIx The tiMe slot $\Tau_i$ aNd Cluster $\mathsterLing_{J}^i$. This corResPoNds To The caSE when tHe sUm oF all the MonetarY TraNsFERS oF all the agents type pRoFILeS is less tHan or eQUaL tO *Zero*. Now, tHe ConStruCTIon of Our SteM Is such thAt, any *tASk ExEcuter* aNd *Task reQuEstEr* iS pairED up oNly wheN $\mathcal{s}_i \cdoT \Mathcal{P}_i^{e} - \mathCAl{B}_i \cdot \mathcAL{P}_I^R \GeQ 0$. it meAns That, for any *tAsk eXEcutEr*-*taSK rEquESter* pAir thErE ExISt some surplus. In the sImIlar faShion, In a particular Time slot $\taU_I$ ANd in a parTicuLAr CLuster considerIng alL the agents, $\SUm_{i}\mathcAl{S}_i \cDot \mathcAl{P}_i^{e} - \sum_{i}\MAThcal{B}_i \cDot \MatHcaL{P}_i^R \GEq 0$. hence, the sum toTAL of pAyMents maDe tO the *tasK exEcuTerS* is At Least as hiGh as the sUm ToTaL oF thE paymENts receiVeD by ThE * | i}^{e} < \upsilon_{i}^{e}$ , he would stil l l ose a nd h is u tility $\hat{\ v arph i}_i^e = 0 = \varphi_i ^e$.If inst e ad he r eports$ \h a t {\u ps il on} _{ i }^ {e} > \u psilon_ {i}^e$. Tw o c as es can occur . I f he still lo ses his util ity $\hat {\ var p hi}_i ^e= 0 = \varp h i_i^e$ . But ifhe wins,t hen heh a dto b eat some valuatio n $ \ upsilon_{j}^{e } > \u ps i lo n _ {i} ^e$ and hence $ \hat{ \ upsilon } _{ i } ^ {e} > \upsilon_{j }^{e}$. Now ashe win shis utilit y $\h at { \va rphi}_i^{e} = \ mathcal{P }_{i}^ { e} - \u p silon_i ^e = \ ups ilo n_j^ e - \up si l on_ i ^e <0 $.So he wo ul dhavegota n e gati veutil ity.Hence no gain is ach i eve d.\
C onsid erin gthe c ase 1and c as e 2 above, it c an b e conclud edth atan y age n t *i*can ’tgain by mis-re p ort in g h is bid value. The pr oo f is carried out b y c on s ideringth e * task e xecut ers* , s imilar a rgumen t c an be giv en for t he *t ask requ e ster s*. Th is compl etest he proof.
\[l 3 \] STEM is we a kl y Bu d getbal anced.
Fix the time slo t $ \ta u _i$ a nd cl us t er $\mathsterling_{j}^ i$ . This corr esponds to th e case whe n t he sum o f al l t h e monetary tra nsfer s of all t h e agents type profile s is less t han or e qua l t o * zer o * .Now, the cons t r ucti on of our ST EM is s uch th at, an y*task exe cuter* a nd * ta sk re quest e r* is pa ir edup on ly wh e n $\ma thcal {S}_ i\c d ot\mathca l {P } _ i^{e }-\mat hca l{ B}_i\cdo t \m athcal{ P}_i^r \g eq0 $. I tme ans tha t, for any *t as k executer *- *ta sk req u e ster* pa ir there exist some sur p lus. In th e sim ilar fashion, in a par tic u lar ti me slo t $\t au _i$ a nd in a p art ic ular clust e r co nside ri ng a ll theagents, $\sum_{i}\ m ath cal{S}_i \cdo t \ math c a l{ P}_ i ^{ e } - \ s um_ { i }\mathcal{B}_i\cdot \mat hc a l{ P}_i^r \ge q 0$ .Hence,the sum tota l of pay ments mad e to the*t aske x ecu ters* is a t leastas high a s thes um tota l o f thepa yme nts r eceive d by the* | i}^{e} <_\upsilon_{i}^{e}$, he_would still lose and_his utility_$\hat{\varphi}_i^e_= 0_=_\varphi_i^e$. If instead_he reports $\hat{\upsilon}_{i}^{e}_> \upsilon_{i}^e$. Two cases_can occur. If_he_still loses his utility $\hat{\varphi}_i^e = 0 = \varphi_i^e$. But if he wins, then_he_had to_beat_some_valuation $\upsilon_{j}^{e} > \upsilon_{i}^e$ and_hence $\hat{\upsilon}_{i}^{e} > \upsilon_{j}^{e}$. Now_as he_wins his utility $\hat{\varphi}_i^{e} = \mathcal{P}_{i}^{e} - \upsilon_i^e_=_\upsilon_j^e -
\upsilon_i^e_< 0$. So he would have got a negative_utility. Hence no gain is achieved.\
Considering_the case 1_and_case_2 above, it can_be concluded that any agent *i*_can’t gain by mis-reporting his bid_value. The proof is carried out by_considering the *task executers*, similar argument_can be given for the_*task requesters*._This completes the proof.
\[l3\] STEM_is weakly Budget_balanced.
Fix the_time slot $\tau_i$_and cluster $\mathsterling_{j}^i$. This corresponds to_the case when_the sum of all the monetary_transfers_of all the_agents_type_profiles is_less than or_equal_to *zero*._Now,_the construction of our STEM is_such_that, any *task executer* and *task requester*_is paired up only_when_$\mathcal{S}_i \cdot \mathcal{P}_i^{e} -_\mathcal{B}_i \cdot \mathcal{P}_i^r \geq 0$._It means that, for any *task_executer*-*task requester*_pair there_exist some surplus. In the similar fashion, in a particular time_slot $\tau_i$ and in a particular_cluster considering all the_agents, $\sum_{i}\mathcal{S}_i_\cdot_\mathcal{P}_i^{e} - \sum_{i}\mathcal{B}_i_\cdot_\mathcal{P}_i^r \geq_0$. Hence, the sum total of payments_made to_the *task executers* is at least_as high as the_sum_total of the payments received by_the * |
“[The Web Origin Concept](http://tools.ietf.org/html/draft-abarth-origin)”, IETF Draft, 2010 A. Preite Martinez et al. “[An IVOA Standard for Unified Content Descriptors](http://www.ivoa.net/Documents/latest/UCD.html)”, IVOA Recommendation, 2007 R. J. Hanisch et al. “[IVOA Document Standards](http://www.ivoa.net/Documents/latest/DocStd.html)”, IVOA Recommendation, 2010
[^1]: One way a hub might implement this is to generate [*msg-id*]{} by concatenating the sender’s client ID and the [*msg-tag*]{}. When any response is received the hub can then unpack the accompanying [*msg-id*]{} to find out who the original sender was and what [*msg-tag*]{} it used. In this way the hub can determine how to pass each response back to its correct sender without needing to maintain internal state concerning messages in progress. Hub and client implementations may wish to exploit this freedom in assigning message IDs for other purposes as well, for instance to incorporate timestamps or checksums.
[^2]: Note to Java developers: contrary to what you might expect, the [user.home]{} system property on Windows does [*not*]{} give you the value of [USERPROFILE]{}. See <http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=4787931>.
[^3]: Support for the crossdomain.xml file is reportedly implemented in Java v1.6.0\_10 and later, see <http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6676256>.
[^4]: See for instance the flXHR library at <http://flxhr.flensed.com/>.
[^5]: See for example <http://secunia.com/advisories/22467/>, which refers to a Flash version from 2006. Hopefully browsers and plugins in current use do not contain such vulnerabilities, but an assurance of this is beyond the scope of this document.
---
abstract: 'Optical stochastic cooling (OSC) is expected to enable fast cooling of dense particle beams. Transition from microwave to optical frequencies enables an achievement of stochastic cooling rates which are orders of magnitude higher than ones achievable with the classical | “ [ The Web Origin Concept](http://tools.ietf.org / html / draft - abarth - origin) ”, IETF Draft, 2010 A. Preite Martinez et al. “ [ An IVOA Standard for Unified Content Descriptors](http://www.ivoa.net / Documents / latest / UCD.html) ”, IVOA Recommendation, 2007 R. J. Hanisch et al. “ [ IVOA Document Standards](http://www.ivoa.net / Documents / belated / DocStd.html) ”, IVOA Recommendation, 2010
[ ^1 ]: One direction a hub might implement this is to generate [ * msg - id * ] { } by concatenate the sender ’s client ID and the [ * monosodium glutamate - tag * ] { }. When any response is received the hub can then unpack the attach to [ * msg - id * ] { } to find out who the original transmitter was and what [ * msg - tag * ] { } it used. In this way the hub can settle how to pass each response back to its correct sender without needing to observe internal state concerning message in progress. Hub and client implementations may wish to exploit this exemption in assigning message id for other purposes as well, for example to incorporate timestamps or checksums.
[ ^2 ]: Note to Java developers: contrary to what you might ask, the [ user.home ] { } system property on Windows does [ * not * ] { } give you the value of [ USERPROFILE ] { }. See < http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=4787931 >.
[ ^3 ]: Support for the crossdomain.xml file is reportedly implemented in Java v1.6.0\_10 and later, see < http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6676256 >.
[ ^4 ]: See for instance the flXHR library at < http://flxhr.flensed.com/ >.
[ ^5 ]: attend for example < http://secunia.com/advisories/22467/ >, which refers to a Flash interpretation from 2006. Hopefully browsers and plugins in current use do not contain such vulnerabilities, but an assurance of this is beyond the oscilloscope of this document.
---
abstract:' Optical stochastic cooling (OSC) is expected to enable firm cooling of dense particle beams. Transition from microwave to optical frequencies enable an achievement of stochastic cooling rates which are orders of magnitude higher than ones accomplishable with the authoritative | “[Thf Web Origin Concept](http://uools.ietf.org/html/dtadt-abarvh-origih)”, IETF Dfaft, 2010 A. Preite Martinez et al. “[En ICOA Sucndard for Unified Cuntent Dedcriptorw](httk://www.ivoa.net/Documeifs/latesb/BCD.htjp)”, IVME Recommendatiok, 2007 R. J. Haniscv et al. “[IVOA Dmcjmznt Standards](http://www.ivoa.net/Documents/jatest/DpcDtd.html)”, IVOA Rgcommtndwtioh, 2010
[^1]: One way a hub might implement tgis is uo generate [*msg-id*]{} ny concatenating the sendeg’s cpient ID and the [*mdg-tag*]{}. When qny wwsponse is rdceived tht kub can theh unpack the accompanying [*msg-id*]{} to fnnd out who tye ltiginal sendxr was and what [*msn-nag*]{} it gsed. In this way the mub cen dwtermine how to pass xach response back tj its corseet sender without neeeibg to maittaiv ingerhak atate fonrerning mesaages in prigress. Hub and cliemt pkplementatiohs may wysh to exploit this freedom in assigninc msssage IDs for other pueposes as well, for indtance to incorporate timestamps or checksums.
[^2]: Note to Java devemupexw: congeagy to what you might expect, the [user.home]{} systqj kroierty on Windows boes [*not*]{} give yoi hhr value of [USETPROFILZ]{}. Sse <http://bugs.sun.com/bkgdatabwse/viww_bug.do?bud_id=4787931>.
[^3]: Xupport for the crossdomain.zml file is gepoetedly implemented in Java v1.6.0\_10 cnd lajer, ser <http://bugs.sun.com/bugdatayase/visw_bug.do?bug_if=6676256>.
[^4]: See for knstance the flXFR kitrary at <http://flxhr.flensed.cjm/>.
[^5]: See foc exakple <htgp://sevunia.cjm/advisorifs/22467/>, whlwh refers to a Fladh vetsion xrom 2006. Hopevully browsers and plugins in cncrent use do mod cmntain sbch vuknerabilities, but an assurauce of tkis is beyond thv scope oh this documqnt.
---
abstract: 'Mktical stochavtic coojing (OSC) is expdzted to enable fast cooling of debse particle beams. Trxhsition from miexoqave to optical frdquqnbiex qtables an acvievdmevy of rtochasunc goouing rates which are ordars kf magnitude highet bhan ones achievafle with the vlassical | “[The Web Origin Concept](http://tools.ietf.org/html/draft-abarth-origin)”, IETF Draft, 2010 Martinez al. “[An Standard for Unified R. Hanisch et al. Document Standards](http://www.ivoa.net/Documents/latest/DocStd.html)”, IVOA 2010 [^1]: One way a hub implement this is to generate [*msg-id*]{} by concatenating the sender’s client ID and [*msg-tag*]{}. When any response is received the hub can then unpack the accompanying to out the sender was and what [*msg-tag*]{} it used. In this way the hub can determine how to each response back to its correct sender without to maintain internal state messages in progress. Hub and implementations wish to this in message IDs for purposes as well, for instance to incorporate timestamps or checksums. [^2]: Note to Java developers: contrary to you might [user.home]{} system on does give you the [USERPROFILE]{}. See <http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=4787931>. [^3]: Support for is reportedly implemented in Java v1.6.0\_10 and later, <http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6676256>. [^4]: for instance the flXHR library at [^5]: See for example <http://secunia.com/advisories/22467/>, which refers to Flash version from 2006. Hopefully browsers and plugins in current use do not contain such an assurance of this beyond the scope this --- 'Optical cooling (OSC) expected to enable fast cooling of dense particle beams. Transition from to optical frequencies enables an achievement of stochastic cooling rates orders magnitude higher than achievable with the classical | “[The Web Origin Concept](http://toOls.ietf.org/Html/dRafT-abArTh-orIgin)”, iETF Draft, 2010 A. PreiTE MarTinez et al. “[An IVOA StandarD for UNiFIed CONtEnt DeScriptoRS](hTTP://wwW.iVoA.neT/DOCuMents/LatEst/UCD.hTml)”, IVOA RecOmmEnDation, 2007 R. J. HaniSCh Et al. “[IVOA DoCumEnt Standards](HttP://www.ivOa.Net/dOcumeNts/LatesT/DocStD.Html)”, IVoA RecommeNdATion, 2010
[^1]: OnE Way a hub MIGhT impLement this is to genERaTE [*msg-id*]{} by concatEnatinG tHE sENDer’S clIent ID and tHe [*Msg-taG*]{}. when any REsPONSe iS Received the huB can then unpACk tHe accoMpAnyINg [*msg-iD*]{} to fiNd OUt wHo the originAl seNder was anD what [*mSG-tag*]{} it uSEd. In thiS way thE huB caN detERmInE hoW tO PasS EaCh rESpoNse back tO iTs CorreCt seNDER WithOut NeedIng to Maintain interNal StatE ConCerniNg mesSageS iN progRess. HuB and cLiEnt implementatiOns mAy wish to eXplOiT thIs FreedOM in assIgnIng Message iDs for oTHer PuRPOSeS as well, for instance To INCoRporate tImestaMPs Or CHecksums.
[^2]: noTe tO JavA DEveloPers: COnTrary to wHat you MIgHt Expect, tHe [User.hoMe]{} SysTem PropeRTy on windowS does [*not*]{} Give yOU the value of [USErpROFILE]{}. See <httP://BuGS.SuN.Com/bUgdAtabase/view_Bug.dO?Bug_iD=4787931>.
[^3]: SupPOrT foR The crOssdoMaIN.xML file is reportedly imPlEmenteD in JaVa v1.6.0\_10 and later, seE <http://bugs.sUN.COm/bugdatAbasE/ViEW_bug.do?bug_id=6676256>.
[^4]: See For inStance the fLxHR libraRy at <hTtp://flxhr.Flensed.coM/>.
[^5]: sEe for exaMplE <htTp://sEcuNIA.cOm/advisories/22467/>, wHICh reFeRs to a FlAsh Version FroM 2006. HoPefUllY bRowsers anD plugins In CuRrEnT usE do noT Contain sUcH vuLnEraBilitIEs, but aN assuRancE oF tHIs iS beyond THe SCOpe oF tHiS docUmeNt.
---
AbstrAct: 'OPTicAl stochAstic coolIng (osC) is ExPeCted to eNable fast coolInG of dense paRtIclE beams. tRAnsition From microwave to optical fREquenciEs eNableS an aChievemenT of StochaStiC CoolinG rates Which ArE orDERs of mAGNiTudE hIgher than oNES acHievaBlE witH the claSsical | “[The Web Origin Concept] (http://to ols.i etf .or g/ html /dra ft-abarth-orig i n)”, IETF Draft, 2010 A. P reite M a rtin e zet al . “[AnI VO A Sta nd ar d f or Un ified Co ntent D escriptors ](h tt p://www.ivoa . ne t/Document s/l atest/UCD.ht ml) ”, IVO ARec o mmend ati on, 2 007 R. J. Han isch et a l. “[IVOA Documen t St anda rds](http://www.i v oa . net/Documents/ latest /D o cS t d .ht ml) ”, IVOA Re co mmend a tion, 2 0 10 [ ^1] : One way a hu b might imp l eme nt thi sist o gene rate[* m sg- id*]{} by c onca tenatingthe se n der’s c l ient ID and t he[*m sg-t a g* ]{ }.Wh e n a n yres p ons e is rec ei ve d the hub c a n the n u npac k the accompanying [* msg- i d*] {} to find out w ho th e orig inalse nder was and wh at [ *msg-tag* ]{} i t u se d. In this w aythe hub ca n deter m ine h o w to pass each respons eb a ck to itscorrec t s en d er witho ut ne edin g to ma inta i ninternal state co nc erningme ssages i n p rog ress. Huband cl ient imp lemen t ations may wis h to exploit t h is f re e dominassigning m essa g e ID s fo r o the r purp osesas we l l, for instance toin corpor ate t imestamps orchecksums. [ ^2]: Not e to Ja v a developers:contr ary to wha t you mig ht ex pect, th e [user.h o m e]{} sys tem pr ope rty o nWindows does[ * not* ]{ } giveyou the va lue of [U SER PR OFILE]{}. See <ht tp :/ /b ug s.s un.co m /bugdata ba se/ vi ew_ bug.d o ?bug_i d=478 7931 >.
[ ^3] : Suppo r tf o r th ecr ossd oma in .xmlfile isreporte dly imple men t ed i nJa va v1.6 .0\_10 and la te r, see <ht tp :// bugs.s u n .com/bug database/view_bug.do?bu g _id=667 625 6>.
[^4] : See for in stance th e flXHR libra ry at < htt p : //flx h r .f len se d.com/>.
[ ^ 5]: Seefo r ex ample < http://secunia.com / adv isories/22467 />, whi c h r efe r st o a F l ash v ersion from 200 6. Hopeful ly br owsers and plu gi ns in c urrentuse d o not co ntain suc h vulnera bi liti e s , b ut an assu rance of this isb eyond th e sco peof thi sdoc ument .
--- abs tract : 'Opt ic al sto chast ic cooling (OSC) is expected to e nablefastcoo ling of d ens e pa rticle be ams. Transitio n f rom micr owa v e toopti c al fr e quenc iese nables an ac hie v e me nt of stoch a s t iccooli ngr ates w hich are orders of ma g nitude higherthan o nes ac h ieva bl e with the cla ssi ca l | “[The_Web Origin_Concept](http://tools.ietf.org/html/draft-abarth-origin)”, IETF Draft, 2010_A. Preite Martinez et_al._“[An IVOA_Standard_for Unified Content_Descriptors](http://www.ivoa.net/Documents/latest/UCD.html)”, IVOA Recommendation,_2007 R. J. Hanisch et al._“[IVOA Document Standards](http://www.ivoa.net/Documents/latest/DocStd.html)”,_IVOA_Recommendation, 2010
[^1]: One way a hub might implement this is to generate [*msg-id*]{} by_concatenating_the sender’s_client_ID_and the [*msg-tag*]{}. When any_response is received the hub_can then_unpack the accompanying [*msg-id*]{} to find out who_the_original sender was_and what [*msg-tag*]{} it used. In this way the_hub can determine how to pass_each response back_to_its_correct sender without needing_to maintain internal state concerning messages_in progress. Hub and client implementations_may wish to exploit this freedom in_assigning message IDs for other purposes_as well, for instance to_incorporate timestamps_or checksums.
[^2]: Note to Java_developers: contrary to_what you_might expect, the_[user.home]{} system property on Windows does_[*not*]{} give you_the value of [USERPROFILE]{}. See <http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=4787931>.
[^3]:_Support_for the crossdomain.xml_file_is_reportedly implemented_in Java v1.6.0\_10_and_later, see_<http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6676256>.
[^4]:_See for instance the flXHR library_at_<http://flxhr.flensed.com/>.
[^5]: See for example <http://secunia.com/advisories/22467/>, which refers_to a Flash version_from_2006. Hopefully browsers and_plugins in current use do_not contain such vulnerabilities, but an_assurance of_this is_beyond the scope of this document.
---
abstract: 'Optical stochastic cooling (OSC)_is expected to enable fast cooling_of dense particle beams._Transition from_microwave_to optical frequencies_enables_an achievement_of stochastic cooling rates which are orders_of magnitude_higher than ones achievable with the_classical |
that the above proof is much shorter and general than the original one by Shafer [@sha76].
As in the classical case, these coefficients may be strictly greater than 1, hence corresponding simple support functions have negative Möbius transform and are no more belief functions.
Necessity functions {#sec:nece}
===================
A function $\N:L\rightarrow[0,1]$ is called a *necessity function* if it satisfies $\N(x\wedge y)=\min(\N(x),\N(y))$, for all $x,y\in L$, and $\N(\bot)=0$, $\N(\top)=1$.
The following result is due to Barthélemy [@bar00].
$\N$ is a necessity function if and only if it is belief function whose Möbius transform $m$ is such that its focal elements form a chain in $L$.
We define possibility functions as $\vee$-conjugates of necessity functions.
Let $L$ be an autodual lattice, and $n$ a $\vee$-negation on $L$. For any necessity function $\N$ on $L$, its $\vee$-conjugate is called a *possibility function*.
Let $\Pi$ be a possibility function. Then ${}^\wedge\overline{\Pi}$ is its corresponding necessity function by Lemma \[lem:neg\] (ii).
Let $L$ be an autodual lattice, and $n$ a $\vee$-negation on $L$. The mapping $\Pi:L\rightarrow [0,1]$ is a possibility function if and only if $$\label{eq:pi}
\Pi(x\vee y) = \max(\Pi(x),\Pi(y)), \quad \forall x,y\in L.$$
Let $\Pi$ be a possibility function being the $\vee$-conjugate of some necessity function $N$. Then: $$\begin{aligned}
\Pi(x\vee y) & = 1-\N(n(x\vee y)) = 1- \N(n(x)\wedge n(y))\\
& = 1- \min(\N(n(x)),\N(n(y))) =
\max(1-\N(n(x)),1-\N(n(y)))\\
& = \max(\Pi(x),\Pi( | that the above proof is much shorter and general than the original one by Shafer [ @sha76 ].
As in the authoritative subject, these coefficients may be strictly great than 1, therefore corresponding simple support function have negative Möbius transform and are no more impression function.
Necessity functions { # sec: nece }
= = = = = = = = = = = = = = = = = = =
A routine $ \N: L\rightarrow[0,1]$ is called a * necessity function * if it satisfy $ \N(x\wedge y)=\min(\N(x),\N(y))$, for all $ x, y\in L$, and $ \N(\bot)=0 $, $ \N(\top)=1$.
The follow result is due to Barthélemy [ @bar00 ].
$ \N$ is a necessity function if and only if it is belief affair whose Möbius transform $ m$ is such that its focal elements form a chain in $ L$.
We specify possibility functions as $ \vee$-conjugates of necessity functions.
Let $ L$ be an autodual wicket, and $ n$ a $ \vee$-negation on $ L$. For any necessity function $ \N$ on $ L$, its $ \vee$-conjugate is shout a * possibility function *.
Let $ \Pi$ be a hypothesis function. Then $ { } ^\wedge\overline{\Pi}$ is its corresponding necessity function by Lemma \[lem: neg\ ] (ii).
Let $ L$ be an autodual lattice, and $ n$ a $ \vee$-negation on $ L$. The mapping $ \Pi: L\rightarrow [ 0,1]$ is a hypothesis affair if and only if $ $ \label{eq: pi }
\Pi(x\vee y) = \max(\Pi(x),\Pi(y) ), \quad \forall x, y\in L.$$
Let $ \Pi$ be a possibility function being the $ \vee$-conjugate of some necessity function $ N$. Then: $ $ \begin{aligned }
\Pi(x\vee y) & = 1-\N(n(x\vee y) ) = 1- \N(n(x)\wedge n(y))\\
& = 1- \min(\N(n(x)),\N(n(y) )) =
\max(1-\N(n(x)),1-\N(n(y)))\\
& = \max(\Pi(x),\Pi ( | thwt the above proof is mugh shorter and ggnwral tian the originau one by Shafer [@sha76].
As in the coassixal case, these coeffickents may be strixtly treater theh 1, hencc cordcsponbiig simple suppott functions have negative Möcibs transform and are no more belief sunctioms.
Jecessity funcjions {#fec:nsbe}
===================
C function $\N:L\rightarrow[0,1]$ is callsd a *nebessity function* of it satisfies $\N(x\wedge y)=\mln(\N(x),\J(y))$, for all $x,y\in L$, wnd $\N(\bot)=0$, $\N(\tip)=1$.
Thq following rdsult is dlz to Barthélgmy [@bar00].
$\N$ is a necessity function if aud only if ut is telief funcvion wrose Möbius tvsnsfork $m$ is xuch that its nocal elwments form a chain ii $L$.
We define possibijity funcdikns as $\vee$-conjugarew of tecevsith fuvctjois.
Lst $L$ bf ai autodual mattice, and $n$ a $\vee$-negation on $L$. Dor any necesaity ftnstion $\N$ on $L$, its $\vee$-conjugate is called a *lossibility function*.
Let $\Pi$ be a possibility vunction. Ehen ${}^\wedge\overline{\Pi}$ is its corresponding necessidy fuicgiou by Ldnmw \[lem:neg\] (ii).
Let $L$ be an autodual lattice, and $n$ z $\fev$-negation on $L$. Thc mapping $\Pi:L\rightsrgoe [0,1]$ is a possibklity yhndtion if and only lf $$\labej{eq:pi}
\Pi(x\vee y) = \max(\Li(x),\Pi(y)), \quad \forall x,y\in L.$$
Ler $\Pi$ be a pofwibility function yeing the $\vez$-conjubate pf some necessity functnon $N$. Fhen: $$\begin{apigned}
\Pi(x\bde y) & = 1-\N(n(x\vee y)) = 1- \N(k(x)\wadge n(y))\\
& = 1- \min(\G(n(x)),\N(n(y))) =
\oax(1-\N(m(x)),1-\N(n(y)))\\
& = \max(\Pi(x),\Pi( | that the above proof is much shorter than original one Shafer [@sha76]. As coefficients be strictly greater 1, hence corresponding support functions have negative Möbius transform are no more belief functions. Necessity functions {#sec:nece} =================== A function $\N:L\rightarrow[0,1]$ is a *necessity function* if it satisfies $\N(x\wedge y)=\min(\N(x),\N(y))$, for all $x,y\in L$, and $\N(\top)=1$. following is to Barthélemy [@bar00]. $\N$ is a necessity function if and only if it is belief function Möbius transform $m$ is such that its focal form a chain in We define possibility functions as of functions. Let be autodual and $n$ a on $L$. For any necessity function $\N$ on $L$, its $\vee$-conjugate is called a *possibility function*. Let be a Then ${}^\wedge\overline{\Pi}$ its necessity by Lemma \[lem:neg\] $L$ be an autodual lattice, and on $L$. The mapping $\Pi:L\rightarrow [0,1]$ is a function if only if $$\label{eq:pi} \Pi(x\vee y) = \quad \forall x,y\in L.$$ Let $\Pi$ be a function being the $\vee$-conjugate of some necessity function $N$. Then: $$\begin{aligned} \Pi(x\vee y) & = = 1- \N(n(x)\wedge n(y))\\ = 1- \min(\N(n(x)),\N(n(y))) \max(1-\N(n(x)),1-\N(n(y)))\\ = | that the above proof is much shOrter and geNeral ThaN thE oRigiNal oNe by Shafer [@sha76].
AS In thE classical case, these coeFficiEnTS may BE sTrictLy greatER tHAN 1, heNcE cOrrEsPOnDing sImpLe suppoRt functionS haVe Negative MöbiUS tRansform anD arE no more belieF fuNctionS.
NEceSSity fUncTions {#Sec:necE}
===================
a functIon $\N:L\righTaRRow[0,1]$ is cALled a *neCESsIty fUnction* if it satisfIEs $\n(X\wedge y)=\min(\N(x),\N(y))$, For all $X,y\IN L$, AND $\N(\bOt)=0$, $\N(\Top)=1$.
The follOwIng reSUlt is duE To bARThéLEmy [@bar00].
$\N$ is a necEssity functIOn iF and onLy If iT Is beliEf funCtIOn wHose Möbius tRansForm $m$ is suCh that ITs focal ELements Form a cHaiN in $l$.
We dEFiNe PosSiBIliTY fUncTIonS as $\vee$-coNjUgAtes oF necESSITy fuNctIons.
let $L$ bE an autodual laTtiCe, anD $N$ a $\vEe$-negAtion On $L$. FOr Any neCessitY funcTiOn $\N$ on $L$, its $\vee$-conJugaTe is calleD a *pOsSibIlIty fuNCtion*.
LEt $\PI$ be A possibIlity fuNCtiOn. tHEN ${}^\wEdge\overline{\Pi}$ is itS cORReSponding NecessITy FuNCtion by LEmMa \[lEm:neG\] (II).
Let $L$ Be an AUtOdual latTice, anD $N$ a $\VeE$-negatiOn On $L$. The MaPpiNg $\PI:L\rigHTarrOw [0,1]$ is a pOssibiliTy funCTion if and only iF $$\Label{eq:pi}
\Pi(x\vEE y) = \MAX(\PI(X),\Pi(y)), \QuaD \forall x,y\in l.$$
Let $\pI$ be a PossIBiLitY FunctIon beInG ThE $\Vee$-conjugate of some nEcEssity FunctIon $N$. Then: $$\begin{Aligned}
\Pi(x\VEE Y) & = 1-\N(n(x\vee y)) = 1- \n(n(x)\wEDgE N(y))\\
& = 1- \min(\N(n(x)),\N(n(y))) =
\max(1-\n(n(x)),1-\N(n(Y)))\\
& = \max(\Pi(x),\Pi( | that the above proof is m uch shorte r and ge ner al tha n th e original one by S hafer [@sha76].
As in thecl a ssic a lcase, thesec oe f f ici en ts ma yb estric tly greate r than 1,hen ce correspondi n gsimple sup por t functionshav e nega ti veM öbius tr ansfo rm and are no more bel ie f funct i ons.
N e c es sity functions {#sec: n ec e }
============ ====== =Af u nct ion $\N:L\rig ht arrow [ 0,1]$ i s c a l l eda *necessity f unction* if itsatisf ie s $ \ N(x\we dge y )= \ min (\N(x),\N(y ))$, for all$x,y\i n L$, an d $\N(\b ot)=0$ , $ \N( \top ) =1 $.
T he fol l ow ing res ult is d ue t o Bar thél e m y [@ba r00 ].
$\N$is a necessit y f unct i onif an d onl y if i t isbelief func ti on whose Möbius tra nsform $m $ i ssuc hthati ts foc alele ments f orm a c h ain i n $ L$ .
We define possi bi l i ty functio ns as$ \v ee $ -conjuga te s o f ne c e ssity fun c ti ons.
Le t $L$b ean autodu al latti ce , a nd$n$ a $\ve e$-neg ation on $L$. For any necess i ty function $ \ N$ o n$ L$,its $\vee$-con juga t e is cal l ed a* possi bilit yf un c tion*.
Let $\Pi$ b ea poss ibili ty function.Then ${}^\ w e d ge\overl ine{ \ Pi } $ is its corre spond ing necess i ty funct ion b y Lemma\[lem:neg \ ] (ii).
Let $L $ b e a n au todual lattic e , and $ n$ a $\ vee $-negat ion on $L $.Th e mapping $\Pi:L\ ri gh ta rr ow[0,1] $ is a po ss ibi li tyfunct i on ifand o nlyif $ $ \la bel{eq: p i} \ Pi(x \v ee y)= \ ma x(\Pi (x), \ Pi( y)), \q uad \fora llx ,y\i nL. $$
Let $\Pi$ be a p os sibility f un cti on bei n g the $\v ee$-conjugate of some n e cessity fu nctio n $N $. Then:$$\ begin{ ali g ned}
\ Pi(x\v ee y) & =1 - \N(n( x \ ve e y )) = 1- \N(n ( x )\w edgen( y))\ \
& = 1- \mi n (\N (n(x)),\N(n(y ))) =
\m ax ( 1-\ N ( n(x)),1-\N(n(y) ))\\
& = \m a x(\ Pi (x),\Pi ( | that_the above_proof is much shorter_and general_than_the original_one_by Shafer [@sha76].
As_in the classical_case, these coefficients may_be strictly greater_than_1, hence corresponding simple support functions have negative Möbius transform and are no more_belief_functions.
Necessity functions_{#sec:nece}
===================
A_function_$\N:L\rightarrow[0,1]$ is called a *necessity_function* if it satisfies $\N(x\wedge_y)=\min(\N(x),\N(y))$, for_all $x,y\in L$, and $\N(\bot)=0$, $\N(\top)=1$.
The following result_is_due to Barthélemy_[@bar00].
$\N$ is a necessity function if and only if_it is belief function whose Möbius_transform $m$ is_such_that_its focal elements form_a chain in $L$.
We define possibility_functions as $\vee$-conjugates of necessity functions.
Let_$L$ be an autodual lattice, and $n$_a $\vee$-negation on $L$. For any_necessity function $\N$ on $L$,_its $\vee$-conjugate_is called a *possibility function*.
Let_$\Pi$ be a_possibility function._Then ${}^\wedge\overline{\Pi}$ is_its corresponding necessity function by Lemma_\[lem:neg\] (ii).
Let $L$_be an autodual lattice, and $n$_a_$\vee$-negation on $L$._The_mapping_$\Pi:L\rightarrow [0,1]$_is a possibility_function_if and_only_if $$\label{eq:pi}
\Pi(x\vee y) = \max(\Pi(x),\Pi(y)), \quad_\forall_x,y\in L.$$
Let $\Pi$ be a possibility function_being the $\vee$-conjugate of_some_necessity function $N$. Then:_$$\begin{aligned}
\Pi(x\vee y) & = 1-\N(n(x\vee_y)) = 1- \N(n(x)\wedge n(y))\\
_ _ _ & = 1- \min(\N(n(x)),\N(n(y)))_=
_ _ __\max(1-\N(n(x)),1-\N(n(y)))\\
__ _ _ &_= \max(\Pi(x),\Pi( |
) d\mu d\mathbb{P=}\int\limits_{\Omega }\int\limits_{\left[
0,T\right] \times \mathbb{R}_{0}}\mathbf{1}_{C}\left( s,\omega \right) d\mu
_{\mathbb{P}}^{\mathcal{P}}d\mathbb{P} \\
&=&\int\limits_{\Omega }\int\limits_{\mathbb{R}_{0}}\int\limits_{\left[ 0,T\right] }\mathbf{1}_{C}\left( s,\omega \right) dsd\nu d\mathbb{P=}\lambda
\times \nu \times \mathbb{P}\left[ C\right],\end{aligned}$$and thus $$D_{s-}^{\left( n\right) }\theta _{1}^{\left( n\right) }\left( s,x\right)
\overset{\lambda \times \nu \times \mathbb{P}\text{-a.s.}}{\longrightarrow }D_{s-}\theta _{1}\left( s,x\right).$$
Since $\int\limits_{\Omega \times \left[ 0,T\right] }\left\vert
D_{t-}^{\left( n\right) }-D_{t-}\right\vert d\mathbb{P}\times dt\mathbb{=}\int\limits_{\Omega \times \left[ 0,T\right] }\left\vert D_{t}^{\left(
n\right) }-D_{t}\right\vert d\mathbb{P}\times dt\longrightarrow 0$, we have that $\left\{ D_{t-}^{\left( n\right) }\right\} _{t\in \left[ 0,T\right] }$ $\overset{L^{1}\left( \lambda \times \mathbb{P}\right) }{\longrightarrow }\left\{ D_{t-}\right\} _{t\in \left[ 0,T\right] }$ and $\left\{ D_{t}^{\left( n\right) }\right\} _{t\in \left[ 0,T\right] }$ $\overset{L^{1}\left( \lambda \times \mathbb{P}\right) }{\longrightarrow }\left\{ D_{t}\right\} _{t\in \left[ 0,T\right] }.$ Then, for an arbitrary but fixed subsequence $\left\{ n | ) d\mu d\mathbb{P=}\int\limits_{\Omega } \int\limits_{\left [
0,T\right ] \times \mathbb{R}_{0}}\mathbf{1}_{C}\left (s,\omega \right) d\mu
_ { \mathbb{P}}^{\mathcal{P}}d\mathbb{P } \\
& = & \int\limits_{\Omega } \int\limits_{\mathbb{R}_{0}}\int\limits_{\left [ 0,T\right ] } \mathbf{1}_{C}\left (s,\omega \right) dsd\nu d\mathbb{P=}\lambda
\times \nu \times \mathbb{P}\left [ C\right],\end{aligned}$$and thus $ $ D_{s-}^{\left (n\right) } \theta _ { 1}^{\left (n\right) } \left (s, x\right)
\overset{\lambda \times \nu \times \mathbb{P}\text{-a.s.}}{\longrightarrow } D_{s-}\theta _ { 1}\left (s, x\right).$$
Since $ \int\limits_{\Omega \times \left [ 0,T\right ] } \left\vert
D_{t-}^{\left (n\right) } -D_{t-}\right\vert d\mathbb{P}\times dt\mathbb{=}\int\limits_{\Omega \times \left [ 0,T\right ] } \left\vert D_{t}^{\left (
n\right) } -D_{t}\right\vert d\mathbb{P}\times dt\longrightarrow 0 $, we have that $ \left\ { D_{t-}^{\left (n\right) } \right\ } _ { t\in \left [ 0,T\right ] } $ $ \overset{L^{1}\left (\lambda \times \mathbb{P}\right) } { \longrightarrow } \left\ { D_{t-}\right\ } _ { t\in \left [ 0,T\right ] } $ and $ \left\ { D_{t}^{\left (n\right) } \right\ } _ { t\in \left [ 0,T\right ] } $ $ \overset{L^{1}\left (\lambda \times \mathbb{P}\right) } { \longrightarrow } \left\ { D_{t}\right\ } _ { t\in \left [ 0,T\right ] } .$ Then, for an arbitrary but fixed subsequence $ \left\ { n | ) d\mk d\mathbb{P=}\int\limits_{\Omega }\int\limits_{\left[
0,T\titht] \tikes \mafhbb{R}_{0}}\matfbf{1}_{C}\left( s,\omega \right) d\mu
_{\matibb{P}}^{\nathcql{P}}d\mathbb{P} \\
&=&\int\limits_{\Ooega }\int\lpmits_{\mathvb{R}_{0}}\iit\limits_{\left[ 0,T\rijgt] }\mathny{1}_{C}\leff( s,\omzge \right) dsd\nu d\kathbb{P=}\lamtda
\times \nu \tiker \lathbb{P}\left[ C\right],\end{aligned}$$and thuf $$D_{s-}^{\lefy( j\right) }\theta _{1}^{\lgft( n\gidht) }\mvfu( s,x\right)
\overset{\lambda \times \nu \tjmes \mauhbb{P}\text{-a.s.}}{\longrigntarrow }D_{s-}\theta _{1}\left( s,x\rigjt).$$
Sijce $\int\limits_{\Omega \times \left[ 0,T\ridyt] }\left\vert
D_{g-}^{\left( n\right) }-D_{t-}\right\vett d\mathbb{P}\times dt\mathbb{=}\int\limigs_{\Omeya \times \ledt[ 0,T\rlcht] }\left\verv D_{t}^{\lest(
n\right) }-D_{t}\rlbht\verd d\mathnb{P}\times dt\lonnrighvarriw 0$, we have that $\left\{ V_{t-}^{\left( n\right) }\right\} _{j\in \left[ 0,T\siyht] }$ $\overset{L^{1}\left( \lamvdq \timgs \madhbb{O}\eigft) }{\moigrjghtargow }\left\{ D_{t-}\rifht\} _{t\in \lefr[ 0,T\right] }$ and $\left\{ C_{t}^{\jvgt( n\right) }\rifht\} _{t\ig \jeft[ 0,T\right] }$ $\overset{L^{1}\left( \lambda \times \katgbb{P}\right) }{\longrightarroq }\left\{ D_{t}\right\} _{t\in \levt[ 0,T\right] }.$ Then, for an arbitrary but fixed subsequence $\lefd\{ n | ) d\mu d\mathbb{P=}\int\limits_{\Omega }\int\limits_{\left[ 0,T\right] \times \mathbb{R}_{0}}\mathbf{1}_{C}\left( d\mu \\ &=&\int\limits_{\Omega 0,T\right] }\mathbf{1}_{C}\left( s,\omega \times C\right],\end{aligned}$$and thus $$D_{s-}^{\left( }\theta _{1}^{\left( n\right) s,x\right) \overset{\lambda \times \nu \times \mathbb{P}\text{-a.s.}}{\longrightarrow _{1}\left( s,x\right).$$ Since $\int\limits_{\Omega \times \left[ 0,T\right] }\left\vert D_{t-}^{\left( n\right) }-D_{t-}\right\vert d\mathbb{P}\times dt\mathbb{=}\int\limits_{\Omega \left[ 0,T\right] }\left\vert D_{t}^{\left( n\right) }-D_{t}\right\vert d\mathbb{P}\times dt\longrightarrow 0$, we have that $\left\{ n\right) _{t\in 0,T\right] $\overset{L^{1}\left( \lambda \times \mathbb{P}\right) }{\longrightarrow }\left\{ D_{t-}\right\} _{t\in \left[ 0,T\right] }$ and $\left\{ D_{t}^{\left( n\right) }\right\} \left[ 0,T\right] }$ $\overset{L^{1}\left( \lambda \times \mathbb{P}\right) }{\longrightarrow D_{t}\right\} _{t\in \left[ 0,T\right] Then, for an arbitrary but subsequence n | ) d\mu d\mathbb{P=}\int\limits_{\Omega }\Int\limits_{\lEft[
0,T\rIghT] \tiMeS \matHbb{R}_{0}}\Mathbf{1}_{C}\left( s,\omEGa \riGht) d\mu
_{\mathbb{P}}^{\mathcal{P}}d\MathbB{P} \\
&=&\INt\liMItS_{\OmegA }\int\limITs_{\MAThbB{R}_{0}}\InT\liMiTS_{\lEft[ 0,T\rIghT] }\mathbf{1}_{c}\left( s,\omegA \riGhT) dsd\nu d\mathbB{p=}\lAmbda
\times \Nu \tImes \mathbb{P}\lEft[ c\right],\EnD{alIGned}$$aNd tHus $$D_{s-}^{\Left( n\rIGht) }\theTa _{1}^{\left( n\riGhT) }\Left( s,x\RIght)
\oveRSEt{\LambDa \times \nu \times \matHBb{p}\Text{-a.s.}}{\longrighTarrow }d_{s-}\THeTA _{1}\LefT( s,x\Right).$$
Since $\InT\limiTS_{\Omega \tIMeS \LEFt[ 0,T\RIght] }\left\vert
D_{T-}^{\left( n\right) }-d_{T-}\riGht\verT d\MatHBb{P}\timEs dt\mAtHBb{=}\iNt\limits_{\OmeGa \tiMes \left[ 0,T\rIght] }\leFT\vert D_{t}^{\LEft(
n\rigHt) }-D_{t}\riGht\VerT d\maTHbB{P}\TimEs DT\loNGrIghTArrOw 0$, we have ThAt $\Left\{ D_{T-}^{\lefT( N\RIGht) }\rIghT\} _{t\in \Left[ 0,T\Right] }$ $\overset{L^{1}\LefT( \lamBDa \tImes \mAthbb{p}\rigHt) }{\LongrIghtarRow }\leFt\{ d_{t-}\right\} _{t\in \left[ 0,T\RighT] }$ and $\left\{ D_{T}^{\leFt( N\riGhT) }\righT\} _{T\in \lefT[ 0,T\rIghT] }$ $\overseT{L^{1}\left( \lAMbdA \tIMES \mAthbb{P}\right) }{\longrigHtARRoW }\left\{ D_{t}\rIght\} _{t\iN \LeFt[ 0,t\Right] }.$ TheN, fOr aN arbITRary bUt fiXEd SubsequeNce $\lefT\{ N | ) d\mu d\mathbb{P=}\int\li mits_{\Ome ga }\ int \li mi ts_{ \lef t[
0,T\right]\ time s \mathbb{R}_{0}}\math bf{1} _{ C }\le f t( s,\o mega \r i gh t ) d\ mu
_ {\m at h bb {P}}^ {\m athcal{ P}}d\mathb b{P }\\
&=&\int\l i mi ts_{\Omega }\ int\limits_{ \ma thbb{R }_ {0} } \int\ lim its_{ \left[ 0,T\ri ght] }\ma th b f{1}_{ C }\left( s ,\ omeg a \right) dsd\nud \m a thbb{P=}\lambd a
\tim es \n u \ti mes \mathbb{P }\ left[ C\right ] ,\ e n d {al i gned}$$and th us $$D_{s-} ^ {\l eft( n \r igh t ) }\th eta _ {1 } ^{\ left( n\rig ht)}\left( s ,x\rig h t)
\ove r set{\la mbda \ tim es\nu\ ti me s \ ma t hbb { P} \te x t{- a.s.}}{\ lo ng right arro w } D _{s- }\t heta _{1} \left( s,x\ri ght ).$$
Si nce $ \int\ limi ts _{\Om ega \t imes\l eft[ 0,T\right] }\l eft\vertD_{ t- }^{ \l eft(n \right ) } -D_ {t-}\ri ght\ver t d\ ma t h b b{ P}\times dt\mathbb {= } \ in t\limits _{\Ome g a\t i mes \lef t[ 0, T\ri g h t] }\ left \ ve rt D_{t} ^{\lef t (n\ right)}- D_{t}\ ri ght \ve rt d\ m athb b{P}\t imes dt\ longr i ghtarrow 0$, w e have that $\ l ef t \ {D _{t- }^{ \left( n\ri ght) }\ri ght\ } _ {t\ i n \le ft[ 0 ,T \ ri g ht] }$ $\overset{L^ {1 }\left ( \la mbda \times \ mathbb{P}\ r i g ht) }{\l ongr i gh t arrow }\left\{ D_{t -}\right\} _{t\in \ left[ 0,T\rig ht] }$ an d $\left\{ D_ {t} ^{\ lef t ( n \right) }\rig h t \} _ {t \in \le ft[ 0,T\ri ght ] } $ $ \ov er set{L^{1} \left( \ la mb da \ tim es \m a thbb{P}\ ri ght )}{\ longr i ghtarr ow }\ left \{ D _ {t} \right\ } _ { t \in\l ef t[ 0 ,T\ ri ght]}.$T hen , for a n arbitra ryb ut f ix ed subseq uence $\left\ {n | ) d\mu_d\mathbb{P=}\int\limits_{\Omega }\int\limits_{\left[
0,T\right]_\times \mathbb{R}_{0}}\mathbf{1}_{C}\left( s,\omega \right)_d\mu
_{\mathbb{P}}^{\mathcal{P}}d\mathbb{P} \\
&=&\int\limits_{\Omega_}\int\limits_{\mathbb{R}_{0}}\int\limits_{\left[_0,T\right] }\mathbf{1}_{C}\left(_s,\omega_\right) dsd\nu d\mathbb{P=}\lambda
\times_\nu \times \mathbb{P}\left[_C\right],\end{aligned}$$and thus $$D_{s-}^{\left( n\right)_}\theta _{1}^{\left( n\right)_}\left(_s,x\right)
\overset{\lambda \times \nu \times \mathbb{P}\text{-a.s.}}{\longrightarrow }D_{s-}\theta _{1}\left( s,x\right).$$
Since $\int\limits_{\Omega \times \left[ 0,T\right] }\left\vert
D_{t-}^{\left( n\right)_}-D_{t-}\right\vert_d\mathbb{P}\times dt\mathbb{=}\int\limits_{\Omega_\times_\left[_0,T\right] }\left\vert D_{t}^{\left(
n\right) }-D_{t}\right\vert d\mathbb{P}\times_dt\longrightarrow 0$, we have that $\left\{_D_{t-}^{\left( n\right)_}\right\} _{t\in \left[ 0,T\right] }$ $\overset{L^{1}\left( \lambda \times_\mathbb{P}\right)_}{\longrightarrow }\left\{ D_{t-}\right\}__{t\in \left[ 0,T\right] }$ and $\left\{ D_{t}^{\left( n\right) }\right\}__{t\in \left[ 0,T\right] }$ $\overset{L^{1}\left( \lambda_\times \mathbb{P}\right) }{\longrightarrow_}\left\{_D_{t}\right\}__{t\in \left[ 0,T\right] }.$_Then, for an arbitrary but fixed_subsequence $\left\{ n |
only three components may be nonzero,
$$\begin{aligned}
\eta^{(xx)(xy)}
= & \frac{1}{2 \hbar} \sum_n \int \frac{d^2 \pi}{(2 \pi \hbar)^2} \pi^{x 2} \Omega_{n {\vec \pi} z} f_{n {\vec \pi}}, \label{eq:eta2da} \\
\eta^{(xx)(yy)}
= & \frac{1}{\hbar} \sum_n \int \frac{d^2 \pi}{(2 \pi \hbar)^2} \pi^x \pi^y \Omega_{n {\vec \pi} z} f_{n {\vec \pi}}, \label{eq:eta2db} \\
\eta^{(xy)(yy)}
= & \frac{1}{2 \hbar} \sum_n \int \frac{d^2 \pi}{(2 \pi \hbar)^2} \pi^{y 2} \Omega_{n {\vec \pi} z} f_{n {\vec \pi}}. \label{eq:eta2dc}\end{aligned}$$
\[eq:eta2d\]
Furthermore, if a system is rotationally invariant, there is only one nonzero component $\eta_{\rm H} = \eta^{(xx)(xy)} = \eta^{(xy)(yy)}$, $$\eta_{\rm H}
= \frac{1}{4 \hbar} \sum_n \int \frac{d^2 \pi}{(2 \pi \hbar)^2} {\vec \pi}^2 \Omega_{n {\vec \pi} z} f_{n {\vec \pi}}. \label{eq:etah}$$ In three dimensions, all the components of the HV should vanish if a system is rotationally invariant. On the other hand, if a system is axially invariant along the $z$ axis, two components are independent,
$$\begin{aligned}
\eta^{(xx)(xy)}
= & \frac{1}{2 \hbar} \sum_n \int \frac{d^3 \pi}{(2 \pi \hbar)^3} \pi^{x 2} \Omega_{n {\vec \pi} z} f_{n {\vec \pi}}, \label{eq:eta3da} \\
| only three components may be nonzero,
$ $ \begin{aligned }
\eta^{(xx)(xy) }
= & \frac{1}{2 \hbar } \sum_n \int \frac{d^2 \pi}{(2 \pi \hbar)^2 } \pi^{x 2 } \Omega_{n { \vec \pi } z } f_{n { \vec \pi } }, \label{eq: eta2da } \\
\eta^{(xx)(yy) }
= & \frac{1}{\hbar } \sum_n \int \frac{d^2 \pi}{(2 \pi \hbar)^2 } \pi^x \pi^y \Omega_{n { \vec \pi } z } f_{n { \vec \pi } }, \label{eq: eta2db } \\
\eta^{(xy)(yy) }
= & \frac{1}{2 \hbar } \sum_n \int \frac{d^2 \pi}{(2 \pi \hbar)^2 } \pi^{y 2 } \Omega_{n { \vec \pi } z } f_{n { \vec \pi } }. \label{eq: eta2dc}\end{aligned}$$
\[eq: eta2d\ ]
Furthermore, if a system is rotationally invariant, there be merely one nonzero component $ \eta_{\rm H } = \eta^{(xx)(xy) } = \eta^{(xy)(yy)}$, $ $ \eta_{\rm H }
= \frac{1}{4 \hbar } \sum_n \int \frac{d^2 \pi}{(2 \pi \hbar)^2 } { \vec \pi}^2 \Omega_{n { \vec \pi } z } f_{n { \vec \pi } }. \label{eq: etah}$$ In three dimensions, all the components of the HV should fly if a system is rotationally invariant. On the early hand, if a system is axially invariant along the $ z$ bloc, two component are independent,
$ $ \begin{aligned }
\eta^{(xx)(xy) }
= & \frac{1}{2 \hbar } \sum_n \int \frac{d^3 \pi}{(2 \pi \hbar)^3 } \pi^{x 2 } \Omega_{n { \vec \pi } z } f_{n { \vec \pi } }, \label{eq: eta3da } \\ | onpy three components may ne nonzero,
$$\begin{aligned}
\eta^{(xx)(sy)}
= & \frxc{1}{2 \hbar} \sum_n \int \frac{d^2 \pi}{(2 \pi \hvar)^2} \pu^{x 2} \Omega_{n {\vec \pi} z} f_{n {\vec \pi}}, \lwbel{eq:etq2da} \\
\eta^{(xx)(yy)}
= & \frac{1}{\hbar} \sum_h \int \hrac{d^2 \pi}{(2 \pi \hbat)^2} \pi^x \pi^y \Omaga_{n {\vec \pi} z} x_{n {\vzc \pi}}, \label{eq:eta2db} \\
\eta^{(xy)(yy)}
= & \frac{1}{2 \hbar} \sim_j \int \frac{d^2 \pi}{(2 \pi \nfar)^2} \lp^{y 2} \Omega_{n {\vec \pi} z} f_{n {\vec \pi}}. \lzbel{eq:eua2dc}\end{aligned}$$
\[eq:ets2d\]
Furthermore, if a system ls rltationally invariwnt, there iw onjt one nonzeru componenu $\zta_{\rm H} = \etz^{(xx)(xy)} = \eta^{(xy)(yy)}$, $$\eta_{\rm H}
= \frac{1}{4 \hcar} \sbm_n \int \frax{d^2 \pi}{(2 \ki \hbar)^2} {\vec \'i}^2 \Omeda_{n {\vec \pi} z} f_{n {\vec \[i}}. \labek{eq:etah}$$ In thrce dikenwions, all the componeits of the HV should vanish ix c system is rotationaolt invdriatt. Ov thd ouhec hznd, if a aystem is zxially invqriant along the $z$ sxyw, two componehts arq yndependent,
$$\begin{aligned}
\eta^{(xx)(xy)}
= & \frab{1}{2 \hbzr} \sum_n \int \frac{d^3 \pi}{(2 \pi \hbar)^3} \pi^{x 2} \Omega_{n {\vec \pi} z} f_{n {\dec \pi}}, \label{eq:eta3da} \\
| only three components may be nonzero, $$\begin{aligned} & \hbar} \sum_n \frac{d^2 \pi}{(2 \pi \pi} f_{n {\vec \pi}}, \\ \eta^{(xx)(yy)} = \frac{1}{\hbar} \sum_n \int \frac{d^2 \pi}{(2 \pi \pi^x \pi^y \Omega_{n {\vec \pi} z} f_{n {\vec \pi}}, \label{eq:eta2db} \\ \eta^{(xy)(yy)} = \frac{1}{2 \hbar} \sum_n \int \frac{d^2 \pi}{(2 \pi \hbar)^2} \pi^{y 2} \Omega_{n {\vec \pi} f_{n \pi}}. \[eq:eta2d\] if a system is rotationally invariant, there is only one nonzero component $\eta_{\rm H} = \eta^{(xx)(xy)} \eta^{(xy)(yy)}$, $$\eta_{\rm H} = \frac{1}{4 \hbar} \sum_n \int \pi}{(2 \pi \hbar)^2} {\vec \Omega_{n {\vec \pi} z} f_{n \pi}}. In three all components the HV should if a system is rotationally invariant. On the other hand, if a system is axially invariant along $z$ axis, are independent, \eta^{(xx)(xy)} & \hbar} \sum_n \int \pi \hbar)^3} \pi^{x 2} \Omega_{n {\vec {\vec \pi}}, \label{eq:eta3da} \\ | only three components may be nOnzero,
$$\begiN{aligNed}
\Eta^{(Xx)(Xy)}
= & \frAc{1}{2 \hbAr} \sum_n \int \frac{d^2 \PI}{(2 \pi \hBar)^2} \pi^{x 2} \Omega_{n {\vec \pi} z} f_{n {\veC \pi}}, \laBeL{Eq:etA2Da} \\
\Eta^{(xx)(Yy)}
= & \frac{1}{\hBAr} \SUM_n \iNt \FrAc{d^2 \Pi}{(2 \PI \hBar)^2} \pi^X \pi^Y \Omega_{n {\Vec \pi} z} f_{n {\veC \pi}}, \LaBel{eq:eta2db} \\
\etA^{(Xy)(Yy)}
= & \frac{1}{2 \hbar} \Sum_N \int \frac{d^2 \pi}{(2 \pI \hbAr)^2} \pi^{y 2} \OMeGa_{n {\VEc \pi} z} F_{n {\vEc \pi}}. \lAbel{eq:ETa2dc}\enD{aligned}$$
\[eQ:eTA2d\]
FurtHErmore, iF A SyStem Is rotationally invARiANt, there is only oNe nonzErO CoMPOneNt $\eTa_{\rm H} = \eta^{(xx)(Xy)} = \Eta^{(xy)(YY)}$, $$\eta_{\rm H}
= \FRaC{1}{4 \HBAr} \sUM_n \int \frac{d^2 \pi}{(2 \pI \hbar)^2} {\vec \pi}^2 \OMEga_{N {\vec \pi} Z} f_{N {\veC \Pi}}. \labeL{eq:etAh}$$ iN thRee dimensioNs, alL the compoNents oF The HV shOUld vaniSh if a sYstEm iS rotATiOnAllY iNVarIAnT. On THe oTher hand, If A sYstem Is axIALLY invAriAnt aLong tHe $z$ axis, two comPonEnts ARe iNdepeNdent,
$$\BegiN{aLigneD}
\eta^{(xx)(Xy)}
= & \fraC{1}{2 \hBar} \sum_n \int \frac{d^3 \Pi}{(2 \pi \Hbar)^3} \pi^{x 2} \OmEga_{N {\vEc \pI} z} F_{n {\vec \PI}}, \label{Eq:eTa3dA} \\
| only three components may be nonzer o,
$ $\b egi n{ alig ned}
\eta^{(xx)( x y)} = & \frac{1}{2 \hbar } \su m_ n \in t \ frac{ d^2 \pi } {( 2 \pi \ hb ar) ^2 } \ pi^{x 2} \Omega _{n {\vec\pi }z} f_{n {\ve c \ pi}}, \lab el{ eq:eta2da} \ \
\eta^ {( xx) ( yy)} = & \f rac{1} { \hbar} \sum_n \ in t \frac { d^2 \pi } { (2 \pi \hbar)^2} \pi^x\ pi ^ y \Omega_{n {\ vec \p i} z} f _{n {\ vec \pi}}, \ label { eq:eta2 d b} \ \
\ eta^{(xy)(yy) }
= & \fr a c{1 }{2 \h ba r}\ sum_n\int\f r ac{ d^2 \pi}{(2 \pi \hbar)^2 } \pi^ { y 2} \O m ega_{n{\vec\pi } z } f_ { n{\ vec \ p i}} . \ lab e l{e q:eta2dc }\ en d{ali gned } $ $
\[e q:e ta2d \]
F urthermore, i f a sys t emis ro tatio nall yinvar iant,there i s only one nonz erocomponent $\ et a_{ \r m H}= \eta^ {(x x)( xy)} =\eta^{( x y)( yy ) } $ ,$$\eta_{\rm H}
= \ f r ac {1}{4 \h bar} \ s um _n \int \fr ac {d^ 2 \p i } {(2 \ pi \ h ba r)^2} {\ vec \p i }^ 2\Omega_ {n {\vec \ pi} z} f_{n {\ve c \pi} }. \labe l{eq: e tah}$$ In thre e dimensions,a ll t he comp one nts of theHV s h ould van i sh if a sys tem i sr ot a tionally invariant. O n theother hand, if a s ystem is a x i a lly inva rian t a l ong the $z$ ax is, t wo compone n ts are i ndepe ndent,
$$\begin{ a l igned}
\e ta^ {(x x)( x y )}
= & \frac{ 1 } {2 \ hb ar} \su m_n \int \ fra c{d ^3\pi }{ (2 \pi \h bar)^3}\p i^ {x 2 } \ Omega _ {n {\vec \ pi} z } f _{n { \ vec \p i}},\lab el {e q :et a3da} \ \
| only_three components_may be nonzero,
$$\begin{aligned}
_\eta^{(xx)(xy)}
_=_& \frac{1}{2_\hbar}_\sum_n \int \frac{d^2_\pi}{(2 \pi \hbar)^2}_\pi^{x 2} \Omega_{n {\vec_\pi} z} f_{n_{\vec_\pi}}, \label{eq:eta2da} \\
\eta^{(xx)(yy)}
= & \frac{1}{\hbar} \sum_n \int \frac{d^2 \pi}{(2 \pi_\hbar)^2}_\pi^x \pi^y_\Omega_{n_{\vec_\pi} z} f_{n {\vec \pi}},_\label{eq:eta2db} \\
\eta^{(xy)(yy)}
_= &_\frac{1}{2 \hbar} \sum_n \int \frac{d^2 \pi}{(2 \pi \hbar)^2}_\pi^{y_2} \Omega_{n {\vec_\pi} z} f_{n {\vec \pi}}. \label{eq:eta2dc}\end{aligned}$$
\[eq:eta2d\]
Furthermore, if a system_is rotationally invariant, there is only_one nonzero component_$\eta_{\rm_H}_= \eta^{(xx)(xy)} = \eta^{(xy)(yy)}$,_$$\eta_{\rm H}
= \frac{1}{4 \hbar}_\sum_n \int \frac{d^2 \pi}{(2 \pi \hbar)^2}_{\vec \pi}^2 \Omega_{n {\vec \pi} z} f_{n_{\vec \pi}}. \label{eq:etah}$$ In three dimensions,_all the components of the_HV should_vanish if a system is_rotationally invariant. On_the other_hand, if a_system is axially invariant along the_$z$ axis, two_components are independent,
$$\begin{aligned}
\eta^{(xx)(xy)}
_=_& \frac{1}{2 \hbar}_\sum_n_\int_\frac{d^3 \pi}{(2_\pi \hbar)^3} \pi^{x_2}_\Omega_{n {\vec_\pi}_z} f_{n {\vec \pi}}, \label{eq:eta3da} \\
|
1^0=\omega_2(\mathbf{k}_1) \\ k_2^0=\omega_{\lambda'}(\mathbf{k}_2)}}\,,\end{aligned}$$
where the energy of the final electron is denoted as $E_2$. Note the division by the group velocity of the first and second polarization state (see also Ref. [@Colladay:2001wk]), respectively, where in the standard theory $|\mathbf{v}_{\mathrm{gr},\,1}|=|\mathbf{v}_{\mathrm{gr},\,2}|=1$. The tensor $(|\mathcal{M}(k_1,k_2)|^2\big)^{\mu\nu\varrho\sigma}$ is given by the trace term of Eq. (5.81) of Ref. [@PeskinSchroeder1995] with some modifications due to the Lorentz-violating kinematics.
![Chosen coordinate system for the phase space integration, where the initial photon three-momentum $\mathbf{k}_1$ lies along the third axis of the coordinate system. For the outgoing photon momentum $\mathbf{k}_2$, spherical coordinates are chosen with the azimuthal angle $\vartheta$ corresponding to the angle between $\mathbf{k}_2$ and the third axis. Cases are treated with the three-vector $\boldsymbol{\zeta}$ having equal or unequal components.[]{data-label="fig:phase-space-integration"}](consistency-modmax-parity-odd_v1_fig4_coordinates.pdf)
The purely algebraic part of the calculation, that includes computation of traces, contraction of indices and inserting kinematical relations, is performed with `Form` [@Vermaseren:2000nd]. The subsequent phase space calculation is done numerically with `C++`, since the resulting matrix element squared contains hundreds of terms. The limit of zero Lorentz violation has to be taken with care and “long double” precision does not suffice here. Therefore, the GNU Multiple Precision Arithmetic Library `GMP` [@GMP:2011] is used with its `C++` interface described in Sec. 12 of the reference previously mentioned.
The first idea was to choose the coordinate system such that $\boldsymbol{\zeta}$ lies along the third axis. Then the phase space should have been integrated with cylindrical coordinates $(k_{2,\bot},\varphi,k_{2,\parallel})$. To cover a general situation, where the initial photon momentum $\mathbf{k}_1$ points in an arbitrary | 1 ^ 0=\omega_2(\mathbf{k}_1) \\ k_2 ^ 0=\omega_{\lambda'}(\mathbf{k}_2)}}\,,\end{aligned}$$
where the energy of the final electron is denoted as $ E_2$. Note the part by the group speed of the first and second polarization state (visualize besides Ref. [ @Colladay:2001wk ]), respectively, where in the standard theory $ |\mathbf{v}_{\mathrm{gr},\,1}|=|\mathbf{v}_{\mathrm{gr},\,2}|=1$. The tensor $ (|\mathcal{M}(k_1,k_2)|^2\big)^{\mu\nu\varrho\sigma}$ is given by the trace condition of Eq. (5.81) of Ref. [ @PeskinSchroeder1995 ] with some modifications due to the Lorentz - violating kinematics.
! [ choose coordinate system for the phase space consolidation, where the initial photon three - momentum $ \mathbf{k}_1 $ lies along the third axis of the coordinate system. For the outgoing photon momentum $ \mathbf{k}_2 $, ball-shaped coordinates are chosen with the azimuthal slant $ \vartheta$ corresponding to the slant between $ \mathbf{k}_2 $ and the third bloc. Cases are treated with the three - vector $ \boldsymbol{\zeta}$ having equal or inadequate components.[]{data - label="fig: phase - space - integration"}](consistency - modmax - parity - odd_v1_fig4_coordinates.pdf)
The strictly algebraic part of the calculation, that include computation of tracing, contraction of indices and inserting kinematical sexual intercourse, is performed with 'Form '[ @Vermaseren:2000nd ]. The subsequent phase space calculation is done numerically with 'C++ `, since the resulting matrix chemical element squared contains hundreds of terms. The limit of zero Lorentz violation has to be taken with care and “ long double ” preciseness does not suffice here. consequently, the GNU Multiple Precision Arithmetic Library 'GMP '[ @GMP:2011 ] is used with its 'C++ 'interface described in Sec. 12 of the reference previously mention.
The first idea was to choose the coordinate system such that $ \boldsymbol{\zeta}$ lie down along the third axis. Then the phase space should have been integrated with cylindrical coordinates $ (k_{2,\bot},\varphi, k_{2,\parallel})$. To brood a general situation, where the initial photon momentum $ \mathbf{k}_1 $ points in an arbitrary | 1^0=\omeha_2(\mathbf{k}_1) \\ k_2^0=\omega_{\lambda'}(\mxthbf{k}_2)}}\,,\end{aligneb}$$
qhere vhe enedgy of tfe final electron is denoted aw $E_2$. Nite the division by thd group vvlocity od tht first and seconv polaridction dtatz (wee also Ref. [@Cokladay:2001wk]), raspectively, whard nn the standard theory $|\mathbf{v}_{\mathrm{dr},\,1}|=|\mathbg{v}_{\lathrm{gr},\,2}|=1$. The tgnsor $(|\iathdal{M}(k_1,k_2)|^2\big)^{\mu\nu\varrho\sigma}$ is given gy the urace term of Eq. (5.81) pf Ref. [@PeskinSchroeder1995] with somf modifications duf to the Lotsnts-ciolating kivematics.
![Chosen coordinzte system for the phase space kntegxation, wherg fhf initial phiton nhree-momentum $\mathbf{k}_1$ lies akong the third axms od the coordinate systxm. For the outgoing khoton momancum $\mathbf{k}_2$, spherical ciorditatev ard chuseh xitg the wzijuthal angme $\vartheta$ corresponding to tne qngle between $\mathbs{k}_2$ and the third axis. Cases are treated wpth fhe three-vector $\boldsymvol{\zeta}$ having equal lr unequaj components.[]{data-label="fig:phase-space-integration"}](conshstenry-oodnaw-parkry-ldd_v1_fig4_coordinates.pdf)
The purely algebraic pare og nhe calculation, tmat includes compuyahipg of traces, cuntraccjoh of indices and ijsertind kinwmatical welayions, is performed with `Forn` [@Vermaseren:2000ud]. Rhe subsequent phade space cauculstion is done numerically wich `C++`, sjnce the redulting mzgrix element squxrec wontains hundreds of terms. The limiv of vero Lordntz violaeion has tl be bdken with care and “long dmuble” preclsion does not suffice here. Thecxfore, the GNU Mgltpple Precnsion Srithmetic Lifrary `GMP` [@GMP:2011] ns used cith igs `C++` interrace devcribed in Fec. 12 of the raverence prevmously megtiobed.
Tye firsg idea was to cnoose the coordinatw system such that $\boussymbol{\zeta}$ liew aoong the third sxir. Trej vhe prdse space shmuld haxr beev integrateb dith cylindrical coordindtes $(k_{2,\bot},\varphi,k_{2,\parallrl})$. To covet a generwl situation, ehere the initial khoton mooentuk $\mwthbf{k}_1$ points in an arbitrary | 1^0=\omega_2(\mathbf{k}_1) \\ k_2^0=\omega_{\lambda'}(\mathbf{k}_2)}}\,,\end{aligned}$$ where the energy of electron denoted as Note the division the and second polarization (see also Ref. respectively, where in the standard theory The tensor $(|\mathcal{M}(k_1,k_2)|^2\big)^{\mu\nu\varrho\sigma}$ is given by the trace term of Eq. (5.81) of [@PeskinSchroeder1995] with some modifications due to the Lorentz-violating kinematics. ![Chosen coordinate system for phase integration, the photon three-momentum $\mathbf{k}_1$ lies along the third axis of the coordinate system. For the outgoing photon $\mathbf{k}_2$, spherical coordinates are chosen with the azimuthal $\vartheta$ corresponding to the between $\mathbf{k}_2$ and the third Cases treated with three-vector having or unequal components.[]{data-label="fig:phase-space-integration"}](consistency-modmax-parity-odd_v1_fig4_coordinates.pdf) purely algebraic part of the calculation, that includes computation of traces, contraction of indices and inserting kinematical is performed [@Vermaseren:2000nd]. The phase calculation done numerically with the resulting matrix element squared contains The limit of zero Lorentz violation has to taken with and “long double” precision does not here. Therefore, the GNU Multiple Precision Arithmetic Library [@GMP:2011] is used with its `C++` interface described in Sec. 12 of the reference previously first idea was to the coordinate system that lies the axis. Then phase space should have been integrated with cylindrical coordinates $(k_{2,\bot},\varphi,k_{2,\parallel})$. To a general situation, where the initial photon momentum $\mathbf{k}_1$ points arbitrary | 1^0=\omega_2(\mathbf{k}_1) \\ k_2^0=\omega_{\lambda'}(\mAthbf{k}_2)}}\,,\end{aLigneD}$$
whEre ThE eneRgy oF the final electROn is Denoted as $E_2$. Note the divisIon by ThE GrouP VeLocitY of the fIRsT ANd sEcOnD poLaRIzAtion StaTe (see alSo Ref. [@CollaDay:2001Wk]), Respectively, WHeRe in the staNdaRd theory $|\mathBf{v}_{\Mathrm{Gr},\,1}|=|\MatHBf{v}_{\maThrM{gr},\,2}|=1$. ThE tensoR $(|\MathcaL{M}(k_1,k_2)|^2\big)^{\mu\Nu\VArrho\sIGma}$ is giVEN bY the Trace term of Eq. (5.81) of ReF. [@peSKinSchroeder1995] wiTh some MoDIfICAtiOns Due to the LoReNtz-viOLating kINeMATIcs.
![cHosen coordinaTe system for THe pHase spAcE inTEgratiOn, wheRe THe iNitial photoN thrEe-momentuM $\mathbF{K}_1$ lies alONg the thIrd axiS of The CoorDInAtE syStEM. FoR ThE ouTGoiNg photon MoMeNtum $\mAthbF{K}_2$, SPHeriCal CoorDinatEs are chosen wiTh tHe azIMutHal anGle $\vaRtheTa$ CorreSpondiNg to tHe Angle between $\matHbf{k}_2$ And the thiRd aXiS. CaSeS are tREated wIth The Three-veCtor $\bolDSymBoL{\ZETa}$ Having equal or unequAl COMpOnents.[]{daTa-labeL="FiG:pHAse-space-InTegRatiON"}](ConsiStenCY-mOdmax-parIty-odd_V1_FiG4_cOordinaTeS.pdf)
ThE pUreLy aLgebrAIc paRt of thE calculaTion, tHAt includes compUTation of traceS, CoNTRaCTion Of iNdices and inSertINg kiNemaTIcAl rELatioNs, is pErFOrMEd with `Form` [@VermasereN:2000nD]. The suBsequEnt phase space CalculatioN IS Done numeRicaLLy WIth `C++`, since the reSultiNg matrix elEMent squaRed coNtains huNdreds of tERMs. The limIt oF zeRo LOreNTZ vIolation has to BE TakeN wIth care And “Long douBle” PreCisIon DoEs not suffIce here. THeReFoRe, The gNU MuLTiple PreCiSioN ARitHmetiC librarY `GMP` [@GmP:2011] is UsEd WIth Its `C++` intERfACE desCrIbEd in sec. 12 Of The reFereNCe pReviousLy mentionEd.
THE firSt IdEa was to Choose the coorDiNate system SuCh tHat $\bolDSYmbol{\zetA}$ lies along the third axis. THEn the phAse Space ShouLd have beeN inTegratEd wITh cyliNdricaL coorDiNatES $(K_{2,\bot},\vARPhI,k_{2,\pArAllel})$. To covER A geNeral SiTuatIon, wherE the initial photon mOMenTum $\mathbf{k}_1$ poiNts In an ARBiTraRY | 1^0=\omega_2(\mathbf{k}_1) \\ k_2^0= \omeg a_{ \la mb da'} (\ma thbf{k}_2)}}\, , \end {aligned}$$
where the ener gy of t h efinal electr o ni s de no te d a s$ E_ 2$. N ote the di vision bythe g roup velocit y o f the firs t a nd second po lar izatio nsta t e (se e a lso R ef. [@ C ollada y:2001wk] ), respec t ively,w h er e in the standard the o ry $|\mathbf{v}_{ \mathr m{ g r} , \ ,1} |=| \mathbf{v} _{ \math r m{gr},\ , 2} | = 1 $.T he tensor $(| \mathcal{M} ( k_1 ,k_2)| ^2 \bi g )^{\mu \nu\v ar r ho\ sigma}$ isgive n by thetracet erm ofE q. (5.8 1) ofRef . [ @Pes k in Sc hro ed e r19 9 5] wi t h s ome modi fi ca tions due t o theLor entz -viol ating kinemat ics .
! [ Cho sen c oordi nate s ystem for t he ph as e space integra tion , where t hein iti al phot o n thre e-m ome ntum $\ mathbf{ k }_1 $l i e salong the third ax is o fthe coor dinate sy st e m. For t he ou tgoi n g phot on m o me ntum $\m athbf{ k }_ 2$ , spher ic al coo rd ina tes arec hose n with the azi mutha l angle $\varth e ta$ correspon d in g to theang le between$\ma t hbf{ k}_2 $ a ndt he th ird a xi s .C ases are treated wi th the t hree- vector $\bold symbol{\ze t a } $ having equ a lo r unequal comp onent s.[]{data- l abel="fi g:pha se-space -integrat i o n"}](con sis ten cy- mod m a x- parity-odd_v1 _ f ig4_ co ordinat es. pdf)
T hepur ely al ge braic par t of the c al cu la tio n, th a t includ es co mp uta tiono f trac es, c ontr ac ti o n o f indic e sa n d in se rt ingkin em atica l re l ati ons, is performe d w i th ` Fo rm ` [@Ver maseren:2000n d] . The subs eq uen t phas e space ca lculation is done numer i cally w ith `C++ `, s ince theres ulting ma t rix el ementsquar ed co n t ainsh u nd red sof terms.T h e l imitof zer o Loren tz violation has t o be taken with c are and “ lo ngd ou b le” p r eci s i on does not suf fice here. T h er efore, the GNU M ultiple Precis ion A r ithmeti c Library `GMP` [@ GM P:20 1 1 ] i s used wit h its `C ++` inter f ace d e sc ribed in Sec.12 of therefere n ceprevi ouslyme ntione d.
T he first i dea was to choose the c oordin ate s yst em such t hat $\b oldsymbol {\ze ta}$ liesalo ngthe t hir d axis . Th e nthe phase spa c e shouldh av e b e e nintegratedw i t h c ylind ric a l coor dina tes $(k_{2,\bot}, \ varphi,k_{2,\p aral l e l}) $.T o co ve r a general si tua ti o n , whereth e initial p hoton mo me n tum $ \mathb f{k}_1 $ point s in an arb itra ry | 1^0=\omega_2(\mathbf{k}_1) \\_k_2^0=\omega_{\lambda'}(\mathbf{k}_2)}}\,,\end{aligned}$$
where the_energy of the final_electron is_denoted_as $E_2$._Note_the division by_the group velocity_of the first and_second polarization state_(see_also Ref. [@Colladay:2001wk]), respectively, where in the standard theory $|\mathbf{v}_{\mathrm{gr},\,1}|=|\mathbf{v}_{\mathrm{gr},\,2}|=1$. The tensor $(|\mathcal{M}(k_1,k_2)|^2\big)^{\mu\nu\varrho\sigma}$ is given_by_the trace_term_of_Eq. (5.81) of Ref. [@PeskinSchroeder1995] with_some modifications due to the_Lorentz-violating kinematics.
![Chosen_coordinate system for the phase space integration, where_the_initial photon three-momentum_$\mathbf{k}_1$ lies along the third axis of the coordinate_system. For the outgoing photon momentum_$\mathbf{k}_2$, spherical coordinates_are_chosen_with the azimuthal angle_$\vartheta$ corresponding to the angle between_$\mathbf{k}_2$ and the third axis. Cases_are treated with the three-vector $\boldsymbol{\zeta}$ having_equal or unequal components.[]{data-label="fig:phase-space-integration"}](consistency-modmax-parity-odd_v1_fig4_coordinates.pdf)
The purely algebraic_part of the calculation, that_includes computation_of traces, contraction of indices_and inserting kinematical_relations, is_performed with `Form`_[@Vermaseren:2000nd]. The subsequent phase space calculation_is done numerically_with `C++`, since the resulting matrix_element_squared contains hundreds_of_terms._The limit_of zero Lorentz_violation_has to_be_taken with care and “long double”_precision_does not suffice here. Therefore, the GNU_Multiple Precision Arithmetic Library_`GMP`_[@GMP:2011] is used with_its `C++` interface described in_Sec. 12 of the reference previously_mentioned.
The first_idea was_to choose the coordinate system such that $\boldsymbol{\zeta}$ lies along the_third axis. Then the phase space_should have been integrated_with cylindrical_coordinates_$(k_{2,\bot},\varphi,k_{2,\parallel})$. To cover_a_general situation,_where the initial photon momentum $\mathbf{k}_1$ points_in an_arbitrary |
on the typical scales of the disk. However the non-detection of an extended disk at 3mm argues against a significant free-free emission component at 1mm. Furthermore, there is no detection of a thermal outflow-like component either at 1mm or at 3mm.
If the 1mm flux of the disk can be attributed to dust emission, a mass budget (M$_{dust}$) can be derived provided that values for the temperature (T$_{dust}$) and for the emissivity of dust ($\kappa$) are assumed. IRAS observations reveal significant amounts of warm dust in 4C31.04. From the flux ratio measured by IRAS S$_{\nu}$(100$\mu$)/S$_{\nu}$(60$\mu$)$\sim$3.5, we derive a range for T$_{dust}$=25–30K for a corresponding range of $\beta$=2–1 in the wavelength dependency of $\kappa \sim \lambda^{-\beta}$. Taking $\kappa$(850$\mu$)=0.9cm$^{2}$gr$^{-1}$ from Klaas et al. ([@kla01]), and T$_{dust}$=30K (i.e., the solution for $\beta$=1) we derive M$_{dust}\sim$6$\times$10$^{7}$M$_{\sun}$ for the warm dust. To better estimate the total dust mass we have fitted the fluxes measured by IRAS (at 100$\mu$ and 60$\mu$) and the PdBI (at 1.3mm) using a two component model for the dust temperatures. Any plausible fit to the observations gives a value for M$_{dust}$ of (4–7)$\times$10$^{8}$M$_{\sun}$. The higher (lower) value of M$_{dust}$ corresponds to the fit with $\beta$=1.5 ($\beta$=1) and temperatures of $\sim$16K(21K) and $\sim$32K(39K) for the cold and warm dust components, respectively. The cold dust component seems to contain the bulk of the mass: M$_{dust}^{cold}>$0.95$\times$M$_{dust}^{total}$).
Taking a molecular gas–to–dust ratio $\sim$100 (Seaquist et al. [@sea04]), we foresee that the total molecular gas mass of the disk might be M$_{gas}$$\sim$(4–7)$\times$10$^{10}$M$_{\sun}$. If we allow for a minor contribution | on the typical scales of the disk. However the non - detection of an unfold disk at 3 mm argue against a meaning free - free emission part at 1 mm. Furthermore, there be no detection of a thermal escape - like component either at 1 mm or at 3 mm.
If the 1 millimeter flux of the disk can be attributed to dust discharge, a mass budget (M$_{dust}$) can be derived provided that values for the temperature (T$_{dust}$) and for the emissivity of dust ($ \kappa$) are wear. IRAS observations reveal significant sum of warm dust in 4C31.04. From the flux ratio measure by IRAS S$_{\nu}$(100$\mu$)/S$_{\nu}$(60$\mu$)$\sim$3.5, we derive a range for T$_{dust}$=25–30 K for a corresponding range of $ \beta$=2–1 in the wavelength dependency of $ \kappa \sim \lambda^{-\beta}$. consume $ \kappa$(850$\mu$)=0.9cm$^{2}$gr$^{-1}$ from Klaas et al. ([ @kla01 ]), and T$_{dust}$=30 K (i.e., the solution for $ \beta$=1) we derive M$_{dust}\sim$6$\times$10$^{7}$M$_{\sun}$ for the warm dust. To better estimate the entire dust mass we have fitted the fluxes measured by IRAS (at 100$\mu$ and 60$\mu$) and the PdBI (at 1.3 mm) using a two component model for the dust temperature. Any plausible fit to the observations feed a value for M$_{dust}$ of (4–7)$\times$10$^{8}$M$_{\sun}$. The eminent (lower) value of M$_{dust}$ corresponds to the fit with $ \beta$=1.5 ($ \beta$=1) and temperatures of $ \sim$16K(21 K) and $ \sim$32K(39 potassium) for the cold and quick dust components, respectively. The cold debris component seems to contain the bulk of the mass: M$_{dust}^{cold}>$0.95$\times$M$_{dust}^{total}$).
Taking a molecular gas – to – dust ratio $ \sim$100 (Seaquist et al. [ @sea04 ]), we foresee that the total molecular gas bulk of the disk might be M$_{gas}$$\sim$(4–7)$\times$10$^{10}$M$_{\sun}$. If we give up for a minor contribution | on the typical scales of tme disk. However jhw non-dxtectioh of an dxtended disk at 3mm argues ajainwt a wignificant free-free eoission clmponent at 1nn. Furthermore, therc is hl decertion of a therkal outflof-like componend dicher at 1mm or at 3mm.
If the 1mm flux of the dixk can be attriboted uo qust vmlssion, a mass budget (M$_{dust}$) can bs derivtd provided that vslues for the temperature (H$_{dush}$) and for the emisdivity of doat ($\hqppa$) are assjmed. IRAS observations reveal significant amounts of darm bust in 4C31.04. Ftoj hve flux ratmo meafured by IRAS S$_{\nu}$(100$\mu$)/S$_{\tu}$(60$\mu$)$\sim$3.5, we derive a rsngx foe T$_{dust}$=25–30K for a corres'onding range of $\betw$=2–1 in the farelength dependency od $\jappa \sim \lamcea^{-\bdta}$. Tekihg $\kapoa$(850$\mn$)=0.9cm$^{2}$gr$^{-1}$ from Ilaas et al. ([@jla01]), and T$_{dust}$=30K (i.e., tne wolution for $\geta$=1) wq qerive M$_{dust}\sim$6$\times$10$^{7}$M$_{\sun}$ for the warm dlst. Fo better estimate the rotal dust mass we hage fitted the fluxes measured by IRAS (at 100$\mu$ and 60$\mu$) and the PdBI (xt 1.3nm) jwijg a two component model for the dust temperaehrts. Sny plausible nit to the observayilnx gives a valug for M$_{bhsf}$ of (4–7)$\times$10$^{8}$M$_{\sun}$. The higher (lowee) value os M$_{dist}$ corresponds to the fit qith $\beta$=1.5 ($\bena$=1) abd temperatures of $\sim$16K(21K) and $\rim$32K(39L) for the cold and warm dust comlonents, resoectively. Ghe cold dust coopokend seems uu contain the bulh of the nass: M$_{dust}^{culd}>$0.95$\tomes$M$_{dtst}^{total}$).
Taning a molecular gas–to–dudt rajio $\sik$100 (Seaquist et al. [@sea04]), we foresee that the total molecular gds kass of che dixk might be M$_{das}$$\sim$(4–7)$\times$10$^{10}$M$_{\suu}$. If we cllow wor a minog contribntion | on the typical scales of the disk. non-detection an extended at 3mm argues component 1mm. Furthermore, there no detection of thermal outflow-like component either at 1mm at 3mm. If the 1mm flux of the disk can be attributed to emission, a mass budget (M$_{dust}$) can be derived provided that values for the (T$_{dust}$) for emissivity dust ($\kappa$) are assumed. IRAS observations reveal significant amounts of warm dust in 4C31.04. From the ratio measured by IRAS S$_{\nu}$(100$\mu$)/S$_{\nu}$(60$\mu$)$\sim$3.5, we derive a for T$_{dust}$=25–30K for a range of $\beta$=2–1 in the dependency $\kappa \sim Taking from et al. ([@kla01]), T$_{dust}$=30K (i.e., the solution for $\beta$=1) we derive M$_{dust}\sim$6$\times$10$^{7}$M$_{\sun}$ for the warm dust. To better estimate the dust mass fitted the measured IRAS 100$\mu$ and 60$\mu$) PdBI (at 1.3mm) using a two the dust temperatures. Any plausible fit to the gives a for M$_{dust}$ of (4–7)$\times$10$^{8}$M$_{\sun}$. The higher value of M$_{dust}$ corresponds to the fit with ($\beta$=1) and temperatures of $\sim$16K(21K) and $\sim$32K(39K) for the cold and warm dust components, respectively. dust component seems to the bulk of mass: Taking molecular ratio $\sim$100 et al. [@sea04]), we foresee that the total molecular gas mass the disk might be M$_{gas}$$\sim$(4–7)$\times$10$^{10}$M$_{\sun}$. If we allow for a | on the typical scales of the diSk. However tHe non-DetEctIoN of aN extEnded disk at 3mm aRGues Against a significant freE-free EmISsioN CoMponeNt at 1mm. FURtHERmoRe, ThEre Is NO dEtectIon Of a therMal outflow-LikE cOmponent eithER aT 1mm or at 3mm.
IF thE 1mm flux of the DisK can be AtTriBUted tO duSt emiSsion, a MAss budGet (M$_{dust}$) cAn BE derivED providED ThAt vaLues for the temperaTUrE (t$_{dust}$) and for the EmissiViTY oF DUst ($\KapPa$) are assumEd. iRAS oBServatiONs REVEal SIgnificant amoUnts of warm dUSt iN 4C31.04. From ThE flUX ratio MeasuReD By IrAS S$_{\nu}$(100$\mu$)/S$_{\nu}$(60$\Mu$)$\siM$3.5, we derive A range FOr T$_{dust}$=25–30k For a corResponDinG raNge oF $\BeTa$=2–1 In tHe WAveLEnGth DEpeNdency of $\KaPpA \sim \lAmbdA^{-\BETA}$. TakIng $\KappA$(850$\mu$)=0.9cm$^{2}$Gr$^{-1}$ from Klaas et Al. ([@kLa01]), anD t$_{duSt}$=30K (i.e., The soLutiOn For $\beTa$=1) we deRive M$_{DuSt}\sim$6$\times$10$^{7}$M$_{\sun}$ fOr thE warm dust. to bEtTer EsTimatE The totAl dUst Mass we hAve fittED thE fLUXEs Measured by IRAS (at 100$\mu$ AnD 60$\MU$) aNd the PdBi (at 1.3mm) uSInG a TWo componEnT moDel fOR The duSt teMPeRatures. ANy plauSIbLe Fit to thE oBservaTiOns GivEs a vaLUe foR M$_{dust}$ Of (4–7)$\times$10$^{8}$M$_{\Sun}$. ThE Higher (lower) valUE of M$_{dust}$ correSPoNDS tO The fIt wIth $\beta$=1.5 ($\beta$=1) And tEMperAturES oF $\siM$16k(21K) and $\Sim$32K(39K) FoR ThE Cold and warm dust compOnEnts, reSpectIvely. The cold dUst componeNT SEems to coNtaiN ThE Bulk of the mass: M$_{Dust}^{cOld}>$0.95$\times$M$_{dUSt}^{total}$).
TAking A moleculAr gas–to–duST Ratio $\sim$100 (seaQuiSt eT al. [@SEA04]), wE foresee that tHE TotaL mOleculaR gaS mass of The DisK miGht Be m$_{gas}$$\sim$(4–7)$\tiMes$10$^{10}$M$_{\sun}$. IF wE aLlOw For A minoR ContribuTiOn | on the typical scales ofthe disk.Howev erthe n on-d etec tion of an ext e nded disk at 3mm argues ag ainst a sign i fi cantfree-fr e ee m iss io ncom po n en t at1mm . Furth ermore, th ere i s no detecti o nof a therm aloutflow-like co mponen teit h er at 1m m orat 3mm .
If t he 1mm fl ux of the disk ca n be att ributed to dust e m is s ion, a mass bu dget ( M$ _ {d u s t}$ ) c an be deri ve d pro v ided th a tv a l ues for the tempe rature (T$_ { dus t}$) a nd fo r the e missi vi t y o f dust ($\k appa $) are as sumed. IRAS ob s ervatio ns rev eal si gnif i ca nt am ou n tso fwar m du st in 4C 31 .0 4. Fr om t h e f luxrat io m easur ed by IRAS S$ _{\ nu}$ ( 100 $\mu$ )/S$_ {\nu }$ (60$\ mu$)$\ sim$3 .5 , we derive a r ange for T$_{ dus t} $=2 5– 30K f o r a co rre spo nding r ange of $\b et a $ = 2– 1 in the wavelengt hd e pe ndency o f $\ka p pa \ s im \lamb da ^{- \bet a } $. Ta king $\ kappa$(8 50$\mu $ )= 0. 9cm$^{2 }$ gr$^{- 1} $ f rom Klaa s etal. ([ @kla01]) , and T$_{dust}$=30K (i.e., the so l ut i o nf or $ \be ta$=1) we d eriv e M$_ {dus t }\ sim $ 6$\ti mes$1 0$ ^ {7 } $M$_{\sun}$ for the w arm du st. T o better esti mate the t o t a l dust m assw eh ave fitted the flux es measure d by IRAS (at100$\mu$ and 60$\ m u $) and t hePdB I ( at1 . 3m m) using a tw o comp on ent mod elfor the du sttem per at ures. Any plausib le f it t o t he ob s ervation sgiv es avalue for M$ _{dus t}$of ( 4 –7) $\times $ 10 $ ^ {8}$ M$ _{ \sun }$. T he hi gher (lo wer) va lue of M$ _{d u st}$ c or respond s to the fitwi th $\beta$ =1 .5($\bet a $ =1) andtemperatures of $\sim$1 6 K(21K)and $\si m$32 K(39K) fo r t he col d a n d warm dustcompo ne nts , respe c t iv ely .The cold d u s t c ompon en t se ems tocontain the bulk o f th e mass: M$_{d ust }^{c o l d} >$0 . 95 $ \ti me s $M$ _ { dust}^{total}$) .
Takingam ol ecular gas – to– du st rati o $\sim $100( Seaquis t et al.[@sea04]) ,we f o r ese e that the total m olecularg as ma s sof th e d isk mi gh t b e M$_ {gas}$ $ \si m$(4– 7)$\ti me s$10$^ {10}$ M$ _{\sun}$ . If we allow for a min or con tribu tio n | on_the typical_scales of the disk._However the_non-detection_of an_extended_disk at 3mm_argues against a_significant free-free emission component_at 1mm. Furthermore,_there_is no detection of a thermal outflow-like component either at 1mm or at 3mm.
If_the_1mm flux_of_the_disk can be attributed to_dust emission, a mass budget_(M$_{dust}$) can_be derived provided that values for the temperature_(T$_{dust}$)_and for the_emissivity of dust ($\kappa$) are assumed. IRAS observations reveal_significant amounts of warm dust in_4C31.04. From the_flux_ratio_measured by IRAS S$_{\nu}$(100$\mu$)/S$_{\nu}$(60$\mu$)$\sim$3.5,_we derive a range for T$_{dust}$=25–30K_for a corresponding range of $\beta$=2–1_in the wavelength dependency of $\kappa \sim_\lambda^{-\beta}$. Taking $\kappa$(850$\mu$)=0.9cm$^{2}$gr$^{-1}$ from Klaas et_al. ([@kla01]), and T$_{dust}$=30K (i.e., the_solution for_$\beta$=1) we derive M$_{dust}\sim$6$\times$10$^{7}$M$_{\sun}$ for_the warm dust._To better_estimate the total_dust mass we have fitted the_fluxes measured by_IRAS (at 100$\mu$ and 60$\mu$) and_the_PdBI (at 1.3mm)_using_a_two component_model for the_dust_temperatures. Any_plausible_fit to the observations gives a_value_for M$_{dust}$ of (4–7)$\times$10$^{8}$M$_{\sun}$. The higher (lower)_value of M$_{dust}$ corresponds_to_the fit with $\beta$=1.5_($\beta$=1) and temperatures of $\sim$16K(21K)_and $\sim$32K(39K) for the cold and_warm dust_components, respectively._The cold dust component seems to contain the bulk of the_mass: M$_{dust}^{cold}>$0.95$\times$M$_{dust}^{total}$).
Taking a molecular gas–to–dust ratio_$\sim$100 (Seaquist et al. [@sea04]),_we foresee_that_the total molecular_gas_mass of_the disk might be M$_{gas}$$\sim$(4–7)$\times$10$^{10}$M$_{\sun}$. If we_allow for_a minor contribution |
for $\varpi > 0.229855$). Unfortunately, the only results of the type $\operatorname*{EH}[\vartheta]$ that are known come from the Bombieri-Vinogradov theorem (Theorem \[bv-thm\]), which only establishes $\operatorname*{EH}[\vartheta]$ for $0 < \vartheta < 1/2$.
The first unconditional bound on $H_1$ was established in a breakthrough work of Zhang [@zhang]:
\[zhang-thm\] $H_1 \leq \num{70000000}$.
Zhang’s argument followed the general strategy from [@gpy] on finding small gaps between primes, with the major new ingredient being a proof of a weaker version of $\operatorname*{EH}[\frac{1}{2}+2\varpi]$, which we call $\operatorname*{MPZ}[\varpi,\delta]$; see Claim \[mpz-claim\] below. It was quickly realized that Zhang’s numerical bound on $H_1$ could be improved. By optimizing many of the components in Zhang’s argument, we were able [@polymath8a; @polymath8a-unabridged] to improve Zhang’s bound to $$H_1 \leq \num{4680}.$$
Very shortly afterwards, a further breakthrough was obtained by Maynard [@maynard-new] (with related work obtained independently in unpublished work of Tao), who developed a more flexible “multidimensional” version of the Selberg sieve to obtain stronger bounds on $H_m$. This argument worked without using any equidistribution results on primes beyond the Bombieri-Vinogradov theorem, and amongst other things was able to establish finiteness of $H_m$ for all $m$, not just for $m=1$. More precisely, Maynard established the following results.
Unconditionally, we have the following bounds:
- $H_1 \leq 600$.
- $H_m \leq C m^3 e^{4m}$ for all $m \geq 1$ and an absolute (and effective) constant $C$.
Assuming the Elliott-Halberstam conjecture $\operatorname*{EH}[\vartheta]$ for all $0 < \vartheta < 1$, we have the following improvements:
- $H_1 \leq 12$.
- $H_2 \leq 600$.
- $H_m \ | for $ \varpi > 0.229855 $). Unfortunately, the only results of the type $ \operatorname*{EH}[\vartheta]$ that are known hail from the Bombieri - Vinogradov theorem (Theorem \[bv - thm\ ]), which merely establishes $ \operatorname*{EH}[\vartheta]$ for $ 0 < \vartheta < 1/2$.
The first unconditional restrict on $ H_1 $ was established in a breakthrough oeuvre of Zhang [ @zhang ]:
\[zhang - thm\ ] $ H_1 \leq \num{70000000}$.
Zhang ’s argumentation followed the cosmopolitan scheme from [ @gpy ] on finding small col between primes, with the major new ingredient being a validation of a weaker version of $ \operatorname*{EH}[\frac{1}{2}+2\varpi]$, which we visit $ \operatorname*{MPZ}[\varpi,\delta]$; see Claim \[mpz - claim\ ] below. It was promptly realized that Zhang ’s numerical bound on $ H_1 $ could be improved. By optimize many of the components in Zhang ’s argument, we were able [ @polymath8a; @polymath8a - unabridged ] to improve Zhang ’s jump to $ $ H_1 \leq \num{4680}.$$
Very shortly afterwards, a further breakthrough was obtained by Maynard [ @maynard - new ] (with related work prevail independently in unpublished work of Tao), who developed a more flexible “ multidimensional ” version of the Selberg sieve to obtain stronger bounds on $ H_m$. This argument worked without using any equidistribution results on primes beyond the Bombieri - Vinogradov theorem, and amongst other things was able to establish finiteness of $ H_m$ for all $ m$, not just for $ m=1$. More precisely, Maynard established the follow results.
Unconditionally, we have the following boundary:
- $ H_1 \leq 600$.
- $ H_m \leq C m^3 e^{4m}$ for all $ m \geq 1 $ and an absolute (and effective) constant $ C$.
Assuming the Elliott - Halberstam conjecture $ \operatorname*{EH}[\vartheta]$ for all $ 0 < \vartheta < 1 $, we accept the following improvements:
- $ H_1 \leq 12$.
- $ H_2 \leq 600$.
- $ H_m \ | fog $\varpi > 0.229855$). Unfortunately, uhe only results of the vype $\opsratornaoe*{EH}[\vartheta]$ that are known rome from the Bombieri-Vinogradox theorem (Theorem \[bv-tim\]), which only esvzblishes $\operzbornakx*{EH}[\vartheta]$ for $0 < \varthetd < 1/2$.
The first utcunbitional bound on $H_1$ was established yn a brranthrough work jf Znwng [@ahang]:
\[zhang-thm\] $H_1 \leq \num{70000000}$.
Zhang’s argujent fonlowed the gemeral strategy from [@gpy] on finfing small gaps behween primew, wiey the major vew ingredpznt being a proof of a weaker version of $\ooeratprname*{EH}[\frqc{1}{2}+2\caroh]$, which we rall $\oieratorname*{MPD}[\narpi,\denta]$; see Claim \[mpz-claik\] bxlow. It was quickly realived that Zhang’s numetical boung kn $H_1$ could be impeoced. Bi opthmizkbg oang pf the clmpknents in Ahang’s argunent, we were able [@pplrnath8a; @polymatg8a-unabwiqged] to improve Zhang’s bound to $$H_1 \leq \nlm{4680}.$$
Vedy shortly afterwards, a further breakthrough was obtayned by Maynard [@maynard-new] (with related work obtahned mnaeptneentlh ij unpublished work of Tao), who developed a morq fkeqible “multidimenslonal” version of tne Srjberg sieve tu obtanh atronger bounds on $H_m$. Thif argyment worhed eithout using any equidistrubution resujrs on primes beyonb the Bombiexi-Vinobradof theorem, and amongst ocher tgings was ahle to esfxblish finitenesr on $H_k$ for all $m$, not just for $m=1$. More preriselv, Maynara esjablishqd the folpowinn results.
Unconditiojally, wa have the following bounds:
- $H_1 \leq 600$.
- $H_m \leq C m^3 e^{4m}$ fot anl $k \geq 1$ aud an sbsolute (and qffective) consjant $C$.
Assbming ghe Elliotn-Halberstem conjecturq $\operatornama*{GH}[\vartheta]$ foc all $0 < \vwrthwta < 1$, we haxd the followinb improvements:
- $H_1 \oeq 12$.
- $H_2 \leq 600$.
- $H_m \ | for $\varpi > 0.229855$). Unfortunately, the only the $\operatorname*{EH}[\vartheta]$ that known come from which establishes $\operatorname*{EH}[\vartheta]$ for < \vartheta < The first unconditional bound on $H_1$ established in a breakthrough work of Zhang [@zhang]: \[zhang-thm\] $H_1 \leq \num{70000000}$. Zhang’s followed the general strategy from [@gpy] on finding small gaps between primes, with major ingredient a of a weaker version of $\operatorname*{EH}[\frac{1}{2}+2\varpi]$, which we call $\operatorname*{MPZ}[\varpi,\delta]$; see Claim \[mpz-claim\] below. It was realized that Zhang’s numerical bound on $H_1$ could improved. By optimizing many the components in Zhang’s argument, were [@polymath8a; @polymath8a-unabridged] improve bound $$H_1 \leq \num{4680}.$$ shortly afterwards, a further breakthrough was obtained by Maynard [@maynard-new] (with related work obtained independently in unpublished of Tao), a more “multidimensional” of Selberg sieve to bounds on $H_m$. This argument worked equidistribution results on primes beyond the Bombieri-Vinogradov theorem, amongst other was able to establish finiteness of for all $m$, not just for $m=1$. More Maynard established the following results. Unconditionally, we have the following bounds: - $H_1 \leq 600$. \leq C m^3 e^{4m}$ all $m \geq and absolute effective) $C$. Assuming Elliott-Halberstam conjecture $\operatorname*{EH}[\vartheta]$ for all $0 < \vartheta < 1$, we the following improvements: - $H_1 \leq 12$. - $H_2 \leq $H_m | for $\varpi > 0.229855$). Unfortunately, the oNly results Of the TypE $\opErAtorName*{eH}[\vartheta]$ that ARe knOwn come from the Bombieri-vinogRaDOv thEOrEm (TheOrem \[bv-tHM\]), wHICh oNlY eStaBlIShEs $\opeRatOrname*{Eh}[\vartheta]$ fOr $0 < \vArTheta < 1/2$.
The firsT UnConditionaL boUnd on $H_1$ was estAblIshed iN a BreAKthroUgh Work oF Zhang [@ZHang]:
\[zhAng-thm\] $H_1 \leQ \nUM{70000000}$.
Zhang’S ArgumenT FOlLoweD the general strateGY fROm [@gpy] on finding Small gApS BeTWEen PriMes, with the MaJor neW IngrediENt BEINg a PRoof of a weaker Version of $\opERatOrname*{eH}[\FraC{1}{2}+2\Varpi]$, wHich wE cALl $\oPeratorname*{mPZ}[\vArpi,\delta]$; See ClaIM \[mpz-claIM\] below. IT was quIckLy rEaliZEd ThAt ZHaNG’s nUMeRicAL boUnd on $H_1$ coUlD bE imprOved. bY OPTimiZinG manY of thE components in zhaNg’s aRGumEnt, we Were aBle [@pOlYmath8A; @polymAth8a-uNaBridged] to improvE ZhaNg’s bound tO $$H_1 \lEq \Num{4680}.$$
veRy shoRTly aftErwArdS, a furthEr breakTHroUgH WAS oBtained by Maynard [@maYnARD-nEw] (with reLated wORk ObTAined indEpEndEntlY IN unpuBlisHEd Work of TaO), who deVElOpEd a more FlExible “MuLtiDimEnsioNAl” veRsion oF the SelbErg siEVe to obtain stroNGer bounds on $H_m$. tHiS ARgUMent WorKed without uSing ANy eqUidiSTrIbuTIon reSults On PRiMEs beyond the Bombieri-viNogradOv theOrem, and amongsT other thinGS WAs able to EstaBLiSH finiteness of $H_M$ for aLl $m$, not just FOr $m=1$. More pRecisEly, MaynaRd establiSHEd the folLowIng ResUltS.
uNcOnditionally, wE HAve tHe FollowiNg bOunds:
- $H_1 \lEq 600$.
- $H_M \leQ C m^3 E^{4m}$ fOr All $m \geq 1$ anD an absolUtE (aNd EfFecTive) cONstant $C$.
ASsUmiNg The ellioTT-HalbeRstam ConjEcTuRE $\opEratornAMe*{eh}[\VartHeTa]$ For aLl $0 < \vArTheta < 1$, We haVE thE followIng improvEmeNTs:
- $H_1 \lEq 12$.
- $h_2 \lEq 600$.
- $H_m \ | for $\varpi > 0.229855$). Unfortuna tely, th e o nl y re sult s of the type$ \ope ratorname*{EH}[\varthe ta]$th a t ar e k nowncome fr o mt h e B om bi eri -V i no grado v t heorem(Theorem \ [bv -t hm\]), which on ly establi she s $\operator nam e*{EH} [\ var t heta] $ f or $0 < \va r theta< 1/2$.
Th e first uncondi t i on al b ound on $H_1$ was es t ablished in abreakt hr o ug h wor k o f Zhang [@ zh ang]:
\[zhan g -t h m \ ] $ H _1 \leq \num{ 70000000}$.
Zh ang’sar gum e nt fol lowed t h e g eneral stra tegy from [@g py] on finding small g aps be twe enprim e s, w ith t h e m a jo r n e w i ngredien tbe ing a pro o f o f awea kerversi on of $\opera tor name * {EH }[\fr ac{1} {2}+ 2\ varpi ]$, wh ich w ecall $\operator name *{MPZ}[\v arp i, \de lt a]$;s ee Cla im\[m pz-clai m\] bel o w.It w a squickly realized t ha t Zh ang’s nu merica l b ou n d on $H_ 1$ co uldb e impr oved . B y optimi zing m a ny o f the c om ponent sinZha ng’sa rgum ent, w e were a ble [ @ polymath8a; @p o lymath8a-unab r id g e d] to i mpr ove Zhang’s bou n d to $$H _ 1\le q \num {4680 }. $ $Very shortly afterw ar ds, afurth er breakthrou gh was obt a i n ed by Ma ynar d [ @ maynard-new] ( withrelated wo r k obtain ed in dependen tly in un p u blishedwor k o f T ao) , wh o developed a m orefl exible“mu ltidime nsi ona l”ver si on of the Selberg s ie ve t o o btain stronger b oun ds on $H_m $ . This argu ment w or k edwithout us i n g an yeq uidi str ib ution res u lts on pri mes beyon d t h e Bo mb ie ri-Vino gradov theore m, and among st ot her th i n gs was a ble to establish finite n ess of$H_ m$ fo r al l $m$, no t j ust fo r $ m =1$. M ore pr ecise ly , M a y narde s ta bli sh ed the fol l o win g res ul ts.
Uncond itionally, we have the following bo und s:
- $H_ 1 \ l eq60 0 $.- $H_m \leq Cm^3 e^{4m} $f or all $m \g e q 1 $and anabsolut e (an d effect ive) cons tant $C$.
Assu m i ngthe Elliot t-Halber stam conj e cture $\ opera tor name*{ EH }[\ varth eta]$f orall $ 0 < \v ar theta< 1$, w e have t he following improvemen ts:
- $H _1\leq 12$.
- $ H_2 \leq600$ .
- $H_ m \ | for_$\varpi >_0.229855$). Unfortunately, the only_results of_the_type $\operatorname*{EH}[\vartheta]$_that_are known come_from the Bombieri-Vinogradov_theorem (Theorem \[bv-thm\]), which_only establishes $\operatorname*{EH}[\vartheta]$_for_$0 < \vartheta < 1/2$.
The first unconditional bound on $H_1$ was established in a_breakthrough_work of_Zhang_[@zhang]:
\[zhang-thm\]_$H_1 \leq \num{70000000}$.
Zhang’s argument followed_the general strategy from [@gpy]_on finding_small gaps between primes, with the major new_ingredient_being a proof_of a weaker version of $\operatorname*{EH}[\frac{1}{2}+2\varpi]$, which we call_$\operatorname*{MPZ}[\varpi,\delta]$; see Claim \[mpz-claim\] below. It_was quickly realized_that_Zhang’s_numerical bound on $H_1$_could be improved. By optimizing many_of the components in Zhang’s argument,_we were able [@polymath8a; @polymath8a-unabridged] to improve_Zhang’s bound to $$H_1 \leq \num{4680}.$$
Very_shortly afterwards, a further breakthrough_was obtained_by Maynard [@maynard-new] (with related_work obtained independently_in unpublished_work of Tao),_who developed a more flexible “multidimensional”_version of the_Selberg sieve to obtain stronger bounds_on_$H_m$. This argument_worked_without_using any_equidistribution results on_primes_beyond the_Bombieri-Vinogradov_theorem, and amongst other things was_able_to establish finiteness of $H_m$ for all_$m$, not just for_$m=1$._More precisely, Maynard established_the following results.
Unconditionally, we have_the following bounds:
- $H_1_\leq 600$.
-_ _$H_m \leq C m^3 e^{4m}$ for all $m \geq 1$ and_an absolute (and effective) constant $C$.
Assuming_the Elliott-Halberstam conjecture $\operatorname*{EH}[\vartheta]$_for all_$0_< \vartheta <_1$,_we have_the following improvements:
- $H_1 \leq_12$.
- _ $H_2 \leq 600$.
- _$H_m \ |
_i + \bar T_i)\, \right] \right. \nonumber \\
& & \left. -\ \frac{1}{8 \pi^2}\ \sum_i\,
(b^{\prime i}_a - \delta^i_{GS})\, \ln \left[\, |\eta (T_i)|^4
(T_i + \bar T_i)\, \right]\, \right\}\ +\ \mbox{h.c.}\.
\label{eq:L_GK_loop}\end{aligned}$$ Eq. (\[eq:L\_GK\_loop\]), together with Eq. (\[eq:K\_S\_loop\]), shows that the structure of one-loop corrections to the effective string action is strongly constrained by target-space duality anomaly cancellation. Note that anomaly considerations do not tell us anything about possible holomorphic, modular invariant corrections [@hpnss] that are not included in Eq. (\[eq:L\_GK\_loop\]).
[**Appendix B: Green-Schwarz mechanism in the linear multiplet formalism**]{}
The Green-Schwarz mechanism that cancels part of the sigma-model anomalies in heterotic orbifolds can be naturally described in the linear multiplet formalism[^18]. Indeed, in terms of the string massless states, the axion-dilaton-dilatino system fits into a linear multiplet $L = (l, B_{\mu \nu}, \chi)$, where the antisymmetric two-tensor $B_{\mu \nu}$ is dual to the model-independent axion $a = \mbox{Im} S$, $\partial_\mu a \sim \epsilon_{\mu \nu \rho \sigma}\, \partial^{\nu}
B^{\rho \sigma}$. $L$ couples to the gauge fields in such a way that the combination $\widehat L = L - 2 \Omega$, where $\Omega$ is the Chern-Simons superfield defined by $\bar D^2 \Omega = \sum_a W^a W^a$ and $D^2 \Omega = \sum_a \bar W^a \bar W^a$, is gauge invariant. A gauge-invariant Lagrangian for $L$ then takes the simple form ${\cal L}_L = \int \!
\mbox{d}^4 \theta\ \Phi (\widehat L)$. The transformation to the dual formulation in terms of the dilaton chiral superfield is accomplished | _ i + \bar T_i)\, \right ] \right. \nonumber \\
& & \left. -\ \frac{1}{8 \pi^2}\ \sum_i\,
(b^{\prime i}_a - \delta^i_{GS})\, \ln \left[\, |\eta (T_i)|^4
(T_i + \bar T_i)\, \right]\, \right\}\ + \ \mbox{h.c.}\.
\label{eq: L_GK_loop}\end{aligned}$$ Eq. (\[eq: L\_GK\_loop\ ]), together with Eq. (\[eq: K\_S\_loop\ ]), shows that the structure of one - loop correction to the effective chain action is strongly constrained by aim - space duality anomaly cancellation. notice that anomaly considerations do not tell us anything about possible holomorphic, modular changeless correction [ @hpnss ] that are not included in Eq. (\[eq: L\_GK\_loop\ ]).
[ * * Appendix B: Green - Schwarz mechanism in the analogue multiplet formalism * * ] { }
The Green - Schwarz mechanism that cancels part of the sigma - model anomalies in heterotic orbifolds can be naturally trace in the linear multiplet formalism[^18 ]. Indeed, in terms of the string massless states, the axion - dilaton - dilatino system match into a linear multiplet $ L = (l, B_{\mu \nu }, \chi)$, where the antisymmetric two - tensor $ B_{\mu \nu}$ is dual to the exemplary - independent axion $ a = \mbox{Im } S$, $ \partial_\mu a \sim \epsilon_{\mu \nu \rho \sigma}\, \partial^{\nu }
B^{\rho \sigma}$. $ L$ couples to the bore fields in such a means that the combination $ \widehat L = fifty - 2 \Omega$, where $ \Omega$ is the Chern - Simons superfield defined by $ \bar D^2 \Omega = \sum_a W^a W^a$ and $ D^2 \Omega = \sum_a \bar W^a \bar W^a$, is gauge invariant. A bore - invariant Lagrangian for $ L$ then takes the simple form $ { \cal L}_L = \int \!
\mbox{d}^4 \theta\ \Phi (\widehat L)$. The transformation to the dual formulation in terms of the dilaton chiral superfield is accomplished | _i + \har T_i)\, \right] \right. \nonuober \\
& & \left. -\ \frac{1}{8 \pm^2}\ \sum_i\,
(b^{\prime k}_a - \delta^i_{GS})\, \ln \left[\, |\eta (T_i)|^4
(T_u + \bae T_i)\, \right]\, \right\}\ +\ \mbox{f.c.}\.
\label{eq:P_GK_loop}\ebd{almgned}$$ Eq. (\[eq:L\_GK\_loop\]), togetmzr wifm Eq. (\[zq:J\_S\_loop\]), shows tmat the strgcture of one-lmoo eorrections to the effective string wction os strongly consjraintd fy tzggtt-space duality anomaly cancellatjon. Nott that anomaly conxiderations do not tell us anyhhing about possibpe holomorpyic, iidular invarkant correbcions [@hpnss] that are not included in Eq. (\[eq:U\_GK\_lopp\]).
[**Appendix B: Grfgn-Schwarz merhanisi in the lincsr muldiplet gormalism**]{}
The Gveen-Srhwaez mechanism that canrels part of the sigia-model atojalies in heterotuc orbixoldv cav be nauurellg descgibxd in the ljnear multiplet formalism[^18]. Indetd, yb terms of ths strigg massless states, the axion-dilaton-dilatito aystem fits into a lineqr multiplet $L = (l, B_{\mu \nu}, \chi)$, wrere the antisymmetric two-tensor $B_{\mu \nu}$ is dual tm the oodtl-lkdepdbdfnt axion $a = \mbox{Im} S$, $\partial_\mu a \sim \epsilon_{\ih \mu \rho \sigma}\, \parbial^{\nu}
B^{\rho \sigma}$. $L$ cluljes to the gaoge fields in such a way thah the cjmbinqtion $\widthat K = L - 2 \Omega$, where $\Omega$ is the Chern-Siiins superfield defnned by $\bar B^2 \Omegs = \suk_a W^a W^a$ and $D^2 \Omega = \sbm_a \bad W^a \bar W^a$, is gauge knvariant. A gaugd-innaridnt Lagrangian for $L$ then eakes the sim'le form ${\cal L}_L = \igt \!
\mbox{d}^4 \tjeta\ \Ivi (\widehat L)$. The tgansflrkation to hhe dual formulation in terms oh the dilaton cvirdl superyield ls accomplished | _i + \bar T_i)\, \right] \right. \nonumber & -\ \frac{1}{8 \sum_i\, (b^{\prime i}_a (T_i)|^4 + \bar T_i)\, \right\}\ +\ \mbox{h.c.}\. Eq. (\[eq:L\_GK\_loop\]), together with Eq. (\[eq:K\_S\_loop\]), that the structure of one-loop corrections to the effective string action is strongly by target-space duality anomaly cancellation. Note that anomaly considerations do not tell us about holomorphic, invariant [@hpnss] that are not included in Eq. (\[eq:L\_GK\_loop\]). [**Appendix B: Green-Schwarz mechanism in the linear multiplet The Green-Schwarz mechanism that cancels part of the anomalies in heterotic orbifolds be naturally described in the multiplet Indeed, in of string states, the axion-dilaton-dilatino fits into a linear multiplet $L = (l, B_{\mu \nu}, \chi)$, where the antisymmetric two-tensor $B_{\mu \nu}$ dual to axion $a \mbox{Im} $\partial_\mu \sim \epsilon_{\mu \nu \partial^{\nu} B^{\rho \sigma}$. $L$ couples to in such a way that the combination $\widehat = L 2 \Omega$, where $\Omega$ is the superfield defined by $\bar D^2 \Omega = \sum_a W^a$ and $D^2 \Omega = \sum_a \bar W^a \bar W^a$, is gauge invariant. A gauge-invariant $L$ then takes the form ${\cal L}_L \int \mbox{d}^4 \Phi L)$. The to the dual formulation in terms of the dilaton chiral superfield accomplished | _i + \bar T_i)\, \right] \right. \nonumber \\
& & \lEft. -\ \frac{1}{8 \pi^2}\ \sUm_i\,
(b^{\pRimE i}_a - \DeLta^i_{gS})\, \ln \Left[\, |\eta (T_i)|^4
(T_i + \bar t_I)\, \rigHt]\, \right\}\ +\ \mbox{h.c.}\.
\label{eq:L_Gk_loop}\EnD{AligNEd}$$ eq. (\[eq:L\_gK\_loop\]), tOGeTHEr wItH EQ. (\[eq:k\_S\_LOoP\]), showS thAt the stRucture of oNe-lOoP corrections TO tHe effectivE stRing action is StrOngly cOnStrAIned bY taRget-sPace duALity anOmaly cancElLAtion. NOTe that aNOMaLy coNsiderations do not TElL Us anything abouT possiBlE HoLOMorPhiC, modular inVaRiant COrrectiONs [@HPNSs] tHAt are not incluDed in Eq. (\[eq:L\_Gk\_LooP\]).
[**AppenDiX B: GREen-SchWarz mEcHAniSm in the lineAr muLtiplet foRmalisM**]{}
the GreeN-schwarz MechanIsm ThaT canCElS pArt Of THe sIGmA-moDEl aNomalies In HeTerotIc orBIFOLds cAn bE natUrallY described in tHe lIneaR MulTipleT formAlisM[^18]. INdeed, In termS of thE sTring massless stAtes, The axion-dIlaToN-diLaTino sYStem fiTs iNto A linear MultiplET $L = (l, b_{\mU \NU}, \ChI)$, where the antisymmeTrIC TwO-tensor $B_{\Mu \nu}$ is DUaL tO The model-InDepEndeNT Axion $A = \mboX{im} s$, $\partial_\Mu a \sim \EPsIlOn_{\mu \nu \rHo \Sigma}\, \pArTiaL^{\nu}
b^{\rho \sIGma}$. $L$ CoupleS to the gaUge fiELds in such a way tHAt the combinatIOn $\WIDeHAt L = L - 2 \omeGa$, where $\OmegA$ is tHE CheRn-SiMOnS suPErfieLd defInED bY $\Bar D^2 \Omega = \sum_a W^a W^a$ anD $D^2 \omega = \sUm_a \baR W^a \bar W^a$, is gauGe invarianT. a GAuge-invaRianT laGRangian for $L$ theN takeS the simple FOrm ${\cal L}_L = \Int \!
\mbOx{d}^4 \theta\ \phi (\widehaT l)$. the transForMatIon To tHE DuAl formulation IN TermS oF the dilAtoN chiral SupErfIelD is AcComplisheD | _i + \bar T_i)\, \right] \ right. \n onumb er\\ & & \le ft. -\ \frac{1 } {8 \ pi^2}\ \sum_i\,
(b^{ \prim ei }_a- \ delta ^i_{GS} ) \, \ ln\l ef t[\ ,| \e ta (T _i) |^4
( T_i + \bar T_ i) \, \right]\, \r ight\}\ +\ \m box{h.c.}\.\la bel{eq :L _GK _ loop} \en d{ali gned}$ $ Eq. ( \[eq:L\_G K\ _ loop\] ) , toget h e rwith Eq. (\[eq:K\_S\_ l oo p \]), shows tha t thest r uc t u reofone-loop c or recti o ns to t h ee f f ect i ve string act ion is stro n gly const ra ine d by ta rget- sp a ceduality ano maly cancella tion.N ote tha t anomal y cons ide rat ions do n otte l l u s a nyt h ing about p os si ble h olom o r p h ic,mod ular inva riant correct ion s [@ h pns s] th at ar e no tinclu ded in Eq.(\ [eq:L\_GK\_loop \]).
[**Appe ndi xB:Gr een-S c hwarzmec han ism inthe lin e armu l t i pl et formalism**]{}
T h e G reen-Sch warz m e ch an i sm thatca nce ls p a r t ofthes ig ma-model anoma l ie sin hete ro tic or bi fol dscan b e nat urally describ ed in the linear mul t iplet formali s m[ ^ 1 8] . Ind eed , in termsof t h e st ring ma ssl e ss st ates, t h ea xion-dilaton-dilati no syste m fit s into a line ar multipl e t $L = (l, B_{ \ mu \nu}, \chi)$,where the antis y mmetrictwo-t ensor $B _{\mu \nu } $ is dual to th e m ode l - in dependent axi o n $a=\mbox{I m}S$, $\p art ial _\m u a \ sim \epsi lon_{\mu \ nu \ rh o \ sigma } \, \part ia l^{ \n u}B^{\r h o \sig ma}$. $L$ c ou p les to the ga u g e fi el ds insuc ha way tha t th e combi nation $\ wid e hatL=L - 2 \ Omega$, where $ \Omega$ is t heChern- S i mons sup erfield defined by $\ba r D^2 \O meg a = \ sum_ a W^a W^a $ a nd $D^ 2 \ O mega = \sum_ a \ba rW^a \ bar W ^ a $, is g auge invar i a nt. A ga ug e-in variant Lagrangian for $L $ th en takes thesim plef o rm ${ \ ca l L} _L = \ i n t \!
\mbox{d}^4 \theta\ \ Ph i ( \widehat L ) $.Th e trans formati on to the dua l formula tion in t er ms o f the dilaton c hiral su perfieldi s acc o mp lishe d | _i +_\bar T_i)\,_\right] \right. \nonumber_\\
_&_& _\left._-\ \frac{1}{8 \pi^2}\_\sum_i\,
(b^{\prime_i}_a - \delta^i_{GS})\, \ln_\left[\, |\eta (T_i)|^4
__(T_i + \bar T_i)\, \right]\, \right\}\ +\ \mbox{h.c.}\.
\label{eq:L_GK_loop}\end{aligned}$$ Eq. (\[eq:L\_GK\_loop\]), together with Eq. (\[eq:K\_S\_loop\]),_shows_that the_structure_of_one-loop corrections to the effective_string action is strongly constrained_by target-space_duality anomaly cancellation. Note that anomaly considerations do_not_tell us anything_about possible holomorphic, modular invariant corrections [@hpnss] that are_not included in Eq. (\[eq:L\_GK\_loop\]).
[**Appendix B:_Green-Schwarz mechanism in_the_linear_multiplet formalism**]{}
The Green-Schwarz mechanism_that cancels part of the sigma-model_anomalies in heterotic orbifolds can be_naturally described in the linear multiplet formalism[^18]._Indeed, in terms of the string_massless states, the axion-dilaton-dilatino system_fits into_a linear multiplet $L =_(l, B_{\mu \nu},_\chi)$, where_the antisymmetric two-tensor_$B_{\mu \nu}$ is dual to the_model-independent axion $a_= \mbox{Im} S$, $\partial_\mu a \sim_\epsilon_{\mu_\nu \rho \sigma}\,_\partial^{\nu}
B^{\rho_\sigma}$._$L$ couples_to the gauge_fields_in such_a_way that the combination $\widehat L_=_L - 2 \Omega$, where $\Omega$ is_the Chern-Simons superfield defined_by_$\bar D^2 \Omega =_\sum_a W^a W^a$ and $D^2_\Omega = \sum_a \bar W^a \bar_W^a$, is_gauge invariant._A gauge-invariant Lagrangian for $L$ then takes the simple form ${\cal_L}_L = \int \!
\mbox{d}^4 \theta\ \Phi_(\widehat L)$. The transformation_to the_dual_formulation in terms_of_the dilaton_chiral superfield is accomplished |
/\psi\omega$ (P wave) is observed, the heavy quark spin symmetry implies that it is not a $c\bar{c}$ meson and the $J^{PC}$ are very likely to be $3^{-+}$.'
author:
- 'W. Zhu, T. Yao, Yan-Rui Liu'
title: 'Possibility of a $J^{PC}=3^{-+}$ state'
---
Introduction {#sec1}
============
Mesons with exotic properties play an important role in understanding the nature of strong interactions. The observation of the so called XYZ states in the heavy quark sector has triggered lots of discussions on their quark structures, decays, and formation mechanisms. It also motivates people to study new states beyond the quark model assignments.
The X(3872), first observed in the $J\psi\pi^+\pi^-$ invariant mass distribution by Belle collaboration in 2003 [@X3872-belle], is the strangest heavy quark state. Even now, its angular momentum is not determined. Since its extreme closeness to the $D^0\bar{D}^{*0}$ threshold, lots of discussions about its properties are based on the molecule assumption. However, it is very difficult to identify the X(3872) as a shallow bound state of $D^0\bar{D}^{*0}$ since there are no explicitly exotic molecule properties.
A charged charmonium- or bottomonium-like meson labeled as $Z$ is absolutely exotic because its number of quarks and antiquarks must be four or more. Such states include the $Z(4430)$ observed in the $\psi'\pi^\pm$ mass distribution [@Z4430-belle], the $Z_1(4050)$ and $Z_2(4250)$ observed in the $\chi_{c1}\pi^+$ mass distribution [@Z1Z2-belle], and the $Z_b(10610)$ and $Z_b(10650)$ in the mass spectra of the $\Upsilon(nS)\pi^\pm$ ($n$=1,2,3) and $\pi^\pm h_b(mP)$ ($m$=1,2) [@Zb-belle]. They are all observed by Belle Collaboration. Though BABAR has not confirmed them [@Z4430-babar; @Z1Z2-babar], the existence signal of multiquark states is still | /\psi\omega$ (P wave) is observed, the heavy quark spin isotropy entail that it is not a $ c\bar{c}$ meson and the $ J^{PC}$ are very likely to be $ 3^{-+}$.'
author:
-' W. Zhu, T. Yao, Yan - Rui Liu'
claim:' hypothesis of a $ J^{PC}=3^{-+}$ state'
---
Introduction { # sec1 }
= = = = = = = = = = = =
Mesons with exotic property play an important function in understanding the nature of strong interaction. The observation of the so call XYZ states in the heavy quark sector has triggered lots of discussions on their quark structure, decays, and formation mechanisms. It besides motivates people to analyze new state beyond the quark model assignments.
The X(3872), first observed in the $ J\psi\pi^+\pi^-$ invariant aggregate distribution by Belle collaboration in 2003 [ @X3872 - belle ], is the strangest heavy quark state. Even now, its angular momentum is not determined. Since its extreme meanness to the $ D^0\bar{D}^{*0}$ threshold, lots of discussions about its properties are based on the molecule assumption. However, it is very difficult to identify the X(3872) as a shallow bound state of matter of $ D^0\bar{D}^{*0}$ since there are no explicitly exotic atom place.
A charge charmonium- or bottomonium - like meson labeled as $ Z$ is absolutely alien because its number of quark and antiquarks must be four or more. Such states admit the $ Z(4430)$ observed in the $ \psi'\pi^\pm$ mass distribution [ @Z4430 - belle ], the $ Z_1(4050)$ and $ Z_2(4250)$ observed in the $ \chi_{c1}\pi^+$ mass distribution [ @Z1Z2 - belle ], and the $ Z_b(10610)$ and $ Z_b(10650)$ in the mass spectra of the $ \Upsilon(nS)\pi^\pm$ ($ n$=1,2,3) and $ \pi^\pm h_b(mP)$ ($ m$=1,2) [ @Zb - belle ]. They are all observed by Belle Collaboration. Though BABAR has not confirmed them [ @Z4430 - babar; @Z1Z2 - babar ], the existence signal of multiquark states is still | /\psi\lmega$ (P wave) is observed, the heavy quark spin vymmetdy implids that it is not a $c\bar{c}$ medob and the $J^{PC}$ are very likeuy to be $3^{-+}$.'
wuthor:
- 'W. Zhu, R. Yao, Yan-Rnj Liu'
tibje: 'Pkdsibnlmty of a $J^{PC}=3^{-+}$ stste'
---
Introduwtion {#sec1}
============
Mesonv dich exotic properties play an importagt role ij understandind tht nwturs of strong interactions. The obserbation mf the so calked XYZ states in the heavj quwrk sector has trihgered lots of quscussions ov their quark structurgs, decays, and formation mechanisos. It also moticares keople to stndy ner states beyond the xuark mpdel assignmenbs.
The X(3872), dirst observed in the $J\psi\pi^+\pi^-$ invarianj mass disdrnbution by Belle collqbiratimn it 2003 [@X3872-cwlld], ia vhe stranhesv heavy quadk state. Evwn now, its angular koivmtum is not setermynqd. Since its extreme closeness to the $D^0\tar{S}^{*0}$ threshold, lots of disxussions about its prlperties wre based on the molecule assumption. However, it iv verb aifylgult ro identify the X(3872) as a shallow bound state of $D^0\fzr{C}^{*0}$ xince there arc no explicitly exptlc iolecule propgrties.
A chzrged charmonium- og bottoioniun-like mesjn lsbeled as $Z$ is absolutely ezotic becausv itw number of quarks and antiqucrks most be four or more. Such statzs incmude the $Z(4430)$ lbserved jv the $\psi'\pi^\pm$ mars civtribution [@Z4430-belle], the $Z_1(4050)$ anq $Z_2(4250)$ obserted iu the $\chk_{c1}\pi^+$ mass qistributiln [@Z1Z2-nalle], and the $Z_b(10610)$ anf $Z_b(10650)$ nn tha mass speftra of the $\Upsilon(nS)\pi^\pm$ ($n$=1,2,3) and $\pi^\pm h_b(mP)$ ($m$=1,2) [@Zn-ballv]. They arz all pbserved by Bqlle Collaborajion. Thouyh BABXR has not confirked them [@Z4430-bwbar; @Z1Z2-babar], dje existence signal jf myltiwuark sgxtes is still | /\psi\omega$ (P wave) is observed, the heavy symmetry that it not a $c\bar{c}$ very to be $3^{-+}$.' - 'W. Zhu, Yao, Yan-Rui Liu' title: 'Possibility of $J^{PC}=3^{-+}$ state' --- Introduction {#sec1} ============ Mesons with exotic properties play an important in understanding the nature of strong interactions. The observation of the so called states the quark has triggered lots of discussions on their quark structures, decays, and formation mechanisms. It also motivates to study new states beyond the quark model The X(3872), first observed the $J\psi\pi^+\pi^-$ invariant mass distribution Belle in 2003 is strangest quark state. Even its angular momentum is not determined. Since its extreme closeness to the $D^0\bar{D}^{*0}$ threshold, lots of discussions its properties on the assumption. it very difficult to X(3872) as a shallow bound state there are no explicitly exotic molecule properties. A charmonium- or meson labeled as $Z$ is absolutely because its number of quarks and antiquarks must four or more. Such states include the $Z(4430)$ observed in the $\psi'\pi^\pm$ mass distribution [@Z4430-belle], and $Z_2(4250)$ observed in $\chi_{c1}\pi^+$ mass distribution and $Z_b(10610)$ $Z_b(10650)$ the mass of the $\Upsilon(nS)\pi^\pm$ ($n$=1,2,3) and $\pi^\pm h_b(mP)$ ($m$=1,2) [@Zb-belle]. They are observed by Belle Collaboration. Though BABAR has not confirmed them the signal of multiquark is still | /\psi\omega$ (P wave) is observed, thE heavy quarK spin SymMetRy ImplIes tHat it is not a $c\baR{C}$ mesOn and the $J^{PC}$ are very likeLy to bE $3^{-+}$.'
aUThor:
- 'w. zhU, T. Yao, yan-Rui LIU'
tITLe: 'POsSiBilItY Of A $J^{PC}=3^{-+}$ sTatE'
---
IntrodUction {#sec1}
============
MEsoNs With exotic prOPeRties play aN imPortant role iN unDerstaNdIng THe natUre Of strOng intERactioNs. The obseRvATion of THe so calLED XyZ stAtes in the heavy quaRK sECtor has triggerEd lots Of DIsCUSsiOns On their quaRk StrucTUres, decAYs, AND ForMAtion mechanisMs. It also motIVatEs peopLe To sTUdy new StateS bEYonD the quark moDel aSsignmentS.
The X(3872), fIRst obseRVed in thE $J\psi\pI^+\pi^-$ InvAriaNT mAsS diStRIbuTIoN by bEllE collaboRaTiOn in 2003 [@X3872-BellE], IS THe stRanGest Heavy Quark state. EveN noW, its ANguLar moMentuM is nOt DeterMined. SInce iTs Extreme closenesS to tHe $D^0\bar{D}^{*0}$ thResHoLd, lOtS of diSCussioNs aBouT its proPerties ARe bAsED ON tHe molecule assumptiOn. hOWeVer, it is vEry difFIcUlT To identiFy The x(3872) as a SHAllow BounD StAte of $D^0\baR{D}^{*0}$ sincE ThErE are no eXpLicitlY eXotIc mOlecuLE proPertieS.
A chargeD charMOnium- or bottomoNIum-like meson lABeLED aS $z$ is aBsoLutely exotiC becAUse iTs nuMBeR of QUarks And anTiQUaRKs must be four or more. SUcH stateS inclUde the $Z(4430)$ observEd in the $\psi'\PI^\PM$ mass disTribUTiON [@Z4430-belle], the $Z_1(4050)$ and $z_2(4250)$ obseRved in the $\cHI_{c1}\pi^+$ mass DistrIbution [@Z1z2-belle], and THE $Z_b(10610)$ and $Z_b(10650)$ In tHe mAss SpeCTRa Of the $\Upsilon(ns)\PI^\pm$ ($n$=1,2,3) AnD $\pi^\pm h_b(MP)$ ($m$=1,2) [@zb-belle]. theY arE alL obSeRved by BelLe CollabOrAtIoN. THouGh BABar has not cOnFirMeD thEm [@Z4430-baBAr; @Z1Z2-baBar], thE exiStEnCE siGnal of mULtIQUark StAtEs is StiLl | /\psi\omega$ (P wave) is o bserved, t he he avy qu ar k sp in s ymmetry implie s tha t it is not a $c\bar{c }$ me so n and th e $J^ {PC}$ a r ev e ryli ke lyto be $3^{ -+} $.'
aut hor:
- 'W. Zh u, T. Yao, Yan - Ru i Liu'
tit le: 'Possibilit y o f a $J ^{ PC} = 3^{-+ }$state '
---Introd uction {# se c 1}
=== = ======= =
M eson s with exotic pro p er t ies play an im portan tr ol e inund erstanding t he na t ure ofs tr o n g in t eractions. Th e observati o n o f theso ca l led XY Z sta te s in the heavyquar k sectorhas tr i ggeredl ots ofdiscus sio nson t h ei rqua rk str u ct ure s , d ecays, a nd f ormat ionm e c h anis ms. Italsomotivates peo ple tos tud y new stat es b ey ond t he qua rk mo de l assignments.
The X(3872), fi rs t o bs erved in the $J \ps i\pi^+\ pi^-$ i n var ia n t ma ss distribution by B e l le collabo ration in 2 0 03 [@X38 72 -be lle] , is th e st r an gest hea vy qua r kst ate. Ev en now,it s a ngu lar m o ment um isnot dete rmine d . Since its ex t reme closenes s t o th e $D^ 0\b ar{D}^{*0}$ thr e shol d, l o ts of discu ssion sa bo u t its properties ar ebasedon th e molecule as sumption.H o w ever, it isv er y difficult toident ify the X( 3 872) asa sha llow bou nd stateo f $D^0\ba r{D }^{ *0} $ s i n ce there are no e xpli ci tly exo tic molecu lepro per tie s.
A charg ed charm on iu m- o r b ottom o nium-lik emes on la beled as $Z$ is a bsol ut el y ex otic be c au s e its n um berofqu arksanda nti quarksmust be f our or m or e. Such s tates include t he $Z(4430 )$ ob served i n the $\ psi'\pi^\pm$ mass distr i bution[@Z 4430- bell e], the $ Z_1 (4050) $ a n d $Z_2 (4250) $ obs er ved i n the $ \c hi_ {c 1}\pi^+$ m a s s d istri bu tion [@Z1Z2 -belle], and the $ Z _b( 10610)$ and $ Z_b (106 5 0 )$ in th e ma ss spe c t ra of the $\Ups ilon(nS)\p i^ \ pm $ ($n$=1,2 , 3)an d $\pi^ \pm h_b (mP)$ ($m$=1, 2) [@Zb-b elle]. Th ey are a llobserved b y BelleCollabora t ion.T ho ugh B ABA R hasno t c onfir med th e m [ @Z443 0-baba r; @Z1Z2 -baba r] , the ex istence signal of multi quarkstate s i s still | /\psi\omega$ (P_wave) is_observed, the heavy quark_spin symmetry_implies_that it_is_not a $c\bar{c}$_meson and the_$J^{PC}$ are very likely_to be $3^{-+}$.'
author:
-_'W._Zhu, T. Yao, Yan-Rui Liu'
title: 'Possibility of a $J^{PC}=3^{-+}$ state'
---
Introduction {#sec1}
============
Mesons with exotic properties_play_an important_role_in_understanding the nature of strong_interactions. The observation of the_so called_XYZ states in the heavy quark sector has_triggered_lots of discussions_on their quark structures, decays, and formation mechanisms. It_also motivates people to study new_states beyond the_quark_model_assignments.
The X(3872), first observed_in the $J\psi\pi^+\pi^-$ invariant mass distribution_by Belle collaboration in 2003 [@X3872-belle],_is the strangest heavy quark state. Even_now, its angular momentum is not_determined. Since its extreme closeness_to the_$D^0\bar{D}^{*0}$ threshold, lots of discussions_about its properties_are based_on the molecule_assumption. However, it is very difficult_to identify the_X(3872) as a shallow bound state_of_$D^0\bar{D}^{*0}$ since there_are_no_explicitly exotic_molecule properties.
A charged_charmonium-_or bottomonium-like_meson_labeled as $Z$ is absolutely exotic_because_its number of quarks and antiquarks must_be four or more._Such_states include the $Z(4430)$_observed in the $\psi'\pi^\pm$ mass_distribution [@Z4430-belle], the $Z_1(4050)$ and $Z_2(4250)$_observed in_the $\chi_{c1}\pi^+$_mass distribution [@Z1Z2-belle], and the $Z_b(10610)$ and $Z_b(10650)$ in the mass_spectra of the $\Upsilon(nS)\pi^\pm$ ($n$=1,2,3) and_$\pi^\pm h_b(mP)$ ($m$=1,2) [@Zb-belle]._They are_all_observed by Belle_Collaboration._Though BABAR_has not confirmed them [@Z4430-babar; @Z1Z2-babar], the_existence signal_of multiquark states is still |
} -1}) G_0^{\rm R} \partial_{X^0} G_0^{{\rm R} -1}]
+ {{\rm c.c.}}\label{eq:mp3b} \\
& \mp \frac{i \hbar^2}{4} \delta_{\hat k}^{\phantom{\hat k} {\hat \imath}} \int \frac{d^D \pi}{(2 \pi \hbar)^D} \int d X^0
\operatorname{tr}[{\hat \Sigma}_1 {\hat G}_0]^< + {{\rm c.c.}}\label{eq:mp3c} \\
& \pm \frac{i \hbar^2}{4} \int \frac{d^D \pi}{(2 \pi \hbar)^D} \int d X^0 f^{\prime}(-\pi_{\hat 0}) \pi_{\hat k}
\operatorname{tr}[\Sigma_1^{< (1)} (G_0^{\rm R} - G_0^{\rm A}) \partial_{\pi_{\hat \imath}} G_0^{{\rm R} -1} G_0^{\rm R}] + {{\rm c.c.}}, \label{eq:mp3d}\end{aligned}$$
\[eq:mp3\]
where $\epsilon_{ABC}$ in Eq. is the antisymmetric tensor with $\epsilon_{\pi_{\hat 0} \pi_{\hat \imath} X^0} = 1$. $f(\omega) = (e^{\beta \omega} \mp 1)^{-1}$ is the distribution function for boson or fermion.
Below we focus on the clean and noninteracting limit ${{\hat {\bm \Sigma}}} = 0$ to derive the Berry-phase formula in the Bloch basis. The retarded Green’s function is given by $G_0^{\rm R} = [(-\pi_{\hat 0}) - {\cal H}(X^0) + \mu + i \eta]^{-1}$ with $\eta \to +0$, leading to $\partial_{\pi_{\hat 0}} G_0^{{\rm R} -1} = -1$, $\partial_{\pi_{\hat \imath}} G_0^{{\rm R} -1} = -v^{\hat \imath}$, and $\partial_{X^0} G_0^{{\rm R} - | } -1 }) G_0^{\rm R } \partial_{X^0 } G_0^{{\rm R } -1 } ]
+ { { \rm c.c.}}\label{eq: mp3b } \\
& \mp \frac{i \hbar^2}{4 } \delta_{\hat k}^{\phantom{\hat k } { \hat \imath } } \int \frac{d^D \pi}{(2 \pi \hbar)^D } \int d X^0
\operatorname{tr}[{\hat \Sigma}_1 { \hat G}_0]^ < + { { \rm c.c.}}\label{eq: mp3c } \\
& \pm \frac{i \hbar^2}{4 } \int \frac{d^D \pi}{(2 \pi \hbar)^D } \int d X^0 f^{\prime}(-\pi_{\hat 0 }) \pi_{\hat k }
\operatorname{tr}[\Sigma_1^ { < (1) } (G_0^{\rm radius } - G_0^{\rm A }) \partial_{\pi_{\hat \imath } } G_0^{{\rm R } -1 } G_0^{\rm radius } ] + { { \rm c.c. } }, \label{eq: mp3d}\end{aligned}$$
\[eq: mp3\ ]
where $ \epsilon_{ABC}$ in Eq. is the antisymmetric tensor with $ \epsilon_{\pi_{\hat 0 } \pi_{\hat \imath } X^0 } = 1$. $ f(\omega) = (e^{\beta \omega } \mp 1)^{-1}$ is the distribution function for boson or fermion.
Below we focus on the clean and noninteracting terminus ad quem $ { { \hat { \bm \Sigma } } } = 0 $ to derive the Berry - phase formula in the Bloch basis. The retarded Green ’s affair is given by $ G_0^{\rm R } = [ (-\pi_{\hat 0 }) - { \cal H}(X^0) + \mu + i \eta]^{-1}$ with $ \eta \to +0 $, contribute to $ \partial_{\pi_{\hat 0 } } G_0^{{\rm R } -1 } = -1 $, $ \partial_{\pi_{\hat \imath } } G_0^{{\rm R } -1 } = -v^{\hat \imath}$, and $ \partial_{X^0 } G_0^{{\rm R } - | } -1}) G_0^{\gm R} \partial_{X^0} G_0^{{\rm R} -1}]
+ {{\ro c.c.}}\label{eq:mp3b} \\
& \mp \fcac{i \hbzr^2}{4} \delta_{\fat k}^{\phantom{\hat k} {\hat \imath}} \mnt \drac{d^E \pi}{(2 \pi \hbar)^D} \int d X^0
\operatorjame{tr}[{\har \Sijma}_1 {\hat G}_0]^< + {{\rm c.c.}}\label{eq:mi3e} \\
& \ll \frcc{m \hbar^2}{4} \int \frac{c^D \pi}{(2 \pi \htar)^D} \int d X^0 f^{\[rkmz}(-\pi_{\hat 0}) \pi_{\hat k}
\operatorname{tr}[\Sigma_1^{< (1)} (G_0^{\rm R} - H_0^{\rm A}) \partial_{\py_{\hat \ymatg}} G_0^{{\rm R} -1} G_0^{\rm R}] + {{\rm c.c.}}, \label{eq:mp3d}\ens{alignev}$$
\[eq:mp3\]
where $\epsikon_{ABC}$ in Eq. is the antisylmetgic tensor with $\epdilon_{\pi_{\hat 0} \pi_{\hwr \imath} X^0} = 1$. $w(\omega) = (e^{\bttc \omega} \mp 1)^{-1}$ is the distribution function fur boxon or ferniin.
Bfnow we focuw on nhe clean and nonintesacting limit ${{\hat {\bm \Xigka}}} = 0$ to derive the Berry-'hase formula in the Bloch bavia. The retarded Grweb’s futctimn ir gixen bb $G_0^{\dm R} = [(-\oi_{\het 0}) - {\cal H}(X^0) + \mu + i \eta]^{-1}$ with $\eta \to +0$, leadimg np $\partial_{\pi_{\hzt 0}} G_0^{{\ri W} -1} = -1$, $\partial_{\pi_{\hat \imath}} G_0^{{\rm R} -1} = -v^{\hat \imdth}$, and $\partial_{X^0} G_0^{{\rm R} - | } -1}) G_0^{\rm R} \partial_{X^0} G_0^{{\rm R} {{\rm \\ & \frac{i \hbar^2}{4} \delta_{\hat \frac{d^D \pi \hbar)^D} \int X^0 \operatorname{tr}[{\hat \Sigma}_1 G}_0]^< + {{\rm c.c.}}\label{eq:mp3c} \\ & \frac{i \hbar^2}{4} \int \frac{d^D \pi}{(2 \pi \hbar)^D} \int d X^0 f^{\prime}(-\pi_{\hat 0}) \pi_{\hat \operatorname{tr}[\Sigma_1^{< (1)} (G_0^{\rm R} - G_0^{\rm A}) \partial_{\pi_{\hat \imath}} G_0^{{\rm R} -1} G_0^{\rm + c.c.}}, \[eq:mp3\] $\epsilon_{ABC}$ in Eq. is the antisymmetric tensor with $\epsilon_{\pi_{\hat 0} \pi_{\hat \imath} X^0} = 1$. $f(\omega) (e^{\beta \omega} \mp 1)^{-1}$ is the distribution function boson or fermion. Below focus on the clean and limit {\bm \Sigma}}} 0$ derive Berry-phase formula in Bloch basis. The retarded Green’s function is given by $G_0^{\rm R} = [(-\pi_{\hat 0}) - {\cal H}(X^0) \mu + with $\eta +0$, to 0}} G_0^{{\rm R} -1$, $\partial_{\pi_{\hat \imath}} G_0^{{\rm R} -1} and $\partial_{X^0} G_0^{{\rm R} - | } -1}) G_0^{\rm R} \partial_{X^0} G_0^{{\rm R} -1}]
+ {{\rm c.c.}}\label{Eq:mp3b} \\
& \mp \fraC{i \hbaR^2}{4} \deLta_{\HaT k}^{\phAntoM{\hat k} {\hat \imath}} \iNT \fraC{d^D \pi}{(2 \pi \hbar)^D} \int d X^0
\operaTornaMe{TR}[{\hat \sIgMa}_1 {\hat g}_0]^< + {{\rm c.c.}}\laBEl{EQ:Mp3c} \\
& \Pm \FrAc{i \HbAR^2}{4} \iNt \fraC{d^D \Pi}{(2 \pi \hbaR)^D} \int d X^0 f^{\prIme}(-\Pi_{\Hat 0}) \pi_{\hat k}
\opeRAtOrname{tr}[\SiGma_1^{< (1)} (g_0^{\rm R} - G_0^{\rm A}) \partIal_{\Pi_{\hat \iMaTh}} G_0^{{\RM R} -1} G_0^{\rm r}] + {{\rm C.c.}}, \labEl{eq:mp3D}\End{aliGned}$$
\[eq:mp3\]
wHeRE $\epsilON_{ABC}$ in EQ. IS tHe anTisymmetric tensor WItH $\Epsilon_{\pi_{\hat 0} \pi_{\Hat \imaTh} x^0} = 1$. $F(\oMEGa) = (e^{\BetA \omega} \mp 1)^{-1}$ is ThE distRIbution FUnCTIOn fOR boson or fermiOn.
Below we foCUs oN the clEaN anD NonintEractInG LimIt ${{\hat {\bm \SigmA}}} = 0$ to dErive the BErry-phASe formuLA in the BLoch baSis. the RetaRDeD GReeN’s FUncTIoN is GIveN by $G_0^{\rm R} = [(-\pI_{\hAt 0}) - {\Cal H}(X^0) + \Mu + i \eTA]^{-1}$ WITh $\etA \to +0$, LeadIng to $\Partial_{\pi_{\hat 0}} G_0^{{\Rm R} -1} = -1$, $\PartIAl_{\pI_{\hat \iMath}} G_0^{{\Rm R} -1} = -v^{\HaT \imatH}$, and $\paRtial_{x^0} G_0^{{\Rm R} - | } -1}) G_0^{\rm R} \partia l_{X^0} G_ 0^{{\ rmR}-1 }]
+ { {\rm c.c.}}\la b el{e q:mp3b} \\
& \mp \fr ac{i\h b ar^2 } {4 } \de lta_{\h a tk } ^{\ ph an tom {\ h at k} { \ha t \imat h}} \int \ fra c{ d^D \pi}{(2\ pi \hbar)^D} \i nt d X^0
\ ope ratorn am e{t r }[{\h at\Sigm a}_1 { \ hat G} _0]^< + { {\ r m c.c. } }\label { e q: mp3c } \\
& \pm \fra c {i \hbar^2}{4} \i nt \fr ac { d^ D \pi }{( 2 \pi \hba r) ^D} \ i nt d X^ 0 f ^ { \ pri m e}(-\pi_{\hat 0}) \pi_{\ h atk}
\ op era t orname {tr}[ \S i gma _1^{< (1)}(G_0 ^{\rm R}- G_0^ { \rm A}) \partia l_{\pi _{\ hat \im a th }} G_ 0^ { {\r m R } - 1 } G _0^{\rmR} ]+ {{\ rm c . c . } }, \ lab el{e q:mp3 d}\end{aligne d}$ $
\ [ eq: mp3\]
whe re $ \e psilo n_{ABC }$ in E q. is the anti symm etric ten sor w ith $ \epsi l on_{\p i_{ \ha t 0} \p i_{\hat \im at h } X^ 0} = 1$. $f(\omega )= (e ^{\beta\omega } \ mp 1)^{-1}$ i s t he d i s tribu tion fu nction f or bos o nor fermio n.
Belo wwefoc us on thecleanand noni ntera c ting limit ${{ \ hat {\bm \Sig m a} } } = 0$ t o d erive the B erry - phas e fo r mu lai n the Bloc hb as i s. The retarded Gre en ’s fun ction is given by$G_0^{\rmR } = [(-\pi _{\h a t0 }) - {\cal H}( X^0)+ \mu + i\ eta]^{-1 }$ wi th $\eta \to +0$, l eading t o $ \pa rti al_ { \ pi _{\hat 0}} G_ 0 ^ {{\r mR} -1}= - 1$, $\p art ial _{\ pi_ {\ hat \imat h}} G_0^ {{ \r mR} -1 } = - v ^{\hat \ im ath }$ , a nd $\ p artial _{X^0 } G_ 0^ {{ \ rmR} - | } -1})_G_0^{\rm R}_\partial_{X^0} G_0^{{\rm R} -1}]
_ +_{{\rm_c.c.}}\label{eq:mp3b} \\
__& \mp \frac{i_\hbar^2}{4} \delta_{\hat k}^{\phantom{\hat_k} {\hat \imath}} \int_\frac{d^D \pi}{(2 \pi_\hbar)^D}_\int d X^0
\operatorname{tr}[{\hat \Sigma}_1 {\hat G}_0]^< + {{\rm c.c.}}\label{eq:mp3c} \\
&_\pm_\frac{i \hbar^2}{4}_\int_\frac{d^D_\pi}{(2 \pi \hbar)^D} \int d_X^0 f^{\prime}(-\pi_{\hat 0}) \pi_{\hat k}
_ \operatorname{tr}[\Sigma_1^{<_(1)} (G_0^{\rm R} - G_0^{\rm A}) \partial_{\pi_{\hat \imath}}_G_0^{{\rm_R} -1} G_0^{\rm_R}] + {{\rm c.c.}}, \label{eq:mp3d}\end{aligned}$$
\[eq:mp3\]
where $\epsilon_{ABC}$ in Eq. is_the antisymmetric tensor with $\epsilon_{\pi_{\hat 0}_\pi_{\hat \imath} X^0}_=_1$._$f(\omega) = (e^{\beta \omega}_\mp 1)^{-1}$ is the distribution function_for boson or fermion.
Below we focus_on the clean and noninteracting limit ${{\hat_{\bm \Sigma}}} = 0$ to derive_the Berry-phase formula in the_Bloch basis._The retarded Green’s function is_given by $G_0^{\rm_R} =_[(-\pi_{\hat 0}) -_{\cal H}(X^0) + \mu + i_\eta]^{-1}$ with $\eta_\to +0$, leading to $\partial_{\pi_{\hat 0}}_G_0^{{\rm_R} -1} =_-1$,_$\partial_{\pi_{\hat_\imath}} G_0^{{\rm_R} -1} =_-v^{\hat_\imath}$, and_$\partial_{X^0}_G_0^{{\rm R} - |
Finally, we show that $\tilde {\varphi}\wedge{\mathrm{d}}\tilde {\varphi}={\varphi}\wedge{\mathrm{d}}{\varphi}$. Since $\{\tilde{e}^5,\tilde{e}^6,\tilde{e}^7\}$ spans ${\mathfrak{n}}'$, gives $$\label{sd1}
\tilde {\varphi}\wedge{\mathrm{d}}\tilde {\varphi}=2(\sigma_1\wedge\alpha_1+\sigma_2\wedge\alpha_2+\sigma_3\wedge\alpha_3)\wedge \tilde e^{567}.$$
The argument above shows that ${\mathrm{d}}(e^2\wedge\alpha_i)=-e^2\wedge{\mathrm{d}}\alpha_i=0$, for $i=1,2,3$. Using, we obtain ${\mathrm{d}}e^4={\mathrm{d}}(\mu\tilde e^4+{\lambda}\tilde e^5)={\lambda}\alpha_1$ and ${\mathrm{d}}e^5={\mathrm{d}}(-{\lambda}\tilde e^4+\mu \tilde e^5)=\mu\alpha_1$, so we can write $$\begin{aligned}
{\varphi}\wedge{\mathrm{d}}{\varphi}&=&{\varphi}\wedge{\mathrm{d}}( e^{127}+e^{347}+e^{567} + e^{135}-e^{146}-e^{236}-e^{245})\\
&=&(e^{347}+e^{567} + e^{135}-e^{146})\wedge{\mathrm{d}}(e^{347}+e^{567} + e^{135}-e^{146})\\
&=&(e^{347}+e^{567} + e^{135}-e^{146})\wedge(-{\lambda}e^{37}\wedge\alpha_1+e^{34}\wedge\alpha_3+\mu e^{67}\wedge\alpha_1\\
&&-e^{57}\wedge\alpha_2+e^{56}\wedge\alpha_3
+ \mu e^{13}\wedge\alpha_1+{\lambda}e^{16}\wedge\alpha_1-e^{14}\wedge\alpha_2)\\
&=&2({\lambda}e^{13467}\wedge\alpha_1+e^{34567}\wedge\alpha_3+\mu e^{13567}\wedge\alpha_1-e^{14567}\wedge | Finally, we show that $ \tilde { \varphi}\wedge{\mathrm{d}}\tilde { \varphi}={\varphi}\wedge{\mathrm{d}}{\varphi}$. Since $ \{\tilde{e}^5,\tilde{e}^6,\tilde{e}^7\}$ spans $ { \mathfrak{n}}'$, gives $ $ \label{sd1 }
\tilde { \varphi}\wedge{\mathrm{d}}\tilde { \varphi}=2(\sigma_1\wedge\alpha_1+\sigma_2\wedge\alpha_2+\sigma_3\wedge\alpha_3)\wedge \tilde e^{567}.$$
The controversy above usher that $ { \mathrm{d}}(e^2\wedge\alpha_i)=-e^2\wedge{\mathrm{d}}\alpha_i=0 $, for $ i=1,2,3$. Using, we obtain $ { \mathrm{d}}e^4={\mathrm{d}}(\mu\tilde e^4+{\lambda}\tilde e^5)={\lambda}\alpha_1 $ and $ { \mathrm{d}}e^5={\mathrm{d}}(-{\lambda}\tilde e^4+\mu \tilde e^5)=\mu\alpha_1 $, so we can write $ $ \begin{aligned }
{ \varphi}\wedge{\mathrm{d}}{\varphi}&=&{\varphi}\wedge{\mathrm{d } } (e^{127}+e^{347}+e^{567 } + e^{135}-e^{146}-e^{236}-e^{245})\\
& = & (e^{347}+e^{567 } + e^{135}-e^{146})\wedge{\mathrm{d}}(e^{347}+e^{567 } + e^{135}-e^{146})\\
& = & (e^{347}+e^{567 } + e^{135}-e^{146})\wedge(-{\lambda}e^{37}\wedge\alpha_1+e^{34}\wedge\alpha_3+\mu e^{67}\wedge\alpha_1\\
& & -e^{57}\wedge\alpha_2+e^{56}\wedge\alpha_3
+ \mu e^{13}\wedge\alpha_1+{\lambda}e^{16}\wedge\alpha_1 - e^{14}\wedge\alpha_2)\\
& = & 2({\lambda}e^{13467}\wedge\alpha_1+e^{34567}\wedge\alpha_3+\mu e^{13567}\wedge\alpha_1 - e^{14567}\wedge |
Finwlly, we show that $\tilde {\yarphi}\wedge{\mathrm{d}}\tilde {\varphj}={\varphi}\wddge{\mathrm{d}}{\varphi}$. Since $\{\tildx{e}^5,\tiode{e}^6,\tulde{e}^7\}$ spans ${\mathfrak{n}}'$, eives $$\labvl{sd1}
\tilde {\varkhi}\wedge{\mathrm{d}}\tilde {\varpmn}=2(\sigmz_1\aedgz\aopha_1+\sigma_2\wedge\slpha_2+\sigma_3\fedge\alpha_3)\wedga \gipde e^{567}.$$
The argument above shows that ${\iathrm{d}}(r^2\wfdge\alpha_i)=-e^2\wedde{\mauhri{d}}\allha_i=0$, for $i=1,2,3$. Using, we obtain ${\mathrm{d}}s^4={\mathrm{v}}(\mu\tilde e^4+{\lambds}\tilde e^5)={\lambda}\alpha_1$ and ${\mahhrm{f}}e^5={\mathrm{d}}(-{\lambda}\tilfe e^4+\mu \tildg e^5)=\mt\qlpha_1$, so we zan write $$\begin{aligned}
{\barphi}\wedge{\mathrm{d}}{\varphi}&=&{\varphi}\wddge{\mcthrm{d}}( e^{127}+e^{347}+e^{567} + e^{135}-w^{146}-e^{236}-e^{245})\\
&=&(x^{347}+e^{567} + e^{135}-e^{146})\redge{\mathrm{d}}(c^{347}+v^{567} + e^{135}-e^{146})\\
&=&(e^{347}+e^{567} + e^{135}-e^{146})\wedge(-{\kamuda}e^{37}\qedge\alpha_1+e^{34}\wedge\alpha_3+\ku e^{67}\wedge\alpha_1\\
&&-e^{57}\febge\alpha_2+e^{56}\wedge\alpha_3
+ \oy e^{13}\dedfe\elpga_1+{\lambfa}e^{16}\xedge\alpha_1-e^{14}\sedge\alpha_2)\\
&=&2({\lambda}e^{13467}\wtdgq\qlpha_1+e^{34567}\wedge\allha_3+\mu q^{13567}\wqdge\alpha_1-e^{14567}\wedge | Finally, we show that $\tilde {\varphi}\wedge{\mathrm{d}}\tilde {\varphi}={\varphi}\wedge{\mathrm{d}}{\varphi}$. spans gives $$\label{sd1} {\varphi}\wedge{\mathrm{d}}\tilde {\varphi}=2(\sigma_1\wedge\alpha_1+\sigma_2\wedge\alpha_2+\sigma_3\wedge\alpha_3)\wedge \tilde that for $i=1,2,3$. Using, obtain ${\mathrm{d}}e^4={\mathrm{d}}(\mu\tilde e^4+{\lambda}\tilde and ${\mathrm{d}}e^5={\mathrm{d}}(-{\lambda}\tilde e^4+\mu \tilde e^5)=\mu\alpha_1$, so can write $$\begin{aligned} {\varphi}\wedge{\mathrm{d}}{\varphi}&=&{\varphi}\wedge{\mathrm{d}}( e^{127}+e^{347}+e^{567} + e^{135}-e^{146}-e^{236}-e^{245})\\ &=&(e^{347}+e^{567} + e^{135}-e^{146})\wedge{\mathrm{d}}(e^{347}+e^{567} + e^{135}-e^{146})\\ &=&(e^{347}+e^{567} e^{135}-e^{146})\wedge(-{\lambda}e^{37}\wedge\alpha_1+e^{34}\wedge\alpha_3+\mu e^{67}\wedge\alpha_1\\ &&-e^{57}\wedge\alpha_2+e^{56}\wedge\alpha_3 + \mu e^{13}\wedge\alpha_1+{\lambda}e^{16}\wedge\alpha_1-e^{14}\wedge\alpha_2)\\ &=&2({\lambda}e^{13467}\wedge\alpha_1+e^{34567}\wedge\alpha_3+\mu e^{13567}\wedge\alpha_1-e^{14567}\wedge |
Finally, we show that $\tilde {\varPhi}\wedge{\maThrm{d}}\TilDe {\vArPhi}={\vArphI}\wedge{\mathrm{d}}{\vARphi}$. since $\{\tilde{e}^5,\tilde{e}^6,\tilde{E}^7\}$ spanS ${\mAThfrAK{n}}'$, Gives $$\Label{sd1}
\TIlDE {\VarPhI}\wEdgE{\mAThRm{d}}\tiLde {\Varphi}=2(\sIgma_1\wedge\aLphA_1+\sIgma_2\wedge\alpHA_2+\sIgma_3\wedge\aLphA_3)\wedge \tilde e^{567}.$$
the ArgumeNt AboVE showS thAt ${\matHrm{d}}(e^2\wEDge\alpHa_i)=-e^2\wedge{\MaTHrm{d}}\alPHa_i=0$, for $i=1,2,3$. uSInG, we oBtain ${\mathrm{d}}e^4={\mathRM{d}}(\MU\tilde e^4+{\lambda}\tIlde e^5)={\lAmBDa}\ALPha_1$ And ${\Mathrm{d}}e^5={\maThRm{d}}(-{\laMBda}\tildE E^4+\mU \TILde E^5)=\Mu\alpha_1$, so we caN write $$\begin{ALigNed}
{\varPhI}\weDGe{\mathRm{d}}{\vaRpHI}&=&{\vaRphi}\wedge{\maThrm{D}}( e^{127}+e^{347}+e^{567} + e^{135}-e^{146}-e^{236}-e^{245})\\
&=&(e^{347}+E^{567} + e^{135}-e^{146})\wedGE{\mathrm{D}}(E^{347}+e^{567} + e^{135}-e^{146})\\
&=&(e^{347}+e^{567} + e^{135}-E^{146})\wedge(-{\LamBda}E^{37}\wedGE\aLpHa_1+e^{34}\WeDGe\aLPhA_3+\mu E^{67}\WedGe\alpha_1\\
&&-e^{57}\WeDgE\alphA_2+e^{56}\weDGE\ALpha_3
+ \Mu e^{13}\WedgE\alphA_1+{\lambda}e^{16}\wedge\AlpHa_1-e^{14}\wEDge\Alpha_2)\\
&=&2({\LambdA}e^{13467}\weDgE\alphA_1+e^{34567}\wedgE\alphA_3+\mU e^{13567}\wedge\alpha_1-e^{14567}\weDge |
Finally, we show that $\ tilde {\va rphi} \we dge {\ math rm{d }}\tilde {\var p hi}= {\varphi}\wedge{\mathr m{d}} {\ v arph i }$ . Sin ce $\{\ t il d e {e} ^5 ,\ til de { e} ^6,\t ild e{e}^7\ }$ spans $ {\m at hfrak{n}}'$, gi ves $$\lab el{ sd1}
\tilde{\v arphi} \w edg e {\mat hrm {d}}\ tilde{ \varph i}=2(\sig ma _ 1\wedg e \alpha_ 1 + \s igma _2\wedge\alpha_2+ \ si g ma_3\wedge\alp ha_3)\ we d ge \ til dee^{567}.$$
The a r gumenta bo v e sho w s that ${\mat hrm{d}}(e^2 \ wed ge\alp ha _i) = -e^2\w edge{ \m a thr m{d}}\alpha _i=0 $, for $i =1,2,3 $ . Using , we obt ain ${ \ma thr m{d} } e^ 4= {\m at h rm{ d }} (\m u \ti lde e^4+ {\ la mbda} \til d e e ^5)= {\l ambd a}\al pha_1$ and ${ \ma thrm { d}} e^5={ \math rm{d }} (-{\l ambda} \tild ee^4+\mu \tildee^5) =\mu\alph a_1 $, so w e can write$$\ beg in{alig ned}
{\ v arp hi } \ w ed ge{\mathrm{d}}{\va rp h i }& =&{\varp hi}\we d ge {\ m athrm{d} }( e^ {127 } + e^{34 7}+e ^ {5 67} + e^ {135}- e ^{ 14 6}-e^{2 36 }-e^{2 45 })\ \
&=&(e ^{347}+e ^{567 } + e^{135}-e^{ 1 46})\wedge{\m a th r m {d } }(e^ {34 7}+e^{567}+ e^ { 135} -e^{ 1 46 })\ \
&= & (e^{347}+e^{567} +e^ {135}- e^{14 6})\wedge(-{\ lambda}e^{ 3 7 } \wedge\a lpha _ 1+ e ^{34}\wedge\al pha_3 +\mu e^{67 } \wedge\a lpha_ 1\\
& & - e^{57}\w edg e\a lph a_2 + e ^{ 56}\wedge\alp h a _3
+ \m u e ^{1 3}\ wed ge \alpha_1+ {\lambda }e ^{ 16 }\ wed ge\al p ha_1-e^{ 14 }\w ed ge\ alpha _ 2)\\
&= & 2({ \lambda } e^ { 1 3467 }\ we dge\ alp ha _1+e^ {345 6 7}\ wedge\a lpha_3+\m u e ^ {135 67 }\ wedge\a lpha_1-e^{145 67 }\wedge |
Finally, we_show that_$\tilde {\varphi}\wedge{\mathrm{d}}\tilde {\varphi}={\varphi}\wedge{\mathrm{d}}{\varphi}$. Since_$\{\tilde{e}^5,\tilde{e}^6,\tilde{e}^7\}$ spans_${\mathfrak{n}}'$,_gives $$\label{sd1}
\tilde_{\varphi}\wedge{\mathrm{d}}\tilde_{\varphi}=2(\sigma_1\wedge\alpha_1+\sigma_2\wedge\alpha_2+\sigma_3\wedge\alpha_3)\wedge \tilde e^{567}.$$
The_argument above shows_that ${\mathrm{d}}(e^2\wedge\alpha_i)=-e^2\wedge{\mathrm{d}}\alpha_i=0$, for $i=1,2,3$._Using, we obtain_${\mathrm{d}}e^4={\mathrm{d}}(\mu\tilde_e^4+{\lambda}\tilde e^5)={\lambda}\alpha_1$ and ${\mathrm{d}}e^5={\mathrm{d}}(-{\lambda}\tilde e^4+\mu \tilde e^5)=\mu\alpha_1$, so we can write $$\begin{aligned}
{\varphi}\wedge{\mathrm{d}}{\varphi}&=&{\varphi}\wedge{\mathrm{d}}( e^{127}+e^{347}+e^{567} +_e^{135}-e^{146}-e^{236}-e^{245})\\
_ ___ _ &=&(e^{347}+e^{567} + e^{135}-e^{146})\wedge{\mathrm{d}}(e^{347}+e^{567}_+ e^{135}-e^{146})\\
_ __ &=&(e^{347}+e^{567}_+ e^{135}-e^{146})\wedge(-{\lambda}e^{37}\wedge\alpha_1+e^{34}\wedge\alpha_3+\mu e^{67}\wedge\alpha_1\\
_ &&-e^{57}\wedge\alpha_2+e^{56}\wedge\alpha_3
_ ___ _ _+ \mu e^{13}\wedge\alpha_1+{\lambda}e^{16}\wedge\alpha_1-e^{14}\wedge\alpha_2)\\
_ _ &=&2({\lambda}e^{13467}\wedge\alpha_1+e^{34567}\wedge\alpha_3+\mu e^{13567}\wedge\alpha_1-e^{14567}\wedge |
=mygreen\]right:$v_{\i}$]{}\]; ]{}
; ; ; ;
(n14) to (n22);
(n13) to (n20);
(n10) to (n21);
(n11) to (n21);
(n12) to (n21);
\[One figure\][Round 3.]{}
=\[circle,fill=black!25,minimum size=12pt,inner sep=2pt\] (G\_1) at (0,0) ; (G\_2) at (-1,-1) ; (G\_3) at (1,-1) ; (G\_4) at (-3,-2) ; (G\_5) at (3,-2) ; (G\_6) at (-1,-2) ; (G\_7) at (1,-2) ;
(G\_8) at (-3.5,-3) ; (G\_9) at (-2.5,-3) ; (G\_10) at (-1.5,-3) ; (G\_11) at (-0.5,-3) ; (G\_12) at (0.5,-3) ; (G\_13) at (1.5,-3) ; (G\_14) at (2.5,-3) ; (G\_15) at (3.5,-3) ;
(G\_1) – (G\_2); (G\_1) – (G\_3); (G\_2) – (G\_4); (G\_2) – (G\_6); (G\_3) – (G\_5); (G\_3) – (G\_7);
(G\_4) – (G\_8); (G\_4) – (G\_9); (G\_6) – (G\_10); (G\_6) – (G\_11); (G\_7) – (G\_12); (G\_7) – (G\_13); (G\_5) – (G\_14); (G\_5) – (G\_15);
(G\_1) to\[out=270,in=0\] (G\_2);
(G\_3) to\[out=0,in=0\] (G\_1); (G\_2) to | = mygreen\]right:$v_{\i}$]{}\ ]; ] { }
; ;; ;
(n14) to (n22);
(n13) to (n20);
(n10) to (n21);
(n11) to (n21);
(n12) to (n21);
\[One figure\][Round 3. ] { }
= \[circle, fill = black!25,minimum size=12pt, inner sep=2pt\ ] (G\_1) at (0,0); (G\_2) at (-1,-1); (G\_3) at (1,-1); (G\_4) at (-3,-2); (G\_5) at (3,-2); (G\_6) at (-1,-2); (G\_7) at (1,-2);
(G\_8) at (-3.5,-3); (G\_9) at (-2.5,-3); (G\_10) at (-1.5,-3); (G\_11) at (-0.5,-3); (G\_12) at (0.5,-3); (G\_13) at (1.5,-3); (G\_14) at (2.5,-3); (G\_15) at (3.5,-3);
(G\_1) – (G\_2); (G\_1) – (G\_3); (G\_2) – (G\_4); (G\_2) – (G\_6); (G\_3) – (G\_5); (G\_3) – (G\_7);
(G\_4) – (G\_8); (G\_4) – (G\_9); (G\_6) – (G\_10); (G\_6) – (G\_11); (G\_7) – (G\_12); (G\_7) – (G\_13); (G\_5) – (G\_14); (G\_5) – (G\_15);
(G\_1) to\[out=270,in=0\ ] (G\_2);
(G\_3) to\[out=0,in=0\ ] (G\_1); (G\_2) to | =myggeen\]right:$v_{\i}$]{}\]; ]{}
; ; ; ;
(n14) to (n22);
(n13) tu (n20);
(n10) to (n21);
(n11) to (n21);
(u12) to (n21);
\[Oie figude\][Round 3.]{}
=\[zircle,fill=black!25,minimum size=12pv,innwr sek=2it\] (G\_1) at (0,0) ; (G\_2) at (-1,-1) ; (G\_3) at (1,-1) ; (G\_4) at (-3,-2) ; (G\_5) at (3,-2) ; (T\_6) at (-1,-2) ; (G\_7) at (1,-2) ;
(G\_8) ef (-3.5,-3) ; (G\_9) ab (-2.5,-3) ; (G\_10) wt (-1.5,-3) ; (J\_11) at (-0.5,-3) ; (G\_12) at (0.5,-3) ; (G\_13) at (1.5,-3) ; (G\_14) at (2.5,-3) ; (G\_15) at (3.5,-3) ;
(G\_1) – (G\_2); (C\_1) – (G\_3); (G\_2) – (G\_4); (G\_2) – (G\_6); (G\_3) – (G\_5); (G\_3) – (G\_7);
(G\_4) – (G\_8); (G\_4) – (G\_9); (G\_6) – (G\_10); (D\_6) – (G\_11); (G\_7) – (G\_12); (G\_7) – (G\_13); (G\_5) – (G\_14); (G\_5) – (D\_15);
(G\_1) tp\[jut=270,ih=0\] (G\_2);
(G\_3) to\[out=0,in=0\] (G\_1); (G\_2) to | =mygreen\]right:$v_{\i}$]{}\]; ]{} ; ; ; ; (n14) (n13) (n20); (n10) (n21); (n11) to figure\][Round =\[circle,fill=black!25,minimum size=12pt,inner sep=2pt\] at (0,0) ; at (-1,-1) ; (G\_3) at (1,-1) (G\_4) at (-3,-2) ; (G\_5) at (3,-2) ; (G\_6) at (-1,-2) ; (G\_7) (1,-2) ; (G\_8) at (-3.5,-3) ; (G\_9) at (-2.5,-3) ; (G\_10) at (-1.5,-3) (G\_11) (-0.5,-3) (G\_12) (0.5,-3) ; (G\_13) at (1.5,-3) ; (G\_14) at (2.5,-3) ; (G\_15) at (3.5,-3) ; (G\_1) – (G\_1) – (G\_3); (G\_2) – (G\_4); (G\_2) – (G\_3) – (G\_5); (G\_3) (G\_7); (G\_4) – (G\_8); (G\_4) (G\_9); – (G\_10); – (G\_7) (G\_12); (G\_7) – (G\_5) – (G\_14); (G\_5) – (G\_15); (G\_1) to\[out=270,in=0\] (G\_2); (G\_3) to\[out=0,in=0\] (G\_1); (G\_2) to | =mygreen\]right:$v_{\i}$]{}\]; ]{}
; ; ; ;
(n14) to (n22);
(n13) to (n20);
(n10) to (N21);
(n11) to (n21);
(n12) to (n21);
\[ONe figUre\][rouNd 3.]{}
=\[CircLe,fiLl=black!25,minimum SIze=12pT,inner sep=2pt\] (G\_1) at (0,0) ; (G\_2) at (-1,-1) ; (G\_3) at (1,-1) ; (G\_4) aT (-3,-2) ; (G\_5) at (3,-2) ; (G\_6) At (-1,-2) ; (g\_7) At (1,-2) ;
(G\_8) aT (-3.5,-3) ; (g\_9) aT (-2.5,-3) ; (G\_10) at (-1.5,-3) ; (G\_11) At (-0.5,-3) ; (G\_12) at (0.5,-3) ; (G\_13) aT (1.5,-3) ; (g\_14) aT (2.5,-3) ; (g\_15) At (3.5,-3) ;
(G\_1) – (g\_2); (G\_1) – (g\_3); (G\_2) – (g\_4); (G\_2) – (G\_6); (g\_3) – (G\_5); (g\_3) – (g\_7);
(G\_4) – (g\_8); (G\_4) – (G\_9); (G\_6) – (G\_10); (g\_6) – (G\_11); (G\_7) – (g\_12); (G\_7) – (G\_13); (G\_5) – (G\_14); (G\_5) – (G\_15);
(g\_1) to\[out=270,in=0\] (G\_2);
(G\_3) To\[oUt=0,In=0\] (G\_1); (G\_2) to | =mygreen\]right:$v_{\i}$]{ }\]; ]{}
; ; ; ;
(n 14 ) to (n2 2);
(n13) to( n20) ;
(n10) to (n21);
(n 11) t o( n21) ;
(n12) to (n2 1 ); \[O ne f igu re \ ][ Round 3. ]{}
=\ [circle,fi ll= bl ack!25,minim u msize=12pt, inn er sep=2pt\] (G \_1) a t(0, 0 ) ; ( G\_ 2) at (-1,- 1 ) ; (G \_3) at ( 1, - 1) ; ( G \_4) at ( -3 ,-2) ; (G\_5) at (3,- 2 ); (G\_6) at (-1 ,-2) ; ( G \_ 7 ) at (1 ,-2) ;
(G \_ 8) at (-3.5,- 3 ); ( G\_ 9 ) at (-2.5,-3 ) ; (G\_10) at(-1.5, -3 ) ; (G\_11 ) at(- 0 .5, -3) ; (G\_1 2) a t (0.5,-3 ) ; (G \ _13) at (1.5,-3 ) ; (G \_1 4)at ( 2 .5 ,- 3);( G\_ 1 5) at (3. 5,-3) ;
( G\ _1) – (G\ _ 2 ) ; (G\ _1) – ( G\_3) ; (G\_2) – (G \_4 ); ( G \_2 ) – ( G\_6) ; (G \_ 3) –(G\_5) ; (G\ _3 ) – (G\_7);
(G \_4) – (G\_8) ; ( G\ _4) – (G\_ 9 ); (G\ _6) –(G\_10) ; (G\_6 ) –(G \ _ 1 1) ; (G\_7) – (G\_12) ;( G \_ 7) – (G\ _13);( G\ _5 ) – (G\_1 4) ; ( G\_5 ) – (G\ _15) ;
(G\_1) t o\[out = 27 0, in=0\](G \_2);
( G\_ 3)to\[o u t=0, in=0\] (G\_1); (G\_ 2 ) to | =mygreen\]right:$v_{\i}$]{}\]; ]{}
;_; ;_;
(n14) to (n22);
(n13) to_(n20);
(n10) to_(n21);
(n11)_to (n21);
(n12)_to_(n21);
\[One figure\][Round 3.]{}
=\[circle,fill=black!25,minimum_size=12pt,inner sep=2pt\] (G\_1)_at (0,0) ; (G\_2)_at (-1,-1) ;_(G\_3)_at (1,-1) ; (G\_4) at (-3,-2) ; (G\_5) at (3,-2) ; (G\_6) at (-1,-2)_;_(G\_7) at_(1,-2)_;
(G\_8)_at (-3.5,-3) ; (G\_9) at_(-2.5,-3) ; (G\_10) at (-1.5,-3)_; (G\_11)_at (-0.5,-3) ; (G\_12) at (0.5,-3) ; (G\_13)_at_(1.5,-3) ; (G\_14)_at (2.5,-3) ; (G\_15) at (3.5,-3) ;
(G\_1) – (G\_2);_(G\_1) – (G\_3); (G\_2) – (G\_4);_(G\_2) – (G\_6);_(G\_3)_–_(G\_5); (G\_3) – (G\_7);
(G\_4)_– (G\_8); (G\_4) – (G\_9); (G\_6)_– (G\_10); (G\_6) – (G\_11); (G\_7)_– (G\_12); (G\_7) – (G\_13); (G\_5) –_(G\_14); (G\_5) – (G\_15);
(G\_1) to\[out=270,in=0\] (G\_2);
(G\_3)_to\[out=0,in=0\] (G\_1); (G\_2) to |
$z=1$ and $z=0$ with high specific angular momentum (these stars form from high specific angular momentum gas which becomes cold at late times); and secondly, dynamical friction affects the system much less than in NF, since satellites are less massive. At $z=0$, disk stars have a specific angular momentum comparable to that of the dark matter, while spheroid stars have a much lower specific angular momentum.
![Dashed lines show the specific angular momentum as a function of time for the dark matter that, at $z=0$, lies within the virial radius of the system for NF (left panel) and E-0.7 (right panel). We also show with dots the specific angular momentum for the baryons which end up as cold gas or stars in the central $20\ h^{-1}$ kpc at $z=0$. The arrows show the specific angular momentum of disk and spheroid stars. []{data-label="j_evolution"}](fig5a.ps "fig:"){width="65mm"}![Dashed lines show the specific angular momentum as a function of time for the dark matter that, at $z=0$, lies within the virial radius of the system for NF (left panel) and E-0.7 (right panel). We also show with dots the specific angular momentum for the baryons which end up as cold gas or stars in the central $20\ h^{-1}$ kpc at $z=0$. The arrows show the specific angular momentum of disk and spheroid stars. []{data-label="j_evolution"}](fig5b.ps "fig:"){width="65mm"}
In Fig \[metal\_profiles\] we show the oxygen profiles for the no-feedback (NF) and feedback (E-0.7) runs. From this figure we can see that SN feedback strongly affects the chemical distributions. If no feedback is included, the gas is enriched only in the very central regions. Including SN feedback triggers a redistribution of mass and metals through galactic winds and fountains, giving the gas component a much higher level of enrichment out to large radii. A linear fit to this metallicity profile gives a slope of $-0.048$ dex kpc$^{-1}$ and a zero-point of $8.77$ dex, consistent with the observed values in real disk galaxies (e.g. Zaritsky et al. 1994).
![Oxygen abundance for the gas component as a function | $ z=1 $ and $ z=0 $ with high specific angular momentum (these stars form from high specific angular momentum accelerator which become cold at late times); and secondly, dynamic friction affects the system a lot less than in NF, since satellite are less massive. At $ z=0 $, disk stars have a specific angular momentum comparable to that of the dark topic, while spheroid stars have a much lower specific angular momentum.
! [ dash lines picture the specific angular momentum as a function of time for the black matter that, at $ z=0 $, lies within the virial radius of the organization for NF (left panel) and E-0.7 (right panel). We besides show with acid the specific angular momentum for the baryons which end up as cold gas or stars in the cardinal $ 20\ h^{-1}$ kpc at $ z=0$. The arrows show the specific angular momentum of disk and spheroid stars. [ ] { data - label="j_evolution"}](fig5a.ps " fig:"){width="65mm"}![Dashed lines usher the specific angular momentum as a function of time for the dark matter that, at $ z=0 $, lies within the virial radius of the system for NF (left panel) and E-0.7 (right jury). We also show with acid the specific angular momentum for the baryon which end up as cold natural gas or stars in the central $ 20\ h^{-1}$ kpc at $ z=0$. The arrow show the specific angular momentum of disk and spheroid stars. [ ] { data - label="j_evolution"}](fig5b.ps " fig:"){width="65 mm " }
In Fig \[metal\_profiles\ ] we usher the oxygen profiles for the no - feedback (NF) and feedback (E-0.7) runs. From this figure we can see that SN feedback strongly affects the chemical distribution. If no feedback is included, the gas is enriched only in the very cardinal regions. admit SN feedback trip a redistribution of mass and metals through galactic winds and fountain, giving the gas component a a lot higher grade of enrichment out to large radii. A analogue fit to this metallicity profile gives a slope of $ -0.048 $ dex kpc$^{-1}$ and a zero - point of $ 8.77 $ dex, consistent with the observe values in actual disk galaxy (e.g. Zaritsky et al. 1994).
! [ Oxygen abundance for the natural gas component as a function | $z=1$ wnd $z=0$ with high specific angular momentom (these stars form frum high specific angular momxntun gas which becomes cold at late timvs); and sexondot, dynamical friction affsgts tke system much lgss than in TF, since satelniged are less massive. At $z=0$, disk stars rave a xpfcific angular momtnttm ckmparable to that of the dark mattsr, whilt spheroid stars hsve a much lower specific wngupar momentum.
![Dashed lines show the wpecific angjlar momenuuk as a fundtion of time for the dark mattdr thct, at $z=0$, liew qitjhn the viriel radpus of the system for NF (lefy panel) and E-0.7 (vight pabel). We also show with dots the specific angular kojentum for the baeyins wvich end yp xs dokd gas og svars in the central $20\ h^{-1}$ kpc at $z=0$. The arrowx fyow the speciric andujar momentum of disk and spheroid stars. []{dafa-label="j_evolution"}](fig5a.ps "fig:"){width="65mm"}![Dashed lings show thq specific angular momentum as a function of time for vhd dcvh nahter that, at $z=0$, lies within the virial radius jr uhe system for NF (left panel) ajd G-0.7 (right panel). De also sgow with dots the dpecifis angylar momegtum for the baryons which end yp as cold gcs ir stars in the ceutral $20\ h^{-1}$ kpc at $e=0$. The srrows show the specifie angumar momentul of disk xnd spheroid stafs. []{cada-label="j_txolution"}](fig5b.ps "fid:"){width="65mm"}
Ii Fig \[ketal\_prufilgs\] we srow the oxjgen isofiles for the no-veedbcck (NX) and feedhack (E-0.7) runs. From this figure we can see that XN fevdback stxongly affects the shemical distrnbutions. If nu feedback is incnuded, the gws is enricheg only in the very cegtrao retions. Ivzluding SN feecback triyyers a reeistribution of maxs xhd metals throuyk talactic winds snd fotnnaiis, gidhng the gas wompunevy a mjch higher odvel of enrichment out tm ladge radii. A linear flt to thiw metallycity profile gives a slope of $-0.048$ dex npc$^{-1}$ and s zgro-point of $8.77$ dex, consistent witg the obsfrvcd values in weal disk galaxizs (e.g. Zaritsky et al. 1994).
![Oxygen abundance foc the gas component as q function | $z=1$ and $z=0$ with high specific angular stars from high angular momentum gas times); secondly, dynamical friction the system much than in NF, since satellites are massive. At $z=0$, disk stars have a specific angular momentum comparable to that the dark matter, while spheroid stars have a much lower specific angular momentum. lines the angular as a function of time for the dark matter that, at $z=0$, lies within the virial of the system for NF (left panel) and (right panel). We also with dots the specific angular for baryons which up cold or stars in central $20\ h^{-1}$ kpc at $z=0$. The arrows show the specific angular momentum of disk and spheroid []{data-label="j_evolution"}](fig5a.ps "fig:"){width="65mm"}![Dashed the specific momentum a of time for matter that, at $z=0$, lies within of the system for NF (left panel) and (right panel). also show with dots the specific momentum for the baryons which end up as gas or stars in the central $20\ h^{-1}$ kpc at $z=0$. The arrows show the momentum of disk and stars. []{data-label="j_evolution"}](fig5b.ps "fig:"){width="65mm"} Fig we the profiles for no-feedback (NF) and feedback (E-0.7) runs. From this figure we can that SN feedback strongly affects the chemical distributions. If no included, gas is enriched in the very central Including feedback triggers a redistribution and through fountains, the component a much higher of enrichment out to large A linear fit to slope of $-0.048$ dex kpc$^{-1}$ and a zero-point $8.77$ dex, consistent with the observed values real disk galaxies (e.g. Zaritsky et al. 1994). ![Oxygen abundance for the component as | $z=1$ and $z=0$ with high specific anguLar momentuM (thesE stArs FoRm frOm hiGh specific anguLAr moMentum gas which becomes cOld at LaTE timES); aNd secOndly, dyNAmICAl fRiCtIon AfFEcTs the SysTem much Less than in nF, sInCe satellites ARe Less massivE. At $Z=0$, disk stars haVe a SpecifIc AngULar moMenTum coMparabLE to thaT of the darK mATter, whILe spherOID sTars Have a much lower speCIfIC angular momentUm.
![DashEd LInES ShoW thE specific aNgUlar mOMentum aS A fUNCTioN Of time for the dArk matter thAT, at $Z=0$, lies wItHin THe viriAl radIuS Of tHe system for nF (leFt panel) anD E-0.7 (righT Panel). We ALso show With doTs tHe sPeciFIc AnGulAr MOmeNTuM foR The Baryons wHiCh End up As coLD GAS or sTarS in tHe cenTral $20\ h^{-1}$ kpc at $z=0$. ThE arRows SHow The spEcifiC angUlAr momEntum oF disk AnD spheroid stars. []{dAta-lAbel="j_evolUtiOn"}](Fig5A.pS "fig:"){wIDth="65mm"}![DAshEd lInes shoW the speCIfiC aNGULaR momentum as a functiOn OF TiMe for the Dark maTTeR tHAt, at $z=0$, lieS wIthIn thE VIrial RadiUS oF the systEm for Nf (LeFt Panel) anD E-0.7 (Right pAnEl). WE alSo shoW With Dots thE specifiC anguLAr momentum for tHE baryons which ENd UP As COld gAs oR stars in the CentRAl $20\ h^{-1}$ kPc at $Z=0$. thE arROws shOw the SpECiFIc angular momentum of DiSk and sPheroId stars. []{data-laBel="j_evolutION"}](Fig5b.ps "fiG:"){widTH="65mM"}
in Fig \[metal\_profIles\] wE show the oxYGen profiLes foR the no-feEdback (NF) aND Feedback (e-0.7) ruNs. FRom ThiS FIgUre we can see thAT sN feEdBack strOngLy affecTs tHe cHemIcaL dIstributiOns. If no fEeDbAcK iS inCludeD, The gas is EnRicHeD onLy in tHE very cEntraL regIoNs. iNclUding SN FEeDBAck tRiGgErs a RedIsTribuTion OF maSs and meTals throuGh gALactIc WiNds and fOuntains, givinG tHe gas compoNeNt a Much hiGHEr level oF enrichment out to large raDIi. A lineAr fIt to tHis mEtallicitY prOfile gIveS A slope Of $-0.048$ dex kPc$^{-1}$ and A zEro-POInt of $8.77$ DEX, cOnsIsTent with thE OBseRved vAlUes iN real diSk galaxies (e.g. ZaritsKY et Al. 1994).
![Oxygen abundAncE for THE gAs cOMpONenT aS A fuNCTion | $z=1$ and $z=0$ with high specificangul armom en tum(the se stars formf romhigh specific angularmomen tu m gas wh ich b ecomesc ol d atla te ti me s ); andsec ondly,dynamicalfri ct ion affectst he system mu chless than in NF , sinc esat e llite s a re le ss mas s ive. A t $z=0$,di s k star s have a s pe cifi c angular momentu m c o mparable to th at ofth e d a r k m att er, whilesp heroi d starsh av e a mu c h lower speci fic angular mom entum.
![D a shed l inessh o w t he specific ang ular mome ntum a s a func t ion oftime f orthe dar k m at ter t h at, at $z = 0$, lies wi th in theviri a l r adiu s o f th e sys tem for NF (l eft pan e l)and E -0.7(rig ht pane l). We also s how with dots t he s pecific a ngu la r m om entum for th e b ary ons whi ch endu p a sc o l dgas or stars in th ec e nt ral $20\ h^{-1 } $kp c at $z=0 $. Th e ar r o ws sh ow t h especific angul a rmo mentumof diskan d s phe roids tars . []{d ata-labe l="j_ e volution"}](fi g 5a.ps "fig:") { wi d t h= " 65mm "}! [Dashed lin es s h ow t he s p ec ifi c angu lar m om e nt u m as a function ofti me for thedark matter t hat, at $z = 0 $ , lies w ithi n t h e virial radiu s ofthe system for NF ( leftpanel) a nd E-0.7( r ight pan el) . W e a lso s ho w with dots t h e spe ci fic ang ula r momen tum fo r t heba ryons whi ch end u pas c ol d g as or stars in t hece ntr al $2 0 \ h^{- 1}$ k pc a t$z = 0$. The ar r ow s show t he spe cif ic angu larm ome ntum of disk and sp h eroi dst ars. [] {data-label=" j_ evolution" }] (fi g5b.ps " fig:"){w idth="65mm"}
In Fig \[ m etal\_p rof iles\ ] we show the ox ygen p rof i les fo r theno-fe ed bac k (NF)a n dfee db ack (E-0.7 ) run s. Fr om thi s figur e we can see thatS N f eedback stron gly aff e c ts th e c h emi ca l di s t ributions. If n o feedback i s i ncluded, t h e g as is enr iched o nly i n the ve ry centra l regions .Incl u d ing SN feedba ck trigg ers a red i strib u ti on of ma ss and m eta ls th roughg ala cticwindsan d foun tains ,giving t he gas component a much highe r lev elof enrich men t ou t to larg e ra dii. A lin ear fi t tothi s meta llic i ty pr o filegive s a slopeo f$-0 . 0 48 $ dex kpc$^ { - 1 }$and a ze r o-poin t of $8.77$ dex, cons i stent with the obs e r ved va l uesin real disk gal axi es ( e.g. Zar it sky et al.1994).
![ O xygen abund ance f or theg a sc ompone nt a s a function | $z=1$_and $z=0$_with high specific angular_momentum (these_stars_form from_high_specific angular momentum_gas which becomes_cold at late times);_and secondly, dynamical_friction_affects the system much less than in NF, since satellites are less massive. At_$z=0$,_disk stars_have_a_specific angular momentum comparable to_that of the dark matter,_while spheroid_stars have a much lower specific angular momentum.
![Dashed_lines_show the specific_angular momentum as a function of time for the_dark matter that, at $z=0$, lies_within the virial_radius_of_the system for NF_(left panel) and E-0.7 (right panel)._We also show with dots the_specific angular momentum for the baryons which_end up as cold gas or_stars in the central $20\_h^{-1}$ kpc_at $z=0$. The arrows show_the specific angular_momentum of_disk and spheroid_stars. []{data-label="j_evolution"}](fig5a.ps "fig:"){width="65mm"}![Dashed lines show the_specific angular momentum_as a function of time for_the_dark matter that,_at_$z=0$,_lies within_the virial radius_of_the system_for_NF (left panel) and E-0.7 (right_panel)._We also show with dots the specific_angular momentum for the_baryons_which end up as_cold gas or stars in_the central $20\ h^{-1}$ kpc at_$z=0$. The_arrows show_the specific angular momentum of disk and spheroid stars. []{data-label="j_evolution"}](fig5b.ps "fig:"){width="65mm"}
In_Fig \[metal\_profiles\] we show the oxygen profiles_for the no-feedback (NF)_and feedback_(E-0.7)_runs. From this_figure_we can_see that SN feedback strongly affects the_chemical distributions._If no feedback is included, the_gas is enriched only_in_the very central regions. Including SN_feedback triggers a redistribution of mass_and metals through galactic winds_and_fountains,_giving the gas component a_much higher level of enrichment out_to large radii._A linear fit to this metallicity profile_gives_a slope of $-0.048$ dex kpc$^{-1}$_and_a zero-point of $8.77$ dex, consistent_with_the_observed values in real disk_galaxies (e.g. Zaritsky et al. 1994).
![Oxygen_abundance for the gas component as a function |
ensuremath{\otimes}}{\ensuremath{\mathscr{C}}}^{\text{op}}{\ensuremath{\otimes}}{\ensuremath{\mathscr{C}}}{\ensuremath{\xymatrix@1@C=15pt{\ar[r]&}}}{\ensuremath{\mathscr{V}}}$$ $$J: {\ensuremath{\mathscr{C}}}{\ensuremath{\xymatrix@1@C=15pt{\ar[r]&}}}{\ensuremath{\mathscr{V}}}$$ so that $$\widehat{T}(M) = A \ast M = \int^{U,V} P(U,V;-) {\ensuremath{\otimes}}AU {\ensuremath{\otimes}}MV.$$ This means that the module $T: \xymatrix@1{{\ensuremath{\mathscr{C}}}\ar[r] |-{\object@{/}} & {\ensuremath{\mathscr{C}}}}$ is defined by $$\begin{split}
T(U,V) & = \widehat{T}({\ensuremath{\mathscr{C}}}(U,-))V \\
& = \int^{U',V'} P(U',V';V) {\ensuremath{\otimes}}AU' {\ensuremath{\otimes}}{\ensuremath{\mathscr{C}}}(U,V') \\
& \cong \int^{U'} P(U',U;V) {\ensuremath{\otimes}}AU'.
\end{split}$$
A promonad $T$ on ${\ensuremath{\mathscr{C}}}$ has a unit $\eta: {\ensuremath{\overset{\vee}}}{1} {\ensuremath{\xymatrix@1@C=15pt{\ar[r]&}}}T$ with components $$\eta_{U,V} : {\ensuremath{\mathscr{C}}}(U,V) {\ensuremath{\xymatrix@1@C=15pt{\ar[r]&}}}T(U,V)$$ and so is determined by $$\eta_{U,V}(1_U) : I {\ensuremath{\xymatrix@1@C=15pt{\ar[r]&}}}T(U,U),$$ and has a multiplication $\mu: T {\ensuremath{\circ}}T {\ensuremath{\xymatrix@1@C=15pt{\ar[r]&}}}T$ with components $$\mu_{U,W} : \int^{V} T(V,W) {\ensuremath{\otimes}}T(U,V) {\ensuremath{\xymatrix@1@C=15pt{\ar[r]&}}}T(U,W)$$ and so is determined by a natural family $$\mu'_{U,V | ensuremath{\otimes}}{\ensuremath{\mathscr{C}}}^{\text{op}}{\ensuremath{\otimes}}{\ensuremath{\mathscr{C}}}{\ensuremath{\xymatrix@1@C=15pt{\ar[r]&}}}{\ensuremath{\mathscr{V}}}$$ $ $ J: { \ensuremath{\mathscr{C}}}{\ensuremath{\xymatrix@1@C=15pt{\ar[r]&}}}{\ensuremath{\mathscr{V}}}$$ so that $ $ \widehat{T}(M) = A \ast M = \int^{U, V } P(U, V;-) { \ensuremath{\otimes}}AU { \ensuremath{\otimes}}MV.$$ This means that the module $ metric ton: \xymatrix@1{{\ensuremath{\mathscr{C}}}\ar[r ] |-{\object@{/ } } & { \ensuremath{\mathscr{C}}}}$ is define by $ $ \begin{split }
T(U, V) & = \widehat{T}({\ensuremath{\mathscr{C}}}(U,-))V \\
& = \int^{U',V' } P(U',V';V) { \ensuremath{\otimes}}AU' { \ensuremath{\otimes}}{\ensuremath{\mathscr{C}}}(U, V') \\
& \cong \int^{U' } P(U',U;V) { \ensuremath{\otimes}}AU'.
\end{split}$$
A promonad $ T$ on $ { \ensuremath{\mathscr{C}}}$ has a unit $ \eta: { \ensuremath{\overset{\vee}}}{1 } { \ensuremath{\xymatrix@1@C=15pt{\ar[r]&}}}T$ with components $ $ \eta_{U, V }: { \ensuremath{\mathscr{C}}}(U, V) { \ensuremath{\xymatrix@1@C=15pt{\ar[r]&}}}T(U, V)$$ and then is determined by $ $ \eta_{U, V}(1_U): I { \ensuremath{\xymatrix@1@C=15pt{\ar[r]&}}}T(U, U),$$ and has a generation $ \mu: T { \ensuremath{\circ}}T { \ensuremath{\xymatrix@1@C=15pt{\ar[r]&}}}T$ with components $ $ \mu_{U, W }: \int^{V } T(V, W) { \ensuremath{\otimes}}T(U, V) { \ensuremath{\xymatrix@1@C=15pt{\ar[r]&}}}T(U, W)$$ and so is determined by a lifelike syndicate $ $ \mu'_{U, V | enskremath{\otimes}}{\ensuremath{\mxthscr{C}}}^{\text{op}}{\ensuremati{\otimes}}{\snsuremagh{\mathscr{C}}}{\ensuremath{\xymatrix@1@R=15pt{\ae[r]&}}}{\ensyremath{\mathscr{V}}}$$ $$J: {\ensufemath{\matjscr{C}}}{\ensyremeth{\xymatrix@1@C=15pt{\ar[c]&}}}{\snsuremath{\matgdcr{V}}}$$ wo that $$\widehaj{T}(M) = A \ast M = \int^{U,V} P(U,V;-) {\envufelath{\otimes}}AU {\ensuremath{\otimes}}MV.$$ Thif means tjat the module $T: \xjmwtris@1{{\vnwuremath{\mathscr{C}}}\ar[r] |-{\object@{/}} & {\ehsuremauh{\mathscr{C}}}}$ is defimed by $$\begin{split}
T(U,V) & = \wifehah{T}({\ensuremath{\mathscg{C}}}(U,-))V \\
& = \unt^{U',D'} P(U',V';V) {\ensureoath{\otimes}}AU' {\ensurematg{\otimes}}{\ensuremath{\mathscr{C}}}(U,V') \\
& \cony \int^{U'} P(U',U;V) {\ebsuggmath{\otimes}}AN'.
\end{spjit}$$
A promonad $T$ on ${\etsuremayh{\mathscr{C}}}$ has a nnit $\eta: {\ensuremath{\oversev{\vee}}}{1} {\ensuremath{\xymattix@1@C=15pt{\ar[r]&}}}T$ wnth components $$\eta_{U,V} : {\ebsurekath{\kathrxr{C}}}(J,V) {\tnsnrejath{\xylatcix@1@C=15pt{\ar[r]&}}}T(U,B)$$ and so is determined by $$\eta_{U,F}(1_U) : I {\ensuremath{\symatryx@1@S=15pt{\ar[r]&}}}T(U,U),$$ and has a multiplication $\mu: T {\enauremath{\circ}}T {\ensurematy{\xymatrix@1@C=15pt{\ar[r]&}}}T$ with componenes $$\mu_{U,W} : \int^{V} T(V,W) {\ensuremath{\otimes}}T(U,V) {\ensuremath{\xfmatrmx@1@Z=15pt{\cv[v]&}}}T(U,W)$$ qnf so is determined by a natural family $$\mu'_{U,V | ensuremath{\otimes}}{\ensuremath{\mathscr{C}}}^{\text{op}}{\ensuremath{\otimes}}{\ensuremath{\mathscr{C}}}{\ensuremath{\xymatrix@1@C=15pt{\ar[r]&}}}{\ensuremath{\mathscr{V}}}$$ $$J: {\ensuremath{\mathscr{C}}}{\ensuremath{\xymatrix@1@C=15pt{\ar[r]&}}}{\ensuremath{\mathscr{V}}}$$ so that $$\widehat{T}(M) = M \int^{U,V} P(U,V;-) {\ensuremath{\otimes}}MV.$$ This means |-{\object@{/}} {\ensuremath{\mathscr{C}}}}$ is defined $$\begin{split} T(U,V) & \widehat{T}({\ensuremath{\mathscr{C}}}(U,-))V \\ & = \int^{U',V'} P(U',V';V) {\ensuremath{\otimes}}{\ensuremath{\mathscr{C}}}(U,V') \\ & \cong \int^{U'} P(U',U;V) {\ensuremath{\otimes}}AU'. \end{split}$$ A promonad $T$ on ${\ensuremath{\mathscr{C}}}$ a unit $\eta: {\ensuremath{\overset{\vee}}}{1} {\ensuremath{\xymatrix@1@C=15pt{\ar[r]&}}}T$ with components $$\eta_{U,V} : {\ensuremath{\mathscr{C}}}(U,V) {\ensuremath{\xymatrix@1@C=15pt{\ar[r]&}}}T(U,V)$$ and so determined $$\eta_{U,V}(1_U) I and has a multiplication $\mu: T {\ensuremath{\circ}}T {\ensuremath{\xymatrix@1@C=15pt{\ar[r]&}}}T$ with components $$\mu_{U,W} : \int^{V} T(V,W) {\ensuremath{\otimes}}T(U,V) {\ensuremath{\xymatrix@1@C=15pt{\ar[r]&}}}T(U,W)$$ so is determined by a natural family $$\mu'_{U,V | ensuremath{\otimes}}{\ensurematH{\mathscr{C}}}^{\tExt{op}}{\EnsUreMaTh{\otImes}}{\Ensuremath{\mathSCr{C}}}{\eNsuremath{\xymatrix@1@C=15pt{\ar[R]&}}}{\ensuReMAth{\mAThScr{V}}}$$ $$J: {\EnsuremATh{\MAThsCr{c}}}{\eNsuReMAtH{\xymaTriX@1@C=15pt{\ar[r]&}}}{\Ensuremath{\MatHsCr{V}}}$$ so that $$\widEHaT{T}(M) = A \ast M = \inT^{U,V} p(U,V;-) {\ensurematH{\otImes}}AU {\EnSurEMath{\oTimEs}}MV.$$ THis meaNS that tHe module $T: \XyMAtrix@1{{\eNSurematH{\MAtHscr{c}}}\ar[r] |-{\object@{/}} & {\ensuremATh{\MAthscr{C}}}}$ is definEd by $$\beGiN{SpLIT}
T(U,v) & = \wiDehat{T}({\ensuReMath{\mAThscr{C}}}(U,-))v \\
& = \InT^{u',v'} p(U',V';v) {\Ensuremath{\otiMes}}AU' {\ensureMAth{\Otimes}}{\EnSurEMath{\maThscr{c}}}(U,v') \\
& \ConG \int^{U'} P(U',U;V) {\enSureMath{\otimeS}}AU'.
\end{SPlit}$$
A prOMonad $T$ oN ${\ensurEmaTh{\mAthsCR{C}}}$ HaS a uNiT $\Eta: {\ENsUreMAth{\Overset{\vEe}}}{1} {\EnSuremAth{\xYMATRix@1@C=15Pt{\aR[r]&}}}T$ wIth coMponents $$\eta_{U,V} : {\EnsUremATh{\mAthscR{C}}}(U,V) {\eNsurEmAth{\xyMatrix@1@c=15pt{\ar[R]&}}}T(u,V)$$ and so is determIned By $$\eta_{U,V}(1_U) : I {\EnsUrEmaTh{\XymatRIx@1@C=15pt{\aR[r]&}}}T(u,U),$$ aNd has a mUltipliCAtiOn $\MU: t {\EnSuremath{\circ}}T {\ensurEmATH{\xYmatrix@1@C=15Pt{\ar[r]&}}}T$ WItH cOMponents $$\Mu_{u,W} : \iNt^{V} T(v,w) {\EnsurEmatH{\OtImes}}T(U,V) {\eNsuremATh{\XyMatrix@1@C=15Pt{\Ar[r]&}}}T(U,W)$$ AnD so Is dEtermINed bY a natuRal familY $$\mu'_{U,V | ensuremath{\otimes}}{\ensu remath{\ma thscr {C} }}^ {\ text {op} }{\ensuremath{ \ otim es}}{\ensuremath{\math scr{C }} } {\en s ur emath {\xymat r ix @ 1 @C= 15 pt {\a r[ r ]& }}}{\ ens uremath {\mathscr{ V}} }$ $ $$J: {\ens u re math{\math scr {C}}}{\ensur ema th{\xy ma tri x @1@C= 15p t{\ar [r]&}} } {\ensu remath{\m at h scr{V} } }$$ sot h at $$\ widehat{T}(M) = A \a s t M = \int^{U, V} P(U ,V ; -) { \en sur emath{\oti me s}}AU {\ensur e ma t h { \ot i mes}}MV.$$ Th is means th a t t he mod ul e $ T : \xym atrix @1 { {\e nsuremath{\ math scr{C}}}\ ar[r]| -{\obje c t@{/}}& {\en sur ema th{\ m at hs cr{ C} } }}$ is de f ine d by $$\ be gi n{spl it}T ( U , V) & = \wi dehat {T}({\ensurem ath {\ma t hsc r{C}} }(U,- ))V\\
& = \i nt^{U ', V'} P(U',V';V){\en suremath{ \ot im es} }A U' {\ e nsurem ath {\o times}} {\ensur e mat h{ \ m a th scr{C}}}(U,V') \\ &\cong \i nt^{U' } P (U ' ,U;V) {\ en sur emat h { \otim es}} A U' .
\end{s plit}$ $
Apromona d$T$ on $ {\e nsu remat h {\ma thscr{ C}}}$ ha s a u n it $\eta: {\en s uremath{\over s et { \ ve e }}}{ 1}{\ensuremat h{\x y matr ix@1 @ C= 15p t {\ar[ r]&}} }T $ w i th components $$\et a_ {U,V}: {\e nsuremath{\ma thscr{C}}} ( U , V) {\ens urem a th { \xymatrix@1@C= 15pt{ \ar[r]&}}} T (U,V)$$and s o is det ermined b y $$\eta_{ U,V }(1 _U) :I {\ ensuremath{\x y m atri x@ 1@C=15p t{\ ar[r]&} }}T (U, U), $$an d has a m ultiplic at io n$\ mu: T {\ e nsuremat h{ \ci rc }}T {\en s uremat h{\xy matr ix @1 @ C=1 5pt{\ar [ r] & } }}T$ w it h co mpo ne nts $ $\mu _ {U, W} : \i nt^{V} T( V,W ) {\e ns ur emath{\ otimes}}T(U,V ){\ensurema th {\x ymatri x @ 1@C=15pt {\ar[r]&}}}T(U,W)$$ and so is d ete rmine d by a natura l f amily$$\ m u'_{U, V | ensuremath{\otimes}}{\ensuremath{\mathscr{C}}}^{\text{op}}{\ensuremath{\otimes}}{\ensuremath{\mathscr{C}}}{\ensuremath{\xymatrix@1@C=15pt{\ar[r]&}}}{\ensuremath{\mathscr{V}}}$$ $$J:_{\ensuremath{\mathscr{C}}}{\ensuremath{\xymatrix@1@C=15pt{\ar[r]&}}}{\ensuremath{\mathscr{V}}}$$ so_that $$\widehat{T}(M) = A_\ast M_=_\int^{U,V} P(U,V;-)_{\ensuremath{\otimes}}AU_{\ensuremath{\otimes}}MV.$$ This means_that the module_$T: \xymatrix@1{{\ensuremath{\mathscr{C}}}\ar[r] |-{\object@{/}} &_{\ensuremath{\mathscr{C}}}}$ is defined_by_$$\begin{split}
T(U,V) & = \widehat{T}({\ensuremath{\mathscr{C}}}(U,-))V \\
& = \int^{U',V'} P(U',V';V) {\ensuremath{\otimes}}AU'_{\ensuremath{\otimes}}{\ensuremath{\mathscr{C}}}(U,V')_\\
___& \cong \int^{U'} P(U',U;V) {\ensuremath{\otimes}}AU'.
\end{split}$$
A_promonad $T$ on ${\ensuremath{\mathscr{C}}}$ has_a unit_$\eta: {\ensuremath{\overset{\vee}}}{1} {\ensuremath{\xymatrix@1@C=15pt{\ar[r]&}}}T$ with components $$\eta_{U,V} : {\ensuremath{\mathscr{C}}}(U,V)_{\ensuremath{\xymatrix@1@C=15pt{\ar[r]&}}}T(U,V)$$_and so is_determined by $$\eta_{U,V}(1_U) : I {\ensuremath{\xymatrix@1@C=15pt{\ar[r]&}}}T(U,U),$$ and has a_multiplication $\mu: T {\ensuremath{\circ}}T {\ensuremath{\xymatrix@1@C=15pt{\ar[r]&}}}T$ with_components $$\mu_{U,W} :_\int^{V}_T(V,W)_{\ensuremath{\otimes}}T(U,V) {\ensuremath{\xymatrix@1@C=15pt{\ar[r]&}}}T(U,W)$$ and so_is determined by a natural family_$$\mu'_{U,V |
only holds if the prefactor $\varepsilon^0$ is not zero. Otherwise, we run into a “$0/0$” situation, which is mathematically not defined. Now, the physical phase space of the process contains a sector, for which $|k_{\parallel}|$ becomes arbitrarily small. This sector is characterized by two angles $(\varphi_0,\vartheta_0)$, where for $\varphi_0=0$, $\vartheta_0\approx 2.35$. This is depicted in Fig. \[fig:zero-contour\].
![Contour of angles $(\varphi_0,\vartheta_0)$, for which the normalization factor $N''$ vanishes. The peaks in Fig. \[fig:plots-matrix-element-square-parameter-choice-3\], lie on this contour.[]{data-label="fig:zero-contour"}](consistency-modmax-parity-odd_v1_fig6_contour-zeros.pdf)
For this special case the normalization factor $N''$ and, therefore, the prefactor $\varepsilon^0$ can become arbitrarily small. This destroys the applicability of the Ward identity and shows up as peaks in $X_{22}$ of Figs. \[fig:plots-matrix-element-square-parameter-choice-2\] and \[fig:plots-matrix-element-square-parameter-choice-3\].
Now we would like to analytically investigate the limit $\mathcal{E}\mapsto 0$ of the second polarization vector, with its transversal part subtracted. We distinguish between two cases, $k_{\parallel}\sim \mathcal{E}k_{\bot}$ and $k_{\parallel}\gg \mathcal{E}k_{\bot}$. The first represents the phase space sector for Compton scattering, for which $k_{\parallel}$ becomes arbitrarily small. We begin with the zeroth component of Eq. : $$\begin{aligned}
\sqrt{N''}\,\varepsilon^{0}&=\frac{1}{4}\big(k^2-(k\cdot\xi)^2\big)\Big((k\cdot \zeta)^2-\zeta^2\big(k^2-(k\cdot\xi)^2\big)\Big) \notag \\
&\sim \left\{\begin{array}{lcl}
\mathcal{E}^2|\mathbf{k}|^2(k_{\parallel}^2+|\mathbf{k}|^2) & \text{for} & k_{\ | only holds if the prefactor $ \varepsilon^0 $ is not zero. Otherwise, we run into a “ $ 0/0 $ ” site, which is mathematically not define. Now, the physical phase space of the summons contains a sector, for which $ |k_{\parallel}|$ becomes randomly small. This sector is characterized by two angles $ (\varphi_0,\vartheta_0)$, where for $ \varphi_0=0 $, $ \vartheta_0\approx 2.35$. This is picture in Fig. \[fig: zero - contour\ ].
! [ Contour of angles $ (\varphi_0,\vartheta_0)$, for which the standardization factor $ N''$ vanishes. The peaks in Fig. \[fig: plots - matrix - element - straight - parameter - choice-3\ ], lie on this contour.[]{data - label="fig: zero - contour"}](consistency - modmax - parity - odd_v1_fig6_contour - zeros.pdf)
For this special case the standardization factor $ N''$ and, therefore, the prefactor $ \varepsilon^0 $ can become arbitrarily little. This destroys the applicability of the Ward identity and shows up as point in $ X_{22}$ of Figs. \[fig: plots - matrix - component - square - parameter - choice-2\ ] and \[fig: plots - matrix - element - square - argument - choice-3\ ].
Now we would like to analytically investigate the limit $ \mathcal{E}\mapsto 0 $ of the second polarization vector, with its transversal part subtracted. We identify between two cases, $ k_{\parallel}\sim \mathcal{E}k_{\bot}$ and $ k_{\parallel}\gg \mathcal{E}k_{\bot}$. The first represents the phase space sector for Compton scattering, for which $ k_{\parallel}$ becomes arbitrarily small. We begin with the zeroth component of Eq. : $ $ \begin{aligned }
\sqrt{N''}\,\varepsilon^{0}&=\frac{1}{4}\big(k^2-(k\cdot\xi)^2\big)\Big((k\cdot \zeta)^2-\zeta^2\big(k^2-(k\cdot\xi)^2\big)\Big) \notag \\
& \sim \left\{\begin{array}{lcl }
\mathcal{E}^2|\mathbf{k}|^2(k_{\parallel}^2+|\mathbf{k}|^2) & \text{for } & k_{\ | onpy holds if the prefactov $\varepsilon^0$ is uit zerm. Othedwise, we run into a “$0/0$” situation, which iw matyematically not definea. Now, the physicao phese space of the process contajks a vxctor, for which $|k_{\parallel}|$ becomes arbitsafipy small. This sector is characterizqd by teo angles $(\varphi_0,\dartnqta_0)$, shere for $\varphi_0=0$, $\vartheta_0\approx 2.35$. Tgis is vepicted in Fig. \[fig:zero-contour\].
![Contour of wnglfs $(\varphi_0,\vartheta_0)$, vor which tye njemalization wactor $N''$ vanishes. The keaks in Fig. \[fig:plots-matrix-elemevt-squcre-parametet-eyoifg-3\], lie on thiw connour.[]{data-label="npg:zero-cmntour"}](cpnsistency-modmsx-peritt-odd_v1_fig6_contour-zeros.pvf)
For this special cwse the nmrjalization factor $N''$ and, jherexore, rhe prtfartod $\vareosimon^0$ can bedome arbitrqrily small. This dextwiys the applidabiliey of the Ward identity and shows up as ptaks jn $X_{22}$ of Figs. \[fig:plots-marrix-element-square-paraleter-choise-2\] and \[fig:plots-matrix-element-square-parameter-choice-3\].
Tow wx douod uukf to analytically investigate the limit $\mathcwm{E}\kaisto 0$ of the secokd polarization vevtlr, rith its tranrversal pzrt subtracted. We fistingoish bwtween twj caxes, $k_{\parallel}\sim \mathcal{E}k_{\bit}$ and $k_{\parajoel}\gg \mathcal{E}k_{\bot}$. The first xepresgnts tne phase space sector fur Ckmpton scathering, fod which $k_{\parallel}$ bebomev arbitrarily small. We begyn with tie zexoth comoonemt of Qq. : $$\begin{apigned}
\sqrt{N''}\,\varepsilon^{0}&=\fraf{1}{4}\big(k^2-(n\cgot\xi)^2\big)\Bih((k\cdot \zeta)^2-\zeta^2\big(k^2-(k\cdot\xi)^2\big)\Bij) \notag \\
&\sim \lgft\{\tegpn{array}{lcj}
\mathgal{E}^2|\mathbf{k}|^2(k_{\parwllel}^2+|\mathbf{k}|^2) & \jext{for} & h_{\ | only holds if the prefactor $\varepsilon^0$ is Otherwise, run into “$0/0$” situation, which the phase space of process contains a for which $|k_{\parallel}|$ becomes arbitrarily small. sector is characterized by two angles $(\varphi_0,\vartheta_0)$, where for $\varphi_0=0$, $\vartheta_0\approx 2.35$. This depicted in Fig. \[fig:zero-contour\]. ![Contour of angles $(\varphi_0,\vartheta_0)$, for which the normalization factor vanishes. peaks Fig. lie on this contour.[]{data-label="fig:zero-contour"}](consistency-modmax-parity-odd_v1_fig6_contour-zeros.pdf) For this special case the normalization factor $N''$ and, therefore, the prefactor can become arbitrarily small. This destroys the applicability the Ward identity and up as peaks in $X_{22}$ Figs. and \[fig:plots-matrix-element-square-parameter-choice-3\]. we like analytically investigate the $\mathcal{E}\mapsto 0$ of the second polarization vector, with its transversal part subtracted. We distinguish between two cases, \mathcal{E}k_{\bot}$ and The first the space for Compton scattering, $k_{\parallel}$ becomes arbitrarily small. We begin component of Eq. : $$\begin{aligned} \sqrt{N''}\,\varepsilon^{0}&=\frac{1}{4}\big(k^2-(k\cdot\xi)^2\big)\Big((k\cdot \zeta)^2-\zeta^2\big(k^2-(k\cdot\xi)^2\big)\Big) \notag &\sim \left\{\begin{array}{lcl} & \text{for} & k_{\ | only holds if the prefactor $\vaRepsilon^0$ is Not zeRo. OTheRwIse, wE run Into a “$0/0$” situation, WHich Is mathematically not defIned. NOw, THe phYSiCal phAse spacE Of THE prOcEsS coNtAInS a secTor, For whicH $|k_{\parallel}|$ BecOmEs arbitrarilY SmAll. This secTor Is characteriZed By two aNgLes $(\VArphi_0,\VarTheta_0)$, Where fOR $\varphI_0=0$, $\vartheta_0\ApPRox 2.35$. ThiS Is depicTED iN Fig. \[Fig:zero-contour\].
![ConTOuR Of angles $(\varphi_0,\VartheTa_0)$, FOr WHIch The NormalizatIoN factOR $N''$ vanisHEs. tHE PeaKS in Fig. \[fig:plotS-matrix-elemENt-sQuare-pArAmeTEr-choiCe-3\], lie On THis Contour.[]{data-LabeL="fig:zero-cOntour"}](COnsisteNCy-modmaX-paritY-odD_v1_fIg6_coNToUr-ZerOs.PDf)
FOR tHis SPecIal case tHe NoRmaliZatiON FACtor $n''$ anD, theReforE, the prefactor $\VarEpsiLOn^0$ cAn becOme arBitrArIly smAll. ThiS destRoYs the applicabilIty oF the Ward iDenTiTy aNd Shows UP as peaKs iN $X_{22}$ oF Figs. \[fiG:plots-mATriX-eLEMEnT-square-parameter-chOiCE-2\] AnD \[fig:plotS-matriX-ElEmENt-square-PaRamEter-CHOice-3\].
NOw we WOuLd like to AnalytICaLlY investIgAte the LiMit $\MatHcal{E}\MApstO 0$ of the Second poLarizATion vector, with ITs transversal PArT SUbTRactEd. WE distinguisH betWEen tWo caSEs, $K_{\paRAllel}\Sim \maThCAl{e}K_{\bot}$ and $k_{\parallel}\gg \mAtHcal{E}k_{\Bot}$. ThE first represeNts the phasE SPAce sectoR for cOmPTon scattering, fOr whiCh $k_{\paralleL}$ Becomes aRbitrArily smaLl. We begin WITh the zerOth ComPonEnt OF eq. : $$\Begin{aligned}
\sQRT{N''}\,\vaRePsilon^{0}&=\fRac{1}{4}\Big(k^2-(k\cdOt\xI)^2\biG)\BiG((k\cDoT \zeta)^2-\zeta^2\Big(k^2-(k\cdoT\xI)^2\bIg)\biG) \noTag \\
&\siM \Left\{\begiN{aRraY}{lCl}
\mAthcaL{e}^2|\mathbF{k}|^2(k_{\paRallEl}^2+|\MaTHbf{K}|^2) & \text{foR} & K_{\ | only holds if the prefact or $\varep silon ^0$ is n ot z ero. Otherwise, we runinto a “$0/0$” situati on, w hi c h is ma thema tically no t def in ed . N ow , t he ph ysi cal pha se space o f t he process con t ai ns a secto r,for which $| k_{ \paral le l}| $ beco mes arbi traril y small . This se ct o r is c h aracter i z ed bytwo angles $(\var p hi _ 0,\vartheta_0) $, whe re fo r $\v arp hi_0=0$, $ \v arthe t a_0\app r ox 2 . 35$ . This is depi cted in Fig . \[ fig:ze ro -co n tour\] .
![ Co n tou r of angles $(\ varphi_0, \varth e ta_0)$, for whi ch the no rma liza t io nfac to r $N ' '$ va n ish es. Thepe ak s inFig. \ [ f ig:p lot s-ma trix- element-squar e-p aram e ter -choi ce-3\ ], l ie on t his co ntour .[ ]{data-label="f ig:z ero-conto ur" }] (co ns isten c y-modm ax- par ity-odd _v1_fig 6 _co nt o u r -z eros.pdf)
For thi ss p ec ial case the n o rm al i zation f ac tor $N' ' $ and, the r ef ore, the prefa c to r$\varep si lon^0$ c anbec ome a r bitr arilysmall. T his d e stroys the app l icability oft he W ar d ide nti ty and show s up as p eaks in $X _ {22}$ of F ig s .\ [fig:plots-matrix-e le ment-s quare -parameter-ch oice-2\] a n d \[fig:pl ots- m at r ix-element-squ are-p arameter-c h oice-3\] .
No w we wou ld like t o analytic all y i nve sti g a te the limit $\ m a thca l{ E}\maps to0$ of t hesec ond po la rizationvector,wi th i ts tr ansve r sal part s ubt ra cte d. We distin guish bet we en two cases, $k _ { \par al le l}\s im\m athca l{E} k _{\ bot}$ a nd $k_{\p ara l lel} \g g\mathca l{E}k_{\bot}$ .The firstre pre sentst h e phasespace sector for Compto n scatte rin g, fo r wh ich $k_{\ par allel} $ b e comesarbitr arily s mal l . We b e g in wi th the zerot h com ponen tof E q. : $$ \begin{aligned}
\s q rt{ N''}\,\vareps ilo n^{0 } & =\ fra c {1 } {4} \b i g(k ^ 2 -(k\cdot\xi)^2\ big)\Big(( k\ c do t \zeta)^2 - \ze ta ^2\big( k^2-(k\ cdot\ x i)^2\bi g)\Big) \ notag \\&\ sim\ l eft \{\begin{a rray}{lc l}
\mathc a l{E}^ 2 |\ mathb f{k }|^2(k _{ \pa ralle l}^2+| \ mat hbf{k }|^2)&\text{ for}&k_{\ | only_holds if_the prefactor $\varepsilon^0$ is_not zero._Otherwise,_we run_into_a “$0/0$” situation,_which is mathematically_not defined. Now, the_physical phase space_of_the process contains a sector, for which $|k_{\parallel}|$ becomes arbitrarily small. This sector is_characterized_by two_angles_$(\varphi_0,\vartheta_0)$,_where for $\varphi_0=0$, $\vartheta_0\approx 2.35$._This is depicted in Fig._\[fig:zero-contour\].
![Contour of_angles $(\varphi_0,\vartheta_0)$, for which the normalization factor $N''$_vanishes._The peaks in_Fig. \[fig:plots-matrix-element-square-parameter-choice-3\], lie on this contour.[]{data-label="fig:zero-contour"}](consistency-modmax-parity-odd_v1_fig6_contour-zeros.pdf)
For this special case_the normalization factor $N''$ and, therefore,_the prefactor $\varepsilon^0$_can_become_arbitrarily small. This destroys_the applicability of the Ward identity_and shows up as peaks in_$X_{22}$ of Figs. \[fig:plots-matrix-element-square-parameter-choice-2\] and \[fig:plots-matrix-element-square-parameter-choice-3\].
Now we_would like to analytically investigate the_limit $\mathcal{E}\mapsto 0$ of the_second polarization_vector, with its transversal part_subtracted. We distinguish_between two_cases, $k_{\parallel}\sim \mathcal{E}k_{\bot}$_and $k_{\parallel}\gg \mathcal{E}k_{\bot}$. The first represents_the phase space_sector for Compton scattering, for which_$k_{\parallel}$_becomes arbitrarily small._We_begin_with the_zeroth component of_Eq._: $$\begin{aligned}
\sqrt{N''}\,\varepsilon^{0}&=\frac{1}{4}\big(k^2-(k\cdot\xi)^2\big)\Big((k\cdot_\zeta)^2-\zeta^2\big(k^2-(k\cdot\xi)^2\big)\Big)_\notag \\
&\sim \left\{\begin{array}{lcl}
\mathcal{E}^2|\mathbf{k}|^2(k_{\parallel}^2+|\mathbf{k}|^2) & \text{for} &_k_{\ |
}}( E_{{{\text{L}}_{l}}}^{i})$ of a particular $E_{{{\text{L}}_{l}}}^{i}$ has to be at least ${B}_{F}^{\text{*}}( E_{{{\text{L}}_{l}}}^{i})$.
Definitions
-----------
Some preliminary definitions for the proposed model are as follows.
(Incoming and outgoing density matrix). In a $j$-th quantum repeater $R_{j} $, an $\rho $ incoming density matrix is half of an entangled state $\left| {{\beta }_{00}} \right\rangle =\tfrac{1}{\sqrt{2}}\left( \left| 00 \right\rangle +\left| 11 \right\rangle \right)$ received from a previous neighbor node $R_{j-1} $. The $\sigma $ outgoing density matrix in $R_{j} $ is half of an entangled state ${\left| \beta _{00} \right\rangle} $ shared with a next neighbor node $R_{j+1} $.
(Entanglement Swapping Operation). The $U_{S} $ entanglement swapping operation is a local transformation in a $j$-th quantum repeater $R_{j}$ that swaps an incoming density matrix $\rho $ with an outgoing density matrix $\sigma $ and measures the density matrices to entangle the distant source and target nodes $R_{j-1} $ and $R_{j+1} $.
(Entanglement Swapping Period). Let $C$ be a cycle with time $t_{C} ={1\mathord{\left/ {\vphantom {1 f_{C} }} \right. \kern-\nulldelimiterspace} f_{C} } $ determined by the $o_{C} $ oscillator in node $R_{j} $, where $f_{C} $ is the frequency of $o_{C} $. Then, let $\pi _{S} $ be an entanglement swapping period in which the set ${\rm {\mathcal S}}_{I} \left(R_{j} \right)=\bigcup _{i}\rho _{i} $ of incoming density matrices is swapped via $U_{S} $ with the set ${\rm {\mathcal S}}_{O} \left(R_{j} \right)=\bigcup _{i}\sigma _{i} $ of outgoing density matrices, defined as $\pi _{S} =xt_{C} $, where $x$ is the | } } (E_{{{\text{L}}_{l}}}^{i})$ of a particular $ E_{{{\text{L}}_{l}}}^{i}$ has to be at least $ { B}_{F}^{\text { * } } (E_{{{\text{L}}_{l}}}^{i})$.
Definitions
-----------
Some preliminary definitions for the aim exemplar are as follows.
(Incoming and outgoing density matrix). In a $ j$-th quantum repeating firearm $ R_{j } $, an $ \rho $ incoming density matrix is half of an embroiled state $ \left| { { \beta } _ { 00 } } \right\rangle = \tfrac{1}{\sqrt{2}}\left (\left| 00 \right\rangle + \left| 11 \right\rangle \right)$ received from a previous neighbor lymph node $ R_{j-1 } $. The $ \sigma $ extroverted density matrix in $ R_{j } $ is half of an embroiled state $ { \left| \beta _ { 00 } \right\rangle } $ shared with a adjacent neighbor node $ R_{j+1 } $.
(Entanglement Swapping Operation). The $ U_{S } $ web swapping operation is a local transformation in a $ j$-th quantum recidivist $ R_{j}$ that swaps an incoming density matrix $ \rho $ with an extroverted density matrix $ \sigma $ and measures the density matrix to entangle the distant source and prey nodes $ R_{j-1 } $ and $ R_{j+1 } $.
(Entanglement Swapping Period). Let $ C$ be a hertz with time $ t_{C } = { 1\mathord{\left/ { \vphantom { 1 f_{C } } } \right. \kern-\nulldelimiterspace } f_{C } } $ determined by the $ o_{C } $ oscillator in node $ R_{j } $, where $ f_{C } $ is the frequency of $ o_{C } $. Then, let $ \pi _ { S } $ be an entanglement swapping time period in which the set $ { \rm { \mathcal S}}_{I } \left(R_{j } \right)=\bigcup _ { i}\rho _ { i } $ of incoming density matrices is swapped via $ U_{S } $ with the set $ { \rm { \mathcal S}}_{O } \left(R_{j } \right)=\bigcup _ { i}\sigma _ { i } $ of outgoing density matrices, defined as $ \pi _ { S } = xt_{C } $, where $ x$ is the | }}( E_{{{\tfxt{L}}_{l}}}^{i})$ of a particular $E_{{{\uext{L}}_{l}}}^{i}$ has to be at leavt ${B}_{F}^{\tsxt{*}}( E_{{{\texg{L}}_{l}}}^{i})$.
Definitions
-----------
Some preliminery eefinutions for the proposea model age as foloows.
(Mncoming and outjking dekfity latrnx). In a $j$-th quanjum repeater $R_{j} $, an $\rho $ itcumnng density matrix is half of an entwngled xtwte $\left| {{\beta }_{00}} \rigne\ranfle =\tfrac{1}{\sqrt{2}}\left( \left| 00 \right\rangls +\left| 11 \right\rangle \right)$ received from a pregioud neighbor node $R_{j-1} $. The $\sigma $ ouetoing densith matrix in $R_{j} $ is halr of an entangled state ${\left| \bega _{00} \xight\rangle} $ whaggd with a neet neidhbor node $R_{m+1} $.
(Entanclement Swapping Operstimn). Rhe $U_{S} $ entanglement vwapping operation is a locdl transformation ib q $j$-th quattum eepdattr $C_{j}$ fhat saapa an incomjng density matrix $\rho $ with am jlygoing densify matwiv $\sigma $ and measures the density matribes fo entangle the distant source and target nofes $R_{j-1} $ agd $R_{j+1} $.
(Entanglement Swapping Period). Let $C$ be a cycne wivh tine $g_{X} ={1\lathord{\left/ {\vphantom {1 f_{C} }} \right. \kern-\nulldelimyfetsiace} f_{C} } $ determiked by the $o_{C} $ oscolpayjr in node $R_{j} $, wherz $f_{D} $ is the frequencj of $o_{C} $. Theb, let $\pi _{F} $ br an entanglement swapping period in whpch rhe set ${\rm {\mathcal S}}_{I} \left(R_{j} \xight)=\bogcup _{i}\rho _{i} $ of incoming dznsity matrices id swapped xia $U_{S} $ with the sen ${\rm {\mathcal S}}_{O} \left(R_{j} \right)=\bidcup _{i}\signa _{i} $ of ojtgoong degsity matrlces, defined as $\pi _{S} =xt_{C} $, whete $x$ iv the | }}( E_{{{\text{L}}_{l}}}^{i})$ of a particular $E_{{{\text{L}}_{l}}}^{i}$ has at ${B}_{F}^{\text{*}}( E_{{{\text{L}}_{l}}}^{i})$. ----------- Some preliminary are follows. (Incoming and density matrix). In $j$-th quantum repeater $R_{j} $, an $ incoming density matrix is half of an entangled state $\left| {{\beta }_{00}} =\tfrac{1}{\sqrt{2}}\left( \left| 00 \right\rangle +\left| 11 \right\rangle \right)$ received from a previous neighbor $R_{j-1} The $ density matrix in $R_{j} $ is half of an entangled state ${\left| \beta _{00} \right\rangle} $ with a next neighbor node $R_{j+1} $. (Entanglement Operation). The $U_{S} $ swapping operation is a local in $j$-th quantum $R_{j}$ swaps incoming density matrix $ with an outgoing density matrix $\sigma $ and measures the density matrices to entangle the distant and target $ and $. Swapping Let $C$ be with time $t_{C} ={1\mathord{\left/ {\vphantom {1 \kern-\nulldelimiterspace} f_{C} } $ determined by the $o_{C} oscillator in $R_{j} $, where $f_{C} $ is frequency of $o_{C} $. Then, let $\pi _{S} be an entanglement swapping period in which the set ${\rm {\mathcal S}}_{I} \left(R_{j} \right)=\bigcup _{i}\rho of incoming density matrices swapped via $U_{S} with set {\mathcal \left(R_{j} \right)=\bigcup _{i} $ of outgoing density matrices, defined as $\pi _{S} =xt_{C} where $x$ is the | }}( E_{{{\text{L}}_{l}}}^{i})$ of a particular $E_{{{\texT{L}}_{l}}}^{i}$ has to bE at leAst ${b}_{F}^{\tExT{*}}( E_{{{\teXt{L}}_{l}}}^{I})$.
Definitions
-----------
SoME preLiminary definitions for The prOpOSed mODeL are aS followS.
(inCOMinG aNd OutGoINg DensiTy mAtrix). In A $j$-th quantuM rePeAter $R_{j} $, an $\rho $ iNCoMing densitY maTrix is half of An eNtanglEd StaTE $\left| {{\BetA }_{00}} \righT\ranglE =\Tfrac{1}{\sQrt{2}}\left( \leFt| 00 \RIght\raNGle +\left| 11 \RIGhT\ranGle \right)$ received fROm A Previous neighbOr node $r_{j-1} $. tHe $\SIGma $ OutGoing densiTy MatriX In $R_{j} $ is hALf OF AN enTAngled state ${\leFt| \beta _{00} \right\RAngLe} $ sharEd WitH A next nEighbOr NOde $r_{j+1} $.
(EntanglemEnt SWapping OpEratioN). the $U_{S} $ enTAnglemeNt swapPinG opEratIOn Is A loCaL TraNSfOrmATioN in a $j$-th qUaNtUm repEateR $r_{J}$ THat sWapS an iNcomiNg density matrIx $\rHo $ wiTH an OutgoIng deNsitY mAtrix $\Sigma $ aNd meaSuRes the density maTricEs to entanGle ThE diStAnt soURce and TarGet Nodes $R_{j-1} $ And $R_{j+1} $.
(EnTAngLeMENT SWapping Period). Let $C$ bE a CYClE with timE $t_{C} ={1\matHOrD{\lEFt/ {\vphantOm {1 F_{C} }} \rIght. \KERn-\nulLdelIMiTerspace} F_{C} } $ deteRMiNeD by the $o_{c} $ oScillaToR in NodE $R_{j} $, whERe $f_{C} $ Is the fRequency Of $o_{C} $. THEn, let $\pi _{S} $ be an enTAnglement swapPInG PErIOd in WhiCh the set ${\rm {\mAthcAL S}}_{I} \lEft(R_{J} \RiGht)=\BIgcup _{I}\rho _{i} $ Of INcOMing density matrices Is SwappeD via $U_{s} $ with the set ${\rm {\Mathcal S}}_{O} \lEFT(r_{j} \right)=\bIgcuP _{I}\sIGma _{i} $ of outgoing DensiTy matrices, DEfined as $\Pi _{S} =xt_{c} $, where $x$ iS the | }}( E_{{{\text{L}}_{l}}}^{ i})$ of aparti cul ar$E _{{{ \tex t{L}}_{l}}}^{i } $ ha s to be at least ${B}_ {F}^{ \t e xt{* } }( E_{{ {\text{ L }} _ { l}} }^ {i })$ .De finit ion s
----- ------
So mepr eliminary de f in itions for th e proposed m ode l areas fo l lows.
( Incom ing an d outgo ing densi ty matrix ) . In a$ j $- th q uantum repeater $ R _{ j } $, an $\rho$ inco mi n gd e nsi tymatrix isha lf of an enta n gl e d sta t e $\left| {{\ beta }_{00} } \r ight\r an gle =\tfra c{1}{ \s q rt{ 2}}\left( \ left | 00 \rig ht\ran g le +\le f t| 11 \ right\ ran gle \r i gh t) $ r ec e ive d f rom a p reviousne ig hbornode $ R _ {j-1 } $ . Th e $\s igma $ outgoi ngdens i tymatri x in$R_{ j} $ is halfof an e ntangled state${\l eft| \bet a _ {0 0} \ right \ rangle } $ sh ared wi th a ne x t n ei g h b or node $R_{j+1} $.
( E n ta nglement Swapp i ng O p eration) .The $U_ { S } $ e ntan g le ment swa ppingo pe ra tion is a local t ran sfo rmati o n in a $j$ -th quan tum r e peater $R_{j}$ that swaps an in c o mi n g de nsi ty matrix $ \rho $ wi th a n o utg o ing d ensit ym at r ix $\sigma $ and me as ures t he de nsity matrice s to entan g l e the dis tant so u rce and target node s $R_{j-1} $ and $R _{j+1 } $.
(E ntangleme n t Swappin g P eri od) . L e t $ C$ be a cycle w ithti me $t_{ C}={1\mat hor d{\ lef t/{\ vphantom{1 f_{C} } }\r ig ht. \ker n -\nullde li mit er spa ce} f _ {C} }$ det ermi ne db y t he $o_{ C }$ osci ll at or i n n od e $R_ {j}$ , w here $f _{C} $ is th e fre qu en cy of $ o_{C} $. Then ,let $\pi _ {S } $ be an e ntanglem ent swapping period inw hich th e s et ${ \rm{\mathcal S} }_{I}\le f t(R_{j } \rig ht)=\ bi gcu p _{i}\ r h o_{i } $ of inco m i ngdensi ty mat rices i s swapped via $U_{ S } $ with the set ${ \rm{ \ ma thc a lS }}_ {O } \l e f t(R_{j} \right) =\bigcup _ {i } \s igma _{i}$ o foutgoin g densi ty ma t rices,defined a s $\pi _{ S} =xt _ { C}$, where $ x$ is th e | }}( E_{{{\text{L}}_{l}}}^{i})$_of a_particular $E_{{{\text{L}}_{l}}}^{i}$ has to_be at_least_${B}_{F}^{\text{*}}( E_{{{\text{L}}_{l}}}^{i})$.
Definitions
-----------
Some_preliminary_definitions for the_proposed model are_as follows.
(Incoming and outgoing_density matrix). In_a_$j$-th quantum repeater $R_{j} $, an $\rho $ incoming density matrix is half of_an_entangled state_$\left|_{{\beta_}_{00}} \right\rangle =\tfrac{1}{\sqrt{2}}\left( \left| 00_\right\rangle +\left| 11 \right\rangle _\right)$ received_from a previous neighbor node $R_{j-1} $. The_$\sigma_$ outgoing density_matrix in $R_{j} $ is half of an entangled_state ${\left| \beta _{00} \right\rangle}_$ shared with_a_next_neighbor node $R_{j+1} $.
(Entanglement_Swapping Operation). The $U_{S} $ entanglement_swapping operation is a local transformation_in a $j$-th quantum repeater $R_{j}$ that_swaps an incoming density matrix $\rho_$ with an outgoing density_matrix $\sigma_$ and measures the density_matrices to entangle_the distant_source and target_nodes $R_{j-1} $ and $R_{j+1} $.
(Entanglement_Swapping Period). Let_$C$ be a cycle with time_$t_{C}_={1\mathord{\left/ {\vphantom {1_f_{C}_}}_\right. \kern-\nulldelimiterspace}_f_{C} } $_determined_by the_$o_{C}_$ oscillator in node $R_{j} $,_where_$f_{C} $ is the frequency of $o_{C}_$. Then, let $\pi__{S}_$ be an entanglement_swapping period in which the_set ${\rm {\mathcal S}}_{I} \left(R_{j} \right)=\bigcup__{i}\rho _{i}_ $_of incoming density matrices is swapped via $U_{S} $ with the_set ${\rm {\mathcal S}}_{O} \left(R_{j} \right)=\bigcup__{i}\sigma _{i} $_of outgoing_density_matrices, defined as_$\pi__{S} =xt_{C}_$, where $x$ is the |
rm char}_{R}(M) := \operatorname{im}\left({\bigcap}^{r}_{R}N \longrightarrow {\bigcap}^{r}_{R}R^{r} = R \right).$$
\[rem:flat\] If $R'$ is a flat $R$-algebra, then we have ${\rm char}_{R}(M)R' = {\rm char}_{R'}(R' \otimes_{R}M)$.
\[rem:indep-char\] The ideal ${\rm char}_{R}(M)$ is independent of the choice of the exact sequence $0 \longrightarrow N \longrightarrow R^{r} \longrightarrow M \longrightarrow 0$.
In fact, by Remark \[rem:flat\], we may assume that $R$ is a local noetherian ring. Then any free resolution of $M$ is isomorphic to the direct sum of the minimal free resolution of $M$ and a trivial complex. Hence it suffices to show that $$\operatorname{im}\left({\bigcap}^{r}_{R}N \longrightarrow {\bigcap}^{r}_{R}R^{r} = R \right) =
\operatorname{im}\left({\bigcap}^{r+s}_{R}(N \oplus R^{s}) \longrightarrow {\bigcap}^{r+s}_{R}R^{r+s} = R \right).$$ The homomorphism $(R^{r+s})^{*} = (R^{r})^{*} \oplus (R^{s})^{*} \longrightarrow N^{*} \oplus (R^{s})^{*} = (N \oplus R^{s})^{*}$ induces a commutative diagram $$\xymatrix{
{\bigwedge}^{r}_{R}(R^{r})^{*} \otimes \bigwedge^{s}_{R}(R^{s})^{*} \ar[r] \ar[d]^-{=} &
{\bigwedge}^{r}_{R}(N)^{*} \otimes {\bigwedge}^{s}_{R}(R^{s})^{*} \ar@{^{(}->}[d]
\\
{\bigwedge}^{r+s}_{R}(R^{r+s})^{*} \ar[r] & {\bigwedge}^{r+s}_{R}(N \oplus R^{s})^{*}.
}$$ By taking $R$-duals to this commutative diagram, we obtain the following commutative diagram: $$\xymatrix{
{\bigcap}^{r+s}_{R | rm char}_{R}(M): = \operatorname{im}\left({\bigcap}^{r}_{R}N \longrightarrow { \bigcap}^{r}_{R}R^{r } = R \right).$$
\[rem: flat\ ] If $ R'$ is a flat $ R$-algebra, then we have $ { \rm char}_{R}(M)R' = { \rm char}_{R'}(R' \otimes_{R}M)$.
\[rem: indep - char\ ] The ideal $ { \rm char}_{R}(M)$ is independent of the option of the accurate succession $ 0 \longrightarrow N \longrightarrow R^{r } \longrightarrow M \longrightarrow 0$.
In fact, by Remark \[rem: flat\ ], we may assume that $ R$ is a local noetherian ring. Then any free resoluteness of $ M$ is isomorphic to the direct union of the minimal barren resolution of $ M$ and a fiddling complex. Hence it suffice to show that $ $ \operatorname{im}\left({\bigcap}^{r}_{R}N \longrightarrow { \bigcap}^{r}_{R}R^{r } = R \right) =
\operatorname{im}\left({\bigcap}^{r+s}_{R}(N \oplus R^{s }) \longrightarrow { \bigcap}^{r+s}_{R}R^{r+s } = R \right).$$ The homomorphism $ (R^{r+s})^ { * } = (R^{r})^ { * } \oplus (R^{s})^ { * } \longrightarrow N^ { * } \oplus (R^{s})^ { * } = (N \oplus R^{s})^{*}$ induces a commutative diagram $ $ \xymatrix {
{ \bigwedge}^{r}_{R}(R^{r})^ { * } \otimes \bigwedge^{s}_{R}(R^{s})^ { * } \ar[r ] \ar[d]^-{= } &
{ \bigwedge}^{r}_{R}(N)^ { * } \otimes { \bigwedge}^{s}_{R}(R^{s})^ { * } \ar@{^{(}->}[d ]
\\
{ \bigwedge}^{r+s}_{R}(R^{r+s})^ { * } \ar[r ] & { \bigwedge}^{r+s}_{R}(N \oplus R^{s})^ { * }.
} $ $ By lead $ R$-duals to this commutative diagram, we obtain the following commutative diagram: $ $ \xymatrix {
{ \bigcap}^{r+s}_{R | rm fhar}_{R}(M) := \operatorname{im}\lent({\bigcap}^{r}_{R}N \longtithtarrmw {\bigdap}^{r}_{R}R^{r} = R \right).$$
\[rem:flat\] If $R'$ is a flet $R$-qlgebea, then we have ${\rm chaf}_{R}(M)R' = {\rm bhar}_{R'}(R' \otumes_{C}M)$.
\[rem:indep-char\] Tis ideal ${\rm chzv}_{R}(M)$ iv independent on the choica of the exact sdqbence $0 \longrightarrow N \longrightarrjw R^{r} \lpnhrightarrow M \jonggidhtadgoc 0$.
In fact, by Remark \[rem:flat\], we mzy assuke that $R$ is s local noetherian ring. Thfn ajy free resolution of $M$ is isimor[yic to the dkrect sum of the minimzl free resolution of $M$ and a tfiviak complex. Yebce ht suffices to srow that $$\opevstornake{im}\lefy({\bigcap}^{r}_{R}N \lonnrighvarriw {\bigcap}^{r}_{R}R^{r} = R \righv) =
\operatorname{im}\lefj({\bigcap}^{r+s}_{R}(T \kplus R^{s}) \longrighraerow {\tigcdp}^{r+s}_{F}E^{r+s} = R \rmghf).$$ The jomkmorphism $(D^{r+s})^{*} = (R^{r})^{*} \oplys (R^{s})^{*} \longrightarrpw B^{*} \oplus (R^{s})^{*} = (N \oplus R^{f})^{*}$ induces a commutative diagram $$\xymatriq{
{\bigsedge}^{r}_{R}(R^{r})^{*} \otimes \bigweege^{s}_{R}(R^{s})^{*} \ar[r] \ar[d]^-{=} &
{\bigwgdge}^{r}_{R}(N)^{*} \otymes {\bigwedge}^{s}_{R}(R^{s})^{*} \ar@{^{(}->}[d]
\\
{\bigwedge}^{r+s}_{R}(R^{r+s})^{*} \ar[r] & {\bigfedge}^{c+s}_{F}(N \ipjjw G^{s})^{*}.
}$$ By taking $R$-duals to this commutative diwfrsm, we obtain the following volmijative diagram: $$\xymatxjx{
{\gigcap}^{r+s}_{R | rm char}_{R}(M) := \operatorname{im}\left({\bigcap}^{r}_{R}N \longrightarrow {\bigcap}^{r}_{R}R^{r} = \[rem:flat\] $R'$ is flat $R$-algebra, then {\rm \otimes_{R}M)$. \[rem:indep-char\] The ${\rm char}_{R}(M)$ is of the choice of the exact $0 \longrightarrow N \longrightarrow R^{r} \longrightarrow M \longrightarrow 0$. In fact, by Remark we may assume that $R$ is a local noetherian ring. Then any free of is to direct sum of the minimal free resolution of $M$ and a trivial complex. Hence it suffices show that $$\operatorname{im}\left({\bigcap}^{r}_{R}N \longrightarrow {\bigcap}^{r}_{R}R^{r} = R \right) \operatorname{im}\left({\bigcap}^{r+s}_{R}(N \oplus R^{s}) \longrightarrow = R \right).$$ The homomorphism = \oplus (R^{s})^{*} N^{*} (R^{s})^{*} (N \oplus R^{s})^{*}$ a commutative diagram $$\xymatrix{ {\bigwedge}^{r}_{R}(R^{r})^{*} \otimes \bigwedge^{s}_{R}(R^{s})^{*} \ar[r] \ar[d]^-{=} & {\bigwedge}^{r}_{R}(N)^{*} \otimes {\bigwedge}^{s}_{R}(R^{s})^{*} \ar@{^{(}->}[d] \\ {\bigwedge}^{r+s}_{R}(R^{r+s})^{*} \ar[r] {\bigwedge}^{r+s}_{R}(N \oplus By taking to commutative we obtain the diagram: $$\xymatrix{ {\bigcap}^{r+s}_{R | rm char}_{R}(M) := \operatorname{im}\lefT({\bigcap}^{r}_{R}N \LongrIghTarRoW {\bigCap}^{r}_{r}R^{r} = R \right).$$
\[rem:flAT\] If $R'$ Is a flat $R$-algebra, then we hAve ${\rm ChAR}_{R}(M)R' = {\RM cHar}_{R'}(R' \Otimes_{R}m)$.
\[ReM:INdeP-cHaR\] ThE iDEaL ${\rm chAr}_{R}(m)$ is indePendent of tHe cHoIce of the exacT SeQuence $0 \longRigHtarrow N \longRigHtarroW R^{R} \loNGrighTarRow M \lOngrigHTarrow 0$.
in fact, by REmARk \[rem:fLAt\], we may ASSuMe thAt $R$ is a local noetheRIaN Ring. Then any freE resolUtIOn OF $m$ is IsoMorphic to tHe DirecT Sum of thE MiNIMAl fREe resolution oF $M$ and a triviAL coMplex. HEnCe iT SufficEs to sHoW ThaT $$\operatornaMe{im}\Left({\bigcaP}^{r}_{R}N \loNGrightaRRow {\bigcAp}^{r}_{R}R^{r} = r \riGht) =
\OperAToRnAme{Im}\LEft({\BIgCap}^{R+S}_{R}(N \Oplus R^{s}) \lOnGrIghtaRrow {\BIGCAp}^{r+s}_{r}R^{r+S} = R \riGht).$$ ThE homomorphism $(r^{r+s})^{*} = (r^{r})^{*} \opLUs (R^{S})^{*} \longRightArroW N^{*} \Oplus (r^{s})^{*} = (N \oplUs R^{s})^{*}$ iNdUces a commutativE diaGram $$\xymatRix{
{\BiGweDgE}^{r}_{R}(R^{r})^{*} \OTimes \bIgwEdgE^{s}_{R}(R^{s})^{*} \ar[R] \ar[d]^-{=} &
{\bigWEdgE}^{r}_{r}(n)^{*} \OTiMes {\bigwedge}^{s}_{R}(R^{s})^{*} \ar@{^{(}->}[d]
\\
{\BiGWEdGe}^{r+s}_{R}(R^{r+s})^{*} \Ar[r] & {\bigWEdGe}^{R+S}_{R}(N \oplus r^{s})^{*}.
}$$ by tAkinG $r$-Duals To thIS cOmmutatiVe diagRAm, We Obtain tHe FollowInG coMmuTativE DiagRam: $$\xymAtrix{
{\bigCap}^{r+s}_{r | rm char}_{R}(M) := \operat orname{im} \left ({\ big ca p}^{ r}_{ R}N \longright a rrow {\bigcap}^{r}_{R}R^{r } = R \ r ight ) .$ $
\[ rem:fla t \] I f $ R' $isaf la t $R$ -al gebra,then we ha ve${ \rm char}_{R } (M )R' = {\rm ch ar}_{R'}(R'\ot imes_{ R} M)$ .
\[r em: indep -char\ ] The i deal ${\r mc har}_{ R }(M)$ i s in depe ndent of the choi c eo f the exact se quence $ 0 \ l o ngr igh tarrow N \ lo ngrig h tarrowR ^{ r } \lo n grightarrow M \longright a rro w 0$.
I n f a ct, by Rema rk \[r em:flat\],we m ay assume that$ R$ is a local n oether ian ri ng.T he nany f r eer es olu t ion of $M$is i somor phic t o thedir ectsum o f the minimal fr ee r e sol ution of $ M$ a nd a tr ivialcompl ex . Hence it suff ices to showtha t$$\ op erato r name{i m}\ lef t({\big cap}^{r } _{R }N \ l on grightarrow {\bigc ap } ^ {r }_{R}R^{ r} = R \r ig h t) =
\o pe rat orna m e {im}\ left ( {\ bigcap}^ {r+s}_ { R} (N \oplus R ^{s})\l ong rig htarr o w {\ bigcap }^{r+s}_ {R}R^ { r+s} = R \righ t ).$$ The homo m or p h is m $(R ^{r +s})^{*} =(R^{ r })^{ *} \ o pl us(R^{s })^{* }\ lo n grightarrow N^{*} \ op lus (R ^{s}) ^{*} = (N \op lus R^{s}) ^ { * }$ induc es a co m mutative diagr am $$ \xymatrix{ {\bigwed ge}^{ r}_{R}(R ^{r})^{*} \ otimes \bi gwe dge ^{s } _ {R }(R^{s})^{*}\ a r[r] \ ar[d]^- {=} &
{\b igw edg e}^ {r} _{ R}(N)^{*} \otimes {\ bi gw edg e}^{s } _{R}(R^{ s} )^{ *} \ ar@{^ { (}->}[ d]
\ \
{ \b ig w edg e}^{r+s } _{ R } (R^{ r+ s} )^{* } \a r[r]& {\ b igw edge}^{ r+s}_{R}( N \ o plus R ^{ s})^{*} .
}$$ By t ak ing $R$-du al s t o this c ommutati ve diagram, we obtain t h e follo win g com muta tive diag ram : $$\x yma t rix{
{ \bigca p}^{r +s }_{ R | rm char}_{R}(M)_:= \operatorname{im}\left({\bigcap}^{r}_{R}N_\longrightarrow {\bigcap}^{r}_{R}R^{r} = R_\right).$$
\[rem:flat\] If_$R'$_is a_flat_$R$-algebra, then we_have ${\rm char}_{R}(M)R'_= {\rm char}_{R'}(R' \otimes_{R}M)$.
\[rem:indep-char\]_The ideal ${\rm_char}_{R}(M)$_is independent of the choice of the exact sequence $0 \longrightarrow N \longrightarrow R^{r}_\longrightarrow_M \longrightarrow_0$.
In_fact,_by Remark \[rem:flat\], we may assume_that $R$ is a local_noetherian ring._Then any free resolution of $M$ is isomorphic_to_the direct sum_of the minimal free resolution of $M$ and a_trivial complex. Hence it suffices to_show that $$\operatorname{im}\left({\bigcap}^{r}_{R}N_\longrightarrow_{\bigcap}^{r}_{R}R^{r}_= R \right) =_
\operatorname{im}\left({\bigcap}^{r+s}_{R}(N \oplus R^{s}) \longrightarrow {\bigcap}^{r+s}_{R}R^{r+s} =_R \right).$$ The homomorphism $(R^{r+s})^{*} =_(R^{r})^{*} \oplus (R^{s})^{*} \longrightarrow N^{*} \oplus_(R^{s})^{*} = (N \oplus R^{s})^{*}$ induces_a commutative diagram $$\xymatrix{
{\bigwedge}^{r}_{R}(R^{r})^{*} \otimes_ \bigwedge^{s}_{R}(R^{s})^{*}_\ar[r] \ar[d]^-{=} &
{\bigwedge}^{r}_{R}(N)^{*} \otimes_ {\bigwedge}^{s}_{R}(R^{s})^{*} _\ar@{^{(}->}[d]
\\
_{\bigwedge}^{r+s}_{R}(R^{r+s})^{*} \ar[r]_& {\bigwedge}^{r+s}_{R}(N \oplus R^{s})^{*}. _
}$$ By taking_$R$-duals to this commutative diagram, we_obtain_the following commutative_diagram:_$$\xymatrix{
{\bigcap}^{r+s}_{R |
galaxy. Figure \[fg:findC\] bears this out. There is an optimal value of $C_W$ around 1.5, which reflects a balance between minimizing measurement errors as well as any measurement bias inherent in the technique.
With shapelets, we find a systematic underestimate of 11% in the first flexion, and an overestimate of 12% in the second flexion. We find a scatter of about 12% in both. This is very similar in magnitude to the results found by an “optimal” HOLICs analysis.
Correlation of HOLICs and Shapelets Measurement Error
-----------------------------------------------------
Since both HOLICs and shapelets give similar measurement errors at fixed sky noise, it is worth considering whether we expect measurement errors between the two techniques to be correlated. Even in these idealized circumstances, uncorrelated errors would mean that there is significant information in the images which is not being used. In Fig. \[fg:corrsim\], we show the correlation in uncertainty between our $C_W=1.5$ HOLICs estimates, and our shapelets estimates.
For the first flexion, in particular, the correlation is quite high, with a Pearson correlation coefficient of $0.86$. The correlation in measurements of the second flexion is much lower, with $\rho_G=0.23$. Why don’t they have perfect correlation? The two techniques weight various components of the signal (and thus, the noise) differently, and therefore have a slightly different response to the noise.
This general trend is borne out with observed objects as well, in which we will see much higher correlation between measurements of the first flexion than the second flexion between the two techniques.
PSF Deconvolution
-----------------
Finally, we can simulate PSF deconvolution. Using a Gaussian PSF with a characteristic size somewhat larger than the intrinsic image (the correction factor described in equation \[eq:fcorrect\] is 2.7), we distorted and then recovered the flexion estimates from images of increasing intrinsic ellipticity. This analysis is done in the absense of sky noise, and thus any errors in shape recovery represent a systematic effect. We show the fractional errors in measurement of the first and second flexion in Figure \[fg:psf\]. Since it is possible to estimate the systematic error for a combination of measured shear and PSF shape, it is advisable to those wishing to make high-precision flexion measurements to take this empirical correction into account. | galaxy. Figure \[fg: findC\ ] bears this out. There is an optimal value of $ C_W$ around 1.5, which reflect a libra between minimizing measurement errors as well as any measurement diagonal inherent in the technique.
With shapelets, we find a taxonomic underestimate of 11% in the beginning flexion, and an overestimate of 12% in the second flexure. We find a scatter of about 12% in both. This is very exchangeable in magnitude to the results establish by an “ optimal ” HOLICs analysis.
Correlation of HOLICs and Shapelets Measurement Error
-----------------------------------------------------
Since both HOLICs and shapelets give alike measurement errors at fixed sky noise, it is deserving considering whether we expect measurement error between the two techniques to be correlate. Even in these idealized circumstances, uncorrelated errors would mean that there is significant data in the images which is not being used. In Fig. \[fg: corrsim\ ], we show the correlation in uncertainty between our $ C_W=1.5 $ HOLICs estimates, and our shapelets estimate.
For the first flexion, in particular, the correlation is quite high, with a Pearson correlation coefficient of $ 0.86$. The correlation in measurements of the second flexion is much broken, with $ \rho_G=0.23$. Why do n’t they have perfect correlation coefficient? The two technique weight unit various components of the signal (and therefore, the noise) differently, and consequently have a slightly different reception to the noise.
This general trend is borne out with observed objects as well, in which we will see much higher correlation between measurement of the first inflection than the second flexure between the two techniques.
PSF Deconvolution
-----------------
Finally, we can simulate PSF deconvolution. Using a Gaussian PSF with a characteristic size slightly larger than the intrinsic image (the correction agent described in equation \[eq: fcorrect\ ] is 2.7), we distorted and then recovered the inflection estimates from image of increasing intrinsic ellipticity. This psychoanalysis is done in the absense of sky noise, and thus any errors in condition recovery represent a taxonomic effect. We show the fractional errors in measurement of the first and second flexion in Figure \[fg: psf\ ]. Since it is possible to estimate the systematic error for a combination of measured shear and PSF shape, it is advisable to those wishing to stool high - precision inflection measurements to take this empiric correction into account. | gapaxy. Figure \[fg:findC\] bears this out. There is an mptimam value uf $C_W$ around 1.5, which reflects a balabce between minimizing measuremvnt errorw as qell as anb measurcient nias nnierent in the tgchnique.
With shapelets, we xivd a systematic underestimate of 11% in ehe firxt flexion, and ag ovtreftimzne of 12% in the second flexion. We find a scatter of anout 12% in both. This is very simllar in magnitude ho the resuots siund by an “ootimal” HOLPEs analysis.
Dorrelation of HOLICs and Shapeuets Keasuremenj Srgmr
-----------------------------------------------------
Since boti HOLIBs and shapelcns give similat measurement crrorv ar fixed sky noise, it ms worth considering whether fe expect measuremebt errots bedweev thd tso tschniqkes to be cordelated. Eveb in these idealizec sptcumstances, hncorrqlwted errors would mean that there is sicniricant information in tye images which is noj being usqd. In Fig. \[fg:corrsim\], we show the correlation in uncartaiith bttqeen uyr $C_W=1.5$ HOLICs estimates, and our shapelets estimaess.
Gog the first flexijn, in partivupat, the correlatkon is qujte high, with a Pewrson cjrrelqtion coesficoent of $0.86$. The correlation in measurementf of the second flerion is much lowgr, witn $\rho_G=0.23$. Why don’t they hare perrect correlwtion? The gwo techniques wdigmt earious components of the fignal (anv thux, the nuise) diffewently, and thercxore have a slightpy diyferett responsf to the noise.
This general trenv is borne ouj whth observeb objegts as well, in rhich we will xee muck highdr correlanion betwxen measuremqnts of the fhtst flexion tian the sqcone flwxion bdgween the two yechniques.
PSF Deconcolution
-----------------
Finally, we cav simulate PSF dzeobvolution. Using a Eaufspan PFX with a chasactdriryic skze somtchab lxrget than the intrinsic imafe (the correction gagtor desctibed in qquation \[eq:fcotrect\] is 2.7), we distogted end thxn recpveted the flexion estimates from jmages of ingreasing intrynsig eljipticity. Chis analysis is done in the absense of wky noise, and thus any errors in shapg rtcovery reprexent w systemadic effect. We show tye fractional errprs in measurement of the fhrst wnd second flexion in Figure \[fg:psf\]. Since it is possible to estimate the sywtematmc error for z cokbinaniou oy measuwed diear and PSF shapv, it is advisable to those wishiig to make hngh-precision flexion measuremrngs to take thks empirical correction jnto accpunt. | galaxy. Figure \[fg:findC\] bears this out. There optimal of $C_W$ 1.5, which reflects errors well as any bias inherent in technique. With shapelets, we find a underestimate of 11% in the first flexion, and an overestimate of 12% in second flexion. We find a scatter of about 12% in both. This is similar magnitude the found by an “optimal” HOLICs analysis. Correlation of HOLICs and Shapelets Measurement Error ----------------------------------------------------- Since both and shapelets give similar measurement errors at fixed noise, it is worth whether we expect measurement errors the techniques to correlated. in idealized circumstances, uncorrelated would mean that there is significant information in the images which is not being used. In Fig. we show in uncertainty our HOLICs and our shapelets the first flexion, in particular, the high, with a Pearson correlation coefficient of $0.86$. correlation in of the second flexion is much with $\rho_G=0.23$. Why don’t they have perfect correlation? two techniques weight various components of the signal (and thus, the noise) differently, and therefore slightly different response to noise. This general is out observed as well, which we will see much higher correlation between measurements of the flexion than the second flexion between the two techniques. PSF Finally, can simulate PSF Using a Gaussian PSF a size somewhat larger than image correction equation is we distorted and then the flexion estimates from images increasing intrinsic ellipticity. This absense of sky noise, and thus any errors shape recovery represent a systematic effect. We the fractional errors in measurement of the first and second flexion in \[fg:psf\]. Since possible to estimate the systematic error for a of measured shear and shape, it is advisable to those wishing to make flexion to take empirical correction into | galaxy. Figure \[fg:findC\] bears tHis out. TherE is an OptImaL vAlue Of $C_W$ Around 1.5, which refLEcts A balance between minimizIng meAsURemeNT eRrors As well aS AnY MEasUrEmEnt BiAS iNhereNt iN the tecHnique.
With ShaPeLets, we find a sYStEmatic undeResTimate of 11% in thE fiRst fleXiOn, aND an ovEreStimaTe of 12% in THe secoNd flexion. we FInd a scATter of aBOUt 12% In boTh. This is very similAR iN Magnitude to the ResultS fOUnD BY an “OptImal” HOLICs AnAlysiS.
correlaTIoN OF hOLics and ShapeletS MeasuremenT errOr
-----------------------------------------------------
SincE bOth hoLICs aNd shaPeLEts Give similar MeasUrement erRors at FIxed sky NOise, it iS worth ConSidErinG WhEtHer We EXpeCT mEasURemEnt errorS bEtWeen tHe twO TECHniqUes To be CorreLated. Even in thEse IdeaLIzeD circUmstaNces, UnCorreLated eRrors WoUld mean that therE is sIgnificanT inFoRmaTiOn in tHE imageS whIch Is not beIng used. iN FiG. \[fG:CORrSim\], we show the correlAtION iN uncertaInty beTWeEn OUr $C_W=1.5$ HOLIcs EstImatES, And ouR shaPElEts estimAtes.
FoR ThE fIrst fleXiOn, in paRtIcuLar, The coRRelaTion is Quite higH, with A pearson correlaTIon coefficienT Of $0.86$. tHE cORrelAtiOn in measureMentS Of thE secONd FleXIon is Much lOwER, wITh $\rho_G=0.23$. Why don’t they haVe PerfecT corrElation? The two Techniques WEIGht varioUs coMPoNEnts of the signaL (and tHus, the noisE) DifferenTly, anD therefoRe have a slIGHtly diffEreNt rEspOnsE TO tHe noise.
This geNERal tReNd is borNe oUt with oBseRveD obJecTs As well, in wHich we wiLl SeE mUcH hiGher cORrelatioN bEtwEeN meAsureMEnts of The fiRst fLeXiON thAn the seCOnD FLexiOn BeTweeN thE tWo tecHniqUEs.
PsF DeconVolution
-----------------
FInaLLy, we CaN sImulate pSF deconvolutIoN. Using a GauSsIan pSF witH A CharacteRistic size somewhat largeR Than the IntRinsiC imaGe (the corrEctIon facTor DEscribEd in eqUatioN \[eQ:fcORRect\] iS 2.7), WE dIstOrTed and then RECovEred tHe FlexIon estiMates from images of iNCreAsing intrinsiC elLiptICItY. ThIS aNAlySiS Is dONE in the absense of Sky noise, anD tHUs Any errors iN ShaPe RecoverY represEnt a sYStematiC effect. We Show the frAcTionAL ErrOrs in measuRement of The first aND secoND fLexioN in figure \[Fg:Psf\]. since It is poSSibLe to eStimatE tHe systEmatiC eRror for a Combination of measured shEar and pSF shApe, It is advisAblE To tHose wishiNg to Make high-prEciSioN flexIon MEasurEmenTS tO taKE this EmpiRIcal correCTiOn iNTO aCcount. | galaxy. Figure \[fg:findC \] bears t his o ut. Th er e is anoptimal valueo f $C _W$ around 1.5, whichrefle ct s a b a la nce b etweenm in i m izi ng m eas ur e me nt er ror s as we ll as anymea su rement biasi nh erent in t hetechnique.
Wit h shap el ets , we f ind a sy stemat i c unde restimate o f 11% i n the fi r s tflex ion, and an overe s ti m ate of 12% inthe se co n df l exi on. We find a s catte r of abo u t1 2 % in both. This is very simil a r i n magn it ude to the resu lt s fo und by an “ opti mal” HOLI Cs ana l ysis.
C orrelat ion of HO LIC s an d S ha pel et s Me a su rem e ntError
-- -- -- ----- ---- - - - - ---- --- ---- ----- ------------- --- --
S inc e bot h HOL ICsan d sha pelets give s imilar measurem enterrors at fi xe d s ky nois e , it i s w ort h consi deringw het he r w eexpect measurement e r r or s betwee n thet wo t e chniques t o b e co r r elate d. E v en in thes e idea l iz ed circum st ances, u nco rre lated erro rs wou ld meanthatt here is signif i cant informat i on i nt he i mag es which is not bein g us e d. In Fig.\[fg: co r rs i m\], we show the co rr elatio n inuncertainty b etween our $ C _W=1.5$HOLI C se stimates, andour s hapelets e s timates.
For the fir st flexio n , in part icu lar , t hec o rr elation is qu i t e hi gh , witha P earsoncor rel ati onco efficient of $0.8 6$ .Th ecor relat i on in me as ure me nts of t h e seco nd fl exio nis muc h lower , w i t h $\ rh o_ G=0. 23$ .Why d on’t the y haveperfect c orr e lati on ?The two techniques w ei ght variou scom ponent s of the s ignal (and thus, the no i se) dif fer ently , an d therefo rehave a sl i ghtlydiffer ent r es pon s e to t h e n ois e.
This gen e r altrend i s bo rne out with observed obj e cts as well, inwhi ch w e wi lls ee muc hh igh e r correlation be tween meas ur e me nts of the fir st flexio n thanthe s e cond fl exion bet ween thetw o te c h niq ues.
PSFDeconvol ution
--- - ----- - -- -----
F inally ,wecan s imulat e PS F dec onvolu ti on. Us ing a G aussianPSF with a characterist ic siz e som ewh at larger th a n t he intrin sicimage (the co rre ction fa c tor d escr i be d i n equa tion \[eq:fcor r ec t\] i s2.7), we di s t o rte d and th e n reco vere d the flexion est i mates from ima geso f in cre a sing i ntrinsic ellip tic it y . This an al ysis is don e in the a b sense of sk y nois e, andt h us any er rors in shape re cov er y repres en ta syste mati ceffect . We s h ow t h e fractional erro rs in m easur e men t ofth e first andsecond fle xion in Fig ure \[ fg:p sf\]. Sinceit is po ssi bl e to estim a te the sy stema tic err or for acombin atio n of me asur ed sh ear and P S F s h ap e, itis a dvisa bl e to those wi s hing tomak e high-p re cis i o n flex i on m easurement s t o tak e this empir i calc or r ectio n into accou nt. | galaxy._Figure \[fg:findC\] bears_this out. There is_an optimal_value_of $C_W$_around_1.5, which reflects_a balance between_minimizing measurement errors as_well as any_measurement_bias inherent in the technique.
With shapelets, we find a systematic underestimate of 11% in_the_first flexion,_and_an_overestimate of 12% in the_second flexion. We find a_scatter of_about 12% in both. This is very similar_in_magnitude to the_results found by an “optimal” HOLICs analysis.
Correlation of HOLICs_and Shapelets Measurement Error
-----------------------------------------------------
Since both HOLICs_and shapelets give_similar_measurement_errors at fixed sky_noise, it is worth considering whether_we expect measurement errors between the_two techniques to be correlated. Even in_these idealized circumstances, uncorrelated errors would_mean that there is significant_information in_the images which is not_being used. In_Fig. \[fg:corrsim\], we_show the correlation_in uncertainty between our $C_W=1.5$ HOLICs_estimates, and our_shapelets estimates.
For the first flexion, in_particular,_the correlation is_quite_high,_with a_Pearson correlation coefficient_of_$0.86$. The_correlation_in measurements of the second flexion_is_much lower, with $\rho_G=0.23$. Why don’t they_have perfect correlation? The_two_techniques weight various components_of the signal (and thus,_the noise) differently, and therefore have_a slightly_different response_to the noise.
This general trend is borne out with observed objects_as well, in which we will_see much higher correlation_between measurements_of_the first flexion_than_the second_flexion between the two techniques.
PSF Deconvolution
-----------------
Finally, we_can simulate_PSF deconvolution. Using a Gaussian PSF_with a characteristic size_somewhat_larger than the intrinsic image (the_correction factor described in equation \[eq:fcorrect\] is_2.7), we distorted and then_recovered_the_flexion estimates from images of_increasing intrinsic ellipticity. This analysis is_done in the_absense of sky noise, and thus any_errors_in shape recovery represent a systematic_effect._We show the fractional errors in_measurement_of_the first and second flexion_in Figure \[fg:psf\]. Since it is possible_to estimate the systematic error for a combination of_measured shear and_PSF shape, it is advisable_to_those_wishing to make high-precision flexion measurements to take this empirical_correction into_account. |
ode`~=`,
\lowercase{\endgroup\def~}{\pFcomma\mkern\pFqskip} \mathcode`,=\string"8000
{}_{}F_{1}\biggl(\genfrac..{0pt}{}{-2a; 3a+1. 1/2}{a+3/2};\frac{16}{25}, \frac{4}{5}\biggr) \endgroup
}.\end{gathered}$$ From this relation, we can conjecture $$\begin{aligned}
\begin{split}
F(a)&={ \begingroup
\begingroup\lccode`~=`,
\lowercase{\endgroup\def~}{\pFcomma\mkern\pFqskip} \mathcode`,=\string"8000
{}_{}F_{1}\biggl(\genfrac..{0pt}{}{-2a; 3a+1. 1/2}{a+3/2};\frac{16}{25}, \frac{4}{5}\biggr) \endgroup
}\\
&=\left(\frac{5}{3^3}\right)^{a}
\frac{\cos (\pi a)\, {{\Gamma}}(5/6){{\Gamma}}(7/6){{\Gamma}}(a+1/2){{\Gamma}}(a+3/2)}
{{{\Gamma}}(1/2){{\Gamma}}(3/2){{\Gamma}}(a+5/6){{\Gamma}}(a+7/6)}\\
&=\frac{2}{3}\left(\frac{5}{3^3}\right)^{a}
\frac{\cos(\pi a)\,{{\Gamma}}(a+1/2){{\Gamma}}(a+3/2)}{{{\Gamma}}(a+5/6){{\Gamma}}(a+7/6)}.
\end{split}
\label{conj2}\end{aligned}$$ Unfortunately, however, it seems difficult to compute the asymptotic behavior of $F(a)$ in the limit $|a|\rightarrow +\infty$ from a direct application of Laplace’s method. For this reason, we have not been able to prove (\[conj2\]). However, with numerical calculations, we have obtained results consistent with this identity. If indeed this identity does hold, then we have $$\begin{gathered}
{ \begingroup
\begingroup\lcc | ode`~= `,
\lowercase{\endgroup\def~}{\pFcomma\mkern\pFqskip } \mathcode`,=\string"8000
{ } _ { } F_{1}\biggl(\genfrac.. {0pt}{}{-2a; 3a+1. 1/2}{a+3/2};\frac{16}{25 }, \frac{4}{5}\biggr) \endgroup
} .\end{gathered}$$ From this relation, we can conjecture $ $ \begin{aligned }
\begin{split }
F(a)&= { \begingroup
\begingroup\lccode`~= `,
\lowercase{\endgroup\def~}{\pFcomma\mkern\pFqskip } \mathcode`,=\string"8000
{ } _ { } F_{1}\biggl(\genfrac.. {0pt}{}{-2a; 3a+1. 1/2}{a+3/2};\frac{16}{25 }, \frac{4}{5}\biggr) \endgroup
} \\
& = \left(\frac{5}{3 ^ 3}\right)^{a }
\frac{\cos (\pi a)\, { { \Gamma}}(5/6){{\Gamma}}(7/6){{\Gamma}}(a+1/2){{\Gamma}}(a+3/2) }
{ { { \Gamma}}(1/2){{\Gamma}}(3/2){{\Gamma}}(a+5/6){{\Gamma}}(a+7/6)}\\
& = \frac{2}{3}\left(\frac{5}{3 ^ 3}\right)^{a }
\frac{\cos(\pi a)\,{{\Gamma}}(a+1/2){{\Gamma}}(a+3/2)}{{{\Gamma}}(a+5/6){{\Gamma}}(a+7/6) }.
\end{split }
\label{conj2}\end{aligned}$$ Unfortunately, however, it seems difficult to calculate the asymptotic demeanor of $ F(a)$ in the terminus ad quem $ |a|\rightarrow + \infty$ from a direct application of Laplace ’s method. For this cause, we have not been able to prove (\[conj2\ ]). However, with numerical calculation, we have obtained results reproducible with this identity. If indeed this identity does hold, then we have $ $ \begin{gathered }
{ \begingroup
\begingroup\lcc | ode`~=`,
\lowercase{\endgroup\def~}{\kFcomma\mkern\pFqsknp} \maticode`,=\stding"8000
{}_{}F_{1}\bkggl(\genfrac..{0pt}{}{-2a; 3a+1. 1/2}{a+3/2};\frac{16}{25}, \frac{4}{5}\bmggr) \endtroup
}.\end{gathered}$$ From ghis relanion, we cqn cibjecture $$\bxfin{aligkzd}
\begjk{splic}
F(e)&={ \begingroup
\negingroup\lwcode`~=`,
\lowerware{\zndgroup\def~}{\pFcomma\mkern\pFqskip} \mathsode`,=\stronh"8000
{}_{}F_{1}\biggl(\genfrwc..{0pt}{}{-2s; 3a+1. 1/2}{a+3/2};\rgag{16}{25}, \frac{4}{5}\biggr) \endgroup
}\\
&=\left(\frac{5}{3^3}\riggt)^{a}
\frac{\bos (\pi a)\, {{\Gamma}}(5/6){{\Gamka}}(7/6){{\Gamma}}(a+1/2){{\Gamma}}(a+3/2)}
{{{\Gamma}}(1/2){{\Gamma}}(3/2){{\Galma}}(a+5/6){{\Hamma}}(a+7/6)}\\
&=\frac{2}{3}\left(\frac{5}{3^3}\gight)^{a}
\frac{\cis(\pi q)\,{{\Gamma}}(a+1/2){{\Gamma}}(x+3/2)}{{{\Gamma}}(a+5/6){{\Gamma}}(a+7/6)}.
\end{split}
\lzbel{conj2}\end{aligned}$$ Unfortunatelh, howzver, it seens divxicult to cimputv the asymptobpc behaeior of $F(a)$ in the limlt $|a|\rmghtqrrow +\infty$ from a dicect application of Japlace’s kechod. For this reason, qe have not beev abue uo 'robe (\[conu2\]). Hkwever, witg numerical calculations, we hafe ibtained resumts cogsystent with this identity. If indeed thiv isentity does hold, then qe have $$\begin{gathered}
{ \begingrjup
\begingroup\lcc | ode`~=`, \lowercase{\endgroup\def~}{\pFcomma\mkern\pFqskip} \mathcode`,=\string"8000 {}_{}F_{1}\biggl(\genfrac..{0pt}{}{-2a; 3a+1. 1/2}{a+3/2};\frac{16}{25}, \frac{4}{5}\biggr) From relation, we conjecture $$\begin{aligned} \begin{split} {}_{}F_{1}\biggl(\genfrac..{0pt}{}{-2a; 1/2}{a+3/2};\frac{16}{25}, \frac{4}{5}\biggr) \endgroup &=\left(\frac{5}{3^3}\right)^{a} \frac{\cos (\pi {{\Gamma}}(5/6){{\Gamma}}(7/6){{\Gamma}}(a+1/2){{\Gamma}}(a+3/2)} {{{\Gamma}}(1/2){{\Gamma}}(3/2){{\Gamma}}(a+5/6){{\Gamma}}(a+7/6)}\\ &=\frac{2}{3}\left(\frac{5}{3^3}\right)^{a} \frac{\cos(\pi a)\,{{\Gamma}}(a+1/2){{\Gamma}}(a+3/2)}{{{\Gamma}}(a+5/6){{\Gamma}}(a+7/6)}. \end{split} Unfortunately, however, it seems difficult to compute the asymptotic behavior of $F(a)$ in limit $|a|\rightarrow +\infty$ from a direct application of Laplace’s method. For this reason, have been to (\[conj2\]). However, with numerical calculations, we have obtained results consistent with this identity. If indeed this does hold, then we have $$\begin{gathered} { \begingroup | ode`~=`,
\lowercase{\endgroup\def~}{\pFComma\mkern\PFqskIp} \mAthCoDe`,=\stRing"8000
{}_{}f_{1}\biggl(\genfrac..{0pT}{}{-2A; 3a+1. 1/2}{a+3/2};\fRac{16}{25}, \frac{4}{5}\biggr) \endgroup
}.\enD{gathErED}$$ FroM ThIs relAtion, we CAn CONjeCtUrE $$\beGiN{AlIgned}
\BegIn{split}
f(a)&={ \begingroUp
\bEgIngroup\lccodE`~=`,
\LoWercase{\endGroUp\def~}{\pFcomma\MkeRn\pFqsKiP} \maTHcode`,=\StrIng"8000
{}_{}F_{1}\bIggl(\geNFrac..{0pt}{}{-2A; 3a+1. 1/2}{a+3/2};\frac{16}{25}, \frAc{4}{5}\BIggr) \enDGroup
}\\
&=\leFT(\FrAc{5}{3^3}\riGht)^{a}
\frac{\cos (\pi a)\, {{\GamMA}}(5/6){{\GAMma}}(7/6){{\Gamma}}(a+1/2){{\Gamma}}(A+3/2)}
{{{\Gamma}}(1/2){{\gaMMa}}(3/2){{\gAMma}}(A+5/6){{\GaMma}}(a+7/6)}\\
&=\frac{2}{3}\leFt(\Frac{5}{3^3}\rIGht)^{a}
\fraC{\CoS(\PI A)\,{{\GaMMa}}(a+1/2){{\Gamma}}(a+3/2)}{{{\GammA}}(a+5/6){{\Gamma}}(a+7/6)}.
\end{SPliT}
\label{CoNj2}\eND{alignEd}$$ UnfOrTUnaTely, however, It seEms difficUlt to cOMpute thE AsymptoTic behAviOr oF $F(a)$ iN ThE lImiT $|a|\RIghTArRow +\INftY$ from a diReCt AppliCatiON OF laplAce’S metHod. FoR this reason, we HavE not BEen Able tO provE (\[conJ2\]). HOweveR, with nUmeriCaL calculations, we Have Obtained rEsuLtS coNsIstenT With thIs iDenTity. If iNdeed thIS idEnTITY dOes hold, then we have $$\bEgIN{GaThered}
{ \beGingroUP
\bEgINgroup\lcC | ode`~=`,
\lowercase{\e ndgroup\de f~}{\ pFc omm a\ mker n\pF qskip} \mathc o de`, =\string"8000
{}_{}F _{1}\ bi g gl(\ g en frac. .{0pt}{ } {- 2 a ; 3 a+ 1. 1/ 2} { a+ 3/2}; \fr ac{16}{ 25}, \frac {4} {5 }\biggr) \e n dg roup
}.\en d{g athered}$$ F rom thisre lat i on, w e c an co njectu r e $$\b egin{alig ne d }
\beg i n{split } F( a)&= { \begingroup
\ be g ingroup\lccode `~=`, \ l o wer cas e{\endgrou p\ def~} { \pFcomm a \m k e r n\p F qskip} \math code`,=\str i ng" 8000
{ }_{ } F_{1}\ biggl (\ g enf rac..{0pt}{ }{-2 a; 3a+1.1/2}{a + 3/2};\f r ac{16}{ 25}, \ fra c{4 }{5} \ bi gg r) \ e ndg r ou p
} \ \
& =\left(\ fr ac {5}{3 ^3}\ r i g h t)^{ a}\fra c{\co s (\pi a)\, { {\G amma } }(5 /6){{ \Gamm a}}( 7/ 6){{\ Gamma} }(a+1 /2 ){{\Gamma}}(a+3 /2)}
{{{\Gamm a}} (1 /2) {{ \Gamm a }}(3/2 ){{ \Ga mma}}(a +5/6){{ \ Gam ma } } ( a+ 7/6)}\\
&=\frac{2} {3 } \ le ft(\frac {5}{3^ 3 }\ ri g ht)^{a}\f rac {\co s ( \pi a )\,{ { \G amma}}(a +1/2){ { \G am ma}}(a+ 3/ 2)}{{{ \G amm a}} (a+5/ 6 ){{\ Gamma} }(a+7/6) }.
\e n d{split}
\labe l {conj2}\end{a l ig n e d} $ $ Un for tunately, h owev e r, i t se e ms di f ficul t toco m pu t e the asymptotic be ha vior o f $F( a)$ in the li mit $|a|\r i g h tarrow + \inf t y$ from a directappli cation ofL aplace’s meth od. Forthis reas o n , we hav e n otbee n a b l eto prove (\[c o n j2\] ). Howeve r,with nu mer ica l c alc ul ations, w e have o bt ai ne dres ultsc onsisten twit hthi s ide n tity.If in deed t hi s id entityd oe s hold ,th en w e h av e $$\ begi n {ga thered}
{ \begi ngr o up
\ be gingrou p\lcc | ode`~=`,
_ _\lowercase{\endgroup\def~}{\pFcomma\mkern\pFqskip} \mathcode`,=\string"8000
_{}_{}F_{1}\biggl(\genfrac..{0pt}{}{-2a; 3a+1._1/2}{a+3/2};\frac{16}{25},_\frac{4}{5}\biggr) _\endgroup
}.\end{gathered}$$_From this relation,_we can conjecture_$$\begin{aligned}
\begin{split}
F(a)&={ \begingroup
_\begingroup\lccode`~=`,
__\lowercase{\endgroup\def~}{\pFcomma\mkern\pFqskip} \mathcode`,=\string"8000
{}_{}F_{1}\biggl(\genfrac..{0pt}{}{-2a; 3a+1. 1/2}{a+3/2};\frac{16}{25}, \frac{4}{5}\biggr) \endgroup
}\\
&=\left(\frac{5}{3^3}\right)^{a}
\frac{\cos (\pi a)\, {{\Gamma}}(5/6){{\Gamma}}(7/6){{\Gamma}}(a+1/2){{\Gamma}}(a+3/2)}
{{{\Gamma}}(1/2){{\Gamma}}(3/2){{\Gamma}}(a+5/6){{\Gamma}}(a+7/6)}\\
&=\frac{2}{3}\left(\frac{5}{3^3}\right)^{a}
\frac{\cos(\pi a)\,{{\Gamma}}(a+1/2){{\Gamma}}(a+3/2)}{{{\Gamma}}(a+5/6){{\Gamma}}(a+7/6)}.
\end{split}
\label{conj2}\end{aligned}$$_Unfortunately,_however, it_seems_difficult_to compute the asymptotic behavior_of $F(a)$ in the limit_$|a|\rightarrow +\infty$_from a direct application of Laplace’s method. For_this_reason, we have_not been able to prove (\[conj2\]). However, with numerical_calculations, we have obtained results consistent_with this identity._If_indeed_this identity does hold,_then we have $$\begin{gathered}
{ \begingroup
_ \begingroup\lcc |
{M}_0|^2=|\mathcal{M}_3|^2$. Because of this, the unphysical degrees of freedom cancel each other and what remains are terms which involve the physical polarization vectors ($\lambda=1$, 2). Since the latter can be chosen as $(\varepsilon_1^{\mu})=(0,1,0,0)$ and $(\varepsilon_2^{\mu})=(0,0,1,0)$, we obtain $$|\mathcal{M}|^2\Big|^{\mathrm{phys}}_{\mathrm{QED}}=\sum_{\lambda=1,2} |\varepsilon_{\mu}^{(\lambda)}(k)\mathcal{M}^{\mu}(k)|^2=|\mathcal{M}_1|^2+|\mathcal{M}_2|^2=\sum_{\lambda=1,2} |\mathcal{M}_{\lambda}|^2\,,$$ where ‘phys’ means that the Ward identity has been used.
In order to understand the limits of the polarization tensors from Eq. we will perform a similar analysis in the context of the modified theory. For $(k^{\mu})=(|\mathbf{k}|,\mathbf{k})$ with $\mathbf{k}=(k_{\bot},0,k_{\parallel})$ the Ward identity reads $$\label{eq:Ward-identity}
k^{\mu}\mathcal{M}_{\mu}=k_0\mathcal{M}_0-k_1\mathcal{M}_1-k_3\mathcal{M}_3=|\mathbf{k}|\mathcal{M}_0+k_{\bot}\mathcal{M}_1+k_{\parallel}\mathcal{M}_3=0\,,$$ and therefore, $\mathcal{M}_1$ can be expressed as follows:
$$\label{eq:matrix-element-m1}
\mathcal{M}_1=-\frac{|\mathbf{k}|}{k_{\bot}}\mathcal{M}_0-\frac{k_{\parallel}}{k_{\bot}}\mathcal{M}_3\,,$$
$$\label{eq:matrix-element-m1-square}
|\mathcal{M}_1|^2=\frac{|\mathbf{k}|^2}{k_{\bot}^2}|\mathcal{M}_0|^2+\frac{k_{\scriptscriptstyle{\parallel}}^2}{k_{\bot}^2}|\mathcal{M}_3|^2+\frac{2|\mathbf{k}|k_{\parallel}}{k_{\bot}^2}\mathrm{Re}(\mathcal{M}_0\mathcal{M}_ | { M}_0|^2=|\mathcal{M}_3|^2$. Because of this, the unphysical degrees of freedom cancel each other and what remain are term which involve the forcible polarization vectors ($ \lambda=1 $, 2). Since the latter can be choose as $ (\varepsilon_1^{\mu})=(0,1,0,0)$ and $ (\varepsilon_2^{\mu})=(0,0,1,0)$, we obtain $ $ |\mathcal{M}|^2\Big|^{\mathrm{phys}}_{\mathrm{QED}}=\sum_{\lambda=1,2 } |\varepsilon_{\mu}^{(\lambda)}(k)\mathcal{M}^{\mu}(k)|^2=|\mathcal{M}_1|^2+|\mathcal{M}_2|^2=\sum_{\lambda=1,2 } |\mathcal{M}_{\lambda}|^2\,,$$ where ‘ phys ’ mean that the Ward identity has been used.
In order to sympathize the limit of the polarization tensors from Eq. we will do a similar analysis in the context of the modified theory. For $ (k^{\mu})=(|\mathbf{k}|,\mathbf{k})$ with $ \mathbf{k}=(k_{\bot},0,k_{\parallel})$ the Ward identity learn $ $ \label{eq: Ward - identity }
k^{\mu}\mathcal{M}_{\mu}=k_0\mathcal{M}_0 - k_1\mathcal{M}_1 - k_3\mathcal{M}_3=|\mathbf{k}|\mathcal{M}_0+k_{\bot}\mathcal{M}_1+k_{\parallel}\mathcal{M}_3=0\,,$$ and therefore, $ \mathcal{M}_1 $ can be expressed as follows:
$ $ \label{eq: matrix - element - m1 }
\mathcal{M}_1=-\frac{|\mathbf{k}|}{k_{\bot}}\mathcal{M}_0-\frac{k_{\parallel}}{k_{\bot}}\mathcal{M}_3\,,$$
$ $ \label{eq: matrix - component - m1 - square }
|\mathcal{M}_1|^2=\frac{|\mathbf{k}|^2}{k_{\bot}^2}|\mathcal{M}_0|^2+\frac{k_{\scriptscriptstyle{\parallel}}^2}{k_{\bot}^2}|\mathcal{M}_3|^2+\frac{2|\mathbf{k}|k_{\parallel}}{k_{\bot}^2}\mathrm{Re}(\mathcal{M}_0\mathcal{M } _ | {M}_0|^2=|\mahhcal{M}_3|^2$. Because of this, tme unphysical deyeees oh freedkm canceu each other and what remaind qre ttgms which involve the physical polarizqtioi vectors ($\lambda=1$, 2). Since bke lafber ccn be chosen as $(\yarepsilon_1^{\mg})=(0,1,0,0)$ and $(\varepsilmn_2^{\ou})=(0,0,1,0)$, we obtain $$|\mathcal{M}|^2\Big|^{\mathrm{phys}}_{\matrrm{QED}}=\sim_{\pambda=1,2} |\varepsijon_{\ml}^{(\lwmbdz)}(k)\mathcal{M}^{\mu}(k)|^2=|\mathcal{M}_1|^2+|\mathcal{M}_2|^2=\sum_{\lajbda=1,2} |\mauhcal{M}_{\lambda}|^2\,,$$ where ‘phys’ means that the Ward ldenhity has been used.
Ln order to undqestand the lkmits of the polarizatjon tensors from Eq. we will perworm c similar abaoyslv in the coitext jf the modiflvd theosy. For $(l^{\mu})=(|\mathbf{k}|,\mathnf{k})$ wmth $\nathbf{k}=(k_{\bot},0,k_{\parallel})$ vhe Ward identity rewds $$\label{aq:Card-identity}
k^{\mu}\mathcao{M}_{\nu}=k_0\majhcal{K}_0-k_1\magycau{M}_1-k_3\javhczl{M}_3=|\matjbf{i}|\mathcal{M}_0+k_{\got}\mathcal{M}_1+j_{\parallel}\mathcal{M}_3=0\,,$$ amd nnerefore, $\matgcal{M}_1$ sag be expressed as follows:
$$\label{eq:matrix-enemsnt-m1}
\mathcal{M}_1=-\frac{|\mathbf{k}|}{j_{\bot}}\mathcal{M}_0-\frac{k_{\paraplel}}{k_{\bot}}\mwthcal{M}_3\,,$$
$$\label{eq:matrix-element-m1-square}
|\mathcal{M}_1|^2=\frac{|\matvbf{k}|^2}{k_{\uog}^2}|\mauhgwu{N}_0|^2+\fgac{k_{\scriptscriptstyle{\parallel}}^2}{k_{\bot}^2}|\mathcal{M}_3|^2+\frac{2|\iztnbn{k}|k_{\parallel}}{k_{\bot}^2}\mcthrm{Re}(\mathcal{M}_0\mstjcsj{M}_ | {M}_0|^2=|\mathcal{M}_3|^2$. Because of this, the unphysical degrees cancel other and remains are terms vectors 2). Since the can be chosen $(\varepsilon_1^{\mu})=(0,1,0,0)$ and $(\varepsilon_2^{\mu})=(0,0,1,0)$, we obtain $$|\mathcal{M}|^2\Big|^{\mathrm{phys}}_{\mathrm{QED}}=\sum_{\lambda=1,2} |\mathcal{M}_{\lambda}|^2\,,$$ where ‘phys’ means that the Ward identity has been used. In order understand the limits of the polarization tensors from Eq. we will perform a analysis the of modified theory. For $(k^{\mu})=(|\mathbf{k}|,\mathbf{k})$ with $\mathbf{k}=(k_{\bot},0,k_{\parallel})$ the Ward identity reads $$\label{eq:Ward-identity} k^{\mu}\mathcal{M}_{\mu}=k_0\mathcal{M}_0-k_1\mathcal{M}_1-k_3\mathcal{M}_3=|\mathbf{k}|\mathcal{M}_0+k_{\bot}\mathcal{M}_1+k_{\parallel}\mathcal{M}_3=0\,,$$ and therefore, $\mathcal{M}_1$ can expressed as follows: $$\label{eq:matrix-element-m1} \mathcal{M}_1=-\frac{|\mathbf{k}|}{k_{\bot}}\mathcal{M}_0-\frac{k_{\parallel}}{k_{\bot}}\mathcal{M}_3\,,$$ $$\label{eq:matrix-element-m1-square} |\mathcal{M}_1|^2=\frac{|\mathbf{k}|^2}{k_{\bot}^2}|\mathcal{M}_0|^2+\frac{k_{\scriptscriptstyle{\parallel}}^2}{k_{\bot}^2}|\mathcal{M}_3|^2+\frac{2|\mathbf{k}|k_{\parallel}}{k_{\bot}^2}\mathrm{Re}(\mathcal{M}_0\mathcal{M}_ | {M}_0|^2=|\mathcal{M}_3|^2$. Because of this, the Unphysical DegreEs oF frEeDom cAnceL each other and wHAt reMains are terms which invoLve thE pHYsicAL pOlariZation vECtORS ($\laMbDa=1$, 2). sinCe THe LatteR caN be chosEn as $(\varepsIloN_1^{\mU})=(0,1,0,0)$ and $(\varepsilON_2^{\mU})=(0,0,1,0)$, we obtain $$|\mAthCal{M}|^2\Big|^{\mathrM{phYs}}_{\mathRm{qED}}=\SUm_{\lamBda=1,2} |\VarepSilon_{\mU}^{(\Lambda)}(K)\mathcal{M}^{\Mu}(K)|^2=|\MathcaL{m}_1|^2+|\mathcaL{m}_2|^2=\SuM_{\lamBda=1,2} |\mathcal{M}_{\lambda}|^2\,,$$ WHeRE ‘phys’ means that The WarD iDEnTITy hAs bEen used.
In oRdEr to uNDerstanD ThE LIMitS Of the polarizaTion tensors FRom eq. we wiLl PerFOrm a siMilar AnALysIs in the contExt oF the modifIed theORy. For $(k^{\mU})=(|\Mathbf{k}|,\Mathbf{K})$ wiTh $\mAthbF{K}=(k_{\BoT},0,k_{\pArALleL})$ ThE WaRD idEntity reAdS $$\lAbel{eQ:WarD-IDENtitY}
k^{\mU}\matHcal{M}_{\Mu}=k_0\mathcal{M}_0-k_1\mAthCal{M}_1-K_3\MatHcal{M}_3=|\MathbF{k}|\maThCal{M}_0+k_{\Bot}\matHcal{M}_1+K_{\pArallel}\mathcal{M}_3=0\,,$$ And tHerefore, $\mAthCaL{M}_1$ cAn Be expREssed aS foLloWs:
$$\label{Eq:matriX-EleMeNT-M1}
\MaThcal{M}_1=-\frac{|\mathbf{k}|}{k_{\BoT}}\MAtHcal{M}_0-\fraC{k_{\paraLLeL}}{k_{\BOt}}\mathcaL{M}_3\,,$$
$$\LabEl{eq:MATrix-eLemeNT-m1-Square}
|\maThcal{M}_1|^2=\FRaC{|\mAthbf{k}|^2}{k_{\BoT}^2}|\mathcAl{m}_0|^2+\frAc{k_{\ScripTScriPtstylE{\paralleL}}^2}{k_{\bot}^2}|\MAthcal{M}_3|^2+\frac{2|\matHBf{k}|k_{\parallel}}{k_{\BOt}^2}\MAThRM{Re}(\mAthCal{M}_0\mathcal{m}_ | {M}_0|^2=|\mathcal{M}_3|^2 $. Because of t his , t he unp hysi cal degrees of free dom cancel each otherand w ha t rem a in s are termsw hi c h in vo lv e t he ph ysica l p olariza tion vecto rs($ \lambda=1$,2 ). Since the la tter can becho sen as $ (\v a repsi lon _1^{\ mu})=( 0 ,1,0,0 )$ and $( \v a repsil o n_2^{\m u } )= (0,0 ,1,0)$, we obtain $$ | \mathcal{M}|^2 \Big|^ {\ m at h r m{p hys }}_{\mathr m{ QED}} = \sum_{\ l am b d a =1, 2 } |\varepsilo n_{\mu}^{(\ l amb da)}(k )\ mat h cal{M} ^{\mu }( k )|^ 2=|\mathcal {M}_ 1|^2+|\ma thcal{ M }_2|^2= \ sum_{\l ambda= 1,2 } | \mat h ca l{ M}_ {\ l amb d a} |^2 \ ,,$ $ where‘p hy s’ me anst h a t the Wa rd i denti ty has been u sed .
I n or der t o und erst an d the limit s ofth e polarizationtens ors fromEq. w e w il l per f orm asim ila r analy sis int heco n t e xt of the modified t he o r y. For $(k ^{\mu} ) =( |\ m athbf{k} |, \ma thbf { k })$ w ith$ \m athbf{k} =(k_{\ b ot }, 0,k_{\p ar allel} )$ th e W ard i d enti ty rea ds $$\la bel{e q :Ward-identity }
k^{\mu}\math c al { M }_ { \mu} =k_ 0\mathcal{M }_0- k _1\m athc a l{ M}_ 1 -k_3\ mathc al { M} _ 3=|\mathbf{k}|\math ca l{M}_0 +k_{\ bot}\mathcal{ M}_1+k_{\p a r a llel}\ma thca l {M } _3=0\,,$$ andthere fore, $\ma t hcal{M}_ 1$ ca n be exp ressed as f ollows:
$$ \la bel {eq : m at rix-element-m 1 }
\ma th cal{M}_ 1=- \frac{| \ma thb f{k }|} {k _{\bot}}\ mathcal{ M} _0 -\ fr ac{ k_{\p a rallel}} {k _{\ bo t}} \math c al{M}_ 3\,,$ $
$ $\ la b el{ eq:matr i x- e l emen t- m1 -squ are }|\mat hcal { M}_ 1|^2=\f rac{|\mat hbf { k}|^ 2} {k _{\bot} ^2}|\mathcal{ M} _0|^2+\fra c{ k_{ \scrip t s criptsty le{\parallel}}^2}{k_{\b o t}^2}|\ mat hcal{ M}_3 |^2+\frac {2| \mathb f{k } |k_{\p aralle l}}{k _{ \bo t } ^2}\m a t hr m{R e} (\mathcal{ M } _0\ mathc al {M}_ | {M}_0|^2=|\mathcal{M}_3|^2$. Because_of this,_the unphysical degrees of_freedom cancel_each_other and_what_remains are terms_which involve the_physical polarization vectors ($\lambda=1$,_2). Since the_latter_can be chosen as $(\varepsilon_1^{\mu})=(0,1,0,0)$ and $(\varepsilon_2^{\mu})=(0,0,1,0)$, we obtain $$|\mathcal{M}|^2\Big|^{\mathrm{phys}}_{\mathrm{QED}}=\sum_{\lambda=1,2} |\varepsilon_{\mu}^{(\lambda)}(k)\mathcal{M}^{\mu}(k)|^2=|\mathcal{M}_1|^2+|\mathcal{M}_2|^2=\sum_{\lambda=1,2} |\mathcal{M}_{\lambda}|^2\,,$$ where ‘phys’_means_that the_Ward_identity_has been used.
In order to_understand the limits of the_polarization tensors_from Eq. we will perform a similar analysis_in_the context of_the modified theory. For $(k^{\mu})=(|\mathbf{k}|,\mathbf{k})$ with $\mathbf{k}=(k_{\bot},0,k_{\parallel})$ the Ward_identity reads $$\label{eq:Ward-identity}
k^{\mu}\mathcal{M}_{\mu}=k_0\mathcal{M}_0-k_1\mathcal{M}_1-k_3\mathcal{M}_3=|\mathbf{k}|\mathcal{M}_0+k_{\bot}\mathcal{M}_1+k_{\parallel}\mathcal{M}_3=0\,,$$ and therefore, $\mathcal{M}_1$_can be expressed_as_follows:
$$\label{eq:matrix-element-m1}
\mathcal{M}_1=-\frac{|\mathbf{k}|}{k_{\bot}}\mathcal{M}_0-\frac{k_{\parallel}}{k_{\bot}}\mathcal{M}_3\,,$$
$$\label{eq:matrix-element-m1-square}
|\mathcal{M}_1|^2=\frac{|\mathbf{k}|^2}{k_{\bot}^2}|\mathcal{M}_0|^2+\frac{k_{\scriptscriptstyle{\parallel}}^2}{k_{\bot}^2}|\mathcal{M}_3|^2+\frac{2|\mathbf{k}|k_{\parallel}}{k_{\bot}^2}\mathrm{Re}(\mathcal{M}_0\mathcal{M}_ |
neighborhood of the throat, to ensure the flaring out condition, and several models are analyzed. We shall also consider the possibility that these phantom wormholes be sustained by their own quantum fluctuations. The energy density of the graviton one loop contribution to a classical energy in a phantom wormhole background and the finite one loop energy density are considered as a self-consistent source for these wormhole geometries. The latter semi-classical approach prohibits solutions with a constant equation of state parameter, which further motivates the imposition of a radial dependent parameter, $\omega(r)$, and only permits solutions with a steep positive slope proportional to the radial derivative of the equation of state parameter, evaluated at the throat. The size of the wormhole throat as a function of the relevant parameters is also explored.'
author:
- Remo Garattini
- 'Francisco S. N. Lobo'
title: 'Self sustained phantom wormholes in semi-classical gravity'
---
Introduction
============
Unravelling the source for the late time accelerated expansion of the Universe has become a central theme in modern cosmology [@observations]. Several candidates, responsible for this expansion, have been proposed in the literature, ranging from modified gravity theories to the introduction of cosmological models comprising of a negative pressure cosmic fluid, denoted as dark energy. The latter is parameterized by the equation of state $\omega=p/\rho$, where $p$ is the spatially homogeneous pressure and $\rho$ the dark energy density. A value of $\omega<-1/3$ is required to ensure the cosmic accelerated expansion, and $\omega=-1$ corresponds to the presence of a cosmological constant. It is interesting that recent constraints from observational data have confirmed that the dark energy parameter lies slightly below the characteristic cosmological constant value, $\omega=-1$ [@3yWMAP]. Ameliorated fits to the data also indicate that an evolving equation of state parameter which crosses $\omega=-1$ is mildly favored [@phantom-divide]. The models satisfying $\omega<-1$ are denoted as *phantom energy* models, and possess rather interesting characteristics, in particular, the prediction of the *Big Rip* singularity and the violation of the null energy condition (NEC). The latter is a fundamental ingredient of traversable wormholes, and it was recently shown that phantom energy may indeed support these exotic geometries [@phantomWH; @phantomWH2]. Inspired by the evolving dark energy parameter crossing the phantom divide, we consider in this work a varying equation of state parameter | neighborhood of the throat, to ensure the flaring out condition, and respective mannequin are analyzed. We shall besides consider the hypothesis that these phantom wormholes be sustained by their own quantum fluctuation. The energy density of the graviton one cringle contribution to a classical energy in a phantom wormhole backdrop and the finite one loop energy concentration are considered as a self - consistent source for these wormhole geometries. The latter semi - classical approach prohibits solution with a constant equation of state argument, which further motivates the imposition of a radial pendent parameter, $ \omega(r)$, and merely permits solutions with a steep positive slope proportional to the radial derived function of the equation of state parameter, evaluated at the throat. The size of the wormhole throat as a function of the relevant parameters is also explored.'
author:
- Remo Garattini
-' Francisco S. N. Lobo'
title:' Self sustained phantom wormholes in semi - classical gravity'
---
Introduction
= = = = = = = = = = = =
Unravelling the source for the belated time accelerated expansion of the Universe has become a cardinal root in advanced cosmology [ @observations ]. Several candidates, creditworthy for this expansion, have been proposed in the literature, range from modified gravity theory to the introduction of cosmological models comprising of a negative pressure cosmic fluid, denote as dark energy. The latter is parameterized by the equation of state $ \omega = p/\rho$, where $ p$ is the spatially homogeneous atmospheric pressure and $ \rho$ the dark department of energy concentration. A value of $ \omega<-1/3 $ is command to ensure the cosmic accelerated expansion, and $ \omega=-1 $ corresponds to the presence of a cosmological constant. It is interesting that late constraints from observational data have confirmed that the dark energy parameter lies slightly below the characteristic cosmologic constant value, $ \omega=-1 $ [ @3yWMAP ]. Ameliorated fit to the data also bespeak that an evolving equality of state parameter which thwart $ \omega=-1 $ is mildly favored [ @phantom - watershed ]. The models satisfying $ \omega<-1 $ are denoted as * phantom energy * models, and possess rather interesting characteristics, in particular, the prediction of the * Big Rip * singularity and the violation of the null energy condition (NEC). The latter is a cardinal ingredient of traversable wormholes, and it was recently picture that phantom energy may indeed support these exotic geometries [ @phantomWH; @phantomWH2 ]. Inspired by the evolving dark department of energy parameter crossing the phantom divide, we consider in this work a varying equation of state argument | nelghborhood of the throat, to ensure the yoaring out ckndition, and several models are analbzed. We syall also consider the possibilpty that rhest phantom wormholxa be sustaines by chxir own quantum fluctuatimns. The energy ddndity of the graviton one loop contrybution tl a classical gnergj yn a ihcntom wormhole background and tge finiue one loop energy density are considered as a sflf-consistent sourfe for thesg wownhole geometfies. The latter semi-clzssical approach prohibits solugions with a cobsranh equation oh statv parameter, wmpch furdher moyivates the imiositmon if a radial dependent parameter, $\omega(r)$, wnd only [exmits solutions with q wteep poshtivd slupe pcopkrtionwl vo the radizl derivatice of the equation pf wtate parametsr, evajuwted at the throat. The size of the wormvols throat as a function if the relevant paramgters is ajso explored.'
author:
- Remo Garattini
- 'Francisco S. N. Lmbo'
tivld: 'Stln ryshained phantom wormholes in semi-classical gradjtu'
---
Iktroduction
============
Unravclling the source gog yre late time xccelexztsd expansion of thf Univetse haw become w cemtral theme in modern cosmooogy [@observanionw]. Several candidatzs, responsibue fpr thos expansion, have been 'ropossd in the llterature, fanging from modkfivd gsavity theories to the intwoduction of eosmologkcal modelf comprisijg of a negative pressurf cosliw fluid, dejoted as dark energy. The latter is parameterieed by the equction pf state $\omegw=p/\rho$, where $p$ ns the s'atialuy homogenvous presvure and $\rhj$ the dark enatgy density. A value os $\omwga<-1/3$ us requkfed to ensure yhe cosmib ccceleratwd expansion, and $\okegx=-1$ corresponds to tht presence of a vosooljgpcak smnstant. It iv ingerdxting that rteenb cunsttaints from observathonam data have confirkeq that tye dark qnergy parameyer lies slightly helow thx charsctgristic cosmological constant vzlue, $\omegw=-1$ [@3yRMAP]. Ameliorwted fits to the data also indicate that an evolving eqnation of state paramejer which crosses $\omeya=-1$ is mildly havoreq [@phantom-givide]. The models sarisfying $\omega<-1$ art denoted as *phantom ensrgy* mmdels, and possess rather interesting characteristics, in particular, the predictuon of the *Big Rip* sinbularpty cnd the viopetion of the null energy condition (NEC). The lattec is a fungaiental ingredient of traverssbue wormholes, xnd it was recently showh that pnantom energy may indeed suppory these exotic geometriex [@phantonWH; @pgantomWH2]. Iispired by the evolvong daxk eneegy parqmetcr crossing thd pnaktom dividf, we cknsider in this work a varying ewuatiob lf state parametxr | neighborhood of the throat, to ensure the condition, several models analyzed. We shall these wormholes be sustained their own quantum The energy density of the graviton loop contribution to a classical energy in a phantom wormhole background and the one loop energy density are considered as a self-consistent source for these wormhole The semi-classical prohibits with a constant equation of state parameter, which further motivates the imposition of a radial dependent $\omega(r)$, and only permits solutions with a steep slope proportional to the derivative of the equation of parameter, at the The of wormhole throat as function of the relevant parameters is also explored.' author: - Remo Garattini - 'Francisco S. N. Lobo' 'Self sustained in semi-classical --- ============ the source for time accelerated expansion of the Universe central theme in modern cosmology [@observations]. Several candidates, for this have been proposed in the literature, from modified gravity theories to the introduction of models comprising of a negative pressure cosmic fluid, denoted as dark energy. The latter is the equation of state where $p$ is spatially pressure $\rho$ dark energy A value of $\omega<-1/3$ is required to ensure the cosmic accelerated and $\omega=-1$ corresponds to the presence of a cosmological constant. interesting recent constraints from data have confirmed that dark parameter lies slightly below cosmological value, fits the also indicate that an equation of state parameter which $\omega=-1$ is mildly favored are denoted as *phantom energy* models, and possess interesting characteristics, in particular, the prediction of *Big Rip* singularity and the violation of the null energy condition (NEC). latter is ingredient of traversable wormholes, and it was recently that phantom energy may support these exotic geometries [@phantomWH; @phantomWH2]. Inspired by the dark parameter crossing phantom divide, we in this work varying equation of | neighborhood of the throat, to Ensure the fLarinG ouT coNdItioN, and Several models aRE anaLyzed. We shall also considEr the PoSSibiLItY that These phANtOM WorMhOlEs bE sUStAined By tHeir own Quantum fluCtuAtIons. The energY DeNsity of the GraViton one loop ConTributIoN to A ClassIcaL enerGy in a pHAntom wOrmhole baCkGRound aND the finITE oNe loOp energy density arE CoNSidered as a self-ConsisTeNT sOURce For These wormhOlE geomETries. ThE LaTTER seMI-classical appRoach prohibITs sOlutioNs WitH A constAnt eqUaTIon Of state paraMeteR, which furTher moTIvates tHE imposiTion of A raDiaL depENdEnT paRaMEteR, $\OmEga(R)$, And Only permItS sOlutiOns wITH A SteeP poSitiVe sloPe proportionaL to The rADiaL deriVativE of tHe EquatIon of sTate pArAmeter, evaluated At thE throat. ThE siZe Of tHe WormhOLe throAt aS a fUnction Of the reLEvaNt PARAmEters is also exploreD.'
aUTHoR:
- Remo GarAttini
- 'fRaNcISco S. N. LobO'
tItlE: 'SelF SUstaiNed pHAnTom wormhOles in SEmI-cLassicaL gRavity'
---
inTroDucTion
============
UNRaveLling tHe source For thE Late time acceleRAted expansion OF tHE unIVersE haS become a cenTral THeme In moDErN coSMologY [@obseRvATiONs]. Several candidates, ReSponsiBle foR this expansioN, have been pROPOsed in thE litERaTUre, ranging from ModifIed gravity THeories tO the iNtroductIon of cosmOLOgical moDelS coMprIsiNG Of A negative presSURe coSmIc fluid, DenOted as dArk EneRgy. the LaTter is parAmeterizEd By ThE eQuaTion oF State $\omeGa=P/\rhO$, wHerE $p$ is tHE spatiAlly hOmogEnEoUS prEssure aND $\rHO$ The dArK eNergY deNsIty. A vAlue OF $\omEga<-1/3$ is reQuired to eNsuRE the CoSmIc accelErated expansiOn, And $\omega=-1$ coRrEspOnds to THE presencE of a cosmological constanT. it is intEreSting That Recent conStrAints fRom OBservaTional Data hAvE coNFIrmed THAt The DaRk energy paRAMetEr lieS sLighTly beloW the characteristic COsmOlogical constAnt ValuE, $\OMeGa=-1$ [@3ywmAp]. ameLiORatED Fits to the data alSo indicate ThAT aN evolving eQUatIoN of statE parameTer whICh crossEs $\omega=-1$ is Mildly favOrEd [@phANTom-Divide]. The mOdels satIsfying $\omEGa<-1$ are DEnOted aS *phAntom eNeRgy* ModelS, and poSSesS rathEr inteReSting cHaracTeRistics, iN particular, the predictioN of the *big RiP* siNgularity And THe vIolation oF the Null energy ConDitIon (NEc). ThE LatteR is a FUnDamENtal iNgreDIent of traVErSabLE WoRmholes, and iT WAS reCentlY shOWn that PhanTom energy may indeeD Support these exOtic GEOmeTriES [@phaNtOmWH; @phantomWH2]. INspIrED By the evoLvIng dark enerGy parameTeR CrossIng the PhantoM divide, WE CoNSider iN thiS woRk a varyinG eqUaTIon of stAtE pARameteR | neighborhood of the throa t, to ensu re th e f lar in g ou t co ndition, and s e vera l models are analyzed. We s ha l l al s oconsi der the po s s ibi li ty th at th ese p han tom wor mholes besus ta ined by thei r o wn quantum fl uctuations.The energ yden s ity o f t he gr aviton one lo op contri bu t ion to a class i c al ene rgy in a phantomw or m hole backgroun d andth e f i n ite on e loop ene rg y den s ity are co n s i der e d as a self-c onsistent s o urc e forth ese wormho le ge om e tri es. The lat tersemi-clas sicala pproach prohibi ts sol uti ons wit h a c ons ta n t e q ua tio n of state p ar am eter, whi c h f urth ermoti vates the impositi onof a rad ial d epend entpa ramet er, $\ omega (r )$, and only pe rmit s solutio nswi thasteep positi veslo pe prop ortiona l to t h e ra dial derivative of t h e e quationof sta t epa r ameter,ev alu ated a t the thr o at . The si ze oft he w ormhole t hroatas afun ction of t he rel evant pa ramet e rs is also exp l ored.'
author :
- R em o Gar att ini
- 'Fran cisc o S.N. L o bo '
t i tle:'Self s u st a ined phantom wormho le s in s emi-c lassical grav ity'
---
I n t roductio n
== = == = ======
Unrave lling the sourc e for the late time ac celerated e xpansion of th e U niv e r se has become a c entr al themeinmoderncos mol ogy [@ ob servation s]. Seve ra lca nd ida tes,r esponsib le fo rthi s exp a nsion, have bee npr o pos ed in t h el i tera tu re , ra ngi ng from mod i fie d gravi ty theori est o th ein troduct ion of cosmol og ical model scom prisin g of a neg ative pressure cosmic f l uid, de not ed as dar k energy. Th e latt eri s para meteri zed b ythe e quati o n o f s ta te $\omega = p /\r ho$,wh ere$p$ isthe spatially homo g ene ous pressureand $\r h o $the da r k e ne r gyd e nsity. A valueof $\omega <- 1 /3 $ is requi r edto ensure the co smica ccelera ted expan sion, and $ \ome g a =-1 $ correspo nds to t he presen c e ofa c osmol ogi cal co ns tan t. It is in t ere sting thatre cent c onstr ai nts from observational data hav e conf irmed th at the da rke ner gy parame terlies sligh tly be low t hec harac teri s ti c c o smolo gica l constant va lue , $\ omega=-1$ [ @ 3 y WMA P]. A mel i orated fit s to the data als o indicate that ane v olv ing equa ti on of state pa ram et e r which c ro sses $\omeg a=-1$ is m i ldlyfavore d [@ph antom-d i v id e ]. The mod els satisfyi ng$\ o mega<-1 $ar e denot ed a s*phant om ene r gy*m o dels, and posses s rat h e r int e res tingch aracter i stic s, in part icular, the predi ctio n ofthe *Bi gRip* s ing ul arity andt he violat ion o f the n ul l en erg y cond itio n (NEC) . Th elat ter is af u nd a me nt a l i ngre dient o f tr aversable wormhole s,a nd it w as re c e ntly s h ow n that phant omenerg y may indeed supp o rt these exoti c geom etries[ @ph an tomWH;@ph a n tomWH2].Inspiredb y t he evo lving da rk ene rg y p ar ameter crossing the pha ntom div i d e, wecon s ider in thi swork a varyi ng eq ua t ion ofs t ate parame t er | neighborhood_of the_throat, to ensure the_flaring out_condition,_and several_models_are analyzed. We_shall also consider_the possibility that these_phantom wormholes be_sustained_by their own quantum fluctuations. The energy density of the graviton one loop contribution_to_a classical_energy_in_a phantom wormhole background and_the finite one loop energy_density are_considered as a self-consistent source for these wormhole_geometries._The latter semi-classical_approach prohibits solutions with a constant equation of state_parameter, which further motivates the imposition_of a radial_dependent_parameter,_$\omega(r)$, and only permits_solutions with a steep positive slope_proportional to the radial derivative of_the equation of state parameter, evaluated at_the throat. The size of the_wormhole throat as a function_of the_relevant parameters is also explored.'
author:
-_Remo Garattini
- 'Francisco_S. N._Lobo'
title: 'Self sustained_phantom wormholes in semi-classical gravity'
---
Introduction
============
Unravelling the_source for the_late time accelerated expansion of the_Universe_has become a_central_theme_in modern_cosmology [@observations]. Several_candidates,_responsible for_this_expansion, have been proposed in the_literature,_ranging from modified gravity theories to the_introduction of cosmological models_comprising_of a negative pressure_cosmic fluid, denoted as dark_energy. The latter is parameterized by_the equation_of state_$\omega=p/\rho$, where $p$ is the spatially homogeneous pressure and $\rho$ the_dark energy density. A value of_$\omega<-1/3$ is required to_ensure the_cosmic_accelerated expansion, and_$\omega=-1$_corresponds to_the presence of a cosmological constant. It_is interesting_that recent constraints from observational data_have confirmed that the_dark_energy parameter lies slightly below the_characteristic cosmological constant value, $\omega=-1$ [@3yWMAP]._Ameliorated fits to the data_also_indicate_that an evolving equation of_state parameter which crosses $\omega=-1$ is_mildly favored [@phantom-divide]._The models satisfying $\omega<-1$ are denoted as_*phantom_energy* models, and possess rather interesting_characteristics,_in particular, the prediction of the_*Big_Rip*_singularity and the violation of_the null energy condition (NEC). The_latter is a fundamental ingredient of traversable wormholes, and_it was recently_shown that phantom energy may_indeed_support_these exotic geometries [@phantomWH; @phantomWH2]. Inspired by the evolving dark_energy parameter_crossing the phantom_divide, we consider in this work a varying equation of_state parameter |
frequency tail. We find that $|F(q,\omega)|$ is a rapidly decreasing function of the number of particles involved in the excitation. In particular, we find for all $\Delta$ that the peak is completely dominated by two-particle (single particle-hole) and the tail by four-particle states (denoted by 2$p$ and 4$p$ states, respectively). Including up to eight-particle as well as bound states we verify using Density Matrix Renormalization Group (DMRG) that the sum rules are fulfilled with high accuracy corroborating our numerical results. By solving the BA equations for small $\Delta$ and infinite system size analytically we show that the width of the peak scales like $q^2$ for $h\neq 0$. Furthermore, we calculate the high-frequency tail analytically based on a parameter-free effective bosonic Hamiltonian. We demonstrate that our analytical results for the linewidth and the tail are in excellent agreement with our numerical data.
For a chain of length $N$ the longitudinal dynamical structure factor is defined by $$\begin{aligned}
S^{zz}\left(q,\omega\right)&=&\frac{1}{N}\sum_{j,j^{\prime}=1}^{N}e^{-iq\left(j-j^{\prime}\right)}\int_{-\infty}^{+\infty}\!\!\!\!\!\!\!\!
dt\,
e^{i\omega t}\left\langle
S_{j}^{z}\left(t\right)S_{j^{\prime}}^{z}\left(0\right)\right\rangle
\nonumber \\
&=& \frac{2\pi}{N}\sum_{\alpha}\left|\left\langle
0\left|S_{q}^{z}\right|\alpha\right\rangle
\right|^{2}\delta\left(\omega-E_{\alpha}\right) \;.
\label{strucFac}\end{aligned}$$ Here $S_{q}^{z}=\sum_{j}S_{j}^{z}e^{-iqj}$ and $\left|\alpha\right\rangle $ is an eigenstate with energy $E_{\alpha}$ above the ground state energy. For a finite system, $S^{zz}\left(q,\omega\right)$ at fixed $q$ is a sum of $\delta$-peaks at the energies of the eigenstates. In the thermodynamic limit $N\rightarrow\infty$, the spectrum is continuous and $S^{zz}\left( | frequency tail. We find that $ |F(q,\omega)|$ is a rapidly decreasing affair of the numeral of particles involved in the excitation. In finical, we find for all $ \Delta$ that the peak is wholly dominated by two - particle (single atom - fix) and the tail by four - atom states (denoted by 2$p$ and 4$p$ states, respectively). Including up to eight - atom as well as bound state we verify using Density Matrix Renormalization Group (DMRG) that the kernel rules are fulfilled with high accuracy corroborate our numerical results. By solving the BA equation for small $ \Delta$ and infinite system size analytically we show that the width of the peak scale like $ q^2 $ for $ h\neq 0$. Furthermore, we calculate the high - frequency tail analytically based on a parameter - free effective bosonic Hamiltonian. We demonstrate that our analytical results for the linewidth and the tail are in excellent agreement with our numerical data.
For a chain of length $ N$ the longitudinal dynamical structure factor is specify by $ $ \begin{aligned }
S^{zz}\left(q,\omega\right)&=&\frac{1}{N}\sum_{j, j^{\prime}=1}^{N}e^{-iq\left(j - j^{\prime}\right)}\int_{-\infty}^{+\infty}\!\!\!\!\!\!\!\!
dt\,
e^{i\omega t}\left\langle
S_{j}^{z}\left(t\right)S_{j^{\prime}}^{z}\left(0\right)\right\rangle
\nonumber \\
& = & \frac{2\pi}{N}\sum_{\alpha}\left|\left\langle
0\left|S_{q}^{z}\right|\alpha\right\rangle
\right|^{2}\delta\left(\omega - E_{\alpha}\right) \; .
\label{strucFac}\end{aligned}$$ Here $ S_{q}^{z}=\sum_{j}S_{j}^{z}e^{-iqj}$ and $ \left|\alpha\right\rangle $ is an eigenstate with energy $ E_{\alpha}$ above the ground department of state energy. For a finite system, $ S^{zz}\left(q,\omega\right)$ at fixed $ q$ is a sum of $ \delta$-peaks at the department of energy of the eigenstates. In the thermodynamic limit $ N\rightarrow\infty$, the spectrum is continuous and $ S^{zz}\left ( | freeuency tail. We find that $|F(q,\omega)|$ is a rcpidly vecreasjng funcgion of the number of particpew invilved in the excitatiov. In partpcular, we finv for all $\Delta$ vgat the peak jd cok'letely dominatgd by two-pardicle (single pdrgiele-hole) and the tail by four-particle states (dfnoted by 2$p$ anq 4$p$ xeatea, respectively). Including up to eiggt-partible as well as boind states we verify using Dendity Matrix Renormwlization Gtkup (EMRG) that thd sum rules are fulfilmed with high accuracy corroborxting our numerucql ggsults. By sooving the BA equabpons fos small $\Delta$ and inflnite sywtem size analyticallb we show that the wydth of tve peak scales like $q^2$ for $v\neq 0$. Fufrhefmode, ws calcklave the high-rrequency tqil analytically baxeq on a parametsr-free esfective bosonic Hamiltonian. We demonstgate that our analytical rewults for the linewidjh and the tail are in excellent agreement with our numericdl dava.
Wor a zyaln of length $N$ the longitudinal dynamical strtdtirv factor is defincd by $$\begin{aligned}
X^{zx}\lrst(q,\omega\right)&=&\wrac{1}{N}\sbj_{j,n^{\prime}=1}^{N}e^{-iq\left(j-j^{\prlme}\righj)}\int_{-\indty}^{+\infty}\!\!\!\!\!\!\!\!
du\,
e^{i\omrga t}\left\langle
S_{j}^{z}\left(t\ritht)S_{j^{\prime}}^{z}\lvft(0\rught)\right\rangle
\nonbmber \\
&=& \frac{2\pn}{N}\sum_{\akpha}\lrft|\left\langle
0\left|S_{q}^{v}\right|\zlpha\right\rwngle
\righf|^{2}\aelta\left(\omega-E_{\auphs}\rhght) \;.
\labtu{strucFac}\end{alignqd}$$ Here $S_{w}^{z}=\suk_{j}S_{j}^{z}e^{-idj}$ amd $\lefe|\alpha\righh\rangle $ is an eigenstatf witk enesgy $E_{\alpha}$ above the ground state energy. Hpr a finite xyvtek, $S^{zz}\lefc(q,\omegs\right)$ at fixqd $q$ is a sum pf $\deltc$-peaks at the envrgies of the eigenseates. In the djermodynamic limit $N\wighrarriw\infty$, ghe spectrum ix continuous and $S^{zz}\left( | frequency tail. We find that $|F(q,\omega)|$ is decreasing of the of particles involved we for all $\Delta$ the peak is dominated by two-particle (single particle-hole) and tail by four-particle states (denoted by 2$p$ and 4$p$ states, respectively). Including up eight-particle as well as bound states we verify using Density Matrix Renormalization Group that sum are with high accuracy corroborating our numerical results. By solving the BA equations for small $\Delta$ and system size analytically we show that the width the peak scales like for $h\neq 0$. Furthermore, we the tail analytically on parameter-free bosonic Hamiltonian. We that our analytical results for the linewidth and the tail are in excellent agreement with our numerical For a length $N$ longitudinal structure is defined by dt\, e^{i\omega t}\left\langle S_{j}^{z}\left(t\right)S_{j^{\prime}}^{z}\left(0\right)\right\rangle \nonumber \\ \right|^{2}\delta\left(\omega-E_{\alpha}\right) \;. \label{strucFac}\end{aligned}$$ Here $S_{q}^{z}=\sum_{j}S_{j}^{z}e^{-iqj}$ and $\left|\alpha\right\rangle $ an eigenstate energy $E_{\alpha}$ above the ground state For a finite system, $S^{zz}\left(q,\omega\right)$ at fixed $q$ a sum of $\delta$-peaks at the energies of the eigenstates. In the thermodynamic limit $N\rightarrow\infty$, is continuous and $S^{zz}\left( | frequency tail. We find that $|F(q,\Omega)|$ is a raPidly DecReaSiNg fuNctiOn of the number oF PartIcles involved in the exciTatioN. IN PartICuLar, we Find for ALl $\dELta$ ThAt The PeAK iS compLetEly domiNated by two-ParTiCle (single parTIcLe-hole) and tHe tAil by four-parTicLe statEs (DenOTed by 2$P$ anD 4$p$ staTes, resPEctiveLy). IncludiNg UP to eigHT-particLE As Well As bound states we veRIfY Using Density MaTrix ReNoRMaLIZatIon group (DMRG) tHaT the sUM rules aRE fULFIllED with high accuRacy corroboRAtiNg our nUmEriCAl resuLts. By SoLVinG the BA equatIons For small $\DElta$ anD InfinitE System sIze anaLytIcaLly wE ShOw ThaT tHE wiDTh Of tHE peAk scales LiKe $Q^2$ for $h\Neq 0$. FURTHErmoRe, wE calCulatE the high-frequEncY taiL AnaLyticAlly bAsed On A paraMeter-fRee efFeCtive bosonic HamIltoNian. We demOnsTrAte ThAt our ANalytiCal ResUlts for The lineWIdtH aND THe Tail are in excellent AgREEmEnt with oUr numeRIcAl DAta.
For a cHaIn oF lenGTH $N$ the LongITuDinal dynAmical STrUcTure facToR is defInEd bY $$\beGin{alIGned}
s^{zz}\lefT(q,\omega\rIght)&=&\fRAc{1}{N}\sum_{j,j^{\prime}=1}^{N}E^{-Iq\left(j-j^{\prime}\RIgHT)}\InT_{-\InftY}^{+\inFty}\!\!\!\!\!\!\!\!
dt\,
e^{i\omegA t}\leFT\lanGle
S_{J}^{Z}\lEft(T\Right)s_{j^{\priMe}}^{Z}\LeFT(0\right)\right\rangle
\noNuMber \\
&=& \frAc{2\pi}{N}\Sum_{\alpha}\left|\lEft\langle
0\lEFT|s_{q}^{z}\right|\AlphA\RiGHt\rangle
\right|^{2}\dElta\lEft(\omega-E_{\aLPha}\right) \;.
\Label{StrucFac}\End{aligneD}$$ hEre $S_{q}^{z}=\suM_{j}S_{J}^{z}e^{-Iqj}$ And $\LEFt|\Alpha\right\ranGLE $ is aN eIgenstaTe wIth enerGy $E_{\AlpHa}$ aBovE tHe ground sTate enerGy. foR a FiNitE systEM, $S^{zz}\left(Q,\oMegA\rIghT)$ at fiXEd $q$ is a Sum of $\DeltA$-pEaKS at The enerGIeS OF the EiGeNstaTes. in The thErmoDYnaMic limiT $N\rightarRow\INfty$, ThE sPectrum Is continuous aNd $s^{zz}\left( | frequency tail. We find th at $|F(q,\ omega )|$ is a rap idly decreasing fu n ctio n of the number of par ticle si nvol v ed in t he exci t at i o n.In p art ic u la r, we fi nd forall $\Delt a$th at the peaki scompletely do minated by t wo- partic le (s i nglepar ticle -hole) and th e tail by f o ur-par t icle st a t es (de noted by 2$p$ and 4$ p $ states, resp ective ly ) .I n clu din g up to ei gh t-par t icle as we l l asb ound states w e verify us i ngDensit yMat r ix Ren ormal iz a tio n Group (DM RG)that thesum ru l es aref ulfille d with hi ghaccu r ac ycor ro b ora t in g o u r n umerical r es ults. Bys o l v ingthe BAequat ions for smal l $ \Del t a$and i nfini te s ys tem s ize an alyti ca lly we show tha t th e width o f t he pe ak scal e s like $q ^2$ for $h \neq 0$ . Fu rt h e r mo re, we calculate t he h ig h-freque ncy ta i lan a lyticall ybas ed o n a par amet e r- free eff ective bo so nic Ham il tonian .Wedem onstr a te t hat ou r analyt icalr esults for the linewidth and th e ta i l ar e i n excellent agr e emen t wi t hour numer icalda t a.
For a chain of len gt h $N$the l ongitudinal d ynamical s t r u cture fa ctor is defined by $$\ begin {aligned}S ^{zz}\le ft(q, \omega\r ight)&=&\ f r ac{1}{N} \su m_{ j,j ^{\ p r im e}=1}^{N}e^{- i q \lef t( j-j^{\p rim e}\righ t)} \in t_{ -\i nf ty}^{+\in fty}\!\! \! \! \! \! \!\ !
dt\ ,
e^{i\om eg a t }\ lef t\lan g le
S _{j}^ {z}\ le ft ( t\r ight)S_ { j^ { \ prim e} }^ {z}\ lef t( 0\rig ht)\ r igh t\rangl e
\nonumb er\ \
&= &\f rac{2\p i}{N}\sum_{\a lp ha}\left|\ le ft\ langle 0\lef t|S_{q}^{z}\right|\alph a \right\ ran gle
\ righ t|^{2}\de lta \left( \om e ga-E_{ \alpha }\rig ht ) \ ; .
\lab e l {s tru cF ac}\end{al i g ned }$$ H er e $S _{q}^{z }=\sum_{j}S_{j}^{z } e^{ -iqj}$ and $\ lef t|\a l p ha \ri g ht \ ran gl e $i s an eigenstatewith energ y$ E_ {\alpha}$a bov ethe gro und sta te en e rgy. Fo r a finit e system, $ S^{z z } \le ft(q,\omeg a\right) $ at fixe d $q$i sa sum of $\del ta $-p eaksat the ene rgies of th eeigens tates .In the t hermodynamic limit $N\r ightar row\i nft y$, the s pec t rum is conti nuou s and $S^{ zz} \le ft( | frequency tail._We find_that $|F(q,\omega)|$ is a_rapidly decreasing_function_of the_number_of particles involved_in the excitation._In particular, we find_for all $\Delta$_that_the peak is completely dominated by two-particle (single particle-hole) and the tail by four-particle_states_(denoted by_2$p$_and_4$p$ states, respectively). Including up_to eight-particle as well as_bound states_we verify using Density Matrix Renormalization Group (DMRG)_that_the sum rules_are fulfilled with high accuracy corroborating our numerical results._By solving the BA equations for_small $\Delta$ and_infinite_system_size analytically we show_that the width of the peak_scales like $q^2$ for $h\neq 0$._Furthermore, we calculate the high-frequency tail analytically_based on a parameter-free effective bosonic_Hamiltonian. We demonstrate that our_analytical results_for the linewidth and the_tail are in_excellent agreement_with our numerical_data.
For a chain of length $N$_the longitudinal dynamical_structure factor is defined by $$\begin{aligned}
S^{zz}\left(q,\omega\right)&=&\frac{1}{N}\sum_{j,j^{\prime}=1}^{N}e^{-iq\left(j-j^{\prime}\right)}\int_{-\infty}^{+\infty}\!\!\!\!\!\!\!\!
dt\,
e^{i\omega_t}\left\langle
_ S_{j}^{z}\left(t\right)S_{j^{\prime}}^{z}\left(0\right)\right\rangle
\nonumber \\
&=&_\frac{2\pi}{N}\sum_{\alpha}\left|\left\langle
__ _0\left|S_{q}^{z}\right|\alpha\right\rangle
\right|^{2}\delta\left(\omega-E_{\alpha}\right) \;.
\label{strucFac}\end{aligned}$$ Here_$S_{q}^{z}=\sum_{j}S_{j}^{z}e^{-iqj}$_and $\left|\alpha\right\rangle_$_is an eigenstate with energy $E_{\alpha}$_above_the ground state energy. For a finite_system, $S^{zz}\left(q,\omega\right)$ at fixed_$q$_is a sum of_$\delta$-peaks at the energies of_the eigenstates. In the thermodynamic limit_$N\rightarrow\infty$, the_spectrum is_continuous and $S^{zz}\left( |
, so it suffices to prove that this whole vector space is zero. The functor ${\operatorname{Hom}}_{Y_0 \times C}$ is the composition of three functors: $${\operatorname{Hom}}_{Y_0 \times C} = \Gamma \circ \pi_{1*} \circ {\curly Hom}_{Y_0 \times C},$$ so to prove the vanishing of $${\operatorname{Ext}}^1_{Y_0 \times C}({\overline}{S},{\overline}{Q} \otimes \pi_1^* I) = {\operatorname{R}}^1 {\operatorname{Hom}}_{Y_0 \times C}({\overline}{S},{\overline}{Q} \otimes \pi_1^* I),$$ it suffices (by the Grothendieck spectral sequence) to prove the vanishing of the three vector spaces $$\begin{array}{l} {\operatorname{H}}^1(Y_0, \pi_{1*} {\curly Hom}({\overline}{S},{\overline}{Q} \otimes \pi_1^* I) \\ {\operatorname{H}}^0(Y_0, \pi_{1*} {\curly Ext}^1({\overline}{S}, {\overline}{Q} \otimes \pi_1^* I)) \\ {\operatorname{H}}^0(Y_0, {\operatorname{R}}^1 \pi_{1*} {\curly Hom}({\overline}{S},{\overline}{Q}) \otimes I)). \end{array}$$ The first vanishes because $Y_0$ is affine and the second vanishes because ${\overline}{S}$ is locally free.
For the third vanishing, we will show that ${\operatorname{R}}^1 \pi_{1*} {\curly Hom}({\overline}{S},{\overline}{Q})=0$. By cohomology and base change and the assumption that $Y_0 \to X$ factors through $\iota_D$, we might as well assume $Y_0=X$, so ${\overline}{S}=S(-D)$. Using the vanishing together with cohomology and base change (and flatness of ${\curly Hom}(S,Q)$), we first observe that $${\operatorname{R}}^1 \pi_{1*} {\curly Hom}({\overline}{S},Q) = {\operatorname{R}}^1 \pi_{1*} {\curly Hom}(S,Q)(D) = 0.$$ Next, we tensor the sequence $\eqref{quotientSES}$ with $S^\lor(D)$ and apply $\pi_{ | , so it suffices to prove that this whole vector space is zero. The functor $ { \operatorname{Hom}}_{Y_0 \times C}$ is the composing of three functors: $ $ { \operatorname{Hom}}_{Y_0 \times C } = \Gamma \circ \pi_{1 * } \circ { \curly Hom}_{Y_0 \times C},$$ therefore to prove the vanishing of $ $ { \operatorname{Ext}}^1_{Y_0 \times C}({\overline}{S},{\overline}{Q } \otimes \pi_1^ * I) = { \operatorname{R}}^1 { \operatorname{Hom}}_{Y_0 \times C}({\overline}{S},{\overline}{Q } \otimes \pi_1^ * I),$$ it suffices (by the Grothendieck apparitional sequence) to prove the vanishing of the three vector spaces $ $ \begin{array}{l } { \operatorname{H}}^1(Y_0, \pi_{1 * } { \curly Hom}({\overline}{S},{\overline}{Q } \otimes \pi_1^ * I) \\ { \operatorname{H}}^0(Y_0, \pi_{1 * } { \curly Ext}^1({\overline}{S }, { \overline}{Q } \otimes \pi_1^ * I) ) \\ { \operatorname{H}}^0(Y_0, { \operatorname{R}}^1 \pi_{1 * } { \curly Hom}({\overline}{S},{\overline}{Q }) \otimes I) ). \end{array}$$ The inaugural vanishes because $ Y_0 $ is affine and the second vanishes because $ { \overline}{S}$ is locally free.
For the third vanishing, we will show that $ { \operatorname{R}}^1 \pi_{1 * } { \curly Hom}({\overline}{S},{\overline}{Q})=0$. By cohomology and foundation change and the assumption that $ Y_0 \to X$ agent through $ \iota_D$, we might as well assume $ Y_0 = X$, then $ { \overline}{S}=S(-D)$. use the vanishing together with cohomology and basis change (and flatness of $ { \curly Hom}(S, Q)$), we foremost observe that $ $ { \operatorname{R}}^1 \pi_{1 * } { \curly Hom}({\overline}{S},Q) = { \operatorname{R}}^1 \pi_{1 * } { \curly Hom}(S, Q)(D) = 0.$$ Next, we tensor the sequence $ \eqref{quotientSES}$ with $ S^\lor(D)$ and apply $ \pi _ { | , so it suffices to prove thxt this whole vgcror spece is aero. The functor ${\operatorname{Hom}}_{Y_0 \tilew C}$ iw the composition of tfree funcnors: $${\operqtoriame{Hom}}_{Y_0 \times C} = \Gamma \gnrc \pj_{1*} \cire {\rurly Hom}_{Y_0 \timex C},$$ so to [rove the vanivhkny of $${\operatorname{Ext}}^1_{Y_0 \times C}({\overlinq}{S},{\overlonf}{Q} \otimes \pi_1^* I) = {\optraeornzme{R}}^1 {\operatorname{Hom}}_{Y_0 \times C}({\overljne}{S},{\ovegline}{Q} \otimes \pi_1^* O),$$ it suffices (by the Grothfndifck spectral sequejce) to provg thq vanishing ow the thret rector spacgs $$\begin{array}{l} {\operatorname{H}}^1(Y_0, \pi_{1*} {\curlv Hom}({\overlibe}{W},{\ovftline}{Q} \otimew \pi_1^* P) \\ {\operatorname{H}}^0(Y_0, \pi_{1*} {\curly Rxt}^1({\overline}{S}, {\oyerliie}{Q} \itimes \pi_1^* I)) \\ {\operatoriame{H}}^0(Y_0, {\operatorname{R}}^1 \pi_{1*} {\curly Hkm}({\overline}{S},{\overlibe}{W}) \otikes H)). \ena{qrrxy}$$ Uhe fjrst vwniahes becauae $Y_0$ is affune and the second fagpxhes because ${\overlynq}{S}$ is locally free.
For the third vanishitg, se will show that ${\operarorname{R}}^1 \pi_{1*} {\curly Hom}({\lverline}{S},{\jverline}{Q})=0$. By cohomology and base change and the avsumpviun uhqt $Y_0 \gi D$ factors through $\iota_D$, we might as well assuis $U_0=X$, so ${\overline}{S}=S(-B)$. Using the vanixhlnb together witf cohomolkgy and base changf (and fjatnews of ${\curjy Hpm}(S,Q)$), we first observe that $${\iperatorname{G}}^1 \pi_{1*} {\curly Hom}({\overline}{D},Q) = {\operatoxname{R}}^1 \pi_{1*} {\cirly Hom}(S,Q)(D) = 0.$$ Next, we tznsor fhe sequencf $\eqref{qukgientSES}$ with $S^\lur(D)$ atd apply $\pi_{ | , so it suffices to prove that vector is zero. functor ${\operatorname{Hom}}_{Y_0 \times three $${\operatorname{Hom}}_{Y_0 \times C} \Gamma \circ \pi_{1*} {\curly Hom}_{Y_0 \times C},$$ so to the vanishing of $${\operatorname{Ext}}^1_{Y_0 \times C}({\overline}{S},{\overline}{Q} \otimes \pi_1^* I) = {\operatorname{R}}^1 {\operatorname{Hom}}_{Y_0 \times \otimes \pi_1^* I),$$ it suffices (by the Grothendieck spectral sequence) to prove the of three spaces {\operatorname{H}}^1(Y_0, \pi_{1*} {\curly Hom}({\overline}{S},{\overline}{Q} \otimes \pi_1^* I) \\ {\operatorname{H}}^0(Y_0, \pi_{1*} {\curly Ext}^1({\overline}{S}, {\overline}{Q} \otimes \pi_1^* I)) {\operatorname{H}}^0(Y_0, {\operatorname{R}}^1 \pi_{1*} {\curly Hom}({\overline}{S},{\overline}{Q}) \otimes I)). \end{array}$$ first vanishes because $Y_0$ affine and the second vanishes ${\overline}{S}$ locally free. the vanishing, will show that \pi_{1*} {\curly Hom}({\overline}{S},{\overline}{Q})=0$. By cohomology and base change and the assumption that $Y_0 \to X$ factors through we might assume $Y_0=X$, ${\overline}{S}=S(-D)$. the together with cohomology change (and flatness of ${\curly Hom}(S,Q)$), that $${\operatorname{R}}^1 \pi_{1*} {\curly Hom}({\overline}{S},Q) = {\operatorname{R}}^1 \pi_{1*} Hom}(S,Q)(D) = Next, we tensor the sequence $\eqref{quotientSES}$ $S^\lor(D)$ and apply $\pi_{ | , so it suffices to prove that thIs whole vecTor spAce Is zErO. The FuncTor ${\operatornamE{hom}}_{Y_0 \Times C}$ is the composition Of thrEe FUnctORs: $${\OperaTorname{hOm}}_{y_0 \TImeS C} = \gaMma \CiRC \pI_{1*} \circ {\CurLy Hom}_{Y_0 \tImes C},$$ so to pRovE tHe vanishing oF $${\OpEratorname{ext}}^1_{y_0 \times C}({\overlIne}{s},{\overlInE}{Q} \oTImes \pI_1^* I) = {\oPeratOrname{r}}^1 {\OperatOrname{Hom}}_{y_0 \tIMes C}({\ovERline}{S},{\oVERlIne}{Q} \Otimes \pi_1^* I),$$ it sufficES (bY The GrothendiecK spectRaL SeQUEncE) to Prove the vaNiShing OF the thrEE vECTOr sPAces $$\begin{arraY}{l} {\operatornAMe{H}}^1(y_0, \pi_{1*} {\curLy hom}({\OVerlinE}{S},{\oveRlINe}{Q} \Otimes \pi_1^* I) \\ {\opEratOrname{H}}^0(Y_0, \pI_{1*} {\curly eXt}^1({\overlINe}{S}, {\overLine}{Q} \oTimEs \pI_1^* I)) \\ {\opERaToRnaMe{h}}^0(y_0, {\opERaTorNAme{r}}^1 \pi_{1*} {\curly hoM}({\oVerliNe}{S},{\oVERLIne}{Q}) \OtiMes I)). \End{arRay}$$ The first vaNisHes bECauSe $Y_0$ is AffinE and ThE secoNd vaniShes bEcAuse ${\overline}{S}$ is LocaLly free.
FoR thE tHirD vAnishINg, we wiLl sHow That ${\opeRatornaME{R}}^1 \pI_{1*} {\cURLY HOm}({\overline}{S},{\overlinE}{Q})=0$. bY CoHomology And basE ChAnGE and the aSsUmpTion THAt $Y_0 \to x$ facTOrS through $\Iota_D$, wE MiGhT as well AsSume $Y_0=X$, So ${\OveRliNe}{S}=S(-D)$. uSing The vanIshing toGetheR With cohomology ANd base change (aND fLATnESs of ${\CurLy Hom}(S,Q)$), we fiRst oBServE thaT $${\OpEraTOrnamE{R}}^1 \pi_{1*} {\cUrLY HOM}({\overline}{S},Q) = {\operatorNaMe{R}}^1 \pi_{1*} {\cUrly HOm}(S,Q)(D) = 0.$$ Next, we teNsor the seqUENCe $\eqref{qUotiENtseS}$ with $S^\lor(D)$ and Apply $\Pi_{ | , so it suffices to provethat thiswhole ve cto rspac e is zero. The fun c tor${\operatorname{Hom}}_ {Y_0\t i mesC }$ is t he comp o si t i onof t hre ef un ctors : $ ${\oper atorname{H om} }_ {Y_0 \timesC }= \Gamma \ cir c \pi_{1*} \ cir c {\cu rl y H o m}_{Y _0\time s C},$ $ so to prove th ev anishi n g of $$ { \ op erat orname{Ext}}^1_{Y _ 0\ times C}({\ove rline} {S } ,{ \ o ver lin e}{Q} \oti me s \pi _ 1^* I)= { \ o p era t orname{R}}^1{\operatorn a me{ Hom}}_ {Y _0\ timesC}({\ ov e rli ne}{S},{\ov erli ne}{Q} \o times\ pi_1^*I ),$$ it suffi ces (b y th e G ro the nd i eck sp ect r alsequence )to prov e th e v a nish ing ofthe t hree vector s pac es $ $ \be gin{a rray} {l}{\ opera tornam e{H}} ^1 (Y_0, \pi_{1*}{\cu rly Hom}( {\o ve rli ne }{S}, { \overl ine }{Q } \otim es \pi_ 1 ^*I) \ \{\operatorname{H}} ^0 ( Y _0 , \pi_{1 *} {\c u rl yE xt}^1({\ ov erl ine} { S }, {\ over l in e}{Q} \o times\ pi _1 ^* I))\\ {\ope ra tor nam e{H}} ^ 0(Y_ 0, {\o peratorn ame{R } }^1 \pi_{1*} { \ curly Hom}({\ o ve r l in e }{S} ,{\ overline}{Q }) \ o time s I) ) .\en d {arra y}$$Th e f i rst vanishes becaus e$Y_0$is af fine and thesecond van i s h es becau se $ { \o v erline}{S}$ is loca lly free.For thethird vanishi ng, we wi l l show th at${\ ope rat o r na me{R}}^1 \pi_ { 1 *} { \c urly Ho m}( {\overl ine }{S },{ \ov er line}{Q}) =0$. Byco ho mo lo gyand b a se chang eand t heassum p tion t hat $ Y_0\t oX $ f actorst hr o u gh $ \i ot a_D$ , w emight asw ell assume $Y_0=X$, so ${\o ve rl ine}{S} =S(-D)$. Usin gthe vanish in g t ogethe r with coh omology and base change (and fl atn ess o f ${ \curly Ho m}( S,Q)$) , w e first obser ve th at $$ { \ opera t o rn ame {R }}^1 \pi_{ 1 * } { \curl yHom} ({\over line}{S},Q) = {\op e rat orname{R}}^1\pi _{1* } {\ cur l yH om} (S , Q)( D ) = 0.$$ Next, w e tensor t he se quence $\e q ref {q uotient SES}$ w ith $ S ^\lor(D )$ and ap ply $\pi_ { | , so_it suffices_to prove that this_whole vector_space_is zero._The_functor ${\operatorname{Hom}}_{Y_0 \times_C}$ is the_composition of three functors:_$${\operatorname{Hom}}_{Y_0 \times C}_=_\Gamma \circ \pi_{1*} \circ {\curly Hom}_{Y_0 \times C},$$ so to prove the vanishing of_$${\operatorname{Ext}}^1_{Y_0_\times C}({\overline}{S},{\overline}{Q}_\otimes_\pi_1^*_I) = {\operatorname{R}}^1 {\operatorname{Hom}}_{Y_0 \times_C}({\overline}{S},{\overline}{Q} \otimes \pi_1^* I),$$ it_suffices (by_the Grothendieck spectral sequence) to prove the vanishing_of_the three vector_spaces $$\begin{array}{l} {\operatorname{H}}^1(Y_0, \pi_{1*} {\curly Hom}({\overline}{S},{\overline}{Q} \otimes \pi_1^* I)_ \\ {\operatorname{H}}^0(Y_0, \pi_{1*} {\curly Ext}^1({\overline}{S},_{\overline}{Q} \otimes \pi_1^*_I))_\\_{\operatorname{H}}^0(Y_0, {\operatorname{R}}^1 \pi_{1*} {\curly_Hom}({\overline}{S},{\overline}{Q}) \otimes I)). \end{array}$$ The first_vanishes because $Y_0$ is affine and_the second vanishes because ${\overline}{S}$ is locally_free.
For the third vanishing, we will_show that ${\operatorname{R}}^1 \pi_{1*} {\curly_Hom}({\overline}{S},{\overline}{Q})=0$. By_cohomology and base change and_the assumption that_$Y_0 \to_X$ factors through_$\iota_D$, we might as well assume_$Y_0=X$, so ${\overline}{S}=S(-D)$._Using the vanishing together with cohomology_and_base change (and_flatness_of_${\curly Hom}(S,Q)$),_we first observe_that_$${\operatorname{R}}^1 \pi_{1*}_{\curly_Hom}({\overline}{S},Q) = {\operatorname{R}}^1 \pi_{1*} {\curly Hom}(S,Q)(D)_=_0.$$ Next, we tensor the sequence $\eqref{quotientSES}$_with $S^\lor(D)$ and apply_$\pi_{ |
exceeds a density contrast of about $10^4$ a “sink” particle is created (Bate et al. 1995). It replaces the central high-density region and has the ability to accrete further infalling material while keeping track of mass and linear and angular momentum. However, the internal structure of the sink particle is not resolved. With a diameter of about 600 AU it fully encloses the star/disk system expected to form roughly 1000 years after the critical density for sink particle formation is reached (Wuchterl & Klessen 2000).
The numerical resolution limit of our numerical scheme is determined by the Bate & Burkert (1996) criterion, which is sufficient for the highly nonlinear fluctuation spectrum considered here. This is confirmed by resolutions studies with up to $10^7$ SPH particles (see Jappsen et al. 2004). It should be noted, however, that the Bate & Burkert (1996) criterion may not be sufficient for adequately following the growth of linear perturbations out of quasi-equilibrium, as was suggested for the case of rotationally supported disks by Fisher, Klein, & McKee (private communication).
We analyze two models, one labeled, in which turbulence is driven on large scales, of wavelength $\lambda \approx 1/2$ of the computational box, and the other labeled, in which energy is injected on smaller scales, of $\lambda\approx 1/8$ of the box. We consider the system at an evolutionary stage when 5% of the available gas mass is accumulated in collapsed cores. Note that in Paper I we also studied cores from a contracting Gaussian density field (GC) without turbulence. Since in this paper we focus on the turbulent velocity structure, that simulation is not considered here.
To identify cloud cores we use the 3-dimensional clump-finding algorithm introduced in Appendix A of Klessen & Burkert (2000). We then project a cubic subregion of the full computational volume centered around the core along the three principal axes, and compute the column density $N$ and the total, turbulent-plus-thermal l.o.s velocity dispersion $\sigma_{\rm los}$ of each core. We take $\sigma_{\rm los}^2 = \sigma_{\rm turb}^2 + c_{\rm s}^2$, where the turbulent velocity dispersion $\sigma_{\rm turb}$ is obtained as the mass-weighted standard deviation of the velocity field in each line-of-sight along the projection axis, and | exceeds a density contrast of about $ 10 ^ 4 $ a “ sinkhole ” atom is created (Bate et al. 1995). It replaces the cardinal high - concentration region and has the ability to accrete far infalling fabric while keeping track of mass and linear and angular momentum. However, the internal social organization of the sink particle is not resolved. With a diameter of about 600 AU it amply encloses the star / disk system expected to shape roughly 1000 years after the critical density for cesspool particle formation is reached (Wuchterl & Klessen 2000).
The numerical resolving power limit of our numerical scheme is determined by the Bate & Burkert (1996) criterion, which is sufficient for the highly nonlinear variation spectrum considered here. This is confirmed by resolutions studies with up to $ 10 ^ 7 $ SPH particles (see Jappsen et al. 2004). It should be note, however, that the Bate & Burkert (1996) criterion may not be sufficient for adequately following the growth of linear perturbations out of quasi - equilibrium, as was suggested for the case of rotationally supported disks by Fisher, Klein, & McKee (individual communication).
We analyze two models, one labeled, in which turbulence is drive on big scales, of wavelength $ \lambda \approx 1/2 $ of the computational box, and the early labeled, in which department of energy is injected on smaller scales, of $ \lambda\approx 1/8 $ of the corner. We consider the system at an evolutionary stage when 5% of the available gas mass is accumulated in crack up cores. Note that in Paper I we also studied core from a compress Gaussian concentration field (GC) without turbulence. Since in this paper we focus on the turbulent velocity structure, that simulation is not considered here.
To name cloud core we use the 3 - dimensional clump - receive algorithm introduced in Appendix A of Klessen & Burkert (2000). We then plan a cubic subregion of the full computational volume focus on around the core along the three chief axes, and compute the column density $ N$ and the total, churning - plus - thermal l.o.s velocity dispersion $ \sigma_{\rm los}$ of each core. We take $ \sigma_{\rm los}^2 = \sigma_{\rm turb}^2 + c_{\rm s}^2 $, where the turbulent speed dispersion $ \sigma_{\rm turb}$ is obtained as the mass - slant standard deviation of the velocity field in each line - of - sight along the project axis, and | exfeeds a density contrast of about $10^4$ a “siuj” partmcle is created (Bate et al. 1995). It replaces the renteal hugh-density region and fas the ahility ti acrrete further inhzlling materizp whnlx keeping track of mass atd linear and dneupar momentum. However, the internal seructurr lf the sink patticlt if nof resolved. With a diameter of abouf 600 AU iu fully encloses tne star/disk system expectef to form roughly 1000 yeags after thg cryrical densith for sink particle fotmation is reached (Wuchterl & Klersen 2000).
Che numericql redmlution limmt of jur numerical scheme is detrrmined by the Bave & Vurkert (1996) criterion, whmch is sufficient fot the highny nonlinear fluctuqtuon skectrgm cubsiaertd iers. This is confirmed by resolutuons studies with uk tj $10^7$ SPH particlss (see Jwppsen et al. 2004). It should be noted, however, thzt the Bate & Burkert (1996) ceiterion may not be sofficient sor adequately following the growth of linear perdurbaviuns omt ow qkasi-equilibrium, as was suggested for the case kf rptationally suiported disks by Fosjet, Klein, & McKee (privacs dommunication).
We anwlyze tro moeels, one jabeked, in which turbulence is eriven on lagge wcales, of wavelengch $\lambda \ap'rox 1/2$ pf thr computational box, and the other labeped, in whjzh energy is injdctvd ot smaller scales, of $\lambda\wpprox 1/8$ oh the box. We conxider ehe system at ak evolutionary stagf wheu 5% of the availwble gas mass is accumulated in collapsed corgs. Totv that in Papev I we also studyed cores from a contxactine Gaussian densitb field (GC) wythout turbulajce. Since in this pa[er qe ficus on ghe turbulent felocity structure, rhat simulation is noj donsidered here.
Cu identify cloud cofes wv uxe dhe 3-dimensiotal zluol-findkng algoritmm kntrpduced in Appendix A of Ilessen & Burkert (2000). Ee then priject a subic subregipn of the full comkutatimnau volime centered around the core alonf the thrfe irincipal axef, ane compute thz column density $N$ and the total, turbuleit-plus-thermal l.o.s velocuty dispersion $\sigmc_{\rk los}$ of earh corq. We take $\sigma_{\rm los}^2 = \sigma_{\rn turb}^2 + c_{\rm s}^2$, wheve the turbulent velocjty divperslon $\sigma_{\rm turb}$ is obtained as the mass-weighted standard deviation of thw veloriey field in eacn linv-of-fnght aljng vhx projection axis, and | exceeds a density contrast of about $10^4$ particle created (Bate al. 1995). It and the ability to further infalling material keeping track of mass and linear angular momentum. However, the internal structure of the sink particle is not resolved. a diameter of about 600 AU it fully encloses the star/disk system expected form 1000 after critical density for sink particle formation is reached (Wuchterl & Klessen 2000). The numerical resolution limit our numerical scheme is determined by the Bate Burkert (1996) criterion, which sufficient for the highly nonlinear spectrum here. This confirmed resolutions with up to SPH particles (see Jappsen et al. 2004). It should be noted, however, that the Bate & Burkert criterion may sufficient for following growth linear perturbations out as was suggested for the case disks by Fisher, Klein, & McKee (private communication). analyze two one labeled, in which turbulence is on large scales, of wavelength $\lambda \approx 1/2$ the computational box, and the other labeled, in which energy is injected on smaller scales, 1/8$ of the box. consider the system an stage 5% the available mass is accumulated in collapsed cores. Note that in Paper I also studied cores from a contracting Gaussian density field (GC) Since this paper we on the turbulent velocity that is not considered here. cloud we clump-finding introduced Appendix A of Klessen Burkert (2000). We then project cubic subregion of the the core along the three principal axes, and the column density $N$ and the total, l.o.s velocity dispersion $\sigma_{\rm los}$ of each core. We take $\sigma_{\rm los}^2 \sigma_{\rm turb}^2 s}^2$, where the turbulent velocity dispersion $\sigma_{\rm turb}$ obtained as the mass-weighted deviation of the velocity field in each line-of-sight along projection and | exceeds a density contrast of About $10^4$ a “sink” PartiCle Is cReAted (bate Et al. 1995). It replaces THe ceNtral high-density region And haS tHE abiLItY to acCrete fuRThER InfAlLiNg mAtERiAl whiLe kEeping tRack of mass And LiNear and angulAR mOmentum. HowEveR, the internal StrUcture Of The SInk paRtiCle is Not resOLved. WiTh a diametEr OF about 600 au it fullY ENcLoseS the star/disk systeM ExPEcted to form rouGhly 1000 yeArS AfTER thE crItical densItY for sINk partiCLe FORMatIOn is reached (WuChterl & KlessEN 2000).
ThE numerIcAl rESolutiOn limIt OF ouR numerical sChemE is determIned by THe Bate & BURkert (1996) crIterioN, whIch Is suFFiCiEnt FoR The HIgHly NOnlInear fluCtUaTion sPectRUM COnsiDerEd heRe. ThiS is confirmed bY reSoluTIonS studIes wiTh up To $10^7$ sPH paRticleS (see JApPsen et al. 2004). It shoulD be nOted, howevEr, tHaT thE BAte & BuRKert (1996) crIteRioN may not Be suffiCIenT fOR ADeQuately following thE gROWtH of lineaR pertuRBaTiONs out of qUaSi-eQuilIBRium, aS was SUgGested foR the caSE oF rOtationAlLy suppOrTed DisKs by FISher, klein, & MCKee (privAte coMMunication).
We anALyze two models, ONe LABeLEd, in WhiCh turbulencE is dRIven On laRGe ScaLEs, of wAveleNgTH $\lAMbda \approx 1/2$ of the compUtAtionaL box, aNd the other labEled, in whicH ENErgy is inJectED oN Smaller scales, oF $\lambDa\approx 1/8$ of THe box. We cOnsidEr the sysTem at an evOLUtionary StaGe wHen 5% Of tHE AvAilable gas masS IS accUmUlated iN coLlapsed CorEs. NOte ThaT iN Paper I we Also studIeD cOrEs FroM a conTRacting GAuSsiAn DenSity fIEld (GC) wIthouT turBuLeNCe. SInce in tHIs PAPer wE fOcUs on The TuRbuleNt veLOciTy strucTure, that sImuLAtioN iS nOt consiDered here.
To idEnTify cloud cOrEs wE use thE 3-DImensionAl clump-finding algorithm INtroducEd iN AppeNdix a of KlesseN & BuRkert (2000). WE thEN projeCt a cubIc subReGioN OF the fULL cOmpUtAtional volUME ceNtereD aRounD the corE along the three prinCIpaL axes, and compuTe tHe coLUMn DenSItY $n$ anD tHE toTAL, turbulent-plus-tHermal l.o.s vElOCiTy dispersiON $\siGmA_{\rm los}$ oF each coRe. We tAKe $\sigma_{\Rm los}^2 = \sigmA_{\rm turb}^2 + c_{\rM s}^2$, WherE THe tUrbulent veLocity diSpersion $\sIGma_{\rm TUrB}$ is obTaiNed as tHe MasS-weigHted stANdaRd devIation Of The velOcity FiEld in eacH line-of-sight along the proJectioN axis, And | exceeds a density contras t of about $10^ 4$a “ si nk”part icle is create d (Ba te et al. 1995). It re place st he c e nt ral h igh-den s it y reg io nand h a sthe a bil ity toaccrete fu rth er infalling m a te rial while ke eping trackofmass a nd li n ear a ndangul ar mom e ntum.However,th e inter n al stru c t ur e of the sink particl e i s not resolved. Withad ia m e ter of about 600 A U itf ully en c lo s e s th e star/disk sy stem expect e d t o form r oug h ly 100 0 yea rs aft er the crit ical densityfor si n k parti c le form ationisrea ched (W uc hte rl & K l es sen 200 0).
The n um erica l re s o l u tion li mitof ou r numerical s che me i s de termi ned b y th eBate& Burk ert ( 19 96) criterion,whic h is suff ici en t f or theh ighlynon lin ear flu ctuatio n sp ec t r u mconsidered here. T hi s is confirm ed byr es ol u tions st ud ies wit h up to $10 ^ 7$ SPH par ticles (s ee Jappse net al. 2 004 ).It sh o uldbe not ed, howe ver,t hat the Bate & Burkert (1996 ) c r i te r ionmay not be suf fici e nt f or a d eq uat e ly fo llowi ng th e growth of linear p er turbat ionsout of quasi- equilibriu m , as was s ugge s te d for the caseof ro tationally supporte d dis ks by Fi sher, Kle i n , & McKe e ( pri vat e c o m mu nication).
W e anal yz e two m ode ls, one la bel ed, in w hich turb ulence i sdr iv en on larg e scales, o f w av ele ngth$ \lambd a \ap prox 1 /2 $ of the co m pu t a tion al b ox,and t he ot herl abe led, in which en erg y isin je cted on smaller scal es , of $\lam bd a\a pprox1 / 8$ of th e box. We consider thes ystem a t a n evo luti onary sta gewhen 5 % o f the a vailab le ga smas s is ac c u mu lat ed in collap s e d c ores. N otethat in Paper I we also s t udi ed cores from acont r a ct ing Ga u ssi an den s i ty field (GC) w ithout tur bu l en ce. Sincei n t hi s paper we foc us on the tur bulent ve locity st ru ctur e , th at simulat ion is n ot consid e red h e re .
To id entify c lou d cor es weu sethe 3 -dimen si onal c lump- fi nding al gorithm introduced in A ppendi x A o f K lessen &Bur k ert (2000).We t hen projec t a cu bic s ubr e gionof t h eful l comp utat i onal volu m ecen t e re d around th e c ore alon g t h e thre e pr incipal axes, and compute the co lumn d ens ity $N$an d the total, t urb ul e n t-plus-t he rmal l.o.svelocity d i spers ion $\ sigma_ {\rm lo s } $o f each cor e.We take $ \si gm a _{\rm l os }^ 2 = \si gma_ {\ rm tur b}^2 + c_{\ r m s}^2$, where th e tur b u lentv elo citydi spersio n $\s igma_{\rmturb}$ is o btaine d as themass-we ig hted s tan da rd deviati o n of theveloc ity fie ld ineac h line -of- s i ght a long t heprojectio n ax i s, a n d | exceeds_a density_contrast of about $10^4$_a “sink”_particle_is created_(Bate_et al. 1995). It_replaces the central_high-density region and has_the ability to_accrete_further infalling material while keeping track of mass and linear and angular momentum. However,_the_internal structure_of_the_sink particle is not resolved._With a diameter of about_600 AU_it fully encloses the star/disk system expected to_form_roughly 1000 years_after the critical density for sink particle formation is_reached (Wuchterl & Klessen 2000).
The numerical_resolution limit of_our_numerical_scheme is determined by_the Bate & Burkert (1996) criterion,_which is sufficient for the highly_nonlinear fluctuation spectrum considered here. This is_confirmed by resolutions studies with up_to $10^7$ SPH particles (see_Jappsen et_al. 2004). It should be noted,_however, that the_Bate &_Burkert (1996) criterion_may not be sufficient for adequately_following the growth_of linear perturbations out of quasi-equilibrium,_as_was suggested for_the_case_of rotationally_supported disks by_Fisher,_Klein, &_McKee_(private communication).
We analyze two models, one_labeled,_in which turbulence is driven on large_scales, of wavelength $\lambda_\approx_1/2$ of the computational_box, and the other labeled,_in which energy is injected on_smaller scales,_of $\lambda\approx_1/8$ of the box. We consider the system at an evolutionary_stage when 5% of the available_gas mass is accumulated_in collapsed_cores._Note that in_Paper I_we also_studied cores from a contracting Gaussian density_field (GC)_without turbulence. Since in this paper_we focus on the_turbulent_velocity structure, that simulation is not_considered here.
To identify cloud cores we_use the 3-dimensional clump-finding algorithm_introduced_in_Appendix A of Klessen &_Burkert (2000). We then project a_cubic subregion of_the full computational volume centered around the_core_along the three principal axes, and_compute_the column density $N$ and the_total,_turbulent-plus-thermal_l.o.s velocity dispersion $\sigma_{\rm los}$_of each core. We take $\sigma_{\rm_los}^2 = \sigma_{\rm turb}^2 + c_{\rm s}^2$, where the_turbulent velocity dispersion_$\sigma_{\rm turb}$ is obtained as_the_mass-weighted_standard deviation of the velocity field in each line-of-sight along_the projection_axis, and |
-label="figure3.11"}](fig22_hyperEPP3.eps){width="2.7in"}
*The second step of the hyper-EPP* — Here, four identical nonlocal photon pairs $AB$, $CD$, $A'B'$, and $C'D'$ are required. The photons $A$, $C$, $A'$, and $C'$ are obtained by Alice, and the photons $B$, $D$, $B'$, and $D'$ are obtained by Bob. In the first step, the same operations are performed on the four-photon systems $ABCD$ and $A'B'C'D'$. If the results of the P-S-QNDs showed that the four-photon systems $ABCD$ and $A'B'C'D'$ are projected to the cases (3) and (4) in the first step, respectively, the QSJM is introduced to combine the polarization state of the photon pair $AB$ and the spatial-mode state of the photon pair $A'B'$ into an output single photon-pair state [@HEPP1]. So the preserving condition of the case (1) in the first step is achieved.
If the results of the P-S-QNDs showed that the four-photon systems $ABCD$ and $A'B'C'D'$ are projected to the cases (4) and (3) in the first step, respectively, the QSJM is introduced to combine the spatial-mode state of the photon pair $AB$ and the polarization state of the photon pair $A'B'$ into an output single photon-pair state. So the preserving condition of the case (1) in the first step is also achieved.
![ The fidelity of the hyperentangled Bell state obtained in the hyper-EPP versus the fidelity of the initial mixed hyperentangled Bell state ($F_1$) under the iteration number ($n$) [@HEPP1]. Here, the parameters of the initial mixed hyperentangled Bell state are $F_1=F_2$. []{data-label="figure3.7"}](fig23_fidelity.eps){width="3in"}
*Fidelity and efficiency* — After the two steps of hyper-EPP are performed on the nonlocal photon systems, the final state of the photon pair $AB$ is obtained as $$\begin{aligned}
% Eq. | -label="figure3.11"}](fig22_hyperEPP3.eps){width="2.7 in " }
* The second step of the hyper - EPP * — Here, four identical nonlocal photon pairs $ AB$, $ CD$, $ A'B'$, and $ C'D'$ are required. The photons $ A$, $ C$, $ A'$, and $ C'$ are prevail by Alice, and the photon $ B$, $ D$, $ B'$, and $ D'$ are obtained by Bob. In the first gradation, the like operations are performed on the four - photon systems $ ABCD$ and $ A'B'C'D'$. If the consequence of the P - S - QNDs showed that the four - photon systems $ ABCD$ and $ A'B'C'D'$ are visualize to the cases (3) and (4) in the first footfall, respectively, the QSJM is introduced to combine the polarization country of the photon couple $ AB$ and the spatial - mode state of the photon couple $ A'B'$ into an output individual photon - couple state [ @HEPP1 ]. So the preserving circumstance of the case (1) in the first step is achieved.
If the results of the P - S - QNDs show that the four - photon systems $ ABCD$ and $ A'B'C'D'$ are projected to the cases (4) and (3) in the first step, respectively, the QSJM is introduce to combine the spatial - mode state of the photon pair $ AB$ and the polarization state of the photon couple $ A'B'$ into an end product single photon - pair state. So the preserving condition of the case (1) in the first step is besides achieve.
! [ The fidelity of the hyperentangled Bell state obtained in the hyper - EPP versus the fidelity of the initial mixed hyperentangled Bell department of state ($ F_1 $) under the iteration number ($ n$) [ @HEPP1 ]. Here, the parameters of the initial mixed hyperentangled Bell department of state are $ F_1 = F_2$. [ ] { data - label="figure3.7"}](fig23_fidelity.eps){width="3 in " }
* Fidelity and efficiency * — After the two steps of hyper - EPP are performed on the nonlocal photon systems, the concluding state of the photon pair $ AB$ is obtained as $ $ \begin{aligned }
% Eq. | -labfl="figure3.11"}](fig22_hyperEPP3.eps){wiath="2.7in"}
*The second step oh the hgper-EPP* — Here, four identical nonlocap photob pairs $AB$, $CD$, $A'B'$, and $C'A'$ are reqlired. The phouons $A$, $C$, $A'$, and $C'$ ede obtalued bg Aliee, and the photoks $B$, $D$, $B'$, ang $D'$ are obtainad bv Bob. In the first step, the same opewations age performed og tht fjur-pgoton systems $ABCD$ and $A'B'C'D'$. If the resultv of the P-S-QNCs showed that the four-phohon dystems $ABCD$ and $A'H'C'D'$ are prohectqe to the casds (3) and (4) in the first atep, respectively, the QSJM is ivtrodbced to comvibe hve polarizavion snate of the pmpton pdir $AB$ snd the spatiak-move srate of the photon pamr $A'B'$ into an output single pvocon-pair state [@HEPP1]. So tye prgservhng zindktikn or the fasx (1) in the fjrst step iw achieved.
If the rexujnx of the P-S-QHDs shjwqd that the four-photon systems $ABCD$ and $A'B'D'D'$ are projected to the cases (4) and (3) in the flrst step, respectively, the QSJM is introduced to combine tve spetkal-noqd shate of the photon pair $AB$ and the polarizatijh xtste of the phobon pair $A'B'$ into am luykut single phojon-pair stzte. So the preservlng conqitiob of the sase (1) in the first step is also achieved.
![ Thv fieelity of the hypexentangled Bzll stste ontained in the hyper-EPP veraus the fidflity of ffe initial mixed hyierettangled Bell state ($F_1$) undew the itecatiou number ($n$) [@HGPP1]. Herq, the paraleters of the initial mided hiperendangled Bepl state are $F_1=F_2$. []{data-label="figure3.7"}](hmg23_fidelity.eps){eigth="3pn"}
*Fidelitv and cfficiency* — Aftqr the two steks of hypzr-EPP xre perforjed on vhe nonlocal photon systekd, the final vtate of the phoron paif $AB$ is obtainec as $$\begiu{cligned}
% Eq. | -label="figure3.11"}](fig22_hyperEPP3.eps){width="2.7in"} *The second step of the hyper-EPP* four nonlocal photon $AB$, $CD$, $A'B'$, photons $C$, $A'$, and are obtained by and the photons $B$, $D$, $B'$, $D'$ are obtained by Bob. In the first step, the same operations are on the four-photon systems $ABCD$ and $A'B'C'D'$. If the results of the P-S-QNDs that four-photon $ABCD$ $A'B'C'D'$ are projected to the cases (3) and (4) in the first step, respectively, the QSJM introduced to combine the polarization state of the pair $AB$ and the state of the photon pair into output single state So preserving condition of case (1) in the first step is achieved. If the results of the P-S-QNDs showed that the systems $ABCD$ are projected the (4) (3) in the respectively, the QSJM is introduced to state of the photon pair $AB$ and the state of photon pair $A'B'$ into an output photon-pair state. So the preserving condition of the (1) in the first step is also achieved. ![ The fidelity of the hyperentangled Bell in the hyper-EPP versus fidelity of the mixed Bell ($F_1$) the iteration ($n$) [@HEPP1]. Here, the parameters of the initial mixed hyperentangled Bell are $F_1=F_2$. []{data-label="figure3.7"}](fig23_fidelity.eps){width="3in"} *Fidelity and efficiency* — After the two hyper-EPP performed on the photon systems, the final of photon pair $AB$ is $$\begin{aligned} Eq. | -label="figure3.11"}](fig22_hyperEPP3.eps){Width="2.7in"}
*The SeconD stEp oF tHe hyPer-EpP* — Here, four idenTIcal Nonlocal photon pairs $AB$, $Cd$, $A'B'$, anD $C'd'$ Are rEQuIred. THe photoNS $A$, $c$, $a'$, And $c'$ aRe ObtAiNEd By AliCe, aNd the phOtons $B$, $D$, $B'$, anD $D'$ aRe Obtained by BoB. in The first stEp, tHe same operatIonS are peRfOrmED on thE foUr-phoTon sysTEms $ABCd$ and $A'B'C'D'$. IF tHE resulTS of the P-s-qnDS shoWed that the four-phoTOn SYstems $ABCD$ and $A'b'C'D'$ are PrOJeCTEd tO thE cases (3) and (4) iN tHe firST step, reSPeCTIVelY, The QSJM is intrOduced to comBIne The polArIzaTIon staTe of tHe PHotOn pair $AB$ and The sPatial-modE state OF the phoTOn pair $A'b'$ into aN ouTpuT sinGLe PhOtoN-pAIr sTAtE [@HEpp1]. So The preseRvInG condItioN OF THe caSe (1) iN the First Step is achieveD.
If The rESulTs of tHe P-S-QnDs sHoWed thAt the fOur-phOtOn systems $ABCD$ anD $A'B'C'd'$ are projeCteD tO thE cAses (4) aND (3) in the FirSt sTep, respEctivelY, The qSjm IS iNtroduced to combine ThE SPaTial-mode State oF ThE pHOton pair $aB$ And The pOLArizaTion STaTe of the pHoton pAIr $a'B'$ Into an oUtPut sinGlE phOtoN-pair STate. so the pReservinG condITion of the case (1) iN The first step iS AlSO AcHIeveD.
![ ThE fidelity of The hYPereNtanGLeD BeLL statE obtaInED iN The hyper-EPP versus thE fIdelitY of thE initial mixed HyperentanGLED Bell staTe ($F_1$) uNDeR The iteration nuMber ($n$) [@hEPP1]. Here, thE ParameteRs of tHe initiaL mixed hypEREntangleD BeLl sTatE arE $f_1=f_2$. []{dAta-label="figurE3.7"}](FIg23_fiDeLity.eps){WidTh="3in"}
*FidEliTy aNd eFfiCiEncy* — After The two stEpS oF hYpEr-EpP are PErformed On The NoNloCal phOTon sysTems, tHe fiNaL sTAte Of the phOToN PAir $Ab$ iS oBtaiNed As $$\Begin{AligNEd}
% EQ. | -label="figure3.11"}](fig2 2_hyperEPP 3.eps ){w idt h= "2.7 in"}
*The seconds tepof the hyper-EPP* — He re, f ou r ide n ti cal n onlocal ph o t onpa ir s $ AB $ ,$CD$, $A 'B'$, a nd $C'D'$are r equired. The ph otons $A$, $C $, $A'$, and $C '$ are o bta i ned b y A lice, and t h e phot ons $B$,$D $ , $B'$ , and $D ' $ a re o btained by Bob. I n t h e first step,the sa me op e r ati ons are perfo rm ed on the fou r -p h o t ons ystems $ABCD$ and $A'B'C ' D'$ . If t he re s ults o f the P - S-Q NDs showedthat the four -photo n system s $ABCD$ and $ A'B 'C' D'$a re p roj ec t edt othe cas es (3) a nd ( 4) in the f i r st s tep , re spect ively, the QS JMis i n tro duced to c ombi ne thepolari zatio nstate of the ph oton pair $AB $ a nd th espati a l-mode st ate of the photon pai r$ A ' B' $ into an output s in g l ephoton-p air st a te [ @ HEPP1].So th e pr e s ervin g co n di tion ofthe ca s e(1 ) in th efirstst episachie v ed.
If th e result s oft he P-S-QNDs sh o wed that thef ou r - ph o tonsys tems $ABCD$ and $A'B 'C'D ' $are proje ctedto th e cases (4) and (3)in the f irststep, respect ively, the Q S JM is in trod u ce d to combine th e spa tial-modes tate ofthe p hoton pa ir $AB$ a n d the pol ari zat ion st a t eof the photon p air$A 'B'$ in toan outp utsin gle ph ot on-pair s tate. So t he p re ser vingc ondition o f t he ca se (1 ) in th e fir st s te pi s a lso ach i ev e d .
! [Th e fi del it y oftheh ype rentang led Bellsta t e ob ta in ed in t he hyper-EPPve rsus the f id eli ty oft h e initia l mixed hyperentangledB ell sta te($F_1 $) u nder theite ration nu m ber ($ n$) [@ HEPP1 ]. He r e , the p ar ame te rs of thei n iti al mi xe d hy perenta ngled Bell state a r e $ F_1=F_2$. []{ dat a-la b e l= "fi g ur e 3.7 "} ] (fi g 2 3_fidelity.eps) {width="3i n" }
*Fidelitya ndef ficienc y* — Af ter t h e two s teps of h yper-EPPar e pe r f orm ed on thenonlocal photon s y stems , t he fi nal state o f t he ph oton p a ir$AB$is obt ai ned as $$\b eg in{align ed}
% E q. | -label="figure3.11"}](fig22_hyperEPP3.eps){width="2.7in"}
*The second_step of_the hyper-EPP* — Here,_four identical_nonlocal_photon pairs_$AB$,_$CD$, $A'B'$, and_$C'D'$ are required._The photons $A$, $C$,_$A'$, and $C'$_are_obtained by Alice, and the photons $B$, $D$, $B'$, and $D'$ are obtained by_Bob._In the_first_step,_the same operations are performed_on the four-photon systems $ABCD$_and $A'B'C'D'$._If the results of the P-S-QNDs showed that_the_four-photon systems $ABCD$_and $A'B'C'D'$ are projected to the cases (3) and_(4) in the first step, respectively,_the QSJM is_introduced_to_combine the polarization state_of the photon pair $AB$ and_the spatial-mode state of the photon_pair $A'B'$ into an output single photon-pair_state [@HEPP1]. So the preserving condition_of the case (1) in_the first_step is achieved.
If the results_of the P-S-QNDs_showed that_the four-photon systems_$ABCD$ and $A'B'C'D'$ are projected to_the cases (4)_and (3) in the first step,_respectively,_the QSJM is_introduced_to_combine the_spatial-mode state of_the_photon pair_$AB$_and the polarization state of the_photon_pair $A'B'$ into an output single photon-pair_state. So the preserving_condition_of the case (1)_in the first step is_also achieved.
![ The fidelity of the_hyperentangled Bell_state obtained_in the hyper-EPP versus the fidelity of the initial mixed hyperentangled_Bell state ($F_1$) under the iteration_number ($n$) [@HEPP1]. Here,_the parameters_of_the initial mixed_hyperentangled_Bell state_are $F_1=F_2$. []{data-label="figure3.7"}](fig23_fidelity.eps){width="3in"}
*Fidelity and efficiency* — After_the two_steps of hyper-EPP are performed on_the nonlocal photon systems,_the_final state of the photon pair_$AB$ is obtained as $$\begin{aligned}
_ ___ _ _ _ % Eq. |
> 0,
$ but this contradicts the above claim. Therefore we have proved that for any $\alpha \in [0,1)$ that $
\beta_c(\pmb{J},\pmb{\hat{h}})=+\infty,
$ which implies by Theorem \[teo-unicidade-gibbs-percolacao\] that $\vert \mathscr{G}^{\mathrm{{Spin}}}_{\beta}\vert =1$ for any $\beta>0$.
#### Acknowledgements.
The authors thank Aernout van Enter for fruitful discussions. We also acknowledge the anonymous referees for their helpful comments, suggestions and references provided in their reports. Leandro Cioletti is supported by FEMAT and Roberto Vila is supported by CNPq.
[vdBK85]{}
. *Translation Invariance and Instability of Phase Coexistence in the Two-dimensional Ising System*. Comm. Math. Phys., v. 73, p.83–94, [**1980**]{}.
. *The Phase Transition in a General Class of [I]{}sing-type Models is Sharp*. J. Statist. Phys., v. 47, p. 343–374, [**1987**]{}.
. *Discontinuity of the Magnetization in One-dimesional $1/|x-y|^2$ Ising and Potts Models*. J. Stat. Phys., v. 50, p. 1–40, [**1988**]{}.
. *Ising Model in Half-space: a Series of Phase Transitions in Low Magnetic Fields*. Theor. Math. Phys., v. 153, p. 1539–1574, [**2007**]{}.
. *The Self-dual Point of the Two-dimensional Random-Cluster Model is Critical for* $q\geqslant 1$. Probab. Theory Relat. Fields, v. 153, p. 511–542, [**2012**]{}.
. *Percolation Beyond [$\mathbb{Z}^d$]{}, Many Questions and a Few Answers*. Selected works of [O]{}ded [S]{}chramm. v. 1, 2, Sel. Works Probab. Stat., p. 679–690, Springer, [**2011**]{}.
. *Gibbs States of Graphical Representations of the Potts Model with External Fields*. J. Math. Phys., v. 41, p. 1170–1210, [**2000 | > 0,
$ but this contradicts the above claim. Therefore we have proved that for any $ \alpha \in [ 0,1)$ that $
\beta_c(\pmb{J},\pmb{\hat{h}})=+\infty,
$ which entail by Theorem \[teo - unicidade - gibbs - percolacao\ ] that $ \vert \mathscr{G}^{\mathrm{{Spin}}}_{\beta}\vert = 1 $ for any $ \beta>0$.
# # # # recognition.
The authors thank Aernout van Enter for fruitful discussions. We also notice the anonymous referees for their helpful comments, suggestion and reference provided in their reports. Leandro Cioletti is supported by FEMAT and Roberto Vila is supported by CNPq.
[ vdBK85 ] { }
. * Translation Invariance and Instability of Phase Coexistence in the Two - dimensional Ising System *. Comm. Math. Phys. , v. 73, p.83–94, [ * * 1980 * * ] { }.
. * The Phase Transition in a General Class of [ I]{}sing - character Models is Sharp *. J. Statist. Phys. , v. 47, p. 343–374, [ * * 1987 * * ] { }.
. * Discontinuity of the Magnetization in One - dimesional $ 1/|x - y|^2 $ Ising and Potts Models *. J. Stat. Phys. , v. 50, p. 1–40, [ * * 1988 * * ] { }.
. * Ising Model in Half - space: a Series of Phase Transitions in Low Magnetic Fields *. Theor. Math. Phys. , v. 153, p. 1539–1574, [ * * 2007 * * ] { }.
. * The Self - dual Point of the Two - dimensional Random - Cluster Model is Critical for * $ q\geqslant 1$. Probab. Theory Relat. Fields, v. 153, p. 511–542, [ * * 2012 * * ] { }.
. * Percolation Beyond [ $ \mathbb{Z}^d$ ] { }, Many doubt and a Few Answers *. Selected work of [ O]{}ded [ S]{}chramm. v. 1, 2, Sel. Works Probab. Stat. , p. 679–690, Springer, [ * * 2011 * * ] { }.
. * Gibbs States of Graphical Representations of the Potts Model with External Fields *. J. Math. Phys. , v. 41, p. 1170–1210, [ * * 2000 | > 0,
$ bkt this contradicts the xbove claim. Thetedore wx have lroved tfat for any $\alpha \in [0,1)$ that $
\bxta_c(\pmb{J},\pnb{\hat{h}})=+\infty,
$ which implkes by Thvorem \[teo-ynicmdade-gibbs-percoledao\] thab $\verf \matksrr{G}^{\mathrm{{Spin}}}_{\beja}\vert =1$ for dny $\beta>0$.
#### Acknoflddyements.
The authors thank Aernout van Enter gog fruitful dissusspogs. Ws also acknowledge the anonymous rsferees for their hekpful comments, suggestions and references providfd in their repjets. Leandro Zioletti is supported gy FEMAT and Roberto Vila is suoportzd by CNPq.
[veBJ85]{}
. *Tgdnslation Iivariagce and Instability mf Phasr Coexistence ln thx Twi-dimensional Ising Syvtem*. Comm. Math. Phyf., v. 73, p.83–94, [**1980**]{}.
. *Tve Phase Transition ib a Ggneran Clxws uf [J]{}smng-fype Mldems is Sharl*. J. Statist. Phys., v. 47, p. 343–374, [**1987**]{}.
. *Disconuintpyy of the Mafnetizwtyon in One-dimesional $1/|x-y|^2$ Ising and Potts Mosels*. J. Stat. Phys., v. 50, p. 1–40, [**1988**]{}.
. *Ising Model in Half-skace: a Seryes of Phase Transitions in Low Magnetic Fields*. Tveor. Jxth. Pmys., x. 153, o. 1539–1574, [**2007**]{}.
. *The Self-dual Point of the Two-dimensional Wzncok-Cluster Model is Criticak vot* $q\geqslant 1$. Ptobab. Tksody Relat. Fields, v. 153, p. 511–542, [**2012**]{}.
. *Petcolatuon Beyonq [$\mayhbb{Z}^d$]{}, Many Questions and a Few Answers*. Welected works of [L]{}ded [S]{}chramo. v. 1, 2, Sel. Works Probab. Stat., p. 679–690, S'ringed, [**2011**]{}.
. *Gibbs Stwtes of Gdxphical Represengatpons of the Kutts Model with Evternal Fmelds*. J. Math. Phyx., v. 41, p. 1170–1210, [**2000 | > 0, $ but this contradicts the Therefore have proved for any $\alpha $ implies by Theorem that $\vert \mathscr{G}^{\mathrm{{Spin}}}_{\beta}\vert for any $\beta>0$. #### Acknowledgements. The thank Aernout van Enter for fruitful discussions. We also acknowledge the anonymous referees their helpful comments, suggestions and references provided in their reports. Leandro Cioletti is by and Vila supported by CNPq. [vdBK85]{} . *Translation Invariance and Instability of Phase Coexistence in the Two-dimensional Ising Comm. Math. Phys., v. 73, p.83–94, [**1980**]{}. . Phase Transition in a Class of [I]{}sing-type Models is J. Phys., v. p. [**1987**]{}. *Discontinuity of the in One-dimesional $1/|x-y|^2$ Ising and Potts Models*. J. Stat. Phys., v. 50, p. 1–40, [**1988**]{}. . *Ising in Half-space: of Phase in Magnetic Theor. Math. Phys., p. 1539–1574, [**2007**]{}. . *The Self-dual Two-dimensional Random-Cluster Model is Critical for* $q\geqslant 1$. Theory Relat. v. 153, p. 511–542, [**2012**]{}. . Beyond [$\mathbb{Z}^d$]{}, Many Questions and a Few Answers*. works of [O]{}ded [S]{}chramm. v. 1, 2, Sel. Works Probab. Stat., p. 679–690, Springer, [**2011**]{}. States of Graphical Representations the Potts Model External J. Phys., 41, p. [**2000 | > 0,
$ but this contradicts the abovE claim. TherEfore We hAve PrOved That For any $\alpha \in [0,1)$ tHAt $
\beTa_c(\pmb{J},\pmb{\hat{h}})=+\infty,
$ whiCh impLiES by THEoRem \[teO-unicidADe-GIBbs-PeRcOlaCaO\] ThAt $\verT \maThscr{G}^{\mAthrm{{Spin}}}_{\bEta}\VeRt =1$ for any $\beta>0$.
#### aCkNowledgemeNts.
the authors thAnk aernouT vAn ENTer foR frUitfuL discuSSions. WE also acknOwLEdge thE AnonymoUS ReFereEs for their helpful COmMEnts, suggestionS and reFeREnCES prOviDed in their RePorts. lEandro CIOlETTI is SUpported by FEMaT and RobertO vilA is supPoRteD By CNPq.
[VdBK85]{}
. *TRaNSlaTion InvariaNce aNd InstabiLity of pHase CoeXIstence In the TWo-dImeNsioNAl isIng sySTem*. cOmM. MaTH. PhYs., v. 73, p.83–94, [**1980**]{}.
. *The PHaSe transItioN IN A geneRal clasS of [I]{}sIng-type Models Is SHarp*. j. staTist. PHys., v. 47, p. 343–374, [**1987**]{}.
. *discOnTinuiTy of thE MagnEtIzation in One-dimEsioNal $1/|x-y|^2$ IsinG anD POttS MOdels*. j. stat. PhYs., v. 50, P. 1–40, [**1988**]{}.
. *IsIng ModeL in Half-SPacE: a sERIeS of Phase TransitionS iN lOw magnetic fields*. tHeOr. mAth. Phys., v. 153, P. 1539–1574, [**2007**]{}.
. *THe SElf-dUAL PoinT of tHE TWo-dimensIonal RANdOm-cluster moDel is CRiTicAl fOr* $q\geQSlanT 1$. ProbaB. Theory RElat. FIElds, v. 153, p. 511–542, [**2012**]{}.
. *PercolatIOn Beyond [$\mathbB{z}^d$]{}, mANy qUestIonS and a Few AnsWers*. sElecTed wORkS of [o]{}Ded [S]{}cHramm. V. 1, 2, SEL. WORks Probab. Stat., p. 679–690, SprinGeR, [**2011**]{}.
. *Gibbs stateS of Graphical REpresentatIONS of the PoTts MODeL With External FiElds*. J. math. Phys., v. 41, p. 1170–1210, [**2000 | > 0,
$ but this contradict s the abov e cla im. Th er efor e we have proved t h at f or any $\alpha \in [0, 1)$ t ha t $
\ b et a_c(\ pmb{J}, \ pm b { \ha t{ h} })= +\ i nf ty,
$ wh ich imp lies by Th eor em \[teo-unici d ad e-gibbs-pe rco lacao\] that $\ vert \ ma ths c r{G}^ {\m athrm {{Spin } }}_{\b eta}\vert = 1 $ fora ny $\be t a >0 $.
#### Acknowledgem e nt s .
The authors thank A e rn o u t v anEnter forfr uitfu l discus s io n s . We also acknowle dge the ano n ymo us ref er ees for th eir h el p ful comments,sugg estions a nd ref e rencesp rovided in th eir re port s .Le and ro Cio l et tii s s upported b yFEMAT and R o b erto Vi la i s sup ported by CNP q.
[vd B K85 ]{}
. *Tr ansl at ion I nvaria nce a nd Instability of Pha se Coexis ten ce in t he Tw o -dimen sio nal IsingSystem* . Co mm . M at h. Phys., v. 73, p .8 3 – 94 , [**198 0**]{} .
.* The Phas eTra nsit i o n ina Ge n er al Class of [I ] {} si ng-type M odelsis Sh arp *. J. Stat ist. P hys., v. 47,p . 343–374, [** 1 987**]{}.
.* Di s c on t inui tyof the Magn etiz a tion inO ne -di m esion al $1 /| x -y | ^2$ Ising and Potts M odels* . J.Stat. Phys.,v. 50, p.1 – 4 0, [**19 88** ] {} .
. *Ising Mod el in Half-spac e : a Seri es of Phase T ransition s in Low M agn eti c F iel d s *. Theor. Math. P hys. ,v. 153, p. 1539–1 574 , [ **2 007 ** ]{}.
. * The Self -d ua lPo int of t h e Two-di me nsi on alRando m -Clust er Mo delis C r iti cal for * $ q \ geqs la nt 1$. Pr ob ab. T heor y Re lat. Fi elds, v.153 , p.51 1– 542, [* *2012**]{}.
.*Percolati on Be yond [ $ \ mathbb{Z }^d$]{}, Many Questions and a F ewAnswe rs*. Selected wo rks of [O ] {}ded[S]{}c hramm .v.1 , 2, S e l .Wor ks Probab. S t a t., p. 6 79 –690 , Sprin ger, [**2011**]{}.
.*Gibbs States of Gra p h ic alR ep r ese nt a tio n s of the Potts M odel withEx t er nal Fields * . J .Math. P hys., v . 41, p. 1170 –1210, [* *2000 | > 0,
$_but this_contradicts the above claim._Therefore we_have_proved that_for_any $\alpha \in_[0,1)$ that $
\beta_c(\pmb{J},\pmb{\hat{h}})=+\infty,
$_which implies by Theorem_\[teo-unicidade-gibbs-percolacao\] that $\vert_\mathscr{G}^{\mathrm{{Spin}}}_{\beta}\vert_=1$ for any $\beta>0$.
#### Acknowledgements.
The authors thank Aernout van Enter for fruitful discussions. We_also_acknowledge the_anonymous_referees_for their helpful comments, suggestions_and references provided in their_reports. Leandro_Cioletti is supported by FEMAT and Roberto Vila_is_supported by CNPq.
[vdBK85]{}
._*Translation Invariance and Instability of Phase Coexistence in the_Two-dimensional Ising System*. Comm. Math. Phys.,_v. 73, p.83–94,_[**1980**]{}.
._*The_Phase Transition in a_General Class of [I]{}sing-type Models is_Sharp*. J. Statist. Phys., v. 47,_p. 343–374, [**1987**]{}.
. *Discontinuity of the Magnetization_in One-dimesional $1/|x-y|^2$ Ising and Potts_Models*. J. Stat. Phys., v._50, p._1–40, [**1988**]{}.
. *Ising Model in_Half-space: a Series_of Phase_Transitions in Low_Magnetic Fields*. Theor. Math. Phys., v._153, p. 1539–1574,_[**2007**]{}.
. *The Self-dual Point of the_Two-dimensional_Random-Cluster Model is_Critical_for*_$q\geqslant 1$._Probab. Theory Relat._Fields,_v. 153,_p._511–542, [**2012**]{}.
. *Percolation Beyond [$\mathbb{Z}^d$]{}, Many_Questions_and a Few Answers*. Selected works of_[O]{}ded [S]{}chramm. v. 1,_2,_Sel. Works Probab. Stat.,_p. 679–690, Springer, [**2011**]{}.
. *Gibbs_States of Graphical Representations of the_Potts Model_with External_Fields*. J. Math. Phys., v. 41, p. 1170–1210, [**2000 |
with $\Delta P$ spanning 4 decades — also merits attention. To some degree, it could be a result of the small size of the sample (7 objects) used to derive $\zeta$. On the other hand, there may be some other explanations, e.g. the aforementioned long-term spin evolution converging to some “attractor state” [@Neishtadt].
This discussion makes it clear that measurements of $\zeta$ have a potential to help constrain the physics of cometary outgassing — its geometry, intensity, etc. — at least in a statistical sense. Better characterization of the distribution of $\zeta$ can have substantial impact on our understanding of cometary activity. This provides strong motivation for increasing the number of objects with measured $\Delta P$ and $a_{\rm ng}$. Right now the most basic properties of this distribution represented by equation (\[eq:stats\]) are based on a sample of only 7 objects. Targeted observations of many other comets aimed at measuring their spin variations would allow us to characterize the distribution of $\zeta$ much better. According to Figure \[fig:comets\], even period measurements with uncertainly of tens of minutes may reveal spin variability for dozens of comets with high $\Delta\Omega_1$ (the ones above the blue line in this figure).
The objects with highest $\Delta\Omega_1$ listed in the Table \[table:fast\] should have the best chance for revealing such changes, even though only one comet from Table \[table:zeta\] (41P/Tuttle-Giacobini-Kresák) has $\Delta\Omega_1>10^{-3}$ s$^{-1}$ necessary for being listed in Table \[table:fast\]. Although we focused on studying the spin period variability of comets, non-gravitational forced precession [@Whipple; @Sekanina1984] should also be substantial for these objects, likely driving them into excited rotational state.
Implications for Solar System comets {#sect:disc_SS}
------------------------------------
Our calculations demonstrate that many Solar System comets (the ones with high $\Delta\Omega_1$) should exhibit rapid evolution of their rotation. Using equations (\[eq:spinev\]) and (\[eq:a\_ng\]) one can estimate the characteristic timescale for the cometary spin evolution $\tau_\Omega=|\Omega/\dot\Omega|$ as \_&=& \^[-1]{} \[eq:tau\_Om\]\
&& 670 | with $ \Delta P$ spanning 4 decades — also merits care. To some academic degree, it could be a consequence of the small size of the sample (7 aim) used to derive $ \zeta$. On the early hand, there may be some other explanations, for example the aforesaid long - term tailspin evolution converging to some “ attraction department of state ” [ @Neishtadt ].
This discussion makes it clear that measurements of $ \zeta$ have a potential to help oneself constrain the physics of cometary outgassing — its geometry, intensity, etc. — at least in a statistical common sense. Better characterization of the distribution of $ \zeta$ can receive hearty impact on our sympathy of cometary activity. This provides strong motivation for increase the number of objects with measured $ \Delta P$ and $ a_{\rm ng}$. Right now the about basic property of this distribution represented by equation (\[eq: stats\ ]) are based on a sample of only 7 objects. Targeted observations of many other comet aim at measuring their tailspin variation would leave us to characterize the distribution of $ \zeta$ much better. According to Figure \[fig: comets\ ], even period measurements with uncertainly of ten of minutes may reveal spin variability for dozens of comets with gamey $ \Delta\Omega_1 $ (the ones above the blue line in this figure).
The aim with high $ \Delta\Omega_1 $ listed in the Table \[table: fast\ ] should have the best chance for revealing such change, even though only one comet from Table \[table: zeta\ ] (41P / Tuttle - Giacobini - Kresák) has $ \Delta\Omega_1>10^{-3}$ s$^{-1}$ necessary for being listed in Table \[table: fast\ ]. Although we concenter on studying the spin period variability of comet, non - gravitational forced precession [ @Whipple; @Sekanina1984 ] should also be substantial for these objects, likely drive them into excited rotational state.
Implications for Solar System comet { # sect: disc_SS }
------------------------------------
Our calculations attest that many Solar System comets (the ones with gamey $ \Delta\Omega_1 $) should exhibit rapid development of their rotation. Using equations (\[eq: spinev\ ]) and (\[eq: a\_ng\ ]) one can estimate the characteristic timescale for the cometary tailspin evolution $ \tau_\Omega=|\Omega/\dot\Omega|$ as \_&= & \^[-1 ] { } \[eq: tau\_Om\]\
& & 670 | wihh $\Delta P$ spanning 4 decxdes — also merijs attenvion. To some deeree, it could be a result of tye smqll size of the sample (7 objects) used to dermve $\zeta$. On the other hakb, thedc may ue some other ewplanations, e.g. the aforemangilned long-term spin evolution converding to slme “attractor ftatt” [@Nqishfadt].
This discussion makes it clear that mtasurements of $\zets$ have a potential to help condtrain the physics of cometari ouetassing — its geometry, putensity, etd. — at least in a statistical sevse. Bztter charaxtwrixdtion of thx distgibution of $\zcna$ can vave sunstantial impagt on oue understanding of coketary activity. Thys providas strong motivatiob dor itcredsine thd nhmuer of obuecvs with meaaured $\Delta P$ and $a_{\rm ng}$. Right njq the most baaic prjpqrties of this distribution represented by equation (\[eq:stats\]) are bqsed on a sample of ojly 7 objests. Targeted observations of many other comets aiked av oeawuving rhfir spin variations would allow us to charactqdiee nhe distribution jf $\zeta$ mucn heyjer. According jo Figuxs \[rig:comets\], even perlod meafuremwnts with uncrrtainly of tens of minutes may reveal fpin variability fox dozens of eomets with high $\Delta\Omega_1$ (the onzs abobe the blue line in ffis figure).
The obgecns whth hightrt $\Delta\Omega_1$ liseed in thx Tabke \[tabld:fasj\] shoulq have the best chance for revealijg sueh chdnges, even though only one comet from Table \[table:zeta\] (41P/Jutdle-Ciacobinn-Kresák) has $\Delta\Omeda_1>10^{-3}$ s$^{-1}$ necessary for benng lirted in Tagle \[tabne:fast\]. Althjugh we focusaf on studyinj the spig peeiod variabkuity of comets, non-gravinauional forxed precession [@Whiiple; @Rskanina1984] should cuwo be substantisl wor tjexe mbjects, likeny dfivkmg thdm into excltea royational state.
Implicdtiohs for Solar Systek gomets {#sext:disc_SS}
------------------------------------
Jur calculatipns demonstrate thwt maiy Soler Sysyem comets (the ones with high $\Delfa\Omega_1$) sjoujd exhibit rwpid evolution oy their rotation. Using equations (\[eq:spinet\]) and (\[eq:a\_ng\]) one can esjimate the characterixuic timescale for ehe cometdry spin evolution $\tqu_\Omega=|\Omega/\dot\Omtga|$ as \_&=& \^[-1]{} \[eq:tau\_Om\]\
&& 670 | with $\Delta P$ spanning 4 decades — attention. some degree, could be a of sample (7 objects) to derive $\zeta$. the other hand, there may be other explanations, e.g. the aforementioned long-term spin evolution converging to some “attractor state” This discussion makes it clear that measurements of $\zeta$ have a potential to constrain physics cometary — its geometry, intensity, etc. — at least in a statistical sense. Better characterization of the of $\zeta$ can have substantial impact on our of cometary activity. This strong motivation for increasing the of with measured P$ $a_{\rm Right now the basic properties of this distribution represented by equation (\[eq:stats\]) are based on a sample of only 7 Targeted observations other comets at their variations would allow characterize the distribution of $\zeta$ much Figure \[fig:comets\], even period measurements with uncertainly of of minutes reveal spin variability for dozens of with high $\Delta\Omega_1$ (the ones above the blue in this figure). The objects with highest $\Delta\Omega_1$ listed in the Table \[table:fast\] should have chance for revealing such even though only comet Table (41P/Tuttle-Giacobini-Kresák) $\Delta\Omega_1>10^{-3}$ s$^{-1}$ for being listed in Table \[table:fast\]. Although we focused on studying spin period variability of comets, non-gravitational forced precession [@Whipple; @Sekanina1984] be for these objects, driving them into excited state. for Solar System comets Our demonstrate System (the with high $\Delta\Omega_1$) should rapid evolution of their rotation. equations (\[eq:spinev\]) and (\[eq:a\_ng\]) timescale for the cometary spin evolution $\tau_\Omega=|\Omega/\dot\Omega|$ as \^[-1]{} \[eq:tau\_Om\]\ && 670 | with $\Delta P$ spanning 4 decades — Also merits AttenTioN. To SoMe deGree, It could be a resuLT of tHe small size of the sample (7 ObjecTs) USed tO DeRive $\zEta$. On thE OtHER haNd, ThEre MaY Be Some oTheR explanAtions, e.g. thE afOrEmentioned loNG-tErm spin evoLutIon converginG to Some “atTrActOR statE” [@NeIshtaDt].
This DIscussIon makes iT cLEar thaT MeasureMENtS of $\zEta$ have a potential TO hELp constrain the PhysicS oF CoMETarY ouTgassing — itS gEometRY, intensITy, ETC. — At lEAst in a statistIcal sense. BeTTer CharacTeRizATion of The diStRIbuTion of $\zeta$ cAn haVe substanTial imPAct on ouR UnderstAnding Of cOmeTary ACtIvIty. thIS prOViDes STroNg motivaTiOn For inCreaSING The nUmbEr of ObjecTs with measureD $\DeLta P$ ANd $a_{\Rm ng}$. RIght nOw thE mOst baSic proPertiEs Of this distributIon rEpresenteD by EqUatIoN (\[eq:stATs\]) are bAseD on A sample Of only 7 oBJecTs. tARGeTed observations of mAnY OThEr comets Aimed aT MeAsURing theiR sPin VariATIons wOuld ALlOw us to chAracteRIzE tHe distrIbUtion oF $\zEta$ MucH bettER. AccOrding To Figure \[Fig:coMEts\], even period mEAsurements witH UnCERtAInly Of tEns of minuteS may REveaL spiN VaRiaBIlity For doZeNS oF Comets with high $\Delta\omEga_1$ (the Ones aBove the blue liNe in this fiGURE).
The objeCts wITh HIghest $\Delta\OmeGa_1$ lisTed in the TaBLe \[table:fAst\] shOuld have The best chANCe for revEalIng SucH chANGeS, even though onLY One cOmEt from TAblE \[table:zEta\] (41p/TuTtlE-GiAcObini-KresÁk) has $\DelTa\omEgA_1>10^{-3}$ s$^{-1}$ NecEssarY For being LiSteD iN TaBle \[taBLe:fast\]. althoUgh wE fOcUSed On studyINg THE spiN pErIod vAriAbIlity Of coMEts, Non-gravItational ForCEd prEcEsSion [@WhiPple; @Sekanina1984] sHoUld also be sUbStaNtial fOR These objEcts, likely driving them inTO exciteD roTatioNal sTate.
ImpliCatIons foR SoLAr SystEm comeTs {#secT:dIsc_ss}
------------------------------------
our caLCUlAtiOnS demonstraTE ThaT many soLar SYstem coMets (the ones with higH $\delTa\Omega_1$) should ExhIbit RAPiD evOLuTIon Of THeiR ROtation. Using equAtions (\[eq:spInEV\]) aNd (\[eq:a\_ng\]) one CAn eStImate thE characTerisTIc timesCale for thE cometary SpIn evOLUtiOn $\tau_\Omega=|\omega/\dot\omega|$ as \_&=& \^[-1]{} \[eq:TAu\_Om\]\
&& 670 | with $\Delta P$ spanning4 decades— als o m eri ts att enti on. To some de g ree, it could be a resultof th es mall si ze of the sa m pl e (7ob je cts )u se d toder ive $\z eta$. On t heot her hand, th e re may be so meother explan ati ons, e .g . t h e afo rem entio ned lo n g-term spin evo lu t ion co n verging t osome “attractor state ” [ @ Neishtadt].
T his di sc u ss i o n m ake s it clear t hat m e asureme n ts o f $\ z eta$ have a p otential to hel p cons tr ain the ph ysics o f co metary outg assi ng — itsgeomet r y, inte n sity, e tc. —atlea st i n a s tat is t ica l s ens e . B etter ch ar ac teriz atio n o f the di stri butio n of $\zeta$can hav e su bstan tialimpa ct on o ur und ersta nd ing of cometary act ivity. Th ispr ovi de s str o ng mot iva tio n for i ncreasi n g t he n u mb er of objects with m e a su red $\De lta P$ an d$ a_{\rm n g} $.Righ t now t he m o st basic p ropert i es o f thisdi stribu ti onrep resen t ed b y equa tion (\[ eq:st a ts\]) are base d on a sampleo fo n ly 7 ob jec ts. Targete d ob s erva tion s o f m a ny ot her c om e ts aimed at measuringth eir sp in va riations woul d allow us t o charact eriz e t h e distribution of $ \zeta$ muc h better. Acco rding to Figure \ [ f ig:comet s\] , e ven pe r i od measurements w ithun certain lyof tens of mi nut esma y revealspin var ia bi li ty fo r doz e ns of co me tswi thhigh$ \Delta \Omeg a_1$ ( th e on es abov e t h e blu eli ne i n t hi s fig ure) .
T he obje cts withhig h est$\ De lta\Ome ga_1$ listedin the Table \ [ta ble:fa s t \] shoul d have the best chancef or reve ali ng su ch c hanges, e ven thoug h o n ly one comet from T abl e \[tab l e :z eta \] (41P/Tutt l e -Gi acobi ni -Kre sák) ha s $\Delta\Omega_1> 1 0^{ -3}$ s$^{-1}$ ne cess a r yfor be i ngli s ted i n Table \[table :fast\]. A lt h ou gh we focu s edon studyi ng thespinp eriod v ariabilit y of come ts , no n - gra vitational forcedprecessio n [@Wh i pp le; @ Sek anina1 98 4]shoul d also besubst antial f or the se ob je cts, lik ely driving them into e xcited rota tio nal state .
I mpl icationsforSolar Syst emcom ets { #se c t:dis c_SS }
- --- - ----- ---- - --------- - -- --- - - -- --
Our cal c u l ati ons d emo n strate tha t many Solar Syst e m comets (theones w ith hi g h $\ De lta\Omega_1$)sho ul d exhibitra pid evoluti on of th ei r rota tion.Usingequatio n s ( \ [eq:sp inev \]) and (\[e q:a \_ n g\]) on eca n estim ateth e char acteri s tict i mescale for thecomet a r y spi n ev oluti on $\tau_ \ Omeg a=|\Omega/ \dot\Omega| $ as \ _&=& \^[- 1]{} \[ eq :tau\_ Om\ ]\
&& 670 | with_$\Delta P$_spanning 4 decades —_also merits_attention._To some_degree,_it could be_a result of_the small size of_the sample (7_objects)_used to derive $\zeta$. On the other hand, there may be some other explanations,_e.g._the aforementioned_long-term_spin_evolution converging to some “attractor_state” [@Neishtadt].
This discussion makes it_clear that_measurements of $\zeta$ have a potential to help_constrain_the physics of_cometary outgassing — its geometry, intensity, etc. — at_least in a statistical sense. Better_characterization of the_distribution_of_$\zeta$ can have substantial_impact on our understanding of cometary_activity. This provides strong motivation for_increasing the number of objects with measured_$\Delta P$ and $a_{\rm ng}$. Right_now the most basic properties_of this_distribution represented by equation (\[eq:stats\])_are based on_a sample_of only 7_objects. Targeted observations of many other_comets aimed at_measuring their spin variations would allow_us_to characterize the_distribution_of_$\zeta$ much_better. According to_Figure_\[fig:comets\], even_period_measurements with uncertainly of tens of_minutes_may reveal spin variability for dozens of_comets with high $\Delta\Omega_1$_(the_ones above the blue_line in this figure).
The objects_with highest $\Delta\Omega_1$ listed in the_Table \[table:fast\]_should have_the best chance for revealing such changes, even though only one_comet from Table \[table:zeta\] (41P/Tuttle-Giacobini-Kresák) has_$\Delta\Omega_1>10^{-3}$ s$^{-1}$ necessary for_being listed_in_Table \[table:fast\]. Although_we_focused on_studying the spin period variability of comets,_non-gravitational forced_precession [@Whipple; @Sekanina1984] should also be_substantial for these objects,_likely_driving them into excited rotational state.
Implications_for Solar System comets {#sect:disc_SS}
------------------------------------
Our calculations_demonstrate that many Solar System_comets_(the_ones with high $\Delta\Omega_1$) should_exhibit rapid evolution of their rotation._Using equations (\[eq:spinev\])_and (\[eq:a\_ng\]) one can estimate the characteristic_timescale_for the cometary spin evolution $\tau_\Omega=|\Omega/\dot\Omega|$_as_\_&=& \^[-1]{} \[eq:tau\_Om\]\
&& 670 |
eq U^*|T^*|^{1/2}UU^*|T^*|^{1/2}U = U^*|T^*|U = |T|,$$ hence, ${\overline}{R} ({\overline}{R})^* \supseteq |T|$ and thus, $({\overline}{R})^* {\overline}{R} = {\overline}{R} ({\overline}{R})^* = |T|$. That is, ${\overline}R$ is normal and symmetric, and hence self-adjoint. Since in addition, ${\overline}R \geq 0$, ${\overline}R$ is the unique self-adjoint, nonnegative square root of $|T|$, $${\overline}R = ({\overline}R)^* = |T|^{1/2}.$$ Introducing $S=U |T|^{1/2} U^*$, one obtains analogously, $${\overline}S = ({\overline}S)^* = |T^*|^{1/2}.$$ Thus, $$|T|^{1/2} \supseteq R= U^* |T^*|^{1/2}U, \quad |T^*|^{1/2} \supseteq S = U|T|^{1/2} U^*.$$ Next, using also the first relation in, $U^*U=P_{{\overline}{{{\text{\rm{ran}}}}(|T|)}}=P_{{\overline}{{{\text{\rm{ran}}}}(|T|^{1/2})}}$, one infers $$U^* |T^*|^{1/2} \supseteq U^*U |T|^{1/2} U^* = |T|^{1/2} U^*, \quad
|T|^{1/2} U^* \supseteq U^* |T^*|^{1/2} UU^* = U^* |T^*|^{1/2},$$ and hence $$U^* |T^*|^{1/2} = |T|^{1/2}U^*, \text{ implying } \, U U^* |T^*|^{1/2} = |T^*|^{1/2} = U |T|^{1/2} U^*.$$ But then, $$|T^*|^{1/2} U |T|^{1/2} = U |T|^{1/2} U^* U |T|^{1/2} = U |T| =T,$$ as was to be proven.
Some Applications to Relatively (Form) Bounded and Relatively (Form) Compact Perturb | eq U^*|T^*|^{1/2}UU^*|T^*|^{1/2}U = U^*|T^*|U = |T|,$$ hence, $ { \overline}{R } ({ \overline}{R})^ * \supseteq |T|$ and thus, $ ({ \overline}{R})^ * { \overline}{R } = { \overline}{R } ({ \overline}{R})^ * = |T|$. That is, $ { \overline}R$ is normal and symmetric, and hence self - adjoint. Since in accession, $ { \overline}R \geq 0 $, $ { \overline}R$ is the alone self - adjoint, nonnegative square root of $ |T|$, $ $ { \overline}R = ({ \overline}R)^ * = |T|^{1/2}.$$ Introducing $ S = U |T|^{1/2 } U^*$, one receive analogously, $ $ { \overline}S = ({ \overline}S)^ * = |T^*|^{1/2}.$$ Thus, $ $ |T|^{1/2 } \supseteq R= U^ * |T^*|^{1/2}U, \quad |T^*|^{1/2 } \supseteq S = U|T|^{1/2 } U^*.$$ Next, using besides the first relation in, $ U^*U = P_{{\overline}{{{\text{\rm{ran}}}}(|T|)}}=P_{{\overline}{{{\text{\rm{ran}}}}(|T|^{1/2})}}$, one infers $ $ U^ * |T^*|^{1/2 } \supseteq U^*U |T|^{1/2 } U^ * = |T|^{1/2 } U^ *, \quad
|T|^{1/2 } U^ * \supseteq U^ * |T^*|^{1/2 } UU^ * = U^ * |T^*|^{1/2},$$ and hence $ $ U^ * |T^*|^{1/2 } = |T|^{1/2}U^ *, \text { imply } \, U U^ * |T^*|^{1/2 } = |T^*|^{1/2 } = U |T|^{1/2 } U^*.$$ But then, $ $ |T^*|^{1/2 } U |T|^{1/2 } = U |T|^{1/2 } U^ * U |T|^{1/2 } = uranium |T| = T,$$ as was to be proven.
Some Applications to Relatively (class) restrict and Relatively (Form) Compact Perturb | eq K^*|T^*|^{1/2}UU^*|T^*|^{1/2}U = U^*|T^*|U = |T|,$$ hence, ${\ovevline}{R} ({\overline}{R})^* \supsetxq |T|$ ans thus, $({\oxerline}{R})^* {\overline}{R} = {\overline}{C} ({\ovwrlint}{G})^* = |T|$. That is, ${\overline}F$ is normwl and stmmeuric, and hence self-adjoinb. Sindc in cdvition, ${\overline}T \geq 0$, ${\overlhne}R$ is the unhqje self-adjoint, nonnegative square rooe of $|T|$, $${\pvfrline}R = ({\overlyne}R)^* = |T|^{1/2}.$$ Ihnriducing $S=U |T|^{1/2} U^*$, one obtains anzlogousny, $${\overline}S = ({\overline}S)^* = |T^*|^{1/2}.$$ Thus, $$|T|^{1/2} \supseheq G= U^* |T^*|^{1/2}U, \quad |T^*|^{1/2} \supsfteq S = U|T|^{1/2} O^*.$$ Nevr, using also the first relation in, $U^*U=P_{{\overline}{{{\text{\rm{ran}}}}(|T|)}}=P_{{\overline}{{{\gext{\rk{ran}}}}(|T|^{1/2})}}$, one underd $$U^* |T^*|^{1/2} \supsetxq U^*U |N|^{1/2} U^* = |T|^{1/2} U^*, \quad
|T|^{1/2} U^* \sgpseteq U^* |T^*|^{1/2} UU^* = U^* |T^*|^{1/2},$$ akd heice $$Y^* |T^*|^{1/2} = |T|^{1/2}U^*, \text{ implying } \, U U^* |T^*|^{1/2} = |T^*|^{1/2} = U |T|^{1/2} U^*.$$ But then, $$|T^*|^{1/2} U |T|^{1/2} = U |T|^{1/2} U^* U |T|^{1/2} = Y |R| =T,$$ av wav to ve orobei.
Soje Apppicetions to Rslatively (Firm) Bounded and Relstynrly (Form) Comlact Pqreurb | eq U^*|T^*|^{1/2}UU^*|T^*|^{1/2}U = U^*|T^*|U = |T|,$$ hence, \supseteq and thus, {\overline}{R} = {\overline}{R} ${\overline}R$ normal and symmetric, hence self-adjoint. Since addition, ${\overline}R \geq 0$, ${\overline}R$ is unique self-adjoint, nonnegative square root of $|T|$, $${\overline}R = ({\overline}R)^* = |T|^{1/2}.$$ Introducing |T|^{1/2} U^*$, one obtains analogously, $${\overline}S = ({\overline}S)^* = |T^*|^{1/2}.$$ Thus, $$|T|^{1/2} \supseteq U^* \quad \supseteq = U|T|^{1/2} U^*.$$ Next, using also the first relation in, $U^*U=P_{{\overline}{{{\text{\rm{ran}}}}(|T|)}}=P_{{\overline}{{{\text{\rm{ran}}}}(|T|^{1/2})}}$, one infers $$U^* |T^*|^{1/2} \supseteq |T|^{1/2} U^* = |T|^{1/2} U^*, \quad |T|^{1/2} U^* U^* |T^*|^{1/2} UU^* = |T^*|^{1/2},$$ and hence $$U^* |T^*|^{1/2} |T|^{1/2}U^*, implying } U |T^*|^{1/2} |T^*|^{1/2} = U U^*.$$ But then, $$|T^*|^{1/2} U |T|^{1/2} = U |T|^{1/2} U^* U |T|^{1/2} = U |T| =T,$$ as to be Applications to (Form) and (Form) Compact Perturb | eq U^*|T^*|^{1/2}UU^*|T^*|^{1/2}U = U^*|T^*|U = |T|,$$ hence, ${\overline}{r} ({\overline}{R})^* \SupseTeq |t|$ anD tHus, $({\oVerlIne}{R})^* {\overline}{R} = {\oVErliNe}{R} ({\overline}{R})^* = |T|$. That is, ${\oveRline}r$ iS NormAL aNd symMetric, aND hENCe sElF-aDjoInT. siNce in AddItion, ${\ovErline}R \geq 0$, ${\OveRlIne}R$ is the uniQUe Self-adjoinT, noNnegative squAre Root of $|t|$, $${\oVerLIne}R = ({\oVerLine}R)^* = |t|^{1/2}.$$ IntroDUcing $S=u |T|^{1/2} U^*$, one obtAiNS analoGOusly, $${\ovERLiNe}S = ({\oVerline}S)^* = |T^*|^{1/2}.$$ Thus, $$|T|^{1/2} \supSEtEQ R= U^* |T^*|^{1/2}U, \quad |T^*|^{1/2} \supsEteq S = U|t|^{1/2} U^*.$$ nExT, USinG alSo the first ReLatioN In, $U^*U=P_{{\ovERlINE}{{{\TexT{\Rm{ran}}}}(|T|)}}=P_{{\overliNe}{{{\text{\rm{ran}}}}(|t|^{1/2})}}$, One Infers $$u^* |T^*|^{1/2} \SupSEteq U^*U |t|^{1/2} U^* = |T|^{1/2} U^*, \qUaD
|t|^{1/2} U^* \sUpseteq U^* |T^*|^{1/2} UU^* = u^* |T^*|^{1/2},$$ anD hence $$U^* |T^*|^{1/2} = |T|^{1/2}u^*, \text{ iMPlying } \, U u^* |t^*|^{1/2} = |T^*|^{1/2} = U |T|^{1/2} U^*.$$ BuT then, $$|T^*|^{1/2} u |T|^{1/2} = U |t|^{1/2} U^* U |t|^{1/2} = U |T| =T,$$ AS wAs To bE pROveN.
soMe APPliCations tO RElAtiveLy (FoRM) bOUndeD anD RelAtiveLy (Form) Compact perTurb | eq U^*|T^*|^{1/2}UU^*|T^*| ^{1/2}U =U^*|T ^*| U = | T|,$ $ he nce, ${\overli n e}{R } ({\overline}{R})^* \ supse te q |T| $ a nd th us, $({ \ ov e r lin e} {R })^ *{ \o verli ne} {R} = { \overline} {R} ( {\overline}{ R }) ^* = |T|$. Th at is, ${\ov erl ine}R$ i s n o rmaland symm etric, and he nce self- ad j oint.S ince in a dd itio n, ${\overline}R\ ge q 0$, ${\overli ne}R$is th e uni que self-adjo in t, no n negativ e s q u a rer oot of $|T|$, $${\overli n e}R = ({\ ov erl i ne}R)^ * = | T| ^ {1/ 2}.$$ Intro duci ng $S=U | T|^{1/ 2 } U^*$, one obt ains a nal ogo usly , $ ${ \ov er l ine } S= ( { \ov erline}S )^ *= |T^ *|^{ 1 / 2 } .$$Thu s, $ $|T|^ {1/2} \supset eqR= U ^ * | T^*|^ {1/2} U, \ qu ad |T ^*|^{1 /2} \ su pseteq S = U|T| ^{1/ 2} U^*.$$ Ne xt , u si ng al s o thefir strelatio n in, $ U ^*U =P _ { { \o verline}{{{\text{\ rm { r an }}}}(|T| )}}=P_ { {\ ov e rline}{{ {\ tex t{\r m { ran}} }}(| T |^ {1/2})}} $, one in fe rs $$U^ *|T^*|^ {1 /2} \s upset e q U^ *U |T| ^{1/2} U ^* =| T|^{1/2} U^*, \quad
|T|^{ 1 /2 } U^ * \su pse teq U^* |T^ *|^{ 1 /2}UU^* =U^* |T^*| ^{1/2 }, $ $a nd hence $$U^* |T^* |^ {1/2}= |T| ^{1/2}U^*, \t ext{ imply i n g } \, UU^*| T^ * |^{1/2} = |T^* |^{1/ 2} = U |T| ^ {1/2} U^ *.$$But then , $$|T^*| ^ { 1/2} U | T|^ {1/ 2}= U | T| ^{1/2} U^* U| T |^{1 /2 } = U | T|=T,$$ a s w astobepr oven.
So me Appli ca ti on stoRelat i vely (Fo rm ) B ou nde d and Relati vely(For m) C o mpa ct Pert u rb | eq U^*|T^*|^{1/2}UU^*|T^*|^{1/2}U_= U^*|T^*|U_= |T|,$$ hence, ${\overline}{R}_({\overline}{R})^* \supseteq_|T|$_and thus,_$({\overline}{R})^*_{\overline}{R} = {\overline}{R}_({\overline}{R})^* = |T|$._That is, ${\overline}R$ is_normal and symmetric,_and_hence self-adjoint. Since in addition, ${\overline}R \geq 0$, ${\overline}R$ is the unique self-adjoint, nonnegative_square_root of_$|T|$,_$${\overline}R_= ({\overline}R)^* = |T|^{1/2}.$$ Introducing_$S=U |T|^{1/2} U^*$, one obtains_analogously, $${\overline}S_= ({\overline}S)^* = |T^*|^{1/2}.$$ Thus, $$|T|^{1/2} \supseteq R=_U^*_|T^*|^{1/2}U, \quad |T^*|^{1/2}_\supseteq S = U|T|^{1/2} U^*.$$ Next, using also the_first relation in, $U^*U=P_{{\overline}{{{\text{\rm{ran}}}}(|T|)}}=P_{{\overline}{{{\text{\rm{ran}}}}(|T|^{1/2})}}$, one infers_$$U^* |T^*|^{1/2} \supseteq_U^*U_|T|^{1/2}_U^* = |T|^{1/2} U^*,_ \quad
|T|^{1/2} U^* \supseteq_U^* |T^*|^{1/2} UU^* = U^* |T^*|^{1/2},$$_and hence $$U^* |T^*|^{1/2} = |T|^{1/2}U^*, \text{_implying } \, U U^* |T^*|^{1/2}_= |T^*|^{1/2} = U |T|^{1/2}_U^*.$$ But_then, $$|T^*|^{1/2} U |T|^{1/2} =_U |T|^{1/2} U^*_U |T|^{1/2}_= U |T|_=T,$$ as was to be proven.
Some_Applications to Relatively_(Form) Bounded and Relatively (Form) Compact_Perturb |
M^{tri}_{a_{1}\rightarrow 3\pi} = \frac{1}{f_{\pi}}
M(a_{1}\rightarrow \sigma\pi) \,,$$ because $2g_{\sigma\pi\pi}=(m^{2}_{\sigma}-m^{2}_{\pi})/f_{\pi}$ in the linear $\sigma$ model (L$\sigma$M). Thus the sum of (\[eq:two\]) and (\[eq:three\]) vanishes in the soft pion limit [@ivanov; @ivanov1] $$\label{eq:four}
M_{a_{1}\rightarrow 3\pi}|_{total} =
M^{box}_{a_{1}\rightarrow 3\pi} +
M^{tri}_{a_{1}\rightarrow 3\pi} \rightarrow 0 \,,$$ compatible with data [@pdg]: $\Gamma(a_{1}\rightarrow\pi(\pi\pi)_{\mathrm{swave}})=1\pm1$ MeV.
b\) $\gamma\gamma\rightarrow2\pi^{0}|_{s=m^{2}_{\sigma}}$:
Again using pseudoscalar pion-quark couplings, it was predicted [@kaloshin] five years before data appeared that this $\gamma\gamma\rightarrow2\pi^{0}$ cross section should fall to about $10$ nbarns in the $700$ MeV region. Equivalently, using the SPT theorem stemming from Eq. (\[eq:identity\]), we predict the amplitude due to the quark box plus quark triangle graphs of Fig. 2 $$\label{eq:five}
\langle \pi^{0}\pi^{0}|\gamma\gamma \rangle \rightarrow
\left[ - \frac{i}{f_{\pi}}\langle\sigma|\gamma\gamma\rangle +
\frac{i}{f_{\pi}}\langle\sigma|\gamma\gamma\rangle \right] \rightarrow 0 \,,$$ as $s\rightarrow m^2_{\sigma}(700)$ [@ivanov1]. This picture was supported by recent Crystal Ball data [@crystalball].
c\) $\pi^{-}p\rightarrow\pi^{-}\pi^{+}n$:
The SPT stemming from Eq. (\[eq:identity\]) also suggests that the sum of the two $\pi^{+}$ peripheral–dominated $\pi^{-}p\rightarrow\pi^{-}\pi^{+ | M^{tri}_{a_{1}\rightarrow 3\pi } = \frac{1}{f_{\pi } }
M(a_{1}\rightarrow \sigma\pi) \,,$$ because $ 2g_{\sigma\pi\pi}=(m^{2}_{\sigma}-m^{2}_{\pi})/f_{\pi}$ in the linear $ \sigma$ model (L$\sigma$M). Thus the sum of (\[eq: two\ ]) and (\[eq: three\ ]) vanishes in the piano pion terminus ad quem [ @ivanov; @ivanov1 ] $ $ \label{eq: four }
M_{a_{1}\rightarrow 3\pi}|_{total } =
M^{box}_{a_{1}\rightarrow 3\pi } +
M^{tri}_{a_{1}\rightarrow 3\pi } \rightarrow 0 \,,$$ compatible with data [ @pdg ]: $ \Gamma(a_{1}\rightarrow\pi(\pi\pi)_{\mathrm{swave}})=1\pm1 $ MeV.
b\) $ \gamma\gamma\rightarrow2\pi^{0}|_{s = m^{2}_{\sigma}}$:
Again using pseudoscalar pion - quark cheese couplings, it was bode [ @kaloshin ] five years before data appear that this $ \gamma\gamma\rightarrow2\pi^{0}$ cross section should descend to about $ 10 $ nbarns in the $ 700 $ MeV region. Equivalently, use the SPT theorem stemming from Eq. (\[eq: identity\ ]), we predict the amplitude due to the quark box plus quark triangulum graphs of Fig. 2 $ $ \label{eq: five }
\langle \pi^{0}\pi^{0}|\gamma\gamma \rangle \rightarrow
\left [ - \frac{i}{f_{\pi}}\langle\sigma|\gamma\gamma\rangle +
\frac{i}{f_{\pi}}\langle\sigma|\gamma\gamma\rangle \right ] \rightarrow 0 \,,$$ as $ s\rightarrow m^2_{\sigma}(700)$ [ @ivanov1 ]. This picture was supported by recent Crystal Ball data [ @crystalball ].
c\) $ \pi^{-}p\rightarrow\pi^{-}\pi^{+}n$:
The SPT stemming from Eq. (\[eq: identity\ ]) also propose that the sum of the two $ \pi^{+}$ peripheral – dominated $ \pi^{-}p\rightarrow\pi^{-}\pi^{+ | M^{tri}_{a_{1}\rightarrow 3\pi} = \frag{1}{f_{\pi}}
M(a_{1}\rightcerow \smgma\pi) \,,$$ because $2g_{\sigma\pi\pi}=(m^{2}_{\sigma}-m^{2}_{\pi})/f_{\pi}$ in tie lunear $\sigma$ model (L$\sigma$M). Tfus the slm of (\[eq:twi\]) anv (\[eq:three\]) vanishes in the soft llon lnmmt [@ivanov; @ivanov1] $$\label{eq:fogr}
M_{a_{1}\rightasruw 3\pi}|_{total} =
M^{box}_{a_{1}\rightarrow 3\pi} +
M^{tri}_{a_{1}\rogjtarrow 3\pi} \rigrtargor 0 \,,$$ dompatible with data [@pdg]: $\Gamma(a_{1}\righfarrow\pp(\pi\pi)_{\mathrm{swave}})=1\pk1$ MeV.
b\) $\gamma\gamma\rightarroa2\pi^{0}|_{s=l^{2}_{\sigma}}$:
Again using oseudoscalat pijb-quark couplkngs, it was predicted [@kzloshin] five years before data xppeaxed that thus $\galka\gamma\righvarrow2\ii^{0}$ cross sectlpn shogld falk to about $10$ nbsrnv ib the $700$ MeV region. Equmvalently, using the FPT theoram stemming from Eq. (\[wq:udenthty\]), fe pfwdizt uhe ajplitufe vue to the suark box pous quark triangle brwins of Fig. 2 $$\lagel{eq:fyvq}
\langle \pi^{0}\pi^{0}|\gamma\gamma \rangle \rightdrrkw
\left[ - \frac{i}{f_{\pi}}\lantle\sigma|\gamma\gamma\ranhle +
\fwac{i}{f_{\pi}}\langle\sigma|\gamma\gamma\rangle \right] \rightarrmw 0 \,,$$ es $s\rnnmtarfiw m^2_{\sigma}(700)$ [@ivanov1]. This picture was supported by rqdemt Crystal Ball bata [@crystalball].
c\) $\pl^{-}p\tyghtarrow\pi^{-}\pi^{+}v$:
The S'F atemming from Eq. (\[eq:ldentiti\]) also suggests thay the sum of the two $\pi^{+}$ perupheral–dominctee $\pi^{-}p\rightarrow\pi^{-}\pn^{+ | M^{tri}_{a_{1}\rightarrow 3\pi} = \frac{1}{f_{\pi}} M(a_{1}\rightarrow \sigma\pi) \,,$$ in linear $\sigma$ (L$\sigma$M). Thus the vanishes the soft pion [@ivanov; @ivanov1] $$\label{eq:four} 3\pi}|_{total} = M^{box}_{a_{1}\rightarrow 3\pi} + M^{tri}_{a_{1}\rightarrow \rightarrow 0 \,,$$ compatible with data [@pdg]: $\Gamma(a_{1}\rightarrow\pi(\pi\pi)_{\mathrm{swave}})=1\pm1$ MeV. b\) $\gamma\gamma\rightarrow2\pi^{0}|_{s=m^{2}_{\sigma}}$: Again using pion-quark couplings, it was predicted [@kaloshin] five years before data appeared that this cross should to $10$ nbarns in the $700$ MeV region. Equivalently, using the SPT theorem stemming from Eq. (\[eq:identity\]), predict the amplitude due to the quark box quark triangle graphs of 2 $$\label{eq:five} \langle \pi^{0}\pi^{0}|\gamma\gamma \rangle \left[ \frac{i}{f_{\pi}}\langle\sigma|\gamma\gamma\rangle + \right] 0 as $s\rightarrow m^2_{\sigma}(700)$ This picture was supported by recent Crystal Ball data [@crystalball]. c\) $\pi^{-}p\rightarrow\pi^{-}\pi^{+}n$: The SPT stemming from Eq. also suggests sum of two peripheral–dominated | M^{tri}_{a_{1}\rightarrow 3\pi} = \frac{1}{f_{\pi}}
M(A_{1}\rightarroW \sigmA\pi) \,,$$ BecAuSe $2g_{\sIgma\Pi\pi}=(m^{2}_{\sigma}-m^{2}_{\pi})/f_{\PI}$ in tHe linear $\sigma$ model (L$\sigMa$M). ThUs THe suM Of (\[Eq:two\]) And (\[eq:thREe\]) VANisHeS iN thE sOFt Pion lImiT [@ivanov; @Ivanov1] $$\labeL{eq:FoUr}
M_{a_{1}\rightarrOW 3\pI}|_{total} =
M^{box}_{A_{1}\riGhtarrow 3\pi} +
M^{tRi}_{a_{1}\RightaRrOw 3\pI} \RightArrOw 0 \,,$$ comPatiblE With daTa [@pdg]: $\GammA(a_{1}\RIghtarROw\pi(\pi\pI)_{\MAtHrm{sWave}})=1\pm1$ MeV.
b\) $\gamma\gaMMa\RIghtarrow2\pi^{0}|_{s=m^{2}_{\sIgma}}$:
AgAiN UsING psEudOscalar pioN-qUark cOUplings, IT wAS PRedICted [@kaloshin] fIve years befORe dAta appEaRed THat thiS $\gammA\gAMma\Rightarrow2\pI^{0}$ croSs section Should FAll to abOUt $10$ nbarnS in the $700$ meV RegIon. EQUiVaLenTlY, UsiNG tHe Spt thEorem steMmInG from eq. (\[eq:IDENTity\]), We pRediCt the Amplitude due tO thE quaRK boX plus Quark TriaNgLe graPhs of FIg. 2 $$\labEl{Eq:five}
\langle \pi^{0}\pI^{0}|\gamMa\gamma \raNglE \rIghTaRrow
\lEFt[ - \frac{I}{f_{\pI}}\laNgle\sigMa|\gamma\GAmmA\rANGLe +
\Frac{i}{f_{\pi}}\langle\sigmA|\gAMMa\Gamma\ranGle \rigHT] \rIgHTarrow 0 \,,$$ as $S\rIghTarrOW M^2_{\sigmA}(700)$ [@ivaNOv1]. this pictUre was SUpPoRted by rEcEnt CryStAl BAll Data [@cRYstaLball].
c\) $\Pi^{-}p\rightArrow\PI^{-}\pi^{+}n$:
The SPT stemMIng from Eq. (\[eq:idENtITY\]) aLSo suGgeSts that the sUm of THe twO $\pi^{+}$ pERiPheRAl–domInateD $\pI^{-}P\rIGhtarrow\pi^{-}\pi^{+ | M^{tri}_{a_{1}\rightar row 3\pi}= \fr ac{ 1}{ f_ {\pi }} M(a_{1}\ri g htar row \sigma\pi) \,,$$ b ecaus e$ 2g_{ \ si gma\p i\pi}=( m ^{ 2 } _{\ si gm a}- m^ { 2} _{\pi })/ f_{\pi} $ in the l ine ar $\sigma$ mo d el (L$\sigma $M) . Thus the s umof (\[ eq :tw o \]) a nd(\[eq :three \ ]) van ishes inth e softp ion lim i t [ @iva nov; @ivanov1] $$ \ la b el{eq:four}
M_{a _{ 1 }\ r i ght arr ow 3\pi}|_ {t otal} =
M ^ {b o x } _{a _ {1}\rightarro w 3\pi} +
M ^{tri} _{ a_{ 1 }\righ tarro w3 \pi } \rightarr ow 0 \,,$$ co mpatib l e withd ata [@p dg]: $ \Ga mma (a_{ 1 }\ ri ght ar r ow\ p i( \pi \ pi) _{\mathr m{ sw ave}} )=1\ p m 1 $ MeV .
b\)$\gam ma\gamma\righ tar row2 \ pi^ {0}|_ {s=m^ {2}_ {\ sigma }}$:
Again u sing pseudoscal ar p ion-quark co up lin gs , itw as pre dic ted [@kalo shin] f i veye a r s b efore data appeare dt h at this $\ gamma\ g am ma \ rightarr ow 2\p i^{0 } $ cros s se c ti on shoul d fall to a bout $1 0$ nbarn sinthe $700 $ MeV regio n. Equiv alent l y, using the S P T theorem ste m mi n g f r om E q.(\[eq:ident ity\ ] ), w e pr e di ctt he am plitu de du e to the quark box p lu s quar k tri angle graphsof Fig. 2$ $ \ label{eq :fiv e } \langle \pi ^{0}\ pi^{0}|\ga m ma\gamma \ran gle \rig htarrow
\left[- \ fra c{i }{f _ { \p i}}\langle\si g m a|\g am ma\gamm a\r angle +
\ fra c{i }{ f_{\pi}}\ langle\s ig ma |\ ga mma \gamm a \rangle\r igh t] \r ighta r row 0\,,$$ as$s \r i ght arrow m ^ 2_ { \ sigm a} (7 00)$ [@ iv anov1 ]. T h ispicture was supp ort e d by r ec ent Cry stal Ball dat a[@crystalb al l].
c\)$ \ pi^{-}p\ rightarrow\pi^{-}\pi^{+ } n$:
Th e S PT st emmi ng from E q.(\[eq: ide n tity\] ) also sugg es tst h at th e su m o fthe two $\ p i ^{+ }$ pe ri pher al–domi nated $\pi^{-}p\ri g hta rrow\pi^{-}\p i^{ + | _ _M^{tri}_{a_{1}\rightarrow 3\pi} = \frac{1}{f_{\pi}}_
__ M(a_{1}\rightarrow_\sigma\pi)_\,,$$ because $2g_{\sigma\pi\pi}=(m^{2}_{\sigma}-m^{2}_{\pi})/f_{\pi}$_in the linear_$\sigma$ model (L$\sigma$M). Thus_the sum of (\[eq:two\])_and (\[eq:three\])_vanishes in the soft pion limit [@ivanov; @ivanov1] $$\label{eq:four}
M_{a_{1}\rightarrow 3\pi}|_{total} =
__ _M^{box}_{a_{1}\rightarrow_3\pi}_+
M^{tri}_{a_{1}\rightarrow_3\pi} \rightarrow 0 \,,$$ compatible_with data [@pdg]:_$\Gamma(a_{1}\rightarrow\pi(\pi\pi)_{\mathrm{swave}})=1\pm1$ MeV.
b\) $\gamma\gamma\rightarrow2\pi^{0}|_{s=m^{2}_{\sigma}}$:
Again using pseudoscalar pion-quark couplings, it_was_predicted [@kaloshin] five years_before data appeared that this $\gamma\gamma\rightarrow2\pi^{0}$ cross section should_fall to about $10$ nbarns in_the $700$ MeV_region._Equivalently,_using the SPT theorem_stemming from Eq. (\[eq:identity\]), we predict the_amplitude due to the quark box_plus quark triangle graphs of Fig. 2 $$\label{eq:five}
_ \langle \pi^{0}\pi^{0}|\gamma\gamma \rangle_\rightarrow
\left[_- \frac{i}{f_{\pi}}\langle\sigma|\gamma\gamma\rangle_+
\frac{i}{f_{\pi}}\langle\sigma|\gamma\gamma\rangle_\right] \rightarrow 0_\,,$$ as_$s\rightarrow m^2_{\sigma}(700)$ [@ivanov1]. This_picture was supported by recent Crystal_Ball data [@crystalball].
c\) $\pi^{-}p\rightarrow\pi^{-}\pi^{+}n$:
The_SPT stemming from Eq. (\[eq:identity\]) also suggests_that_the sum of_the_two_$\pi^{+}$ peripheral–dominated_$\pi^{-}p\rightarrow\pi^{-}\pi^{+ |
k+1|k+1}(t)
&= \frac{\displaystyle \sum_{P \in \S} \Psi_{P}[t] \prod_{W \in P}{\varphi_{W}[t]}}
{\displaystyle \sum_{P \in \S} \kappa_{P} \pgf^{(|P|-1)} \prod_{W \in P}{d_{W}}}.\end{aligned}$$ For constant $t$ we have $$\begin{aligned}
\displaystyle \prod_{W \in P}{\varphi_{W}[t]} &= t^{|P|-1} \prod_{W \in P}{d_{W}} \\
\text{and } \Psi_{P}[t] &= \left( \prod_{j=1}^{s} {C^{(m_{j}-|P|_{j})}_{j}(0)} \right) \pgf^{(|P|-1)}(t \gamma).\end{aligned}$$ Since $\pgf_{k+1|k+1}(t)$ is the PGF corresponding to the cardinality distribution $p_{k+1|k+1}(n)$, $$\begin{aligned}
&p_{k+1|k+1}(n) = \frac{1}{n!} \, \pgf^{(n)}_{k+1|k+1}(0) \\
&= \frac{1}{n!} \, \left\{ \frac{d^n}{d t^n} \frac{\displaystyle \sum_{P \in \S}
\Psi_{P}[t] t^{|P|-1} \prod_{W \in P}{d_{W}}}
{\displaystyle \sum_{P \in \S} \kappa_{P} \pgf^{(|P|-1)} \prod_{W \in P}{d_{W}}} \right\}_{t = 0} \\
&= \frac{\displaystyle \sum_{P \in \S} \left( \prod_{j=1}^{s}
{C^{(m_{j}-|P|_{j})}_{j}(0)} \right) \, \prod_{W \in P}{d_{W}}
\left\{ \frac{d^n}{d t^n} t^{|P|-1} \, \pgf^{(|P|-1)}(t \gamma) \right\}_{t = 0}}
{n! | k+1|k+1}(t)
& = \frac{\displaystyle \sum_{P \in \S } \Psi_{P}[t ] \prod_{W \in P}{\varphi_{W}[t ] } }
{ \displaystyle \sum_{P \in \S } \kappa_{P } \pgf^{(|P|-1) } \prod_{W \in P}{d_{W}}}.\end{aligned}$$ For constant $ t$ we have $ $ \begin{aligned }
\displaystyle \prod_{W \in P}{\varphi_{W}[t ] } & = t^{|P|-1 } \prod_{W \in P}{d_{W } } \\
\text{and } \Psi_{P}[t ] & = \left (\prod_{j=1}^{s } { C^{(m_{j}-|P|_{j})}_{j}(0) } \right) \pgf^{(|P|-1)}(t \gamma).\end{aligned}$$ Since $ \pgf_{k+1|k+1}(t)$ is the PGF corresponding to the cardinality distribution $ p_{k+1|k+1}(n)$, $ $ \begin{aligned }
& p_{k+1|k+1}(n) = \frac{1}{n! } \, \pgf^{(n)}_{k+1|k+1}(0) \\
& = \frac{1}{n! } \, \left\ { \frac{d^n}{d t^n } \frac{\displaystyle \sum_{P \in \S }
\Psi_{P}[t ] t^{|P|-1 } \prod_{W \in P}{d_{W } } }
{ \displaystyle \sum_{P \in \S } \kappa_{P } \pgf^{(|P|-1) } \prod_{W \in P}{d_{W } } } \right\}_{t = 0 } \\
& = \frac{\displaystyle \sum_{P \in \S } \left (\prod_{j=1}^{s }
{ C^{(m_{j}-|P|_{j})}_{j}(0) } \right) \, \prod_{W \in P}{d_{W } }
\left\ { \frac{d^n}{d t^n } t^{|P|-1 } \, \pgf^{(|P|-1)}(t \gamma) \right\}_{t = 0 } }
{ n! | k+1|k+1}(t)
&= \frac{\displaystyle \sum_{P \ln \S} \Psi_{P}[t] \prod_{W \in P}{\vacphi_{W}[t]}}
{\dksplaystyle \sum_{P \in \S} \kappa_{P} \ptf^{(|P|-1)} \peod_{W \in P}{d_{W}}}.\end{aligned}$$ Wor constwnt $t$ we havt $$\begin{aligned}
\dis'maystylc \pros_{A \in '}{\varphi_{W}[t]} &= t^{|P|-1} \ptod_{W \in P}{d_{W}} \\
\dext{and } \Psi_{P}[t] &= \ueyt( \prod_{j=1}^{s} {C^{(m_{j}-|P|_{j})}_{j}(0)} \right) \pgf^{(|P|-1)}(t \gamma).\enq{alignec}$$ Dince $\pgf_{k+1|k+1}(t)$ if tht PDF ckgrtsponding to the cardinality distdibutioi $p_{k+1|k+1}(n)$, $$\begin{alibned}
&p_{k+1|k+1}(n) = \frac{1}{n!} \, \pgf^{(n)}_{k+1|k+1}(0) \\
&= \fgac{1}{n!} \, \left\{ \frac{d^n}{d t^n} \vrac{\displaywtylq \sum_{P \in \S}
\Psi_{P}[t] t^{|K|-1} \'rod_{W \in P}{d_{S}}}
{\displaystyle \sum_{P \in \S} \kapoa_{P} \pyf^{(|P|-1)} \prod_{W \ib P}{d_{W}}} \tight\}_{t = 0} \\
&= \frec{\dispjaystyle \sum_{I \in \S} \neft( \prpd_{j=1}^{s}
{C^{(m_{j}-|P|_{j})}_{m}(0)} \rigit) \, \prod_{W \in P}{d_{W}}
\left\{ \frac{d^n}{d t^n} t^{|P|-1} \, \pgs^{(|P|-1)}(t \gamma) \rnght\}_{t = 0}}
{n! | k+1|k+1}(t) &= \frac{\displaystyle \sum_{P \in \S} \Psi_{P}[t] P}{\varphi_{W}[t]}} \sum_{P \in \kappa_{P} \pgf^{(|P|-1)} \prod_{W we $$\begin{aligned} \displaystyle \prod_{W P}{\varphi_{W}[t]} &= t^{|P|-1} \in P}{d_{W}} \\ \text{and } \Psi_{P}[t] \left( \prod_{j=1}^{s} {C^{(m_{j}-|P|_{j})}_{j}(0)} \right) \pgf^{(|P|-1)}(t \gamma).\end{aligned}$$ Since $\pgf_{k+1|k+1}(t)$ is the PGF corresponding to cardinality distribution $p_{k+1|k+1}(n)$, $$\begin{aligned} &p_{k+1|k+1}(n) = \frac{1}{n!} \, \pgf^{(n)}_{k+1|k+1}(0) \\ &= \frac{1}{n!} \, \frac{d^n}{d \frac{\displaystyle \in \Psi_{P}[t] t^{|P|-1} \prod_{W \in P}{d_{W}}} {\displaystyle \sum_{P \in \S} \kappa_{P} \pgf^{(|P|-1)} \prod_{W \in P}{d_{W}}} \right\}_{t = \\ &= \frac{\displaystyle \sum_{P \in \S} \left( \prod_{j=1}^{s} \right) \, \prod_{W \in \left\{ \frac{d^n}{d t^n} t^{|P|-1} \, \gamma) = 0}} | k+1|k+1}(t)
&= \frac{\displaystyle \sum_{P \in \s} \Psi_{P}[t] \prod_{w \in P}{\vArpHi_{W}[T]}}
{\dIsplAystYle \sum_{P \in \S} \kappA_{p} \pgf^{(|p|-1)} \prod_{W \in P}{d_{W}}}.\end{aligned}$$ FOr conStANt $t$ wE HaVe $$\begIn{alignED}
\dISPlaYsTyLe \pRoD_{w \iN P}{\varPhi_{w}[t]} &= t^{|P|-1} \proD_{W \in P}{d_{W}} \\
\texT{anD } \PSi_{P}[t] &= \left( \prod_{J=1}^{S} {C^{(M_{j}-|P|_{j})}_{j}(0)} \right) \Pgf^{(|p|-1)}(t \gamma).\end{alIgnEd}$$ SincE $\pGf_{k+1|K+1}(T)$ is thE PGf corrEspondINg to thE cardinalItY DistriBUtion $p_{k+1|K+1}(N)$, $$\BeGin{aLigned}
&p_{k+1|k+1}(n) = \frac{1}{n!} \, \pgF^{(N)}_{k+1|K+1}(0) \\
&= \Frac{1}{n!} \, \left\{ \frac{d^N}{d t^n} \frAc{\DIsPLAysTylE \sum_{P \in \S}
\PsI_{P}[T] t^{|P|-1} \prOD_{W \in P}{d_{W}}}
{\DIsPLAYstYLe \sum_{P \in \S} \kappA_{P} \pgf^{(|P|-1)} \prod_{W \IN P}{d_{w}}} \right\}_{T = 0} \\
&= \fRac{\DIsplayStyle \SuM_{p \in \s} \left( \prod_{j=1}^{s}
{c^{(m_{j}-|P|_{J})}_{j}(0)} \right) \, \prOd_{W \in P}{D_{w}}
\left\{ \frAC{d^n}{d t^n} t^{|p|-1} \, \pgf^{(|P|-1)}(t \GamMa) \rIght\}_{T = 0}}
{N! | k+1|k+1}(t)
&= \frac{\dis playstyle\sum_ {P\in \ S} \ Psi_ {P}[t] \prod_{ W \in P}{\varphi_{W}[t]}}
{ \dis p la ystyl e \sum_ { P\ i n \ S} \ kap pa _ {P } \pg f^{ (|P|-1) } \prod_{W \i nP}{d_{W}}}.\ e nd {aligned}$ $ F or constant$t$ we ha ve $$ \ begin {al igned }
\dis p laysty le \prod_ {W \in P} { \varphi _ { W} [t]} &= t^{|P|-1} \pr o d_ { W \in P}{d_{W} } \\
\ te x t{ a n d } \P si_{P}[t]&= \lef t ( \prod _ {j = 1 } ^{s } {C^{(m_{j}-| P|_{j})}_{j } (0) } \rig ht ) \ p gf^{(| P|-1) }( t \g amma).\end{ alig ned}$$ Si nce $\ p gf_{k+1 | k+1}(t) $ is t hePGF cor r es po ndi ng tot he ca r din ality di st ri butio n $p _ { k + 1|k+ 1}( n)$, $$\b egin{aligned}
&p _{k+ 1 |k+ 1}(n) = \f rac{ 1} {n!}\, \pg f^{(n )} _{k+1|k+1}(0) \ \
&= \frac{1} {n! }\,\l eft\{ \frac{ d^n }{d t^n} \ frac{\d i spl ay s t y le \sum_{P \in \S} \P si_{P}[t ] t^{| P |- 1} \prod_{W \ inP}{d _ { W}}} { \d isplayst yle \s u m_ {P \in \S }\kappa _{ P}\pg f^{(| P |-1) } \pro d_{W \in P}{d _ {W}}} \right\} _ {t = 0} \\
&= \f r a c{ \ disp lay style \sum_ {P \ i n \S } \l e ft ( \ p rod_{ j=1}^ {s } {C^{(m_{j}-|P|_{ j} )}_{j} (0)}\right) \, \p rod_{W \in P } {d_{W}}
\l e ft\{ \frac{d^n }{d t ^n} t^{|P| - 1} \, \p gf^{( |P|-1)}( t \gamma) \ right\}_ {t= 0 }} { n! | k+1|k+1}(t)
&=_\frac{\displaystyle \sum_{P_\in \S} \Psi_{P}[t] \prod_{W_\in P}{\varphi_{W}[t]}}
__ __ _{\displaystyle \sum_{P \in_\S} \kappa_{P} \pgf^{(|P|-1)} \prod_{W_\in P}{d_{W}}}.\end{aligned}$$ For_constant_$t$ we have $$\begin{aligned}
\displaystyle \prod_{W \in P}{\varphi_{W}[t]} &= t^{|P|-1} \prod_{W \in P}{d_{W}} \\
\text{and }_\Psi_{P}[t]_&= \left(_\prod_{j=1}^{s}_{C^{(m_{j}-|P|_{j})}_{j}(0)}_\right) \pgf^{(|P|-1)}(t \gamma).\end{aligned}$$ Since $\pgf_{k+1|k+1}(t)$_is the PGF corresponding to_the cardinality_distribution $p_{k+1|k+1}(n)$, $$\begin{aligned}
&p_{k+1|k+1}(n) = \frac{1}{n!} \, \pgf^{(n)}_{k+1|k+1}(0) \\
&=_\frac{1}{n!}_\, \left\{ \frac{d^n}{d_t^n} \frac{\displaystyle \sum_{P \in \S}
_\Psi_{P}[t] t^{|P|-1} \prod_{W \in P}{d_{W}}}
_ {\displaystyle_\sum_{P_\in_\S} \kappa_{P} \pgf^{(|P|-1)} \prod_{W_\in P}{d_{W}}} \right\}_{t = 0} \\
&=_\frac{\displaystyle \sum_{P \in \S} \left( \prod_{j=1}^{s}_
{C^{(m_{j}-|P|_{j})}_{j}(0)} \right) \,_\prod_{W \in P}{d_{W}}
_ \left\{ \frac{d^n}{d t^n} t^{|P|-1}_\, \pgf^{(|P|-1)}(t_\gamma) \right\}_{t = 0}}
_ {n! |
:StructureOfHomogeneousRicciSolitonsAndTheAlekseevskiiConjecture]. The results obtained there are essential in our proof and we briefly recall those which we need.
We first observe that it suffices to prove the theorem for simply-connected manifolds. Now consider a simply-connected, expanding, semi-algebraic Ricci soliton on $G/K$. As $G/K$ is endowed with a $G$-invariant metric, $Ad(K)$ is contained in a compact subgroup of $Aut(G)$ and so we have a decomposition $\mathfrak g = \mathfrak p \oplus \mathfrak k$, where $\mathfrak p$ is an $Ad(K)$-complement to $\mathfrak k$. We fix the point $p = eK \in M = G/K$ and naturally identify $\mathfrak p$ with $T_pM$ as follows $$X\in\mathfrak p \quad \leftrightarrow \quad {{\left. \frac{d}{ds} \right|_{s=0}}}exp(sX)\cdot p = {{\left. \frac{d}{ds} \right|_{s=0}}}exp(sX)K.$$ Although there is more than one choice of $\mathfrak p$ that one can make, we apply the work [@LauretLafuente:StructureOfHomogeneousRicciSolitonsAndTheAlekseevskiiConjecture] in the sequel and so we choose, as they do, to have $B(\mathfrak k,\mathfrak p)=0$, where $B$ is the Killing form of $\mathfrak g$.
As $G/K$ admits an expanding Ricci soliton, we know from [@LauretLafuente:StructureOfHomogeneousRicciSolitonsAndTheAlekseevskiiConjecture] that the group $G$ decomposes as $N\rtimes U$ where $N$ is the nilradical and $U$ is a reductive subgroup which contains the stabilizer $K$. Thus the underlying manifold of $M$ may be considered as $N \times U/K$ and we naturally identify the point $p=eK\in G/K$ with $(e,eK)\in N \times U/K$. The subalgebra $\mathfrak u$ contains a subspace $\mathfrak h$ which is complementary to $\mathfrak k$, and so we have $T_pM \simeq \mathfrak p = | : StructureOfHomogeneousRicciSolitonsAndTheAlekseevskiiConjecture ]. The results obtained there are essential in our proof and we concisely echo those which we need.
We first note that it suffice to prove the theorem for simply - connected manifold. Now consider a merely - connected, expanding, semi - algebraic Ricci soliton on $ G / K$. As $ thousand / K$ is endowed with a $ G$-invariant metric, $ Ad(K)$ is check in a compendious subgroup of $ Aut(G)$ and so we have a decay $ \mathfrak g = \mathfrak phosphorus \oplus \mathfrak k$, where $ \mathfrak p$ is an $ Ad(K)$-complement to $ \mathfrak k$. We cook the point $ p = eK \in M = G / K$ and naturally identify $ \mathfrak p$ with $ T_pM$ as postdate $ $ X\in\mathfrak p \quad \leftrightarrow \quad { { \left. \frac{d}{ds } \right|_{s=0}}}exp(sX)\cdot p = { { \left. \frac{d}{ds } \right|_{s=0}}}exp(sX)K.$$ Although there cost more than one choice of $ \mathfrak p$ that one can stool, we apply the work [ @LauretLafuente: StructureOfHomogeneousRicciSolitonsAndTheAlekseevskiiConjecture ] in the sequel and so we choose, as they do, to have $ B(\mathfrak k,\mathfrak p)=0 $, where $ B$ is the Killing mannequin of $ \mathfrak g$.
As $ G / K$ admits an expanding Ricci soliton, we know from [ @LauretLafuente: StructureOfHomogeneousRicciSolitonsAndTheAlekseevskiiConjecture ] that the group $ G$ decomposes as $ N\rtimes U$ where $ N$ is the nilradical and $ U$ is a reductive subgroup which contains the stabilizer $ K$. Thus the underlying manifold of $ M$ may be considered as $ N \times U / K$ and we naturally identify the point $ p = eK\in G / K$ with $ (e, eK)\in N \times U / K$. The subalgebra $ \mathfrak u$ contains a subspace $ \mathfrak h$ which is complementary to $ \mathfrak k$, and so we suffer $ T_pM \simeq \mathfrak p = | :StrkctureOfHomogeneousRicciRolitonsAndTheAlekseevvkiiCohjecture]. The results obtained there ere wssenupal in our proof and de brieflj recall rhost which we need.
We first onfervs thac mt suffices to krove the thaorem for simpny-zounected manifolds. Now consider a sim[ly-connrched, expanding, femi-sjgebdaic Ricci soliton on $G/K$. As $G/K$ is sndowed with a $G$-invatiant metric, $Ad(K)$ is contaijed ln a compact subgrlup of $Aut(G)$ and wo we have a decomposiuimn $\mathfrai g = \mathfrak p \oplus \mathfrak y$, whexe $\mathfrak p$ is dn $Ad(K)$-complxment no $\mathfrak k$. We fix dhe poimt $p = eK \in M = G/K$ ane naturally identify $\kathfrak p$ with $T_pI$ as follmwa $$X\in\mathfrak p \wuqd \lextrichtafeow \quzd {{\lsft. \frwc{d}{vs} \right|_{s=0}}}exl(sX)\cdot p = {{\oeft. \frac{d}{ds} \right|_{s=0}}}txp(fQ)L.$$ Although tgere if iore than one choice of $\mathfrak p$ that ons can make, we apply the work [@LauretLafuente:SjructureOfRomogeneousRicciSolitonsAndTheAlekseevskiiConjectgre] ii ghe scqueu ajd so we choose, as they do, to have $B(\mathfrak h,\jauhfgak p)=0$, where $B$ is bhe Killing form og $\layrfrak g$.
As $G/K$ xdmits an expanding Ricci slliton, re kniw from [@LwureyLafuente:StructureOfHomogenwousRicciSolptonwAndTheAlekseevskinConjecture] chat tne grpup $G$ decomposes as $N\rtnmes U$ where $N$ is the nilrzaical and $U$ is a recuwtive subgroup which contayns the svabilnzer $K$. Tfus jhe undqrlying majifold of $M$ may be consifered av $N \times K/K$ and we naturally identify thx point $p=eK\in G/N$ wpth $(e,eK)\in N \tikes U/K$. The sufalgebra $\mathftak u$ concains x subspace $\mathfrek h$ which if complementasi to $\mathfrak k$, and sj we havw $T_pM \skoeq \mathfrak p = | :StructureOfHomogeneousRicciSolitonsAndTheAlekseevskiiConjecture]. The results obtained there are essential proof we briefly those which we it to prove the for simply-connected manifolds. consider a simply-connected, expanding, semi-algebraic Ricci on $G/K$. As $G/K$ is endowed with a $G$-invariant metric, $Ad(K)$ is contained a compact subgroup of $Aut(G)$ and so we have a decomposition $\mathfrak g \mathfrak \oplus k$, $\mathfrak p$ is an $Ad(K)$-complement to $\mathfrak k$. We fix the point $p = eK \in = G/K$ and naturally identify $\mathfrak p$ with as follows $$X\in\mathfrak p \leftrightarrow \quad {{\left. \frac{d}{ds} \right|_{s=0}}}exp(sX)\cdot = \frac{d}{ds} \right|_{s=0}}}exp(sX)K.$$ there more one choice of p$ that one can make, we apply the work [@LauretLafuente:StructureOfHomogeneousRicciSolitonsAndTheAlekseevskiiConjecture] in the sequel and so we choose, they do, $B(\mathfrak k,\mathfrak where is Killing form of As $G/K$ admits an expanding Ricci from [@LauretLafuente:StructureOfHomogeneousRicciSolitonsAndTheAlekseevskiiConjecture] that the group $G$ decomposes as U$ where is the nilradical and $U$ is reductive subgroup which contains the stabilizer $K$. Thus underlying manifold of $M$ may be considered as $N \times U/K$ and we naturally identify $p=eK\in G/K$ with $(e,eK)\in \times U/K$. The $\mathfrak contains subspace h$ which complementary to $\mathfrak k$, and so we have $T_pM \simeq \mathfrak = | :StructureOfHomogeneousRicCiSolitonsandThEAlEksEeVskiIConJecture]. The resuLTs obTained there are essentiaL in ouR pROof aND wE brieFly recaLL tHOSe wHiCh We nEeD.
we First ObsErve thaT it sufficeS to PrOve the theoreM FoR simply-conNecTed manifolds. now ConsidEr A siMPly-coNneCted, eXpandiNG, semi-aLgebraic RIcCI solitON on $G/K$. As $g/k$ Is EndoWed with a $G$-invarianT MeTRic, $Ad(K)$ is contaiNed in a CoMPaCT SubGroUp of $Aut(G)$ anD sO we haVE a decomPOsITIOn $\mAThfrak g = \mathfrAk p \oplus \matHFraK k$, wherE $\mAthFRak p$ is An $Ad(K)$-CoMPleMent to $\mathfRak k$. we fix the pOint $p = ek \In M = G/K$ anD NaturalLy idenTifY $\maThfrAK p$ WiTh $T_PM$ AS foLLoWs $$X\IN\maThfrak p \qUaD \lEftriGhtaRROW \Quad {{\LefT. \fraC{d}{ds} \rIght|_{s=0}}}exp(sX)\cdoT p = {{\lEft. \fRAc{d}{Ds} \rigHt|_{s=0}}}exP(sX)K.$$ alThougH there Is morE tHan one choice of $\mAthfRak p$ that oNe cAn MakE, wE applY The worK [@LaUreTLafuenTe:StrucTUreofhOMOgEneousRicciSolitonSANDthEAlekseeVskiiCONjEcTUre] in the SeQueL and SO We choOse, aS ThEy do, to haVe $B(\matHFrAk K,\mathfrAk P)=0$, where $b$ iS thE KiLling FOrm oF $\mathfRak g$.
As $G/K$ AdmitS An expanding RicCI soliton, we knoW FrOM [@laURetLAfuEnte:StructuReOfhOmogEneoUSRIccIsolitOnsAnDTHEALEkseevskiiConjecturE] tHat the Group $g$ decomposes as $n\rtimes U$ whERE $n$ is the niLradICaL And $U$ is a reductiVe subGroup which COntains tHe staBilizer $K$. thus the unDERlying maNifOld Of $M$ May BE CoNsidered as $N \tiMES U/K$ aNd We naturAllY identiFy tHe pOinT $p=ek\iN G/K$ with $(e,ek)\in N \timeS U/k$. THe SuBalGebra $\MAthfrak u$ CoNtaInS a sUbspaCE $\mathfRak h$ wHich Is CoMPleMentary TO $\mATHfraK k$, AnD so wE haVe $t_pM \siMeq \mAThfRak p = | :StructureOfHomogeneousRic ciSolitons AndTh eAl eks ee vski iCon jecture]. Ther esul ts obtained there areessen ti a l in ou r pro of andw eb r ief ly r eca ll th ose w hic h we ne ed.
We fi rst o bserve thati tsuffices t o p rove the the ore m forsi mpl y -conn ect ed ma nifold s . Nowconsideras imply- c onnecte d , e xpan ding, semi-algebr a ic Ricci solitonon $G/ K$ . A s $G/ K$is endowed w ith a $G$-inv a ri a n t me t ric, $Ad(K)$is containe d in a com pa cts ubgrou p of$A u t(G )$ and so w e ha ve a deco mposit i on $\ma t hfrak g = \ma thf rak p \ o pl us \m at h fra k k $,w her e $\math fr ak p$ i s an $ A d (K)$ -co mple mentto $\mathfrak k$ . We fix thepoint $p=eK \i n M =G/K$an d naturally ide ntif y $\mathf rak p $ w it h $T_ p M$ asfol low s $$X\i n\mathf r akp\ q ua d \leftrightarrow\q u a d{{\left. \frac { d} {d s } \right |_ {s= 0}}} e x p(sX) \cdo t p = {{\le ft. \f r ac {d }{ds} \ ri ght|_{ s= 0}} }ex p(sX) K .$$Althou gh there is m o re than one ch o ice of $\math f ra k p$ that on e can make, wea pply the wo rk[ @Laur etLaf ue n te : StructureOfHomogene ou sRicci Solit onsAndTheAlek seevskiiCo n j e cture] i n th e s e quel and so we choo se, as the y do, tohave$B(\math frak k,\m a t hfrak p) =0$ , w her e $ B $ i s the Killing f ormof $\math fra k g$.
As$G/ K$adm it s an expa nding Ri cc iso li ton , wek now from [ @La ur etL afuen t e:Stru cture OfHo mo ge n eou sRicciS o li t o nsAn dT he Alek see vs kiiCo njec t ure ] thatthe group $G $ dec om po ses as$N\rtimes U$wh ere $N$ is t henilrad i c al and $ U$ is a reductive subgr o up whic h c ontai ns t he stabil ize r $K$. Th u s theunderl yingma nif o l d of$ M $may b e consider e d as $N \ ti mesU/K$ an d we naturally ide n tif y the point $ p=e K\in G /K $ w i th $(e ,e K )\i n N \times U/K$.The subalg eb r a$\mathfrak u$co ntainsa subsp ace $ \ mathfra k h$ whic h is comp le ment a r y t o $\mathfr ak k$, a nd so weh ave $ T _p M \si meq \math fr akp = | :StructureOfHomogeneousRicciSolitonsAndTheAlekseevskiiConjecture]. The_results obtained_there are essential in_our proof_and_we briefly_recall_those which we_need.
We first observe_that it suffices to_prove the theorem_for_simply-connected manifolds. Now consider a simply-connected, expanding, semi-algebraic Ricci soliton on $G/K$. As $G/K$_is_endowed with_a_$G$-invariant_metric, $Ad(K)$ is contained in_a compact subgroup of $Aut(G)$_and so_we have a decomposition $\mathfrak g = \mathfrak_p_\oplus \mathfrak k$,_where $\mathfrak p$ is an $Ad(K)$-complement to $\mathfrak k$._We fix the point $p =_eK \in M_=_G/K$_and naturally identify $\mathfrak_p$ with $T_pM$ as follows $$X\in\mathfrak_p \quad \leftrightarrow \quad {{\left._\frac{d}{ds} \right|_{s=0}}}exp(sX)\cdot p = {{\left. \frac{d}{ds} \right|_{s=0}}}exp(sX)K.$$_Although there is more than one_choice of $\mathfrak p$ that_one can_make, we apply the work_[@LauretLafuente:StructureOfHomogeneousRicciSolitonsAndTheAlekseevskiiConjecture] in the_sequel and_so we choose,_as they do, to have $B(\mathfrak_k,\mathfrak p)=0$, where_$B$ is the Killing form of_$\mathfrak_g$.
As $G/K$ admits_an_expanding_Ricci soliton,_we know from_[@LauretLafuente:StructureOfHomogeneousRicciSolitonsAndTheAlekseevskiiConjecture]_that the_group_$G$ decomposes as $N\rtimes U$ where_$N$_is the nilradical and $U$ is a_reductive subgroup which contains_the_stabilizer $K$. Thus the_underlying manifold of $M$ may_be considered as $N \times U/K$_and we_naturally identify_the point $p=eK\in G/K$ with $(e,eK)\in N \times U/K$. The subalgebra_$\mathfrak u$ contains a subspace $\mathfrak_h$ which is complementary_to $\mathfrak_k$,_and so we_have_$T_pM \simeq_\mathfrak p = |
}^0}(P_n,J).$$
Denote by ${\operatorname{Perf}}(\hat{S})\subset D(\hat{S})$ the full triangulated subcategory which is generated by the DG $\hat{S}^0$-module $\hat{S}$.
Assume that the DG algebra ${{\mathcal C}}$ is admissible and finite dimensional. Then
a\) The contravariant functor $\nabla :D(\hat{S})\to D({{\mathcal C}})$ is full and faithful on the subcategory ${\operatorname{Perf}}(\hat{S})$.
b\) $\nabla (\hat{S})$ is isomorphic to $k$.
Denote by $m\subset \hat{S}^0$ the maximal ideal and put $S_n:=\hat{S}^0/m^n\hat{S}^0$. Since the DG algebra ${{\mathcal C}}$ is finite dimensional $S_n$ is also finite dimensional for all $n$. We need a few lemmas.
Let $K$ be a DG $\hat{S}^0$-module such that $\dim
_kK<\infty$. Then the natural morphism of complexes $$\Hom _{\hat{S}^0}(K,B{{\mathcal C}}\otimes {{\mathcal C}})\to \Hom _{\hat{S}^0}(K,J)$$ is a quasi-isomorphism.
Notice that since the algebra $\hat{S}$ is local, every element $x\in m$ acts on $K$ as a nilpotent operator. Hence in particular $m^nK=0$ for $n>>0$. For the same reason the DG $\hat{S}^0$-module $K$ has a filtration with subquotients isomorphic to $k$. Thus we may prove the assertion by induction on $\dim K$. If $K=k$, then this is part b) of Proposition 13.4. Otherwise we can find a short exact sequence of DG $\hat{S}^0$-modules $$0\to M\to K\to N\to 0,$$ such that $\dim M,\dim N <\dim K$.
[**Sublemma.**]{} [*The sequence of complexes $$0\to \Hom _{\hat{S}^0}(N, B{{\mathcal C}}\otimes {{\mathcal C}})\to \Hom _{\hat{S}^0}(K, B{{\mathcal C}}\otimes
{{\mathcal C}}) \to \Hom _{\hat{S}^0}(M, | } ^0}(P_n, J).$$
Denote by $ { \operatorname{Perf}}(\hat{S})\subset D(\hat{S})$ the full triangulated subcategory which is generated by the DG $ \hat{S}^0$-module $ \hat{S}$.
Assume that the DG algebra $ { { \mathcal C}}$ is admissible and finite dimensional. Then
a\) The contravariant functor $ \nabla: D(\hat{S})\to D({{\mathcal C}})$ is wide and close on the subcategory $ { \operatorname{Perf}}(\hat{S})$.
b\) $ \nabla (\hat{S})$ is isomorphic to $ k$.
Denote by $ m\subset \hat{S}^0 $ the maximal ideal and put $ S_n:=\hat{S}^0 / m^n\hat{S}^0$. Since the DG algebra $ { { \mathcal C}}$ is finite dimensional $ S_n$ is besides finite dimensional for all $ n$. We need a few lemma.
Let $ K$ be a DG $ \hat{S}^0$-module such that $ \dim
_ kK<\infty$. Then the lifelike morphism of building complex $ $ \Hom _ { \hat{S}^0}(K, B{{\mathcal C}}\otimes { { \mathcal C}})\to \Hom _ { \hat{S}^0}(K, J)$$ is a quasi - isomorphism.
Notice that since the algebra $ \hat{S}$ is local, every component $ x\in m$ acts on $ K$ as a nilpotent operator. therefore in particular $ m^nK=0 $ for $ n>>0$. For the same reason the DG $ \hat{S}^0$-module $ K$ has a filtration with subquotients isomorphous to $ k$. Thus we may prove the affirmation by induction on $ \dim K$. If $ potassium = k$, then this is part b) of Proposition 13.4. Otherwise we can find a short accurate sequence of DG $ \hat{S}^0$-modules $ $ 0\to M\to K\to N\to 0,$$ such that $ \dim M,\dim N < \dim K$.
[ * * Sublemma. * * ] { } [ * The sequence of complex $ $ 0\to \Hom _ { \hat{S}^0}(N, B{{\mathcal C}}\otimes { { \mathcal C}})\to \Hom _ { \hat{S}^0}(K, B{{\mathcal C}}\otimes
{ { \mathcal C } }) \to \Hom _ { \hat{S}^0}(M, | }^0}(P_n,J).$$
Fenote by ${\operatorname{Pevf}}(\hat{S})\subset D(\haj{S})$ the fnll trizngulatea subcategory which is genereted by tye DG $\hat{S}^0$-module $\hat{S}$.
Arsume than the DG qlgeura ${{\mathcal C}}$ is admissinje ahf fiuive dimensional. Jhen
a\) The cottravariant futcgox $\nabla :D(\hat{S})\to D({{\mathcal C}})$ is full agd faitnfkl on the subcwtegpwy ${\olvrctorname{Perf}}(\hat{S})$.
b\) $\nabla (\hat{S})$ is isomorkhic to $k$.
Denote by $m\subset \hat{S}^0$ the maximal ldeap and put $S_n:=\hat{S}^0/m^n\jat{S}^0$. Since jge QT algebra ${{\maghcal C}}$ is finite dimehsional $S_n$ is also finite dimenrionak for all $b$. Qe jged a few lenmas.
Lvt $K$ be a DG $\mst{S}^0$-modgle sucn that $\dim
_kK<\innty$. Tien rhe natural morphism mf complexes $$\Hom _{\hwt{S}^0}(K,B{{\mathwam C}}\otimes {{\mathcal C}})\ro \Hok _{\had{S}^0}(K,J)$$ us x qhaxi-jsomorohiam.
Notice tgat since tye algebra $\hat{S}$ is kosql, every elemsnt $x\ig i$ acts on $K$ as a nilpotent operator. Henbe ih particular $m^nK=0$ for $n>>0$. Dor the same reason tje DG $\hat{F}^0$-module $K$ has a filtration with subquotients isommrphir go $j$. Bhus qe may prove the assertion by induction on $\dim H$. Ig $L=k$, then this if part b) of Pgoljsition 13.4. Othetwise wz czn find a short exwct seqoence if DG $\hat{F}^0$-modiles $$0\to M\to K\to N\to 0,$$ such tyat $\dim M,\dim B <\dim K$.
[**Sublemma.**]{} [*Thz sequence oy compkexes $$0\to \Hom _{\hat{S}^0}(N, B{{\mathcal E}}\otimea {{\mathcal C}})\ho \Hom _{\haf{R}^0}(K, B{{\mathcal C}}\otioes
{{\kadhcal C}}) \uu \Hom _{\hat{S}^0}(M, | }^0}(P_n,J).$$ Denote by ${\operatorname{Perf}}(\hat{S})\subset D(\hat{S})$ the full which generated by DG $\hat{S}^0$-module $\hat{S}$. ${{\mathcal is admissible and dimensional. Then a\) contravariant functor $\nabla :D(\hat{S})\to D({{\mathcal C}})$ full and faithful on the subcategory ${\operatorname{Perf}}(\hat{S})$. b\) $\nabla (\hat{S})$ is isomorphic to Denote by $m\subset \hat{S}^0$ the maximal ideal and put $S_n:=\hat{S}^0/m^n\hat{S}^0$. Since the DG ${{\mathcal is dimensional is also finite dimensional for all $n$. We need a few lemmas. Let $K$ be a $\hat{S}^0$-module such that $\dim _kK<\infty$. Then the natural of complexes $$\Hom _{\hat{S}^0}(K,B{{\mathcal {{\mathcal C}})\to \Hom _{\hat{S}^0}(K,J)$$ is quasi-isomorphism. that since algebra is every element $x\in acts on $K$ as a nilpotent operator. Hence in particular $m^nK=0$ for $n>>0$. For the same reason DG $\hat{S}^0$-module a filtration subquotients to Thus we may assertion by induction on $\dim K$. this is part b) of Proposition 13.4. Otherwise can find short exact sequence of DG $\hat{S}^0$-modules M\to K\to N\to 0,$$ such that $\dim M,\dim <\dim K$. [**Sublemma.**]{} [*The sequence of complexes $$0\to \Hom _{\hat{S}^0}(N, B{{\mathcal C}}\otimes {{\mathcal C}})\to \Hom C}}\otimes {{\mathcal C}}) \to _{\hat{S}^0}(M, | }^0}(P_n,J).$$
Denote by ${\operatorname{PeRf}}(\hat{S})\subsEt D(\haT{S})$ tHe fUlL triAnguLated subcategoRY whiCh is generated by the DG $\haT{S}^0$-modUlE $\Hat{S}$.
aSsUme thAt the DG ALgEBRa ${{\mAtHcAl C}}$ Is ADmIssibLe aNd finitE dimensionAl. THeN
a\) The contravARiAnt functor $\NabLa :D(\hat{S})\to D({{\maThcAl C}})$ is fUlL anD FaithFul On the SubcatEGory ${\opEratornamE{PERf}}(\hat{S})$.
B\) $\Nabla (\haT{s})$ Is IsomOrphic to $k$.
Denote by $M\SuBSet \hat{S}^0$ the maxiMal ideAl ANd PUT $S_n:=\Hat{s}^0/m^n\hat{S}^0$. SinCe The DG ALgebra ${{\mAThCAL c}}$ is FInite dimensioNal $S_n$ is also FIniTe dimeNsIonAL for alL $n$. We nEeD A feW lemmas.
Let $K$ Be a Dg $\hat{S}^0$-moduLe such THat $\dim
_kk<\Infty$. ThEn the nAtuRal MorpHIsM oF coMpLExeS $$\hoM _{\haT{s}^0}(K,B{{\Mathcal C}}\OtImEs {{\matHcal c}})\TO \hOm _{\haT{S}^0}(K,j)$$ is a Quasi-Isomorphism.
NoTicE thaT SinCe the AlgebRa $\haT{S}$ Is locAl, everY elemEnT $x\in m$ acts on $K$ as a NilpOtent operAtoR. HEncE iN partICular $m^NK=0$ fOr $n>>0$. for the sAme reasON thE Dg $\HAT{S}^0$-Module $K$ has a filtratIoN WItH subquotIents iSOmOrPHic to $k$. ThUs We mAy prOVE the aSserTIoN by inducTion on $\DIm k$. IF $K=k$, then ThIs is paRt B) of proPositIOn 13.4. OtHerwisE we can fiNd a shORt exact sequencE Of DG $\hat{S}^0$-modulES $$0\tO m\To k\To N\tO 0,$$ suCh that $\dim M,\dIm N <\dIM K$.
[**SuBlemMA.**]{} [*THe sEQuencE of coMpLExES $$0\to \Hom _{\hat{S}^0}(N, B{{\mathcal c}}\oTimes {{\mAthcaL C}})\to \Hom _{\hat{S}^0}(K, B{{\Mathcal C}}\otIMES
{{\mathcal c}}) \to \HOM _{\hAT{S}^0}(M, | }^0}(P_n,J).$$
Denote by${\operato rname {Pe rf} }( \hat {S}) \subset D(\hat { S})$ the full triangulated subc at e gory wh ich i s gener a te d byth eDG$\ h at {S}^0 $-m odule $ \hat{S}$.
As su me that theD Galgebra ${ {\m athcal C}}$isadmiss ib lea nd fi nit e dim ension a l. The n
a\) Th ec ontrav a riant f u n ct or $ \nabla :D(\hat{S} ) \t o D({{\mathcalC}})$is fu l l an d f aithful on t he su b categor y $ { \ o per a torname{Perf} }(\hat{S})$ .
b \) $\n ab la( \hat{S })$ i si som orphic to $ k$.
Denote b y $m\s u bset \h a t{S}^0$ the m axi mal ide a lan d p ut $S_ n := \ha t {S} ^0/m^n\h at {S }^0$. Sin c e t he D G a lgeb ra ${ {\mathcal C}} $ i s fi n ite dime nsion al $ S_ n$ is alsofinit edimensional for all $n$. Wenee da f ew lemm a s.
Le t $ K$be a DG $\hat{ S }^0 $- m o d ul e such that $\dim_k K < \i nfty$. T hen th e n at u ral morp hi smof c o m plexe s $$ \ Ho m _{\hat {S}^0} ( K, B{ {\mathc al C}}\o ti mes {{ \math c al C }})\to \Hom _{ \hat{ S }^0}(K,J)$$ is a quasi-isomo r ph i s m.
Not ice that since the alge bra$ \h at{ S }$ is loca l, ev e ry element $x\in m$ a cts on $K$as a nilpoten t operator . H ence inpart i cu l ar $m^nK=0$ fo r $n> >0$. For t h e same r eason the DG$\hat{S}^ 0 $ -module$K$ ha s a fi l t ra tion with sub q u otie nt s isomo rph ic to $ k$. Th uswema y prove t he asser ti on b yind uctio n on $\di mK$. I f $ K=k$, then t his i s pa rt b ) of Propos i ti o n 13. 4. O ther wis ewe ca n fi n d a shortexact seq uen c e of D G$\hat{S }^0$-modules$$ 0\to M\toK\ toN\to 0 , $ $ such t hat $\dim M,\dim N <\di m K$.
[ **S ublem ma.* *]{} [*Th e s equenc e o f compl exes $ $0\to \ Hom _ {\hat { S }^ 0}( N, B{{\mathc a l C} }\oti me s {{ \mathca l C}})\to \Hom _{\ h at{ S}^0}(K, B{{\ mat hcal C }} \ot i me s
{{ \m a thc a l C}}) \to \Hom_{\hat{S}^ 0} ( M, | }^0}(P_n,J).$$
Denote by_${\operatorname{Perf}}(\hat{S})\subset D(\hat{S})$_the full triangulated subcategory_which is_generated_by the_DG_$\hat{S}^0$-module $\hat{S}$.
Assume that_the DG algebra_${{\mathcal C}}$ is admissible_and finite dimensional._Then
a\)_The contravariant functor $\nabla :D(\hat{S})\to D({{\mathcal C}})$ is full and faithful on the subcategory_${\operatorname{Perf}}(\hat{S})$.
b\)_$\nabla (\hat{S})$_is_isomorphic_to $k$.
Denote by $m\subset \hat{S}^0$_the maximal ideal and put_$S_n:=\hat{S}^0/m^n\hat{S}^0$. Since_the DG algebra ${{\mathcal C}}$ is finite dimensional_$S_n$_is also finite_dimensional for all $n$. We need a few lemmas.
Let_$K$ be a DG $\hat{S}^0$-module such_that $\dim
_kK<\infty$. Then_the_natural_morphism of complexes $$\Hom__{\hat{S}^0}(K,B{{\mathcal C}}\otimes {{\mathcal C}})\to \Hom _{\hat{S}^0}(K,J)$$_is a quasi-isomorphism.
Notice that since the_algebra $\hat{S}$ is local, every element $x\in_m$ acts on $K$ as a_nilpotent operator. Hence in particular_$m^nK=0$ for_$n>>0$. For the same reason_the DG $\hat{S}^0$-module_$K$ has_a filtration with_subquotients isomorphic to $k$. Thus we_may prove the_assertion by induction on $\dim K$._If_$K=k$, then this_is_part_b) of_Proposition 13.4. Otherwise_we_can find_a_short exact sequence of DG $\hat{S}^0$-modules_$$0\to_M\to K\to N\to 0,$$ such that $\dim_M,\dim N <\dim K$.
[**Sublemma.**]{}_[*The_sequence of complexes $$0\to_\Hom _{\hat{S}^0}(N, B{{\mathcal C}}\otimes {{\mathcal_C}})\to \Hom _{\hat{S}^0}(K, B{{\mathcal C}}\otimes
{{\mathcal C}})_\to \Hom__{\hat{S}^0}(M, |
_{i-1,j} ) \vf_i(\bar{\vx},\bar{\vu}) \notag + (t_{i}-t_{i-1}) \left( \vA_i \partial_{t_j}\!\bar{\vx} + \vB_i \partial_{t_j}\!\bar{\vu} \right).$$ For simplicity in notation, we drop this dependencies on time, $z$. Using the continuity condition of the state trajectory, the order of derivatives on the right hand side of the equation is changed which is resulted in. Furthermore, since the control input trajectory in SLQ comprises a time-varying feedforward term and a time-varying linear state feedback, its sensitivity can be calculated as. Moreover, for the initial condition of this equation we have $\partial_{t_j}\bar{\vx}(z=0) = \mathbf{0}$ due to the fixed initial state. Based on Lemma 1, the second order approximation of the intermediate cost $L(\vx,\vu)$ around the nominal trajectories (which are in turn a function of $t_j$) can be written as $$\begin{aligned}
& L(z,\bar{\vx}(s)+\delta\!\vx,\bar{\vu}(s)+\delta\!\vu) \simeq
q(z) +
\begin{bmatrix} \vq(z) \\ \vr(z) \end{bmatrix}^\top
\begin{bmatrix} \partial_{t_j}\vx(z) \\ \partial_{t_j}\vu(z) \end{bmatrix} (t_j-\bar{t}_j)
\notag \\
& \hspace{2mm}
+\begin{bmatrix} \delta\!\vx \\ \delta\!\vu \end{bmatrix}^\top\!
\left(
\begin{bmatrix} \vq(z) \\ \vr(z) \end{bmatrix} +
\begin{bmatrix} \vQ(z) & \vP(z) \\ \vP(z)^\top & \vR(z) \end{bmatrix}
\begin{bmatrix} \partial_{t_j}\vx(z) \\ \partial_{t_j}\vu(z) \end{bmatrix}
(t_j-\bar{t}_j) \right) \notag \\
& \hspace{2mm}
| _ { i-1,j }) \vf_i(\bar{\vx},\bar{\vu }) \notag + (t_{i}-t_{i-1 }) \left (\vA_i \partial_{t_j}\!\bar{\vx } + \vB_i \partial_{t_j}\!\bar{\vu } \right).$$ For simplicity in notation, we drop this dependencies on meter, $ z$. use the continuity condition of the state trajectory, the order of derivative on the right hand english of the equality is changed which is resulted in. Furthermore, since the control input trajectory in SLQ comprises a prison term - varying feedforward term and a fourth dimension - vary linear state feedback, its sensitivity can be calculated as. Moreover, for the initial condition of this equality we have $ \partial_{t_j}\bar{\vx}(z=0) = \mathbf{0}$ due to the fixed initial state. base on Lemma 1, the second order approximation of the average cost $ L(\vx,\vu)$ around the nominative trajectories (which are in turn a function of $ t_j$) can be written as $ $ \begin{aligned }
& L(z,\bar{\vx}(s)+\delta\!\vx,\bar{\vu}(s)+\delta\!\vu) \simeq
q(z) +
\begin{bmatrix } \vq(z) \\ \vr(z) \end{bmatrix}^\top
\begin{bmatrix } \partial_{t_j}\vx(z) \\ \partial_{t_j}\vu(z) \end{bmatrix } (t_j-\bar{t}_j)
\notag \\
& \hspace{2 mm }
+ \begin{bmatrix } \delta\!\vx \\ \delta\!\vu \end{bmatrix}^\top\!
\left (
\begin{bmatrix } \vq(z) \\ \vr(z) \end{bmatrix } +
\begin{bmatrix } \vQ(z) & \vP(z) \\ \vP(z)^\top & \vR(z) \end{bmatrix }
\begin{bmatrix } \partial_{t_j}\vx(z) \\ \partial_{t_j}\vu(z) \end{bmatrix }
(t_j-\bar{t}_j) \right) \notag \\
& \hspace{2 mm } | _{i-1,j} ) \vf_i(\bar{\vx},\bar{\vu}) \notag + (t_{i}-u_{i-1}) \left( \vA_i \partico_{t_j}\!\bar{\tx} + \vB_i \partial_{g_j}\!\bar{\vu} \right).$$ For simplicity ib notqtion, we drop this depdndencies on time, $z$. Uwung the coifinuity condiflon oy vhe state trajegtory, the osder of derivadixed on the right hand side of the equwtion ix fhanged which ys rtsujted pn. Furthermore, since the controm input trajectory im SLQ comprises a time-varylng veedforward term ajd a time-vatgind linear statd feedback, its sensitibity can be calculated as. Moreoxer, fpr the inijnql fmndition of this equation we have $\pastial_{t_j}\nar{\vx}(z=0) = \mathbf{0}$ dux to the fixed initial stete. Based on Lemma 1, jhe second oxder approximation of tye injermegiatd cort $M(\ve,\vu)$ arounf tie nominal frajectoriew (which are in turn a dunction of $t_n$) can fe written as $$\begin{aligned}
& L(z,\bar{\vx}(s)+\delta\!\vq,\bar{\bu}(s)+\delta\!\vu) \simeq
q(z) +
\begun{bmatrix} \vq(z) \\ \vr(z) \enf{bmatrix}^\tjp
\begin{bmatrix} \partial_{t_j}\vx(z) \\ \partial_{t_j}\vu(z) \end{bmadrix} (v_j-\car{u}_j)
\kjgqg \\
& \hspace{2mm}
+\begin{bmatrix} \delta\!\vx \\ \delta\!\vu \end{fjauriq}^\top\!
\left(
\begin{bmabrix} \vq(z) \\ \vr(z) \end{bkahrov} +
\begin{bmatrib} \vQ(z) & \vP(a) \\ \vP(z)^\top & \vR(z) \end{hmatrix}
\fegin{vmatrix} \pwrtisl_{t_j}\vx(z) \\ \partial_{t_j}\vu(z) \end{bmqtrix}
(t_j-\bar{t}_l) \rught) \notag \\
& \hspace{2lm}
| _{i-1,j} ) \vf_i(\bar{\vx},\bar{\vu}) \notag + (t_{i}-t_{i-1}) \left( + \partial_{t_j}\!\bar{\vu} \right).$$ simplicity in notation, time, Using the continuity of the state the order of derivatives on the hand side of the equation is changed which is resulted in. Furthermore, since control input trajectory in SLQ comprises a time-varying feedforward term and a time-varying state its can calculated as. Moreover, for the initial condition of this equation we have $\partial_{t_j}\bar{\vx}(z=0) = \mathbf{0}$ due the fixed initial state. Based on Lemma 1, second order approximation of intermediate cost $L(\vx,\vu)$ around the trajectories are in a of can be written $$\begin{aligned} & L(z,\bar{\vx}(s)+\delta\!\vx,\bar{\vu}(s)+\delta\!\vu) \simeq q(z) + \begin{bmatrix} \vq(z) \\ \vr(z) \end{bmatrix}^\top \begin{bmatrix} \partial_{t_j}\vx(z) \\ \partial_{t_j}\vu(z) \end{bmatrix} (t_j-\bar{t}_j) \\ & \delta\!\vx \\ \end{bmatrix}^\top\! \begin{bmatrix} \\ \vr(z) \end{bmatrix} \vQ(z) & \vP(z) \\ \vP(z)^\top & \partial_{t_j}\vx(z) \\ \partial_{t_j}\vu(z) \end{bmatrix} (t_j-\bar{t}_j) \right) \notag \\ \hspace{2mm} | _{i-1,j} ) \vf_i(\bar{\vx},\bar{\vu}) \notag + (t_{i}-t_{i-1}) \lEft( \vA_i \partIal_{t_j}\!\Bar{\Vx} + \vb_i \PartIal_{t_J}\!\bar{\vu} \right).$$ For SImplIcity in notation, we drop tHis dePeNDencIEs On timE, $z$. Using THe CONtiNuItY coNdITiOn of tHe sTate traJectory, the OrdEr Of derivativeS On The right haNd sIde of the equaTioN is chaNgEd wHIch is ResUlted In. FurtHErmore, Since the cOnTRol inpUT trajecTORy In SLq comprises a time-vaRYiNG feedforward teRm and a TiME-vARYinG liNear state fEeDback, ITs sensiTIvITY Can BE calculated as. moreover, for THe iNitial CoNdiTIon of tHis eqUaTIon We have $\partiAl_{t_j}\Bar{\vx}(z=0) = \matHbf{0}$ due TO the fixED initiaL state. basEd oN LemMA 1, tHe SecOnD OrdER aPprOXimAtion of tHe InTermeDiatE COST $L(\vx,\Vu)$ aRounD the nOminal trajectOriEs (whICh aRe in tUrn a fUnctIoN of $t_j$) Can be wRitteN aS $$\begin{aligned}
& L(z,\Bar{\vX}(s)+\delta\!\vx,\Bar{\Vu}(S)+\deLtA\!\vu) \siMEq
q(z) +
\beGin{BmaTrix} \vq(z) \\ \Vr(z) \end{bMAtrIx}^\TOP
\BeGin{bmatrix} \partial_{t_J}\vX(Z) \\ \PaRtial_{t_j}\vU(z) \end{bMAtRiX} (T_j-\bar{t}_j)
\nOtAg \\
& \hSpacE{2MM}
+\begiN{bmaTRiX} \delta\!\vx \\ \Delta\!\vU \EnD{bMatrix}^\tOp\!
\Left(
\beGiN{bmAtrIx} \vq(z) \\ \VR(z) \enD{bmatrIx} +
\begin{bMatriX} \VQ(z) & \vP(z) \\ \vP(z)^\top & \vR(Z) \End{bmatrix}
\begIN{bMATrIX} \parTiaL_{t_j}\vx(z) \\ \partiAl_{t_j}\VU(z) \enD{bmaTRiX}
(t_j-\BAr{t}_j) \rIght) \nOtAG \\
& \hSPace{2mm}
| _{i-1,j} ) \vf_i(\bar{\vx} ,\bar{\vu} ) \no tag +(t _{i} -t_{ i-1}) \left( \ v A_i\partial_{t_j}\!\bar{\ vx} + \ v B_i\ pa rtial _{t_j}\ ! \b a r {\v u} \ rig ht ) .$ $ For si mplicit y in notat ion ,we drop this de pendencies on time, $z$.Usi ng the c ont i nuity co nditi on oft he sta te trajec to r y, the order o f de riva tives on the righ t h a nd side of the equat io n i s cha nge d which is r esult e d in. F u rt h e r mor e , since the c ontrol inpu t tr ajecto ry in SLQ co mpris es a t ime-varying fee dforwardterm a n d a tim e -varyin g line arsta te f e ed ba ck, i t s s e ns iti v ity can beca lc ulate d as . M o reov er, for theinitial condi tio n of thi s equ ation weha ve $\ partia l_{t_ j} \bar{\vx}(z=0)= \m athbf{0}$ du etoth e fix e d init ial st ate. Ba sed onL emm a1 , th e second order app ro x i ma tion ofthe in t er me d iate cos t$L( \vx, \ v u)$ a roun d t he nomin al tra j ec to ries (w hi ch are i n t urn a fu n ctio n of $ t_j$) ca n bew ritten as $$\b e gin{aligned}& L ( z ,\ b ar{\ vx} (s)+\delta\ !\vx , \bar {\vu } (s )+\ d elta\ !\vu) \ s im e q
q(z) +
\begin{bm at rix} \ vq(z) \\ \vr(z) \e nd{bmatrix } ^ \ top
\be gin{ b ma t rix} \partial_ {t_j} \vx(z) \\\ partial_ {t_j} \vu(z) \ end{bmatr i x } (t_j-\ bar {t} _j)
\n o t ag \\
& \hspac e { 2mm} +\begin {bm atrix}\de lta \!\ vx\\ \delta\! \vu \end {b ma tr ix }^\ top\!
\left(\b egi n{ bma trix} \vq(z) \\ \ vr(z )\e n d{b matrix} +\ b egin {b ma trix } \ vQ (z) & \vP ( z)\\ \vP( z)^\top & \v R (z)\e nd {bmatri x}
\begin{bma tr ix} \parti al _{t _j}\vx ( z ) \\ \pa rtial_{t_j}\vu(z) \end{ b matrix}
( t_j-\ bar{ t}_j) \ri ght ) \not ag\ \
& \h space{ 2mm}
| _{i-1,j} )_\vf_i(\bar{\vx},\bar{\vu}) \notag_+ (t_{i}-t_{i-1}) \left( \vA_i_\partial_{t_j}\!\bar{\vx} +_\vB_i_\partial_{t_j}\!\bar{\vu} \right).$$_For_simplicity in notation,_we drop this_dependencies on time, $z$._Using the continuity_condition_of the state trajectory, the order of derivatives on the right hand side of_the_equation is_changed_which_is resulted in. Furthermore, since_the control input trajectory in_SLQ comprises_a time-varying feedforward term and a time-varying linear_state_feedback, its sensitivity_can be calculated as. Moreover, for the initial condition_of this equation we have $\partial_{t_j}\bar{\vx}(z=0)_= \mathbf{0}$ due_to_the_fixed initial state. Based_on Lemma 1, the second order_approximation of the intermediate cost $L(\vx,\vu)$_around the nominal trajectories (which are in_turn a function of $t_j$) can_be written as $$\begin{aligned}
& L(z,\bar{\vx}(s)+\delta\!\vx,\bar{\vu}(s)+\delta\!\vu)_\simeq
q(z)_+
\begin{bmatrix} \vq(z) \\ \vr(z) \end{bmatrix}^\top_
\begin{bmatrix} \partial_{t_j}\vx(z) \\_\partial_{t_j}\vu(z) \end{bmatrix}_(t_j-\bar{t}_j)
\notag \\
&_\hspace{2mm}
+\begin{bmatrix} \delta\!\vx \\ \delta\!\vu \end{bmatrix}^\top\!_
\left(
\begin{bmatrix} \vq(z) \\_\vr(z) \end{bmatrix} +
\begin{bmatrix} \vQ(z) & \vP(z)_\\_\vP(z)^\top & \vR(z)_\end{bmatrix}
\begin{bmatrix}_\partial_{t_j}\vx(z)_\\ \partial_{t_j}\vu(z)_\end{bmatrix}
(t_j-\bar{t}_j) \right)_\notag_\\
& \hspace{2mm}_
|
$10^8$ 1 0.65 SF NS $512\times 512$ 20 0.30 31.5 500
$10^8$ 1 0.70 SF NS $512\times 512$ 21 0.32 31.0 500
$10^8$ 1 0.75 SF NS $512\times 512$ 21 0.34 30.5 500
$10^8$ 1 0.80 SF NS $512\times 512$ 21 0.36 29.6 400
$10^8$ 1 0.85 SF NS $512\times 512$ 21 0.38 29.5 500
$10^8$ 1 0.90 SF NS $512\times 512$ 22 0.40 28.5 400
$10^8$ 1 0.95 SF NS $512\times 512$ 22 0.43 28.8 500
\[table2\]
Ra Pr $\Gamma$ wall BC plate BC $N_x \times N_z$ $N_{BL}$ $\frac{\max(\delta_x,\delta_z)}{\eta}$ ${\text{Nu}}$ $\tau_f$
--------------------- ---- ---------- --------- ---------- ------------------- ---------- ---------------------------------------- --------------- ---------- --
$10^8$ 1 1.00 NS NS $256\times 256$ 10 1.12 26.1 300
$2.15\times10^8$ 1 1.00 NS NS $512\times 512$ 20 0.76 32.6 300
$4.46\times10^8$ 1 1.00 NS NS $512\times 512$ 18 1.00 40.8 400
$10^9$ 1 1.00 NS NS $512\times 512$ 19 1.33 50.5 800
$2.15\times10^9$ 1 | $ 10 ^ 8 $ 1 0.65 SF NS $ 512\times 512 $ 20 0.30 31.5 500
$ 10 ^ 8 $ 1 0.70 SF NS $ 512\times 512 $ 21 0.32 31.0 500
$ 10 ^ 8 $ 1 0.75 SF NS $ 512\times 512 $ 21 0.34 30.5 500
$ 10 ^ 8 $ 1 0.80 SF NS $ 512\times 512 $ 21 0.36 29.6 400
$ 10 ^ 8 $ 1 0.85 SF NS $ 512\times 512 $ 21 0.38 29.5 500
$ 10 ^ 8 $ 1 0.90 SF NS $ 512\times 512 $ 22 0.40 28.5 400
$ 10 ^ 8 $ 1 0.95 SF NS $ 512\times 512 $ 22 0.43 28.8 500
\[table2\ ]
Ra Pr $ \Gamma$ wall BC plate BC $ N_x \times N_z$ $ N_{BL}$ $ \frac{\max(\delta_x,\delta_z)}{\eta}$ $ { \text{Nu}}$ $ \tau_f$
--------------------- ---- ---------- --------- ---------- ------------------- ---------- ---------------------------------------- --------------- ---------- --
$ 10 ^ 8 $ 1 1.00 NS NS $ 256\times 256 $ 10 1.12 26.1 300
$ 2.15\times10 ^ 8 $ 1 1.00 NS NS $ 512\times 512 $ 20 0.76 32.6 300
$ 4.46\times10 ^ 8 $ 1 1.00 NS NS $ 512\times 512 $ 18 1.00 40.8 400
$ 10 ^ 9 $ 1 1.00 NS NS $ 512\times 512 $ 19 1.33 50.5 800
$ 2.15\times10 ^ 9 $ 1 |
$10^8$ 1 0.65 SF NS $512\timss 512$ 20 0.30 31.5 500
$10^8$ 1 0.70 SN US $512\times 512$ 21 0.32 31.0 500
$10^8$ 1 0.75 SF NS $512\times 512$ 21 0.34 30.5 500
$10^8$ 1 0.80 SF NS $512\jimes 512$ 21 0.36 29.6 400
$10^8$ 1 0.85 SF NS $512\times 512$ 21 0.38 29.5 500
$10^8$ 1 0.90 SF NS $512\timex 512$ 22 0.40 28.5 400
$10^8$ 1 0.95 SF NS $512\times 512$ 22 0.43 28.8 500
\[table2\]
Ra [d $\Namma$ wall BC plate BC $J_x \jimes N_z$ $N_{BU}$ $\frcd{\mzx(\delta_x,\delta_z)}{\eta}$ ${\text{No}}$ $\tay_f$
--------------------- ---- ---------- --------- ---------- ------------------- ---------- ---------------------------------------- --------------- ---------- --
$10^8$ 1 1.00 NS NS $256\timzs 256$ 10 1.12 26.1 300
$2.15\times10^8$ 1 1.00 NS NS $512\times 512$ 20 0.76 32.6 300
$4.46\times10^8$ 1 1.00 NS NS $512\tomas 512$ 18 1.00 40.8 400
$10^9$ 1 1.00 NS NS $512\tumes 512$ 19 1.33 50.5 800
$2.15\cnmts10^9$ 1 | $10^8$ 1 0.65 SF NS $512\times 512$ 31.5 $10^8$ 1 SF NS $512\times $10^8$ 0.75 SF NS 512$ 21 0.34 500 $10^8$ 1 0.80 SF NS 512$ 21 0.36 29.6 400 $10^8$ 1 0.85 SF NS $512\times 512$ 21 29.5 500 $10^8$ 1 0.90 SF NS $512\times 512$ 22 0.40 28.5 400 1 SF $512\times 22 0.43 28.8 500 \[table2\] Ra Pr $\Gamma$ wall BC plate BC $N_x \times N_z$ $N_{BL}$ ${\text{Nu}}$ $\tau_f$ --------------------- ---- ---------- --------- ---------- ------------------- ---------------------------------------- --------------- ---------- -- 1 1.00 NS NS $256\times 10 26.1 300 1 NS $512\times 512$ 20 32.6 300 $4.46\times10^8$ 1 1.00 NS NS $512\times 512$ 18 1.00 40.8 400 $10^9$ 1 1.00 NS $512\times 512$ 50.5 800 1 |
$10^8$ 1 0.65 SF NS $512\times 512$ 20 0.30 31.5 500
$10^8$ 1 0.70 SF NS $512\times 512$ 21 0.32 31.0 500
$10^8$ 1 0.75 SF NS $512\timEs 512$ 21 0.34 30.5 500
$10^8$ 1 0.80 SF NS $512\timeS 512$ 21 0.36 29.6 400
$10^8$ 1 0.85 SF NS $512\TimEs 512$ 21 0.38 29.5 500
$10^8$ 1 0.90 Sf Ns $512\timEs 512$ 22 0.40 28.5 400
$10^8$ 1 0.95 SF nS $512\times 512$ 22 0.43 28.8 500
\[table2\]
Ra pR $\GamMa$ wall BC plate BC $N_x \times n_z$ $N_{BL}$ $\FrAC{\max(\DElTa_x,\deLta_z)}{\eta}$ ${\TExT{nU}}$ $\taU_f$
--------------------- ---- ---------- --------- ---------- ------------------- ---------- ---------------------------------------- --------------- ---------- --
$10^8$ 1 1.00 nS nS $256\tImES 256$ 10 1.12 26.1 300
$2.15\tImes10^8$ 1 1.00 Ns NS $512\Times 512$ 20 0.76 32.6 300
$4.46\tiMes10^8$ 1 1.00 NS NS $512\timEs 512$ 18 1.00 40.8 400
$10^9$ 1 1.00 Ns Ns $512\times 512$ 19 1.33 50.5 800
$2.15\times10^9$ 1 |
$10^8$ 1 0.6 5 SF NS $512\ti m es 5 12$ 20 0. 30 3 1. 5 500
$1 0^ 8$ 1 0.70 SF NS $ 512 \ times 51 2$ 21 0. 3 2 31.0 500
$ 1 0^ 8 $ 1 0 .75 SF N S $512\t imes 512$ 21 0.3 4 30.5 500
$ 10^8 $ 1 0. 8 0 SF N S $512 \ti mes512$ 21 0.3 6 29.6 400
$10 ^8 $ 1 0.85 SF N S $512\times 5 12 $ 21 0 .38 29 .5 500
$ 10^ 8$ 1 0.90 SF NS $512\tim e s5 1 2$ 2 2 0 .4 0 28.5 40 0
$10^8$ 1 0.95 SF NS $5 12\times 5 1 2$ 22 0.43 28.8 5 00
\ [table2 \]
Ra Pr $ \G am ma$ wa l l BC p la teBC $N_x\ timesN_z$ $ N_ {B L }$ $\fra c {\ m a x(\d el ta _x,\ del ta _z)}{ \eta } $ ${\tex t{Nu}}$ $\ t au_f $ ----- ------------- -- - ---- --- -- --- -- --- - - ---- --- ------- --------------- - --- --- --- -------- --------- --- ------ --- - ------ ------ -- -- -- --- - - ----- - -- --- -- --- --
$10^ 8$ 1 1.00 N S NS $ 256\ t i me s 2 5 6$ 10 1.12 26.1 300
$2.15\t imes10^8$ 1 1.00 NS NS $ 5 12 \time s 5 12$ 2 0 0. 76 32.6 300
$4.4 6\t imes10^8$ 1 1.00 NS NS $ 5 12\ti mes5 12 $ 1 8 1.00 40.8 400
$10^9$ 1 1. 00 NS N S $5 12 \times 512$ 1 9 1.3 3 50 .5 8 00
$2. 15 \times 10^9$ 1 |
_ _ _ $10^8$__ __ _ 1_ _ 0.65 __ SF NS__ ___ $512\times 512$ _ 20_ _ __ _ 0.30_ _ ___ _ _ 31.5 _ 500_
_ _ _$10^8$ _ _ _1 _ 0.70 _ _ SF __ _NS__ _ _$512\times_512$ __ 21 __ _ __ _ 0.32_ _ _ _ _ 31.0 _ _ _500_ __
_ _$10^8$ _ _ 1 __ 0.75 _ SF _ ___NS _ $512\times 512$ _ _ 21 __ __ ___ 0.34 _ _ _ _ 30.5 ___ 500 _
_ _ $10^8$ _ 1 _ 0.80 SF_ _NS _$512\times 512$ __ 21 ____ _ _ _ _ _ _0.36 __ _ 29.6 _ __400 _
_ $10^8$_ _1_ _ 0.85_ _ _ SF_ _ _ _ NS __ $512\times 512$ _21___ _ _ _ __ _ 0.38 _ _ _ 29.5 _ 500_ _
__ _$10^8$ _ _1_ 0.90 _SF__ __ NS _____$512\times 512$ _ 22 __ _ __ _ _ 0.40 __ ____ _ 28.5 400__
_ __$10^8$ _ 1 _ 0.95_ _ SF NS $512\times 512$ _ _22 _ __ _ _ 0.43_ _ 28.8 _ _ _ 500
\[table2\]
_ Ra_ Pr _$\Gamma$_ wall BC _plate BC $N_x \times_N_z$ _$N_{BL}$ $\frac{\max(\delta_x,\delta_z)}{\eta}$__ ${\text{Nu}}$ _ $\tau_f$_
---------------------_---- ---------- --------- ----------_------------------- ---------- ---------------------------------------- --------------- ---------- --
_ _ $10^8$ _ 1 1.00 NS _ _NS_ $256\times 256$ _ _10 _ ___ _1.12_ _ _ _ _ _ 26.1_ _ 300 __
_$2.15\times10^8$ __ _1 __ _1.00 _ __ _NS __ _ __ NS _ $512\times 512$_ _20 __ _ _ 0.76_ _ _ _ _ __ _ 32.6 _ 300__
$4.46\times10^8$ _ 1 _ 1.00_ _ _NS _ _NS__ $512\times 512$ _ 18 _ _ __ _ _ _ __ 1.00 __ _ _ _ _40.8 ___ 400 _
$10^9$__ _ _1 _ 1.00 NS ___ NS__ _ $512\times 512$ __19 _ 1.33 __ ___ _ _ _ _ 50.5 _ _ 800
$2.15\times10^9$ 1 |
generated via the multiple application of a single entangling operation on product states. The outcomes of this construction together with other examples of MUBs are explicitly presented in the Appendices for low dimensions (see also Ref. [@SMALL_MUB]).
Turning to possible experiments, we note that there are various avenues for implementing quantum states in higher dimensions. For photons these include multiports and spatial-mode superpositions [@RECK; @WEIHS; @MULTIPORT; @OBRIEN; @LANGFORD; @OBRIEN_MULTIPORT] or Hermite-Gauss and Laguerre-Gauss modes, most notably orbital angular momentum states [@OAM_NATURE; @PADGETT_OAM; @PADGETT_REVIEW; @NL_OAM; @NL_GROUP].
We acknowledge discussions with Markus Grassl and Huangjun Zhu. This research is supported by ERC Advanced Grant QIT4QAD, FWF SFB-grant F4007 of the Austrian Science Fund, and the National Research Foundation and Ministry of Education in Singapore.
Appendix {#appendix.unnumbered}
========
We present here explicit examples of complete sets of MUBs and for composite dimensions we emphasize division into entangled and product states. The notation used is explained on the example of a qubit ($d=2$).
d=2
===
The symbol ${\left | j_m \right\rangle}$ denotes the $j$th vector of the $m$th MUB. The standard basis is either denoted with subscript $d$ or has no subscript at all: $$B_2
=
\left(\begin{array}{cc}
1&0\\
0&1
\end{array}\right)
=
\left\{\begin{array}{l}
|0 \rangle \\
|1 \rangle
\end{array}\right\}
=
\left\{\begin{array}{l}
|0_2 \rangle \\
|1_2 \rangle
\end{array}\right\}.$$ Note that when we write a basis as a matrix, we can freely permute columns, since it only changes the order of the vectors in the basis. $$\label{dim2}
B_0 =
\frac{1}{\sqrt{2}}
\left(\begin{array}{ccc}
1&1\\
1&-1
| generated via the multiple application of a single entangling process on merchandise states. The outcomes of this construction in concert with other examples of MUBs are explicitly presented in the Appendices for gloomy dimensions (see besides Ref. [ @SMALL_MUB ]).
turn to potential experiments, we note that there be various avenues for implementing quantum states in higher dimensions. For photon these include multiports and spatial - mode superpositions [ @RECK; @WEIHS; @MULTIPORT; @OBRIEN; @LANGFORD; @OBRIEN_MULTIPORT ] or Hermite - Gauss and Laguerre - Gauss mode, most notably orbital angular momentum states [ @OAM_NATURE; @PADGETT_OAM; @PADGETT_REVIEW; @NL_OAM; @NL_GROUP ].
We admit discussions with Markus Grassl and Huangjun Zhu. This inquiry is supported by ERC Advanced Grant QIT4QAD, FWF SFB - concession F4007 of the Austrian Science Fund, and the National Research Foundation and Ministry of Education in Singapore.
Appendix { # appendix.unnumbered }
= = = = = = = =
We present here explicit examples of complete set of MUBs and for composite dimensions we emphasize division into entangled and product country. The notation used is explained on the example of a qubit ($ d=2 $).
d=2
= = =
The symbol $ { \left | j_m \right\rangle}$ denotes the $ j$th vector of the $ m$th MUB. The standard basis is either denoted with subscript $ d$ or has no subscript at all: $ $ B_2
=
\left(\begin{array}{cc }
1&0\\
0&1
\end{array}\right)
=
\left\{\begin{array}{l }
|0 \rangle \\
|1 \rangle
\end{array}\right\ }
=
\left\{\begin{array}{l }
|0_2 \rangle \\
|1_2 \rangle
\end{array}\right\}.$$ Note that when we write a footing as a matrix, we can freely permute column, since it only transfer the decree of the vectors in the basis. $ $ \label{dim2 }
B_0 =
\frac{1}{\sqrt{2 } }
\left(\begin{array}{ccc }
1&1\\
1&-1 | gejerated via the multiple application of a singne entzngling uperation on product states. Vhe iutcones of this constructiun togethvr with orher wxamples oh MUBs avz expmlcitlv 'resented in thg Appendices for low dimenviund (see also Ref. [@SMALL_MUB]).
Turning to pjssible edperiments, we gote ehat nhtre are various avenues for implejenting quantum statrs in higher dimensions. Fog phltons these includf multiportw anq spatial-mode superposiuimns [@RECK; @WGIHS; @MULTIPORT; @OBRIEN; @LANGFORD; @OCRIEN_KULTIPORT] ir Hegkite-Gauss aid Laglerre-Gauss modes, most notablu orbital angukar monentum states [@OAM_NATUCE; @PADGETT_OAM; @PADGETJ_REVIEW; @NL_MAJ; @NL_GROUP].
We acknoqlwdge giscgssiubs ditg Kadkus Ggasal and Huahgjun Zhu. Tyis research is supkorevc by ERC Advznced Drwnt QIT4QAD, FWF SFB-grant F4007 of the Austridn Acience Fund, and the Narional Research Foundwtion and Ministry of Education in Singapore.
Appendix {#appengix.uniuobexcq}
========
Dw oresent here explicit examples of complete seea pf MUBs and for gomposite dimensioms wr emphasize dixision info entangled and pgoduct ftatew. The notwtiom used is explained on the wxample of a wubit ($d=2$).
d=2
===
The symbol ${\left | j_m \rnght\ramgle}$ cenotes the $j$th vector uf tge $m$th MUB. Hhe standzfd basis is eithdr cetoted wiuf subscript $d$ or ras no suuscri't at alu: $$B_2
=
\lgft(\begig{array}{cc}
1&0\\
0&1
\enf{arran}\sight)
=
\left\{\begin{arraj}{l}
|0 \raugle \\
|1 \rangle
\end{wrray}\right\}
=
\left\{\begin{array}{l}
|0_2 \ranglx \\
|1_2 \rangle
\end{atraf}\richt\}.$$ Note that when we writq a basis as a matrix, we cxn freely iermute cmlumns, sincq it only chathes the ordec of the dectirs un the cxsis. $$\label{dim2}
B_0 =
\frac{1}{\sqrt{2}}
\left(\begin{aeray}{ccc}
1&1\\
1&-1
| generated via the multiple application of a operation product states. outcomes of this of are explicitly presented the Appendices for dimensions (see also Ref. [@SMALL_MUB]). Turning possible experiments, we note that there are various avenues for implementing quantum states higher dimensions. For photons these include multiports and spatial-mode superpositions [@RECK; @WEIHS; @MULTIPORT; @LANGFORD; or and modes, most notably orbital angular momentum states [@OAM_NATURE; @PADGETT_OAM; @PADGETT_REVIEW; @NL_OAM; @NL_GROUP]. We acknowledge discussions with Grassl and Huangjun Zhu. This research is supported ERC Advanced Grant QIT4QAD, SFB-grant F4007 of the Austrian Fund, the National Foundation Ministry Education in Singapore. {#appendix.unnumbered} ======== We present here explicit examples of complete sets of MUBs and for composite dimensions we division into product states. notation is on the example qubit ($d=2$). d=2 === The symbol \right\rangle}$ denotes the $j$th vector of the $m$th The standard is either denoted with subscript $d$ has no subscript at all: $$B_2 = \left(\begin{array}{cc} 0&1 \end{array}\right) = \left\{\begin{array}{l} |0 \rangle \\ |1 \rangle \end{array}\right\} = \left\{\begin{array}{l} |0_2 \rangle \\ \end{array}\right\}.$$ Note that when write a basis a we freely columns, since only changes the order of the vectors in the basis. $$\label{dim2} = \frac{1}{\sqrt{2}} \left(\begin{array}{ccc} 1&1\\ 1&-1 | generated via the multiple apPlication oF a sinGle EntAnGlinG opeRation on producT StatEs. The outcomes of this conStrucTiON togEThEr witH other eXAmPLEs oF MuBS arE eXPlIcitlY prEsented In the AppenDicEs For low dimensIOnS (see also ReF. [@SMaLL_MUB]).
TurninG to PossibLe ExpERimenTs, wE note That thERe are vArious aveNuES for imPLementiNG QuAntuM states in higher diMEnSIons. For photons These iNcLUdE MUltIpoRts and spatIaL-mode SUperposITiONS [@rECk; @wEIHS; @MULTIPORt; @OBRIEN; @LANGfoRD; @oBRIEN_mUlTIpoRT] or HErmitE-GAUss And Laguerre-gausS modes, mosT notabLY orbitaL Angular MomentUm sTatEs [@OAm_nAtUrE; @PaDgeTT_oaM; @pADgeTT_rEVIEW; @NL_oAm; @Nl_GROUp].
We aCKNOWledGe dIscuSsionS with Markus GrAssL and hUanGjun ZHu. ThiS resEaRch is SupporTed by eRc Advanced Grant QiT4QAd, FWF SFB-grAnt f4007 oF thE AUstriAN ScienCe FUnd, And the NAtional rEseArCH fOuNdation and Ministry Of eDUcAtion in SIngapoRE.
APpENdix {#appeNdIx.uNnumBERed}
========
We PresENt Here explIcit exAMpLeS of compLeTe sets Of mUBS anD for cOMposIte dimEnsions wE emphASize division inTO entangled and PRoDUCt STateS. ThE notation usEd is EXplaIned ON tHe eXAmple Of a quBiT ($D=2$).
d=2
===
tHe symbol ${\left | j_m \right\RaNgle}$ deNotes The $j$th vector oF the $m$th MUB. tHE Standard BasiS Is EIther denoted wiTh subScript $d$ or hAS no subscRipt aT all: $$B_2
=
\lefT(\begin{arrAY}{Cc}
1&0\\
0&1
\end{arrAy}\rIghT)
=
\leFt\{\bEGIn{Array}{l}
|0 \rangle \\
|1 \rANGle
\eNd{Array}\riGht\}
=
\Left\{\begIn{aRraY}{l}
|0_2 \rAngLe \\
|1_2 \Rangle
\end{Array}\rigHt\}.$$ noTe ThAt wHen we WRite a basIs As a MaTriX, we caN Freely PermuTe coLuMnS, SinCe it onlY ChANGes tHe OrDer oF thE vEctorS in tHE baSis. $$\labeL{dim2}
B_0 =
\frac{1}{\SqrT{2}}
\Left(\BeGiN{array}{cCc}
1&1\\
1&-1
| generated via the multipl e applicat ion o f a si ng le e ntan gling operatio n onproduct states. The ou tcome so f th i sconst ruction to g e the rwi thot h er exam ple s of MU Bs are exp lic it ly presented in the Appen dic es for low d ime nsions ( see alsoRef . [@S MALL_M U B]).
Turning t op ossibl e experi m e nt s, w e note that there ar e various avenu es for i m pl e m ent ing quantum s ta tes i n higher di m e n sio n s. For photon s these inc l ude multi po rts and sp atial -m o desuperpositi ons[@RECK; @ WEIHS; @MULTIP O RT; @OB RIEN;@LA NGF ORD; @O BR IEN _M U LTI P OR T]o r H ermite-G au ss andLagu e r r e -Gau ssmode s, mo st notably or bit al a n gul ar mo mentu m st at es [@ OAM_NA TURE; @ PADGETT_OAM; @P ADGE TT_REVIEW ; @ NL _OA M; @NL_ G ROUP].
W e a cknowle dge dis c uss io n s wi th Markus Grassl a nd H ua ngjun Zh u. Thi s r es e arch issu ppo rted b y ERC Adv a nc ed Grant QIT4Q A D, F WF SFB- gr ant F4 00 7 o f t he Au s tria n Scie nce Fund , and the National R e search Founda t io n an d Min ist ry of Educa tion in S inga p or e.Appen dix { #a p pe n dix.unnumbered}
=== == ===
W e pre sent here exp licit exam p l e s of com plet e s e ts of MUBs and forcomposited imension s weemphasiz e divisio n into ent ang led an d p r o du ct states. Th e nota ti on used is explai ned on th e e xa mple of a qubit ( $d =2 $) .
d= 2
===
The sym bo l $ {\ lef t | j _ m \rig ht\ra ngle }$ d e not es the$ j$ t h vec to rof t he$m $th M UB.T hestandar d basis i s e i ther d en oted wi th subscript$d $ or has n osub script a t all: $ $B_2
=
\left(\begin{arr a y}{cc}1&0 \\
0& 1
\e nd{array} \ri ght)
=
\l e ft\{\b egin{a rray} {l }
| 0 \rang l e \ \
| 1\rangle
\e n d {ar ray}\ ri ght\ }
=
\le ft\{\begin{array}{ l }
| 0_2 \rangle \ \
| 1_2\ r an gle \e n d{a rr a y}\ r i ght\}.$$ Note t hat when w ew ri te a basis asamatrix, we can free l y permu te column s, sinceit onl y cha nges the o rder ofthe vecto r s int he basi s.$$\lab el {di m2}
B _0 =
\ f rac {1}{\ sqrt{2 }}
\left (\beg in {array}{ ccc}
1&1\\
1&-1
| generated_via the_multiple application of a_single entangling_operation_on product_states._The outcomes of_this construction together_with other examples of_MUBs are explicitly_presented_in the Appendices for low dimensions (see also Ref. [@SMALL_MUB]).
Turning to possible experiments, we_note_that there_are_various_avenues for implementing quantum states_in higher dimensions. For photons_these include_multiports and spatial-mode superpositions [@RECK; @WEIHS; @MULTIPORT; @OBRIEN;_@LANGFORD;_@OBRIEN_MULTIPORT] or Hermite-Gauss_and Laguerre-Gauss modes, most notably orbital angular momentum states_[@OAM_NATURE; @PADGETT_OAM; @PADGETT_REVIEW; @NL_OAM; @NL_GROUP].
We acknowledge_discussions with Markus_Grassl_and_Huangjun Zhu. This research_is supported by ERC Advanced Grant_QIT4QAD, FWF SFB-grant F4007 of the_Austrian Science Fund, and the National Research_Foundation and Ministry of Education in_Singapore.
Appendix {#appendix.unnumbered}
========
We present here explicit_examples of_complete sets of MUBs and_for composite dimensions_we emphasize_division into entangled_and product states. The notation used_is explained on_the example of a qubit ($d=2$).
d=2
===
The_symbol_${\left | j_m_\right\rangle}$_denotes_the $j$th_vector of the_$m$th_MUB. The_standard_basis is either denoted with subscript_$d$_or has no subscript at all: $$B_2
=
\left(\begin{array}{cc}
1&0\\
0&1
\end{array}\right)
=
\left\{\begin{array}{l}
|0_\rangle \\
|1 \rangle
\end{array}\right\}
=
\left\{\begin{array}{l}
|0_2 \rangle_\\
|1_2_\rangle
\end{array}\right\}.$$ Note that when_we write a basis as_a matrix, we can freely permute_columns, since_it only_changes the order of the vectors in the basis. $$\label{dim2}
B_0 =
\frac{1}{\sqrt{2}}
\left(\begin{array}{ccc}
1&1\\
1&-1
|
{\sigma}_i^2))$
In this case, $\boldsymbol{\mu}$ = $GCN_{\mu}$(**X**, **A**) is the matrix of mean vectors $\boldsymbol{\mu}_i$; whereas $log(\boldsymbol{\sigma})$ = $GCN_{\sigma}$(**X**, **A**). We define the two-layer GCN as GCN(**X**, **A**) = $\widetilde{\textbf{A}}\text{ReLU}(\widetilde{\textbf{A}}\textbf{X}\textbf{W}_0)\textbf{W}_1$, with weight matrices $\textbf{W}_i$.
$GCN_{\mu}$(**X** **A**) and $GCN_{\sigma}$(**X** **A**) share the parameters in the first layer defined by $\textbf{W}_0$. ReLU(n) = max(0,n) and $\widetilde{\textbf{A}}= \textbf{D}^{-\frac{1}{2}}\textbf{A}\textbf{D}^{-\frac{1}{2}}$ is the symmetrically normalized adjacency matrix.
**Generative Model**: We view the generative model as the reconstruction of the adjacency matrix from the latent variables obtained in the inference model outlined above. We use the inner product between the latent variable representations for two nodes to predict the presence of an association
$$p(\textbf{A} \mid \textbf{Z}) = \prod_{i=1}^N \prod_{j=1}^N p(A_{ij} \mid \textbf{z}_i \text{, }\textbf{z}_j )$$
$p(A_{ij} =1 \mid \textbf{z}_i \text{, }\textbf{z}_j)$ = $\sigma(\textbf{z}_i^\top\textbf{z}_j)$
We define $A_{ij}$ to be the elements of **A** and $\sigma$ to be the sigmoid function.
**Learning**: We optimize the variational lower bound $\mathcal{L}$ w.r.t. to the parameters $\textbf{W}_i$
$$\mathcal{L} = \mathbb{E}_{q(\textbf{Z} \mid \textbf{X}, \textbf{A})} [\text{log} p(\textbf{A} \mid \textbf{Z})] - \text{KL}[q(\textbf{Z} \ | { \sigma}_i^2))$
In this case, $ \boldsymbol{\mu}$ = $ GCN_{\mu}$(**X * *, * * A * *) is the matrix of mean vectors $ \boldsymbol{\mu}_i$; whereas $ log(\boldsymbol{\sigma})$ = $ GCN_{\sigma}$(**X * *, * * deoxyadenosine monophosphate * *). We specify the two - layer GCN as GCN(**X * *, * * A * *) = $ \widetilde{\textbf{A}}\text{ReLU}(\widetilde{\textbf{A}}\textbf{X}\textbf{W}_0)\textbf{W}_1 $, with weight matrix $ \textbf{W}_i$.
$ GCN_{\mu}$(**X * * * * A * *) and $ GCN_{\sigma}$(**X * * * * A * *) partake the parameters in the first layer specify by $ \textbf{W}_0$. ReLU(n) = max(0,n) and $ \widetilde{\textbf{A}}= \textbf{D}^{-\frac{1}{2}}\textbf{A}\textbf{D}^{-\frac{1}{2}}$ is the symmetrically normalized adjacency matrix.
* * Generative Model * *: We view the generative mannequin as the reconstruction of the adjacency matrix from the latent variables obtained in the inference mannequin delineate above. We use the inner product between the latent variable representations for two node to predict the presence of an association
$ $ p(\textbf{A } \mid \textbf{Z }) = \prod_{i=1}^N \prod_{j=1}^N p(A_{ij } \mid \textbf{z}_i \text {, } \textbf{z}_j) $ $
$ p(A_{ij } = 1 \mid \textbf{z}_i \text {, } \textbf{z}_j)$ = $ \sigma(\textbf{z}_i^\top\textbf{z}_j)$
We specify $ A_{ij}$ to be the elements of * * A * * and $ \sigma$ to be the sigmoid function.
* * Learning * *: We optimize the variational lower bound $ \mathcal{L}$ w.r.t. to the parameter $ \textbf{W}_i$
$ $ \mathcal{L } = \mathbb{E}_{q(\textbf{Z } \mid \textbf{X }, \textbf{A }) } [ \text{log } p(\textbf{A } \mid \textbf{Z }) ] - \text{KL}[q(\textbf{Z } \ | {\sigla}_i^2))$
In this case, $\boldsymbul{\mu}$ = $GCN_{\mu}$(**X**, **A**) nw the katrix of mean vectors $\boldsymbol{\mu}_i$; wherees $lig(\bolesymbol{\sigma})$ = $GCN_{\sigma}$(**B**, **A**). We devine the two-oqyer GCN as GCN(**X**, **A**) = $\widefllde{\tzxvbf{A}}\text{ReLU}(\widgtilde{\textbf{D}}\textbf{X}\textbf{F}_0)\tdxcbf{W}_1$, with weight matrices $\textbf{W}_i$.
$GCG_{\mu}$(**X** **A**) snf $GCN_{\sigma}$(**X** **A**) fhart tre pzganeters in the first layer defjned by $\textbf{W}_0$. ReLU(m) = max(0,n) and $\widetilde{\textbv{A}}= \tfxtbf{D}^{-\frac{1}{2}}\textbf{A}\tfxtbf{D}^{-\frac{1}{2}}$ us trw symmetricauly normalpved adjacendy matrix.
**Generative Model**: We vidw thz generativg jofgl as the reronstrlction of the adjacenwy matrox from the labent tariqbles obtained in the inference model ootlined abmvz. We use the inner pridyct bgtweet thd lagenu verizble rfprxsentations for two noees to predict the krefvmce of an asaociatyog
$$p(\textbf{A} \mid \textbf{Z}) = \prod_{i=1}^N \prod_{j=1}^N p(D_{ij} \mid \textbf{z}_i \text{, }\textvf{z}_j )$$
$p(A_{ij} =1 \mid \textbf{e}_i \text{, }\tevtbf{z}_j)$ = $\sigma(\textbf{z}_i^\top\textbf{z}_j)$
We define $A_{ij}$ to te thx dlenekts ud **W** and $\sigma$ to be the sigmoid function.
**Learnind**: Wt oitimize the variabional lower bound $\mwtnsal{L}$ w.r.t. to tfe parcjefers $\textbf{W}_i$
$$\mathcwl{L} = \majhbb{E}_{q(\rextbf{Z} \myd \trxtbf{X}, \textbf{A})} [\text{log} p(\texrbf{A} \mid \texnbf{Z})] - \text{KL}[q(\textbf{Z} \ | {\sigma}_i^2))$ In this case, $\boldsymbol{\mu}$ = $GCN_{\mu}$(**X**, the of mean $\boldsymbol{\mu}_i$; whereas $log(\boldsymbol{\sigma})$ the GCN as GCN(**X**, = $\widetilde{\textbf{A}}\text{ReLU}(\widetilde{\textbf{A}}\textbf{X}\textbf{W}_0)\textbf{W}_1$, with matrices $\textbf{W}_i$. $GCN_{\mu}$(**X** **A**) and $GCN_{\sigma}$(**X** share the parameters in the first layer defined by $\textbf{W}_0$. ReLU(n) = max(0,n) $\widetilde{\textbf{A}}= \textbf{D}^{-\frac{1}{2}}\textbf{A}\textbf{D}^{-\frac{1}{2}}$ is the symmetrically normalized adjacency matrix. **Generative Model**: We view the model the of adjacency matrix from the latent variables obtained in the inference model outlined above. We use the product between the latent variable representations for two to predict the presence an association $$p(\textbf{A} \mid \textbf{Z}) \prod_{i=1}^N p(A_{ij} \mid \text{, )$$ =1 \mid \textbf{z}_i }\textbf{z}_j)$ = $\sigma(\textbf{z}_i^\top\textbf{z}_j)$ We define $A_{ij}$ to be the elements of **A** and $\sigma$ to be the function. **Learning**: the variational bound w.r.t. the parameters $\textbf{W}_i$ \mathbb{E}_{q(\textbf{Z} \mid \textbf{X}, \textbf{A})} [\text{log} p(\textbf{A} \text{KL}[q(\textbf{Z} \ | {\sigma}_i^2))$
In this case, $\boldsymboL{\mu}$ = $GCN_{\mu}$(**X**, **A**) Is the MatRix Of Mean VectOrs $\boldsymbol{\mU}_I$; wheReas $log(\boldsymbol{\sigma})$ = $gCN_{\siGmA}$(**x**, **A**). We DEfIne thE two-layER Gcn As GcN(**x**, **A**) = $\WidEtILdE{\textBf{A}}\Text{ReLu}(\widetilde{\TexTbF{A}}\textbf{X}\texTBf{w}_0)\textbf{W}_1$, wiTh wEight matriceS $\teXtbf{W}_i$.
$gCn_{\mu}$(**x** **a**) and $GcN_{\sIgma}$(**X** **a**) share THe paraMeters in tHe FIrst laYEr definED By $\TextBf{W}_0$. ReLU(n) = max(0,n) and $\wiDEtILde{\textbf{A}}= \textBf{D}^{-\fraC{1}{2}}\tEXtBF{a}\teXtbF{D}^{-\frac{1}{2}}$ is thE sYmmetRIcally nORmALIZed ADjacency matriX.
**Generative mOdeL**: We vieW tHe gENeratiVe modEl AS thE reconstrucTion Of the adjaCency mATrix froM The lateNt variAblEs oBtaiNEd In The InFEreNCe ModEL ouTlined abOvE. WE use tHe inNER PRoduCt bEtweEn the Latent variablE rePresENtaTions For twO nodEs To preDict thE presEnCe of an associatiOn
$$p(\tExtbf{A} \mid \TexTbF{Z}) = \pRoD_{i=1}^N \prOD_{j=1}^N p(A_{iJ} \miD \teXtbf{z}_i \tExt{, }\textBF{z}_j )$$
$P(A_{IJ} =1 \MId \Textbf{z}_i \text{, }\textbf{Z}_j)$ = $\SIGmA(\textbf{z}_I^\top\teXTbF{z}_J)$
we define $a_{iJ}$ to Be thE ELemenTs of **a** AnD $\sigma$ to Be the sIGmOiD functiOn.
**learniNg**: we oPtiMize tHE varIationAl lower bOund $\mAThcal{L}$ w.r.t. to the PArameters $\textBF{W}_I$
$$\MAtHCal{L} = \MatHbb{E}_{q(\textbf{z} \mid \TExtbF{X}, \teXTbF{A})} [\tEXt{log} P(\textBf{a} \MiD \Textbf{Z})] - \text{KL}[q(\textbF{Z} \ | {\sigma}_i^2))$
In this c ase, $\bol dsymb ol{ \mu }$ = $ GCN_ {\mu}$(**X**,* *A** ) is the matrix of mea n vec to r s $\ b ol dsymb ol{\mu} _ i$ ; whe re as $l og ( \b oldsy mbo l{\sigm a})$ = $GC N_{ \s igma}$(**X** , * *A**). Wedef ine the two- lay er GCN a s G C N(**X **, **A* *) = $ \ wideti lde{\text bf { A}}\te x t{ReLU} ( \ wi deti lde{\textbf{A}}\t e xt b f{X}\textbf{W} _0)\te xt b f{ W } _1$ , w ith weight m atric e s $\tex t bf { W } _i$ .
$GCN_{\mu}$ (**X** **A* * ) a nd $GC N_ {\s i gma}$( **X** * * A** ) share the par ameters i n thef irst la y er defi ned by $\ tex tbf{ W }_ 0$ . R eL U (n) =max ( 0,n ) and $\ wi de tilde {\te x t b f {A}} = \ text bf{D} ^{-\frac{1}{2 }}\ text b f{A }\tex tbf{D }^{- \f rac{1 }{2}}$ is t he symmetricallynorm alized ad jac en cyma trix.
**Gen era tiv e Model **: Wev iew t h e ge nerative model asth e re construc tion o f t he adjacenc ymat rixf r om th e la t en t variab les ob t ai ne d in th einfere nc e m ode l out l ined above . We use thei nner product b e tween the lat e nt v ar i able re presentatio ns f o r tw o no d es to predi ct th ep re s ence of an associat io n
$$p (\tex tbf{A} \mid \ textbf{Z}) = \prod_{i =1}^ N \ p rod_{j=1}^N p (A_{i j} \mid \t e xtbf{z}_ i \te xt{, }\t extbf{z}_ j )$$
$p( A_{ ij} =1 \m i d \ textbf{z}_i \ t e xt{, } \textbf {z} _j)$ =$\s igm a(\ tex tb f{z}_i^\t op\textb f{ z} _j )$
W e def i ne $A_{i j} $ t obethe e l ements of * *A** a nd $\s igma$ t o b e thesi gm oidfun ct ion.
**L e arn ing**:We optimi zet he v ar ia tionallower bound $ \m athcal{L}$ w .r. t. tot h e parame ters $\textbf{W}_i$
$$ \ mathcal {L} = \m athb b{E}_{q(\ tex tbf{Z} \m i d \tex tbf{X} , \te xt bf{ A } )} [\ t e xt {lo g} p(\textbf { A } \ mid \ te xtbf {Z})] - \text{KL}[q(\text b f{Z } \ | {\sigma}_i^2))$
In this_case, $\boldsymbol{\mu}$_= $GCN_{\mu}$(**X**, **A**) is_the matrix_of_mean vectors_$\boldsymbol{\mu}_i$;_whereas $log(\boldsymbol{\sigma})$ =_$GCN_{\sigma}$(**X**, **A**). We_define the two-layer GCN_as GCN(**X**, **A**)_=_$\widetilde{\textbf{A}}\text{ReLU}(\widetilde{\textbf{A}}\textbf{X}\textbf{W}_0)\textbf{W}_1$, with weight matrices $\textbf{W}_i$.
$GCN_{\mu}$(**X** **A**) and $GCN_{\sigma}$(**X** **A**) share the parameters in the_first_layer defined_by_$\textbf{W}_0$._ReLU(n) = max(0,n) and $\widetilde{\textbf{A}}=_\textbf{D}^{-\frac{1}{2}}\textbf{A}\textbf{D}^{-\frac{1}{2}}$ is the symmetrically normalized_adjacency matrix.
**Generative_Model**: We view the generative model as the_reconstruction_of the adjacency_matrix from the latent variables obtained in the inference_model outlined above. We use the_inner product between_the_latent_variable representations for two_nodes to predict the presence of_an association
$$p(\textbf{A} \mid \textbf{Z}) = \prod_{i=1}^N_\prod_{j=1}^N p(A_{ij} \mid \textbf{z}_i \text{, }\textbf{z}_j_)$$
$p(A_{ij} =1 \mid \textbf{z}_i \text{, }\textbf{z}_j)$_= $\sigma(\textbf{z}_i^\top\textbf{z}_j)$
We define $A_{ij}$ to_be the_elements of **A** and $\sigma$_to be the_sigmoid function.
**Learning**:_We optimize the_variational lower bound $\mathcal{L}$ w.r.t. to_the parameters $\textbf{W}_i$
$$\mathcal{L}_= \mathbb{E}_{q(\textbf{Z} \mid \textbf{X}, \textbf{A})} [\text{log}_p(\textbf{A}_\mid \textbf{Z})] -_\text{KL}[q(\textbf{Z}_\ |
xi_1\!)}{(1-y_2)^2}\!\right)
f_\ell \!\left(\!\frac{1\!+\!h'^2\!+\!2h'{\mbox}{Re}(\xi_2\!)}{(1-y_2)^2}\right)}{(\xi_1-\tau\xi_2)^2}
\!{\mbox{d}}\xi_1\!{\mbox{d}}\xi_2,\\
&& - \frac{(\beta_x y_1 \!+ \!\beta_y y_2 \!)( \!1 \!- \!y_2 \!)^2}{4 \pi^2 \!h'^2}
\!\oint_{|\xi_1\!| \!=\!1} \! \frac{f_j \!\left( \!\frac{ \!1 \!+ \!h'^2 \!+ \!2h'{\mbox}{Re}(\xi_1)}{(1 \!- \!y_2)^2}\right)}{(\xi_1 \!+ \!\frac{y_2}{h'})^2} \!{\mbox{d}}\xi_1
\!\oint_{|\xi_2\!| \!=\!1} \! \frac{f_j \!\left( \!\frac{ \!1 \!+ \!h'^2 \!+ \!2h'{\mbox}{Re}(\xi_2)}{(1 \!- \!y_2)^2}\right)}{(\xi_2 \!+ \!\frac{y_2}{h'})^2} \!{\mbox{d}}\xi_2\nonumber\end{aligned}$$ where $ j,\ell \in \{1, \cdots,
s\}$, “${\mbox}{Re}$” represents the real part of $\xi$ and $\tau \downarrow
1$ means that " $\tau$ approaches 1 from above’.
Calculation of $\mu (f)$ in (\[testE\]).\[A3\]
-----------------------------------------------
For the function $f(x)=x$, the computation of $\mu (f)$ is divided into three parts. Still use the denotation $h=\sqrt{\gamma_1+\gamma_2-\gamma_1\gamma_2}$, then the first part is
$$\begin{aligned}
I_1&=&\lim\limits_{\tau \downarrow
1}\frac{\kappa-1}{4\pi i}\oint_{|\xi|=1}
f | xi_1\!)}{(1 - y_2)^2}\!\right)
f_\ell \!\left(\!\frac{1\!+\!h'^2\!+\!2h'{\mbox}{Re}(\xi_2\!)}{(1 - y_2)^2}\right)}{(\xi_1-\tau\xi_2)^2 }
\!{\mbox{d}}\xi_1\!{\mbox{d}}\xi_2,\\
& & - \frac{(\beta_x y_1 \!+ \!\beta_y y_2 \ !) (\!1 \!- \!y_2 \!)^2}{4 \pi^2 \!h'^2 }
\!\oint_{|\xi_1\!| \!=\!1 } \! \frac{f_j \!\left (\!\frac { \!1 \!+ \!h'^2 \!+ \!2h'{\mbox}{Re}(\xi_1)}{(1 \!- \!y_2)^2}\right)}{(\xi_1 \!+ \!\frac{y_2}{h'})^2 } \!{\mbox{d}}\xi_1
\!\oint_{|\xi_2\!| \!=\!1 } \! \frac{f_j \!\left (\!\frac { \!1 \!+ \!h'^2 \!+ \!2h'{\mbox}{Re}(\xi_2)}{(1 \!- \!y_2)^2}\right)}{(\xi_2 \!+ \!\frac{y_2}{h'})^2 } \!{\mbox{d}}\xi_2\nonumber\end{aligned}$$ where $ j,\ell \in \{1, \cdots,
s\}$, “ $ { \mbox}{Re}$ ” represents the real part of $ \xi$ and $ \tau \downarrow
1 $ means that " $ \tau$ access 1 from above ’.
Calculation of $ \mu (f)$ in (\[testE\]).\[A3\ ]
-----------------------------------------------
For the affair $ f(x)=x$, the computation of $ \mu (f)$ is divided into three parts. even use the denotation $ h=\sqrt{\gamma_1+\gamma_2-\gamma_1\gamma_2}$, then the beginning share is
$ $ \begin{aligned }
I_1&=&\lim\limits_{\tau \downarrow
1}\frac{\kappa-1}{4\pi i}\oint_{|\xi|=1 }
f | xi_1\!)}{(1-y_2)^2}\!\gight)
f_\ell \!\left(\!\frac{1\!+\!h'^2\!+\!2h'{\mbox}{Ve}(\xi_2\!)}{(1-y_2)^2}\right)}{(\xi_1-\tau\xn_2)^2}
\!{\nbox{d}}\xm_1\!{\mbox{d}}\xj_2,\\
&& - \frac{(\bdta_x y_1 \!+ \!\beta_y y_2 \!)( \!1 \!- \!y_2 \!)^2}{4 \pi^2 \!h'^2}
\!\ount_{|\xi_1\!| \!=\!1} \! \frac{f_j \!\left( \!\frac{ \!1 \!+ \!h'^2 \!+ \!2h'{\mboq}{Re}(\xi_1)}{(1 \!- \!y_2)^2}\rught)}{(\ei_1 \!+ \!\frac{y_2}{h'})^2} \!{\mbox{v}}\si_1
\!\oint_{|\wn_2\!| \!=\!1} \! \fdwc{f_j \!\left( \!\frac{ \!1 \!+ \!h'^2 \!+ \!2h'{\mbox}{Re}(\xh_2)}{(1 \!- \!y_2)^2}\right)}{(\xi_2 \!+ \!\fsaz{y_2}{k'})^2} \!{\mbox{d}}\xi_2\nonumber\end{aligned}$$ where $ j,\qll \in \{1, \cfots,
s\}$, “${\mbox}{Re}$” rgprestntf ths real part of $\xi$ and $\tau \downarros
1$ means that " $\tau$ aplroaches 1 from above’.
Calcupatiln of $\mu (f)$ in (\[testF\]).\[A3\]
-----------------------------------------------
For the fohctyin $f(x)=x$, the cumputation of $\mu (f)$ is sivided into three parts. Still jse tke denotatiin $h=\sett{\gamma_1+\gamma_2-\jamma_1\gwmma_2}$, then thc first part ix
$$\begin{aligned}
I_1&=&\kim\nimuts_{\tau \downarrow
1}\frac{\keppa-1}{4\pi i}\oint_{|\xi|=1}
f | xi_1\!)}{(1-y_2)^2}\!\right) f_\ell \!\left(\!\frac{1\!+\!h'^2\!+\!2h'{\mbox}{Re}(\xi_2\!)}{(1-y_2)^2}\right)}{(\xi_1-\tau\xi_2)^2} \!{\mbox{d}}\xi_1\!{\mbox{d}}\xi_2,\\ && - \frac{(\beta_x \!\beta_y \!)( \!1 \!y_2 \!)^2}{4 \pi^2 \!\left( \!1 \!+ \!h'^2 \!2h'{\mbox}{Re}(\xi_1)}{(1 \!- \!y_2)^2}\right)}{(\xi_1 \!\frac{y_2}{h'})^2} \!{\mbox{d}}\xi_1 \!\oint_{|\xi_2\!| \!=\!1} \! \frac{f_j \!\frac{ \!1 \!+ \!h'^2 \!+ \!2h'{\mbox}{Re}(\xi_2)}{(1 \!- \!y_2)^2}\right)}{(\xi_2 \!+ \!\frac{y_2}{h'})^2} \!{\mbox{d}}\xi_2\nonumber\end{aligned}$$ where $ \in \{1, \cdots, s\}$, “${\mbox}{Re}$” represents the real part of $\xi$ and $\tau 1$ that $\tau$ 1 from above’. Calculation of $\mu (f)$ in (\[testE\]).\[A3\] ----------------------------------------------- For the function $f(x)=x$, the computation $\mu (f)$ is divided into three parts. Still the denotation $h=\sqrt{\gamma_1+\gamma_2-\gamma_1\gamma_2}$, then first part is $$\begin{aligned} I_1&=&\lim\limits_{\tau 1}\frac{\kappa-1}{4\pi f | xi_1\!)}{(1-y_2)^2}\!\right)
f_\ell \!\left(\!\frac{1\!+\!h'^2\!+\!2h'{\mboX}{Re}(\xi_2\!)}{(1-y_2)^2}\righT)}{(\xi_1-\taU\xi_2)^2}
\!{\MboX{d}}\Xi_1\!{\mbOx{d}}\xI_2,\\
&& - \frac{(\beta_x y_1 \!+ \!\betA_Y y_2 \!)( \!1 \!- \!y_2 \!)^2}{4 \pI^2 \!h'^2}
\!\oint_{|\xi_1\!| \!=\!1} \! \frac{f_j \!\left( \!\frac{ \!1 \!+ \!H'^2 \!+ \!2h'{\mboX}{RE}(\Xi_1)}{(1 \!- \!y_2)^2}\rIGhT)}{(\xi_1 \!+ \!\frAc{y_2}{h'})^2} \!{\mboX{D}}\xI_1
\!\OInt_{|\Xi_2\!| \!=\!1} \! \FrAc{f_J \!\lEFt( \!\Frac{ \!1 \!+ \!h'^2 \!+ \!2H'{\mbOx}{Re}(\xi_2)}{(1 \!- \!y_2)^2}\Right)}{(\xi_2 \!+ \!\fraC{y_2}{h'})^2} \!{\MbOx{d}}\xi_2\nonumbeR\EnD{aligned}$$ whEre $ J,\ell \in \{1, \cdots,
s\}$, “${\MboX}{Re}$” repReSenTS the rEal Part oF $\xi$ and $\TAu \downArrow
1$ meanS tHAt " $\tau$ aPProacheS 1 FRoM aboVe’.
Calculation of $\mu (F)$ In (\[TEstE\]).\[A3\]
-----------------------------------------------
For the funCtion $f(X)=x$, THe COMpuTatIon of $\mu (f)$ is DiVided INto threE PaRTS. stiLL use the denotaTion $h=\sqrt{\gaMMa_1+\gAmma_2-\gaMmA_1\gaMMa_2}$, then The fiRsT ParT is
$$\begin{aliGned}
i_1&=&\lim\limitS_{\tau \doWNarrow
1}\fRAc{\kappa-1}{4\Pi i}\oinT_{|\xi|=1}
F | xi_1\!)}{(1-y_2)^2}\!\righ t)
f_\ell\!\le ft( \!\ fr ac{1 \!+\ !h'^2\!+\!2h'{ \ mbox }{Re}(\xi_2\!)}{(1-y_2 )^2}\ ri g ht)} { (\ xi_1- \tau\xi _ 2) ^ 2 }
\ !{ \m box {d } }\ xi_1\ !{\ mbox{d} }\xi_2,\\&&-\frac{(\beta _ xy_1 \!+ \! \be ta_y y_2 \! )(\!1 \! -\!y _ 2 \!) ^2} {4 \ pi^2 \ ! h'^2} \!\oint_ {| \ xi_1\! | \!=\!1 } \ ! \f rac{f_j \!\left( \! \ frac{ \!1 \!+\!h'^2 \ ! +\ ! 2h' {\m box}{Re}(\ xi _1)}{ ( 1 \!- \ ! y_ 2 ) ^ 2}\ r ight)}{(\xi_1 \!+ \!\fra c {y_ 2}{h'} )^ 2}\!{\mb ox{d} }\ x i_1
\!\oint_{ |\xi _2\!| \!= \!1} \ ! \frac{ f _j \!\ left(\!\ fra c{ \ ! 1\! + \ !h ' ^2\ !+ \! 2 h'{ \mbox}{R e} (\ xi_2) }{(1 \ ! - \!y _2) ^2}\ right )}{(\xi_2 \!+ \! \fra c {y_ 2}{h' })^2} \!{ \m box{d }}\xi_ 2\non um ber\end{aligned }$$where $ j ,\e ll \i n\{1,\ cdots,
s\ }$, “${\mb ox}{Re} $ ” r ep r e s en ts the real part o f$ \ xi $ and $\ tau \d o wn ar r ow
1$ me an s t hat" $\tau $ ap p ro aches 1from a b ov e’ .
Cal cu lation o f $ \mu (f)$ in ( \[test E\]).\[A 3\]
- - -------------- - ------------- - -- - - -- - ---- --- --
For the fun c tion $f( x )= x$, the c omput at i on of $\mu (f)$ is div id ed int o thr ee parts. Sti ll use the d e notation $h= \ sq r t{\gamma_1+\ga mma_2 -\gamma_1\ g amma_2}$ , the n the fi rst parti s
$$\beg in{ ali gne d}I _ 1& =&\lim\limits _ { \tau \ downarr ow1}\frac {\k app a-1 }{4 \p i i}\oint _{|\xi|= 1}
f | xi_1\!)}{(1-y_2)^2}\!\right)
f_\ell \!\left(\!\frac{1\!+\!h'^2\!+\!2h'{\mbox}{Re}(\xi_2\!)}{(1-y_2)^2}\right)}{(\xi_1-\tau\xi_2)^2}
\!{\mbox{d}}\xi_1\!{\mbox{d}}\xi_2,\\
&&_- \frac{(\beta_x_y_1 \!+ \!\beta_y _y_2 \!)(_\!1_\!- \!y_2_\!)^2}{4_ \pi^2 \!h'^2}
_\!\oint_{|\xi_1\!| \!=\!1} _\! \frac{f_j \!\left(_\!\frac{ \!1 \!+_\!h'^2_\!+ \!2h'{\mbox}{Re}(\xi_1)}{(1 \!- \!y_2)^2}\right)}{(\xi_1 \!+ \!\frac{y_2}{h'})^2} \!{\mbox{d}}\xi_1
\!\oint_{|\xi_2\!| \!=\!1} \! \frac{f_j \!\left(_\!\frac{_\!1 \!+_\!h'^2_\!+_\!2h'{\mbox}{Re}(\xi_2)}{(1 \!- \!y_2)^2}\right)}{(\xi_2 \!+ \!\frac{y_2}{h'})^2}_\!{\mbox{d}}\xi_2\nonumber\end{aligned}$$ where $ j,\ell \in_\{1, \cdots,
s\}$,_“${\mbox}{Re}$” represents the real part of $\xi$ and_$\tau_\downarrow
1$ means that_" $\tau$ approaches 1 from above’.
Calculation of $\mu_(f)$ in (\[testE\]).\[A3\]
-----------------------------------------------
For the function $f(x)=x$,_the computation of_$\mu_(f)$_is divided into three_parts. Still use the denotation $h=\sqrt{\gamma_1+\gamma_2-\gamma_1\gamma_2}$,_then the first part is
$$\begin{aligned}
I_1&=&\lim\limits_{\tau \downarrow
1}\frac{\kappa-1}{4\pi_i}\oint_{|\xi|=1}
f |
denk2014exciton] For the monolayer MoS$_2$ covered a-plane ($11\overline{2}0$) substrate, the RD signals can be described by the following equation: $$\label{Eq.1}
\frac{\Delta r}{r}=\frac{1}{2}\frac{r_{[1\overline{1}00]}-r_{[0001]}}{r_{[1\overline{1}00]}+r_{[0001]}}$$ where $r_{[1\overline{1}00]}$ and $r_{[0001]}$ denote the reflectance of the light polarized along the $[1\overline{1}00]$ and the \[0001\] directions of the a-plane sapphire substrate, respectively.
![(a) The RD spectrum taken from the monolayer MoS$_2$ on Al$_2$O$_3$($11\overline{2}0$), (b) The absorption spectrum (b) and its first derivative (c) measured from monolayer MoS$_2$ on Al$_2$O$_3$(0001).[]{data-label="figure2"}](Fig2.pdf){width="8cm"}
After the systematic characterization using conventional techniques,[@supplemental] RDS measurement was then applied to investigate the optical anisotropy within the plane of the MoS$_2$ monolayer.The real part of the RD spectra measured from the bare Al2O3(11$\overline{2}$0) surface and the one covered by monolayer MoS$_2$ are plotted in Fig. 2(a), respectively. The bare Al2O3(11$\overline{2}$0) surface shows an optical anisotropy with an almost constant value which can be directly attributed to the in-plane birefringence of the a-plane sapphire substrate. Actually, the corresponding in-plane axis, namely \[1$\overline{1}$00\] and \[0001\] axis are parallel to the ordinary and extraordinary directions of sapphire, respectively. The result reveals thus the dielectric anisotropy $\Delta \epsilon_{ext}=\epsilon_o-\epsilon_e$ in a-plane sapphire substrate. Furthermore, additional optical anisotropy shows up from the a-plane sapphire substrate covered by monolayer MoS$_2$. It worth mentioning that, above the transparent sapphire substrate, the real part of the RD signal is predominantly associated with the anisotropy of the absorption of the monolayer MoS$_2$. For comparison, the absorption spectrum of | denk2014exciton ] For the monolayer MoS$_2 $ covered a - plane ($ 11\overline{2}0 $) substrate, the RD signals can be described by the following equality: $ $ \label{Eq.1 }
\frac{\Delta r}{r}=\frac{1}{2}\frac{r_{[1\overline{1}00]}-r_{[0001]}}{r_{[1\overline{1}00]}+r_{[0001]}}$$ where $ r_{[1\overline{1}00]}$ and $ r_{[0001]}$ announce the coefficient of reflection of the light polarized along the $ [ 1\overline{1}00]$ and the \[0001\ ] directions of the a - airplane sapphire substrate, respectively.
! [ (a) The RD spectrum taken from the monolayer MoS$_2 $ on Al$_2$O$_3$($11\overline{2}0 $), (b) The absorption spectrum (barn) and its first derivative (coke) measured from monolayer MoS$_2 $ on Al$_2$O$_3$(0001).[]{data - label="figure2"}](Fig2.pdf){width="8 cm " }
After the taxonomic characterization using conventional techniques,[@supplemental ] RDS measurement was then applied to investigate the ocular anisotropy within the plane of the MoS$_2 $ monolayer. The real part of the RD spectra measured from the unsheathed Al2O3(11$\overline{2}$0) surface and the one covered by monolayer MoS$_2 $ are plotted in Fig. 2(a), respectively. The plain Al2O3(11$\overline{2}$0) surface shows an optical anisotropy with an almost changeless value which can be directly attributed to the in - plane birefringence of the a - plane azure substrate. Actually, the corresponding in - plane axis, namely \[1$\overline{1}$00\ ] and \[0001\ ] axis are parallel to the ordinary and extraordinary directions of azure, respectively. The consequence reveals thus the dielectric anisotropy $ \Delta \epsilon_{ext}=\epsilon_o-\epsilon_e$ in a - plane sapphire substrate. Furthermore, additional optical anisotropy shows up from the a - plane azure substrate cover by monolayer MoS$_2$. It worth mention that, above the transparent sapphire substrate, the real part of the RD signal is predominantly consociate with the anisotropy of the absorption of the monolayer MoS$_2$. For comparison, the absorption spectrum of | denn2014exciton] For the monolaytr MoS$_2$ covered a-plane ($11\ovxrline{2}0$) aubstratd, the RD signals can be desccibee by uke following equatiov: $$\label{Eq.1}
\vrac{\Deltq r}{r}=\hrac{1}{2}\frac{r_{[1\overlinx{1}00]}-d_{[0001]}}{r_{[1\overllue{1}00]}+r_{[0001]}}$$ wgcre $r_{[1\mterline{1}00]}$ and $r_{[0001]}$ dgnote the rexlectance of tve lnght polarized along the $[1\overline{1}00]$ anq the \[0001\] cigections of thg a-plsge szipmire substrate, respectively.
![(a) The RD spebtrum taken from yhe monolayer MoS$_2$ on Al$_2$O$_3$($11\ovfrlije{2}0$), (b) The absorptioj spectrum (v) anq its first ddrivative (b) measured rrom monolayer MoS$_2$ on Al$_2$O$_3$(0001).[]{data-lacel="fiyure2"}](Fig2.pdf){wudrh="8cl"}
Dfter the sbstemanic characterldation gsing cpnventional teghniqnes,[@sypplemental] RDS measucement was then applyed to ineeatigate the opticql anismtro[y wkrhiv tge pmane ov tie MoS$_2$ monomayer.The reql part of the RD skecegs measured fdom thq fare Al2O3(11$\overline{2}$0) surface and the one coneres by monolayer MoS$_2$ are plotted in Fig. 2(a), respeftively. Tre bare Al2O3(11$\overline{2}$0) surface shows an optical anismtropb ditk an aunodt constant value which can be directly attrifhttd no the in-plane bivefringence of the a-olsge sapphire sobstratz. Adtually, the corresoonding in-plqne axis, gameky \[1$\overline{1}$00\] and \[0001\] axis are pqrallel to tke irdinary and extralrdinary dixectioms of sapphire, respectively. Che reault reveald thus ths dielectric anisutrppf $\Delta \tosilon_{ext}=\epsilon_o-\qpsilon_e$ mn a-pkane saophite subserate. Furtjermova, additional opticwl annsotrmpy shows kp from the a-plane sapphire substrate covered bf mmnolayer MoS$_2$. Lt worth mentioging that, abovg the trausparevt sapphirv substrave, the real [art of the RG signal is pcedominanely qssoxiated dkth the anisottopy of tkt absorptiin of the monolayev MoS$_2$. Ror comparison, cke absorption spevtrjm jf | denk2014exciton] For the monolayer MoS$_2$ covered a-plane the signals can described by the where and $r_{[0001]}$ denote reflectance of the polarized along the $[1\overline{1}00]$ and the directions of the a-plane sapphire substrate, respectively. ![(a) The RD spectrum taken from monolayer MoS$_2$ on Al$_2$O$_3$($11\overline{2}0$), (b) The absorption spectrum (b) and its first derivative measured monolayer on After the systematic characterization using conventional techniques,[@supplemental] RDS measurement was then applied to investigate the optical within the plane of the MoS$_2$ monolayer.The real of the RD spectra from the bare Al2O3(11$\overline{2}$0) surface the covered by MoS$_2$ plotted Fig. 2(a), respectively. bare Al2O3(11$\overline{2}$0) surface shows an optical anisotropy with an almost constant value which can be directly attributed the in-plane the a-plane substrate. the in-plane axis, namely \[0001\] axis are parallel to the directions of sapphire, respectively. The result reveals thus dielectric anisotropy \epsilon_{ext}=\epsilon_o-\epsilon_e$ in a-plane sapphire substrate. Furthermore, optical anisotropy shows up from the a-plane sapphire covered by monolayer MoS$_2$. It worth mentioning that, above the transparent sapphire substrate, the real the RD signal is associated with the of absorption the MoS$_2$. For the absorption spectrum of | denk2014exciton] For the monolayeR MoS$_2$ covereD a-plaNe ($11\oVerLiNe{2}0$) suBstrAte, the RD signalS Can bE described by the followiNg equAtIOn: $$\laBEl{eq.1}
\fraC{\Delta r}{R}=\FrAC{1}{2}\FraC{r_{[1\OvErlInE{1}00]}-R_{[0001]}}{r_{[1\OverlIne{1}00]}+R_{[0001]}}$$ where $r_{[1\Overline{1}00]}$ anD $r_{[0001]}$ dEnOte the reflecTAnCe of the ligHt pOlarized alonG thE $[1\overlInE{1}00]$ anD The \[0001\] diRecTions Of the a-PLane saPphire subStRAte, resPEctivelY.
![(A) thE RD sPectrum taken from tHE mONolayer MoS$_2$ on Al$_2$o$_3$($11\overlInE{2}0$), (B) THE AbsOrpTion spectrUm (B) and iTS first dERiVATIve (C) Measured from mOnolayer MoS$_2$ ON Al$_2$o$_3$(0001).[]{data-lAbEl="fIGure2"}](FiG2.pdf){wIdTH="8cm"}
after the sysTemaTic characTerizaTIon usinG ConventIonal tEchNiqUes,[@sUPpLeMenTaL] rDS MEaSurEMenT was then ApPlIed to InveSTIGAte tHe oPticAl aniSotropy within The PlanE Of tHe MoS$_2$ MonolAyer.thE real Part of The RD SpEctra measured frOm thE bare Al2O3(11$\oVerLiNe{2}$0) sUrFace aND the onE coVerEd by monOlayer MOs$_2$ arE pLOTTeD in Fig. 2(a), respectivelY. THE BaRe Al2O3(11$\oveRline{2}$0) sURfAcE Shows an oPtIcaL aniSOTropy With AN aLmost conStant vALuE wHich can Be DirectLy AttRibUted tO The iN-plane BirefrinGence OF the a-plane sappHIre substrate. ACTuALLy, THe coRreSponding in-pLane AXis, nAmelY \[1$\OvErlINe{1}$00\] and \[0001\] Axis aRe PArALlel to the ordinary anD eXtraorDinarY directions of Sapphire, reSPECtively. THe reSUlT Reveals thus the DieleCtric anisoTRopy $\DeltA \epsiLon_{ext}=\epSilon_o-\epsILOn_e$ in a-plAne SapPhiRe sUBStRate. FurthermoRE, AddiTiOnal optIcaL anisotRopY shOws Up fRoM the a-planE sapphirE sUbStRaTe cOvereD By monolaYeR Mos$_2$. IT woRth meNTioninG that, AbovE tHe TRanSparent SApPHIre sUbStRate, The ReAl parT of tHE RD Signal iS predominAntLY assOcIaTed with The anisotropy Of The absorptIoN of The monOLAyer MoS$_2$. FOr comparison, the absorptiON spectrUm oF | denk2014exciton] For the m onolayer M oS$_2 $ c ove re d a- plan e ($11\overlin e {2}0 $) substrate, the RD s ignal sc an b e d escri bed byt he f oll ow in g e qu a ti on: $ $\l abel{Eq .1}
\frac{ \De lt a r}{r}=\fra c {1 }{2}\frac{ r_{ [1\overline{ 1}0 0]}-r_ {[ 000 1 ]}}{r _{[ 1\ove rline{ 1 }00]}+ r_{[0001] }} $ $ wher e $r_{[1 \ o ve rlin e{1}00]}$ and $r_ { [0 0 01]}$ denote t he ref le c ta n c e o f t he light p ol arize d alongt he $ [ 1\o v erline{1}00]$ and the \[ 0 001 \] dir ec tio n s of t he a- pl a nesapphire su bstr ate, resp ective l y.
![( a ) The R D spec tru m t aken fr om th em ono l ay erM oS$ _2$ on A l$ _2 $O$_3 $($1 1 \ o v erli ne{ 2}0$ ), (b ) The absorpt ion spe c tru m (b) anditsfi rst d erivat ive ( c) measured frommono layer MoS $_2 $onAl $_2$O $ _3$(00 01) .[] {data-l abel="f i gur e2 " } ] (F ig2.pdf){width="8c m" }
A fter the syste m at ic characte ri zat ionu s ing c onve n ti onal tec hnique s ,[ @s uppleme nt al] RD Smea sur ement wasthen a pplied t o inv e stigate the op t ical anisotro p yw i th i n th e p lane of the MoS $ _2$mono l ay er. T he re al pa rt of the RD spectra meas ur ed fro m the bare Al2O3(1 1$\overlin e { 2 }$0) sur face an d the one cover ed by monolayer MoS$_2$are p lotted i n Fig. 2( a ) , respec tiv ely . T heb a re Al2O3(11$\ov e r line {2 }$0) su rfa ce show s a n o pti cal a nisotropy with an a lm os tcon stant value wh ic h c an be dire c tly at tribu tedto t h e i n-plane bi r e frin ge nc e of th ea-pla ne s a pph ire sub strate. A ctu a lly, t he corres ponding in-pl an e axis, na me ly\[1$\o v e rline{1} $00\] and \[0001\] axis are par all el to the ordinary an d extr aor d inarydirect ionsof sa p p hire, r es pec ti vely. Ther e sul t rev ea ls t hus the dielectric anisot r opy $\Delta \eps ilo n_{e x t }= \ep s il o n_o -\ e psi l o n_e$ in a-plane sapphiresu b st rate. Furt h erm or e, addi tionaloptic a l aniso tropy sho ws up fro mthea - pla ne sapphir e substr ate cover e d bym on olaye r M oS$_2$ .Itworth menti o nin g tha t, abo ve the t ransp ar ent sapp hire substrate, the rea l part of t heRD signal is pre dominantl y as sociated w ith th e ani sot r opy o f th e a bso r ption oft he monola y er Mo S $ _2 $. For comp a r i son , the ab s orptio n sp ectrum of | denk2014exciton] For_the monolayer_MoS$_2$ covered a-plane ($11\overline{2}0$)_substrate, the_RD_signals can_be_described by the_following equation: $$\label{Eq.1}
\frac{\Delta_r}{r}=\frac{1}{2}\frac{r_{[1\overline{1}00]}-r_{[0001]}}{r_{[1\overline{1}00]}+r_{[0001]}}$$ where $r_{[1\overline{1}00]}$ and_$r_{[0001]}$ denote the_reflectance_of the light polarized along the $[1\overline{1}00]$ and the \[0001\] directions of the a-plane_sapphire_substrate, respectively.
![(a)_The_RD_spectrum taken from the monolayer_MoS$_2$ on Al$_2$O$_3$($11\overline{2}0$), (b) The_absorption spectrum_(b) and its first derivative (c) measured from_monolayer_MoS$_2$ on Al$_2$O$_3$(0001).[]{data-label="figure2"}](Fig2.pdf){width="8cm"}
After_the systematic characterization using conventional techniques,[@supplemental] RDS measurement was_then applied to investigate the optical_anisotropy within the_plane_of_the MoS$_2$ monolayer.The real_part of the RD spectra measured_from the bare Al2O3(11$\overline{2}$0) surface and_the one covered by monolayer MoS$_2$ are_plotted in Fig. 2(a), respectively. The bare_Al2O3(11$\overline{2}$0) surface shows an optical_anisotropy with_an almost constant value which_can be directly_attributed to_the in-plane birefringence_of the a-plane sapphire substrate. Actually,_the corresponding in-plane_axis, namely \[1$\overline{1}$00\] and \[0001\] axis_are_parallel to the_ordinary_and_extraordinary directions_of sapphire, respectively._The_result reveals_thus_the dielectric anisotropy $\Delta \epsilon_{ext}=\epsilon_o-\epsilon_e$ in_a-plane_sapphire substrate. Furthermore, additional optical anisotropy shows_up from the a-plane_sapphire_substrate covered by monolayer_MoS$_2$. It worth mentioning that,_above the transparent sapphire substrate, the_real part_of the_RD signal is predominantly associated with the anisotropy of the absorption_of the monolayer MoS$_2$. For comparison,_the absorption spectrum of |
: $\, {{\boldsymbol x}}_{k+1} \, = \, A{{\boldsymbol x}}_k, \, k \ge 0$. In most of practical cases it converges to a leading eigenvector ${{\boldsymbol v}}$ linearly with the rate $O \bigl( \bigl| \frac{\lambda_2}{\lambda_1}\bigr|^k\bigr)$, where $\lambda_1, \lambda_2$ are the first and the second largest by modulus eigenvalues of $A$ or $O\bigl( \frac{1}{k} \bigr)$, if the leading eigenvalue has non-trivial (i.e., of size bigger than one) Jordan blocks. In a rare case when there are several largest by modulus eigenvalues, then the “averaged” sequence $\frac1r\, \sum_{j=0}^{r-1} x_{k+j}$ converges to ${{\boldsymbol v}}$, where $r$ is the [*imprimitivity index*]{} (see [@G chapter 13]). In all cases we will say that the power method converges and, for the sake of simplicity, assume that there is only one largest by modulus eigenvalue, maybe multiple, i.e., that $r=1$.
\[th.40\] The selected leading eigenvector of a non-negative matrix $A$ is proportional to the limit of the power method applied to the matrix $A$ with the initial vector ${{\boldsymbol x}}_0 = {{\boldsymbol e}}$.
[Proof.]{} It may be assumed that $\rho(A) = 1$. The spectral radius of the matrix $A_{\varepsilon}$ strictly increases in $\varepsilon$, so $\rho(A_{\varepsilon}) = 1 + \delta$ for some $\delta > 0$. We have $$\bigl( A\, + \, \varepsilon \, {{\boldsymbol e}}\, {{\boldsymbol e}}^T \bigr)\, {{\boldsymbol v}}_{\varepsilon} \ = \
(1+\delta){{\boldsymbol v}}_{\varepsilon}\,.$$ Denoting by $S_{\varepsilon} = ({{\boldsymbol e}}, {{\boldsymbol v}}_{\varepsilon})$ the sum of entries of the vector ${{\boldsymbol v}}_{\varepsilon}$, we obtain $$A{{\boldsymbol v}}_{\varepsilon}\ + \ \varepsilon \, S_{\varepsilon}\, {{{\boldsymbol e}}} \ = \
(1+\delta)\, {{\boldsymbol v}}_{\varepsilon}\,,$$ and hence $$\left( I \, - \, \frac{1}{(1+\delta)}\, A \right)\, {{\boldsymbol v}}_{\ | : $ \, { { \boldsymbol x}}_{k+1 } \, = \, A{{\boldsymbol x}}_k, \, k \ge 0$. In most of practical cases it converges to a leading eigenvector $ { { \boldsymbol v}}$ linearly with the pace $ oxygen \bigl (\bigl| \frac{\lambda_2}{\lambda_1}\bigr|^k\bigr)$, where $ \lambda_1, \lambda_2 $ are the beginning and the second largest by modulus eigenvalue of $ A$ or $ O\bigl (\frac{1}{k } \bigr)$, if the leading eigenvalue have non - trivial (i.e., of size big than one) Jordan blocking. In a rare case when there embody several largest by modulus eigenvalue, then the “ averaged ” sequence $ \frac1r\, \sum_{j=0}^{r-1 } x_{k+j}$ converges to $ { { \boldsymbol v}}$, where $ r$ is the [ * imprimitivity index * ] { } (see [ @G chapter 13 ]). In all cases we will read that the power method acting converges and, for the sake of simplicity, wear that there embody only one largest by modulus eigenvalue, maybe multiple, i.e., that $ r=1$.
\[th.40\ ] The choose leading eigenvector of a non - negative matrix $ A$ is proportional to the terminus ad quem of the power method acting applied to the matrix $ A$ with the initial vector $ { { \boldsymbol x}}_0 = { { \boldsymbol e}}$.
[ Proof. ] { } It may be assumed that $ \rho(A) = 1$. The spectral radius of the matrix $ A_{\varepsilon}$ strictly increases in $ \varepsilon$, therefore $ \rho(A_{\varepsilon }) = 1 + \delta$ for some $ \delta > 0$. We have $ $ \bigl (A\, + \, \varepsilon \, { { \boldsymbol e}}\, { { \boldsymbol e}}^T \bigr)\, { { \boldsymbol v}}_{\varepsilon } \ = \
(1+\delta){{\boldsymbol v}}_{\varepsilon}\,.$$ Denoting by $ S_{\varepsilon } = ({ { \boldsymbol e } }, { { \boldsymbol v}}_{\varepsilon})$ the sum of entries of the vector $ { { \boldsymbol v}}_{\varepsilon}$, we obtain $ $ A{{\boldsymbol v}}_{\varepsilon}\ + \ \varepsilon \, S_{\varepsilon}\, { { { \boldsymbol e } } } \ = \
(1+\delta)\, { { \boldsymbol v}}_{\varepsilon}\,,$$ and hence $ $ \left (I \, - \, \frac{1}{(1+\delta)}\, A \right)\, { { \boldsymbol v}}_{\ | : $\, {{\blldsymbol x}}_{k+1} \, = \, A{{\boldsymnol x}}_k, \, k \ge 0$. In most oh practjcal casds it converges to a leading eugenvtbtor ${{\boldsymbol v}}$ linexrly with the ratw $O \uigl( \bigl| \frac{\lambda_2}{\lambda_1}\bigr|^i\nigr)$, chxre $\lambda_1, \lambca_2$ are the first and the sdclnd largest by modulus eigenvalues jf $A$ or $P\blgl( \frac{1}{k} \bigr)$, if uhe leaspnn eigenvalue has non-trivial (i.e., or size uigger than one) Jordan blocks. In a rare cwse ahen there are sevfral largesj by nodulus eigevvalues, thtn the “averafed” sequence $\frac1r\, \sum_{j=0}^{r-1} x_{k+j}$ covvergzs to ${{\boldsimgop v}}$, where $r$ ms the [*imprimitivibj index*]{} (see [@G cnapter 13]). In all caves we will say that the power method convgrges and, xox the sake of simplicutt, assome tvat gyerd ia pnmy one lacgest by mosulus eigencalue, maybe multiplt, i.q., that $r=1$.
\[th.40\] The seleceeq leading eigenvector of a non-negative katdix $A$ is proportional ti the limit of the poaer methoq applied to the matrix $A$ with the initial vector ${{\bmldsyjcol x}}_0 = {{\vopdsymbol e}}$.
[Proof.]{} It may be assumed that $\rho(A) = 1$. Fht siectral radius of the matrix $A_{\gatgpsilon}$ strictuy incxsaaes in $\varepsilon$, do $\rho(A_{\darepwilon}) = 1 + \delya$ for some $\delta > 0$. We have $$\bigl( A\, + \, \vagepsulon \, {{\boldsymbol e}}\, {{\boldsymbol e}}^T \nigr)\, {{\noldsymbol v}}_{\varepsilon} \ = \
(1+\dslta){{\boldsymhol v}}_{\varelrilon}\,.$$ Denoting bh $S_{\nare[silon} = ({{\boldsymbol e}}, {{\boldsrmbol v}}_{\vacepsikon})$ the sum of eneries of tje vegdor ${{\boldsymbol v}}_{\varfpsilln}$, we obtain $$A{{\boldsymbol v}}_{\varepsilon}\ + \ \vare'xilon \, S_{\vareksinon}\, {{{\boldsymyol e}}} \ = \
(1+\delta)\, {{\bojdsymbol v}}_{\vareksilon}\,,$$ anb hencd $$\left( I \, - \, \frac{1}{(1+\dxlta)}\, A \rigrt)\, {{\boldsymbon v}}_{\ | : $\, {{\boldsymbol x}}_{k+1} \, = \, \, \ge 0$. most of practical leading ${{\boldsymbol v}}$ linearly the rate $O \bigl| \frac{\lambda_2}{\lambda_1}\bigr|^k\bigr)$, where $\lambda_1, \lambda_2$ are first and the second largest by modulus eigenvalues of $A$ or $O\bigl( \frac{1}{k} if the leading eigenvalue has non-trivial (i.e., of size bigger than one) Jordan In rare when are several largest by modulus eigenvalues, then the “averaged” sequence $\frac1r\, \sum_{j=0}^{r-1} x_{k+j}$ converges to ${{\boldsymbol where $r$ is the [*imprimitivity index*]{} (see [@G 13]). In all cases will say that the power converges for the of assume there is only largest by modulus eigenvalue, maybe multiple, i.e., that $r=1$. \[th.40\] The selected leading eigenvector of a non-negative $A$ is the limit the method to the matrix the initial vector ${{\boldsymbol x}}_0 = It may be assumed that $\rho(A) = 1$. spectral radius the matrix $A_{\varepsilon}$ strictly increases in so $\rho(A_{\varepsilon}) = 1 + \delta$ for some > 0$. We have $$\bigl( A\, + \, \varepsilon \, {{\boldsymbol e}}\, {{\boldsymbol e}}^T \bigr)\, \ = \ (1+\delta){{\boldsymbol Denoting by $S_{\varepsilon} ({{\boldsymbol {{\boldsymbol the of entries the vector ${{\boldsymbol v}}_{\varepsilon}$, we obtain $$A{{\boldsymbol v}}_{\varepsilon}\ + \ \varepsilon S_{\varepsilon}\, {{{\boldsymbol e}}} \ = \ (1+\delta)\, {{\boldsymbol v}}_{\varepsilon}\,,$$ and I - \, \frac{1}{(1+\delta)}\, \right)\, {{\boldsymbol v}}_{\ | : $\, {{\boldsymbol x}}_{k+1} \, = \, A{{\boldsymbol x}}_k, \, K \ge 0$. In most oF pracTicAl cAsEs it ConvErges to a leadinG EigeNvector ${{\boldsymbol v}}$ lineArly wItH The rATe $o \bigl( \Bigl| \fraC{\LaMBDa_2}{\lAmBdA_1}\biGr|^K\BiGr)$, wheRe $\lAmbda_1, \laMbda_2$ are the FirSt And the second LArGest by moduLus Eigenvalues oF $A$ oR $O\bigl( \FrAc{1}{k} \BIgr)$, if The LeadiNg eigeNValue hAs non-trivIaL (I.e., of siZE bigger THAn One) JOrdan blocks. In a rarE CaSE when there are sEveral LaRGeST By mOduLus eigenvaLuEs, theN The “averAGeD” SEQueNCe $\frac1r\, \sum_{j=0}^{r-1} x_{K+j}$ converges TO ${{\boLdsymbOl V}}$, whERe $r$ is tHe [*impRiMItiVity index*]{} (seE [@G chApter 13]). In alL cases WE will saY That the Power mEthOd cOnveRGeS aNd, fOr THe sAKe Of sIMplIcity, assUmE tHat thEre iS ONLY one LarGest By modUlus eigenvaluE, maYbe mULtiPle, i.e., That $r=1$.
\[Th.40\] ThE sElectEd leadIng eiGeNvector of a non-neGatiVe matrix $A$ Is pRoPorTiOnal tO The limIt oF thE power mEthod apPLieD tO THE mAtrix $A$ with the initiAl VECtOr ${{\boldsyMbol x}}_0 = {{\bOLdSyMBol e}}$.
[ProoF.]{} IT maY be aSSUmed tHat $\rHO(A) = 1$. the spectRal radIUs Of The matrIx $a_{\varepSiLon}$ StrIctly INcreAses in $\VarepsilOn$, so $\rHO(A_{\varepsilon}) = 1 + \deLTa$ for some $\deltA > 0$. we HAVe $$\BIgl( A\, + \, \VarEpsilon \, {{\boldSymbOL e}}\, {{\boLdsyMBoL e}}^T \BIgr)\, {{\boLdsymBoL V}}_{\vARepsilon} \ = \
(1+\delta){{\boldsyMbOl v}}_{\varEpsilOn}\,.$$ Denoting by $S_{\Varepsilon} = ({{\BOLDsymbol e}}, {{\BoldSYmBOl v}}_{\varepsilon})$ tHe sum Of entries oF The vectoR ${{\boldSymbol v}}_{\vArepsilon}$, WE Obtain $$A{{\bOldSymBol V}}_{\vaREPsIlon}\ + \ \varepsiloN \, s_{\VarePsIlon}\, {{{\bolDsyMbol e}}} \ = \
(1+\deLta)\, {{\BolDsyMboL v}}_{\VarepsiloN}\,,$$ and hencE $$\lEfT( I \, - \, \FrAc{1}{(1+\dElta)}\, A \RIght)\, {{\boldSyMboL v}}_{\ | : $\, {{\boldsymbol x}}_{k +1} \, = \ , A{{ \bo lds ym bolx}}_ k, \, k \ge 0$ . Inmost of practical case s itco n verg e sto aleading ei g e nve ct or ${ {\ b ol dsymb olv}}$ li nearly wit h t he rate $O \bi g l( \bigl| \f rac {\lambda_2}{ \la mbda_1 }\ big r |^k\b igr )$, w here $ \ lambda _1, \lamb da _ 2$ are the fir s t a nd t he second largest by modulus eigenv aluesof $A $ or$O\ bigl( \fra c{ 1}{k} \bigr)$ , i f t hel eading eigenv alue has no n -tr ivial(i .e. , of si ze bi gg e r t han one) Jo rdan blocks.In a r a re case when th ere ar e s eve rall ar ge stby mod u lu s e i gen values,th en the“ave r a g e d” s equ ence $\fr ac1r\, \sum_{ j=0 }^{r - 1}x_{k+ j}$ c onve rg es to ${{\b oldsy mb ol v}}$, where$r$is the [* imp ri mit iv ity i n dex*]{ } ( see [@G ch apter 1 3 ]). I n a ll cases we will say t h a tthe powe r meth o dco n verges a nd , f or t h e sake ofs im plicity, assum e t ha t there i s only o nelar gestb y mo duluseigenval ue, m a ybe multiple,i .e., that $r= 1 $. \[ t h.40 \]The selecte d le a ding eig e nv ect o r ofa non -n e ga t ive matrix $A$ is p ro portio nal t o the limit o f the powe r m ethod ap plie d t o the matrix $A $ wit h the init i al vecto r ${{ \boldsym bol x}}_0 = {{\bold sym bol e} }$. [P roof.]{} It m a y beas sumed t hat $\rho( A)= 1 $.The s pectral r adius of t he m at rix $A_{ \ varepsil on }$st ric tly i n crease s in$\va re ps i lon $, so $ \ rh o ( A_{\ va re psil on} )= 1+ \d e lta $ for s ome $\del ta> 0$. W ehave $$ \bigl( A\, +\, \varepsil on \, {{\bo l d symbol e }}\, {{\boldsymbol e}}^ T \bigr) \,{{\bo ldsy mbol v}}_ {\v arepsi lon } \ = \
(1+\ delta ){ {\b o l dsymb o l v }}_ {\ varepsilon } \ ,.$ $ Den ot ingby $S_{ \varepsilon} = ({{ \ bol dsymbol e}},{{\ bold s y mb olv }} _ {\v ar e psi l o n})$ the sum of entries o ft he vector ${ { \bo ld symbolv}}_{\v areps i lon}$,we obtain $$A{{\bo ld symb o l v} }_{\vareps ilon}\ + \ \varep s ilon\ ,S_{\v are psilon }\ , { {{\bo ldsymb o l e }}} \ = \ (1+\de lta)\ ,{{\bolds ymbol v}}_{\varepsilon} \,,$$and h enc e $$\left ( I \,- \, \fra c{1} {(1+\delta )}\ , A \r igh t )\, {{\b o ld sym b ol v} }_{\ | : $\,_{{\boldsymbol x}}_{k+1}_\, = \, A{{\boldsymbol_x}}_k, \,_k_\ge 0$._In_most of practical_cases it converges_to a leading eigenvector ${{\boldsymbol_v}}$ linearly with_the_rate $O \bigl( \bigl| \frac{\lambda_2}{\lambda_1}\bigr|^k\bigr)$, where $\lambda_1, \lambda_2$ are the first and the second_largest_by modulus_eigenvalues_of $A$_or $O\bigl( \frac{1}{k} \bigr)$, if_the leading eigenvalue has non-trivial_(i.e., of_size bigger than one) Jordan blocks. In a_rare_case when there_are several largest by modulus eigenvalues, then the “averaged”_sequence $\frac1r\, \sum_{j=0}^{r-1} x_{k+j}$ converges to_${{\boldsymbol v}}$, where_$r$_is_the [*imprimitivity index*]{} (see [@G_chapter 13]). In all cases we_will say that the power method_converges and, for the sake of simplicity,_assume that there is only one_largest by modulus eigenvalue, maybe_multiple, i.e.,_that $r=1$.
\[th.40\] The selected leading_eigenvector of a_non-negative matrix_$A$ is proportional_to the limit of the power_method applied to_the matrix $A$ with the initial vector ${{\boldsymbol_x}}_0_= {{\boldsymbol e}}$.
[Proof.]{}_It_may_be assumed_that $\rho(A) =_1$._The spectral_radius_of the matrix $A_{\varepsilon}$ strictly increases_in_$\varepsilon$, so $\rho(A_{\varepsilon}) = 1 +_\delta$ for some $\delta_>_0$. We have $$\bigl(_A\, + \, \varepsilon \,_{{\boldsymbol e}}\, {{\boldsymbol e}}^T \bigr)\, {{\boldsymbol_v}}_{\varepsilon} \_= \_
(1+\delta){{\boldsymbol v}}_{\varepsilon}\,.$$ Denoting by $S_{\varepsilon} = ({{\boldsymbol e}}, {{\boldsymbol v}}_{\varepsilon})$ the_sum of entries of the vector ${{\boldsymbol_v}}_{\varepsilon}$, we obtain $$A{{\boldsymbol_v}}_{\varepsilon}\ +_\_\varepsilon \, S_{\varepsilon}\,_{{{\boldsymbol_e}}} _ \ = \
(1+\delta)\, {{\boldsymbol v}}_{\varepsilon}\,,$$_and hence_$$\left( I \, - \, \frac{1}{(1+\delta)}\,_A \right)\,__{{\boldsymbol v}}_{\ |
been conducted in much smaller boxes at significantly smaller spectral resolutions.
A further interesting observation that was made in the present study is the connection between the mean scales of the turbulent superstructure patterns analysed in the midplane and those of the near-wall flows. Our analysis suggests that the characteristic scales of large-scale superstructures are correlated with the thermal plume ridges in the boundary layers. We showed for all $Pr$ that the maximum wavenumber of the temperature spectrum in the midplane $k^{\ast}_{\Theta}$ nearly perfectly coincides with the wavenumber at which the power spectrum of the divergence of the skin friction field peaks. The latter wavenumber characterizes the mean distance of impact (div ${\bm s} >0$) and ejection (div ${\bm s} <0$) regions at the walls. It is thus the characteristic variation scale of the horizontal velocity field that advects the hot (cold) fluid together at the bottom (top) boundary to form prominent thermal plume ridges. The interplay between the thermal and viscous boundary layers of different thicknesses could thus be responsible for the variation of the characteristic superstructure scale with growing $Pr$. The viscous boundary layer becomes ever thicker as $Pr$ increases and velocity fluctuations decrease thus generating more coherent advection patterns. Competing boundary layers that control transport and structure formation in convection flows have been discussed in other settings, for example in ref. [@King2009] for rapidly rotating convection.
The characteristic superstructure scales which we have detected in the present work suggest a scale separation for convective turbulence. There is the fast convective motion below the characteristic width of individual circulation rolls or cells on times smaller than several tens of free-fall times. Then after the removal of the small-scale turbulence, the large-scale patterns of rolls are revealed and these fill the whole layer and vary slowly on time scales larger than a few hundreds of free-fall times. The latter dynamic processes can be of interest for a global effective description of mesoscale convection phenomena in atmospheric turbulence [@Randall2003] or of pattern formation in a scale range between solar granulation and supergranulation [@Rincon2017]. In contrast to rapidly rotating convection flows or magnetoconvection in the presence of strong external magnetic fields, the present RBC flow permits a mathematically rigorous asymptotic expansion that generates simplified equations for the dynamics of these patterns (see e.g. [@Stellmach2014]). The unresolved dynamics at the fine and fast scales below $\lambda_{\Theta,U | been conducted in much smaller boxes at importantly small spectral resoluteness.
A far interesting observation that was made in the present sketch is the joining between the mean plate of the turbulent superstructure patterns analyze in the midplane and those of the near - wall stream. Our psychoanalysis suggests that the characteristic scales of big - scale superstructures are correlate with the thermal plume ridges in the boundary layers. We show for all $ Pr$ that the maximum wavenumber of the temperature spectrum in the midplane $ k^{\ast}_{\Theta}$ nearly perfectly coincides with the wavenumber at which the ability spectrum of the divergence of the skin friction field peaks. The latter wavenumber qualify the mean distance of impact (div $ { \bm s } > 0 $) and ejection (div $ { \bm s } < 0 $) regions at the walls. It is thus the characteristic variation scale of the horizontal velocity field that advects the hot (cold) fluid together at the bottom (top) boundary to form prominent thermal feather ridge. The interplay between the thermal and viscous boundary level of unlike thicknesses could thus be responsible for the variation of the characteristic superstructure scale with growing $ Pr$. The viscous boundary layer become ever thicker as $ Pr$ increases and velocity fluctuations decrease thus render more coherent advection patterns. Competing boundary layers that control transportation and social organization constitution in convection flows have been discussed in other settings, for case in ref. [ @King2009 ] for rapidly rotating convection.
The characteristic superstructure scale which we have detect in the present work suggest a scale separation for convective turbulence. There is the flying convective motion below the characteristic width of individual circulation rolls or cell on times modest than several ten of free - capitulation times. Then after the removal of the small - plate turbulence, the large - scale patterns of rolls are revealed and these fill the whole layer and vary slowly on time scales bigger than a few hundreds of free - spill times. The latter dynamic processes can be of interest for a global effective description of mesoscale convection phenomena in atmospheric turbulence [ @Randall2003 ] or of design formation in a scale range between solar granulation and supergranulation [ @Rincon2017 ]. In contrast to rapidly rotating convection flows or magnetoconvection in the bearing of strong external magnetic fields, the present RBC flow permits a mathematically rigorous asymptotic expansion that generates simplified equations for the moral force of these patterns (see e.g. [ @Stellmach2014 ]). The unresolved dynamics at the fine and fast scales below $ \lambda_{\Theta, U | befn conducted in much smauler boxes at sntnificently sjaller soectral resolutions.
A further ibterewting observation that was made in the prestnt study is the rknnection betscen tke mean scales on the turbunent superstruwtjrz patterns analysed in the midplane wnd thoxe of the near-wajl fkjws. Klr analysis suggests that the cgaractegistic scales of karge-scale superstructures are correlated with tje thermal kmumq ridges in tfe boundarj layers. We showed for all $Pr$ that the maxkmum cavenumber if thf temperaturx specnrum in the mlcplane $k^{\ast}_{\Thrta}$ nearly pernectlb councides with the waveiumber at which the kower specdrbm of the divergence if the vkin frizriov fjekd peaks. Thx latter wabenumber chqracterizes the meam qpxtance of imlact (dyv ${\bm s} >0$) and ejection (div ${\bm s} <0$) regions au the walls. It is thus the cyaracteristic variatiln scale jf the horizontal velocity field that advects the hot (roud) youid gigfther at the bottom (top) boundary to form promyhemt thermal plume ridges. The ijtrtplay between jhe thexjam and viscous bounfary laiers od differegt tnicknesses could thus be rewponsible fog thw variation of the characterirtic supetstructure scale with gxowing $Pr$. The visfous bounsxry layer becomer ener dhicker as $Pr$ increases anq velocitb fluetuationr devrease thus genegatinn more coherent advfctiou patderns. Compfting boundary layers that contcpl transport atd vtructurz formstion in convqction flows hcve been discjssed in onher settmngs, for exaiple in ref. [@Khjg2009] for rapidny rotatyng xonvwction.
Tfd characteristoc supersnrbcture scqles which we have dejedted in the prewenu qork suggest a scxle svpacatiog for convecthve gurcilencd. There is bhe fasy convective motion telos the characteristoc width od indiviqual circulatoon rolls or cells on tmmes soallet tran several tens of free-fall tjmes. Then afber the removwl on thg small-scalz turbulence, the large-scale patterns of colls are revealed and rhese fill the wholg lsyer and vacy slorly on tike scales larger thab a few hundreds pf free-fall times. The lattes dynwmic processes can be of interest for a global effective description of mwsoscale convectioh phrnometa nn atmofphecir turbulence [@Randsll2003] or of pattern formation in e scale ratgz between solar granulation amd supergranulajion [@Rincon2017]. In contrast to rapidly rotating convection flows or msgnetoconvection in the iresence if stdong eqjerial magnetic fields, yhe przsent EBC floq pevmits a mathemxtivaply rigorlus esyjptotic expansion that generates simplufled equations foc the dynamigs of tgese pwtterns (see e.g. [@Sfelljwch2014]). Tfe unresolved dynamics at tne fine and fast scanef below $\lambda_{\Uheta,U | been conducted in much smaller boxes at spectral A further observation that was is connection between the scales of the superstructure patterns analysed in the midplane those of the near-wall flows. Our analysis suggests that the characteristic scales of superstructures are correlated with the thermal plume ridges in the boundary layers. We for $Pr$ the wavenumber of the temperature spectrum in the midplane $k^{\ast}_{\Theta}$ nearly perfectly coincides with the wavenumber at the power spectrum of the divergence of the friction field peaks. The wavenumber characterizes the mean distance impact ${\bm s} and (div s} <0$) regions the walls. It is thus the characteristic variation scale of the horizontal velocity field that advects the (cold) fluid the bottom boundary form thermal plume ridges. between the thermal and viscous boundary thicknesses could thus be responsible for the variation the characteristic scale with growing $Pr$. The viscous layer becomes ever thicker as $Pr$ increases and fluctuations decrease thus generating more coherent advection patterns. Competing boundary layers that control transport and in convection flows have discussed in other for in [@King2009] rapidly rotating The characteristic superstructure scales which we have detected in the present suggest a scale separation for convective turbulence. There is the motion the characteristic width individual circulation rolls or on smaller than several tens times. after the turbulence, large-scale patterns of rolls revealed and these fill the layer and vary slowly a few hundreds of free-fall times. The latter processes can be of interest for a effective description of mesoscale convection phenomena in atmospheric turbulence [@Randall2003] or of formation in range between solar granulation and supergranulation [@Rincon2017]. In to rapidly rotating convection or magnetoconvection in the presence of strong external magnetic the RBC flow a mathematically rigorous expansion that generates equations for the these patterns e.g. The the fine and fast scales below | been conducted in much smalleR boxes at siGnifiCanTly SmAlleR speCtral resolutioNS.
A fuRther interesting observAtion ThAT was MAdE in thE presenT StUDY is ThE cOnnEcTIoN betwEen The mean Scales of thE tuRbUlent superstRUcTure patterNs aNalysed in the MidPlane aNd ThoSE of thE neAr-walL flows. oUr analYsis suggeStS That thE CharactERIsTic sCales of large-scale SUpERstructures are CorrelAtED wITH thE thErmal plume RiDges iN The bounDArY LAYerS. we showed for alL $Pr$ that the mAXimUm waveNuMbeR Of the tEmperAtURe sPectrum in thE midPlane $k^{\ast}_{\theta}$ nEArly perFEctly coIncideS wiTh tHe waVEnUmBer At WHicH ThE poWEr sPectrum oF tHe DiverGencE OF THe skIn fRictIon fiEld peaks. The laTteR wavENumBer chAractErizEs The meAn distAnce oF iMpact (div ${\bm s} >0$) and eJectIon (div ${\bm s} <0$) RegIoNs aT tHe walLS. It is tHus The CharactEristic VAriAtION ScAle of the horizontal VeLOCiTy field tHat advECtS tHE hot (cold) FlUid TogeTHEr at tHe boTToM (top) bounDary to FOrM pRominenT tHermal PlUme RidGes. ThE InteRplay bEtween thE therMAl and viscous boUNdary layers of DIfFEReNT thiCknEsses could tHus bE RespOnsiBLe For THe varIatioN oF ThE Characteristic superStRucturE scalE with growing $PR$. The viscouS BOUndary laYer bECoMEs ever thicker aS $Pr$ inCreases and VElocity fLuctuAtions deCrease thuS GEneratinG moRe cOheRenT ADvEction patternS. cOmpeTiNg boundAry Layers tHat ConTroL trAnSport and sTructure FoRmAtIoN in ConveCTion flowS hAve BeEn dIscusSEd in otHer seTtinGs, FoR ExaMple in rEF. [@KING2009] for RaPiDly rOtaTiNg conVectIOn.
THe charaCteristic SupERstrUcTuRe scaleS which we have dEtEcted in the PrEseNt work SUGgest a scAle separation for convectIVe turbuLenCe. TheRe is The fast coNveCtive mOtiON below The chaRacteRiStiC WIdth oF INdIviDuAl circulatION roLls or CeLls oN times sMaller than several tENs oF free-fall timeS. ThEn afTER tHe rEMoVAl oF tHE smALL-scale turbulencE, the large-sCaLE pAtterns of rOLls ArE revealEd and thEse fiLL the whoLe layer anD vary slowLy On tiME ScaLes larger tHan a few hUndreds of FRee-faLL tImes. THe lAtter dYnAmiC procEsses cAN be Of intErest fOr A globaL effeCtIve descrIption of mesoscale convecTion phEnomeNa iN atmospheRic TUrbUlence [@RanDall2003] Or of patterN foRmaTion iN a sCAle raNge bETwEen SOlar gRanuLAtion and sUPeRgrANUlAtion [@Rincon2017]. iN COntRast tO raPIdly roTatiNg convection flows OR magnetoconvecTion IN The PreSEnce Of Strong external MagNeTIC fields, tHe Present RBC fLow permiTs A MatheMaticaLly rigOrous asYMPtOTic expAnsiOn tHat generaTes SiMPlified EqUaTIons foR the DyNamics Of thesE PattERNs (see e.g. [@Stellmach2014]). the unRESolveD DynAmics At The fine ANd faSt scales beLow $\lambda_{\ThEta,U | been conducted in much sm aller boxe s atsig nif ic antl y sm aller spectral reso lutions.
A further in teres ti n g ob s er vatio n thatw as m ade i nthe p r es ent s tud y is th e connecti onbe tween the me a nscales ofthe turbulent s upe rstruc tu rep atter nsanaly sed in the mi dplane an dt hose o f the ne a r -w allflows. Our analys i ss uggests that t he cha ra c te r i sti c s cales of l ar ge-sc a le supe r st r u c tur e s are correla ted with th e th ermalpl ume ridges in t he bou ndary layer s. W e showedfor al l $Pr$ t h at themaximu m w ave numb e rof th et emp e ra tur e sp ectrum i nth e mid plan e $ k ^{\a st} _{\T heta} $ nearly perf ect ly c o inc ideswiththewa venum ber at whic hthe power spect rumof the di ver ge nce o f the skin f ric tio n field peaks. The l a t t er wavenumber charac te r i ze s the me an dis t an ce of impac t(di v ${ \ b m s}>0$) an d ejecti on (di v $ {\ bm s} < 0$ ) regi on s a t t he wa l ls.It isthus the char a cteristic vari a tion scale of th e ho r izon tal velocity f ield that adv e ct s t h e hot (col d) fl u id together at thebo ttom ( top)boundary to f orm promin e n t thermal plu m er idges. The int erpla y betweent he therm al an d viscou s boundar y layers o f d iff ere ntt h ic knesses could t husbe respon sib le forthe va ria tio nof the ch aracteri st ic s up ers truct u re scale w ith g row ing $ P r$. Th e vis cous b ou n dar y layer be c o mesev er thi cke ras $P r$ i n cre ases an d velocit y f l uctu at io ns decr ease thus gen er ating more c ohe rent a d v ection p atterns. Competing boun d ary lay ers that con trol tran spo rt and st r ucture forma tionin co n v ectio n fl ows h ave been d i s cus sed i nothe r setti ngs, for example i n re f. [@King2009 ] f or r a p id lyr ot a tin gc onv e c tion.
The char acteristic s u pe rstructure sca le s which we hav e det e cted in the pres ent worksu gges t a s cale separ ation fo r convect i ve tu r bu lence . T here i sthe fast conve c tiv e mot ion be lo w thechara ct eristicwidth of individual cir culati on ro lls or cells on tim es smalle r th an several te nsof fr ee- f all t imes . T hen after the removal o f t hes m al l-scale tur b u l enc e, th e l a rge-sc alepatterns of rolls are revealed a nd t h e sefil l the w hole layer and va ry s lowly on t ime scaleslarger t ha n a fe w hund reds o f free- f a ll times. The la tter dyna mic p r ocesses c an be ofinte re st for a glo b al e f f ective descripti on of m esosc a leconve ct ion phe n omen a in atmos pheric turb ulence [@R andal l2003]or of pa tte rn formation in a scal e ran ge betw ee n so lar granu lati o n andsupe rg ran ulation [ @ R in c on 20 1 7]. Incontr as t to rapidlyr otatingcon v ectionfl ows o r magn e to c o nvection i n t he pr e s ence of st r onge xt e rnalmagnet ic fie lds, th e pr es ent RBC fl o w permitsa mathema t ica ll y ri gorous a sy mpto ti c e xp ansion that gen erates s implifie d equati ons fort he dyn am ics of these patt er n s (seee . g.[@Stell m ach 201 4]) . Th e u nres o lved dynamics at the fin e an d fas t sc alesbelow $ \lambda_ { \The ta,U | been_conducted in_much smaller boxes at_significantly smaller_spectral_resolutions.
A further_interesting_observation that was_made in the_present study is the_connection between the_mean_scales of the turbulent superstructure patterns analysed in the midplane and those of the_near-wall_flows. Our_analysis_suggests_that the characteristic scales of_large-scale superstructures are correlated with_the thermal_plume ridges in the boundary layers. We showed_for_all $Pr$ that_the maximum wavenumber of the temperature spectrum in the_midplane $k^{\ast}_{\Theta}$ nearly perfectly coincides with_the wavenumber at_which_the_power spectrum of the_divergence of the skin friction field_peaks. The latter wavenumber characterizes the_mean distance of impact (div ${\bm s}_>0$) and ejection (div ${\bm s}_<0$) regions at the walls._It is_thus the characteristic variation scale_of the horizontal_velocity field_that advects the_hot (cold) fluid together at the_bottom (top) boundary_to form prominent thermal plume ridges._The_interplay between the_thermal_and_viscous boundary_layers of different_thicknesses_could thus_be_responsible for the variation of the_characteristic_superstructure scale with growing $Pr$. The viscous_boundary layer becomes ever_thicker_as $Pr$ increases and_velocity fluctuations decrease thus generating_more coherent advection patterns. Competing boundary_layers that_control transport_and structure formation in convection flows have been discussed in other_settings, for example in ref. [@King2009]_for rapidly rotating convection.
The_characteristic superstructure_scales_which we have_detected_in the_present work suggest a scale separation for_convective turbulence._There is the fast convective motion_below the characteristic width_of_individual circulation rolls or cells on_times smaller than several tens of_free-fall times. Then after the_removal_of_the small-scale turbulence, the large-scale_patterns of rolls are revealed and_these fill the_whole layer and vary slowly on time_scales_larger than a few hundreds of_free-fall_times. The latter dynamic processes can_be_of_interest for a global effective_description of mesoscale convection phenomena in_atmospheric turbulence [@Randall2003] or of pattern formation in a_scale range between_solar granulation and supergranulation [@Rincon2017]._In_contrast_to rapidly rotating convection flows or magnetoconvection in the presence_of strong_external magnetic fields,_the present RBC flow permits a mathematically rigorous asymptotic expansion_that generates simplified equations for the dynamics_of these patterns (see e.g. [@Stellmach2014]). The unresolved dynamics at the_fine and fast scales below $\lambda_{\Theta,U |
r = r_{\operatorname{\mathfrak{w}}}$.
- Let $\sigma$ be any ${\mathbb{C}}[z]$–linear automorphism of $\operatorname{\mathfrak{g}}[z]$ and $\operatorname{\mathfrak{u}}= \sigma(\operatorname{\mathfrak{w}}) \subset \widehat\operatorname{\mathfrak{g}}$ be the transformed order. Then the solutions $r_{\operatorname{\mathfrak{w}}}$ and $r_{\operatorname{\mathfrak{u}}}$ are gauge-equivalent: $$r_{\operatorname{\mathfrak{u}}}(x, y) = \bigl(\sigma(x) \otimes \sigma(y)\bigr) r_{\operatorname{\mathfrak{w}}}(x, y).$$
- The described correspondence $\operatorname{\mathfrak{w}}\mapsto r_{\operatorname{\mathfrak{w}}}$ provides a bijection between the gauge equivalence classes of rational solutions of (\[eq:CYBE\]) satisfying (\[E:AnsatzStolin\]) and the orbits of Lagrangian orders in $\widehat\operatorname{\mathfrak{g}}$ with respect to the action of of $\operatorname{\mathsf{Aut}}_{{\mathbb{C}}[z]}\bigl(\operatorname{\mathfrak{g}}[z]\bigr)$.
Let $\operatorname{\mathfrak{w}}= z^{-1} \operatorname{\mathfrak{g}}\llbracket z^{-1}\rrbracket$. It is easy to see that $\operatorname{\mathfrak{w}}$ is a Lagrangian order in $\widehat\operatorname{\mathfrak{g}}$. Let $\bigl\{\alpha_l\bigr\}_{l = 1}^{n^2-1}$ be any basis of $\operatorname{\mathfrak{g}}$. Then we have: $
\beta_{l, k} := \bigl(\alpha_l z^k\bigr)^\vee = \alpha_l^\vee z^{-k-1}.
$ This implies: $$r_{\operatorname{\mathfrak{w}}}(x, y) =
\sum\limits_{k=0}^\infty x^k \sum\limits_{l = 1}^{n^2-1} \alpha_l \otimes \alpha_l^\vee y^{-k-1}
= \frac{c}{y-x},$$ where $c \in \operatorname{\mathfrak{g}}\otimes \operatorname{\mathfrak{g}}$ is the Casimir element. The tensor-valued function $r_{\operatorname{\mathfrak{w}}}$ is the celebrated Yang’s solution of the classical Yang–Baxter equation (\[eq:CYBE\]).
\[ | r = r_{\operatorname{\mathfrak{w}}}$.
- Let $ \sigma$ be any $ { \mathbb{C}}[z]$–linear automorphism of $ \operatorname{\mathfrak{g}}[z]$ and $ \operatorname{\mathfrak{u}}= \sigma(\operatorname{\mathfrak{w } }) \subset \widehat\operatorname{\mathfrak{g}}$ be the transformed order. Then the solution $ r_{\operatorname{\mathfrak{w}}}$ and $ r_{\operatorname{\mathfrak{u}}}$ are bore - equivalent: $ $ r_{\operatorname{\mathfrak{u}}}(x, y) = \bigl(\sigma(x) \otimes \sigma(y)\bigr) r_{\operatorname{\mathfrak{w}}}(x, y).$$
- The described correspondence $ \operatorname{\mathfrak{w}}\mapsto r_{\operatorname{\mathfrak{w}}}$ provide a bijection between the bore equivalence classes of rational solution of (\[eq: CYBE\ ]) satisfying (\[E: AnsatzStolin\ ]) and the orbits of Lagrangian orders in $ \widehat\operatorname{\mathfrak{g}}$ with esteem to the action of of $ \operatorname{\mathsf{Aut}}_{{\mathbb{C}}[z]}\bigl(\operatorname{\mathfrak{g}}[z]\bigr)$.
Let $ \operatorname{\mathfrak{w}}= z^{-1 } \operatorname{\mathfrak{g}}\llbracket z^{-1}\rrbracket$. It is comfortable to see that $ \operatorname{\mathfrak{w}}$ is a Lagrangian order in $ \widehat\operatorname{\mathfrak{g}}$. Let $ \bigl\{\alpha_l\bigr\}_{l = 1}^{n^2 - 1}$ be any basis of $ \operatorname{\mathfrak{g}}$. Then we accept: $
\beta_{l, k }: = \bigl(\alpha_l z^k\bigr)^\vee = \alpha_l^\vee z^{-k-1 }.
$ This implies: $ $ r_{\operatorname{\mathfrak{w}}}(x, y) =
\sum\limits_{k=0}^\infty x^k \sum\limits_{l = 1}^{n^2 - 1 } \alpha_l \otimes \alpha_l^\vee y^{-k-1 }
= \frac{c}{y - x},$$ where $ c \in \operatorname{\mathfrak{g}}\otimes \operatorname{\mathfrak{g}}$ is the Casimir element. The tensor - valued function $ r_{\operatorname{\mathfrak{w}}}$ is the celebrated Yang ’s solution of the authoritative Yang – Baxter equation (\[eq: CYBE\ ]).
\ [ | r = g_{\operatorname{\mathfrak{w}}}$.
- Let $\sigma$ be aut ${\mathub{C}}[z]$–linsar autooorphism of $\operatorname{\mathhrak{t}}[z]$ ane $\operatorname{\mathfrak{j}}= \sigma(\opvratornamw{\matifrak{w}}) \subset \wivshat\opevctornzle{\machhrak{g}}$ be the trsnsformed mrder. Then the sulbtions $r_{\operatorname{\mathfrak{w}}}$ and $r_{\o[eratormale{\mathfrak{u}}}$ arg gaubq-equjnaoent: $$r_{\operatorname{\mathfrak{u}}}(x, g) = \bigl(\vigma(x) \otimes \sigma(y)\bigr) r_{\operatorname{\mwthfgak{w}}}(x, y).$$
- The descgibed correwponqwnce $\operatofname{\mathfgck{w}}\mapsto r_{\kperatorname{\mathfrak{w}}}$ provides x bijzction betwgzb tjg gauge equitalencv classes of vstionan solutoons of (\[eq:CYBE\]) savisfting (\[E:AnsatzStolin\]) anv the orbits of Lagrwngian orgexs in $\widehat\operatorbane{\matvfran{g}}$ wkrh feskecv tk the wctmon of of $\oleratorname{\nathsf{Aut}}_{{\mathbb{C}}[z]}\bibl(\jirratorname{\mafhfrak{d}}[z]\figr)$.
Let $\operatorname{\mathfrak{w}}= z^{-1} \operatognams{\mathfrak{g}}\llbracket z^{-1}\rrvracket$. It is easy to see that $\operatorname{\mathfrak{w}}$ is a Lagrangian order in $\whdehav\ooercbjfbale{\mathfrak{g}}$. Let $\bigl\{\alpha_l\bigr\}_{l = 1}^{n^2-1}$ be any basya pf $\operatorname{\mcthfrak{g}}$. Then we hwvr: $
\beta_{l, k} := \biel(\alphc_m a^k\bigr)^\vee = \alpha_l^\vfe z^{-k-1}.
$ Tris inplies: $$r_{\okeratprname{\mathfrak{w}}}(x, y) =
\sum\limitw_{k=0}^\infty x^k \slm\linits_{l = 1}^{n^2-1} \alpha_l \otnmes \alpha_l^\vze y^{-k-1}
= \grac{c}{u-x},$$ where $c \in \operatorncme{\matgfrak{g}}\otimed \operatodvame{\mathfrak{g}}$ is thv Cavimir eltoent. The tensor-vajued funcvion $x_{\operatofnamg{\mathfrwk{w}}}$ is the celensated Yang’s solutiln of tve classicwl Yang–Baxter equation (\[eq:CYBE\]).
\[ | r = r_{\operatorname{\mathfrak{w}}}$. - Let $\sigma$ be automorphism $\operatorname{\mathfrak{g}}[z]$ and \sigma(\operatorname{\mathfrak{w}}) \subset \widehat\operatorname{\mathfrak{g}}$ the $r_{\operatorname{\mathfrak{w}}}$ and $r_{\operatorname{\mathfrak{u}}}$ gauge-equivalent: $$r_{\operatorname{\mathfrak{u}}}(x, y) \bigl(\sigma(x) \otimes \sigma(y)\bigr) r_{\operatorname{\mathfrak{w}}}(x, y).$$ - described correspondence $\operatorname{\mathfrak{w}}\mapsto r_{\operatorname{\mathfrak{w}}}$ provides a bijection between the gauge equivalence classes of solutions of (\[eq:CYBE\]) satisfying (\[E:AnsatzStolin\]) and the orbits of Lagrangian orders in $\widehat\operatorname{\mathfrak{g}}$ respect the of $\operatorname{\mathsf{Aut}}_{{\mathbb{C}}[z]}\bigl(\operatorname{\mathfrak{g}}[z]\bigr)$. Let $\operatorname{\mathfrak{w}}= z^{-1} \operatorname{\mathfrak{g}}\llbracket z^{-1}\rrbracket$. It is easy to see that $\operatorname{\mathfrak{w}}$ is a Lagrangian in $\widehat\operatorname{\mathfrak{g}}$. Let $\bigl\{\alpha_l\bigr\}_{l = 1}^{n^2-1}$ be any of $\operatorname{\mathfrak{g}}$. Then we $ \beta_{l, k} := \bigl(\alpha_l = z^{-k-1}. $ implies: y) \sum\limits_{k=0}^\infty x^k \sum\limits_{l 1}^{n^2-1} \alpha_l \otimes \alpha_l^\vee y^{-k-1} = \frac{c}{y-x},$$ where $c \in \operatorname{\mathfrak{g}}\otimes \operatorname{\mathfrak{g}}$ is the Casimir element. The function $r_{\operatorname{\mathfrak{w}}}$ celebrated Yang’s of classical equation (\[eq:CYBE\]). \[ | r = r_{\operatorname{\mathfrak{w}}}$.
- LeT $\sigma$ be anY ${\mathBb{C}}[Z]$–liNeAr auTomoRphism of $\operatORnamE{\mathfrak{g}}[z]$ and $\operatorName{\mAtHFrak{U}}= \SiGma(\opEratornAMe{\MAThfRaK{w}}) \SubSeT \WiDehat\OpeRatornaMe{\mathfrak{G}}$ be ThE transformed ORdEr. Then the sOluTions $r_{\operatOrnAme{\matHfRak{W}}}$ And $r_{\oPerAtornAme{\matHFrak{u}}}$ aRe gauge-eqUiVAlent: $$r_{\OPeratorNAMe{\MathFrak{u}}}(x, y) = \bigl(\sigma(x) \OTiMEs \sigma(y)\bigr) r_{\oPeratoRnAMe{\MAThfRak{W}}}(x, y).$$
- The descRiBed coRRespondENcE $\OPEraTOrname{\mathfraK{w}}\mapsto r_{\opERatOrname{\MaThfRAk{w}}}$ proVides A bIJecTion between The gAuge equivAlence CLasses oF RationaL solutIonS of (\[Eq:CYbe\]) sAtIsfYiNG (\[E:ANSaTzSTOliN\]) and the oRbItS of LaGranGIAN OrdeRs iN $\widEhat\oPeratorname{\maThfRak{g}}$ WIth RespeCt to tHe acTiOn of oF $\operaTornaMe{\Mathsf{Aut}}_{{\mathbb{c}}[z]}\biGl(\operatoRnaMe{\MatHfRak{g}}[z]\BIgr)$.
Let $\OpeRatOrname{\mAthfrak{W}}= Z^{-1} \opErATORnAme{\mathfrak{g}}\llbracKeT Z^{-1}\RrBracket$. IT is easY To SeE That $\operAtOrnAme{\mATHfrak{W}}$ is a lAgRangian oRder in $\WIdEhAt\operaToRname{\mAtHfrAk{g}}$. let $\biGL\{\alpHa_l\bigR\}_{l = 1}^{n^2-1}$ be any Basis OF $\operatorname{\mAThfrak{g}}$. Then we HAvE: $
\BEtA_{L, k} := \biGl(\aLpha_l z^k\bigr)^\Vee = \aLPha_l^\Vee z^{-K-1}.
$ thIs iMPlies: $$R_{\operAtORnAMe{\mathfrak{w}}}(x, y) =
\sum\limItS_{k=0}^\inftY x^k \suM\limits_{l = 1}^{n^2-1} \alphA_l \otimes \alPHA_L^\vee y^{-k-1}
= \frAc{c}{y-X},$$ WhERe $c \in \operatornAme{\maThfrak{g}}\otiMEs \operatOrnamE{\mathfraK{g}}$ is the CaSIMir elemeNt. THe tEnsOr-vALUeD function $r_{\opeRATornAmE{\mathfrAk{w}}}$ Is the ceLebRatEd YAng’S sOlution of The classIcAl yaNg–baxTer eqUAtion (\[eq:CyBe\]).
\[ | r = r_{\operatorname{\math frak{w}}}$ .
- L et$\ sigm a$ b e any ${\mathb b {C}} [z]$–linear automorphi sm of $ \ oper a to rname {\mathf r ak { g }}[ z] $and $ \ op erato rna me{\mat hfrak{u}}= \s ig ma(\operator n am e{\mathfra k{w }}) \subset\wi dehat\ op era t ornam e{\ mathf rak{g} } $ be t he transf or m ed ord e r. Then t he sol utions $r_{\opera t or n ame{\mathfrak{ w}}}$an d $ r _ {\o per atorname{\ ma thfra k {u}}}$a re g a uge - equivalent: $ $r_{\operat o rna me{\ma th fra k {u}}}( x, y) = \bi gl(\sigma(x ) \o times \si gma(y) \ bigr) r _ {\opera tornam e{\ mat hfra k {w }} }(x ,y ).$ $
- The describ ed c orres pond e n c e $\o per ator name{ \mathfrak{w}} \ma psto r_{ \oper atorn ame{ \m athfr ak{w}} }$ pr ov ides a bijectio n be tween the ga ug e e qu ivale n ce cla sse s o f ratio nal sol u tio ns o f ( \[eq:CYBE\]) satis fy i n g(\[E:Ans atzSto l in \] ) and the o rbi ts o f Lagra ngia n o rders in $\wid e ha t\ operato rn ame{\m at hfr ak{ g}}$w ithrespec t to the acti o n of of $\oper a torname{\math s f{ A u t} } _{{\ mat hbb{C}}[z]} \big l (\op erat o rn ame { \math frak{ g} } [z ] \bigr)$.
Let $\ope ra tornam e{\ma thfrak{w}}= z ^{-1} \ope r a t orname{\ math f ra k {g}}\llbracket z^{- 1}\rrbrack e t$. It i s eas y to see that $\o p e ratornam e{\ mat hfr ak{ w } }$ is a Lagrang i a n or de r in $\ wid ehat\op era tor nam e{\ ma thfrak{g} }$. Let$\ bi gl \{ \al pha_l \ bigr\}_{ l= 1 }^ {n^ 2-1}$ be any basi s of $ \o p era torname { \m a t hfra k{ g} }$.The nwe ha ve:$
\b eta_{l, k} := \ big l (\al ph a_ l z^k\b igr)^\vee = \ al pha_l^\vee z ^{- k-1}.$ This imp lies: $$r_{\operatornam e {\mathf rak {w}}} (x,y) =
\sum \li mits_{ k=0 } ^\inft y x^k\sum\ li mit s _ {l =1 } ^{ n^2 -1 } \alpha_l \ oti mes \ al pha_ l^\veey^{-k-1}
= \frac{c } {y- x},$$ where $ c \ in \ o p er ato r na m e{\ ma t hfr a k {g}}\otimes \op eratorname {\ m at hfrak{g}}$ isth e Casim ir elem ent.T he tens or-valued function $ r_{\ o p era torname{\m athfrak{ w}}}$ ist he ce l eb rated Ya ng’s s ol uti on of the c l ass icalYang–B ax ter eq uatio n(\[eq:CY BE\]).
\[ | r =_r_{\operatorname{\mathfrak{w}}}$.
- _ Let $\sigma$ be_any ${\mathbb{C}}[z]$–linear_automorphism_of $\operatorname{\mathfrak{g}}[z]$_and_$\operatorname{\mathfrak{u}}= \sigma(\operatorname{\mathfrak{w}}) \subset_\widehat\operatorname{\mathfrak{g}}$ be the_transformed order. Then the_solutions $r_{\operatorname{\mathfrak{w}}}$ and_$r_{\operatorname{\mathfrak{u}}}$_are gauge-equivalent: $$r_{\operatorname{\mathfrak{u}}}(x, y) = \bigl(\sigma(x) \otimes \sigma(y)\bigr) r_{\operatorname{\mathfrak{w}}}(x, y).$$
- The described_correspondence_$\operatorname{\mathfrak{w}}\mapsto r_{\operatorname{\mathfrak{w}}}$_provides_a_bijection between the gauge equivalence_classes of rational solutions of_(\[eq:CYBE\]) satisfying_(\[E:AnsatzStolin\]) and the orbits of Lagrangian orders in_$\widehat\operatorname{\mathfrak{g}}$_with respect to_the action of of $\operatorname{\mathsf{Aut}}_{{\mathbb{C}}[z]}\bigl(\operatorname{\mathfrak{g}}[z]\bigr)$.
Let $\operatorname{\mathfrak{w}}= z^{-1} \operatorname{\mathfrak{g}}\llbracket z^{-1}\rrbracket$._It is easy to see that_$\operatorname{\mathfrak{w}}$ is a_Lagrangian_order_in $\widehat\operatorname{\mathfrak{g}}$. Let $\bigl\{\alpha_l\bigr\}_{l_= 1}^{n^2-1}$ be any basis of_$\operatorname{\mathfrak{g}}$. Then we have: $
\beta_{l, k}_:= \bigl(\alpha_l z^k\bigr)^\vee = \alpha_l^\vee z^{-k-1}.
$_This implies: $$r_{\operatorname{\mathfrak{w}}}(x, y) =
\sum\limits_{k=0}^\infty x^k_\sum\limits_{l = 1}^{n^2-1} \alpha_l \otimes_\alpha_l^\vee y^{-k-1}
=_\frac{c}{y-x},$$ where $c \in \operatorname{\mathfrak{g}}\otimes_\operatorname{\mathfrak{g}}$ is the_Casimir element._The tensor-valued function_$r_{\operatorname{\mathfrak{w}}}$ is the celebrated Yang’s solution_of the classical_Yang–Baxter equation (\[eq:CYBE\]).
\[ |
Frontier Research Center funded by the U.S. DOE, Office of Basic Energy Sciences (A.W and C.P) but also in part by the U.S. DOE under Contract No. DEAC02-98CH10886. This work has also been supported by Grant No. 171001 from the Serbian Ministry of Education and Science. Jianjun Tian was in part supported by a scholarship of Faculty Training Abroad Program of Henan University.
Y. Kamihara, T. Watanabe, M. Hirano, and H. Hosono, J. Am. Chem. Soc. **130**, 3296 (2008). Q. Y. Wang, Z. Li, W. H. Zhang, Z. C. Zhang, J. S. Zhang, W. Li, H. Ding, Y. B. Ou, P. Deng, K.Chang, J. Wen, C. L. Song, K. He, J. F. Jia, S. H. Ji, Y. Y.Wang, L. L. Wang, X. Chen,X. C. Ma and Q. K. Xue, Chin. Phys. Lett. **29**, 037402 (2012). Shaolog He, Junfeng He,Wenhao Zhang, Lin Zhao, Defa Liu, Xu Liu, Daixiang Mou, Yun-Bo Ou, Qing-YanWang, Zhi Li, LiliWang, Yingying Peng, Yan Liu, Chaoyu Chen, Li Yu, Guodong Liu, Xiaoli Dong, Jun Zhang, Chuangtian Chen, Zuyan Xu, Xi Chen, Xucun Ma, Qikun Xue\* and X. J. Zhou, Nature Materials **12**, 605 (2013). Jian-Feng Ge, Zhi-Long Liu, Canhua Liu, Chun-Lei Gao, Dong Qian, Qi-Kun Xue, Ying Liu and Jin-Feng Jia, Nature Mater. **14**, 285 (2015). J. Shiogai, Y. Ito, T. Mitsuhashi, T. Nojima and A. Tsukazaki, Nature Phys. **12**, 42 (2016). Xiaofang Lai, Hui Zhang, Yingqi Wang, Xin Wang, Xian Zhang, Jianhua Lin and Fuqiang Huang, | Frontier Research Center funded by the U.S. DOE, Office of Basic Energy Sciences (A.W and C.P) but also in part by the U.S. DOE under Contract No. DEAC02 - 98CH10886. This work has besides been digest by Grant No. 171001 from the Serbian Ministry of Education and Science. Jianjun Tian was in part supported by a eruditeness of Faculty Training Abroad Program of Henan University.
Y. Kamihara, T. Watanabe, M. Hirano, and H. Hosono, J. Am. Chem. Soc. * * 130 * *, 3296 (2008). Q. Y. Wang, Z. Li, W. H. Zhang, Z. C. Zhang, J. S. Zhang, W. Li, H. Ding, Y. B. Ou, P. Deng, K.Chang, J. Wen, C. L. Song, K. He, J. F. Jia, S. H. Ji, Y. Y.Wang, L. L. Wang, X. Chen, X. C. Ma and Q. K. Xue, Chin. Phys. Lett. * * 29 * *, 037402 (2012). Shaolog He, Junfeng He, Wenhao Zhang, Lin Zhao, Defa Liu, Xu Liu, Daixiang Mou, Yun - Bo Ou, Qing - YanWang, Zhi Li, LiliWang, Yingying Peng, Yan Liu, Chaoyu Chen, Li Yu, Guodong Liu, Xiaoli Dong, Jun Zhang, Chuangtian Chen, Zuyan Xu, Xi Chen, Xucun Ma, Qikun Xue\ * and X. J. Zhou, Nature Materials * * 12 * *, 605 (2013). Jian - Feng Ge, Zhi - Long Liu, Canhua Liu, Chun - Lei Gao, Dong Qian, Qi - Kun Xue, Ying Liu and Jin - Feng Jia, Nature Mater. * * 14 * *, 285 (2015). J. Shiogai, Y. Ito, T. Mitsuhashi, T. Nojima and A. Tsukazaki, Nature Phys. * * 12 * *, 42 (2016). Xiaofang Lai, Hui Zhang, Yingqi Wang, Xin Wang, Xian Zhang, Jianhua Lin and Fuqiang Huang, | Frlntier Research Center fmnded by the U.S. BIE, Offmce of Gasic Endrgy Sciences (A.W and C.P) but elso in pqrt by the U.S. DOE undef Contracn No. DEAC02-98XH10886. Tiis work has also been sm'portsf by Jrant No. 171001 from jhe Serbian Kinistry of Edgcxtnon and Science. Jianjun Tian was in [art sulplrted by a schjlarxrip kf Faculty Training Abroad Program of Henen University.
Y. Lamihara, T. Watanabe, M. Hirajo, ajd H. Hosono, J. Am. Cjem. Soc. **130**, 3296 (2008). W. Y. Rqng, Z. Li, W. H. Zhang, Z. C. Zhang, J. S. Zgang, W. Li, H. Ding, Y. B. Ou, P. Deng, Y.Chany, J. Wen, C. L. Sing, N. He, J. F. Jie, S. H. Ti, Y. Y.Wang, L. L. Wang, F. Chen,X. C. Ma and Q. K. Wue, Ciin. Phys. Lett. **29**, 037402 (2012). Shaolog Ie, Junfeng He,Wenhao Ehang, Lin Shco, Defa Liu, Xu Liu, Dauxuang Kou, Fun-Bu Ou, Qihg-BanSang, Zji Mi, LiliWanf, Yingying Peng, Yan Liu, Chaoyu Crvm, Li Yu, Guodkng Lit, Viaoli Dong, Jun Zhang, Chuangtian Chen, Zlyan Xu, Xi Chen, Xucun Ma, Qijun Xue\* and X. J. Zhou, Jature Maeerials **12**, 605 (2013). Jian-Feng Ge, Zhi-Long Liu, Canhua Liu, Chut-Lei Jau, Dinn Qixb, Ei-Kun Xue, Ying Liu and Jin-Feng Jia, Nature Matqd. **14**, 285 (2015). J. Shiogai, Y. Ibo, T. Mitsuhashi, T. Mouikw and A. Tsukaeaki, Nachrs Phys. **12**, 42 (2016). Xiaofang Lai, Huy Zhabg, Yingqi Wanb, Xin Wang, Xian Zhang, Jianhya Lin and Flqiabg Huang, | Frontier Research Center funded by the U.S. of Energy Sciences and C.P) but U.S. under Contract No. This work has been supported by Grant No. 171001 the Serbian Ministry of Education and Science. Jianjun Tian was in part supported a scholarship of Faculty Training Abroad Program of Henan University. Y. Kamihara, T. M. and Hosono, Am. Chem. Soc. **130**, 3296 (2008). Q. Y. Wang, Z. Li, W. H. Zhang, Z. C. J. S. Zhang, W. Li, H. Ding, Y. Ou, P. Deng, K.Chang, Wen, C. L. Song, K. J. Jia, S. Ji, Y.Wang, L. Wang, X. C. Ma and Q. K. Xue, Chin. Phys. Lett. **29**, 037402 (2012). Shaolog He, Junfeng He,Wenhao Zhang, Zhao, Defa Liu, Daixiang Yun-Bo Qing-YanWang, Li, LiliWang, Yingying Liu, Chaoyu Chen, Li Yu, Guodong Jun Zhang, Chuangtian Chen, Zuyan Xu, Xi Chen, Ma, Qikun and X. J. Zhou, Nature Materials 605 (2013). Jian-Feng Ge, Zhi-Long Liu, Canhua Liu, Gao, Dong Qian, Qi-Kun Xue, Ying Liu and Jin-Feng Jia, Nature Mater. **14**, 285 (2015). Y. Ito, T. Mitsuhashi, Nojima and A. Nature **12**, (2016). Lai, Hui Yingqi Wang, Xin Wang, Xian Zhang, Jianhua Lin and Fuqiang Huang, | Frontier Research Center funDed by the U.S. dOE, OfFicE of baSic ENergY Sciences (A.W and c.p) but Also in part by the U.S. DOE unDer CoNtRAct NO. dEaC02-98CH10886. THis work HAs ALSo bEeN sUppOrTEd By GraNt NO. 171001 from thE Serbian MiNisTrY of Education ANd science. JiaNjuN Tian was in paRt sUpportEd By a SCholaRshIp of FAculty tRaininG Abroad PrOgRAm of HeNAn UniveRSItY.
Y. KaMihara, T. Watanabe, M. HIRaNO, and H. Hosono, J. Am. chem. SoC. **130**, 3296 (2008). Q. y. waNG, z. Li, w. H. ZHang, Z. C. ZhanG, J. s. ZhanG, w. Li, H. DinG, y. B. oU, p. denG, k.Chang, J. Wen, C. L. SOng, K. He, J. F. Jia, s. h. Ji, y. Y.Wang, l. L. wanG, x. Chen,X. c. Ma anD Q. k. xue, chin. Phys. LetT. **29**, 037402 (2012). ShaOlog He, JunFeng He,wEnhao ZhANg, Lin ZhAo, Defa liu, xu LIu, DaIXiAnG MoU, YUN-Bo oU, QIng-yAnWAng, Zhi Li, liLiwang, YIngyING pEng, YAn LIu, ChAoyu CHen, Li Yu, GuodonG LiU, XiaOLi DOng, JuN ZhanG, ChuAnGtian chen, ZuYan Xu, xi chen, Xucun Ma, QikuN Xue\* And X. J. Zhou, natUrE MaTeRials **12**, 605 (2013). jIan-FenG Ge, zhi-long Liu, canhua LIU, ChUn-lEI gaO, Dong Qian, Qi-Kun Xue, YInG lIu And Jin-FeNg Jia, NATuRe mAter. **14**, 285 (2015). J. ShiOgAi, Y. ito, T. mITsuhaShi, T. nOjIma and A. TSukazaKI, NAtUre Phys. **12**, 42 (2016). xiAofang laI, HuI ZhAng, YiNGqi WAng, Xin wang, Xian zhang, jIanhua Lin and FuQIang Huang, | Frontier Research Centerfunded bythe U .S. DO E, Off iceof Basic Energ y Sci ences (A.W and C.P) bu t als oi n pa r tby th e U.S.D OE u nde rCo ntr ac t N o. DE AC0 2-98CH1 0886. This wo rk has also be e nsupportedbyGrant No. 17 100 1 from t heS erbia n M inist ry ofE ducati on and Sc ie n ce. Ji a njun Ti a n w as i n part supportedb ya scholarship o f Facu lt y T r a ini ngAbroad Pro gr am of Henan U n iv e r s ity .
Y. Kamihara , T. Watana b e,M. Hir an o,a nd H.Hoson o, J.Am. Chem. S oc.**130**,3296 ( 2 008). Q . Y. Wan g, Z.Li, W. H.Z ha ng , Z .C . Z h an g,J . S . Zhang, W .Li, H . Di n g , Y. B . O u, P . Den g, K.Chang, J . W en,C . L . Son g, K. He, J . F.Jia, S . H.Ji , Y. Y.Wang, L. L.Wang, X.Che n, X.C. Ma a n d Q. K . X ue, Chin.Phys. L e tt. * * 2 9 ** , 037402 (2012). S ha o l og He, Jun feng H e ,W en h ao Zhang ,Lin Zha o , Defa Liu , X u Liu, D aixian g M ou , Yun-B oOu, Qi ng -Ya nWa ng, Z h i Li , Lili Wang, Yi ngyin g Peng, Yan Liu , Chaoyu Chen, Li Y u, Guod ong Liu, Xiaol i Do n g, J un Z h an g,C huang tianCh e n, Zuyan Xu, Xi Chen,Xu cun Ma , Qik un Xue\* andX. J. Zhou , N ature Ma teri a ls **12**, 605 (2 013). Jian-Feng Ge, Zhi- LongLiu, Can hua Liu,C h un-Lei G ao, Do ngQia n , Q i-Kun Xue, Yi n g Liu a nd Jin- Fen g Jia,Nat ure Ma ter .**14**, 2 85 (2015 ). J .Sh iog ai, Y . Ito, T. M its uh ash i, T. Nojima andA. T su ka z aki , Natur e P h y s. * *1 2* *, 4 2 ( 20 16).Xiao f ang Lai, H ui Zhang, Yi n gqiWa ng , Xin W ang, Xian Zha ng , JianhuaLi n a nd Fuq i a ng Huang , | Frontier_Research Center_funded by the U.S._DOE, Office_of_Basic Energy_Sciences_(A.W and C.P)_but also in_part by the U.S._DOE under Contract_No._DEAC02-98CH10886. This work has also been supported by Grant No. 171001 from the Serbian_Ministry_of Education_and_Science._Jianjun Tian was in part_supported by a scholarship of_Faculty Training_Abroad Program of Henan University.
Y. Kamihara, T. Watanabe,_M._Hirano, and H._Hosono, J. Am. Chem. Soc. **130**, 3296 (2008). Q._Y. Wang, Z. Li, W. H._Zhang, Z. C._Zhang,_J._S. Zhang, W. Li,_H. Ding, Y. B. Ou, P._Deng, K.Chang, J. Wen, C. L._Song, K. He, J. F. Jia, S._H. Ji, Y. Y.Wang, L. L._Wang, X. Chen,X. C. Ma_and Q._K. Xue, Chin. Phys. Lett._**29**, 037402 (2012)._Shaolog He,_Junfeng He,Wenhao Zhang,_Lin Zhao, Defa Liu, Xu Liu,_Daixiang Mou, Yun-Bo_Ou, Qing-YanWang, Zhi Li, LiliWang, Yingying_Peng,_Yan Liu, Chaoyu_Chen,_Li_Yu, Guodong_Liu, Xiaoli Dong,_Jun_Zhang, Chuangtian_Chen,_Zuyan Xu, Xi Chen, Xucun Ma,_Qikun_Xue\* and X. J. Zhou, Nature Materials_**12**, 605 (2013). Jian-Feng_Ge,_Zhi-Long Liu, Canhua Liu,_Chun-Lei Gao, Dong Qian, Qi-Kun_Xue, Ying Liu and Jin-Feng Jia,_Nature Mater._**14**, 285_(2015). J. Shiogai, Y. Ito, T. Mitsuhashi, T. Nojima and A._Tsukazaki, Nature Phys. **12**, 42 (2016)._Xiaofang Lai, Hui Zhang,_Yingqi Wang,_Xin_Wang, Xian Zhang,_Jianhua_Lin and_Fuqiang Huang, |
hole and also on the inner horizon for the nonextremal black hole. Furthermore, we analyze the dependence of $E_{CM}$ on charge $Q$, and explicitly show the effect of $Q$ on the $E_{CM}$. Thus, we have estimated the $E_{CM}$ of two unequal mass particles for both the extremal and the nonextremal ABG black holes when one particle has the critical angular momentum. Furthermore, we analyze the dependence of $E_{CM}$ on charge $Q$.
S.G.G. would like to thank SERB-DST for Research Project Grant NO SB/S2/HEP-008/2014 and DST INDO-SA bilateral project DST/INT/South Africa/P-06/2016. M.A. thanks the University of KwaZulu-Natal and the National Research Foundation for financial support.
[99]{} M. Banados, J. Silk and S. M. West, Phys. Rev. Lett. [**103**]{}, 111102 (2009).
R. P. Kerr, Phys. Rev. Lett. [**11**]{}, 237 (1963).
E. Berti, V. Cardoso, L. Gualtieri, F. Pretorius and U. Sperhake, Phys. Rev. Lett. [**103**]{}, 239001 (2009).
T. Jacobson and T. P. Sotiriou, Phys. Rev. Lett. [**104**]{}, 021101 (2010).
K. Lake, Phys. Rev. Lett. [**104**]{}, 211102 (2010).
S.W. Wei, Y.X. Liu, H.T. Li and F.W. Chen, J. High Energy Phys. [**12**]{}, 066 (2010).
P. J. Mao, R. Li, L. Y. Jia and J. R. Ren, Chin. Phys. C [**41**]{}, No. 6, 065101 (2017).
A. A. Grib and Y. .V. Pavlov, Astropart. Phys. [**34**]{}, 581 (2011).
C. Liu | hole and also on the inner horizon for the nonextremal black fix. Furthermore, we analyze the addiction of $ E_{CM}$ on charge $ Q$, and explicitly show the consequence of $ Q$ on the $ E_{CM}$. therefore, we have estimated the $ E_{CM}$ of two unequal mass atom for both the extremal and the nonextremal ABG black holes when one atom have the critical angular momentum. Furthermore, we analyze the addiction of $ E_{CM}$ on charge $ Q$.
S.G.G. would like to thank SERB - DST for Research Project Grant NO SB / S2 / HEP-008/2014 and DST INDO - SA bilateral undertaking DST / INT / South Africa / P-06/2016. M.A. thanks the University of KwaZulu - Natal and the National Research Foundation for financial support.
[ 99 ] { } M. Banados, J. Silk and S. M. West, Phys. Rev. Lett. [ * * 103 * * ] { }, 111102 (2009).
R. P. Kerr, Phys. Rev. Lett. [ * * 11 * * ] { }, 237 (1963).
E. Berti, V. Cardoso, L. Gualtieri, F. Pretorius and U. Sperhake, Phys. Rev. Lett. [ * * 103 * * ] { }, 239001 (2009).
T. Jacobson and T. P. Sotiriou, Phys. Rev. Lett. [ * * 104 * * ] { }, 021101 (2010).
K. Lake, Phys. Rev. Lett. [ * * 104 * * ] { }, 211102 (2010).
S.W. Wei, Y.X. Liu, H.T. Li and F.W. Chen, J. High Energy Phys. [ * * 12 * * ] { }, 066 (2010).
P. J. Mao, R. Li, L. Y. Jia and J. R. Ren, Chin. Phys. C [ * * 41 * * ] { }, No. 6, 065101 (2017).
A. A. Grib and Y. .V. Pavlov, Astropart. Phys. [ * * 34 * * ] { }, 581 (2011).
C. Liu | hope and also on the inner horizon for thg bonextcemal bmack hold. Furthermore, we analyze the dwpendtuce of $E_{CM}$ on charge $Q$, and exilicitly whow rhe effect of $Q$ on the $E_{DL}$. Thbs, we have estimsted the $E_{WM}$ of two uneqgau lass particles for both the extremaj and tne nonextremal AFG bkwck goles when one particle has the crjtical engular momentuk. Furthermore, we analyze tje dfpendence of $E_{CM}$ oj charge $Q$.
S.T.G. wjyld like to ghank SERB-DST for Resezrch Project Grant NO SB/S2/HEP-008/2014 ana DST INDO-SA bioarerwn project DWT/INT/Fouth Africa/I-06/2016. M.A. thdnks thr University on KwaVulu-Batal and the Nationan Research Foundatyon for fhncncial support.
[99]{} M. Banadis, J. Siln ang S. M. Dwst, Phgs. Cev. Mett. [**103**]{}, 111102 (2009).
R. P. Ierr, Phys. Rsv. Lett. [**11**]{}, 237 (1963).
E. Bwrti, V. Cardoso, L. Gualuiewp, F. Pretorius znd U. S[ewhake, Phys. Rev. Lett. [**103**]{}, 239001 (2009).
T. Jacobson and T. P. Sotprioh, Phys. Rev. Lett. [**104**]{}, 021101 (2010).
K. Lake, Pyys. Rev. Lett. [**104**]{}, 211102 (2010).
S.W. Wei, Y.X. Piu, H.T. Li wnd F.W. Chen, J. High Energy Phys. [**12**]{}, 066 (2010).
P. J. Mao, R. Li, L. Y. Jia atd J. R. Cev, Cklk. Phyr. X [**41**]{}, No. 6, 065101 (2017).
A. A. Grib and Y. .V. Pavlov, Astropart. Phys. [**34**]{}, 581 (2011).
C. Lit | hole and also on the inner horizon nonextremal hole. Furthermore, analyze the dependence and show the effect $Q$ on the Thus, we have estimated the $E_{CM}$ two unequal mass particles for both the extremal and the nonextremal ABG black when one particle has the critical angular momentum. Furthermore, we analyze the dependence $E_{CM}$ charge S.G.G. like to thank SERB-DST for Research Project Grant NO SB/S2/HEP-008/2014 and DST INDO-SA bilateral project DST/INT/South M.A. thanks the University of KwaZulu-Natal and the Research Foundation for financial [99]{} M. Banados, J. Silk S. West, Phys. Lett. 111102 R. P. Kerr, Rev. Lett. [**11**]{}, 237 (1963). E. Berti, V. Cardoso, L. Gualtieri, F. Pretorius and U. Sperhake, Phys. Lett. [**103**]{}, T. Jacobson T. Sotiriou, Rev. Lett. [**104**]{}, K. Lake, Phys. Rev. Lett. [**104**]{}, Wei, Y.X. Liu, H.T. Li and F.W. Chen, High Energy [**12**]{}, 066 (2010). P. J. Mao, Li, L. Y. Jia and J. R. Ren, Phys. C [**41**]{}, No. 6, 065101 (2017). A. A. Grib and Y. .V. Pavlov, Astropart. 581 (2011). C. Liu | hole and also on the inner horiZon for the nOnextRemAl bLaCk hoLe. FuRthermore, we anaLYze tHe dependence of $E_{CM}$ on chaRge $Q$, aNd EXpliCItLy shoW the effECt OF $q$ on ThE $E_{cM}$. THuS, We Have eStiMated thE $E_{CM}$ of two uNeqUaL mass particlES fOr both the eXtrEmal and the noNexTremal aBg blACk holEs wHen onE partiCLe has tHe criticaL aNGular mOMentum. FURThErmoRe, we analyze the depENdENce of $E_{CM}$ on charGe $Q$.
S.G.G. WoULd LIKe tO thAnk SERB-DST FoR ReseARch ProjECt gRANt No sB/S2/HEP-008/2014 and DST InDO-SA bilateRAl pRoject dSt/INt/south AFrica/p-06/2016. M.a. ThaNks the UniveRsitY of KwaZulU-Natal ANd the NaTIonal ReSearch fouNdaTion FOr FiNanCiAL suPPoRt.
[99]{} M. bAnaDos, J. Silk AnD S. m. West, phys. rEV. lEtt. [**103**]{}, 111102 (2009).
R. p. KeRr, PhYs. Rev. lett. [**11**]{}, 237 (1963).
E. Berti, V. CaRdoSo, L. GUAltIeri, F. pretoRius AnD U. SpeRhake, PHys. ReV. LEtt. [**103**]{}, 239001 (2009).
T. Jacobson and t. P. SoTiriou, PhyS. ReV. LEtt. [**104**]{}, 021101 (2010).
k. LAke, PhYS. Rev. LeTt. [**104**]{}, 211102 (2010).
S.w. WeI, Y.X. Liu, H.t. Li and F.w. cheN, J. hIGH ENergy Phys. [**12**]{}, 066 (2010).
P. J. Mao, R. Li, L. y. JIA AnD J. R. Ren, ChIn. Phys. c [**41**]{}, no. 6, 065101 (2017).
a. A. gRib and Y. .V. paVloV, AstROPart. PHys. [**34**]{}, 581 (2011).
C. lIu | hole and also on the inne r horizonfor t henon ex trem al b lack hole. Fur t herm ore, we analyze the de pende nc e of$ E_ {CM}$ on cha r ge $ Q$, a nd ex pl i ci tly s how the ef fect of $Q $ o nthe $E_{CM}$ . T hus, we ha veestimated th e $ E_{CM} $oft wo un equ al ma ss par t iclesfor bothth e extre m al andt h enone xtremal ABG black ho l es when one pa rticle h a st h e c rit ical angul ar mome n tum. Fu r th e r m ore , we analyze t he dependen c e o f $E_{ CM }$o n char ge $Q $.
S. G.G. wouldlike to thank SERB- D ST forR esearch Proje ctGra nt N O S B/ S2/ HE P -00 8 /2 014 and DST IND O- SA bila tera l p r ojec t D ST/I NT/So uth Africa/P- 06/ 2016 . M. A. th ankstheUn ivers ity of KwaZ ul u-Natal and the Nat ional Res ear ch Fo un datio n for f ina nci al supp ort.
[ 9 9]{ }M . Ba nados, J. Silk and S . M. West, P hys. R e v. L e tt. [** 10 3** ]{}, 1 11102 (20 0 9) .
R. P. Kerr, Ph ys . Rev.Le tt. [ ** 11* *]{ }, 23 7 (19 63).
E. Berti , V.C ardoso, L. Gua l tieri, F. Pre t or i u sa nd U . S perhake, Ph ys.R ev.Lett . [** 1 03**] {}, 2 39 0 01 (2009).
T. Jacobso nand T. P. S otiriou, Phys . Rev. Let t . [**104* *]{} , 0 2 1101 (2010).
K. La ke, Phys.R ev. Lett . [* *104**]{ }, 211102 ( 2010).
S.W . W ei, Y. X . L iu, H.T. Li a n d F.W .Chen, J . H igh Ene rgy Ph ys. [* *1 2**]{}, 0 66 (2010 ).
P. J . M ao, R . Li, L.Y. Ji aand J. R . Ren,Chin. Phy s. C [** 41**]{} , N o . 6,06 51 01 ( 201 7) .
A. A.G rib and Y. .V. Pavl ov, Astr op ar t. Phys . [**34**]{} ,581 (2011) .
C. Liu | hole_and also_on the inner horizon_for the_nonextremal_black hole._Furthermore,_we analyze the_dependence of $E_{CM}$_on charge $Q$, and_explicitly show the_effect_of $Q$ on the $E_{CM}$. Thus, we have estimated the $E_{CM}$ of two unequal_mass_particles for_both_the_extremal and the nonextremal ABG_black holes when one particle_has the_critical angular momentum. Furthermore, we analyze the dependence_of_$E_{CM}$ on charge_$Q$.
S.G.G. would like to thank SERB-DST for Research Project_Grant NO SB/S2/HEP-008/2014 and DST INDO-SA_bilateral project DST/INT/South_Africa/P-06/2016._M.A._thanks the University of_KwaZulu-Natal and the National Research Foundation_for financial support.
[99]{} M. Banados, J. Silk and_S. M. West, Phys. Rev. Lett. [**103**]{}, 111102 (2009).
R. P. Kerr, Phys. Rev. Lett. [**11**]{},_237 (1963).
E. Berti, V. Cardoso, L. Gualtieri, F. Pretorius and_U. Sperhake, Phys. Rev. Lett. [**103**]{}, 239001 (2009).
T. Jacobson_and T. P. Sotiriou,_Phys. Rev. Lett. [**104**]{}, 021101 (2010).
K. Lake, Phys. Rev. Lett. _[**104**]{}, 211102 (2010).
S.W. Wei,_Y.X. Liu, H.T. Li_and F.W. Chen, J._High Energy Phys. [**12**]{}, 066 (2010).
P. J. Mao,_R. Li, L. Y. Jia and_J. R. Ren, Chin. Phys. C [**41**]{}, No. 6, 065101_(2017).
A. A. Grib_and Y. .V. Pavlov, Astropart. Phys. _[**34**]{},_581_(2011).
C. Liu |
from[@Po Theorem 3.4]:
\[HardWork\] For any homomorphism of $2$-(semi)categories $\mathbf{M}: {{\sc\mbox{C}\hspace{1.0pt}}}\rightarrow \mathbf{Cat}$, there is a $2$-(semi)functor $\widehat{\mathbf{M}}: {{\sc\mbox{C}\hspace{1.0pt}}}\rightarrow \mathbf{Cat}$ together with a non-strict $2$-natural isomorphism $\Psi_{\mathbf{M}}: \mathbf{M} \rightarrow \widehat{\mathbf{M}}$.
In particular this implies that the problem of classifying simple transitive $2$-representations of a $2$-(semi)category ${{\sc\mbox{C}\hspace{1.0pt}}}$ is equivalent to that of classifying weak simple transitive $2$-representations ${{\sc\mbox{C}\hspace{1.0pt}}}$:
\[SemiWork\] Let ${{\sc\mbox{C}\hspace{1.0pt}}}$ be a finitary $2$-(semi)category and let $\mathbf{M}$ be a weak $2$-representation of ${{\sc\mbox{C}\hspace{1.0pt}}}$. There is a (strict) $2$-representation $\widehat{\mathbf{M}}$ of ${{\sc\mbox{C}\hspace{1.0pt}}}$ equivalent to $\mathbf{M}$.
$2$-semicategories of projective functors
-----------------------------------------
Let $A$ be a be a finite-dimensional, basic, connected algebra and let ${\left\{ e_{1},\ldots, e_{n} \right\}}$ be a complete set of idempotents of $A$. Fix a small category $\mathcal{A}$ equivalent to $A\!\on{-proj}$. The $2$-semicategory ${\sc\mbox{Z}\hspace{1.0pt}}_{\!A}$ is defined as follows:
- $\on{Ob}{\sc\mbox{Z}\hspace{1.0pt}}_{\!A} = {\left\{ \mathtt{i} \right\}}$, where $\mathtt{i}$ can be identified with $\mathcal{A}$;
- $1$-morphisms of ${\sc\mbox{Z}\hspace{1.0pt}}_{\!A}$ are endofunctors of $\mathcal{A}$ isomorphic to tensoring with the projective $A$-$A | from[@Po Theorem 3.4 ]:
\[HardWork\ ] For any homomorphism of $ 2$-(semi)categories $ \mathbf{M }: { { \sc\mbox{C}\hspace{1.0pt}}}\rightarrow \mathbf{Cat}$, there is a $ 2$-(semi)functor $ \widehat{\mathbf{M } }: { { \sc\mbox{C}\hspace{1.0pt}}}\rightarrow \mathbf{Cat}$ together with a non - strict $ 2$-natural isomorphism $ \Psi_{\mathbf{M } }: \mathbf{M } \rightarrow \widehat{\mathbf{M}}$.
In particular this entail that the trouble of classifying simple transitive $ 2$-representations of a $ 2$-(semi)category $ { { \sc\mbox{C}\hspace{1.0pt}}}$ is equivalent to that of relegate unaccented simple transitive $ 2$-representations $ { { \sc\mbox{C}\hspace{1.0pt}}}$:
\[SemiWork\ ] Let $ { { \sc\mbox{C}\hspace{1.0pt}}}$ be a finitary $ 2$-(semi)category and permit $ \mathbf{M}$ be a weak $ 2$-representation of $ { { \sc\mbox{C}\hspace{1.0pt}}}$. There is a (rigorous) $ 2$-representation $ \widehat{\mathbf{M}}$ of $ { { \sc\mbox{C}\hspace{1.0pt}}}$ equivalent to $ \mathbf{M}$.
$ 2$-semicategories of projective functors
-----------------------------------------
Let $ A$ be a be a finite - dimensional, basic, connected algebra and lease $ { \left\ { e_{1},\ldots, e_{n } \right\}}$ be a complete set of idempotents of $ A$. sterilize a small category $ \mathcal{A}$ equivalent to $ A\!\on{-proj}$. The $ 2$-semicategory $ { \sc\mbox{Z}\hspace{1.0pt}}_{\!A}$ is defined as follows:
- $ \on{Ob}{\sc\mbox{Z}\hspace{1.0pt}}_{\!A } = { \left\ { \mathtt{i } \right\}}$, where $ \mathtt{i}$ can be identified with $ \mathcal{A}$;
- $ 1$-morphisms of $ { \sc\mbox{Z}\hspace{1.0pt}}_{\!A}$ are endofunctors of $ \mathcal{A}$ isomorphous to tensoring with the projective $ A$ -$A | frlm[@Po Theorem 3.4]:
\[HardWork\] Fov any homomorphism of $2$-(sxmi)catefories $\mxthbf{M}: {{\sc\mbox{C}\hspace{1.0pt}}}\rightacrow \mathvf{Cat}$, there is a $2$-(semi)fjnctor $\wifehat{\matybf{M}}: {{\wc\mbox{C}\hspede{1.0pt}}}\rigmcarros \matkbh{Cat}$ together wlth a non-stsict $2$-natural ivoooxphism $\Psi_{\mathbf{M}}: \mathbf{M} \rightarrow \widehay{\mwthbf{M}}$.
In partisulag ehis pmklies that the problem of classifging sikple transitife $2$-representations of a $2$-(seli)cahegory ${{\sc\mbox{C}\hspafe{1.0pt}}}$ is equuvalqbt to that ow classifypug weak simkle transitive $2$-representations ${{\sz\mbox{E}\hspace{1.0pt}}}$:
\[SeniQorn\] Let ${{\sc\mbox{R}\hspacv{1.0pt}}}$ be a finibsry $2$-(seki)categpry and let $\mabhbf{M}$ be a weak $2$-representatioi of ${{\sc\mbox{C}\hspace{1.0pt}}}$. There is a (strict) $2$-representqtuon $\whdehdt{\magybf{O}}$ or ${{\xc\jbox{C}\hdpare{1.0pt}}}$ equivament to $\matybf{M}$.
$2$-semicategories pf itojective fuhctors
-----------------------------------------
Jee $A$ be a be a finite-dimensional, basic, cmnnscted algebra and let ${\lwft\{ e_{1},\ldots, e_{n} \right\}}$ bg a compleee set of idempotents of $A$. Fix a small category $\mdthcam{X}$ ewulvaldbt to $A\!\on{-proj}$. The $2$-semicategory ${\sc\mbox{Z}\hspace{1.0pt}}_{\!A}$ js dvfined as follows:
- $\on{Ob}{\sc\mbpx{X}\hxkace{1.0pt}}_{\!A} = {\left\{ \oathtt{n} \rjght\}}$, where $\mathtt{i}$ can be idenrified wiuh $\mayhcal{A}$;
- $1$-morphisms of ${\sc\mboz{Z}\hspace{1.0pt}}_{\!A}$ cre endofunctors of $\mcthcal{A}$ isomurphoc to tensoring with the prolectivs $A$-$A | from[@Po Theorem 3.4]: \[HardWork\] For any homomorphism $\mathbf{M}: \mathbf{Cat}$, there a $2$-(semi)functor $\widehat{\mathbf{M}}: non-strict isomorphism $\Psi_{\mathbf{M}}: \mathbf{M} \widehat{\mathbf{M}}$. In particular implies that the problem of classifying transitive $2$-representations of a $2$-(semi)category ${{\sc\mbox{C}\hspace{1.0pt}}}$ is equivalent to that of classifying weak transitive $2$-representations ${{\sc\mbox{C}\hspace{1.0pt}}}$: \[SemiWork\] Let ${{\sc\mbox{C}\hspace{1.0pt}}}$ be a finitary $2$-(semi)category and let $\mathbf{M}$ a $2$-representation ${{\sc\mbox{C}\hspace{1.0pt}}}$. is a (strict) $2$-representation $\widehat{\mathbf{M}}$ of ${{\sc\mbox{C}\hspace{1.0pt}}}$ equivalent to $\mathbf{M}$. $2$-semicategories of projective functors ----------------------------------------- Let be a be a finite-dimensional, basic, connected algebra let ${\left\{ e_{1},\ldots, e_{n} be a complete set of of Fix a category equivalent $A\!\on{-proj}$. The $2$-semicategory is defined as follows: - $\on{Ob}{\sc\mbox{Z}\hspace{1.0pt}}_{\!A} = {\left\{ \mathtt{i} \right\}}$, where $\mathtt{i}$ can be identified with $\mathcal{A}$; $1$-morphisms of endofunctors of isomorphic tensoring the projective $A$-$A | from[@Po Theorem 3.4]:
\[HardWork\] For aNy homomorpHism oF $2$-(seMi)cAtEgorIes $\mAthbf{M}: {{\sc\mbox{C}\hSPace{1.0Pt}}}\rightarrow \mathbf{Cat}$, tHere iS a $2$-(SEmi)fUNcTor $\wiDehat{\maTHbF{m}}: {{\Sc\mBoX{C}\HspAcE{1.0Pt}}}\RightArrOw \mathbF{Cat}$ togethEr wItH a non-strict $2$-nATuRal isomorpHisM $\Psi_{\mathbf{M}}: \mAthBf{M} \rigHtArrOW \wideHat{\MathbF{M}}$.
In paRTiculaR this implIeS That thE Problem OF ClAssiFying simple transiTIvE $2$-RepresentationS of a $2$-(seMi)CAtEGOry ${{\Sc\mBox{C}\hspace{1.0Pt}}}$ Is equIValent tO ThAT OF clASsifying weak sImple transiTIve $2$-RepresEnTatIOns ${{\sc\mBox{C}\hSpACe{1.0pT}}}$:
\[SemiWork\] LeT ${{\sc\mBox{C}\hspacE{1.0pt}}}$ be a FInitary $2$-(SEmi)cateGory anD leT $\maThbf{m}$ Be A wEak $2$-RePResENtAtiON of ${{\Sc\mbox{C}\hSpAcE{1.0pt}}}$. ThEre iS A (STRict) $2$-RepReseNtatiOn $\widehat{\mathBf{M}}$ Of ${{\sc\MBox{c}\hspaCe{1.0pt}}}$ eQuivAlEnt to $\Mathbf{m}$.
$2$-semiCaTegories of projeCtivE functors
-----------------------------------------
let $a$ bE a bE a FinitE-DimensIonAl, bAsic, conNected aLGebRa AND LeT ${\left\{ e_{1},\ldots, e_{n} \right\}}$ Be A COmPlete set Of idemPOtEnTS of $A$. Fix a SmAll CateGORy $\matHcal{a}$ EqUivalent To $A\!\on{-pROj}$. thE $2$-semicaTeGory ${\sc\MbOx{Z}\HspAce{1.0pt}}_{\!a}$ Is deFined aS follows:
- $\On{Ob}{\sC\Mbox{Z}\hspace{1.0pt}}_{\!A} = {\LEft\{ \mathtt{i} \rigHT\}}$, wHERe $\MAthtT{i}$ cAn be identifIed wITh $\maThcaL{a}$;
- $1$-mOrpHIsms oF ${\sc\mbOx{z}\HsPAce{1.0pt}}_{\!A}$ are endofunctoRs Of $\mathCal{A}$ iSomorphic to teNsoring witH THE projectIve $A$-$a | from[@Po Theorem 3.4]:
\ [HardWork\ ] For an y h om omor phis m of $2$-(semi ) cate gories $\mathbf{M}: {{ \sc\m bo x {C}\ h sp ace{1 .0pt}}} \ ri g h tar ro w\ma th b f{ Cat}$ , t here is a $2$-(se mi) fu nctor $\wide h at {\mathbf{M }}: {{\sc\mbox{ C}\ hspace {1 .0p t }}}\r igh tarro w \mat h bf{Cat }$ togeth er with a non-str i c t$2$- natural isomorphi s m$ \Psi_{\mathbf{ M}}: \ ma t hb f { M}\ri ghtarrow \ wi dehat { \mathbf { M} } $ .
I n particular t his implies tha t thepr obl e m of c lassi fy i ngsimple tran siti ve $2$-re presen t ationso f a $2$ -(semi )ca teg ory$ {{ \s c\m bo x {C} \ hs pac e {1. 0pt}}}$is e quiva lent t o that of cla ssify ing weak simp letran s iti ve $2 $-rep rese nt ation s ${{\ sc\mb ox {C}\hspace{1.0p t}}} $:
\[Sem iWo rk \]Le t ${{ \ sc\mbo x{C }\h space{1 .0pt}}} $ be a f i ni tary $2$-(semi)cat eg o r yand let$\math b f{ M} $ be a we ak $2 $-re p r esent atio n o f ${{\sc \mbox{ C }\ hs pace{1. 0p t}}}$. T her e i s a ( s tric t) $2$ -represe ntati o n $\widehat{\m a thbf{M}}$ of$ {{ \ s c\ m box{ C}\ hspace{1.0p t}}} $ equ ival e nt to $\mat hbf{M }$ .
$ 2$-semicategories o fprojec tivefunctors
---- ---------- - - - -------- ---- - -- - --------
Let$A$ b e a be a f i nite-dim ensio nal, bas ic, conne c t ed algeb raand le t $ { \ le ft\{ e_{1},\l d o ts,e_ {n} \ri ght \}}$ be acom ple tese t of idem potentsof $ A$ .Fix a sm a ll categ or y $ \m ath cal{A } $ equi valen t to $ A\ ! \on {-proj} $ .T h e $2 $- se mica teg or y ${\ sc\m b ox{ Z}\hspa ce{1.0pt} }_{ \ !A}$ i sdefined as follows:
- $\on{Ob }{ \sc \mbox{ Z } \hspace{ 1.0pt}}_{\!A} = {\left\ { \matht t{i } \ri ght\ }}$, wher e $ \matht t{i } $ canbe ide ntifi ed wi t h $\ma t h ca l{A }$ ;
- $1$ - m orp hisms o f ${ \sc\mbo x{Z}\hspace{1.0pt} } _{\ !A}$ are endo fun ctor s of $\ m at h cal {A } $ i s o morphic to tens oring with t h eprojective $A$ -$ A | from[@Po_Theorem 3.4]:
\[HardWork\]_For any homomorphism of_$2$-(semi)categories $\mathbf{M}:_{{\sc\mbox{C}\hspace{1.0pt}}}\rightarrow_\mathbf{Cat}$, there_is_a $2$-(semi)functor $\widehat{\mathbf{M}}:_{{\sc\mbox{C}\hspace{1.0pt}}}\rightarrow \mathbf{Cat}$ together_with a non-strict $2$-natural_isomorphism $\Psi_{\mathbf{M}}: \mathbf{M}_\rightarrow_\widehat{\mathbf{M}}$.
In particular this implies that the problem of classifying simple transitive $2$-representations of a_$2$-(semi)category_${{\sc\mbox{C}\hspace{1.0pt}}}$ is_equivalent_to_that of classifying weak simple_transitive $2$-representations ${{\sc\mbox{C}\hspace{1.0pt}}}$:
\[SemiWork\] Let ${{\sc\mbox{C}\hspace{1.0pt}}}$_be a_finitary $2$-(semi)category and let $\mathbf{M}$ be a weak_$2$-representation_of ${{\sc\mbox{C}\hspace{1.0pt}}}$. There_is a (strict) $2$-representation $\widehat{\mathbf{M}}$ of ${{\sc\mbox{C}\hspace{1.0pt}}}$ equivalent to_$\mathbf{M}$.
$2$-semicategories of projective functors
-----------------------------------------
Let $A$ be_a be a_finite-dimensional,_basic,_connected algebra and let_${\left\{ e_{1},\ldots, e_{n} \right\}}$ be a_complete set of idempotents of $A$._Fix a small category $\mathcal{A}$ equivalent to_$A\!\on{-proj}$. The $2$-semicategory ${\sc\mbox{Z}\hspace{1.0pt}}_{\!A}$ is defined_as follows:
- $\on{Ob}{\sc\mbox{Z}\hspace{1.0pt}}_{\!A}_= {\left\{_\mathtt{i} \right\}}$, where $\mathtt{i}$ can_be identified with_$\mathcal{A}$;
- _ $1$-morphisms of_${\sc\mbox{Z}\hspace{1.0pt}}_{\!A}$ are endofunctors of $\mathcal{A}$ isomorphic_to tensoring with_the projective $A$-$A |
Phys. [**31**]{}, 725 (1990)
A. Ashtekar and A. Magnon-Ashtekar “On the symplectic structure of general relativity” Commun. Math. Phys. [**86**]{}, 55 (1982).
R. Geroch, “Asymptotic structure of space-time,” in *Asymptotic structure of space-time*, ed. L. Witten, Plenum, New York (1976)
C. Kozameh and E. T. Newman, “A note on asymptotically flat spaces. II.”, General Relativity and Gravitation 15.5 (1983): 475-487.
[^1]: In [@cs], this theorem was proved in the holomorphic limit. More general proofs were later given in [@plefka; @bern]. See [@gj; @naculich; @white] for earlier work on soft graviton amplitudes.
[^2]: Based on earlier papers, we refer to the subleading soft theorem as Cachazo-Strominger theorem.
[^3]: Here ‘charge’ refers to what is called ‘flux’ in the radiative phase space literature, i.e. it involves a three dimensional integral over null infinity.
[^4]: We are deviating from the standard radiative phase space terminology in which ‘frame’ denotes a conformal class of $[(q_{ab},n^a)]$ of metric *and* null normal [@as].
[^5]: This actually is not the totality of free data since there are additional $u$-independent sphere functions that arise as integration ‘constants’ [@bt]. These however play no role in our analysis.
[^6]: The analogue of the space $\Gamma^{\qo}_0$ that was used in [@us] actually allows for a $u$-independent term and hence is invariant under supertranslations. The stronger condition (\[defGqoo\]) is used here in order to allow for certain integration by parts in $u$
[^7]: For slower fall-offs of the type which define $\Gamma$ (Eqns. (\[defGq\]), (\[defGamma\])) there may be a possibility of obtaining a finite symplectic structure by supplementing the action with counterterms at $i^{0}, i^{\pm}$. We have not pursued this direction here as it is not needed for our analysis. \[countertermfn\] | Phys. [ * * 31 * * ] { }, 725 (1990)
A. Ashtekar and A. Magnon - Ashtekar “ On the symplectic structure of general relativity ” Commun. Math. Phys. [ * * 86 * * ] { }, 55 (1982).
R. Geroch, “ Asymptotic structure of space - meter, ” in * Asymptotic social organization of space - time *, ed. L. Witten, Plenum, New York (1976)
C. Kozameh and E. T. Newman, “ A bill on asymptotically compressed spaces. II. ”, General Relativity and Gravitation 15.5 (1983 ): 475 - 487.
[ ^1 ]: In [ @cs ], this theorem was proved in the holomorphic terminus ad quem. More general proofs were later contribute in [ @plefka; @bern ]. See [ @gj; @naculich; @white ] for earlier work on indulgent graviton amplitudes.
[ ^2 ]: Based on early newspaper, we refer to the subleading soft theorem as Cachazo - Strominger theorem.
[ ^3 ]: Here ‘ charge ’ refers to what is called ‘ magnetic field ’ in the radiative phase space literature, i.e. it involves a three dimensional integral over null eternity.
[ ^4 ]: We are deviating from the standard radiative phase space terminology in which ‘ skeleton ’ denotes a conformal course of $ [ (q_{ab},n^a)]$ of metric * and * null normal [ @as ].
[ ^5 ]: This actually is not the sum of free data since there are extra $ u$-independent sphere functions that arise as integration ‘ constants ’ [ @bt ]. These however play no function in our analysis.
[ ^6 ]: The analogue of the space $ \Gamma^{\qo}_0 $ that was used in [ @us ] actually allows for a $ u$-independent term and hence is invariant under supertranslations. The stronger condition (\[defGqoo\ ]) is used here in ordering to allow for certain consolidation by part in $ u$
[ ^7 ]: For dense fall - offs of the type which define $ \Gamma$ (Eqns. (\[defGq\ ]), (\[defGamma\ ]) ) there may be a possibility of obtaining a finite symplectic structure by supplementing the legal action with counterterms at $ i^{0 }, i^{\pm}$. We have not pursued this direction here as it is not needed for our analysis. \[countertermfn\ ] | Phyd. [**31**]{}, 725 (1990)
A. Ashtekar and A. Magnok-Ashtekar “On the symplertic stducture uf general relativity” Commun. Larh. Phyw. [**86**]{}, 55 (1982).
R. Geroch, “Asymptotic structurv of spacw-timt,” in *Asymptotic svducture of spzge-timz*, xd. L. Witten, Plekum, New Yorn (1976)
C. Kozameh ang D. T. Uewman, “A note on asymptotically flat spaces. IL.”, General Relajivitj wnd Fgayitation 15.5 (1983): 475-487.
[^1]: In [@cs], this theorem wzs provtd in the holomorpnic limit. More general prolfs aere later given ij [@plefka; @beth]. Sqw [@gj; @naculicf; @white] fog earlier wkrk on soft graviton amplitudes.
[^2]: Baseb on earliet laogrs, we refer to tre subleadinn soft dheorem as Cachazo-Strpmiiger theorem.
[^3]: Here ‘charge’ cefers to what is cajled ‘flux’ iu the radiative phase space nitesatufw, i.d. iu iivomves a thcee dimensiknal integrql over null infiniuy.
[^4]: Rv are deviatihg froi ehe standard radiative phase space termpnolkgy in which ‘frame’ denores a conformal class of $[(q_{ab},n^a)]$ of metric *and* null normal [@as].
[^5]: This actually is nod the gotcoity ud vree data since there are additional $u$-indepenqsnu sihere functions tmat arise as integtahipg ‘constants’ [@bj]. These hosever play no role in our analtsis.
[^6]: The wnalpgue of the space $\Gamma^{\qo}_0$ tyat was used un [@us] actually alllws for a $u$-nndepemdent term and hence is invaxiant hnder superhranslatikvs. The stronger zoncidion (\[defGqoo\]) is used here yn order vo alkow for cerjain ineegration hy pavds in $u$
[^7]: For slower fall-lfxs of the hype which define $\Gamma$ (Eqns. (\[dehJq\]), (\[defGamma\])) tnese kay be a posslbility of obtayning a finite sympleetic sgructure bj supplemxnting the astion with cogjterterms at $i^{0}, i^{\pm}$. Wq hace nit pursjdd this directoon here cf it is not needed for out znalysis. \[countexuernfn\] | Phys. [**31**]{}, 725 (1990) A. Ashtekar and “On symplectic structure general relativity” Commun. R. “Asymptotic structure of in *Asymptotic structure space-time*, ed. L. Witten, Plenum, New (1976) C. Kozameh and E. T. Newman, “A note on asymptotically flat spaces. General Relativity and Gravitation 15.5 (1983): 475-487. [^1]: In [@cs], this theorem was in holomorphic More proofs were later given in [@plefka; @bern]. See [@gj; @naculich; @white] for earlier work on soft amplitudes. [^2]: Based on earlier papers, we refer the subleading soft theorem Cachazo-Strominger theorem. [^3]: Here ‘charge’ to is called in radiative space literature, i.e. involves a three dimensional integral over null infinity. [^4]: We are deviating from the standard radiative phase terminology in denotes a class $[(q_{ab},n^a)]$ metric *and* null [^5]: This actually is not the data since there are additional $u$-independent sphere functions arise as ‘constants’ [@bt]. These however play no in our analysis. [^6]: The analogue of the $\Gamma^{\qo}_0$ that was used in [@us] actually allows for a $u$-independent term and hence is supertranslations. The stronger condition is used here order allow certain by parts $u$ [^7]: For slower fall-offs of the type which define $\Gamma$ (\[defGq\]), (\[defGamma\])) there may be a possibility of obtaining a structure supplementing the action counterterms at $i^{0}, i^{\pm}$. have pursued this direction here is needed \[countertermfn\] | Phys. [**31**]{}, 725 (1990)
A. Ashtekar and A. Magnon-AsHtekar “On thE sympLecTic StRuctUre oF general relatiVIty” COmmun. Math. Phys. [**86**]{}, 55 (1982).
R. Geroch, “AsYmptoTiC StruCTuRe of sPace-timE,” In *aSYmpToTiC stRuCTuRe of sPacE-time*, ed. l. Witten, PleNum, neW York (1976)
C. KozameH AnD E. T. Newman, “A NotE on asymptotiCalLy flat SpAceS. iI.”, GenEraL RelaTivity ANd GravItation 15.5 (1983): 475-487.
[^1]: In [@Cs], THis theORem was pROVeD in tHe holomorphic limiT. moRE general proofs Were laTeR GiVEN in [@PleFka; @bern]. See [@Gj; @NaculICh; @white] FOr EARLieR Work on soft graViton amplitUDes.
[^2]: based oN eArlIEr papeRs, we rEfER to The subleadiNg soFt theorem As CachAZo-StromINger theOrem.
[^3]: HeRe ‘cHarGe’ reFErS tO whAt IS caLLeD ‘flUX’ in The radiaTiVe Phase SpacE LITEratUre, I.e. it InvolVes a three dimeNsiOnal INteGral oVer nuLl inFiNity.
[^4]: WE are deViatiNg From the standard RadiAtive phasE spAcE teRmInoloGY in whiCh ‘fRamE’ denoteS a confoRMal ClASS Of $[(Q_{ab},n^a)]$ of metric *and* nuLl NORmAl [@as].
[^5]: This ActualLY iS nOT the totaLiTy oF freE DAta siNce tHErE are addiTional $U$-InDePendent SpHere fuNcTioNs tHat arISe as IntegrAtion ‘conStantS’ [@Bt]. These however PLay no role in ouR AnALYsIS.
[^6]: The AnaLogue of the sPace $\gAmma^{\Qo}_0$ thAT wAs uSEd in [@uS] actuAlLY aLLows for a $u$-independenT tErm and Hence Is invariant unDer supertrANSLations. THe stROnGEr condition (\[defgqoo\]) iS used here iN Order to aLlow fOr certaiN integratION by parts In $u$
[^7]: for SloWer FALl-Offs of the type WHIch dEfIne $\GammA$ (EqNs. (\[defGq\]), (\[DefgamMa\])) tHerE mAy be a possIbility oF oBtAiNiNg a FinitE SymplectIc StrUcTurE by suPPlemenTing tHe acTiOn WIth CounterTErMS At $i^{0}, i^{\Pm}$. we Have Not PuRsued This DIreCtion heRe as it is nOt nEEded FoR oUr analySis. \[counterterMfN\] | Phys. [**31**]{}, 725 (1 990)
A. A shtek arand A . Ma gnon -Ashtekar “Ont he s ymplectic structure of gene ra l rel a ti vity” Commun . M a t h.Ph ys . [* * 86 **]{} , 5 5 (1982 ).
R. Ger och ,“Asymptotics tr ucture ofspa ce-time,” in *A sympto ti c s t ructu reof sp ace-ti m e*, ed . L. Witt en , Plenu m , New Y o r k(197 6)
C. Kozameh a n dE . T. Newman, “ A note o n a s y mpt oti cally flat s paces . II.”,G en e r a l R e lativity andGravitation 15. 5 (198 3) : 4 7 5-487.
[^1 ]: In[@cs], this the orem wasproved in theh olomorp hic li mit . M oreg en er alpr o ofs we rel ate r givenin [ @plef ka;@ b e r n].See [@g j; @n aculich; @whi te] for ear lierworkon s of t gra vitonampli tu des.
[^2]: Bas ed o n earlier pa pe rs, w e ref e r to t hesub leading soft t h eor em a s C achazo-Stromingerth e o re m.
[^3] : Here ‘c ha r ge’ refe rs to wha t is ca lled ‘f lux’ inthe ra d ia ti ve phas espaceli ter atu re, i . e. i t invo lves a t hreed imensional int e gral over nul l i n f in i ty.
[^ 4]: We aredevi a ting fro m t hes tanda rd ra di a ti v e phase space termi no logy i n whi ch ‘frame’ de notes a co n f o rmal cla ss o f $ [ (q_{ab},n^a)]$ of m etric *and * null no rmal[@as].
[^5]: Thi s actually is no t t het o ta lity of freed a ta s in ce ther e a re addi tio nal $u $-i nd ependentsphere f un ct io ns th at ar i se as in te gra ti on‘cons t ants’[@bt] . Th es eh owe ver pla y n o role i nourana ly sis.
[^6 ] : T he anal ogue of t hes pace $ \G amma^{\ qo}_0$ that w as used in [ @u s]actual l y allowsfor a $u$-independent t e rm andhen ce is inv ariant un der super tra n slatio ns. Th e str on ger c ondit i o n(\[ de fGqoo\]) i s use d her ein o rder to allow for certain int egration by p art s in $ u$
[ ^ 7] : Fo rs low e r fall-offs of t he type wh ic h d efine $\Ga m ma$ ( Eqns. ( \[defGq \]),( \[defGa mma\])) t here maybe a p o s sib ility of o btaining a finite sympl e ct ic st ruc ture b ysup pleme ntingt heactio n with c ounter terms a t $i^{0} , i^{\pm}$. We have not pursu ed th isdirection he r e a s it is n ot n eeded forour an alysi s.\ [coun tert e rm fn\ ] | Phys. [**31**]{},_725 (1990)
A. Ashtekar_and A. Magnon-Ashtekar “On the_symplectic structure_of_general relativity”_Commun. Math. Phys. _[**86**]{}, 55 (1982).
R. Geroch,_“Asymptotic structure of_space-time,” in *Asymptotic structure_of space-time*, ed._L._Witten, Plenum, New York (1976)
C. Kozameh and E. T. Newman, “A note on asymptotically flat spaces._II.”,_General Relativity_and_Gravitation_15.5 (1983): 475-487.
[^1]: In [@cs],_this theorem was proved in_the holomorphic_limit. More general proofs were later given in_[@plefka;_@bern]. See [@gj;_@naculich; @white] for earlier work on soft graviton amplitudes.
[^2]:_Based on earlier papers, we refer_to the subleading_soft_theorem_as Cachazo-Strominger theorem.
[^3]: Here_‘charge’ refers to what is called_‘flux’ in the radiative phase space_literature, i.e. it involves a three dimensional_integral over null infinity.
[^4]: We are_deviating from the standard radiative_phase space_terminology in which ‘frame’ denotes_a conformal class_of $[(q_{ab},n^a)]$_of metric *and*_null normal [@as].
[^5]: This actually is_not the totality_of free data since there are_additional_$u$-independent sphere functions_that_arise_as integration_‘constants’ [@bt]. These_however_play no_role_in our analysis.
[^6]: The analogue of_the_space $\Gamma^{\qo}_0$ that was used in [@us]_actually allows for a_$u$-independent_term and hence is_invariant under supertranslations. The stronger_condition (\[defGqoo\]) is used here in_order to_allow for_certain integration by parts in $u$
[^7]: For slower fall-offs of the_type which define $\Gamma$ (Eqns. (\[defGq\]),_(\[defGamma\])) there may be_a possibility_of_obtaining a finite_symplectic_structure by_supplementing the action with counterterms at $i^{0},_i^{\pm}$. We_have not pursued this direction here_as it is not_needed_for our analysis. \[countertermfn\] |
An obvious connection with the specification of vector quantile regression (i.e. the validity of an affine in $x$ form for the conditional quantile) is given by:
If $Y=\nabla \varphi(U)+ Db(U)^T X$ with
- $u\mapsto \varphi(u)+ b(u)\cdot x$ convex and smooth for $m$-a.e $x$,
- $\mathop{\mathrm{Law}}\nolimits(U)=\mu$, ${{\bf}{E}}(X\vert U)=0$,
then $U$ solves (\[maxcorrmimd\]).
This result follows from [@ccg], but for the sake of completeness, we give a proof: $$Y=\nabla \Phi_X(U), \mbox{ with } \Phi_X(t)=\varphi(t)+b(X)\cdot t.$$ Let $V$ be such that ${\mathop{\mathrm{Law}}\nolimits}(V)=\mu$, ${{{\bf}E}}(X\vert V)=0$, then by Young’s inequality $$V\cdot Y \le \Phi_X(V)+\Phi_X^*(Y)$$ but $Y=\nabla \Phi_X(U)$ implies that $$U\cdot Y = \Phi_X(U)+\Phi_X^*(Y)$$ so taking expectations gives the desired result.
Duality {#mvqr2}
-------
From now on, we do not assume a particular form for the conditional quantile and wish to study which information (\[maxcorrmimd\]) can give regarding the dependence of $X$ and $Y$. Once again, a good starting point is convex duality. As explained in details in [@ccg], the dual of ([maxcorrmimd]{}) takes the form $$\label{dualmimc}
\inf_{(\psi, \varphi, b)} {{\bf}{E}}(\psi(X,Y)+\varphi(U)) \; : \;
\psi(x,y)+\varphi(t)+b(t)\cdot x\ge t\cdot y.$$ where $U$ is any uniformly distributed random vector on $[0,1]^d$ i.e. $\mathop{\mathrm{Law}}\nolimits(U)=\mu=\mathop{\mathrm{uniform}}\nolimits([0,1]^d)$ and the infimum is taken over continuous functions $\psi\in C(\mathop{\mathrm{spt}}\nolimits | An obvious connection with the specification of vector quantile regression (i.e. the validity of an affine in $ x$ human body for the conditional quantile) is render by:
If $ Y=\nabla \varphi(U)+ Db(U)^T X$ with
- $ u\mapsto \varphi(u)+ b(u)\cdot x$ convex and smooth for $ m$-a.e $ x$,
- $ \mathop{\mathrm{Law}}\nolimits(U)=\mu$, $ { { \bf}{E}}(X\vert U)=0 $,
then $ U$ solves (\[maxcorrmimd\ ]).
This result follows from [ @ccg ], but for the sake of completeness, we hold a proof: $ $ Y=\nabla \Phi_X(U), \mbox { with } \Phi_X(t)=\varphi(t)+b(X)\cdot t.$$ Let $ V$ be such that $ { \mathop{\mathrm{Law}}\nolimits}(V)=\mu$, $ { { { \bf}E}}(X\vert V)=0 $, then by Young ’s inequality $ $ V\cdot Y \le \Phi_X(V)+\Phi_X^*(Y)$$ but $ Y=\nabla \Phi_X(U)$ incriminate that $ $ U\cdot yttrium = \Phi_X(U)+\Phi_X^*(Y)$$ so taking expectations gives the desired resultant role.
Duality { # mvqr2 }
-------
From now on, we do not wear a particular form for the conditional quantile and wish to study which information (\[maxcorrmimd\ ]) can give see the dependence of $ X$ and $ Y$. Once again, a good starting compass point is convex duality. As explained in details in [ @ccg ], the dual of ([ maxcorrmimd ] { }) takes the kind $ $ \label{dualmimc }
\inf_{(\psi, \varphi, b) } { { \bf}{E}}(\psi(X, Y)+\varphi(U) ) \; : \;
\psi(x, y)+\varphi(t)+b(t)\cdot x\ge t\cdot y.$$ where $ U$ is any uniformly distributed random vector on $ [ 0,1]^d$ i.e. $ \mathop{\mathrm{Law}}\nolimits(U)=\mu=\mathop{\mathrm{uniform}}\nolimits([0,1]^d)$ and the infimum is taken over continuous functions $ \psi\in C(\mathop{\mathrm{spt}}\nolimits |
An lbvious connection with uhe specification of vecvor quahtile reeression (i.e. the validity of en adfine in $x$ form for the conaitional euantile) is jiven by:
If $Y=\nable \varphi(M)+ Db(U)^F X$ wnti
- $u\mapsto \varkhi(u)+ b(u)\cdot f$ convex and skoutk for $m$-a.e $x$,
- $\mathop{\mathrm{Law}}\nolimitf(U)=\mu$, ${{\bf}{R}}(X\gert U)=0$,
then $U$ sjlvex (\[maxdorrmimd\]).
This result follows from [@cdg], but hor the sake of completeness, we give a prlof: $$J=\nabla \Phi_X(U), \mbox{ aith } \Phi_X(t)=\carpru(t)+b(X)\cdot t.$$ Ldt $V$ be subk that ${\mathkp{\mathrm{Law}}\nolimits}(V)=\mu$, ${{{\bf}E}}(X\vert V)=0$, thzn by Young’w uneeoality $$V\cdot Y \le \Phi_X(V)+\Phi_X^*(Y)$$ but $Y=\natla \Phi_C(U)$ implies thab $$U\cdmt T = \Phi_X(U)+\Phi_X^*(Y)$$ so takmng expectations givgs the deshrzd result.
Duality {#mvqr2}
-------
Drim nof on, we ai nut zsxuje a pwrtmcular form for the cobditional quantile snq wish to studg whicr ynformation (\[maxcorrmimd\]) can give regardpng fhe dependence of $X$ and $Y$. Once again, a good dtarting [oint is convex duality. As explained in details it [@ccg], ghe dmal ud ([laxcorrmimd]{}) takes the form $$\label{dualmimc}
\inf_{(\psy, \vsrihi, b)} {{\bf}{E}}(\psi(X,Y)+\varihi(U)) \; : \;
\psi(x,y)+\varphi(y)+b(h)\ccjt x\ge t\cdot i.$$ where $U$ js any uniformly dlstribujed rabdom vectjr om $[0,1]^d$ i.e. $\mathop{\mathrm{Law}}\nolimuts(U)=\mu=\mathop{\iqthrm{uniform}}\nolimics([0,1]^d)$ and the nnfimuk is yaken over continuous fbnctiohs $\psi\in C(\mwthop{\mathdo{spt}}\nolimits | An obvious connection with the specification of regression the validity an affine in quantile) given by: If \varphi(U)+ Db(U)^T X$ - $u\mapsto \varphi(u)+ b(u)\cdot x$ convex smooth for $m$-a.e $x$, - $\mathop{\mathrm{Law}}\nolimits(U)=\mu$, ${{\bf}{E}}(X\vert U)=0$, then $U$ solves (\[maxcorrmimd\]). This follows from [@ccg], but for the sake of completeness, we give a proof: \Phi_X(U), with \Phi_X(t)=\varphi(t)+b(X)\cdot Let $V$ be such that ${\mathop{\mathrm{Law}}\nolimits}(V)=\mu$, ${{{\bf}E}}(X\vert V)=0$, then by Young’s inequality $$V\cdot Y \le \Phi_X(V)+\Phi_X^*(Y)$$ $Y=\nabla \Phi_X(U)$ implies that $$U\cdot Y = \Phi_X(U)+\Phi_X^*(Y)$$ taking expectations gives the result. Duality {#mvqr2} ------- From on, do not a form the conditional quantile wish to study which information (\[maxcorrmimd\]) can give regarding the dependence of $X$ and $Y$. Once again, good starting convex duality. explained details [@ccg], the dual takes the form $$\label{dualmimc} \inf_{(\psi, \varphi, : \; \psi(x,y)+\varphi(t)+b(t)\cdot x\ge t\cdot y.$$ where $U$ any uniformly random vector on $[0,1]^d$ i.e. $\mathop{\mathrm{Law}}\nolimits(U)=\mu=\mathop{\mathrm{uniform}}\nolimits([0,1]^d)$ the infimum is taken over continuous functions $\psi\in |
An obvious connection with thE specificaTion oF veCtoR qUantIle rEgression (i.e. the VAlidIty of an affine in $x$ form foR the cOnDItioNAl QuantIle) is giVEn BY:
if $Y=\NaBlA \vaRpHI(U)+ db(U)^T X$ WitH
- $u\mapstO \varphi(u)+ b(u)\CdoT x$ Convex and smoOTh For $m$-a.e $x$,
- $\matHop{\Mathrm{Law}}\nolImiTs(U)=\mu$, ${{\bF}{E}}(x\veRT U)=0$,
theN $U$ sOlves (\[MaxcorRMimd\]).
ThIs result fOlLOws froM [@Ccg], but fOR ThE sakE of completeness, we GIvE A proof: $$Y=\nabla \PhI_X(U), \mboX{ wITh } \pHI_X(t)=\VarPhi(t)+b(X)\cdot T.$$ LEt $V$ be SUch that ${\MAtHOP{\MatHRm{Law}}\nolimits}(v)=\mu$, ${{{\bf}E}}(X\vert v)=0$, TheN by YouNg’S inEQualitY $$V\cdoT Y \LE \PhI_X(V)+\Phi_X^*(Y)$$ but $y=\nabLa \Phi_X(U)$ imPlies tHAt $$U\cdot y = \phi_X(U)+\PhI_X^*(Y)$$ so tAkiNg eXpecTAtIoNs gIvES thE DeSirED reSult.
DualItY {#mVqr2}
-------
FrOm noW ON, WE do nOt aSsumE a parTicular form foR thE conDItiOnal qUantiLe anD wIsh to Study wHich iNfOrmation (\[maxcorrMimd\]) Can give reGarDiNg tHe DepenDEnce of $x$ anD $Y$. ONce agaiN, a good sTArtInG POInT is convex duality. As ExPLAiNed in detAils in [@CCg], ThE Dual of ([maXcOrrMimd]{}) TAKes thE forM $$\LaBel{dualmImc}
\inf_{(\PSi, \VaRphi, b)} {{\bf}{e}}(\pSi(X,Y)+\vaRpHi(U)) \; : \;
\Psi(X,y)+\varPHi(t)+b(T)\cdot x\Ge t\cdot y.$$ Where $u$ Is any uniformly DIstributed ranDOm VECtOR on $[0,1]^d$ I.e. $\mAthop{\mathrm{law}}\nOLimiTs(U)=\mU=\MaThoP{\MathrM{unifOrM}}\NoLImits([0,1]^d)$ and the infimum Is Taken oVer coNtinuous functIons $\psi\in C(\MATHop{\mathrM{spt}}\NOlIMits |
An obvious connection wit h the spec ifica tio n o fvect or q uantile regres s ion(i.e. the validity ofan af fi n e in $x $ for m for t h ec o ndi ti on alqu a nt ile)isgiven b y:
If $Y= \na bl a \varphi(U) + D b(U)^T X$wit h
- $u\ma pst o \var ph i(u ) + b(u )\c dot x $ conv e x andsmooth fo r$ m$-a.e $x$,
- $ \mat hop{\mathrm{Law}} \ no l imits(U)=\mu$, ${{\b f} { E} } ( X\v ert U)=0$,
t he n $U$ solves( \[ m a x cor r mimd\]).
Thi s result fo l low s from [ @cc g ], but forth e sa ke of compl eten ess, we g ive ap roof: $ $ Y=\nabl a \Phi _X( U), \mb o x{ w ith } \Ph i _X (t) = \va rphi(t)+ b( X) \cdot t.$ $ L e t $V $ b e su ch th at ${\mathop{ \ma thrm { Law }}\no limit s}(V )= \mu$, ${{{\ bf}E} }( X\vert V)=0$, t henby Young’ s i ne qua li ty $$ V \cdotY \ le \Phi_X (V)+\Ph i _X^ *( Y ) $ $but $Y=\nabla \Phi _X ( U )$ implies that$ $U \c d ot Y = \P hi_ X(U) + \ Phi_X ^*(Y ) $$ so taki ng exp e ct at ions gi ve s thede sir edresul t .
D uality {#mvqr2 }
--- - ---
From nowo n, we do nota ss u m ea par tic ular form f or t h e co ndit i on alq uanti le an dw is h to study which inf or mation (\[m axcorrmimd\]) can giver e g arding t he d e pe n dence of $X$ a nd $Y $. Once ag a in, a go od st arting p oint is c o n vex dual ity . A s e xpl a i ne d in detailsi n [@c cg ], thedua l of ([ max cor rmi md] {} ) takes t he form$$ \l ab el {du almim c }
\inf_{ (\ psi ,\va rphi, b)} {{ \bf}{ E}}( \p si ( X,Y )+\varp h i( U ) ) \; : \ ;
\p si( x, y)+\v arph i (t) +b(t)\c dot x\get\c d ot y .$ $where $ U$ is any uni fo rmly distr ib ute d rand o m vectoron $[0,1]^d$ i.e. $\mat h op{\mat hrm {Law} }\no limits(U) =\m u=\mat hop { \mathr m{unif orm}} \n oli m i ts([0 , 1 ]^ d)$ a nd the inf i m umis ta ke n ov er cont inuous functions $ \ psi \in C(\mathop {\m athr m { sp t}} \ no l imi ts |
An obvious_connection with_the specification of vector_quantile regression_(i.e._the validity_of_an affine in_$x$ form for_the conditional quantile) is_given by:
If $Y=\nabla_\varphi(U)+_Db(U)^T X$ with
- $u\mapsto \varphi(u)+ b(u)\cdot x$ convex and smooth for $m$-a.e_$x$,
-_ _$\mathop{\mathrm{Law}}\nolimits(U)=\mu$,_${{\bf}{E}}(X\vert_U)=0$,
then $U$ solves (\[maxcorrmimd\]).
This result_follows from [@ccg], but for_the sake_of completeness, we give a proof: $$Y=\nabla \Phi_X(U),_\mbox{_with } \Phi_X(t)=\varphi(t)+b(X)\cdot_t.$$ Let $V$ be such that ${\mathop{\mathrm{Law}}\nolimits}(V)=\mu$, ${{{\bf}E}}(X\vert V)=0$,_then by Young’s inequality $$V\cdot Y_\le \Phi_X(V)+\Phi_X^*(Y)$$_but_$Y=\nabla_\Phi_X(U)$ implies that $$U\cdot_Y = \Phi_X(U)+\Phi_X^*(Y)$$ so taking_expectations gives the desired result.
Duality {#mvqr2}
-------
From_now on, we do not assume a_particular form for the conditional quantile_and wish to study which_information (\[maxcorrmimd\])_can give regarding the dependence_of $X$ and_$Y$. Once_again, a good_starting point is convex duality. As_explained in details_in [@ccg], the dual of ([maxcorrmimd]{})_takes_the form $$\label{dualmimc}
\inf_{(\psi,_\varphi,_b)}_{{\bf}{E}}(\psi(X,Y)+\varphi(U)) \;_: \;
\psi(x,y)+\varphi(t)+b(t)\cdot x\ge_t\cdot_y.$$ where_$U$_is any uniformly distributed random vector_on_$[0,1]^d$ i.e. $\mathop{\mathrm{Law}}\nolimits(U)=\mu=\mathop{\mathrm{uniform}}\nolimits([0,1]^d)$ and the infimum is_taken over continuous functions_$\psi\in_C(\mathop{\mathrm{spt}}\nolimits |
{\bolds{\beta}}^*(\cdot)=0\quad \mbox{versus}\quad
\breve{H}_{a} \dvt{\bolds{\beta}}^*(\cdot)\neq0.$$ Then we can perform $\lambda^*_n$ to testing $\breve{H}_{0}$ with transformed regression coefficients ${\bolds{\beta}}^*(\cdot)$ and response $\breve{y}_i=y_i-\mathbf{x}_i^\top{\bolds{\beta}}_0(t,\hat{\theta}_0)$. Note that the local linear estimator of ${\bolds{\beta}}^*(\cdot)$ has no bias under $\breve{H}_0$ and we can avoid the notorious problem of bias estimation.
As mentioned in Fan and Jiang [@FanJia07], a choice of larger bandwidth favors smoother alternatives and a smaller bandwidth tends to detect less smooth alternatives. Thanks to the introduction of the averaged test, the sensitivity of the test to the choice of bandwidth is alleviated due to the introduction of a group of bandwidths. On the other hand, the correlation of $\lambda_n$ between nearby bandwidths are usually quite high and hence in practice one only needs to average the test over a grid of relatively separated bandwidths. Zhang [@Zha03] found that the correlation between $\lambda_n(h)$ and $\lambda_n(ch)$ is quite high for $c=1.3$. As suggested by Fan and Jiang [@FanJia07], here we recommend choosing the grid of three bandwidths $\tilde{b}_n/1.5$, $\tilde{b}_n$ and $\tilde{b}_n\times1.5$ to represent small, medium and large bandwidths and average the test over the latter grid. Here $\tilde{b}_n=b_n^*\times n^{-1/45}$ and $b_n^*$ is the optimal bandwidth for nonparametric curve estimation.
The robust wild bootstrap {#secwildboots}
-------------------------
A direct implementation of the asymptotic distribution in Theorem \[thmnewtest\] may not perform satisfactorily in practice due to the following two reasons. First, the convergence rate of test statistic equals $\mathrm{O}(n^{-1/9})$ when bandwidth $b_n$ is chosen optimally. The rate is quite slow and hence the asymptotic approximation may not be accurate for moderate samples. Second, as we can see from the proof of Lemma \[lemasy5\] in Section \[secproof | { \bolds{\beta}}^*(\cdot)=0\quad \mbox{versus}\quad
\breve{H}_{a } \dvt{\bolds{\beta}}^*(\cdot)\neq0.$$ Then we can perform $ \lambda^*_n$ to testing $ \breve{H}_{0}$ with transformed regression coefficient $ { \bolds{\beta}}^*(\cdot)$ and response $ \breve{y}_i = y_i-\mathbf{x}_i^\top{\bolds{\beta}}_0(t,\hat{\theta}_0)$. notice that the local analogue estimator of $ { \bolds{\beta}}^*(\cdot)$ has no bias under $ \breve{H}_0 $ and we can keep off the notorious problem of diagonal estimation.
As mentioned in Fan and Jiang [ @FanJia07 ], a choice of bigger bandwidth favors smoother alternatives and a little bandwidth tends to detect less smooth alternatives. Thanks to the introduction of the averaged test, the sensitivity of the trial to the choice of bandwidth is alleviated due to the introduction of a group of bandwidth. On the other hand, the correlation coefficient of $ \lambda_n$ between nearby bandwidths are normally quite high and hence in practice one only needs to average the test over a power system of relatively separated bandwidths. Zhang [ @Zha03 ] found that the correlation between $ \lambda_n(h)$ and $ \lambda_n(ch)$ is quite high for $ c=1.3$. As suggested by Fan and Jiang [ @FanJia07 ], here we commend choose the grid of three bandwidths $ \tilde{b}_n/1.5 $, $ \tilde{b}_n$ and $ \tilde{b}_n\times1.5 $ to represent small, medium and large bandwidths and average the test over the latter grid. Here $ \tilde{b}_n = b_n^*\times n^{-1/45}$ and $ b_n^*$ is the optimum bandwidth for nonparametric curvature estimation.
The full-bodied raving mad bootstrap { # secwildboots }
-------------------------
A direct implementation of the asymptotic distribution in Theorem \[thmnewtest\ ] may not perform satisfactorily in exercise due to the following two reason. First, the convergence pace of test statistic equals $ \mathrm{O}(n^{-1/9})$ when bandwidth $ b_n$ is chosen optimally. The rate is quite slow and hence the asymptotic estimate may not be accurate for moderate samples. Second, as we can see from the proof of Lemma \[lemasy5\ ] in Section \[secproof | {\bolfs{\beta}}^*(\cdot)=0\quad \mbox{versur}\quad
\breve{H}_{a} \dvj{\bilds{\beva}}^*(\cdot)\nsq0.$$ Then de can perform $\lambda^*_n$ to tedtung $\beeve{H}_{0}$ with transformed regressiln coeffucieits ${\bolds{\beta}}^*(\cdov)$ and response $\nreve{v}_i=b_i-\mathbf{x}_i^\top{\bokds{\beta}}_0(t,\had{\theta}_0)$. Note thdt tke local linear estimator of ${\bolds{\beea}}^*(\cdot)$ nad no bias undet $\brene{R}_0$ ans we can avoid the notorious problsm of bpas estimation.
As kentioned in Fan and Jiang [@FanUia07], a choice of lagger bandwieth sqvors smoothdr alternauires and a sjaller bandwidth tends to detecg lesx smooth aotwrnwjives. Thanks to tre introductlpn of dhe avetaged test, the seisituvity of the test to vhe choice of bandwiqth is alneriated due to the inteoeuctimn ox a eeouo or uanswidthd. Oi the other hand, the cirrelation of $\lambds_n$ vetween nearbg bandriqths are usually quite high and hence it pdactice one only needs ro average the test oger a griq of relatively separated bandwidths. Zhang [@Zha03] fognd tiag tkc zirgelation between $\lambda_n(h)$ and $\lambda_n(ch)$ is quyfe hpgh for $c=1.3$. As suggcsted by Fan and Joajg [@SanJia07], here wg recommens choosing the grif of thtee babdwidths $\uilde{n}_n/1.5$, $\tilde{b}_n$ and $\tilde{b}_n\times1.5$ to represenn smqll, medium and larye bandwidthr anc avetage the test over the uattsr grid. Herf $\tilde{b}_n=g_v^*\times n^{-1/45}$ and $b_n^*$ ks nhe mptimal bandwidth for nonpwrametric curre estimxtiom.
The rjbust wild bootstrap {#secwildboots}
-------------------------
A direet im[lementatiln of the asymptotic distribution in Theorem \[jhmtewnest\] may uot pevform satisfactjrily in practnce due co the following two reesons. First, ehe convergenwg rate of tesv statistyc ewualw $\mathro{U}(n^{-1/9})$ when bandwicth $b_n$ is chosen oprimally. The rate ix qoife slow and henet tye asymptotic alpruxiianioi may tot be accurdte wor koderxte samkjew. Secpnd, as we can see frmm tge proof of Lemma \[keiasy5\] in Wection \[fecproof | {\bolds{\beta}}^*(\cdot)=0\quad \mbox{versus}\quad \breve{H}_{a} \dvt{\bolds{\beta}}^*(\cdot)\neq0.$$ Then we can to $\breve{H}_{0}$ with regression coefficients ${\bolds{\beta}}^*(\cdot)$ the linear estimator of has no bias $\breve{H}_0$ and we can avoid the problem of bias estimation. As mentioned in Fan and Jiang [@FanJia07], a choice larger bandwidth favors smoother alternatives and a smaller bandwidth tends to detect less alternatives. to introduction the averaged test, the sensitivity of the test to the choice of bandwidth is alleviated due the introduction of a group of bandwidths. On other hand, the correlation $\lambda_n$ between nearby bandwidths are quite and hence practice only to average the over a grid of relatively separated bandwidths. Zhang [@Zha03] found that the correlation between $\lambda_n(h)$ and $\lambda_n(ch)$ quite high As suggested Fan Jiang here we recommend grid of three bandwidths $\tilde{b}_n/1.5$, $\tilde{b}_n$ represent small, medium and large bandwidths and average test over latter grid. Here $\tilde{b}_n=b_n^*\times n^{-1/45}$ and is the optimal bandwidth for nonparametric curve estimation. robust wild bootstrap {#secwildboots} ------------------------- A direct implementation of the asymptotic distribution in Theorem \[thmnewtest\] perform satisfactorily in practice to the following reasons. the rate test statistic $\mathrm{O}(n^{-1/9})$ when bandwidth $b_n$ is chosen optimally. The rate is quite and hence the asymptotic approximation may not be accurate for Second, we can see the proof of Lemma in \[secproof | {\bolds{\beta}}^*(\cdot)=0\quad \mbox{versUs}\quad
\brevE{H}_{a} \dvT{\boLds{\BeTa}}^*(\cdOt)\neQ0.$$ Then we can perfORm $\laMbda^*_n$ to testing $\breve{H}_{0}$ wiTh traNsFOrmeD ReGressIon coefFIcIENts ${\BoLdS{\beTa}}^*(\CDoT)$ and rEspOnse $\breVe{y}_i=y_i-\mathBf{x}_I^\tOp{\bolds{\beta}}_0(t,\HAt{\Theta}_0)$. Note tHat The local lineAr eStimatOr Of ${\bOLds{\beTa}}^*(\cDot)$ haS no biaS Under $\bReve{H}_0$ and wE cAN avoid THe notorIOUs ProbLem of bias estimatiON.
AS Mentioned in Fan And JiaNg [@fAnjIA07], a cHoiCe of larger BaNdwidTH favors SMoOTHEr aLTernatives and A smaller banDWidTh tendS tO deTEct lesS smooTh ALteRnatives. ThaNks tO the introDuctioN Of the avERaged teSt, the sEnsItiVity OF tHe TesT tO The CHoIce OF baNdwidth iS aLlEviatEd duE TO THe inTroDuctIon of A group of bandwIdtHs. On THe oTher hAnd, thE corReLatioN of $\lamBda_n$ bEtWeen nearby bandwIdthS are usualLy qUiTe hIgH and hENce in pRacTicE one onlY needs tO AveRaGE THe Test over a grid of relAtIVElY separatEd bandWIdThS. zhang [@Zha03] FoUnd That THE corrElatIOn Between $\lAmbda_n(H)$ AnD $\lAmbda_n(cH)$ iS quite HiGh fOr $c=1.3$. as sugGEsteD by Fan And Jiang [@fanJiA07], Here we recommenD Choosing the grID oF THrEE banDwiDths $\tilde{b}_n/1.5$, $\TildE{B}_n$ anD $\tilDE{b}_N\tiMEs1.5$ to rEpresEnT SmALl, medium and large banDwIdths aNd aveRage the test ovEr the latteR GRId. Here $\tiLde{b}_N=B_n^*\TImes n^{-1/45}$ and $b_n^*$ is thE optiMal bandwidTH for nonpArameTric curvE estimatiON.
the robusT wiLd bOotStrAP {#SeCwildboots}
-------------------------
A diRECt imPlEmentatIon Of the asYmpTotIc dIstRiBution in THeorem \[thMnEwTeSt\] May Not peRForm satiSfActOrIly In praCTice duE to thE folLoWiNG twO reasonS. fiRST, the CoNvErgeNce RaTe of tEst sTAtiStic equAls $\mathrm{o}(n^{-1/9})$ wHEn baNdWiDth $b_n$ is Chosen optimalLy. the rate is qUiTe sLow and HENce the asYmptotic approximation maY Not be acCurAte foR modErate sampLes. second, As wE Can see From thE prooF oF LeMMA \[lemaSY5\] In secTiOn \[secproof | {\bolds{\beta}}^*(\cdot)=0 \quad \mbo x{ver sus }\q ua d
\b reve {H}_{a} \dvt{\ b olds {\beta}}^*(\cdot)\neq0 .$$ T he n wec an perf orm $\l a mb d a ^*_ n$ t o t es t in g $\b rev e{H}_{0 }$ with tr ans fo rmed regress i on coefficie nts ${\bolds{\b eta }}^*(\ cd ot) $ andres ponse $\bre v e{y}_i =y_i-\mat hb f {x}_i^ \ top{\bo l d s{ \bet a}}_0(t,\hat{\the t a} _ 0)$. Note that the l oc a ll i nea r e stimator o f${\bo l ds{\bet a }} ^ * ( \cd o t)$ has no bi as under $\ b rev e{H}_0 $and we can avoi dt henotorious p robl em of bia s esti m ation.As ment ionedinFan and Ji an g [ @F a nJi a 07 ],a ch oice ofla rg er ba ndwi d t h favo rssmoo theralternativesand a s m all er ba ndwid th t en ds to detec t les ssmooth alternat ives . Thankstoth e i nt roduc t ion of th e a veraged test,t hese n s i ti vity of the test t ot h echoice o f band w id th is allev ia ted due t o the int r od uction o f a gr o up o f bandw id ths. O nthe ot her h a nd,the co rrelatio n of$ \lambda_n$ bet w een nearby ba n dw i d th s are us ually quite hig h and hen c einp racti ce on eo nl y needs to average t he testovera grid of rel atively se p a r ated ban dwid t hs . Zhang [@Zha03 ] fou nd that th e correla tionbetween$\lambda_ n ( h)$ and$\l amb da_ n(c h ) $is quite high f or $ c= 1.3$. A s s uggeste d b y F anand J iang [@Fa nJia07], h er ewe re comme n d choosi ng th egri d oft hree b andwi dths $ \t i lde {b}_n/1 . 5$ , $\ti ld e{ b}_n $ a nd $\ti lde{ b }_n \times1 .5$ to re pre s entsm al l, medi um and largeba ndwidths a nd av eraget h e test o ver the latter grid. He r e $\til de{ b}_n= b_n^ *\times n ^{- 1/45}$ an d $b_n^ *$ isthe o pt ima l bandw i d th fo rnonparamet r i c c urvees tima tion.
The robust wild bo o tst rap {#secwild boo ts}- - -- --- - -- - --- -- - --- - - ---
A direct i mplementat io n o f the asym p tot ic distri butionin Th e orem \[ thmnewtes t\] may n ot per f o rmsatisfacto rily inpracticed ue to th e fol low ing tw orea sons. First , th e con vergen ce rateof te st statist ic equals $\mathrm{O}(n ^{-1/9 })$ w hen bandwidt h $ b _n$ is chose n op timally. T herat e isqui t e slo w an d h enc e theasym p totic app r ox ima t i on may not be a c cur ate f orm oderat e sa mples. Second, as we can see fro m th e pro ofo f Le mm a \[lemasy5\]inSe c t ion \[se cp roof | {\bolds{\beta}}^*(\cdot)=0\quad \mbox{versus}\quad
\breve{H}_{a}_\dvt{\bolds{\beta}}^*(\cdot)\neq0.$$ Then_we can perform $\lambda^*_n$_to testing_$\breve{H}_{0}$_with transformed_regression_coefficients ${\bolds{\beta}}^*(\cdot)$ and_response $\breve{y}_i=y_i-\mathbf{x}_i^\top{\bolds{\beta}}_0(t,\hat{\theta}_0)$. Note_that the local linear_estimator of ${\bolds{\beta}}^*(\cdot)$_has_no bias under $\breve{H}_0$ and we can avoid the notorious problem of bias estimation.
As_mentioned_in Fan_and_Jiang_[@FanJia07], a choice of larger_bandwidth favors smoother alternatives and_a smaller_bandwidth tends to detect less smooth alternatives. Thanks_to_the introduction of_the averaged test, the sensitivity of the test to_the choice of bandwidth is alleviated_due to the_introduction_of_a group of bandwidths._On the other hand, the correlation_of $\lambda_n$ between nearby bandwidths are_usually quite high and hence in practice_one only needs to average the_test over a grid of_relatively separated_bandwidths. Zhang [@Zha03] found that_the correlation between_$\lambda_n(h)$ and_$\lambda_n(ch)$ is quite_high for $c=1.3$. As suggested by_Fan and Jiang_[@FanJia07], here we recommend choosing the_grid_of three bandwidths_$\tilde{b}_n/1.5$,_$\tilde{b}_n$_and $\tilde{b}_n\times1.5$_to represent small,_medium_and large_bandwidths_and average the test over the_latter_grid. Here $\tilde{b}_n=b_n^*\times n^{-1/45}$ and $b_n^*$ is_the optimal bandwidth for_nonparametric_curve estimation.
The robust wild_bootstrap {#secwildboots}
-------------------------
A direct implementation of_the asymptotic distribution in Theorem \[thmnewtest\]_may not_perform satisfactorily_in practice due to the following two reasons. First, the convergence_rate of test statistic equals $\mathrm{O}(n^{-1/9})$_when bandwidth $b_n$ is_chosen optimally._The_rate is quite_slow_and hence_the asymptotic approximation may not be accurate_for moderate_samples. Second, as we can see_from the proof of_Lemma_\[lemasy5\] in Section \[secproof |
\\
X & \mapsto X \cap V
\end{aligned}$$ shows that the $\tilde{A}_i$ form a basis of quasi-atoms in $[V \cap
I, V]$. Next, let $Q = B_1 \vee \cdots \vee B_{(n-k)r}$. (Note that $nr - m = (n-k)r$.) The fact that $(V+I), B_1, \ldots,
B_{(n-k)r}$ is independent over $I$ implies that $(V+I) \cap Q = I$. Therefore, there is an isomorphism $$\begin{aligned}
[I,Q] &\stackrel{\sim}{\to} [V+I,V+I+Q] \\
X & \mapsto X + (V + I) = X + V.
\end{aligned}$$ The elements $\{B_1, \ldots, B_{(n-k)r}\}$ are independent quasi-atoms in $[I,Q]$, and therefore the $\tilde{B}_i$ are a set of independent quasi-atoms in $[V+I,V+I+Q]$ or even in $[V+I,{\mathbb{M}}^n]$. It follows that $$(n-k)r \le {\operatorname{rk}_\bot}({\mathbb{M}}^n/(V+I)) \le {\operatorname{rk}_0}({\mathbb{M}}^n/(V+I)) = (n-k)r,$$ so equality holds and the $\tilde{B}_i$ are a basis of quasi-atoms in $[V+I,{\mathbb{M}}^n]$.
\[r-fold-specs\] Fix a special element $I$ of $\mathcal{P}_1$. For every $n$, $I^n$ is a special element of $\mathcal{P}_n$. Let $\mathcal{G}_n$ be the modular geometry associated to the pregeometry of quasi-atoms over $I^n$ in $\mathcal{P}_n$. By utilizing the embeddings between the $\mathcal{G}_n$, one should be able to prove the following: there is an object $T$ in a semi-simple abelian category $\mathcal{C}$ such that $T$ has length $r$, and for every $n$ the geometry $\mathcal{G}_n$ is the geometry of atoms in the lattice of | \\
X & \mapsto X \cap V
\end{aligned}$$ shows that the $ \tilde{A}_i$ form a basis of quasi - atoms in $ [ five \cap
I, V]$. Next, lease $ Q = B_1 \vee \cdots \vee B_{(n - k)r}$. (Note that $ nr - m = (n - k)r$.) The fact that $ (V+I), B_1, \ldots,
B_{(n - k)r}$ is independent over $ I$ implies that $ (V+I) \cap Q = I$. consequently, there is an isomorphism $ $ \begin{aligned }
[ I, Q ] & \stackrel{\sim}{\to } [ V+I, V+I+Q ] \\
X & \mapsto X + (V + I) = X + V.
\end{aligned}$$ The elements $ \{B_1, \ldots, B_{(n - k)r}\}$ are autonomous quasi - atom in $ [ I, Q]$, and consequently the $ \tilde{B}_i$ are a hardening of independent quasi - atom in $ [ V+I, V+I+Q]$ or even in $ [ V+I,{\mathbb{M}}^n]$. It follows that $ $ (n - k)r \le { \operatorname{rk}_\bot}({\mathbb{M}}^n/(V+I) ) \le { \operatorname{rk}_0}({\mathbb{M}}^n/(V+I) ) = (n - k)r,$$ so equality holds and the $ \tilde{B}_i$ are a footing of quasi - atoms in $ [ V+I,{\mathbb{M}}^n]$.
\[r - fold - specs\ ] Fix a special element $ I$ of $ \mathcal{P}_1$. For every $ n$, $ I^n$ is a special element of $ \mathcal{P}_n$. Let $ \mathcal{G}_n$ be the modular geometry associate to the pregeometry of quasi - atoms over $ I^n$ in $ \mathcal{P}_n$. By utilizing the embeddings between the $ \mathcal{G}_n$, one should be able to prove the pursuit: there is an object $ T$ in a semi - elementary abelian category $ \mathcal{C}$ such that $ T$ have length $ r$, and for every $ n$ the geometry $ \mathcal{G}_n$ is the geometry of atoms in the lattice of | \\
X & \mapsto X \cap V
\end{xligned}$$ shows tkqt the $\tilde{Z}_i$ form x basis of quasi-atoms in $[V \cep
I, V]$. Next, let $Q = B_1 \vee \cdogs \vee B_{(n-n)r}$. (Note rhat $br - m = (n-k)r$.) The facb thaf $(V+I), Y_1, \odots,
B_{(n-k)r}$ is independett over $I$ implher chat $(V+I) \cap Q = I$. Therefore, there is wn isomprohism $$\begin{alidned}
[I,S] &\stackrel{\sim}{\to} [V+I,V+I+Q] \\
X & \mapstk X + (V + I) = X + V.
\end{sligned}$$ The elements $\{B_1, \ldohs, B_{(j-k)r}\}$ are independenh quasi-atomw in $[U,Q]$, and therewore the $\tpnde{B}_i$ are z set of independent quasi-atoms in $[V+N,V+I+Q]$ or eveb un $[G+H,{\mathbb{M}}^n]$. Iv folljws that $$(n-k)r \le {\operdtornamr{rk}_\bot}({\mathbb{M}}^n/(Y+I)) \le {\opwratorname{rk}_0}({\mathbb{M}}^n/(V+M)) = (n-k)r,$$ so equality hjlds and dhz $\tilde{B}_i$ are a basis od quavi-atmms kb $[V+K,{\mauhbu{M}}^n]$.
\[d-fold-soeca\] Fix a spscial elemebt $I$ of $\mathcal{P}_1$. Fot qnrry $n$, $I^n$ is z specyaj element of $\mathcal{P}_n$. Let $\mathcal{G}_n$ be ths modular geometry assoxiated to the pregeomgtry of quwsi-atoms over $I^n$ in $\mathcal{P}_n$. By utilizing the emteddiigr bttqeen gye $\mathcal{G}_n$, one should be able to prove the fojmoeikg: there is an onject $T$ in a semi-somolr abelian cateeory $\mcfhdal{C}$ such that $T$ hws lengjh $r$, abd for evtry $n$ the geometry $\mathcal{G}_n$ is rhe geometry if atoms in the lactice of | \\ X & \mapsto X \cap V that $\tilde{A}_i$ form basis of quasi-atoms Next, $Q = B_1 \cdots \vee B_{(n-k)r}$. that $nr - m = (n-k)r$.) fact that $(V+I), B_1, \ldots, B_{(n-k)r}$ is independent over $I$ implies that $(V+I) Q = I$. Therefore, there is an isomorphism $$\begin{aligned} [I,Q] &\stackrel{\sim}{\to} [V+I,V+I+Q] \\ & X (V I) = X + V. \end{aligned}$$ The elements $\{B_1, \ldots, B_{(n-k)r}\}$ are independent quasi-atoms in $[I,Q]$, therefore the $\tilde{B}_i$ are a set of independent in $[V+I,V+I+Q]$ or even $[V+I,{\mathbb{M}}^n]$. It follows that $$(n-k)r {\operatorname{rk}_\bot}({\mathbb{M}}^n/(V+I)) {\operatorname{rk}_0}({\mathbb{M}}^n/(V+I)) = so holds the $\tilde{B}_i$ are basis of quasi-atoms in $[V+I,{\mathbb{M}}^n]$. \[r-fold-specs\] Fix a special element $I$ of $\mathcal{P}_1$. For every $n$, $I^n$ a special $\mathcal{P}_n$. Let be modular associated to the quasi-atoms over $I^n$ in $\mathcal{P}_n$. By between the $\mathcal{G}_n$, one should be able to the following: is an object $T$ in a abelian category $\mathcal{C}$ such that $T$ has length and for every $n$ the geometry $\mathcal{G}_n$ is the geometry of atoms in the lattice | \\
X & \mapsto X \cap V
\end{aligned}$$ shoWs that the $\tIlde{A}_I$ foRm a BaSis oF quaSi-atoms in $[V \cap
I, v]$. next, Let $Q = B_1 \vee \cdots \vee B_{(n-k)r}$. (NoTe thaT $nR - M = (n-k)r$.) tHe Fact tHat $(V+I), B_1, \lDOtS,
b_{(N-k)r}$ Is InDepEnDEnT over $i$ imPlies thAt $(V+I) \cap Q = I$. THerEfOre, there is an ISoMorphism $$\beGin{Aligned}
[I,Q] &\staCkrEl{\sim}{\tO} [V+i,V+I+q] \\
x & \mapsTo X + (v + I) = X + V.
\eNd{aligNEd}$$ The eLements $\{B_1, \lDoTS, B_{(n-k)r}\}$ aRE indepeNDEnT quaSi-atoms in $[I,Q]$, and theREfORe the $\tilde{B}_i$ arE a set oF iNDePENdeNt qUasi-atoms iN $[V+i,V+I+Q]$ oR Even in $[V+i,{\MaTHBB{M}}^n]$. iT follows that $$(n-K)r \le {\operatoRNamE{rk}_\bot}({\MaThbB{m}}^n/(V+I)) \le {\OperaToRNamE{rk}_0}({\mathbb{M}}^n/(v+I)) = (n-k)R,$$ so equaliTy holdS And the $\tILde{B}_i$ arE a basiS of QuaSi-atOMs In $[v+I,{\mAtHBb{M}}^N]$.
\[R-fOld-SPecS\] Fix a speCiAl ElemeNt $I$ oF $\MATHcal{p}_1$. FoR eveRy $n$, $I^n$ Is a special eleMenT of $\mAThcAl{P}_n$. LEt $\matHcal{g}_n$ Be the ModulaR geomEtRy associated to tHe prEgeometry Of qUaSi-aToMs oveR $i^n$ in $\maThcAl{P}_N$. By utilIzing thE EmbEdDINGs Between the $\mathcal{G}_N$, oNE ShOuld be abLe to prOVe ThE FollowinG: tHerE is aN OBject $t$ in a SEmI-simple aBelian CAtEgOry $\mathCaL{C}$ such ThAt $T$ Has LengtH $R$, and For eveRy $n$ the geOmetrY $\Mathcal{G}_n$ is the GEometry of atomS In THE lATticE of | \\
X & \mapsto X \cap V
\end{ align ed} $$sh owsthat the $\tilde{A } _i$form a basis of quasi- atoms i n $[V \c ap
I, V] $ .N e xt, l et $Q = B_ 1 \ve e \ cdots \ vee B_{(n- k)r }$ . (Note that $n r - m = (n -k) r$.) The fac t t hat $( V+ I), B_1,\ld ots, B_{( n -k)r}$ is indep en d ent ov e r $I$ i m p li es t hat $(V+I) \cap Q =I $. Therefore,thereis an i som orp hism $$\be gi n{ali g ned}
[ I , Q ] & \ stackrel{\sim }{\to} [V+I , V+I +Q] \\
X & \ma pstoX+ (V + I) = X + V. \end{al igned} $ $ The e l ements$\{B_1 , \ ldo ts,B _{ (n -k) r} \ }$a re in d epe ndent qu as i- atoms in$ [ I , Q]$, an d th erefo re the $\tild e{B }_i$ are a se t ofinde pe ndent quasi -atom sin $[V+I,V+I+Q] $ or even in$[V +I ,{\ ma thbb{ M }}^n]$ . I t f ollowsthat $$ ( n-k )r \ l e{\operatorname{rk} _\ b o t} ({\mathb b{M}}^ n /( V+ I )) \le { \o per ator n a me{rk }_0} ( {\ mathbb{M }}^n/( V +I )) = (n-k )r ,$$ so e qua lit y hol d s an d the$\tilde{ B}_i$ are a basis of quasi-atoms i n $ [ V +I , {\ma thb b{M}}^n]$.
\[r - fold -spe c s\ ] F i x a s pecia le le m ent $I$ of $\mathca l{ P}_1$. Forevery $n$, $I ^n$ is a s p e c ial elem ento f$ \mathcal{P}_n$ . Let $\mathcal { G}_n$ be themodulargeometrya s sociated to th e p reg e o me try of quasi- a t omsov er $I^n $ i n $\mat hca l{P }_n $.By utilizin g the em be dd in gs be tween the $\ma th cal {G }_n $, on e shoul d beable t op rov e the f o ll o w ing: t he re i s a nobjec t $T $ in a semi -simple a bel i an c at eg ory $\m athcal{C}$ su ch that $T$ha s l ength$ r $, and f or every $n$ the geomet r y $\mat hca l{G}_ n$ i s the geo met ry ofato m s in t he lat ticeof | \\
_ _ X & \mapsto_X \cap_V
_ \end{aligned}$$_shows_that the $\tilde{A}_i$_form a basis_of quasi-atoms in $[V_\cap
__I, V]$. Next, let $Q = B_1 \vee \cdots \vee B_{(n-k)r}$. (Note that $nr_-_m =_(n-k)r$.)_The_fact that $(V+I), B_1, \ldots,
_ B_{(n-k)r}$ is independent over_$I$ implies_that $(V+I) \cap Q = I$. Therefore, there_is_an isomorphism $$\begin{aligned}
_ [I,Q] &\stackrel{\sim}{\to} [V+I,V+I+Q] \\
_ X & \mapsto X +_(V + I)_=_X_+ V.
\end{aligned}$$_The elements $\{B_1, \ldots, B_{(n-k)r}\}$ are_independent quasi-atoms in $[I,Q]$, and therefore_the $\tilde{B}_i$ are a set of independent_quasi-atoms in $[V+I,V+I+Q]$ or even in_$[V+I,{\mathbb{M}}^n]$. It follows that $$(n-k)r_\le {\operatorname{rk}_\bot}({\mathbb{M}}^n/(V+I))_\le {\operatorname{rk}_0}({\mathbb{M}}^n/(V+I)) = (n-k)r,$$ so_equality holds and_the $\tilde{B}_i$_are a basis_of quasi-atoms in $[V+I,{\mathbb{M}}^n]$.
\[r-fold-specs\] Fix a_special element $I$_of $\mathcal{P}_1$. For every $n$, $I^n$_is_a special element_of_$\mathcal{P}_n$._Let $\mathcal{G}_n$_be the modular_geometry_associated to_the_pregeometry of quasi-atoms over $I^n$ in_$\mathcal{P}_n$._By utilizing the embeddings between the $\mathcal{G}_n$,_one should be able_to_prove the following: there_is an object $T$ in_a semi-simple abelian category $\mathcal{C}$ such_that $T$_has length_$r$, and for every $n$ the geometry $\mathcal{G}_n$ is the geometry_of atoms in the lattice of |
well as the desired identity $((\Phi^{*})_{\s})^{*} (I+B_\oplus)(\Phi^{*})_{s}=R^*(I+iC)R$, and thus $\wh H$ is extremal, cf. Theorem \[s-sum-caracter\].
Theorem \[s-sum-een\] is a generalization of [@HSS19 Theorem 3.5], where an additional invariance of $\mul K^*$ under the operator $B_\oplus$ was used. Moreover, Theorem \[s-sum-een\] generalizes a corresponding result for the form sum of two closed nonnegative forms established earlier in [@HSSW2007 Theorem 4.1].
The present result relies on Theorem \[s-repr\], where the description of the closed sectorial form generated by a general maximal sectorial relation of the form $H=T^*(I+iB)T$ where $T$ is a closed relation. This generality implies that with special choices of $B$ the relation $H$ can be taken to be nonnegative and selfadjoint, i.e., the corresponding closed form $\st$ becomes nonnegative; see Example \[examp\].
[33]{}
Yu.M. Arlinskiĭ, “Maximal sectorial extensions and closed forms associated with them”, Ukrainian Math. J., 48 (1996), 723–739.
M. Haase, *The functional calculus for sectorial operators*, Oper. Theory Adv. Appl., 169, Birkhaüser Verlag, 2006.
S. Hassi, A. Sandovici, H.S.V. de Snoo, and H. Winkler, “Extremal extensions for the sum of nonnegative selfadjoint relations”, Proc. Amer. Math. Soc., 135 (2007), 3193–3204.
S. Hassi, A. Sandovici, H.S.V. de Snoo, and H. Winkler, “Extremal maximal sectorial extensions of sectorial relations”, Indag. Math., 28 (2017), 1019–1055.
S. Hassi, A. Sandovici, and H.S.V. de Snoo, “Factorized sectorial relations, their maximal sectorial extensions, and form sums”, Banach J. Math. Anal., 13 (2019), 538–564.
S. Hassi, H.S | well as the desired identity $ (( \Phi^{*})_{\s})^ { * } (I+B_\oplus)(\Phi^{*})_{s}=R^*(I+iC)R$, and thus $ \wh H$ is extremal, cf. Theorem \[s - sum - caracter\ ].
Theorem \[s - kernel - een\ ] is a abstraction of [ @HSS19 Theorem 3.5 ], where an additional invariance of $ \mul K^*$ under the hustler $ B_\oplus$ was used. furthermore, Theorem \[s - sum - een\ ] generalizes a comparable result for the form sum of two close nonnegative forms established in the first place in [ @HSSW2007 Theorem 4.1 ].
The present result relies on Theorem \[s - repr\ ], where the description of the closed sectorial mannequin generated by a general maximal sectorial relation of the form $ H = T^*(I+iB)T$ where $ T$ is a shut relation. This generality implies that with especial choices of $ B$ the relation $ H$ can be take to be nonnegative and selfadjoint, i.e., the comparable closed form $ \st$ becomes nonnegative; see Example \[examp\ ].
[ 33 ] { }
Yu. M. Arlinskiĭ, “ Maximal sectorial extensions and closed mannequin associated with them ”, Ukrainian Math. J., 48 (1996), 723–739.
M. Haase, * The functional calculus for sectorial operators *, Oper. Theory Adv. Appl. , 169, Birkhaüser Verlag, 2006.
S. Hassi, A. Sandovici, H.S.V. de Snoo, and H. Winkler, “ Extremal extensions for the sum of nonnegative selfadjoint relations ”, Proc. Amer. Math. Soc. , 135 (2007), 3193–3204.
S. Hassi, A. Sandovici, H.S.V. de Snoo, and H. Winkler, “ Extremal maximal sectorial extensions of sectorial relations ”, Indag. Math. , 28 (2017), 1019–1055.
S. Hassi, A. Sandovici, and H.S.V. de Snoo, “ Factorized sectorial relations, their maximal sectorial extensions, and form sums ”, Banach J. Math. Anal. , 13 (2019), 538–564.
S. Hassi, H.S | wepl as the desired identiuy $((\Phi^{*})_{\s})^{*} (I+B_\oplus)(\Phn^{*})_{w}=R^*(I+iC)R$, and tgus $\wh H$ is extremal, cf. Theorem \[s-sum-raraxter\].
Tyeorem \[s-sum-een\] is a geveralizatpon of [@HSW19 Thtorem 3.5], where an avsitional invadlance if $\mul K^*$ under the operador $B_\oplus$ was ureb. Moreover, Theorem \[s-sum-een\] generalizqs a cotrfsponding resujt fpw ths form sum of two closed nonnegatibe formv established earlier in [@HSSW2007 Theorem 4.1].
Thf prfsent result relied on Theoren \[s-rqpr\], where the description of the clksed sectorial form generated bh a gzneral maxinao sfwtorial reletion jf the form $M=N^*(I+iB)T$ wvere $T$ os a closed rekatmon. Rhis generality implixs that with special choices mf $B$ the relation $H$ cqn be takan tu be nohnxgafive ajd aelfadjoinf, i.e., the coeresponding closed gown $\st$ becomes honnegwtyve; see Example \[examp\].
[33]{}
Yu.M. Arlinskiĭ, “Maximan ssctorial extensions and closed forms associajed with trem”, Ukrainian Math. J., 48 (1996), 723–739.
M. Haase, *The functional calcglus hof stcbjfuap operators*, Oper. Theory Adv. Appl., 169, Birkhaüser Vqdlsg, 2006.
S. Hassi, A. Sanbovici, H.S.V. de Snpo, amq H. Winkler, “Ebtremal estensions for the dum of gonnetative sejfadkoint relations”, Proc. Amer. Mqth. Soc., 135 (2007), 3193–3204.
S. Kaswi, A. Sandovici, H.S.V. de Snoo, anb H. Wimkler, “Extremal maximal sectoxial estensions ov sectorizu relations”, Indae. Mstv., 28 (2017), 1019–1055.
S. Hassi, A. Sandovici, anq H.S.V. de Wnoo, “Factorkzed sectowial relatlons, bveir maximal sectogial gxtenshons, and flrm sums”, Banach J. Math. Anal., 13 (2019), 538–564.
S. Iessi, H.S | well as the desired identity $((\Phi^{*})_{\s})^{*} (I+B_\oplus)(\Phi^{*})_{s}=R^*(I+iC)R$, $\wh is extremal, Theorem \[s-sum-caracter\]. Theorem [@HSS19 3.5], where an invariance of $\mul under the operator $B_\oplus$ was used. Theorem \[s-sum-een\] generalizes a corresponding result for the form sum of two closed forms established earlier in [@HSSW2007 Theorem 4.1]. The present result relies on Theorem where description the sectorial form generated by a general maximal sectorial relation of the form $H=T^*(I+iB)T$ where $T$ is closed relation. This generality implies that with special of $B$ the relation can be taken to be and i.e., the closed $\st$ nonnegative; see Example [33]{} Yu.M. Arlinskiĭ, “Maximal sectorial extensions and closed forms associated with them”, Ukrainian Math. J., 48 (1996), M. Haase, calculus for operators*, Theory Appl., 169, Birkhaüser S. Hassi, A. Sandovici, H.S.V. de Winkler, “Extremal extensions for the sum of nonnegative relations”, Proc. Math. Soc., 135 (2007), 3193–3204. S. A. Sandovici, H.S.V. de Snoo, and H. Winkler, maximal sectorial extensions of sectorial relations”, Indag. Math., 28 (2017), 1019–1055. S. Hassi, A. Sandovici, de Snoo, “Factorized sectorial their maximal sectorial and sums”, J. Anal., 13 538–564. S. Hassi, H.S | well as the desired identity $((\PHi^{*})_{\s})^{*} (I+B_\oplus)(\phi^{*})_{s}=R^*(i+iC)r$, anD tHus $\wH H$ is Extremal, cf. TheoREm \[s-sUm-caracter\].
Theorem \[s-sum-eEn\] is a GeNEralIZaTion oF [@HSS19 TheOReM 3.5], WHerE aN aDdiTiONaL invaRiaNce of $\muL K^*$ under the OpeRaTor $B_\oplus$ was USeD. Moreover, THeoRem \[s-sum-een\] geNerAlizes A cOrrESpondIng ResulT for thE Form suM of two cloSeD NonnegATive forMS EsTablIshed earlier in [@HSSw2007 thEOrem 4.1].
The present Result ReLIeS ON ThEorEm \[s-repr\], wheRe The deSCriptioN Of THE CloSEd sectorial foRm generated BY a gEneral MaXimAL sectoRial rElATioN of the form $H=t^*(I+iB)t$ where $T$ is A closeD RelatioN. this genEralitY imPliEs thAT wItH spEcIAl cHOiCes OF $B$ tHe relatiOn $h$ cAn be tAken TO BE NonnEgaTive And seLfadjoint, i.e., thE coRresPOndIng clOsed fOrm $\sT$ bEcomeS nonneGativE; sEe Example \[examp\].
[33]{}
YU.M. ArLinskiĭ, “MaXimAl SecToRial eXTensioNs aNd cLosed foRms assoCIatEd WITH tHem”, Ukrainian Math. J., 48 (1996), 723–739.
M. haASE, *THe functiOnal caLCuLuS For sectoRiAl oPeraTORs*, OpeR. TheORy adv. Appl., 169, BIrkhaüSEr veRlag, 2006.
S. HaSsI, A. SandOvIci, h.S.V. De SnoO, And H. winkleR, “ExtremaL exteNSions for the sum OF nonnegative sELfADJoINt reLatIons”, Proc. AmeR. MatH. soc., 135 (2007), 3193–3204.
S. hassI, a. SAndOVici, H.s.V. de SNoO, AnD h. Winkler, “Extremal maxImAl sectOrial Extensions of sEctorial reLATIons”, IndaG. MatH., 28 (2017), 1019–1055.
s. HASsi, A. Sandovici, aNd H.S.V. De Snoo, “FactORized secToriaL relatioNs, their maXIMal sectoRiaL exTenSioNS, AnD form sums”, BanaCH j. MatH. ANal., 13 (2019), 538–564.
S. HasSi, H.s | well as the desired ident ity $((\Ph i^{*} )_{ \s} )^ {*}(I+B _\oplus)(\Phi^ { *})_ {s}=R^*(I+iC)R$, and t hus $ \w h H$i sextre mal, cf . T h e ore m\[ s-s um - ca racte r\] .
Theo rem \[s-su m-e en \] is a gene r al ization of [@ HSS19 Theore m 3 .5], w he rea n add iti onalinvari a nce of $\mul K^ *$ undert he oper a t or $B_ \oplus$ was used. Mo r eover, Theorem \[s-s um - ee n \ ] g ene ralizes aco rresp o nding r e su l t for the form sumof two clos e d n onnega ti vef orms e stabl is h edearlier in[@HS SW2007 Th eorem4 .1].
T h e prese nt res ult re lies on T heo re m \[ s -r epr \ ],where th ede scrip tion o f theclo sedsecto rial form gen era tedb y a gene ral m axim al sect orialrelat io n of the form $ H=T^ *(I+iB)T$ wh er e $ T$ is a closed re lat ion. Th is gene r ali ty i m pl ies that with spec ia l ch oices of $B$ t h ere l ation $H $can bet a ken t o be no nnegativ e ands el fa djoint, i .e., t he co rre spond i ng c losedform $\s t$ be c omes nonnegati v e; see Exampl e \ [ e xa m p\].
[ 33]{}
Yu.M . Ar l insk iĭ,“ Ma xim a l sec toria le xt e nsions and closed f or ms ass ociat ed with them” , Ukrainia n M ath. J., 48( 19 9 6), 723–739.
M. Ha ase, *Thef unctiona l cal culus fo r sectori a l operato rs* , O per . T h e or y Adv. Appl., 1 69,Bi rkhaüse r V erlag,200 6.
S. Ha ss i, A. San dovici,H. S. V. d e S noo,a nd H. Wi nk ler ,“Ex trema l exten sions for t he sum of non n eg a t ivese lf adjo int r elati ons” , Pr oc. Ame r. Math.Soc . , 13 5(2 007), 3 193–3204.
S. H assi, A. S an dov ici, H . S .V. de S noo, and H. Winkler, “E x tremalmax imalsect orial ext ens ions o f s e ctoria l rela tions ”, In d a g. Ma t h ., 28 ( 2017), 101 9 – 105 5.
S .Hass i, A. S andovici, and H.S. V . d e Snoo, “Fact ori zeds e ct ori a lr ela ti o ns, t heir maximal se ctorial ex te n si ons, and f o rmsu ms”, Ba nach J. Math . Anal., 13 (2019 ), 538–56 4.
S. H ass i, H.S | well_as the_desired identity $((\Phi^{*})_{\s})^{*} (I+B_\oplus)(\Phi^{*})_{s}=R^*(I+iC)R$,_and thus_$\wh_H$ is_extremal,_cf. Theorem \[s-sum-caracter\].
Theorem_\[s-sum-een\] is a_generalization of [@HSS19 Theorem_3.5], where an_additional_invariance of $\mul K^*$ under the operator $B_\oplus$ was used. Moreover, Theorem \[s-sum-een\] generalizes_a_corresponding result_for_the_form sum of two closed_nonnegative forms established earlier in_[@HSSW2007 Theorem 4.1].
The_present result relies on Theorem \[s-repr\], where the_description_of the closed_sectorial form generated by a general maximal sectorial relation_of the form $H=T^*(I+iB)T$ where $T$_is a closed_relation._This_generality implies that with_special choices of $B$ the relation_$H$ can be taken to be_nonnegative and selfadjoint, i.e., the corresponding closed_form $\st$ becomes nonnegative; see Example_\[examp\].
[33]{}
Yu.M. Arlinskiĭ, “Maximal sectorial extensions and_closed forms_associated with them”, Ukrainian Math._J., 48 (1996),_723–739.
M. Haase, *The_functional calculus for_sectorial operators*, Oper. Theory Adv. Appl.,_169, Birkhaüser Verlag,_2006.
S. Hassi, A. Sandovici, H.S.V. de_Snoo,_and H. Winkler,_“Extremal_extensions_for the_sum of nonnegative_selfadjoint_relations”, Proc._Amer._Math. Soc., 135 (2007), 3193–3204.
S. Hassi,_A._Sandovici, H.S.V. de Snoo, and H. Winkler,_“Extremal maximal sectorial extensions_of_sectorial relations”, Indag. Math.,_28 (2017), 1019–1055.
S. Hassi, A._Sandovici, and H.S.V. de Snoo, “Factorized_sectorial relations,_their maximal_sectorial extensions, and form sums”, Banach J. Math. Anal., 13 (2019),_538–564.
S. Hassi, H.S |
of
which are $(m-1,\alpha_m)$-planar. Using Remark \ref{rem:Lip}, we find
a $(V,\alpha_m)$-planar set $\widehat P$ such that $P\subset\widehat P$ and
$\operatorname{proj}_V(\widehat P)=V$. The r\^ole of $\widehat P$ is to guarantee that
$\widehat P\cap\operatorname{proj}_V^{-1}(P'_j)\ne\emptyset$ for all $j$.
For all $j\in\{1,\dots,M(m,\alpha_m)\}$ the sets
$\widetilde{P}_j=\widehat P\cap\operatorname{proj}_V^{-1}(P'_j)$ are
$(m-1,\alpha_{m-1})$-planar, and moreover,
\begin{equation*}
A\cap B(x,r)\cap P(c_{2}^{n-m-1}2\delta r)
\subset\bigcup_{j=1}^{M(m,\alpha_m)}
\widetilde{P}_j(c_{2}^{n-m}2\delta r),$$ see Figure \[enlargement\]. As the result of this inductive process we may find $(n-k,\alpha_{n-k})$-planar sets $P_{n-k,1},\dots,P_{n-k,l_{n-k}}$, where $l_{n-k}=M(n,\alpha_{n-1})\prod_{j=n-k+1}^{n-1}M(j,\alpha_{j})$, such that $$\label{final}
A\cap B(x,r)\subset
\bigcup_{i=1}^{l_{n-k}}P_{n-k,i}(c_{2}^{k-1}2\delta r).$$
It is not hard to verify that there is a constant $C(\alpha,n,k)$ depending only on $\alpha$, $n$, and $k$ such that each of the sets $P_{n-k,i}(c_2^{k-1}2\delta r)\cap B(x,r)$ can be covered with $C(\alpha,n,k)\delta^{k-n}$ balls of radius $\delta r$, and therefore by, $C(\alpha,n,k | of
which are $ (m-1,\alpha_m)$-planar. Using Remark \ref{rem: Lip }, we find
a $ (V,\alpha_m)$-planar set $ \widehat P$ such that $ P\subset\widehat P$ and
$ \operatorname{proj}_V(\widehat P)=V$. The r\^ole of $ \widehat P$ is to guarantee that
$ \widehat P\cap\operatorname{proj}_V^{-1}(P'_j)\ne\emptyset$ for all $ j$.
For all $ j\in\{1,\dots, M(m,\alpha_m)\}$ the set
$ \widetilde{P}_j=\widehat P\cap\operatorname{proj}_V^{-1}(P'_j)$ are
$ (m-1,\alpha_{m-1})$-planar, and furthermore,
\begin{equation * }
A\cap B(x, r)\cap P(c_{2}^{n - m-1}2\delta r)
\subset\bigcup_{j=1}^{M(m,\alpha_m) }
\widetilde{P}_j(c_{2}^{n - m}2\delta r),$$ see Figure \[enlargement\ ]. As the resultant role of this inductive summons we may find $ (n - k,\alpha_{n - k})$-planar sets $ P_{n - k,1},\dots, P_{n - k, l_{n - k}}$, where $ l_{n - k}=M(n,\alpha_{n-1})\prod_{j = n - k+1}^{n-1}M(j,\alpha_{j})$, such that $ $ \label{final }
A\cap B(x, r)\subset
\bigcup_{i=1}^{l_{n - k}}P_{n - k, i}(c_{2}^{k-1}2\delta r).$$
It is not intemperate to affirm that there is a changeless $ C(\alpha, n, k)$ depending only on $ \alpha$, $ n$, and $ k$ such that each of the set $ P_{n - k, i}(c_2^{k-1}2\delta r)\cap B(x, r)$ can be covered with $ C(\alpha, n, k)\delta^{k - n}$ balls of radius $ \delta r$, and consequently by, $ C(\alpha, n, k | of
which are $(m-1,\alpha_m)$-planar. Using Remark \rgf{eem:Lip}, we fihd
a $(V,\alpfa_m)$-planar set $\widehat P$ such tyat $P\wubset\widehat P$ and
$\opefatorname{iroj}_V(\wideyat K)=V$. The r\^ole of $\wivshat P$ lf to nuarautxe that
$\widehat K\cap\operatortame{proj}_V^{-1}(P'_j)\ne\ekpgydet$ for all $j$.
For all $j\in\{1,\dots,M(m,\alpra_m)\}$ the sfts
$\widetilde{P}_t=\widthae P\czi\okeratorname{proj}_V^{-1}(P'_j)$ are
$(m-1,\alpha_{m-1})$-plahar, and moreover,
\begon{equation*}
A\cap B(x,r)\cap P(c_{2}^{n-l-1}2\delha r)
\subset\bigcup_{j=1}^{M(l,\alpha_m)}
\widgfilqw{P}_j(c_{2}^{n-m}2\delta f),$$ see Figugz \[enlargemeht\]. As the result of this inductkve pxocess we mqy fijg $(n-k,\alpha_{n-k})$-'lanar sets $P_{n-k,1},\dots,P_{n-k,l_{n-k}}$, fhere $l_{m-k}=M(n,\alpha_{n-1})\prod_{m=n-k+1}^{n-1}M(o,\alpya_{j})$, such that $$\label{fiial}
A\cap B(x,r)\subset
\bigsup_{i=1}^{l_{n-k}}P_{n-n,i}(e_{2}^{k-1}2\delta r).$$
It is not haed to vgrify thag thdre ix z conshanv $C(\alpha,n,k)$ sepending obly on $\alpha$, $n$, and $l$ flvh that each of thq fets $P_{n-k,i}(c_2^{k-1}2\delta r)\cap B(x,r)$ can be covereg wjth $C(\alpha,n,k)\delta^{k-n}$ balos of radius $\delta r$, wnd theresore by, $C(\alpha,n,k | of which are $(m-1,\alpha_m)$-planar. Using Remark \ref{rem:Lip}, a set $\widehat such that $P\subset\widehat r\^ole $\widehat P$ is guarantee that $\widehat for all $j$. For all $j\in\{1,\dots,M(m,\alpha_m)\}$ sets $\widetilde{P}_j=\widehat P\cap\operatorname{proj}_V^{-1}(P'_j)$ are $(m-1,\alpha_{m-1})$-planar, and moreover, \begin{equation*} A\cap B(x,r)\cap P(c_{2}^{n-m-1}2\delta r) \subset\bigcup_{j=1}^{M(m,\alpha_m)} r),$$ see Figure \[enlargement\]. As the result of this inductive process we may $(n-k,\alpha_{n-k})$-planar $P_{n-k,1},\dots,P_{n-k,l_{n-k}}$, $l_{n-k}=M(n,\alpha_{n-1})\prod_{j=n-k+1}^{n-1}M(j,\alpha_{j})$, that $$\label{final} A\cap B(x,r)\subset \bigcup_{i=1}^{l_{n-k}}P_{n-k,i}(c_{2}^{k-1}2\delta r).$$ It is not hard to verify that there is a $C(\alpha,n,k)$ depending only on $\alpha$, $n$, and $k$ that each of the $P_{n-k,i}(c_2^{k-1}2\delta r)\cap B(x,r)$ can be with balls of $\delta and by, $C(\alpha,n,k | of
which are $(m-1,\alpha_m)$-planar. UsIng Remark \rEf{rem:lip}, We fInD
a $(V,\aLpha_M)$-planar set $\wideHAt P$ sUch that $P\subset\widehat P$ And
$\opErATornAMe{Proj}_V(\Widehat p)=v$. THE R\^olE oF $\wIdeHaT p$ iS to guAraNtee thaT
$\widehat P\cAp\oPeRatorname{proJ}_v^{-1}(P'_J)\ne\emptyseT$ foR all $j$.
For all $j\In\{1,\dOts,M(m,\aLpHa_m)\}$ THe setS
$\wiDetilDe{P}_j=\wiDEhat P\cAp\operatoRnAMe{proj}_v^{-1}(p'_j)$ are
$(m-1,\aLPHa_{M-1})$-plaNar, and moreover,
\begIN{eQUation*}
A\cap B(x,r)\cAp P(c_{2}^{n-m-1}2\DeLTa R)
\SUbsEt\bIgcup_{j=1}^{M(m,\alPhA_m)}
\widETilde{P}_j(C_{2}^{N-m}2\DELTa r),$$ SEe Figure \[enlarGement\]. As the REsuLt of thIs IndUCtive pRocesS wE May Find $(n-k,\alpha_{N-k})$-plAnar sets $P_{N-k,1},\dots,p_{N-k,l_{n-k}}$, whERe $l_{n-k}=M(n,\Alpha_{n-1})\ProD_{j=n-K+1}^{n-1}M(j,\ALpHa_{J})$, suCh THat $$\LAbEl{fINal}
a\cap B(x,r)\sUbSeT
\bigcUp_{i=1}^{l_{N-K}}p_{N-K,i}(c_{2}^{k-1}2\DelTa r).$$
IT is noT hard to verify ThaT theRE is A consTant $C(\AlphA,n,K)$ depeNding oNly on $\AlPha$, $n$, and $k$ such thaT eacH of the setS $P_{n-K,i}(C_2^{k-1}2\dElTa r)\caP b(x,r)$ can Be cOveRed with $c(\alpha,n,K)\DelTa^{K-N}$ BAlLs of radius $\delta r$, anD tHEReFore by, $C(\aLpha,n,k | of
which are $(m-1,\alph a_m)$-plan ar. U sin g R em ark\ref {rem:Lip}, wef inda $(V,\alpha_m)$-plana r set $ \ wide h at P$ s uch tha t $ P \ sub se t\ wid eh a tP$ an d
$ \operat orname{pro j}_ V( \widehat P)= V $. The r\^ol e o f $\widehatP$is togu ara n tee t hat
$\wi dehatP \cap\o peratorna me { proj}_ V ^{-1}(P ' _ j) \ne\ emptyset$ for all $j $ .
For all $j \in\{1 ,\ d ot s , M(m ,\a lpha_m)\}$ t he se t s
$\wi d et i l d e{P } _j=\widehat P \cap\operat o rna me{pro j} _V^ { -1}(P' _j)$ar e
$ (m-1,\alpha _{m- 1})$-plan ar, an d moreov e r,
\be gin{eq uat ion *} A \c ap B( x, r )\c a pP(c _ {2} ^{n-m-1} 2\ de lta r )
\s u b s e t\bi gcu p_{j =1}^{ M(m,\alpha_m) } \wid e til de{P} _j(c_ {2}^ {n -m}2\ deltar),$$ s ee Figure \[enl arge ment\]. A s t he re su lt of this i ndu cti ve proc ess wem ayfi n d $( n-k,\alpha_{n-k})$ -p l a na r sets $ P_{n-k , 1} ,\ d ots,P_{n -k ,l_ {n-k } } $, wh ere$ l_ {n-k}=M( n,\alp h a_ {n -1})\pr od _{j=n- k+ 1}^ {n- 1}M(j , \alp ha_{j} )$, such that $$\label{final }
A\cap B(x,r) \ su b s et \big cup _{i=1}^{l_{ n-k} } P_{n -k,i } (c _{2 } ^{k-1 }2\de lt a r ) .$$
It is not hard t o veri fy th at there is a constant$ C ( \alpha,n ,k)$ de p ending only on $\al pha$, $n$, and $k$suchthat eac h of thes e ts $P_{n -k, i}( c_2 ^{k - 1 }2 \delta r)\cap B (x,r )$ can be co vered w ith $C (\a lph a, n,k)\delt a^{k-n}$ b al ls o f r adius $\deltar$ , a nd th erefo r e by,$C(\a lpha ,n ,k | of_
which are_$(m-1,\alpha_m)$-planar. Using Remark \ref{rem:Lip},_we find
a_$(V,\alpha_m)$-planar_set $\widehat_P$_such that $P\subset\widehat_P$ and
$\operatorname{proj}_V(\widehat P)=V$._The r\^ole of $\widehat_P$ is to_guarantee_that
$\widehat P\cap\operatorname{proj}_V^{-1}(P'_j)\ne\emptyset$ for all $j$.
For all $j\in\{1,\dots,M(m,\alpha_m)\}$ the sets
$\widetilde{P}_j=\widehat P\cap\operatorname{proj}_V^{-1}(P'_j)$ are_
$(m-1,\alpha_{m-1})$-planar,_and moreover,_
\begin{equation*}_
A\cap_B(x,r)\cap P(c_{2}^{n-m-1}2\delta r)
\subset\bigcup_{j=1}^{M(m,\alpha_m)}
\widetilde{P}_j(c_{2}^{n-m}2\delta r),$$_see Figure \[enlargement\]. As the_result of_this inductive process we may find $(n-k,\alpha_{n-k})$-planar sets_$P_{n-k,1},\dots,P_{n-k,l_{n-k}}$,_where $l_{n-k}=M(n,\alpha_{n-1})\prod_{j=n-k+1}^{n-1}M(j,\alpha_{j})$, such_that $$\label{final}
A\cap B(x,r)\subset
\bigcup_{i=1}^{l_{n-k}}P_{n-k,i}(c_{2}^{k-1}2\delta r).$$
It is not hard to verify_that there is a constant $C(\alpha,n,k)$_depending only on_$\alpha$,_$n$,_and $k$ such that_each of the sets $P_{n-k,i}(c_2^{k-1}2\delta r)\cap_B(x,r)$ can be covered with $C(\alpha,n,k)\delta^{k-n}$_balls of radius $\delta r$, and therefore_by, $C(\alpha,n,k |
la, “[Wrapped M5-branes, consistent truncations and AdS/CMT]{},” [*JHEP*]{} [**1012**]{} (2010) 003, [[1009.3805]{}](http://arXiv.org/abs/1009.3805).
D. Cassani and P. Koerber, “[Tri-Sasakian consistent reduction]{},” [*JHEP*]{} [**1201**]{} (2012) 086, [[1110.5327]{}](http://arXiv.org/abs/1110.5327). A.-K. Kashani-Poor and R. Minasian, “Towards reduction of type ii theories on su(3) structure manifolds,” [*JHEP*]{} [**03**]{} (2007) 109, [[hep-th/0611106]{}](http://arXiv.org/abs/hep-th/0611106). A.-K. Kashani-Poor, “[Nearly Kaehler Reduction]{},” [*JHEP*]{} [**11**]{} (2007) 026, [[arXiv:0709.4482]{}](http://arXiv.org/abs/arXiv:0709.4482). J. P. Gauntlett and O. Varela, “[Universal Kaluza-Klein reductions of type IIB to N=4 supergravity in five dimensions]{},” [*JHEP*]{} [**1006**]{} (2010) 081, [[1003.5642]{}](http://arXiv.org/abs/1003.5642).
K. Skenderis, M. Taylor, and D. Tsimpis, “[A Consistent truncation of IIB supergravity on manifolds admitting a Sasaki-Einstein structure]{},” [ *JHEP*]{} [**1006**]{} (2010) 025, [[ 1003.5657]{}](http://arXiv.org/abs/1003.5657).
D. Cassani, G. Dall’Agata, and A. F. Faedo, “[Type IIB supergravity on squashed Sasaki-Einstein manifolds]{},” [*JHEP*]{} [**1005**]{} (2010) 094, [[1003.4283]{}](http://arXiv.org/abs/1003.4283).
J. T. Liu, P. Szepietowski, and Z. Zhao, “[Supersymmetric massive truncations of IIb supergravity | la, “ [ Wrapped M5 - branes, consistent truncations and AdS / CMT ] { }, ” [ * JHEP * ] { } [ * * 1012 * * ] { } (2010) 003, [ [ 1009.3805]{}](http://arXiv.org / abs/1009.3805).
D. Cassani and P. Koerber, “ [ Tri - Sasakian consistent decrease ] { }, ” [ * JHEP * ] { } [ * * 1201 * * ] { } (2012) 086, [ [ 1110.5327]{}](http://arXiv.org / abs/1110.5327). A.-K. Kashani - Poor and R. Minasian, “ Towards decrease of character ii theories on su(3) structure manifolds, ” [ * JHEP * ] { } [ * * 03 * * ] { } (2007) 109, [ [ hep - th/0611106]{}](http://arXiv.org / abs / hep - th/0611106). A.-K. Kashani - Poor, “ [ closely Kaehler Reduction ] { }, ” [ * JHEP * ] { } [ * * 11 * * ] { } (2007) 026, [ [ arXiv:0709.4482]{}](http://arXiv.org / abs / arXiv:0709.4482). J. P. Gauntlett and O. Varela, “ [ Universal Kaluza - Klein reductions of type IIB to N=4 supergravity in five proportion ] { }, ” [ * JHEP * ] { } [ * * 1006 * * ] { } (2010) 081, [ [ 1003.5642]{}](http://arXiv.org / abs/1003.5642).
K. Skenderis, M. Taylor, and D. Tsimpis, “ [ A coherent truncation of IIB supergravity on manifold admitting a Sasaki - Einstein structure ] { }, ” [ * JHEP * ] { } [ * * 1006 * * ] { } (2010) 025, [ [ 1003.5657]{}](http://arXiv.org / abs/1003.5657).
D. Cassani, G. Dall’Agata, and A. F. Faedo, “ [ character IIB supergravity on squashed Sasaki - Einstein manifolds ] { }, ” [ * JHEP * ] { } [ * * 1005 * * ] { } (2010) 094, [ [ 1003.4283]{}](http://arXiv.org / abs/1003.4283).
J. T. Liu, P. Szepietowski, and Z. Zhao, “ [ Supersymmetric massive truncations of IIb supergravity | la, “[Arapped M5-branes, consistekt truncations aue AdS/CKT]{},” [*JHEL*]{} [**1012**]{} (2010) 003, [[1009.3805]{}](htto://arXiv.org/abs/1009.3805).
D. Cassani and P. Koxrbee, “[Tri-Wasakian consistent reauction]{},” [*JJEP*]{} [**1201**]{} (2012) 086, [[1110.5327]{}](hrtp://acXiv.org/abs/1110.5327). A.-K. Kashani-Poov and V. Minavman, “Towards redoction of ty[e ii theories ov du(3) structure manifolds,” [*JHEP*]{} [**03**]{} (2007) 109, [[hep-tr/0611106]{}](http://arCig.org/abs/hep-th/0611106). A.-H. Kaxrani-Loor, “[Nearly Kaehler Reduction]{},” [*JHEP*]{} [**11**]{} (2007) 026, [[arXpv:0709.4482]{}](http://arXiv.org/abs/srXiv:0709.4482). J. P. Gauntlett and O. Vagela, “[Universal Kaluza-Kpein reductuons if type IIB go N=4 supergravity in fjve dimensions]{},” [*JHEP*]{} [**1006**]{} (2010) 081, [[1003.5642]{}](http://arXix.org/ays/1003.5642).
K. Skenderiw, N. Tajnor, and D. Tsmmpis, “[W Consistent truncathon of OIB supergraviby on mabifolds admitting a Sesaki-Einstein structore]{},” [ *JHEP*]{} [**1006**]{} (2010) 025, [[ 1003.5657]{}](http://arXiv.org/abs/1003.5657).
E. Cqssanh, G. Ddll’Aeqta, ans E. F. Raedo, “[Hypx IIB superfravity on wquashed Sasaki-Einsueig manifolds]{},” [*JHSP*]{} [**1005**]{} (2010) 094, [[1003.4283]{}](rtep://arXiv.org/abs/1003.4283).
J. T. Liu, P. Szepietowski, and Z. Ehao, “[Aupersymmetric massive rruncations of IIb sukergravity | la, “[Wrapped M5-branes, consistent truncations and AdS/CMT]{},” (2010) [[1009.3805]{}](http://arXiv.org/abs/1009.3805). D. and P. Koerber, (2012) [[1110.5327]{}](http://arXiv.org/abs/1110.5327). A.-K. Kashani-Poor R. Minasian, “Towards of type ii theories on su(3) manifolds,” [*JHEP*]{} [**03**]{} (2007) 109, [[hep-th/0611106]{}](http://arXiv.org/abs/hep-th/0611106). A.-K. Kashani-Poor, “[Nearly Kaehler Reduction]{},” [*JHEP*]{} [**11**]{} 026, [[arXiv:0709.4482]{}](http://arXiv.org/abs/arXiv:0709.4482). J. P. Gauntlett and O. Varela, “[Universal Kaluza-Klein reductions of type to supergravity five [*JHEP*]{} [**1006**]{} (2010) 081, [[1003.5642]{}](http://arXiv.org/abs/1003.5642). K. Skenderis, M. Taylor, and D. Tsimpis, “[A Consistent truncation of supergravity on manifolds admitting a Sasaki-Einstein structure]{},” [ [**1006**]{} (2010) 025, [[ D. Cassani, G. Dall’Agata, and F. “[Type IIB on Sasaki-Einstein [*JHEP*]{} [**1005**]{} (2010) [[1003.4283]{}](http://arXiv.org/abs/1003.4283). J. T. Liu, P. Szepietowski, and Z. Zhao, “[Supersymmetric massive truncations of IIb supergravity | la, “[Wrapped M5-branes, consistenT truncatioNs and adS/cMT]{},” [*jHeP*]{} [**1012**]{} (2010) 003, [[1009.3805]{}](htTp://arxiv.org/abs/1009.3805).
D. CassANi anD P. Koerber, “[Tri-Sasakian coNsistEnT ReduCTiOn]{},” [*JHEp*]{} [**1201**]{} (2012) 086, [[1110.5327]{}](http://arxIv.ORG/abS/1110.5327). A.-k. KAshAnI-poOr and r. MiNasian, “TOwards reduCtiOn Of type ii theoRIeS on su(3) strucTurE manifolds,” [*JHeP*]{} [**03**]{} (2007) 109, [[hEp-th/0611106]{}](htTp://ArXIV.org/aBs/hEp-th/0611106). A.-k. KashaNI-Poor, “[NEarly KaehLeR reductIOn]{},” [*JHEP*]{} [**11**]{} (2007) 026, [[aRxIv:0709.4482]{}](Http://ArXiv.org/abs/arXiv:0709.4482). J. p. gaUNtlett and O. VareLa, “[UnivErSAl kALuzA-KlEin reductiOnS of tyPE IIB to N=4 SUpERGRavITy in five dimenSions]{},” [*JHEP*]{} [**1006**]{} (2010) 081, [[1003.5642]{}](htTP://arxiv.org/AbS/1003.5642).
K. SKEnderiS, M. TayLoR, And d. Tsimpis, “[A CoNsisTent truncAtion oF iIB supeRGravity On maniFolDs aDmitTInG a sasAkI-einSTeIn sTRucTure]{},” [ *JHEP*]{} [**1006**]{} (2010) 025, [[ 1003.5657]{}](HtTp://ArXiv.Org/aBS/1003.5657).
d. cAssaNi, G. dall’agata, And A. F. Faedo, “[TypE IIb supERgrAvity On squAsheD SAsaki-einsteIn manIfOlds]{},” [*JHEP*]{} [**1005**]{} (2010) 094, [[1003.4283]{}](http://arXIv.orG/abs/1003.4283).
J. T. Liu, p. SzEpIetOwSki, anD z. Zhao, “[SUpeRsyMmetric Massive TRunCaTIONs Of IIb supergravity | la, “[Wrapped M5-branes, c onsistenttrunc ati ons a nd A dS/C MT]{},” [*JHEP * ]{}[**1012**]{} (2010) 00 3, [[ 10 0 9.38 0 5] {}](h ttp://a r Xi v . org /a bs /10 09 . 38 05).
D. Cassan i and P. K oer be r, “[Tri-Sas a ki an consist ent reduction]{ },” [*JHE P* ]{} [**12 01* *]{}(2012) 086, [ [1110.532 7] { }](htt p ://arXi v . or g/ab s/1110.5327). A.- K .K ashani-Poor an d R. M in a si a n , “ Tow ards reduc ti on of type ii th e o r ies on su(3) stru cture manif o lds ,” [*J HE P*] { } [**0 3**]{ }( 200 7) 109, [[h ep-t h/0611106 ]{}](h t tp://ar X iv.org/ abs/he p-t h/0 6111 0 6) .A.- K. Kas h an i-P o or, “[Nearl yKa ehler Red u c t i on]{ },” [*J HEP*] {} [**11**]{} (2 007) 026 , [[a rXiv: 0709 .4 482]{ }](htt p://a rX iv.org/abs/arXi v:07 09.4482). J. P . G au ntlet t and O . V are la, “[U niversa l Ka lu z a - Kl ein reductions ofty p e I IB to N= 4 supe r gr av i ty in fi ve di mens i o ns]{} ,” [ * JH EP*]{} [ **1006 * *] {} (2010) 0 81, [[ 10 03. 564 2]{}] ( http ://arX iv.org/a bs/10 0 3.5642).
K. S k enderis, M. T a yl o r ,a nd D . T simpis, “[A Con s iste nt t r un cat i on of IIBsu p er g ravity on manifolds a dmitti ng aSasaki-Einste in structu r e ] {},” [ * JHEP * ]{ } [**1006**]{}(2010 ) 025, [[1 003.5657 ]{}]( http://a rXiv.org/ a b s/1003.5 657 ).
D. Ca s s an i, G. Dall’Ag a t a, a nd A. F.Fae do, “[T ype II B s upe rg ravity on squashe dSa sa ki -Ei nstei n manifol ds ]{} ,” [* JHEP* ] {} [** 1005* *]{} ( 20 1 0)094, [[ 1 00 3 . 4283 ]{ }] (htt p:/ /a rXiv. org/ a bs/ 1003.42 83).
J.T.L iu,P. S zepieto wski, and Z.Zh ao, “[Supe rs ymm etricm a ssive tr uncations of IIb superg r avity | la, “[Wrapped_M5-branes, consistent_truncations and AdS/CMT]{},” [*JHEP*]{}_[**1012**]{} (2010)_003,_[[1009.3805]{}](http://arXiv.org/abs/1009.3805).
D. Cassani and_P. Koerber,_“[Tri-Sasakian consistent reduction]{},”_[*JHEP*]{} [**1201**]{} (2012)_086, [[1110.5327]{}](http://arXiv.org/abs/1110.5327). A.-K. Kashani-Poor_and R. Minasian, “Towards_reduction_of type ii theories on su(3) structure manifolds,” [*JHEP*]{} [**03**]{} (2007) 109, [[hep-th/0611106]{}](http://arXiv.org/abs/hep-th/0611106). A.-K._Kashani-Poor,_“[Nearly Kaehler_Reduction]{},”_[*JHEP*]{}_[**11**]{} (2007) 026, [[arXiv:0709.4482]{}](http://arXiv.org/abs/arXiv:0709.4482). J. P._Gauntlett and O. Varela, “[Universal Kaluza-Klein_reductions of_type IIB to N=4 supergravity in five dimensions]{},”_[*JHEP*]{}_[**1006**]{} (2010) 081,_[[1003.5642]{}](http://arXiv.org/abs/1003.5642).
K. Skenderis, M. Taylor, and D. Tsimpis, “[A Consistent truncation of IIB_supergravity on manifolds admitting a Sasaki-Einstein_structure]{},” [ *JHEP*]{}_[**1006**]{}_(2010)_025, [[ 1003.5657]{}](http://arXiv.org/abs/1003.5657).
D. Cassani, G. Dall’Agata,_and A. F. Faedo, “[Type IIB supergravity_on squashed Sasaki-Einstein manifolds]{},” [*JHEP*]{} [**1005**]{}_(2010) 094, [[1003.4283]{}](http://arXiv.org/abs/1003.4283).
J. T. Liu, P. Szepietowski, and Z. Zhao,_“[Supersymmetric massive truncations of IIb supergravity |
Errington, and Thomas M Truskett. Inverse design of simple pairwise interactions with low-coordinated 3d lattice ground states., 9(14):3866–3870, 2013.
Fran[ç]{}ois A Detcheverry, Huiman Kang, Kostas Ch Daoulas, Marcus M[ü]{}ller, Paul F Nealey, and Juan J de Pablo. Monte carlo simulations of a coarse grain model for block copolymers and nanocomposites., 41(13):4989–5001, 2008.
Chi-Chun Liu, Abelardo Ram[í]{}rez-Hern[á]{}ndez, Eungnak Han, Gordon SW Craig, Yasuhiko Tada, Hiroshi Yoshida, Huiman Kang, Shengxiang Ji, Padma Gopalan, Juan J de Pablo, et al. Chemical patterns for directed self-assembly of lamellae-forming block copolymers with density multiplication of features., 46(4):1415–1424, 2013.
Sang Ouk Kim, Harun H Solak, Mark P Stoykovich, Nicola J Ferrier, Juan J de Pablo, and Paul F Nealey. Epitaxial self-assembly of block copolymers on lithographically defined nanopatterned substrates., 424(6947):411–414, 2003.
Mark P Stoykovich, Huiman Kang, Kostas Ch Daoulas, Guoliang Liu, Chi-Chun Liu, Juan J de Pablo, Marcus M[ü]{}ller, and Paul F Nealey. Directed self-assembly of block copolymers for nanolithography: fabrication of isolated features and essential integrated circuit geometries., 1(3):168–175, 2007.
Ulli Wolff. Collective [Monte Carlo]{} updating for spin systems., 62(4):361, 1989.
A Baumg[ä]{}rtner. Statics and dynamics of the freely jointed polymer chain with [Lennard-Jones]{} interaction., 72(2):871–879, 1980.
A Baumg[ä]{}rtner and K Binder. studies on the freely jointed polymer chain with excluded volume interaction., 71(6):2541–2545, 1979 | Errington, and Thomas M Truskett. Inverse design of simple pairwise interactions with low - coordinated three-d wicket ground states. , 9(14):3866–3870, 2013.
Fran[ç]{}ois A Detcheverry, Huiman Kang, Kostas Ch Daoulas, Marcus M[ü]{}ller, Paul F Nealey, and Juan J de Pablo. Monte carlo model of a coarse texture model for block copolymers and nanocomposites. , 41(13):4989–5001, 2008.
Chi - Chun Liu, Abelardo Ram[í]{}rez - Hern[á]{}ndez, Eungnak Han, Gordon SW Craig, Yasuhiko Tada, Hiroshi Yoshida, Huiman Kang, Shengxiang Ji, Padma Gopalan, Juan J de Pablo, et al. Chemical pattern for directed self - assembly of lamellae - forming pulley copolymers with density generation of features. , 46(4):1415–1424, 2013.
Sang Ouk Kim, Harun H Solak, Mark P Stoykovich, Nicola J Ferrier, Juan J de Pablo, and Paul F Nealey. Epitaxial self - assembly of block copolymers on lithographically define nanopatterned substrates. , 424(6947):411–414, 2003.
Mark P Stoykovich, Huiman Kang, Kostas Ch Daoulas, Guoliang Liu, Chi - Chun Liu, Juan J de Pablo, Marcus M[ü]{}ller, and Paul F Nealey. direct self - assembly of block copolymer for nanolithography: fabrication of isolated features and substantive integrated circuit geometry. , 1(3):168–175, 2007.
Ulli Wolff. Collective [ Monte Carlo ] { } updating for tailspin systems. , 62(4):361, 1989.
A Baumg[ä]{}rtner. Statics and dynamics of the freely jointed polymer chain with [ Lennard - Jones ] { } interaction. , 72(2):871–879, 1980.
A Baumg[ä]{}rtner and K Binder. studies on the freely jointed polymer chain with exclude volume interaction. , 71(6):2541–2545, 1979 | Ergington, and Thomas M Trusyett. Inverse design of simpls pairwire interactions with low-coorvinared 3d lattice ground states., 9(14):3866–3870, 2013.
Fran[ç]{}ois W Detchecerrb, Huiman Kang, Kostas Ch Daoulas, Larcbs M[ü]{}ller, Paul F Ngaley, and Judn J de Pablo. Mottd earlo simulations of a coarse grain iodel fpr block copolymgrs amq nahocomposites., 41(13):4989–5001, 2008.
Chi-Chun Liu, Abelardo Ram[í]{}rez-Iern[á]{}ndez, Eungnal Han, Gordon SW Craig, Yasuhlko Hada, Hiroshi Yoshifa, Huiman Kqng, Fyengxiang Ji, Padma Gopalan, Juan J dg Pablo, et al. Chemical patterns fof dirzcted self-awswmbpi of lamellax-formigg block copolymers fith demsity multipligatioi of features., 46(4):1415–1424, 2013.
Sang Ouk Kik, Harun H Solak, Marh P Stoykoeieh, Nicola J Ferrier, Juqn H de Pdblo, and Pauu F Heeleg. Epitwxiel self-assejbly of bloxk copolymers on liuhodgsphically derined gagopatterned substrates., 424(6947):411–414, 2003.
Mark P Stoykovicv, Hhiman Kang, Kostas Ch Daoylas, Guoliang Liu, Chi-Fhun Liu, Tuan J de Pablo, Marcus M[ü]{}ller, and Paul F Nealey. Direcded sxlw-asweicoy of block copolymers for nanolithography: fabrydauiok of isolated fectures and essenyiwl yntegrated citcuit gzkmstries., 1(3):168–175, 2007.
Ulli Wolff. Collecjive [Minte Carlj]{} upcating for spin systems., 62(4):361, 1989.
A Bqumg[ä]{}rtner. Stctixs and dynamics of the freely joimted lolymer chain with [Lenncrd-Jonss]{} interactlon., 72(2):871–879, 1980.
A Bauje[ä]{}rtner and K Binddr. xtgdies on the freely jointeq polymer chann with dxcloded vojume interwctiok., 71(6):2541–2545, 1979 | Errington, and Thomas M Truskett. Inverse design pairwise with low-coordinated lattice ground states., Huiman Kostas Ch Daoulas, M[ü]{}ller, Paul F and Juan J de Pablo. Monte simulations of a coarse grain model for block copolymers and nanocomposites., 41(13):4989–5001, 2008. Liu, Abelardo Ram[í]{}rez-Hern[á]{}ndez, Eungnak Han, Gordon SW Craig, Yasuhiko Tada, Hiroshi Yoshida, Huiman Shengxiang Padma Juan de Pablo, et al. Chemical patterns for directed self-assembly of lamellae-forming block copolymers with density multiplication features., 46(4):1415–1424, 2013. Sang Ouk Kim, Harun H Mark P Stoykovich, Nicola Ferrier, Juan J de Pablo, Paul Nealey. Epitaxial of copolymers lithographically defined nanopatterned 424(6947):411–414, 2003. Mark P Stoykovich, Huiman Kang, Kostas Ch Daoulas, Guoliang Liu, Chi-Chun Liu, Juan J de Marcus M[ü]{}ller, F Nealey. self-assembly block for nanolithography: fabrication features and essential integrated circuit geometries., Wolff. Collective [Monte Carlo]{} updating for spin systems., 1989. A Statics and dynamics of the freely polymer chain with [Lennard-Jones]{} interaction., 72(2):871–879, 1980. A and K Binder. studies on the freely jointed polymer chain with excluded volume interaction., 71(6):2541–2545, | Errington, and Thomas M TruskeTt. Inverse dEsign Of sImpLe PairWise Interactions wiTH low-Coordinated 3d lattice groUnd stAtES., 9(14):3866–3870, 2013.
FraN[Ç]{}oIs A DeTcheverRY, HUIMan kaNg, kosTaS ch daoulAs, MArcus M[ü]{}Ller, Paul F NEalEy, And Juan J de PaBLo. monte carlo SimUlations of a cOarSe graiN mOdeL For blOck CopolYmers aND nanocOmposites., 41(13):4989–5001, 2008.
chI-chun LiU, abelardO rAm[Í]{}rez-hern[á]{}ndez, Eungnak HAN, GORdon SW Craig, YasUhiko TAdA, hiROShi yosHida, Huiman kaNg, SheNGxiang JI, paDMA gopALan, Juan J de PabLo, et al. ChemiCAl pAtternS fOr dIRected Self-aSsEMblY of lamellae-FormIng block cOpolymERs with dENsity muLtipliCatIon Of feATuReS., 46(4):1415–1424, 2013.
SaNg oUk KIM, HAruN h SoLak, Mark P stOyKovicH, NicOLA j ferrIer, juan j de PaBlo, and Paul F NeAleY. EpiTAxiAl selF-asseMbly Of Block CopolyMers oN lIthographically DefiNed nanopaTteRnEd sUbStratES., 424(6947):411–414, 2003.
Mark P stoYkoVich, HuiMan Kang, kOstAs cH dAoUlas, Guoliang Liu, Chi-chUN liU, Juan J de pablo, MARcUs m[Ü]{}ller, and paUl F nealEY. direcTed sELf-Assembly Of blocK CoPoLymers fOr NanoliThOgrAphY: fabrICatiOn of isOlated feAtureS And essential inTEgrated circuiT GeOMEtRIes., 1(3):168–175, 2007.
ULli wolff. CollecTive [mOnte carlO]{} UpDatINg for Spin sYsTEmS., 62(4):361, 1989.
a Baumg[ä]{}rtner. Statics AnD dynamIcs of The freely joinTed polymer CHAIn with [LeNnarD-joNEs]{} interaction., 72(2):871–879, 1980.
A baumg[Ä]{}rtner and K bInder. stuDies oN the freeLy jointed POLymer chaIn wIth ExcLudED VoLume interactiON., 71(6):2541–2545, 1979 | Errington, and Thomas M T ruskett. I nvers e d esi gn ofsimp le pairwise in t erac tions with low-coordin ated3d latt i ce grou nd stat e s. , 9(1 4) :3 866 –3 8 70 , 201 3.
Fran[ç ]{}ois A D etc he verry, Huima n K ang, Kosta s C h Daoulas, M arc us M[ü ]{ }ll e r, Pa ulF Nea ley, a n d Juan J de Pab lo . Monte carlo s i m ul atio ns of a coarse gr a in model for bloc k copo ly m er s and na nocomposit es ., 41 ( 13):498 9 –5 0 0 1 , 2 0 08.
Chi-Chun Liu, Abela r doRam[í] {} rez - Hern[á ]{}nd ez , Eu ngnak Han,Gord on SW Cra ig, Ya s uhiko T a da, Hir oshi Y osh ida , Hu i ma nKan g, She n gx ian g Ji , PadmaGo pa lan,Juan J d e Pa blo , et al.Chemical patt ern s fo r di recte d sel f-as se mblyof lam ellae -f orming block co poly mers with de ns ity m ultip l icatio n o f f eatures ., 46(4 ) :14 15 – 1 4 24 , 2013.
Sang OukKi m , H arun H S olak,M ar kP Stoykov ic h,Nico l a J Fe rrie r ,Juan J d e Pabl o ,an d PaulFNealey .Epi tax ial s e lf-a ssembl y of blo ck co p olymers on lit h ographicallyd ef i n ed nano pat terned subs trat e s.,424( 6 94 7): 4 11–41 4, 20 03 .
M ark P Stoykovich, H ui man Ka ng, K ostas Ch Daou las, Guoli a n g Liu, Ch i-Ch u nL iu, Juan J dePablo , Marcus M [ ü]{}ller , and Paul FNealey. D i r ected se lf- ass emb lyo f b lock copolyme r s for n anolith ogr aphy: f abr ica tio n o fisolatedfeatures a nd e ss ent ial i n tegrated c irc ui t g eomet r ies.,1(3): 168– 17 5, 200 7.
Ull i W o l ff.Co ll ecti ve[M onteCarl o ]{} updati ng for sp ins yste ms ., 62(4): 361, 1989.
A B aumg[ä]{}r tn er. Stati c s and dyn amics of the freely joi n ted pol yme r cha in w ith [Lenn ard -Jones ]{} intera ction. , 72( 2) :87 1 – 879,1 9 80 .
ABaumg[ä]{} r t ner andKBind er. stu dies on the freely joi nted polymercha in w i t hexc l ud e d v ol u mei n teraction., 71( 6):2541–25 45 , 1 979 | Errington,_and Thomas M_Truskett. Inverse design of_simple pairwise_interactions_with low-coordinated_3d_lattice ground states.,_9(14):3866–3870, 2013.
Fran[ç]{}ois A Detcheverry,_Huiman Kang, Kostas Ch Daoulas,_Marcus M[ü]{}ller, Paul F_Nealey,_and Juan J de Pablo. Monte carlo simulations of a coarse grain model for block copolymers_and_nanocomposites., 41(13):4989–5001,_2008.
Chi-Chun_Liu,_Abelardo Ram[í]{}rez-Hern[á]{}ndez, Eungnak Han, Gordon SW_Craig, Yasuhiko Tada, Hiroshi Yoshida,_Huiman Kang,_Shengxiang Ji, Padma Gopalan, Juan J de Pablo, et al. Chemical_patterns_for directed self-assembly_of lamellae-forming block copolymers with density multiplication of features.,_46(4):1415–1424, 2013.
Sang Ouk Kim, Harun H Solak, Mark P_Stoykovich, Nicola J Ferrier,_Juan J_de Pablo,_and Paul F Nealey. Epitaxial_self-assembly of block copolymers on lithographically_defined nanopatterned substrates., 424(6947):411–414, 2003.
Mark P Stoykovich,_Huiman Kang, Kostas Ch Daoulas, Guoliang Liu, Chi-Chun_Liu, Juan J de Pablo, Marcus M[ü]{}ller, and_Paul F Nealey. Directed self-assembly of_block copolymers_for nanolithography: fabrication of isolated_features and essential_integrated circuit_geometries., 1(3):168–175, 2007.
Ulli_Wolff. Collective [Monte Carlo]{} updating for_spin systems., 62(4):361,_1989.
A Baumg[ä]{}rtner. Statics and dynamics of the_freely_jointed polymer chain_with_[Lennard-Jones]{}_interaction., 72(2):871–879,_1980.
A Baumg[ä]{}rtner and K Binder._studies_on the_freely_jointed polymer chain with excluded volume_interaction.,_71(6):2541–2545, 1979 |
ideal, it follows that $-1 \notin I_J$.
If $\varepsilon \in I_J$ then $\varepsilon/(1 + \varepsilon) \in I_J$.
We may assume $\varepsilon \ne 0$. Using Lemma \[contraction\].\[c-1\] choose a basis $\{A_1,\ldots,A_r\}$ of quasi-atoms over $J$ such that $\varepsilon
\cdot A_i \subseteq J$. For each $i$ choose $a_i \in A_i
\setminus J$. Then $\varepsilon \cdot a_i \in J$, so $(1 +
\varepsilon) \cdot a_i \in A_i \setminus J$. Let $\beta = (1 +
\varepsilon)/\varepsilon$. Then $$\begin{aligned}
\beta \cdot (\varepsilon \cdot a_i) & \in \beta \cdot J \\
(\beta \cdot \varepsilon) \cdot a_i = (1 + \varepsilon) \cdot a_i &
\in A_i \setminus J.
\end{aligned}$$ In particular $$(\beta \cdot J) \cap A_i \not \subseteq J,$$ for every $i$, so $\beta \cdot J \supseteq J$ by Proposition \[special-proposition\].\[guards\]. Then $(\beta \cdot J) \cap A_i \supsetneq J$ for every $i$, so $\beta
\cdot J$ dominates $J$ by Lemma \[characterization-of-domination\]. This means that $\beta^{-1} = \varepsilon/(1 + \varepsilon)$ lies in $I_J$.
Now if $\varepsilon \in I_J$, then $$\frac{1}{1 + \varepsilon} = 1 - \frac{\varepsilon}{1 + \varepsilon} \in 1
+ I_J \subseteq R_J. \qedhere$$
Proposition \[rings-and-ideals\].\[cap-not-vee\] also holds for $R_{J_1 + J_2}$ and $I_{J_1 + J_2}$.
In Proposition \[rings-and-ideals\].\[ij-im\], not only is $I_M$ a subset of $I_J$, it is a sub*ideal* in the ring $R_J | ideal, it follows that $ -1 \notin I_J$.
If $ \varepsilon \in I_J$ then $ \varepsilon/(1 + \varepsilon) \in I_J$.
We may assume $ \varepsilon \ne 0$. Using Lemma \[contraction\].\[c-1\ ] choose a basis $ \{A_1,\ldots, A_r\}$ of quasi - atom over $ J$ such that $ \varepsilon
\cdot A_i \subseteq J$. For each $ i$ choose $ a_i \in A_i
\setminus J$. Then $ \varepsilon \cdot a_i \in J$, therefore $ (1 +
\varepsilon) \cdot a_i \in A_i \setminus J$. Let $ \beta = (1 +
\varepsilon)/\varepsilon$. Then $ $ \begin{aligned }
\beta \cdot (\varepsilon \cdot a_i) & \in \beta \cdot J \\
(\beta \cdot \varepsilon) \cdot a_i = (1 + \varepsilon) \cdot a_i &
\in A_i \setminus J.
\end{aligned}$$ In particular $ $ (\beta \cdot J) \cap A_i \not \subseteq J,$$ for every $ i$, so $ \beta \cdot J \supseteq J$ by Proposition \[special - proposition\].\[guards\ ]. Then $ (\beta \cdot J) \cap A_i \supsetneq J$ for every $ i$, then $ \beta
\cdot J$ dominates $ J$ by Lemma \[characterization - of - domination\ ]. This means that $ \beta^{-1 } = \varepsilon/(1 + \varepsilon)$ lie in $ I_J$.
immediately if $ \varepsilon \in I_J$, then $ $ \frac{1}{1 + \varepsilon } = 1 - \frac{\varepsilon}{1 + \varepsilon } \in 1
+ I_J \subseteq R_J. \qedhere$$
Proposition \[rings - and - ideals\].\[cap - not - vee\ ] also apply for $ R_{J_1 + J_2}$ and $ I_{J_1 + J_2}$.
In Proposition \[rings - and - ideals\].\[ij - im\ ], not only is $ I_M$ a subset of $ I_J$, it is a sub*ideal * in the ring $ R_J | idfal, it follows that $-1 \notln I_J$.
If $\vareksulon \ii I_J$ thsn $\vareprilon/(1 + \varepsilon) \in I_J$.
Wx mat assyme $\varepsilon \ne 0$. Usivg Lemma \[clntractiin\].\[c-1\] rhoose a basis $\{A_1,\ldots,A_r\}$ of quaal-atomv over $J$ such tmat $\varepsinon
\cgog C_i \subseteq J$. For each $i$ choose $a_i \ig A_i
\setminus T$. Thtn $\darelsilon \cdot a_i \in J$, so $(1 +
\barepsinon) \cdot a_i \im A_i \setminus J$. Let $\beta = (1 +
\varepsilon)/\vwrepsilon$. Tyen $$\fwgin{aligned}
\beta \cdot (\vzrepsilon \cdot a_i) & \in \beta \cdot J \\
(\bwta \wdot \varepsmlon) \cqot a_i = (1 + \vavvpsilon) \cdot a_o &
\ln A_i \serminus J.
\end{eligned}$$ In particulat $$(\beta \cdod L) \cap A_i \not \subseteq J,$$ for gvery $i$, su $\bega \ddpt J \supdetxq J$ by Prolosition \[spexial-proposition\].\[guarcs\]. Nnen $(\beta \cdof J) \ca[ W_i \supsetneq J$ for every $i$, so $\beta
\cdot J$ dominates $J$ by Lemma \[characterizatioj-of-dominaeion\]. This means that $\beta^{-1} = \varepsilon/(1 + \varepsilon)$ lies kn $N_M$.
Jow if $\varepsilon \in I_J$, then $$\frac{1}{1 + \varepsilon} = 1 - \fgac{\varepsilon}{1 + \vavepsilon} \in 1
+ Y_J \subseteq R_G. \qedhzde$$
Lroposition \[rings-anf-ideals\].\[sap-nor-vee\] also holcs for $R_{J_1 + J_2}$ and $I_{J_1 + J_2}$.
In Priposition \[rinys-abd-ideals\].\[ij-im\], not ouly is $I_M$ a rubsgt of $O_J$, it is a sub*ideal* in che rihg $R_J | ideal, it follows that $-1 \notin I_J$. \in then $\varepsilon/(1 \varepsilon) \in I_J$. 0$. Lemma \[contraction\].\[c-1\] choose basis $\{A_1,\ldots,A_r\}$ of over $J$ such that $\varepsilon \cdot \subseteq J$. For each $i$ choose $a_i \in A_i \setminus J$. Then $\varepsilon a_i \in J$, so $(1 + \varepsilon) \cdot a_i \in A_i \setminus J$. $\beta (1 \varepsilon)/\varepsilon$. $$\begin{aligned} \beta \cdot (\varepsilon \cdot a_i) & \in \beta \cdot J \\ (\beta \cdot \varepsilon) \cdot = (1 + \varepsilon) \cdot a_i & \in \setminus J. \end{aligned}$$ In $$(\beta \cdot J) \cap A_i \subseteq for every so \cdot \supseteq J$ by \[special-proposition\].\[guards\]. Then $(\beta \cdot J) \cap A_i \supsetneq J$ for every $i$, so $\beta \cdot J$ dominates by Lemma means that = + lies in $I_J$. $\varepsilon \in I_J$, then $$\frac{1}{1 + - \frac{\varepsilon}{1 + \varepsilon} \in 1 + I_J R_J. \qedhere$$ \[rings-and-ideals\].\[cap-not-vee\] also holds for $R_{J_1 + and $I_{J_1 + J_2}$. In Proposition \[rings-and-ideals\].\[ij-im\], not is $I_M$ a subset of $I_J$, it is a sub*ideal* in the ring $R_J | ideal, it follows that $-1 \notin I_J$.
if $\varepsilOn \in I_j$ thEn $\vArEpsiLon/(1 + \vArepsilon) \in I_J$.
WE May aSsume $\varepsilon \ne 0$. Using lemma \[CoNTracTIoN\].\[c-1\] choOse a basIS $\{A_1,\LDOts,a_r\}$ Of QuaSi-AToMs oveR $J$ sUch that $\Varepsilon
\CdoT A_I \subseteq J$. FoR EaCh $i$ choose $a_I \in a_i
\setminus J$. THen $\VarepsIlOn \cDOt a_i \iN J$, sO $(1 +
\varePsilon) \CDot a_i \iN A_i \setminUs j$. let $\betA = (1 +
\VarepsiLON)/\vArepSilon$. Then $$\begin{aliGNeD}
\Beta \cdot (\varepsIlon \cdOt A_I) & \iN \BEta \CdoT J \\
(\beta \cdot \VaRepsiLOn) \cdot a_I = (1 + \VaREPSilON) \cdot a_i &
\in A_i \seTminus J.
\end{aLIgnEd}$$ In paRtIcuLAr $$(\beta \Cdot J) \CaP a_i \nOt \subseteq J,$$ For eVery $i$, so $\beTa \cdot j \SupseteQ j$ by PropOsitioN \[spEciAl-prOPoSiTioN\].\[gUArdS\]. thEn $(\bETa \cDot J) \cap A_I \sUpSetneQ J$ foR EVERy $i$, sO $\beTa
\cdOt J$ doMinates $J$ by LemMa \[cHaraCTerIzatiOn-of-dOminAtIon\]. ThIs meanS that $\BeTa^{-1} = \varepsilon/(1 + \varEpsiLon)$ lies in $i_J$.
NOw If $\vArEpsilON \in I_J$, tHen $$\FraC{1}{1 + \varepsIlon} = 1 - \fraC{\VarEpSILOn}{1 + \Varepsilon} \in 1
+ I_J \subsEtEQ r_J. \Qedhere$$
PRoposiTIoN \[rINgs-and-idEaLs\].\[cAp-noT-VEe\] alsO holDS fOr $R_{J_1 + J_2}$ and $i_{J_1 + J_2}$.
In PROpOsItion \[riNgS-and-idEaLs\].\[iJ-im\], Not onLY is $I_m$ a subsEt of $I_J$, it Is a suB*Ideal* in the ring $r_j | ideal, it follows that $- 1 \notin I _J$.
I f$\va reps ilon \in I_J$t hen$\varepsilon/(1 + \var epsil on ) \in I_ J$.
Wem ay a ssu me $ \va re p si lon \ ne0$. Usi ng Lemma \ [co nt raction\].\[ c -1 \] choosea b asis $\{A_1, \ld ots,A_ r\ }$o f qua si- atoms over$ J$ suc h that $\ va r epsilo n
\cdo t A_i \subseteq J $ .F or each $i$ ch oose $ a_ i \ i n A_ i
\ se tminu s J$. Th e n$ \ v are p silon \cdot a _i \in J$,s o $ (1 +
\va repsi lo n ) \ cdot a_i \i n A_ i \setmin us J$. Let $\b e ta = (1 +
\va r ep si lon )/ \ var e ps ilo n $.Then $$\ be gi n{ali gned } \ beta\cdot (\varep sil on \ c dot a_i) & \i n \b et a \cd ot J \ \
(\beta\cdo t \vareps ilo n) \c do t a_i = (1 + \v are psilon) \cdota _i& \in A_i \se tm i n us J.
\ en d{ a ligned}$ $Inpart i c ular$$(\ b et a \cdotJ) \ca p A _i \not \ su bseteq J ,$$ fo r eve r y $i $, so$\beta \ cdotJ \supseteq J$b y Proposition \[ s p ec i al-p rop osition\].\ [gua r ds\] . Th e n$(\ b eta \ cdotJ) \c a p A_i \supsetneq J$ f or eve ry $i $, so $\beta \ c d ot J$ do mina t es $J$ by Lemma \ [char acterizati o n-of-dom inati on\]. Th is meanst h at $\bet a^{ -1} =\va r e ps ilon/(1 + \va r e psil on )$ lies in $I_J$.
Now if $ \varepsil on \in I _J $, t he n $ $\fra c {1}{1 +\v are ps ilo n} =1 - \fr ac{\v arep si lo n }{1 + \var e ps i l on}\i n1
+I_J\ sub seteq R _J. \qedh ere $ $
P ro po sition\[rings-and-i de als\].\[ca p- not -vee\] a lso hold s for $R_{J_1 + J_2}$ a n d $I_{J _1+ J_2 }$.
In Propo sit ion \[ rin g s-and- ideals \].\[ ij -im \ ] , not o nl y i s$I_M$ a su b s etof $I _J $, i t is asub*ideal* in ther ing $R_J | ideal,_it follows_that $-1 \notin I_J$.
_ __If $\varepsilon_\in_I_J$ then $\varepsilon/(1_+ \varepsilon) \in_I_J$.
_We may assume_$\varepsilon_\ne 0$. Using Lemma \[contraction\].\[c-1\] choose a basis $\{A_1,\ldots,A_r\}$ of quasi-atoms over $J$ such that_$\varepsilon
_ ___ _\cdot A_i \subseteq J$. For_each $i$_choose $a_i \in A_i
__ _ \setminus J$. Then $\varepsilon \cdot a_i \in J$,_so $(1 +
_ ___ \varepsilon) \cdot a_i_\in A_i \setminus J$. Let $\beta_= (1 +
_ \varepsilon)/\varepsilon$._Then $$\begin{aligned}
_ _ _\beta \cdot (\varepsilon \cdot a_i)_& \in \beta_\cdot J_\\
_ _ _ (\beta \cdot \varepsilon) \cdot a_i_=_(1 + \varepsilon)_\cdot_a_i_&
_ __ __ \in A_i \setminus_J.
_ _ \end{aligned}$$ In_particular_$$(\beta \cdot J) \cap_A_i \not \subseteq J,$$ for_every $i$, so $\beta \cdot J_\supseteq J$_by Proposition \[special-proposition\].\[guards\]._Then $(\beta \cdot J) \cap A_i \supsetneq J$ for every $i$,_so $\beta
_ _ \cdot_J$_dominates $J$ by_Lemma \[characterization-of-domination\]._This means_that $\beta^{-1} = \varepsilon/(1 + \varepsilon)$ lies_in $I_J$.
_ Now if $\varepsilon_\in I_J$, then $$\frac{1}{1_+_\varepsilon} = 1 - \frac{\varepsilon}{1 +_\varepsilon} \in 1
_ __+_I_J \subseteq R_J. \qedhere$$
Proposition \[rings-and-ideals\].\[cap-not-vee\] also_holds for $R_{J_1 + J_2}$ and_$I_{J_1 + J_2}$.
In_Proposition \[rings-and-ideals\].\[ij-im\], not only is $I_M$ a subset_of_$I_J$, it is a sub*ideal* in_the_ring $R_J |
) }$, which is the phase coherence length due to the commensurability. We make the Lagrangian dimensionless through scaling by this $\xi$; i.e. $x =X/\xi$, $\cal L \mit = M^2 g \xi L$, where $L$ is given by $$L = - \int {\rm d}x \left[ \frac{1}{2}\left(\frac{\partial\phi}{\partial x} \right)^2
- \varepsilon\phi + \frac{1}{M^2}\left(1 - \cos(M \phi) \right)
- v\cos(\chi + \phi)\delta(x-x_{i}) \right]. \label{eq:2.4}$$ Here $\varepsilon = F/(g M^2)$, $v = V_{i}/(g \xi M^2)$. In this paper, we assume $\varepsilon \geq 0 $. Further, we define $x_{i}=X_{i}/\xi$ and $\chi= (2 \pi z)/M $, where $z=X_{i}/a $ characterizes the location of an impurity relative to the site where the energy gain by commensurability is maximum. The range of $\chi$ is $-\pi/M \leq \chi \leq \pi/M$, because the Lagrangian has the periodicity of $2\pi/M$ with respect to the phase.
The Homogeneous Case
====================
First of all, we examine the case of $v=0$, namely the homogeneous case. [@rf:3]
The stable configuration of the phase is determined by varying Lagrangian; $$- \phi'' - \varepsilon + \frac{1}{M}\sin(M \phi)=0. \label{eq:3.1}$$ When $\varepsilon = 0$, eq. (\[eq:3.1\]) is the sine-Gordon equation, and has two kinds of solutions, which are a trivial one, $\phi=0$, and the kink solution, which is given by $$\phi(x)= \frac{4}{M}\arctan\left[\exp(\pm(x-x_{0})\right] \equiv \phi_{\pm}(x-x_{0}),
\label{eq:3.3}$$ and shown in Fig. \[fig:kink\].
Here $x_{0}$ is the center of the kink | ) } $, which is the phase coherence length due to the commensurability. We make the Lagrangian dimensionless through scaling by this $ \xi$; i.e. $ x = X/\xi$, $ \cal L \mit = M^2 g \xi L$, where $ L$ is given by $ $ lambert = - \int { \rm d}x \left [ \frac{1}{2}\left(\frac{\partial\phi}{\partial x } \right)^2
- \varepsilon\phi + \frac{1}{M^2}\left(1 - \cos(M \phi) \right)
- v\cos(\chi + \phi)\delta(x - x_{i }) \right ]. \label{eq:2.4}$$ Here $ \varepsilon = F/(g M^2)$, $ five = V_{i}/(g \xi M^2)$. In this paper, we assume $ \varepsilon \geq 0 $. Further, we specify $ x_{i}=X_{i}/\xi$ and $ \chi= (2 \pi z)/M $, where $ z = X_{i}/a $ characterizes the localization of an impurity proportional to the site where the energy gain by commensurability is maximum. The range of $ \chi$ is $ -\pi / M \leq \chi \leq \pi / M$, because the Lagrangian suffer the periodicity of $ 2\pi / M$ with obedience to the phase.
The Homogeneous Case
= = = = = = = = = = = = = = = = = = = =
First of all, we examine the case of $ v=0 $, namely the homogeneous case. [ @rf:3 ]
The stable configuration of the phase is determine by varying Lagrangian; $ $ - \phi" - \varepsilon + \frac{1}{M}\sin(M \phi)=0. \label{eq:3.1}$$ When $ \varepsilon = 0 $, eq. (\[eq:3.1\ ]) is the sine - Gordon equation, and has two kind of solutions, which are a trivial one, $ \phi=0 $, and the crick solution, which is given by $ $ \phi(x)= \frac{4}{M}\arctan\left[\exp(\pm(x - x_{0})\right ] \equiv \phi_{\pm}(x - x_{0 }),
\label{eq:3.3}$$ and show in Fig. \[fig: kink\ ].
Here $ x_{0}$ is the center of the kink | ) }$, wjich is the phase coherekce length due to the cmmmenshrabilith. We make the Lagrangian dimxnsiinlesw through scaling by tfis $\xi$; i.e. $x =X/\xi$, $\cql L \nit = M^2 g \xm L$, wherc $L$ ia givzn by $$L = - \int {\rm d}x \left[ \xrac{1}{2}\left(\frac{\pastkap\phi}{\partial x} \right)^2
- \varepsilon\phi + \frac{1}{M^2}\lrfh(1 - \cos(M \phi) \ridht)
- n\cjs(\chj + \phi)\delta(x-x_{i}) \right]. \label{eq:2.4}$$ Here $\barepsinon = F/(g M^2)$, $v = F_{i}/(g \xi M^2)$. In this paper, we wssule $\varepsilon \geq 0 $. Further, wg desune $x_{i}=X_{i}/\xi$ avd $\chi= (2 \pi z)/M $, where $z=S_{i}/a $ characterizes the location of au impurity tzoatlee to the smte whvre the energn gain ty commrnsurability ix meximym. The range of $\chi$ iv $-\pi/M \leq \chi \leq \ki/M$, becausa che Lagrangian has thw periogicidy ow $2\pi/O$ wjti rsspect to the phase.
Fhe Homogenwous Case
====================
First of akl, qe examine ths case os $v=0$, namely the homogeneous case. [@rf:3]
The stdbls configuration of the phase is determined bi varying Jagrangian; $$- \phi'' - \varepsilon + \frac{1}{M}\sin(M \phi)=0. \label{ex:3.1}$$ Whei $\xartpwilon = 0$, fq. (\[eq:3.1\]) is the sine-Gordon equation, and has two kyhdx pf solutions, wmich are a trivial oje, $\khi=0$, and the kivk solbfikn, which is given hy $$\phi(x)= \frac{4}{N}\arctan\lest[\exl(\pm(x-x_{0})\right] \equiv \phi_{\pm}(x-x_{0}),
\lavel{eq:3.3}$$ and shjqn in Fig. \[fig:kink\].
Hexe $x_{0}$ is the eenter of tne kink | ) }$, which is the phase coherence to commensurability. We the Lagrangian dimensionless i.e. =X/\xi$, $\cal L = M^2 g L$, where $L$ is given by = - \int {\rm d}x \left[ \frac{1}{2}\left(\frac{\partial\phi}{\partial x} \right)^2 - \varepsilon\phi + \frac{1}{M^2}\left(1 \cos(M \phi) \right) - v\cos(\chi + \phi)\delta(x-x_{i}) \right]. \label{eq:2.4}$$ Here $\varepsilon = F/(g $v V_{i}/(g M^2)$. this paper, we assume $\varepsilon \geq 0 $. Further, we define $x_{i}=X_{i}/\xi$ and $\chi= (2 \pi $, where $z=X_{i}/a $ characterizes the location of impurity relative to the where the energy gain by is The range $\chi$ $-\pi/M \chi \leq \pi/M$, the Lagrangian has the periodicity of $2\pi/M$ with respect to the phase. The Homogeneous Case ==================== First all, we case of namely homogeneous [@rf:3] The stable the phase is determined by varying - \varepsilon + \frac{1}{M}\sin(M \phi)=0. \label{eq:3.1}$$ When $\varepsilon 0$, eq. is the sine-Gordon equation, and has kinds of solutions, which are a trivial one, and the kink solution, which is given by $$\phi(x)= \frac{4}{M}\arctan\left[\exp(\pm(x-x_{0})\right] \equiv \phi_{\pm}(x-x_{0}), \label{eq:3.3}$$ and shown \[fig:kink\]. Here $x_{0}$ is center of the | ) }$, which is the phase coherence lEngth due to The coMmeNsuRaBiliTy. We Make the LagrangIAn diMensionless through scalIng by ThIS $\xi$; i.E. $X =X/\Xi$, $\cal l \mit = M^2 g \xI l$, wHERe $L$ Is GiVen By $$l = - \InT {\rm d}x \LefT[ \frac{1}{2}\leFt(\frac{\partIal\PhI}{\partial x} \rigHT)^2
- \vArepsilon\pHi + \fRac{1}{M^2}\left(1 - \cos(M \Phi) \Right)
- v\CoS(\chI + \Phi)\deLta(X-x_{i}) \riGht]. \labEL{eq:2.4}$$ HerE $\varepsilOn = f/(G M^2)$, $v = V_{i}/(g \XI M^2)$. In thiS PApEr, we Assume $\varepsilon \gEQ 0 $. FURther, we define $x_{I}=X_{i}/\xi$ aNd $\CHi= (2 \PI Z)/M $, wHerE $z=X_{i}/a $ charaCtErizeS The locaTIoN OF An iMPurity relativE to the site wHEre The eneRgY gaIN by comMensuRaBIliTy is maximum. the rAnge of $\chi$ Is $-\pi/M \lEQ \chi \leq \PI/M$, becauSe the LAgrAngIan hAS tHe PerIoDIciTY oF $2\pi/m$ WitH respect To ThE phasE.
The hOMOGeneOus case
====================
first Of all, we examinE thE casE Of $v=0$, NamelY the hOmogEnEous cAse. [@rf:3]
THe staBlE configuration oF the Phase is deTerMiNed By VaryiNG LagraNgiAn; $$- \pHi'' - \varepSilon + \frAC{1}{M}\sIn(m \PHI)=0. \lAbel{eq:3.1}$$ When $\varepsilOn = 0$, EQ. (\[Eq:3.1\]) Is the sinE-GordoN EqUaTIon, and haS tWo kInds OF SolutIons, WHiCh are a trIvial oNE, $\pHi=0$, And the kInK solutIoN, whIch Is givEN by $$\pHi(x)= \fraC{4}{M}\arctan\Left[\eXP(\pm(x-x_{0})\right] \equiV \Phi_{\pm}(x-x_{0}),
\label{eQ:3.3}$$ AnD SHoWN in FIg. \[fIg:kink\].
Here $x_{0}$ Is thE CentEr of THe KinK | ) }$, which is the phase c oherence l ength du e t othecomm ensurability.W e ma ke the Lagrangian dime nsion le s s th r ou gh sc aling b y t h i s $ \x i$ ; i .e . $ x =X/ \xi $, $\ca l L \mit = M^ 2g \xi L$, wh e re $L$ is gi ven by $$L = -\in t {\rm d }x\ left[ \ frac{ 1}{2}\ l eft(\f rac{\part ia l \phi}{ \ partial x }\rig ht)^2
- \varepsi l on \ phi + \frac{1} {M^2}\ le f t( 1 - \ cos (M \phi) \ ri ght)- v\cos( \ ch i + \p h i)\delta(x-x_ {i}) \right ] . \ label{ eq :2. 4 }$$ He re $\ va r eps ilon = F/(g M^2 )$, $v =V_{i}/ ( g \xi M ^ 2)$. In thispap er, wea ss um e $ \v a rep s il on\ geq 0 $. Fu rt he r, we def i n e $x_{ i}= X_{i }/\xi $ and $\chi=(2\piz )/M $, w here$z=X _{ i}/a$ char acter iz es the location ofan impuri tyre lat iv e tot he sit e w her e the e nergy g a inby c o mm ensurability is ma xi m u m. The ran ge of$ \c hi $ is $-\p i/ M \ leq\ c hi \l eq \ p i/ M$, beca use th e L ag rangian h as the p eri odi cityo f $2 \pi/M$ with re spect to the phase.The Homogeneo u sC a se ==== === =========== ==
F irst ofa ll , w e exam ine t he ca s e of $v=0$, namelyth e homo geneo us case. [@rf :3]
The s t a b le confi gura t io n of the phaseis de termined b y varying Lagr angian;$$- \phi' ' - \varep sil on+ \ fra c { 1} {M}\sin(M \ph i ) =0.\l abel{eq :3. 1}$$ Wh en$\v are psi lo n = 0$, e q. (\[eq :3 .1 \] )isthe s i ne-Gordo nequ at ion , and has tw o kin ds o fso l uti ons, wh i ch a re a t ri vial on e, $\ph i=0$ , an d the k ink solut ion , whi ch i s given by $$\phi(x) =\frac{4}{M }\ arc tan\le f t [\exp(\p m(x-x_{0})\right] \equi v \phi_{ \pm }(x-x _{0} ),
\labe l{e q:3.3} $$a nd sho wn inFig.\[ fig : k ink\] .
H ere $ x_{0}$ ist h e c enter o f th e kink | ) }$,_which is_the phase coherence length_due to_the_commensurability. We_make_the Lagrangian dimensionless_through scaling by_this $\xi$; i.e. $x_=X/\xi$, $\cal L_\mit_= M^2 g \xi L$, where $L$ is given by $$L = - \int_{\rm_d}x \left[__\frac{1}{2}\left(\frac{\partial\phi}{\partial_x} \right)^2
- \varepsilon\phi +_\frac{1}{M^2}\left(1 - \cos(M \phi) \right)
-_v\cos(\chi +_\phi)\delta(x-x_{i}) \right]. \label{eq:2.4}$$ Here $\varepsilon = F/(g M^2)$,_$v_= V_{i}/(g \xi_M^2)$. In this paper, we assume $\varepsilon \geq 0_$. Further, we define $x_{i}=X_{i}/\xi$ and_$\chi= (2 \pi_z)/M_$,_where $z=X_{i}/a $ characterizes_the location of an impurity relative_to the site where the energy_gain by commensurability is maximum. The range_of $\chi$ is $-\pi/M \leq \chi_\leq \pi/M$, because the Lagrangian_has the_periodicity of $2\pi/M$ with respect_to the phase.
The_Homogeneous Case
====================
First_of all, we_examine the case of $v=0$, namely_the homogeneous case. [@rf:3]
The_stable configuration of the phase is_determined_by varying Lagrangian;_$$-_\phi''_- \varepsilon_+ \frac{1}{M}\sin(M \phi)=0._\label{eq:3.1}$$_When $\varepsilon_=_0$, eq. (\[eq:3.1\]) is the sine-Gordon equation,_and_has two kinds of solutions, which are_a trivial one, $\phi=0$,_and_the kink solution, which_is given by $$\phi(x)= \frac{4}{M}\arctan\left[\exp(\pm(x-x_{0})\right]_\equiv \phi_{\pm}(x-x_{0}),
\label{eq:3.3}$$ and shown in_Fig. \[fig:kink\].
Here $x_{0}$_is the_center of the kink |
fact, for $X=X_0$, $w=0$ should give the Gepner model $(2)^4$.
In the smooth universal covering space $\wt\MMM^{K3}_{SCFT}$ (see [(\[wendland:cover\])]{}), the lift of each family $\FFF_{X}$ can be described within a fixed geometric interpretation, i.e. we once and for all choose null vectors $\upn,\,\up$ that generate $H^0(X,\Z),\, H^4(X,\Z)$ to use [(\[wendland:deco\])]{}. Above, we have also specified a complex structure and a normalized Khler class, or equivalently the three-plane $\Sigma$. We focus on the Fermat quartic $X_0$ in the following, so $\Sigma_{X_0}=\Span_\R(\Om_{X_0},\omega_{FS})$ as in Ex. \[wendland:sthk\]. Moreover, $B=\beta\omega_{FS}$, which allows us to cast all four-planes in our family into the following form: $$\begin{aligned}
\la{wendland:mho}
&&x=\Omega_X\oplus\mho_{\beta,V}\quad
\mbox{ where with }\quad \no{\omega_{FS}}=4,\;
\beta\in\R,\;V\in\R^+\colon\nonumber\\
&&\quad\quad\quad\mho_{\beta,V}=\Span_\R\{\omega_{FS} - 4\beta\up,\,
\upn+\beta\omega_{FS}+(V-2\beta^2)\up\}.\end{aligned}$$ The parameters $(\beta,V)\in\R\times \R^+$ of the theories under investigation can be conveniently combined into a parameter $\tau$ on the upper half plane $\H$: $$\tau:= \beta+i\sqrt{{ {\textstyle {V\over 2}} }} \;\;\in\;\;
\H=\left\{ z\in\C \bigm| \im(z)>0 \right\}.$$ To reproduce Witten’s ${\S}^2\simeq\C\cup\{\infty\}$ we now have to divide out all dualities that leave invariant the subvariety of $\wt{\mathcal M}^{K3}_{SCFT}$ given by the four-planes $x$ as in [(\[wendland:mho\])]{}, and then | fact, for $ X = X_0 $, $ w=0 $ should give the Gepner model $ (2)^4$.
In the smooth universal joint cover space $ \wt\MMM^{K3}_{SCFT}$ (see [ (\[wendland: cover\ ]) ] { }), the ski tow of each kin $ \FFF_{X}$ can be described within a fixed geometric rendition, i.e. we once and for all choose null vectors $ \upn,\,\up$ that generate $ H^0(X,\Z),\, H^4(X,\Z)$ to use [ (\[wendland: deco\ ]) ] { }. Above, we have besides specify a complex structure and a anneal Khler class, or equivalently the three - plane $ \Sigma$. We concentrate on the Fermat quartic $ X_0 $ in the following, so $ \Sigma_{X_0}=\Span_\R(\Om_{X_0},\omega_{FS})$ as in Ex. \[wendland: sthk\ ]. Moreover, $ B=\beta\omega_{FS}$, which allows us to cast all four - airplane in our family into the following form: $ $ \begin{aligned }
\la{wendland: mho }
& & x=\Omega_X\oplus\mho_{\beta, V}\quad
\mbox { where with } \quad \no{\omega_{FS}}=4,\;
\beta\in\R,\;V\in\R^+\colon\nonumber\\
& & \quad\quad\quad\mho_{\beta, V}=\Span_\R\{\omega_{FS } - 4\beta\up,\,
\upn+\beta\omega_{FS}+(V-2\beta^2)\up\}.\end{aligned}$$ The parameters $ (\beta, V)\in\R\times \R^+$ of the theory under investigation can be conveniently combined into a argument $ \tau$ on the upper half airplane $ \H$: $ $ \tau:= \beta+i\sqrt { { { \textstyle { V\over 2 } } } } \;\;\in\;\;
\H=\left\ { z\in\C \bigm| \im(z)>0 \right\}.$$ To reproduce Witten ’s $ { \S}^2\simeq\C\cup\{\infty\}$ we now have to divide out all duality that leave invariant the subvariety of $ \wt{\mathcal M}^{K3}_{SCFT}$ given by the four - planes $ x$ as in [ (\[wendland: mho\ ]) ] { }, and then | faft, for $X=X_0$, $w=0$ should give uhe Gepner model $(2)^4$.
Nb the vmooth universxl covering space $\wt\MMM^{K3}_{SCFT}$ (swe [(\[webdland:cover\])]{}), the lift ow each falily $\FFF_{Z}$ cai be described wmfhin a nnxed fcometxir interpretatiok, i.e. we once and for all cvousz null vectors $\upn,\,\up$ that generate $H^0(V,\Z),\, H^4(X,\Z)$ yo use [(\[wendland:dgco\])]{}. Anjve, sv mave also specified a complex stducture and a normalozed Khler class, or equivapentpy the three-plane $\Digma$. We foxus jb the Fermat quartic $X_0$ in the follkwing, so $\Sigma_{X_0}=\Span_\R(\Om_{X_0},\omega_{FS})$ xs in Ex. \[wendlane:srhk\]. Koreover, $B=\bxta\omeda_{FS}$, which allows us to casy all four-plancs in oue family into the folnowing form: $$\begin{ajigned}
\la{wanbland:mho}
&&x=\Omega_X\oplus\myo_{\veta,V}\xuad
\kbox{ qhefe sivh }\suad \nl{\omxga_{FS}}=4,\;
\beta\in\D,\;V\in\R^+\colon\ninumber\\
&&\quad\quad\quad\khj_{\veta,V}=\Span_\R\{\omefa_{FS} - 4\feea\up,\,
\upn+\beta\omega_{FS}+(V-2\beta^2)\up\}.\end{aligned}$$ Tht parzmeters $(\beta,V)\in\R\times \R^+$ of the theories undet investigwtion can be conveniently combined into a parametar $\tan$ un uhc jppfr half plane $\H$: $$\tau:= \beta+i\sqrt{{ {\textstyle {V\over 2}} }} \;\;\on\;\;
\H=\left\{ z\in\C \binm| \im(z)>0 \right\}.$$ To relrldise Witten’s ${\S}^2\skmeq\C\cbl\{\ihfty\}$ we now have tl dividg out qll dualiuies yhat leave invariant the suvvariety of $\ct{\mqthcal M}^{K3}_{SCFT}$ giveu by the foux-planex $x$ ax in [(\[wendland:mho\])]{}, and thzn | fact, for $X=X_0$, $w=0$ should give the $(2)^4$. the smooth covering space $\wt\MMM^{K3}_{SCFT}$ each $\FFF_{X}$ can be within a fixed interpretation, i.e. we once and for choose null vectors $\upn,\,\up$ that generate $H^0(X,\Z),\, H^4(X,\Z)$ to use [(\[wendland:deco\])]{}. Above, we also specified a complex structure and a normalized Khler class, or equivalently the $\Sigma$. focus the quartic $X_0$ in the following, so $\Sigma_{X_0}=\Span_\R(\Om_{X_0},\omega_{FS})$ as in Ex. \[wendland:sthk\]. Moreover, $B=\beta\omega_{FS}$, which allows us cast all four-planes in our family into the form: $$\begin{aligned} \la{wendland:mho} &&x=\Omega_X\oplus\mho_{\beta,V}\quad where with }\quad \no{\omega_{FS}}=4,\; \beta\in\R,\;V\in\R^+\colon\nonumber\\ - \upn+\beta\omega_{FS}+(V-2\beta^2)\up\}.\end{aligned}$$ The $(\beta,V)\in\R\times of theories under investigation be conveniently combined into a parameter $\tau$ on the upper half plane $\H$: $$\tau:= \beta+i\sqrt{{ {\textstyle {V\over }} \;\;\in\;\; \bigm| \im(z)>0 To Witten’s we now have out all dualities that leave invariant $\wt{\mathcal M}^{K3}_{SCFT}$ given by the four-planes $x$ as [(\[wendland:mho\])]{}, and | fact, for $X=X_0$, $w=0$ should give the GePner model $(2)^4$.
IN the sMooTh uNiVersAl coVering space $\wt\Mmm^{K3}_{SCfT}$ (see [(\[wendland:cover\])]{}), the lIft of EaCH famILy $\fFF_{X}$ cAn be desCRiBED wiThIn A fiXeD GeOmetrIc iNterpreTation, i.e. we OncE aNd for all chooSE nUll vectors $\Upn,\,\Up$ that generaTe $H^0(x,\Z),\, H^4(X,\Z)$ tO uSe [(\[wENdlanD:deCo\])]{}. AboVe, we haVE also sPecified a CoMPlex stRUcture aND A nOrmaLized Khler class, or EQuIValently the thrEe-planE $\SIGmA$. wE foCus On the FermaT qUartiC $x_0$ in the fOLlOWINg, sO $\sigma_{X_0}=\Span_\R(\Om_{x_0},\omega_{FS})$ as iN ex. \[wEndlanD:sThk\]. mOreoveR, $B=\betA\oMEga_{fS}$, which alloWs us To cast all Four-plANes in ouR Family iNto the FolLowIng fORm: $$\BeGin{AlIGneD}
\La{WenDLanD:mho}
&&x=\OmeGa_x\oPlus\mHo_{\beTA,v}\QUad
\mBox{ WherE with }\Quad \no{\omega_{FS}}=4,\;
\BetA\in\R,\;v\In\R^+\Colon\NonumBer\\
&&\qUaD\quad\Quad\mhO_{\beta,v}=\SPan_\R\{\omega_{FS} - 4\beta\Up,\,
\upN+\beta\omegA_{FS}+(v-2\bEta^2)\Up\}.\End{alIGned}$$ ThE paRamEters $(\beTa,V)\in\R\tIMes \r^+$ oF THE tHeories under investIgATIoN can be coNvenieNTlY cOMbined inTo A paRameTER $\tau$ oN the UPpEr half plAne $\H$: $$\taU:= \BeTa+I\sqrt{{ {\teXtStyle {V\OvEr 2}} }} \;\;\iN\;\;
\H=\lEft\{ z\iN\c \bigM| \im(z)>0 \riGht\}.$$ To repRoducE witten’s ${\S}^2\simeq\C\CUp\{\infty\}$ we now hAVe TO DiVIde oUt aLl dualities That LEave InvaRIaNt tHE subvArietY oF $\Wt{\MAthcal M}^{K3}_{SCFT}$ given by ThE four-pLanes $X$ as in [(\[wendland:Mho\])]{}, and then | fact, for $X=X_0$, $w=0$should giv e the Ge pne rmode l $( 2)^4$.
In the smoo th universal coveringspace $ \ wt\M M M^ {K3}_ {SCFT}$ (s e e [( \[ we ndl an d :c over\ ])] {}), th e lift ofeac hfamily $\FFF _ {X }$ can bedes cribed withi n a fixed g eom e tricint erpre tation , i.e.we once a nd for al l choose n ul l ve ctors $\upn,\,\up $ t h at generate $H ^0(X,\ Z) , \, H ^4( X,\ Z)$ to use [ (\[we n dland:d e co \ ] ) ]{} . Above, we ha ve also spe c ifi ed a c om ple x struc turean d anormalizedKhle r class,or equ i valentl y the th ree-pl ane $\ Sigm a $. W e f oc u s o n t heF erm at quart ic $ X_0$in t h e f ollo win g, s o $\S igma_{X_0}=\S pan _\R( \ Om_ {X_0} ,\ome ga_{ FS })$ a s in E x. \[ we ndland:sthk\].More over, $B= \be ta \om eg a_{FS } $, whi chall ows usto cast all f o u r -p lanes in our famil yi n to the fol lowing fo rm : $$\begi n{ ali gned } \la{w endl a nd :mho}
&& x=\Ome g a_ X\ oplus\m ho _{\bet a, V}\ qua d
\mb o x{ w here w ith }\qu ad \n o {\omega_{FS}}= 4 ,\;
\beta\in\ R ,\ ; V \i n \R^+ \co lon\nonumbe r\\& &\qu ad\q u ad \qu a d\mho _{\be ta , V} = \Span_\R\{\omega_{F S} - 4\b eta\u p,\,
\upn+\b eta\omega_ { F S }+(V-2\b eta^ 2 )\ u p\}.\end{align ed}$$ The param e ters $(\ beta, V)\in\R\ times \R^ + $ of thethe ori esund e r i nvestigationc a n be c onvenie ntl y combi ned in toa p ar ameter $\ tau$ onth eup pe r h alf p l ane $\H$ :$$\ ta u:= \bet a +i\sqr t{{ { \tex ts ty l e { V\over2 }} } } \; \; \i n\;\ ; \H =\lef t\{z \in \C \big m| \im(z) >0\ righ t\ }. $$ To r eproduce Witt en ’s ${\S}^2 \s ime q\C\cu p \ {\infty\ }$ we now have to divid e out al l d ualit iesthat leav e i nvaria ntt he sub variet y of$\ wt{ \ m athca l M} ^{K 3} _{SCFT}$ g i v enby th efour -planes $x$ as in [(\[wen d lan d:mho\])]{},and the n | fact,_for $X=X_0$,_$w=0$ should give the_Gepner model_$(2)^4$.
In_the smooth_universal_covering space $\wt\MMM^{K3}_{SCFT}$_(see [(\[wendland:cover\])]{}), the_lift of each family_$\FFF_{X}$ can be_described_within a fixed geometric interpretation, i.e. we once and for all choose null vectors $\upn,\,\up$_that_generate $H^0(X,\Z),\,_H^4(X,\Z)$_to_use [(\[wendland:deco\])]{}. Above, we have_also specified a complex structure_and a_normalized Khler class, or equivalently the three-plane $\Sigma$._We_focus on the_Fermat quartic $X_0$ in the following, so $\Sigma_{X_0}=\Span_\R(\Om_{X_0},\omega_{FS})$ as_in Ex. \[wendland:sthk\]. Moreover, $B=\beta\omega_{FS}$, which allows_us to cast_all_four-planes_in our family into_the following form: $$\begin{aligned}
\la{wendland:mho}
&&x=\Omega_X\oplus\mho_{\beta,V}\quad
\mbox{ where with_}\quad \no{\omega_{FS}}=4,\;
\beta\in\R,\;V\in\R^+\colon\nonumber\\
&&\quad\quad\quad\mho_{\beta,V}=\Span_\R\{\omega_{FS} - 4\beta\up,\,
\upn+\beta\omega_{FS}+(V-2\beta^2)\up\}.\end{aligned}$$ The_parameters $(\beta,V)\in\R\times \R^+$ of the theories under_investigation can be conveniently combined into_a parameter $\tau$ on the_upper half_plane $\H$: $$\tau:= \beta+i\sqrt{{ {\textstyle_{V\over 2}} }}_\;\;\in\;\;
\H=\left\{_z\in\C \bigm| \im(z)>0_\right\}.$$ To reproduce Witten’s ${\S}^2\simeq\C\cup\{\infty\}$ we_now have to_divide out all dualities that leave_invariant_the subvariety of_$\wt{\mathcal_M}^{K3}_{SCFT}$_given by_the four-planes $x$_as_in [(\[wendland:mho\])]{},_and_then |
0\_[2,Ø]{}\^2+ \^|Ø\_T|\
&C\_0\^T \^n(t)\_[p]{}dt \_[L\^(0,T;L\^(Ø))]{} + \_0\_[2,Ø]{}\^2+ \^|Ø\_T|\
& CT\^\^n\_[p,Ø\_T]{} \_[L\^(0,T;L\^(Ø))]{} + \_0\_[2,Ø]{}\^2+ \^|Ø\_T|\
& \^n\_[p,Ø\_T]{}\^p + C T \_[L\^(0,T;L\^(Ø))]{}\^[p’]{}+ \_0\_[2,Ø]{}\^2+ \^|Ø\_T|
where $C$ is a constant due to the Sobolev embedding $W^{1,p}\hookrightarrow L^{\frac{4}{4-p}}(\O)$, Korn’s and Young’s inequalities, that is unessential for our aims. Finally, straightforward computations yield the assertion.
\[lemma1\] Let $p\in \left[\frac{5}{3},2\right)$. For every function $\bv$ sufficiently smooth and space-periodic with respect to $\Omega$ it holds \_[3,Ø]{}\^3 C\_[S]{}\^3 \_[2,Ø]{}\^\_[,Ø]{}\^ with $C_{S} $ independent of $\bv$.
Since $$\frac{1}{3}= \frac{3p-4}{3p-2}\frac{1}{2}+\frac{2}{3p-2}\frac{8-3p}{12},$$ interpolating we get \[conti-lem\] \_[3,Ø]{} \_[2,Ø]{}\^\_[,Ø]{}\^. Then the Korn inequality, the Sobolev embedding $W^{1, \frac{4}{4-p}}_{\per}(\O)\hookrightarrow L_{\per}^{\frac{12}{8-3p}}(\O)$ and the Poincaré inequality imply $$\|\n\bv\|_{3,\O} \leq C_{S} \|\nabla \bv\|_{2,\O}^{\frac{3p-4}{3p-2}}\|\nabla\cD\bv\|_{\frac{4}{4-p},\O}^{\frac{2}{3p-2}}$$ where $C_{S}$ is a constant depending on $|\O|$ and $p$. Raising to the power $3$ the proof is concluded.
We recall the following Hö | 0\_[2,Ø]{}\^2 + \^|Ø\_T|\
& C\_0\^T \^n(t)\_[p]{}dt \_[L\^(0,T;L\^(Ø) ) ] { } + \_0\_[2,Ø]{}\^2 + \^|Ø\_T|\
& CT\^\^n\_[p, Ø\_T ] { } \_[L\^(0,T;L\^(Ø) ) ] { } + \_0\_[2,Ø]{}\^2 + \^|Ø\_T|\
& \^n\_[p, Ø\_T]{}\^p + C T \_[L\^(0,T;L\^(Ø))]{}\^[p’]{}+ \_0\_[2,Ø]{}\^2 + \^|Ø\_T|
where $ C$ is a constant due to the Sobolev embedding $ W^{1,p}\hookrightarrow L^{\frac{4}{4 - p}}(\O)$, Korn ’s and Young ’s inequalities, that is inessential for our aim. Finally, straightforward calculation render the assertion.
\[lemma1\ ] Let $ p\in \left[\frac{5}{3},2\right)$. For every routine $ \bv$ sufficiently placid and space - periodic with esteem to $ \Omega$ it holds \_[3,Ø]{}\^3 C\_[S]{}\^3 \_[2,Ø]{}\^\_[,Ø]{}\^ with $ C_{S } $ independent of $ \bv$.
Since $ $ \frac{1}{3}= \frac{3p-4}{3p-2}\frac{1}{2}+\frac{2}{3p-2}\frac{8 - 3p}{12},$$ interpolate we get \[conti - lem\ ] \_[3,Ø ] { } \_[2,Ø]{}\^\_[,Ø]{}\^. Then the Korn inequality, the Sobolev embedding $ W^{1, \frac{4}{4 - p}}_{\per}(\O)\hookrightarrow L_{\per}^{\frac{12}{8 - 3p}}(\O)$ and the Poincaré inequality incriminate $ $ \|\n\bv\|_{3,\O } \leq C_{S } \|\nabla \bv\|_{2,\O}^{\frac{3p-4}{3p-2}}\|\nabla\cD\bv\|_{\frac{4}{4 - p},\O}^{\frac{2}{3p-2}}$$ where $ C_{S}$ is a changeless depending on $ |\O|$ and $ p$. Raising to the power $ 3 $ the proof is concluded.
We remember the following Hö | 0\_[2,Ø]{}\^2+ \^|Ø\_T|\
&C\_0\^H \^n(t)\_[p]{}dt \_[L\^(0,T;L\^(Ø))]{} + \_0\_[2,Ø]{}\^2+ \^|Ø\_T|\
& CT\^\^n\_[p,Ø\_T]{} \_[L\^(0,T;U\^(Ø))]{} + \_0\_[2,Ø]{}\^2+ \^|Ø\_T|\
& \^n\_[p,Ø\_T]{}\^p + C T \_[L\^(0,T;L\^(Ø))]{}\^[p’]{}+ \_0\_[2,Ø]{}\^2+ \^|Ø\_V|
where $D$ is a cunstant due to the Sobolev elbwddint $W^{1,p}\hookrightarrow L^{\frxc{4}{4-p}}(\O)$, Korn’d and Yoyng’s unequalitixa, that lf unsdsenciel for our aims. Finally, sdraightforward cum'utations yield the assertion.
\[lemma1\] Lqt $p\in \kevt[\frac{5}{3},2\right)$. Fot evegy fundniin $\bv$ sufficiently smooth and space-ptriodic with respevt to $\Omega$ it holds \_[3,Ø]{}\^3 C\_[S]{}\^3 \_[2,Ø]{}\^\_[,Ø]{}\^ aith $C_{S} $ independent ov $\bv$.
Since $$\ftzc{1}{3}= \seac{3p-4}{3p-2}\frac{1}{2}+\fraz{2}{3p-2}\frac{8-3p}{12},$$ inuexpolating wg get \[conti-lem\] \_[3,Ø]{} \_[2,Ø]{}\^\_[,Ø]{}\^. Then the Korn ivequakity, the Sibileg embedding $X^{1, \frac{4}{4-i}}_{\per}(\O)\hookrighbsrrow N_{\per}^{\frav{12}{8-3p}}(\O)$ and the Polncaré inwquality imply $$\|\n\bv\|_{3,\O} \lxq C_{S} \|\nabla \bv\|_{2,\O}^{\frac{3p-4}{3k-2}}\|\nabla\cD\bv\|_{\xrcc{4}{4-p},\O}^{\frac{2}{3p-2}}$$ where $C_{S}$ is a consjant gepeveine oh $|\P|$ znd $p$. Gaiaing to ths power $3$ thw proof is concludec.
Wq recall the fkllowigg Hö | 0\_[2,Ø]{}\^2+ \^|Ø\_T|\ &C\_0\^T \^n(t)\_[p]{}dt \_[L\^(0,T;L\^(Ø))]{} + \_0\_[2,Ø]{}\^2+ CT\^\^n\_[p,Ø\_T]{} + \_0\_[2,Ø]{}\^2+ & \^n\_[p,Ø\_T]{}\^p + where is a constant to the Sobolev $W^{1,p}\hookrightarrow L^{\frac{4}{4-p}}(\O)$, Korn’s and Young’s inequalities, is unessential for our aims. Finally, straightforward computations yield the assertion. \[lemma1\] Let \left[\frac{5}{3},2\right)$. For every function $\bv$ sufficiently smooth and space-periodic with respect to $\Omega$ holds C\_[S]{}\^3 with $ independent of $\bv$. Since $$\frac{1}{3}= \frac{3p-4}{3p-2}\frac{1}{2}+\frac{2}{3p-2}\frac{8-3p}{12},$$ interpolating we get \[conti-lem\] \_[3,Ø]{} \_[2,Ø]{}\^\_[,Ø]{}\^. Then the Korn the Sobolev embedding $W^{1, \frac{4}{4-p}}_{\per}(\O)\hookrightarrow L_{\per}^{\frac{12}{8-3p}}(\O)$ and the inequality imply $$\|\n\bv\|_{3,\O} \leq \|\nabla \bv\|_{2,\O}^{\frac{3p-4}{3p-2}}\|\nabla\cD\bv\|_{\frac{4}{4-p},\O}^{\frac{2}{3p-2}}$$ where $C_{S}$ is constant on $|\O|$ $p$. to power $3$ the is concluded. We recall the following Hö | 0\_[2,Ø]{}\^2+ \^|Ø\_T|\
&C\_0\^T \^n(t)\_[p]{}dt \_[L\^(0,T;L\^(Ø))]{} + \_0\_[2,Ø]{}\^2+ \^|Ø\_T|\
& CT\^\^n\_[p,Ø\_T]{} \_[L\^(0,T;l\^(Ø))]{} + \_0\_[2,Ø]{}\^2+ \^|Ø\_T|\
& \^n\_[p,Ø\_T]{}\^p + c T \_[L\^(0,T;L\^(ø))]{}\^[p’]{}+ \_0\_[2,Ø]{}\^2+ \^|ø\_T|
wHeRe $C$ iS a coNstant due to the sObolEv embedding $W^{1,p}\hookrightArrow l^{\fRAc{4}{4-p}}(\O)$, kOrN’s and young’s iNEqUALitIeS, tHat Is UNeSsentIal For our aIms. Finally, StrAiGhtforward coMPuTations yieLd tHe assertion.
\[lEmmA1\] Let $p\iN \lEft[\FRac{5}{3},2\riGht)$. for evEry funCTion $\bv$ SufficienTlY Smooth ANd space-PERiOdic With respect to $\OmegA$ It HOlds \_[3,Ø]{}\^3 C\_[S]{}\^3 \_[2,Ø]{}\^\_[,Ø]{}\^ with $C_{s} $ indepEnDEnT OF $\bv$.
sinCe $$\frac{1}{3}= \frac{3P-4}{3p-2}\Frac{1}{2}+\fRAc{2}{3p-2}\frac{8-3P}{12},$$ InTERPolATing we get \[contI-lem\] \_[3,Ø]{} \_[2,Ø]{}\^\_[,Ø]{}\^. Then THe KOrn ineQuAliTY, the SoBolev EmBEddIng $W^{1, \frac{4}{4-p}}_{\peR}(\O)\hoOkrightarRow L_{\peR}^{\Frac{12}{8-3p}}(\O)$ aND the PoiNcaré iNeqUalIty iMPlY $$\|\n\Bv\|_{3,\O} \LeQ c_{S} \|\nABlA \bv\|_{2,\o}^{\FraC{3p-4}{3p-2}}\|\nabla\CD\Bv\|_{\Frac{4}{4-p},\o}^{\fraC{2}{3P-2}}$$ WHEre $C_{s}$ is A conStant Depending on $|\O|$ aNd $p$. raisINg tO the pOwer $3$ tHe prOoF is coNcludeD.
We reCaLl the following HÖ | 0\_[2,Ø]{}\^2+ \^|Ø\_T|\
& C\_0\^T \^ n(t)\ _[p ]{} dt \_[ L\^( 0,T;L\^(Ø))]{} + \_ 0\_[2,Ø]{}\^2+ \^|Ø\_T |\
&CT \ ^\^n \ _[ p,Ø\_ T]{} \_ [ L\ ^ ( 0,T ;L \^ (Ø) )] { }+ \_0 \_[ 2,Ø]{}\ ^2+ \^|Ø\_ T|\
& \^n\_[p,Ø\_ T ]{ }\^p + C T \_ [L\^(0,T;L\^ (Ø) )]{}\^ [p ’]{ } + \_0 \_[ 2,Ø]{ }\^2+\ ^|Ø\_T |
where$C $ is ac onstant d ue tothe Sobolev embed d in g $W^{1,p}\hook righta rr o wL ^ {\f rac {4}{4-p}}( \O )$, K o rn’s an d Y o u n g’s inequalities, that is un e sse ntialfo r o u r aims . Fin al l y,straightfor ward computat ions y i eld the asserti on.
\ [le mma 1\]L et $ p\i n\ lef t [\ fra c {5} {3},2\ri gh t) $. Fo r ev e r y func tio n $\ bv$ s ufficiently s moo th a n d s pace- perio dicwi th re spectto $\ Om ega$ it holds \ _[3, Ø]{}\^3 C \_[ S] {}\ ^3 \_[2 , Ø]{}\^ \_[ ,Ø] {}\^ wi th $C_{ S } $ i n d e pe ndent of $\bv$.
S in c e $ $\frac{1 }{3}=\ fr ac { 3p-4}{3p -2 }\f rac{ 1 } {2}+\ frac { 2} {3p-2}\f rac{8- 3 p} {1 2},$$ i nt erpola ti ngweget \ [ cont i-lem\ ] \_[3,Ø ]{} \ _ [2,Ø]{}\^\_[,Ø ] {}\^. Then th e K o r ni nequ ali ty, the Sob olev embe ddin g $ W^{ 1 , \fr ac{4} {4 - p} } _{\per}(\O)\hookrig ht arrowL_{\p er}^{\frac{12 }{8-3p}}(\ O ) $ and the Poi n ca r é inequality i mply$$\|\n\bv\ | _{3,\O}\leqC_{S} \| \nabla \b v \ |_{2,\O} ^{\ fra c{3 p-4 } { 3p -2}}\|\nabla\ c D \bv\ |_ {\frac{ 4}{ 4-p},\O }^{ \fr ac{ 2}{ 3p -2}}$$ wh ere $C_{ S} $is a co nstan t dependi ng on $ |\O |$ an d $p$.Raisi ng t oth e po wer $3$ th e proo fis con clu de d.
W e re c all the fo llowing H ö | 0\_[2,Ø]{}\^2+ \^|Ø\_T|\
&C\_0\^T_\^n(t)\_[p]{}dt \_[L\^(0,T;L\^(Ø))]{}_+ \_0\_[2,Ø]{}\^2+ \^|Ø\_T|\
& CT\^\^n\_[p,Ø\_T]{}_\_[L\^(0,T;L\^(Ø))]{} +_\_0\_[2,Ø]{}\^2+_\^|Ø\_T|\
& \^n\_[p,Ø\_T]{}\^p_+_C T \_[L\^(0,T;L\^(Ø))]{}\^[p’]{}+_\_0\_[2,Ø]{}\^2+ \^|Ø\_T|
where $C$_is a constant due_to the Sobolev_embedding_$W^{1,p}\hookrightarrow L^{\frac{4}{4-p}}(\O)$, Korn’s and Young’s inequalities, that is unessential for our aims. Finally, straightforward_computations_yield the_assertion.
\[lemma1\]_Let_$p\in \left[\frac{5}{3},2\right)$. For every function_$\bv$ sufficiently smooth and space-periodic_with respect_to $\Omega$ it holds \_[3,Ø]{}\^3 C\_[S]{}\^3 \_[2,Ø]{}\^\_[,Ø]{}\^ with_$C_{S}_$ independent of_$\bv$.
Since $$\frac{1}{3}= \frac{3p-4}{3p-2}\frac{1}{2}+\frac{2}{3p-2}\frac{8-3p}{12},$$ interpolating we get \[conti-lem\] \_[3,Ø]{} \_[2,Ø]{}\^\_[,Ø]{}\^._Then the Korn inequality, the Sobolev_embedding $W^{1, \frac{4}{4-p}}_{\per}(\O)\hookrightarrow_L_{\per}^{\frac{12}{8-3p}}(\O)$_and_the Poincaré inequality imply_$$\|\n\bv\|_{3,\O} \leq C_{S} \|\nabla \bv\|_{2,\O}^{\frac{3p-4}{3p-2}}\|\nabla\cD\bv\|_{\frac{4}{4-p},\O}^{\frac{2}{3p-2}}$$ where_$C_{S}$ is a constant depending on_$|\O|$ and $p$. Raising to the power_$3$ the proof is concluded.
We recall_the following Hö |
bolometric luminosity —
is a few times higher than the 0.5–10.0 keV luminosity. In some cases, the published disk temperatures were “effective temperatures” — converted from “color temperatures” by application of a color correction factor, which attempts to account for effects such as spectral hardening from radiative transfer through a disk (Shimura & Takahara 1995; Merloni, Fabian, & Ross 2000; Makishima et al. 2000). In these cases, we converted the effective temperature to a color temperature. Color temperatures are compared directly to color temperatures in this work. This introduces no significant temperature bias; a recent study has shown that the correction factor for IMBHs should be very similar to that sometimes applied to stellar-mass black holes (Fabian, Miller, & Ross 2004).
The lower energy bound of the 0.5–10.0 keV range is somewhat higher than the present lower energy bounds of [*Chandra*]{} and [*XMM-Newton*]{}; however, history suggests that as all X-ray detectors age, the lower energy bound gradually increases. This range was chosen to be forward–looking, and to avoid any biases inherent in relying too heavily on the lowest bins in the [*Chandra*]{} and [*XMM-Newton*]{} bandpasses.
To understand the properties of the ULX sources within the context of BHCs in their brightest states, we plotted the luminosity and disk temperature of each ULX, and the corresponding data for the five most luminous observations of each BHC in our sample (see Figure 1). In selecting the brightest BHC observations, we are attempting to select those phases wherein each source is closest to its Eddington luminosity. In all cases, the errors on the disk color temperatures are 90% confidence errors. For the ULX sources, the luminosity errors are the 90% confidence errors in the measured flux. A review of the literature shows that constraints on the distance to given Galactic sources can change considerably over time with refined measurements, especially when extinction is particularly high. To be conservative, the luminosity errors on the stellar–mass black holes were set by taking the best-fit measured flux, and enforcing fractional errors of $\pm 30$%. This value is somewhat arbitrary, but certainly greater than most quoted errors, and represents a best-guess value based on the broad literature.
The difference between the IMBHC ULXs and standard, stellar-mass BHCs is shown clearly in Figure 1. | bolometric luminosity —
is a few times higher than the 0.5–10.0 keV luminosity. In some cases, the published phonograph record temperature were “ effective temperatures ” — converted from “ color temperature ” by application of a color discipline factor, which attempts to account for consequence such as spectral hardening from radiative transfer through a magnetic disk (Shimura & Takahara 1995; Merloni, Fabian, & Ross 2000; Makishima et al. 2000). In these cases, we converted the effective temperature to a coloring material temperature. Color temperature are compared directly to color temperatures in this work. This precede no significant temperature bias; a recent sketch has shown that the correction factor for IMBHs should be very similar to that sometimes use to stellar - mass black holes (Fabian, Miller, & Ross 2004).
The lower department of energy bound of the 0.5–10.0 keV range is somewhat higher than the present lower energy limit of [ * Chandra * ] { } and [ * XMM - Newton * ] { }; however, history suggest that as all X - ray detectors age, the lower energy constipate gradually addition. This range was choose to be forward – looking, and to avoid any biases built-in in relying too heavily on the lowest bank identification number in the [ * Chandra * ] { } and [ * XMM - Newton * ] { } bandpasses.
To understand the properties of the ULX sources within the context of BHCs in their brightest states, we plotted the luminosity and magnetic disk temperature of each ULX, and the corresponding data for the five most luminous observation of each BHC in our sample distribution (determine Figure 1). In selecting the brightest BHC observations, we are try to select those phases wherein each source is close to its Eddington luminosity. In all cases, the errors on the disk color temperature are 90% confidence error. For the ULX sources, the luminosity errors are the 90% assurance errors in the measured flux. A review of the literature read that constraints on the distance to given Galactic reference can change considerably over time with refined measurement, especially when extinction is particularly high. To be button-down, the luminosity errors on the stellar – mass black holes were put by taking the best - fit mensural flux, and enforcing fractional errors of $ \pm 30$%. This value is somewhat arbitrary, but certainly greater than most quote errors, and represents a best - guess value based on the broad literature.
The difference between the IMBHC ULXs and standard, leading - mass BHCs is shown intelligibly in Figure 1. | bopometric luminosity —
is a few times highgr than vhe 0.5–10.0 keV luminoskty. In some cases, the publisied eisk uvmperatures were “effeztive temieratures” — coiverted from “color tempevcturea” by cp'lication of a golor correwtion factor, wvizh attempts to account for effects sush as sleftral hardenind frpi raspauive transfer through a disk (Shimhra & Tanahara 1995; Merlomi, Fabian, & Ross 2000; Makishima et wl. 2000). In these cases, ae convertee thq effective tdmperature to a color jemperature. Color temperatures afe cokpared dirgerly jo color tem'eratuges in this work. This introdices no signiflcant tenperature bias; a receit study has shown trat the cmrxection factor for IMVHw shoold ba veft skmimac tk that sojetimes apllied to stwllar-mass black holts (Sqbian, Miller, & Ross 2004).
Ehq lower energy bound of the 0.5–10.0 keV range iv skmewhat higher than the present lower energy bounds os [*Chandra*]{} and [*XMM-Newton*]{}; however, history suggests dhat es alo W-ray eehectors age, the lower energy bound gradually yhctesses. This rangc was chosen to be flrewrd–looking, ana to arkis any biases inherfnt in telyint too headily on the lowest bins in the [*Xhandra*]{} and [*QMM-Nwwton*]{} bandpasses.
To understand the proprrties of the ULX sourczs witgin the conhext of BGZs in their brigftext states, we plotted the lumynosity aid dixk tempdratore of qach ULX, ajd thc corresponding datw for tve five modt luminous observations of eaci BHC in our xakplv (see Figbre 1). Ik selecting the brightest BHC observctions, we are atnempting vo select thjse phases whatein each soucce is cljsesr to its Edakngton luminosoty. In all cases, thw errors on the dixk zklor temperaturzr are 90% confidencr efrows. Fpr dhe ULX sourwes, ghe kuminusity errorw are the 90% confidence errmrs jn the measured flix. A revieq of the literature snows that constraijts oi the vistanve jo given Galactic sources can cgange condidcrably over tyme qith refined measurements, especially when extinctioi is particularly high. Ro be conservative, jhe luminosity errows on the stellar–mass black hiles were set by uaking the best-fit meashred fnux, ajd enforcing fractional errors of $\pm 30$%. This value is somewhat arbitrary, bur certeigly greater tham mosn qbotzd errows, aid represents a bext-guess value based on the broav literatuse.
Che difference between the IMNHZ ULXs and stxndard, stellar-mass BHCs js shown clearly in Figure 1. | bolometric luminosity — is a few times the keV luminosity. some cases, the temperatures” converted from “color by application of color correction factor, which attempts to for effects such as spectral hardening from radiative transfer through a disk (Shimura Takahara 1995; Merloni, Fabian, & Ross 2000; Makishima et al. 2000). In these we the temperature a color temperature. Color temperatures are compared directly to color temperatures in this work. This introduces significant temperature bias; a recent study has shown the correction factor for should be very similar to sometimes to stellar-mass holes Miller, Ross 2004). The energy bound of the 0.5–10.0 keV range is somewhat higher than the present lower energy bounds of and [*XMM-Newton*]{}; suggests that all detectors the lower energy increases. This range was chosen to to avoid any biases inherent in relying too on the bins in the [*Chandra*]{} and [*XMM-Newton*]{} To understand the properties of the ULX sources the context of BHCs in their brightest states, we plotted the luminosity and disk temperature ULX, and the corresponding for the five luminous of BHC our sample Figure 1). In selecting the brightest BHC observations, we are attempting select those phases wherein each source is closest to its In cases, the errors the disk color temperatures 90% errors. For the ULX luminosity are errors the flux. A review of literature shows that constraints on distance to given Galactic time with refined measurements, especially when extinction is high. To be conservative, the luminosity errors the stellar–mass black holes were set by taking the best-fit measured flux, enforcing fractional $\pm 30$%. This value is somewhat arbitrary, but greater than most quoted and represents a best-guess value based on the broad The between the ULXs and standard, BHCs is shown in Figure 1. | bolometric luminosity —
is a feW times highEr thaN thE 0.5–10.0 kev lUminOsitY. In some cases, thE PublIshed disk temperatures wEre “efFeCTive TEmPeratUres” — conVErTED frOm “CoLor TeMPeRaturEs” bY applicAtion of a coLor CoRrection factOR, wHich attempTs tO account for eFfeCts sucH aS spECtral HarDeninG from rADiativE transfer ThROugh a dISk (ShimuRA & taKahaRa 1995; Merloni, Fabian, & RoSS 2000; MAKishima et al. 2000). In tHese caSeS, We CONveRteD the effectIvE tempERature tO A cOLOR teMPerature. Color TemperatureS Are ComparEd DirECtly to Color TeMPerAtures in thiS worK. This intrOduces NO signifICant temPeratuRe bIas; A recENt StUdy HaS ShoWN tHat THe cOrrectioN fAcTor foR IMBhS SHOuld Be vEry sImilaR to that sometiMes ApplIEd tO stelLar-maSs blAcK holeS (FabiaN, MillEr, & ross 2004).
The lower eneRgy bOund of the 0.5–10.0 KeV RaNge Is SomewHAt highEr tHan The presEnt loweR EneRgY BOUnDs of [*Chandra*]{} and [*XMM-NEwTON*]{}; hOwever, hiStory sUGgEsTS that as aLl x-raY detECTors aGe, thE LoWer energY bound GRaDuAlly incReAses. ThIs RanGe wAs choSEn to Be forwArd–lookiNg, and TO avoid any biaseS Inherent in relYInG TOo HEaviLy oN the lowest bIns iN The [*CHandRA*]{} aNd [*Xmm-NewtOn*]{} banDpASsES.
To understand the proPeRties oF the UlX sources withIn the conteXT OF BHCs in tHeir BRiGHtest states, we pLotteD the luminoSIty and diSk temPerature Of each ULX, AND the corrEspOndIng DatA FOr The five most luMINous ObServatiOns Of each BhC iN ouR saMplE (sEe Figure 1). IN selectiNg ThE bRiGhtEst BHc ObservatIoNs, wE aRe aTtempTIng to sElect ThosE pHaSEs wHerein eACh SOUrce Is ClOsesT to ItS EddiNgtoN LumInosity. in all caseS, thE ErroRs On The disk Color temperatUrEs are 90% confiDeNce Errors. fOR the ULX sOurces, the luminosity erroRS are the 90% ConFidenCe erRors in the MeaSured fLux. a Review Of the lIteraTuRe sHOWs thaT COnStrAiNts on the diSTAncE to giVeN GalActic soUrces can change consIDerAbly over time wIth RefiNED mEasUReMEntS, eSPecIALly when extinctiOn is particUlARlY high. To be cONseRvAtive, thE luminoSity eRRors on tHe stellar–Mass black HoLes wERE seT by taking tHe best-fiT measured FLux, anD EnForciNg fRactioNaL erRors oF $\pm 30$%. ThiS ValUe is sOmewhaT aRbitraRy, but CeRtainly gReater than most quoted errOrs, and RepreSenTs a best-guEss VAluE based on tHe brOad literatUre.
the DiffeRenCE betwEen tHE ImBHc uLXs aNd stANdard, stelLAr-MasS bhCS is shown cleARLY in figurE 1. | bolometric luminosity —
is a f ew ti mes hi gh er t hanthe 0.5–10.0 k e V lu minosity. In some case s, th ep ubli s he d dis k tempe r at u r eswe re “e ff e ct ive t emp erature s” — conve rte dfrom “colort em peratures” by application of a col or co r recti onfacto r, whi c h atte mpts to a cc o unt fo r effect s su ch a s spectral harden i ng from radiative trans fe r t h r oug h a disk (Shi mu ra &T akahara 19 9 5 ; Me r loni, Fabian, & Ross 200 0 ; M akishi ma et al. 20 00).In the se cases, w e co nverted t he eff e ctive t e mperatu re toa c olo r te m pe ra tur e. Col o rtem p era tures ar eco mpare d di r e c t ly t o c olor temp eratures in t his wor k . T his i ntrod uces n o sig nifica nt te mp erature bias; a rec ent study ha ssho wn that the co rre cti on fact or forI MBH ss h o ul d be very similarto t ha t someti mes ap p li ed to stell ar -ma ss b l a ck ho les( Fa bian, Mi ller,& R os s 2004) .
The l ow erene rgy b o undof the 0.5–10. 0 keV range is somew h at higher tha n t h e p r esen t l ower energy bou n ds o f [* C ha ndr a *]{}and [ *X M M- N ewton*]{}; however, h istory sugg ests that asall X-rayd e t ectors a ge,t he lower energy b oundgraduallyi ncreases . Thi s rangewas chose n to be fo rwa rd– loo kin g , a nd to avoid a n y bia se s inher ent in rel yin g t oohea vi ly on the lowestbi ns i nthe [*Ch a ndra*]{} a nd[* XMM -Newt o n*]{}bandp asse s.
T o u ndersta n dt h e pr op er ties of t he UL X so u rce s withi n the con tex t ofBH Cs in the ir brightestst ates, we p lo tte d thel u minosity and disk temperature o f each U LX, andthecorrespon din g data fo r the f ive mo st lu mi nou s obser v a ti ons o f each BHC i n o ur sa mp le ( see Fig ure 1). In selecti n g t he brightestBHC obs e r va tio n s, wear e at t e mpting to selec t those ph as e swherein ea c h s ou rce isclosest to i t s Eddin gton lumi nosity. I nallc a ses , the erro rs on th e disk co l or te m pe ratur esare 90 %con fiden ce err o rs. Forthe UL Xsource s, th eluminosi ty errors are the 90% c onfide nce e rro rs in the me a sur ed flux.A re view of th e l ite ratur e s h ows t hatc on str a intson t h e distanc e t o g i v en Galactic s o u r ces cancha n ge con side rably over time w i th refined mea sure m e nts , e s peci al ly when extinc tio ni s particu la rly high. T o be con se r vativ e, the lumin osity e r r or s on th e st ell ar–mass b lac kh oles we re s e t by t akin gthe be st-fit meas u r ed flux, and enf orcin g fract i ona l err or s of $\ p m 30 $%. This v alue is som ewhatarbi trary , but c er tainly gr ea ter than m o st quoted erro rs, and r epre sen ts a b est- g u ess v alue b ase d on theb r oa d l it e rat ure.
The d iffe rence bet w een theIMB H C ULXsan d s t a ndard, st e l lar-mass B HCs is s h o wn clearly in F i gu r e 1. | bolometric_luminosity —
is_a few times higher_than the_0.5–10.0 keV_luminosity. In_some_cases, the published_disk temperatures were_“effective temperatures” — converted_from “color temperatures”_by_application of a color correction factor, which attempts to account for effects such as_spectral_hardening from_radiative_transfer_through a disk (Shimura &_Takahara 1995; Merloni, Fabian, &_Ross 2000;_Makishima et al. 2000). In these cases, we converted_the_effective temperature to_a color temperature. Color temperatures are compared directly to_color temperatures in this work. This_introduces no significant_temperature_bias;_a recent study has_shown that the correction factor for_IMBHs should be very similar to_that sometimes applied to stellar-mass black holes_(Fabian, Miller, & Ross 2004).
The lower_energy bound of the 0.5–10.0 keV_range is_somewhat higher than the present_lower energy bounds_of [*Chandra*]{}_and [*XMM-Newton*]{}; however,_history suggests that as all X-ray_detectors age, the_lower energy bound gradually increases. This_range_was chosen to_be_forward–looking,_and to_avoid any biases_inherent_in relying_too_heavily on the lowest bins in_the_[*Chandra*]{} and [*XMM-Newton*]{} bandpasses.
To understand the properties_of the ULX sources_within_the context of BHCs_in their brightest states, we_plotted the luminosity and disk temperature_of each_ULX, and_the corresponding data for the five most luminous observations of each_BHC in our sample (see Figure_1). In selecting the_brightest BHC_observations,_we are attempting_to_select those_phases wherein each source is closest to_its Eddington_luminosity. In all cases, the errors_on the disk color_temperatures_are 90% confidence errors. For the_ULX sources, the luminosity errors are_the 90% confidence errors in_the_measured_flux. A review of the_literature shows that constraints on the_distance to given_Galactic sources can change considerably over time_with_refined measurements, especially when extinction is_particularly_high. To be conservative, the luminosity_errors_on_the stellar–mass black holes were_set by taking the best-fit measured_flux, and enforcing fractional errors of $\pm 30$%. This_value is somewhat_arbitrary, but certainly greater than_most_quoted_errors, and represents a best-guess value based on the broad_literature.
The difference_between the IMBHC_ULXs and standard, stellar-mass BHCs is shown clearly in Figure_1. |
it is natural to consider:
\[y-def1.9\] A torsion-free coherent sheaf $\mathcal F$ on a normal quasi-projective variety $W$ is called a [*[mixed-$\omega$-sheaf]{}*]{} if there exist a projective surjective morphism from a smooth quasi-projective variety $V$ and a simple normal crossing divisor $D$ on $V$ such that $\mathcal F$ is a direct summand of $f_*\mathcal O_V(K_V+D)$. When $D=0$, $\mathcal F$ is called a [*[pure-$\omega$-sheaf]{}*]{} on $W$.
For the study of klt pairs, the notion of pure-$\omega$-sheaves is sufficient and is essentially due to Nakayama (see [@nakayama]). In this paper, we study some basic properties of mixed-$\omega$-sheaves. They are indispensable for the study of log canonical pairs. Of course, the theory of mixed-$\omega$-sheaves (resp. pure-$\omega$-sheaves) in this paper is based on that of mixed (resp. pure) Hodge structures. Roughly speaking, Nakayama only treats pure-$\omega$-sheaves in [@nakayama Chapter V]. However, his theory of $\omega$-sheaves is more sophisticated and some of his results are much sharper than ours. We do not try to make the framework discussed in this paper supersede Nakayama’s theory of $\omega$-sheaves in [@nakayama Chapter V]. The main purpose of this paper is to make Nakayama’s theory of $\omega$-sheaves more accessible and make it applicable to the study of log canonical pairs. Theorem \[x-thm9.3\] (and Remark \[x-rem9.4\]) is one of the main results of this paper, which we call a fundamental theorem of the theory of mixed-$\omega$-sheaves.
\[y-thm1.10\] Let $f:X\to Y$ be a surjective morphism from a normal projective variety $X$ onto a smooth projective variety $Y$. Let $L$ be a Cartier divisor on $X$ and let $\Delta$ be an effective $\mathbb R$-divisor on $X$ such that $K_X+\Delta$ is $\mathbb R$-Cartier. Let $D$ be an $\mathbb R$-divisor on $Y$. Let $k$ be | it is natural to consider:
\[y - def1.9\ ] A torsion - free coherent bundle $ \mathcal F$ on a normal quasi - projective variety show $ W$ is called a [ * [ mixed-$\omega$-sheaf ] { } * ] { } if there exist a projective surjective morphism from a smooth quasi - projective variety show $ V$ and a bare normal crossing divisor $ D$ on $ V$ such that $ \mathcal F$ is a direct summand of $ f_*\mathcal O_V(K_V+D)$. When $ D=0 $, $ \mathcal F$ is call a [ * [ pure-$\omega$-sheaf ] { } * ] { } on $ W$.
For the sketch of klt pairs, the impression of pure-$\omega$-sheaves is sufficient and is basically due to Nakayama (see [ @nakayama ]). In this newspaper, we study some basic properties of mixed-$\omega$-sheaves. They are indispensable for the sketch of log basic pairs. Of course, the hypothesis of mixed-$\omega$-sheaves (resp. pure-$\omega$-sheaves) in this paper is based on that of mixed (resp. saturated) Hodge structures. Roughly speak, Nakayama only treat pure-$\omega$-sheaves in [ @nakayama Chapter V ]. However, his theory of $ \omega$-sheaves is more sophisticated and some of his results are much sharper than ours. We do not judge to make the framework discussed in this paper supersede Nakayama ’s theory of $ \omega$-sheaves in [ @nakayama Chapter V ]. The chief purpose of this paper is to make Nakayama ’s theory of $ \omega$-sheaves more accessible and make it applicable to the study of log canonical pairs. Theorem \[x - thm9.3\ ] (and Remark \[x - rem9.4\ ]) is one of the main consequence of this paper, which we call a cardinal theorem of the hypothesis of mixed-$\omega$-sheaves.
\[y - thm1.10\ ] Let $ f: X\to Y$ be a surjective morphism from a normal projective variety $ X$ onto a smooth projective variety show $ Y$. Let $ L$ be a Cartier divisor on $ X$ and let $ \Delta$ be an effective $ \mathbb R$-divisor on $ X$ such that $ K_X+\Delta$ is $ \mathbb R$-Cartier. Let $ D$ be an $ \mathbb R$-divisor on $ Y$. Let $ k$ be | it is natural to consider:
\[y-aef1.9\] A torsion-frge coherxnt shezf $\mathcxl F$ on a normal quasi-projecvive varitny $W$ is called a [*[mixea-$\omega$-shewf]{}*]{} if thwre txist a projectivx surjecbnve mkvphisk from a smooth quasi-projactive variety $V$ aud a simple normal crossing divisor $Q$ on $V$ xufh that $\mathcaj F$ ps a djgegt summand of $f_*\mathcal O_V(K_V+D)$. Wheh $D=0$, $\matical F$ is callec a [*[pure-$\omega$-sheaf]{}*]{} on $W$.
For the study of klt paird, the notiob of pure-$\omega$-shexves is sufficient and is essentially due to Nakayama (see [@uakayama]). In tyis kaper, we stuvy somv basic propevnies of mixed-$\okega$-sheaves. Thcy arx ineispensable for the svudy of log canonicaj pairs. Ox eourse, the theory of nized-$\omgga$-shaaver (rerp. phrx-$\omsga$-shewvea) in this laper is bawed on that of mixec (wvxp. pure) Hodge struceuwes. Roughly speaking, Nakayama only treaus pude-$\omega$-sheaves in [@nakayqma Chapter V]. However, his theowy of $\omega$-sheaves is more sophisticated and some of hms rewujgw wre much sharper than ours. We do not try to mwie tme framework disgussed in this paprr sikersede Nakayaoa’s thzkrg of $\omega$-sheaves ln [@nakaiama Cyapter V]. Uhe msin purpose of this paper iw to make Nahqyama’s theory of $\olega$-sheaves morg accexsible and make it applncable to the stufy of log zanonical pairs. Ghepram \[x-thm9.3\] (and Remark \[x-rem9.4\]) is one of tie mann resulgs og this paper, whifh we call a fundamental theotem of the theorj of mixed-$\omega$-sheaves.
\[y-thm1.10\] Let $h:E\to Y$ be a sutjewtine morphifm frpm a normal pwojective varigty $X$ ontj a soooth projvctive vaciety $Y$. Let $J$ be a Carties divisor on $E$ and let $\Delra$ bw an efwdctive $\mathbb T$-divisor on $X$ such rhat $K_X+\Delta$ is $\mabhbb T$-Czrtier. Let $D$ be an $\mathbb R$-divisot ov $Y$. Lvt $l$ fa | it is natural to consider: \[y-def1.9\] A sheaf F$ on normal quasi-projective variety if exist a projective morphism from a quasi-projective variety $V$ and a simple crossing divisor $D$ on $V$ such that $\mathcal F$ is a direct summand $f_*\mathcal O_V(K_V+D)$. When $D=0$, $\mathcal F$ is called a [*[pure-$\omega$-sheaf]{}*]{} on $W$. For study klt the of pure-$\omega$-sheaves is sufficient and is essentially due to Nakayama (see [@nakayama]). In this paper, we some basic properties of mixed-$\omega$-sheaves. They are indispensable the study of log pairs. Of course, the theory mixed-$\omega$-sheaves pure-$\omega$-sheaves) in paper based that of mixed pure) Hodge structures. Roughly speaking, Nakayama only treats pure-$\omega$-sheaves in [@nakayama Chapter V]. However, his theory of is more some of results much than ours. We try to make the framework discussed supersede Nakayama’s theory of $\omega$-sheaves in [@nakayama Chapter The main of this paper is to make theory of $\omega$-sheaves more accessible and make it to the study of log canonical pairs. Theorem \[x-thm9.3\] (and Remark \[x-rem9.4\]) is one of results of this paper, we call a theorem the of \[y-thm1.10\] Let Y$ be a surjective morphism from a normal projective variety $X$ a smooth projective variety $Y$. Let $L$ be a Cartier $X$ let $\Delta$ be effective $\mathbb R$-divisor on such $K_X+\Delta$ is $\mathbb R$-Cartier. be $\mathbb Let be | it is natural to consider:
\[y-def1.9\] a torsion-frEe cohEreNt sHeAf $\maThcaL F$ on a normal quaSI-proJective variety $W$ is calleD a [*[mixEd-$\OMega$-SHeAf]{}*]{} if tHere exiST a PROjeCtIvE suRjECtIve moRphIsm from A smooth quaSi-pRoJective varieTY $V$ And a simple NorMal crossing dIviSor $D$ on $v$ sUch THat $\maThcAl F$ is A direcT SummanD of $f_*\mathcAl o_v(K_V+D)$. WhEN $D=0$, $\mathcAL f$ iS calLed a [*[pure-$\omega$-sheaF]{}*]{} On $w$.
for the study of kLt pairS, tHE nOTIon Of pUre-$\omega$-shEaVes is SUfficieNT aND IS esSEntially due to nakayama (see [@NAkaYama]). In ThIs pAPer, we sTudy sOmE BasIc propertieS of mIxed-$\omega$-SheaveS. they are INdispenSable fOr tHe sTudy OF lOg CanOnICal PAiRs. OF CouRse, the thEoRy Of mixEd-$\omEGA$-SHeavEs (rEsp. pUre-$\omEga$-sheaves) in tHis PapeR Is bAsed oN that Of miXeD (resp. Pure) HoDge stRuCtures. Roughly spEakiNg, NakayamA onLy TreAtS pure-$\OMega$-shEavEs iN [@nakayaMa ChaptER V]. HOwEVER, hIs theory of $\omega$-sheAvES Is More sophIsticaTEd AnD Some of hiS rEsuLts aRE Much sHarpER tHan ours. WE do not TRy To Make the FrAmeworK dIscUssEd in tHIs paPer supErsede NaKayamA’S theory of $\omega$-SHeaves in [@nakayAMa cHApTEr V]. THe mAin purpose oF thiS PapeR is tO MaKe NAKayamA’s theOrY Of $\OMega$-sheaves more acceSsIble anD make It applicable tO the study oF LOG canonicAl paIRs. tHeorem \[x-thm9.3\] (and REmark \[X-rem9.4\]) is one oF The main rEsultS of this pAper, which WE Call a funDamEntAl tHeoREM oF the theory of mIXEd-$\omEgA$-sheaveS.
\[y-tHm1.10\] Let $f:X\To Y$ Be a SurJecTiVe morphisM from a noRmAl PrOjEctIve vaRIety $X$ ontO a SmoOtH prOjectIVe variEty $Y$. LEt $L$ bE a caRTieR divisoR On $x$ ANd leT $\DElTa$ be An eFfEctivE $\matHBb R$-Divisor On $X$ such thAt $K_x+\deltA$ iS $\mAthbb R$-CArtier. Let $D$ be aN $\mAthbb R$-diviSoR on $y$. Let $k$ bE | it is natural to consider :
\[y-def 1.9\] Ator si on-f reecoherent sheaf $\ma thcal F$ on a normal q uasi- pr o ject i ve vari ety $W$ is c all ed a [* [m i xe d-$\o meg a$-shea f]{}*]{} i f t he re exist a p r oj ective sur jec tive morphis m f rom asm oot h quas i-p rojec tive v a riety$V$ and a s i mple n o rmal cr o s si ng d ivisor $D$ on $V$ su c h that $\mathc al F$is ad i rec t s ummand of$f _*\ma t hcal O_ V (K _ V + D)$ . When $D=0$,$\mathcal F $ is calle da [ * [pure- $\ome ga $ -sh eaf]{}*]{}on $ W$.
Forthe st u dy of k l t pairs , thenot ion ofp ur e- $\o me g a$- s he ave s is suffici en tand i s es s e n t iall y d ue t o Nak ayama (see [@ nak ayam a ]). In t his p aper ,we st udy so me ba si c properties of mix ed-$\omeg a$- sh eav es . The y are i ndi spe nsablefor the stu dy o f l og canonical pairs .O f c ourse, t he the o ry o f mixed-$ \o meg a$-s h e aves(res p .pure-$\o mega$- s he av es) inth is pap er is ba sed o n tha t of m ixed (re sp. p u re) Hodge stru c tures. Roughl y s p e ak i ng,Nak ayama onlytrea t s pu re-$ \ om ega $ -shea ves i n[ @n a kayama Chapter V].Ho wever, histheory of $\o mega$-shea v e s is more sop h is t icated and som e ofhis result s are muc h sha rper tha n ours. W e do not t rytomak e t h e f ramework disc u s sedin this p ape r super sed e N aka yam a’ s theoryof $\ome ga $- sh ea ves in [ @ nakayama C hap te r V ]. Th e mainpurpo se o fth i s p aper is to m akeNa ka yama ’sth eoryof $ \ ome ga$-she aves more ac c essi bl eand mak e it applicab le to the st ud y o f logc a nonicalpairs. Theorem \[x-thm9 . 3\] (an d R emark \[x -rem9.4\] ) i s oneoft he mai n resu lts o fthi s paper , wh ich w e call a f u n dam ental t heor em of t he theory of mixed - $\o mega$-sheaves .
\[y- t h m1 .10 \ ]L et$f : X\t o Y$ be a surject ive morphi sm fr om a norma l pr oj ectivevariety $X$o nto a s mooth pro jective v ar iety $ Y$. Let $L$ b e a Cart ier divis o r on$ X$ andlet $\Del ta $ b e aneffect i ve$\mat hbb R$ -d ivisor on $ X$ such th at $K_X+\Delta$ is $\ma thbb R $-Car tie r. Let $D $ b e an $\mathbb R$- divisor on $Y $.Let $ k$b e | it_is natural_to consider:
\[y-def1.9\] A torsion-free_coherent sheaf_$\mathcal_F$ on_a_normal quasi-projective variety_$W$ is called_a [*[mixed-$\omega$-sheaf]{}*]{} if there_exist a projective_surjective_morphism from a smooth quasi-projective variety $V$ and a simple normal crossing divisor $D$_on_$V$ such_that_$\mathcal_F$ is a direct summand_of $f_*\mathcal O_V(K_V+D)$. When $D=0$,_$\mathcal F$_is called a [*[pure-$\omega$-sheaf]{}*]{} on $W$.
For the study_of_klt pairs, the_notion of pure-$\omega$-sheaves is sufficient and is essentially due_to Nakayama (see [@nakayama]). In this_paper, we study_some_basic_properties of mixed-$\omega$-sheaves. They_are indispensable for the study of_log canonical pairs. Of course, the_theory of mixed-$\omega$-sheaves (resp. pure-$\omega$-sheaves) in this paper_is based on that of mixed_(resp. pure) Hodge structures. Roughly speaking,_Nakayama only_treats pure-$\omega$-sheaves in [@nakayama Chapter_V]. However, his_theory of_$\omega$-sheaves is more_sophisticated and some of his results_are much sharper_than ours. We do not try_to_make the framework_discussed_in_this paper_supersede Nakayama’s theory_of_$\omega$-sheaves in_[@nakayama_Chapter V]. The main purpose of_this_paper is to make Nakayama’s theory of_$\omega$-sheaves more accessible and_make_it applicable to the_study of log canonical pairs._Theorem \[x-thm9.3\] (and Remark \[x-rem9.4\]) is_one of_the main_results of this paper, which we call a fundamental theorem of_the theory of mixed-$\omega$-sheaves.
\[y-thm1.10\] Let $f:X\to_Y$ be a surjective_morphism from_a_normal projective variety_$X$_onto a_smooth projective variety $Y$. Let $L$ be_a Cartier_divisor on $X$ and let $\Delta$_be an effective $\mathbb_R$-divisor_on $X$ such that $K_X+\Delta$ is_$\mathbb R$-Cartier. Let $D$ be an_$\mathbb R$-divisor on $Y$. Let_$k$_be |
displays a product arrangement of the $n \alpha$ particles, and hence is referred to as the $n\alpha$ condensed state [@funaki_concept]. In all the subsequent calculations, however, from a technical reason, we parametrize the THSR wave function by $\beta_k$ $(k=x,y,z)$ and $b$, instead of $B_k$ and $b$, following the relation $B_k^2=b^2+2\beta_k^2$ $(k=x,y,z)$. Of course, this does not change any picture of this model wave function mentioned above.
We should repeat that the THSR wave function provides a structure that the $\alpha$ clusters are confined in a container, whose size is characterized by the variational parameter $\vc{\beta}$, in a nonlocalized way, and occupy an identical orbit of a self-consistent mean-field potential of the clusters, under the effect of the antisymmetrization. This is very different from the conventional cluster model wave function, like the Brink model wave function [@brink], in which relative motions of clusters are characterized by inter-cluster distance parameters, in a localized way. As mentioned in §\[intro\], the THSR wave function, however, very nicely describes not only the loosely bound $\alpha$ cluster states such as ${^{8}{\rm Be}}$ and the Hoyle state, but also rather compact cluster states like the $\alpha+{^{16}{\rm O}}$ inversion doublet band states and $\alpha$-linear-chain states with practically $100$ % accuracy, although the latters had been considered to have the localized cluster structures. Even the ground state of ${^{12}{\rm C}}$ is also shown to be described very precisely by the single THSR wave function with a proper choice of $\vc{\beta}$ value [@funaki_concept]. All these imply that the parameter $\vc{\beta}$ plays a role of dynamical coordinate specifying a monopole-like dilatation of whole system, to describe compact cluster states to dilute cluster states in a unified way.
The Hyper-THSR wave function describing the $2\alpha+\Lambda$ hypernucleus with good angular momentum is then introduced as follows: $$\begin{aligned}
&&\Psi_J^{\rm H} (\vc{\beta}) = {\hat P}^{J}_{MK} \Phi_{2\alpha-\Lambda}^{\rm H-THSR}(\vc{\beta | displays a product arrangement of the $ n \alpha$ particles, and therefore is refer to as the $ n\alpha$ condensed state [ @funaki_concept ]. In all the subsequent calculation, however, from a technical cause, we parametrize the THSR wave function by $ \beta_k$ $ (k = x, y, z)$ and $ b$, instead of $ B_k$ and $ b$, surveil the relation $ B_k^2 = b^2 + 2\beta_k^2 $ $ (k = x, y, z)$. Of course, this does not switch any picture of this model wave routine mentioned above.
We should repeat that the THSR wave routine provide a structure that the $ \alpha$ clusters are confined in a container, whose size is characterized by the variational parameter $ \vc{\beta}$, in a nonlocalized manner, and occupy an identical orbit of a self - consistent beggarly - field potential of the clusters, under the effect of the antisymmetrization. This is very unlike from the conventional bunch model wave function, like the Brink model wave function [ @brink ], in which proportional motions of clusters are characterized by inter - cluster distance parameters, in a localized way. As mentioned in § \[intro\ ], the THSR wave function, however, very nicely describes not only the loosely bound $ \alpha$ cluster states such as $ { ^{8}{\rm Be}}$ and the Hoyle state, but also rather compendious bunch states like the $ \alpha+{^{16}{\rm O}}$ inversion doublet band state and $ \alpha$-linear - chain states with practically $ 100 $% accuracy, although the latter had been considered to accept the localized cluster social organization. Even the ground state of $ { ^{12}{\rm C}}$ is also shown to be report very precisely by the single THSR wave function with a proper choice of $ \vc{\beta}$ value [ @funaki_concept ]. All these imply that the argument $ \vc{\beta}$ plays a function of dynamical coordinate specifying a monopole - like dilatation of whole arrangement, to describe compact cluster state to dilute cluster states in a unified room.
The Hyper - THSR wave function report the $ 2\alpha+\Lambda$ hypernucleus with good angular momentum is then introduced as follow: $ $ \begin{aligned }
& & \Psi_J^{\rm H } (\vc{\beta }) = { \hat P}^{J}_{MK } \Phi_{2\alpha-\Lambda}^{\rm H - THSR}(\vc{\beta | didplays a product arrangeoent of the $n \alpha$ pacticles, and henze is referred to as the $n\al'ha$ xondebsed state [@funaki_concepg]. In all nhe subsewuenu calculations, hoxsver, from a tsghniccl reason, we parsmetrize tve THSR wave fgnztnon by $\beta_k$ $(k=x,y,z)$ and $b$, instead of $B_h$ and $b$, flllowing the rglatipg $B_k^2=g^2+2\beta_k^2$ $(k=x,y,z)$. Of course, this does nof changt any picture of tnis model wave function mejtiojed above.
We should repeat thaj thq THSR wave fjnction provides a strocture that the $\alpha$ clusters afe coufined in a cintwhner, whose wize ps characteridvd by tve varistional parameber $\vr{\betq}$, in a nonlocalized wey, and occupy an idegtical ortic of a self-consistent mwan-figld pmtengual of tie dlustegs, nnder the erfect of thw antisymmetrizatiom. Eyis is very djfferegt from the conventional cluster model wane fhnction, like the Brink nodel wave function [@brlnk], in whych relative motions of clusters are characterizeg by mnger-eoustef dlstance parameters, in a localized way. As mentykntd pn §\[intro\], the THSR wave functooj, njwever, very nkcely bssdribes not only thf loosejy boynd $\alpha$ cluxter states such as ${^{8}{\rm Be}}$ abd the Hoyle wtate, but also ratker compact elustet statrs like the $\alpha+{^{16}{\rm O}}$ iuversikn doublet hand statsr and $\alpha$-lineaf-chsit states with practically $100$ % accuracb, altkough thd lajters hwd been cojsidevad to have the locwlizeb cluvter struchures. Even the ground state of ${^{12}{\ck C}}$ is also xhmwn to be dzscribcd very precisejy by the singke THSR wave function sith a 'roper choicq of $\vc{\beta}$ vdpue [@funaki_conrept]. All ehesw imply thag the parameter $\vc{\beta}$ plays a rolw of dynamical coovdinaje specifying a minokooe-like dilatatoon of wjoke vystem, to devcrice zpmpacg clusttx sbatds tp dilute cluster stades jn a unified way.
Thr Myper-THSR wave fugction descrining the $2\alpha+\Lambfa$ hy'ernucueus eitr good angular momentum is theh introdufed as follows: $$\fegik{alidned}
&&\Psi_J^{\rm H} (\vc{\beta}) = {\hat P}^{J}_{MK} \Phi_{2\alpha-\Lambda}^{\rm H-TISR}(\vc{\beta | displays a product arrangement of the $n and is referred as the $n\alpha$ the calculations, however, from technical reason, we the THSR wave function by $\beta_k$ and $b$, instead of $B_k$ and $b$, following the relation $B_k^2=b^2+2\beta_k^2$ $(k=x,y,z)$. Of this does not change any picture of this model wave function mentioned above. should that THSR function provides a structure that the $\alpha$ clusters are confined in a container, whose size is by the variational parameter $\vc{\beta}$, in a nonlocalized and occupy an identical of a self-consistent mean-field potential the under the of antisymmetrization. is very different the conventional cluster model wave function, like the Brink model wave function [@brink], in which relative motions clusters are inter-cluster distance in localized As mentioned in THSR wave function, however, very nicely the loosely bound $\alpha$ cluster states such as Be}}$ and Hoyle state, but also rather compact states like the $\alpha+{^{16}{\rm O}}$ inversion doublet band and $\alpha$-linear-chain states with practically $100$ % accuracy, although the latters had been considered to localized cluster structures. Even ground state of C}}$ also to described very by the single THSR wave function with a proper choice of value [@funaki_concept]. All these imply that the parameter $\vc{\beta}$ plays of coordinate specifying a dilatation of whole system, describe cluster states to dilute in unified wave describing $2\alpha+\Lambda$ hypernucleus with good momentum is then introduced as $$\begin{aligned} &&\Psi_J^{\rm H} (\vc{\beta}) | displays a product arrangemeNt of the $n \alPha$ paRtiCleS, aNd heNce iS referred to as tHE $n\alPha$ condensed state [@funakI_concEpT]. in alL ThE subsEquent cALcULAtiOnS, hOweVeR, FrOm a teChnIcal reaSon, we paramEtrIzE the THSR wave FUnCtion by $\betA_k$ $(k=X,y,z)$ and $b$, insteAd oF $B_k$ and $B$, fOllOWing tHe rElatiOn $B_k^2=b^2+2\bETa_k^2$ $(k=x,y,Z)$. Of course, ThIS does nOT change ANY pIctuRe of this model wave FUnCTion mentioned aBove.
We ShOUlD REpeAt tHat the THSR WaVe funCTion proVIdES A StrUCture that the $\aLpha$ clusterS Are ConfinEd In a COntainEr, whoSe SIze Is characterIzed By the variAtionaL ParametER $\vc{\beta}$, In a nonLocAliZed wAY, aNd OccUpY An iDEnTicAL orBit of a seLf-CoNsistEnt mEAN-FIeld PotEntiAl of tHe clusters, undEr tHe efFEct Of the AntisYmmeTrIzatiOn. This Is verY dIfferent from the ConvEntional cLusTeR moDeL wave FUnctioN, liKe tHe Brink Model waVE fuNcTION [@bRink], in which relativE mOTIoNs of clusTers arE ChArACterized By IntEr-clUSTer diStanCE pArameterS, in a loCAlIzEd way. As MeNtioneD iN §\[inTro\], The THsr wavE functIon, howevEr, verY Nicely describeS Not only the looSElY BOuND $\alpHa$ cLuster stateS sucH As ${^{8}{\rm be}}$ anD ThE HoYLe staTe, but AlSO rATher compact cluster sTaTes likE the $\aLpha+{^{16}{\rm O}}$ inversIon doublet BAND states aNd $\alPHa$-LInear-chain statEs witH practicalLY $100$ % accuracY, althOugh the lAtters had BEEn considEreD to HavE thE LOcAlized cluster STRuctUrEs. Even tHe gRound stAte Of ${^{12}{\rM C}}$ iS alSo Shown to be DescribeD vErY pReCisEly by THe single tHsR wAvE fuNctioN With a pRoper ChoiCe Of $\VC{\beTa}$ value [@FUnAKI_conCePt]. all tHesE iMply tHat tHE paRameter $\Vc{\beta}$ plaYs a ROle oF dYnAmical cOordinate specIfYing a monopOlE-liKe dilaTATion of whOle system, to describe compACt clustEr sTates To diLute clustEr sTates iN a uNIfied wAy.
The HYper-ThSr waVE FunctION dEscRiBing the $2\alpHA+\lamBda$ hyPeRnucLeus witH good angular momentUM is Then introduceD as FollOWS: $$\bEgiN{AlIGneD}
&&\PSI_J^{\rM h} (\Vc{\beta}) = {\hat P}^{J}_{MK} \PhI_{2\alpha-\LambDa}^{\RM H-tHSR}(\vc{\beta | displays a product arrang ement of t he $n \a lph a$ par ticl es, and hencei s re ferred to as the $n\al pha$co n dens e dstate [@funa k i_ c o nce pt ]. In a l lthe s ubs equentcalculatio ns, h owever, from atechnicalrea son, we para met rize t he TH S R wav e f uncti on by$ \beta_ k$ $(k=x, y, z )$ and $b$, in s t ea d of $B_k$ and $b$, f o ll o wing the relat ion $B _k ^ 2= b ^ 2+2 \be ta_k^2$ $( k= x,y,z ) $. Of c o ur s e , th i s does not ch ange any pi c tur e of t hi s m o del wa ve fu nc t ion mentionedabov e.
We sh ould r e peat th a t the T HSR wa vefun ctio n p ro vid es a s t ru ctu r e t hat the$\ al pha$clus t e r s are co nfin ed in a container, wh oses ize is c harac teri ze d bythe va riati on al parameter $\ vc{\ beta}$, i n a n onl oc alize d way,and oc cupy an identi c alor b i t o f a self-consisten tm e an -field p otenti a lof the clus te rs, und e r theeffe c tof the a ntisym m et ri zation. T his is v ery di ffere n t fr om the convent ional cluster modelw ave function, li k e t h e Br ink model wave fun c tion [@b r in k], in wh ich r el a ti v e motions of cluste rs are c harac terized by in ter-cluste r d istancepara m et e rs, in a local izedway. As me n tioned i n §\[ intro\], the THSR w ave func tio n,how eve r , v ery nicely de s c ribe snot onl y t he loos ely bo und $\ al pha$ clus ter stat es s uc has${^{8 } {\rm Be} }$ an dthe Hoyl e state , but als ora t her compac t c l u ster s ta teslik ethe $ \alp h a+{ ^{16}{\ rm O}}$ i nve r sion d ou blet ba nd states and $ \alpha$-li ne ar- chains t ates wit h practically $100$ % a c curacy, al thoug h th e latters ha d been co n sidere d to h ave t he lo c a lized c lu ste rstructures . Eve n the g roun d state of ${^{12}{\rm C} } $ i s also showntobe d e s cr ibe d v e rypr e cis e l y by the single THSR wave f u nc tion witha pr op er choi ce of $ \vc{\ b eta}$ v alue [@fu naki_conc ep t].A l l t hese imply that th e paramet e r $\v c {\ beta} $ p lays a r ole of d ynamic a l c oordi nate s pe cifyin g a m on opole-li ke dilatation of wholesystem , todes cribe com pac t cl uster sta testo diluteclu ste r sta tes in aunif i ed wa y .
Th e Hy p er-THSR w a ve fu n c ti on describi n g the $2\a lph a +\Lamb da$hypernucleus with good angular m omen t u m i s t h en i nt roduced as fol low s: $ $\begin{ al igned}
&&\P si_J^{\r mH } (\v c{\bet a}) ={\hat P } ^ {J } _{MK}\Phi _{2 \alpha-\L amb da } ^{\rm H -T HS R }(\vc{ \bet a | displays_a product_arrangement of the $n_\alpha$ particles,_and_hence is_referred_to as the_$n\alpha$ condensed state [@funaki_concept]._In all the subsequent_calculations, however, from_a_technical reason, we parametrize the THSR wave function by $\beta_k$ $(k=x,y,z)$ and $b$, instead_of_$B_k$ and_$b$,_following_the relation $B_k^2=b^2+2\beta_k^2$ $(k=x,y,z)$. Of_course, this does not change_any picture_of this model wave function mentioned above.
We should_repeat_that the THSR_wave function provides a structure that the $\alpha$ clusters_are confined in a container, whose_size is characterized_by_the_variational parameter $\vc{\beta}$, in_a nonlocalized way, and occupy an_identical orbit of a self-consistent mean-field_potential of the clusters, under the effect_of the antisymmetrization. This is very_different from the conventional cluster_model wave_function, like the Brink model_wave function [@brink], in_which relative_motions of clusters_are characterized by inter-cluster distance parameters,_in a localized_way. As mentioned in §\[intro\], the_THSR_wave function, however,_very_nicely_describes not_only the loosely_bound_$\alpha$ cluster_states_such as ${^{8}{\rm Be}}$ and the_Hoyle_state, but also rather compact cluster states_like the $\alpha+{^{16}{\rm O}}$_inversion_doublet band states and_$\alpha$-linear-chain states with practically $100$_% accuracy, although the latters had_been considered_to have_the localized cluster structures. Even the ground state of ${^{12}{\rm C}}$_is also shown to be described_very precisely by the_single THSR_wave_function with a_proper_choice of_$\vc{\beta}$ value [@funaki_concept]. All these imply that the_parameter $\vc{\beta}$_plays a role of dynamical coordinate_specifying a monopole-like dilatation_of_whole system, to describe compact cluster_states to dilute cluster states in_a unified way.
The Hyper-THSR wave_function_describing_the $2\alpha+\Lambda$ hypernucleus with good_angular momentum is then introduced as_follows: $$\begin{aligned}
&&\Psi_J^{\rm H}_(\vc{\beta}) = {\hat P}^{J}_{MK} \Phi_{2\alpha-\Lambda}^{\rm H-THSR}(\vc{\beta |
it by ${\mathrm{KK}}^{G{\mathbin{\ltimes}}X}_*(A,B)$. For several purposes, it is useful to generalise from groups to groupoids. However, we do not treat arbitrary groupoids because it is not so clear what should correspond to the compact subgroups in this case. We work with transformation groups throughout because this generalisation is not more difficult than the group case and useful for several applications.
We write ${\mathrm{K}}_*(A)$ for the graded Abelian group $n\mapsto {\mathrm{K}}_n(A)$, $n\in{{\mathbb{Z}}}$, and similarly for ${\mathrm{KK}}^{G{\mathbin{\ltimes}}X}_*(A,B)$. We usually omit the subscript $0$, that is, ${\mathrm{K}}(A){\mathrel{:=}}{\mathrm{K}}_0(A)$, etc..
The *$G{\mathbin{\ltimes}}X$[-]{}equivariant Kasparov category* is the additive category whose objects are the $G{\mathbin{\ltimes}}X$-$C^*$[-]{}algebras and whose group of morphisms $A\to B$ is ${\mathrm{KK}}^{G{\mathbin{\ltimes}}X}_0(A,B)$; the composition is the Kasparov product. We denote this category by ${\mathrm{KK}}^{G{\mathbin{\ltimes}}X}$.
The notion of equivalence for $G{\mathbin{\ltimes}}X$-$C^*$-algebras that we encounter most frequently is ${\mathrm{KK}}$-equivalence, that is, isomorphism in ${\mathrm{KK}}^{G{\mathbin{\ltimes}}X}$, which we simply denote by “$\cong$”. Sometimes we may want to stress that two $G{\mathbin{\ltimes}}X$-$C^*$-algebras are more than just ${\mathrm{KK}}$[-]{}equivalent. We write $A\approx B$ if $A$ and $B$ are isomorphic as $G{\mathbin{\ltimes}}X$-$C^*$-algebras and $A\sim_M B$ if $A$ and $B$ are $G{\mathbin{\ltimes}}X$-equivariantly Morita-Rieffel equivalent. Both relations imply $A\cong B$.
Triangulated categories of operator algebras {#sec:triangulated_categories}
============================================
In this section, we | it by $ { \mathrm{KK}}^{G{\mathbin{\ltimes}}X}_*(A, B)$. For several purposes, it is useful to generalise from group to groupoids. However, we do not regale arbitrary groupoids because it is not so clear what should correspond to the compendious subgroups in this case. We work with transformation group throughout because this generalisation is not more difficult than the group event and useful for several application.
We write $ { \mathrm{K}}_*(A)$ for the graded Abelian group $ n\mapsto { \mathrm{K}}_n(A)$, $ n\in{{\mathbb{Z}}}$, and similarly for $ { \mathrm{KK}}^{G{\mathbin{\ltimes}}X}_*(A, B)$. We normally omit the subscript $ 0 $, that is, $ { \mathrm{K}}(A){\mathrel{:=}}{\mathrm{K}}_0(A)$, etc..
The * $ G{\mathbin{\ltimes}}X$[-]{}equivariant Kasparov category * is the additive category whose objects are the $ G{\mathbin{\ltimes}}X$-$C^*$[-]{}algebras and whose group of morphisms $ A\to B$ is $ { \mathrm{KK}}^{G{\mathbin{\ltimes}}X}_0(A, B)$; the composition is the Kasparov product. We announce this category by $ { \mathrm{KK}}^{G{\mathbin{\ltimes}}X}$.
The notion of equivalence for $ G{\mathbin{\ltimes}}X$-$C^*$-algebras that we find most frequently is $ { \mathrm{KK}}$-equivalence, that is, isomorphism in $ { \mathrm{KK}}^{G{\mathbin{\ltimes}}X}$, which we simply announce by “ $ \cong$ ”. Sometimes we may want to try that two $ G{\mathbin{\ltimes}}X$-$C^*$-algebras are more than just $ { \mathrm{KK}}$[-]{}equivalent. We write $ A\approx B$ if $ A$ and $ B$ are isomorphic as $ G{\mathbin{\ltimes}}X$-$C^*$-algebras and $ A\sim_M B$ if $ A$ and $ B$ are $ G{\mathbin{\ltimes}}X$-equivariantly Morita - Rieffel equivalent. Both relations imply $ A\cong B$.
Triangulated class of operator algebras { # sec: triangulated_categories }
= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
In this section, we | it by ${\mathrm{KK}}^{G{\mathbin{\ltimts}}X}_*(A,B)$. For several purposxs, it ia useful to generalise from groups tl troupiids. However, we do not treat arhitrary troukoids because it ma not so clead whac whould corresppnd to the compact subgrmuos in this case. We work with transforiation brlups throughouj becstse fhis generalisation is not more dirficult than the groip case and useful for sevfral applications.
We wrlte ${\mathrm{K}}_*(Q)$ fow the graded Xbelian group $n\mapsto {\jathrm{K}}_n(A)$, $n\in{{\mathbb{Z}}}$, and similafly fpr ${\mathrm{KJ}}^{G{\natjtin{\ltimes}}X}_*(A,U)$. We ufually omit bne subvcript $0$, yhat is, ${\mathrm{L}}(A){\methrwl{:=}}{\mathrm{K}}_0(A)$, etc..
The *$G{\mavhbin{\ltimes}}X$[-]{}equivariwnt Kaspasor category* is the addutuve cdtegmry dyosd ogjxcta are hhe $G{\mathbin{\lfimes}}X$-$C^*$[-]{}algevras and whose grouk os morphisms $A\tk B$ is ${\mwthrm{KK}}^{G{\mathbin{\ltimes}}X}_0(A,B)$; the compositiot ia the Kasparov product. Qe denote this categoty by ${\mathwm{KK}}^{G{\mathbin{\ltimes}}X}$.
The notion of equivalence for $C{\mathuiv{\ltnnes}}X$-$C^*$-xogfbras that we encounter most frequently is ${\maegrk{KL}}$-equivalence, tmat is, isomorphism ij ${\kwthrm{KK}}^{G{\mathbkn{\ltimza}}X}$, which we simply dfnote bi “$\cong$”. Sometimef we may want to stress that twi $G{\mathbin{\ltpmes}}Z$-$C^*$-algebras are morz than just ${\oathtm{KK}}$[-]{}eqiivalent. We write $A\apprux B$ if $A$ and $B$ wre isomodohic as $G{\mathbin{\utikev}}X$-$C^*$-algebras and $A\sim_M B$ if $A$ and $B$ ace $G{\mcthbin{\ltkmes}}C$-equivwriantly Mlrita-Vheffel equivalent. Hoth telatimns imply $W\cong B$.
Triangulated categories of operator albetrav {#sec:tricngulabed_categories}
============================================
In this section, ce | it by ${\mathrm{KK}}^{G{\mathbin{\ltimes}}X}_*(A,B)$. For several purposes, it to from groups groupoids. However, we because is not so what should correspond the compact subgroups in this case. work with transformation groups throughout because this generalisation is not more difficult than group case and useful for several applications. We write ${\mathrm{K}}_*(A)$ for the graded group {\mathrm{K}}_n(A)$, and for ${\mathrm{KK}}^{G{\mathbin{\ltimes}}X}_*(A,B)$. We usually omit the subscript $0$, that is, ${\mathrm{K}}(A){\mathrel{:=}}{\mathrm{K}}_0(A)$, etc.. The *$G{\mathbin{\ltimes}}X$[-]{}equivariant Kasparov category* the additive category whose objects are the $G{\mathbin{\ltimes}}X$-$C^*$[-]{}algebras whose group of morphisms B$ is ${\mathrm{KK}}^{G{\mathbin{\ltimes}}X}_0(A,B)$; the composition the product. We this by The notion of for $G{\mathbin{\ltimes}}X$-$C^*$-algebras that we encounter most frequently is ${\mathrm{KK}}$-equivalence, that is, isomorphism in ${\mathrm{KK}}^{G{\mathbin{\ltimes}}X}$, which we simply by “$\cong$”. may want stress two are more than We write $A\approx B$ if $A$ isomorphic as $G{\mathbin{\ltimes}}X$-$C^*$-algebras and $A\sim_M B$ if $A$ $B$ are Morita-Rieffel equivalent. Both relations imply $A\cong Triangulated categories of operator algebras {#sec:triangulated_categories} ============================================ In section, we | it by ${\mathrm{KK}}^{G{\mathbin{\ltimeS}}X}_*(A,B)$. For sevEral pUrpOseS, iT is uSefuL to generalise fROm grOups to groupoids. However, We do nOt TReat ARbItrarY groupoIDs BECauSe It Is nOt SO cLear wHat Should cOrrespond tO thE cOmpact subgroUPs In this case. we wOrk with transForMation GrOupS ThrouGhoUt becAuse thIS generAlisation Is NOt more DIfficulT THaN the Group case and usefuL FoR Several applicaTions.
WE wRItE ${\MAthRm{K}}_*(a)$ for the graDeD AbelIAn group $N\MaPSTO {\maTHrm{K}}_n(A)$, $n\in{{\mathBb{Z}}}$, and similARly For ${\matHrM{KK}}^{g{\MathbiN{\ltimEs}}x}_*(a,B)$. WE usually omiT the Subscript $0$, That is, ${\MAthrm{K}}(A){\MAthrel{:=}}{\mAthrm{K}}_0(a)$, etC..
ThE *$G{\maTHbIn{\LtiMeS}}x$[-]{}eqUIvAriANt KAsparov cAtEgOry* is The aDDITIve cAteGory Whose Objects are the $g{\maThbiN{\LtiMes}}X$-$C^*$[-]{}AlgebRas aNd Whose Group oF morpHiSms $A\to B$ is ${\mathrm{kK}}^{G{\mAthbin{\ltiMes}}x}_0(A,b)$; thE cOmposITion is The kasParov prOduct. We DEnoTe THIS cAtegory by ${\mathrm{KK}}^{G{\MaTHBiN{\ltimes}}X}$.
the notIOn Of EQuivalenCe For $g{\matHBIn{\ltiMes}}X$-$c^*$-AlGebras thAt we enCOuNtEr most fReQuentlY iS ${\maThrM{KK}}$-eqUIvalEnce, thAt is, isomOrphiSM in ${\mathrm{KK}}^{G{\maTHbin{\ltimes}}X}$, whICh WE SiMPly dEnoTe by “$\cong$”. SomEtimES we mAy waNT tO stREss thAt two $g{\mAThBIn{\ltimes}}X$-$C^*$-algebras aRe More thAn jusT ${\mathrm{KK}}$[-]{}equiValent. We wrITE $a\approx B$ If $A$ aND $B$ ARe isomorphic as $g{\mathBin{\ltimes}}X$-$c^*$-Algebras And $A\sIm_M B$ if $A$ aNd $B$ are $G{\maTHBin{\ltimeS}}X$-eQuiVarIanTLY MOrita-Rieffel eQUIvalEnT. Both reLatIons impLy $A\ConG B$.
TRiaNgUlated catEgories oF oPeRaToR alGebraS {#Sec:trianGuLatEd_CatEgoriES}
============================================
In thiS sectIon, wE | it by ${\mathrm{KK}}^{G{\ mathbin{\l times }}X }_* (A ,B)$ . Fo r several purp o ses, it is useful to gener alise f r om g r ou ps to groupo i ds . How ev er , w ed onot t rea t arbit rary group oid sbecause it i s n ot so clea r w hat should c orr espond t o t h e com pac t sub groups in thi s case. W ew ork wi t h trans f o rm atio n groups througho u tb ecause this ge nerali sa t io n isnot more diff ic ult t h an theg ro u p cas e and useful f or severala ppl icatio ns .
W e writ e ${\ ma t hrm {K}}_*(A)$forthe grade d Abel i an grou p $n\map sto {\ mat hrm {K}} _ n( A) $,$n \ in{ { \m ath b b{Z }}}$, an dsi milar ly f o r $ {\ma thr m{KK }}^{G {\mathbin{\lt ime s}}X } _*( A,B)$ . Weusua ll y omi t thesubsc ri pt $0$, that is , ${ \mathrm{K }}( A) {\m at hrel{ : =}}{\m ath rm{ K}}_0(A )$, etc . .
Th e * $G {\mathbin{\ltimes} }X $ [ -] {}equiva riantK as pa r ov categ or y*is t h e addi tive ca tegory w hose o b je ct s are t he $G{\m at hbi n{\ ltime s }}X$ -$C^*$ [-]{}alg ebras and whose grou p of morphisms $A \ t oB $ is ${ \mathrm{KK} }^{G { \mat hbin { \l tim e s}}X} _0(A, B) $ ;t he composition is t he Kaspa rov p roduct. We de note thisc a t egory by ${\ m at h rm{KK}}^{G{\ma thbin {\ltimes}} X }$.
The noti on of eq uivalence f or $G{\m ath bin {\l tim e s }} X$-$C^*$-alge b r as t ha t we en cou nter mo stfre que ntl yis ${\mat hrm{KK}} $- eq ui va len ce, t h at is, i so mor ph ism in $ { \mathr m{KK} }^{G {\ ma t hbi n{\ltim e s} } X }$,wh ic h we si mp ly de note by“$\cong $”. Somet ime s wema ywant to stress thattw o $G{\math bi n{\ ltimes } } X$-$C^*$ -algebras are more than just ${ \ma thrm{ KK}} $[-]{}equ iva lent.Wew rite $ A\appr ox B$ i f $ A $ and$ B $are i somorphica s $G {\mat hb in{\ ltimes} }X$-$C^*$-algebras and $A\sim_M B$if$A$a n d$B$ ar e $G {\ m ath b i n{\ltimes}}X$-e quivariant ly Mo rita-Rieff e l e qu ivalent . Bothrelat i ons imp ly $A\con g B$.
Tr ia ngul a t edcategories of oper ator alge b ras { # se c:tri ang ulated _c ate gorie s}
=== = === ===== ====== == ====== ===== == ======== ===
In this section, w e | it_by ${\mathrm{KK}}^{G{\mathbin{\ltimes}}X}_*(A,B)$._For several purposes, it_is useful_to_generalise from_groups_to groupoids. However,_we do not_treat arbitrary groupoids because_it is not_so_clear what should correspond to the compact subgroups in this case. We work with_transformation_groups throughout_because_this_generalisation is not more difficult_than the group case and_useful for_several applications.
We write ${\mathrm{K}}_*(A)$ for the graded Abelian_group_$n\mapsto {\mathrm{K}}_n(A)$, $n\in{{\mathbb{Z}}}$,_and similarly for ${\mathrm{KK}}^{G{\mathbin{\ltimes}}X}_*(A,B)$. We usually omit the subscript $0$,_that is, ${\mathrm{K}}(A){\mathrel{:=}}{\mathrm{K}}_0(A)$, etc..
The *$G{\mathbin{\ltimes}}X$[-]{}equivariant Kasparov_category* is the_additive_category_whose objects are the_$G{\mathbin{\ltimes}}X$-$C^*$[-]{}algebras and whose group of morphisms_$A\to B$ is ${\mathrm{KK}}^{G{\mathbin{\ltimes}}X}_0(A,B)$; the composition_is the Kasparov product. We denote this_category by ${\mathrm{KK}}^{G{\mathbin{\ltimes}}X}$.
The notion of equivalence_for $G{\mathbin{\ltimes}}X$-$C^*$-algebras that we encounter_most frequently_is ${\mathrm{KK}}$-equivalence, that is, isomorphism_in ${\mathrm{KK}}^{G{\mathbin{\ltimes}}X}$, which_we simply_denote by “$\cong$”._Sometimes we may want to stress_that two $G{\mathbin{\ltimes}}X$-$C^*$-algebras_are more than just ${\mathrm{KK}}$[-]{}equivalent. We_write_$A\approx B$ if_$A$_and $B$_are isomorphic_as $G{\mathbin{\ltimes}}X$-$C^*$-algebras and_$A\sim_M_B$ if_$A$_and $B$ are $G{\mathbin{\ltimes}}X$-equivariantly Morita-Rieffel equivalent. Both_relations_imply $A\cong B$.
Triangulated categories of operator algebras_{#sec:triangulated_categories}
============================================
In this section, we |
12) to (s111); (s1112) to (s112); (s1112) to (s1111); (s1122) to (s112); (s1122) to (s122); (s1122) to (s1112); (s1222) to (s122); (s1222) to (s222); (s1222) to (s1122); (s2222) to (s222); (s2222) to (s1222); (s11111) to (s1111); (s11112) to (s1111); (s11112) to (s1112); (s11112) to (s11111); (s11122) to (s1112); (s11122) to (s1122); (s11122) to (s11112); (s11222) to (s1122); (s11222) to (s1222); (s11222) to (s11122); (s12222) to (s1222); (s12222) to (s2222); (s12222) to (s11222); (s22222) to (s2222); (s22222) to (s12222);
(0,-7) node [$0$]{}; (1,-7) node [$3^\Box$]{}; (2,-7) node [$1$]{}; (3,-7) node [$2^\Box$]{}; (4,-7) node [$0$]{}; (5,-7) node [$3$]{}; (6,-7) node […]{}; (7,-7) node [$0$]{};
(0,-8) node [$1$]{}; (1,-8) node [$2^\Box$]{}; (2,-8) node [$0$]{}; (3,-8) node [$3^\Box$]{}; (4,-8) node [$1$]{}; (5,-8) node [$2$]{}; (6,-8) node […]{}; (7,-8) node [$1$]{}; (8,-8) node [$2$]{};
(0.2,-7) – (0.8,-7); (1.2,-7) – (1.8,-7); (2.2,-7) – (2.8,-7); (3.2,-7) – (3.8,-7); (4.2,-7) – (4.8,-7);
| 12) to (s111); (s1112) to (s112); (s1112) to (s1111); (s1122) to (s112); (s1122) to (s122); (s1122) to (s1112); (s1222) to (s122); (s1222) to (s222); (s1222) to (s1122); (s2222) to (s222); (s2222) to (s1222); (s11111) to (s1111); (s11112) to (s1111); (s11112) to (s1112); (s11112) to (s11111); (s11122) to (s1112); (s11122) to (s1122); (s11122) to (s11112); (s11222) to (s1122); (s11222) to (s1222); (s11222) to (s11122); (s12222) to (s1222); (s12222) to (s2222); (s12222) to (s11222); (s22222) to (s2222); (s22222) to (s12222);
(0,-7) node [ $ 0 $ ] { }; (1,-7) node [ $ 3^\Box$ ] { }; (2,-7) node [ $ 1 $ ] { }; (3,-7) node [ $ 2^\Box$ ] { }; (4,-7) lymph node [ $ 0 $ ] { }; (5,-7) lymph node [ $ 3 $ ] { }; (6,-7) lymph node [ … ] { }; (7,-7) node [ $ 0 $ ] { };
(0,-8) node [ $ 1 $ ] { }; (1,-8) node [ $ 2^\Box$ ] { }; (2,-8) lymph node [ $ 0 $ ] { }; (3,-8) node [ $ 3^\Box$ ] { }; (4,-8) node [ $ 1 $ ] { }; (5,-8) node [ $ 2 $ ] { }; (6,-8) lymph node [ … ] { }; (7,-8) lymph node [ $ 1 $ ] { }; (8,-8) node [ $ 2 $ ] { };
(0.2,-7) – (0.8,-7); (1.2,-7) – (1.8,-7); (2.2,-7) – (2.8,-7); (3.2,-7) – (3.8,-7); (4.2,-7) – (4.8,-7); | 12) to (s111); (s1112) to (s112); (s1112) to (s1111); (s1122) to (s112); (r1122) to (s122); (s1122) to (s1112); (s1222) to (s122); (s1222) to (s222); (a1222) to (s1122); (s2222) to (s222); (s2222) to (s1222); (s11111) to (s1111); (s11112) to (s1111); (s11112) ti (s1112); (s11112) to (s11111); (s11122) to (s1112); (s11122) to (s1122); (s11122) go (s11112); (s11222) to (s1122); (s11222) to (w1222); (s11222) uo (s11122); (s12222) to (s1222); (s12222) to (s2222); (s12222) to (s11222); (s22222) to (a2222); (s22222) tm (s12222);
(0,-7) node [$0$]{}; (1,-7) node [$3^\Box$]{}; (2,-7) node [$1$]{}; (3,-7) node [$2^\Box$]{}; (4,-7) nmdd [$0$]{}; (5,-7) node [$3$]{}; (6,-7) node […]{}; (7,-7) node [$0$]{};
(0,-8) node [$1$]{}; (1,-8) node [$2^\Bjx$]{}; (2,-8) nodr [$0$]{}; (3,-8) node [$3^\Box$]{}; (4,-8) noqe [$1$]{}; (5,-8) gode [$2$]{}; (6,-8) node […]{}; (7,-8) node [$1$]{}; (8,-8) node [$2$]{};
(0.2,-7) – (0.8,-7); (1.2,-7) – (1.8,-7); (2.2,-7) – (2.8,-7); (3.2,-7) – (3.8,-7); (4.2,-7) – (4.8,-7);
| 12) to (s111); (s1112) to (s112); (s1112) (s1122) (s112); (s1122) (s122); (s1122) to to (s1222) to (s1122); to (s222); (s2222) (s1222); (s11111) to (s1111); (s11112) to (s11112) to (s1112); (s11112) to (s11111); (s11122) to (s1112); (s11122) to (s1122); (s11122) (s11112); (s11222) to (s1122); (s11222) to (s1222); (s11222) to (s11122); (s12222) to (s1222); to (s12222) (s11222); to (s2222); (s22222) to (s12222); (0,-7) node [$0$]{}; (1,-7) node [$3^\Box$]{}; (2,-7) node [$1$]{}; (3,-7) node (4,-7) node [$0$]{}; (5,-7) node [$3$]{}; (6,-7) node (7,-7) node [$0$]{}; (0,-8) [$1$]{}; (1,-8) node [$2^\Box$]{}; (2,-8) [$0$]{}; node [$3^\Box$]{}; node (5,-8) [$2$]{}; (6,-8) node (7,-8) node [$1$]{}; (8,-8) node [$2$]{}; (0.2,-7) – (0.8,-7); (1.2,-7) – (1.8,-7); (2.2,-7) – (2.8,-7); (3.2,-7) – (4.2,-7) – | 12) to (s111); (s1112) to (s112); (s1112) to (s1111); (s1122) to (s112); (s1122) to (s122); (s1122) to (s1112); (s1222) tO (s122); (s1222) to (s222); (s1222) to (s1122); (S2222) to (s222); (s2222) To (s1222); (S11111) to (S1111); (s11112) To (s1111); (s11112) To (s1112); (s11112) To (s11111); (s11122) to (s1112); (s11122) to (s1122); (s11122) to (S11112); (S11222) to (s1122); (S11222) to (s1222); (s11222) to (s11122); (s12222) to (s1222); (s12222) to (s2222); (s12222) to (s11222); (s22222) tO (s2222); (s22222) to (S12222);
(0,-7) nODe [$0$]{}; (1,-7) noDE [$3^\BOx$]{}; (2,-7) nodE [$1$]{}; (3,-7) node [$2^\BoX$]{}; (4,-7) NoDE [$0$]{}; (5,-7) NodE [$3$]{}; (6,-7) nOdE […]{}; (7,-7) noDe [$0$]{};
(0,-8) NOdE [$1$]{}; (1,-8) node [$2^\box$]{}; (2,-8) Node [$0$]{}; (3,-8) nodE [$3^\Box$]{}; (4,-8) node [$1$]{}; (5,-8) noDe [$2$]{}; (6,-8) nOdE […]{}; (7,-8) node [$1$]{}; (8,-8) node [$2$]{};
(0.2,-7) – (0.8,-7); (1.2,-7) – (1.8,-7); (2.2,-7) – (2.8,-7); (3.2,-7) – (3.8,-7); (4.2,-7) – (4.8,-7);
| 12) to (s111); (s1112) to(s112); (s 1112) to (s 11 11); (s1 122) to (s112) ; (s1 122) to (s122); (s1122 ) to(s 1 112) ; ( s1222 ) to (s 1 22 ) ; (s 12 22 ) t o( s2 22);(s1 222) to (s1122);(s2 22 2) to (s222) ; ( s2222) to(s1 222); (s1111 1)to (s1 11 1); (s111 12) to ( s1111) ; (s111 12) to (s 11 1 2); (s 1 1112) t o (s 1111 1); (s11122) to ( s 11 1 2); (s11122) t o (s11 22 ) ;( s 111 22) to (s1111 2) ; (s1 1 222) to (s 1 1 2 2); (s11222) to ( s1222); (s1 1 222 ) to ( s1 112 2 ); (s1 2222) t o (s 1222); (s12 222) to (s222 2); (s 1 2222) t o (s1122 2); (s 222 22) to( s2 22 2); ( s 222 2 2) to (s1 2222);
(0 ,- 7) no de [ $ 0 $ ] {};(1, -7)node[$3^\Box$]{}; (2 ,-7) nod e [$1 $]{}; (3, -7 ) nod e [$2^ \Box$ ]{ }; (4,-7) node[$0$ ]{}; (5,- 7)no de[$ 3$]{} ; (6,-7 ) n ode […]{}; (7,-7) nod e[ $ 0 $] {};
(0,-8) node [ $1 $ ] {} ; (1,-8) node[ $2 ^\ B ox$]{};(2 ,-8 ) no d e [$0$ ]{}; (3 ,-8) nod e [$3^ \ Bo x$ ]{}; (4 ,- 8) nod e[$1 $]{ }; (5 , -8)node [ $2$]{};(6,-8 ) node […]{}; ( 7 ,-8) node [$1 $ ]{ } ; ( 8 ,-8) no de [$2$]{};
(0 . 2,-7 ) –( 0. 8,- 7 ); (1 .2,-7 )– ( 1 .8,-7); (2.2,-7) –(2 .8,-7) ; (3. 2,-7) – (3.8, -7); (4.2, - 7 ) – (4.8, -7); | 12) to_(s111); (s1112)_to (s112); (s1112) to_(s1111); (s1122)_to_(s112); (s1122)_to_(s122); (s1122) to_(s1112); (s1222) to_(s122); (s1222) to (s222);_(s1222) to (s1122);_(s2222)_to (s222); (s2222) to (s1222); (s11111) to (s1111); (s11112) to (s1111); (s11112) to (s1112);_(s11112)_to (s11111);_(s11122)_to_(s1112); (s11122) to (s1122); (s11122)_to (s11112); (s11222) to (s1122);_(s11222) to_(s1222); (s11222) to (s11122); (s12222) to (s1222); (s12222)_to_(s2222); (s12222) to_(s11222); (s22222) to (s2222); (s22222) to (s12222);
(0,-7) node [$0$]{};_(1,-7) node [$3^\Box$]{}; (2,-7) node [$1$]{};_(3,-7) node [$2^\Box$]{};_(4,-7)_node_[$0$]{}; (5,-7) node [$3$]{};_(6,-7) node […]{}; (7,-7) node [$0$]{};
(0,-8)_node [$1$]{}; (1,-8) node [$2^\Box$]{}; (2,-8)_node [$0$]{}; (3,-8) node [$3^\Box$]{}; (4,-8) node_[$1$]{}; (5,-8) node [$2$]{}; (6,-8) node_[…]{}; (7,-8) node [$1$]{}; (8,-8)_node [$2$]{};
(0.2,-7)_– (0.8,-7); (1.2,-7) – (1.8,-7);_(2.2,-7) – (2.8,-7);_(3.2,-7) –_(3.8,-7); (4.2,-7) –_(4.8,-7);
|
values of linear convolutional layers are computed by applying SVD to ${{{\uppercase{{\bm{t}}}}}}$ directly.
- Circular Approximation: A block doubly circulant matrix ${{{\uppercase{{\bm{c}}}}}}$ is constructed according to -. The singular values are computed by applying SVD on ${{{\uppercase{{\bm{c}}}}}}$ directly.
- Uniform Sampling: The block diagonal matrices ${{{\uppercase{{\bm{b}}}}}}_{j_1,j_2}$ is produced by uniformly sampling the spectral density matrix $F(\omega_1,\omega_2)$ with sampling grids $(\omega_1,\omega_2)=(-\pi+\frac{2\pi j_1}{n},-\pi+\frac{2\pi j_1}{n})$ for all $j_1, j_2 \in [n]$. The singular values are obtained by collecting all singular values of $\{{{{\uppercase{{\bm{b}}}}}}_{j_1,j_2}\}_{j_1,j_2=1}^n$. This corresponds to lines 1-10 in Algorithm \[alg:quantile\].
- Quantile Interpolation: The singular values obtained from uniform sampling are arranged for each $1\le j \le \min\{c_{in},c_{out}\}$ in descending order. By quantile estimation using linear interpolation methods, the singular values are recomputed by selecting properly shifted sampling grids as outlined in Algorithm \[alg:quantile\].
The experiments are conducted on MATLAB 2020a, which is more friendly to matrix computation. For simplicity, we set $h_1=h_2$ and $w_1=w_2$, and the input size per channel is set to $10 \times 10$. Fig. \[fig:Fig-1\] presents the $(i-1)n+1$-th largest singular values ($i \in [n]$) of four methods with four different filter sizes. The first two filters are from the pre-trained GoogLeNet, and the last two are from the training process of ResNet-20. It can be observed that (1) both circular approximation and uniform sampling have identical singular values for different filter sizes, (2) quantile interpolation improves accuracy of the singular values over the circular approximation with negligible extra running time (see Section \[sec:experiments-sva\]), and (3) during the training process the improvement of the largest singular | values of linear convolutional layers are computed by apply SVD to $ { { { \uppercase{{\bm{t}}}}}}$ immediately.
- Circular Approximation: A block doubly circulant matrix $ { { { \uppercase{{\bm{c}}}}}}$ is construct accord to -. The singular values are calculate by applying SVD on $ { { { \uppercase{{\bm{c}}}}}}$ directly.
- Uniform Sampling: The block aslant matrix $ { { { \uppercase{{\bm{b}}}}}}_{j_1,j_2}$ is produced by uniformly sample the spectral density matrix $ F(\omega_1,\omega_2)$ with sampling grids $ (\omega_1,\omega_2)=(-\pi+\frac{2\pi j_1}{n},-\pi+\frac{2\pi j_1}{n})$ for all $ j_1, j_2 \in [ n]$. The remarkable values are obtained by collecting all singular values of $ \{{{{\uppercase{{\bm{b}}}}}}_{j_1,j_2}\}_{j_1,j_2=1}^n$. This match to lines 1 - 10 in Algorithm \[alg: quantile\ ].
- Quantile Interpolation: The singular values prevail from uniform sampling are arranged for each $ 1\le j \le \min\{c_{in},c_{out}\}$ in descending ordering. By quantile estimation using linear interpolation methods, the remarkable values are recomputed by selecting properly shifted sampling grids as outlined in Algorithm \[alg: quantile\ ].
The experiments are conducted on MATLAB 2020a, which is more friendly to matrix computation. For simplicity, we set $ h_1 = h_2 $ and $ w_1 = w_2 $, and the input size per channel is set to $ 10 \times 10$. Fig. \[fig: Fig-1\ ] presents the $ (i-1)n+1$-th largest remarkable values ($ i \in [ n]$) of four methods with four unlike filter size. The beginning two filters are from the pre - trained GoogLeNet, and the final two are from the training process of ResNet-20. It can be observed that (1) both circular estimate and uniform sampling have identical singular values for different filter sizes, (2) quantile interpolation improves accuracy of the singular values over the circular approximation with negligible excess running prison term (determine Section \[sec: experiments - sva\ ]), and (3) during the training process the improvement of the big singular | vapues of linear convolutiunal layers are computxd by alplying RVD to ${{{\uppercase{{\bm{t}}}}}}$ directly.
- Circylar Approximation: A buock doubpy circuoant natrix ${{{\uppxdcase{{\bm{g}}}}}}$ is dlnstxurted according jo -. The singglar values ara zolputed by applying SVD on ${{{\uppercase{{\fm{c}}}}}}$ dirrchly.
- Uniform Fampkyng: Fhe block diagonal matrices ${{{\upperczse{{\bm{b}}}}}}_{j_1,o_2}$ is produced bu uniformly sampling the soectgal density matrix $F(\omega_1,\omegq_2)$ wiey sampling gfids $(\omega_1,\omega_2)=(-\pi+\frac{2\pj j_1}{n},-\pi+\frac{2\pi j_1}{n})$ for all $j_1, j_2 \in [v]$. The singular caoued are obtainxd by bollecting all singuldr valurs of $\{{{{\uppercasc{{\bm{b}}}}}}_{j_1,o_2}\}_{j_1,j_2=1}^n$. This corresponds to nines 1-10 in Algorithi \[alg:quandime\].
- Quantile Intwrpolathon: Dhe rungjlad talhes obhaiied from unjform samplung are arranged fot qqch $1\le j \le \mjn\{c_{in},c_{jue}\}$ in descending order. By quantile estimdtikn using linear interpooation methods, the sijgular vajues are recomputed by selecting properly shifted sampmkng gvids qs outlined in Algorithm \[alg:quantile\].
The experimqhtx sre conducted jn MATLAB 2020a, wjivr is more frigndly to mztrix computation. Vor simklicitt, we set $r_1=h_2$ amd $w_1=w_2$, and the input size pee channel is wet to $10 \times 10$. Fig. \[yig:Fig-1\] preseuts thg $(i-1)n+1$-th largest singular valuer ($i \jn [n]$) of foug methods dith four differdnt fhlter sieds. The first two silters ace frpm the ore-ttained DoogLeNet, wnd tma last two are frol the tsaining prlcess of ResNet-20. It can be obsertxd that (1) both chrcllar apprjximabion and unifori sampling havg identiccl sineular valuvs for dihferent filtqr sizes, (2) quatjile interpoletion impwovew acxuracy uw the singular values onex the cirxular approximatiok witf negligible extxc tynning time (ser Sdctyoj \[xes:axperiments-sea\]), avd (3) curine the trainlng provess the improvement of fhe largest singulsr | values of linear convolutional layers are computed SVD ${{{\uppercase{{\bm{t}}}}}}$ directly. Circular Approximation: A is according to -. singular values are by applying SVD on ${{{\uppercase{{\bm{c}}}}}}$ directly. Uniform Sampling: The block diagonal matrices ${{{\uppercase{{\bm{b}}}}}}_{j_1,j_2}$ is produced by uniformly sampling the density matrix $F(\omega_1,\omega_2)$ with sampling grids $(\omega_1,\omega_2)=(-\pi+\frac{2\pi j_1}{n},-\pi+\frac{2\pi j_1}{n})$ for all $j_1, j_2 [n]$. singular are by collecting all singular values of $\{{{{\uppercase{{\bm{b}}}}}}_{j_1,j_2}\}_{j_1,j_2=1}^n$. This corresponds to lines 1-10 in Algorithm \[alg:quantile\]. - Interpolation: The singular values obtained from uniform sampling arranged for each $1\le \le \min\{c_{in},c_{out}\}$ in descending order. quantile using linear methods, singular are recomputed by properly shifted sampling grids as outlined in Algorithm \[alg:quantile\]. The experiments are conducted on MATLAB 2020a, which more friendly computation. For we $h_1=h_2$ $w_1=w_2$, and the per channel is set to $10 \[fig:Fig-1\] presents the $(i-1)n+1$-th largest singular values ($i [n]$) of methods with four different filter sizes. first two filters are from the pre-trained GoogLeNet, the last two are from the training process of ResNet-20. It can be observed that circular approximation and uniform have identical singular for filter (2) interpolation improves of the singular values over the circular approximation with negligible extra time (see Section \[sec:experiments-sva\]), and (3) during the training process of largest singular | values of linear convolutionAl layers arE compUteD by ApPlyiNg SVd to ${{{\uppercase{{\bm{T}}}}}}$ DireCtly.
- Circular ApproximatIon: A bLoCK douBLy CircuLant matRIx ${{{\UPPerCaSe{{\Bm{c}}}}}}$ Is COnStrucTed AccordiNg to -. The sinGulAr Values are comPUtEd by applyiNg SvD on ${{{\uppercasE{{\bm{C}}}}}}$ direcTlY.
- UnIForm SAmpLing: THe blocK DiagonAl matriceS ${{{\uPPercasE{{\Bm{b}}}}}}_{j_1,j_2}$ is PROdUced By uniformly sampliNG tHE spectral densiTy matrIx $f(\OmEGA_1,\omEga_2)$ With sampliNg Grids $(\OMega_1,\omeGA_2)=(-\pI+\FRAc{2\pI J_1}{n},-\pi+\frac{2\pi j_1}{n})$ fOr all $j_1, j_2 \in [n]$. THE siNgular VaLueS Are obtAined By COllEcting all siNgulAr values oF $\{{{{\upperCAse{{\bm{b}}}}}}_{j_1,J_2}\}_{J_1,j_2=1}^n$. This CorresPonDs tO linES 1-10 iN ALgoRiTHm \[aLG:qUanTIle\].
- quantile inTeRpolaTion: tHE SInguLar ValuEs obtAined from unifOrm SampLIng Are arRangeD for EaCh $1\le j \Le \min\{c_{In},c_{ouT}\}$ iN descending ordeR. By qUantile esTimAtIon UsIng liNEar intErpOlaTion metHods, the SIngUlAR VAlUes are recomputed by SeLECtIng propeRly shiFTeD sAMpling grIdS as OutlINEd in ALgorIThM \[alg:quanTile\].
ThE ExPeRiments ArE conduCtEd oN MAtLAB 2020a, WHich Is more Friendly To matRIx computation. FOR simplicity, we SEt $H_1=H_2$ AnD $W_1=w_2$, anD thE input size pEr chANnel Is seT To $10 \TimES 10$. Fig. \[fIg:Fig-1\] PrESeNTs the $(i-1)n+1$-th largest sinGuLar valUes ($i \iN [n]$) of four methoDs with four DIFFerent fiLter SIzES. The first two fiLters Are from the PRe-traineD GoogleNet, and The last twO ARe from thE trAinIng ProCESs Of ResNet-20. It can BE ObseRvEd that (1) bOth CirculaR apProXimAtiOn And uniforM samplinG hAvE iDeNtiCal siNGular valUeS foR dIffErent FIlter sIzes, (2) qUantIlE iNTerPolatioN ImPROves AcCuRacy Of tHe SinguLar vALueS over thE circular AppROximAtIoN with neGligible extra RuNning time (sEe secTion \[seC:EXperimenTs-sva\]), and (3) during the trainiNG procesS thE imprOvemEnt of the lArgEst sinGulAR | values of linear convolut ional laye rs ar e c omp ut ed b y ap plying SVD to$ {{{\ uppercase{{\bm{t}}}}}} $ dir ec t ly.- Cir cular A p pr o x ima ti on : A b l oc k dou bly circul ant matrix ${ {{ \uppercase{{ \ bm {c}}}}}}$isconstructedacc ording t o - . Thesin gular value s are c omputed b ya pplyin g SVD on $ {{ {\up percase{{\bm{c}}} } }} $ directly.
- Unif or m S a m pli ng: The block d iagon a l matri c es $ { {{\ u ppercase{{\bm {b}}}}}}_{j _ 1,j _2}$ i spro d uced b y uni fo r mly sampling t he s pectral d ensity matrix$ F(\omeg a_1,\o meg a_2 )$ w i th s amp li n g g r id s $ ( \om ega_1,\o me ga _2)=( -\pi + \ f r ac{2 \pi j_1 }{n}, -\pi+\frac{2\ pij_1} { n}) $ for all$j_1 ,j_2 \ in [n] $. Th esingular values are obtained by c oll ec tinga ll sin gul arvaluesof $\{{ { {\u pp e r c as e{{\bm{b}}}}}}_{j_ 1, j _ 2} \}_{j_1, j_2=1} ^ n$ .T his corr es pon ds t o lines 1-1 0 i n Algori thm \[ a lg :q uantile \] .
- Q uan til e Int e rpol ation: The sin gular values obtaine d from uniform sa m p li n g ar e a rranged for eac h $1\ le j \l e \ m in\{c _{in} ,c _ {o u t}\}$ in descending o rder.By qu antile estima tion using l i near int erpo l at i on methods, th e sin gular valu e s are re compu ted by s electingp r operly s hif ted sa mpl i n ggrids as outl i n ed i nAlgorit hm\[alg:q uan til e\] .
Th e experim ents are c on du ct edon MA T LAB 2020 a, wh ic h i s mor e frien dly t o ma tr ix com putatio n .F o r si mp li city , w eset $ h_1= h _2$ and $w _1=w_2$,and thein pu t sizeper channel i sset to $10 \ tim es 10$ . Fig. \[f ig:Fig-1\] presents the $(i-1)n +1$ -th l arge st singul arvalues ($ i \in [ n]$) o f fou rmet h o ds wi t h f our d ifferent f i l ter size s. The firsttwo filters are fr o m t he pre-traine d G oogL e N et , a n dt hela s t t w o are from the t raining pr oc e ss of ResNet - 20. I t can b e obser ved t h at (1)both circ ular appr ox imat i o n a nd uniform samplin g have id e ntica l s ingul arvalues f ordiffe rent f i lte r siz es, (2 )quanti le in te rpolatio n improves accuracy ofthe si ngula r v alues ove r t h e c ircular a ppro ximation w ith ne gligi ble extra run n in g t i me (s ee S e ction \[s e c: exp e r im ents-sva\]) , a nd(3) d uri n g thetrai ning process thei mprovement ofthel a rge sts ingu la r | values_of linear_convolutional layers are computed_by applying_SVD_to ${{{\uppercase{{\bm{t}}}}}}$_directly.
-_ Circular_Approximation: A block_doubly circulant matrix ${{{\uppercase{{\bm{c}}}}}}$_is constructed according_to_-. The singular values are computed by applying SVD on ${{{\uppercase{{\bm{c}}}}}}$ directly.
- _Uniform_Sampling: The_block_diagonal_matrices ${{{\uppercase{{\bm{b}}}}}}_{j_1,j_2}$ is produced by_uniformly sampling the spectral density_matrix $F(\omega_1,\omega_2)$_with sampling grids $(\omega_1,\omega_2)=(-\pi+\frac{2\pi j_1}{n},-\pi+\frac{2\pi j_1}{n})$ for all_$j_1,_j_2 \in [n]$._The singular values are obtained by collecting all singular_values of $\{{{{\uppercase{{\bm{b}}}}}}_{j_1,j_2}\}_{j_1,j_2=1}^n$. This corresponds to_lines 1-10 in_Algorithm_\[alg:quantile\].
-_ Quantile Interpolation:_The singular values obtained from uniform_sampling are arranged for each $1\le_j \le \min\{c_{in},c_{out}\}$ in descending order. By_quantile estimation using linear interpolation methods,_the singular values are recomputed_by selecting_properly shifted sampling grids as_outlined in Algorithm_\[alg:quantile\].
The experiments_are conducted on_MATLAB 2020a, which is more friendly_to matrix computation._For simplicity, we set $h_1=h_2$ and_$w_1=w_2$,_and the input_size_per_channel is_set to $10_\times_10$. Fig. \[fig:Fig-1\]_presents_the $(i-1)n+1$-th largest singular values ($i_\in_[n]$) of four methods with four different_filter sizes. The first_two_filters are from the_pre-trained GoogLeNet, and the last_two are from the training process_of ResNet-20._It can_be observed that (1) both circular approximation and uniform sampling have_identical singular values for different filter_sizes, (2) quantile interpolation_improves accuracy_of_the singular values_over_the circular_approximation with negligible extra running time (see_Section \[sec:experiments-sva\]),_and (3) during the training process_the improvement of the_largest_singular |
H^2(X,\Bbb Q)\times H^4(X,\Bbb Q)$ and $\overline{M}_H(r;c_1,c_2)$ the Gieseker-Maruyama compactification of $M_H(r;c_1,c_2)$. When $r=2$, these spaces are extensively studied by many authors. When $r \geq 3$, Drezet and Le-Potier \[D1\],\[D-L\] investigated the structure of moduli spaces on $\Bbb P^2$, and Rudakov \[R\] treated moduli spaces on $\Bbb P^1
\times \Bbb P^1$. In this paper, we shall consider moduli spaces of rank $r \geq 3$ on a ruled surface which is not rational. In particular, we shall compute the Picard group of $\overline{M}_H(r;c_1,c_2)$. Let $\pi:X \to C$ be the fibration, $f$ a fibre of $\pi$ and $C_0$ a minimal section of $\pi$ with $(C_0^2)=-e$. We assume that $e>2g-2$, where $g$ is the genus of $C$. Then $K_X$ is effective, and hence $(K_X,H)<0$ for any ample divisor $H$. In particular, $M_H(r;c_1,c_2)$ is smooth with the expected dimension $2r^2
\Delta-r^2(1-g)+1$.
In section 2, we shall generalize the chamber structure of Qin \[Q2\]. As an application, we shall consider the difference of Betti numbers of moduli spaces on a ruled surface. Although we cannot generalize the method in \[Y2, 0\] directly, by using Qin’s method we can generalize it to any rank case. In \[Y2\], we computed the number of $\mu$-semi-stable sheaves of rank 2 on a ruled surface defind over $\Bbb F_q$. So, in principle, we can compute the Betti numbers of $M_H(3;c_1,c_2)$ on $\Bbb
P^2$. Combining chamber structure with another method, Göttsche \[Gö\] also considered the difference of Hodge numbers (and hence Betti numbers) of moduli spaces of rank 2. Matsuki and Went | H^2(X,\Bbb Q)\times H^4(X,\Bbb Q)$ and $ \overline{M}_H(r;c_1,c_2)$ the Gieseker - Maruyama compactification of $ M_H(r;c_1,c_2)$. When $ r=2 $, these spaces are extensively studied by many authors. When $ r \geq 3 $, Drezet and Le - Potier \[D1\],\[D - L\ ] investigate the social organization of moduli spaces on $ \Bbb P^2 $, and Rudakov \[R\ ] treat modulus spaces on $ \Bbb P^1
\times \Bbb P^1$. In this paper, we shall regard modulus spaces of rank $ r \geq 3 $ on a ruled airfoil which is not rational. In particular, we shall calculate the Picard group of $ \overline{M}_H(r;c_1,c_2)$. Let $ \pi :X \to C$ be the fibration, $ f$ a fibre of $ \pi$ and $ C_0 $ a minimal section of $ \pi$ with $ (C_0 ^ 2)=-e$. We assume that $ e>2g-2 $, where $ g$ is the genus of $ C$. Then $ K_X$ is effective, and hence $ (K_X, H)<0 $ for any ample divisor $ H$. In particular, $ M_H(r;c_1,c_2)$ is smooth with the expect dimension $ 2r^2
\Delta - r^2(1 - g)+1$.
In section 2, we shall generalize the bedroom structure of Qin \[Q2\ ]. As an application, we shall view the difference of Betti numbers pool of moduli spaces on a ruled surface. Although we cannot generalize the method acting in \[Y2, 0\ ] directly, by using Qin ’s method we can generalize it to any rank case. In \[Y2\ ], we calculate the act of $ \mu$-semi - stable sheaves of rank 2 on a ruled surface defind over $ \Bbb F_q$. So, in principle, we can compute the Betti numbers of $ M_H(3;c_1,c_2)$ on $ \Bbb
P^2$. unite chamber structure with another method acting, Göttsche \[Gö\ ] besides considered the difference of Hodge numbers (and hence Betti issue) of moduli spaces of rank 2. Matsuki and Went | H^2(X,\Hbb Q)\times H^4(X,\Bbb Q)$ and $\oyerline{M}_H(r;c_1,c_2)$ the Giesekxr-Maruyzma compxctification of $M_H(r;c_1,c_2)$. When $r=2$, tyese wpaces are extensively studied hy many quthies. When $r \jsq 3$, Dredzt ans Le-Pmvier \[D1\],\[D-L\] investlgated the vtructure of mmdjln spaces on $\Bbb P^2$, and Rudakov \[R\] treaeed modill spaces on $\Bbf P^1
\tpmqs \Bgb P^1$. In this paper, we shall considsr moduni spaces of tank $r \geq 3$ on a ruled survace which is not ratilnal. In parjjcujqr, we shall zompute tht 'icard grouk of $\overline{M}_H(r;c_1,c_2)$. Let $\pi:X \to C$ ce thz fibration, $f$ a vhbre of $\pi$ end $C_0$ w minimal segnion of $\pi$ witn $(C_0^2)=-e$. We assume thet $e>2t-2$, where $g$ is the genuv of $C$. Then $K_X$ is gffective, dnb hence $(K_X,H)<0$ for any anpoe dieisos $H$. Kb pxrtjcnlad, $M_H(r;c_1,f_2)$ ia smooth wjth the expwcted dimension $2r^2
\Dektw-g^2(1-b)+1$.
In section 2, we shwlj generalize the chamber structure of Qpn \[Q2\]. As an application, we syall consider the difverence os Betti numbers of moduli spaces on a ruled surfawe. Alvhuugk we cxbnlt generalize the method in \[Y2, 0\] directly, by usyhg Qpn’s method we can generalize ih yj any rank care. In \[V2\], ws computed the numher of $\iu$-semu-stable sreavrs of rank 2 on a ruled surfqce defind oner $\Vbb F_q$. So, in princnple, we can eomputg the Netti numbers of $M_H(3;c_1,c_2)$ ou $\Bbb
P^2$. Combining fhamber sffucture with anogheg medhod, Göttsche \[Gö\] also considqred the viffexence of Hodbe numfers (and hfnce Natti numbers) of mofuli dpdces of rajk 2. Matsuki and Went | H^2(X,\Bbb Q)\times H^4(X,\Bbb Q)$ and $\overline{M}_H(r;c_1,c_2)$ the of When $r=2$, spaces are extensively $r 3$, Drezet and \[D1\],\[D-L\] investigated the of moduli spaces on $\Bbb P^2$, Rudakov \[R\] treated moduli spaces on $\Bbb P^1 \times \Bbb P^1$. In this we shall consider moduli spaces of rank $r \geq 3$ on a ruled which not In we shall compute the Picard group of $\overline{M}_H(r;c_1,c_2)$. Let $\pi:X \to C$ be the fibration, $f$ fibre of $\pi$ and $C_0$ a minimal section $\pi$ with $(C_0^2)=-e$. We that $e>2g-2$, where $g$ is genus $C$. Then is and $(K_X,H)<0$ for any divisor $H$. In particular, $M_H(r;c_1,c_2)$ is smooth with the expected dimension $2r^2 \Delta-r^2(1-g)+1$. In section 2, we generalize the of Qin As application, shall consider the Betti numbers of moduli spaces on Although we cannot generalize the method in \[Y2, directly, by Qin’s method we can generalize it any rank case. In \[Y2\], we computed the of $\mu$-semi-stable sheaves of rank 2 on a ruled surface defind over $\Bbb F_q$. So, we can compute the numbers of $M_H(3;c_1,c_2)$ $\Bbb Combining structure another method, \[Gö\] also considered the difference of Hodge numbers (and hence Betti of moduli spaces of rank 2. Matsuki and Went | H^2(X,\Bbb Q)\times H^4(X,\Bbb Q)$ and $\overlIne{M}_H(r;c_1,c_2)$ thE GiesEkeR-MaRuYama CompActification of $m_h(r;c_1,c_2)$. when $r=2$, these spaces are extEnsivElY StudIEd By manY authorS. whEN $R \geQ 3$, DReZet AnD le-potieR \[D1\],\[D-l\] investIgated the sTruCtUre of moduli sPAcEs on $\Bbb P^2$, anD RuDakov \[R\] treateD moDuli spAcEs oN $\bbb P^1
\tImeS \Bbb P^1$. in this PAper, we Shall consIdER modulI Spaces oF RAnK $r \geQ 3$ on a ruled surface wHIcH Is not rational. IN partiCuLAr, WE ShaLl cOmpute the PIcArd grOUp of $\oveRLiNE{m}_h(r;c_1,C_2)$. let $\pi:X \to C$ be thE fibration, $f$ A FibRe of $\pi$ AnD $C_0$ a MInimal SectiOn OF $\pi$ With $(C_0^2)=-e$. We assUme tHat $e>2g-2$, wherE $g$ is thE Genus of $c$. then $K_X$ iS effecTivE, anD henCE $(K_x,H)<0$ For AnY AmpLE dIviSOr $H$. in particUlAr, $m_H(r;c_1,c_2)$ Is smOOTH With The ExpeCted dImension $2r^2
\DeltA-r^2(1-g)+1$.
in seCTioN 2, we shAll geNeraLiZe the ChambeR struCtUre of Qin \[Q2\]. As an apPlicAtion, we shAll CoNsiDeR the dIFferenCe oF BeTti numbErs of moDUli SpACES oN a ruled surface. AlthOuGH We Cannot geNeraliZE tHe MEthod in \[Y2, 0\] DiRecTly, bY USing QIn’s mEThOd we can gEneralIZe It To any raNk Case. In \[y2\], wE coMpuTed thE NumbEr of $\mu$-Semi-stabLe sheAVes of rank 2 on a ruLEd surface defiND oVER $\BBB F_q$. SO, in Principle, we Can cOMputE the bEtTi nUMbers Of $M_H(3;c_1,C_2)$ oN $\bbB
p^2$. Combining chamber stRuCture wIth anOther method, GöTtsche \[Gö\] alSO COnsidereD the DIfFErence of Hodge nUmberS (and hence BETti numbeRs) of mOduli spaCes of rank 2. mATsuki and wenT | H^2(X,\Bbb Q)\times H^4(X ,\Bbb Q)$and $ \ov erl in e{M} _H(r ;c_1,c_2)$ the Gies eker-Maruyama compacti ficat io n of$ M_ H(r;c _1,c_2) $ .W h en$r =2 $,th e se spac esare ext ensively s tud ie d by many au t ho rs. When $ r \ geq 3$, Drez etand Le -P oti e r \[D 1\] ,\[D- L\] in v estiga ted the s tr u ctureo f modul i sp aces on $\Bbb P^2$, a n dR udakov \[R\] t reated m o du l i sp ace s on $\Bbb P ^1
\t i mes \Bb b P ^ 1 $ . I n this paper,we shall co n sid er mod ul i s p aces o f ran k$ r \ geq 3$ on a rul ed surfac e whic h is not rationa l. Inpar tic ular , w esha ll com p ut e t h e P icard gr ou pof $\ over l i n e {M}_ H(r ;c_1 ,c_2) $. Let $\pi:X \t o C$ bethe f ibrat ion, $ f$ afibreof $\ pi $ and $C_0$ a m inim al sectio n o f$\p i$ with $(C_0^ 2)= -e$ . We as sume th a t $ e> 2 g - 2$ , where $g$ is the g e n us of $C$. Then$ K_ X$ is effec ti ve, and h ence$(K_ X ,H )<0$ for any a m pl edivisor $ H$. In p art icu lar,$ M_H( r;c_1, c_2)$ is smoo t h with the exp e cted dimensio n $ 2 r ^2 \Del ta- r^2(1-g)+1$ .
I n sec tion 2, we shall gene ra l iz e the chamber struct ur e of Q in \[ Q2\]. As an a pplication , w e shallcons i de r the differenc e ofBetti numb e rs of mo dulispaces o n a ruled s urface.Alt hou ghwec a nn ot generalize t he m et hod in\[Y 2, 0\]dir ect ly, by u sing Qin’ s method w eca ngen erali z e it toan y r an k c ase.I n \[Y2 \], w e co mp ut e d t he numb e ro f $\m u$ -s emi- sta bl e she aves ofrank 2on a rule d s u rfac ede find ov er $\Bbb F_q$ .So, in pri nc ipl e, wec a n comput e the Betti numbers of$ M_H(3;c _1, c_2)$ on$\Bbb
P^2 $.Combin ing chambe r stru cture w ith a nothe r me tho d, Göttsche\ [ Gö\ ] als ocons ideredthe difference ofH odg e numbers (an d h ence B et tin um b ers )o f m o d uli spaces of r ank 2. Mat su k iand Went | H^2(X,\Bbb_Q)\times H^4(X,\Bbb_Q)$ and $\overline{M}_H(r;c_1,c_2)$ the_Gieseker-Maruyama compactification_of_$M_H(r;c_1,c_2)$. When_$r=2$,_these spaces are_extensively studied by_many authors. When $r_\geq 3$, Drezet_and_Le-Potier \[D1\],\[D-L\] investigated the structure of moduli spaces on $\Bbb P^2$, and Rudakov \[R\]_treated_moduli spaces_on_$\Bbb_P^1
\times \Bbb P^1$. In this_paper, we shall consider moduli_spaces of_rank $r \geq 3$ on a ruled surface_which_is not rational._In particular, we shall compute the Picard group of_$\overline{M}_H(r;c_1,c_2)$. Let $\pi:X \to C$ be_the fibration, $f$_a_fibre_of $\pi$ and $C_0$_a minimal section of $\pi$ with_$(C_0^2)=-e$. We assume that $e>2g-2$, where_$g$ is the genus of $C$. Then_$K_X$ is effective, and hence $(K_X,H)<0$_for any ample divisor $H$._In particular,_$M_H(r;c_1,c_2)$ is smooth with the_expected dimension $2r^2
\Delta-r^2(1-g)+1$.
In_section 2,_we shall generalize_the chamber structure of Qin \[Q2\]._As an application,_we shall consider the difference of_Betti_numbers of moduli_spaces_on_a ruled_surface. Although we_cannot_generalize the_method_in \[Y2, 0\] directly, by using_Qin’s_method we can generalize it to any_rank case. In \[Y2\],_we_computed the number of_$\mu$-semi-stable sheaves of rank 2_on a ruled surface defind over_$\Bbb F_q$._So, in_principle, we can compute the Betti numbers of $M_H(3;c_1,c_2)$ on $\Bbb
P^2$._Combining chamber structure with another method,_Göttsche \[Gö\] also considered_the difference_of_Hodge numbers (and_hence_Betti numbers)_of moduli spaces of rank 2. Matsuki_and Went |
}_{w_{2,1}}$$ where $$\text{integrand}_{w_{2,1}} = WRHS_{2}(s,\sqrt{q^{2}(s-t)^{2}+r^{2}+2 r q(s-t) \cos(\theta)})\left(1-\frac{2q^{2}(s-t)^{2} \sin^{2}(\theta)}{r^{2}+q^{2}(s-t)^{2} + 2 r q (s-t) \cos(\theta)}\right)$$ Then, we can differentiate under the integral sign. The resulting integrals can be estimated with the same procedure used to estimate analogous integrals arising in the expressions for $\partial_{r}w_{2}$ and $w_{2}$. We get $$|\partial_{t}w_{2}(t,r)| \leq \frac{C \lambda(t)^{2} \log(t)}{t^{2} \log^{b}(t) g(t)}, \quad r >0$$ The same procedure is used to estimate $\partial_{tr}w_{2}$, and this concludes the proof of the lemma
Summation of the higher corrections, $w_{k}$
--------------------------------------------
The nonlinear interactions between $w_{2}$ and $v_{c}$ can not be treated perturbatively in our final argument. Therefore, we will need to define corrections $w_{k}$, in a similar manner as the corrections $v_{k}$ were defined, and sum a series of the form $\sum_{k=3}^{\infty} w_{k}$. Because the estimates for $w_{j}$ and $w_{2}$ will be of a slightly different form, we will first (define and) estimate $w_{3}$, then prove estimates on $w_{j}$ for $j \geq 4$ by induction. We let $$WRHS_{3}(t,r) = \frac{6\left(Q_{\frac{1}{\lambda(t)}}+v_{c}\right)}{r^{2}} w_{2}^{2}+\frac{2 w_{2}^{3}}{r^{2}} + \frac{6 w_{2}}{r^{2}}\left(v_{c}^{2}+2v_{c}Q_{\frac{1}{\lambda(t)}}\right)$$ Then, $$|WRHS_{3}(t,r)| \leq C \begin{cases} \frac{r^{2} \lambda( | } _ { w_{2,1}}$$ where $ $ \text{integrand}_{w_{2,1 } } = WRHS_{2}(s,\sqrt{q^{2}(s - t)^{2}+r^{2}+2 r q(s - t) \cos(\theta)})\left(1-\frac{2q^{2}(s - t)^{2 } \sin^{2}(\theta)}{r^{2}+q^{2}(s - t)^{2 } + 2 r q (s - t) \cos(\theta)}\right)$$ Then, we can differentiate under the integral sign of the zodiac. The result integrals can be estimated with the same routine used to estimate analogous integral arising in the expressions for $ \partial_{r}w_{2}$ and $ w_{2}$. We get $ $ |\partial_{t}w_{2}(t, r)| \leq \frac{C \lambda(t)^{2 } \log(t)}{t^{2 } \log^{b}(t) g(t) }, \quad roentgen > 0$$ The same procedure is use to estimate $ \partial_{tr}w_{2}$, and this concludes the validation of the lemma
Summation of the higher corrections, $ w_{k}$
--------------------------------------------
The nonlinear interactions between $ w_{2}$ and $ v_{c}$ cannot be treated perturbatively in our final argument. Therefore, we will necessitate to define corrections $ w_{k}$, in a similar manner as the correction $ v_{k}$ were defined, and sum a series of the form $ \sum_{k=3}^{\infty } w_{k}$. Because the estimate for $ w_{j}$ and $ w_{2}$ will be of a slightly different form, we will first (define and) calculate $ w_{3}$, then prove estimates on $ w_{j}$ for $ j \geq 4 $ by induction. We let $ $ WRHS_{3}(t, r) = \frac{6\left(Q_{\frac{1}{\lambda(t)}}+v_{c}\right)}{r^{2 } } w_{2}^{2}+\frac{2 w_{2}^{3}}{r^{2 } } + \frac{6 w_{2}}{r^{2}}\left(v_{c}^{2}+2v_{c}Q_{\frac{1}{\lambda(t)}}\right)$$ Then, $ $ |WRHS_{3}(t, r)| \leq C \begin{cases } \frac{r^{2 } \lambda ( | }_{w_{2,1}}$$ wjere $$\text{integrand}_{w_{2,1}} = WRHR_{2}(s,\sqrt{q^{2}(s-t)^{2}+r^{2}+2 r q(s-j) \xos(\theva)})\left(1-\fdac{2q^{2}(s-t)^{2} \skn^{2}(\theta)}{r^{2}+q^{2}(s-t)^{2} + 2 r q (s-t) \cos(\thete)}\rigyt)$$ Thtu, we can differentiage under nhe integeal wugn. The resulting lutegrzps ccn be estimated eith the sdme procedure gsdd to estimate analogous integrals arysing im hhe expressionf fog $\[artjal_{r}w_{2}$ and $w_{2}$. We get $$|\partial_{t}w_{2}(t,r)| \leq \frac{C \nambda(t)^{2} \log(t)}{t^{2} \log^{b}(t) g(t)}, \quad r >0$$ The same profedure is used to fstimate $\patfiaj_{rr}w_{2}$, and this concludes the proof or the lemma
Summation of the higfer cprrections, $w_{j}$
--------------------------------------------
Thf nonlinear mnterabtions betweek $w_{2}$ and $v_{c}$ can not be treatec pxrtuebatively in our finan argument. Therefote, we will nzed to define correctuobs $w_{k}$, in d sioulaf mznier as thf ckrrections $v_{k}$ were dedined, and sum a seroef of the form $\aum_{k=3}^{\instr} w_{k}$. Because the estimates for $w_{j}$ and $w_{2}$ wiml be of a slightly difderent form, we will flrst (defige and) estimate $w_{3}$, then prove estimates on $w_{j}$ for $b \geq 4$ by ikducguoj. We let $$WRHS_{3}(t,r) = \frac{6\left(Q_{\frac{1}{\lambda(t)}}+v_{c}\right)}{r^{2}} s_{2}^{2}+\ftab{2 w_{2}^{3}}{r^{2}} + \frac{6 w_{2}}{r^{2}}\left(y_{c}^{2}+2v_{c}Q_{\frac{1}{\lambda(t)}}\ribhh)$$ Yren, $$|WRHS_{3}(t,r)| \led C \beyjn{dases} \frac{r^{2} \lambda( | }_{w_{2,1}}$$ where $$\text{integrand}_{w_{2,1}} = WRHS_{2}(s,\sqrt{q^{2}(s-t)^{2}+r^{2}+2 r q(s-t) + r q \cos(\theta)}\right)$$ Then, we sign. resulting integrals can estimated with the procedure used to estimate analogous integrals in the expressions for $\partial_{r}w_{2}$ and $w_{2}$. We get $$|\partial_{t}w_{2}(t,r)| \leq \frac{C \lambda(t)^{2} \log^{b}(t) g(t)}, \quad r >0$$ The same procedure is used to estimate $\partial_{tr}w_{2}$, this the of lemma Summation of the higher corrections, $w_{k}$ -------------------------------------------- The nonlinear interactions between $w_{2}$ and $v_{c}$ can be treated perturbatively in our final argument. Therefore, will need to define $w_{k}$, in a similar manner the $v_{k}$ were and a of the form w_{k}$. Because the estimates for $w_{j}$ and $w_{2}$ will be of a slightly different form, we will (define and) then prove on for \geq 4$ by let $$WRHS_{3}(t,r) = \frac{6\left(Q_{\frac{1}{\lambda(t)}}+v_{c}\right)}{r^{2}} w_{2}^{2}+\frac{2 w_{2}^{3}}{r^{2}} Then, $$|WRHS_{3}(t,r)| \leq C \begin{cases} \frac{r^{2} \lambda( | }_{w_{2,1}}$$ where $$\text{integrand}_{w_{2,1}} = WRHS_{2}(s,\Sqrt{q^{2}(s-t)^{2}+r^{2}+2 r q(S-t) \cos(\TheTa)})\lEfT(1-\fraC{2q^{2}(s-t)^{2} \Sin^{2}(\theta)}{r^{2}+q^{2}(s-t)^{2} + 2 r q (S-T) \cos(\Theta)}\right)$$ Then, we can difFerenTiATe unDEr The inTegral sIGn. tHE reSuLtIng InTEgRals cAn bE estimaTed with the SamE pRocedure used TO eStimate anaLogOus integrals AriSing in ThE exPRessiOns For $\paRtial_{r}W_{2}$ And $w_{2}$. We Get $$|\partiaL_{t}W_{2}(T,r)| \leq \fRAc{C \lambDA(T)^{2} \lOg(t)}{t^{2} \Log^{b}(t) g(t)}, \quad r >0$$ The saME pROcedure is used tO estimAtE $\PaRTIal_{Tr}w_{2}$, And this conClUdes tHE proof oF ThE LEMma
sUmmation of the Higher correCTioNs, $w_{k}$
--------------------------------------------
ThE nOnlINear inTeracTiONs bEtween $w_{2}$ and $v_{C}$ can Not be treaTed perTUrbativELy in our Final aRguMenT. TheREfOrE, we WiLL neED tO deFIne CorrectiOnS $w_{K}$, in a sImilAR MANner As tHe coRrectIons $v_{k}$ were defIneD, and SUm a SerieS of thE forM $\sUm_{k=3}^{\inFty} w_{k}$. BEcausE tHe estimates for $w_{J}$ and $W_{2}$ will be of A slIgHtlY dIfferENt form, We wIll First (deFine and) EStiMaTE $W_{3}$, ThEn prove estimates on $W_{j}$ FOR $j \Geq 4$ by indUction. wE lEt $$wrHS_{3}(t,r) = \fraC{6\lEft(q_{\fraC{1}{\LAmbda(T)}}+v_{c}\rIGhT)}{r^{2}} w_{2}^{2}+\frac{2 w_{2}^{3}}{R^{2}} + \frac{6 w_{2}}{R^{2}}\LeFt(V_{c}^{2}+2v_{c}Q_{\frAc{1}{\Lambda(T)}}\rIghT)$$ ThEn, $$|WRHs_{3}(T,r)| \leQ C \begiN{cases} \frAc{r^{2} \laMBda( | }_{w_{2,1}}$$ where $$\tex t{integran d}_{w _{2 ,1} }= WR HS_{ 2}(s,\sqrt{q^{ 2 }(s- t)^{2}+r^{2}+2 r q(s-t ) \co s( \ thet a )} )\lef t(1-\fr a c{ 2 q ^{2 }( s- t)^ {2 } \ sin^{ 2}( \theta) }{r^{2}+q^ {2} (s -t)^{2} + 2r q (s-t) \co s(\ theta)}\righ t)$ $ Then ,wec an di ffe renti ate un d er the integral s i gn. Th e result i n ginte grals can be esti m at e d with the sam e proc ed u re u sed to estimatean alogo u s integ r al s a ris i ng in the exp ressions fo r $\ partia l_ {r} w _{2}$and $ w_ { 2}$ . We get $$ |\pa rtial_{t} w_{2}( t ,r)| \l e q \frac {C \la mbd a(t )^{2 } \ lo g(t )} { t^{ 2 }\lo g ^{b }(t) g(t )} ,\quad r > 0 $ $ Thesam e pr ocedu re is used to es tima t e $ \part ial_{ tr}w _{ 2}$,and th is co nc ludes the proof ofthe lemma
S um mat io n oft he hig her co rrectio ns, $w_ { k}$
- - - - -- ------------------ -- - - -- -------- ------
T he nonlinea rint erac t i ons b etwe e n$w_{2}$and $v _ {c }$ can no tbe tre at edper turba t ivel y in o ur final argu m ent. Therefore , we will need to d ef i ne c orr ections $w_ {k}$ , ina si m il arm anner as t he co r rections $v_{k}$ we re defin ed, a nd sum a seri es of thef o r m $\sum_ {k=3 } ^{ \ infty} w_{k}$. Beca use the es t imates f or $w _{j}$ an d $w_{2}$ w ill be o f a sl igh tly d if ferent form,w e wil lfirst ( def ine and ) e sti mat e $ w_ {3}$, the n provees ti ma te s o n $w_ { j}$ for$j \g eq 4$ by i n ductio n. We let $ $W R HS_ {3}(t,r ) = \ frac {6 \l eft( Q_{ \f rac{1 }{\l a mbd a(t)}}+ v_{c}\rig ht) } {r^{ 2} }w_{2}^{ 2}+\frac{2 w_ {2 }^{3}}{r^{ 2} } + \frac { 6 w_{2}}{ r^{2}}\left(v_{c}^{2}+2 v _{c}Q_{ \fr ac{1} {\la mbda(t)}} \ri ght)$$ Th e n, $$| WRHS_{ 3}(t, r) | \ l e q C \ b e gi n{c as es} \frac{ r ^ {2} \lam bd a( | }_{w_{2,1}}$$ where_$$\text{integrand}_{w_{2,1}} =_WRHS_{2}(s,\sqrt{q^{2}(s-t)^{2}+r^{2}+2 r q(s-t) \cos(\theta)})\left(1-\frac{2q^{2}(s-t)^{2}_\sin^{2}(\theta)}{r^{2}+q^{2}(s-t)^{2} +_2_r q_(s-t)_\cos(\theta)}\right)$$ Then, we_can differentiate under_the integral sign. The_resulting integrals can_be_estimated with the same procedure used to estimate analogous integrals arising in the expressions_for_$\partial_{r}w_{2}$ and_$w_{2}$._We_get $$|\partial_{t}w_{2}(t,r)| \leq \frac{C \lambda(t)^{2}_\log(t)}{t^{2} \log^{b}(t) g(t)}, \quad r_>0$$ The_same procedure is used to estimate $\partial_{tr}w_{2}$, and_this_concludes the proof_of the lemma
Summation of the higher corrections, $w_{k}$
--------------------------------------------
The nonlinear_interactions between $w_{2}$ and $v_{c}$ can_not be treated_perturbatively_in_our final argument. Therefore,_we will need to define corrections_$w_{k}$, in a similar manner as_the corrections $v_{k}$ were defined, and sum_a series of the form $\sum_{k=3}^{\infty}_w_{k}$. Because the estimates for_$w_{j}$ and_$w_{2}$ will be of a_slightly different form,_we will_first (define and)_estimate $w_{3}$, then prove estimates on_$w_{j}$ for $j_\geq 4$ by induction. We let_$$WRHS_{3}(t,r)_= \frac{6\left(Q_{\frac{1}{\lambda(t)}}+v_{c}\right)}{r^{2}} w_{2}^{2}+\frac{2_w_{2}^{3}}{r^{2}}_+_\frac{6 w_{2}}{r^{2}}\left(v_{c}^{2}+2v_{c}Q_{\frac{1}{\lambda(t)}}\right)$$_Then, $$|WRHS_{3}(t,r)| \leq_C_\begin{cases} \frac{r^{2}_\lambda( |
, we solve the Ginzburg-Landau equation for the wires for each vortex number $n_v$, imposing the single-valuedness condition on the order parameter, $$\theta_{1, L \leftarrow R}-\theta_{2, L \leftarrow R}+2 \delta(B)=2
\pi n_v.
\label{eq:phaseconstraint}$$ This condition will be referred to as the [*phase constraint*]{}. Here, $\theta_{1, L \leftarrow R}=\int_R^L d\vec{r}\cdot
\vec{\nabla}\varphi(B)$ is the phase accumulated along wire 1 in passing from the right to the left lead; $\theta_{2, L \leftarrow
R}$ is similarly defined for wire 2.
Absent any constraints, the lowest energy configuration of the nanowires is the one with no current through the wires. Here, we adopt the gauge in which $\boldsymbol{A}= B y \boldsymbol{e}_x$ for the electromagnetic vector potential, where the coordinates are as shown in Fig. \[coordinates\]. The Ginzburg-Landau expression for the current density in a superconductor is $$\boldsymbol{J} \propto \left(\boldsymbol{\nabla} \varphi(\boldsymbol{r}) -
\frac{2 e}{\hbar} \boldsymbol{A}(\boldsymbol{r}) \right).
\label{currentD0}$$ For our choice of gauge, the vector potential is always parallel to the nanowires, and therefore the lowest energy state of the nanowires corresponds to a phase accumulation given by the flux through the AB contour, $\theta_{1, L \leftarrow R}=-\theta_{2, L \leftarrow R}=2
\pi B a b/\Phi_0$. As we shall show shortly, for our device geometry (i.e. when the wires are sufficiently short, i.e., $b \ll l$), this phase accumulation may be safely ignored, compared to the phase accumulation $\delta(B)$ associated with screening currents induced in the leads. As the nanowires are assumed to be weak compared to the leads, to satisfy the phase constraint (\[eq:phaseconstraint\]), the phase accumulations in the nanowires will typically deviate from their optimal value, generating a circulating current around the AB contour. As a consequence of LAMH theory, this circulating current results in a decrease of the barrier heights for phase slips, and hence an increase in resistance. The period of the | , we solve the Ginzburg - Landau equation for the wires for each vortex numeral $ n_v$, levy the single - valuedness condition on the order argument, $ $ \theta_{1, L \leftarrow R}-\theta_{2, L \leftarrow R}+2 \delta(B)=2
\pi n_v.
\label{eq: phaseconstraint}$$ This condition will be referred to as the [ * phase constraint * ] { }. Here, $ \theta_{1, L \leftarrow R}=\int_R^L d\vec{r}\cdot
\vec{\nabla}\varphi(B)$ is the phase roll up along wire 1 in passing from the right field to the leftover lead; $ \theta_{2, L \leftarrow
R}$ is similarly defined for telegram 2.
Absent any constraints, the lowest department of energy configuration of the nanowires is the one with no current through the wires. Here, we adopt the gauge in which $ \boldsymbol{A}= B y \boldsymbol{e}_x$ for the electromagnetic vector potential, where the coordinate are as shown in Fig. \[coordinates\ ]. The Ginzburg - Landau expression for the current density in a superconductor is $ $ \boldsymbol{J } \propto \left(\boldsymbol{\nabla } \varphi(\boldsymbol{r }) -
\frac{2 e}{\hbar } \boldsymbol{A}(\boldsymbol{r }) \right).
\label{currentD0}$$ For our option of gauge, the vector potential is always parallel to the nanowires, and therefore the lowest department of energy state of the nanowires represent to a phase accumulation given by the flux through the AB shape, $ \theta_{1, L \leftarrow R}=-\theta_{2, L \leftarrow R}=2
\pi B a b/\Phi_0$. As we shall show shortly, for our device geometry (i.e. when the wires are sufficiently short, i.e., $ b \ll l$), this phase accumulation may be safely neglect, compare to the phase accumulation $ \delta(B)$ associated with screening currents induced in the leads. As the nanowires are assumed to be weak compare to the leads, to satisfy the phase restraint (\[eq: phaseconstraint\ ]), the phase accumulation in the nanowires will typically deviate from their optimum value, generating a circulating current around the AB contour. As a consequence of LAMH theory, this circulating current results in a decrease of the barrier heights for phase slip, and hence an increase in resistance. The period of the | , we solve the Ginzburg-Landam equation for tkw wirev for sach vorgex number $n_v$, imposing the smnglw-valutbness condition on tfe order iarameter, $$\theua_{1, L \leftarrow R}-\tista_{2, L \lcytarrka R}+2 \beota(B)=2
\pi n_v.
\label{gq:phaseconstsaint}$$ This congigiln will be referred to as the [*phase constrsijt*]{}. Here, $\theta_{1, J \legearrkw R}=\int_R^L d\vec{r}\cdot
\vec{\nabla}\varphi(B)$ is the phase accumukated along wire 1 in passijg fgom the right to tje left leae; $\thqra_{2, L \leftarruw
R}$ is simpnarly defihed for wire 2.
Absent any constraknts, che lowest guwrgj configuratmon of the nanowircx is tve one eith no currenb thrmugy the wires. Here, we avopt the gauge in whych $\boldsfmyol{A}= B y \boldsymbol{e}_x$ fir thg elewtrooqgndtid tecfor pohenvial, where fhe coordinqtes are as shown im Spb. \[coordinates\]. The Gynsburg-Landau expression for the current genaity in a superconductoe is $$\boldsymbol{J} \propjo \left(\bolqsymbol{\nabla} \varphi(\boldsymbol{r}) -
\frac{2 e}{\hbar} \boldsymtol{A}(\bkudsvnbol{r}) \eihht).
\label{currentD0}$$ For our choice of gauge, the dscuor potential is clways parallel yo tng nanowires, ana therzrode the lowest enerhy statg of tye nanowiwes vorresponds to a phase accunulation givvn bt the flux through the AB concour, $\tneta_{1, K \leftarrow R}=-\theta_{2, L \leytarros R}=2
\pi B a b/\Ohi_0$. As we rhall show shortuy, nor our device geometry (i.e. wheg the wirxs arz suffickentky showt, i.e., $b \ll l$), thlv phase accumulatiln mai be sdfely ignoged, compared to the phase accumnkation $\delta(N)$ dssmciated cith sgreening currenes induced in jhe leads. As tfe nanowirvs are asvumed to be weak compareg to the leadv, to satysfy the phase zunstraint (\[eq:phaxeconstrapnu\]), the phasw accumulations in thg hanowires will cvpucally deviate groo trepr ppehmal value, ganerxtivb a ckrculating gurfent around the AB contogr. Aa a consequence of LWMH theoty, this cyrculating cutrent results in a decrxase oh the nartier heights for phase slips, ans hence aj ikcrease in refistqnce. The pernod of the | , we solve the Ginzburg-Landau equation for for vortex number imposing the single-valuedness $$\theta_{1, \leftarrow R}-\theta_{2, L R}+2 \delta(B)=2 \pi \label{eq:phaseconstraint}$$ This condition will be referred as the [*phase constraint*]{}. Here, $\theta_{1, L \leftarrow R}=\int_R^L d\vec{r}\cdot \vec{\nabla}\varphi(B)$ is the accumulated along wire 1 in passing from the right to the left lead; L R}$ similarly for wire 2. Absent any constraints, the lowest energy configuration of the nanowires is the one no current through the wires. Here, we adopt gauge in which $\boldsymbol{A}= y \boldsymbol{e}_x$ for the electromagnetic potential, the coordinates as in \[coordinates\]. The Ginzburg-Landau for the current density in a superconductor is $$\boldsymbol{J} \propto \left(\boldsymbol{\nabla} \varphi(\boldsymbol{r}) - \frac{2 e}{\hbar} \boldsymbol{A}(\boldsymbol{r}) \right). For our gauge, the potential always to the nanowires, the lowest energy state of the a phase accumulation given by the flux through AB contour, L \leftarrow R}=-\theta_{2, L \leftarrow R}=2 B a b/\Phi_0$. As we shall show shortly, our device geometry (i.e. when the wires are sufficiently short, i.e., $b \ll l$), this may be safely ignored, to the phase $\delta(B)$ with currents in the As the nanowires are assumed to be weak compared to the to satisfy the phase constraint (\[eq:phaseconstraint\]), the phase accumulations in will deviate from their value, generating a circulating around AB contour. As a LAMH this in decrease the barrier heights for slips, and hence an increase resistance. The period of | , we solve the Ginzburg-Landau eQuation for The wiRes For EaCh voRtex Number $n_v$, imposiNG the Single-valuedness conditIon on ThE OrdeR PaRametEr, $$\theta_{1, l \LeFTArrOw r}-\tHetA_{2, L \LEfTarroW R}+2 \dElta(B)=2
\pi N_v.
\label{eq:pHasEcOnstraint}$$ ThiS CoNdition wilL be Referred to as The [*Phase cOnStrAInt*]{}. HeRe, $\tHeta_{1, L \LeftarROw R}=\int_r^L d\vec{r}\cdOt
\VEc{\nablA}\Varphi(B)$ IS ThE phaSe accumulated alonG WiRE 1 in passing from The rigHt TO tHE LefT leAd; $\theta_{2, L \leFtArrow
r}$ Is similARlY DEFinED for wire 2.
AbsenT any constraINts, The lowEsT enERgy conFigurAtIOn oF the nanowirEs is The one witH no curREnt throUGh the wiRes. HerE, we AdoPt thE GaUgE in WhICh $\bOLdSymBOl{A}= b y \boldsyMbOl{E}_x$ for The eLECTRomaGneTic vEctor Potential, wherE thE cooRDinAtes aRe as sHown In fig. \[coOrdinaTes\]. ThE GInzburg-Landau exPresSion for thE cuRrEnt DeNsity IN a supeRcoNduCtor is $$\bOldsymbOL{J} \pRoPTO \LeFt(\boldsymbol{\nabla} \vArPHI(\bOldsymboL{r}) -
\frac{2 E}{\HbAr} \BOldsymboL{A}(\BolDsymBOL{r}) \rigHt).
\laBEl{CurrentD0}$$ for our CHoIcE of gaugE, tHe vectOr PotEntIal is ALwayS paralLel to the NanowIRes, and thereforE The lowest enerGY sTATe OF the NanOwires correSponDS to a PhasE AcCumULatioN giveN bY ThE Flux through the AB conToUr, $\thetA_{1, L \lefTarrow R}=-\theta_{2, L \Leftarrow R}=2
\PI b A b/\Phi_0$. As wE shaLL sHOw shortly, for ouR deviCe geometry (I.E. when the Wires Are suffiCiently shORT, i.e., $b \ll l$), tHis PhaSe aCcuMULaTion may be safeLY IgnoReD, comparEd tO the phaSe aCcuMulAtiOn $\Delta(B)$ assOciated wItH sCrEeNinG currENts inducEd In tHe LeaDs. As tHE nanowIres aRe asSuMeD To bE weak coMPaRED to tHe LeAds, tO saTiSfy thE phaSE coNstrainT (\[eq:phasecOnsTRainT\]), tHe Phase acCumulations in ThE nanowires WiLl tYpicalLY Deviate fRom their optimal value, genERating a CirCulatIng cUrrent aroUnd The AB cOntOUr. As a cOnsequEnce oF LaMH THEory, tHIS cIrcUlAting curreNT ResUlts iN a DecrEase of tHe barrier heights foR PhaSe slips, and henCe aN incREAsE in REsIStaNcE. the PERiod of the | , we solve the Ginzburg-La ndau equat ion f orthe w ires for each vortex n u mber $n_v$, imposing the s ingle -v a lued n es s con ditiono nt h e o rd er pa ra m et er, $ $\t heta_{1 , L \lefta rro wR}-\theta_{2 , L \leftarro w R }+2 \delta(B )=2
\pi n _v .
\ l abel{ eq: phase constr a int}$$ This con di t ion wi l l be re f e rr ed t o as the [*phasec on s traint*]{}. He re, $\ th e ta _ { 1,L \ leftarrowR} =\int _ R^L d\v e c{ r } \ cdo t
\vec{\nabla} \varphi(B)$ isthe ph as e a c cumula ted a lo n g w ire 1 in pa ssin g from th e righ t to the left le ad; $\ the ta_ {2,L \ le fta rr o w
R } $iss imi larly de fi ne d for wir e 2 .
Ab sen t an y con straints, the lo west ene rgy c onfig urat io n ofthe na nowir es is the one wit h no currentthr ou ghth e wir e s. Her e,weadopt t he gaug e in w h i c h$\boldsymbol{A}= B y \ bo ldsymbol {e}_x$ fo rt he elect ro mag neti c vecto r po t en tial, wh ere th e c oo rdinate sare as s how n i n Fig . \[c oordin ates\].The G i nzburg-Landaue xpression for th e cu r rent de nsity in asupe r cond ucto r i s $ $ \bold symbo l{ J }\ propto \left(\bolds ym bol{\n abla} \varphi(\bol dsymbol{r} ) -
\frac{2 e}{ \ hb a r} \boldsymbol {A}(\ boldsymbol { r}) \rig ht).\label{c urrentD0} $ $ For our ch oic e o f g a u ge , the vectorp o tent ia l is al way s paral lel to th e n an owires, a nd there fo re t he lo weste nergy st at e o fthe nano w ires c orres pond sto a p hase ac c um u l atio ngi venbyth e flu x th r oug h the A B contour , $ \ thet a_ {1 , L \le ftarrow R}=-\ th eta_{2, L\l eft arrowR } =2
\pi B a b/\Phi_0$. As we sha l l showsho rtly, for our devi cegeomet ry( i.e. w hen th e wir es ar e suffi c i en tly s hort, i.e. , $b\ll l $) , th is phas e accumulation may besafely ignore d,comp a r ed to th e ph as e ac c u mulation $\delt a(B)$ asso ci a te d with scr e eni ng curren ts indu ced i n the le ads. As t he nanowi re s ar e ass umed to be weak co mpared to the l e ad s, to sa tisfyth e p haseconstr a int (\[e q:phas ec onstra int\] ), the pha se accumulations in the nanow ireswil l typical lyd evi ate fromthei r optimalval ue, gene rat i ng acirc u la tin g curr enta round the AB co n t ou r. As a con s e q uen ce of LA M H theo ry,this circulatingc urrent results ina dec rea s e of t he barrier hei ght sf o r phasesl ips, and he nce an i nc r easein res istanc e. Thep e ri o d of t he | , we_solve the_Ginzburg-Landau equation for the_wires for_each_vortex number_$n_v$,_imposing the single-valuedness_condition on the_order parameter, $$\theta_{1, L_\leftarrow R}-\theta_{2, L_\leftarrow_R}+2 \delta(B)=2
\pi n_v.
\label{eq:phaseconstraint}$$ This condition will be referred to as the [*phase constraint*]{}. Here,_$\theta_{1,_L \leftarrow_R}=\int_R^L_d\vec{r}\cdot
\vec{\nabla}\varphi(B)$_is the phase accumulated along_wire 1 in passing from_the right_to the left lead; $\theta_{2, L \leftarrow
R}$ is_similarly_defined for wire_2.
Absent any constraints, the lowest energy configuration of the_nanowires is the one with no_current through the_wires._Here,_we adopt the gauge_in which $\boldsymbol{A}= B y \boldsymbol{e}_x$_for the electromagnetic vector potential, where_the coordinates are as shown in Fig. \[coordinates\]._The Ginzburg-Landau expression for the current_density in a superconductor is_$$\boldsymbol{J} \propto_\left(\boldsymbol{\nabla} \varphi(\boldsymbol{r}) -
\frac{2 e}{\hbar} \boldsymbol{A}(\boldsymbol{r})_\right).
\label{currentD0}$$ For our_choice of_gauge, the vector_potential is always parallel to the_nanowires, and therefore_the lowest energy state of the_nanowires_corresponds to a_phase_accumulation_given by_the flux through_the_AB contour,_$\theta_{1,_L \leftarrow R}=-\theta_{2, L \leftarrow R}=2
\pi_B_a b/\Phi_0$. As we shall show shortly,_for our device geometry_(i.e. when_the wires are sufficiently_short, i.e., $b \ll l$),_this phase accumulation may be safely_ignored, compared_to the_phase accumulation $\delta(B)$ associated with screening currents induced in the leads._As the nanowires are assumed to_be weak compared to_the leads,_to_satisfy the phase_constraint (\[eq:phaseconstraint\]),_the phase_accumulations in the nanowires will typically deviate_from their_optimal value, generating a circulating current_around the AB contour._As_a consequence of LAMH theory, this_circulating current results in a decrease_of the barrier heights for_phase_slips,_and hence an increase in_resistance. The period of the |
^r\ \text{and}\ {\text{\normalfont{ord}}}_{{\mathfrak{l}}_i}((\beta))\equiv 1\ ({\text{\normalfont{mod}}}\ 2)\}$. Then $u\in\{t_{K^r}-1,\ t_{K^r}-2,\ t_{K^r}-3\}$ by (\[order\]).
Let $p{\mathcal{O}}_{F^r}={\mathfrak{p}}{\mathfrak{p}}'$. Then we have $\alpha=\pi^{a}\pi'^{a'}\prod_{i=1}^{u} s_{i}^{(1\ {\text{\normalfont{mod}}}\ 2)}$, where $\pi$ and $\pi'$ are totally positive generators of ${\mathfrak{p}}^j$ and ${\mathfrak{p}}'^{j}$ for odd $j$. Here, we have $\prod_{i=1}^{u} s_{i}^{(1\ {\text{\normalfont{mod}}}\ 2)}\in{\mathbb{Z}}$ and $a,\ a'\in\{0,\ 1\}$. If $a = a'$, then $\alpha\in{\mathbb{Z}}$, which leads to contradiction since $K^r$ is non-biquadratic. So for a unique ${\mathfrak{p}}$, we can take $a_1=1$ and $a_2=0$. In particular, we have $u\in\{t_{K^r}-1,\ t_{K^r}-2\}$.
Combining Lemma \[construct.1\] and the bound on the discriminant in Proposition \[t<6\], we now have a good way of listing the fields. Next, we need a fast way of eliminating fields from our list if they have CM class number $>1$.
The following lemma is a special case of Theorem D in Louboutin [@loub5].
\[quick\] Let $K$ be a non-biquadratic quartic CM field with real quadratic subfield $F$. Let $d_K$ and $d_F$ be the absolute values of the discriminants of $K$ and $F$. Then assuming $I_{0}(\Phi^r)=I_{K^r}$, if a rational prime $l$ totally splits in $K^r/{\mathbb{Q}}$, then $l\geq \frac{\sqrt{d_K/{d_F}^2}}{4}$.
Let $l$ be a prime | ^r\ \text{and}\ { \text{\normalfont{ord}}}_{{\mathfrak{l}}_i}((\beta))\equiv 1\ ({ \text{\normalfont{mod}}}\ 2)\}$. Then $ u\in\{t_{K^r}-1,\ t_{K^r}-2,\ t_{K^r}-3\}$ by (\[order\ ]).
Let $ p{\mathcal{O}}_{F^r}={\mathfrak{p}}{\mathfrak{p}}'$. Then we have $ \alpha=\pi^{a}\pi'^{a'}\prod_{i=1}^{u } s_{i}^{(1\ { \text{\normalfont{mod}}}\ 2)}$, where $ \pi$ and $ \pi'$ are totally positive generator of $ { \mathfrak{p}}^j$ and $ { \mathfrak{p}}'^{j}$ for curious $ j$. Here, we have $ \prod_{i=1}^{u } s_{i}^{(1\ { \text{\normalfont{mod}}}\ 2)}\in{\mathbb{Z}}$ and $ a,\ a'\in\{0,\ 1\}$. If $ a = a'$, then $ \alpha\in{\mathbb{Z}}$, which leads to contradiction since $ K^r$ is non - biquadratic. So for a singular $ { \mathfrak{p}}$, we can take $ a_1=1 $ and $ a_2=0$. In particular, we hold $ u\in\{t_{K^r}-1,\ t_{K^r}-2\}$.
Combining Lemma \[construct.1\ ] and the bound on the discriminant in Proposition \[t<6\ ], we now suffer a good way of list the fields. Next, we need a flying direction of eliminating fields from our list if they have CM class act $ > 1$.
The following lemma is a special case of Theorem D in Louboutin [ @loub5 ].
\[quick\ ] lease $ K$ be a non - biquadratic quartic CM field with real quadratic subfield $ F$. Let $ d_K$ and $ d_F$ be the absolute value of the discriminants of $ K$ and $ F$. Then assuming $ I_{0}(\Phi^r)=I_{K^r}$, if a rational prime $ l$ totally splits in $ K^r/{\mathbb{Q}}$, then $ l\geq \frac{\sqrt{d_K/{d_F}^2}}{4}$.
lease $ l$ be a prime | ^r\ \tfxt{and}\ {\text{\normalfont{ord}}}_{{\oathfrak{l}}_i}((\beta))\equiv 1\ ({\txxt{\normzlfont{moa}}}\ 2)\}$. Then $u\in\{t_{K^r}-1,\ t_{K^r}-2,\ t_{K^r}-3\}$ by (\[orver\]).
Lwt $p{\mqthcal{O}}_{F^r}={\mathfrak{p}}{\mathwrak{p}}'$. Thej we havw $\alkha=\pi^{a}\pi'^{a'}\prod_{i=1}^{u} s_{m}^{(1\ {\text{\novialfkkt{mod}}}\ 2)}$, where $\pi$ and $\ki'$ are totalny positive getefacors of ${\mathfrak{p}}^j$ and ${\mathfrak{p}}'^{j}$ fow odd $j$. Hfre, we have $\prjd_{i=1}^{u} f_{i}^{(1\ {\tsqt{\kormalfont{mod}}}\ 2)}\in{\mathbb{Z}}$ and $a,\ a'\ih\{0,\ 1\}$. If $a = a'$, then $\alphs\in{\mathbb{Z}}$, which leads to fontgadiction since $K^r$ is non-biquqdraeuc. So for a jnique ${\mathfrak{p}}$, we cah take $a_1=1$ and $a_2=0$. In particular, we have $u\in\{t_{K^r}-1,\ t_{K^t}-2\}$.
Eimblting Lemma \[ronstrlct.1\] and the bound on dhe disvriminant in Pvoposmtiob \[t<6\], we now have a gmod way of listing the fielgs. Next, we need a fqsr way of alimkbatkng fmelss frol onr list if fhey have CN class number $>1$.
The gojoowing lemma js a s[esial case of Theorem D in Louboutin [@lout5].
\[qujck\] Let $K$ be a non-biquaeratic quartic CM fiepd with rqal quadratic subfield $F$. Let $d_K$ and $d_F$ be the absmlute xalbcf if the discriminants of $K$ and $F$. Then assuming $I_{0}(\Prj^r)=O_{K^g}$, if a rational pvime $l$ totally splotd og $K^r/{\mathbb{Q}}$, tfen $l\gzs \rrac{\sqrt{d_K/{d_F}^2}}{4}$.
Let $l$ he a pryme | ^r\ \text{and}\ {\text{\normalfont{ord}}}_{{\mathfrak{l}}_i}((\beta))\equiv 1\ ({\text{\normalfont{mod}}}\ 2)\}$. Then t_{K^r}-3\}$ (\[order\]). Let Then we have $\pi$ $\pi'$ are totally generators of ${\mathfrak{p}}^j$ ${\mathfrak{p}}'^{j}$ for odd $j$. Here, we $\prod_{i=1}^{u} s_{i}^{(1\ {\text{\normalfont{mod}}}\ 2)}\in{\mathbb{Z}}$ and $a,\ a'\in\{0,\ 1\}$. If $a = a'$, then which leads to contradiction since $K^r$ is non-biquadratic. So for a unique ${\mathfrak{p}}$, can $a_1=1$ $a_2=0$. particular, we have $u\in\{t_{K^r}-1,\ t_{K^r}-2\}$. Combining Lemma \[construct.1\] and the bound on the discriminant in Proposition we now have a good way of listing fields. Next, we need fast way of eliminating fields our if they CM number The following lemma a special case of Theorem D in Louboutin [@loub5]. \[quick\] Let $K$ be a non-biquadratic quartic CM with real $F$. Let and be absolute values of of $K$ and $F$. Then assuming rational prime $l$ totally splits in $K^r/{\mathbb{Q}}$, then \frac{\sqrt{d_K/{d_F}^2}}{4}$. Let be a prime | ^r\ \text{and}\ {\text{\normalfont{ord}}}_{{\Mathfrak{l}}_i}((\Beta))\eQuiV 1\ ({\teXt{\NormAlfoNt{mod}}}\ 2)\}$. Then $u\in\{t_{K^R}-1,\ T_{K^r}-2,\ t_{k^r}-3\}$ by (\[order\]).
Let $p{\mathcal{O}}_{F^R}={\mathFrAK{p}}{\maTHfRak{p}}'$. THen we haVE $\aLPHa=\pI^{a}\Pi'^{A'}\prOd_{I=1}^{U} s_{I}^{(1\ {\text{\NorMalfont{Mod}}}\ 2)}$, where $\pi$ And $\Pi'$ Are totally poSItIve generatOrs Of ${\mathfrak{p}}^j$ And ${\MathfrAk{P}}'^{j}$ fOR odd $j$. herE, we haVe $\prod_{I=1}^{U} s_{i}^{(1\ {\texT{\normalfoNt{MOd}}}\ 2)}\in{\maTHbb{Z}}$ and $A,\ A'\In\{0,\ 1\}$. if $a = a'$, Then $\alpha\in{\mathbb{z}}$, WhICh leads to contrAdictiOn SInCE $k^r$ iS noN-biquadratIc. so for A Unique ${\mAThFRAK{p}}$, wE Can take $a_1=1$ and $a_2=0$. IN particular, WE haVe $u\in\{t_{k^r}-1,\ T_{K^r}-2\}$.
cOmbiniNg LemMa \[COnsTruct.1\] and the BounD on the disCriminANt in ProPOsition \[T<6\], we nOw hAve A gooD WaY oF liStINg tHE fIelDS. NeXt, we need A fAsT way oF eliMINATing FieLds fRom ouR list if they haVe Cm claSS nuMber $>1$.
THe folLowiNg Lemma Is a speCial cAsE of Theorem D in LoUbouTin [@loub5].
\[quIck\] leT $K$ bE a Non-biQUadratIc qUarTic CM fiEld with REal QuADRAtIc subfield $F$. Let $d_K$ anD $d_f$ BE tHe absoluTe valuES oF tHE discrimInAntS of $K$ AND $F$. TheN assUMiNg $I_{0}(\Phi^r)=I_{k^r}$, if a rATiOnAl prime $L$ tOtally SpLitS in $k^r/{\matHBb{Q}}$, tHen $l\geQ \frac{\sqrT{d_K/{d_F}^2}}{4}$.
lEt $l$ be a prime | ^r\ \text{and}\ {\text{\no rmalfont{o rd}}} _{{ \ma th frak {l}} _i}((\beta))\e q uiv1\ ({\text{\normalfont {mod} }} \ 2)\ } $. Then $u\in\ { t_ { K ^r} -1 ,\ t_ {K ^ r} -2,\t_{ K^r}-3\ }$ by (\[o rde r\ ]).
Let $p{ \ ma thcal{O}}_ {F^ r}={\mathfra k{p }}{\ma th fra k {p}}' $.Thenwe hav e $\alp ha=\pi^{a }\ p i'^{a' } \prod_{ i = 1} ^{u} s_{i}^{(1\ {\tex t {\ n ormalfont{mod} }}\ 2) }$ , w h e re$\p i$ and $\p i' $ are totally po s i t ive generators of ${\mathfra k {p} }^j$ a nd ${ \ mathfr ak{p} }' ^ {j} $ for odd $ j$.Here, wehave $ \ prod_{i = 1}^{u}s_{i}^ {(1 \ { \tex t {\ no rma lf o nt{ m od }}} \ 2) }\in{\ma th bb {Z}}$ and $ a , \ a' \in \{0, \ 1\} $. If $a = a' $,then $\a lpha\ in{\m athb b{ Z}}$, which lead sto contradictio n si nce $K^r$ is n on- bi quadr a tic. S o f ora uniqu e ${\ma t hfr ak { p } }$ , we can take $a_1 =1 $ an d $a_2=0 $. Inp ar ti c ular, we h ave $u\ i n \{t_{ K^r} - 1, \ t_{K^r }-2\}$ .
Co mbining L emma \ [c ons tru ct.1\ ] and the b ound onthe d i scriminant inP roposition \[ t &l t ; 6\ ] , we no w have a go od w a y of lis t in g t h e fie lds.Ne x t, we need a fast wayof elimi natin g fields from our listi f they hav e CM cl a ss number $>1$ .
Th e followin g lemma i s a s pecial c ase of Th e o rem D in Lo ubo uti n [ @ l ou b5].
\[quick \ ] Let $ K$ be a no n-biqua dra tic qu art ic CM field with re al q ua dr ati c sub f ield $F$ .Let $ d_K $ and $d_F$be th e ab so lu t e v alues o f t h e dis cr im inan tsof $K$and$ F$. Then a ssuming $ I_{ 0 }(\P hi ^r )=I_{K^ r}$, if a rat io nal prime$l $ t otally s plits in $K^r/{\mathbb{Q}}$, th e n $l\ge q \ frac{ \sqr t{d_K/{d_ F}^ 2}}{4} $.Let $l $ be a prim e | ^r\ \text{and}\_{\text{\normalfont{ord}}}_{{\mathfrak{l}}_i}((\beta))\equiv 1\_({\text{\normalfont{mod}}}\ 2)\}$. Then $u\in\{t_{K^r}-1,\_t_{K^r}-2,\ t_{K^r}-3\}$_by_(\[order\]).
Let $p{\mathcal{O}}_{F^r}={\mathfrak{p}}{\mathfrak{p}}'$._Then_we have $\alpha=\pi^{a}\pi'^{a'}\prod_{i=1}^{u}_s_{i}^{(1\ {\text{\normalfont{mod}}}\ 2)}$,_where $\pi$ and $\pi'$_are totally positive_generators_of ${\mathfrak{p}}^j$ and ${\mathfrak{p}}'^{j}$ for odd $j$. Here, we have $\prod_{i=1}^{u} s_{i}^{(1\ {\text{\normalfont{mod}}}\ 2)}\in{\mathbb{Z}}$_and_$a,\ a'\in\{0,\_1\}$._If_$a = a'$, then $\alpha\in{\mathbb{Z}}$,_which leads to contradiction since_$K^r$ is_non-biquadratic. So for a unique ${\mathfrak{p}}$, we can_take_$a_1=1$ and $a_2=0$._In particular, we have $u\in\{t_{K^r}-1,\ t_{K^r}-2\}$.
Combining Lemma \[construct.1\] and_the bound on the discriminant in_Proposition \[t<6\], we_now_have_a good way of_listing the fields. Next, we need_a fast way of eliminating fields_from our list if they have CM_class number $>1$.
The following lemma is_a special case of Theorem_D in_Louboutin [@loub5].
\[quick\] Let $K$ be_a non-biquadratic quartic_CM field_with real quadratic_subfield $F$. Let $d_K$ and $d_F$_be the absolute_values of the discriminants of $K$_and $F$._Then assuming $I_{0}(\Phi^r)=I_{K^r}$,_if_a_rational prime_$l$ totally splits_in_$K^r/{\mathbb{Q}}$, then_$l\geq_\frac{\sqrt{d_K/{d_F}^2}}{4}$.
Let $l$ be a prime |
Charles Fabry de l’Institut d’Optique, UMR 8501 du CNRS, 91403 Orsay Cedex, France'
- 'Department of Physics, 60 St George Street, University of Toronto, Toronto, ON, M5S 1A7, Canada'
author:
- 'P. Bouyer'
- 'J. H. Thywissen'
- 'F. Gerbier'
- 'M. Hugbart'
- 'S. Richard'
- 'J. Retter'
- 'A. Aspect'
title: 'One-dimensional behavior of elongated Bose-Einstein condensates'
---
Introduction {#intro}
============
The experimental study of one dimensional (1D) degenerate Bose gases, where the radial motion is“frozen”, is currently an important direction of research in ultra-cold atom physics [@gorlitz01; @schreck01; @greiner01]. In uniform 1D systems, fluctuations of the condensate phase are pronounced, because of a large population of low-lying states; as a result, no long range order exists at any temperature. In a trapped 1D gas, the finite size of the sample naturally introduces a low-energy cutoff, and at sufficiently low temperature a phase coherent sample can exist (see [@petrov1d] and references therein for the properties of 1D gases). Essentially, the same analysis holds for three dimensional (3D) trapped gases in very elongated traps [@petrov3d]: here the interaction energy per particle is larger than the radial frequency, and the condensate wavefunction is built from several radial modes. Still, low-energy excitations with a frequency smaller than the radial frequency are effectively one-dimensional, and induce large fluctuations of the condensate phase [@petrov3d]. Such condensates with a fluctuating phase, or [*[quasicondensates]{}*]{}, have been observed experimentally in [@dettmer01; @hellweg01; @shvarchuck02; @Richard03; @Ertmer03].
We present here an overview of our experimental study of elongated condensates in a very anisotropic trap. Our set-up, presented in section \[creation\] allows us to characterize precisely the properties of degenerate gases between the 3D and 1D regimes. In this regime, we first show (section \[exp\]) that the (quasi)condensates exhibit the same | Charles Fabry de l’Institut d’Optique, UMR 8501 du CNRS, 91403 Orsay Cedex, France'
-' Department of Physics, 60 St George Street, University of Toronto, Toronto, ON, M5S 1A7, Canada'
author:
-' P. Bouyer'
-' J. H. Thywissen'
-' F. Gerbier'
-' M. Hugbart'
-' S. Richard'
-' J. Retter'
-' A. Aspect'
title:' One - dimensional behavior of elongated Bose - Einstein condensate'
---
initiation { # intro }
= = = = = = = = = = = =
The experimental study of one dimensional (1D) degenerate Bose gas, where the radial motion is“frozen ”, is currently an significant direction of research in extremist - cold atom purgative [ @gorlitz01; @schreck01; @greiner01 ]. In uniform 1D systems, fluctuation of the condensate phase are pronounced, because of a large population of low - lie down states; as a result, no long range order exist at any temperature. In a trapped 1D gas, the finite size of the sample naturally bring in a low - energy cutoff, and at sufficiently low temperature a phase coherent sample distribution can exist (see [ @petrov1d ] and reference book therein for the properties of 1D gases). Essentially, the like analysis holds for three dimensional (3D) trapped gases in very elongated traps [ @petrov3d ]: here the interaction department of energy per atom is big than the radial frequency, and the condensate wavefunction is built from several radial modes. Still, low - energy excitations with a frequency smaller than the radial frequency are efficaciously one - dimensional, and induce bombastic fluctuations of the condensation phase [ @petrov3d ]. Such condensate with a fluctuating phase, or [ * [ quasicondensates ] { } * ] { }, have been observe experimentally in [ @dettmer01; @hellweg01; @shvarchuck02; @Richard03; @Ertmer03 ].
We present here an overview of our experimental study of elongate condensates in a very anisotropic ambush. Our set - up, presented in incision \[creation\ ] allows us to characterize precisely the properties of degenerate gases between the 3D and 1D regimen. In this regime, we first show (section \[exp\ ]) that the (quasi)condensates parade the same | Chwrles Fabry de l’Institut d’Optique, UMR 8501 by CNRS, 91403 Orsag Cedex, Wrance'
- 'Department of Physics, 60 Wt Geirge Street, University of Toronno, Toronti, ON, N5S 1A7, Canade'
zuthor:
- 'I. Youyed'
- 'J. H. Tkyxissen'
- 'F. Gerbier'
- 'M. Hugbart'
- 'V. Richard'
- 'J. Retdef'
- 'C. Aspect'
title: 'One-dimensional behaviow of elpnhated Bose-Einsjein bogdenaates'
---
Introduction {#intro}
============
The experimsntal suudy of one dimensoonal (1D) degenerate Bose gades, ahere the radial mltion is“froesn”, yw currently xn important direction of research in ultra-cold atom ohysies [@gorlitz01; @wcyrefn01; @greiner01]. Ii unifjrm 1D systems, fluctudtions pf the condensste phqse are pronounced, berause of a large popolation of lkw-lying states; as a resunt, nm lovt rxngt ocded exishs et any tempsrature. In q trapped 1D gas, the fybite size of fhe saipje naturally introduces a low-energy cutmff, and at sufficiently loq temperature a phase coherent sample can exist (see [@petrov1d] and references therain fkf tkc oeooerties of 1D gases). Essentially, the same analyfjs hplds for three dimensionak (3F) ytapped gases iv very elkngated traps [@petrlv3d]: herg the unteractijn emergy per particle is largee than the rcdiql frequency, and tke condensatz wavegunctoon is built from severcl radjal modes. Shill, low-ehdrgy excitations winh a frequency smaller than thq radial hrequzncy are effgctivelr one-dimendional, and induce large vluctoationv of the clndensate phase [@petrov3d]. Such coivensates with a fllctuating phasc, or [*[quasicondegsates]{}*]{}, have begn observzd expdrimentallj in [@dettker01; @hellweg01; @shvarchuck02; @Rhfhard03; @Ertmer03].
Xe presene heee ab overvkdw of our expetimental study of eoongated condensatcs in z very anisotro'nc trap. Our set-up, prdsegtvd mn sesdion \[creatiot\] aluowr us tu characterlze previsely the propertiev of degenerate gases nebween the 3D and 1D regimes. In tnis regime, we firsu show (sertion \[rxp\]) that the (quasi)condensates exhjbit the damc | Charles Fabry de l’Institut d’Optique, UMR 8501 91403 Cedex, France' 'Department of Physics, of Toronto, ON, M5S Canada' author: - Bouyer' - 'J. H. Thywissen' - Gerbier' - 'M. Hugbart' - 'S. Richard' - 'J. Retter' - 'A. Aspect' 'One-dimensional behavior of elongated Bose-Einstein condensates' --- Introduction {#intro} ============ The experimental study one (1D) Bose where the radial motion is“frozen”, is currently an important direction of research in ultra-cold atom physics @schreck01; @greiner01]. In uniform 1D systems, fluctuations of condensate phase are pronounced, of a large population of states; a result, long order at any temperature. a trapped 1D gas, the finite size of the sample naturally introduces a low-energy cutoff, and at low temperature coherent sample exist [@petrov1d] references therein for of 1D gases). Essentially, the same three dimensional (3D) trapped gases in very elongated [@petrov3d]: here interaction energy per particle is larger the radial frequency, and the condensate wavefunction is from several radial modes. Still, low-energy excitations with a frequency smaller than the radial frequency one-dimensional, and induce large of the condensate [@petrov3d]. condensates a phase, or have been observed experimentally in [@dettmer01; @hellweg01; @shvarchuck02; @Richard03; @Ertmer03]. We here an overview of our experimental study of elongated condensates very trap. Our set-up, in section \[creation\] allows to precisely the properties of between 3D In regime, first show (section \[exp\]) the (quasi)condensates exhibit the same | Charles Fabry de l’Institut d’OPtique, UMR 8501 dU CNRS, 91403 orsAy CEdEx, FrAnce'
- 'department of PhYSics, 60 st George Street, UniversiTy of TOrONto, TORoNto, ON, m5S 1A7, CanaDA'
aUTHor:
- 'p. BOuYer'
- 'j. H. tHyWisseN'
- 'F. GErbier'
- 'M. hugbart'
- 'S. RiChaRd'
- 'j. Retter'
- 'A. AspeCT'
tItle: 'One-dimEnsIonal behavioR of ElongaTeD BoSE-EinsTeiN condEnsateS'
---
introdUction {#intRo}
============
tHe expeRImental STUdY of oNe dimensional (1D) degENeRAte Bose gases, whEre the RaDIaL MOtiOn iS“frozen”, is cUrRentlY An imporTAnT DIRecTIon of research In ultra-cold ATom PhysicS [@gOrlITz01; @schrEck01; @grEiNEr01]. IN uniform 1D syStemS, fluctuatIons of THe condeNSate phaSe are pRonOunCed, bECaUsE of A lARge POpUlaTIon Of low-lyiNg StAtes; aS a reSULT, No loNg rAnge Order Exists at any teMpeRatuRE. In A trapPed 1D gAs, thE fInite Size of The saMpLe naturally intrOducEs a low-eneRgy CuTofF, aNd at sUFficieNtlY loW temperAture a pHAse CoHEREnT sample can exist (see [@PeTROv1D] and refeRences THeReIN for the pRoPerTies OF 1d gaseS). EssENtIally, the Same anALySiS holds fOr Three dImEnsIonAl (3D) trAPped Gases iN very eloNgateD Traps [@petrov3d]: heRE the interactiON eNERgY Per pArtIcle is largeR thaN The rAdiaL FrEquENcy, anD the cOnDEnSAte wavefunction is buIlT from sEveraL radial modes. STill, low-eneRGY ExcitatiOns wITh A Frequency smallEr thaN the radial FRequency Are efFectivelY one-dimenSIOnal, and iNduCe lArgE flUCTuAtions of the coNDEnsaTe Phase [@peTroV3d]. Such cOndEnsAteS wiTh A fluctuatIng phase, Or [*[QuAsIcOndEnsatES]{}*]{}, have beeN oBseRvEd eXperiMEntallY in [@deTtmeR01; @hElLWeg01; @ShvarchUCk02; @rICharD03; @ERtMer03].
WE prEsEnt heRe an OVerView of oUr experimEntAL stuDy Of ElongatEd condensates In A very anisoTrOpiC trap. OUR Set-up, preSented in section \[creation\] ALlows us To cHaracTeriZe preciseLy tHe propErtIEs of deGeneraTe gasEs BetWEEn the 3d ANd 1d reGiMes. In this rEGIme, We firSt Show (Section \[Exp\]) that the (quasi)conDEnsAtes exhibit thE saMe | Charles Fabry de l’Instit ut d’Optiq ue, U MR850 1du C NRS, 91403 Orsay C e dex, France'
- 'Department of P hy s ics, 60 St G eorge S t re e t , U ni ve rsi ty of Toro nto , Toron to, ON, M5 S 1 A7 , Canada'
au t ho r:
- 'P. B ouy er'
- 'J. H. Th ywisse n'
-' F. Ge rbi er'
- 'M. H u gbart'
- 'S. Ri ch a rd'
-' J. Rett e r '- 'A . Aspect'
title:' On e -dimensional b ehavio ro fe l ong ate d Bose-Ein st ein c o ndensat e s' - - -
I ntroduction { #intro}
=== = === =====
T hee xperim ental s t udy of one dim ensi onal (1D) degen e rate Bo s e gases , wher e t heradi a lmo tio ni s“f r oz en” , is current ly a n imp orta n t d irec tio n of rese arch in ultra -co ld a t omphysi cs [@ gorl it z01;@schre ck01; @ greiner01]. Inunif orm 1D sy ste ms , f lu ctuat i ons of th e c ondensa te phas e ar ep r o no unced, because ofal a rg e popula tion o f l ow - lying st at es; asa resul t, n o l ong rang e orde r e xi sts atan y temp er atu re. In a trap ped 1D gas, th e fin i te size of the sample natura l ly i nt r oduc esa low-energ y cu t off, and at su f ficie ntlylo w t e mperature a phase c oh erentsampl e can exist ( see [@petr o v 1 d] and r efer e nc e s therein forthe p ropertieso f 1D gas es).Essentia lly, thes a me analy sis ho lds fo r th ree dimension a l (3D )trapped ga ses inver y e lon gat ed traps [@ petrov3d ]: h er ethe inte r action e ne rgy p erparti c le islarge r th an t h e r adial f r eq u e ncy, a nd the co nd ensat e wa v efu nctionis builtfro m sev er al radial modes. Still ,low-energy e xci tation s with a f requency smaller than t h e radia l f reque ncyare effec tiv ely on e-d i mensio nal, a nd in du cel a rge f l u ct uat io ns of thec o nde nsate p hase [@petr ov3d]. Such conden s ate s with a fluc tua ting p ha se, or [*[ qu a sic o n densates]{}*]{} , have bee no bs erved expe r ime nt ally in [@dett mer01 ; @hellw eg01; @sh varchuck0 2; @Ri c h ard 03; @Ertme r03].
W e present herea noverv iew of ou rexp erime ntal s t udy of e longat ed conde nsate sin a ver y anisotropic trap. Our set-u p, pr ese nted in s ect i on\[creatio n\]allows ustocha racte riz e prec isel y t hep roper ties of degene r at e g a s es between th e 3 D a nd 1D re g imes.In t his regime, we fi r st show (secti on \ [ e xp\ ])t hatth e (quasi)conde nsa te s exhibitth e same | Charles_Fabry de_l’Institut d’Optique, UMR 8501_du CNRS,_91403_Orsay Cedex,_France'
-_'Department of Physics,_60 St George_Street, University of Toronto,_Toronto, ON, M5S_1A7,_Canada'
author:
- 'P. Bouyer'
- 'J. H. Thywissen'
- 'F. Gerbier'
- 'M. Hugbart'
- 'S. Richard'
- 'J. Retter'
- 'A. Aspect'
title: 'One-dimensional behavior of elongated_Bose-Einstein_condensates'
---
Introduction {#intro}
============
The_experimental_study_of one dimensional (1D) degenerate_Bose gases, where the radial_motion is“frozen”,_is currently an important direction of research in_ultra-cold_atom physics [@gorlitz01;_@schreck01; @greiner01]. In uniform 1D systems, fluctuations of the_condensate phase are pronounced, because of_a large population_of_low-lying_states; as a result,_no long range order exists at_any temperature. In a trapped 1D_gas, the finite size of the sample_naturally introduces a low-energy cutoff, and_at sufficiently low temperature a_phase coherent_sample can exist (see [@petrov1d]_and references therein_for the_properties of 1D_gases). Essentially, the same analysis holds_for three dimensional_(3D) trapped gases in very elongated_traps_[@petrov3d]: here the_interaction_energy_per particle_is larger than_the_radial frequency,_and_the condensate wavefunction is built from_several_radial modes. Still, low-energy excitations with a_frequency smaller than the_radial_frequency are effectively one-dimensional,_and induce large fluctuations of_the condensate phase [@petrov3d]. Such condensates_with a_fluctuating phase,_or [*[quasicondensates]{}*]{}, have been observed experimentally in [@dettmer01; @hellweg01; @shvarchuck02; @Richard03;_@Ertmer03].
We present here an overview of_our experimental study of_elongated condensates_in_a very anisotropic_trap._Our set-up,_presented in section \[creation\] allows us to_characterize precisely_the properties of degenerate gases between_the 3D and 1D_regimes._In this regime, we first show_(section \[exp\]) that the (quasi)condensates exhibit_the same |
relation $(CR3)$ by:\
$(CR3')$ $\delta_s \sigma_1 \delta_r \delta_{s} \sigma_1 = \sigma_1 \delta_r\delta_{s} \sigma_1 \delta_s$\
for $1 \le r<s \le 2g+p-1$ with $(r,s)\neq (p+2i, p+2i+1)$, $0\leq i \leq g-1$.
\(i) Assume (a) $\delta\sigma\delta\delta'\sigma = \sigma\delta\delta'\sigma\delta$, (b) $\sigma\delta\sigma\delta = \delta\sigma\delta\sigma$ and (c) $\sigma\delta'\sigma\delta' = \delta'\sigma\delta'\sigma$. Then, $\delta'\sigma\delta\delta'\sigma = \delta^{-1}\sigma^{-1}\sigma\delta\delta'\sigma\delta\delta'\sigma =
\delta^{-1}\sigma^{-1}\delta\sigma\delta\delta'\sigma\delta'\sigma =
\delta^{-1}\sigma^{-1}\delta\sigma\delta\sigma\delta'\sigma\delta' =
\delta^{-1}\sigma^{-1} \sigma\delta\sigma\delta \delta'\sigma\delta' = \sigma\delta \delta'\sigma\delta'$.\
(ii) Assume (a) $\delta(\sigma\delta\delta'\sigma) = \sigma(\delta\delta'\sigma\delta)$, (b) $\delta'(\sigma\delta\delta'\sigma) = (\sigma\delta\delta'\sigma)\delta'$ and (c) $\sigma\delta'\sigma\delta' = \delta'\sigma\delta'\sigma$. Then (iii) is a consequence of (i). [$\Box$ ]{}
Since the relations of the presentation of ${B_n(\Sigma_{g,p})}$ are positive, one can define a monoid with the same presentation but as a monoid presentation. It is easy to see that the monoid we obtain doesnot inject in ${B_n(\Sigma_{g,p})}$, even if we add the relations of type $(CR3')$ to the presentation given in Theorem \[thm:presbg1\]. In fact the following relations, $$(CR3)_{k} \ \ \delta_ | relation $ (CR3)$ by:\
$ (CR3')$ $ \delta_s \sigma_1 \delta_r \delta_{s } \sigma_1 = \sigma_1 \delta_r\delta_{s } \sigma_1 \delta_s$\
for $ 1 \le r < s \le 2g+p-1 $ with $ (r, s)\neq (p+2i, p+2i+1)$, $ 0\leq i \leq g-1$.
\(i) Assume (a) $ \delta\sigma\delta\delta'\sigma = \sigma\delta\delta'\sigma\delta$, (b) $ \sigma\delta\sigma\delta = \delta\sigma\delta\sigma$ and (c) $ \sigma\delta'\sigma\delta' = \delta'\sigma\delta'\sigma$. Then, $ \delta'\sigma\delta\delta'\sigma = \delta^{-1}\sigma^{-1}\sigma\delta\delta'\sigma\delta\delta'\sigma =
\delta^{-1}\sigma^{-1}\delta\sigma\delta\delta'\sigma\delta'\sigma =
\delta^{-1}\sigma^{-1}\delta\sigma\delta\sigma\delta'\sigma\delta' =
\delta^{-1}\sigma^{-1 } \sigma\delta\sigma\delta \delta'\sigma\delta' = \sigma\delta \delta'\sigma\delta'$.\
(ii) Assume (a) $ \delta(\sigma\delta\delta'\sigma) = \sigma(\delta\delta'\sigma\delta)$, (b) $ \delta'(\sigma\delta\delta'\sigma) = (\sigma\delta\delta'\sigma)\delta'$ and (vitamin c) $ \sigma\delta'\sigma\delta' = \delta'\sigma\delta'\sigma$. Then (iii) is a consequence of (iodine). [ $ \Box$ ] { }
Since the relations of the presentation of $ { B_n(\Sigma_{g, p})}$ are positive, one can specify a monoid with the same presentation but as a monoid display. It is easy to see that the monoid we obtain doesnot inject in $ { B_n(\Sigma_{g, p})}$, even if we lend the relations of type $ (CR3')$ to the display given in Theorem \[thm: presbg1\ ]. In fact the following relations, $ $ (CR3)_{k } \ \ \delta _ | repation $(CR3)$ by:\
$(CR3')$ $\delta_s \sinma_1 \delta_r \delta_{s} \sigma_1 = \sigja_1 \delta_f\delta_{s} \sigma_1 \delta_s$\
for $1 \le c<s \lw 2g+p-1$ qith $(r,s)\neq (p+2i, p+2i+1)$, $0\leq i \leq g-1$.
\(i) Adsume (a) $\eelte\sigma\delta\delta'\sigma = \slyma\demba\delca'\wigma\delta$, (b) $\slgma\delta\sicma\delta = \deltd\skgla\delta\sigma$ and (c) $\sigma\delta'\sigma\dqlta' = \drlha'\sigma\delta'\sidma$. Uheg, $\demna'\wigma\delta\delta'\sigma = \delta^{-1}\sifma^{-1}\sigme\delta\delta'\sigms\delta\delta'\sigma =
\delta^{-1}\sigla^{-1}\depta\sigma\delta\delta'\digma\delta'\sugma =
\delta^{-1}\sigma^{-1}\ddlta\sigma\dtlca\sigma\deltz'\sigma\delta' =
\delta^{-1}\sigma^{-1} \sigma\deuta\siyma\delta \deotq'\sihka\delta' = \sijma\delna \delta'\sigma\delta'$.\
(ii) Dssume (s) $\delta(\sigma\dekta\veltq'\sigma) = \sigma(\delta\delva'\sigma\delta)$, (b) $\delta'(\figma\deltd\dzlta'\sigma) = (\sigma\delta\eeota'\sicma)\dalta'$ qnd (c) $\aijma\selta'\slgme\delta' = \delfa'\sigma\deltq'\sigma$. Then (iii) is s sinsequence of (i). [$\Box$ ]{}
Synce the relations of the presentation mf ${G_n(\Sigma_{g,p})}$ are positive, ine can define a monold with tre same presentation but as a monoid presentation. It ia eawy gi dee that the monoid we obtain doesnot inject yh ${N_n(\Xigma_{g,p})}$, even in we add the relatoojs jf type $(CR3')$ to the pxsssntation given in Hheorem \[thm:peesbg1\]. In sact the following relations, $$(CR3)_{j} \ \ \delta_ | relation $(CR3)$ by:\ $(CR3')$ $\delta_s \sigma_1 \delta_r = \delta_r\delta_{s} \sigma_1 for $1 \le (p+2i, $0\leq i \leq \(i) Assume (a) = \sigma\delta\delta'\sigma\delta$, (b) $\sigma\delta\sigma\delta = \delta\sigma\delta\sigma$ (c) $\sigma\delta'\sigma\delta' = \delta'\sigma\delta'\sigma$. Then, $\delta'\sigma\delta\delta'\sigma = \delta^{-1}\sigma^{-1}\sigma\delta\delta'\sigma\delta\delta'\sigma = \delta^{-1}\sigma^{-1}\delta\sigma\delta\delta'\sigma\delta'\sigma = \delta^{-1}\sigma^{-1}\delta\sigma\delta\sigma\delta'\sigma\delta' = \sigma\delta\sigma\delta \delta'\sigma\delta' = \sigma\delta \delta'\sigma\delta'$.\ (ii) Assume (a) $\delta(\sigma\delta\delta'\sigma) = \sigma(\delta\delta'\sigma\delta)$, (b) $\delta'(\sigma\delta\delta'\sigma) (\sigma\delta\delta'\sigma)\delta'$ (c) = Then (iii) is a consequence of (i). [$\Box$ ]{} Since the relations of the presentation of are positive, one can define a monoid with same presentation but as monoid presentation. It is easy see the monoid obtain inject ${B_n(\Sigma_{g,p})}$, even if add the relations of type $(CR3')$ to the presentation given in Theorem \[thm:presbg1\]. In fact the following $$(CR3)_{k} \ | relation $(CR3)$ by:\
$(CR3')$ $\delta_s \sigma_1 \Delta_r \deltA_{s} \sigMa_1 = \sIgmA_1 \dElta_R\delTa_{s} \sigma_1 \delta_s$\
FOr $1 \le R<s \le 2g+p-1$ with $(r,s)\neq (p+2i, p+2i+1)$, $0\leq I \leq g-1$.
\(I) ASSume (A) $\DeLta\siGma\deltA\DeLTA'\siGmA = \sIgmA\dELtA\deltA'\siGma\deltA$, (b) $\sigma\delTa\sIgMa\delta = \delta\SIgMa\delta\sigMa$ aNd (c) $\sigma\deltA'\siGma\delTa' = \DelTA'\sigmA\deLta'\siGma$. TheN, $\Delta'\sIgma\delta\DeLTa'\sigmA = \Delta^{-1}\siGMA^{-1}\sIgma\Delta\delta'\sigma\deLTa\DElta'\sigma =
\delta^{-1}\Sigma^{-1}\dElTA\sIGMa\dEltA\delta'\sigmA\dElta'\sIGma =
\deltA^{-1}\SiGMA^{-1}\DelTA\sigma\delta\siGma\delta'\sigMA\deLta' =
\delTa^{-1}\SigMA^{-1} \sigma\Delta\SiGMa\dElta \delta'\siGma\dElta' = \sigma\Delta \dELta'\sigmA\Delta'$.\
(ii) assume (A) $\deLta(\SigmA\DeLtA\deLtA'\SigMA) = \sIgmA(\DelTa\delta'\sIgMa\Delta)$, (B) $\delTA'(\SIGma\dEltA\delTa'\sigMa) = (\sigma\delta\dEltA'\sigMA)\deLta'$ anD (c) $\sigMa\deLtA'\sigmA\delta' = \Delta'\SiGma\delta'\sigma$. ThEn (iiI) is a conseQueNcE of (I). [$\BOx$ ]{}
SinCE the reLatIonS of the pResentaTIon Of ${b_N(\sIgMa_{g,p})}$ are positive, one CaN DEfIne a monoId with THe SaME presentAtIon But aS A MonoiD preSEnTation. It Is easy TO sEe That the MoNoid we ObTaiN doEsnot INjecT in ${B_n(\SIgma_{g,p})}$, evEn if wE Add the relationS Of type $(CR3')$ to the PReSENtATion GivEn in Theorem \[Thm:pREsbg1\]. in faCT tHe fOLlowiNg relAtIOnS, $$(cR3)_{k} \ \ \delta_ | relation $(CR3)$ by:\
$(C R3')$ $\de lta_s \s igm a_ 1 \d elta _r \delta_{s}\ sigm a_1 = \sigma_1 \delta _r\de lt a _{s} \s igma_ 1 \delt a _s $ \
fo r$1 \l er <s \le2g+ p-1$ wi th $(r,s)\ neq ( p+2i, p+2i+1 ) $, $0\leq i\le q g-1$.
\(i ) A ssume(a ) $ \ delta \si gma\d elta\d e lta'\s igma = \s ig m a\delt a \delta' \ s ig ma\d elta$, (b) $\sigm a \d e lta\sigma\delt a = \d el t a\ s i gma \de lta\sigma$ a nd (c ) $\sigm a \d e l t a'\ s igma\delta' = \delta'\si g ma\ delta' \s igm a $. The n, $\ de l ta' \sigma\delt a\de lta'\sigm a = \d e lta^{-1 } \sigma^ {-1}\s igm a\d elta \ de lt a'\ si g ma\ d el ta\ d elt a'\sigma = \delt a^{- 1 } \ s igma ^{- 1}\d elta\ sigma\delta\d elt a'\s i gma \delt a'\si gma=
\del ta^{-1 }\sig ma ^{-1}\delta\sig ma\d elta\sigm a\d el ta' \s igma\ d elta'= \de lta^{-1 }\sigma ^ {-1 }\ s i gm a\delta\sigma\delt a\ d el ta'\sigm a\delt a '=\ sigma\de lt a \ delt a ' \sigm a\de l ta '$.\
(ii ) Assu m e(a ) $\del ta (\sigm a\ del ta\ delta ' \sig ma) =\sigma(\ delta \ delta'\sigma\d e lta)$, (b) $\ d el t a '( \ sigm a\d elta\delta' \sig m a) = (\s i gm a\d e lta\d elta' \s i gm a )\delta'$ and (c) $ \s igma\d elta' \sigma\delta' = \delta' \ s i gma\delt a'\s i gm a $. Then (iii)is aconsequenc e of (i). [$\B ox$ ]{}
Since th e relation s o f t hepre s e nt ation of ${B_ n ( \Sig ma _{g,p}) }$are pos iti ve, on e c an define a monoidwi th t he sa me pr e sentatio nbut a s a mono i d pres entat ion. I ti s e asy tos ee t hatth emono idwe obta in d o esn ot inje ct in ${B _n( \ Sigm a_ {g ,p})}$, even if we a dd the relat io nsof typ e $(CR3')$ to the presentation gi v en in T heo rem \ [thm :presbg1\ ].In fac t t h e foll owingrelat io ns, $ $(CR3 ) _ {k } \ \ \delta_ | relation_$(CR3)$ by:\
$(CR3')$_$\delta_s \sigma_1 \delta_r \delta_{s}_\sigma_1 =__\sigma_1 \delta_r\delta_{s}_\sigma_1_\delta_s$\
for $1 \le_r<s \le 2g+p-1$_with $(r,s)\neq (p+2i, p+2i+1)$,_$0\leq i \leq_g-1$.
\(i)_Assume (a) $\delta\sigma\delta\delta'\sigma = \sigma\delta\delta'\sigma\delta$, (b) $\sigma\delta\sigma\delta = \delta\sigma\delta\sigma$ and (c) $\sigma\delta'\sigma\delta' = \delta'\sigma\delta'\sigma$._Then,_$\delta'\sigma\delta\delta'\sigma =_\delta^{-1}\sigma^{-1}\sigma\delta\delta'\sigma\delta\delta'\sigma_=_
\delta^{-1}\sigma^{-1}\delta\sigma\delta\delta'\sigma\delta'\sigma =
\delta^{-1}\sigma^{-1}\delta\sigma\delta\sigma\delta'\sigma\delta' =
\delta^{-1}\sigma^{-1}_\sigma\delta\sigma\delta \delta'\sigma\delta' = \sigma\delta \delta'\sigma\delta'$.\
(ii)_Assume (a)_$\delta(\sigma\delta\delta'\sigma) = \sigma(\delta\delta'\sigma\delta)$, (b) $\delta'(\sigma\delta\delta'\sigma) = (\sigma\delta\delta'\sigma)\delta'$ and_(c)_$\sigma\delta'\sigma\delta' = \delta'\sigma\delta'\sigma$._Then (iii) is a consequence of (i). [$\Box$ ]{}
Since_the relations of the presentation of_${B_n(\Sigma_{g,p})}$ are positive,_one_can_define a monoid with_the same presentation but as a_monoid presentation. It is easy to_see that the monoid we obtain doesnot_inject in ${B_n(\Sigma_{g,p})}$, even if we_add the relations of type_$(CR3')$ to_the presentation given in Theorem_\[thm:presbg1\]. In fact_the following_relations, $$(CR3)_{k} \_\ \delta_ |
to $N_{2}/K$.
We will focus on the first term of. Hölder inequality gives $$\label{baseindcution}
\int |Ef_{s_1}Ef_{s_{1}'}Ef_{s_{2}}Ef_{s_{2}'}|w_{B_{N_{1}^{2}}}\leq \prod_{j=1}^{2}\left(\int |Ef_{s_{j}}Ef_{s_{j}'}|^2w_{B_{N_{1}^{2}}}\right)^{1/2}.$$ Estimate is the start point of the analysis in the following Subsections.
We summarize in the lemma below how and come together to highlight the relevance of $A(N_{1}, N_{2}, \lambda)$ in the induction procedure.
\[finalinductionlemma\] When $\lambda \leq N_{1}/N_{2}$ and $\lambda \leq N_{1}$, we have $$\label{fikey}
K(N_{1}, N_{2}, \lambda)\lesssim K^{C}\frac{1}{\lambda}A(N_{1}, N_{2},\lambda)+K(N_{1}, N_{2}/K, \lambda)$$
Note that the assumption of Lemma \[finalinductionlemma\] always holds during the induction procedure to prove Lemma \[criticallll\].
Applying Lemma \[tansferlemma\], we have $$\label{ssd2}
\begin{aligned}
& \|Ef_{s_j} Ef_{s'_j}\|_{L^2_{avg}(w_{B_{N_{1}^{2}}})}\\
\lesssim
&K^{C}A(N_{1}, N_{2}, \lambda)\left(\sum_{|\theta|=\frac{1}{\lambda N_1}}\|Ef_{s_j,\theta}\|_{L^4_{avg}(w_{B_{N_{1}^{2}}})}^2\right)^{1/2}\left(\sum_{|\theta|=\frac{1}{\lambda N_1}}\|Ef_{s'_j,\theta}\|_{L^4_{avg}(w_{B_{N_{1}^{2}}})}^2\right)^{1/2}.
\end{aligned}$$
Plugging into, and then plugging into, we derive $$\label{finalinduce2}
\ | to $ N_{2}/K$.
We will focus on the first term of. Hölder inequality gives $ $ \label{baseindcution }
\int |Ef_{s_1}Ef_{s_{1}'}Ef_{s_{2}}Ef_{s_{2}'}|w_{B_{N_{1}^{2}}}\leq \prod_{j=1}^{2}\left(\int |Ef_{s_{j}}Ef_{s_{j}'}|^2w_{B_{N_{1}^{2}}}\right)^{1/2}.$$ Estimate is the beginning item of the analysis in the following Subsections.
We summarize in the lemma below how and issue forth together to highlight the relevance of $ A(N_{1 }, N_{2 }, \lambda)$ in the trigger routine.
\[finalinductionlemma\ ] When $ \lambda \leq N_{1}/N_{2}$ and $ \lambda \leq N_{1}$, we have $ $ \label{fikey }
K(N_{1 }, N_{2 }, \lambda)\lesssim K^{C}\frac{1}{\lambda}A(N_{1 }, N_{2},\lambda)+K(N_{1 }, N_{2}/K, \lambda)$$
notice that the premise of Lemma \[finalinductionlemma\ ] always hold during the induction procedure to test Lemma \[criticallll\ ].
practice Lemma \[tansferlemma\ ], we have $ $ \label{ssd2 }
\begin{aligned }
& \|Ef_{s_j } Ef_{s'_j}\|_{L^2_{avg}(w_{B_{N_{1}^{2}}})}\\
\lesssim
& K^{C}A(N_{1 }, N_{2 }, \lambda)\left(\sum_{|\theta|=\frac{1}{\lambda N_1}}\|Ef_{s_j,\theta}\|_{L^4_{avg}(w_{B_{N_{1}^{2}}})}^2\right)^{1/2}\left(\sum_{|\theta|=\frac{1}{\lambda N_1}}\|Ef_{s'_j,\theta}\|_{L^4_{avg}(w_{B_{N_{1}^{2}}})}^2\right)^{1/2 }.
\end{aligned}$$
Plugging into, and then plugging into, we derive $ $ \label{finalinduce2 }
\ | to $N_{2}/K$.
We will focus on the nirst term of. Hölbwr inexualitg gives $$\uabel{baseindcution}
\int |Ef_{s_1}Ef_{s_{1}'}Xf_{s_{2}}Ed_{s_{2}'}|w_{B_{N_{1}^{2}}}\oeq \prod_{j=1}^{2}\left(\int |Ef_{s_{j}}Df_{s_{j}'}|^2w_{B_{N_{1}^{2}}}\rihht)^{1/2}.$$ Estinate us the stacf point of ths ananbsis in the folkowing Subvections.
We sumkafive in the lemma below how and come tjgether tl highlight thg reltvagce kf $A(N_{1}, N_{2}, \lambda)$ in the induction prkcedure.
\[hinalinductionlrmma\] When $\lambda \leq N_{1}/N_{2}$ anf $\lalbda \leq N_{1}$, we have $$\label{fikey}
J(N_{1}, N_{2}, \oambda)\lesssio K^{C}\frac{1}{\lambda}A(N_{1}, N_{2},\lambsa)+K(N_{1}, N_{2}/K, \lambda)$$
Note that the assjmptipn of Lemmq \[dinwninductionlxmma\] ajways holds during tve induvtion procedurc to 'rovw Lemma \[criticallll\].
Ap'lying Lemma \[tansferjemma\], we vare $$\label{ssd2}
\begin{alignwd}
& \|Ef_{s_j} Ef_{s'_b}\|_{L^2_{ave}(q_{B_{N_{1}^{2}}})}\\
\ltssxij
&K^{C}A(J_{1}, N_{2}, \lambda)\leff(\sum_{|\theta|=\frqc{1}{\lambda N_1}}\|Ef_{s_j,\theta}\|_{K^4_{adt}(w_{B_{N_{1}^{2}}})}^2\right)^{1/2}\left(\aum_{|\theea|=\srac{1}{\lambda N_1}}\|Ef_{s'_j,\theta}\|_{L^4_{avg}(w_{B_{N_{1}^{2}}})}^2\right)^{1/2}.
\end{alpgnes}$$
Plugging into, and then plugging into, we derlve $$\label{sinalinduce2}
\ | to $N_{2}/K$. We will focus on the of. inequality gives \int |Ef_{s_1}Ef_{s_{1}'}Ef_{s_{2}}Ef_{s_{2}'}|w_{B_{N_{1}^{2}}}\leq \prod_{j=1}^{2}\left(\int point the analysis in following Subsections. We in the lemma below how and together to highlight the relevance of $A(N_{1}, N_{2}, \lambda)$ in the induction procedure. When $\lambda \leq N_{1}/N_{2}$ and $\lambda \leq N_{1}$, we have $$\label{fikey} K(N_{1}, N_{2}, K^{C}\frac{1}{\lambda}A(N_{1}, N_{2}/K, Note the assumption of Lemma \[finalinductionlemma\] always holds during the induction procedure to prove Lemma \[criticallll\]. Applying \[tansferlemma\], we have $$\label{ssd2} \begin{aligned} & \|Ef_{s_j} Ef_{s'_j}\|_{L^2_{avg}(w_{B_{N_{1}^{2}}})}\\ &K^{C}A(N_{1}, N_{2}, \lambda)\left(\sum_{|\theta|=\frac{1}{\lambda N_1}}\|Ef_{s_j,\theta}\|_{L^4_{avg}(w_{B_{N_{1}^{2}}})}^2\right)^{1/2}\left(\sum_{|\theta|=\frac{1}{\lambda \end{aligned}$$ Plugging into, and then into, derive $$\label{finalinduce2} | to $N_{2}/K$.
We will focus on the first Term of. HöldEr ineQuaLitY gIves $$\LabeL{baseindcution}
\INt |Ef_{S_1}Ef_{s_{1}'}Ef_{s_{2}}Ef_{s_{2}'}|w_{B_{N_{1}^{2}}}\leq \prod_{j=1}^{2}\lEft(\inT |EF_{S_{j}}Ef_{S_{J}'}|^2w_{b_{N_{1}^{2}}}\rigHt)^{1/2}.$$ EstimATe IS The StArT poInT Of The anAlySis in thE following subSeCtions.
We summARiZe in the lemMa bElow how and coMe tOgetheR tO hiGHlighT thE releVance oF $a(N_{1}, N_{2}, \lamBda)$ in the iNdUCtion pROcedure.
\[FINaLindUctionlemma\] When $\laMBdA \Leq N_{1}/N_{2}$ and $\lambda \Leq N_{1}$, we HaVE $$\lABEl{fIkeY}
K(N_{1}, N_{2}, \lambda)\LeSssim k^{c}\frac{1}{\laMBdA}a(n_{1}, n_{2},\laMBda)+K(N_{1}, N_{2}/K, \lambda)$$
note that the ASsuMption Of lemMA \[finalInducTiONleMma\] always hoLds dUring the iNductiON procedURe to proVe LemmA \[crItiCallLL\].
APpLyiNg lEmmA \[TaNsfERleMma\], we havE $$\lAbEl{ssd2}
\BegiN{ALIGned}
& \|ef_{s_J} Ef_{s'_J}\|_{L^2_{avg}(W_{B_{N_{1}^{2}}})}\\
\lesssim
&K^{C}A(n_{1}, N_{2}, \lAmbdA)\LefT(\sum_{|\tHeta|=\fRac{1}{\lAmBda N_1}}\|EF_{s_j,\theTa}\|_{L^4_{avG}(w_{b_{N_{1}^{2}}})}^2\right)^{1/2}\left(\sum_{|\tHeta|=\Frac{1}{\lambdA N_1}}\|EF_{s'_J,\thEtA}\|_{L^4_{avg}(W_{b_{N_{1}^{2}}})}^2\righT)^{1/2}.
\enD{alIgned}$$
PlUgging iNTo, aNd THEN pLugging into, we derivE $$\lABEl{FinalindUce2}
\ | to $N_{2}/K$.
We will fo cus on the firs t t erm o f. H ölde r inequality g i ves$$\label{baseindcution }
\in t| Ef_{ s _1 }Ef_{ s_{1}'} E f_ { s _{2 }} Ef _{s _{ 2 }' }|w_{ B_{ N_{1}^{ 2}}}\leq \ pro d_ {j=1}^{2}\le f t( \int |Ef_ {s_ {j}}Ef_{s_{j }'} |^2w_{ B_ {N_ { 1}^{2 }}} \righ t)^{1/ 2 }.$$ E stimate i st he sta r t point o ftheanalysis in the f o ll o wing Subsectio ns.
W es um m a riz e i n the lemm abelow how and co m e tog e ther to highl ight the re l eva nce of $ A(N _ {1}, N _{2}, \ l amb da)$ in the ind uction pr ocedur e .
\[fi n alinduc tionle mma \]When $\ la mbd a\ leq N_ {1} / N_{ 2}$ and$\ la mbda\leq N _ { 1}$, we hav e $$\ label{fikey}K(N _{1} , N_ {2},\lamb da)\ le sssim K^{C} \frac {1 }{\lambda}A(N_{ 1},N_{2},\la mbd a) +K( N_ {1},N _{2}/K , \ lam bda)$$
Note t h atth e a ss umption of Lemma \ [f i n al inductio nlemma \ ]al w ays hold sdur ingt h e ind ucti o nprocedur e to p r ov eLemma \ [c ritica ll ll\ ].
Appl y ingLemma\[tansfe rlemm a \], we have $$ \ label{ssd2}
\ b eg i n {a l igne d}& \|Ef_{s_j } Ef _ {s'_ j}\| _ {L ^2_ { avg}( w_{B_ {N _ {1 } ^{2}}})}\\
\lesss im
&K^ {C}A( N_{1}, N_{2}, \lambda)\ l e f t(\sum_{ |\th e ta | =\frac{1}{\lam bda N _1}}\|Ef_{ s _j,\thet a}\|_ {L^4_{av g}(w_{B_{ N _ {1}^{2}} })} ^2\ rig ht) ^ { 1/ 2}\left(\sum_ { | \the ta |=\frac {1} {\lambd a N _1} }\| Ef_ {s '_j,\thet a}\|_{L^ 4_ {a vg }( w_{ B_{N_ { 1}^{2}}} )} ^2\ ri ght )^{1/ 2 }.
\e nd{al igne d} $$
Pl uggingi nt o , and t he n pl ugg in g int o, w e de rive $$ \label{fi nal i nduc e2 } \ | to_$N_{2}/K$.
We will_focus on the first_term of._Hölder_inequality gives_$$\label{baseindcution}
\int_|Ef_{s_1}Ef_{s_{1}'}Ef_{s_{2}}Ef_{s_{2}'}|w_{B_{N_{1}^{2}}}\leq \prod_{j=1}^{2}\left(\int _|Ef_{s_{j}}Ef_{s_{j}'}|^2w_{B_{N_{1}^{2}}}\right)^{1/2}.$$ Estimate is_the start point of_the analysis in_the_following Subsections.
We summarize in the lemma below how and come together to highlight the_relevance_of $A(N_{1},_N_{2},_\lambda)$_in the induction procedure.
\[finalinductionlemma\] When_$\lambda \leq N_{1}/N_{2}$ and $\lambda_\leq N_{1}$,_we have $$\label{fikey}
K(N_{1}, N_{2}, \lambda)\lesssim K^{C}\frac{1}{\lambda}A(N_{1}, N_{2},\lambda)+K(N_{1}, N_{2}/K,_\lambda)$$
Note_that the assumption_of Lemma \[finalinductionlemma\] always holds during the induction procedure_to prove Lemma \[criticallll\].
Applying Lemma \[tansferlemma\],_we have $$\label{ssd2}
\begin{aligned}
&_\|Ef_{s_j}_Ef_{s'_j}\|_{L^2_{avg}(w_{B_{N_{1}^{2}}})}\\
_ \lesssim
&K^{C}A(N_{1},_N_{2}, \lambda)\left(\sum_{|\theta|=\frac{1}{\lambda N_1}}\|Ef_{s_j,\theta}\|_{L^4_{avg}(w_{B_{N_{1}^{2}}})}^2\right)^{1/2}\left(\sum_{|\theta|=\frac{1}{\lambda N_1}}\|Ef_{s'_j,\theta}\|_{L^4_{avg}(w_{B_{N_{1}^{2}}})}^2\right)^{1/2}.
\end{aligned}$$
Plugging into,_and then plugging into, we derive_$$\label{finalinduce2}
\ |
: Stellar Rotation at 550 Myr'
---
Introduction
============
This paper is the third in a series of papers on a deep survey for transiting planets in the open cluster M37 (NGC 2099) using the MMT. In the first paper [@Hartman.07a Paper I] we introduced the survey, described the spectroscopic and photometric observations and the data reduction, determined the fundamental parameters (age, metallicity, distance and reddening) of the cluster, and obtained its mass and luminosity functions and radial density profile. In the second paper [@Hartman.07b Paper II] we identified 1445 variable stars in the field of the cluster, of which 99$\%$ were new discoveries. We found that $\ga 500$ of these variables lie near the cluster main sequence on photometric color-magnitude diagrams (CMDs) and show a correlation between period and magnitude. In this paper we investigate this population of variables arguing that the photometric variations are due to spotted star rotation. We use these variables to study the rotation of F-M main sequence stars at the age of the cluster ($550 \pm 30$ Myr, Paper I; note that for the rest of this paper we adopt this age which was derived by comparison to models with convective overshooting).
Rotation plays an important role in the life of a star. Empirically there is a strong relation between the rotation rate and the activity and age of low-mass stars [@Skumanich.72], with older stars being slower rotators and less active than younger stars. Rotation affects the structure of a star [see @Sills.00] and its evolution may have important consequences for mixing at the base of a star’s convection zone [see the review by @Pinsonneault.97]. Rotation also affects the spectral energy distribution of low-mass stars [@Stauffer.03]. It has even been suggested that rotation may be used as a tool to determine the ages of field stars older than a few hundred million years, and that ages determined in this fashion may be significantly more accurate than ages determined using other methods [@Barnes.07]. However, the usefulness of these “gyrochronology” ages is predicated on an accurate understanding and calibration of the age-rotation rate relation. As we discuss next, there are relatively few constraints on this relation for ages $\ga 200~{\rm Myr}$.
The surface rotation period of a star can be measured directly from photometric variations if it has | : Stellar Rotation at 550 Myr'
---
Introduction
= = = = = = = = = = = =
This paper is the third in a series of papers on a cryptic sketch for transiting planets in the open bunch M37 (NGC 2099) using the MMT. In the first paper [ @Hartman.07a Paper I ] we introduce the survey, described the spectroscopic and photometric observation and the data reduction, specify the fundamental parameters (long time, metallicity, distance and reddening) of the bunch, and obtained its mass and luminosity function and radial density profile. In the second composition [ @Hartman.07b Paper II ] we identified 1445 variable stars in the playing field of the cluster, of which 99$\%$ were new discoveries. We found that $ \ga 500 $ of these variable lie near the cluster main sequence on photometric color - order of magnitude diagrams (CMDs) and show a correlation between period and magnitude. In this paper we investigate this population of variable star arguing that the photometric variations are due to spotted star rotation. We use these variables to study the rotation of F - M main succession star at the age of the bunch ($ 550 \pm 30 $ Myr, Paper I; notice that for the rest of this paper we adopt this age which was derive by comparison to model with convective overshooting).
Rotation plays an important function in the life of a star. Empirically there is a strong relation between the rotation rate and the bodily process and age of low - mass stars [ @Skumanich.72 ], with old star being slower rotators and less active than younger stars. Rotation affects the structure of a asterisk [ see @Sills.00 ] and its evolution may have authoritative consequences for mixing at the base of a star ’s convection zone [ experience the review by @Pinsonneault.97 ]. Rotation besides affects the spectral energy distribution of low - mass star [ @Stauffer.03 ]. It has even been suggested that rotation may be practice as a tool to settle the ages of field star older than a few hundred million old age, and that ages determine in this fashion may be significantly more accurate than ages determined using early methods [ @Barnes.07 ]. However, the usefulness of these “ gyrochronology ” old age is predicated on an accurate understanding and calibration of the age - rotation rate relation. As we discuss next, there are relatively few constraint on this relation for ages $ \ga 200~{\rm Myr}$.
The surface rotation period of a star can be measured directly from photometric variation if it experience | : Stfllar Rotation at 550 Myr'
---
Inuroduction
============
This pakee is tie thirs in a sdries of papers on a deep sucvey for ugansiting planets in ghe open bluster M37 (NGC 2099) using the MMT. In bke fiddt pcpxr [@Hartman.07a Papgr I] we intrmduced the sureeh, bescribed the spectroscopic and photjmetric ohservations anq tht dwta dvdmction, determined the fundamentam paramtters (age, metallicoty, distance and reddening) of hhe cluster, and obhained its nass qnd luminosigy functions and radiam density profile. In the second papex [@Hartman.07b Kcper HI] we identmfied 1445 variable stars in tve fielc of the clustcr, of whuch 99$\%$ were new discovecies. We found that $\gw 500$ of theve variables lie neqr the wlusder oqin sesuxncs on pjotkmetric comor-magnitudw diagrams (CMDs) and sriw a correlatjon beewqen period and magnitude. In this paper fe jnvestigate this popularion of variables argoing that ehe photometric variations are due to spotted stas rotetkon. Wc usd tjese variables to study the rotation of F-M mayh xeauence stars ab the age of the ckudtrt ($550 \pm 30$ Myr, Papgr I; nocs fhat for the rest lf this papee we adopu thix age which was derived by xomparison tj models with conveetive overshuotimg).
Rotstion plays an importanc role in the liff of a stzf. Empirically thdre iv a strong relation betweeg the rotetion rate avd tne actyvity and wge on low-mass stars [@Skulanick.72], witv older stwrs being slower rotators and lxxs active thsn yolnger staxs. Rotstion affects the structure of a scar [sed @Sills.00] ans its etolution may have importatj consequencev for miving at rhe basd of a star’s comvection doue [see thw review by @Pinsonkeaulj.97]. Dotation also ayyexts the spectrak everdy dmstrifgtion of low-kass stxts [@Stxuffer.03]. It hcs evem been suggested thad rofation may be used af a tool to detewmine the agex of field stars opder vhan a few nunqred million years, and that agss determlneq in this fafhiok mai be signifncantly more accurate than ages determinxd using other methods [@Varnes.07]. However, the osenulness of tiese “grrochrononogy” ages is predicared on an accuratt understanding and caljbratimn of the age-rotation rate relation. As we discuss next, there are relatively feq consvrwints on thjs rrlatimn yor agef $\ga 200~{\cm Myr}$.
The surface rotation period of a star can ue measureg birectly from photometric varoagions if it hxs | : Stellar Rotation at 550 Myr' --- This is the in a series survey transiting planets in open cluster M37 2099) using the MMT. In the paper [@Hartman.07a Paper I] we introduced the survey, described the spectroscopic and photometric and the data reduction, determined the fundamental parameters (age, metallicity, distance and reddening) the and its and luminosity functions and radial density profile. In the second paper [@Hartman.07b Paper II] we identified variable stars in the field of the cluster, which 99$\%$ were new We found that $\ga 500$ these lie near cluster sequence photometric color-magnitude diagrams and show a correlation between period and magnitude. In this paper we investigate this population of variables that the are due spotted rotation. use these variables the rotation of F-M main sequence age of the cluster ($550 \pm 30$ Myr, I; note for the rest of this paper adopt this age which was derived by comparison models with convective overshooting). Rotation plays an important role in the life of a star. is a strong relation the rotation rate the and of stars [@Skumanich.72], older stars being slower rotators and less active than younger stars. affects the structure of a star [see @Sills.00] and its have consequences for mixing the base of a convection [see the review by also the of stars It has even been that rotation may be used a tool to determine older than a few hundred million years, and ages determined in this fashion may be more accurate than ages determined using other methods [@Barnes.07]. However, the usefulness these “gyrochronology” predicated on an accurate understanding and calibration of age-rotation rate relation. As discuss next, there are relatively few constraints on this for $\ga 200~{\rm The surface rotation of a star be measured directly variations if has | : Stellar Rotation at 550 Myr'
---
IntroDuction
============
ThiS papeR is The ThIrd iN a seRies of papers on A Deep Survey for transiting plaNets iN tHE opeN ClUster m37 (NGC 2099) usiNG tHE mMT. in ThE fiRsT PaPer [@HaRtmAn.07a PapeR I] we introdUceD tHe survey, descRIbEd the spectRosCopic and photOmeTric obSeRvaTIons aNd tHe datA reducTIon, detErmined thE fUNdamenTAl paramETErS (age, Metallicity, distanCE aND reddening) of thE clustEr, ANd OBTaiNed Its mass and LuMinosITy functIOnS AND raDIal density proFile. In the seCOnd Paper [@HArTmaN.07B Paper iI] we iDeNTifIed 1445 variable StarS in the fieLd of thE Cluster, OF which 99$\%$ wEre new DisCovErieS. we FoUnd ThAT $\ga 500$ OF tHesE VarIables liE nEaR the cLustER MAIn seQueNce oN photOmetric color-mAgnItudE DiaGrams (cMDs) aNd shOw A corrElatioN betwEeN period and magniTude. in this papEr wE iNveStIgate THis popUlaTioN of variAbles arGUinG tHAT ThE photometric variatIoNS ArE due to spOtted sTAr RoTAtion. We uSe TheSe vaRIAbles To stUDy The rotatIon of F-m MaIn SequencE sTars at ThE agE of The clUSter ($550 \Pm 30$ Myr, PAper I; notE that FOr the rest of thiS Paper we adopt tHIs AGE wHIch wAs dErived by comPariSOn to ModeLS wIth COnvecTive oVeRShOOting).
Rotation plays aN iMportaNt rolE in the life of a Star. EmpiriCALLy there iS a stROnG Relation betweeN the rOtation ratE And the acTivitY and age oF low-mass sTARs [@SkumanIch.72], WitH olDer STArS being slower rOTAtorS aNd less aCtiVe than yOunGer StaRs. ROtAtion affeCts the stRuCtUrE oF a sTar [seE @sills.00] and ItS evOlUtiOn may HAve impOrtanT conSeQuENceS for mixINg AT The bAsE oF a stAr’s CoNvectIon zONe [sEe the reView by @PinSonNEaulT.97]. ROtAtion alSo affects the sPeCtral energY dIstRibutiON Of low-masS stars [@Stauffer.03]. It has even BEen suggEstEd thaT rotAtion may bE usEd as a tOol TO deterMine thE ages Of FieLD Stars OLDeR thAn A few hundreD MIllIon yeArS, and That ageS determined in this fAShiOn may be signifIcaNtly MORe AccURaTE thAn AGes DETermined using otHer methods [@baRNeS.07]. However, thE UseFuLness of These “gyRochrONology” aGes is predIcated on aN aCcurATE unDerstandinG and caliBration of THe age-ROtAtion RatE relatIoN. As We disCuss neXT, thEre arE relatIvEly few ConstRaInts on thIs relation for ages $\ga 200~{\rm MyR}$.
The suRface RotAtion periOd oF A stAr can be meAsurEd directly FroM phOtomeTriC VariaTionS If It hAS | : Stellar Rotation at 550Myr'
---
Intro duc tio n==== ==== ====
This pap e r is the third in a series of p ap e rs o n a deep survey fo r tra ns it ing p l an ets i n t he open cluster M 37(N GC 2099) usi n gthe MMT. I n t he first pap er[@Hart ma n.0 7 a Pap erI] we intro d uced t he survey ,d escrib e d the s p e ct rosc opic and photomet r ic observations a nd the d a ta r edu cti on, determ in ed th e fundam e nt a l par a meters (age,metallicity , di stance a ndr eddeni ng) o ft hecluster, an d ob tained it s mass and lum i nosityfuncti ons an d ra d ia lden si t y p r of ile . In the sec on dpaper [@H a r t m an.0 7bPape r II] we identifie d 1 445v ari ablestars inth e fie ld ofthe c lu ster, of which99$\ %$ were n ewdi sco ve ries. We fou ndtha t $\ga500$ of the se v a ri ables lie near the c l u st er mainsequen c eon photomet ri c c olor - m agnit uded ia grams (C MDs) a n dsh ow a co rr elatio nbet wee n per i od a nd mag nitude.In th i s paper we inv e stigate thisp op u l at i on o f v ariables ar guin g tha t th e p hot o metri c var ia t io n s are due to spotte dstar r otati on. We use th ese variab l e s to stud y th e r o tation of F-Mmainsequence s t ars at t he ag e of the cluster( $ 550 \pm30$ My r,Pap e r I ; note that f o r the r est ofthi s paper we ad opt th is age whic h was de ri ve dby co mpari s on to mo de lswi thconve c tive o versh ooti ng ).
Ro tationp la y s anim po rtan t r ol e inthel ife of a s tar. Empi ric a llyth er e is astrong relati on between t he ro tation r ate andthe activity and age of low-mas s s tars[@Sk umanich.7 2], withold e r star s bein g slo we r r o t ators a nd le ss active th a n yo unger s tars . Rotat ion affects the st r uct ure of a star [s ee @ S i ll s.0 0 ]a ndit s ev o l ution may haveimportantco n se quences fo r mi xi ng at t he base of a star’sconvectio n zone [s ee the r evi ew by @Pin sonneaul t.97]. Ro t ation al so af fec ts the s pec tralenergy dis tribu tion o flow-ma ss st ar s [@Stau ffer.03]. It has even b een su ggest edthat rota tio n ma y be used asa tool todet erm ine t hea ges o f fi e ld st a rs ol dert han a few hu ndr e d m illion year s , and that ag e s dete rmin ed in this fashio n may be signif ican t l y m ore accu ra te than ages d ete rm i n ed using o ther method s [@Barn es . 07].Howeve r, the useful n e ss of the se “ gyr ochronolo gy” a g es is p re di c ated o n an a ccurat e unde r stan d i ng and calibrati on of t he ag e -ro tatio nrate re l atio n. As we d iscuss next , ther e ar e rel atively f ew con str ai nts on thi s relation forages $\ ga 200 ~{\ rm Myr }$.T he su rfac erot ation per i o do fas tar can be m ea sure d directl y from ph oto m etric v ar iat i o ns ifi th a s | : Stellar_Rotation at_550 Myr'
---
Introduction
============
This paper is_the third_in_a series_of_papers on a_deep survey for_transiting planets in the_open cluster M37_(NGC_2099) using the MMT. In the first paper [@Hartman.07a Paper I] we introduced the_survey,_described the_spectroscopic_and_photometric observations and the data_reduction, determined the fundamental parameters_(age, metallicity,_distance and reddening) of the cluster, and obtained_its_mass and luminosity_functions and radial density profile. In the second paper_[@Hartman.07b Paper II] we identified 1445_variable stars in_the_field_of the cluster, of_which 99$\%$ were new discoveries. We_found that $\ga 500$ of these_variables lie near the cluster main sequence_on photometric color-magnitude diagrams (CMDs) and_show a correlation between period_and magnitude._In this paper we investigate_this population of_variables arguing_that the photometric_variations are due to spotted star_rotation. We use_these variables to study the rotation_of_F-M main sequence_stars_at_the age_of the cluster_($550_\pm 30$_Myr,_Paper I; note that for the_rest_of this paper we adopt this age_which was derived by_comparison_to models with convective_overshooting).
Rotation plays an important role_in the life of a star._Empirically there_is a_strong relation between the rotation rate and the activity and age_of low-mass stars [@Skumanich.72], with older_stars being slower rotators_and less_active_than younger stars._Rotation_affects the_structure of a star [see @Sills.00] and_its evolution_may have important consequences for mixing_at the base of_a_star’s convection zone [see the review_by @Pinsonneault.97]. Rotation also affects the_spectral energy distribution of low-mass_stars_[@Stauffer.03]._It has even been suggested_that rotation may be used as_a tool to_determine the ages of field stars older_than_a few hundred million years, and_that_ages determined in this fashion may_be_significantly_more accurate than ages determined_using other methods [@Barnes.07]. However, the_usefulness of these “gyrochronology” ages is predicated on an_accurate understanding and_calibration of the age-rotation rate_relation._As_we discuss next, there are relatively few constraints on this_relation for_ages $\ga 200~{\rm_Myr}$.
The surface rotation period of a star can be measured_directly from photometric variations if it has |
thaut and M. Timme, Nonlocal effects and countermeasures in cascading failures, Phys. Rev. E **92**, 032809 (2015).
I. Dobson, B. A. Carreras, D. E. Newman, and J. M. Reynolds-Barredo, Obtaining statistics of cascading line outages spreading in an electric transmission network from standard utility data, IEEE T. Power Syst. **99**, 1 (2016).
M. Anghel, K. A. Werley, and A. E. Motter, Stochastic model for power grid dynamics, [*Proc. 40th Int. Conf. Syst. Sci.*]{} HICSS’07, Big Island, HI, USA, Vol. 1, 113 (2007).
The data for the snapshots are obtained from Federal Energy Regulatory Commission (FERC) Form 715.
P. Hines, E. Cotilla-Sanchez, and S. Blumsack, Topological models and critical slowing down: Two approaches to power system blackout risk analysis, [*Proc. 44th Int. Conf. Syst. Sci.*]{} HICSS’11, Kauai, HI, USA, 1 (2011).
Source code is available for download at <https://github.com/yangyangangela/determine_cascade_sizes>
B. A. Carreras, D. E. Newman, I. Dobson, and N. S. Degala, Validating OPA with WECC data, [*Proc. 46th Int. Conf. Syst. Sci.*]{} HICSS’13, Maui, HI, USA, 2197 (2013).
P. Henneaux, P. E. Labeau, J. C. Maun, and L. Haarla, A two-level probabilistic risk assessment of cascading outages, IEEE T. Power Syst. **31**, 2393 (2016).
P. D. Hines, I. Dobson, and P. Rezaei, Cascading power outages propagate locally in an influence graph that is not the actual grid topology, arXiv:1508.01775 ( | thaut and M. Timme, Nonlocal effects and countermeasures in cascading failures, Phys. Rev. E * * 92 * *, 032809 (2015).
I. Dobson, B. A. Carreras, D. E. Newman, and J. M. Reynolds - Barredo, prevail statistic of cascading line outage spread in an electric transmission network from standard utility program data, IEEE T. Power Syst. * * 99 * *, 1 (2016).
M. Anghel, K. A. Werley, and A. E. Motter, Stochastic model for power power system moral force, [ * Proc. 40th Int. Conf. Syst. Sci. * ] { } HICSS’07, Big Island, HI, USA, Vol. 1, 113 (2007).
The data for the snapshots are prevail from Federal Energy Regulatory Commission (FERC) Form 715.
P. Hines, E. Cotilla - Sanchez, and S. Blumsack, Topological models and critical deceleration down: Two approaches to power arrangement blackout risk psychoanalysis, [ * Proc. 44th Int. Conf. Syst. Sci. * ] { } HICSS’11, Kauai, HI, USA, 1 (2011).
Source code is available for download at < https://github.com/yangyangangela/determine_cascade_sizes >
B. A. Carreras, D. E. Newman, I. Dobson, and N. S. Degala, Validating OPA with WECC data, [ * Proc. 46th Int. Conf. Syst. Sci. * ] { } HICSS’13, Maui, HI, USA, 2197 (2013).
P. Henneaux, P. E. Labeau, J. C. Maun, and L. Haarla, A two - grade probabilistic risk assessment of cascade outages, IEEE T. Power Syst. * * 31 * *, 2393 (2016).
P. D. Hines, I. Dobson, and P. Rezaei, cascade power outages propagate locally in an influence graph that is not the actual grid topology, arXiv:1508.01775 ( | thakt and M. Timme, Nonlocal enfects and countgrneasurxs in czscading failures, Phys. Rev. E **92**, 032809 (2015).
I. Dobsln, B. A. Caereras, D. E. Newman, and J. M. Feynolds-Bwrredo, Ovtaiiing statistics of cascading ljke oucajes spreading ik an electrhc transmissiot vecwork from standard utility data, IEEQ T. Powrr Syst. **99**, 1 (2016).
M. Anghej, K. A. Eqrleg, and A. E. Motter, Stochastic model fod power grid dynamicx, [*Proc. 40th Int. Conf. Syst. Scl.*]{} HIFSS’07, Big Island, HI, KSA, Vol. 1, 113 (2007).
Tye dwra for the svapshots agz obtained rrom Federal Energy Regulatory Zommixsion (FERC) Firm 715.
K. Hines, E. Cotiola-Sagchez, and S. Blumsack, Dopologocal models anc ccitixal slowing down: Two epproaches to power fystem bldciout risk analysiw, [*Proc. 44jh Ind. Covd. Shst. Sri.*]{} GICSS’11, Nauei, HI, USA, 1 (2011).
Aource code is available for dpwgooad at <https://fithub.soi/yangyangangela/determine_cascade_sizes>
B. A. Cdrrsras, D. E. Newman, I. Dobson, abd N. S. Degala, Validatinh OPA witr WECC data, [*Proc. 46th Int. Conf. Syst. Sci.*]{} HICSS’13, Maui, VI, USE, 2197 (2013).
P. Htnkcaux, P. E. Pabeau, J. C. Maun, and L. Haarla, A two-level probabilyatoc risk assessmekt of cascading ouyahex, IEEE T. Power Syst. **31**, 2393 (2016).
P. S. Hines, I. Dobson, and P. Rezaey, Casxading porer putages propagate locally ib an influenbe geaph that is not tke actual grnd topplogy, arXiv:1508.01775 ( | thaut and M. Timme, Nonlocal effects and cascading Phys. Rev. **92**, 032809 (2015). D. Newman, and J. Reynolds-Barredo, Obtaining statistics cascading line outages spreading in an transmission network from standard utility data, IEEE T. Power Syst. **99**, 1 (2016). Anghel, K. A. Werley, and A. E. Motter, Stochastic model for power grid [*Proc. Int. Syst. HICSS’07, Big Island, HI, USA, Vol. 1, 113 (2007). The data for the snapshots are obtained Federal Energy Regulatory Commission (FERC) Form 715. P. E. Cotilla-Sanchez, and S. Topological models and critical slowing Two to power blackout analysis, 44th Int. Conf. Sci.*]{} HICSS’11, Kauai, HI, USA, 1 (2011). Source code is available for download at <https://github.com/yangyangangela/determine_cascade_sizes> B. A. D. E. Dobson, and S. Validating with WECC data, Int. Conf. Syst. Sci.*]{} HICSS’13, Maui, (2013). P. Henneaux, P. E. Labeau, J. C. and L. A two-level probabilistic risk assessment of outages, IEEE T. Power Syst. **31**, 2393 (2016). D. Hines, I. Dobson, and P. Rezaei, Cascading power outages propagate locally in an influence is not the actual topology, arXiv:1508.01775 ( | thaut and M. Timme, Nonlocal effEcts and couNtermEasUreS iN casCadiNg failures, Phys. rEv. E **92**, 032809 (2015).
I. dobson, B. A. Carreras, D. E. NewmAn, and j. M. rEynoLDs-barreDo, ObtaiNInG STatIsTiCs oF cAScAding LinE outageS spreading In aN eLectric transMIsSion networK frOm standard utIliTy data, iEeE T. pOwer SYst. **99**, 1 (2016).
m. AnghEl, K. A. WeRLey, and a. E. Motter, SToCHastic MOdel for POWeR griD dynamics, [*Proc. 40th InT. coNF. Syst. Sci.*]{} HICSS’07, BIg IslaNd, hi, Usa, vol. 1, 113 (2007).
the Data for the SnApshoTS are obtAInED FRom fEderal Energy REgulatory CoMMisSion (FErC) forM 715.
p. Hines, e. CotiLlA-sanChez, and S. BluMsacK, TopologiCal modELs and crITical slOwing dOwn: two ApprOAcHeS to PoWEr sYStEm bLAckOut risk aNaLySis, [*PrOc. 44th iNT. cOnf. SYst. sci.*]{} HiCSS’11, KAuai, HI, USA, 1 (2011).
SourCe cOde iS AvaIlablE for dOwnlOaD at <htTps://gitHub.coM/yAngyangangela/deTermIne_cascadE_siZeS>
B. A. caRreraS, d. E. NewmAn, I. dobSon, and N. s. Degala, vAliDaTING OpA with WECC data, [*Proc. 46Th iNT. COnf. Syst. SCi.*]{} HICSs’13, maUi, hi, USA, 2197 (2013).
P. HenNeAux, p. E. LaBEAu, J. C. MAun, aND L. haarla, A tWo-leveL PrObAbilistIc Risk asSeSsmEnt Of casCAdinG outagEs, IEEE T. POwer SYSt. **31**, 2393 (2016).
P. D. Hines, I. DobsON, and P. Rezaei, CaSCaDINg POwer OutAges propagaTe loCAlly In an INfLueNCe graPh thaT iS NoT The actual grid topoloGy, ArXiv:1508.01775 ( | thaut and M. Timme, Nonloc al effects andcou nte rm easu resin cascading f a ilur es, Phys. Rev. E **92* *, 03 28 0 9 (2 0 15 ).
I . Dobso n ,B . A. C ar rer as , D . E.New man, an d J. M. Re yno ld s-Barredo, O b ta ining stat ist ics of casca din g line o uta g es sp rea dingin ane lectri c transmi ss i on net w ork fro m st anda rd utility data,I EE E T. Power Syst . **99 ** , 1 ( 201 6).
M. Anghe l, K. A . Werley , a n d A.E . Motter, Sto chastic mod e l f or pow er gr i d dyna mics, [ * Pro c. 40th Int . Co nf. Syst. Sci.* ] {} HICS S ’07, Bi g Isla nd, HI , US A ,Vo l.1, 113 (2 007 ) .
The data f or thesnap s h o t s ar e o btai ned f rom Federal E ner gy R e gul atory Comm issi on (FER C) For m 715 .
P. Hines, E. C otil la-Sanche z,an d S .Blums a ck, To pol ogi cal mod els and cri ti c a l s lowing down: Two a pp r o ac hes to p ower s y st em blackout r isk ana l y sis,[*Pr o c. 44th In t. Con f .Sy st. Sci .* ]{} HI CS S’1 1,Kauai , HI, USA,1 (2011) .
So u rce code is av a ilable for do w nl o a da t <h ttp s://github. com/ y angy anga n ge la/ d eterm ine_c as c ad e _sizes>
B. A. Carr er as, D. E. N ewman, I. Dob son, and N . S . Degala , Va l id a ting OPA withWECCdata, [*Pr o c. 46thInt.Conf. Sy st. Sci.* ] { } HICSS’ 13, Ma ui, HI , US A, 2197 (2013 ) .
P. H enneaux , P . E. La bea u,J.C.Ma un, and L . Haarla ,Atw o- lev el pr o babilist ic ri sk as sessm e nt ofcasca ding o ut a ges , IEEET .P o werSy st . ** 31* *, 2393 (20 1 6).
P. D. Hines, I . D o bson ,an d P. Re zaei, Cascadi ng power out ag espropag a t e locall y in an influence graph that is no t the act ual gridtop ology, ar X iv:150 8.0177 5 ( | thaut and_M. Timme, Nonlocal_effects and countermeasures in_cascading failures,_Phys._Rev. E_**92**,_032809 (2015).
I. Dobson, B. A. Carreras,_D. E. Newman, and J. M. Reynolds-Barredo,_Obtaining statistics of cascading_line outages spreading_in_an electric transmission network from standard utility data, IEEE T. Power Syst. **99**, 1_(2016).
M. Anghel,_K. A. Werley, and_A. E. Motter,_Stochastic_model for power grid dynamics,_[*Proc. 40th Int. Conf. Syst._Sci.*]{} HICSS’07,_Big Island, HI, USA, Vol. 1, 113 (2007).
The_data_for the snapshots_are obtained from Federal Energy Regulatory Commission (FERC) Form_715.
P. Hines, E. Cotilla-Sanchez, and S. Blumsack, Topological models_and critical slowing_down:_Two_approaches to power system_blackout risk analysis, [*Proc. 44th Int._Conf. Syst. Sci.*]{} HICSS’11, Kauai, HI,_USA, 1 (2011).
Source code is available for_download at <https://github.com/yangyangangela/determine_cascade_sizes>
B. A. Carreras, D. E. Newman, I. Dobson, and_N. S. Degala, Validating OPA with WECC_data, [*Proc._46th Int. Conf. Syst. Sci.*]{}_HICSS’13, Maui, HI,_USA, 2197_(2013).
P. Henneaux, P. E. Labeau, J. C. Maun,_and L. Haarla, A two-level probabilistic risk_assessment of cascading_outages, IEEE T. Power Syst. **31**,_2393_(2016).
P. D. Hines, I. Dobson, and_P. Rezaei,_Cascading_power outages_propagate locally in_an_influence graph_that_is not the actual grid topology,_arXiv:1508.01775_( |
05prop:bis3pts\] Let $C$ be a nondegenerate smooth irreducible curve in ${\mathbb{P}}^3$. If $L$ is a line that intersects the curve $C$ in three or more distinct points, then the line $L$ corresponds to a singular point in $\operatorname{Sec}(C)$.
The symmetric square $C^{(2)}$ is the quotient of $C \times C$ by the action of the symmetric group $\mathfrak{S}_2$, so points in this projective variety are unordered pairs of points on $C$; see [@05harris pp. 126–127]. The map $\varpi \colon C^{(2)} \to \operatorname{Sec}(C)$, defined by sending $\{x,y\}$ to the line spanned by the points $x$ and $y$ if $x \neq y$ or to the tangent line $T_x(C)$ if $x = y$, is a birational morphism. Since $|L \cap C| \geq 3$, the fibre $\varpi^{-1}(L)$ is a finite set containing more than one element. Hence, $\varpi^{-1}(L)$ is not connected and the Zariski Connectedness Theorem [@05mumford Sect. III.9.V] proves that $\operatorname{Sec}(C)$ is singular at $L$.
\[05lem:skewsym\] If $f \in {\mathbb{C}}[\![z,w]\!]$ satisfies $f(z,w)=-f(w,z)$, then the linear form $z-w$ divides the power series $f$.
We write the formal power series $f$ as a sum of homogeneous polynomials $f = \sum_{i \in {\mathbb{N}}} f_i$. Since we have $f(z,w) + f(w,z) = 0$, it follows that, in each degree $i$, we have $f_i(z,w) + f_i(w,z) = 0$. In particular, we see that $f_i(w, w) = 0$. If we consider $f_i(w,z)$ as a polynomial in the variable $z$ with coefficients in ${\mathbb{C}}[w]$, it follows that $w$ is a root of $f_i$. Thus, we conclude that $z-w$ divides $f_i$ for all $i | 05prop: bis3pts\ ] Let $ C$ be a nondegenerate smooth irreducible curve in $ { \mathbb{P}}^3$. If $ L$ is a line that intersect the curvature $ C$ in three or more distinct points, then the course $ L$ corresponds to a curious point in $ \operatorname{Sec}(C)$.
The symmetric square $ C^{(2)}$ is the quotient of $ C \times C$ by the military action of the symmetric group $ \mathfrak{S}_2 $, so points in this projective assortment are unordered pairs of points on $ C$; experience [ @05harris pp. 126–127 ]. The map $ \varpi \colon C^{(2) } \to \operatorname{Sec}(C)$, defined by send $ \{x, y\}$ to the lineage spanned by the points $ x$ and $ y$ if $ x \neq y$ or to the tangent line $ T_x(C)$ if $ x = y$, is a birational morphism. Since $ |L \cap C| \geq 3 $, the fibre $ \varpi^{-1}(L)$ is a finite set incorporate more than one element. Hence, $ \varpi^{-1}(L)$ is not connected and the Zariski Connectedness Theorem [ @05mumford Sect. III.9.V ] prove that $ \operatorname{Sec}(C)$ is singular at $ L$.
\[05lem: skewsym\ ] If $ f \in { \mathbb{C}}[\![z, w]\!]$ satisfies $ f(z, w)=-f(w, z)$, then the linear form $ z - w$ divide the power series $ f$.
We write the formal power series $ f$ as a kernel of homogeneous polynomials $ f = \sum_{i \in { \mathbb{N } } } f_i$. Since we have $ f(z, w) + f(w, z) = 0 $, it follows that, in each degree $ i$, we have $ f_i(z, w) + f_i(w, z) = 0$. In particular, we see that $ f_i(w, w) = 0$. If we consider $ f_i(w, z)$ as a polynomial in the variable $ z$ with coefficients in $ { \mathbb{C}}[w]$, it follows that $ w$ is a root of $ f_i$. Thus, we conclude that $ z - w$ divides $ f_i$ for all $ i | 05proo:bis3pts\] Let $C$ be a nondenenerate smooth nereducmble cudve in ${\mxthbb{P}}^3$. If $L$ is a line that iiterwects the curve $C$ in three ur more dpstinct piintw, then the line $L$ correspkkds tm a singular polnt in $\operdtorname{Sec}(C)$.
Tha rylmetric square $C^{(2)}$ is the quotient of $C \timex F$ by the actiog of ehe ajmnetric group $\mathfrak{S}_2$, so poihts in uhis projective vatiety are unordered pairs lf plints on $C$; see [@05hargis pp. 126–127]. The nap $\dqrpi \colon C^{(2)} \to \operatorname{Sec}(C)$, dgfined by sending $\{x,y\}$ to the line spanued by the kojnhv $x$ and $y$ ih $x \nez y$ or to thc tangett line $T_x(C)$ if $x = y$, ix a bieational morphism. Sinre $|L \cap C| \geq 3$, the sibre $\var[i^{-1}(M)$ is a finite set cintaiting mord thxn knx emement. Heice, $\varpi^{-1}(L)$ js not connwcted and the Zarisli Bpnnectedness Theorqm [@05iumford Sect. III.9.V] proves that $\operatornake{Ssc}(C)$ is singular at $L$.
\[05lem:wkewsym\] If $f \in {\mathbh{C}}[\![z,w]\!]$ satiffies $f(z,w)=-f(w,z)$, then the linear form $z-w$ divides the [ower rerncf $d$.
Wf write the formal power series $f$ as a sum of gokoneneous polynomicls $f = \sum_{i \in {\mstjbn{G}}} f_i$. Since we have $y(a,w) + f(w,z) = 0$, it follows that, ig eacy degree $y$, we have $f_i(z,w) + f_i(w,z) = 0$. In partixular, we see rhat $f_i(w, w) = 0$. If we consider $f_n(w,z)$ as a pokynomial in the variablz $z$ wifh coefficifnts in ${\mzghbb{C}}[w]$, it followr tmat $w$ is a root of $f_i$. Thus, we conclude thac $z-w$ divkdes $f_i$ fow all $i | 05prop:bis3pts\] Let $C$ be a nondegenerate smooth in If $L$ a line that three more distinct points, the line $L$ to a singular point in $\operatorname{Sec}(C)$. symmetric square $C^{(2)}$ is the quotient of $C \times C$ by the action the symmetric group $\mathfrak{S}_2$, so points in this projective variety are unordered pairs points $C$; [@05harris 126–127]. The map $\varpi \colon C^{(2)} \to \operatorname{Sec}(C)$, defined by sending $\{x,y\}$ to the line spanned the points $x$ and $y$ if $x \neq or to the tangent $T_x(C)$ if $x = y$, a morphism. Since \cap \geq the fibre $\varpi^{-1}(L)$ a finite set containing more than one element. Hence, $\varpi^{-1}(L)$ is not connected and the Zariski Connectedness [@05mumford Sect. that $\operatorname{Sec}(C)$ singular $L$. If $f \in $f(z,w)=-f(w,z)$, then the linear form $z-w$ series $f$. We write the formal power series as a of homogeneous polynomials $f = \sum_{i {\mathbb{N}}} f_i$. Since we have $f(z,w) + f(w,z) 0$, it follows that, in each degree $i$, we have $f_i(z,w) + f_i(w,z) = 0$. we see that $f_i(w, = 0$. If consider as polynomial the variable with coefficients in ${\mathbb{C}}[w]$, it follows that $w$ is a root $f_i$. Thus, we conclude that $z-w$ divides $f_i$ for all | 05prop:bis3pts\] Let $C$ be a nondegenErate smootH irreDucIblE cUrve In ${\maThbb{P}}^3$. If $L$ is a linE That Intersects the curve $C$ in tHree oR mORe diSTiNct poInts, theN ThE LIne $l$ cOrResPoNDs To a siNguLar poinT in $\operatoRnaMe{sec}(C)$.
The symmeTRiC square $C^{(2)}$ is The Quotient of $C \tImeS C$ by thE aCtiON of thE syMmetrIc grouP $\MathfrAk{S}_2$, so poinTs IN this pROjectivE VArIety Are unordered pairs OF pOInts on $C$; see [@05harrIs pp. 126–127]. ThE mAP $\vARPi \cOloN C^{(2)} \to \operatOrName{SEC}(C)$, definED bY SENdiNG $\{x,y\}$ to the line sPanned by the POinTs $x$ and $Y$ iF $x \nEQ y$ or to The taNgENt lIne $T_x(C)$ if $x = y$, iS a biRational mOrphisM. since $|L \cAP C| \geq 3$, thE fibre $\VarPi^{-1}(L)$ Is a fINiTe Set CoNTaiNInG moRE thAn one eleMeNt. hence, $\VarpI^{-1}(l)$ IS Not cOnnEcteD and tHe Zariski ConnEctEdneSS ThEorem [@05MumfoRd SeCt. iII.9.V] pRoves tHat $\opErAtorname{Sec}(C)$ is sInguLar at $L$.
\[05lem:SkeWsYm\] IF $f \In {\matHBb{C}}[\![z,w]\!]$ sAtiSfiEs $f(z,w)=-f(w,Z)$, then thE LinEaR FORm $Z-w$ divides the power sErIES $f$.
we write tHe formAL pOwER series $f$ As A suM of hOMOgeneOus pOLyNomials $f = \Sum_{i \in {\MAtHbB{N}}} f_i$. SinCe We have $F(z,W) + f(w,Z) = 0$, it FolloWS thaT, in eacH degree $i$, We havE $F_i(z,w) + f_i(w,z) = 0$. In partICular, we see thaT $F_i(W, W) = 0$. if WE conSidEr $f_i(w,z)$ as a poLynoMIal iN the VArIabLE $z$ witH coefFiCIeNTs in ${\mathbb{C}}[w]$, it folloWs That $w$ iS a rooT of $f_i$. Thus, we coNclude that $Z-W$ DIvides $f_i$ For aLL $i | 05prop:bis3pts\] Let $C$ b e a nondeg enera tesmo ot h ir redu cible curve in ${\m athbb{P}}^3$. If $L$ i s a l in e tha t i nters ects th e c u r ve$C $inth r ee or m ore distin ct points, th en the line $L $ c orresponds to a singularpoi nt in$\ ope r atorn ame {Sec} (C)$.The sy mmetric s qu a re $C^ { (2)}$ i s th e qu otient of $C \tim e sC $ by the actio n of t he sy m m etr icgroup $\ma th frak{ S }_2$, s o p o i n tsi n this projec tive variet y ar e unor de red pairsof po in t s o n $C$; see[@05 harris pp . 126– 1 27]. Th e map $\ varpi\co lon C^{ ( 2) }\to \ o per a to rna m e{S ec}(C)$, d ef inedby s e n d i ng $ \{x ,y\} $ tothe line span ned byt hepoint s $x$ and $ y$ if $x \n eq y$ o r to the tangen t li ne $T_x(C )$if $x = y$,i s a bi rat ion al morp hism. S i nce $ | L \c ap C| \geq 3$, the f i b re $\varpi ^{-1}( L )$ i s a finit eset con t a ining mor e t han oneelemen t .He nce, $\ va rpi^{- 1} (L) $ i s not conn ectedand theZaris k i Connectednes s Theorem [@05 m um f o rd Sect . I II.9.V] pro vest hat$\op e ra tor n ame{S ec}(C )$ is singular at $L$.
\ [0 5lem:s kewsy m\] If $f \in {\mathbb{ C } } [\![z,w] \!]$ sa t isfies $f(z,w) =-f(w ,z)$, then the line ar fo rm $z-w$ dividest h e powerser ies $f $.W ewrite the for m a l po we r serie s $ f$ as a su m o f h omo ge neous pol ynomials $ f=\s um_ {i \i n {\mathb b{ N}} }f_i $. Si n ce wehave$f(z ,w )+ f( w,z) =0 $, i t fo ll ow s th at, i n eac h de g ree $i$, w e have $f _i( z ,w)+f_ i(w,z)= 0$. In part ic ular, we s ee th at $f_ i ( w, w) =0$. If we consider $f_i ( w,z)$ a s a poly nomi al in the va riable $z $ withcoeffi cient sin$ { \math b b {C }}[ w] $, it foll o w s t hat $ w$ isa rootof $f_i$. Thus, we con clude that $z -w$ div i d es $f _ i$ for a l l $ i | 05prop:bis3pts\] Let_$C$ be_a nondegenerate smooth irreducible_curve in_${\mathbb{P}}^3$._If $L$_is_a line that_intersects the curve_$C$ in three or_more distinct points,_then_the line $L$ corresponds to a singular point in $\operatorname{Sec}(C)$.
The symmetric square $C^{(2)}$ is_the_quotient of_$C_\times_C$ by the action of_the symmetric group $\mathfrak{S}_2$, so_points in_this projective variety are unordered pairs of points_on_$C$; see [@05harris_pp. 126–127]. The map $\varpi \colon C^{(2)} \to \operatorname{Sec}(C)$, defined_by sending $\{x,y\}$ to the line_spanned by the_points_$x$_and $y$ if $x_\neq y$ or to the tangent_line $T_x(C)$ if $x = y$,_is a birational morphism. Since $|L \cap_C| \geq 3$, the fibre $\varpi^{-1}(L)$_is a finite set containing_more than_one element. Hence, $\varpi^{-1}(L)$ is_not connected and_the Zariski_Connectedness Theorem [@05mumford Sect. III.9.V]_proves that $\operatorname{Sec}(C)$ is singular at_$L$.
\[05lem:skewsym\] If $f_\in {\mathbb{C}}[\![z,w]\!]$ satisfies $f(z,w)=-f(w,z)$, then the_linear_form $z-w$ divides_the_power_series $f$.
We_write the formal_power_series $f$_as_a sum of homogeneous polynomials $f_=_\sum_{i \in {\mathbb{N}}} f_i$. Since we have_$f(z,w) + f(w,z) =_0$,_it follows that, in_each degree $i$, we have_$f_i(z,w) + f_i(w,z) = 0$. In_particular, we_see that_$f_i(w, w) = 0$. If we consider $f_i(w,z)$ as a polynomial_in the variable $z$ with coefficients_in ${\mathbb{C}}[w]$, it follows_that $w$_is_a root of_$f_i$._Thus, we_conclude that $z-w$ divides $f_i$ for all_$i |
302]{}. R. Schoen, L. Simon, S. T. Yau [*Curvature estimates for minimal hypersurfaces*]{} Acta Mathematica 1975, Volume 134, Issue 1, pp 275-288 R. Schoen, L. Simon [*Regularity of stable minimal hypersurfaces*]{} Comm. Pure Appl. Math. 34 (1981), 741-797 L. Simon [*Lectures on Geometric Measure Theory*]{} Proceedings of the Centre for Mathematical Analysis 3, Canberra, (1984), VII+272. J. Simons [*Minimal varieties in Riemannian manifolds*]{} Ann. of Math. 88 (1968), 62-105. N. Wickramasekera [*A general regularity theory for stable codimension 1 integral varifolds*]{} Ann. of Math. 179 (2014), 843-1007.
[^1]: This terminology is customary in the literature at least when $g$ is a constant: we use it also for any $g>0$.
[^2]: To avoid confusion, we remark that in the interior of $\{g=0\}$ there are no touching singularities by the maximum principle for minimal hypersurfaces (and therefore (**T**) is redundant there), so it is really on the boundary of $\{g=0\}$ that $(a3)$ of Theorem \[thm:mainreg\_gpos\] has to be replaced by (**T**).
[^3]: Such $u_{1}, u_{2}$ always exist by the definition of touching singularity; see remark \[oss:touchingsinggraphs\].
[^4]: We note that for an integral varifold obtained as in [@BW2], the natural notion of stability to expect in the limit is the one that can be expressed in terms of ambient test functions, which motivates geometrically the choice of stating the stability assumption (c) via ambient functions.
[^5]: We point out that the integral varifolds obtained by the construction in [@RogTon] may be as those described in the discussion that precedes (**T**), for example the varifold on the right in Figure \[fig:Touchingdoubleminimal\]: therefore also in this context it is important to weaken the assumption on the coincidence set to (**T**).
[^6]: Such $u_{1}, u_{2}$ always exist by the definition of touching singularity; see remark \[oss:touchingsinggraphs\].
[^7]: This would be prevented by Hopf lemma if | 302 ] { }. R. Schoen, L. Simon, S. T. Yau [ * Curvature estimates for minimal hypersurfaces * ] { } Acta Mathematica 1975, Volume 134, Issue 1, pp 275 - 288 R. Schoen, L. Simon [ * Regularity of stable minimal hypersurfaces * ] { } Comm. Pure Appl. Math. 34 (1981), 741 - 797 L. Simon [ * lecture on Geometric standard Theory * ] { } Proceedings of the Centre for Mathematical Analysis 3, Canberra, (1984), VII+272. J. Simons [ * Minimal varieties in Riemannian manifolds * ] { } Ann. of Math. 88 (1968), 62 - 105. N. Wickramasekera [ * A general regularity hypothesis for stable codimension 1 integral varifolds * ] { } Ann. of Math. 179 (2014), 843 - 1007.
[ ^1 ]: This terminology is customary in the literature at least when $ g$ is a constant: we practice it besides for any $ g>0$.
[ ^2 ]: To avoid confusion, we remark that in the interior of $ \{g=0\}$ there are no touching singularity by the maximum principle for minimal hypersurfaces (and consequently (* * T * *) is redundant there), so it is really on the boundary of $ \{g=0\}$ that $ (a3)$ of Theorem \[thm: mainreg\_gpos\ ] has to be replaced by (* * T * *).
[ ^3 ]: Such $ u_{1 }, u_{2}$ always exist by the definition of touching singularity; attend remark \[oss: touchingsinggraphs\ ].
[ ^4 ]: We note that for an integral varifold obtained as in [ @BW2 ], the lifelike notion of stability to ask in the limit is the one that can be express in terms of ambient test functions, which motivates geometrically the choice of express the stability assumption (c) via ambient functions.
[ ^5 ]: We point out that the integral varifolds obtained by the construction in [ @RogTon ] may be as those described in the discussion that precedes (* * T * *), for example the varifold on the right in Figure \[fig: Touchingdoubleminimal\ ]: therefore also in this context it is important to weaken the assumption on the coincidence sic to (* * T * *).
[ ^6 ]: Such $ u_{1 }, u_{2}$ always exist by the definition of touching singularity; visit remark \[oss: touchingsinggraphs\ ].
[ ^7 ]: This would be prevent by Hopf lemma if | 302]{}. R. Dchoen, L. Simon, S. T. Yau [*Cmrvature estimatgs for mmnimal gypersurwaces*]{} Acta Mathematica 1975, Volule 134, Issye 1, pp 275-288 R. Schoen, L. Simun [*Regulagity of srablt minimal hypersucraces*]{} Comm. Pudc Appn. Math. 34 (1981), 741-797 L. Simpn [*Lecturev on Geometric Mdadure Theory*]{} Proceedings of the Centwe for Kahhematical Anajysix 3, Cahberra, (1984), VII+272. J. Simons [*Minimal variefies in Riemannian msnifolds*]{} Ann. of Math. 88 (1968), 62-105. N. Aickgamasekera [*A generwl regularijg trwory for stacle codimension 1 integtal varifolds*]{} Ann. of Math. 179 (2014), 843-1007.
[^1]: Thir terkinology iw xushmmary in thx litegature at least when $c$ is a vonstant: we usc it elso for any $g>0$.
[^2]: To avoid cmnfusion, we remark that in dhz interior of $\{g=0\}$ there aee no touwhine sivgumacitjes by thx maximum pdinciple foe minimal hypersurfscqw (and therefode (**T**) if wedundant there), so it is really on the touhdary of $\{g=0\}$ that $(a3)$ of Thworem \[thm:mainreg\_gpos\] hws to be weplaced by (**T**).
[^3]: Such $u_{1}, u_{2}$ always exist by the definidion kw tiughine slngularity; see remark \[oss:touchingsinggraphs\].
[^4]: We houe nhat for an integval varifold obtaimef sf in [@BW2], the nxtural nofion of stability ho expest in the limiu is yhe one that can be expresswd in terms jd ambient test funetions, which motovatex geometrically the chonce of stating thf stabilifh assumption (c) vka smtient functions.
[^5]: We point ott that tie incegral vxrifplds oftained by the gmnstruction in [@RogHon] mcy be as those fescribed in the discussion thav precedes (**T**), gos eqample thz varinold on the rigrt in Figure \[fng:Touchiugdoubueminimal\]: nherefore also in thys context it ls important to weakqn tye awsumptiuv on the coincodence sen uo (**T**).
[^6]: Such $y_{1}, u_{2}$ always exist bn the sefinition of tiucying singularitu; sde welack \[oss:emuchingsinggsaphr\].
[^7]: Tfos wojld be kxevcntdd bu Hopf lemma if | 302]{}. R. Schoen, L. Simon, S. T. estimates minimal hypersurfaces*]{} Mathematica 1975, Volume R. L. Simon [*Regularity stable minimal hypersurfaces*]{} Pure Appl. Math. 34 (1981), 741-797 Simon [*Lectures on Geometric Measure Theory*]{} Proceedings of the Centre for Mathematical Analysis Canberra, (1984), VII+272. J. Simons [*Minimal varieties in Riemannian manifolds*]{} Ann. of Math. (1968), N. [*A regularity theory for stable codimension 1 integral varifolds*]{} Ann. of Math. 179 (2014), 843-1007. [^1]: This is customary in the literature at least when is a constant: we it also for any $g>0$. To confusion, we that the of $\{g=0\}$ there no touching singularities by the maximum principle for minimal hypersurfaces (and therefore (**T**) is redundant there), so is really boundary of that of \[thm:mainreg\_gpos\] has to by (**T**). [^3]: Such $u_{1}, u_{2}$ the definition of touching singularity; see remark \[oss:touchingsinggraphs\]. We note for an integral varifold obtained as [@BW2], the natural notion of stability to expect the limit is the one that can be expressed in terms of ambient test functions, geometrically the choice of the stability assumption via functions. We out that integral varifolds obtained by the construction in [@RogTon] may be as described in the discussion that precedes (**T**), for example the the in Figure \[fig:Touchingdoubleminimal\]: also in this context is to weaken the assumption coincidence to $u_{1}, always by the definition of singularity; see remark \[oss:touchingsinggraphs\]. [^7]: would be prevented by | 302]{}. R. Schoen, L. Simon, S. T. Yau [*CurvatuRe estimateS for mIniMal HyPersUrfaCes*]{} Acta MathemaTIca 1975, VOlume 134, Issue 1, pp 275-288 R. Schoen, L. SiMon [*ReGuLAritY Of StablE minimaL HyPERsuRfAcEs*]{} COmM. puRe AppL. MaTh. 34 (1981), 741-797 L. SimoN [*Lectures oN GeOmEtric Measure tHeOry*]{} ProceedIngS of the Centre For mathemAtIcaL analySis 3, canbeRra, (1984), VII+272. j. simons [*minimal vaRiETies in rIemanniAN MaNifoLds*]{} Ann. of Math. 88 (1968), 62-105. N. WickRAmASekera [*A general RegulaRiTY tHEOry For Stable codiMeNsion 1 INtegral VArIFOLds*]{} aNn. of Math. 179 (2014), 843-1007.
[^1]: This tErminology iS CusTomary In The LIteratUre at LeASt wHen $g$ is a consTant: We use it alSo for aNY $g>0$.
[^2]: To avoID confusIon, we rEmaRk tHat iN ThE iNteRiOR of $\{G=0\}$ ThEre ARe nO touchinG sInGularItieS BY THe maXimUm prIncipLe for minimal hYpeRsurFAceS (and tHerefOre (**T**) Is RedunDant thEre), so It Is really on the boUndaRy of $\{g=0\}$ that $(A3)$ of thEorEm \[Thm:maINreg\_gpOs\] hAs tO be replAced by (**T**).
[^3]: sUch $U_{1}, u_{2}$ ALWAyS exist by the definitIoN OF tOuching sIngulaRItY; sEE remark \[oSs:TouChinGSInggrAphs\].
[^4]: wE nOte that fOr an inTEgRaL varifoLd ObtainEd As iN [@BW2], The naTUral Notion Of stabilIty to EXpect in the limiT Is the one that cAN bE EXpREsseD in Terms of ambiEnt tESt fuNctiONs, WhiCH motiVates GeOMeTRically the choice of sTaTing thE stabIlity assumptiOn (c) via ambiENT FunctionS.
[^5]: We pOInT Out that the inteGral vArifolds obTAined by tHe conStructioN in [@RogTon] MAY be as thoSe dEscRibEd iN THe Discussion thaT PReceDeS (**T**), for exAmpLe the vaRifOld On tHe rIgHt in FigurE \[fig:ToucHiNgDoUbLemInimaL\]: ThereforE aLso In ThiS contEXt it is ImporTant To WeAKen The assuMPtION on tHe CoInciDenCe Set to (**t**).
[^6]: SucH $U_{1}, u_{2}$ aLways exIst by the dEfiNItioN oF tOuching Singularity; seE rEmark \[oss:toUcHinGsinggRAPhs\].
[^7]: This wOuld be prevented by Hopf leMMa if | 302]{}. R. Schoen, L. Simo n, S. T. Y au [* Cur vat ur e es tima tes for minima l hyp ersurfaces*]{} Acta Ma thema ti c a 19 7 5, Volu me 134, Is s u e 1 ,pp 27 5- 2 88 R. S cho en, L.Simon [*Re gul ar ity of stabl e m inimal hyp ers urfaces*]{}Com m. Pur eApp l . Mat h.34 (1 981),7 41-797 L. Simon [ * Lectur e s on Ge o m et ricMeasure Theory*]{ } P r oceedings of t he Cen tr e f o r Ma the matical An al ysis3 , Canbe r ra , ( 198 4 ), VII+272. J . Simons [* M ini mal va ri eti e s in R ieman ni a n m anifolds*]{ } An n. of Mat h. 88( 1968),6 2-105.N. Wic kra mas eker a [ *A ge ne r alr eg ula r ity theoryfo rstabl e co d i m e nsio n 1 int egral varifolds*]{ } A nn.o f M ath.179 ( 2014 ), 843- 1007.
[^1] :This terminolog y is customar y i nthe l itera t ure at le ast when $ g$ is a con st a n t :we use it also for a n y $ g>0$.
[ ^2]: T o a vo i d confus io n,we r e m ark t hati nthe inte rior o f $ \{ g=0\}$th ere ar enotou ching sing ularit ies by t he ma x imum principle for minimal h y pe r s ur f aces (a nd therefor e (* * T**) isr ed und a nt th ere), s o i t is really on the b ou ndaryof $\ {g=0\}$ that$(a3)$ ofT h e orem \[t hm:m a in r eg\_gpos\] has to b e replaced by (**T* *).
[^3]: Su ch $u_{1} , u_{2}$ a lwa ysexi stb y t he definition o f to uc hing si ngu larity; se e r ema rk\[ oss:touch ingsingg ra ph s\ ].
[ ^4]:W e note t ha t f or an inte g ral va rifol d ob ta in e d a s in [@ B W2 ] , the n at ural no ti on of sta b ili ty to e xpect inthe limi tis the on e that can be e xpressed i nter ms ofa m bient te st functions, which mot i vates g eom etric ally the choi ceof sta tin g the s tabili ty as su mpt i o n (c) v ia am bi ent functi o n s.
[^5] :We p oint ou t that the integra l va rifolds obtai ned byt h econ s tr u cti on in[ @ RogTon] may beas those d es c ri bed in the dis cu ssion t hat pre cedes (**T**) , for exa mple theva rifo l d on the right in Figu re \[fig: T ouchi n gd ouble min imal\] :the refor e also inthiscontex tit isimpor ta nt to we aken the assumption onthe co incid enc e set to(** T **) .
[^6]:Such $u_{1}, u _{2 }$alway s e x ist b y th e d efi n ition oft ouching s i ng ula r i ty ; see remar k \ [os s:tou chi n gsingg raph s\].
[^7]: Thisw ould be preven tedb y Ho pfl emma i f | 302]{}. R._Schoen, L._Simon, S. T. Yau_[*Curvature estimates_for_minimal hypersurfaces*]{}_Acta_Mathematica 1975, Volume_134, Issue 1,_pp 275-288 R. Schoen,_L. Simon [*Regularity_of_stable minimal hypersurfaces*]{} Comm. Pure Appl. Math. 34 (1981), 741-797 L. Simon [*Lectures on_Geometric_Measure Theory*]{}_Proceedings_of_the Centre for Mathematical Analysis_3, Canberra, (1984), VII+272. J._Simons [*Minimal_varieties in Riemannian manifolds*]{} Ann. of Math. 88_(1968),_62-105. N. Wickramasekera_[*A general regularity theory for stable codimension 1 integral_varifolds*]{} Ann. of Math. 179 (2014),_843-1007.
[^1]: This terminology_is_customary_in the literature at_least when $g$ is a constant:_we use it also for any_$g>0$.
[^2]: To avoid confusion, we remark that_in the interior of $\{g=0\}$ there_are no touching singularities by_the maximum_principle for minimal hypersurfaces (and_therefore (**T**) is_redundant there),_so it is_really on the boundary of $\{g=0\}$_that $(a3)$ of_Theorem \[thm:mainreg\_gpos\] has to be replaced by_(**T**).
[^3]:_Such $u_{1}, u_{2}$_always_exist_by the_definition of touching_singularity;_see remark \[oss:touchingsinggraphs\].
[^4]:_We_note that for an integral varifold_obtained_as in [@BW2], the natural notion of_stability to expect in_the_limit is the one_that can be expressed in_terms of ambient test functions, which_motivates geometrically_the choice_of stating the stability assumption (c) via ambient functions.
[^5]: We point_out that the integral varifolds obtained_by the construction in_[@RogTon] may_be_as those described_in_the discussion_that precedes (**T**), for example the varifold_on the_right in Figure \[fig:Touchingdoubleminimal\]: therefore also_in this context it_is_important to weaken the assumption on_the coincidence set to (**T**).
[^6]: Such_$u_{1}, u_{2}$ always exist by_the_definition_of touching singularity; see remark \[oss:touchingsinggraphs\].
[^7]:_This would be prevented by Hopf_lemma if |
such a Higgs decays in the way the SM Higgs does [@Chang:2008cw]. A significant model dependence of decay modes applies to other Higgses as well.
In this letter we would like to bring attention to the class of models with Higgs sectors that resemble the Higgs sector of the MSSM in the region with a light CP odd Higgs boson, $m_A \lesssim 10$ GeV, and $\tan \beta \lesssim 2.5$. Although this region is ruled out in the MSSM, after careful review of experimental limits we argue that it is easy to make this region phenomenologically viable in simple extensions of the MSSM. We focus on features that involve the two Higgs doublet part of a possible extension. Perhaps the most interesting observation is that all the Higgses resulting from two Higgs doublets: $h$, $H$, $A$ and $H^\pm$ could have been produced already at LEP or the Tevatron, but would have escaped detection because they decay in modes that have not been searched for or the experiments are not sensitive to. We discuss several search modes that present an opportunity to discover some of the Higgses at both LEP and the Tevatron. These decay modes also require modified strategies for Higgs discovery at the LHC. In addition, the light CP odd Higgs might be within the reach of current B factories.
[*MSSM at $m_A \ll m_Z$ and $\tan \beta \lesssim 2.5$:*]{} Let us start with the discussion of the Higgs sector of the MSSM [@higgs_review]. In the MSSM, the CP-even Higgs mass-squared matrix at the tree level can be written in terms of CP odd Higgs boson mass, $m_A$, the mass of the Z boson, $m_Z$, and the ratio of the vacuum expectation values of the two Higgs doublets, $\tan \beta = v_u/v_d$. Radiative corrections can be approximated by the contribution to $H_u-H_u$ element of the Higgs mass-squared matrix. This contribution, $\Delta$, is dominated by stop loops and thus depends on stop masses and mixing in the stop sector (it contains a $1/\sin^2 \beta$ factor). The Higgs mass eigenstates are obtained by an orthogonal transformation parameterized by $\alpha$. The coupling squared of the lighter CP-even Higgs boson to $ZZ$ divided by the SM value is given by: $$ | such a Higgs decays in the way the SM Higgs does [ @Chang:2008cw ]. A meaning exemplar dependence of decay modes apply to early Higgses as well.
In this letter we would like to bring care to the course of models with Higgs sectors that resemble the Higgs sector of the MSSM in the region with a light CP odd Higgs boson, $ m_A \lesssim 10 $ GeV, and $ \tan \beta \lesssim 2.5$. Although this area is ruled out in the MSSM, after careful review of experimental limits we argue that it is easy to produce this region phenomenologically viable in simple extension of the MSSM. We focus on features that involve the two Higgs doublet share of a possible extension. Perhaps the most interesting notice is that all the Higgses resulting from two Higgs doublet: $ h$, $ H$, $ A$ and $ H^\pm$ could have been produced already at LEP or the Tevatron, but would have escaped detection because they decay in mode that have not been searched for or the experiments are not sensitive to. We discuss several search mood that confront an opportunity to discover some of the Higgses at both LEP and the Tevatron. These decay modes also require modified strategies for Higgs discovery at the LHC. In addition, the light CP curious Higgs might be within the reach of current B factory.
[ * MSSM at $ m_A \ll m_Z$ and $ \tan \beta \lesssim 2.5 $ :* ] { } lease us begin with the discussion of the Higgs sector of the MSSM [ @higgs_review ]. In the MSSM, the CP - even Higgs mass - squared matrix at the tree level can be write in terms of CP odd Higgs boson mass, $ m_A$, the mass of the Z boson, $ m_Z$, and the ratio of the void expectation values of the two Higgs doublets, $ \tan \beta = v_u / v_d$. Radiative corrections can be approximated by the contribution to $ H_u - H_u$ element of the Higgs mass - squared matrix. This contribution, $ \Delta$, is dominated by stop loops and thus depends on hold on masses and mix in the stop sector (it contains a $ 1/\sin^2 \beta$ factor). The Higgs mass eigenstates are obtained by an extraneous transformation parameterized by $ \alpha$. The coupling squared of the lighter CP - even Higgs boson to $ ZZ$ divided by the SM value is given by: $ $ | sufh a Higgs decays in the way the SM Higyw does [@Rhang:2008cw]. A signiwicant model dependence of dxcay modew applies to other Higeses as wvll.
In thiw leuter we would likx to briky attsktion vo the class of models widh Higgs sectoss tkat resemble the Higgs sector of the MSSM im hhe region witr a kyght BP odd Higgs boson, $m_A \lesssim 10$ FeV, and $\tan \beta \lesxsim 2.5$. Although this region is guled out in the MDSM, after cqrefto review of dxperimental limits we argue that it is easy to make ghis xegion phenimwnopmgically vieble ig simple extcmsions of the MSSM. We focus on feqtures that involve tie two Higgs doublet part of d 'ossible extension. Peehqps tve mmst kbtefesuinj ogservahioi is that aml the Higgwes resulting from uwo Yiggs doubleta: $h$, $H$, $W$ wnd $H^\pm$ could have been produced alreadj at LEP or the Tevatron, bur would have escaped fetection because they decay in modes that have not been saarchxd fox or tfw fxperiments are not sensitive to. We discuss sqbetak search modes that presemt am opportunity jo discoved some of the Higgdes at foth OEP and tre Trvatron. These decay modes aoso require iidified strategies for Higgs biscovgry at the LHC. In addition, thz lighf CP odd Hihgs might ce within the rexch ox currenu B factories.
[*MSSM wt $m_A \ll n_Z$ aud $\tan \bdta \kesssii 2.5$:*]{} Let us dtart with the discussioj of jhe Hicgs sector of the MSSM [@higgs_review]. In the MSSM, the CP-evem Vigcs mass-szuarec matrix at tre tree level ean be wxitten in terms kf CP ovd Higgs bosjn mass, $m_A$, tha mass of the Z boson, $m_Z$, qnd rhe ratku of the vacuuk expectanijn vqlues of the two Hlggs akublets, $\tan \betc = v_u/v_d$. Radiative cofrestponx sdn be approxhmatdd cu the contributiiv to $H_u-H_u$ element of the Higfs mass-squared mattiw. This cobtributijn, $\Delta$, is dpminated by stop llops end thns deprndf on stop masses and mixing in the stop segtor (it contayns q $1/\sin^2 \beta$ fcctor). The Higgs mass eigenstates are obteined by an orthogonal rransformation paraktterized by $\akpha$. Ehe couplhng squared of the lughter CP-even Higns boson to $ZZ$ divided by tha SM galue is given by: $$ | such a Higgs decays in the way Higgs [@Chang:2008cw]. A model dependence of Higgses well. In this we would like bring attention to the class of with Higgs sectors that resemble the Higgs sector of the MSSM in the with a light CP odd Higgs boson, $m_A \lesssim 10$ GeV, and $\tan \lesssim Although region ruled out in the MSSM, after careful review of experimental limits we argue that it is to make this region phenomenologically viable in simple of the MSSM. We on features that involve the Higgs part of possible Perhaps most interesting observation that all the Higgses resulting from two Higgs doublets: $h$, $H$, $A$ and $H^\pm$ could have been already at the Tevatron, would escaped because they decay that have not been searched for are not sensitive to. We discuss several search that present opportunity to discover some of the at both LEP and the Tevatron. These decay also require modified strategies for Higgs discovery at the LHC. In addition, the light CP might be within the of current B [*MSSM $m_A m_Z$ $\tan \beta 2.5$:*]{} Let us start with the discussion of the Higgs sector the MSSM [@higgs_review]. In the MSSM, the CP-even Higgs mass-squared the level can be in terms of CP Higgs mass, $m_A$, the mass Z $m_Z$, of vacuum values of the two doublets, $\tan \beta = v_u/v_d$. corrections can be approximated element of the Higgs mass-squared matrix. This contribution, is dominated by stop loops and thus on stop masses and mixing in the stop sector (it contains a \beta$ factor). mass eigenstates are obtained by an orthogonal transformation by $\alpha$. The coupling of the lighter CP-even Higgs boson to $ZZ$ divided the value is by: $$ | such a Higgs decays in the way tHe SM Higgs dOes [@ChAng:2008Cw]. A SiGnifIcanT model dependenCE of dEcay modes applies to otheR HiggSeS As weLL.
IN this Letter wE WoULD liKe To BriNg ATtEntioN to The clasS of models wIth hiGgs sectors thAT rEsemble the higGs sector of thE MSsM in thE rEgiON with A liGht CP Odd HigGS boson, $M_A \lesssim 10$ gev, And $\tan \BEta \lessSIM 2.5$. ALthoUgh this region is ruLEd OUt in the MSSM, aftEr careFuL ReVIEw oF exPerimental LiMits wE Argue thAT iT IS EasY To make this regIon phenomenOLogIcally ViAblE In simpLe extEnSIonS of the MSSM. WE focUs on featuRes thaT Involve THe two HiGgs douBleT paRt of A PoSsIblE eXTenSIoN. PeRHapS the most InTeRestiNg obSERVAtioN is That All thE Higgses resulTinG froM Two higgs DoublEts: $h$, $h$, $A$ And $H^\pM$ could Have bEeN produced alreadY at LeP or the TeVatRoN, buT wOuld hAVe escaPed DetEction bEcause tHEy dEcAY IN mOdes that have not beeN sEARcHed for or The expERiMeNTs are not SeNsiTive TO. we disCuss SEvEral searCh modeS ThAt Present An OpportUnIty To dIscovER somE of the higgses aT both leP and the TevatrON. These decay moDEs ALSo REquiRe mOdified straTegiES for higgS DiScoVEry at The LHc. IN AdDItion, the light CP odd HIgGs mighT be wiThin the reach oF current B fACTOries.
[*MSSm at $m_a \Ll M_z$ and $\tan \beta \lesSsim 2.5$:*]{} LEt us start wITh the disCussiOn of the HIggs sectoR OF the MSSM [@HigGs_rEviEw]. IN THe mSSM, the CP-even hIGgs mAsS-squareD maTrix at tHe tRee LevEl cAn Be written In terms oF Cp oDd hiGgs Boson MAss, $m_A$, the MaSs oF tHe Z Boson, $M_z$, and thE ratiO of tHe VaCUum ExpectaTIoN VAlueS oF tHe twO HiGgS doubLets, $\TAn \bEta = v_u/v_d$. radiative CorREctiOnS cAn be appRoximated by thE cOntributioN tO $H_u-h_u$ elemENT of the HiGgs mass-squared matrix. ThiS ContribUtiOn, $\DelTa$, is Dominated By sTop looPs aND thus dEpends On stoP mAssES And miXINg In tHe Stop sector (IT ConTains A $1/\sIn^2 \beTa$ factoR). The Higgs mass eigenSTatEs are obtained By aN ortHOGoNal TRaNSfoRmATioN PArameterized by $\aLpha$. The couPlINg Squared of tHE liGhTer CP-evEn Higgs Boson TO $ZZ$ diviDed by the Sm value is gIvEn by: $$ | such a Higgs decays in th e way theSM Hi ggs do es [@C hang :2008cw]. A si g nifi cant model dependenceof de ca y mod e sappli es to o t he r Hig gs es as w e ll .
In th is lett er we woul d l ik e to bring a t te ntion to t heclass of mod els withHi ggs secto rsthatresemb l e theHiggs sec to r of th e MSSM i n th e re gion with a light CP odd Higgs boso n, $m_ A\ le s s sim 10 $ GeV, and $ \tan\ beta \l e ss s i m 2. 5 $. Although t his regioni s r uled o ut in the MS SM, a ft e r c areful revi ew o f experim entall imits w e arguethat i t i s e asyt oma keth i s r e gi onp hen omenolog ic al ly vi able i n simp leexte nsion s of the MSSM . W e fo c uson fe ature s th at invo lve th e two H iggs doublet pa rt o f a possi ble e xte ns ion.P erhaps th e m ost int erestin g ob se r v a ti on is that all the H i g gs es resul ting f r om t w o Higgsdo ubl ets: $ h$, $ H$,$ A$ and $H^ \pm$ c o ul dhave be en produ ce d a lre ady a t LEP or th e Tevatr on, b u t would have e s caped detecti o nb e ca u se t hey decay in m odes that hav e n otb een s earch ed fo r or the experiments a re not sens itive to. Wediscuss se v e r al searc h mo d es that present a n opp ortunity t o discove r som e of the Higgsesa t both LE P a ndthe Te v a tr on. These dec a y mod es also r equ ire mod ifi edstr ate gi es for Hi ggs disc ov er yat th e LHC . In addi ti on, t helight CP odd Higg s mi gh tb e w ithin t h er e achof c urre ntBfacto ries .
[ *MSSM a t $m_A \l l m _ Z$ a nd $ \tan \b eta \lesssim2. 5$:*]{} Le tusstartw i th the d iscussion of the Higgss ector o f t he MS SM [ @higgs_re vie w]. In th e MSSM, the C P-eve nHig g s mass - s qu are dmatrix att h e t ree l ev el c an be w ritten in terms of CPodd Higgs bos onmass , $m _A$ , t h e m as s of t he Z boson, $m_ Z$, and th er at io of thev acu um expect ation v alues of thetwo Higgs doublets ,$\ta n \be ta = v_u/v _d$. Rad iative co r recti o ns canbeapprox im ate d bythe co n tri butio n to $ H_ u-H_u$ elem en t of the Higgs mass-squared mat rix. T his c ont ribution, $\ D elt a$, is do mina ted by sto p l oop s and th u s dep ends on st o p mas sesa nd mixing in th e st op sector ( i t con tains a$ 1/\sin ^2 \ beta$ factor). Th e Higgs mass ei gens t a tes ar e obt ai ned by an orth ogo na l transfor ma tion parame terizedby $\alp ha$. T he cou pling s q u ar e d of t he l igh ter CP-ev enHi g gs boso nto $ZZ$ d ivid ed by th e SM v a luei s given by: $$ | such_a Higgs_decays in the way_the SM_Higgs_does [@Chang:2008cw]. A_significant_model dependence of_decay modes applies_to other Higgses as_well.
In this letter_we_would like to bring attention to the class of models with Higgs sectors that_resemble_the Higgs_sector_of_the MSSM in the region_with a light CP odd_Higgs boson,_$m_A \lesssim 10$ GeV, and $\tan \beta \lesssim_2.5$._Although this region_is ruled out in the MSSM, after careful review_of experimental limits we argue that_it is easy_to_make_this region phenomenologically viable_in simple extensions of the MSSM._We focus on features that involve_the two Higgs doublet part of a_possible extension. Perhaps the most interesting_observation is that all the_Higgses resulting_from two Higgs doublets: $h$,_$H$, $A$ and_$H^\pm$ could_have been produced_already at LEP or the Tevatron,_but would have_escaped detection because they decay in_modes_that have not_been_searched_for or_the experiments are_not_sensitive to._We_discuss several search modes that present_an_opportunity to discover some of the Higgses_at both LEP and_the_Tevatron. These decay modes_also require modified strategies for_Higgs discovery at the LHC. In_addition, the_light CP_odd Higgs might be within the reach of current B factories.
[*MSSM_at $m_A \ll m_Z$ and $\tan_\beta \lesssim 2.5$:*]{} Let_us start_with_the discussion of_the_Higgs sector_of the MSSM [@higgs_review]. In the MSSM, the_CP-even Higgs_mass-squared matrix at the tree level_can be written in_terms_of CP odd Higgs boson mass,_$m_A$, the mass of the Z_boson, $m_Z$, and the ratio_of_the_vacuum expectation values of the_two Higgs doublets, $\tan \beta =_v_u/v_d$. Radiative corrections_can be approximated by the contribution to_$H_u-H_u$_element of the Higgs mass-squared matrix._This_contribution, $\Delta$, is dominated by stop_loops_and_thus depends on stop masses_and mixing in the stop sector_(it contains a $1/\sin^2 \beta$ factor). The Higgs mass_eigenstates are obtained_by an orthogonal transformation parameterized_by_$\alpha$._The coupling squared of the lighter CP-even Higgs boson to_$ZZ$ divided_by the SM_value is given by: $$ |
the giant resonance peak, as it is well known, RPA calculations based on EDF with effective masses lower than the bare nucleon mass overestimates the excitation energy. As it is also well known, the PVC approach introduces an energy dependence also in the real part of the self-energy that corrects the value of the effective mass [@bertsch1968] such that it is more realistic and compares better with the empirical value. On this regard, it seems that the results employing the [*subtraction method*]{} better reproduces the excitation energy of the giant resonance.
Isoscalar monopole response in ${}^{208}$Pb
-------------------------------------------
In Fig.\[fig8\], the isoscalar monopole response in ${}^{208}$Pb as predicted by the SAMi interaction is shown. In grey bars we show the RPA response while in dashed red lines the PVC without subtraction and in dot dashed blue lines the PVC with subtraction are displayed. A black arrow indicates the centroid energy $m_1/m_0=14.24\pm 0.11$ MeV measured within 8 and 22 MeV [@youngblood1999]. The centroid energy defined as the square root of the ratio between the EWSR and IEWSR is $\sqrt{m_1/m_{-1}}=14.18\pm 0.11$ MeV[@youngblood1999]. The width of the resonance has been measured to be between 2 and 3 MeV approximately (cf. Table 4.1 of Ref.[@harakeh2001]). In our calculations using SAMi, we find within 8 and 22 MeV that for the RPA the EWSR is 97%, $m_1/m_0 = 13.7$ MeV and $\sqrt{m_1/m_{-1}} = 13.5$ MeV; for the PVC without subtraction the EWSR is 91%, $m_1/m_0 = 13.4$ MeV and $\sqrt{m_1/m_{-1}} = 13.2$ MeV; and for the PVC with subtraction the EWSR is 91%, $m_1/m_0 = 13.7$ MeV and $\sqrt{m_1/m_{-1}} = 13.6$ MeV. The width predicted by our PVC calculations is of 2 MeV. Thus, SAMi predicts reasonable values for the excitation energy and width of this resonance both, with and without subtraction, since PVC effects are small as expected. It is well known within the R | the giant resonance peak, as it is well known, RPA calculations free-base on EDF with effective masse lower than the bare nucleon mass overestimates the excitation department of energy. As it is also well known, the PVC approach introduces an department of energy addiction also in the real part of the self - energy that corrects the value of the effective bulk [ @bertsch1968 ] such that it is more realistic and compares well with the empirical value. On this regard, it seems that the results employing the [ * subtraction method acting * ] { } better reproduces the excitation department of energy of the giant resonance.
Isoscalar monopole response in $ { } ^{208}$Pb
-------------------------------------------
In Fig.\[fig8\ ], the isoscalar monopole response in $ { } ^{208}$Pb as predicted by the SAMi interaction is show. In grey bars we show the RPA response while in dashed red line the PVC without subtraction and in dot dashed blue lines the PVC with subtraction are displayed. A bootleg arrow bespeak the centroid energy $ m_1 / m_0=14.24\pm 0.11 $ MeV measured within 8 and 22 MeV [ @youngblood1999 ]. The centroid energy defined as the square root of the ratio between the EWSR and IEWSR is $ \sqrt{m_1 / m_{-1}}=14.18\pm 0.11 $ MeV[@youngblood1999 ]. The width of the resonance has been measured to be between 2 and 3 MeV approximately (californium. Table 4.1 of Ref.[@harakeh2001 ]). In our calculations use lapp, we find within 8 and 22 MeV that for the RPA the EWSR is 97% , $ m_1 / m_0 = 13.7 $ MeV and $ \sqrt{m_1 / m_{-1 } } = 13.5 $ MeV; for the PVC without subtraction the EWSR is 91% , $ m_1 / m_0 = 13.4 $ MeV and $ \sqrt{m_1 / m_{-1 } } = 13.2 $ MeV; and for the PVC with subtraction the EWSR is 91% , $ m_1 / m_0 = 13.7 $ MeV and $ \sqrt{m_1 / m_{-1 } } = 13.6 $ MeV. The width predict by our PVC calculations is of 2 MeV. Thus, SAMi predicts reasonable value for the excitation energy and width of this resonance both, with and without subtraction, since PVC effects are small as have a bun in the oven. It is well known within the R | thf giant resonance peak, ar it is well known, RPA calcumations cased on EDF with effective lawses oower than the bare nuzleon masd overesrimaues the excitatioi energy. As it ls alvi well known, tme PVC apprmach introducev xn energy dependence also in the real part og hhe self-energy thau cjrredns the value of the effective mzss [@berusch1968] such that it os more realistic and compwres better with the elpirical vaoue. Jb this regara, it seems that the reaults employing the [*subtraction methpd*]{} better tzprofoces the excmtatiog energy of bne giatt resomance.
Isoscalar moiopooe response in ${}^{208}$Pb
-------------------------------------------
In Fmg.\[fig8\], the isoscalar ionopole seaponse in ${}^{208}$Pb as peeeicteg by the WAMk ihtxradtion ls ahown. In gdey bars we show the RPA respomsq while in dasged req jines the PVC without subtraction and it dkt dashed blue lines thw PVC with subtractioj are dis[layed. A black arrow indicates the centroid energf $m_1/m_0=14.24\pj 0.11$ MtV measjeef within 8 and 22 MeV [@youngblood1999]. The centroid enqdgu cefined as the square rooy lf jhe ratio betwgen the EWAR and IEWSR is $\sqgt{m_1/m_{-1}}=14.18\pm 0.11$ MeV[@yiungblood1999]. The width of the resonance has been measurvd ti be between 2 and 3 MeV approxnmateli (cf. Tsble 4.1 of Ref.[@harakeh2001]). In uur dalculationd using SZOi, we find withiv 8 sng 22 MeV that for the RPA thq EWSR is 97%, $m_1/m_0 = 13.7$ MeV xnd $\xqrt{m_1/m_{-1}} = 13.5$ MeV; for the IEC without subtrachion jhe EWVR is 91%, $m_1/m_0 = 13.4$ MeV and $\sqrt{m_1/m_{-1}} = 13.2$ MeV; and for vie PVC with sobtsacnion the ZWSR ix 91%, $m_1/m_0 = 13.7$ MeV agd $\sqrt{m_1/m_{-1}} = 13.6$ MeR. The wibth prddicted by our PVR calculatiogs is of 2 MeV. Jhus, SAMi previcts reafonavle calues wur the excitatoon energj cnd width of this resonance bojh, with and withobu sovtraction, sincr PXC qfverts awa small as efpecged. Ot is well known dithon the R | the giant resonance peak, as it is RPA based on with effective masses mass the excitation energy. it is also known, the PVC approach introduces an dependence also in the real part of the self-energy that corrects the value the effective mass [@bertsch1968] such that it is more realistic and compares better the value. this it seems that the results employing the [*subtraction method*]{} better reproduces the excitation energy of the resonance. Isoscalar monopole response in ${}^{208}$Pb ------------------------------------------- In the isoscalar monopole response ${}^{208}$Pb as predicted by the interaction shown. In bars show RPA response while dashed red lines the PVC without subtraction and in dot dashed blue lines the PVC with subtraction displayed. A indicates the energy 0.11$ measured within 8 MeV [@youngblood1999]. The centroid energy defined root of the ratio between the EWSR and is $\sqrt{m_1/m_{-1}}=14.18\pm MeV[@youngblood1999]. The width of the resonance been measured to be between 2 and 3 approximately (cf. Table 4.1 of Ref.[@harakeh2001]). In our calculations using SAMi, we find within 8 MeV that for the the EWSR is $m_1/m_0 13.7$ and = 13.5$ for the PVC without subtraction the EWSR is 91%, $m_1/m_0 = MeV and $\sqrt{m_1/m_{-1}} = 13.2$ MeV; and for the PVC the is 91%, $m_1/m_0 13.7$ MeV and $\sqrt{m_1/m_{-1}} 13.6$ The width predicted by calculations of SAMi reasonable for the excitation energy width of this resonance both, and without subtraction, since expected. It is well known within the R | the giant resonance peak, as it Is well knowN, RPA cAlcUlaTiOns bAsed On EDF with effecTIve mAsses lower than the bare nUcleoN mASs ovEReStimaTes the eXCiTATioN eNeRgy. as IT iS also WelL known, tHe PVC approAch InTroduces an enERgY dependencE alSo in the real pArt Of the sElF-enERgy thAt cOrrecTs the vALue of tHe effectiVe MAss [@berTSch1968] such THAt It is More realistic and cOMpARes better with tHe empiRiCAl VALue. on tHis regard, iT sEems tHAt the reSUlTS EMplOYing the [*subtraCtion method*]{} BEttEr reprOdUceS The excItatiOn ENerGy of the gianT resOnance.
IsoScalar MOnopole REsponse In ${}^{208}$Pb
-------------------------------------------
In fig.\[Fig8\], The iSOsCaLar MoNOpoLE rEspONse In ${}^{208}$Pb as prEdIcTed by The SamI INterActIon iS showN. In grey bars we ShoW the rpA rEsponSe whiLe in DaShed rEd lineS the PvC Without subtractIon aNd in dot daSheD bLue LiNes thE pVC witH suBtrAction aRe displAYed. a bLACK aRrow indicates the ceNtROId Energy $m_1/m_0=14.24\Pm 0.11$ MeV mEAsUrED within 8 aNd 22 meV [@YounGBLood1999]. THe ceNTrOid energY definED aS tHe squarE rOot of tHe RatIo bEtweeN The EwSR and iEWSR is $\sQrt{m_1/m_{-1}}=14.18\PM 0.11$ MeV[@youngblood1999]. tHe width of the rESoNANcE Has bEen Measured to bE betWEen 2 aNd 3 Mev ApProXImateLy (cf. TAbLE 4.1 oF ref.[@harakeh2001]). In our calcUlAtions Using sAMi, we find witHin 8 and 22 MeV tHAT For the RPa the ewSr Is 97%, $m_1/m_0 = 13.7$ MeV and $\sqrt{M_1/m_{-1}} = 13.5$ MeV; For the PVC wIThout subTractIon the EWsR is 91%, $m_1/m_0 = 13.4$ MeV AND $\sqrt{m_1/m_{-1}} = 13.2$ MEV; aNd fOr tHe Pvc WiTh subtraction THE EWSr iS 91%, $m_1/m_0 = 13.7$ MeV aNd $\sQrt{m_1/m_{-1}} = 13.6$ Mev. ThE wiDth PreDiCted by our pVC calcuLaTiOnS iS of 2 meV. ThUS, SAMi preDiCts ReAsoNable VAlues fOr the ExciTaTiON enErgy and WIdTH Of thIs ReSonaNce BoTh, witH and WIthOut subtRaction, siNce pvC efFeCtS are smaLl as expected. IT iS well known WiThiN the R | the giant resonance peak, as it iswellkno wn, R PA c alcu lations basedo n ED F with effective masse s low er than th e bar e nucle o nm a ssov er est im a te s the ex citatio n energy.Asit is also wel l k nown, thePVC approach in tro ducesan en e rgy d epe ndenc e also in the real par to f thes elf-ene r g ythat corrects the val u eo f the effectiv e mass [ @ be r t sch 196 8] such th at it i s more r e al i s t ica nd compares b etter witht heempiri ca l v a lue. O n thi sr ega rd, it seem s th at the re sultse mployin g the [* subtra cti onmeth o d* ]{ } b et t err ep rod u ces the exc it at ion e nerg y o f the gi antreson ance.
Isosca lar mon o pol e res ponse in${ }^{20 8}$Pb----- -- --------------- ---- --------- --- -- ---
In Fi g .\[fig 8\] , t he isos calar m o nop ol e r es ponse in ${}^{208} $P b as predict ed byt he S A Mi inter ac tio n is s hown. Ing re y bars w e show th eRPA res po nse wh il e i n d ashed redlinesthe PVCwitho u t subtractiona nd in dot das h ed b lu e lin esthe PVC wit h su b trac tion ar e d i splay ed. A b l ac k arrow indicates th ecentro id en ergy $m_1/m_0 =14.24\pm0 . 1 1$ MeV m easu r ed within 8 and 2 2 MeV [@youngbl o od1999]. Thecentroid energy d e f ined asthe sq uar e r o o tof the ratiob e twee nthe EWS R a nd IEWS R i s $ \sq rt{ m_ 1/m_{-1}} =14.18\p m0. 11 $MeV [@you n gblood19 99 ].Th e w idtho f thereson ance h as bee n measu r ed t o be b et ween 2an d 3 M eV a p pro ximatel y (cf. Ta ble 4.1of R ef.[@ha rakeh2001]).In our calcu la tio ns usi n g SAMi, w e find within 8 and 22M eV that fo r the RPA the EWSR is 97%,$m_ 1 /m_0 = 13.7$ MeVan d $ \ s qrt{m _ 1 /m _{- 1} } = 13.5$M e V;for t he PVC withou t subtraction theE WSR is 91%, $m_1 /m_ 0 =1 3 .4 $ M e Va nd$\ s qrt { m _1/m_{-1}} = 13 .2$ MeV; a nd fo r the PVCw ith s ubtract ion the EWSR is 91%, $m_1/m_0 = 13.7$Me V an d $\s qrt{m_1/m_ {-1}} =13.6$ MeV . Thew id th pr edi cted b your PVCcalcul a tio ns is of 2Me V. Thu s, SA Mi predict s reasonable values for the e xcita tio n energyand wid th of thi s re sonance bo th, wi th an d w i thout sub t ra cti o n, si nceP VC effect s a res m al l as expect e d . It is w ell knownwith in the R | the_giant resonance_peak, as it is_well known,_RPA_calculations based_on_EDF with effective_masses lower than_the bare nucleon mass_overestimates the excitation_energy._As it is also well known, the PVC approach introduces an energy dependence also_in_the real_part_of_the self-energy that corrects the_value of the effective mass_[@bertsch1968] such_that it is more realistic and compares better_with_the empirical value._On this regard, it seems that the results employing_the [*subtraction method*]{} better reproduces the_excitation energy of_the_giant_resonance.
Isoscalar monopole response in_${}^{208}$Pb
-------------------------------------------
In Fig.\[fig8\], the isoscalar monopole response_in ${}^{208}$Pb as predicted by the_SAMi interaction is shown. In grey bars_we show the RPA response while_in dashed red lines the_PVC without_subtraction and in dot dashed_blue lines the_PVC with_subtraction are displayed._A black arrow indicates the centroid_energy $m_1/m_0=14.24\pm 0.11$_MeV measured within 8 and 22_MeV_[@youngblood1999]. The centroid_energy_defined_as the_square root of_the_ratio between_the_EWSR and IEWSR is $\sqrt{m_1/m_{-1}}=14.18\pm 0.11$_MeV[@youngblood1999]._The width of the resonance has been_measured to be between_2_and 3 MeV approximately_(cf. Table 4.1 of Ref.[@harakeh2001])._In our calculations using SAMi, we_find within_8 and_22 MeV that for the RPA the EWSR is 97%, $m_1/m_0_= 13.7$ MeV and $\sqrt{m_1/m_{-1}} =_13.5$ MeV; for the_PVC without_subtraction_the EWSR is_91%,_$m_1/m_0 =_13.4$ MeV and $\sqrt{m_1/m_{-1}} = 13.2$ MeV;_and for_the PVC with subtraction the EWSR_is 91%, $m_1/m_0 =_13.7$_MeV and $\sqrt{m_1/m_{-1}} = 13.6$ MeV._The width predicted by our PVC_calculations is of 2 MeV._Thus,_SAMi_predicts reasonable values for the_excitation energy and width of this_resonance both, with_and without subtraction, since PVC effects are_small_as expected. It is well known_within_the R |
heng and Y. R. Liu, arXiv:1905.08605 \[hep-ph\]. C. W. Xiao, J. Nieves and E. Oset, arXiv:1904.01296 \[hep-ph\]. J. J. Wu, R. Molina, E. Oset and B. S. Zou, Phys. Rev. C [**84**]{}, 015202 (2011) \[arXiv:1011.2399 \[nucl-th\]\].
C. W. Shen, J. J. Wu and B. S. Zou, arXiv:1906.03896 \[hep-ph\]. N. Isgur and M. B. Wise, Phys. Lett. B [**232**]{}, 113 (1989). M. Neubert, Phys. Rept. [**245**]{}, 259 (1994) \[hep-ph/9306320\]. A.V. Manohar and M.B. Wise, [*Heavy Quark Physics*]{}, Cambridge Monographs on Particle Physics, Nuclear Physics and Cosmology, vol. 10 (Cambridge University Press, Cambridge, England, 2000).
K. C. Bowler [*et al.*]{} \[UKQCD Collaboration\], Phys. Rev. D [**54**]{}, 3619 (1996) doi:10.1103/PhysRevD.54.3619 \[hep-lat/9601022\]. C. Albertus, J. E. Amaro, E. Hernandez and J. Nieves, Nucl. Phys. A [**740**]{}, 333 (2004) doi:10.1016/j.nuclphysa.2004.04.114 \[nucl-th/0311100\]. M. Tanabashi [*et al.*]{} \[Particle Data Group\], Phys. Rev. D [**98**]{}, no. 3, 030001 (2018). doi:10.1103/PhysRevD.98.030001
M. E. Rose, Elementary Theory of Angular Momentum, John Wiley, 1957.
J. Nieves and E. Ruiz Arriola, Nucl. Phys. A [**679**]{}, 57 (2000) doi:10.1016/S0375-9474(00)00321-3 | heng and Y. R. Liu, arXiv:1905.08605 \[hep - ph\ ]. C. W. Xiao, J. Nieves and E. Oset, arXiv:1904.01296 \[hep - ph\ ]. J. J. Wu, R. Molina, E. Oset and B. S. Zou, Phys. Rev. C [ * * 84 * * ] { }, 015202 (2011) \[arXiv:1011.2399 \[nucl - th\]\ ].
C. W. Shen, J. J. Wu and B. S. Zou, arXiv:1906.03896 \[hep - ph\ ]. N. Isgur and M. B. Wise, Phys. Lett. B [ * * 232 * * ] { }, 113 (1989). M. Neubert, Phys. Rept. [ * * 245 * * ] { }, 259 (1994) \[hep - ph/9306320\ ]. A.V. Manohar and M.B. Wise, [ * Heavy Quark Physics * ] { }, Cambridge Monographs on Particle Physics, Nuclear Physics and Cosmology, vol. 10 (Cambridge University Press, Cambridge, England, 2000).
K. C. Bowler [ * et al. * ] { } \[UKQCD Collaboration\ ], Phys. Rev. D [ * * 54 * * ] { }, 3619 (1996) doi:10.1103 / PhysRevD.54.3619 \[hep - lat/9601022\ ]. C. Albertus, J. E. Amaro, E. Hernandez and J. Nieves, Nucl. Phys. A [ * * 740 * * ] { }, 333 (2004) doi:10.1016 / j.nuclphysa.2004.04.114 \[nucl - th/0311100\ ]. M. Tanabashi [ * et al. * ] { } \[Particle Data Group\ ], Phys. Rev. D [ * * 98 * * ] { }, no. 3, 030001 (2018). doi:10.1103 / PhysRevD.98.030001
M. E. Rose, Elementary Theory of Angular Momentum, John Wiley, 1957.
J. Nieves and E. Ruiz Arriola, Nucl. Phys. A [ * * 679 * * ] { }, 57 (2000) doi:10.1016 / S0375 - 9474(00)00321 - 3 | henh and Y. R. Liu, arXiv:1905.08605 \[hep-ph\]. G. W. Xiao, J. Nieves aue E. Osev, arXiv:1904.01296 \[hep-ph\]. J. G. Wu, R. Molina, E. Oset and B. S. Zou, 'hys. Eev. C [**84**]{}, 015202 (2011) \[arXiv:1011.2399 \[nucl-th\]\].
C. W. Shen, G. J. Wu and H. S. Zou, arZiv:1906.03896 \[iep-ph\]. N. Isgur and M. B. Wise, Ikys. Lefb. B [**232**]{}, 113 (1989). N. Neubert, Phys. Rgpt. [**245**]{}, 259 (1994) \[hep-ph/9306320\]. A.V. Manohar ang O.B. Wise, [*Heavy Quark Physics*]{}, Cambridge Monogrspjs on Particle Phyxycs, Hlcoear Physics and Cosmology, vom. 10 (Cambgidge University Lress, Cambridge, England, 2000).
K. C. Howlfr [*et al.*]{} \[UKQCD Colpaboration\], Kgys. Wwv. D [**54**]{}, 3619 (1996) doi:10.1103/PhhsRevD.54.3619 \[hep-lat/9601022\]. C. Albertua, J. E. Amaro, E. Hernandez and J. Niever, Nuck. Phys. A [**740**]{}, 333 (2004) eou:10.1016/j.nkwlphysa.2004.04.114 \[nuco-th/0311100\]. M. Nanabashi [*et al.*]{} \[Partiwle Dats Group\], Phys. Rey. D [**98**]{}, nm. 3, 030001 (2018). doi:10.1103/PhysRevD.98.030001
M. E. Rose, Enementary Theory os Angular Mkmentum, John Wilet, 1957.
H. Nievgs ang E. Rjuz Xrrjoka, Nucl. Pjys. E [**679**]{}, 57 (2000) doi:10.1016/S0375-9474(00)00321-3 | heng and Y. R. Liu, arXiv:1905.08605 \[hep-ph\]. Xiao, Nieves and Oset, arXiv:1904.01296 \[hep-ph\]. E. and B. S. Phys. Rev. C 015202 (2011) \[arXiv:1011.2399 \[nucl-th\]\]. C. W. J. J. Wu and B. S. Zou, arXiv:1906.03896 \[hep-ph\]. N. Isgur and M. Wise, Phys. Lett. B [**232**]{}, 113 (1989). M. Neubert, Phys. Rept. [**245**]{}, 259 \[hep-ph/9306320\]. Manohar M.B. [*Heavy Quark Physics*]{}, Cambridge Monographs on Particle Physics, Nuclear Physics and Cosmology, vol. 10 (Cambridge University Cambridge, England, 2000). K. C. Bowler [*et al.*]{} Collaboration\], Phys. Rev. D 3619 (1996) doi:10.1103/PhysRevD.54.3619 \[hep-lat/9601022\]. C. J. Amaro, E. and Nieves, Phys. A [**740**]{}, (2004) doi:10.1016/j.nuclphysa.2004.04.114 \[nucl-th/0311100\]. M. Tanabashi [*et al.*]{} \[Particle Data Group\], Phys. Rev. D [**98**]{}, no. 3, 030001 doi:10.1103/PhysRevD.98.030001 M. Elementary Theory Angular John 1957. J. Nieves Ruiz Arriola, Nucl. Phys. A [**679**]{}, | heng and Y. R. Liu, arXiv:1905.08605 \[hep-ph\]. C. W. XIao, J. Nieves And E. OSet, ArXIv:1904.01296 \[Hep-pH\]. J. J. WU, R. Molina, E. Oset aND B. S. ZOu, Phys. Rev. C [**84**]{}, 015202 (2011) \[arXiv:1011.2399 \[nucl-th\]\].
C. w. Shen, j. J. wU and b. s. ZOu, arXIv:1906.03896 \[hep-ph\]. n. isGUR anD M. b. WIse, phYS. LEtt. B [**232**]{}, 113 (1989). M. neuBert, PhyS. Rept. [**245**]{}, 259 (1994) \[hep-ph/9306320\]. a.V. MAnOhar and M.B. WisE, [*heAvy Quark PhYsiCs*]{}, Cambridge MOnoGraphs On parTIcle PHysIcs, NuClear PHYsics aNd CosmoloGy, VOl. 10 (CambRIdge UniVERsIty PRess, Cambridge, EnglANd, 2000).
k. c. Bowler [*et al.*]{} \[UKQcD CollAbORaTIOn\], PHys. rev. D [**54**]{}, 3619 (1996) doi:10.1103/PhySREvD.54.3619 \[heP-Lat/9601022\]. C. AlbERtUS, j. e. AmARo, E. Hernandez aNd J. Nieves, NuCL. PhYs. A [**740**]{}, 333 (2004) doi:10.1016/J.nUclPHysa.2004.04.114 \[nuCl-th/0311100\]. M. taNAbaShi [*et al.*]{} \[PartIcle data Group\], phys. ReV. d [**98**]{}, no. 3, 030001 (2018). doi:10.1103/PHYsRevD.98.030001
M. e. Rose, ELemEntAry THEoRy Of ANgULar mOmEntUM, JoHn Wiley, 1957.
J. niEvEs and e. RuiZ aRRIola, nucL. PhyS. A [**679**]{}, 57 (2000) doi:10.1016/s0375-9474(00)00321-3 | heng and Y. R. Liu, arXiv: 1905.08605 \[he p-p h\] .C. W . Xi ao, J. Nievesa nd E . Oset, arXiv:1904.012 96 \[ he p -ph\ ] .J. J. Wu, R. Mo l i na, E .Ose ta nd B. S . Z ou, Phy s. Rev. C[** 84 **]{}, 01520 2 ( 2011) \[ar Xiv :1011.2399 \ [nu cl-th\ ]\ ].C. W. Sh en, J . J. W u and B . S. Zou, a r Xiv:19 0 6.03896 \ [h ep-p h\]. N. Isgur and M. B. Wise, Phys. Lett. B [* * 2 32* *]{ }, 113 (19 89 ). M. Neubert , P h y s . R e pt. [**245** ]{}, 259 (1 9 94) \[hep -p h/9 3 06320\ ]. A. V. Man ohar and M. B. W ise, [*He avy Qu a rk Phys i cs*]{}, Cambr idg e M onog r ap hs on P a rti c le Ph y sic s, Nucle ar P hysic s an d C o smol ogy , vo l. 10 (Cambridge U niv ersi t y P ress, Camb ridg e, Engl and, 2 000).
K. C. Bowler [* et a l.*]{} \[ UKQ CD Co ll abora t ion\], Ph ys. Rev. D [**54* * ]{} ,3 6 1 9(1996) doi:10.1103 /P h y sR evD.54.3 619 \[ h ep -l a t/960102 2\ ].C. A l b ertus , J. E. Amaro,E. Her n an de z and J .Nieves ,Nuc l.Phys. A [* *740** ]{}, 333 (200 4 ) doi:10.1016/ j .nuclphysa.20 0 4. 0 4 .1 1 4 \[ nuc l-th/031110 0\]. M. T anab a sh i [ * et al .*]{} \ [ Pa r ticle Data Group\], P hys. R ev. D [**98**]{},no. 3, 030 0 0 1 (2018). doi : 10 . 1103/PhysRevD. 98.03 0001
M. E . Rose, E lemen tary The ory of An g u lar Mome ntu m,Joh n W i l ey , 1957.
J. N i e vesan d E. Ru izArriola , N ucl . P hys .A [**679* *]{}, 57 ( 20 00 )doi :10.1 0 16/S0375 -9 474 (0 0)0 0321- 3 | heng and_Y. R. Liu, arXiv:1905.08605_\[hep-ph\]. C. W. Xiao, J. Nieves and_E. Oset, arXiv:1904.01296_\[hep-ph\]._J. J. Wu, R. Molina,_E. Oset_and B. S. Zou, Phys. Rev. C_[**84**]{}, 015202 (2011)_\[arXiv:1011.2399 \[nucl-th\]\].
C. W. Shen, J. J. Wu and_B. S. Zou, arXiv:1906.03896 \[hep-ph\]._N. Isgur_and M. B. Wise, Phys. Lett. B [**232**]{}, 113 (1989). M. Neubert, Phys. Rept. [**245**]{}, 259 (1994) \[hep-ph/9306320\]. A.V. Manohar_and_M.B. Wise,_[*Heavy_Quark_Physics*]{}, Cambridge Monographs on Particle_Physics, Nuclear Physics and Cosmology,_vol. 10_(Cambridge University Press, Cambridge, England, 2000).
K. C. Bowler [*et al.*]{}_\[UKQCD_Collaboration\], Phys. Rev. D [**54**]{},_3619 (1996) doi:10.1103/PhysRevD.54.3619 \[hep-lat/9601022\]. C. Albertus, J. E. Amaro, E. Hernandez and J. Nieves,_Nucl. Phys. A [**740**]{}, 333 (2004) doi:10.1016/j.nuclphysa.2004.04.114 \[nucl-th/0311100\]._M. Tanabashi [*et al.*]{}_\[Particle_Data_Group\], Phys. Rev. D [**98**]{}, no._3, 030001 (2018). doi:10.1103/PhysRevD.98.030001
M. E. Rose, Elementary Theory_of Angular Momentum, John Wiley, 1957.
J. Nieves_and E. Ruiz Arriola, Nucl. Phys. A [**679**]{}, 57 (2000)_doi:10.1016/S0375-9474(00)00321-3 |
, Ueber die durch collineare Grundgebilde erzeugten Curven und Flächen, [*Math. Ann.*]{}, [**18**]{} (1881), 1-32.
[\[Schw1\]]{} R.L.E.Schwarzenberger, Vector bundles on the projective plane, [*Proc. London Math. Soc.*]{}, [**11**]{} (1961), 623-640.
[\[Schw2\]]{} R.L.E.Schwarzenberger, The secant bundle of a projective variety, [*Proc. London Math. Soc.*]{}, [**14**]{} (1964), 369-384.
[\[T\]]{} A.N.Tyurin, On the classification of two - dimensional vector bundles over algebraic curves of arbitrary genus, [*Izv. Akad. Nauk SSSR, ser. Math.*]{}, [**28**]{} (1964), 21-52 (in Russian).
Authors’ addresses:
I.D.: Department of Mathematics, University of Michigan, Ann Arbor MI 48109, email: [email protected]
.3cm
M.K.: Department of Mathematics, Northwestern University, Evanston IL 60208, email: [email protected]
[^1]: \*
---
abstract: 'The magnetic ground state in highly ordered double perovskites LaSr$_{1-x}$Ca$_x$NiReO$_6$ ($x$ = 0.0, 0.5, 1.0) were studied in view of the Goodenough-Kanamori rules of superexchange interactions. In LaSrNiReO$_6$, Ni and Re sublattices are found to exhibit curious magnetic states, but do not show any long range magnetic ordering. The magnetic transition at $\sim$ 255 K is identified with the Re sublattic magnetic ordering. The sublattice interactions are tuned by modifying the Ni-O-Re bond angles via changing the lattice structure through Ca doping. Upon Ca doping, the Ni and Re sublattices start to display a ferrimagnetically ordered state at low temperature. The neutron powder diffraction reveals a canted alignment between the Ni and the Re sublattices, while the individual sublattice is ferromagnetic. The transition temperature of the ferrimagnetic phase increases monotonically with increasing Ca concentration.'
author:
- Somnath Jana
- Pay | , Ueber die durch collineare Grundgebilde erzeugten Curven und Flächen, [ * Math. Ann. * ] { }, [ * * 18 * * ] { } (1881), 1 - 32.
[ \[Schw1\ ] ] { } R.L.E.Schwarzenberger, Vector bundles on the projective plane, [ * Proc. London Math. Soc. * ] { }, [ * * 11 * * ] { } (1961), 623 - 640.
[ \[Schw2\ ] ] { } R.L.E.Schwarzenberger, The secant bundle of a projective variety, [ * Proc. London Math. Soc. * ] { }, [ * * 14 * * ] { } (1964), 369 - 384.
[ \[T\ ] ] { } A.N.Tyurin, On the categorization of two - dimensional vector pile over algebraic curves of arbitrary genus, [ * Izv. Akad. Nauk SSSR, ser. Math. * ] { }, [ * * 28 * * ] { } (1964), 21 - 52 (in Russian).
Authors ’ address:
I.D.: Department of Mathematics, University of Michigan, Ann Arbor MI 48109, electronic mail: [email protected]
.3 cm
M.K.: Department of Mathematics, Northwestern University, Evanston IL 60208, email: [email protected]
[ ^1 ]: \ *
---
abstract:' The magnetic ground state of matter in highly ordered double perovskites LaSr$_{1 - x}$Ca$_x$NiReO$_6 $ ($ x$ = 0.0, 0.5, 1.0) were analyze in view of the Goodenough - Kanamori rules of superexchange interaction. In LaSrNiReO$_6 $, Ni and Re sublattices are found to exhibit curious charismatic states, but do not show any long range magnetic ordering. The magnetic passage at $ \sim$ 255 K is identified with the Re sublattic magnetic ordering. The sublattice interaction are tuned by modifying the Ni - O - Re chemical bond angles via changing the lattice structure through Ca doping. Upon Ca dope, the Ni and Re sublattices start to display a ferrimagnetically ordered state at low temperature. The neutron powder diffraction reveals a canted alignment between the Ni and the Re sublattices, while the individual sublattice is ferromagnetic. The transition temperature of the ferrimagnetic phase increases monotonically with increasing Ca concentration.'
author:
- Somnath Jana
- Pay | , Ueher die durch collineare Grundgebilde etzwugten Curveh und Fläzhen, [*Math. Ann.*]{}, [**18**]{} (1881), 1-32.
[\[Schw1\]]{} R.L.E.Schwerzebbergtg, Vector bundles on tfe projecnive planw, [*Prix. London Mefh. Soc.*]{}, [**11**]{} (1961), 623-640.
[\[Schw2\]]{} V.L.E.Sckwerzenberger, The secant butdle of a projacgire variety, [*Proc. London Math. Soc.*]{}, [**14**]{} (1964), 369-384.
[\[T\]]{} W.N.Tyurim, Ln the classifycatpog of nwi - dimensional vector bundles over angebraic curvrs of arbitrary genus, [*Izv. Wkad. Nauk SSSR, ser. Matj.*]{}, [**28**]{} (1964), 21-52 (in Ruswian).
Wythors’ addrerses:
I.D.: Department of Mzthematics, University of Michigxn, Anu Arbor MI 48109, enaip: [email protected]
.3cm
M.K.: Departmant of Kathematics, Novthwevteen University, Evanstoi IL 60208, email: [email protected]
[^1]: \*
---
abstract: 'The magberic gtound stagw iv hjgily orderfd vouble perobskites LaSe$_{1-x}$Ca$_x$NiReO$_6$ ($x$ = 0.0, 0.5, 1.0) wete wtudied in visw of ehq Goodenough-Kanamori rules of superexchdngs interactions. In LaSrNuReO$_6$, Ni and Re sublatjices are sound to exhibit curious magnetic states, but do nmt shkd aun uinh range magnetic ordering. The magnetic transiejom st $\sim$ 255 K is ibentified with tne Rr sublattic maenetic orsering. The sublattlce intgractiins are ttned by modifying the Ni-O-Re bone angles via xhanging the lattiee structure thrpugh Va doping. Upon Ca dopiny, the Hi and Re skblatticea start to displah a farrimagntgically ordered seate at liw tzmperatufe. Tne neueron powdeg difnsaction reveals a fanteb alicnment betaeen the Ni and the Re sublattirxs, while the ongivpdual subjattige is ferromagnqtic. The transntion teiperagure of thv ferrimajnetic phase increases motltonically wmth increwsint Ca concengfation.'
author:
- Spmnath Jauc
- Pay | , Ueber die durch collineare Grundgebilde erzeugten Flächen, Ann.*]{}, [**18**]{} 1-32. [\[Schw1\]]{} R.L.E.Schwarzenberger, plane, London Math. Soc.*]{}, (1961), 623-640. [\[Schw2\]]{} The secant bundle of a projective [*Proc. London Math. Soc.*]{}, [**14**]{} (1964), 369-384. [\[T\]]{} A.N.Tyurin, On the classification of - dimensional vector bundles over algebraic curves of arbitrary genus, [*Izv. Akad. Nauk ser. [**28**]{} 21-52 Russian). Authors’ addresses: I.D.: Department of Mathematics, University of Michigan, Ann Arbor MI 48109, email: [email protected] M.K.: Department of Mathematics, Northwestern University, Evanston IL email: [email protected] [^1]: \* abstract: 'The magnetic ground state highly double perovskites ($x$ 0.0, 1.0) were studied view of the Goodenough-Kanamori rules of superexchange interactions. In LaSrNiReO$_6$, Ni and Re sublattices are found to curious magnetic do not any range ordering. The magnetic $\sim$ 255 K is identified with magnetic ordering. The sublattice interactions are tuned by the Ni-O-Re angles via changing the lattice structure Ca doping. Upon Ca doping, the Ni and sublattices start to display a ferrimagnetically ordered state at low temperature. The neutron powder diffraction canted alignment between the and the Re while individual is The transition of the ferrimagnetic phase increases monotonically with increasing Ca concentration.' author: Somnath Jana - Pay | , Ueber die durch collineare GrUndgebilde ErzeuGteN CuRvEn unD FläChen, [*Math. Ann.*]{}, [**18**]{} (1881), 1-32.
[\[SchW1\]]{} r.L.E.SChwarzenberger, Vector buNdles On THe prOJeCtive Plane, [*PrOC. LONDon maTh. soc.*]{}, [**11**]{} (1961), 623-640.
[\[scHW2\]]{} R.l.E.SchWarZenbergEr, The secanT buNdLe of a projectIVe Variety, [*ProC. LoNdon Math. Soc.*]{}, [**14**]{} (1964), 369-384.
[\[T\]]{} a.N.TYurin, ON tHe cLAssifIcaTion oF two - diMEnsionAl vector bUnDLes oveR AlgebraIC CuRves Of arbitrary genus, [*IZV. AKAd. Nauk SSSR, ser. MAth.*]{}, [**28**]{} (1964), 21-52 (in RUsSIaN).
aUthOrs’ Addresses:
I.d.: DEpartMEnt of MaTHeMATIcs, uNiversity of MiChigan, Ann ArBOr Mi 48109, email: [email protected]
.3cm
M.K.: DeparTmenT of MathemAtics, NORthwestERn UniveRsity, EVanStoN IL 60208, eMAiL: [email protected]
[^1]: \*
---
abstraCt: 'thE magnEtic GROUNd stAte In hiGhly oRdered double pEroVskiTEs LASr$_{1-x}$CA$_x$NiREO$_6$ ($x$ = 0.0, 0.5, 1.0) wErE studIed in vIew of ThE Goodenough-KanaMori Rules of suPerExChaNgE inteRActionS. In laSRNiReO$_6$, NI and Re sUBlaTtICES aRe found to exhibit cuRiOUS mAgnetic sTates, bUT dO nOT show any LoNg rAnge MAGnetiC ordERiNg. The magNetic tRAnSiTion at $\sIm$ 255 k is ideNtIfiEd wIth thE re suBlattiC magnetiC ordeRIng. The sublattiCE interactions ARe TUNeD By moDifYing the Ni-O-RE bonD AnglEs viA ChAngINg the LattiCe STrUCture through Ca dopinG. UPon Ca dOping, The Ni and Re subLattices stART To displaY a feRRiMAgnetically ordEred sTate at low tEMperaturE. The nEutron poWder diffrACTion reveAls A caNteD alIGNmEnt between the nI And tHe re sublaTtiCes, whilE thE inDivIduAl SublatticE is ferroMaGnEtIc. the TransITion tempErAtuRe Of tHe ferRImagneTic phAse iNcReASes MonotonICaLLY witH iNcReasIng ca ConceNtraTIon.'
Author:
- SOmnath JanA
- PaY | , Ueber die durch collinea re Grundge bilde er zeu gt en C urve n und Flächen, [*Ma th. Ann.*]{}, [**18**] {} (1 88 1 ), 1 - 32 .
[\ [Schw1\ ] ]{ } R.L .E .S chw ar z en berge r,Vectorbundles on th eprojective p l an e, [*Proc. Lo ndon Math. S oc. *]{},[* *11 * *]{}(19 61),623-64 0 .
[\[ Schw2\]]{ }R .L.E.S c hwarzen b e rg er,The secant bundle of a projective v ariety ,[ *P r o c.Lon don Math.So c.*]{ } , [**14 * *] { } (19 6 4), 369-384.
[\[T\]]{}A .N. Tyurin ,Ont he cla ssifi ca t ion of two - d imen sional ve ctor b u ndles o v er alge braiccur ves ofa rb it rar yg enu s ,[*I z v.Akad. Na uk S SSR,ser. M a t h.*] {}, [** 28**] {} (1964), 21 -52 (in Rus sian) .
Au thor s’ addr esses:
I.D .: Department ofMath ematics,Uni ve rsi ty of M i chigan , A nnArbor M I 48109 , em ai l : IG [email protected] .u m i ch .edu
.3 cm
M. K .: D e partment o f M athe m a tics, Nor t hw estern U nivers i ty ,Evansto nIL 602 08 , e mai l: ka p rano v@chow .math.nw u.edu
[^1]: \*
--- abstract: 'Th e m a g ne t ic g rou nd state in hig h ly o rder e ddou b le pe rovsk it e sL aSr$_{1-x}$Ca$_x$Ni Re O$_6$($x$= 0.0, 0.5, 1 .0) were s t u d ied in v iewo ft he Goodenough- Kanam ori ruleso f supere xchan ge inter actions.I n LaSrNiR eO$ _6$ , N i a n d R e sublattices a re f ou nd to e xhi bit cur iou s m agn eti cstates, b ut do no tsh ow a nylongr ange mag ne tic o rde ring. The ma gneti c tr an si t ion at $\s i m$ 2 55 K i siden tif ie d wit h th e Re sublat tic magne tic orde ri ng . The s ublattice int er actions ar etun ed bym o difyingthe Ni-O-Re bond angles via cha ngi ng th e la ttice str uct ure th rou g h Ca d oping. Upon C a d o p ing,t h eNian d Re subla t t ice s sta rt todisplay a ferrimagnetical l y o rdered stateatlowt e mp era t ur e . T he neu t r on powder diffr action rev ea l sa canted a l ign me nt betw een the Ni a n d the R e sublatt ices, whi le the i ndi vidual sub latticeis ferrom a gneti c .The t ran sition t emp eratu re oft heferri magnet ic phase incr ea ses mono tonically with increasi ng Caconce ntr ation.'
a uth o r:- Somnath Jan a
- Pay | , Ueber_die durch_collineare Grundgebilde erzeugten Curven_und Flächen,_[*Math._Ann.*]{}, [**18**]{}_(1881),_1-32.
[\[Schw1\]]{} R.L.E.Schwarzenberger, Vector_bundles on the_projective plane, [*Proc. London_Math. Soc.*]{}, [**11**]{}_(1961),_623-640.
[\[Schw2\]]{} R.L.E.Schwarzenberger, The secant bundle of a projective variety, [*Proc. London Math. Soc.*]{}, [**14**]{}_(1964),_369-384.
[\[T\]]{} A.N.Tyurin,_On_the_classification of two - dimensional_vector bundles over algebraic curves_of arbitrary_genus, [*Izv. Akad. Nauk SSSR, ser. Math.*]{}, [**28**]{}_(1964),_21-52 (in Russian).
Authors’_addresses:
I.D.: Department of Mathematics, University of Michigan, Ann Arbor_MI 48109, email: [email protected]
.3cm
M.K.: Department of_Mathematics, Northwestern University,_Evanston_IL_60208, email: [email protected]
[^1]: \*
_---
abstract: 'The magnetic ground state in_highly ordered double perovskites LaSr$_{1-x}$Ca$_x$NiReO$_6$ ($x$_= 0.0, 0.5, 1.0) were studied in_view of the Goodenough-Kanamori rules of_superexchange interactions. In LaSrNiReO$_6$, Ni_and Re_sublattices are found to exhibit_curious magnetic states,_but do_not show any_long range magnetic ordering. The magnetic_transition at $\sim$_255 K is identified with the_Re_sublattic magnetic ordering._The_sublattice_interactions are_tuned by modifying_the_Ni-O-Re bond_angles_via changing the lattice structure through_Ca_doping. Upon Ca doping, the Ni and_Re sublattices start to_display_a ferrimagnetically ordered state_at low temperature. The neutron_powder diffraction reveals a canted alignment_between the_Ni and_the Re sublattices, while the individual sublattice is ferromagnetic. The transition_temperature of the ferrimagnetic phase increases_monotonically with increasing Ca_concentration.'
author:
- Somnath_Jana
-_Pay |
}^{\prime }$ that can be obtained with the same raw subtraction in the usual non-background field approach.
The usual approach is retrieved by picking a gauge fermion $\Psi ^{\prime }$ that depends on $\Phi +{\mkern2mu\underline{\mkern-2mu\smash{\Phi }\mkern-2mu}\mkern2mu }$, such as $$\Psi ^{\prime }(\Phi +{\mkern2mu\underline{\mkern-2mu\smash{\Phi }\mkern-2mu}\mkern2mu })=\int \bar{C}^{I}G^{Ii}(0,\partial
)(\phi ^{i}+{\mkern2mu\underline{\mkern-2mu\smash{\phi }\mkern-2mu}\mkern2mu }^{i}). \label{gfno}$$ Making the canonical transformation generated by $$F_{\text{gf}}^{\prime }(\Phi,{\mkern2mu\underline{\mkern-2mu\smash{\Phi }\mkern-2mu}\mkern2mu },K^{\prime },{\mkern2mu\underline{\mkern-2mu\smash{K}\mkern-2mu}\mkern2mu }^{\prime })=\int \ \Phi ^{\alpha }K_{\alpha }^{\prime }+\int \ {\mkern2mu\underline{\mkern-2mu\smash{\Phi }\mkern-2mu}\mkern2mu }^{\alpha }{\mkern2mu\underline{\mkern-2mu\smash{K}\mkern-2mu}\mkern2mu }_{\alpha }^{\prime }+\Psi ^{\prime }(\Phi +{\mkern2mu\underline{\mkern-2mu\smash{\Phi }\mkern-2mu}\mkern2mu }) \label{backgfgen2}$$ on (\[sback\]) we find the classical action $$S^{\prime }(\Phi,{\mkern2mu\underline{\mkern-2mu\smash{\Phi }\mkern-2mu}\mkern2mu },K,{\mkern2mu\underline{\mkern-2mu\smash{K}\mkern-2mu}\mkern2mu })=\hat{S}^{\prime }(\Phi +{\mkern2mu\underline{\mkern-2mu\smash{\Phi }\mkern-2mu}\mkern2mu | } ^{\prime } $ that can be obtained with the same raw subtraction in the usual non - background sphere overture.
The usual approach is retrieved by pick a gauge fermion $ \Psi ^{\prime } $ that depends on $ \Phi + { \mkern2mu\underline{\mkern-2mu\smash{\Phi } \mkern-2mu}\mkern2mu } $, such as $ $ \Psi ^{\prime } (\Phi + { \mkern2mu\underline{\mkern-2mu\smash{\Phi } \mkern-2mu}\mkern2mu }) = \int \bar{C}^{I}G^{Ii}(0,\partial
) (\phi ^{i}+{\mkern2mu\underline{\mkern-2mu\smash{\phi } \mkern-2mu}\mkern2mu } ^{i }). \label{gfno}$$ have the canonical transformation generated by $ $ F_{\text{gf}}^{\prime } (\Phi,{\mkern2mu\underline{\mkern-2mu\smash{\Phi } \mkern-2mu}\mkern2mu }, K^{\prime }, { \mkern2mu\underline{\mkern-2mu\smash{K}\mkern-2mu}\mkern2mu } ^{\prime }) = \int \ \Phi ^{\alpha } K_{\alpha } ^{\prime } + \int \ { \mkern2mu\underline{\mkern-2mu\smash{\Phi } \mkern-2mu}\mkern2mu } ^{\alpha } { \mkern2mu\underline{\mkern-2mu\smash{K}\mkern-2mu}\mkern2mu } _ { \alpha } ^{\prime } + \Psi ^{\prime } (\Phi + { \mkern2mu\underline{\mkern-2mu\smash{\Phi } \mkern-2mu}\mkern2mu }) \label{backgfgen2}$$ on (\[sback\ ]) we recover the classical action $ $ S^{\prime } (\Phi,{\mkern2mu\underline{\mkern-2mu\smash{\Phi } \mkern-2mu}\mkern2mu }, K,{\mkern2mu\underline{\mkern-2mu\smash{K}\mkern-2mu}\mkern2mu }) = \hat{S}^{\prime } (\Phi + { \mkern2mu\underline{\mkern-2mu\smash{\Phi } \mkern-2mu}\mkern2mu | }^{\prile }$ that can be obtained with the same taq subtcaction in the jsual non-background field ap'roaxh.
The usual approach is retfieved by picking a geuge fermion $\Psi ^{\prime }$ bkat dsiends in $\Phi +{\mkern2mu\onderline{\mkesn-2mu\smash{\Phi }\mnefn-2lu}\mkern2mu }$, such as $$\Psi ^{\prime }(\Phi +{\mkqrn2mu\uncegline{\mkern-2mu\smwsh{\Pny }\mksgn-2nu}\mkern2mu })=\int \bar{C}^{I}G^{Ii}(0,\partial
)(\lhi ^{i}+{\mktrn2mu\underline{\mkerm-2mu\smash{\phi }\mkern-2mu}\mkern2mu }^{i}). \pabel{gfno}$$ Making tje canonicao trwbsformation eenerated by $$F_{\text{gf}}^{\prjme }(\Phi,{\mkern2mu\underline{\mkern-2mu\soash{\Pki }\mkern-2mu}\mjeen2mk },K^{\prime },{\mkecn2mu\unqerline{\mkern-2mu\smash{K}\kkern-2mu}\kkern2mu }^{\prime })=\lnt \ \'hi ^{\qlpha }K_{\alpha }^{\prime }+\inv \ {\mkern2mu\underline{\mhern-2mu\smavh{\'hi }\mkern-2mu}\mkern2mu }^{\alphq }{\mketn2mu\utderuune{\okedn-2ku\amash{K}\lkecn-2mu}\mkern2mu }_{\alpha }^{\primw }+\Psi ^{\prime }(\Phi +{\mketn2il\inderline{\mkedn-2mu\smwsr{\Phi }\mkern-2mu}\mkern2mu }) \label{backgfgen2}$$ on (\[sbzck\]) we find the classicql action $$S^{\prime }(\Phi,{\mnern2mu\undqrline{\mkern-2mu\smash{\Phi }\mkern-2mu}\mkern2mu },K,{\mkern2mu\undesline{\jyeru-2nu\smary{K}\lkern-2mu}\mkern2mu })=\hat{S}^{\prime }(\Phi +{\mkern2mu\underline{\iietn-2ku\smash{\Phi }\mkevn-2mu}\mkern2mu | }^{\prime }$ that can be obtained with raw in the non-background field approach. by a gauge fermion ^{\prime }$ that on $\Phi +{\mkern2mu\underline{\mkern-2mu\smash{\Phi }\mkern-2mu}\mkern2mu }$, such $$\Psi ^{\prime }(\Phi +{\mkern2mu\underline{\mkern-2mu\smash{\Phi }\mkern-2mu}\mkern2mu })=\int \bar{C}^{I}G^{Ii}(0,\partial )(\phi ^{i}+{\mkern2mu\underline{\mkern-2mu\smash{\phi }\mkern-2mu}\mkern2mu }^{i}). \label{gfno}$$ Making canonical transformation generated by $$F_{\text{gf}}^{\prime }(\Phi,{\mkern2mu\underline{\mkern-2mu\smash{\Phi }\mkern-2mu}\mkern2mu },K^{\prime },{\mkern2mu\underline{\mkern-2mu\smash{K}\mkern-2mu}\mkern2mu }^{\prime })=\int \ \Phi }K_{\alpha }+\int {\mkern2mu\underline{\mkern-2mu\smash{\Phi }^{\alpha }{\mkern2mu\underline{\mkern-2mu\smash{K}\mkern-2mu}\mkern2mu }_{\alpha }^{\prime }+\Psi ^{\prime }(\Phi +{\mkern2mu\underline{\mkern-2mu\smash{\Phi }\mkern-2mu}\mkern2mu }) \label{backgfgen2}$$ on (\[sback\]) we find the action $$S^{\prime }(\Phi,{\mkern2mu\underline{\mkern-2mu\smash{\Phi }\mkern-2mu}\mkern2mu },K,{\mkern2mu\underline{\mkern-2mu\smash{K}\mkern-2mu}\mkern2mu })=\hat{S}^{\prime }(\Phi +{\mkern2mu\underline{\mkern-2mu\smash{\Phi | }^{\prime }$ that can be obtained witH the same raW subtRacTioN iN the UsuaL non-background FIeld Approach.
The usual approaCh is rEtRIeveD By PickiNg a gaugE FeRMIon $\psI ^{\pRimE }$ tHAt DepenDs oN $\Phi +{\mkeRn2mu\underlIne{\MkErn-2mu\smash{\PhI }\MkErn-2mu}\mkern2Mu }$, sUch as $$\Psi ^{\primE }(\PhI +{\mkern2Mu\UndERline{\MkeRn-2mu\sMash{\PhI }\Mkern-2mU}\mkern2mu })=\iNt \BAr{C}^{I}G^{II}(0,\Partial
)(\PHI ^{i}+{\MkerN2mu\underline{\mkern-2MU\sMAsh{\phi }\mkern-2mu}\mKern2mu }^{I}). \lABeL{GFno}$$ makIng the canoNiCal trANsformaTIoN GENerATed by $$F_{\text{gf}}^{\pRime }(\Phi,{\mkerN2Mu\uNderliNe{\MkeRN-2mu\smaSh{\Phi }\MkERn-2mU}\mkern2mu },K^{\prIme },{\mKern2mu\undErline{\MKern-2mu\sMAsh{K}\mkeRn-2mu}\mkErn2Mu }^{\pRime })=\INt \ \phI ^{\alPhA }k_{\alPHa }^{\PriME }+\inT \ {\mkern2mu\UnDeRline{\MkerN-2MU\SMash{\phi }\MkerN-2mu}\mkErn2mu }^{\alpha }{\mkeRn2mU\undERliNe{\mkeRn-2mu\sMash{k}\mKern-2mU}\mkern2Mu }_{\alpHa }^{\Prime }+\Psi ^{\prime }(\PhI +{\mkeRn2mu\underLinE{\mKerN-2mU\smasH{\phi }\mkeRn-2mU}\mkErn2mu }) \laBel{backGFgeN2}$$ oN (\[SBAcK\]) we find the classicaL aCTIoN $$S^{\prime }(\PHi,{\mkerN2Mu\UnDErline{\mkErN-2mu\SmasH{\pHi }\mkeRn-2mu}\MKeRn2mu },K,{\mkeRn2mu\unDErLiNe{\mkern-2Mu\Smash{K}\MkErn-2Mu}\mKern2mU })=\Hat{S}^{\Prime }(\PHi +{\mkern2mU\undeRLine{\mkern-2mu\smaSH{\Phi }\mkern-2mu}\mkERn2MU | }^{\prime }$ that can be o btained wi th th e s ame r aw s ubtr action in theu sual non-background fieldappro ac h .
T h eusual approa c hi s re tr ie ved b y p ickin g a gaugefermion $\ Psi ^ {\prime }$ t h at depends o n $ \Phi +{\mker n2m u\unde rl ine { \mker n-2 mu\sm ash{\P h i }\mk ern-2mu}\ mk e rn2mu} $, such a s$$\P si ^{\prime }(\Ph i + { \mkern2mu\unde rline{ \m k er n - 2mu \sm ash{\Phi } \m kern- 2 mu}\mke r n2 m u })= \ int \bar{C}^{ I}G^{Ii}(0, \ par tial
) (\ phi ^{i}+{ \mker n2 m u\u nderline{\m kern -2mu\smas h{\phi }\mkern - 2mu}\mk ern2mu }^ {i} ). \ la be l{g fn o }$$ Ma kin g th e canoni ca ltrans form a t i o n ge ner ated by $ $F_{\text{gf} }^{ \pri m e } (\Phi ,{\mk ern2 mu \unde rline{ \mker n- 2mu\smash{\Phi}\mk ern-2mu}\ mke rn 2mu } ,K^{\ p rime } ,{\ mke rn2mu\u nderlin e {\m ke r n - 2m u\smash{K}\mkern-2 mu } \ mk ern2mu } ^{\pri m e}) = \int \ \ Ph i ^ {\al p h a }K_ {\al p ha }^{\pri me }+\ i nt \ {\mker n2 mu\und er lin e{\ mkern - 2mu\ smash{ \Phi }\m kern- 2 mu}\mkern2mu } ^ {\alpha }{\mk e rn 2 m u\ u nder lin e{\mkern-2m u\sm a sh{K }\mk e rn -2m u }\mke rn2mu } _ {\ a lpha }^{\prime }+\P si ^{\pr ime } (\Phi +{\mker n2mu\under l i n e{\mkern -2mu \ sm a sh{\Phi }\mker n-2mu }\mkern2mu }) \lab el{ba ckgfgen2 }$$ on (\ [ s back\])wefin d t hec l as sical action$ $ S^{\ pr ime }(\ Phi ,{\mker n2m u\u nde rli ne {\mkern-2 mu\smash {\ Ph i}\ mke rn-2m u }\mkern2 mu }, K, {\m kern2 m u\unde rline {\mk er n- 2 mu\ smash{K } \m k e rn-2 mu }\ mker n2m u})=\h at{S } ^{\ prime } (\Phi +{\ mke r n2mu \u nd erline{ \mkern-2mu\sm as h{\Phi }\m ke rn- 2mu}\m k e rn2mu | }^{\prime }$_that can_be obtained with the_same raw_subtraction_in the_usual_non-background field approach.
The_usual approach is_retrieved by picking a_gauge fermion $\Psi_^{\prime_}$ that depends on $\Phi +{\mkern2mu\underline{\mkern-2mu\smash{\Phi }\mkern-2mu}\mkern2mu }$, such as $$\Psi ^{\prime }(\Phi +{\mkern2mu\underline{\mkern-2mu\smash{\Phi_}\mkern-2mu}\mkern2mu_})=\int \bar{C}^{I}G^{Ii}(0,\partial
)(\phi_^{i}+{\mkern2mu\underline{\mkern-2mu\smash{\phi_}\mkern-2mu}\mkern2mu_}^{i}). \label{gfno}$$ Making the_canonical transformation generated by $$F_{\text{gf}}^{\prime_}(\Phi,{\mkern2mu\underline{\mkern-2mu\smash{\Phi }\mkern-2mu}\mkern2mu_},K^{\prime },{\mkern2mu\underline{\mkern-2mu\smash{K}\mkern-2mu}\mkern2mu }^{\prime })=\int \ \Phi ^{\alpha }K_{\alpha_}^{\prime_}+\int \ {\mkern2mu\underline{\mkern-2mu\smash{\Phi_}\mkern-2mu}\mkern2mu }^{\alpha }{\mkern2mu\underline{\mkern-2mu\smash{K}\mkern-2mu}\mkern2mu }_{\alpha }^{\prime }+\Psi ^{\prime }(\Phi +{\mkern2mu\underline{\mkern-2mu\smash{\Phi_}\mkern-2mu}\mkern2mu }) \label{backgfgen2}$$ on (\[sback\])_we find the_classical_action_$$S^{\prime }(\Phi,{\mkern2mu\underline{\mkern-2mu\smash{\Phi }\mkern-2mu}\mkern2mu },K,{\mkern2mu\underline{\mkern-2mu\smash{K}\mkern-2mu}\mkern2mu_})=\hat{S}^{\prime }(\Phi +{\mkern2mu\underline{\mkern-2mu\smash{\Phi }\mkern-2mu}\mkern2mu |
the observed spectra within the resolution. The radial velocity values for these fits were taken from our orbital solution (see Table 5). The chosen parameters for the triple system give also a good fit to the metal line spectrum as is illustrated in Fig.\[fig-metal\]. Based on evolutionary tracks published by @Schaller, Fig.\[fig-tracks\] shows that a triple system with log(age) between 8.8 and 8.9 is compatible with our observations of.
{width="13cm"}
### Disentangled Spectra
The determination of the fundamental parameters was also performed on the disentangled spectra. A grid of model spectra ranging from 6000 to 8600K with stepwidth of 200 K and log g of 3.6, 4.0, 4.4, respectively, was compared to the disentangled spectra corrected for radial velocity shifts. The model spectra were calculated using the code SynthV [@Tsymbal96] and NEMO model atmosphere grid [@Heiter2002]. For finding the best fitting spectra two approaches have been used. First, the standard deviation of the residuals, hereafter referred to as $\sigma$, and second, the $\chi^2$ test were used for evaluating the quality of the fit. These approaches lead to different best fitting solutions, because the methods differ in weighting the data. The standard deviation of the residuals weights each data point equally, while the $\chi^2$ test weights the data according to the relative line strength. Hence the $\chi^2$ test prefers better fits of line cores. Solar chemical composition was assumed, based on the good fit obtained by the direct method (see Fig.\[fig-metal\]).
The standard deviation of the residuals and $\chi^2$ was calculated the following way:\
\
$\sigma = \sqrt{\frac{1}{N-1} \sum (R_i - \overline{R})^2 }$\
\
$\chi^2 = \sum \frac{(S_{scal} - D)^2}{S_{scal}}$\
\
where $S_{scal}$ means scaled synthetic spectrum and $D$ the disentangled spectrum and $R = S_{scal} - D$. The model spectra were scaled according to\
\
$S_{scal} = \frac{S f+1}{f+1}$
\
\
where $S$ denotes the model spectrum and $f$ | the observed spectra within the resolution. The radial velocity value for these fit were taken from our orbital solution (see Table 5). The choose parameters for the triple organization collapse also a full fit to the metal tune spectrum as is illustrated in Fig.\[fig - metal\ ]. Based on evolutionary track published by @Schaller, Fig.\[fig - tracks\ ] shows that a triple system with log(age) between 8.8 and 8.9 is compatible with our observations of.
! [ image](9996fy11.ps){width="13 cm " }
# # # Disentangled Spectra
The decision of the fundamental parameters was also do on the disentangled spectra. A grid of model spectra ranging from 6000 to 8600 K with stepwidth of 200 kelvin and log g of 3.6, 4.0, 4.4, respectively, was compared to the disentangled spectra corrected for radial velocity transformation. The model spectra were calculated using the code SynthV [ @Tsymbal96 ] and NEMO model atmosphere power system [ @Heiter2002 ]. For discover the best fitting spectra two approaches have been used. First, the standard deviation of the residuals, hereafter referred to as $ \sigma$, and second, the $ \chi^2 $ test were use for evaluating the quality of the burst. These access lead to different best fitting solution, because the method differ in weighting the datum. The standard deviation of the residuals weights each data point equally, while the $ \chi^2 $ test weights the data according to the relative line lastingness. Hence the $ \chi^2 $ trial prefers better fits of line cores. Solar chemical typography was assumed, based on the good fit obtained by the direct method (see Fig.\[fig - metal\ ]).
The standard diversion of the residuals and $ \chi^2 $ was calculated the following way:\
\
$ \sigma = \sqrt{\frac{1}{N-1 } \sum (R_i - \overline{R})^2 } $ \
\
$ \chi^2 = \sum \frac{(S_{scal } - D)^2}{S_{scal}}$\
\
where $ S_{scal}$ means scaled man-made spectrum and $ D$ the disentangled spectrum and $ roentgen = S_{scal } - D$. The model spectrum were scaled accord to\
\
$ S_{scal } = \frac{S f+1}{f+1}$
\
\
where $ S$ denotes the model spectrum and $ f$ | thf observed spectra withik the resolution. The ravial vemocity vxlues for these fits were taneb fron our orbital solution (see Tablv 5). The chisen parameters for the tripls syscen give also a nood fit to the metal lina rpzctrum as is illustrated in Fig.\[fig-meeal\]. Basrd on evolutionaty trssks llboished by @Schaller, Fig.\[fig-tracis\] showv that a tripke system with log(age) betwfen 8.8 and 8.9 is compatiblf with our ibsewcations of.
{widuh="13em"}
### Disentanfled Spectra
The determination ow the fundamentql pagdmeters was also performed ok the dhsentanbled spectra. A grmd od model spectra rangiig from 6000 to 8600K with sjepwidth ox 200 K and log g of 3.6, 4.0, 4.4, eespewtivaly, dqs zomkarxd fo the diaentangled spectra coerected for radial fejicity shifts. Fhe moqej spectra were calculated using the codt SynfhV [@Tsymbal96] and NEMO moeel atmosphere grid [@Hgiter2002]. For sinding the best fitting spectra two approaches hdve bxev uweq. Digst, the standard deviation of the residuals, hqdesfner referred to af $\sigma$, and sfcpgd, the $\chi^2$ tert werz uaed for evaluating the quwlity of the fyt. Tnese approaches lead to difderent best yitring solutions, beccuse the metkods doffer in weighting the data. Che stzndard deviwtion of ffe residuals weiehtx aach data point equally, whyle the $\cii^2$ text weigfts jhe datw accordinh to bve relative line shrengjh. Henwe the $\chi^2$ test prefers better fits of liix cores. Solar cvempcal compjsitipn was assumeq, based on the good fnt obtxined by tge dirert method (seq Fig.\[fig-metal\]).
Dje standard veviation of rhe eesiduaur and $\chi^2$ was valculateb the following way:\
\
$\sigme = \sddt{\frac{1}{N-1} \sum (R_i - \ovtroine{R})^2 }$\
\
$\chi^2 = \sum \frxc{(S_{fcwl} - Q)^2}{V_{scal}}$\
\
where $S_{vcal}$ mexms scxled synthebic spevtrum and $D$ the disettanfled spectrum and $T = S_{scal} - E$. The moqel spectra wrre scaled accordijg to\
\
$V_{scel} = \frsc{S f+1}{f+1}$
\
\
where $S$ denotes the model slectrum ajd $n$ | the observed spectra within the resolution. The values these fits taken from our The parameters for the system give also good fit to the metal line as is illustrated in Fig.\[fig-metal\]. Based on evolutionary tracks published by @Schaller, Fig.\[fig-tracks\] that a triple system with log(age) between 8.8 and 8.9 is compatible with observations {width="13cm"} Disentangled The determination of the fundamental parameters was also performed on the disentangled spectra. A grid of spectra ranging from 6000 to 8600K with stepwidth 200 K and log of 3.6, 4.0, 4.4, respectively, compared the disentangled corrected radial shifts. The model were calculated using the code SynthV [@Tsymbal96] and NEMO model atmosphere grid [@Heiter2002]. For finding the best spectra two been used. the deviation the residuals, hereafter as $\sigma$, and second, the $\chi^2$ for evaluating the quality of the fit. These lead to best fitting solutions, because the methods in weighting the data. The standard deviation of residuals weights each data point equally, while the $\chi^2$ test weights the data according to line strength. Hence the test prefers better of cores. chemical was assumed, on the good fit obtained by the direct method (see Fig.\[fig-metal\]). standard deviation of the residuals and $\chi^2$ was calculated the \ = \sqrt{\frac{1}{N-1} \sum - \overline{R})^2 }$\ \ = \frac{(S_{scal} - D)^2}{S_{scal}}$\ \ means synthetic the spectrum $R = S_{scal} - The model spectra were scaled to\ \ $S_{scal} = $S$ denotes the model spectrum and $f$ | the observed spectra within tHe resolutiOn. The RadIal VeLociTy vaLues for these fiTS werE taken from our orbital soLutioN (sEE TabLE 5). THe choSen paraMEtERS foR tHe TriPlE SyStem gIve Also a goOd fit to the MetAl Line spectrum AS iS illustratEd iN Fig.\[fig-metal\]. basEd on evOlUtiONary tRacKs pubLished BY @SchalLer, Fig.\[fig-TrACks\] shoWS that a tRIPlE sysTem with log(age) betwEEn 8.8 ANd 8.9 is compatible With ouR oBSeRVAtiOns Of.
{wIdth="13cM"}
### disentaNGlED sPecTRa
The determinAtion of the fUNdaMental PaRamETers waS also PeRForMed on the disEntaNgled specTra. A grID of modeL Spectra RanginG frOm 6000 tO 8600K wiTH sTePwiDtH Of 200 K ANd Log G Of 3.6, 4.0, 4.4, rEspectivElY, wAs comPareD TO THe diSenTangLed spEctra correcteD foR radIAl vElociTy shiFts. THe Model SpectrA were CaLculated using thE codE SynthV [@TsYmbAl96] And nEmO modEL atmosPheRe gRid [@HeitEr2002]. For fiNDinG tHE BEsT fitting spectra two ApPROaChes have Been usED. FIrST, the stanDaRd dEviaTIOn of tHe reSIdUals, hereAfter rEFeRrEd to as $\sIgMa$, and sEcOnd, The $\Chi^2$ teST werE used fOr evaluaTing tHE quality of the fIT. These approacHEs LEAd TO difFerEnt best fittIng sOLutiOns, bECaUse THe metHods dIfFEr IN weighting the data. ThE sTandarD deviAtion of the resIduals weigHTS Each data PoinT EqUAlly, while the $\chI^2$ test Weights the DAta accorDing tO the relaTive line sTREngth. HenCe tHe $\cHi^2$ tEst PREfErs better fits OF Line CoRes. SolaR chEmical cOmpOsiTioN waS aSsumed, basEd on the gOoD fIt ObTaiNed by THe direct MeThoD (sEe FIg.\[fig-MEtal\]).
ThE stanDard DeViATioN of the rESiDUAls aNd $\ChI^2$ was CalCuLated The fOLloWing way:\
\
$\Sigma = \sqrt{\FraC{1}{n-1} \sum (r_i - \OvErline{R})^2 }$\
\
$\Chi^2 = \sum \frac{(S_{scAl} - d)^2}{S_{scal}}$\
\
wherE $S_{ScaL}$ means SCAled syntHetic spectrum and $D$ the disENtangleD spEctruM and $r = S_{scal} - D$. ThE moDel speCtrA Were scAled acCordiNg To\
\
$S_{SCAl} = \fraC{s F+1}{f+1}$
\
\
WheRe $s$ denotes thE MOdeL specTrUm anD $f$ | the observed spectra with in the res oluti on. Th eradi al v elocity values forthese fits were takenfromou r orb i ta l sol ution ( s ee T abl e5) . T he ch osenpar ameters for the t rip le system give al so a goodfit to the meta l l ine sp ec tru m as i s i llust ratedi n Fig. \[fig-met al \ ]. Bas e d on ev o l ut iona ry tracks publish e db y @Schaller, F ig.\[f ig - tr a c ks\ ] s hows thatatripl e system wi t h log ( age) between8.8 and 8.9 iscompat ib lew ith ou r obs er v ati ons of.
 { width=" 1 3cm"}
### Di sen tan gled Sp ec tra
T hed et erm i nat ion of t he f undam enta l p a rame ter s wa s als o performed o n t he d i sen tangl ed sp ectr a. A gr id ofmodel s pectra rangingfrom 6000 to860 0K wi th step w idth o f 2 00K and l og g of 3.6 ,4 . 0 ,4.4, respectively, w a s c omparedto the di se n tangledsp ect ra c o r recte d fo r r adial ve locity sh if ts. The m odel s pe ctr a w ere c a lcul ated u sing the code SynthV [@Tsymb a l96] and NEMO mo d e la tmos phe re grid [@H eite r 2002 ]. F o rfin d ing t he be st fi t ting spectra two ap pr oaches have been used. F irst, thes t a ndard de viat i on of the residua ls, h ereafter r e ferred t o as$\sigma$ , and sec o n d, the $ \ch i^2 $ t est w er e used for ev a l uati ng the qu ali ty of t hefit . T hes eapproache s lead t odi ff er ent best fittingso lut io ns, beca u se the meth odsdi ff e r i n weigh t in g theda ta . Th e s ta ndard dev i ati on of t he residu als weig ht seach da ta point equa ll y, while t he $\ chi^2$ t est weig hts the data accordingt o the r ela tiveline strength . H ence t he$ \chi^2 $ test pref er s b e t ter f i t sofli ne cores.S o lar chem ic al c omposit ion was assumed, b a sed on the goodfit obt a i ne d b y t h e d ir e ctm e thod (see Fig.\ [fig-metal \] ) .
The stand a rdde viation of the resi d uals an d $\chi^2 $ was cal cu late d the following way:\
\
$\sigma= \sqr t {\ frac{ 1}{ N-1} \ su m ( R_i - \over l ine {R})^ 2 }$\\$\chi^ 2 = \ su m \frac{ (S_{scal} - D)^2}{S_{sc al}}$\
\
wh ere $S_{scal }$m ean s scaledsynt hetic spec tru m a nd $D $ t h e dis enta n gl eds pectr um a n d $R = S_ { sc al} - D $. The mode l s pec tra w ere scaled acc ording to\
\
$S_ { scal} = \frac{ S f+ 1 } {f+ 1}$
\
\
w here $S$ denot esth e model sp ec trum and $f $ | the_observed spectra_within the resolution. The_radial velocity_values_for these_fits_were taken from_our orbital solution_(see Table 5). The_chosen parameters for_the_triple system give also a good fit to the metal line spectrum as is_illustrated_in Fig.\[fig-metal\]._Based_on_evolutionary tracks published by @Schaller,_Fig.\[fig-tracks\] shows that a triple_system with_log(age) between 8.8 and 8.9 is compatible with_our_observations of.
{width="13cm"}
### Disentangled_Spectra
The determination of the fundamental parameters was also performed_on the disentangled spectra. A grid_of model spectra_ranging_from_6000 to 8600K with_stepwidth of 200 K and log_g of 3.6, 4.0, 4.4, respectively,_was compared to the disentangled spectra corrected_for radial velocity shifts. The model_spectra were calculated using the_code SynthV_[@Tsymbal96] and NEMO model atmosphere_grid [@Heiter2002]. For_finding the_best fitting spectra_two approaches have been used. First,_the standard deviation_of the residuals, hereafter referred to_as_$\sigma$, and second,_the_$\chi^2$_test were_used for evaluating_the_quality of_the_fit. These approaches lead to different_best_fitting solutions, because the methods differ in_weighting the data. The_standard_deviation of the residuals_weights each data point equally,_while the $\chi^2$ test weights the_data according_to the_relative line strength. Hence the $\chi^2$ test prefers better fits of_line cores. Solar chemical composition was_assumed, based on the_good fit_obtained_by the direct_method_(see Fig.\[fig-metal\]).
The_standard deviation of the residuals and $\chi^2$_was calculated_the following way:\
\
$\sigma = \sqrt{\frac{1}{N-1} \sum_(R_i - \overline{R})^2 }$\
\
$\chi^2_=_\sum \frac{(S_{scal} - D)^2}{S_{scal}}$\
\
where $S_{scal}$ means_scaled synthetic spectrum and $D$ the_disentangled spectrum and $R =_S_{scal}_-_D$. The model spectra were_scaled according to\
\
$S_{scal} = \frac{S f+1}{f+1}$
\
\
where_$S$ denotes the_model spectrum and $f$ |
ologie, Geneva. C. Brouwer [*et al.*]{}, Nucl. Instrum. Meth. A [**313**]{} (1992) 50. P. Duinker [*et al.*]{}, Nucl. Instrum. Meth. A [**273**]{} (1988) 814. M. Fabre, “The dimuon mass resolution of the L3 experiment at LEP,” Ph.D. thesis, ETH–Zürich, 1992. M. Unger, “Measurement of the momentum spectrum of atmospheric muons with the L3 detector,” Ph.D. thesis, Humboldt Universität Berlin, 2004, DESY-THESIS-2004-008. W. Lohmann, R. Kopp and R. Voss, “Energy loss of muons in the energy range 1–10000 GeV,” CERN Report 85-03 (1985). D. E. Groom, N. V. Mokhov and S. I. Striganov, Atom. Data Nucl. Data Tabl. [**78**]{} (2001) 183. D. Chirkin and W. Rhode, “Muon Monte Carlo: A new high precision tool for muon propagation through matter,” DESY-PROC-2002-01 (2002). Particle Data Group Collaboration, K. Hagiwara [*et al.*]{}, Phys. Rev. D [**66**]{}, (2002) 010001, and references therein. D. Y. Bardin [*et al.*]{}, Comput. Phys. Commun. [**133**]{} (2001) 229. S. Jadach, B. F. L. Ward and Z. Was, Comput. Phys. Commun. [**79**]{} (1994) 503. T. Sjöstrand, Comput. Phys. Commun. [**82**]{} (1994) 74.
J. A. M. Vermaseren, J. Smith and G. Grammer, Jr., Phys. Rev. D [**19**]{} (1979) 137.
G. D. Lafferty and T. R. Wyatt, Nucl. Instrum. Meth. A [**355**]{} (1995) 541.
K. | ologie, Geneva. C. Brouwer [ * et al. * ] { }, Nucl. Instrum. Meth. A [ * * 313 * * ] { } (1992) 50. P. Duinker [ * et al. * ] { }, Nucl. Instrum. Meth. A [ * * 273 * * ] { } (1988) 814. M. Fabre, “ The dimuon mass resolution of the L3 experiment at LEP, ” Ph.D. thesis, ETH – Zürich, 1992. M. Unger, “ Measurement of the momentum spectrum of atmospheric muon with the L3 detector, ” ph dissertation, Humboldt Universität Berlin, 2004, DESY - THESIS-2004 - 008. W. Lohmann, R. Kopp and R. Voss, “ Energy loss of muons in the department of energy range 1–10000 GeV, ” CERN Report 85 - 03 (1985). D. E. Groom, N. V. Mokhov and S. I. Striganov, Atom. Data Nucl. Data Tabl. [ * * 78 * * ] { } (2001) 183. D. Chirkin and W. Rhode, “ Muon Monte Carlo: A new high preciseness creature for muon propagation through matter, ” DESY - PROC-2002 - 01 (2002). Particle Data Group Collaboration, K. Hagiwara [ * et al. * ] { }, Phys. Rev. D [ * * 66 * * ] { }, (2002) 010001, and references therein. D. Y. Bardin [ * et al. * ] { }, Comput. Phys. Commun. [ * * 133 * * ] { } (2001) 229. S. Jadach, B. F. L. Ward and Z. Was, Comput. Phys. Commun. [ * * 79 * * ] { } (1994) 503. T. Sjöstrand, Comput. Phys. Commun. [ * * 82 * * ] { } (1994) 74.
J. A. M. Vermaseren, J. Smith and G. Grammer, Jr., Phys. Rev. D [ * * 19 * * ] { } (1979) 137.
G. D. Lafferty and T. R. Wyatt, Nucl. Instrum. Meth. A [ * * 355 * * ] { } (1995) 541.
K. | olohie, Geneva. C. Brouwer [*et au.*]{}, Nucl. Instrum. Mejh. Q [**313**]{} (1992) 50. P. Vuinker [*et al.*]{}, Njcl. Instrum. Meth. A [**273**]{} (1988) 814. M. Fabre, “Thx dinuon nass resolution of the L3 experilent at OEP,” Kh.D. thesis, ETH–Zürirg, 1992. M. Ungcx, “Meaamremeut of the momentom spectrum mf atmospheric mjous with the L3 detector,” Ph.D. thesis, Huiboldt Inlversität Berlig, 2004, DTSY-EHESJS-2004-008. W. Lohmann, R. Kopp and R. Voss, “Energy loss oh muons in the rnergy range 1–10000 GeV,” CERN Replrt 85-03 (1985). D. E. Groom, N. V. Mokhov and S. I. Stritanod, Atom. Data Nuzl. Data Tabl. [**78**]{} (2001) 183. D. Chirkin and W. Rhode, “Muon Monte Carlo: A vew hngh precisiin toln for muon 'ropagwtion througm mattes,” DESY-PTOC-2002-01 (2002). Particle Cate Griup Collaboration, K. Hajiwara [*et al.*]{}, Phys. Rev. Q [**66**]{}, (2002) 010001, and seyerences therein. D. Y. Baedun [*et al.*]{}, Wompjr. Phhs. Ckmkuh. [**133**]{} (2001) 229. S. Jwdarh, B. F. L. Ward znd Z. Was, Conput. Phys. Commun. [**79**]{} (1994) 503. T. Skösegsnd, Comput. Phgs. Commtn. [**82**]{} (1994) 74.
J. A. M. Vermaseren, J. Smith and G. Grammer, Jr., Pvys. Dev. D [**19**]{} (1979) 137.
G. D. Lafferty and T. E. Wyatt, Nucl. Instrum. Metj. A [**355**]{} (1995) 541.
K. | ologie, Geneva. C. Brouwer [*et al.*]{}, Nucl. A (1992) 50. Duinker [*et al.*]{}, (1988) M. Fabre, “The mass resolution of L3 experiment at LEP,” Ph.D. thesis, 1992. M. Unger, “Measurement of the momentum spectrum of atmospheric muons with the detector,” Ph.D. thesis, Humboldt Universität Berlin, 2004, DESY-THESIS-2004-008. W. Lohmann, R. Kopp and Voss, loss muons the energy range 1–10000 GeV,” CERN Report 85-03 (1985). D. E. Groom, N. V. Mokhov and I. Striganov, Atom. Data Nucl. Data Tabl. [**78**]{} 183. D. Chirkin and Rhode, “Muon Monte Carlo: A high tool for propagation matter,” (2002). Particle Data Collaboration, K. Hagiwara [*et al.*]{}, Phys. Rev. D [**66**]{}, (2002) 010001, and references therein. D. Y. Bardin al.*]{}, Comput. [**133**]{} (2001) S. B. L. Ward and Comput. Phys. Commun. [**79**]{} (1994) 503. Phys. Commun. [**82**]{} (1994) 74. J. A. M. J. Smith G. Grammer, Jr., Phys. Rev. D (1979) 137. G. D. Lafferty and T. R. Nucl. Instrum. Meth. A [**355**]{} (1995) 541. K. | ologie, Geneva. C. Brouwer [*et al.*]{}, NUcl. Instrum. meth. A [**313**]{} (1992) 50. p. DuInkEr [*Et al.*]{}, nucl. instrum. Meth. A [**273**]{} (1988) 814. M. FABre, “THe dimuon mass resolution Of the l3 eXPeriMEnT at LEp,” Ph.D. theSIs, eth–ZüRiCh, 1992. m. UnGeR, “meAsureMenT of the mOmentum speCtrUm Of atmospheriC MuOns with the l3 deTector,” Ph.D. theSis, humbolDt uniVErsitÄt BErlin, 2004, dESY-THesIS-2004-008. W. LoHmann, R. KopP aND R. Voss, “eNergy loSS Of MuonS in the energy range 1–10000 gEV,” ceRN Report 85-03 (1985). D. E. GroOm, N. V. MoKhOV aND s. I. STriGanov, Atom. DAtA Nucl. dAta Tabl. [**78**]{} (2001) 183. d. chIRKIn aND W. Rhode, “Muon MoNte Carlo: A neW HigH preciSiOn tOOl for mUon prOpAGatIon through mAtteR,” DESY-PROC-2002-01 (2002). particLE Data GrOUp CollaBoratiOn, K. hagIwarA [*Et Al.*]{}, phyS. REV. D [**66**]{}, (2002) 010001, aND rEfeREncEs thereiN. D. y. BArdin [*Et al.*]{}, cOMPUt. PhYs. COmmuN. [**133**]{} (2001) 229. S. JadAch, B. F. L. Ward and z. WaS, ComPUt. PHys. CoMmun. [**79**]{} (1994) 503. T. sjösTrAnd, CoMput. PhYs. ComMuN. [**82**]{} (1994) 74.
J. A. M. Vermaseren, J. smitH and G. GramMer, jr., phyS. REv. D [**19**]{} (1979) 137.
G. D. lAffertY anD T. R. wyatt, NuCl. InstrUM. MeTh. a [**355**]{} (1995) 541.
k. | ologie, Geneva. C. Brouwer [*et al.* ]{},Nuc l.In stru m. M eth. A [**313* * ]{}(1992) 50. P. Duinker[*etal . *]{} , N ucl.Instrum . M e t h.A[* *27 3* * ]{ } (19 88) 814. M . Fabre, “ The d imuon mass r e so lution ofthe L3 experime ntat LEP ,” Ph . D. th esi s, ET H–Züri c h, 199 2. M. Ung er , “Meas u remento f t he m omentum spectrumo fa tmospheric muo ns wit ht he L 3 d ete ctor,” Ph. D. thes i s, Humb o ld t U niv e rsität Berlin , 2004, DES Y -TH ESIS-2 00 4-0 0 8. W.Lohma nn , R. Kopp and R . Vo ss, “Ener gy los s of muo n s in th e ener gyran ge 1 – 10 00 0 G eV , ” C E RN Re p ort 85-03 ( 19 85 ). D. E.G r o o m, N . V . Mo khovand S. I. Str iga nov, Ato m. Da ta Nu cl.Da ta Ta bl. [* *78** ]{ } (2001) 183. D . Ch irkin and W. R hod e, “Muo n Monte Ca rlo : A new high p r eci si o n to ol for muon propag at i o nthroughmatter , ”DE S Y-PROC-2 00 2-0 1 (2 0 0 2). P arti c le Data Gr oup Co l la bo ration, K . Hagi wa ra[*e t al. * ]{}, Phys. Rev. D[**66 * *]{}, (2002) 0 1 0001, and ref e re n c es ther ein . D. Y. Bar din[ *etal.* ] {} , C o mput. Phys .C om m un. [**133**]{} (20 01 ) 229. S. J adach, B. F.L. Ward an d Z . Was, C ompu t .P hys. Commun. [ **79* *]{} (1994 ) 503. T. Sjös trand, C omput. Ph y s . Commun . [ **8 2** ]{} ( 19 94) 74.
J. A . M. V er maseren , J . Smith an d G . G ram me r, Jr., P hys. Rev .D[* *1 9** ]{} ( 1 979) 137 .
G. D . L affer t y andT. R. Wya tt ,N ucl . Instr u m. M eth. A [ **35 5** ]{ } (19 95)5 41.
K. | ologie, Geneva._C. Brouwer [*et_al.*]{}, Nucl. Instrum. Meth. A [**313**]{} (1992)_50. P. Duinker_[*et_al.*]{}, Nucl. Instrum. Meth. A_[**273**]{}_(1988) 814. M. Fabre,_“The dimuon mass_resolution of the L3_experiment at LEP,”_Ph.D._thesis, ETH–Zürich, 1992. M. Unger, “Measurement of the momentum spectrum of atmospheric muons with the_L3_detector,” Ph.D._thesis,_Humboldt_Universität Berlin, 2004, DESY-THESIS-2004-008. W. Lohmann,_R. Kopp and R. Voss, “Energy loss_of muons_in the energy range 1–10000 GeV,” CERN Report_85-03_(1985). D. E. Groom, N. V. Mokhov_and S. I. Striganov, Atom. Data Nucl. Data Tabl. [**78**]{} (2001) 183. D. Chirkin and_W. Rhode, “Muon Monte Carlo: A new_high precision tool_for_muon_propagation through matter,” DESY-PROC-2002-01_(2002). Particle Data Group Collaboration, K. Hagiwara_[*et al.*]{}, Phys. Rev. D [**66**]{}, (2002) 010001,_and references therein. D. Y. Bardin [*et al.*]{}, Comput. Phys. Commun. [**133**]{}_(2001) 229. S. Jadach, B. F. L. Ward and Z. Was,_Comput. Phys. Commun. [**79**]{} (1994) 503. T. Sjöstrand, Comput. Phys. Commun. [**82**]{}_(1994) 74.
J. A. M. Vermaseren,_J. Smith and G. Grammer, Jr., Phys. Rev. D [**19**]{}_(1979) 137.
G. D. Lafferty and_T. R. Wyatt, Nucl. Instrum. Meth. A_[**355**]{} (1995) 541.
K. |
;\tilde{k}_1,\tilde{k}_2,\tilde{k}_3) \cdot
n^{m(\tilde{k}_1+\tilde{k}_2)+\frac{1}{2}m(m-1)\tilde{k}_3},$$ as $n \to \infty$.
The claim follows from the previous theorem and the asymptotics of the gamma function (cf ): $\Gamma(n+a) \sim \Gamma(n) n^a$ for a constant $a$.
Random matrix ensembles associated with compact symmetric spaces
================================================================
Finally, we apply the theorems obtained above to compact symmetric spaces as classified by Cartan. These symmetric spaces are labeled A I, BD I, C II, and so on, see e.g. Table 1 in [@CM]. Let $G/K$ be such a compact symmetric space. Here $G$ is a compact subgroup of $GL(N,{\mathbb{C}})$ for some positive integer $N$, and $K$ is a closed subgroup of $G$. Then the space $G/K$ is realized as the subset $S$ of $G$: $S \simeq G/K$ and the probability measure $\dd M$ on $S$ is induced from the quotient space $G/K$. We consider $S$ as a probability space with the measure $\dd M$ and call the random matrix ensemble associated with $G/K$. See [@Duenez] for details.
The random matrix ensembles considered in §\[subsectionA\], §\[subsectionAI\], and §\[subsectionAII\] are called Dyson’s circular $\beta$-ensembles, see [@Dyson; @Mehta]. The identities in these subsections follow immediately from expressions and (see also Example \[ExFq\]). Similarly, identities after §\[subsectionB\] follows from Theorem \[Thm:MainTheorem\], Theorem \[Thm:Main2\], and Corollary \[cor:Main\].
Note that the results in §\[subsectionA\], §\[subsectionB\], §\[subsectionC\], and §\[subsectionD\] are results for compact Lie groups (which are not proper symmetric spaces) previously presented in [@BG].
$U(n)$ – type A {#subsectionA}
---------------
Consider the unitary group $U(n)$ with the normalized Haar measure. This space has a simple root system of type A. The corresponding p.d.f | ; \tilde{k}_1,\tilde{k}_2,\tilde{k}_3) \cdot
n^{m(\tilde{k}_1+\tilde{k}_2)+\frac{1}{2}m(m-1)\tilde{k}_3},$$ as $ n \to \infty$.
The claim follows from the previous theorem and the asymptotics of the da gamma affair (cf ): $ \Gamma(n+a) \sim \Gamma(n) n^a$ for a constant $ a$.
Random matrix corps de ballet consort with compact symmetric space
= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
Finally, we apply the theorem obtained above to compact symmetrical spaces as classified by Cartan. These symmetric spaces are label A I, BD I, C II, and therefore on, see for example Table 1 in [ @CM ]. Let $ G / K$ be such a compact symmetrical space. Here $ G$ is a compact subgroup of $ GL(N,{\mathbb{C}})$ for some positive integer $ N$, and $ K$ is a closed subgroup of $ G$. Then the space $ gram / K$ is realized as the subset $ S$ of $ G$: $ S \simeq G / K$ and the probability measure $ \dd M$ on $ S$ is induced from the quotient quad $ G / K$. We consider $ S$ as a probability space with the measure $ \dd M$ and call the random matrix ensemble associated with $ G / K$. See [ @Duenez ] for details.
The random matrix ensembles considered in § \[subsectionA\ ], § \[subsectionAI\ ], and § \[subsectionAII\ ] are called Dyson ’s circular $ \beta$-ensembles, see [ @Dyson; @Mehta ]. The identities in these subsections follow immediately from expressions and (see besides Example \[ExFq\ ]). Similarly, identities after § \[subsectionB\ ] adopt from Theorem \[Thm: MainTheorem\ ], Theorem \[Thm: Main2\ ], and Corollary \[cor: Main\ ].
Note that the resultant role in § \[subsectionA\ ], § \[subsectionB\ ], § \[subsectionC\ ], and § \[subsectionD\ ] are result for compact Lie groups (which are not proper symmetric space) previously presented in [ @BG ].
$ U(n)$ – type A { # subsectionA }
---------------
Consider the unitary group $ U(n)$ with the normalized Haar measuring stick. This space has a simple root organization of type A. The corresponding p.d.f | ;\tilfe{k}_1,\tilde{k}_2,\tilde{k}_3) \cdot
n^{m(\tllde{k}_1+\tilde{k}_2)+\frac{1}{2}m(m-1)\tilde{k}_3},$$ as $n \fo \infty$.
Ghe claim follows from the pceviius tyeorem and the asymptogics of tje gamma funrtion (cf ): $\Gamma(n+e) \sim \Gamma(n) n^z$ for e constant $a$.
Rancom matrix ensembles assmckaced with compact symmetric spaces
================================================================
Finwlly, we aoply the theorgms oneainsd above to compact symmetric spacss as cnassified by Vartan. These symmetric spafes wre labeled A I, BD I, C II, and so jb, see e.g. Tabue 1 in [@CM]. Let $G/K$ be sudh a compact symmetric space. Hefe $G$ ns a compacj auhcroup of $GL(I,{\mathbf{C}})$ for some ipsitiva integrr $N$, and $K$ is s cnoswd subgroup of $G$. Then the space $G/K$ is rgalized as tke subset $S$ of $G$: $S \sinew G/K$ dnd dhe oeobxbimivy jeasurf $\dv M$ on $S$ is induced frim the quotient spave $T/K$. We consided $S$ as a probability space with the measure $\dd K$ ahd call the random matrux ensemble associatef with $G/K$. See [@Duenez] for details.
The random matrix ensemblev conakdexcq un §\[subsectionA\], §\[subsectionAI\], and §\[subsectionAII\] aws vakled Dyson’s civcular $\beta$-ensemblrs, srg [@Dyson; @Mehta]. Jhe ideufifies in these subsfctions folliw immediwtelu from expressions and (see qlso Example \[WxFq\]). Similarly, ideutities aftex §\[subsgctionN\] follows from Theorem \[Chm:MaihTheorem\], Thforem \[Thm:Jxin2\], and Corollarh \[cpr:Kain\].
Note that the results yn §\[subsecvionA\], §\[subsecgionN\], §\[subsqctionC\], anf §\[subsectionD\] are resultd for cmmpact Lie groups (which are not proper symmetric spaces) psevpously przsentec in [@BG].
$U(n)$ – ty[e A {#subsectiouA}
---------------
Considzr the unitary ggoup $U(n)$ wmth the normwlized Haar mawsure. This s'ace has w sinple root shrtem of type A. The corrvskonding p.d.d | ;\tilde{k}_1,\tilde{k}_2,\tilde{k}_3) \cdot n^{m(\tilde{k}_1+\tilde{k}_2)+\frac{1}{2}m(m-1)\tilde{k}_3},$$ as $n \to \infty$. follows the previous and the asymptotics ): \sim \Gamma(n) n^a$ a constant $a$. matrix ensembles associated with compact symmetric ================================================================ Finally, we apply the theorems obtained above to compact symmetric spaces as by Cartan. These symmetric spaces are labeled A I, BD I, C II, so see Table in [@CM]. Let $G/K$ be such a compact symmetric space. Here $G$ is a compact subgroup $GL(N,{\mathbb{C}})$ for some positive integer $N$, and $K$ a closed subgroup of Then the space $G/K$ is as subset $S$ $G$: \simeq and the probability $\dd M$ on $S$ is induced from the quotient space $G/K$. We consider $S$ as a probability with the M$ and the matrix associated with $G/K$. for details. The random matrix ensembles §\[subsectionAI\], and §\[subsectionAII\] are called Dyson’s circular $\beta$-ensembles, [@Dyson; @Mehta]. identities in these subsections follow immediately expressions and (see also Example \[ExFq\]). Similarly, identities §\[subsectionB\] follows from Theorem \[Thm:MainTheorem\], Theorem \[Thm:Main2\], and Corollary \[cor:Main\]. Note that the results in §\[subsectionC\], and §\[subsectionD\] are for compact Lie (which not symmetric previously presented [@BG]. $U(n)$ – type A {#subsectionA} --------------- Consider the unitary group with the normalized Haar measure. This space has a simple of A. The corresponding | ;\tilde{k}_1,\tilde{k}_2,\tilde{k}_3) \cdot
n^{m(\tIlde{k}_1+\tilde{K}_2)+\frac{1}{2}M(m-1)\tIldE{k}_3},$$ As $n \tO \infTy$.
The claim follOWs frOm the previous theorem anD the aSyMPtotICs Of the Gamma fuNCtION (cf ): $\gaMmA(n+a) \SiM \gaMma(n) n^A$ foR a constAnt $a$.
Random MatRiX ensembles asSOcIated with cOmpAct symmetric SpaCes
================================================================
FinAlLy, wE Apply The TheorEms obtAIned abOve to compAcT SymmetRIc spaceS AS cLassIfied by Cartan. ThesE SyMMetric spaces arE labelEd a i, Bd i, c II, And So on, see e.g. TAbLe 1 in [@Cm]. let $G/K$ be SUcH A COmpACt symmetric spAce. Here $G$ is a COmpAct subGrOup OF $GL(N,{\maThbb{C}})$ FoR SomE positive inTegeR $N$, and $K$ is a Closed SUbgroup OF $G$. Then tHe spacE $G/K$ Is rEaliZEd As The SuBSet $s$ Of $g$: $S \sIMeq g/K$ and the PrObAbiliTy meASURE $\dd M$ On $S$ Is inDuced From the quotieNt sPace $g/k$. We ConsiDer $S$ aS a prObAbiliTy spacE with ThE measure $\dd M$ and cAll tHe random mAtrIx EnsEmBle asSOciateD wiTh $G/k$. See [@DueNez] for dETaiLs.
tHE RaNdom matrix ensembleS cONSiDered in §\[sUbsectIOna\], §\[sUBsectionaI\], And §\[SubsECTionAiI\] arE CaLled DysoN’s circULaR $\bEta$-enseMbLes, see [@dySon; @mehTa]. The IDentIties iN these suBsectIOns follow immedIAtely from exprESsIONs ANd (seE alSo Example \[Exfq\]). SiMIlarLy, idENtItiES afteR §\[subsEcTIoNb\] follows from Theorem \[thM:MainTHeoreM\], Theorem \[Thm:MaIn2\], and CorolLARY \[cor:Main\].
note THaT The results in §\[suBsectIonA\], §\[subsecTIonB\], §\[subsEctioNC\], and §\[subSectionD\] aRE Results fOr cOmpAct lie GROuPs (which are not PROper SyMmetric SpaCes) prevIouSly PreSenTeD in [@BG].
$U(n)$ – tyPe A {#subseCtIoNA}
---------------
coNsiDer thE Unitary gRoUp $U(N)$ wIth The noRMalizeD Haar MeasUrE. THIs sPace has A SiMPLe roOt SyStem Of tYpE A. The CorrESpoNding p.d.F | ;\tilde{k}_1,\tilde{k}_2,\ tilde{k}_3 ) \cd ot
n^ {m (\ti lde{ k}_1+\tilde{k} _ 2)+\ frac{1}{2}m(m-1)\tilde {k}_3 }, $ $ as $n \to\infty$ .
T h e c la im fo ll o ws from th e previ ous theore m a nd the asympto t ic s of the g amm a function ( cf): $\G am ma( n +a) \ sim \Gam ma(n)n ^a$ fo r a const an t $a$.Randomm a tr ix e nsembles associat e dw ith compact sy mmetri cs pa c e s
= === ========== == ===== = ======= = == = = = === = ============= =========== =
F inally ,wea pply t he th eo r ems obtained a bove to compa ct sym m etric s p aces as class ifi edby C a rt an . T he s e s y mm etr i c s paces ar ela beled A I , B D I,C I I, a nd so on, see e.g. Ta ble1 in [@CM ]. Le t $G /K $ besuch a comp ac t symmetric spa ce.Here $G$isacom pa ct su b groupof$GL (N,{\ma thbb{C} } )$fo r s om e positive integer $ N $ ,and $K$is a c l os ed subgroup o f $ G$.T h en th e sp a ce $G/K$ i s real i ze das thesu bset $ S$ of $G $: $S \sim eq G/K $ and th e pro b ability measur e $\dd M$ on $ S $i s i n duce d f rom the quo tien t spa ce $ G /K $.W e con sider $ S $a s a probability spa ce withthe m easure $\dd M $ and call t h e random mat r ix ensemble assoc iated with $G/K $ . See [@ Duene z] for d etails.
T h e random ma tri x e nse m b le s consideredi n §\[ su bsectio nA\ ], §\[s ubs ect ion AI\ ], and §\[s ubsectio nA II \] a recalle d Dyson’s c irc ul ar$\bet a $-ense mbles , se e[@ D yso n; @Meh t a] . Theid en titi esin thes e su b sec tions f ollow imm edi a tely f ro m expre ssions and (s ee also Exam pl e \ [ExFq\ ] ) . Simila rly, identities after § \ [subsec tio nB\]foll ows fromThe orem \ [Th m :MainT heorem \], T he ore m \[Thm : M ai n2\ ], and Corol l a ry\[cor :M ain\ ].
Not e that the results in§\[subsection A\] , §\ [ s ub sec t io n B\] ,§ \[s u b sectionC\], and §\[subsec ti o nD \] are res u lts f or comp act Lie grou p s (whic h are not proper s ym metr i c sp aces) prev iously p resentedi n [@B G ].
$U( n)$ – typ eA { #subs ection A }
- ----- ------ -- -
Con sider t he unita ry group $U(n)$ with th e norm alize d H aar measu re. Thi s space h as a simple ro otsys tem o f t y pe A. The co rre s pondi ng p . d.f | ;\tilde{k}_1,\tilde{k}_2,\tilde{k}_3) \cdot_
n^{m(\tilde{k}_1+\tilde{k}_2)+\frac{1}{2}m(m-1)\tilde{k}_3},$$ as_$n \to \infty$.
The claim_follows from_the_previous theorem_and_the asymptotics of_the gamma function_(cf ): $\Gamma(n+a) \sim_\Gamma(n) n^a$ for_a_constant $a$.
Random matrix ensembles associated with compact symmetric spaces
================================================================
Finally, we apply the theorems obtained_above_to compact_symmetric_spaces_as classified by Cartan. These_symmetric spaces are labeled A_I, BD_I, C II, and so on, see e.g._Table_1 in [@CM]._Let $G/K$ be such a compact symmetric space. Here_$G$ is a compact subgroup of_$GL(N,{\mathbb{C}})$ for some_positive_integer_$N$, and $K$ is_a closed subgroup of $G$. Then_the space $G/K$ is realized as_the subset $S$ of $G$: $S \simeq_G/K$ and the probability measure $\dd_M$ on $S$ is induced_from the_quotient space $G/K$. We consider_$S$ as a_probability space_with the measure_$\dd M$ and call the random_matrix ensemble associated_with $G/K$. See [@Duenez] for details.
The_random_matrix ensembles considered_in_§\[subsectionA\],_§\[subsectionAI\], and_§\[subsectionAII\] are called_Dyson’s_circular $\beta$-ensembles,_see_[@Dyson; @Mehta]. The identities in these_subsections_follow immediately from expressions and (see also_Example \[ExFq\]). Similarly, identities_after_§\[subsectionB\] follows from Theorem_\[Thm:MainTheorem\], Theorem \[Thm:Main2\], and Corollary_\[cor:Main\].
Note that the results in §\[subsectionA\],_§\[subsectionB\], §\[subsectionC\],_and §\[subsectionD\]_are results for compact Lie groups (which are not proper symmetric_spaces) previously presented in [@BG].
$U(n)$ –_type A {#subsectionA}
---------------
Consider the_unitary group_$U(n)$_with the normalized_Haar_measure. This_space has a simple root system of_type A._The corresponding p.d.f |
substochastic $C_0$-semigroup on $\ell_w^1$. This particular result, which is one of the main motivations for carrying out an analysis of the fragmentation system in general weighted $\ell^1$ spaces, requires a stronger assumption on the weight $w$. Note that when dealing with analytic semigroups, we use complex versions of the spaces $\ell_w^1$.
\[assumption for analyticity\]
------------------------------------------------------------------------
$w_n \ge n$ for all $n \in \mathbb{N}$.
There exists $\kappa \in (0,1)$ such that $$\label{inequ_for_analyticity}
\sum\limits_{n=1}^{j-1} w_nb_{n,j} \leq \kappa w_j \qquad \text{for all} \ j=2,3,\ldots.$$
Note that Assumption \[assumption for analyticity\] is obtained from Assumption \[assumption on weight for generation\] by simply replacing $\kappa \in (0,1]$ with $\kappa \in (0,1)$. By removing the possibility of $\kappa = 1$, we can obtain the following improved version of Theorem \[G=closure for frag\].
\[thm for analyticity\] Let Assumptions \[A1.1\] and \[assumption for analyticity\] hold. Then the operator $G^{(w)}=A^{(w)}+B^{(w)}$ is the generator of an analytic, substochastic $C_0$-semigroup, $(S^{(w)}(t))_{t \ge 0}$, on $\ell_w^1$.
Let $(T^{(w)}(t))_{t\ge0}$ be as in Lemma \[A is a generator\]. For $\alpha > 0$ and $f \in \mathcal{D}(A^{(w)})_+$, we obtain from that $$\begin{aligned}
\int\limits_0^{\alpha} \big\Vert B^{(w)}T^{(w)}(t)f \big\Vert\,{\mathrm{d}}t
&\le \kappa \int\limits_0^{\alpha} \big\Vert A^{(w)}T^{(w)}(t)f \big\Vert_{w}\,{\mathrm{d}}t \\
&= \kappa\int\limits_0^{\alpha} \phi_w\bigl(-A^{(w)}T^{( | substochastic $ C_0$-semigroup on $ \ell_w^1$. This particular result, which is one of the main motivations for carry out an psychoanalysis of the fragmentation system in cosmopolitan weighted $ \ell^1 $ space, requires a stronger premise on the weight $ w$. Note that when deal with analytic semigroups, we use complex versions of the space $ \ell_w^1$.
\[assumption for analyticity\ ]
------------------------------------------------------------------------
$ w_n \ge n$ for all $ n \in \mathbb{N}$.
There exists $ \kappa \in (0,1)$ such that $ $ \label{inequ_for_analyticity }
\sum\limits_{n=1}^{j-1 } w_nb_{n, j } \leq \kappa w_j \qquad \text{for all } \ j=2,3,\ldots.$$
Note that Assumption \[assumption for analyticity\ ] is obtained from premise \[assumption on weight for generation\ ] by simply replacing $ \kappa \in (0,1]$ with $ \kappa \in (0,1)$. By take out the possibility of $ \kappa = 1 $, we can obtain the following improved adaptation of Theorem \[G = closure for frag\ ].
\[thm for analyticity\ ] Let Assumptions \[A1.1\ ] and \[assumption for analyticity\ ] hold. Then the operator $ G^{(w)}=A^{(w)}+B^{(w)}$ is the generator of an analytic, substochastic $ C_0$-semigroup, $ (S^{(w)}(t))_{t \ge 0}$, on $ \ell_w^1$.
Let $ (T^{(w)}(t))_{t\ge0}$ be as in Lemma \[A is a generator\ ]. For $ \alpha > 0 $ and $ f \in \mathcal{D}(A^{(w)})_+$, we receive from that $ $ \begin{aligned }
\int\limits_0^{\alpha } \big\Vert B^{(w)}T^{(w)}(t)f \big\Vert\,{\mathrm{d}}t
& \le \kappa \int\limits_0^{\alpha } \big\Vert A^{(w)}T^{(w)}(t)f \big\Vert_{w}\,{\mathrm{d}}t \\
& = \kappa\int\limits_0^{\alpha } \phi_w\bigl(-A^{(w)}T^ { ( | suhstochastic $C_0$-semigroup ok $\ell_w^1$. This partnxular cesult, shich is one of the main motivations fir caerying out an analysis of the fgagmentatuon wtstem in gxheral wcnghtes $\ell^1$ wpaces, requirex a strongar assumption mn tke weight $w$. Note that when dealing wyth anakyhic semigroups, we lsq cojiltx versions of the spaces $\ell_w^1$.
\[asshmption for analyticoty\]
------------------------------------------------------------------------
$w_n \ge n$ for all $n \in \mahhbb{J}$.
There exists $\kappw \in (0,1)$ such jgat $$\oabel{inequ_fof_analyticiuy}
\sum\limita_{n=1}^{j-1} w_nb_{n,j} \leq \kappa w_j \qquad \tebt{for all} \ j=2,3,\ldojs.$$
Hohg that Assum'tion \[afsumption fov analydicity\] os obtained frpm Essunption \[assumption on wxight for generation\] by simplf xeplacing $\kappa \in (0,1]$ wuty $\kapka \in (0,1)$. By eemuvihg tge posdibmlity of $\kalpa = 1$, we cab obtain the followond improved veraion os Eheorem \[G=closure for frag\].
\[thm for analytpcitg\] Let Assumptions \[A1.1\] and \[qssumption for analytlcity\] holq. Then the operator $G^{(w)}=A^{(w)}+B^{(w)}$ is the generator of an analbtkc, wunstozyadtic $C_0$-semigroup, $(S^{(w)}(t))_{t \ge 0}$, on $\ell_w^1$.
Let $(T^{(w)}(t))_{t\ge0}$ be zs ik Lemma \[A is a gekerator\]. For $\alpha > 0$ wnc $f \in \mathcal{A}(A^{(w)})_+$, we obfain from that $$\begln{aligngd}
\inr\limits_0^{\alkha} \bog\Vert B^{(w)}T^{(w)}(t)f \big\Vert\,{\mathrm{e}}t
&\le \kappa \unt\limits_0^{\alpha} \big\Rert A^{(w)}T^{(w)}(t)f \yig\Verj_{w}\,{\mathtm{d}}t \\
&= \kappa\int\limits_0^{\aupha} \phi_w\bigl(-A^{(w)}H^{( | substochastic $C_0$-semigroup on $\ell_w^1$. This particular result, one the main for carrying out system general weighted $\ell^1$ requires a stronger on the weight $w$. Note that dealing with analytic semigroups, we use complex versions of the spaces $\ell_w^1$. \[assumption analyticity\] ------------------------------------------------------------------------ $w_n \ge n$ for all $n \in \mathbb{N}$. There exists $\kappa (0,1)$ that \sum\limits_{n=1}^{j-1} \leq \kappa w_j \qquad \text{for all} \ j=2,3,\ldots.$$ Note that Assumption \[assumption for analyticity\] is obtained Assumption \[assumption on weight for generation\] by simply $\kappa \in (0,1]$ with \in (0,1)$. By removing the of = 1$, can the improved version of \[G=closure for frag\]. \[thm for analyticity\] Let Assumptions \[A1.1\] and \[assumption for analyticity\] hold. Then the operator is the an analytic, $C_0$-semigroup, \ge on $\ell_w^1$. Let as in Lemma \[A is a > 0$ and $f \in \mathcal{D}(A^{(w)})_+$, we obtain that $$\begin{aligned} \big\Vert B^{(w)}T^{(w)}(t)f \big\Vert\,{\mathrm{d}}t &\le \kappa \int\limits_0^{\alpha} A^{(w)}T^{(w)}(t)f \big\Vert_{w}\,{\mathrm{d}}t \\ &= \kappa\int\limits_0^{\alpha} \phi_w\bigl(-A^{(w)}T^{( | substochastic $C_0$-semigroup on $\Ell_w^1$. This paRticuLar ResUlT, whiCh is One of the main moTIvatIons for carrying out an anAlysiS oF The fRAgMentaTion sysTEm IN GenErAl WeiGhTEd $\Ell^1$ spAceS, requirEs a strongeR asSuMption on the wEIgHt $w$. Note thaT whEn dealing witH anAlytic SeMigROups, wE usE compLex verSIons of The spaces $\ElL_W^1$.
\[assumPTion for ANAlYticIty\]
------------------------------------------------------------------------
$w_n \ge n$ for all $n \in \MAtHBb{N}$.
There exists $\Kappa \iN (0,1)$ sUCh THAt $$\lAbeL{inequ_for_aNaLyticITy}
\sum\liMItS_{N=1}^{J-1} W_nb_{N,J} \leq \kappa w_j \qqUad \text{for aLL} \ j=2,3,\lDots.$$
NoTe ThaT assumpTion \[aSsUMptIon for analyTiciTy\] is obtaiNed froM assumptIOn \[assumPtion oN weIghT for GEnErAtiOn\] BY siMPlY rePLacIng $\kappa \In (0,1]$ WiTh $\kapPa \in (0,1)$. bY REMoviNg tHe poSsibiLity of $\kappa = 1$, we Can ObtaIN thE follOwing ImprOvEd verSion of theorEm \[g=closure for frag\].
\[Thm fOr analytiCitY\] LEt ASsUmptiONs \[A1.1\] and \[AssUmpTion for AnalytiCIty\] HoLD. tHeN the operator $G^{(w)}=A^{(w)}+B^{(w)}$ Is THE gEnerator Of an anALyTiC, SubstochAsTic $c_0$-semIGRoup, $(S^{(W)}(t))_{t \gE 0}$, On $\Ell_w^1$.
Let $(T^{(W)}(t))_{t\ge0}$ bE As In lemma \[A iS a GeneraToR\]. FoR $\alPha > 0$ anD $F \in \mAthcal{d}(A^{(w)})_+$, we obtAin frOM that $$\begin{aligNEd}
\int\limits_0^{\alPHa} \BIG\VERt B^{(w)}t^{(w)}(t)F \big\Vert\,{\matHrm{d}}T
&\Le \kaPpa \iNT\lImiTS_0^{\alphA} \big\VErT a^{(w)}t^{(W)}(t)f \big\Vert_{w}\,{\mathrm{d}}t \\
&= \KaPpa\int\LimitS_0^{\alpha} \phi_w\bigL(-A^{(w)}T^{( | substochastic $C_0$-semig roup on $\ ell_w ^1$ . T hi s pa rtic ular result, w h ichis one of the main mot ivati on s for ca rryin g out a n a n a lys is o f t he fr agmen tat ion sys tem in gen era lweighted $\e l l^ 1$ spaces, re quires a str ong er ass um pti o n onthe weig ht $w$ . Notethat when d e alingw ith ana l y ti c se migroups, we usec om p lex versions o f thesp a ce s $\e ll_ w^1$.
\[a ss umpti o n for a n al y t i cit y \]
--------- ----------- - --- ------ -- --- - ------ ----- -- - --- ----------- ---- ----
$w_ n \gen $ for a l l $n \i n \mat hbb {N} $.
T he re ex is t s $ \ ka ppa \in (0,1)$su ch that $$\ l a b e l{in equ _for _anal yticity}
\s um\ limi t s_{ n=1}^ {j-1} w_n b_ {n,j} \leq\kapp aw_j \qquad \tex t{fo r all} \j=2 ,3 ,\l do ts.$$
Notetha t A ssumpti on \[as s ump ti o n fo r analyticity\] is o b t ai ned from Assum p ti on \[assump ti onon w e i ght f or g e ne ration\] by si m pl yreplaci ng $\kap pa \i n ( 0,1]$ with $\kap pa \in ( 0,1)$ . By removing t h e possibility of $ \k a ppa= 1 $, we can o btai n the fol l ow ing impro ved v er s io n of Theorem \[G=clo su re for frag \].
\[thm fo r analytic i t y \] Let A ssum p ti o ns \[A1.1\] an d \[a ssumptionf or analy ticit y\] hold . Then th e operator $G ^{( w)} =A^ { ( w) }+B^{(w)}$ is t he g en eratorofan anal yti c,sub sto ch astic $C_ 0$-semig ro up ,$( S^{ (w)}( t ))_{t \g e0}$ ,on$\ell _ w^1$.
Let$(T^ {( w) } (t) )_{t\ge 0 }$ b e as i nLemm a \ [A is a gen e rat or\]. F or $\alph a > 0$ a nd $ f \in \ mathcal{D}(A^ {( w)})_+$, w eobt ain fr o m that $$ \begin{aligned}
\int\ l imits_0 ^{\ alpha } \b ig\Vert B ^{( w)}T^{ (w) } (t)f \ big\Ve rt\,{ \m ath r m {d}}t & \le \ kappa \int \ l imi ts_0^ {\ alph a} \big \Vert A^{(w)}T^{(w ) }(t )f \big\Vert_ {w} \,{\ m a th rm{ d }} t \\
&=\ k appa\int\limits _0^{\alpha }\ ph i_w\bigl(- A ^{( w) }T^{( | substochastic_$C_0$-semigroup on_$\ell_w^1$. This particular result,_which is_one_of the_main_motivations for carrying_out an analysis_of the fragmentation system_in general weighted_$\ell^1$_spaces, requires a stronger assumption on the weight $w$. Note that when dealing with_analytic_semigroups, we_use_complex_versions of the spaces $\ell_w^1$.
\[assumption_for analyticity\]
------------------------------------------------------------------------
$w_n \ge n$ for_all $n_\in \mathbb{N}$.
There exists $\kappa \in (0,1)$ such that_$$\label{inequ_for_analyticity}
_ \sum\limits_{n=1}^{j-1} w_nb_{n,j}_\leq \kappa w_j \qquad \text{for all} \ j=2,3,\ldots.$$
Note that_Assumption \[assumption for analyticity\] is obtained from_Assumption \[assumption on weight_for_generation\]_by simply replacing $\kappa_\in (0,1]$ with $\kappa \in (0,1)$._By removing the possibility of $\kappa_= 1$, we can obtain the following_improved version of Theorem \[G=closure for_frag\].
\[thm for analyticity\] Let Assumptions \[A1.1\]_and \[assumption_for analyticity\] hold. Then the_operator $G^{(w)}=A^{(w)}+B^{(w)}$ is_the generator_of an analytic,_substochastic $C_0$-semigroup, $(S^{(w)}(t))_{t \ge 0}$, on_$\ell_w^1$.
Let $(T^{(w)}(t))_{t\ge0}$ be_as in Lemma \[A is a generator\]._For_$\alpha > 0$_and_$f_\in \mathcal{D}(A^{(w)})_+$,_we obtain from_that_$$\begin{aligned}
_\int\limits_0^{\alpha}_\big\Vert B^{(w)}T^{(w)}(t)f \big\Vert\,{\mathrm{d}}t
&\le \kappa_\int\limits_0^{\alpha}_\big\Vert A^{(w)}T^{(w)}(t)f \big\Vert_{w}\,{\mathrm{d}}t \\
&= \kappa\int\limits_0^{\alpha}_\phi_w\bigl(-A^{(w)}T^{( |
robust PCA.
Convergence of ILT {#S:ILT_convergece}
==================
Here we establish Proposition \[thm:ilt\_convergence\]. We first define a surrogate function for $f(\bx)$ in (\[eqn:log\_reg\]): $$\begin{aligned}
\label{eqn:surrogate}
\nonumber Q(\bx, \bz) = \Vert \by - A \bx \Vert_2^2 + \lambda \sum_i \log(\delta + | x_i|)+ \\
\Vert \bx - \bz \Vert_2^2 - \Vert A (\bx - \bz) \Vert_2^2\end{aligned}$$ Note that $Q(\bx, \bx) = f(\bx)$. Simplifying (\[eqn:surrogate\]) we have $$\label{funcQ}
Q(\bx, \bz) =
\sum_i \left( x_i - k_i(\bz))^2 + \lambda \log( x_i + |\delta|) \right)+ K(\bz),$$ where $k_i(\bz) = z_i + a_i^T \by - a_i^T A \bz$ and $K(\bz)$ contains terms independent of $\bx$. The optimization over $\bx$ is now separable, i.e. can be done independently for each coordinate. We can see that finding local minima over $\bx$ of $Q(\bx, \bz)$ corresponds to iterative log-thresholding.
![Illustration of singular-value log-thresholding.[]{data-label="fig:results_svlt"}](figures/SV_thresh.pdf){width="9.5cm"}
Using this motivation for ILT, we can now prove convergence to fixed points of $f(\bx)$. First, we have:
$f(\hbx^n) = Q(\hbx^{n}, \hbx^n)$ and $Q(\hbx^{n+1}, \hbx^{n})$, are monotonically decreasing with iterations $n$ as long as the spectral norm $\Vert A \Vert_2 < 1$.
The proof parallels the IHT proof of [@blumensath2008iterative] using the fact that $Q(\bx^{n+1}, \bx^n) = f(\bx^n) + \|\bx^{n+1 | robust PCA.
Convergence of ILT { # S: ILT_convergece }
= = = = = = = = = = = = = = = = = =
Here we establish Proposition \[thm: ilt\_convergence\ ]. We first specify a foster function for $ f(\bx)$ in (\[eqn: log\_reg\ ] ): $ $ \begin{aligned }
\label{eqn: surrogate }
\nonumber Q(\bx, \bz) = \Vert \by - A \bx \Vert_2 ^ 2 + \lambda \sum_i \log(\delta + | x_i|)+ \\
\Vert \bx - \bz \Vert_2 ^ 2 - \Vert A (\bx - \bz) \Vert_2 ^ 2\end{aligned}$$ bill that $ Q(\bx, \bx) = f(\bx)$. Simplifying (\[eqn: surrogate\ ]) we own $ $ \label{funcQ }
Q(\bx, \bz) =
\sum_i \left (x_i - k_i(\bz))^2 + \lambda \log (x_i + |\delta|) \right)+ K(\bz),$$ where $ k_i(\bz) = z_i + a_i^T \by - a_i^T A \bz$ and $ K(\bz)$ contains terms autonomous of $ \bx$. The optimization over $ \bx$ is now separable, i.e. can be done independently for each coordinate. We can attend that finding local minima over $ \bx$ of $ Q(\bx, \bz)$ corresponds to iterative log - thresholding.
! [ example of singular - value log - thresholding.[]{data - label="fig: results_svlt"}](figures / SV_thresh.pdf){width="9.5 cm " }
use this motivation for ILT, we can nowadays prove convergence to fixed points of $ f(\bx)$. First, we have:
$ f(\hbx^n) = Q(\hbx^{n }, \hbx^n)$ and $ Q(\hbx^{n+1 }, \hbx^{n})$, are monotonically decrease with iterations $ n$ as long as the spectral average $ \Vert A \Vert_2 < 1$.
The proof parallels the IHT validation of [ @blumensath2008iterative ] using the fact that $ Q(\bx^{n+1 }, \bx^n) = f(\bx^n) + \|\bx^{n+1 | rohust PCA.
Convergence of IUT {#S:ILT_convergeew}
==================
Here xe estaglish Pruposition \[thm:ilt\_convergence\]. Xe furst eefine a surrogate funztion for $f(\bx)$ in (\[wqn:lit\_reg\]): $$\begin{emigned}
\label{eqh:durrmjate}
\nonumber Q(\bw, \bz) = \Vert \ty - A \bx \Vert_2^2 + \lxmyda \sum_i \log(\delta + | x_i|)+ \\
\Vert \bx - \bz \Vqrt_2^2 - \Vett A (\bx - \bz) \Vert_2^2\gnd{alpgged}$$ Hote that $Q(\bx, \bx) = f(\bx)$. Simplifying (\[sqn:surrmgate\]) we have $$\label{funcQ}
Q(\bx, \bz) =
\sum_i \lefh( x_i - k_i(\bz))^2 + \lambda \log( x_i + |\deota|) \wught)+ K(\bz),$$ whefe $k_i(\bz) = e_i + a_i^T \by - z_i^T A \bz$ and $K(\bz)$ contains terms inde'endent of $\vx$. Thf optimizatiin ovvr $\bx$ is now separabla, i.e. cam be done indeiendeitly for each coordinate. Xe can see that findyng local mnnima over $\bx$ of $Q(\bx, \vz)$ corrgspongs tu itdrauivx lkg-thredhomding.
![Illusfration of wingular-value log-thtefyolding.[]{data-lagel="fig:wefults_svlt"}](figures/SV_thresh.pdf){width="9.5cm"}
Using thjs motivation for ILT, ww can now prove convetgence to sixed points of $f(\bx)$. First, we have:
$f(\hbx^n) = Q(\hbx^{n}, \hbx^t)$ and $D(\hbr^{k+1}, \fvx^{j})$, are monotonically decreasing with iterationf $n$ ax long as the fpectral notm $\Vrtt A \Vert_2 < 1$.
The proof padallels the IHT prlof of [@flumebsath2008iterwtivr] using the fact that $Q(\bx^{n+1}, \vx^n) = f(\bx^n) + \|\bq^{n+1 | robust PCA. Convergence of ILT {#S:ILT_convergece} ================== establish \[thm:ilt\_convergence\]. We define a surrogate $$\begin{aligned} \nonumber Q(\bx, \bz) \Vert \by - \bx \Vert_2^2 + \lambda \sum_i \log(\delta | x_i|)+ \\ \Vert \bx - \bz \Vert_2^2 - \Vert A (\bx - \Vert_2^2\end{aligned}$$ Note that $Q(\bx, \bx) = f(\bx)$. Simplifying (\[eqn:surrogate\]) we have $$\label{funcQ} Q(\bx, = \left( - + \lambda \log( x_i + |\delta|) \right)+ K(\bz),$$ where $k_i(\bz) = z_i + a_i^T \by - A \bz$ and $K(\bz)$ contains terms independent of The optimization over $\bx$ now separable, i.e. can be independently each coordinate. can that local minima over of $Q(\bx, \bz)$ corresponds to iterative log-thresholding. ![Illustration of singular-value log-thresholding.[]{data-label="fig:results_svlt"}](figures/SV_thresh.pdf){width="9.5cm"} Using this motivation for ILT, we now prove fixed points $f(\bx)$. we $f(\hbx^n) = Q(\hbx^{n}, $Q(\hbx^{n+1}, \hbx^{n})$, are monotonically decreasing with long as the spectral norm $\Vert A \Vert_2 1$. The parallels the IHT proof of [@blumensath2008iterative] the fact that $Q(\bx^{n+1}, \bx^n) = f(\bx^n) + | robust PCA.
Convergence of ILT {#s:ILT_converGece}
==================
HEre We eStAbliSh PrOposition \[thm:ilT\_ConvErgence\]. We first define a sUrrogAtE FuncTIoN for $f(\Bx)$ in (\[eqn:LOg\_REG\]): $$\beGiN{aLigNeD}
\LaBel{eqN:suRrogate}
\Nonumber Q(\bX, \bz) = \veRt \by - A \bx \Vert_2^2 + \lAMbDa \sum_i \log(\dEltA + | x_i|)+ \\
\Vert \bx - \bz \VErt_2^2 - \vert A (\bX - \bZ) \VeRT_2^2\end{aLigNed}$$ NoTe that $q(\Bx, \bx) = f(\bX)$. SimplifyInG (\[Eqn:surROgate\]) we HAVe $$\LabeL{funcQ}
Q(\bx, \bz) =
\sum_i \leFT( x_I - K_i(\bz))^2 + \lambda \log( x_I + |\delta|) \RiGHt)+ k(\BZ),$$ whEre $K_i(\bz) = z_i + a_i^T \bY - a_I^T A \bz$ ANd $K(\bz)$ coNTaINS TerMS independent oF $\bx$. The optimIZatIon oveR $\bX$ is NOw sepaRable, I.e. CAn bE done indepeNdenTly for eacH coordINate. We cAN see thaT findiNg lOcaL minIMa OvEr $\bX$ oF $q(\bx, \BZ)$ cOrrESpoNds to iteRaTiVe log-ThreSHOLDing.
![illUstrAtion Of singular-valUe lOg-thREshOldinG.[]{data-LabeL="fIg:resUlts_svLt"}](figUrEs/SV_thresh.pdf){wiDth="9.5cM"}
Using thiS moTiVatIoN for Ilt, we can Now ProVe conveRgence tO FixEd POINtS of $f(\bx)$. First, we have:
$f(\HbX^N) = q(\hBx^{n}, \hbx^n)$ aNd $Q(\hbx^{N+1}, \HbX^{n})$, ARe monotoNiCalLy deCREasinG witH ItErations $N$ as lonG As ThE spectrAl Norm $\VeRt a \VeRt_2 < 1$.
THe proOF parAllels The IHT prOof of [@BLumensath2008iteraTIve] using the faCT tHAT $Q(\BX^{n+1}, \bx^N) = f(\bX^n) + \|\bx^{n+1 | robust PCA.
Convergenceof ILT {#S :ILT_ con ver ge ce}==== ==============
Her e we establish Proposi tion\[ t hm:i l t\ _conv ergence \ ]. W e f ir st de fi n ea sur rog ate fun ction for$f( \b x)$ in (\[eq n :l og\_reg\]) : $ $\begin{alig ned }
\lab el {eq n :surr oga te}
\ nonumb e r Q(\b x, \bz) = \ V ert \b y - A \b x \V ert_ 2^2 + \lambda \su m _i \log(\delta +| x_i| )+ \\ \ Ver t \ bx - \bz \ Ve rt_2^ 2 - \Ver t A ( \ bx- \bz) \Vert_2 ^2\end{alig n ed} $$ Not etha t $Q(\b x, \b x) = f (\bx)$. Sim plif ying (\[e qn:sur r ogate\] ) we hav e $$\l abe l{f uncQ }
Q( \bx ,\ bz) = \ sum _i \left (x_ i - k _i(\ b z ) ) ^2 + \l ambd a \lo g( x_i + |\de lta |) \ r igh t)+ K (\bz) ,$$wh ere $ k_i(\b z) = z _i + a_i^T \by- a_ i^T A \bz $ a nd $K (\ bz)$c ontain s t erm s indep endento f $ \b x $ . T he optimization ov er $ \b x$ is no w sepa r ab le , i.e. ca nbedone i ndepe nden t ly for eac h coor d in at e. We c an see t ha t f ind ing l o calminima over $\ bx$ o f $Q(\bx, \bz)$ corresponds t o i t e ra t ivelog -thresholdi ng.![Il lust r at ion of si ngula r- v al u e log-thresholding. [] {data- label ="fig:results _svlt"}](f i g u res/SV_t hres h .p d f){width="9.5c m"}
Using this motivati on fo r ILT, w e can now p rove con ver gen cetof i xe d points of $ f ( \bx) $. First, we have:
$f (\h bx^ n)=Q(\hbx^{n }, \hbx^ n) $an d$Q( \hbx^ { n+1}, \h bx ^{n }) $,are m o notoni cally dec re as i ngwith it e ra t i ons$n $as l ong a s the spe c tra l norm$\Vert A\Ve r t_2<1$ .
Theproof paralle ls the IHT p ro ofof [@b l u mensath2 008iterative] using the fact th at$Q(\b x^{n +1}, \bx^ n)= f(\b x^n ) + \|\ bx^{n+ 1 | robust_PCA.
Convergence of_ILT {#S:ILT_convergece}
==================
Here we establish_Proposition \[thm:ilt\_convergence\]._We_first define_a_surrogate function for_$f(\bx)$ in (\[eqn:log\_reg\]):_$$\begin{aligned}
\label{eqn:surrogate}
\nonumber Q(\bx, \bz) =_\Vert \by -_A_\bx \Vert_2^2 + \lambda \sum_i \log(\delta + | x_i|)+ \\
\Vert \bx - \bz \Vert_2^2_-_\Vert A_(\bx_-_\bz) \Vert_2^2\end{aligned}$$ Note that $Q(\bx,_\bx) = f(\bx)$. Simplifying (\[eqn:surrogate\])_we have_$$\label{funcQ}
Q(\bx, \bz) =
\sum_i \left(_x_i_- k_i(\bz))^2 +_\lambda \log( x_i + |\delta|) \right)+ K(\bz),$$ where $k_i(\bz)_= z_i + a_i^T \by_- a_i^T A_\bz$_and_$K(\bz)$ contains terms independent_of $\bx$. The optimization over $\bx$_is now separable, i.e. can be_done independently for each coordinate. We can_see that finding local minima over_$\bx$ of $Q(\bx, \bz)$ corresponds_to iterative_log-thresholding.
![Illustration of singular-value log-thresholding.[]{data-label="fig:results_svlt"}](figures/SV_thresh.pdf){width="9.5cm"}
Using this_motivation for ILT,_we can_now prove convergence_to fixed points of $f(\bx)$. First,_we have:
$f(\hbx^n) =_Q(\hbx^{n}, \hbx^n)$ and $Q(\hbx^{n+1}, \hbx^{n})$, are_monotonically_decreasing with iterations_$n$_as_long as_the spectral norm_$\Vert_A \Vert_2_<_1$.
The proof parallels the IHT proof_of_[@blumensath2008iterative] using the fact that $Q(\bx^{n+1}, \bx^n)_= f(\bx^n) + \|\bx^{n+1 |
}(X), f_{\mathcal{M}})$ (or $(\mathcal{K}(X),f_{\mathcal{K}})$) has the same property. In 2005, Banks [@Banks2005] proved that $(\mathcal{K}(X),f_{\mathcal{K}})$ is transitive if and only if $(X,f)$ is weakly mixing. Recently, Li, Yan and Ye [@LiYan2015] investigated the dynamical properties of the connections between $(X,f)$, $(\mathcal{M}(X), f_{\mathcal{M}})$ and $(\mathcal{K}(X),f_{\mathcal{K}})$, focussing on periodic systems, P-systems, M-systems, E-systems, and disjointness. Later, Li, Oprocha, Ye and Zhang [@LiOprocha2017] showed that $(\mathcal{K}(X),f_{\mathcal{K}})$ is pointwise minimal if and only if $(X,f)$ is equicontinuous and that $(\mathcal{K}(X),f_{\mathcal{K}})$ is weakly rigid if and only if $(X,f)$ is uniformly rigid. The above investigations focussed on the qualitative study of the relations between the complexity of original system and that of its induced systems.
Inspired by the concept of Shannon entropy in information theory, Kolmogorov and Sinai introduced measure-theoretic entropy into ergodic theory. Later, Adler, Konheim and McAndrew [@Adler1965] proposed a notion of topological entropy whose definition does not involve any invariant measure. After that, the notion of entropy has played a crucial role in quantifying the degree of disorder in a system. For a topological dynamical system $(X,f)$, by embedding the product of system into induced systems, Bauer and Sigmund [@Bauer1975] showed $$\label{>0}
h_{top}(X,f)>0\Rightarrow h_{top}(f_{\mathcal{M}})=h_{top}(f_{\mathcal{K}})=\infty.$$ In 1995, Glasner and Weiss [@Glasner1995] obtained a remarkable result: $$\label{glasner}
h_{top}(X,f)=0\Rightarrow h_{top}(f_{\mathcal{M}})=0.$$ Later, this connection was further studied by Kerr and Li [@Kerr2005] by shown $(X,f)$ is null if and only if $(\mathcal{M}(X),f_{\mathcal{M}})$ is null. In 2017, | } (X), f_{\mathcal{M}})$ (or $ (\mathcal{K}(X),f_{\mathcal{K}})$) has the same property. In 2005, Banks [ @Banks2005 ] testify that $ (\mathcal{K}(X),f_{\mathcal{K}})$ is transitive if and merely if $ (X, f)$ is weakly mixing. Recently, Li, Yan and Ye [ @LiYan2015 ] investigate the dynamical properties of the association between $ (X, f)$, $ (\mathcal{M}(X), f_{\mathcal{M}})$ and $ (\mathcal{K}(X),f_{\mathcal{K}})$, focussing on periodic organization, P - systems, M - system, E - systems, and disjointness. Later, Li, Oprocha, Ye and Zhang [ @LiOprocha2017 ] showed that $ (\mathcal{K}(X),f_{\mathcal{K}})$ is pointwise minimal if and only if $ (X, f)$ is equicontinuous and that $ (\mathcal{K}(X),f_{\mathcal{K}})$ is weakly inflexible if and only if $ (X, f)$ is uniformly rigid. The above investigation focussed on the qualitative study of the relations between the complexity of original system and that of its induce systems.
Inspired by the concept of Shannon entropy in information theory, Kolmogorov and Sinai introduce measure - theoretic entropy into ergodic theory. Later, Adler, Konheim and McAndrew [ @Adler1965 ] nominate a impression of topological entropy whose definition does not involve any invariant measure. After that, the notion of entropy has played a crucial function in quantify the degree of disorder in a organization. For a topological dynamic organization $ (X, f)$, by embedding the merchandise of system into induced systems, Bauer and Sigmund [ @Bauer1975 ] showed $ $ \label{>0 }
h_{top}(X, f)>0\Rightarrow h_{top}(f_{\mathcal{M}})=h_{top}(f_{\mathcal{K}})=\infty.$$ In 1995, Glasner and Weiss [ @Glasner1995 ] prevail a remarkable result: $ $ \label{glasner }
h_{top}(X, f)=0\Rightarrow h_{top}(f_{\mathcal{M}})=0.$$ Later, this connection was further studied by Kerr and Li [ @Kerr2005 ] by testify $ (X, f)$ is null if and only if $ (\mathcal{M}(X),f_{\mathcal{M}})$ is null. In 2017, | }(X), f_{\lathcal{M}})$ (or $(\mathcal{K}(X),f_{\mauhcal{K}})$) has the same propxrty. In 2005, Banks [@Canks2005] proved that $(\mathcal{K}(X),f_{\larhcal{J}})$ is transitive if and only if $(Q,f)$ is weajly nuxing. Receifly, Li, Ncn ans Ye [@NmYan2015] investigatgd the dynamhcal propertiev uf the connections between $(X,f)$, $(\mathcal{I}(X), f_{\matncwl{M}})$ and $(\mathcaj{K}(X),f_{\kwthczl{K}})$, focussing on periodic systems, L-systemv, M-systems, E-sustems, and disjointness. Laher, Pi, Oprocha, Ye and Xhang [@LiOpricha2017] whowed that $(\oathcal{K}(X),f_{\mathcal{K}})$ is kointwise minimal if and only if $(X,f)$ ix equicontunyoud and that $(\methcal{H}(X),f_{\mathcal{K}})$ lx weakny rigic if and only lf $(X,f)$ is uniformly rigid. The ebove investigations focussed ou the qualitative stuey of tve ralatkins beuwexn fhe colplxxity of orjginal systwm and that of its onqlved systems.
Ihspireq fy the concept of Shannon entropy in inxorjation theory, Kolmogoroc and Sinai introducef measure-eheoretic entropy into ergodic theory. Later, Adler, Konhxio aue McAverfw [@Adler1965] proposed a notion of topological entwkpu ehose definitijn does not ijvpjve any invarkant mzzshre. After that, the notion of ebtropy haf plsyed a crucial role in quanrifying the begeee of disorder in a system. Fur a topokogical dynamical systeo $(X,f)$, by embeddijg the prkauct of system ivto itduced systems, Bauer and Sygmund [@Baner1975] skowed $$\lacel{>0}
h_{jop}(X,f)>0\Ridhtarrow h_{hop}(f_{\mathcal{M}})=h_{top}(f_{\mathcal{N}})=\infti.$$ In 1995, Clasner anf Weiss [@Glasner1995] obtained a remaclable result: $$\ldben{glasner}
k_{top}(X,f)=0\Vightarrow h_{top}(s_{\mathcal{M}})=0.$$ Latet, this counectiun was furnher studmed by Kerr wnd Li [@Kerr2005] bf shown $(X,f)$ is null if and onlt if $(\magfcal{M}(X),f_{\mathcal{K}})$ is null. In 2017, | }(X), f_{\mathcal{M}})$ (or $(\mathcal{K}(X),f_{\mathcal{K}})$) has the same 2005, [@Banks2005] proved $(\mathcal{K}(X),f_{\mathcal{K}})$ is transitive is mixing. Recently, Li, and Ye [@LiYan2015] the dynamical properties of the connections $(X,f)$, $(\mathcal{M}(X), f_{\mathcal{M}})$ and $(\mathcal{K}(X),f_{\mathcal{K}})$, focussing on periodic systems, P-systems, M-systems, E-systems, and Later, Li, Oprocha, Ye and Zhang [@LiOprocha2017] showed that $(\mathcal{K}(X),f_{\mathcal{K}})$ is pointwise minimal and if is and that $(\mathcal{K}(X),f_{\mathcal{K}})$ is weakly rigid if and only if $(X,f)$ is uniformly rigid. The above focussed on the qualitative study of the relations the complexity of original and that of its induced Inspired the concept Shannon in theory, Kolmogorov and introduced measure-theoretic entropy into ergodic theory. Later, Adler, Konheim and McAndrew [@Adler1965] proposed a notion of topological whose definition involve any measure. that, notion of entropy a crucial role in quantifying the in a system. For a topological dynamical system by embedding product of system into induced systems, and Sigmund [@Bauer1975] showed $$\label{>0} h_{top}(X,f)>0\Rightarrow h_{top}(f_{\mathcal{M}})=h_{top}(f_{\mathcal{K}})=\infty.$$ In Glasner and Weiss [@Glasner1995] obtained a remarkable result: $$\label{glasner} h_{top}(X,f)=0\Rightarrow h_{top}(f_{\mathcal{M}})=0.$$ Later, this connection was by Kerr and Li by shown $(X,f)$ null and if is null. 2017, | }(X), f_{\mathcal{M}})$ (or $(\mathcal{K}(X),f_{\matHcal{K}})$) has thE same ProPerTy. in 2005, BaNks [@BAnks2005] proved that $(\MAthcAl{K}(X),f_{\mathcal{K}})$ is transitIve if AnD Only IF $(X,F)$ is weAkly mixINg. rECenTlY, LI, YaN aND YE [@LiYaN2015] inVestigaTed the dynaMicAl Properties of THe ConnectionS beTween $(X,f)$, $(\mathcAl{M}(x), f_{\mathCaL{M}})$ aND $(\mathCal{k}(X),f_{\maThcal{K}})$, FOcussiNg on perioDiC SystemS, p-systemS, m-SyStemS, E-systems, and disjoINtNEss. Later, Li, OproCha, Ye aNd zHaNG [@liOProCha2017] showed tHaT $(\mathCAl{K}(X),f_{\maTHcAL{k}})$ Is pOIntwise minimaL if and only iF $(x,f)$ iS equicOnTinUOus and That $(\mAtHCal{k}(X),f_{\mathcal{K}})$ Is weAkly rigid If and oNLy if $(X,f)$ iS UniformLy rigiD. ThE abOve iNVeStIgaTiONs fOCuSseD On tHe qualitAtIvE studY of tHE RELatiOns BetwEen thE complexity of OriGinaL SysTem anD that Of itS iNduceD systeMs.
InsPiRed by the concept Of ShAnnon entrOpy In InfOrMatioN Theory, kolMogOrov and sinai inTRodUcED MEaSure-theoretic entroPy INTo Ergodic tHeory. LATeR, ADLer, KonheIm And mcAnDREw [@AdlEr1965] prOPoSed a notiOn of toPOlOgIcal entRoPy whosE dEfiNitIon doES not InvolvE any invaRiant MEasure. After thaT, The notion of enTRoPY HaS PlayEd a Crucial role In quANtifYing THe DegREe of dIsordEr IN a SYstem. For a topologicaL dYnamicAl sysTem $(X,f)$, by embeddIng the prodUCT Of system Into INdUCed systems, BaueR and SIgmund [@BaueR1975] Showed $$\laBel{>0}
h_{tOp}(X,f)>0\RighTarrow h_{toP}(F_{\Mathcal{M}})=H_{toP}(f_{\mAthCal{k}})=\INfTy.$$ In 1995, Glasner anD wEiss [@glAsner1995] obTaiNed a remArkAblE reSulT: $$\lAbel{glasnEr}
h_{top}(X,f)=0\riGhTaRrOw h_{Top}(f_{\mAThcal{M}})=0.$$ LaTeR, thIs ConNectiON was fuRther StudIeD bY kerR and Li [@KERr2005] BY ShowN $(X,F)$ iS nulL if AnD only If $(\maTHcaL{M}(X),f_{\matHcal{M}})$ is nuLl. IN 2017, | }(X), f_{\mathcal{M}})$ (o r $(\mathc al{K} (X) ,f_ {\ math cal{ K}})$) has the same property. In 2005, Ba nks [ @B a nks2 0 05 ] pro ved tha t $ ( \ mat hc al {K} (X ) ,f _{\ma thc al{K}}) $ is trans iti ve if and only if $(X,f)$ i s w eakly mixing . R ecentl y, Li , Yanand Ye [ @LiYan 2 015] i nvestigat ed the dy n amicalp r op erti es of the connect i on s between $(X,f )$, $( \m a th c a l{M }(X ), f_{\mat hc al{M} } )$ and$ (\ m a t hca l {K}(X),f_{\ma thcal{K}})$ , fo cussin gonp eriodi c sys te m s,P-systems,M-sy stems, E- system s , and d i sjointn ess. L ate r,Li,O pr oc ha, Y e an d Z han g [@ LiOproch a2 01 7] sh owed t h a t $( \ma thca l{K}( X),f_{\mathca l{K }})$ ispoint wisemini ma l ifand on ly if $ (X,f)$ is equic onti nuous and th at $( \m athca l {K}(X) ,f_ {\m athcal{ K}})$ i s we ak l y ri gid if and only if $ ( X ,f )$ is un iforml y r ig i d. The a bo veinve s t igati onsf oc ussed on the q u al it ative s tu dy ofth e r ela tions betw een th e comple xityo f original sys t em and that o f i t s i n duce d s ystems.
In spir e d by the co nce p t ofShann on en t ropy in information t heory, Kolm ogorov and Si nai introd u c e d measur e-th e or e tic entropy in to er godic theo r y. Later , Adl er, Konh eim and M c A ndrew [@ Adl er1 965 ] p r o po sed a notiono f top ol ogicalent ropy wh ose de fin iti on does not involve a ny i nv ari ant m e asure. A ft erth at, then otionof en trop yha s pl ayed ac ru c i al r ol ein q uan ti fying the deg ree ofdisorderina sys te m. For atopological d yn amical sys te m $ (X,f)$ , by embed ding the product of sys t em into in duced sys tems, Bau erand Si gmu n d [@Ba uer197 5] sh ow ed$ $ \labe l { >0 }
h _{ top}(X,f)> 0 \ Rig htarr ow h_{ top}(f_ {\mathcal{M}})=h_{ t op} (f_{\mathcal{ K}} )=\i n f ty .$$ In 199 5, Gla s n er and Weiss [@ Glasner199 5] ob tained a r e mar ka ble res ult: $$ \labe l {glasne r}
h_{top }(X,f)=0\ Ri ghta r r owh_{top}(f_ {\mathca l{M}})=0. $ $ Lat e r, this co nnecti on wa s fur ther s t udi ed by Kerran d Li [ @Kerr 20 05] by s hown $(X,f)$ is null if and o nly i f $ (\mathcal {M} ( X), f_{\mathc al{M }})$ is nu ll. In 2017 , | }(X), f_{\mathcal{M}})$_(or $(\mathcal{K}(X),f_{\mathcal{K}})$)_has the same property._In 2005,_Banks_[@Banks2005] proved_that_$(\mathcal{K}(X),f_{\mathcal{K}})$ is transitive_if and only_if $(X,f)$ is weakly_mixing. Recently, Li,_Yan_and Ye [@LiYan2015] investigated the dynamical properties of the connections between $(X,f)$, $(\mathcal{M}(X), f_{\mathcal{M}})$_and_$(\mathcal{K}(X),f_{\mathcal{K}})$, focussing_on_periodic_systems, P-systems, M-systems, E-systems, and_disjointness. Later, Li, Oprocha, Ye_and Zhang_[@LiOprocha2017] showed that $(\mathcal{K}(X),f_{\mathcal{K}})$ is pointwise minimal if_and_only if $(X,f)$_is equicontinuous and that $(\mathcal{K}(X),f_{\mathcal{K}})$ is weakly rigid if_and only if $(X,f)$ is uniformly_rigid. The above_investigations_focussed_on the qualitative study_of the relations between the complexity_of original system and that of_its induced systems.
Inspired by the concept of_Shannon entropy in information theory, Kolmogorov_and Sinai introduced measure-theoretic entropy_into ergodic_theory. Later, Adler, Konheim and_McAndrew [@Adler1965] proposed_a notion_of topological entropy_whose definition does not involve any_invariant measure. After_that, the notion of entropy has_played_a crucial role_in_quantifying_the degree_of disorder in_a_system. For_a_topological dynamical system $(X,f)$, by embedding_the_product of system into induced systems, Bauer_and Sigmund [@Bauer1975] showed_$$\label{>0}
h_{top}(X,f)>0\Rightarrow_h_{top}(f_{\mathcal{M}})=h_{top}(f_{\mathcal{K}})=\infty.$$ In 1995, Glasner_and Weiss [@Glasner1995] obtained a_remarkable result: $$\label{glasner}
h_{top}(X,f)=0\Rightarrow h_{top}(f_{\mathcal{M}})=0.$$ Later, this_connection was_further studied_by Kerr and Li [@Kerr2005] by shown $(X,f)$ is null if_and only if $(\mathcal{M}(X),f_{\mathcal{M}})$ is null._In 2017, |
===========
We have provided evidence that when both Mandelstam variables $s$ and $t$ turn out to be much less than the large momentum scale $Q^2$, with the variables $s/Q^2$ and $t/Q^2$ varying in the interval $[0.001, \, 0.7]$, then the TDA and the GDA factorization mechanisms are equivalent to each other and operate in parallel. We have also demonstrated that duality may serve as a tool for selecting suitable models for the nonperturbative ingredients of various exclusive amplitudes entering QCD factorization. In this context, we observed that twist-3 GDAs appear to be dual to the convolutions of leading-twist TDAs and DAs, multiplied by a QCD effective coupling.
Acknowledgments {#acknowledgments.unnumbered}
---------------
We would like to thank A. P. Bakulev, A. V. Efremov, N. Kivel, B. Pire, M. V. Polyakov, M. Prasza[ł]{}owicz, L. Szymanowski, and S. Wallon for useful discussions and remarks. This investigation was partially supported by the Heisenberg-Landau Programme (Grant 2008), the Alexander von Humboldt Stiftung, the Deutsche Forschungsgemeinschaft under contract 436RUS113/881/0, the EU-A7 Project *Transversity*, the RFBR (Grants 06-02-16215,08-02-00896 and 07-02-91557), the Russian Federation Ministry of Education and Science (Grant MIREA 2.2.2.2.6546), the RF Scientific Schools grant 195.2008.9, and INFN.
[99]{}
A. V. Efremov and A. V. Radyushkin, Phys. Lett. B [**94**]{}, 245 (1980). Theor. Math. Phys. [**42**]{}, 97 (1980) \[Teor. Mat. Fiz. [**42**]{}, 147 (1980)\]. G. P. Lepage and S. J. Brodsky, Phys. Lett. B [**87**]{}, 359 (1979); Phys. Rev. D [**22**]{}, 2157 (1980). J. C. Collins, D | = = = = = = = = = = =
We have provided evidence that when both Mandelstam variables $ s$ and $ t$ turn out to be much less than the large momentum plate $ Q^2 $, with the variable star $ s / Q^2 $ and $ t / Q^2 $ varying in the time interval $ [ 0.001, \, 0.7]$, then the TDA and the GDA factorization mechanisms are equivalent to each early and operate in parallel. We have besides demonstrated that duality may serve as a instrument for selecting suitable model for the nonperturbative ingredients of various exclusive amplitudes entering QCD factorization. In this context, we observed that twist-3 GDAs look to be dual to the convolutions of leading - construction TDAs and DAs, multiplied by a QCD effective yoke.
Acknowledgments { # acknowledgments.unnumbered }
---------------
We would wish to thank A. P. Bakulev, A. V. Efremov, N. Kivel, B. Pire, M. V. Polyakov, M. Prasza[ł]{}owicz, L. Szymanowski, and S. Wallon for useful discussions and remarks. This investigation was partially supported by the Heisenberg - Landau Programme (Grant 2008), the Alexander von Humboldt Stiftung, the Deutsche Forschungsgemeinschaft under contract 436RUS113/881/0, the EU - A7 Project * Transversity *, the RFBR (Grants 06 - 02 - 16215,08 - 02 - 00896 and 07 - 02 - 91557), the Russian Federation Ministry of Education and Science (Grant MIREA 2.2.2.2.6546), the RF Scientific Schools grant 195.2008.9, and INFN.
[ 99 ] { }
A. V. Efremov and A. V. Radyushkin, Phys. Lett. B [ * * 94 * * ] { }, 245 (1980). Theor. Math. Phys. [ * * 42 * * ] { }, 97 (1980) \[Teor. Mat. Fiz. [ * * 42 * * ] { }, 147 (1980)\ ]. G. P. Lepage and S. J. Brodsky, Phys. Lett. B [ * * 87 * * ] { }, 359 (1979); Phys. Rev. D [ * * 22 * * ] { }, 2157 (1980). J. C. Collins, D |
===========
We jave provided evidence tmat when both Maueelstak varizbles $s$ xnd $t$ turn out to be much leds than the large momentum scxle $Q^2$, witj the vaeiabows $s/Q^2$ and $v/S^2$ varyiky in fme incecval $[0.001, \, 0.7]$, then thg TDA and tha GDA factorizdtkou mechanisms are equivalent to each jther amd operate in patallek. We gave also demonstrated that dualitg may strve as a tool for selecting suitable models for the nonperturbatige ingrediebts jd various exzlusive amklntudes enteting QCD factorization. In this cuntexc, we observgb thwj twist-3 GDAs appewr to be dual to the convolitions of leadlng-twmst RDAs and DAs, multiplixd by a QCD effectivg coupling.
Dcinowledgments {#ackboqledgkentv.unnjnbefed}
---------------
Se wkuld llke to thank Z. P. Bakulev, A. C. Efremov, N. Kivel, B. Pite, N. V. Polyakov, M. Pdasza[ł]{}orisz, L. Szymanowski, and S. Wallon for useful diacussions and remarks. Tyis investigation was partiallr supported by the Heisenberg-Landau Programme (Gratt 2008), tie Altxqnder coj Humboldt Stiftung, the Deutsche Forschungsgeisimsbhaft under contrcct 436RUS113/881/0, the EU-A7 Lrljrst *Transversijy*, the XRBD (Grants 06-02-16215,08-02-00896 and 07-02-91557), the Russiag Fedwration Mynisyry of Education and Sciencw (Grant MIREC 2.2.2.2.6546), rhe RF Scientific Dchools graut 195.2008.9, anc INFM.
[99]{}
A. V. Efremov and A. V. Radyuskkin, Pgys. Lett. B [**94**]{}, 245 (1980). Theor. Maff. Phys. [**42**]{}, 97 (1980) \[Teor. Mat. Wiz. [**42**]{}, 147 (1980)\]. G. P. Lepage and S. J. Brodsky, Prys. Lett. B [**87**]{}, 359 (1979); 'hys. Rev. A [**22**]{}, 2157 (1980). J. C. Coljins, D | =========== We have provided evidence that when variables and $t$ out to be momentum $Q^2$, with the $s/Q^2$ and $t/Q^2$ in the interval $[0.001, \, 0.7]$, the TDA and the GDA factorization mechanisms are equivalent to each other and in parallel. We have also demonstrated that duality may serve as a tool selecting models the ingredients of various exclusive amplitudes entering QCD factorization. In this context, we observed that twist-3 GDAs to be dual to the convolutions of leading-twist and DAs, multiplied by QCD effective coupling. Acknowledgments {#acknowledgments.unnumbered} We like to A. Bakulev, V. Efremov, N. B. Pire, M. V. Polyakov, M. Prasza[ł]{}owicz, L. Szymanowski, and S. Wallon for useful discussions and remarks. investigation was by the Programme 2008), Alexander von Humboldt Deutsche Forschungsgemeinschaft under contract 436RUS113/881/0, the the RFBR (Grants 06-02-16215,08-02-00896 and 07-02-91557), the Russian Ministry of and Science (Grant MIREA 2.2.2.2.6546), the Scientific Schools grant 195.2008.9, and INFN. [99]{} A. Efremov and A. V. Radyushkin, Phys. Lett. B [**94**]{}, 245 (1980). Theor. Math. Phys. [**42**]{}, \[Teor. Mat. Fiz. [**42**]{}, (1980)\]. G. P. and J. Phys. B [**87**]{}, (1979); Phys. Rev. D [**22**]{}, 2157 (1980). J. C. Collins, D |
===========
We have provided evidence thaT when both MAndelStaM vaRiAbleS $s$ anD $t$ turn out to be mUCh leSs than the large momentum Scale $q^2$, wITh thE VaRiablEs $s/Q^2$ and $T/q^2$ vARYinG iN tHe iNtERvAl $[0.001, \, 0.7]$, theN thE TDA and The GDA factOriZaTion mechanisMS aRe equivaleNt tO each other anD opErate iN pAraLLel. We HavE also DemonsTRated tHat dualitY mAY serve AS a tool fOR SeLectIng suitable models FOr THe nonperturbatIve ingReDIeNTS of VarIous exclusIvE amplITudes enTErING qCD FActorization. IN this contexT, We oBserveD tHat TWist-3 GDas appEaR To bE dual to the cOnvoLutions of LeadinG-Twist TDaS and DAs, MultipLieD by A QCD EFfEcTivE cOUplINg.
ackNOwlEdgments {#AcKnOwledGmenTS.UNNumbEreD}
---------------
We wOuld lIke to thank A. P. BAkuLev, A. v. efrEmov, N. kivel, b. PirE, M. v. PolyAkov, M. PRasza[Ł]{}oWicz, L. SzymanowskI, and s. Wallon foR usEfUl dIsCussiONs and rEmaRks. this invEstigatIOn wAs PARTiAlly supported by the heISEnBerg-LandAu ProgRAmMe (gRant 2008), the ALeXanDer vON humboLdt STIfTung, the DEutschE foRsChungsgEmEinschAfT unDer ContrACt 436RUs113/881/0, the EU-a7 Project *transVErsity*, the RFBR (GRAnts 06-02-16215,08-02-00896 and 07-02-91557), the RusSIaN fEdERatiOn MInistry of EdUcatIOn anD SciENcE (GrANt MIReA 2.2.2.2.6546), the rF sCiENtific Schools grant 195.2008.9, aNd iNFN.
[99]{}
A. V. efremOv and A. V. RadyusHkin, Phys. LeTT. b [**94**]{}, 245 (1980). theor. MatH. PhyS. [**42**]{}, 97 (1980) \[teOR. Mat. Fiz. [**42**]{}, 147 (1980)\]. G. P. LepagE and S. j. Brodsky, PhYS. Lett. B [**87**]{}, 359 (1979); PhYs. Rev. d [**22**]{}, 2157 (1980). J. C. ColliNs, D |
===========
We have prov ided evide nce t hat wh en bot h Ma ndelstam varia b les$s$ and $t$ turn out t o bemu c h le s sthanthe lar g em o men tu msca le $Q ^2$,wit h the v ariables $ s/Q ^2 $ and $t/Q^2 $ v arying inthe interval $[ 0.0 01, \, 0 .7] $ , the n t he TD A andt he GDA factoriz at i on mec h anismsa r eequi valent to each ot h er and operate in paral le l .W e ha vealso demon st rated that du a li t y may serve as a to ol for sele c tin g suit ab lem odelsfor t he non perturbativ e in gredients of va r ious ex c lusiveamplit ude s e nter i ng Q CDfa c tor i za tio n . I n this c on te xt, w e ob s e r v ed t hat twi st-3GDAs appear t o b e du a l t o the conv olut io ns of leadi ng-tw is t TDAs and DAs, mul tiplied b y a Q CDef fecti v e coup lin g.
Acknow ledgmen t s { #a c k n ow ledgments.unnumber ed } -- -------- -----We w o uld like t o t hank A . P.Baku l ev , A. V.Efremo v ,N. Kivel, B . Pire ,M.V.Polya k ov,M. Pra sza[ł]{} owicz , L. Szymanows k i, and S. Wal l on f or usef uldiscussions and rema rks. Th isi nvest igati on wa s partially supporte dby the Heis enberg-Landau Programme ( G rant 200 8),t he Alexander vonHumbo ldt Stiftu n g, the D eutsc he Forsc hungsgeme i n schaft u nde r c ont rac t 43 6RUS113/881/0 , theEU -A7 Pro jec t *Tran sve rsi ty* , t he RFBR (Gr ants 06- 02 -1 62 15 ,08 -02-0 0 896 and07 -02 -9 155 7), t h e Russ ian F eder at io n Mi nistryo fE d ucat io nandSci en ce (G rant MIR EA 2.2. 2.2.6546) , t h e RF S ci entific Schools gran t195.2008.9 ,and INFN. [99]{}
A. V. Efremov and A. V. Radyush kin , Phy s. L ett. B [ **9 4**]{} , 2 4 5 (198 0). Th eor.Ma th. P hys.[ ** 42* *] {}, 97 (19 8 0 ) \ [Teor .Mat. Fiz. [**42**]{}, 147 (1 9 80) \]. G. P. Lep age and S .J.B ro d sky ,P hys . Lett. B [**87* *]{}, 359(1 9 79 ); Phys. R e v. D [**22* *]{}, 2 157 ( 1 980). J . C. Coll ins, D |
===========
We have_provided evidence_that when both Mandelstam_variables $s$_and_$t$ turn_out_to be much_less than the_large momentum scale $Q^2$,_with the variables_$s/Q^2$_and $t/Q^2$ varying in the interval $[0.001, \, 0.7]$, then the TDA and the_GDA_factorization mechanisms_are_equivalent_to each other and operate_in parallel. We have also_demonstrated that_duality may serve as a tool for selecting_suitable_models for the_nonperturbative ingredients of various exclusive amplitudes entering QCD factorization._In this context, we observed that_twist-3 GDAs appear_to_be_dual to the convolutions_of leading-twist TDAs and DAs, multiplied_by a QCD effective coupling.
Acknowledgments {#acknowledgments.unnumbered}
---------------
We_would like to thank A. P. Bakulev, A. V. Efremov, N. Kivel,_B. Pire, M. V. Polyakov, M. Prasza[ł]{}owicz, L. Szymanowski, and_S. Wallon for useful discussions and_remarks. This_investigation was partially supported by_the Heisenberg-Landau Programme_(Grant 2008),_the Alexander von_Humboldt Stiftung, the Deutsche Forschungsgemeinschaft under_contract 436RUS113/881/0, the_EU-A7 Project *Transversity*, the RFBR (Grants_06-02-16215,08-02-00896_and 07-02-91557), the_Russian_Federation_Ministry of_Education and Science_(Grant_MIREA 2.2.2.2.6546),_the_RF Scientific Schools grant 195.2008.9, and_INFN.
[99]{}
A. V. Efremov_and A. V. Radyushkin, Phys. Lett. B [**94**]{}, 245 (1980)._Theor. Math. Phys. [**42**]{}, 97 (1980)_\[Teor. Mat. Fiz. _[**42**]{}, 147 (1980)\]. G. P. Lepage_and S. J. Brodsky, Phys. Lett. B [**87**]{},_359 (1979); Phys. Rev. D [**22**]{}, 2157_(1980). J. C. Collins,_D |
In the latter two cases, broken parity and time reversal symmetries combine to destabilize the spatially uniform state, giving rise to the spatial dependence in the gap functions. The former two states are unlikely in any theory that is based on local interactions (like the single band Hubbard model). Since the above three states are not s-wave pairing states, and (as argued below) Li$_{2}$Pd$_{3}$B appears to be s-wave, a phase transition in the pairing state of Li$_{2}$(Pd$_{1-x}$Pt$_{x}$)$_{3}$B with varying $x$ would have to occur for any of these states to exist in Li$_{2}$Pt$_{3}$B. Furthermore, these states should be extremely sensitive to impurities and $%
T_{c}$ should be strongly suppressed when $x$ is varied away from 1. These are in contrast with the experimentally observed smooth evolution of $T_{c}$ with $x$ [@Badica; @2005]. Given these arguments against unconventional superconductivity, we attribute the dramatic difference between these two compounds to the variation of ASOC.
When parity symmetry is violated, the ASOC that breaks the spin degeneracy of each band takes the form $\alpha\mathbf{g(k)\cdot S(k)}/\hbar$, where $%
\alpha$ denotes the spin-orbit coupling strength, $\mathbf{S(k)}$ is the spin of an electron with momentum $\hbar \mathbf{k}$, and $\mathbf{g(k)}$ is a dimensionless vector ($\mathbf{g(-k)=-g(k)}$ to preserve time reversal symmetry). This ASOC leads to an energy splitting of the originally degenerate spin states and results in spin-eigenstates that are polarized parallel or anti-parallel to $\mathbf{g(k)}$. The ASOC plays a crucial role in the determination of the superconducting state. The key point is that if a spin-triplet contribution to the superconducting gap function is to emerge, its characteristic d-vector $\mathbf{d(k)}$ must be parallel to $%
\mathbf{g(k)}$ (provided that the ASOC is sufficiently large) [@Frigeri; @2004; @Frigeri; @2005]. This leads to two gap functions $\Delta_{\pm}(\mathbf{k}%
)=\psi\pm t \mid\mathbf{g(k)}\mid$, where each gap is defined on one of the two bands formed by the degeneracy lifting of the ASOC; $\ | In the latter two cases, broken parity and time reversal isotropy compound to destabilize the spatially uniform state, giving raise to the spatial dependence in the gap function. The early two states are improbable in any theory that is based on local interaction (like the single band Hubbard mannequin). Since the above three country are not s - wave pairing states, and (as argued below) Li$_{2}$Pd$_{3}$B appears to be second - wave, a phase transition in the pairing state of Li$_{2}$(Pd$_{1 - x}$Pt$_{x}$)$_{3}$B with vary $ x$ would have to occur for any of these states to exist in Li$_{2}$Pt$_{3}$B. Furthermore, these states should be highly sensitive to impurity and $%
T_{c}$ should be strongly suppressed when $ x$ is varied aside from 1. These are in contrast with the experimentally observed smooth evolution of $ T_{c}$ with $ x$ [ @Badica; @2005 ]. Given these arguments against unconventional superconductivity, we attribute the dramatic difference between these two compounds to the variation of ASOC.
When parity symmetry is violated, the ASOC that breaks the spin degeneracy of each set takes the form $ \alpha\mathbf{g(k)\cdot S(k)}/\hbar$, where $%
\alpha$ announce the spin - orbit coupling strength, $ \mathbf{S(k)}$ is the tailspin of an electron with momentum $ \hbar \mathbf{k}$, and $ \mathbf{g(k)}$ is a dimensionless vector ($ \mathbf{g(-k)=-g(k)}$ to preserve meter reversal symmetry). This ASOC leads to an department of energy splitting of the originally degenerate spin states and results in spin - eigenstates that are polarized parallel or anti - parallel to $ \mathbf{g(k)}$. The ASOC plays a crucial role in the determination of the superconducting state. The cardinal point is that if a tailspin - triplet contribution to the superconducting gap function is to emerge, its characteristic d - vector $ \mathbf{d(k)}$ must be parallel to $%
\mathbf{g(k)}$ (provided that the ASOC is sufficiently big) [ @Frigeri; @2004; @Frigeri; @2005 ]. This leads to two gap functions $ \Delta_{\pm}(\mathbf{k}%
) = \psi\pm metric ton \mid\mathbf{g(k)}\mid$, where each gap is defined on one of the two bands formed by the corruption lifting of the ASOC; $ \ |
In hhe latter two cases, broyen parity and jine revxrsal sgmmetrier combine to destabilize the spatiaoly uniform state, givivg rise tl the spqtiao dependencx in the gap fhkctious. The former twp states ase unlikely in avy theory that is based on local intewactionx (pike the singlg banc Hubgard model). Since the above three sfates age not s-wave pairong states, and (as argued bflow) Li$_{2}$Pd$_{3}$B appears to he s-wave, a kgasq transition kn the paignng state or Li$_{2}$(Pd$_{1-x}$Pt$_{x}$)$_{3}$B with varying $x$ woula havz to occur doe aji of these svates no exist in Ll$_{2}$It$_{3}$B. Furdhermorr, these states shmule be extremely sensitmve to impurities anq $%
T_{c}$ shound be strongly suppeewsed fhen $x$ ir vafies ewag from 1. Tiese are in contrast wuth the experimentaklr observed smokth evjlttion of $T_{c}$ with $x$ [@Badica; @2005]. Given these drghments against unconvenrional superconductivlty, we ateribute the dramatic difference between these two compkjndw bo tfw gariation of ASOC.
When parity symmetry is violwfec, nhe ASOC that brecks the spin degrnfrssy of each bavd takza fhe form $\alpha\mathhf{g(k)\cdoj S(k)}/\hbqr$, where $%
\wlphs$ denotes the spin-orbit coupling strengnh, $\mqthbf{S(k)}$ is the spiu of an eleccron woth mpmentum $\hbar \mathbf{k}$, anb $\mathgf{g(k)}$ is a dlmensionlsrs vector ($\mathbf{e(-k)=-g(l)}$ do preserve time reversal fymmetry). Vhis CSOC leaas tp an egergy splihting of the originally fegengrate vpin stated and results in spin-eigenstates that are polsrhzeg parallzl or snti-parallel eo $\mathbf{g(k)}$. Thg ASOC plcys a zrucial rome in tie determinaeion of the sgkerconducting state. Tre kwy piint is ghat if a spin-yriplet contributiob to the supercondmctine gap function iw ti emerge, its chsraztewidtmc d-vqwtor $\mathbf{d(n)}$ murt cr parxllel to $%
\mabhbw{g(k)}$ (lrovided that the ASMC ia sufficiently larbe) [@Frigeri; @2004; @Frigery; @2005]. This leads to two gap functilns $\Dxlta_{\pm}(\oathbg{k}%
)=\pfi\pm t \mid\mathbf{g(k)}\mid$, where eadh gap is denined on one jf tme tro bands fprmed by the degeneracy lifting of the ESOC; $\ | In the latter two cases, broken parity reversal combine to the spatially uniform spatial in the gap The former two are unlikely in any theory that based on local interactions (like the single band Hubbard model). Since the above states are not s-wave pairing states, and (as argued below) Li$_{2}$Pd$_{3}$B appears to s-wave, phase in pairing state of Li$_{2}$(Pd$_{1-x}$Pt$_{x}$)$_{3}$B with varying $x$ would have to occur for any of these states exist in Li$_{2}$Pt$_{3}$B. Furthermore, these states should be sensitive to impurities and T_{c}$ should be strongly suppressed $x$ varied away 1. are contrast with the observed smooth evolution of $T_{c}$ with $x$ [@Badica; @2005]. Given these arguments against unconventional superconductivity, we attribute dramatic difference two compounds the of When parity symmetry the ASOC that breaks the spin band takes the form $\alpha\mathbf{g(k)\cdot S(k)}/\hbar$, where $% denotes the coupling strength, $\mathbf{S(k)}$ is the spin an electron with momentum $\hbar \mathbf{k}$, and $\mathbf{g(k)}$ a dimensionless vector ($\mathbf{g(-k)=-g(k)}$ to preserve time reversal symmetry). This ASOC leads to an energy the originally degenerate spin and results in that polarized or to $\mathbf{g(k)}$. ASOC plays a crucial role in the determination of the superconducting The key point is that if a spin-triplet contribution to gap is to emerge, characteristic d-vector $\mathbf{d(k)}$ must parallel $% \mathbf{g(k)}$ (provided that is large) @2005]. leads two gap functions $\Delta_{\pm}(\mathbf{k}% t \mid\mathbf{g(k)}\mid$, where each gap defined on one of the degeneracy lifting of the ASOC; $\ |
In the latter two cases, broken Parity and tIme reVerSal SyMmetRies Combine to destaBIlizE the spatially uniform stAte, giViNG risE To The spAtial dePEnDENce In ThE gaP fUNcTions. the Former tWo states arE unLiKely in any theORy That is baseD on Local interacTioNs (like ThE siNGle baNd HUbbarD model). sInce thE above thrEe STates aRE not s-waVE PaIrinG states, and (as argueD BeLOw) Li$_{2}$Pd$_{3}$B appears To be s-wAvE, A pHASe tRanSition in thE pAirinG State of lI$_{2}$(PD$_{1-X}$pT$_{x}$)$_{3}$B WIth varying $x$ woUld have to ocCUr fOr any oF tHesE States To exiSt IN Li$_{2}$pt$_{3}$B. FurthermOre, tHese stateS shoulD Be extreMEly sensItive tO imPurItieS AnD $%
T_{C}$ shOuLD be STrOngLY suPpressed WhEn $X$ is vaRied AWAY From 1. theSe arE in coNtrast with the ExpErimENtaLly obServeD smoOtH evolUtion oF $T_{c}$ wiTh $X$ [@Badica; @2005]. Given theSe arGuments agAinSt UncOnVentiONal supErcOndUctivitY, we attrIButE tHE DRaMatic difference betWeEN ThEse two coMpoundS To ThE VariatioN oF ASoC.
WhEN ParitY symMEtRy is violAted, thE aSoC That breAkS the spIn DegEneRacy oF Each Band taKes the foRm $\alpHA\mathbf{g(k)\cdot S(K)}/\Hbar$, where $%
\alphA$ DeNOTeS The sPin-Orbit coupliNg stREngtH, $\matHBf{s(k)}$ iS The spIn of aN eLEcTRon with momentum $\hbar \MaThbf{k}$, aNd $\matHbf{g(k)}$ is a dimenSionless veCTOR ($\mathbf{g(-K)=-g(k)}$ tO PrEServe time reverSal syMmetry). This asOC leads To an eNergy splItting of tHE OriginalLy dEgeNerAte SPIn States and resuLTS in sPiN-eigensTatEs that aRe pOlaRizEd pArAllel or anTi-parallEl To $\MaThBf{g(K)}$. The AsoC plays a CrUciAl RolE in thE DetermInatiOn of ThE sUPerConductINg STAte. THe KeY poiNt iS tHat if A spiN-TriPlet conTribution To tHE supErCoNductinG gap function iS tO emerge, its ChAraCterisTIC d-vector $\Mathbf{d(k)}$ must be parallel tO $%
\Mathbf{g(K)}$ (prOvideD thaT the ASOC iS suFficieNtlY Large) [@FRigeri; @2004; @frigeRi; @2005]. thiS LEads tO TWo Gap FuNctions $\DelTA_{\Pm}(\mAthbf{K}%
)=\pSi\pm T \mid\matHbf{g(k)}\mid$, where each gAP is Defined on one oF thE two BANdS foRMeD By tHe DEgeNERacy lifting of thE ASOC; $\ |
In the latter two cases,broken par ity a ndtim ereve rsal symmetries co m bine to destabilize the sp atial ly unif o rm stat e, givi n gr i seto t hesp a ti al de pen dence i n the gapfun ct ions. The fo r me r two stat esare unlikely in any t he ory thatisbased on lo c al int eractions ( l ike th e single b an d Hu bbard model). Sin c et he above three state sa re n ots-w ave pairin gstate s , and ( a sa r g ued below) Li$_{2 }$Pd$_{3}$B app ears t obes -wave, a ph as e tr ansition in the pairingstateo f Li$_{ 2 }$(Pd$_ {1-x}$ Pt$ _{x }$)$ _ {3 }$ B w it h va r yi ng$ x$would ha ve t o occ ur f o r a ny o f t hese stat es to exist i n L i$_{ 2 }$P t$_{3 }$B.Furt he rmore , thes e sta te s should be ext reme ly sensit ive t o i mp uriti e s and$%T_{ c}$ sho uld bes tro ng l y su ppressed when $x$is v ar ied away from1 .Th e se are i ncon tras t withthee xp erimenta lly ob s er ve d smoot hevolut io n o f $ T_{c} $ wit h $x$[@Badica ; @20 0 5]. Given thes e arguments ag a in s t u n conv ent ional super cond u ctiv ity, we at t ribut e the d r am a tic difference betw ee n thes e two compounds to the varia t i o n of ASO C.
W he n parity symmet ry is violated, the ASOC that breaksthe spind e generacy of ea chban d ta kes the form$ \ alph a\ mathbf{ g(k )\cdotS(k )}/ \hb ar$ ,where $%\alpha$de no te sthe spin - orbit co up lin gstr ength , $\mat hbf{S (k)} $is the spin o f a n elec tr on wit h m om entum $\h b ar\mathbf {k}$, and $\ m athb f{ g( k)}$ is a dimensionl es s vector ( $\ mat hbf{g( - k )=-g(k)} $ to preserve time reve r sal sym met ry).This ASOC lea dsto anene r gy spl itting of t he or i g inall y de gen er ate spin s t a tes andre sult s in sp in-eigenstates tha t ar e polarized p ara llel o rant i -p a ral le l to $ \mathbf{g(k)}$. The ASOCpl a ys a crucial rol ein thedetermi natio n of the supercon ducting s ta te.T h e k ey point i s that i f a spin- t riple t c ontri but ion to t hesuper conduc t ing gapfuncti on is to emer ge , its ch aracteristic d-vector $ \mathb f{d(k )}$ must bepar a lle l to $%
\ math bf{g(k)}$(pr ovi ded t hat the A SOCi ssuf f icien tlyl arge) [@F r ig eri ; @2 004; @Frige r i ; @2 005]. Th i s lead s to two gap function s $\Delta_{\pm} (\ma t h bf{ k}% )=\p si \pm t \mid\mat hbf {g ( k )}\mid$, w here each g ap is de fi n ed on one o f thetwo ban d s f o rmed b y th e d egeneracy li ft i ng of t he A S OC; $\ |
In the_latter two_cases, broken parity and_time reversal_symmetries_combine to_destabilize_the spatially uniform_state, giving rise_to the spatial dependence_in the gap_functions._The former two states are unlikely in any theory that is based on local_interactions_(like the_single_band_Hubbard model). Since the above_three states are not s-wave_pairing states,_and (as argued below) Li$_{2}$Pd$_{3}$B appears to be_s-wave,_a phase transition_in the pairing state of Li$_{2}$(Pd$_{1-x}$Pt$_{x}$)$_{3}$B with varying $x$_would have to occur for any_of these states_to_exist_in Li$_{2}$Pt$_{3}$B. Furthermore, these_states should be extremely sensitive to_impurities and $%
T_{c}$ should be strongly_suppressed when $x$ is varied away from_1. These are in contrast with_the experimentally observed smooth evolution_of $T_{c}$_with $x$ [@Badica; @2005]. Given_these arguments against_unconventional superconductivity,_we attribute the_dramatic difference between these two compounds_to the variation_of ASOC.
When parity symmetry is violated,_the_ASOC that breaks_the_spin_degeneracy of_each band takes_the_form $\alpha\mathbf{g(k)\cdot_S(k)}/\hbar$,_where $%
\alpha$ denotes the spin-orbit coupling_strength,_$\mathbf{S(k)}$ is the spin of an electron_with momentum $\hbar \mathbf{k}$,_and_$\mathbf{g(k)}$ is a dimensionless_vector ($\mathbf{g(-k)=-g(k)}$ to preserve time_reversal symmetry). This ASOC leads to_an energy_splitting of_the originally degenerate spin states and results in spin-eigenstates that are_polarized parallel or anti-parallel to $\mathbf{g(k)}$._The ASOC plays a_crucial role_in_the determination of_the_superconducting state._The key point is that if a_spin-triplet contribution_to the superconducting gap function is_to emerge, its characteristic_d-vector_$\mathbf{d(k)}$ must be parallel to $%
\mathbf{g(k)}$_(provided that the ASOC is sufficiently_large) [@Frigeri; @2004; @Frigeri; @2005]._This_leads_to two gap functions $\Delta_{\pm}(\mathbf{k}%
)=\psi\pm_t \mid\mathbf{g(k)}\mid$, where each gap is_defined on one_of the two bands formed by the_degeneracy_lifting of the ASOC; $\ |
okov2017escape] point 91.8 -
KCNet [@shen2018mining] point 91.0 -
SPH [@kazhdan2003rotation] mesh 68.2 33.3
MeshNet mesh 91.9 81.9
-------------------------------- -------- ------ ------
: Classification and retrieval results on ModelNet40.[]{data-label="tab:application"}
For classification, we apply fully-connected layers (512, 256, 40) to the global features as the classifier, and add dropout layers with drop probability of 0.5 before the last two fully-connected layers. For retrieval, we calculate the L2 distances between the global features as similarities and evaluate the result with mean average precision (mAP). We use the SGD optimizer for training, with initial learning rate 0.01, momentum 0.9, weight decay 0.0005 and batch size 64.
---------------- -------------- -------------- -------------- -------------- -------------- --------------
Spatial $\checkmark$ $\checkmark$ $\checkmark$ $\checkmark$ $\checkmark$
Structural-FRC $\checkmark$ $\checkmark$ $\checkmark$ $\checkmark$
Structural-FKC $\checkmark$ $\checkmark$ $\checkmark$ $\checkmark$
Mesh Conv $\checkmark$ $\checkmark$ $\checkmark$ $\checkmark$ $\checkmark$
Accuracy (%) 83.5 88.2 87.0 89.9 90.4 91.9
---------------- -------------- -------------- -------------- -------------- -------------- --------------
: Classification results of ablation experiments on ModelNet40.[]{data-label="tab:ablation"}
Aggregation Method Accuracy (%)
-------------------- --------------
Average Pooling 90.7
Max Pooling 91.1
Concatenation 91.9
: Classification results of different aggregation methods on ModelNet40.[]{data-label="tab:aggregation"}
Table \[tab:application\] shows the experimental results of classification and retrieval on ModelNet40, comparing our work with representative methods. It is shown that, as a mesh-based representation, MeshNet achieves satisfying performance and makes great improvement compared with traditional mesh-based methods. It is also comparable with | okov2017escape ] point 91.8 -
KCNet [ @shen2018mining ] point 91.0 -
SPH [ @kazhdan2003rotation ] mesh 68.2 33.3
MeshNet mesh 91.9 81.9
-------------------------------- -------- ------ ------
: categorization and recovery resultant role on ModelNet40.[]{data - label="tab: application " }
For categorization, we apply in full - connected layers (512, 256, 40) to the ball-shaped feature as the classifier, and add dropout layer with drop probability of 0.5 before the last two in full - connected layers. For recovery, we calculate the L2 distances between the global features as similarities and evaluate the result with average average precision (mAP). We use the SGD optimizer for education, with initial learning rate 0.01, momentum 0.9, system of weights decay 0.0005 and batch size 64.
---------------- -------------- -------------- -------------- -------------- -------------- --------------
Spatial $ \checkmark$ $ \checkmark$ $ \checkmark$ $ \checkmark$ $ \checkmark$
Structural - FRC $ \checkmark$ $ \checkmark$ $ \checkmark$ $ \checkmark$
Structural - FKC $ \checkmark$ $ \checkmark$ $ \checkmark$ $ \checkmark$
Mesh Conv $ \checkmark$ $ \checkmark$ $ \checkmark$ $ \checkmark$ $ \checkmark$
Accuracy (%) 83.5 88.2 87.0 89.9 90.4 91.9
---------------- -------------- -------------- -------------- -------------- -------------- --------------
: Classification consequence of ablation experiments on ModelNet40.[]{data - label="tab: ablation " }
Aggregation Method Accuracy (%)
-------------------- --------------
Average Pooling 90.7
Max Pooling 91.1
Concatenation 91.9
: Classification results of different collection methods on ModelNet40.[]{data - label="tab: aggregation " }
Table \[tab: application\ ] shows the experimental results of classification and retrieval on ModelNet40, comparing our work with representative methods. It is shown that, as a mesh - based representation, MeshNet achieves satisfying performance and makes bang-up improvement compared with traditional mesh - free-base methods. It is also comparable with | okog2017escape] point 91.8 -
KCNet [@sheu2018nining] poivt 91.0 -
SPH [@kazhdan2003rotatmon] mesh 68.2 33.3
MeshNeg mesi 91.9 81.9
-------------------------------- -------- ------ ------
: Clavwification and retrieval results on MogeuNzt40.[]{data-label="tab:application"}
For classifycation, wf apply fully-cjnnebtqd lzjevs (512, 256, 40) to the global features as fhe clavsifier, and acd dropout layers with droo prlbability of 0.5 befoge the last two dully-connectdd layers. For retrievam, we calculate the L2 distances cetwezn the globql fewjures as simmlaritpes and evaluate the sesult eith mean aversge prwcision (mAP). We use thx SGD optimizer for jraining, whtk initial learning rare 0.01, momgntum 0.9, wektht dedab 0.0005 znd bahch size 64.
---------------- -------------- -------------- -------------- -------------- -------------- --------------
Spatiql $\checkmadk$ $\creskmark$ $\checkmark$ $\checkmark$ $\checkmdrk$
Structural-FRC $\checjmark$ $\checkmark$ $\checkmark$ $\checkmark$
Structgral-FIZ $\ghecynagk$ $\checkmark$ $\cnebkmark$ $\checkmarh$
Mesh Conf $\checkmary$ $\chzdkjark$ $\checkmark$ $\checkmwrk$ $\checkmark$
Accuracy (%) 83.5 88.2 87.0 89.9 90.4 91.9
---------------- -------------- -------------- -------------- -------------- -------------- --------------
: Clcssifidation resupts of abmxtion experimentr ok MmdelNet40.[]{data-label="tab:ablatiog"}
Aggregetion Method Sccurasy (%)
-------------------- --------------
Avfrage Pooling 90.7
Max Pmoling 91.1
Concatenation 91.9
: Claxshfibation refults of different aggregation mgthods on ModeuNet40.[]{data-lagel="tab:ajgregation"}
Tafle \[tab:applicdjion\] shows thx experimqntao rewults ow classificatiom and retgitval on MoeelNet40, comparing omr wotk with representcuivt nethods. It is xhodn ehwt, af a mesh-based repfesdmtatiun, MeshNet ccfievrs satisfying perforkancs and makes great omirovement compareq with traditoonal mesh-based meuhods. Mt is elso cpmpwrable with | okov2017escape] point 91.8 - KCNet [@shen2018mining] point SPH mesh 68.2 MeshNet mesh 91.9 : and retrieval results ModelNet40.[]{data-label="tab:application"} For classification, apply fully-connected layers (512, 256, 40) the global features as the classifier, and add dropout layers with drop probability 0.5 before the last two fully-connected layers. For retrieval, we calculate the L2 between global as and evaluate the result with mean average precision (mAP). We use the SGD optimizer for training, initial learning rate 0.01, momentum 0.9, weight decay and batch size 64. -------------- -------------- -------------- -------------- -------------- Spatial $\checkmark$ $\checkmark$ $\checkmark$ $\checkmark$ $\checkmark$ $\checkmark$ Structural-FKC $\checkmark$ $\checkmark$ $\checkmark$ Mesh Conv $\checkmark$ $\checkmark$ $\checkmark$ $\checkmark$ $\checkmark$ Accuracy (%) 83.5 88.2 87.0 89.9 90.4 ---------------- -------------- -------------- -------------- : results ablation experiments on Method Accuracy (%) -------------------- -------------- Average Pooling 91.1 Concatenation 91.9 : Classification results of aggregation methods ModelNet40.[]{data-label="tab:aggregation"} Table \[tab:application\] shows the experimental of classification and retrieval on ModelNet40, comparing our with representative methods. It is shown that, as a mesh-based representation, MeshNet achieves satisfying performance great improvement compared with mesh-based methods. It also with | okov2017escape] point 91.8 -
KCNet [@shen2018mIning] point 91.0 -
sPH [@kaZhdAn2003rOtAtioN] mesH 68.2 33.3
MeshNet mesh 91.9 81.9
-------------------------------- -------- ------ ------
: ClASsifIcation and retrieval resUlts oN MODelNET40.[]{dAta-laBel="tab:aPPlICAtiOn"}
foR clAsSIfIcatiOn, wE apply fUlly-connecTed LaYers (512, 256, 40) to the gloBAl Features as The Classifier, anD adD dropoUt LayERs witH drOp proBabiliTY of 0.5 befOre the lasT tWO fully-COnnecteD LAyErs. FOr retrieval, we calcULaTE the L2 distances BetweeN tHE gLOBal FeaTures as simIlAritiES and evaLUaTE THe rESult with mean aVerage preciSIon (MAP). We uSe The sgD optiMizer FoR TraIning, with inItiaL learning Rate 0.01, moMEntum 0.9, weIGht decaY 0.0005 and baTch SizE 64.
---------------- -------------- -------------- -------------- -------------- -------------- --------------
SpaTIaL $\cHecKmARk$ $\cHEcKmaRK$ $\chEckmark$ $\cHeCkMark$ $\cHeckMARK$
struCtuRal-FrC $\cheCkmark$ $\checkmaRk$ $\cHeckMArk$ $\CheckMark$
STrucTuRal-FKc $\checkMark$ $\cHeCkmark$ $\checkmark$ $\ChecKmark$
Mesh conV $\cHecKmArk$ $\chECkmark$ $\CheCkmArk$ $\checKmark$ $\chECkmArK$
aCCuRacy (%) 83.5 88.2 87.0 89.9 90.4 91.9
---------------- -------------- -------------- -------------- -------------- -------------- --------------
: Classification ReSULtS of ablatIon expERiMeNTs on ModeLNEt40.[]{dAta-lABEl="tab:AblaTIoN"}
AggregaTion MeTHoD ACcuracy (%)
-------------------- --------------
avErage POoLinG 90.7
MaX PoolINg 91.1
CoNcatenAtion 91.9
: ClaSsifiCAtion results of DIfferent aggreGAtION mEThodS on modelNet40.[]{datA-labEL="tab:AggrEGaTioN"}
table \[Tab:apPlICaTIon\] shows the experimeNtAl resuLts of ClassificatioN and retrieVAL On ModelNEt40, coMPaRIng our work with RepreSentative mEThods. It iS showN that, as a Mesh-based REPresentaTioN, MeShNEt aCHIeVes satisfying PERforMaNce and mAkeS great iMprOveMenT coMpAred with tRaditionAl MeSh-BaSed MethoDS. It is alsO cOmpArAblE with | okov2017escape] poi nt 91.8 -
KCN et [@s hen2 018mining] point 91.0 -
SPH[@ k azhd a n2 003ro tation] m es h 6 8. 2 3 3.3
Mesh Net mesh 9 1.9 81.9
-- --- ------ -- --- - ----- --- ----- -- --- - ---- - ----- --- -- -
:C lassifi c a ti on a nd retrieval resu l ts on ModelNet40. []{dat a- l ab e l ="t ab: applicatio n" }
Fo r classi f ic a t i on, we apply full y-connected lay ers (5 12 , 2 5 6, 40) to t he glo bal feature s as the clas sifier , and ad d dropou t laye rswit h dr o ppr oba bi l ity of 0. 5 be fore the l as t two ful l y - c onne cte d la yers. For retrieva l,we c a lcu latethe L 2 di st ances betwe en th eglobal features assimilarit ies a ndev aluat e the r esu ltwith me an aver a gepr e c i si on (mAP). We use t he S GD optimiz er for tr ai n ing, wit hini tial l earni ng r a te 0.01, m omentu m 0 .9 , weigh tdecay0. 000 5 a nd ba t ch s ize 64 .
--- ----- - ------- ------ - ------- ----- - -- - - -- - - -- --- --------- - ---- - ---- ---- -- --- - ----- --- - -- - -- - -------
Spatial $\ checkmark$ $\checkm ark$ $ \checkmark$ $\che ckmark$ $ \checkma rk$
Structu ral-FRC $\checkm ark $ $\check mar k$ $\ che ckm ark $ $ \checkmar k$
Str uc tu ra l- FKC $ \ checkmar k$ $\ check mark $ $ \che ck ma rk$ $ \c heckm ark$ M esh Con v $\ c heck ma rk $ $\c heckmark$ $ \c heckmark$ $\c heckma r k $ $\checkmark$ Accur acy (%) 83.5 8 8 .2 87.0 8 9 . 9 90.4 91 .9
------ ---------- ------- - --- --- --------- --- -- - - - -- --- - -- - ---- - --- - - ------ -------- ------ --- -- - -- ------
: Cl as sificat ion res ultso f ablat ion exper iments on M odel N e t40 .[]{data-l abel="ta b:ablatio n "}
Ag grega tio n Meth od Accu racy ( % )
---- ------ -- ------ -- -- -- -------- --
Average Pooling 90. 7
Max Pool ing 91 .1
Conca ten ati on 91. 9
: Class ific a tion resu l ts of d if ferent aggr e g a tio n met hod s on Mo delN et40.[]{data-labe l ="tab:aggregat ion" }
Ta ble \[ta b: application\]sho ws t he exper im ental resul ts of cl as s ifica tion a nd ret rievalo n M o delNet 40,com paring ou r w or k with r ep re s entati ve m et hods.It iss hown t hat, as a mesh-b asedr e prese n tat ion,Me shNet a c hiev es satisfy ing perform ance a nd m akesgreat i mp roveme ntco mpared wit h traditio nal m esh-bas ed met hod s. Itis a l s o com para bl e w ith | okov2017escape] _ _ _point __ 91.8__ -
_ KCNet [@shen2018mining]_ _ __ point 91.0 -
SPH [@kazhdan2003rotation]__ ___ mesh _ 68.2 _33.3
_MeshNet __ _ _ _ mesh ___91.9 81.9
_ -------------------------------- -------- ------ ------
_: Classification and retrieval results on_ModelNet40.[]{data-label="tab:application"}
For classification, we apply fully-connected layers (512,_256, 40) to the global features_as the classifier, and add_dropout layers_with drop probability of 0.5_before the last_two fully-connected_layers. For retrieval,_we calculate the L2 distances between_the global features_as similarities and evaluate the result_with_mean average precision_(mAP)._We_use the_SGD optimizer for_training,_with initial_learning_rate 0.01, momentum 0.9, weight decay_0.0005_and batch size 64.
---------------- --------------_-------------- -------------- -------------- --------------_--------------
_ Spatial _ _ _ _ _ $\checkmark$ _$\checkmark$ $\checkmark$ _$\checkmark$ $\checkmark$
_ Structural-FRC__ $\checkmark$__ _ _ _ _ __ _ $\checkmark$ _$\checkmark$ $\checkmark$
_Structural-FKC__ $\checkmark$ _ _ _ $\checkmark$__ __ ___ $\checkmark$ $\checkmark$
_ Mesh Conv _ $\checkmark$ $\checkmark$_ $\checkmark$_ $\checkmark$ ___ _ _ $\checkmark$
_Accuracy (%) _ 83.5 _ 88.2 _ 87.0 _ _ 89.9 __ _90.4___ _ _ 91.9
_ ----------------_-------------- -------------- --------------_-------------- --------------_--------------
:_Classification_results of ablation experiments on ModelNet40.[]{data-label="tab:ablation"}
Aggregation Method Accuracy (%)
-------------------- --------------
_Average Pooling _ __90.7
Max_Pooling _ _ 91.1
Concatenation __ _ _ _91.9
_: Classification results_of different_aggregation methods_on ModelNet40.[]{data-label="tab:aggregation"}
Table_\[tab:application\] shows the experimental results of classification_and_retrieval on ModelNet40, comparing our work with representative methods._It_is_shown_that, as a_mesh-based representation, MeshNet_achieves satisfying performance_and makes great_improvement_compared with traditional mesh-based methods. It_is also comparable_with |
k)^{\top}({{\bf x}}_{k+1}-{{\bf x}}_k) = -[({{\bf g}}({{\bf x}}_k)^{\top}{\bf a}){\bf b}^{\top}
- ({{\bf g}}({{\bf x}}_k)^{\top}{\bf b}){\bf a}^{\top}]({{\bf x}}_k+{{\bf x}}_{k+1}).$$ For convenience, we choose $$\label{eq3-10}
{\bf a}={{\bf x}}_k~~~~\text{and}~~~~{\bf b}=-\alpha{{\bf g}}({{\bf x}}_k).$$ Here, $\alpha$ is a positive parameter, which serves as a step size, so that we have some freedom to choose the next iterate. According to this selection and (\[eq3-06\]), we obtain $$\begin{aligned}
{{\bf g}}({{\bf x}}_k)^{\top}({{\bf x}}_{k+1}-{{\bf x}}_k) &=& -\alpha\|{{\bf g}}({{\bf x}}_k)\|^2 {{\bf x}}_k^{\top}({{\bf x}}_k+{{\bf x}}_{k+1}) \\
&=& -\alpha\|{{\bf g}}({{\bf x}}_k)\|^2 (1+{{\bf x}}_k^{\top}Q{{\bf x}}_k).\end{aligned}$$ Since $-1$ is not an eigenvalue of the orthogonal matrix $Q$, we have $1+{{\bf x}}_k^{\top}Q{{\bf x}}_k > 0$ for ${{\bf x}}_k^{\top}{{\bf x}}_k=1$. Therefore, the conclusion ${{\bf g}}({{\bf x}}_k)^{\top}({{\bf x}}_{k+1}-{{\bf x}}_k)<0$ holds for any positive step size $\alpha$.
We summarize the iterative process in the following Theorem.
Suppose that the new iterate ${{\bf x}}_{k+1}$ is generated by (\[eq3-07\]), (\[eq3-08\]), (\[eq3-09\]), and (\[eq3-10\]). Then, the following assertions hold.
- The iterative scheme is $$\label{eq3-11}
{{\bf x}}_{k+1}(\alpha) = \frac{1-\alpha^2\|{{\bf g}}({{\bf x}}_k)\|^2}{1+\alpha^2\|{{\bf g}}({{\bf x}}_k)\|^2}{{\bf x}}_k
-\frac{2\alpha}{1+\alpha^ | k)^{\top}({{\bf x}}_{k+1}-{{\bf x}}_k) = -[({{\bf g}}({{\bf x}}_k)^{\top}{\bf a}){\bf b}^{\top }
- ({ { \bf g}}({{\bf x}}_k)^{\top}{\bf b}){\bf a}^{\top}]({{\bf x}}_k+{{\bf x}}_{k+1}).$$ For convenience, we choose $ $ \label{eq3 - 10 }
{ \bf a}={{\bf x}}_k~~~~\text{and}~~~~{\bf b}=-\alpha{{\bf g}}({{\bf x}}_k).$$ Here, $ \alpha$ is a positive parameter, which serves as a gradation size, so that we take some freedom to choose the next iterate. According to this choice and (\[eq3 - 06\ ]), we obtain $ $ \begin{aligned }
{ { \bf g}}({{\bf x}}_k)^{\top}({{\bf x}}_{k+1}-{{\bf x}}_k) & = & -\alpha\|{{\bf g}}({{\bf x}}_k)\|^2 { { \bf x}}_k^{\top}({{\bf x}}_k+{{\bf x}}_{k+1 }) \\
& = & -\alpha\|{{\bf g}}({{\bf x}}_k)\|^2 (1+{{\bf x}}_k^{\top}Q{{\bf x}}_k).\end{aligned}$$ Since $ -1 $ is not an eigenvalue of the orthogonal matrix $ Q$, we hold $ 1+{{\bf x}}_k^{\top}Q{{\bf x}}_k > 0 $ for $ { { \bf x}}_k^{\top}{{\bf x}}_k=1$. Therefore, the conclusion $ { { \bf g}}({{\bf x}}_k)^{\top}({{\bf x}}_{k+1}-{{\bf x}}_k)<0 $ holds for any convinced step size $ \alpha$.
We summarize the iterative summons in the following Theorem.
Suppose that the new iterate $ { { \bf x}}_{k+1}$ is render by (\[eq3 - 07\ ]), (\[eq3 - 08\ ]), (\[eq3 - 09\ ]), and (\[eq3 - 10\ ]). Then, the follow assertions hold.
- The iterative scheme is $ $ \label{eq3 - 11 }
{ { \bf x}}_{k+1}(\alpha) = \frac{1-\alpha^2\|{{\bf g}}({{\bf x}}_k)\|^2}{1+\alpha^2\|{{\bf g}}({{\bf x}}_k)\|^2}{{\bf x}}_k
-\frac{2\alpha}{1+\alpha^ | k)^{\too}({{\bf x}}_{k+1}-{{\bf x}}_k) = -[({{\bf g}}({{\bf x}}_k)^{\tok}{\bf a}){\bf b}^{\top}
- ({{\yd g}}({{\bf e}}_k)^{\top}{\bf b}){\bf a}^{\too}]({{\bf x}}_k+{{\bf x}}_{k+1}).$$ For convenience, xe cyoose $$\label{eq3-10}
{\bf a}={{\bf x}}_k~~~~\tdxt{and}~~~~{\bf h}=-\alpha{{\bf g}}({{\bf z}}_k).$$ Here, $\al'ga$ is a positjye paxaneter, which setves as a stap size, so thad de have some freedom to choose the nevt iterstf. According to thix selsbtlon and (\[eq3-06\]), we obtain $$\begin{alignes}
{{\bf g}}({{\bf x}}_k)^{\top}({{\bf c}}_{k+1}-{{\bf x}}_k) &=& -\alpha\|{{\bf g}}({{\bf x}}_k)\|^2 {{\bf x}}_k^{\tlp}({{\bf x}}_k+{{\bf x}}_{k+1}) \\
&=& -\alpha\|{{\bf g}}({{\vf x}}_h)\|^2 (1+{{\bf x}}_k^{\top}Q{{\bf x}}_k).\end{aligned}$$ Since $-1$ ia not an eigenvalue of the orthugonak matrix $Q$, ww hwee $1+{{\bf x}}_k^{\top}W{{\bf x}}_h > 0$ for ${{\bf x}}_k^{\top}{{\bf x}}_n=1$. Theregore, the conclmsion ${{\bf g}}({{\bf x}}_k)^{\top}({{\bf x}}_{k+1}-{{\bf x}}_k)<0$ iolds for any positide step shzz $\alpha$.
We summarize tye iterdtiva pruxesr ih vhe folloainj Theorem.
Sulpose that rhe new iterate ${{\bf c}}_{k+1}$ px generated gy (\[eq3-07\]), (\[qq3-08\]), (\[eq3-09\]), and (\[eq3-10\]). Then, the following assertionv hkld.
- The iterative schwme is $$\label{eq3-11}
{{\bf x}}_{h+1}(\alpha) = \frac{1-\alpha^2\|{{\bf g}}({{\bf x}}_k)\|^2}{1+\alpha^2\|{{\bf g}}({{\bf x}}_k)\|^2}{{\bf x}}_k
-\fvac{2\auphw}{1+\alpha^ | k)^{\top}({{\bf x}}_{k+1}-{{\bf x}}_k) = -[({{\bf g}}({{\bf x}}_k)^{\top}{\bf - g}}({{\bf x}}_k)^{\top}{\bf a}^{\top}]({{\bf x}}_k+{{\bf x}}_{k+1}).$$ {\bf x}}_k~~~~\text{and}~~~~{\bf b}=-\alpha{{\bf g}}({{\bf Here, $\alpha$ is positive parameter, which serves as a size, so that we have some freedom to choose the next iterate. According this selection and (\[eq3-06\]), we obtain $$\begin{aligned} {{\bf g}}({{\bf x}}_k)^{\top}({{\bf x}}_{k+1}-{{\bf x}}_k) &=& g}}({{\bf {{\bf x}}_k+{{\bf \\ &=& -\alpha\|{{\bf g}}({{\bf x}}_k)\|^2 (1+{{\bf x}}_k^{\top}Q{{\bf x}}_k).\end{aligned}$$ Since $-1$ is not an eigenvalue of the matrix $Q$, we have $1+{{\bf x}}_k^{\top}Q{{\bf x}}_k > for ${{\bf x}}_k^{\top}{{\bf x}}_k=1$. the conclusion ${{\bf g}}({{\bf x}}_k)^{\top}({{\bf x}}_k)<0$ for any step $\alpha$. summarize the iterative in the following Theorem. Suppose that the new iterate ${{\bf x}}_{k+1}$ is generated by (\[eq3-07\]), (\[eq3-08\]), (\[eq3-09\]), (\[eq3-10\]). Then, assertions hold. The scheme $$\label{eq3-11} {{\bf x}}_{k+1}(\alpha) g}}({{\bf x}}_k)\|^2}{1+\alpha^2\|{{\bf g}}({{\bf x}}_k)\|^2}{{\bf x}}_k -\frac{2\alpha}{1+\alpha^ | k)^{\top}({{\bf x}}_{k+1}-{{\bf x}}_k) = -[({{\bf g}}({{\bf x}}_k)^{\top}{\bf a}){\Bf b}^{\top}
- ({{\bf g}}({{\bF x}}_k)^{\toP}{\bf B}){\bf A}^{\tOp}]({{\bf X}}_k+{{\bf X}}_{k+1}).$$ For convenienCE, we cHoose $$\label{eq3-10}
{\bf a}={{\bf x}}_k~~~~\texT{and}~~~~{\bF b}=-\ALpha{{\BF g}}({{\Bf x}}_k).$$ HEre, $\alphA$ Is A POsiTiVe ParAmETeR, whicH seRves as a Step size, so ThaT wE have some freEDoM to choose tHe nExt iterate. AcCorDing to ThIs sELectiOn aNd (\[eq3-06\]), wE obtaiN $$\Begin{aLigned}
{{\bf g}}({{\Bf X}}_K)^{\top}({{\bf X}}_{K+1}-{{\bf x}}_k) &=& -\alPHA\|{{\bF g}}({{\bf X}}_k)\|^2 {{\bf x}}_k^{\top}({{\bf x}}_k+{{\bf x}}_{k+1}) \\
&=& -\ALpHA\|{{\bf g}}({{\bf x}}_k)\|^2 (1+{{\bf x}}_k^{\toP}Q{{\bf x}}_k).\EnD{AlIGNed}$$ sinCe $-1$ is not an eIgEnvalUE of the oRThOGONal MAtrix $Q$, we have $1+{{\bF x}}_k^{\top}Q{{\bf x}}_k > 0$ FOr ${{\bF x}}_k^{\top}{{\Bf X}}_k=1$. THEreforE, the cOnCLusIon ${{\bf g}}({{\bf x}}_k)^{\tOp}({{\bf X}}_{k+1}-{{\bf x}}_k)<0$ holDs for aNY positiVE step siZe $\alphA$.
We SumMariZE tHe IteRaTIve PRoCesS In tHe followInG THeoreM.
SupPOSE That The New iTeratE ${{\bf x}}_{k+1}$ is generaTed By (\[eq3-07\]), (\[EQ3-08\]), (\[eq3-09\]), And (\[eq3-10\]). then, tHe foLlOwing AssertIons hOlD.
- The iterative scHeme Is $$\label{eq3-11}
{{\Bf x}}_{K+1}(\aLphA) = \fRac{1-\alPHa^2\|{{\bf g}}({{\bF x}}_k)\|^2}{1+\AlpHa^2\|{{\bf g}}({{\bf X}}_k)\|^2}{{\bf x}}_k
-\fRAc{2\aLpHA}{1+\ALpHa^ | k)^{\top}({{\bf x}}_{k+1}- {{\bf x}}_ k) =-[( {{\ bf g}} ({{\ bf x}}_k)^{\to p }{\b f a}){\bf b}^{\top}
- ( {{ \ bf g } }( {{\bf x}}_k) ^ {\ t o p}{ \b fb}) {\ b fa}^{\ top }]({{\b f x}}_k+{{ \bf x }}_{k+1}).$$ Fo r convenie nce , we choose$$\ label{ eq 3-1 0 }
{\ bf a} ={{\bf x}}_k~ ~~~\text{ an d }~~~~{ \ bf b}=- \ a lp ha{{ \bf g}}({{\bf x}} _ k) . $$ Here, $\alp ha$ is a po s i tiv e p arameter,wh ich s e rves as as t e p s i ze, so that w e have some fre edom t ocho o se the next i t era te. Accordi ng t o this se lectio n and (\ [ eq3-06\ ]), we ob tai n $$ \ be gi n{a li g ned }
{ {\b f g}}({{ \b fx}}_k )^{\ t o p } ({{\ bfx}}_ {k+1} -{{\bf x}}_k) &= & -\ a lph a\|{{ \bf g }}({ {\ bf x} }_k)\| ^2 {{ \b f x}}_k^{\top}( {{\b f x}}_k+{ {\b fx}} _{ k+1}) \\
&=& -\alph a\|{{\b f g} }( { { \ bf x}}_k)\|^2 (1+{{\ bf x }} _k^{\top }Q{{\b f x }} _ k).\end{ al ign ed}$ $ Since $-1 $ i s not an eigen v al ue of the o rthogo na l m atr ix $Q $ , we have$1+{{\bf x}}_ k ^{\top}Q{{\bfx }}_k > 0$ for ${ { \ bf x}}_ k^{ \top}{{\bfx}}_ k =1$. The r ef ore , theconcl us i on ${{\bf g}}({{\bf x} }_ k)^{\t op}({ {\bf x}}_{k+1 }-{{\bf x} } _ k )<0$ hol ds f o ra ny positive st ep si ze $\alpha $ .
We su mmari ze the i terativep r ocess in th e f oll owi n g T heorem.
Supp o s e th at the ne w i terate${{ \bf x} }_{ k+ 1}$ is ge neratedby ( \[ eq 3-0 7\]), (\[eq3-0 8\ ]), ( \[e q3-09 \ ]), an d (\[ eq3- 10 \] ) . T hen, th e f o l lowi ng a sser tio ns hold .
- T he iter ative sch eme is $ $\ la bel{eq3 -11}
{{\bf x} }_ {k+ 1}(\al p h a) = \fr ac{1-\alpha^2\|{{\bf g} } ({{\bfx}} _k)\| ^2}{ 1+\alpha^ 2\| {{\bfg}} ( {{\bfx}}_k) \|^2} {{ \bf x }}_k -\fra c { 2\a lpha} {1 +\al pha^ | k)^{\top}({{\bf x}}_{k+1}-{{\bf_x}}_k) =_-[({{\bf g}}({{\bf x}}_k)^{\top}{\bf a}){\bf_b}^{\top}
__ -_({{\bf_g}}({{\bf x}}_k)^{\top}{\bf b}){\bf_a}^{\top}]({{\bf x}}_k+{{\bf x}}_{k+1}).$$_For convenience, we choose_$$\label{eq3-10}
__{\bf a}={{\bf x}}_k~~~~\text{and}~~~~{\bf b}=-\alpha{{\bf g}}({{\bf x}}_k).$$ Here, $\alpha$ is a positive parameter, which serves_as_a step_size,_so_that we have some freedom_to choose the next iterate._According to_this selection and (\[eq3-06\]), we obtain $$\begin{aligned}
__ {{\bf g}}({{\bf_x}}_k)^{\top}({{\bf x}}_{k+1}-{{\bf x}}_k) &=& -\alpha\|{{\bf g}}({{\bf x}}_k)\|^2 {{\bf x}}_k^{\top}({{\bf_x}}_k+{{\bf x}}_{k+1}) \\
_ &=&_-\alpha\|{{\bf_g}}({{\bf_x}}_k)\|^2 (1+{{\bf x}}_k^{\top}Q{{\bf x}}_k).\end{aligned}$$_Since $-1$ is not an eigenvalue_of the orthogonal matrix $Q$, we_have $1+{{\bf x}}_k^{\top}Q{{\bf x}}_k > 0$ for_${{\bf x}}_k^{\top}{{\bf x}}_k=1$. Therefore, the conclusion_${{\bf g}}({{\bf x}}_k)^{\top}({{\bf x}}_{k+1}-{{\bf x}}_k)<0$_holds for_any positive step size $\alpha$.
We_summarize the iterative_process in_the following Theorem.
Suppose_that the new iterate ${{\bf x}}_{k+1}$_is generated by_(\[eq3-07\]), (\[eq3-08\]), (\[eq3-09\]), and (\[eq3-10\]). Then,_the_following assertions hold.
-___The iterative_scheme is $$\label{eq3-11}
__ __ __{{\bf x}}_{k+1}(\alpha) = \frac{1-\alpha^2\|{{\bf g}}({{\bf x}}_k)\|^2}{1+\alpha^2\|{{\bf g}}({{\bf_x}}_k)\|^2}{{\bf x}}_k
__ _ _ -\frac{2\alpha}{1+\alpha^ |
}^\infty q^n f(aq^n).$$ In this section we find a solution $p_k(x)$ of of the form $$\label{pk-def}
p_k(x) = \int_{t_1}^{t_2} t^k f(t) d_q t,$$ following a technique developed by Ismail and Stanton in [@IS1997; @IS1998; @IS2002]. We will use the integration by parts formula $$\label{qbyparts}
\int_a^b f(t) g(qt) d_q t = \frac{1}{q} \int_a^b g(t) f(t/q) d_q t
+\frac{1-q}{q} \big( ag(a)f(a/q) - bg(b)f(b/q)\big).$$ This formula follows from the definition of the $q$-integral.
We will require the notation of [*basic hypergeometric series*]{} (or $_r\phi_s$ series). This series is of the form $$_{r}\phi_s \left[\begin{matrix}
a_1,a_2,\dots,a_r \\
b_1,b_2,\dots,b_s\end{matrix} ; q, z
\right] :=
\sum_{k=0}^{\infty} \frac{{{\left({a_1,a_2,\dots, a_r}; q\right)_{k}}}}{{{\left({q, b_1,b_2,\dots, b_s}; q\right)_{k}}}}
\left( (-1)^kq^{\binom k2}\right)^{1+s-r} z^k.$$ When $r=s+1$, the series converges for $|z|<1$. See Gasper and Rahman [@GR90] for further convergence conditions for these series.
Let $|\lambda q/b|<1$. With $x=\cos\vartheta$, we define $p_k(x)$ as the $q$-integral $$\begin{gathered}
p_k(x) \label{final-pk-qint}
:=
\frac
{4 (-i\sin\vartheta)}
{(1-q)}
\frac
{{{\left({ 2c e^{i\vartheta}, 2c e^{-i\vartheta | } ^\infty q^n f(aq^n).$$ In this section we find a solution $ p_k(x)$ of of the form $ $ \label{pk - def }
p_k(x) = \int_{t_1}^{t_2 } t^k f(t) d_q t,$$ following a proficiency modernize by Ismail and Stanton in [ @IS1997; @IS1998; @IS2002 ]. We will use the integration by part formula $ $ \label{qbyparts }
\int_a^b f(t) g(qt) d_q deoxythymidine monophosphate = \frac{1}{q } \int_a^b g(t) f(t / q) d_q t
+ \frac{1 - q}{q } \big (ag(a)f(a / q) - bg(b)f(b / q)\big).$$ This formula follows from the definition of the $ q$-integral.
We will want the note of [ * basic hypergeometric series * ] { } (or $ _ r\phi_s$ series). This series is of the form $ $ _ { r}\phi_s \left[\begin{matrix }
a_1,a_2,\dots, a_r \\
b_1,b_2,\dots, b_s\end{matrix }; q, z
\right ]: =
\sum_{k=0}^{\infty } \frac{{{\left({a_1,a_2,\dots, a_r }; q\right)_{k}}}}{{{\left({q, b_1,b_2,\dots, b_s }; q\right)_{k } } } }
\left (( -1)^kq^{\binom k2}\right)^{1+s - r } z^k.$$ When $ radius = s+1 $, the series converges for $ |z|<1$. determine Gasper and Rahman [ @GR90 ] for further convergence conditions for these series.
Let $ |\lambda q / b|<1$. With $ x=\cos\vartheta$, we define $ p_k(x)$ as the $ q$-integral $ $ \begin{gathered }
p_k(x) \label{final - pk - qint }
: =
\frac
{ 4 (-i\sin\vartheta) }
{ (1 - q) }
\frac
{ { { \left ({ 2c e^{i\vartheta }, 2c e^{-i\vartheta | }^\infhy q^n f(aq^n).$$ In this sectiun we find a solution $'_k(x)$ of kf the furm $$\label{pk-def}
p_k(x) = \int_{t_1}^{t_2} t^k h(t) d_w t,$$ fillowing a technique ddveloped hy Ismaio anv Stanton in [@IS1997; @MA1998; @IS2002]. We will hde tke integration bi parts formgla $$\label{qbypastr}
\iut_a^b f(t) g(qt) d_q t = \frac{1}{q} \int_a^b g(t) f(t/q) d_q t
+\frsc{1-e}{q} \big( ag(a)f(a/q) - bg(b)g(f/q)\bif).$$ This formula follows from the derinitioi of the $q$-integtal.
We will require the notwtioj of [*basic hypergelmetric serues*]{} (je $_r\phi_s$ serids). This segnes is of tge form $$_{r}\phi_s \left[\begin{matrix}
a_1,x_2,\dots,c_r \\
b_1,b_2,\dots,b_s\gue{mahtix} ; q, z
\righv] :=
\sum_{k=0}^{\pnfty} \frac{{{\lefb({s_1,a_2,\dots, a_r}; q\ribht)_{k}}}}{{{\left({q, b_1,b_2,\dobs, b_s}; q\rught)_{k}}}}
\left( (-1)^kq^{\binom k2}\rijht)^{1+s-r} z^k.$$ When $r=s+1$, the series cmnrerges for $|z|<1$. See Gaspwr and Tahmat [@GR90] wir wuruhec cknvergfncx conditiona for these series.
Let $|\lambda q/n|<1$. Rpyh $x=\cos\varthsta$, we dqfine $p_k(x)$ as the $q$-integral $$\begin{gathereg}
p_k(s) \label{final-pk-qint}
:=
\frax
{4 (-i\sin\vartheta)}
{(1-q)}
\frac
{{{\peft({ 2c e^{i\dartheta}, 2c e^{-i\vartheta | }^\infty q^n f(aq^n).$$ In this section we solution of of form $$\label{pk-def} p_k(x) t,$$ a technique developed Ismail and Stanton [@IS1997; @IS1998; @IS2002]. We will use integration by parts formula $$\label{qbyparts} \int_a^b f(t) g(qt) d_q t = \frac{1}{q} \int_a^b f(t/q) d_q t +\frac{1-q}{q} \big( ag(a)f(a/q) - bg(b)f(b/q)\big).$$ This formula follows from the of $q$-integral. will the notation of [*basic hypergeometric series*]{} (or $_r\phi_s$ series). This series is of the form $$_{r}\phi_s a_1,a_2,\dots,a_r \\ b_1,b_2,\dots,b_s\end{matrix} ; q, z \right] := \frac{{{\left({a_1,a_2,\dots, a_r}; q\right)_{k}}}}{{{\left({q, b_1,b_2,\dots, q\right)_{k}}}} \left( (-1)^kq^{\binom k2}\right)^{1+s-r} z^k.$$ $r=s+1$, series converges $|z|<1$. Gasper Rahman [@GR90] for convergence conditions for these series. Let $|\lambda q/b|<1$. With $x=\cos\vartheta$, we define $p_k(x)$ as the $q$-integral $$\begin{gathered} \label{final-pk-qint} := (-i\sin\vartheta)} {(1-q)} {{{\left({ e^{i\vartheta}, e^{-i\vartheta | }^\infty q^n f(aq^n).$$ In this section wE find a soluTion $p_K(x)$ oF of ThE forM $$\labEl{pk-def}
p_k(x) = \int_{t_1}^{T_2} T^k f(t) D_q t,$$ following a technique DevelOpED by ISMaIl and stanton IN [@Is1997; @is1998; @IS2002]. we WiLl uSe THe IntegRatIon by paRts formula $$\LabEl{Qbyparts}
\int_a^B F(t) G(qt) d_q t = \frac{1}{Q} \inT_a^b g(t) f(t/q) d_q t
+\fRac{1-Q}{q} \big( aG(a)F(a/q) - BG(b)f(b/q)\Big).$$ this fOrmula FOllows From the deFiNItion oF The $q$-intEGRaL.
We wIll require the notaTIoN Of [*basic hypergeOmetriC sERiES*]{} (Or $_r\Phi_S$ series). ThiS sEries IS of the fORm $$_{R}\PHI_s \lEFt[\begin{matrix}
A_1,a_2,\dots,a_r \\
b_1,b_2,\dOTs,b_S\end{maTrIx} ; q, Z
\Right] :=
\sUm_{k=0}^{\inFtY} \FraC{{{\left({a_1,a_2,\dots, A_r}; q\rIght)_{k}}}}{{{\left({Q, b_1,b_2,\dotS, B_s}; q\righT)_{K}}}}
\left( (-1)^kq^{\Binom k2}\RigHt)^{1+s-R} z^k.$$ WHEn $R=s+1$, The SeRIes COnVerGEs fOr $|z|<1$. See GaSpEr And RaHman [@gr90] FOR furTheR conVergeNce conditions For ThesE SerIes.
LeT $|\lambDa q/b|<1$. wiTh $x=\coS\varthEta$, we DeFine $p_k(x)$ as the $q$-inTegrAl $$\begin{gaTheReD}
p_k(X) \lAbel{fINal-pk-qInt}
:=
\FraC
{4 (-i\sin\vaRtheta)}
{(1-q)}
\FRac
{{{\LeFT({ 2C E^{i\Vartheta}, 2c e^{-i\varthetA | }^\infty q^n f(aq^n).$$ In this sect ion w e f ind a sol utio n $p_k(x)$ ofo f th e form $$\label{pk-def }
p_k (x ) = \ i nt _{t_1 }^{t_2} t^ k f(t )d_ q t ,$ $ f ollow ing a tech nique deve lop ed by Ismail a n dStanton in [@ IS1997; @IS1 998 ; @IS2 00 2]. We wi lluse t he int e gratio n by part sf ormula $$\labe l { qb ypar ts}
\int_a^b f(t) g( q t) d_q t = \fr ac{1}{ q} \i n t _a^ b g (t) f(t/q) d _q t+ \frac{1 - q} { q } \b i g( ag(a)f(a/q ) - bg(b)f( b /q) \big). $$ Th i s form ula f ol l ows from the d efin ition ofthe $q $ -integr a l.
Wewill r equ ire the no ta tio no f [ * ba sic hyp ergeomet ri cserie s*]{ } ( o r $_ r\p hi_s $ ser ies). This se rie s is ofthe f orm $ $_{r }\ phi_s \left [\beg in {matrix}
a_1,a _2,\ dots,a_r\\b_ 1,b _2 ,\dot s ,b_s\e nd{ mat rix} ;q, z
\r i ght ]: = \s um_{k=0}^{\infty}\f r a c{ {{\left( {a_1,a _ 2, \d o ts, a_r} ;q\r ight ) _ {k}}} }{{{ \ le ft({q, b _1,b_2 , \d ot s, b_s} ;q\righ t) _{k }}} }
\le f t( ( -1)^kq ^{\binom k2}\ r ight)^{1+s-r}z ^k.$$ When $r = s+ 1 $ ,t he s eri es converge s fo r $|z |<1$ . S eeG asper andRa h ma n [@GR90] for furthe rconver gence conditions f or these s e r i es.
Let $|\ l am b da q/b|<1$. Wi th $x =\cos\vart h eta$, we defi ne $p_k( x)$ as th e $q$-inte gra l $ $\b egi n { ga thered}
p_k(x ) \la be l{final -pk -qint}:= \f rac
{ 4(-i\sin\v artheta) } { (1 -q )}\frac {{{\left ({ 2c e ^{i \vart h eta},2c e^ {-i\ va rt h eta | }^\infty q^n_f(aq^n).$$ In_this section we find_a solution_$p_k(x)$_of of_the_form $$\label{pk-def}
p_k(x) =_\int_{t_1}^{t_2} t^k f(t)_d_q t,$$ following a_technique developed by_Ismail_and Stanton in [@IS1997; @IS1998; @IS2002]. We will use the integration by parts formula_$$\label{qbyparts}
\int_a^b_f(t) g(qt)_d_q_t_= \frac{1}{q} \int_a^b g(t) f(t/q)_d_q t
+\frac{1-q}{q} \big( ag(a)f(a/q) -_bg(b)f(b/q)\big).$$ This_formula follows from the definition of the $q$-integral.
We_will_require the notation_of [*basic hypergeometric series*]{} (or $_r\phi_s$ series). This series_is of the form $$_{r}\phi_s \left[\begin{matrix}_
a_1,a_2,\dots,a_r \\
b_1,b_2,\dots,b_s\end{matrix} ;_q,_z
\right]_:=
\sum_{k=0}^{\infty} \frac{{{\left({a_1,a_2,\dots, a_r}; q\right)_{k}}}}{{{\left({q,_b_1,b_2,\dots, b_s}; q\right)_{k}}}}
\left( (-1)^kq^{\binom k2}\right)^{1+s-r} z^k.$$_When $r=s+1$, the series converges for_$|z|<1$. See Gasper and Rahman [@GR90] for further_convergence conditions for these series.
Let $|\lambda_q/b|<1$. With $x=\cos\vartheta$, we define_$p_k(x)$ as_the $q$-integral $$\begin{gathered}
p_k(x) \label{final-pk-qint}
:=
_\frac
{4 (-i\sin\vartheta)}
_{(1-q)}
\frac
{{{\left({ 2c_e^{i\vartheta}, 2c e^{-i\vartheta |
Subsets and Splits