id
stringlengths 24
24
| idx
int64 0
402
| paragraph
stringlengths 106
17.2k
|
|---|---|---|
60c73e98ee301c1988c78755 | 82 | Figure : Acceptance probability for 3-iodotoluene as a function of the amount of NCMC relaxation. The acceptance rate shown here is analogous to that in Figure except that this test uses a fixed set of MD snapshots as a basis for move proposals, as described in the text. Overall acceptance (black line) increases dramatically up to 10000 NCMC switching steps per cycle, then increases more slowly. The black dashed line marks the acceptance probability of instantaneous MC rotations, given the same set of MD snapshots as starting points. The solid blue line denotes the acceptance probability of substantial rotations, those larger than 45 degrees, and the dashed blue line indicates the overall acceptance probability of instantaneous MC rotations from the same set of snapshots. Thus, NCMC does only modestly worse at sampling substantial rearrangements than sampling all rearrangements, whereas MC has orders of magnitude lower acceptance of substantial rearrangements. |
673de5145a82cea2fae6cf76 | 0 | With over half of global energy wasted as heat, it is critical from both an environmental and economic standpoint to make use of this underutilised energy waste. Thermoelectric (TE) power directly converts heat to electricity through the Seebeck effect in a TE material, with potential applications to energy harvesting and for improving energy efficiency at multiple scales. The performance of a TE material is expressed by the figure of merit ZT : ZT = S 2 σ T |
673de5145a82cea2fae6cf76 | 1 | where S is the Seekbeck coefficient, σ is the electrical conductivity, S 2 σ is the power factor (PF), and κ el and κ latt are the electronic and lattice (phonon) components of the thermal conductivity κ. S, σ and κ el are related through the carrier concentration n, which can be controlled by chemical doping and is typically optimised in heavily-doped semiconductors. κ latt is independent of the electronic transport and must be minimised to achieve a high ZT . Typical materials engineering strategies to optimise ZT include doping and alloying, which can both improve the electrical properties (e.g. by increasing n or modulating the electronic structure to induce "band convergence") and/or reduce the κ latt . A variety of materials have been studied as prospective TEs including chalcogenides, skutterudites, half-Heuslers, zintl compounds, clathrates, oxides, and oxychalcogenides (Ref. and references therein). The current industry-standard materials for low-and high-temperature applications are based on Bi 2 Te 3 (ZT ≃ 1 from 350-450 K, ZT up to 1.9 reported for nanostructured (Bi 1-x Sb x )Te 3 alloys ) and PbTe (ZT up to 2.2 at 915 K with endotaxial nanostructuring ). However, the low abundance of Te precludes widespread adoption, and thus alternatives composed of more abundant elements are critical. The discovery of high ZT in orthorhombic SnSe (up to 2.6 in single crystals and 3.1 in polycrystalline samples ) has led to considerable interest in the group IV-VI sulphides and selenides. The favourable electrical properties of these materials are well established from the use of SnS as a photovoltaic (PV) absorber material, and SnS and SnSe also show unusually low κ latt associated with the high-temperature Pnma → Cmcm phase transition. However, the "headline" ZT in orthorhombic SnSe is only achieved at high temperature, whereas many potential applications of TEs require the recovery of low-grade waste heat, and so SnSe cannot replace Bi 2 Te 3 . |
673de5145a82cea2fae6cf76 | 2 | While the thermoelectric performance of orthorhombic SnS and SnSe is well characterised, the IV-VI chalcogenides display a rich structural chemistry, with several potential phases that could show superior low-temperature thermoelectric performance (Figure ). Among these, the recently-discovered metastable cubic "π" phase of SnS and SnSe has shown considerable potential for photovoltaic applications. A recent study of the IV-VI chalcogenides also revealed new structural-property relationships that suggest the π structure should lend itself to low ) and rocksalt (Fm 3m) phases. These images were prepared using VESTA. κ latt . The potential thermoelectric performance of the π phases has been explored in several computational studies, but significant approximations in the calculations mean these results are at best indicative. |
673de5145a82cea2fae6cf76 | 3 | We have developed a fully ab inito workflow for predicting ZT , by combining state-of-the-art approaches to computing the S, σ , κ el and κ latt , which we and others have validated against multiple classes of TE materials including chalcogenides, oxides and oxychalcogenides. In this study, we apply this workflow to obtain high-quality reference predictions of the thermoelectric performance of π SnS and SnSe. π SnS and SnSe have lower κ latt and higher (absolute) S than the corresponding orthorhombic phases, but at the cost of higher carrier effective masses, stronger electron scattering, and lower σ that requires heavier doping to offset. The majority of the κ latt occurs through intraband tunnelling characteristic of amorphous materials, leading to very low thermal conductivity at and above room temperature. With n-type doping at the level achieved in experiments on p-type SnS and SnSe, π SnSe has a lowtemperature ZT comparable to Bi 2 Te 3 and a high-T ZT competitive with orthorhombic SnSe, resulting in a large average ZT for low-to mid-temperature applications. More generally, these findings demonstrate targeting structural complexity as a strategy for suppressing the low-T κ latt in TE materials to optimise the ZT toward more widespread heat-recovery applications. |
673de5145a82cea2fae6cf76 | 4 | Calculations were performed using plane-wave density-functional theory (DFT) as implemented in the Vienna Ab initio Simulation Package (VASP) code. Initial structures of π SnS and SnSe were taken from previous work and optimised to tight tolerances of 10 -8 eV on the total energy and 10 - eV Å -1 on the forces. PBEsol+D3 was used to describe electron exchange and correlation, as this has been shown to provide an accurate description of the structure and dynamics of these materials. Projectoraugmented wave (PAW) pseudopotentials were used to describe the ion cores, with the valence configurations: Sn -5s 2 5p 2 , S -3s 2 3p 4 , and Se -4s 2 4p 4 . A plane-wave basis with a kineticenergy cutoff of 600 eV was employed with a 4 × 4 × 4 Γ-centred Monkhorst-Pack k-point mesh to model the valence electrons. |
673de5145a82cea2fae6cf76 | 5 | Phonon and κ latt calculations were performed using the Phonopy and Phono3py codes. The second-order (harmonic) force constants were obtained using a displacement step of 10 -2 Å and 2 × 2 × 2 supercell expansions with 512 atoms. The third-order (anharmonic) force constants were computed using a step of 3 × 10 -2 Å and the 64-atom unit cells. Phonon dispersion curves were obtained by interpolating the frequencies along q-point paths between the high-symmetry points in the P2 1 3 Brillouin zone. Atom-projected density of states (PDoS) curves were obtained from frequencies evaluated on uniform Γ-centered q-point grids with 24 × 24 × 24 subdivisions. The "particle-like" contribution to the κ latt was determined by solving the phonon Boltzmann transport equation (BTE) within the single-mode relaxation-time approximation (RTA), using modal properties computed on 10 × 10 × 10 Γ-centered q-point grids. Testing on smaller 8 × 8 × 8 meshes confirmed the RTA and full linarised BTE solution are quantitatively similar (Fig. , ESI † ). The "wave-like" intraband tunnelling contribution to the κ latt was additionally obtained by solving the Wigner transport equation. The S, σ and κ el were determined by solving the electron BTE within the momentum relaxation-time approximation (MRTA) as implemented in the AMSET code. Accurate electronic bandgaps, E g , and uniform band structures were obtained using non-self-consistent HSE06. Energy-and momentumdependent electron relaxation times were estimated from the rates of acoustic deformation potential (ADP), ionized impurity (IMP), piezoelectric (PIE), and polar optical phonon (POP) scattering. The deformation potentials were determined from single-point calculations on deformed structures generated by AMSET. The elastic constant matrices C, high-frequency/static dielectric constants ε ∞ /ε S , Born effective-charge tensors Z * and piezoelectric tensors e were obtained using the finite differences and density-functional perturbation theory (DFPT) approaches in VASP. All properties were computed with PBEsol+D3, apart from the ε ∞ , which were computed using nonself-consistent HSE06. The POP frequencies ω po were calculated by combining the phonon frequencies at q = Γ from Phonopy with the Z * from DFPT. k-point convergence tests and validation of non-self-consistent HSE06 are provided in Table . † Additional details of the calculations and analysis performed in this work, including key equations, are provided in the ESI. † |
673de5145a82cea2fae6cf76 | 6 | The structural chemistry of the Group IV-VI oxides and chalcogenides is strongly influenced by the stereochemical activity of the valence ns 2 lone pair on the tetrel T 2+ cations. This gives rise to five structure types across the TCh series (T = Ge/Sn/Pb, Ch = S/Se/Te), viz. a high-symmetry cubic rocksalt phase (Fm 3m spacegroup), two low-symmetry orthorhombic phases (Pnma, Cmcm), and intermediate-symmetry rhombohedral and cubic π phases (R3m/P2 1 3) (Fig. ). |
673de5145a82cea2fae6cf76 | 7 | The heavier PbCh form the rocksalt structure, in which the Pb 2+ cations adopt a centrosymmetric octahedral environment with six equal Pb-Ch bond lengths and the 6s 2 lone pairs are inactive. For GeS/GeSe and SnS/SnSe, the "strain" induced by placing the Ge 2+ /Sn 2+ in a centrosymmetric environment leads to an energetic preference for the orthorhombic Pnma structure, where the "pseudo-2D" layered structure and three-coordinate geometry allow the 4s 2 /5s 2 lone pairs to project into an interlayer void space. The R3m phase adopted by GeTe, and by SnTe at low temperature, lies between these extremes, with a cation off-centering along the rocksalt {111} direction giving three short and three long T-Te bonds and a partial relaxation of the strain in the highersymmetry phase. |
673de5145a82cea2fae6cf76 | 8 | The π structure is a distorted 2 × 2 × 2 expansion of the eightatom rocksalt conventional cell, and differs significantly from the rhombohedral phase in that it retains a cubic P2 1 3 spacegroup but with a larger primitive cell with n a = 64 atoms. The Pnma structure has one unique Sn site with one short and two long bonds to chalcogen atoms (Table † ). In contrast, the π structure has four Sn sites, with 2 × 4 atoms in rhombohedral-like sites with three equivalent Sn-Ch distances and the other 2 × 12 in sites with three different Sn-Ch bond lengths. |
673de5145a82cea2fae6cf76 | 9 | We predict optimised lattice constants of a = 11.38 and 11.77 Å for π SnS and SnSe, respectively, which are within 2% of the measured a = 11.6 and 11.97 Å. Previous studies have shown that both π phases are dynamically stable (i.e. there are no imaginary modes in the phonon dispersion curves) and energetically metastable with respect to the orthorhombic phases of SnS and the orthorhombic and rocksalt phases of SnSe. Our calculated phonon spectra confirm the dynamical stability, with real frequencies across the Brillouin zone (Figs 2 and S2 † ). |
673de5145a82cea2fae6cf76 | 10 | Using the SM-RTA we predict room-temperature κ latt from "particle-like" transport, κ p , of 0.26 and 0.18 W m -1 K -1 for π SnS and SnSe. These are well below the averaged values of 0.96-5.01 W m -1 K -1 for IV-VI chalcogenides in the other structure types in Fig. obtained from similar calculations. We previously showed that low κ latt in the IV-VI chalcogenides is a balance of low phonon group velocities and a large "phase space" of energy-and momentum-conserving phonon scattering pathways, favoured by structure types with large primitive cells, and strong three-phonon interactions, favoured in structures where the T 2+ are constrained to locally-symmetric environments. Separating the κ p into harmonic (group velocity) and weighted-average lifetime components, κ/τ CRTA and τ CRTA , confirms the large unit cell of the π structure reduces the group velocities by ∼75-80% compared to the Pnma phases (Fig. ). The τ CRTA of π SnS and SnSe are also around 30 and 50% smaller above 300 K (Fig. ). We stress that we only use the CRTA model to determine a weighted-average lifetime, and we do not use this approximation to calculate the κ latt . Comparison of the "phase space" functions N2 (ω) indicate that the π phases show comparable numbers of scattering pathways to the Pnma phases, but with higher spectral weight at low frequency (Fig. ). This does not follow the trend with unit-cell size in our previous study, indicating that the larger numbers of energy-conserving scattering channels enabled by the denser phonon dispersions are offset by momentum conservation restrictions from the higher cubic symmetry. The calculated Pλ place the anharmonicity in the π phases between the Pnma and R3m phases (Fig. ), which can be ascribed to 25% of the Sn atoms being in more constrained rhombohedral-Fig. Analysis of the "particle-like" thermal conductivity, κ p , of π SnS and SnSe using the models employed in our previous studies. The P in (d) are taken from our previous study, but include the P for Pnma SnS from Ref. . like environments (c.f. Table |
673de5145a82cea2fae6cf76 | 11 | where ∆ f ave is the average intraband spacing and f max is the maximum frequency in the dispersion. This condition is close to being met in both π SnS (∆ f ave = 4.4 × 10 -2 THz, f max /(3n a ) = 4 × 10 -2 THz) and SnSe (2.9 × 10 -2 THz, 2.5 × 10 -2 THz). Fig. (a) shows the contributions of the RTA κ p and intraband tunnelling κ w , obtained by solving the Wigner transport equation, to the κ latt of π SnSe. The κ p and κ w become equal around 200 K and the κ w accounts for 78% of the κ latt at 700 K. In contrast, the κ w of Pnma SnSe is almost negligible at 300 K and increases the κ latt by 18% at 700 K (Fig. ). π and Pnma SnS show similar, contrasting, behaviour (Figs S4/S5 † ). The very low κ p confirms the strategy of targeting large, complex unit cells can reduce the κ p below the amorphous limit. A consequence of the shallow temperature dependence of the κ w is that π SnS and SnSe have 70-75% lower thermal conductivity than the Pnma phases at 300 K, and κ latt comparable to the layering b direction, which is the "hard" axis for thermal transport, at high T (Fig. ). |
673de5145a82cea2fae6cf76 | 12 | To the best of our knowledge the κ latt of π SnS and SnSe have not been measured. We previously predicted a room-temperature lattice thermal conductivity of 0.13 W m -1 K -1 for π SnS. However, this calculation did not account for intraband tunneling and is equivalent to the κ p obtained in the present study. Using the data from Ref. , we obtain a κ w of 0.32 W m -1 K -1 and a total κ latt of 0.45 W m -1 K -1 (Fig. † ), which is 26% lower than predicted in the present work. One possible explanation for this discrepancy is the use of a dispersion correction in the present calculations, which has a small but noticeable effect on the dis- persion (Fig. ). This could be due in part to the ∼3.3% smaller equilibrium volume predicted with the dispersion correction, since the phonon frequencies, and, by extension, κ latt , can be sensitive to the volume. Related to this point, we also note that the anharmonicity in the π phases may be indicative of large thermal expansion, which could have a significant impact on the κ latt . However, the high computational cost of calculating the thirdorder force constants for the π phases means it is not feasible to account for this in the present calculations. |
673de5145a82cea2fae6cf76 | 13 | Other computational studies predicted room-temperature κ latt of 5.15 and 2.54/0.86 W m -1 K -1 for π SnS and SnSe using the Slack model, but the larger predictions suggest this model is likely not appropriate for these systems. On the other hand, Ref. predicted κ latt of 0.49 and 0.32 W m -1 K -1 for π SnS and SnSe using the Cahill model for the minimum thermal conductivity, which are remakably similar to the present calculations. |
673de5145a82cea2fae6cf76 | 14 | The calculated electronic band structures and DoS of π SnS and SnSe (Fig. † ) are comparable to previous studies and yield direct bandgaps of E g = 1.61 and 1.28 eV that are an excellent match to experimental measurements and previous calculations. The σ , S and κ el of the π phases were calculated as a function of temperature and an extrinsic carrier concentration n ("doping level") by solving the electron BTE with energy-and monentum-dependent electron scattering rates from four scattering processes common to semiconductors. With this approach, we estimate per-state electron lifetimes, which is an improvement on previous computational studies that have determined the σ and κ el with respect to an unknown, fixed or state-independent relaxation times. The doping level is a key parameter for optimising the electrical properties of semiconductor thermoelectrics. Pnma SnS and SnSe can be both p-type (hole) and n-type (electron) doped at n up to 4 × 10 19 cm -3 . 45,46 Fig. shows the calculated σ , absolute |S|, PF S 2 σ and κ el of the π and Pnma 36,37 phases of SnS and SnSe for p-and n-type doping with n = 10 16 -10 20 cm -3 at T = 700 K. This temperature was chosen to be well below the high-temperature Pnma → Cmcm phase transitions in orthorhombic SnS and SnSe (T = 878/807 K 71 ). |
673de5145a82cea2fae6cf76 | 15 | The π phases are predicted to have 10-20× lower σ with ptype doping, and ∼ 10× lower σ with n-type doping, than their Pnma counterparts (Fig. ). This is partially offset by a 1.5-2× larger |S| (Fig. ), but the net result is that the Pnma phases have up to 8× larger PFs at a reference n = 4 × 10 19 cm -3 . As for the Pnma phases, we predict larger σ but comparable |S| with n-type doping, yielding reasonable PFs of 1.1 and 1.4 mW m -1 K -2 for π SnS and SnSe, respectively, that can be enhanced to 1.7 and 2.3 mW m -1 K -2 at a larger n = 10 20 cm -3 (Fig. ). On the other hand, at higher n the κ el are proportional to the σ through the Wiedemann-Franz law, |
673de5145a82cea2fae6cf76 | 16 | where L is the Lorentz number, which in general is not constant. We therefore predict a negligible κ el of <0.1 W m -1 K -1 for the π phases with n = 4 × 10 19 compared to ∼1 W m -1 K -1 for ntype Pnma SnSe under the same conditions. This analysis shows that optimising the electrical properties of the π phases requires n-type doping at a level comparable to, or ideally higher than, the p-type doping levels reported in experiments on Pnma SnSe. The electrical conductivity is given by: |
673de5145a82cea2fae6cf76 | 17 | where e is the elementary charge and µ is the carrier mobility determined by the lifetime τ and conductivity effective mass m * σ , all of which are obtained from a weighted average over the bands that contribute to the transport. The Seebeck coefficient is related to the Seebeck effective mass m * where q = ±e for hole and electron carriers, respectively, h is the Planck constant and k B is the Boltzmann constant. The m * σ and m * S for the Pnma and π phases of SnS and SnSe were determined according to the procedure in Ref. at a nominal n = 4 × 10 19 cm -1 and T = 700 K and are presented in Table . † We calculate m * σ between 1.1-2.5 m e for the π phases, which are ∼ 5 -10× larger than the 0.15-0.27 m e for the Pnma phases. On the other hand, the m * S of the π phases are 3-4× larger than those of the Pnma phases. These differences partially account for the lower conductivity and higher Seebeck coefficients of the π phases. Furthermore, the m * σ for electrons are around 30 and 45-50% lower than holes in the Pnma and π phases, respectively, whereas the m * S differ by < 10%, which explains why these materials are predicted to show better n-type conductivity and power factors. |
673de5145a82cea2fae6cf76 | 18 | The m * σ and * S can be used to calculate the Fermi surface complexity factor (N * V K * ), where N * V is the effective valley degeneracy, i.e. the number of band extrema involved in the transport, and K * is the effective anisotropy factor that describes the shape of the carrier pockets in the Fermi surface. The (N * V K * ) shows a strong correlation to the maximum power factor, and values above ∼10 are characteristic of materials such as the lead chalcogenides where features of the electronic structure are known to contribute significantly to the thermoelectric performance. We calculate (N * V K * ) = 4.6/5.1 for p-type Pnma SnS/SnSe and 8.2/11 for the n-type materials, while the larger m * σ of the π phases result in 60-80% smaller values of 1.1/1.7 and 3.2/3.7. This implies that the electronic structures of the Pnma phases show features that are beneficial for the thermoelectric properties, and that these are largely absent in the π phases. While we do not calculate the N V or K, the values of (N * V K * ) = 1.1-3.7 predicted for the π phases suggest either a low effective valley degeneracy and/or an effective anisotropy K < 1. |
673de5145a82cea2fae6cf76 | 19 | Furthermore, comparing the µ as a function of n indicates the mobilities are also an order of magnitude lower in the π phases (Figs S9/S10 † ). The µ of both the Pnma and π phases is limited by polar-optic phonon (POP) scattering (Figs S9-S12 † ), for which the scattering rates are proportional to the POP frequency ω po and inversely proportional to the high-frequency and static dielectric constants ε ∞ /ε S . The π phases have larger ω po and smaller ε, which both favour stronger scattering. The lower σ of the π phases can therefore be explained by larger carrier masses and stronger (POP) carrier scattering. |
673de5145a82cea2fae6cf76 | 20 | The temperature dependence of the electrical properties is typically to doping level. At the larger n required to obtain reasonable PFs, the S 2 σ decrease with temperature due to the metallic-like decrease in σ with T typical of heavily-doped ("degenerate") semiconductors (Figs S13/S14 † ). As in Pnma SnS/SnSe, this is partially offset by an increase in the |S|, but for n-type π-SnSe there is a ∼30% reduction in the PFs from ∼2 to 1.4 mW m -1 K -2 from 300-700 K at our reference n = 4 × 10 19 cm -3 . |
673de5145a82cea2fae6cf76 | 21 | There have been a limited number of experimental studies on the π phases for photovoltaic applications. Measurements on thin films of n-type π SnS yielded a conductivity of 0.1 S cm -1 at 673 K and n = 3.7 × 10 17 cm -3 , which is comparable to our predicted 0.7 S cm -1 . Similarly, measurements on Sn(S 0.45 Se 0.55 ) films obtained a p-type conductivity of 2 × 10 -2 S cm -1 at 723 K and n = 10 16 cm -3 , 77 which falls between our predicted σ = 0.9 × 10 -2 and 6 × 10 -2 S cm -1 for π SnS and SnSe. Nair et al. reported that the room-temperature σ of Pnma SnS and SnSe are 2-3 orders of magnitude larger than the corresponding π phases, which is qualitatively consistent with our findings. The Seebeck coefficients of 485 and 314 µV K -1 measured for p-type SnS-SnSe and SnS-SnSe-SnS stacks with n = 9 × 10 15 and 10 15 cm -3 are less than half the maximum we predict. Our calculations on both π structures suggest the S change sign with temperature at low n, which would account for this discrepancy (Figs S13/S14 † ). |
673de5145a82cea2fae6cf76 | 22 | We predict Seebeck coefficients of |S| ≃ 300-600 µV K -1 at carrier concentrations between 10 19 -10 20 cm -3 , which are similar to the calculations in Ref. . However, the electron lifetimes in Ref. were calculated from the deformation potential, and do not account for the dominant POP scattering, so the predicted conductivity is significantly larger than in the present study. Our predictions of higher σ and PFs with n-type are also qualitatively consistent with the "reduced" values from the calculations in Ref. . |
673de5145a82cea2fae6cf76 | 23 | Combining the S, σ , κ el and κ latt allows us to predict the figure of merit ZT as a function of doping level and temperature (Eq. 1) For n-type π SnSe, an industrially-viable ZT > 1 can be obtained at T as low as 300 K, and ZT ≃ 3 comparable to state-of-theart experiments on orthorhombic SnSe 14 can be obtained at high temperature (Fig. ). With n = 4 × 10 19 cm -3 we predict maximum ZT , ZT max , of 1.6 and 2.4 for n-type π SnS and SnSe, respectively, close to the orthorhombic Pnma → Cmcm transition temperatures of 878 and 800 K 71 (Table ). These are larger than the ZT max of the p-type Pnma phases predicted by similar calculations, and very similar to the n-type ZT max (Table ). Larger n = 10 20 cm -1 improve the limiting electrical conductivity and PFs of the π phases, yielding significantly improved ZT max of 2.2 and 3 at these temperatures. Both the Pnma and π phases are predicted to show superior performance with n-type doping, but the differences are much larger in the π phases, with p-type ZT > 1 only attainable at high T and large n (Fig. , Table |
673de5145a82cea2fae6cf76 | 24 | Previous theoretical studies have predicted p-type ZT of 0.6 for π SnS at 800 K and ZT between 0.74 and 1 between 300-900 K. At the same T = 800 K and n = 10 20 cm -3 employed in Ref. , we predict larger p-type ZT = 1.21 and 1.25 for π SnS and SnSe. This study also predicted higher S 2 σ with p-type doping, in contrast to our results, which is likely due to the neglect of POP scattering in calculating the σ as noted above. The calculated ZT in Ref. are obtained by neglecting the κ latt entirely, while Ref. estimates the κ latt with the Slack model and uses a constant electron relaxation time to calculate the σ and κ el , all of which are likely to be poor approximations. However, since neither study specifies the carrier concentration(s) at which the ZT were calculated, it is impossible to compare to our calculations. |
673de5145a82cea2fae6cf76 | 25 | While many flagship thermoelectrics perform best at high tem- perature, the majority of applications require good low-T performance. We predict π SnSe to have ZT = 1.1 and 1.3 at 300 K with n = 4 × 10 19 and 10 20 cm -3 , respectively, which are ∼2-3 × larger than p-type Pnma SnSe (Fig. ). The "device efficiency" η of a TE material is given by: 2 |
673de5145a82cea2fae6cf76 | 26 | where T h and T c are the hot-and cold-side temperatures, respectively, and ZT is the average ZT over the temperature range T c → T h . The continuous increase in the ZT with temperature results in a ZT that increases continuously up to ∼1.9 and 2.3 at T h = 800 K with the two doping levels (Fig. (b)). In contrast, the slower rise in ZT with temperature in Pnma SnSe means ZT > 1 is only achieved when T h ≥ 650 K. Our results therefore show that n-type π SnSe, given a suitably-high doping level, could display viable thermoelectric performance for a wide range of low-to-mid temperature applications. A similar contrast is seen between π and Pnma SnS, i.e. the π phase performs better at low T and therefore shows a higher ZT over a wide temperature range. However, the generally lower ZT means the threshold of ZT = 1 is not reached until T h = 675 K (500 K with n = 10 20 cm -3 ). However, for larger T h = 880 K we predict reasonable ZT of 1.2 and 1.5 at the two doping levels. For applications where sulphide-based TEs are preferred, exploring doping strategies to maximise the n of π SnS may be facile. |
673de5145a82cea2fae6cf76 | 27 | As predicted by a previous analysis of the lattice thermal conductivity of the IV-VI chalcogenides, the complex structure lowers the particle-like transport below the amorphous limit, and the shallow temperature dependence of the intraband tunnelling transport leads to κ latt much lower than other chalcogenides at room temperature and competitive with the κ latt along the layering axis in the orthorhombic phases at high T . The cubic structure supports larger absolute Seebeck coefficients than the Pnma phases, but larger carrier effective masses and stronger polaroptic phonon scattering lead to low conductivity, and n-type doping with carrier concentrations on the order of 4 × 10 19 , and ideally larger, are required to obtain reasonable power factors. If this is achievable, we predict that π SnSe could show ZT > 1, competitive with Bi 2 Te 3 , 9 at room temperature, and ZT as high as 3 at 800 K, matching with state-of-the-art experiments on Pnma SnSe. With a typical 300 K cold-side temperature, this strong low-temperature performance results in a much higher average ZT than we predict for p-type Pnma SnSe, making π SnSe suitable for a wide range of low-to mid-temperature applications. The predicted performance of π SnS is also considerably better than p-type Pnma SnS, such that a ZT > 1 could be achieved for mid-temperature applications if the n can be optimised. |
673de5145a82cea2fae6cf76 | 28 | Despite being discovered almost a decade ago, the potential TE performance of the π monochalcogenides has not yet been uncovered, due to a combination of experiments focusing on photovoltaic applications and computational predictions using unsuitable approximations. Among the latter, there are likely few circumstances where neglecting the κ latt in the ZT equation would be reasonable. The Slack model significantly overestimates the κ latt of the π phases compared to solving the Wigner transport equation, although the Cahill model works remarkably well. Finally, unless a reasonable estimate of an electron relaxation time can be obtained (e.g. from experiments ), our results suggest electronic transport calculations within the constant relaxationtime approximation cannot be relied upon for quantitative predictions. On the other hand, the good match between our predicted σ and experimental measurements suggests phenomenological models for the electron scattering may be adequate provided they capture the dominant scattering mechanism(s) (POP scattering for Pnma and π SnS/SnSe). |
673de5145a82cea2fae6cf76 | 29 | In order to realise our predictions experimentally, two key challenges will need to be addressed, viz. synthesis/stability and optimising the electrical properties. π SnS and SnSe have been prepared in nanoparticulate and thin-film forms with techniques including chemical bath deposition (CBD), solution synthesis, spray pyrolysis, and aerosol-assisted chemical vapour deposition (AACVD). A number of experiments have examined thermal stability, with the consensus that π SnS transforms to the Pnma phase on heating to ≈600 K or higher. If the stability "window" cannot be widened, π SnS and SnSe could only be used for low-temperature applications. |
673de5145a82cea2fae6cf76 | 30 | Literature on the relevant properties of the π phases is currently sparse, and the largest n-type carrier concentrations of 4×10 17 reported to date are orders of magnitude lower than we predict are required to achieve high thermoelectric performance. p-type orthorhombic SnS and SnSe have been prepared with n up to 4 × 10 19 cm -1 , but experiments on n-doped SnSe have only achieved 10 19 cm -3 . Our prediction of superior performance in the n-type Pnma Sn and Ge chalcogenides suggests the exploration of strategies for n-type doping the IV-VI chalcogenides would be worthwhile. The structural preferences of the IV-VI chalcogenides can be understood in terms of the activity of the tetrel lone pair driven by the interaction between the tetrel and chalcogen atoms. Generally, the lighter chalcogenides prefer lower-symmetry phases (e.g. SnS/SnSe and GeS/GeSe adopt the orthorhombic Pnma phase), while heavier chalcogenides prefer higher-symmetry phases (e.g. SnTe, GeTe and the Pb chalcogenides adopt the rhombohedral and rocksalt phases). Previous studies have found that π SnS is closer to the convex hull than π SnSe (∆G = 2.3 and 3 kJ mol -1 per F.U. at 300 K). The existence of a direct conversion pathway to the orthorhombic phase has not, to our knowledge, been established, but we have previously considered conversion to the structurally-similar rocksalt phases and obtained energy barriers of 8.5 and 2.2 kJ mol -1 per F.U. for π SnS and SnSe. Based on this, admittedly incomplete, information, we suggest that the π phases might be stabilised by elements that favour the Pnma/R3m-like local geometry over the six-coordinate geometry in the rocksalt phases. This suggests that a SnS/SnSe solid solution could be used to stabilise π-SnSe, or that alloying with GeS/GeSe could be used to stabilise the π phases of both Sn chalcogenides. |
673de5145a82cea2fae6cf76 | 31 | For adjusting the carrier concentrations, we would expect the π phases to have similar defect chemistry to the orthorhombic phases. Applying similar principles to the above, we might favour dopants that do not form cubic chalcogenide phases, and thus of the two common dopants used for the p-type orthorhombic phases Ag may be preferable to Na. For n-type doping we might consider Group V elements such as Bi and Sb, since the sesquichalcogenides adopt orthorhombic and rhombohedral phases. We note, however, that if the π phases are kinetically formed, then the effect of potential dopants on the kinetics of formation may need to be considered. This is very difficult to address computationally, and we feel it may be better approached experimentally. |
673de5145a82cea2fae6cf76 | 32 | Despite the challenges of stabilising and heavily doping the π phases, in our view the exceptional predicted TE performance, and in particular the ultra-low low-temperature κ latt , warrants further investigation. On the last point, this study evidences the successful strategy of targeting structural complexity as a route to achieving low room-temperature κ latt and high low-temperature ZT . If our understanding of the IV-VI chalcogenides generalises to the related group IV oxides and V-VI sesquioxides/sesquichalcogenides, further investigation of the low-symmetry orthorhombic phases of the Sb 2 Ch 3 and Bi 2 Ch 3 , or ternary IV-V oxides in the pyrochlore and derived structure types, 38 could be rewarding. |
61addca06d4e8fe372abeade | 0 | Chemical structure generators enumerate or generate molecular graphs of organic or bioorganic molecules. They are an integral part of systems for computer-assisted structure elucidation (CASE) and can be used to create molecular libraries for virtual screening , or enumerate chemical spaces in general . The history of chemical graph generators goes back at least to the 1960s DENDRAL project which was aimed at the CASE of organic molecules based on mass spectrometric data . DENDRAL was developed for NASA's Mariner program to search for life on Mars . Its structure generator used substructures as building blocks and was able to deal with overlapping substructures. In the early history of the structure generators, ASSEMBLE was another building block based structure generator . In the field, there is a family of generators based on mathematical theorems such as algorithmic group theory and combinatorics . Besides DENDRAL, MASS was also another good example for the applications of mathematical theorems in structure generation. It was a tool for the mathematical analysis of molecular structures. SMOG was the successor of the MASS algorithm. Many works followed but few examples of practical usability are available even today . Among the currently available structure generators, such as DENDRAL, ASSEMBLE, SMOG, COCON and LSD , MOLGEN constituted the state-of-the-art for decades in terms of speed, completeness and reliability. The first version of MOLGEN was based on the strategy of DENDRAL software and developed to overcome the limitations of DENDRAL . The software is based on the orderly graph generation method . Although MOLGEN is the de facto gold standard in the field, it has the downside of being closed-source software. The algorithm cannot be further developed or modified by scientists based on their interests. The most efficient and fast open-source chemical graph generator was MAYGEN based on the orderly generation method. However, MAYGEN is approximately 3 times slower than MOLGEN. The state of the art of large scale structure generation was recently set by the lab of Jean-Louis Reymond in developing an in-house solution for a nauty-based structure generator, which enabled them to produce the numeration of 166 billion organic small molecules in the chemical universe database GDB-17. To the best of our knowledge, this in-house generator was not released as open-source or otherwise. Thus, there is still the need for an efficient open-source chemical graph generator. In we expressed the hope to "trigger a surge in the development of improved and faster" structure generators. Here we present the novel structure generator surge, based on the principle of the canonical generation path method. Surge is open-source and outperforms MOLGEN 5.0 by orders of magnitude in speed. Furthermore, surge is easily extensible with more features and adaptable to further application. |
61addca06d4e8fe372abeade | 1 | Surge is based on the nauty package for computing automorphism groups of graphs as well as canonical labels. Like nauty, surge is written in a portable subset of C and runs on a considerable number of different systems. Surge is an integration of three existing tools from the nauty suite : a) geng generates simple graphs based on certain boundary conditions, b) vcolg colors vertices in the output of geng and c) multig inserts multi-edges in the output of the first two tools (Figure ). An isomorphism between two graphs is a bijection between their vertex sets that maps edges onto edges. If the graphs have adornments, such as atom types for the vertices or bond multiplicities for the edges, then those adornments must be preserved by the mapping. If the two graphs are the same; i.e., the isomorphism is from a graph to itself, it is called an automorphism. The automorphisms form a group under the operation of function composition, called the automorphism group. The meanings of isomorphism and automorphism are different for each of the three stages in our algorithm. Referring to Figure , at the first stage (which we call a simple graph) there are no vertex or edge adornments and all rotations and reflections, 10 in total, are automorphisms. |
61addca06d4e8fe372abeade | 2 | When vertex adornments are added in the second stage, the atom type becomes significant so only the identity mapping and the reflection through the oxygen atom are automorphisms. In the final stage, edge adornments are added but in this example the automorphism group is not further reduced since the reflection through the oxygen atom preserves both atom type and bond multiplicity. Note how the automorphism groups in the second and third stages are subgroups of the automorphism groups in the previous stages. |
61addca06d4e8fe372abeade | 3 | Input to surge consists of a molecular formula such as C 7 H 12 O 2 S. Based on the element counts, in this case C=7, O=2, S=1, H=12, the atom valencies are used to calculate the plausible range of the number of edges of a connected simple graph representing the topology of a molecule with this formula, with hydrogen atoms omitted. Then geng is called to generate all the connected simple graphs with those parameters, subject also to a maximum degree condition depending on the molecular formula . Geng generates one graph from each isomorphism class and these are passed to the second stage as they are produced, without any need to store them . In this example, there are 10 non-hydrogen atoms and the number of edges is in the range 9-11. |
61addca06d4e8fe372abeade | 4 | Given a simple graph G from the first stage, the second stage assigns elements to vertices in all distinct ways. The element counts must be correct, and we must have valence degree at each ≥ vertex. More onerously, we only want one member of each equivalence class of element assignment under the automorphism group of G. We next explain how this is accomplished. The vertices of G are arbitrarily numbered 1,2,...,n. An element assignment can be represented as a list showing the element assigned to each vertex in order of vertex number. For example, a valid list might be L = (C,C,C,S,O,C,C,C,O,C). Automorphisms of G have an action on lists that permutes their entries. Namely, for list L and automorphism the list (L) assigns the same element to vertex (v) as L assigns to v, for each γ, γ γ vertex v. Thus, |
61addca06d4e8fe372abeade | 5 | If L is a list of elements and is an automorphism, L and (L) give equivalent assignment of γ γ elements to the vertices of G. Our task in this stage is to choose exactly one assignment from each equivalence class. Given a fixed ordering of the elements, for example C < O < S, two lists can be compared lexicographically, for example This algorithm is efficient if the automorphism group Aut(G) is small, but that is not always the case. Therefore, we adopt a more complex approach. An automorphism of G is called minor if there are two leaves (vertices of degree 1) x,y with a common neighbour and the automorphism merely swaps x and y; i.e. (x y). The minor subgroup M Aut(G) is the subgroup generated by ≤ all the minor automorphisms. A flower is a maximal set of leaves with the same neighbour. In the left graph of Figure , the flowers are {1,2,3}, {6,10} and {9,11}. The minor subgroup M consists of all automorphisms that preserve the flowers, such as (1 2 3)(9 11). The order of M is . In addition to 3! × 2! × 2! = 24 M, the automorphism group may contain automorphisms that do not preserve the flowers, such as (6 11)(7 8)(9 10). To capture such automorphisms, we colour the graph as in the right side of Figure . Vertices not in flowers are coloured black. Within each flower, vertices are coloured red, blue, green, … in order of vertex number, using a fixed list of colours that does not include black. Now let N be the group of automorphisms that respect the vertex colours. In the example, N has only the identity and (6 9)(7 8) ). An arbitrary automorphism of G can be obtained by first applying an element of N to capture how the flowers are mapped to each other, and then applying an element of M to capture the movement of leaves within each flower. In both steps the choice is unique, so we have a factorization Aut(G) = NM = { | in N, in M }. γδ γ δ (In the language of group theory, M is a normal subgroup and N is a complete set of coset representatives.) In the example, consider (1 2)(6 11)(7 8)(9 10). It swaps the flowers {6,10} and {9,11} so we choose the element of N which does that, namely = (6 9)(7 8) . Then γ we have to arrange the leaves within the flowers with an element of M, namely =(1 2)( for some in N and in M. Note that in both L and L* the δ γ γ δ elements are in nonincreasing order within each flower, as they are maximized with respect to M. Also recall that the automorphisms in N preserve the order of vertex numbers within the flowers, by virtue of the fact that we coloured the vertices in order of vertex number when we computed N. This means that we can take to be identity, and so L* = (L). This proves that L* δ γ = L, since L = max { (L) | in N }. γ γ In order to implement the condition L = max { (L) | in M }, we don't need to compute M γ γ explicitly. Instead, since M is generated by transpositions, it suffices that within each flower the elements are in decreasing order relative to vertex number. Using the ordering of elements that we have chosen, in the example we just need to enforce the inequalities element(1) ≥ element(2) ≥ element(3), element(6) ≥ element and element(9) ≥ element . The program recursively assigns elements to vertices in order of vertex number and enforces these inequalities as they become active rather than at the end. |
61addca06d4e8fe372abeade | 6 | After the assignment of elements to vertices is complete, the program moves to the next stage of selecting a bond multiplicity for each edge. This is the same type of problem as in the second stage. Instead of a list of elements for each vertex, we have a list of multiplicities for each edge. Instead of Aut(G), we use the subgroup of Aut(G) that preserves the element assignment. Otherwise M and N are defined as before. In the implementation, we don't use nauty to compute N but instead filter the N subgroup from the second stage, rejecting those automorphisms which don't preserve elements and converting the others to their action on the edges. As an example, geng makes 534,493 unlabelled simple graphs in 1.3 seconds for Lysopine C 9 H 18 N 2 O 4 . For these graphs, the second stage subgroup N is trivial 58% of the time and never larger than 72. Assignment of elements to vertices produces 3,012,069,151 vertex-labelled graphs in 90 seconds.The N subgroup for the third stage is trivial 98% of the time and never larger than 24. Finally, the assignment of bond multiplicities produces 5,979,199,394 completed molecules in an additional 100 seconds. As demonstrated by our examples, surge can generate molecular structures very quickly, allowing for the inspection of extremely large sets of isomers. The generation speed is several times faster than even the fastest output format (SMILES). On the other hand, any particular application will likely have stronger restrictions on the structure than just a molecular formula. For example, some substructures may make the molecule unstable or give it chemical properties undesirable in the application. Or, experimental investigation of an unknown compound may have determined some features of the structure, so that only molecules with those features are of interest. For these reasons, surge provides a number of filters to limit the output. The 3-stage generation method allows some of them to be implemented almost for free, and all of them are much more efficient than filtering the output through an external program. For example, restrictions on the number of short rings and the planarity of the molecule can be enforced at Stage 1. Surge also provides some "badlists" of forbidden substructures (many of them inspired by the corresponding feature of MOLGEN). The open-source nature of surge allows for a more advanced feature. By writing small code snippets, the user can insert custom filters into any of the three stages, and also perform such tasks as adding extra elements and command-line options. Several worked examples are provided with the program. |
61addca06d4e8fe372abeade | 7 | Surge is available under a liberal open-source License (Apache 2.0) on GitHub at as well as from . The system can be built with the standard Unix Configure/Make scheme and the resulting stand-alone executable is then run from the command line. By default, surge generates all constitutional isomers of a given molecular formula. Surge can write output in either SDfile or SMILES format. SMILES output is produced very efficiently by constructing a template for each simple graph at the first stage, so that only atom types and bond multiplicity must be filled in before output. We benchmarked surge with the set of molecular formulae given in Table . Since our motivation for developing structure generators is for the generation of large molecules, Table consists of natural products, randomly selected from the natural products database COCONUT . For the list of molecular formulae, surge outperformed MOLGEN by orders of magnitude (Figure ) and MOLGEN terminated at a built-in limit of 2 31 -1 structures. Reported computation times were linearly extrapolated based on the MOLGEN timing for 2 31 -1 structures and the actual number of isomers reported by surge. Note that surge generates between 7 million and 22 million molecules per second for all of these examples. |
61addca06d4e8fe372abeade | 8 | Table : Execution time (seconds) for selected MF of natural products on a compute-optimized c2-standard-4 Google cloud VM. Times for MOLGEN 5.0 were determined with the -noaromaticity flag to achieve comparability. Both MOLGEN and surge were instructed to generate but not to output structures. For randomly selected 10 molecular formulae, 4 options of surge were tested and results are given in Table . These options are -p0:1 At most one cycle of length 5 -P |
61addca06d4e8fe372abeade | 9 | We have presented surge, a structure generator for constitutional isomers based on the canonical generation path method. To the best of our knowledge, surge is the fastest chemical structure generator available. A number of badlist options are available to avoid the generation of potentially unlikely structures. Current limitations include the lack of an aromaticity detection. |
60c7525bf96a002e4a2881ce | 0 | Solid-liquid interfaces possess all of the complexities of gas-solid interfaces; yet, the presence of solvating molecules introduces specific interactions among solvent molecules, solid surfaces, and reactive species that greatly affect the efficacy of a material as a catalyst or adsorbent. These complex, manybody interactions among solvent molecules and surface species couple inherently fast electron-transfer processes with slower long-range changes in the structure of the solvation shells that cumulatively affect the free energies of adsorption, activation, and reaction. Catalytic reactions within zeolites are especially sensitive to these effects, because the pores that confine active sites possess dimensions similar to the solvation shells of surface intermediates (0.5 -2 nm). Consequently, the topology and polarity of these pores induce significant changes in the structure and dynamics of the solvent molecules. Water (H 2 O) molecules within microporous environments form intricate structures through hydrogen bonds (HB) that impact the stability of adsorbates, intrapore diffusion coefficients, and surface reactions. Decades of work have shown that HBs among aqueous solvents and polar surfaces can affect the adsorption of alcohols within microporous solids, transport of H 2 O through carbon nanotubes, and the stability of reactive surface intermediates within zeolites. The role of confined water in biocatalysis, however, is more clearly understood due to the successful characterization of water structures within the catalytic clefts of enzymes. This insight provided opportunities to design optimal solvents and mutations that improve enzymatic performance. Such knowledge does not yet exist for catalytic reactions within solvent-filled pores of zeolites despite the ubiquity and industrial importance of these systems. In zeolite catalysts, silanol defects ((SiOH) x ) and other HB donors stabilize H 2 O molecules within pores and near active sites. These H 2 O molecules form structures that depend on the topology of the zeolite and must reorganize in response to nearby catalytic events. For example, rates of 1-octene epoxidation increase 100-fold when the density of (SiOH) x increases from 0 to ~5 per unit cell within Ti-BEA, because the disruption of H 2 O clusters near Ti active sites greatly increases the entropy of the system. Conversely, glucose isomerization rates are 5 -50-fold greater in hydrophobic Ti-and Sn-BEA than hydrophilic analogues, because H 2 O co-localized near active sites forms HB structures with glucose and decreases the entropy of relevant transition states. In both of these examples, the dependence of epoxidation and glucose isomerization turnover rates on (SiOH) x density reflect changes in the thermodynamic stability of transition states via the H 2 O structures occluded within the BEA pores. Consequently, the catalysis community recognizes that H 2 O structures that form within zeolites greatly affects the stability of surface intermediates. However, we lack a quantitative understanding for how the physicochemical properties of these H 2 O structures depends on the shape and size of the surrounding pore and, more generally, how these structures impact catalysis at solid-liquid interfaces. |
60c7525bf96a002e4a2881ce | 1 | Here, we characterize the structure of H 2 O contained within FAU, BEA, MFI, and CDO zeolites of varying silanol density using a combination of spectroscopic and kinetic methods and molecular dynamics simulations to illustrate the impact of pore topology and polarity. We elucidate how the reorganization of H 2 O solvent structures during epoxidation catalysis changes the free energies of reactive surface intermediates and facilitates reaction. H 2 O molecules form bulk-like structures within the 1.3 nm supercages of FAU zeolites. In contrast, H 2 O molecules coalesce into confined one-dimensional chains as the characteristic pore dimension decreases from the 1.3 to 0.45 nm. Kinetic measurements of 1-alkene epoxidation over Ti-containing zeolites provide a complementary method to evince the effects of confined H 2 O molecules on the stability of species relevant for catalysis. Confined H 2 O molecules dynamically restructure to accommodate the formation of reactive intermediates, and these processes produce large differences (greater than 400-fold) in measured epoxidation rates. Collectively, these data and interpretations demonstrate how solvent molecules may form complex structures within confining environments, whose reorganization during catalysis may have a large impact on the stability of surface intermediates and measured rates and selectivities. These phenomena likely extend to other classes of microporous materials (e.g., carbon nanotubes, metal-organic frameworks, silico-alumino phosphates) and polar and protic solvents (e.g., alcohols, lactones), which present new opportunities to understand and engineer solid-liquid interfaces. |
60c7525bf96a002e4a2881ce | 2 | The presence of silanol defects ((SiOH) x ; i.e., the hydrophility of a material) stabilizes H 2 O within zeolite pores and impacts adsorption energies for H 2 O within BEA and MFI zeolites. The literature, however, lacks an understanding for how the structures that H 2 O forms depends on the combination of zeolite topology and (SiOH) x densities. Moreover, the thermochemical properties of these H 2 O structures will likely exhibit a complex dependence on the pore diameter and (SiOH) x density, because the importance of SiOH functions may change as H 2 O loses opportunities to hydrogen bond via spatial constraints. To understand the role of SiOH functions on stabilizing H 2 O structures within zeolite pores, we measured in situ infrared (IR) spectra and conducted molecular dynamics (MD) simulations of the H 2 O present within both hydrophilic and hydrophobic variants of the Ti-zeolite frameworks when these materials are immersed in liquid water. Titanium substituted zeolites with the FAU (1.3 nm supercages), BEA (0.65 nm pores), MFI (0.55 nm pores) frameworks and siliceous CDO (0.45 nm pores) were synthesized through post-synthetic modification or hydrothermal synthesis to prepare materials with a range of controlled topologies (e.g., 2D pore networks, 3D supercages), mean pore dimensions (0.45 -1.3 nm), and densities of (SiOH) x . Here, we make comparisons primarily among materials that contain the greatest and lowest densities of (SiOH) x (i.e., the most hydrophilic and hydrophobic samples). Materials named Ti-zeolite-OH are the variant of the framework that contains the greatest densities of (SiOH) x (i.e., are the most hydrophilic), while those denoted as Ti-zeolite-F were synthesized to contain low densities of (SiOH) x (i.e., are hydrophobic). The MD simulations of hydrophilic materials involved structures containing ~5 (SiOH) 4 defects per unit cell place at non-adjacent tetrahedral sites but otherwise located randomly. The same (SiOH) 4 density was used in simulations of Si-CDO-OH samples to facilitate equitable comparisons, although this material contains fewer (SiOH) x groups than other Ti-zeolite-OH samples (Fig. ). MD simulations of hydrophobic structures made use of pristine frameworks with zero (SiOH) x groups. Detailed characterization of all materials, the specifics of experimental methods, spectral and rate analysis methods, and computational methods are in the supporting information. |
60c7525bf96a002e4a2881ce | 3 | Bulk H 2 O differs structurally from H 2 O stabilized within the pores of a zeolite. Figure shows IR spectra of bulk water (Fig. ) and of water within Ti-FAU-OH (Fig. ), Ti-BEA-OH (Fig. ), Ti-MFI-OH (Fig. ), and Si-CDO-OH (Fig. ) at 313 K. These spectra show distinct features between 2750 -3750 cm -1 that reflect ν(O-H) of H 2 O molecules within chemically distinguishable bonding configurations. Experimental IR, Raman, and sum-frequency generation spectroscopies and computational predictions demonstrate that condensed-phase H 2 O molecules possess five specific hydrogen-bonding configurations (top of Fig. ; Table ) that include "free" H 2 O that lacks HB (Free; ~3636 cm -1 ), H 2 O that donates two and accepts one HB (DDA; ~3570 cm -1 ), H 2 O that donates one and accepts one HB (DA; ~3470 cm -1 ), H 2 O that both donates and accepts two HB (DDAA; ~3320 cm -1 ), and H 2 O that donates one and accepts two HB (DAA; ~3040 cm -1 ). The vibrational spectra of bulk H 2 O (Fig. ) shows that ν(O-H) of H 2 O in a DDAA bonding configuration dominates the measured spectra, which agrees with prior reports. Within the bulk-fluid phase, H 2 O molecules form dynamic three-dimensional hydrogen bonding structures described well by our MD simulations of TIP5P H 2 O (Fig. ). IR spectra of H 2 O within Ti-FAU-OH (Fig. ) show that H 2 O within the 1.3 nm supercages of hydrophilic FAU and of hydrophobic (pristine) Ti-FAU-F (Fig. ) appear largely similar and possess lineshapes that resemble bulk H 2 O. H 2 O within these supercages possess the spatial freedom to form threedimensional structures (Figs. and); yet, a fraction of H 2 O molecules adopt a DDAA configuration by simultaneously forming HBs with both SiOH defects and other H 2 O. These differences are reflected between the time-averaged spatial distributions of H 2 O in Ti-FAU-F and Ti-FAU-OH: the former exhibits a regular and well-defined boundary (Fig. ) whereas that latter contains irregularities caused by the attractive interactions between H 2 O and randomly sited silanol groups within the FAU structure (Fig. ). Within Ti-BEA-OH and Ti-MFI-OH, H 2 O molecules within the DA configuration account for a much larger fraction of the vibrational spectra (Figs. and). These hydrophilic materials, however, also stabilize greater numbers of DDAA H 2 O molecules than the analogous Ti-BEA-F (Fig. ) and Ti-MFI-F (Fig. ) materials, because H 2 O hydrogen bonds with (SiOH) x defects at the pore walls. Within Ti-BEA-OH, H 2 O molecules with the DDAA configuration congregate at intersections between pores (Figs. 1h, 1l, S4d, and S4g) or at mesoporous grain-boundary defects. In comparison to frameworks with larger pore diameter, H 2 O molecules within Si-CDO-OH (Fig. ) largely exist within the DA hydrogen bonding configuration. Collectively, the combined interpretation of these results suggest that H 2 O molecules form solvent structures that, on average, possess fewer hydrogen bonds per H 2 O molecule and different hydrogen bonding partners as the characteristic diameter of the surrounding pore decreases. To further investigate how topology and surface chemistry of the pores affect the structure of H 2 O, we compared the average number of HBs calculated from vibrational spectra (<N HB > IR ) to those observed during the MD simulation (<N HB > MD ). The experimental measurements and MD simulations show consistent differences among hydrophobic (Ti-zeolite-F) and hydrophilic (Ti-zeolite-OH) zeolite structures. The average number of HB decreases uniformly with the mean diameter of the zeolite pore and with the number of hydrogen bond donating residues on the pore walls. Figure establishes these trends via the level of parity between values of <N HB > IR and <N HB > MD across all materials examined. H 2 O within the bulk fluid phase and Ti-FAU possess an average of 3 -3.6 HBs irrespective of density of (SiOH) 4 . The average number of hydrogen bonds decreases monotonically towards an average value of ~2 as the characteristic pore diameter decreases across the series of Ti-BEA, Ti-MFI, and Si-CDO, because these sub-nanometer siloxane pores do not permit three-dimensional HB networks among H 2 O molecules. Further, values of <N HB > IR and <N HB > MD agree closely with prior reports, where MD simulations of defect-free Si-FAU, Si-MFI, and H + -form Al-BEA show that, on average, H 2 O within the FAU, BEA, and MFI frameworks possess ~3, ~2.7, and ~2 HBs per H 2 O molecule, respectively. |
60c7525bf96a002e4a2881ce | 4 | Analysis of the MD simulations (validated by comparisons to experimental results, Figure ) provide insight to the structural properties of H 2 O within the zeolite frameworks and its dependence on interactions among H 2 O molecules, (SiOH) x defects, and the siloxane pore walls. Ti-FAU-F, on average, stabilizes ~118 H 2 O molecules per unit cell, while Ti-BEA-F contains an average of ~5 H 2 O molecules per unit cell (Table ). In comparison, Ti-MFI-F and stabilizes only ~ 2 H 2 O molecules per unit cell. Strikingly, H 2 O molecules do not enter the Si-CDO-F structure over throughout the course of our 1 µs MD simulations. These low uptakes are qualitatively consistent with reported H 2 O uptakes into these hydrophobic frameworks. The MD simulations indicate that the numbers of H 2 O molecules stabilized within hydrophilic variants of Ti-FAU, -BEA, and -MFI differ only slightly from values in hydrophobic samples (Table ) but Si-CDO-OH possesses ~2 H 2 O per unit cell. Given the small change in the overall volume per unit cell between hydrophilic and hydrophobic samples of the zeolites, the number of water molecules per unit cell for these variants are expected to be similar. However, the hydrophilic samples preferentially stabilize H 2 O molecules near defects and increase the local density of water. Conversely, the hydrophobic samples containing similar number of water molecules would exhibit higher fluctuations in the water molecule positions in absence of any stabilizing interaction with the pore. Together, these data suggest H 2 O completely fills the supercages of FAU but does not fully occupy the sub-nanometer channels of BEA, MFI, and CDO. Instead, H 2 O molecules coalesce into discrete structures distributed throughout these porous architectures. Speciation of the HBs to H 2 O molecules within the MD simulations reveal that the percentage of HBs that H 2 O forms with surface functions (i.e., SiOH or Si-O-Si linkages) increases as characteristic pore diameters decrease and (SiOH) 4 density increases (Tables and). HBs between H 2 O molecules and SiOH functions account for 1% and 6% of all HB interactions within Ti-FAU-F and Ti-FAU-OH, respectively. In contrast, much larger percentages of HBs involve SiOH groups within Ti-BEA-OH (16%), Ti-MFI-OH (35%), and Si-CDO-OH (60%). Further, these trends and further analysis indicate H 2 O molecules form one-dimensional chains (primarily dimers through tetramers) with multiple points of interactions with (SiOH) x defects within BEA, MFI, and CDO structures. The prevalence and structure of these oligomeric chains agrees with IR spectra that show large fractions of H 2 O exists in DA hydrogenbonding configurations inside Ti-BEA, Ti-MFI, and Si-CDO-OH (Figs. and). In summary, concurrent interpretation of the infrared spectra of ν(O-H), parity between <N HB > IR and <N HB > MD , and simulated distributions of HB configurations strongly suggest that H 2 O forms three-dimensional bulk-like structures within FAU but largely one-dimensional oligomers within BEA, MFI, and CDO, the density of which should vary with the activity of liquid H 2 O and density of silanol groups. |
60c7525bf96a002e4a2881ce | 5 | As a consequence of these interactions, the molecular motion of H 2 O molecules are strongly correlated in ways that depend on the size and surface chemistry of the surrounding pores. Although Ti-BEA, Ti-MFI, and Si-CDO zeolites all stabilize H 2 O molecules as oligomeric chains, the translational motion of H 2 O within the pores differs significantly among these frameworks and with the presence of (SiOH) x . Analysis of time-averaged spatial distributions from MD simulations (Section S4.2) show that H 2 O molecules can radially translate up to ~2.2 Å from the centerline of the Ti-BEA-F pores but only ~ 1 and 0.6 Å from the center of pores within Ti-MFI-F and Si-CDO-OH, respectively. Moreover, the space accessed by H 2 O molecules increases significantly with the addition of (SiOH) 4 . These changes in spatial and structural confinement of the chains of H 2 O molecules within these zeolites significantly affects the thermodynamic properties of these solvent structures (vide infra). The thermodynamic stability of hydrogen bonds between H 2 O molecules entrained within zeolites depend strongly on the shape and size of the solvent cluster. We measured IR spectra of H 2 O molecules in the bulk-fluid phase and within pores of hydrophilic and hydrophobic zeolite frameworks (FAU, BEA, MFI, CDO) as functions of temperature to interrogate the changes in enthalpy and entropy that result from the disruption of HBs within H 2 O structures in these distinct environments. Figure shows IR spectra of bulk H 2 O as a function of temperature (303 -343 K) possess ν(O-H) that systematically shift to higher wavenumbers with increasing temperature and shows a clear isosbestic point at 3345 cm -1 , which agrees with previously reported Raman and infrared spectra of H 2 O. The blue shift in ν(O-H) with temperature reflects a change in the populations of H 2 O that exist in one of the five hydrogen bonding configurations (vide supra). The change in enthalpy associated with disrupting a single hydrogen bond among H 2 O molecules (ΔH <HB> ) within the different pore environments was quantified via van't Hoff analysis of the changes in 〈𝑁〉 . In this analysis, the molar extinction coefficients for ν(O-H) of H 2 O within the various HB arrangements are assumed to be the same. The enthalpy change associated with disrupting a H 2 O molecule (ΔH HB ) is then determined by multiplying ΔH <HB> by 〈𝑁〉 , which describes the change in enthalpy associated with perturbing the average H 2 O molecule within a given environment (Section S2.3.1). The corresponding change in entropy due to breaking a hydrogen bond (ΔS HB ) was determined using statistical mechanics (i.e., comparing the populations of H 2 O in the varying bonding configurations; Section S2.3.1). 54 ). These values agree with literature values for the changes in enthalpy (~10 kJ mol -1 ) and entropy (~25 J (mol H 2 O) -1 K -1 ) associated with the disruption of hydrogen bonding interactions within bulk H 2 O. H 2 O within Ti-FAU form nearly three-dimensional structures that result in H 2 O molecules with <N HB > values only slightly less than that for bulk water (<N HB > = 3.3). As a result, values of ΔH HB for H 2 O entrained within Ti-FAU (~9 ± 2 kJ (mol H 2 O) -1 ) resemble that for bulk H 2 O. Within Ti-BEA and Ti-MFI, H 2 O appears to exist as oligomers (vide supra) that merge and gain additional HB at intersections within these three-dimensional pore networks, such that H 2 O molecules possess two to three hydrogen bonds, on average. Consequently, values of ΔH HB within Ti-BEA and Ti-MFI (~5 kJ (mol H 2 O) -1 ) are approximately half the value of those found in Ti-FAU and bulk H 2 O. Interestingly, the value of <N HB > for Si-CDO-OH is the lowest among all structures (<N HB > = 2.0) and ΔH HB approaches zero, because the vibrational spectra of H 2 O show negligible changes with increasing temperature. This behavior suggests that H 2 O oligomers within Si-CDO-OH access multiple configurations with similar energies (Section S2.4). We postulate that the disruption of individual HB between pairs of H 2 O molecules results in concomitant restructuring of the H 2 O oligomers to form HB with nearby (SiOH) x functions, which maintains a constant average of two HB and gives nearly equivalent interaction energies. This interpretation seems consistent with values of ΔH HB that do not depend on the density of (SiOH) x within a given framework despite the significant fraction of HB observed between H 2 O and (SiOH) 4 in simulations (Tables and). Comparisons among these measurements show that the intrinsic enthalpic stabilization afforded by a single hydrogen bond (i.e., values of ΔH <HB> ) does not depend on the topology or defect density of a zeolite, however, the entropy gained when a hydrogen bond cleaves depends strongly on the remaining number of hydrogen bonds to the H 2 O molecule. |
60c7525bf96a002e4a2881ce | 6 | The disruption of H 2 O structures gives rise to an increase in entropy that reflects the increase in the number of accessible microstates of the system. Values of ΔS HB within Ti-FAU are similar that of bulk H 2 O, which reflects the three-dimensional H 2 O clusters (~1.3 nm) that form within the supercages of FAU. ΔS HB values within Ti-FAU do not depend on the density of (SiOH) x , likely because the HBs within these supercages primarily reflect bonds between H 2 O molecules, rather than with (SiOH) x defects (Table ). The disruption of H 2 O within Ti-MFI leads to ΔS HB that are ~12 J mol -1 K -1 greater than Ti-BEA. The greater values of ΔS HB for H 2 O in 0.55 nm pores of Ti-MFI reflect lesser <N HB > compared to those in the 0.65 nm pores of Ti-BEA. Values of ΔS HB also decrease slightly with [(SiOH) x ] for a given zeolite framework, which suggests that (SiOH) x defects provide a stabilizing interaction when HB among H 2 O molecules are disrupted. The differences of ΔS HB among all these zeolite structures may be envisioned as cutting strands of a rope. Within bulk water and Ti-FAU, the ropes possess multiple strands that bind H 2 O molecules even when a single strand fails. In contrast, ropes within Ti-MFI, and to a lesser extent Ti-BEA, contain fewer strands and the loss of a single one may allow H 2 O molecules to break free. Si-CDO-OH differs considerably: measured values of ΔS HB (and ΔH HB ) are nearly zero because a HB between H 2 O and a silanol group replaces each HB lost between H 2 O molecules due to significant spatial confinement within these 0.45 nm pores. |
60c7525bf96a002e4a2881ce | 7 | Collectively, these data and interpretations show that H 2 O molecules coalesce into structures, whose shape and properties depend strongly on the topology of the surrounding pore and to a lesser degree on the [(SiOH) x ]. As the characteristic dimensions of the pores decrease, the H 2 O solvent structures lose the added stability conferred by forming HB with other H 2 O in more than one dimension and become motion among H 2 O molecules becomes increasingly correlated. Therefore, the enthalpic costs and entropic gains associated with breaking each HB depend on zeolite topology. These relationships between the enthalpy and entropy of creating and forming HBs with H 2 O molecules carries catalytic consequences that appear as differences in turnover rates and the stability of surface intermediates at solid-liquid interfaces, as we demonstrate below. |
60c7525bf96a002e4a2881ce | 8 | Kinetic measurements for alkene epoxidation reactions with H 2 O 2 provide an ideal method to examine the catalytic consequences of the structure and dynamics of intraporous H 2 O, because these reactions exhibit rates that depend strongly on the density of (SiOH) x and size of the confining pore. 18,39,40 Figure shows rates of 1-alkene epoxidation are 10 -400-fold greater in zeolites that contain significant densities of (SiOH) x (i.e., are hydrophilic) than the analogous structures with far fewer (SiOH) x . The rates of 1-alkene epoxidation are ~10-fold greater in Ti-FAU-OH than Ti-FAU-F (Fig. ) and depend weakly on the chain length of the 1-alkene. Turnover rates for 1-hexene epoxidation are 60-times higher within Ti-BEA-OH in comparison to Ti-BEA-F (Fig. ), and the difference between rates in hydrophilic and hydrophobic variants increases with 1-alkene chain length to a difference of 150-fold for 1-decene. Finally, the ratio of epoxidation rates within Ti-MFI-OH to those in Ti-MFI-F increase from a ratio of 100x to 400x as the 1-alkene is changed from 1-hexene to 1-decene (Fig. ). These differences in turnover rates among Ti-zeolites and 1-alkenes do not reflect differences in the mechanism for epoxidation or differences in the electronic properties of the active sites. Rather, vast range of epoxidation rates among Ti-zeolite with distinct topologies and (SiOH) x densities reflects differences between the structures of solvent molecules that solvate reactive intermediates and transition states at Ti active sites and that recognize the shape, size, and polarity of the surrounding voids. Silanol functions that neighbor Ti active sites nucleate structures of H 2 O that reorganize in response to the individual elementary steps during epoxidation catalysis. The extent by which solvent molecules must restructure depends on the dimensions of the reactive intermediates, the initial structure of the solvent molecules (that depend on zeolite topology and polarity), and the type and strength of specific interactions between the solvent and reactive species. Consequently, the size of the 1-alkene correlates to the average number of hydrogen bonds and solvent molecules disturbed to accommodate the surface intermediates and transition states for epoxidation. Under reaction conditions where the concentration of H 2 O 2 exceeds that of the 1-alkene, pools of Ti-OOH and Ti-(η 2 -O 2 ) intermediates comprise the most abundant reactive intermediate (MARI; collectively denoted as Ti-OOH) which leads to the following rate expression for alkene epoxidation |
60c7525bf96a002e4a2881ce | 9 | where r E is the rate of epoxidation, [L] is the total number of Ti atoms, T is the absolute temperature in Kelvin, k b , h, and R are the Boltzmann, Planck, and ideal gas constants, respectively, ∆𝐺 ‡ is the apparent free energy of activation, and [alkene] is the fluid-phase concentration of the alkene. |
60c7525bf96a002e4a2881ce | 10 | Born-Haber thermochemical cycles provide a conceptual framework to deconstruct an apparent change in free energy between two states into individual contributions from distinct chemical processes. Scheme 1 shows a series of chemical steps that conveniently deconvolutes measured values of ∆𝐺 ‡ into specific chemical interactions within Ti-zeolites that possess low (Scheme 1a) or high (Scheme 1b) densities of (SiOH) x . In this sequence, fluid-phase alkene molecules enter the pores of the Tizeolite, displace solvent molecules, and interact with the surrounding pore walls, which corresponds to a free energy of adsorption (∆𝐺 ) that depends primarily on the characteristic size of the surrounding pores. Next, solvent molecules localized near Ti active sites must reorganize to accommodate the 1-alkene conformation that positions the C=C with Ti-OOH intermediates prior to O-atom transfer (∆𝐺 ). Finally, the events that lead to O-atom transfer between the Ti-OOH and the C=C is described by an intrinsic free energy of activation (∆𝐺 ‡ ) that depends only on the identity of the active metal atom. In this conceptual framework, values of ∆𝐺 ‡ are given by |
60c7525bf96a002e4a2881ce | 11 | We have shown that ∆𝐺 ‡ does not depend on the size of the confining pore or the density of (SiOH) x , because ∆𝐺 ‡ corresponds to the free energy required to exchange electrons between Ti-OOH species and the C=C following all changes to the configuration of the alkene and solvent. Alkenes preferentially associate with hydrophobic, siloxane regions of the pores upon adsorption within the zeolite; therefore, values of ∆𝐺 primarily reflect differences in the characteristic pore size and do not depend significantly on (SiOH) x density. Then, this framework suitably attributes all changes in ∆𝐺 ‡ caused by solvent restructuring about the transition state complex to the excess interaction between that transition state and surrounding solvation shells, which ∆𝐺 captures. The hydrophobic Ti-zeolites contain exceedingly few (SiOH) x proximate to Ti active sites, and as such, the local solvent structures predominantly consist of non-hydrogen bonded CH 3 CN molecules that require insignificant reorganization energies and offer a convenient point of reference. As such, we set the ∆𝐺 of Ti-zeolite-F materials to be equal to zero for the purposes of these comparisons. This analysis provides differences between ∆𝐺 ‡ that for a given zeolite topology reflect the disruption and reorganization of H 2 O molecules near Ti active sites within the hydrophilic Ti-zeolites. |
60c7525bf96a002e4a2881ce | 12 | Here, ∆𝐺 quantifies the free energy difference caused by the disruption of hydrogen bonding interactions among H 2 O and solvent molecules as these species reorganize to accommodate the formation of the reactive intermediate that precedes the epoxidation transition state, which strongly resembles the transition state itself. . The dashed lines represent the values of ΔH HB and ΔS HB measured via variable temperature IR spectroscopy in the bulk fluid phase (black) or in Ti-FAU (orange), Ti-BEA (blue), and Ti-MFI (purple). The upper line for each of the Ti-zeolites corresponds to the most hydrophilic material, while the lower linear corresponds to the most hydrophobic sample. The shaded region is intended to represent the span of enthalpy -entropy compensation expected within these different pore environments due to the differences in (SiOH) x for a given zeolite topology. |
60c7525bf96a002e4a2881ce | 13 | Figure shows values of values of ΔH excess increase linearly with ΔS excess for 1-alkene epoxidations within Ti-FAU, Ti-BEA, and Ti-MFI catalysts, which reflects enthalpy-entropy (H-S) compensation effects widely reported for catalytic reactions and intimately tied to the dynamic restructuring of HB networks of H 2 O molecules within the zeolite pores (Fig. , Table ). The values of ΔH excess and ΔS excess increase in proportion to one another with increasing 1-alkene chain length, which suggests that the perturbation of H 2 O molecules, and consequently values of ΔH excess and ΔS excess , increase with the number of methylene units of the reactants. Within Ti-FAU, values of ΔG excess (at 313 K, via ΔG excess = ΔH excess + TΔS excess ) lie near 0 kJ mol -1 , which agrees with rate ratios for 1-alkene epoxidation over Ti-FAU-OH and Ti-FAU-F that do not depend on alkene chain length (Fig. ). In contrast, values of ΔG excess become increasingly negative with chain length or (SiOH) x density within Ti-BEA zeolites, because the entropy gain associated with disrupting H 2 O clusters confined within BEA pores overwhelms the enthalpic cost of reorganizing these molecules. These comparisons reveal the molecular origins for epoxidation rate ratios that increase with 1-alkene chain length in Ti-BEA-OH and Ti-BEA-F (Fig. ). The same phenomenon appears also within Ti-MFI materials only to a larger extent with rate ratios between Ti-MFI-OH to Ti-MFI-F that reach a factor of 400 for 1-decene epoxidation (Fig. ). These greater gains in ΔS excess for epoxidation transition states in Ti-MFI reflect the low value for <N HB >, in comparison to other zeolite frameworks, which give rise to larger entropy gains by liberating H 2 O molecules for the minimal enthalpic cost (Table ). |
60c7525bf96a002e4a2881ce | 14 | The observed H-S compensation effects via epoxidation catalysis (i.e., ΔH excess and ΔS excess ) are quantitatively consistent with the experimentally measured changes in enthalpies and entropies associated with disrupting H 2 O solvent structures (i.e., ΔH HB and ΔS HB ; Fig. ) within these same pore environments. Figure shows the H-S compensation that originates from ΔH HB and ΔS HB measured independently via IR spectroscopy (vide supra) within bulk H 2 O and H 2 O filled pores of Ti-zeolite catalysts. The excellent agreement between kinetic and spectroscopic measurements testifies to the control that solvent reorganization processes hold over the stability of adsorbates and reactive intermediates (e.g., epoxidation transition states). Significantly, the H-S compensation relationship obtained from analysis of IR spectra do not rely on the presence of a reactant and do not arise from measurements of specific catalytic reaction rates, therefore, these measurements strongly suggest that many other catalytic reactions will be governed by these same compensation phenomena within pores of these zeolites within aqueous solvents. Here, hydrogen bonded H 2 O molecules within confining pores (<1.3 nm) possess entropies that are much lower than bulk H 2 O as a result of the highly correlated molecular motion imparted by the surrounding pores. The disruption of these rigid H 2 O structures results in an entropic gain that is much larger than observed in bulk H 2 O, which leads to entropy-dominated decreases in the free energy of the system. In summary, this work gives compelling evidence for how the structure of intraporous solvent molecules and the reorganization of these solvent structures to accommodate the formation of reactive surface intermediates lead to consequential changes in the stability of reactive surface intermediates and large changes in the rates of reactions. |
60c7525bf96a002e4a2881ce | 15 | The structures of hydrogen bonded H 2 O solvent clusters depend sensitively on the shape and size of the surrounding void. Within 1.3 nm cavities, H 2 O molecules coalesce into three-dimensional clusters that resemble bulk liquid. In contrast, H 2 O within sub-nanometer pores form short largely one-dimensional chains of molecules that exist throughout the pore structure and possess cross-linking hydrogen bonds primarily at intersections between pores. During liquid-phase catalysis, these H 2 O structures must reorganize to accommodate the formation of reactive surface intermediates which leads to disproportionate increases in entropy that outweigh the enthalpic penalty for disrupting these confined solvent molecules. These phenomena likely affect a wide range of catalytic reactions that proceed within zeolite pores filled with aqueous solvents, and these reactions should therefore follow the enthalpy-entropy compensation relationships described here (Table ). |
60c7525bf96a002e4a2881ce | 16 | The understanding developed here applies broadly, because the chemical interactions among polar SiOH functions, reactive intermediates, metal active sites, and protic solvents are recognized to strongly affect multiple classes of reactions that occur within microporous catalysts including methanol-to-hydrocarbons (MTH), alkene epoxidations, glucose isomerization, alcohol dehydration, transfer hydrogenation, and Baeyer-Villiger oxidation reactions. These data and interpretations clarify how the proximity of hydrogen-bonded solvent structures to reactive metal atoms leads to significant changes in the free energies of reactive intermediates via solvent reorganization. The condensation of protic molecules within zeolite pores may lead to the formation of extended solvent structures even during gas-phase catalysis. As such, similar phenomena should be considered in these circumstances. These concepts should also be considered examine in the context of optimizing rate, selectivities, and yields for catalytic conversions. Our findings demonstrate that the zeolite framework may be viewed as a way to escape the single-factor linear free energy relationships that link the stabilization of the reactant in a solvent to the stability of the transition state. Combinations of solvents (single or multicomponent) and microporous frameworks (zeolite, metal-organic-frameworks, or covalent-organic-frameworks) may be identified that make use of these emergent enthalpy-entropy compensation effects. While previous discoveries in this area often involved some level of serendipity, the comparisons demonstrated here suggest rational approaches for harnessing solvent, zeolite, reactant interactions to improve catalyst performance by significant margins. On-going work within our group seeks to understand how the structure of protic solvents and aqueousorganic solvent mixtures influence catalytic reactions and to exploit the characteristics of these hydrogen bonded solvation shells. |
60c7525bf96a002e4a2881ce | 17 | Ti-BEA-F was synthesized hydrothermally in fluoride media by adapting a previously published procedure. TEAF was dissolved in deionized H 2 O in a polypropylene container and combined with TIPO to produce a clear homogeneous solution. TEOS was then added slowly over a period of one minute to this solution under static conditions, which initially formed a biphasic mixture. The mixture was then stirred for 16 h to produce an opaque homogeneous solution. The lid to the polypropylene container was then removed to evaporate the ethanol and isopropanol that form through the hydrolysis of the TEOS and TIPO, respectively. The solution was left open while stirring until the mass of the solution decreased by 115% of the value of the estimated mass of the alcohols to ensure that the alcohols evaporated completely. Subsequently, deionized H 2 O was added to yield a gel with a final molar composition of 1 SiO 2 : 0.0033 TIPO: 0.56 TEAF: 7 H 2 O. This gel was then loaded into a Teflon-lined stainless-steel autoclave (Parr instruments, 45 cm 3 ) that contained 5 wt. % (relative to SiO 2 within the gel) dealuminated BEA as seeds to promote the formation of the BEA zeolite framework. This autoclave was then sealed and heated to 413 K while rotating (60 rpm) in a convection oven (Yamato, DKN602C) for 25 days. The resultant solids were recovered, washed with H 2 O, and dried for 16 h at 373 K. The dried solids were then heated at 1 K min -1 to 823 K in flowing air (100 cm 3 min -1 ) and held at 823 K for 10 h to produce a bleached-white solid. Ti-MFI was synthesized hydrothermally in either hydroxide or fluoride media. A desired amount of the TiBO precursor was dissolved in 27.7 g of TEOS in a polypropylene bottle with a screw cap to form a homogeneous solution that was subsequently cooled to 273 K. Separately, a mixture of 28.7 g of TPAOH and 50.5 g of H 2 O was cooled to 273 K and was added slowly (over a period of ~1 min) to the TEOS solution, which yielded a biphasic mixture. This solution was warmed to 298 K and stirred for 12 h to produce a clear homogeneous solution, which indicates successful hydrolysis of the metal precursor and TEOS. The cover was then removed to completely evaporate the alcohol (e.g., ethanol, butanol) formed through hydrolysis of TEOS and the metal alkoxides. To ensure the complete evaporation of the alcohols, an additional 15 wt. % of the calculated mass of the alcohols was evaporated over the course of 24 -48 h and deionized H 2 O was added to make up for the excess liquid evaporated. This solution was then loaded into a Teflon-lined stainless-steel autoclave. While in the Teflon liner, a desired amount of HF (warning: HF is extremely dangerous and should be handled very carefully) was added to the synthesis gel and stirred manually with a polypropylene spatula for ~10 seconds prior to gelation. Notably, the addition of at least an equimolar amount of HF, with respect to TPAOH, increases the viscosity of the synthesis gel significantly. These steps yield a gel with the approximate composition of 1 Si : 0.0033 M : 0.43 TPAOH : a HF : 28.3 H 2 O, where a depends on the amount of HF added. A small amount of Ti-MFI seeds (5% by mass relative to SiO 2 ) from a previous synthesis in OH -media were added to promote the crystallization of MFI. These autoclaves were then heated to 443 K while rotating (30 rpm) in a convection oven for 3 -7 days. The resulting solids were recovered by centrifugation, washed with H 2 O, and dried for 16 h at 373 K. The dried solids were then heated in flowing air (100 cm 3 min -1 ) to 823 K at 1 K min -1 and held for 10 h to produce M-MFI materials that were bleached white in appearance. |
60c7525bf96a002e4a2881ce | 18 | Si-CDO was synthesized hydrothermally via the topotactic conversion of a pre-CDO layered silicate that to RUB-37 (i.e., Si-CDO). First, commercial DEDMA-OH solution was dehydrated until a concentration of 23.6% was reached. Then, 5.16 g of fumed silica was combined with 19.02 g of the preconcentrated DEDMA-OH solution. After 15 minutes stirring, a thick, white gel is yielded with a molar composition of 1 SiO 2 : 0.44 DEDMA-OH : 9.45 H 2 O. This gel was transferred into a teflon-lined autoclave, which was heated to 423 K in a convection oven for 13 days. The resulting solids were recovered by filtration, washed with H 2 O, and dried at 353 K for 12 h to yield 3.70 g RUB-36 (i.e., a pre-CDO silicate). RUB-36 was then heated in flowing air to 923 K at 2 K min -1 and held for 4 h to produce Si-CDO, that is bleached white in appearance. |
60c7525bf96a002e4a2881ce | 19 | Band edge energies (Table ) were measured by diffuse reflectance UV-vis spectroscopy (DRUV-vis). In short, samples were intimately ground with magnesium oxide (MgO; Sigma-Aldrich, 99.995%) in a 1:10 ratio of Ti-silicate to MgO by mass. These samples were loaded into a Harrick diffuse-reflectance accessory and spectra were obtained using a spectrophotometer (Agilent, CARY5) with pure MgO as the background. |
60c7525bf96a002e4a2881ce | 20 | Metal contents were quantified using energy dispersive X-ray fluorescence (EDXRF). Briefly, ~30 mg of the catalyst was loaded into a polypropylene sample holder (1 cm diameter) that was sealed with ultralene film. Samples were loaded into a spectrometer (Shimadzu, EDX-7000), whose sample chamber was purged with He (Airgas, Ultra-zero grade). Measurements were taken between 0 -30 keV (100 co-averaged scans), and the relative intensities of the element-specific fluorescence features were used to calculate the percent, by mass, of each element within the sample. |
60c7525bf96a002e4a2881ce | 21 | Attenuated total reflectance infrared (ATR-IR) spectroscopy was used to characterize the vibrational structure of H 2 O within the pores of zeolite samples with different densities of (SiOH) x . Zeolites (~30 mg) were first dispersed into CH 3 CN (~1 cm 3 ) via sonication and were subsequently drip coated onto a ZnSe cylindrical internal reflection element (IRE; International Crystal Labs). The catalyst-coated IRE was loaded into an ATR flow cell (Axiom, TNL-120), which was mounted onto a Fourier-transform infrared spectrometer (Bruker, Vertex 70). The cell was heated by a resistive heating cartridge placed within the wall of the cell, which was controlled by an electronic temperature controller (Watlow, EZ-Zone). Samples were first pretreated at 423 K (5 K min -1 ) for 2 h in flowing He (10 cm 3 min -1 ) to desorb CH 3 CN that was present from the catalyst deposition. Background spectra (2500 scans, 1 cm -1 resolution) were then taken at 303 K in He. H 2 O was subsequently introduced (1 cm -1 min -1 ) using a high-pressure piston pump (SSI, Series 1) and spectra (2500 scans, 1 cm -1 resolution) were acquired at a desired temperature (303 -343 K). |
60c7525bf96a002e4a2881ce | 22 | To quantify the spectral contributions of H 2 O within the pores of the zeolites, 2,4,6-tri-tert-butylpyridine (2,4,6-ttbupy; Sigma-Aldrich, 99%) was used as a probe molecule because it cannot fit into the 12membered ring pores of FAU or BEA. After the spectra of H 2 O were obtained on a given catalyst-coated crystal, a stream of He (10 cm 3 min -1 ) was flowed over the catalyst surface at 423 K (5 K min -1 ) for 2 h to remove all H 2 O. The IRE was subsequently cooled to 303 K and a spectrum of 2,4,6-ttbupy (~1 M) in nhexane (Sigma-Aldrich, 99%) was acquired. Comparisons of the recorded spectra over a zeolite-coated IRE to a clean IRE (Section S2.2) were made to distinguish the spectral contributions from within the zeolite pores from the bulk solution. |
60c7525bf96a002e4a2881ce | 23 | The three-dimensional coordinates of each zeolite framework were obtained from the International Zeolite Association Structure Database. It should be noted that the BEA structure was chosen to model Ti-BEA materials. One Si atom for every 5 unit cells was random selected and substituted for a Ti atom. An additional constraint for Ti placement in Ti-MFI was introduced such that Ti atoms were sited in either T3, T7, T8, T10, or T12 positions as previously characterized by Hijar et al. For modeling zeolite pores with silanol defects, approximately 5 (SiOH) 4 defects per unit cell were randomly placed under the condition that defects may not occupy adjacent tetrahedral sites. Approximately 15 unit cells were constructed and solvated with TIP5P water molecules. The silica forcefields derived by Emami et al. were used to parameterize the zeolite framework. Parameters involving Ti atoms were obtained from UFF. The tleap package of AmberTools was used to construct the MD systems. The system was first minimized for 5,000 steps using the steepest descent method, followed by 45,000 steps using a conjugate gradient method. The system was then heated from 0 K to 313 K at NPT conditions. Production simulations were implemented using the AMBER18 molecular dynamics package using the pmemd GPU accelerated module under constant NPT conditions, periodic boundary conditions, and a 2 femtosecond integration timestep. Temperature (313 K) and pressure (1 atm) were maintained using Berendsen thermostat and barostat, respectively. Bonds involving hydrogen atoms were constrained using the SHAKE algorithm. Electrostatic interactions were treated using the Partial Mesh Ewald method with a nonbonded distance cutoff set at 10 Å. Each system was simulated for 1 µs. Trajectory snapshots were saved every 100 ps during production simulations and were visualized using Visual Molecular Dynamics (VMD) software. The average number of hydrogen bonds formed per water molecule <N HB > MD was calculated as the total number of hydrogen bonds formed normalized by the total number of water molecules within a given zeolite framework. Hydrogen bonds were further differentiated based on the identity of the donor-acceptor pair. Hydrogen bonds that involved the surface silanol groups were classified as water-surface (W-S) hydrogen bonds, while hydrogen bonds involving other water molecules were labeled as water-water (W-W) hydrogen bonds. |
60c7525bf96a002e4a2881ce | 24 | Trajectories were processed using the CPPTRAJ module of AMBER18 and MDtraj Python library. Hydrogen bonds were identified based on a distance cutoff of 3.5 Å and angle cutoff of 120 degrees between the acceptor and donor atom. In-house scripts and the matplotlib Python library were used to generate water distribution plots. |
60c7525bf96a002e4a2881ce | 25 | Rates of alkene epoxidation and oxidant decomposition were measured using batch reactors (100 cm 3 , three-neck round-bottom flasks) equipped with reflux condensers to minimize evaporative losses. An alkene and H 2 O 2 were added to a solution of CH 3 CN and decane (used as an internal standard for GC analysis) and heated to the desired temperature (303 -348 K) while stirring at 700 rpm. The reactions were initiated by the addition of a catalyst, and small aliquots (~500 μL) of the reaction solution were extracted through a syringe filter (0.22 μm, polypropylene) at predetermined time intervals. The concentrations of the organic components within these aliquots were quantified via gas chromatography. All species were identified, and calibration factors were quantified using standards of known concentration. In all reported data, the standard uncertainty for measured reaction rates was below 10%. For all reactions, only the 1,2epoxyalkane products were observed, which reflects the intentionally low conversion over which these experiments were conducted. Rates for the conversion of alkene and oxidant were determined using the method of initial rates and were measured as functions of reactant concentrations. All reported results were obtained at differential conversion (i.e., <0.5% conversion of the limiting reagent); consequently, all reported rates do not reflect artifacts associated with catalyst deactivation. 1-Alkene epoxidation kinetics were measured in the absence of mass-transfer artifacts as indicated by turnover rates that depend linearly on the concentration of the alkene (data not shown). |
61a0855c78480507259b642f | 0 | Solar-driven water splitting utilizing photo(electro)catalytic systems is a possible strategy to secure the future supply of low-entropy energy in form of storable high-energy molecular fuels, such as hydrogen, alcohols or hydrocarbons. However, practically viable and sustainable systems required for large-scale applications should be preferably based on highly abundant and low-cost materials with non-critical availability, i.e., without any possible supply restrictions due to various political, economic or environmental concerns. In terms of sustainability, low cost and non-toxicity, one of the most attractive materials for photo(electro)catalytic applications are polymeric carbon nitrides (CNx). Carbon nitrides are readily available, tunable and chemically robust polymers, that have been widely utilized in a variety of light-driven chemical processes such as hydrogen evolution, CO2 reduction, selective chemical syntheses, or organic pollutants degradation. However, the studies demonstrating visible light-driven water oxidation to dioxygen using CNx-based photoelectrocatalytic systems are still rather rare. This is, on the one hand, due to the notoriously slow kinetics of the oxygen evolution reaction (OER), which typically results in accumulation of oxidizing equivalents that translates in enhanced photocorrosion of CNx. On the other hand, the very fabrication of mechanically robust CNx-based photoelectrodes is challenging, in particular due to the poor adhesion of CNx to conductive substrates, and the very low conductivity of CNx films, hindering efficient charge transport to the external circuit. In our own work, we have been developing a distinct type of hybrid CNx-based photoanodes in which a thin layer of CNx is deposited onto a porous TiO2 layer that acts as an electron-collecting scaffold, overcoming thus at the same time both the problems of low adhesion and low electronic conductivity of CNx. Such photoanodes are capable of visible light-driven oxygen evolution upon deposition of a suitable cocatalyst for water oxidation, typically in form of metal oxide (IrOx, CoOx, NiOx) nanoparticles. Notably, we found out that one of the crucial problems of the metal oxide nanoparticles as cocatalysts is their parasitic absorption of visible light, blocking thus the light absorption by the light absorber. This problem could be only partially overcome by using, for example, ultrasmall (1-2 nm) CoO(OH)x nanoparticles that exhibit a larger bandgap and correspondingly better transparency in the visible range due to quantum size effects. This led us to hypothesize that, in contrast to conventional bulk metal oxide-based water oxidation catalysts, a molecular-scale catalyst might be favorable for preventing the undesired light absorption by the catalyst and enabling also more controllable cocatalyst deposition. With this motivation in mind, we turned our attention to water oxidation catalysts based on well-defined molecular polyoxometalates (POMs), such as [Co4(H2O)2(PW9O34)2] 10─ = CoPOM, a tetra-cobalt-doped polyoxometalate, that have attracted much attention with regards to catalytic applications, and have been previously utilized as cocatalysts on various semiconducting metal oxides (e.g., TiO2, Fe2O3) to fabricate photoanodes for light-driven water splitting. However, to the best of our knowledge, no studies on CNx-based photoanodes comprising molecular POMs cocatalysts for photoelectrocatalytic water-splitting have been reported so far. |
61a0855c78480507259b642f | 1 | Herein, we report for the first time a triadic design of a photoanode consisting of porous TiO2 as an electron collector, CNx as a sensitizer for visible light, and CoPOM as a molecular water oxidation catalyst. The triad photoanode enables visible (λ > 420 nm) light-driven oxidation of water to dioxygen at moderate bias potentials (> 0.2 V vs. RHE). Notably, we show that effective immobilization of CoPOM into the porous structure of the photoanode plays a crucial role in photoanode performance and can be significantly improved using polyethyleneimine (PEI), a cationic polymer that can act as a non-sacrificial, electrostatic linker between the surface of CNx and the CoPOM that are both charged negatively (Scheme 1). The thus achieved optimized immobilization of CoPOM is demonstrated to result in a more efficient transfer of photogenerated holes to water molecules and enhanced oxygen evolution. |
61a0855c78480507259b642f | 2 | The CoPOM complex was synthesized according to literature. Briefly, Na2WO42H2O (50.89 g), NaHPO47H2O (4.60 g) and Co(NO3)26H2O (9.98 g) were dissolved in 50 mL deionized water in a 200 mL round-bottom flask. The pH was adjusted to 7 by HCl under magnetic stirring. The solution was then stirred and refluxed at 100 °C for 2 h and cooled down to room temperature. CoPOM was finally obtained by recrystallization and washed by deionized water. The purity of CoPOM is confirmed by Attenuated total reflection Fourier transform infrared (ATR-IR) spectroscopy. |
61a0855c78480507259b642f | 3 | Preparation of FTO/TiO2 substrates: TiO2 layers on FTO were prepared using an established doctorblading protocol. Briefly, 0.25 g TiO2 powder (Hombikat UV-100, pure anatase) was added to 1.25 mL anhydrous ethanol. The mixture was treated in ultrasonic bath for 10 min to produce a welldispersed suspension. The FTO glass substrates with a size of 1.5 × 2.5 cm were first cleaned by acetone for removing residual organic contaminants by ultrasonication for 20 min. The cleaned FTO glass was then etched in 0.1 M NaOH and rinsed with deionized water. Two FTO glass pieces were placed between microscope glasses and fixed using a 3M scotch tape as frame and spacer, leaving an exposed area of ~ 1.5 cm × 1.5 cm. Then, 200 μL TiO2 suspension was dropped on the microscope glass and gently swept by a glass stick onto the FTO glass pieces. After drying at 70 °C for 20 min, the TiO2 films were pressed at 10 4 N to improve the mechanical stability. All samples were calcined at 450 °C for 30 min in air before any tests or further treatments. The FTO/TiO2 substrates are abbreviated as TiO2 in this report. |
61a0855c78480507259b642f | 4 | CNx was deposited by chemical vapor deposition of urea pyrolysis products according to our previous report. Two pieces of TiO2 electrodes were placed in a Schlenk tube connected to a round-bottom flask containing 1 g of urea. Before the CNx deposition was started, the muffle oven (Carbolite, Germany) was preheated to 425 °C. Then, the reactor was directly placed into the muffle oven and heated at 425 °C for 30 min. Finally, the reactor was cooled down to room temperature in air. The resulting electrodes are denoted as CNx-TiO2. Material characterization: The electronic absorption spectra were measured using a UV-Vis spectrophotometer (UV-2600, Shimadzu, Japan) equipped with the integrating sphere and the absorptance (Abs.) was calculated by the equation: |
61a0855c78480507259b642f | 5 | The baselines were recorded using an FTO glass and a BaSO4 plate as references for transmittance and reflectance, respectively. Scanning Electron Microscopy (SEM) and Energy-dispersive X-ray spectroscopy (EDX) elemental mapping were performed using an NVision 40 (Zeiss Microscopy, Germany) scanning electron microscope equipped with an Octane Elite (EDAX, USA) EDX system. Photoluminescence (PL) spectra were recorded on an RF-6000 spectrofluorophotometer (Shimadzu, Japan) using excitation wavelength of 360 nm with a 400 nm cut-off filter placed in front of the emission detector. Attenuated total reflection Fourier transform infrared (ATR-IR) spectroscopy was performed by the FT-IR spectrometer (Alpha II, Bruker, Germany). X-ray photoelectron spectroscopy (XPS) measurements were performed with monochromatized Al Kα radiation using a PHI Quantera SXM system (ULVAC-PHI, Japan). The binding energies were calibrated based on C 1s peak of adventitious carbon (284.8 eV). |
61a0855c78480507259b642f | 6 | The photoelectrochemical measurements were conducted using a SP-300 BioLogic potentiostat and a typical 3-electrode system consisting of a Pt wire counter electrode, a Ag/AgCl (3.5 M KCl, 0.207 V vs. SHE) reference electrode and tested photoanodes as working electrodes with geometric irradiation area of 0.5 cm 2 . Photoanodes were irradiated by visible light (λ > 420 nm) using a 150 W Xe lamp (L.O.T.-Oriel) with light power density of ~ 150 mW cm -2 , equipped with a KG-3 (LOT-Quantum Design) heat-absorbing filter and a 420 nm longpass optical filter. |
61a0855c78480507259b642f | 7 | The oxygen evolution was recorded by FireSting optical fiber oxygen meter (PyroScience, GmbH) in a home-made air-tight two-compartment cell with the oxygen collection efficiency as approximately 75%, which was estimated by a direct electrolysis using a Pt working electrode. The volume of the photoanode compartment was 5 mL. The oxygen concentrations are not corrected for the losses in the gaseous headspace. The electrolyte was purged with argon before the electrodes were illuminated under applied potential of 1.12 V vs. RHE. The incident monochromatic photon-to-current conversion efficiency (IPCE) was recorded using a photoelectric spectrometer (Instytut Fotonowy Sp. z o.o.) equipped with a tunable monochromatic light source provided with a 150W Xenon lamp and a grating monochromator with a bandwidth of ~ 10 nm. The value of photocurrent density was the difference between current density under irradiation and in the dark in steady-state conditions with a wavelength sampling interval of 10 nm. The IPCE value for each wavelength was calculated according to equation: |
61a0855c78480507259b642f | 8 | where iph is the photocurrent density, h is Planck's constant, c is the velocity of light, P is the light power density, λ is the irradiation wavelength, and q is the elementary charge. The electrolyte for all photoelectrochemical measurements was 0.1 M sodium borate electrolyte with pH value of 8.0. Na2SO3 (0.1 M) was dissolved in the electrolyte when photocurrents were measured in the presence of sacrificial electron donor. All potentials are recalculated and reported vs. RHE. |
61a0855c78480507259b642f | 9 | The immobilization of the negatively charged [Co4(H2O)2(PW9O34)2] 10─ (CoPOM) water oxidation catalyst onto polymeric carbon nitride is challenging since the surface of carbon nitride is known to be negatively charged due to large amount of unprotonated surface Brønsted base moieties that can be protonated only by highly concentrated strong acids. In order to effectively immobilize the anionic CoPOM cocatalyst onto the internal surface of our porous CNx-TiO2 photoelectrodes, we have therefore utilized the layer-by-layer technique demonstrated by Jeon et al. for immobilization of CoPOM onto various metal oxides. The CNx-TiO2 electrodes carrying negative surface net charge were sequentially immersed into a solution of the cationic polyethyleneimine (PEI) and a solution of the anionic CoPOM for the desired number of times to fabricate the CoPOM-PEI-CNx-TiO2 photoanode. |
61a0855c78480507259b642f | 10 | It is known that PEI can be protonated in a wide pH range (pH 3-10) when dissolved in aqueous solutions. The positively charged cationic PEI thus plays a role of an electrostatic linker between CNx and CoPOM that are both charged negatively. For comparison, the CNx-TiO2 electrodes were also only dipped into the CoPOM solution resulting in the reference, linker-free electrodes CoPOM-CNx-TiO2. To evaluate and compare the CoPOM loading of the two CoPOM-containing CNx-TiO2 electrodes, energy dispersive X-ray (EDX) spectra were recorded. The CoPOM-PEI-CNx-TiO2 electrode shows significantly higher concentration of elements contained in CoPOM, with 0.35 at% Co, 0.13 at% P and 0.97 at% W, compared to 0.05 at% Co, 0.07 at% P and 0.03 at% W in case of the PEI-free CoPOM-CNx-TiO2 reference electrode (Supporting Information, Fig. ). The higher CoPOM loading of CoPOM-PEI-CNx-TiO2 was further corroborated by XPS, showing the increase of surface concentration of Co by the factor of 3.4 in photoelectrodes comprising the PEI linker (Supporting Information, Figure ). In addition, the EDX mapping (Supporting Information, Fig. ) depicts the homogeneous distribution of CoPOM within the porous CNx-TiO2 structure. Finally, only in case of CoPOM-PEI-CNx-TiO2, the characteristic IR fingerprint of CoPOM is detectable (Fig. ), which confirms improved CoPOM immobilization compared to electrodes without the cationic PEI linker. We conclude that the more effective immobilization of CoPOM in the CoPOM-PEI-CNx-TiO2 photoanodes is due to beneficial effect of the electrostatic attraction between the positively charged PEI linker and the negatively charged CNx and CoPOM components. |
61a0855c78480507259b642f | 11 | Figure depicts the UV-Vis electronic absorption spectra of TiO2, CNx-TiO2, CoPOM-CNx-TiO2 and CoPOM-PEI-CNx-TiO2. All CNx-containing electrodes exhibit a significant red shift compared to the optical absorption of pristine anatase TiO2 (3.2 eV, ~ 390 nm) and CNx (2.9 eV , ~ 428 nm), which we ascribe to effective sensitization of TiO2 by CNx, including formation of a charge-transfer complex between CNx and TiO2, making thus possible also the direct optical electron transfer from the HOMO of CNx to the conduction band of TiO2. The optical absorption edge of the hybrid photoanodes (~ 2.6 eV, ~ 477 nm) determined from the Tauc plots (Supporting Information, Fig. ) is larger than the value typically obtained in our previous studies (~ 2.3-2.5 eV), which can be explained by the inherent limitations of the Tauc formalism as applied for bandgap determination of hybrid materials, and to the fact that in previous studies we determined the bandgap using the Kubelka-Munk function calculated from diffuse reflectance spectra of corresponding powders, while here we use absorptance data obtained from measurements on complete photoanodes. Notably, the change in electronic absorption properties upon the deposition of the CoPOM catalyst is negligible, which indicates that the parasitic light absorption by the CoPOM catalyst is very low. This clearly highlights the advantage of using the molecular CoPOM as compared to, for example, cobalt oxide catalysts that typically show significant light absorption in the visible due to their fully developed band structure and correspondingly low bandgap, blocking thus partially the visible light absorption by the light absorber. Figure : ATR-FTIR spectra of all CNx-containing electrodes and CoPOM powder (a); UV-Vis electronic absorption spectra of the photoanodes; Abs. = absorptance (b). The non-zero baseline can be ascribed to the differences in internal reflection and scattering at the FTO/TiO2 interface in the transmittance and reflectance measurements modes. The photoelectrocatalytic properties of our photoanodes were investigated under visible light (λ > 420 nm) using an appropriate cutoff-filter, in order to effectively shut off the intrinsic UV light absorption of TiO2. The porous TiO2 layer thus serves solely as an electron-collecting scaffold that transports electrons injected under visible light irradiation from CNX into TiO2 to the underlying FTO glass support, whereas the oxidizing equivalents (i.e., photoholes) photogenerated in CNx should be ideally channeled to the water-oxidizing CoPOM catalyst to drive dioxygen evolution from water. First, potential-dependent photocurrents (Figure ) were measured for all electrodes utilizing illumination by chopped visible light (λ > 420 nm, 150 mW cm -2 ) in borate electrolyte (pH 8). As expected, no photocurrents could be detected at the pristine TiO2 substrate since anatase TiO2 cannot be excited by visible light. The photocurrents recorded for the CNx-TiO2 electrode without any CoPOM catalyst are attributed to the photocorrosion processes presumably at the CNx/TiO2 interface as no oxygen evolution was detected at this electrode under identical experimental conditions (see Figure ). In contrast, the photocurrent values significantly increased in the presence of both CNx sensitizer and CoPOM cocatalyst. Importantly, the polymeric PEI linker-containing CoPOM-PEI-CNx-TiO2 photoanode showed the highest photocurrents within the whole potential range and also the highest monochromatic quantum efficiencies (IPCE) measured at a constant bias potential of 1.12 V vs. RHE (Figure ). Importantly, the CoPOM-PEI-CNx-TiO2 photoanode exhibited also the most negative photocurrent onset potential of 0.2 V vs. RHE, which clearly indicates an improved rectifying behavior due to more effective extraction of photogenerated holes from CNx. Figure shows photocurrent transients under the same visible light irradiation conditions at a constant potential of 0.78 V vs. RHE, and indicates a relatively good short-term stability of the photocurrent response. Interestingly, for the CoPOM-free electrode (black line), the current spikes after switching on the light and negative current overshoots appearing after the light is switched off become significantly more pronounced. Such spikes and overshoots are a typical fingerprint of intense surface recombination processes, indicating that, in the absence of water oxidation catalyst, the photoholes in CNx do not undergo the desired interfacial transfer, but instead accumulate in the CNx layer and subsequently either recombine or induce photocorrosion. In contrast, the current spikes are less pronounced and the overshoots are nearly absent in both CoPOM-containing electrodes, which again indicates that the CoPOM cocatalyst can extract holes generated in the CNx layer and trigger the desired water oxidation reaction. |
61a0855c78480507259b642f | 12 | In order to shed more light on the factors governing the photoresponse of CoPOM-containing photoanodes, we performed an analysis of charge separation (ηsep) and hole transfer (ηtr) efficiencies according to established protocols. This analysis is based on the assumption that the measured photocurrent density in water oxidation can be calculated by multiplying the maximum possible photocurrent (obtained from the absorbed photon flux) by ηsep and ηtr, whereby the hole transfer efficiency ηtr in the presence of a readily oxidizable reducing agent (here Na2SO3) is taken as 100%; for details see the experimental section. The calculated ηtr and ηsep values for the both CoPOM-containing electrode are depicted in Figure , calculated from data in Figures and. As expected, both efficiencies show clear dependence on the applied potential as stronger positive applied bias is beneficial for both charge separation and hole transfer. Notably, the charge separation efficiency ηsep is very similar for both PEI linker-free CoPOM-CNx-TiO2 and PEI-containing CoPOM-PEI-CNx-TiO2, which is also in line with only minor differences in photoluminescence spectra of the corresponding photoanodes that are possibly related to slightly more efficient quenching of emissive states in CNx (Supporting Information, Figure ). In stark contrast, the CoPOM-PEI-CNx-TiO2 exhibits significantly higher charge transfer efficiencies than CoPOM-CNx-TiO2 in the whole potential range. In other words, this data suggests that the superior photoelectrocatalytic behavior of the PEI linker-containing CoPOM-PEI-CNx-TiO2 photoanode arises from a more efficient transfer of photogenerated holes from CNx to water, which can be attributed to the higher CoPOM loading due to the beneficial effect of the cationic PEI polymer linker. However, in addition to its beneficial effect on the CoPOM immobilization, a question arises whether the cationic PEI polymer could potentially also act as a sacrificial electron donor that can be simply more easily oxidized than water by holes from CNx. In order to address directly this issue, we measured the photocurrents from the CNx-TiO2 electrode modified with PEI polymer only (dipped in PEI solution 5 times), and the measurements at PEI-CNx-TiO2 photoanodes were repeated subsequently in four cycles. The deposition of PEI enhanced photocurrents, but a gradual decrease of photocurrents was observed, ending up at the same values as those for the CNx-TiO2 photoanode (Figure ). Hence, in the absence of the CoPOM catalyst, the PEI does extract effectively the holes from CNx, but is thereby oxidatively degraded. As a next step, the same protocol was also applied to the CoPOM-PEI-CNx-TiO2 photoanode. Contrary to the gradual decline of photocurrents observed for PEI-CNx-TiO2, the photocurrents at CoPOM-PEI-CNx-TiO2 remain completely stable over all four cycles (Figure ). This is also in line with the chronoamperometric photocurrent measurements of the three electrodes (Figure ), which show that PEI-CNx-TiO2 exhibits a fast decline of photocurrent due to degradation of PEI in the absence of CoPOM, whereby the photocurrent at CoPOM-PEI-CNx-TiO2 is practically stable. Therefore, we conclude that in the presence of CoPOM, the holes are efficiently transferred from PEI to the CoPOM catalyst where they drive water oxidation, and the cationic PEI linker is thereby effectively stabilized. In order to unambiguously prove the dioxygen evolution at CoPOM modified photoanodes, we performed photoelectrocatalytic OER measurements (Figure ) in a borate solution (pH 8) under prolonged (1 hour) visible light irradiation (λ > 420 nm, 150 mW/cm 2 ). Both CoPOM-containing hybrid photoanodes, CoPOM-CNx-TiO2 and CoPOM-PEI-CNx-TiO2, clearly exhibit OER activity under visible light illumination. This confirms our assumption that charge transfer from CNx to CoPOM is feasible and that the presence of the CoPOM cocatalyst is necessary to trigger the OER. Importantly, at the PEI-containing electrode the oxygen evolution rate was doubled compared to the counterpart photoanode without PEI. Importantly, no oxygen evolution was observed at the CoPOM-free CNx-TiO2 photoanode despite substantial photocurrents that can be ascribed to photocorrosion. This result is also in line with our previous studies that confirmed that the presence of an effective OER catalyst is absolutely necessary to observe oxygen as a product of water oxidation at CNx-TiO2 hybrid photoanodes. On the other hand, CNx-free pristine TiO2 photoanodes modified with CoPOM exhibited neither photocurrents nor oxygen evolution since pristine TiO2 does not absorb in the visible range. The apparent (based on dissolved O2 and uncorrected for losses in the headspace) Faradaic efficiencies (FE) of oxygen evolution for CoPOM-PEI-CNx-TiO2 (15% ± 4%) and for CoPOM-CNx-TiO2 (12% ± 4%) are rather low, which suggest that even in the best photoanodes the overall utilization of holes generated in CNx for water oxidation is still far from optimum, and a substantial portion of holes does not induce the OER but instead contributes to the photocorrosion of CNx. The improved photocurrent onset potential, higher oxygen production rate and FE at the PEI-containing CoPOM-PEI-CNx-TiO2 photoanode clearly confirm the beneficial effect of the cationic PEI polymer that serves as an effective linker by establishing the electrostatic attraction between the CNx sensitizer and CoPOM catalyst that are both negatively charged. This results in a more efficient hole extraction from CNx and more effective utilization of photogenerated holes for water oxidation due to the more effective immobilization (i.e., higher loading) of the CoPOM water oxidation catalyst. |
61a0855c78480507259b642f | 13 | A triad photoanode comprising a molecular cobalt polyoxometalate (CoPOM) embedded in the porous structure of hybrid photoanodes consisting of polymeric carbon nitride deposited onto an electron collecting porous TiO2 layer is reported for the first time. The photoanodes exhibit complete water oxidation to dioxygen under visible (λ > 420 nm) light irradiation, with photocurrents down to relatively low bias potentials of 0.2 V vs. RHE. Importantly, it is demonstrated that polyethyleneimine (PEI), a positively charged cationic polymer that has been previously reported to enable improved deposition of CoPOM onto various metal oxides, can also act as a highly effective electrostatic linker for immobilization of the anionic CoPOM onto the negatively charged surface of carbon nitride. |
61a0855c78480507259b642f | 14 | Mechanistic studies revealed that the optimized deposition of CoPOM using the PEI linker translates directly into improved efficiency of the transfer of photogenerated oxidizing equivalents (holes) to water molecules and thus to enhanced oxygen evolution. On the other hand, the charge separation efficiency in triad photoanodes was largely unaffected by the CoPOM loading, and remained rather low (below 10% at moderate bias potentials), suggesting that primary recombination is a key performance bottleneck in triad photoanodes. Importantly, we also show that the PEI linker is effectively stabilized in the presence of the CoPOM catalyst that efficiently extracts the holes from PEI, preventing thus its oxidative degradation that takes place in the absence of CoPOM. This work thus highlights the importance of careful design of multi-component photoelectrocatalytic systems, and provides a simple protocol for effective immobilization of POM-based catalysts into soft matter-based photoelectrocatalytic architectures for light-driven water oxidation. |
61326a27d5f080805bba0d77 | 0 | Hydrogels have become a very important class of materials, with the global sales reaches approximately 15-20 billion U.S. dollars annually. Since their first demonstrated application in 1960, hydrogels have found uses in many different areas such as contact lenses, wound dressings, cosmetics, drug delivery systems, tissue engineering, agriculture, and hygiene products. Hydrogels can be made of synthetic polymers such as polyethyleneglycol (PEG), natural polymers such as gelatin and collagen, or hybrids of natural and synthetic polymers. The choice of polymers and crosslinkers is crucial, as this affects the final properties of the hydrogel. Pore size, mechanical properties, swelling ratio, swelling rate, biodegradability, biocompatibility, chemical resistance, optical properties, and stimuli responsiveness are additional important hydrogel properties, which must be tailored according to the application. For example, a high swelling ratio and fast swelling kinetics are needed for superabsorbent applications. On the other hand, biodegradability and biocompatibility are more important for drug delivery or tissue engineering applications. Hydrophilicity is a vital property of all hydrogels, with hydrophilicity generally provided by hydrophilic polymers, such PEG or polyacrylamides. Although there are numerous reports on different types of hydrogels, novel formulations that can outperform commercial products are still needed. New hydrogel formulations with superior properties may not be enough for commercialization, as the scalability of the hydrogel synthesis, availability and cost of the starting materials are also critical factors for industrial production. In this paper, we describe a facile method to fabricate a novel hydrogesl based on urazole starting from cheap and readily available precursors. We also demonstrated the potential of urazole based hydrogels as ion-exchange materials. |
61326a27d5f080805bba0d77 | 1 | Urazoles are nitrogen-containing heterocycles that have been mostly used as precursors of triazolinediones (TADs). TADs have recently attracted attention owing to their click and clicklike reactions. Urazoles can readily be oxidized to TADs by strong oxidants. Alternatively, some urazoles can be aerobically oxidized to corresponding TADs by laccase enzyme. Urazole oxidation can also be achieved in situ using a ruthenium photocatalyst. In general, TAD or in situ oxidized urazole forms adducts that urazoles are bound to a material or a polymer via urazole's amide nitrogen. On the other hand, there are only a few reports that urazole is attached to material via its imide nitrogen. It is important to note that urazole group is untouched when attached via imide nitrogen. However, in these previous two reports, urazole oxidized to TAD and reacted further for other purposes. Therefore, urazole were not utilized as functional group. Urazole has rarely been thought as functional group due to TAD's highly successful click reactions. However, urazole protons are highly acidic and might be exploited to obtain functional materials having anionic character. The acidity of urazole was studied in details as proton-transfer reactions . In one study, the acidity of urazole was exploited to obtain novel polymers via N-alkylation. Some urazole derivatives, such as 4-oleyl urazole and 1,2 Diacetyl-4oleyl urazole, have been patented as additives for functional fluids. The potential of urazoles as analogs of prostaglandins in bronchodilation has been demonstrated. Interestingly, urazole was thought to be a precursor of uracil in the pre-RNA world, as it reacts with ribose to form four different ribosides. Recently, Yang et. al. used urazole to synthesize urazole-gold nanoparticles to determine curcumine using a fluorescent spectrometer. Our strategy to obtain a urazole-containing material involves embedding urazole's precursor (semicarbazide) in the polymer backbone rather than attaching urazole directly to a material. The latter strategy is challenging because of the amide protons of urazole. Semicarbazide was cyclized to urazole on the material in a solid state step under strong basic conditions . The resulting urazole-containing gel was a hydrogel, with 87% swelling ratio and 0.91 mPa elastic modulus. Interestingly, the polymer backbone of the material is highly hydrophobic, with the ionic salt character of urazole conferring hydrophilicity. We also demonstrated that a urazole-salt ionic pair can be exploited as an ion exchanger. When the hydrogel was placed in tap water, it successfully removed Ca+2 and Mg+2 from the water. |
61326a27d5f080805bba0d77 | 2 | All reagents were purchased from Sigma-Aldrich unless otherwise noted. THF was purchased as anhydrous and inhibitor free (>99.9). Poly(hexamethylene diisocyanate) -viscosity 1,300-2,200-was kept under nitrogen after opening the bottle. Methods Nuclear Magnetic Resonance (NMR) was done using a Varian 500 MHz spectrometer at ambient temperature. Attenuated total reflection (ATR) FT-IR spectra were recorded using a Perkin-Elmer Spectrum 100 in the region of 4000-650 cm -1 . Four scans were completed with a resolution of 2 cm -1 . A background measurement was performed prior to loading the sample onto the ATR for measurement. Raman spectroscopy (Renishaw in Via Reflex Raman Microscope and Spectrometer). Nd-YAG laser with a power of 0.5 mW at 532 nm was used for the data acquisition and the spectral range was kept from 300 cm -1 to 3000 cm -1 . Thermogravimetric analysis (TGA) was performed (Shimadzu, DTG-60H) from room temperature to 500°C with a heating rate of 10°C min -1 in a nitrogen atmosphere. Mechanical behaviour of the synthesized gel and hydrogel, a tensile test was performed. The test was performed using the Instron 5982 universal tensile system (Instron) with a 200 N static load cell and at tension rate of 5.0 %/ min up to failure of the samples under constant temperature and humidity conditions. Material tensile modulus (E) was determined from the linear portion of the stress-strain curve slope. Moreover, ultimate stress and fracture strain is also reported.Ion chromatography analysis were purchased as a service from water analysis laboratories of Istanbul Munipilacity Water and Sewege Administration (ISKI). Synthesis of semicarbazide gel: In a typical synthesis, 8.6 mL poly(hexamethylene diisocyanate) was added to an oven dried 100-mL round bottom flask together with 50 mL anhydrous tetrahydrofuran (THF) under inert conditions. This mixture was stirred for 10 minutes while keeping flask in an ice bath. 2.10gram ethyl carbazate was dissolved in 20 mL THF and added to the round bottom flask over 3-5 minutes in portions. Then, reaction was kept stirring for 60 more minutes. After, 100 µL water added and the solution was poured into glass petri dishes. Then, a watch glass was put over it for 2 days to let THF evaporate slowly. After, the watch glass was removed and all THF was evaporated for 2 more days. Finally, gel was peeled from the petri dish (ESI Figure ). |
61326a27d5f080805bba0d77 | 3 | In a typical synthesis, 10 grams of semicarbazide gels from previous step were cut into the pieces and placed into 100-mL round bottom flask. In method A, 3 grams of anhydrous K2CO3 and anhydrous ethanol was put and refluxed for 24 hours. After, gel was placed into beaker and kept in 100 mL water for 20 minutes. This process repeated for 4 times. Then, gel was dried under open air for 2-3 days. In method B, gels were placed into aqueous K2CO3 solution (10% w/w) and refluxed for 2 days. Finally, gels were dialyzed against DI water several times to remove K2CO3 completely. |
61326a27d5f080805bba0d77 | 4 | Potassium salt version of urazole gels were placed into DI water. Then, concentrated HCl is added until pH of solution becomes 1. Gels were kept at this pH for 4 hours and then dialyzed against DI water several times to remove HCl and KCl completely. Synthesis of poly(hexamethylene-semicarbazide): In a typical synthesis, 4.3 mL poly(hexamethylene diisocyanate) was added to semicarbazide solution (3.3 gram) in 100 mL anhydrous tetrahydrofuran (THF) under inert conditions. This mixture was stirred for 10 minutes while keeping flask in an ice bath. Then ice bath is removed, and reaction kept stirred for an hour at Scheme 1: Synthesis of semicarbazide gel (3), urazole potassium salt hydroge (4) and urazole gel as free acid (5). ambient temperature. White polymeric sticky material was precipitated in the reaction flask over time. Then-THF decanted and 100mL fresh THF was added and stirred for 30 minutes to remove excess semicarbazide. Finally, THF was decanted, and material dried under vacuum. Isolate yield: 7.43 gram 93%. Synthesis of poly(hexamethylene-urazole): 2 grams of poly(hexamethylene-semicarbazide) were added into oven dried 100 mL round bottom flask. Then, 1 gram of anhydrous K2CO3 and 60 mL anhydrous ethanol were added. Mixture was refluxed for 24 hours. Then, most of the ethanol was decanted carefully and the rest removed under vacuum. Sticky polymer -K2CO3 mixture dissolved in ice-water mixture. Then, 6M HCl added slowly, and white solid was precipitated immediately. More HCl added until pH of solution became 1-2. Product was collected by vacuum filtration and washed with excess DI water to remove acid. Sticky polymeric material was dissolved in CHCl3 and transferred into vial. Finally, chloroform removed under vacuum. Isolated yield: 1.61 grams. Synthesis of triazolinedione gel: In a typical synthesis, 200 mg of urazole gel placed into 3 mL dichloromethane. Then, 100 µL HNO3 was added, and mixture was shaken vigorously for 3-5 minutes. Finally, gels were filtered and washed with excess dichloromethane to remove residual HNO3. Quantification of Urazole Content: 150 mg of 2-naphthol dissolved in 7.5 mL CDCl3. To this solution, one drop of acetonitrile was added to use as integration reference. 1 mL aliquots of this solution were added onto three batches of the TAD-gel that was oxidized and washed 10 minutes ago. After 45 minutes, 0.7 mL of these solutions were taken for NMR analysis. 2-naphthol loss was determined by NMR integration after normalizing acetonitrile peaks in each spectrum. Determination of swelling ratio of hydrogels: Swollen hydrogels were frozen using liquid nitrogen and then freeze dried to complete dryness. Then, freeze dried gels were placed into DI water or phosphate buffer. Masses of wet hydrogel was recorded until it reaches equilibrium. Finallly, swelling ratio was calculated using the formula below. Swelling ratio = (Mass of wet gel -mass of freeze dried gel)/mass of freeze dried gel x 100 Ion-exchange study: 550mL of tap water was collected into a 1L Erlenmeyer flask. 50 mL of it was separated and kept as a control group. 50 grams of dry urazole-potassium gels were added to the tap water. After 2,8 and 120 hours, 10mL aliquots of water were taken for ion-chromtography analysis. Samples were filtered thorough 0.22um or 0.45um filter before analysis Cell Culture. Human skin fibroblast (HSF) cells (ATCC, PCS-201-030) were incubated at 37 °C under in humidified atmosphere air with 5% CO2 and cultured in Dulbecco's modified Eagle's medium (DMEM, Sigma, Germany) supplemented with 5% Penicillin-Streptomycin Ampicillin (PSA), 5% l-glutamine, and 10% fetal bovine serum (FBS) in tissue culture flasks (TPP, Germany). Cells were harvested Trypsin/D-Hanks solution (0.25 w/v %, Gibco) at reach 90% confluency then seeded at a density of 5 × 10 3 cells/well in a 96-well plate for incubation 24 h. Cell Viability Assay Cell viability analysis was measured using tetrazolium salt WST-1 colorimetric assay after treatment of hydrogel extract on seeded cells in a 96-well plate. Autoclaved samples were prepared 0.02 mg based on ISO 10993-12 protocol and incubated in 1 ml DMEM 72 h at 37 °C at 80 rpm. Therefore, seeded cells were cultivated as 100% and 50% extract concentration and negative and positive controls contained no material and 5% DMSO, respectively. After 2h, 6h, 24h, 48h, and 72h incubation, followed by the removed medium from each well was rinsed with PBS to clean the complete extraction residue. The treated cells were cultivated with a medium contained %10 WST-1 at 3h. Consequently the formazan was converted from tetrazolium salt in living cells, medium was measured absorbance at 450 nm with an ELISA reader. The cell viability percentage data calculated with normalized negative control. |
61326a27d5f080805bba0d77 | 5 | The synthesis of the urazole gel is shown in Scheme 1. In the first step, a multifunctional isocyanate, namely (1) poly(hexamethylene diisocyanate), was reacted with the equimolar ethyl carbazate to form an isocyanate semicarbazidecontaining prepolymer intermediate (2). In the same reaction flask, polymerization of this intermediate was initiated via the addition of a small amount of water. The polymerization of poly(urea) was also accompanied with gelation. This process is highly sensitive to moisture and heat. High temperatures or open air conditions results in bubbles in the gel (Electronic Supporting Information: ESI Figure ). However, bubble free gels can be obtained by carrying the gelation reaction at ambient temperatures and inert conditions.(ESI Figure ) The semicarbazide-containing gel (3) was cyclized into a urazole-potassium gel (4) under strong basic conditions at high temperature. Urazole cyclization is efficient and fast when Ethanol/K2CO3 is used. However, ethanol-water switch may damage hydrogels. Therefore, at the gel washing step, solvent exchange should be done carefully and slowly to prevent stress on the gels. Although aqueous K2CO3 cyclization is very slow and less efficient, the gels can be washed easily with water. Urazole potassium gel can be acidified to obtain a urazole gel in a free acid form (5). Interestingly, a urazole-potassium gel (4) can hold a large amount of water owing to its urazole-potassium salt groups. The gels synthesized were characterized by FT-IR and Raman spectroscopy. 1 H-nuclear magnetic resonance (NMR) analysis of urazole cyclization was done using a model compound, as the gels are crosslinked and not soluble. To obtain a soluble model compound, all isocyanates group of poly(hexamethylene diisocyanate) was reacted with ethyl carbazate to form poly(hexamethylene semicarbazide) (ESI, Scheme S1). The model compound is soluble in DMSO, and 1 H-NMR spectrum analysis showed that semicarbazide and urazole formation was successful (Fig. ). FT-IR spectrum of semicarbizde gel (3) and urazole gel (4) reveals that new carbonyl mode at 1767 cm -1 is appeared after urazole formation when gel is in free acid form (Fig. ). Additionally, carbonyl mode of semicarbazide at 1724 cm -1 is disappeared. These FT-IR findings are also in agreement with our previously synthesized urazole compounds in which their structures were also confirmed with 1 H-NMR. Urazole potassium salt has a unique mode at 1597cm -1 and carbonyl mode at 1767 cm -1 is also not present when analysis was done under dry and neat conditions. Urazole formation was determined qualitatively and quantitively by oxidizing them into triazolinediones (TAD). TAD formation can be monitored by naked eye owing to distinct red/pink color of TAD groups. The oxidation can be with strong oxidants such HNO3 or dibromohyndation in dichloromethane. In some cases, semicarbazide derivatives may form diethylazodicarboxylate (DEAD)-like structures upon oxidation, which have a similar color (orange to red) to TADs and similar reactivity to TADs. When we treated the semicarbazide gels (3) with strong oxidizers, including HNO3 and 1,3-Dibromo-5,5dimethylhydantoin, at least for 1 hour, we observed no color change in the material. This indicated that the red color of the gel was due to the urazole group instead of a possible DEAD like by-product. |
61326a27d5f080805bba0d77 | 6 | Quantification of urazole groups can be done by indirectly by determining the TAD groups on the gels. Titrating TAD groups against one of the reactants can be used for quantification. However, the titration method is prone to human error, as reactants take some time to diffuse deep into the gel. NMR analysis is a more reliable method to quantify TADs. A stock solution of a TAD reactive molecule can be reacted with a TAD gel to determine the TAD content of the gel. However, the target molecule should be chosen carefully. For example, furan has a low boiling point (30°C), and some of it could easily escape causing a positive error in TAD quantification. On the other hand, aniline gives 1:2 adducts with TAD that causes a negative error in quantification (ESI, Figure ,S4). Additionally, amine group of aniline can also decompose TAD over time that complicated analysis. to aniline, 1-naphthol also reacts with more than one TAD group. A reliable analysis could be done using 2-naphthol, as it a forms a 1:1 adduct with TAD (ESI Figure ). To determine the TAD content, three different batches of urazole potassium gels were oxidized using HNO3. |
61326a27d5f080805bba0d77 | 7 | Quantification was done by using 2-naphtol as the target molecule. 1 H-NMR results showed that an average of 0.71 mmol active TAD per gram of gel (Table ). It should be noted that some active TADs decomposed over time, without reacting with furan or 2-naphtol, which resulted in a smaller value than the actual value. Theoretically, a maximum of 1.65 mmol of TAD may be present if equimolar ethyl carbazate and triisocyanate are used at the beginning of the synthesis. The swelling ratio of the hydrogel was 87%. Scanning electron microscopy showed that the hydrogel has a porous structure like a sponge, with pore sizes between 12 and 30 µm (Fig. ). |
61326a27d5f080805bba0d77 | 8 | The porous structure and pore sizes of the gels can be tuned by changing the curing conditions, such as temperature, initiator amount, or stoichiometric ratio of isocyanate and ethyl carbazate. Generally, a porous structure is a desired property of hydrogels, as it increases the surface area and aids diffusion of molecules, such as drugs and biologics. We analyzed the thermal stability of the hydrogel using thermogravimetric analysis (TGA). The TGA analysis showed that the semicarbazide gel was thermally the least stable (Fig. ). It should be noted that in the urazole cyclization step, the high temperature probably helped unreacted isocyanate (i.e., isocyanate remaining in the gel network) to react to form extra bonds, resulting in higher thermal stability for urazole gels. The urazole free acid gel was slightly more stable than its potassium salt version below 300°C. This might be due to the hydrogen bonding capability of the urazole group when urazole is in free acid form. Under the same conditions, the urazole-potassium salt has a residual mass of 14.85% of its weight after heating to 700°C, whereas semicarbazide 5.01% and free acid has only 0.05%. The potassium mass contribution to the material is heoretically 12.90% that explains high residual mass for urazole potassium gel. |
61326a27d5f080805bba0d77 | 9 | Figure shows the tensile test results for the semicarbazide gel and urazole-potassium hydrogel. The test was conducted when the was in a swollen state. As can be seen, semicarbazide has a higher elastic modulus than the hydrogel, and it is also more brittle than the hydrogel. However, the hydrogel shows hyperplastic behavior, which is common for hydrogels. The results of each group are consistent with each other, which is a sign of good reproducibility and consistency of gel synthesis. The mechanical properties of the samples, such as Young's modulus and ultimate stress strain at break, are presented (Fig. ). The Young's modulus of the gel and hydrogel was 2.15 MPa and 0.91 MPa, respectively. For some application such as cartilage replacement, choice of hydrogels are limited because the modulus of most hydrogels is in the order of 0.1 MPa. Moreover, the gel and hydrogel ultimate stress is 0.499 MPa and 0.311 MPa, respectively and their strain at break is 25 % and 62.5% , respectively. It means that the hydrogel strain at break is almost 2.5 times the strain at break for the gel while its ultimate stress is 40% less than the gel. It means the although the hydrogel became ductile compared to gel but there is no drastic decrease in its strength. |
61326a27d5f080805bba0d77 | 10 | Figure shows the results of the WST-1 cell viability assay of HSF cells treated with solutions containing 0.032 mg/ml and 0.016 mg/ml of urazole for 2, 6, 24, 48, and 72 hours. As shown in Figure , the hydrogel did not affect the viability of the cells at lower concentrations but reduced cell viability in a dose-and time-dependent manner. Figure shows both while light (left) and confocal microscopy images (right) of the cells on the hydrogel containing 0.032 mg/ml of urazole after 72 h. As can be seen, there was no noticeable change in their morphology. |
61326a27d5f080805bba0d77 | 11 | Figure shows a comparison of while light images of the proliferation of the cells treated with the extraction solution at two concentrations for 72 hours. Consistent with the viability results, the hydrogel did not exhibit a significant toxic or affect cell viability and attachment. Overall, the data in the present study suggest that the prepared hydrogel can be considered a biocompatible material and that it should be further evaluated. We observed that urazole changing its potassium ions with ion in the medium over time. Possible mechanism is shown in scheme 2. It is expected that urazole's unique ionic character selectively binds to higher valent ions and releases its potassium ions. Ion-chromatography analyses shown that urazole potassium gel can remove Ca +2 and Mg +2 ions completely, Na + partially from the tap water in exchange with K + ions (Fig. ). This shows that urazole can be used to reduce water hardness. We also tested effect of urazole on water hardness with commercially available water hardness test kits. The results showed that urazole potassium gel (1% w/w), reduces the water hardness of tap water from 3.6°dH to 1.3°dH when gel was kept in water for 12 hours. When gels were kept for 36 hours, the water hardness reduced to 0.1 °dH for the same amount of the gels. |
62db30f0a7d17e34206889cc | 0 | Reversible phosphorylation is a common post-translational modification seen in over 30% of eukaryotic proteins. Phosphatases work in tandem with kinases to regulate this process in a broad spectrum of organisms. Among the protein phosphatase families, protein tyrosine phosphatases (PTPs) are defined by the active site signature motif called the P-loop (HCX5R) that includes a cysteine residue required for catalysis. Among the subclass of classical, pTyr-specific PTPs, protein tyrosine phosphatase 1B (PTP1B) and Yersinia outer protein H (YopH) are the most studied. PTP1B is a human PTP whose best-known biological role is that of a negative regulator in the insulin signaling pathway, while YopH is the virulence factor of Yersinia pestis, responsible for the bubonic plague. PTPs catalyze dephosphorylation through a two-step mechanism involving the nucleophilic cysteine in the P-loop. Another catalytic residue, an aspartic acid, resides on a different conserved structural loop found among the classical PTPs. This is a mobile loop consisting of about a dozen residues called the WPD-loop, defined by the conserved residues tryptophan, proline, and aspartate found near its center. While mobile loops in proteins are common, PTPs are unusual in having a key catalytic residue residing on such a mobile element. This loop remains mobile upon substrate binding, but favors a closed conformation approximately 8 Å closer to the P-loop, bringing the aspartic acid into position to protonate the leaving group in the first step of catalysis (Figure ). |
62db30f0a7d17e34206889cc | 1 | In this step the nucleophilic cysteine attacks the phosphate ester while the aspartic acid protonates the aryl leaving group. This step is followed by hydrolysis of the cysteinyl-phosphate intermediate, using a water molecule that is activated by the same aspartate, now acting as a general base; both steps require the WPD-loop to be closed. Positioning of the nucleophilic water is assisted by a glutamine residue on the Q-loop, another conserved protein element in classical PTPs. Another loop common to the classical PTPs is the E-loop, which contains a conserved glutamate residue that is usually found in a hydrogen bonding interaction with the conserved P-loop arginine, which provides hydrogen bonding interactions for substrate binding and transition state stabilization. the WPD-loop closes toward the P-loop. This brings a conserved aspartic acid into position to protonate the leaving group, followed by a subsequent rate-determining step where the same residue acts as a general base to activate a water molecule in the hydrolysis of the phosphocysteine intermediate. This figure is adapted from ref. . Copyright 2021 American Chemical Society. |
62db30f0a7d17e34206889cc | 2 | The rate-determining step in the PTP reaction is hydrolysis of the phosphoenzyme, making kcat independent of the original phosphoester substrate. Members of the PTP superfamily share the same catalytic residues, highly superimposable active site structures, and the same mechanism and transition state for both chemical steps. Yet, their rates vary over several orders of magnitude and exhibit different pH-rate dependencies. For example, at 25 °C, YopH and PTP1B exhibit kcat values for the substrate p-nitrophenyl phosphate (pNPP) of 720 s -1 , and 52 s -1 , respectively, at their pH optima and PTP1B has a significantly broader pH-rate profile than YopH. Other PTP superfamily members such as VHR, VHZ and SsoPTP have lower kcat values of 3 -4 s -1 at 25 °C and pH 5.5. If not active site or mechanistic differences, what, then, gives rise to the highly variable kinetics found among PTPs? In light of recent findings of significant protein dynamics in PTPs, it is likely that differences in loop dynamics play a major role. |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.