id
stringlengths 24
24
| idx
int64 0
402
| paragraph
stringlengths 106
17.2k
|
|---|---|---|
632e22362984c91d4d6aa766 | 8 | Sintering process provides a predominantly solid structure via mass transport reactions. In such an irreversible process, surface energy is consumed through particle bonding to lower the system's energy . The sinterability directly decreases with broader particle size distribution (PSD) , as well as presence of agglomerates . Thermodynamically, decrease in the average particle size results in increase in the specific surface area of a powder as well as increase in interfacial free energy density, ๐พ (๐ฝ. ๐ 5! ), higher reactivity of smaller particles, and consequently enhanced sinterability and densification behavior by decrease in porosity. It is noteworthy that the particles with narrower distribution exhibit a lower sintering rate prior to the occurrence of the grain growth but higher densification rate after the grain growth took place . The kinetics of ceramics sintering is depended significantly on the PSD in a way that using smaller particles along with narrower distribution greatly enhance the kinetics of the sintering reactions rate due to the increase in driving force of sintering process, and thermodynamically enhance tendency of the system to lower the chemical potential of the constituents and the Gibb's free energy of the system. |
632e22362984c91d4d6aa766 | 9 | The sintering process of ceramics consists of three major stages. In the initial stage, a strong bonding happened between the tangential particles to reduce the Helmholtz's free energy (A) of the surface atoms in constant volume (short time). In the intermediate stage, interparticle neck growth (long time) happens in order to decrease the system's Gibb's free energy (G) at constant pressure where most of the open pores that are connected to the external surface are annihilated. In the final stage the closed pores are eliminated and the monolith became hermetic and grain growth propagate at higher kinetic rates. Intermediate and final state have been vastly |
632e22362984c91d4d6aa766 | 10 | studied for shrinkage of the green compact by diffusion mechanisms thorough grain boundary and lattice pathways . Therefore, prior to sintering, to achieve a homogenous sintering process with a narrower PSD and eliminate the agglomerates, the as-received powder was wet ball-milled. Figure shows the particle size analysis (PSA) of the as-received and ball-milled 6Sc1CeSZ powders, using polarization intensity differential scattering technique (PIDS). The particle size distribution of both as-sintered and milled powder shows a bimodal gaussian distribution which the mean, median and mode of each individual powder is represented in figure (a). The PSD of the as-received powder consist of 4 peaks with the maximums located at 0.3 ยต๐, 10 ยต๐, 29 ยต๐ and 55 ยต๐, with more than 50% of the particles have a diameter less than 9.7 ยต๐. |
632e22362984c91d4d6aa766 | 11 | The 55 ยต๐ peak attributed to agglomerates of as-received powder which only consist of less than 2% of the total volume percentage. After the milling, the agglomeration peak completely disappeared and the mean particle size decreased by ~50%. However, a slightly more broaden peak, but with skewness closer to unity and with higher uniformity was achieved. The PSD of the ball-milled 6๐๐1๐ถ๐๐๐ shows two peaks with maxima at 300 ๐๐ and the second maxima at 3 ยต๐ with an overall particle range of 100 ๐๐ to 10 ยต๐. The inter quartile coefficient of skewness (IQCS) as a measure of degree of asymmetry of the particle size distribution was calculated from the equation below ] |
632e22362984c91d4d6aa766 | 12 | where ๐ !C , ๐ C$ , and ๐ 4C are lower quartile point, median, and upper quartile point. The IQCS of the as-received and milled PDS data is calculated to be 0.31 and 0.7, respectively. However, after the milling, a more homogenous symmetrical distribution is achieved by lowering the IQCS by more than 125%. where ๐ = and ๐ ; are thickness, and ๐ = and ๐ ; are radius of the green and sintered pellets. |
632e22362984c91d4d6aa766 | 13 | for sample sintered at 1000 ยฐ๐ถ and 1650 ยฐ๐ถ, respectively, where they differentiate by ~30%. The higher radial shrinkage of the samples was attributed to the pressure distribution under uniaxial press. Theoretically, a gaussian pressure distribution is generated under almost every uniaxial press. Therefore, always a stress gradient is produced inside the die press from the center to the pellet's rim. Samples sintered at 1000 ยฐ๐ถ, 1100 ยฐ๐ถ, and 1200 ยฐ๐ถ exhibited higher ๐
: ๐ด ratio (although smaller individual ๐ด% and ๐
%) while the samples sintered at higher temperatures showed a barely perceptible positive slope where ๐
: ๐ด ratio ~1.2. At this point, one can assume that 1200 ยฐ๐ถ could be consider as the temperature, sufficiently high that grain growth kinetic rate is fast enough /activated to eliminate open pores to the surface and make the pellet heterotic. including samples sintered at high temperatures showed some residual porosities. These trapped pores were mostly observed at the 3and 4-grain-boundaries junctions which specifically were responsible for stopping the densification process and reaching the plateau. However, samples sintered in low-temperature regime, showed very low density, with open pores compared to other ceramic oxides such as YSZ. This effect is more tangle when the low density of the samples sintered at 1000 ยฐ๐ถ to 1200 ยฐ๐ถ is concerned where the process was controlled by sluggish diffusion of cations at lower temperatures. Politova et al. discussed that the ๐๐ ! ๐ " -๐๐๐ ! system is prone to lower achievable density. This phenomenon occurs owing to the slower rate of cation diffusion in this system compared to that of anion transport. However, addition of ๐ถ๐ "%,&% enhanced the mobility by isomorphous substitution of the host lattice, ๐๐๐ ! . The enhancement of conductivity was attributed to the lattice parameter changes in the host unit cell where larger cation ๐ถ๐ &% (๐ -. "# = 0.97 ร
) randomly substitute ๐๐ &% (๐ () "# = 0.84 ร
) sites at the eight folded coordination (8๐ถ๐) . The powder particles have a solid-vapor surface energy (๐ฝ. ๐ 5! ). The energy per unit mass can be stored as excess surface area (๐ฝ. ๐๐ 51 ). However it is not thermodynamically possible to consume all the excess surface energy during the sintering, therefore the emergence of the other interfaces add energy to the system such as grain boundaries that cause the densification plateau and the higher relative density couldn't be achieved . Figure shows the effect of the sintering temperature on the density, grain size, and pore size of the as-sintered pellets. The density graph shows a sigmoidal dependence of the density on sintering temperature with a reached plateau around 1500 ยฐ๐ถ, indicating the saturation limit of the densification process. The densification plateau occurrence depends on several factors such as particle size distribution of the powder particle packing, sintering heating rate (here 2 ยฐ๐ถ. ๐๐๐ 51 ), and the sintering atmosphere (air) . |
632e22362984c91d4d6aa766 | 14 | Thermodynamically as-received powder particles have a solid-vapor surface energy (๐ฝ. ๐ 5! ). The energy per unit mass can be stored as excess surface area (๐ฝ. ๐๐ 51 ). However it is not thermodynamically possible to consume all the excess surface energy during the sintering, therefore the emergence of the other interfaces add energy to the system such as grain boundaries that cause the densification plateau and the higher relative density couldn't be achieved and the energy is consumed to decrease the surface energy of the grains by consuming the high energy stored at the grain boundaries by coarsening the microstructure. |
632e22362984c91d4d6aa766 | 15 | It can be observed from figure that upon increasing the sintering temperature from 1000 ยฐ๐ถ to 1400 ยฐ๐ถ apparent density increased. Conventionally sintered pellets at 1500 ยฐ๐ถ and 1600 ยฐ๐ถ showed the highest densification followed by a plateau . However, some ceramics oxides showed a reverse densification behavior which after a point density decreases due to partial formation of liquid phase . However, sintering at ๐ โฅ 1500 ยฐ๐ถ resulted in excessive grain The obtained powder X-ray diffraction (XRD) patterns of the as-sintered pellets at 1000 ยฐ๐ถ to 1650 ยฐ๐ถ for 6 hours were collected at room temperature and presented in a comparative stack plot in figure . The crystal structure and lattice constant were determined from the XRD pattern using the Rietveld method implemented with GSAS-๐ช๐ช software . The average crystallite size was determined from the (111) reflection of the fluorite structure using the Scherrer equation ๐ = ๐
๐ ๐ฝ. ๐๐๐ ๐ where ๐ is the mean size of the crystalline domains, ๐
is shape factor and dimensionless (typically 0.94 ), ๐ is the wavelength of radiation, ๐ฝ is the full width (line broadening) at half maximum intensity (FWHM) in radians, and ๐ is the Bragg angle in radians. It is noteworthy that the shape factor of the XRD peak profiles could be fit by either Gaussian (for rounded tops typically due to strain broadening) or Lorentzian (for sharper tops due to size distribution and dislocations) distributions. The peak shapes of the experimental data Typically follow Voigt distribution which is a convolution of Gaussian and Lorentzian peak shapes where both contributions are equally weighted which resulted to shape factor 0.94 instead of Scherrer's original shape factor value of 0.88 or 0.9. The calculated crystalline size and the lattice parameters can be found in table I. The calculated crystallite size is observed to improve with sintering temperature for all samples from 28 nm to 89 for sintering at 1000 and 1650 respectively for 2 hours. ) which ๐กฮ and ๐กฮฮ can be formed due to the presence of distorted oxygen sublattice in the cubic (fluorite) environment . The XRD data suggested that depending on sintering temperature and time, three phases of cubic, tetragonal and monoclinic were detected. The XRD pattern suggested formation and coexistence of the cubic and tetragonal structure for samples sintered at ๐ > 1200 ยฐ๐ถ. However, evidence of formation monoclinic phase next to cubic and tetragonal structures were observed in the sample sintered at ๐ < 1200 ยฐ๐ถ. The samples sintered at 1000 ยฐ๐ถ and 1100 ยฐ๐ถ contained a distinguished (11 ยฉ 1) peak of monoclinic at ~28ยฐ as given in figure . The volume fraction of the monoclinic phase was calculated to be in the range of ๐ G < 7%. Generally the monoclinic phase formation in zirconia systems depends on the primary sample preparation methods, the sintering conditions, and even following mechanical treatments such as surface polishing/grinding . Therefore, to minimize the effect of the powder XRD samples' preparations, the XRD was carried out on the as-sintered pellets placed on a p-type B-doped silicon zero diffraction plate, and no grinding process had been done. |
632e22362984c91d4d6aa766 | 16 | They also showed that crushing a sample prior to XRD data acquisition increase the monoclinic content by more than 10 ๐ฃ๐๐%. Hence, in this study, a lower cooling rate was applied to minimize the above-mentioned effect. It was noticeable that XRD data showed a reciprocal however small relationship of intensity of (111) peak with increase in sintering temperature for samples sintered at 1000 ยฐ๐ถ, 1100 ยฐ๐ถ, and 1200 ยฐ๐ถ. This change can be attributed to possible convolution of intensity of the (111) peak reflected from all three phases of cubic, tetragonal, and monoclinic at 2๐(ยฐ) equal to 30ยฐ, 29.8ยฐ, and 31ยฐ, respectively. |
632e22362984c91d4d6aa766 | 17 | Figure shows the phase percentage of the detected cubic, tetragonal, monoclinic phases. The samples sintered at 1000 ยฐ๐ถ and 1100 ยฐ๐ถ showed a relatively small weight fraction of monoclinic phase which decreased by increase in sintering temperature and didn't appear in other XRD patterns. Samples sintered at ๐ > 1200 ยฐ๐ถ showed the only two phases of cubic and tetragonal. The weight percentage of the tetragonal decreased by increase in the temperature from ๐ฅ H โ 80 ๐ค๐ก% at 1100 ยฐ๐ถ to ๐ฅ H โ 30 ๐ค๐ก%. This phenomenon implied that increase in sintering temperature is beneficial to maintain the highly-conductive cubic and tetragonal phases of 6Sc1CeZr and avoid formation of monoclinic and (less probable) rhombohedral phases. Raman spectra of the sintered samples from 1000 ยฐ๐ถ to 1650 ยฐ๐ถ with the corresponding phases labeled using square (cubic), circle (tetragonal), and triangle (monoclinic) markers. The ๐ โ ๐ก phase transition happens by oxygen ion displacement from fluorite ideal sites (8C sites) . The X-ray atomic scattering of oxygen is fundamentally smaller than that of ๐๐, ๐ถ๐ and ๐๐ to detect tetragonal regions . Therefore, obtaining Raman data is strongly recommended as a complimentary technique indenting to phase analysis of ๐๐๐ ! -based solid solutions. The Raman spectra of investigated areas readily showed the polymorphism of zirconia with the presence of cubic, tetragonal, and monoclinic phases. The tetragonal phase was detected to be the prevailing phase in all samples. Thus, Raman spectra of the investigated samples is in a good agreement with the obtained XRD data. The primary Raman shift peaks of the tetragonal phase occurred at 265, 331, 469, and 640 ๐๐ 51 and that of the cubic phase was detected at 640 ๐๐ 51 which are in a good agreement with literatures . The monoclinic peaks were detected at 362, 415, 538, and 678 ๐๐ 51 which is in a good agreement with literatures . In samples sintered at The obtained Raman modes didn't considerably differentiate in the location and width of the tetragonal peaks by increase in the sintering temperature. However, the primary cubic peak of the first order active Raman mode at 640 ๐๐ 51 (๐ : = 1000 ยฐ๐ถ) changed to 642 ๐๐ 51 (๐ : = indicator peak. The slight decrease in the average bond length of the ๐๐ 6% โฏ ๐ 65 should cause the increase in the bond strength and consequently shift the Raman modes toward higher vibrational frequencies. The shift can be also related to a mass effect of the constitutional cations on the vibrational modes in binary and ternary systems. In scandia-ceria-zirconia ternary system, ๐๐ (44.955 ๐ข) is more than three times heavier than ๐ถ๐ (140.116 ๐ข). Therefore, the small shift in the modes should be influenced by small dopant amount of 1๐๐๐% ๐ถ๐. Doping ๐๐๐ ! with ๐ถ๐๐ ! shifted the Raman bands related to the stretching modes of ๐๐ -๐ (at ~640 ๐๐ 51 ) shift toward lower wavenumbers implying that the average cation-oxygen bond distances are lengthened as ๐ถ๐ &% content increases. However, addition of ๐๐ ! ๐ " shifted the ๐๐ -๐ (at ~640 ๐๐ 51 ) to higher wavelengths. Since the ๐๐ content is higher than ๐ถ๐, a gradual shift of the main cubic peak to the higher frequencies was observed. Kim et al. observed that the degree of tetragonality of the cell ( ๐ ๐ โ ratio) plays a crucial role in the phase transition in zirconia systems in a way that ๐๐๐ ! solid solution stability increases by a decrease in tetragonality toward unity and vice versa. It has been well-discussed that a significant inductive load effect resulted from the leads and instrument induction always presents in high-frequency data points . The main transport properties of the ionic conductors are represented in the high-frequency regime of the EIS data. Therefore, it is crucial to subtract the inductive load effect from the obtained data to eliminate the instrumentation and leads resistance as Bauerle in his well-known paper discussed in details . This inductance appears as an imaginary part of the impedance data in the positive direction of the imaginary part of the impedance axis (direction), as shown in figure . It is noteworthy that, unlike the samples' behavior, there is not a significant dependence of inductance on temperature. At low temperatures, the effect of the inductance is not effective although it is present, but in high temperatures, the inductance effect is obvious as the highest frequency data points lying below the x-axis. The fitting in this manuscript is restricted to the high-frequency regime of the impedance spectra after subtracting the lead inductance using resistors, capacitors, and constant phase element (CPE). after subtraction of the induction load acquired at 400 ยฐ๐ถ and 800 ยฐ๐ถ, respectively. As can be seen, the inductive load effect at higher temperatures significantly contribute to the sample resistance, contrary to low temperatures. The insets show the fit and R-C circuit used for the fit. As can be seen, the data before subtraction of inductance, need to be fit using a resistor and a CPE element which deviate from an ideal capacitor. However, after removing inductance load, a good fit is achieved just using resistor and capacitor. From the EIS data it was observed that the grain, grain boundary, and total resistivity have analogous behavior. This high electrical conductivity is attributed to the high achieved relative density and most minuscule contribution of grain boundary resistivity to the total resistivity as sintering temperature increased. However, the bulk resistivity is not significantly dependent on grain size and sintering conditions, while two semicircles were observed in the low-temperature regime of the impedance plots. The high-frequency arc corresponds to the grain (bulk) resistance, and the intermediate frequency semicircle corresponds to the grain boundary resistance component. Solid electrolytes such as scandia-ceria-doped zirconia have a negligible electronic conductivity. Although mathematically the electronic conductivity cannot be identically equal to zero, owing to local equilibrium thermodynamic where |
632e22362984c91d4d6aa766 | 18 | However electronic conductivity in fact is as small as 10 51C S. cm 51 .Therefore it can be assumed that the increase in total conductivity is generally attributed to the increase in ionic conductivity due to decrease in interior grain conductivity. Figure shows the dependent of the activation energy and the conductivity pre-exponential factor to the microstructure. In this study, activation energy and pre-exponential factor showed a reciprocal relation with the grain size where finer grains exhibited higher activation energy than coarse grains. analysis. The sample sintered at 1650 ยฐC, while achieved the densification plateau, has the largest grain size, however, it is probable that at ๐ > 1600 ยฐ๐ถ, grain coarsening followed by grain boundary broadening as well as pore size growth. Thus, migrating ions at the grain boundary pathways had been scattered and causing reduction in the measured total conductivity. |
632e22362984c91d4d6aa766 | 19 | The effect of sintering temperature on the relative density, grain size, pore size, phase content balance, and electrical conductivity of the 6 ๐๐๐% ๐๐ ! ๐ " -1 ๐๐๐% ๐ถ๐๐ ! ๐๐ -๐๐๐๐๐ ๐๐๐ ! has been investigated. The samples' density reached the density plateau at 1500 ยฐC while at T>1500 ยฐC densification process stopped and coarsening happened. Then XRD and Raman spectroscopy revealed the presence of minor monoclinic phase for samples sintered at ๐ < 1200 ยฐ๐ถ which decreases with increase in ๐ : . The presence of the cubic phase increased by increase in the sintering temperature. The EIS spectra of the fine-and coarse-grained structures were investigated using circuits consist of R-C and R-CPE elements. The effect of the inductive load has been discussed in interpretation of the Nyquist spectra at intermediate and high temperatures. The total electrical conductivity was measured and presented in an Arrhenius plot. |
650cbd0f60c37f4f763792ef | 0 | Amyloid fibrils are filamentous, ฮฒ -sheet-rich protein aggregates that are implicated in numerous diseases, including Alzheimer's, Parkinson's, Huntington's, and Type II Diabetes. Because of their insoluble and non-crystalline nature, conventional structural characterization methods such as X-ray crystallography and solution-state nuclear magnetic resonance (NMR) spectroscopy cannot be used to determine the molecular structure of fibrils. Fortunately, advances in solid-state NMR (ssNMR) and cryogenic electron microscopy (cryo-EM) have enabled the determination of high-resolution structures of amyloid fibrils prepared in vitro, as well as those harvested from the tissues of patients. These studies indicate that fibrils share a common conformational motif known as a "cross-ฮฒ" structure, in which extended ฮฒ -sheets stack together with their strands aligned perpendicularly to the fibril's long axis. They also demonstrate that fibrils exhibit polymorphism, where a given peptide is capable of forming a variety of distinct fibril structures. Despite the wealth of high-resolution structural information gleaned from ssNMR and cryo-EM, these techniques are not without their pitfalls. Solid-state NMR, for example, requires high sample loads (> 10 mg), expensive isotopic labeling schemes, and long spectral acquisition times that can take several days. To obtain high-resolution structural data, both ssNMR and cryo-EM require relatively extensive sample preparation to ensure that fibrils are both homogeneous and well-ordered. This can introduce bias in determining the molecular structure of amyloids by preferentially selecting for the most abundant fibril polymorphs. In addition, it can be difficult to resolve electron densities in cryo-EM or obtain good chemical shift dispersion in ssNMR for all but the most well-ordered regions of fibrils. Disordered, dynamic, or structurally heterogeneous regions, which could play impor-tant roles in initiating aggregation, sequestering other proteins into amyloid plaques, or aberrantly interacting with biological cells, can be difficult to study with these techniques. |
650cbd0f60c37f4f763792ef | 1 | In contrast, vibrational methods such as Raman spectroscopy do not suffer from these pitfalls. Raman spectroscopy can be used to interrogate a wide variety of samples with little preparation, including fibrils in solution, gels, and fibril films. Data acquisition is relatively fast and robust. In addition, Raman spectroscopy can be used to quantify the distributions of peptide bond and side chain dihedral angles in amyloid fibrils. Similarly, polarized Raman measurements can be used to determine the relative orientation of chemical bonds and functional groups in fibrils. Thus, the conformational sensitivity of Raman spectral features enables the facile differentiation of fibril polymorphs, as well as the ability to robustly monitor the structural evolution of oligomeric precursors that aggregate into fibrils. However, despite its versatility and structural sensitivity, Raman spectroscopy is generally considered to be a "lowresolution" characterization method in the broader amyloid community. One reason for this is because Raman spectroscopy has generally only been used to qualitatively evaluate fibril secondary structures rather than determining threedimensional models like ssNMR. We believe, however, that structural parameters measured using Raman spectroscopy (vide supra) can in fact be harnessed to determine detailed molecular-level structural models of amyloid fibrils. We recognize that this can be accomplished by taking inspiration from ssNMR, in which experimentally measured distances and dihedral angles are used as constraints in energy minimization procedures performed on structural models of amyloid fibrils using molecular dynamics (MD) simulations. To test this idea, we investigated the molecular structures of three fibril polymorphs prepared from amylin 20-29 and amyloid-ฮฒ 25-35 (Aฮฒ 25-35 ) peptides. Amylin 20-29 , de-rives from residues 20 -29 of the 37-amino acid amylin peptide, which forms fibrils implicated in the pathology of Type II Diabetes. . Amyloid-ฮฒ 25-35 (Aฮฒ 25-35 ) derives from residues 25 -35 of the 40-42 residue long Aฮฒ peptide. Fibrils formed from Aฮฒ compose extracellular plaques that have been implicated in the pathology associated with Alzheimer's disease and Cerebral Amyloid Angiopathy. |
650cbd0f60c37f4f763792ef | 2 | Amylin 20-29 (sequence SNNFGAILSS) and Aฮฒ 25-35 (sequence GSNKGAIIGLM) were purchased from GL Biochem (Shanghai, China) at 91% purity and used without further purification. Dimethylsulfoxide (DMSO) was purchased at โฅ99.8% purity from Supelco. Milli-Q grade water (18.2 Mฮฉ cm) was obtained from a Milli-Q ยฎ IQ 7000 Ultrapure Lab Water System from Millipore Sigma. Phosphate buffer saline tablets were purchased from Sigma Aldrich. Phosphate buffer solution was made from sodium phosphate monobasic monohydrate and sodium phosphate dibasic anhydrous purchased from Fisher Scientific. Filters (0.45 ยตm pore-size) were purchased from Fisher Scientific. Mica (Product No. 50, V1 Grade) and silicon substrates (Product No. 21610 -55) were purchased from Ted Pella Inc. An ultra-sharp commercial Atomic Force Microscopy (AFM) probe (160AC, OPUS by MikroMach) and a standard AFM probe (PPP-NCHR, Nanosensorsโข) were both purchased from NanoAndMore USA Corporation. |
650cbd0f60c37f4f763792ef | 3 | Two different fibril polymorphs were prepared from the amylin 20-29 peptide using a modified procedure based on that developed by Madine et al. For both polymorphs, 2 mg of peptide were first disaggregated in 20 ยตL of DMSO after incubation at room temperature (22 ยฐC) for 1 h. To prepare the antiparallel ฮฒ-sheet polymorph (polymorph 1), 980 ยตL of water was slowly added, followed by an additional 1 mL of sodium phosphate buffer (20 mM, pH 7.13). The final solution was 2 mL with a final DMSO concentration of 1.41 ร 10 -4 mM, 10 mM of buffer, and a peptide concentration of 1.13 mM. To prepare the parallel ฮฒ-sheet polymorph (polymorph 2), 1980 ยตL of sodium phosphate buffer (100 mM, pH 7.13) was slowly added to the sample. The final solution was 2 mL with a final DMSO concentration of 1.41 ร 10 -4 mM, 100 mM of buffer, and a peptide concentration of 1.01 mM. The final step for both polymorphs involved filtering solutions using a 0.45 ยตm pore-sized filter to remove any undissolved large aggregates. All starting solutions were visually clear and both samples were incubated at room temperature (22 ยฐC). Aggregates could be visually observed after 7 days of incubation for polymorph 1 and 9 days for polymorph 2. |
650cbd0f60c37f4f763792ef | 4 | Parallel ฮฒ-sheet fibrils were also prepared from Aฮฒ 25-35 peptide. This was accomplished by dissolving 2.12 mg of peptide in 1.06 mL of water. The resulting peptide solution appeared visually clear. After this, 1.06 mL of 2ร phosphate buffer saline (PBS, pH 7.4) was slowly added to the sample for a final peptide concentration of 1 mM. The solution was gently mixed by carefully inverting the sample vial 3 times. Aggregates could be visually observed shortly after preparing the solution and was incubated at 37 ยฐC for 3 days. |
650cbd0f60c37f4f763792ef | 5 | Following incubation, fibrils were harvested from each sample by centrifuging 500 ยตL of solutions for 1 h at 21300รg (Eppendorf Centrifuge 5425). The supernatant for each sample was carefully decanted so that the pellet remained. The pellet was washed twice to remove residual salt crystals by sequentially resuspending it in water and centrifuging. Following this, the pellet was resuspended in 250 ยตL of water. Aliquots of each sample were then diluted, deposited onto a silicon substrate, and allowed to dry in a dust free environment to create a coffee ring for polarized Raman measurements. |
650cbd0f60c37f4f763792ef | 6 | Raman spectra were measured using a Horiba LabRAM HR Evolution Raman microscope (Horiba Scientific) using 633 nm excitation from a Helium-Neon laser. The laser light was focused on the sample using an infinity-corrected achromatic 100ร objective (0.9 NA, MPLN100X, Olympus) with the average power ranging from to 2.5 mW to 17.9 mW. Acquisition times ranged between 30 -360 s per spectrum. Spectral acquisition parameters were carefully chosen to balance maximizing signal-to-noise and mitigating photodegradation of samples. Under the illumination conditions used, we observed no visual signs or spectral signatures of sample degradation (Figure ). The scattered light was collected by the focusing objective in a 180ยฐbackscattering geometry. The scattered light was focused into a spectrometer and dispersed using a 600 gr/mm grating. Spectra were imaged using a Synapse EM CCD camera (1600X200-FV, Horiba Scientific). |
650cbd0f60c37f4f763792ef | 7 | For polarization measurements, fibrils were aligned so that their long axis was parallel to the polarization of the laser light, which we defined as the laboratory coordinate's Z direction. The lab frame's Y coordinate was defined to be along the direction of the laser light's propagation. A half-wave plate was used to rotate the polarization of the incident light along the X direction, but was removed for incident light polarization measurements involving the Z direction. As described in detail by this was done to reduce artifacts introduced by the half-wave plate. A polarizer was used to select either the Z or X polarization component of the scattered light, which was then focused into a spectrometer and dispersed using a 600 gr/mm grating. An optical scrambler was installed before the spectrometer entrance to depolarize the scattered light to minimize the polarization bias of the grating. Polarized Raman measurements were made for the following four incident and scattered light configurations: ZZ, ZX, XZ, and XX (incident and scattered light, respectively). The instrument was benchmarked with Raman polarization measurements of cyclohexane to ensure that our measurements were accurate. We found that the depolarization ratios measured for cyclohexane on our instrument were within experimental error of values reported previously in the literature. E. Atomic Force Microscopy (AFM) Samples were prepared for AFM imaging using a procedure that was modified from Ostapchenko et al. Briefly, a mica disc was exfoliated to create an atomically flat and clean surface. A 10 ยตL aliquot (2.2 mg/mL) of fibril solution was then deposited onto the disc. The sample was incubated on the disc for 10 min in a dust free environment before being washed 3ร with water. The sample was then wicked dry with filter paper and dried overnight prior to imaging in a dust free environment. |
650cbd0f60c37f4f763792ef | 8 | The AFM measurements were performed on an Asylum MFP-3D-BIO AFM instrument (Oxford Instruments) in the AC mode. Amylin 20-29 polymorph 1 (Figure ) was imaged using an ultra-sharp commercial probe with a 26 N m -1 force constant and 300 kHz resonance frequency. Amylin 20-29 polymorph 2 (Figure ) and Aฮฒ 25-35 fibrils (Figure ) were imaged using a standard AFM probe with a 42 N m -1 force constant and 330 kHz resonance frequency. |
650cbd0f60c37f4f763792ef | 9 | The images were analyzed with Gwyddion software (opensource software for Scanning Probe Microscopy Data) and the height traces obtained from Gwyddion were visualized as graphs in Graphpad Prism 9. The average diameter for each polymorph was calculated as the maxima of the height trace across 19 or 23 individual fibril fragments (identified visually). See SI for additional details and images (Figure ). |
650cbd0f60c37f4f763792ef | 10 | For the MD simulations, two 24-mer models of amylin 20-29 protofibrils and one 24-mer model of an Aฮฒ 25-35 protofibril were constructed using VMD. For each model, a single ฮฒsheet containing 12 peptides was first built and then duplicated and stacked on the initial ฮฒ-sheet. The putative fibril structures were constructed based on Raman data. For amylin 20-29 , both an antiparallel and parallel ฮฒ-sheet fibril model were constructed, while only a parallel ฮฒ-sheet fibril was constructed for Aฮฒ 25-35 . The ฮฑ-carbons of the two stacked ฮฒ-sheets were initially separated by 10 ร
for each model. Rather than solvate these models with explicit solvent, we opted to use a generalized Born implicit solvent (GBIS) protocol with a dielectric constant of 3.23, mimicking that of the interior of proteins, and an ion concentration consistent with experimental sample preparation. Implicit solvent enabled us to better approximate the environment of an effectively infinite fibril. These models consist of 3336 and 3720 atoms for amylin 20-29 and Aฮฒ 25-35 fibrils, respectively. |
650cbd0f60c37f4f763792ef | 11 | During the minimization and equilibration simulations, structural constraints obtained from the Raman experimental data were applied to the 24-mer fibril models using NAMD's Collective Variable functionality. In particular, the ฯ-angles of the peptides were harmonically restrained about the ฯangle peaks we observe experimentally (150ยฐamylin 20-29 polymorph 1 and 139ยฐfor amylin 20-29 polymorph 2 and Aฮฒ 25-35 ). The angle between peptide carbonyl bonds and the fibril long axis was harmonically constrained to the values obtained experimentally. Additionally, the constraint force constants were tuned to reproduce the experimentally obtained ฯangle and C--O bond angle distributions. The fibril axes were approximated in these models by using VMD to calculate the inertial tensor of the ฮฑ-carbons and extracting the principal components. The fibril long axes were taken as the resulting principal component perpendicular to the peptides. They were calculated separately for every frame of the simulation trajectories, as atomic fluctuations in the fibrils cause small changes to the axis vectors. |
650cbd0f60c37f4f763792ef | 12 | The NAMD MD package was used to energy minimize and simulate the fibril models. The all-atom CHARMM36m force field was used to calculate potential energies and forces due to its improved treatment of secondary structure compared to CHARMM36. All simulations were performed under a constant temperature and pressure of 298 K and 1 atm, respectively. The Verlet velocity integration algorithm was used with a time step of 1 fs, and the SHAKE algorithm was employed to constrain heavy atom-hydrogen covalent bonds. Non-bonded interactions were calculated for atom pairs using a cutoff of 12 ร
, and a switching function was used at distances greater than 10 ร
to truncate the potential. The particle mesh Ewald method was used to calculate long range electrostatics. Visual and quantitative analysis of MD simulations was performed with Amber's cpptraj tool and VMD. The 24-mer fibril models were first energy minimized for 5000 steps with rigid backbone atoms using the conjugate gradient minimization scheme. Following this, we released the rigid backbone atoms and equilibrated the three fibril models for 2 ns with the aforementioned ฯ-and C--O bond angle constraints for each polymorph taken from the Raman data. Following these constrained equilibration simulations, we released all the constraints and ran another 2 ns to observe the thermodynamic stability of the 24-mer fibrils. We collected ฯand C--O bond angles, as well as inter-strand and inter-sheet distances from the constrained and unconstrained simulations. |
650cbd0f60c37f4f763792ef | 13 | We chose to study fibrils prepared from the amylin 20-29 and Aฮฒ 25-35 peptides because of their small, tractable nature, which makes them ideal for quantitatively analyzing Ra- man spectra without the need for complex isotopic labeling schemes and performing detailed MD simulations. Studies suggest that residues 20 -29 form the amyloidogenic core of amylin. Fibrils prepared from amylin 20-29 can be poised to adopt well-defined morphologies, but there have been conflicting reports on whether it forms parallel, antiparallel, or mixed ฮฒ-sheet fibril structures. Wildtype Aฮฒ consists of two predominate isoforms, the 40-residue Aฮฒ 1-40 and the 42-residue Aฮฒ 1-42 . Although it is a less abundant isoform, Aฮฒ 25-35 has also been found in the brain and has been shown to aggregate into ฮฒ-sheet structures. Despite this, we are unaware of any known structures reported for Aฮฒ 25-35 fibrils. |
650cbd0f60c37f4f763792ef | 14 | AFM imaging (Figure ) reveals that the aggregated samples prepared from amylin 20-29 and Aฮฒ 25-35 peptides resulted in the formation of amyloid fibrils. The two polymorphs produced from amylin 20-29 exhibit similar average diameters of (9.1ยฑ2.9) nm for polymorph 1 (Figure ) and (9.7ยฑ4.4) nm for polymorph 2 (Figure ). They also show similar morphologies consisting of unbranched fibrils that are several microns in length. In contrast, the fibrils prepared from Aฮฒ 25-35 are shorter, being only several hundred nanometers to a few microns in length, and exhibit an average diameter of (5.7 ยฑ 1.8) nm (Figure ). |
650cbd0f60c37f4f763792ef | 15 | To further characterize the fibrils, we used Raman spectroscopy to investigate their molecular structures. As shown in Figure , each polymorph exhibits unique spectral fingerprints in the region between 1100 -1700 cm -1 . The most structurally informative Raman bands in the spectra occur in the Amide I, II, and III regions, located between 1600 -1700 cm -1 , 1500 -1600 cm -1 , and 1200 -1350 cm -1 , respectively. We examined these regions in detail to obtain structural information about the fibrils by performing spectral deconvolution analysis (see SI for details, Figures and). The results of this analysis are summarized in Table , which lists the Raman band assignments for the three polymorphs. |
650cbd0f60c37f4f763792ef | 16 | The most intense features in the spectra shown in Figure occur in the Amide I region. The Amide I vibration consists mainly of amide C--O stretching. . Its structural sensitivity derives from transition dipole coupling between neighboring C--O oscillators that produce a delocalized Amide I normal mode. Coupling also results in characteristic "excitonic splitting" patterns in the Amide I band that are diagnostic of different protein secondary structure elements. For example, the Amide I band for canonical (infinitely long) parallel ฮฒ-sheets is predicted to split into two sub-bands, a high frequency A(0, 0) mode and a low frequency B(ฯ, 0) mode that are both Raman and IR-active. In contrast, the Amide I band is predicted to split into four sub-bands for canonical (infinitely long) antiparallel ฮฒ-sheet structures: the A(0, 0) mode, which is Raman-active and forbidden in the IR; the B 1 (0, ฯ) mode, which is relatively weak in both Raman and IR; the B 2 (ฯ, 0) mode, which is very strong in IR spectra; and the B 3 (ฯ, ฯ), which is essentially forbidden in IR and Raman spectra. In practice, however, it can be difficult to differentiate between parallel and antiparallel ฮฒ-sheet structures using the Amide I band alone. This is because the Amide I band depends on additional factors such as hydrogen bonding patterns of the peptide bond C--O groups, the registry of ฮฒ -strands, and the twist of the ฮฒ -sheets. The intense band at ca. 1670 cm -1 can be assigned to the Amide I A(0, 0) for all polymorphs. The B(ฯ, 0) mode occurs near 1630 cm -1 for both amylin 20-29 polymorphs and is downshifted to ca. 1620 cm -1 in Aฮฒ 25-35 fibrils. We do not observe Raman bands in any of the fibril spectra that can be assigned to the B 1 (0, ฯ) mode characteristic of canonical antiparallel ฮฒ-sheet structures. However, for amylin 20-29 polymorph 1, we observe a band at 1695 cm -1 in the IR spectrum (Figure ), which we assign to the B 1 (0, ฯ) mode. The presence of this band suggests that these fibrils are composed of antiparallel ฮฒ-sheet structures (vide infra). We attribute the absence of the B 1 (0, ฯ) mode in the Raman spectrum to the fact that there is likely some local disordering of the ฮฒ-strand registries in the amylin 20-29 polymorph 1 fibrils. The amylin 20-29 fibrils contain additional bands in the ca. 1600 -1700 cm -1 region. Both polymorphs exhibit bands at ca. 1605 cm -1 , while polymorph 2 exhibits an additional band at 1586 cm -1 . These bands can be assigned to the in-plane ring stretching modes of phenylalanine. Amylin 20-29 fibrils are also expected to contain spectral contributions from asparagine's side chain C--O stretching mode. We see no evidence that the asparagine residues of amylin 20-29 fibrils contribute significant spectral contributions to the 1600 -1700 cm -1 region. However, to confirm that our assignment of 1671 cm -1 bands in amylin 20-29 fibrils is correct, we measured the Raman spectrum of a lyophilized powder of asparagine. The Raman spectrum of asparagine (Figure ) exhibits a strong band at 1640 cm -1 , which can be assigned to the C--O stretching mode of asparagine side chains. This suggests that the spectral contributions of asparagine side chains is negligible for the 1671 cm -1 bands that we assign to the Amide I A(0, 0) of the peptide backbone amide groups. |
650cbd0f60c37f4f763792ef | 17 | The Amide II mode gives rise to a band located in the 1500 -1600 cm -1 region. It consists of an out-of-phase combination of NH in-plane bending and CN stretching. The Amide II is typically strong in IR spectra, but weak in Raman spectra excited with visible wavelengths. Interestingly, Lee and coworkers suggest that the Amide II is enhanced in the Raman spectra of parallel ฮฒ-sheet structures. As seen in Figure , the Amide II is suppressed in the spectrum of amylin 20-29 polymorph 1 (Figure ), but appears in the spectra of the other two fibril polymorphs (Figure ). This suggests that amylin 20-29 polymorph 2 and Aฮฒ 25-35 fibrils consist of parallel ฮฒ-sheet structures. |
650cbd0f60c37f4f763792ef | 18 | The most structurally-sensitive bands in the Raman spectra shown in Figure occur between ca. 1200 -1350 cm -1 in the Amide III region. The canonical Amide III band, as characterized in N-methylacetamide, occurs at ca. 1315 cm -1 and originates from a vibration consisting of an in-phase combination of NH in-plane bending and CN stretching. In peptides and proteins, however, the Amide III region is considerably more complex, consisting of several bands that derive from vibrations containing significant contributions of CN stretching, NH bending, and/or C ฮฑ H bending motions. Asher and coworkers, have assigned the Amide III region of peptides and proteins in detail, identifying three sub-bands called the Amide III 1 , Amide III 2 , and Amide III 3 . They have shown that the Amide III 3 is the most conformationally sensitive, as its frequency can be correlated to the Ramachandran ฯdihedral angles of peptide bonds. For ฮฒ-sheet structures, the Amide III 3 band occurs between ca. 1220 -1240 cm -1 , and it is easily identified due to its relatively strong intensity (even with visible Raman excitation) compared to other bands in the region. Based on peak intensities from our spectral deconvolution analysis (Figures and), we assign the 1236 cm -1 band for amylin 20-29 polymorph 1, the 1224 cm -1 band for amylin 20-29 polymorph 2, and the 1225 cm -1 band for Aฮฒ 25-35 fibrils to Amide III 3 modes that are diagnostic of amyloid fibril ฮฒ-sheet structures. We capitalized on the structural sensitivity of the Amide III 3 band to determine the distribution of Ramachandran ฯangles for the amylin 20-29 and Aฮฒ 25-35 fibril peptide bonds. To do this, we utilized the methodology of Asher and coworkers, which correlates the frequencies of the Amide III 3 band envelope to different ฯ-angles (see SI for details). As shown in Figure , the ฯ-angles for all the fibril polymorphs occur between ca. 120 -160ยฐ. The distribution for amylin 20-29 polymorph 1 (Figure ) is centered at 151ยฐ, well within the range of canonical antiparallel ฮฒ-sheet structures. In contrast, the distributions for the other polymorphs are centered at ca. 140ยฐ(Figure ), indicating that they are composed of parallel ฮฒ-sheet structures. We observe no ฯ-angles in the fibrils that suggest there are ฮฒ-turn structures. This indicates that all three polymorphs consist of extended ฮฒ-strands that assemble into ฮฒ-sheets. All spectra were smoothed with a Savitzky-Golay Filter 66 using a 4 th order polynomial over an 11-point window for visual clarity. The original spectra can be found in Figure . |
650cbd0f60c37f4f763792ef | 19 | We used polarized Raman spectroscopy to gain additional structural insights on amylin 20-29 and Aฮฒ 25-35 fibrils. Raman anisotropy measurements on aligned samples can be employed to determine the relative orientation of chemical bonds and functional groups in fibrils, as first demonstrated by Lednev and coworkers. We prepared anisotropic samples of aligned fibrils for each polymorph using a drop coat deposition method. This method creates a "coffee ring" peptide film that forms because the evaporation of the solvent produces shear forces that move fibrils in the center of the droplet towards the perimeter. As the fibrils approach the perimeter of the droplet, they align themselves parallel to its edge with a high degree of ordering. We aligned the incident laser light to the edge of the coffee ring for the polarized Raman measurements. We designated a laboratory coordinate system (XYZ) where the Z-direction corresponds to the long axis of the fibrils (tangent to the coffee ring edge) and the Y-axis to the propagation direction of the incident laser light (Figure inset). Given the uniaxial symmetry of the fibrils, we only needed to acquire Raman spectra using four different combinations of incident and scattered light polarization configurations to obtain orientation information of chemical functional groups: ZZ, ZX, XZ, and XX (where the first and second letter indicate the direction of the incident and scattered light, respectively, along the laboratory coordinate system). The polarized Raman spectra of the three different fibril polymorphs are shown in Figure . As expected, the crosspolarized spectra (ZX and XZ, shown in blue and purple, respectively) overlap almost perfectly, indicating that there is no photodamage of the samples, no artifacts introduced by the different configurations of the polarization optics used (see Experimental Methods for more details), and that no displacements/rotations occurred in the samples during the measurement. The ZZ (black) and XX (red) spectra show the largest intensity variations. The most striking difference occurs with the Amide I A(0, 0) band, which is most intense in the ZZ spectrum and very weak in the XX spectrum. |
650cbd0f60c37f4f763792ef | 20 | Quantitatively interpreting the polarized spectra to extract orientation information requires knowing the Raman tensors for particular vibrational normal modes. Fortunately, the Raman tensor for the (delocalized) Amide I A(0, 0) mode has been determined. It is oriented parallel to the C--O axis of peptide bonds in ฮฒ-sheet structures. Thus, the Amide I A(0, 0) band is an ideal spectroscopic marker to determine the relative orientation of C--O bonds in amylin 20-29 and Aฮฒ 25-35 fibrils. |
650cbd0f60c37f4f763792ef | 21 | The theory to obtain orientation information from polar-ized Raman spectra has been described in extensive detail elsewhere. The key step is to determine the respective most probable orientation distribution functions, N mp (ฮธ ), for the the Amide I A(0, 0) tensor of each polymorph. The method of finding N mp (ฮธ ) is described in the section Basic Approach to Determine the Most Probable Orientation Distribution Function of the SI. Briefly, for systems such as fibrils with uniaxial symmetry, N mp (ฮธ ) can be estimated with reasonable accuracy by experimentally measuring two order parameters called โจP 2 โฉ and โจP 4 โฉ. As discussed in detail in the SI, these order parameters can be determined from the intensity ratios R 1 = I ZX /I ZZ and R 2 = I XZ /I XX for the Amide I A(0, 0) mode (see Table and S2 and Figures ). Figure -c shows N mp (ฮธ ) for the Amide I A(0, 0) tensors. The distributions are normalized such that multiplying N mp (ฮธ ) by sin ฮธ yields the preferred orientation distribution of the tensors. As shown in Figure -g, preferred orientation distributions of the Amide I A(0, 0) tensors for all the polymorphs are bimodal. The maximum probability for the distributions corresponding to the Amide I A(0, 0) tensors occur at ca. ยฑ10ยฐand ยฑ15ยฐfor amylin 20-29 polymorphs 1 and 2, respectively, and ยฑ11ยฐfor Aฮฒ 25-35 fibrils (Table ). These values indicate that the ฮฒ-strands of the polymorphs are oriented approximately perpendicularly to the fibril long axis, which is the hallmark of the cross-ฮฒ architecture observed in amyloids. The smaller peaks in the C--O bond angle distributions that occur at 90ยฐcan be attributed to local disordering of the cross-ฮฒ structure or because of misalignment of the fibrils filaments in the coffee rings that we prepared. Based on the relative intensities of the peaks, disordering and misalignment accounts for roughly 19% for both amylin 20-29 polymorphs and 33% for Aฮฒ 25-35 fibrils of the total probability for the C--O bond angle distributions. |
650cbd0f60c37f4f763792ef | 22 | where p(ฯ i )/p(ฯ eq ) is the ratio of peptide bonds with ฯangles angles, ฯ i and ฯ eq , respectively. The angle, ฯ eq , is the "equilibrium" or minimum energy ฯ-angle angle for reach polymorph, while R is the molar gas constant, T is the temperature, and โG(ฯ i ) = G(ฯ i ) -G(ฯ eq ) is the apparent Gibbs free energy difference between ฯ i and ฯ eq . From eq. 1, โG(ฯ i ), can be determined: Eq. 2 can be used to determine an apparent free energy landscape for the fibril peptide bonds along the ฯ-angle structure coordinate. As shown in Figure , the apparent energy landscapes behave harmonically around ฯ eq and can be modeled in terms of a simple torsional spring using Hooke's Law: |
650cbd0f60c37f4f763792ef | 23 | where ฮบ is the torsional spring force constant, which reflects the curvature of the harmonic potential wells. A similar approach can also be used to determine the torsional spring constants for the C--O bond angles. These distributions are more complex, and, as shown in Figure , the potential energy landscapes are anharmonic. To calculate the force constants, however, we assume that the potential wells behave harmonically in the region between 5 -20ยฐ, near the equilibrium bond angles. |
650cbd0f60c37f4f763792ef | 24 | Table summarizes the force constants obtained from the experimentally measured ฯ-and C--O angle distributions. It is important to note that fibril misalignment introduces additional uncertainty in determining the force constants for the C--O bond angles, but not the Ramachandran ฯ-angles. The effect of this greater uncertainty is that the apparent force constants determined experimentally for the C--O bond angles are lower than in actuality. This means that the structural con-straints applied in the MD simulations are relaxed and allows a greater ensemble of fibril structures to be sampled in the trajectories, effectively reducing the apparent resolution of our structural models. |
650cbd0f60c37f4f763792ef | 25 | We recognized that the C--O and ฯ-angle distributions obtained from Raman measurements provide structural constraints that can be used to determine molecular structural models of the amylin 20-29 and Aฮฒ 25-35 fibril polymorphs. To accomplish this, we first manually constructed putative structures of the three different fibril polymorphs for MD simulations (see Computational Methods for details). Then, using an analogous approach to ssNMR, we performed an energy minimization procedure using MD wherein we restrained the C--O and ฯ-angles using the equilibrium angles and torsional force constants determined experimentally from Raman measurements (vide supra). |
650cbd0f60c37f4f763792ef | 26 | Figure presents representative snapshots from the resulting ensemble of fibril structures for the three polymorphs from these simulations. All three polymorphs are composed of extended ฮฒ-strands that assemble into sheets. The strands for amylin 20 while those of polymorph 2 and Aฮฒ 25-35 form parallel ฮฒsheet structures. For amylin 20-29 polymorph 1, hydrophobic contacts occur between the side chains of Phe, Ala, and Ile residues within the same ฮฒ-sheet, while the side chains of Leu, Ala, and Ile form hydrophobic zippers between ฮฒsheets (Figure ). In contrast, the side chains of Phe, Ala, Ile, and Leu residues are aligned down the length of the ฮฒ-sheets for polymorph 2, forming a string of hydrophobic interactions (and ฯ-stacking interactions for Phe) between their neigboring residues, while the side chains of Ile, Leu, and Ala residues from opposite ฮฒ-sheets additionally form hydrophobic zippers, similar to polymorph 1 (Figure and). For Aฮฒ 25-35 , hydrophobic interactions occur between the side chains of Ala an Ile residues within the same ฮฒ-sheet, as well as Leu and Ile residues from opposite ฮฒ-sheets (Figure ). In addition, hydrogen bonding interactions are observed between Ser, Asn, and Lys residues between neighboring ฮฒ-strands (Figure ). All three polymorphs obtained from the MD simulation exhibit structural features that are reasonable for amyloid fibrils. The average inter-sheet distances for the three structures are between 9.4 -10.4ร
, and the average inter-strand spacings are between 4.9 -5.1ร
(Table ). These values are in good agreement with the equatorial and meridional spacings observed in experimental fiber diffraction data of amyloid fibrils. Similarly, no unrealistic bond angle distortions or steric clashes are observed, and the (ฯ, ฯ, ฯ) Ramachandran angles are all within allowed values (Table -S6). |
650cbd0f60c37f4f763792ef | 27 | To further validate the models, we examined whether the simulated fibrils were structurally stable. We tested this by removing the structural constraints on the fibril structures and then running additional simulations. The unconstrained structures held together during the simulation production runs and did not dissociate. The C--O bond distributions (Figure ) of the unconstrained structures are bimodal, exhibiting peaks around ยฑ10-15ยฐfrom the fibril axes, in good agreement with our experimental data. Similarly, the median values for the ฯangle distributions (Figure ) are close to our experimental measurements. The distribution of amylin 20-29 polymorph 1 is unimodal and peaked near 147ยฐ. In contrast, the distributions for amylin 20-29 polymorph 2 and Aฮฒ 25-35 appear bimodal, exhibiting a main peak around 139ยฐ(close to experimental measurements) and a smaller peak that is downshifted to ca. 120ยฐ. This indicates that the fibril structures corresponding to amylin 20-29 polymorph 2 and Aฮฒ 25-35 have both slightly evolved compared to the original constrained structures. The reason for these discrepancies is likely due to the limitations of our simulations since we could only model a single protofilament segment. We hypothesize that the structures we simulated are more conformationally flexible (thus allowing the ฯ-angle distributions to evolve) since they are not subjected to lattice packing forces of multiple protofilaments that stabilize real amyloid fibrils. |
650cbd0f60c37f4f763792ef | 28 | Early studies utilizing ssNMR and IR spectroscopy by Landsbury and Griffin suggested that amylin 20-29 fibrils are composed of antiparallel ฮฒ-sheet structures. These finding were later corroborated by Nielsen and coworkers who also employed ssNMR to investigate the structure of amylin 20-29 fibrils. The Landsbury-Griffin-Nielsen studies, however, have been contradicted by others, who suggest that amylin 20-29 fibrils can also form parallel ฮฒ-sheet structures. Middleton and coworkers employed ssNMR and X-ray fiber diffraction to investigate amylin 20-29 fibrils. Their analysis of C cross-polarization MAS and RR ssNMR data revealed that amylin 20-29 can form both parallel and antiparallel ฮฒ-sheet fibril polymorphs, despite electron micrographs of their fibril samples appearing morphologically homogeneous. Similarly, Song et al. published a recent cryo-EM study, which also suggests that amylin 20-29 fibrils can adopt parallel ฮฒ-sheet structures. |
650cbd0f60c37f4f763792ef | 29 | Our work corroborates the findings of Middleton and coworkers that amylin 20-29 forms both parallel and antiparallel ฮฒ-sheet fibril polymorphs. Furthermore, the structures that we determine for both polymorphs exhibit hydrophobic zippers between the side chains of Leu, Ala, and Ile that are consistent with ssNMR data obtained by both the Middleton and Nielsen groups . In addition, the ฮฒ-sheets of both polymorphs also exhibit a slight twist around their fibril axes, in agreement with the antiparallel ฮฒ-sheet structure reported by Nielsen and coworkers . This agreement between our structural models and those derived from ssNMR data highlight the utility and complementary nature of using Raman spectroscopy to quantitatively assess the molecular structures of different amyloid fibrils. |
650cbd0f60c37f4f763792ef | 30 | Regarding Aฮฒ 25-35 , we are unaware of any detailed fibril structures being reported in the literature. In the case of the full-length (1 -40 or 1 -42) wildtype Aฮฒ peptide, residues 25 -29 adopt a bend structure that brings the two ฮฒ-sheets of the fibril protofilaments together. In contrast, Aฮฒ 25-35 is too small to adopt bend or turn structures. Its cross-ฮฒ core structure resembles that of amylin 20-29 polymorph 2. Interestingly, Aฮฒ 25-35 fibrils appear to be more structurally heterogeneous than amylin 20-29 polymorph 2 fibrils, as evidenced by the former's broader ฯ-angle distribution (c.f. Figure and). This may be due to the difference in side chain composition of the two peptides. Amylin 20-29 polymorph 2 adopts parallel ฮฒsheets that allows Phe, Ala, and Leu side chains to be aligned in-register between neighboring ฮฒ-strands, thereby allowing hydrophobic and ฯ-stacking interactions that make the fibril structures more rigid. The MD simulations suggest that this does not occur for Aฮฒ 25-35 fibrils since the side chains hydrophobic residues in the interior between the two ฮฒ-sheets are more disordered (Figure ). This side chain disordering could give rise to the greater conformational heterogeneity observed in the Raman spectra for Aฮฒ 25-35 fibrils compared to amylin 20-29 polymorph 2. |
650cbd0f60c37f4f763792ef | 31 | In this study, we present detailed molecular-level structural models of amylin 20-29 and Aฮฒ 25-35 fibrils determined using MD simulations that are based on C--O bond and ฯdihedral angle constraints measured by Raman spectroscopy. The agreement between our fibril models of amylin 20-29 polymorphs 1 and 2 with those reported by the Landsbury, Griffin, Middleton, and Nielsen groups highlight the potential to develop quantitative molecular structural models of fibrils using Raman spectroscopy. It also highlights the fact that our approach provides structural information that complements gold-standard techniques such as ssNMR and cryo-EM. |
650cbd0f60c37f4f763792ef | 32 | We believe that using experimental parameters measured by Raman spectroscopy to guide MD is more powerful and synergistic than using either technique alone. Namely, the bond and dihedral angle parameters measured by Raman spectroscopy provide structural constraints that help guide the construction of starting models and narrow the conformational phase space sampled over the course of the simulation production runs. In addition, the MD simulations provide a powerful method to visualize the three-dimensional structure of amyloid fibrils in a way that cannot be appreciated by only inspecting the bond and dihedral angle distributions measured by Raman spectroscopy. Finally, our work shows that Raman spectroscopy can be used to quantitatively discriminate between different fibril polymorphs. |
650cbd0f60c37f4f763792ef | 33 | We believe that our approach could be useful in refining fibril structures determined by ssNMR or cryo-EM. This is particularly true for refining disordered or dynamic regions of fibrils, which are difficult to probe using ssNMR and cryo-EM, that could play important roles in initiating aggregation or aberrantly interacting with biological cells. For ssNMR, the final ensemble of structures is visualized by selecting the top 10 or 20 most energetically favorable protein conformers consistent with the experimental constraints applied in the MD simulations. In cryo-EM, disordered, dynamic, or heterogeneous regions could result in poorly resolved electron densities that make it difficult to obtain any structural information at all. In contrast, the widths of dihedral and bond angle distributions measured by Raman spectroscopy naturally give insights into the structural dynamics and conformational heterogeneity of fibrils. This information can be directly incorporated into simulations by parameterizing MD force fields with experimentally measured harmonic force constants, as described above. |
650cbd0f60c37f4f763792ef | 34 | The Raman structural constraints that we utilized can be determined by employing either non-polarized or polarized measurements. Thus, our approach does not inherently rely on aligned fibril samples or polarization measurements. However, more structurally-sensitive Raman spectroscopic markers need to be discovered to enable the determination of higher-resolution fibril models. There are several reports describing Raman spectroscopic markers that can be used to measure amino acid side chain dihedral angles. However, these apply to only a handful of amino acids such as glutamine, asparagine, and tryptophan. Discovering a Ramachandran ฯ-dihedral angle spectroscopic marker would be especially powerful for constraining the structure of the peptide backbone in fibrils. Work by Schweitzer-Stenner and coworkers describes an interesting approach to analyze the Amide I band to determine both Ramachandran ฯ-and ฯangles. However, it remains unclear whether their methodology can be applied to systems beyond tetrapeptides. |
650cbd0f60c37f4f763792ef | 35 | Obtaining site-specific structural information would also be useful in developing our technique. Residue-specific structural information could conceivably be obtained through sitespecific labeling schemes of amide nitrogen and carbon atoms. Since the pathophysiologically-relevant lengths of many amyloidogenic proteins, including amylin, are generally less than 50 amino acids, isotopic editing of peptides could be achieved using solid-phase synthesis. Isotopic labeling of the peptide backbone carbon and nitrogen atoms would decouple the NH bending, CN stretching, and C--O stretching motions in amide vibrations. Isolating the corresponding Amide I and III modes for individual residues could therefore be achieved by determining the difference spectra between the isotopically labeled and unlabeled fibril species. Thus, with these current limitations, it is clear that follow-up studies are needed to increase the utility of our methodology. Despite this, however, we believe that our approach lays a foundation towards potentially using Raman spectroscopy with MD to visualize the three-dimensional structures of amyloid fibrils and other biological macromolecules. Spectroscopy Society of Pittsburgh (DP and MH), and the National Institutes of Health (P20 GM1350007) through the Vermont Center for Cardiovascular and Brain Health pilot grant award (DP, YO, MH, and UN). The Raman microscope used in this work was purchased from funds provided by the National Science Foundation (DMR-1919610). The AFM used in this work was purchased from funds provided by the National Center for Research Resources (S10RR025498). |
6516cb1fa69febde9eebd793 | 0 | The fundamental development of new emitters for organic light-emitting diodes (OLEDs) has always grabbed a lot of attention. Since they play an important role in developing efficient OLEDs in terms of color purity, efficiency, and stability; key aspects for applications in displays, biomedical, and solid-state lighting. Fig. displays the multilayered architecture of OLEDs and their mechanisms allowing direct charge injection, transport, and recombination to form emitting excitons. According to the principle of spin statistics, the excitation of the emitter leads to the formation of 25% singlet excitons and 75% triplet excitons. The emitters based on small molecules present in the active layers of OLEDs are broadly classified into three types, such as fluorescence, phosphorescence, and thermally activated delayed fluorescence (TADF) emitters (Fig. ). Their basic photophysical properties are well reported in many excellent reviews and literatures. The first generation of OLEDs are based on fluorescent emitters (emission comes only from the S1 state after excitation), which harvest only 25% of singlet excitons and therefore reach 25% of maximum internal quantum efficiency (IQE), since 75% of triplet excitons are dark. The external quantum efficiency (EQE) of an OLED device depends on IQE and relates to each other as follows: 9 |
6516cb1fa69febde9eebd793 | 1 | Where,๏จ ๐ผ๐๐ธ and ๏จ ๐๐ข๐ก , denote the internal quantum efficiency and the light-out-coupling efficiency, respectively. Eq. 1 revealed that ๏จ ๐ผ๐๐ธ depends on ๏ง (charge balance factor of injected holes and electrons), ๏จ ๏ง (efficiency of radiative exciton production), and ๏ ๐๐ฟ (photoluminescence quantum yield). In case of fluorescence emitters, ๏จ ๏ง is strongly limited, since the 75% of electrically generated dark triplet excitons. Therefore, there was a search for techniques/emitters to harvest them, that is, the development of second and third generation of emitters. The Second generation of OLEDs are based on phosphorescent emitters (e.g., Iridium(III) and Platinum (II) complexes) that harvest both singlet and triplet excitons, reaching 100% of maximum ๏จ ๐ผ๐๐ธ . Though it overwhelms the limitations of conventional fluorescent OLEDs, its broad emission renders the poor color purity of the devices. In addition, sustainability and cost concerns are always present when using rare-earth materials. The third generation of OLEDs based on TADF organic emitters with a very small singlet-triplet energy gap (ฮEST), which produces both singlet and triplet excitons through an efficient triplet to singlet upconversion process, enables them to reach 100% of ๏จ ๐ผ๐๐ธ . The TADF emitters significantly enhanced the OLED performance, promising low-cost and sustainable emitters. However, the wide emission spectra of TADF emitters still limit the low color purity of the devices, and the long-lived excitons are leading to degradation processes that reduce the ๏จ ๐ธ๐๐ธ and device stability. |
6516cb1fa69febde9eebd793 | 2 | It is in this context that the hyperfluorescence (HF) emitters are emerging. Indeed, they have been referred to as the fourth generation on OLED active layers. HF is defined as the emission mechanism that combines two different emitters, such as TADF and fluorescence emitters (Fig. ). As stated, TADF molecules have a small energy gap (๏EST) between the singlet and the accepting triplet energy states. This permits the upconversion of excited energy from accepting triplet to singlet states, which provides efficient emission from the singlet state. However, if the excited energy produced by the TADF emitter is harvested by a fluorescent molecule following the Fรถrster resonance energy transfer (FRET) process, this will emit four times more photons than conventional fluorescence at the same excitation conditions. In addition, HF emitters display a strong and narrow-band emission, contrary to the broad emission of TADF emitters. Therefore, HF has been established as a promising approach to designing OLEDs with high color purity and enhanced stability. The study of HF emitters and their optoelectronic applications is progressing with momentum. In this mini-review, we systematically documented the molecular design principles of HF emitters, their fundamental photophysical requirements, and the advances in HF-OLED. |
6516cb1fa69febde9eebd793 | 3 | TADF emitter acts as a sensitizer, which plays an important role in harvesting the singlet excitons of fluorescence emitters. Therefore, the TADF emitters should have an shift for efficient energy transfer. Furthermore, an efficient FRET (high kFRET) process requires the maximum overlap between the absorption spectrum of the fluorescence emitter and the emission spectrum of the TADF emitter. Finally, the first singlet excited state (S1) of the fluorescence emitter must be below that of the S1 state of the TADF emitter (Fig. ). The concentration of two emitters is also crucial for efficient HF behavior. Typically, a doping strategy is followed, in which a high amount of TADF emitter and a lesser amount of fluorescence emitter in the active layer is required. |
6516cb1fa69febde9eebd793 | 4 | than the fluorescence and TADF emitters to prevent reverse energy transfer. Moreover, it is also very important to prevent Dexter energy transfer (DET) among triplet states to avoid loss of triplet exciton, which is achieved through efficient kRISC, and the use of less doping concentrations of emitters. Finally, a key requirement is to control carrier trapping, which is essential to forbid the formation of triplet exciton in the dopant fluorescence emitter. The implementation of bulkier groups in fluorescence emitter is the most suitable approach to control it. In the last few years, resonance (MR)-TADF terminal emitters have been greatly used in HF-OLEDs. Concerning the current use of MR-TADF terminal emitters, they should have short-range charge transfer characteristics, a small Stoke shift and efficient kRISc. The FRET processes are evaluated by using simple following mathematical equations: |
6516cb1fa69febde9eebd793 | 5 | where ฯF is the prompt fluorescence lifetime, R0 is the Fรถrster radius, at which the energy transfer efficiency is 50% and R denotes the average intermolecular distance between fluorescence dopant and TADF emitter. where, n, ฮบ 2 and ๏F are the refractive index of the medium, dipole orientation factor, the prompt quantum yield of the TADF emitter, respectively. |
6516cb1fa69febde9eebd793 | 6 | A few representative examples of successful TADF/fluorescence emitter combinations are described as follows: F1 is one of the best fluorescence emitters used for the development of narrow-emitting HF-OLED (Fig. ). The emitter F1/MR-T1 exhibits a strong absorption band at 457 nm and a sharp fluorescence band at 468 nm, with full width at half maximum (FWHM) of about 14 nm in toluene. This is related to the combined effect of its polycyclic structure and the multiple resonance (MR) effect of the boron and nitrogen atoms. In addition, F1 exhibits a ๏PL of 74% in toluene. In fact, F1 and its derivatives are also TADF active and recently been employed in many HF-OLED as terminal emitters with MR-TADF characteristics (discussed later). |
6516cb1fa69febde9eebd793 | 7 | Kwon and co-workers selected TADF emitters T1 and T2 (Fig. Concerning the low-energy part of the visible spectrum, the Kwon group reported a red HF-OLED combining F2 and a TADF emitter T3 (Figs. and). Owing to the sharp emission and high ๏PL of BODIPYs, they selected a BODIPY-based red emitting fluorescence emitter F2 (emission maximum at 620 nm), featuring a high ๏PL of 99% and a small FWHM of 31 nm in solution. However, the deep lowest unoccupied molecular orbital (LUMO) of 3.83 eV may induce electron trapping directly on BODIPY in the emissive layer. Therefore, a TADF emitter with deep LUMO would be best for these BODIPY-based fluorescence emitters, which help reduce the trapping of electrons in the emissive layer. Thus, they selected a TADF emitter T3 (Fig. ) that possesses a deep LUMO (3.9 eV) along with better spectral overlap with the absorption spectrum of F2 and the emission spectrum of T3. In addition, a high ๏PL of 98% and high kFRET of 5.10 ร 10 7 s -1 were obtained for 0.7% doped films. Therefore, the selection of the T3 and F2 combination resulted in an efficient HF-OLED (see below). 57 |
6516cb1fa69febde9eebd793 | 8 | There was a continuous success in designing HF-OLEDs after the first examples reported by the Adachi group. Using a suitable host and the right choice of TADFfluorescence emitter couple , such as T4:F3, T5:F4, T6:F5, and T7:F6 (Figs ), they succeeded to fabricating blue-, green-, yellow-, and red-emitting HF-OLEDs, respectively (Table ). With the limit of ~15-50 wt% doping concentrations of TADF emitters and 1 wt% of fluorescence emitter, the maximum EQE of the blue-, green-, yellow-, and red-emitting HF-OLEDs were 13.4%, 15.8%, 18.0%, and 17.5%, respectively. There was a huge increase in EQE compared to conventional fluorescent emitters in all devices, demonstrating the EQE-enhancing role of HF emitters. No doubt, this approach demonstrates a path for designing several new HF emitters using theTADF-fluorescence emitter couple and their fruitful applications in OLEDs. However, a low EQE (13.4%) was observed for the first blue HF-OLED based on the T4:F3 emitter couple (Table ). This might be attributed to the incomplete FRET from the TADF emitter to the fluorescent emitter, as the TADF emitter (T4) emits at longer wavelength than the fluorescence emitter (F3), decreasing the spectral overlap JF (eq. 4). |
6516cb1fa69febde9eebd793 | 9 | In contrast, Lee group reported a HF-OLED showing improved EQEs, which contain a deep blue TADF emitter (T8) and fluorescence emitter (F3). At 50 wt.% doping concentration of T8 and 0.1 wt.% doping concentration of F3, a deep blue HF-OLED was fabricated with host DPEPO (Table ). This resulted in an improved EQE reaching a value of 18.1%. These studies suggest that the choice of the T8:F3 couple is better than the T4:F3 couple for blue HF-OLED (Figs. and). Further studies revealed that a blue HF-OLED constructed using a T8:F3 couple without a host exhibits EQE of 15.4 %. As the next steps, another TADF emitter T9 was used to sensitize the blue emission of F3. Similar to the previous works, a solution-processed efficient blue HF-OLEDs was prepared, reaching an EQE of 18.8 % using a bulkier host DPOBBPE (Table ). These findings revealed that the nature of the host plays a vital role f in the device's performance by controlling Dexter energy transfer; the bulkier the host, the lesser will be the Dexter energy transfer. Kwon group recently developed two more blue-emitting TADF emitters, T1 and T2 (Fig. ). Both emit deep blue emissions and exhibit high ๏PL. HF-OLEDs were fabricated using F1 as a dopant with host DBFPO. The blue HF-OLEDs achieved using the T1:F1 and T2:F2 couple exhibit impressive EQE of 38.8% and 37.3%, respectively. In addition, the EL spectra of the both blue HF-OLEDs featured a FWHM of 19 nm. Very recently Kwon group reported one blue HF-OLED using T10 as quadrupolar donor-acceptor-donor type TADF emitter and F1 as a fluorescence emitter in DBFPO host (Fig. ) . The HF-OLED device fabricated with this combination exhibits high EQEs of 43.9% of with x/y CIE chromaticity coordinates of 0.12/0.16. This significant increase was attributed to the successful control of the DET process through efficient KRISC and shielded LUMO in T10. In addition, the HF-OLED constructed using the T10:F1 couple displays an EL peak maximum at 473 nm with a narrow FWHM of 21 nm. These investigations suggest an efficient HF-OLED could be developed by the proper selection of host, TADF emitter, and fluorescence terminal emitter. |
6516cb1fa69febde9eebd793 | 10 | Besides blue HF-OLEDs, there were a few efforts to develop other (green, red, and white) HF-OLEDs (Table ). In detail, an efficient green HF-OLED fabricated using TADF emitter T11, fluorescence emitters F7 and F8 with mCP host (Figs. ). The EQEs were 13.5% and 14.6% for T11:F7 and T11:F8 couple emitter, respectively. Here, it could be noted that, EQE of only ~5% was obtained using the traditional Alq3 host with TADF emitter T11 and fluorescence emitters F7 and F8. This indicates the crucial role of the host towards high EQEs. Yellow HF-OLEDs with the TADF emitter T12 and yellow emitting fluorescence emitter F5 (Figs. and), which feature a good overlap of the emission spectrum of T12 with the absorption spectrum of F5, featured a maximum EQE of 19.1% using the mCBP host. Finally, red HF-OLEDs were prepared mixing T13 or T14 with F6 and CBP host. The device composed of T13:F6 exhibited a maximum EQE of 8 % and x/y CIE chromaticity coordinates of 0.61/0.38. |
6516cb1fa69febde9eebd793 | 11 | The device performance with the T14:F6 couple was lower than that of device with the T13:F6; this might be due to their different TADF behavior owing to the presence of different donor groups in T13 and T14. However, the EQE of the devices with T13:F6 couple was higher than those with each component. Recently, another red HF-OLED was reported using red-emitting BODIPY-based fluorescence emitter F2 and TADF emitter T3 (Table ). The fabricated HF-OLED with T3 and F2 shows EQEs of 19.4%. |
6516cb1fa69febde9eebd793 | 12 | The device exhibits an EL peak maximum at 617 nm with x/y CIE chromaticity coordinates of 0.64/ 0.36 and a FWHM of 44 nm. In addition, the device shows a longer device lifetime of 954 h at 3000 cd m -2 . Wang group reported a red HF-OLED using TADF emitter T15 and conventional red fluorescence emitter F6 (Table ). The HF-OLED fabricated using this combination resulted in a maximum EQE of 16.4% and x/y CIE chromaticity coordinates of 0.65/0.35. Recently, HF-OLEDs with MR-TADF materials as terminal emitters have been developed rapidly owing to their strong and narrow emission. MR-TADF materials (Fig. ) are fused polycyclic aromatic compounds with suitably placed electron-deficient and electron-rich centers, enabling their HOMO and LUMO separation on different atoms due to the opposite resonance/mesomeric effects. This reduces ฮEST and triggers TADF emission. The rigid frameworks and short-range charge transfer features of MR-TADF materials make them ideal candidates for OLEDs realizing narrow emission with high color purity and high efficiency. The following are some representative examples of this class of HF -OLEDs. Adachi group used TADF emitter T16 to sensitize the narrow emission of MR-TADF emitter MR-T1 (Figs. and). The presence of a bulky m-terphenyl (instead of carbazole) unit at the para-position of T16 triggers a greenish-blue emission at ~ 475 nm in solution, which is close to the absorption maximum of MRT1, enabling efficient energy transfer between them. In thin film with 0.5 wt.% of MR-T1 and 20 wt.% of T16 the emission maximum observed at 485 nm with a ๏PL of 86%. High kRISC i.e., 9.2ร10 5 s -1 in thin-film, attested an excellent TADF behavior. Moreover, a high FRET efficiency of about 64% from T16 to MR-T1, indicates an efficient excited energy transfer from T16 to MRT1. This combination resulted in an efficient narrow-emitting (FWHM = 19nm) HF-OLED with a maximum EQE of 41% and good stability (Table ). Zhang group reported a blue HF-OLED using the T17: MR-T1 couple. The donor-void-acceptor structural feature of T17 justifies its high ๏PL of 92% and displays an emission maximum at 440 nm in toluene. |
6516cb1fa69febde9eebd793 | 13 | The HF-OLED fabricated with 30 wt% TADF sensitizer T17 and 1 wt% MR-TADF emitter MR-T1, resulted in a maximum EQE of 20.6% with a narrow FWHM of 21 nm that has x/y CIE chromaticity coordinates of 0.140/0.195. In parallel, many researchers exploited B, N doped MR-TADF terminal emitters in HF-OLEDs to achieve narrow emission with high efficiencies and high color purity. Kwon group reported a modified MR-TADF emitter (MR-T2) by attaching the tert-butyl group to the MR-T1 core. As mentioned earlier, incorporating bulky groups into the backbone of the terminal emitter can effectively inhibit the DET process and increase ๏PL. In thin film, MR-T2 displayed a ๏PL of 91.9%. This increase in ๏PL compared to MRT1 is due to the addition of the tert-butyl group. MR-T2 showed a short delayed lifetime of 2.93 ยตs, and kRISC of 2.54 ร 10 5 s -1 . Therefore, they have selected a suitable TADF emitter, T18 to sensitize the blue emission of the MR-TADF emitter, MR-T2. The thin film obtained with 30 wt.% T18 and 1% of MR-T2 has a ๏PL of 97.3% and a kFRET of about 7.77ร 10 7 s -1 . This indicates efficient energy transfer from T18 to MR-T2. Therefore, the blue HF-OLED fabricated with the MR-T2:T18 combination exhibits a maximum EQE of 40.7% along with a narrow FWHM of 19 nm and x/y CIE chromaticity coordinates of 0.12/0.15. Zysman-Colman group reported a different B, N doped MR-TADF emitter, MR-T3. |
6516cb1fa69febde9eebd793 | 14 | and MR-TADF emitter MR-T4 combination. With an aim of developing an efficient TADF emitter, a sterically congested TADF emitter, T20 with multiple donor groups was developed. The presence of multiple donor-acceptor moieties in the same molecule supported the orbital mixing of LE and CT states, ensuing a high kRISC of 2.36 ร 10 6 s -1 . |
6516cb1fa69febde9eebd793 | 15 | Hence, the HF-OLED obtained with the T20: MR-T4 combination records an EQE maximum of 32.5%, a FWHM of <30 nm and x/y CIE coordinates chromaticity of 0.13/0.12. Very recently, the Lee group reported an MR-TADF emitter MR-T5, a sterically shielded and rigid emitter based on triptycene-fused B,N core. In doped film, MR-T5 exhibits an emission maximum at 463 nm with a ๏PL of 99%. They have selected a suitable TADF emitter, T21, which effectively sensitizes the emission of MR-T5. The thin film with 20 wt% T21 and 2 wt% MR-T5 with mCBP:DPEPO exhibits a high kFRET of about 4-6ร10 7 s -1 . The HF-OLED with T21: MR-T5 showed a maximum EQE of 27.5% and a FWHM of 29nm. Moreover, the Zysman-Colman group reported a new acceptor-free MR-TADF molecule, MR-T6. MR-T6 exhibits weak TADF with a ๏PL of 67% and an emission maximum at 441 nm in 3 wt% PMMA film. The HF device obtained with 35 wt% T22 and 1 wt % MR-T6, reached a maximum EQE of 16.5%, with deep-blue x/y CIE chromaticity coordinates of 0.15/0.11. Recently, same group reported green and red HF-OLEDs based on MR-TADF emitter MR-T7 and MR-T8, respectively. 70 MR-T7 with triphenylamine donor groups showed green emission at 551 nm with FWHM of 58 nm and ๏PL of 93%, while MR-T8 with diphenylamine donor groups showed a red emission at 617 nm with FWHM of 56 nm and ๏PL of 60% in the 2 wt% mCP doped films. A green HF-OLED fabricated using 2 wt% MR-T7 as emitter and 10wt% TADF sensitizer T11, recorded a high maximum EQE of 30 % with x/y CIE chromaticity coordinates of 0.424/0.551. Also, a red HF OLED fabricated using 2wt% MR-T8 as emitter and 10 wt% TADF sensitizer T11 and recorded a high maximum EQE of 18% with x/y CIE chromaticity coordinates of 0.585/0.396 (Table ). There is also increasing interest in designing novel organic-based emitters for the fabrication of white OLED (WOLED). They are developed with low cost and have exhibited notable merits of high efficiency. The WOLED device is characterized by the x/y CIE coordinates of 0.33/0.33. Normally, white light is produced by the simultaneous blue, green, and red missions, or the addition of two complementary emissions. This is usually possible with organic emitters that exhibit dual/multiple/broad emissions. In the case of HF-WOLEDs, the white light is obtained by combining the emissions of both TADF and fluorescence emitters, which covers the complete visible range (400-800 nm). Fig. depicts the co-emission mechanisms of the hyperfluorescent emitters to produce white light emission. Here, the TADF emitter performs a dual role, acting itself as a blue emitter and a sensitizer for fluorescence emitter. A HF-WOLED was developed with T23 as blue TADF emitter, and F5 as a yellow fluorescent emitter. Both T23 and F6 (Figs. and), were co-doped with 0.05% doping concentration, and the HF-WOLED device constructed in DPEPO host exhibits |
6516cb1fa69febde9eebd793 | 16 | x/y CIE coordinates of 0.28/0.35 and high EQE of 15.5%. Similarly, another HF-WOLED developed using a TADF emitter T23 and orange-emitting rubrene (F9) without any host, reached EQEs of 7.48%. In addition, stable EL spectra were recorded with the change of the x/y CIE coordinates of 0.36/0.41 to 0.36/0.43 with increase in voltage from 5 V to 8 V. Friend and coworkers reported a HF-WOLED by using a skyblue TADF emitter T24, complemented by yellow and red emitting fluorescence emitters F5 and F1, the device resulted in a high EQE of 21.8% with x/y CIE coordinates of 0.43/0.45. Very recently, the Xu group reported an ultra-thin bilayer concept for the design of HF-WOLED, where a emissive layer contains an ultrathin layer of blue fluorescence emitter (F3), and a layer of TADF emitter (T25 or T26) doped with a yellow fluorescence emitter (F5). The efficiency of these HF-WOLEDs is highly sensitive to the thickness of the layers. At the thinness of 0.1nm of blue emitter layer, the device exhibits dual band EL spectra with x/y CIE coordinates of 0.40/0.50. The maximum EQE values of 20.9%, and 11.8 % are reported for the devices using TADF emitters T25 and T26, respectively. Similarly, the Tang group recently reported a HF-WOLED using a yellow emitter F5 , blue emitter F11 and a TADF emitter, T6 with a pure hydrocarbon host, SF4-TPE. They witnessed with an increase in thickness of the blue layer, the EL intensity becomes stronger and the HF-WOLED constructed using 10 nm thickness of the blue layer exhibited maximum EQE of 17 Interestingly, these efficiencies of HF-OLEDs are much higher compared to reported fluorescence OLEDs. |
6516cb1fa69febde9eebd793 | 17 | Significant advances in solid-state lighting technology have been made with the appealing use of new hyperfluorescent active layers and their respective OLEDs. Here, the Foster resonance energy transfer (FRET) between a TADF and fluorescent emitter plays a primary role in the device efficiency of blue-, green-, red-as well as whiteemitting HF-OLEDs. In addition, circularly polarized HF-OLEDs using chiral emitters are also possible. In contrast to fluorescent OLEDs, all-hyperfluorescent OLEDs have high color purity (owing to their narrow emission) and significantly enhanced EQEs. for OLED is also a big challenge because it is very difficult to design high triplet energy host materials. In this context, the recent work of the Bryce group should be followed, which provides molecular design directions for obtaining very high triplet energy host materials. Recently , it was observed that phosphine oxide-free hosts perform better in OLEDs and therefore, the design of new phosphine oxide-free hosts and their use in devices should be investigated. Most of the reported MR-TADF emitter exhibits weak TADF emission and therefore, the design of efficient MR-TADF terminal emitters with high kRISC is further required to have efficient HF-OLEDs. Finally, dyads based on covalent/non-covalent emitters could be explored; since they have the capacity to control distance towards an enhanced FRET process. |
6746321af9980725cf0f27bc | 0 | In native mass spectrometry (nMS) for structural biology, insource collisional activation is often used to clean up biomolecules and complexes that are adducted by salt and water molecules. This strategy can result in 'cleaner' biomolecules and narrower peaks that increase spectral resolution. These two outcomes result in deconvolved masses that are accurate, and easier to obtain than when the complex is 'adducted,' leading to broad peaks. In-source collisional activation (isCA) can also be used as a pseudo-MS n step to dissociate the precursor macromolecular complex, or dislodge native ligands, for targeted experiments downstream from the source. However, it is also known that isCA can cause native-like biomolecular complexes to restructure such that downstream activation steps and analyses do not reflect the native-like structure. Hence, it is critically important to balance the need for 'cleaner' mass spectra with the need to produce ionized biological macromolecular complexes that produce data reflecting the native-like structure and chemistry. However, it can be difficult to ascertain what degree of isCA is appropriate, while also resisting the natural urge to obtain spectra that look 'nice', i.e., spectra with narrow, resolved peaks. This paper seeks to illustrate why care must be taken with isCA when pursuing nMS, by using surface-induced dissociation and electron capture dissociation as tools to develop appropriate "native" instrument tuning. In particular, this paper assesses native-like tuning for a cyclic IMS QTOF instrument with a commercial implementation of SID installed after the selection quadrupole. |
6746321af9980725cf0f27bc | 1 | Surface-induced dissociation (SID) causes native-like protein complexes to dissociate along the weakest subunit interfaces, providing information about subunit stoichiometry and connectivity, assuming that no intertwined subunits exist and/or that restructuring is not a necessary path to dissociation. However, when isCA causes rearrangement to a non-native-like structure, the protein complexes will no longer produce nativelike SID fragmentation patterns. By this we mean that protein complexes will not dissociate through pathways that indicate their native-like topology, subunit connectivity, etc. Work published by the Wysocki lab in 2015 illustrates this point. As protein complexes were subjected to various degrees of isCA, they underwent structural rearrangement that was reflected in the SID spectrum. Collision-induced dissociation (CID) was not able to detect these structural rearrangements because the CID products are typically a highly charged monomer and the complementary (N-1mer) subunit, which does not possess a native-like topology. Thus, SID proved to be a useful probe of source-induced structural rearrangements and can be used to tune isCA to maintain native-like structures. |
6746321af9980725cf0f27bc | 2 | Electron capture dissociation (ECD) is an ion activation strategy that involves ions interacting with low-energy electrons to induce fragmentation. For proteins, the electron capture process causes the N-Cฮฑ bond to fragment, producing c/z-type fragments, in comparison to the b/y-type fragments typically obtained by CID. Because these fragments generally come from flexible and solvent-exposed regions, the hydrophobic core is often left intact after the ECD process. This can provide information about the gas phase structures of native-like protein complexes. Historically, this technique was limited to FTICR instruments due to the feasibility of trapping electrons in the ICR cell to interact with trapped protein cations. When combined with long trapping times, FTICR-based ECD experiments on small proteins revealed that, on long time scales, proteins can elongate and collapse in ways that affect ECD fragmentation. While those studies took place in an FTICR instrument, the Ruotolo lab recently illustrated that ECD could also be sensitive to source-induced structural changes on an IM-Q-TOF instrument. The work presented here examines a recently developed cyclic IMS QTOF platform that features commercially available SID and ExD devices. We demonstrate which pre-IM regions can be prone to activating native-like proteins and protein complexes, and make recommendations on how to tune those regions. We also demonstrate how isCA can result in quaternary structure rearrangement and how SID can be a sensitive diagnostic tool of that effect. While isCA -fixed SID experiments were performed by the Wysocki lab in the past, those experiments demonstrated the effect of isCA on SID through large jumps between isCA energies. For the first time, this work demonstrates the same effect, in greater detail, by varying isCA in small steps while tracking the effect on the SID pathways for standard protein complexes, while also reporting protein complex drift time changes and monomer charge states, during the isCA -fixed SID experiments. Overall, we illustrate that SID, in particular, and ECD, can serve as independent means to validate an instrument's source tuning to strike the balance between maximum signal intensity and adduct removal, while maintaining native-like structure for protein complexes. Panel C was adapted with permission from reference 32, copyright 2022 American Chemical Society. Ubiquitin from bovine erythrocytes was purchased from Sigma Aldrich (Ubq, monomer, 8.5 kDa, P/N U6253), streptavidin (SA, tetramer, 53 kDa) was purchased from Thermo Scientific (catalog number 21125), human recombinant C-reactive protein (CRP, pentamer, 115 kDa) was purchased from EMD Millipore (P/N 236608), and phosphorylase B (PhosB, dimer, 197 kDa) was purchased from Millipore Sigma (P/N P6635). Tryptophan RNA-binding attenuation protein (TRAP) from Bacillus stearothermophilus (bst) was obtained from Prof. Mark Foster's lab at The Ohio State University. It was expressed and purified as previously described, with the minor alteration that cell lysing was accomplished via sonication after overnight expression. Proteins purchased as lyophilized powder were reconstituted in 18 Mฮฉ water to a protein complex concentration of ~40 ยตM, except for Ubq, which was reconstituted to 100 ยตM. All proteins were buffer exchanged twice into 200 mM ammonium acetate using BioRad Microspin 6 buffer exchange columns and diluted to 1 ยตM for CRP, 2 -3 ยตM for SA, and 4 uM for PhosB in 160/40 ammonium acetate/triethylammonium acetate, except for Ubq which was diluted to 10 ยตM with 200 mM ammonium acetate. TRAP was provided at a protein complex concentration of ~1.5 ยตM in 600 mM ammonium acetate. The 21+ charge state was obtained by spraying from this stock solution. The 15+ charge state was obtained by adding 1 ยตL of TEAA to 4 ยตL of the stock solution. For all complexes, the triethylammonium acetate (Sigma Aldrich, P/N 90358) was added for chemical charge reduction because lower charge states often give native-like fragmentation patterns compared to normal charge states from 200 mM ammonium acetate. Ions were introduced into the instrument via nanoelectrospray ionization, using glass capillaries (Sutter Instruments) pulled on a Sutter Instruments P-97 tip puller (Novato, CA). Electrospray voltages were ramped up to 1.3 kV to induce spray, then were ramped down to 0.7 -1.0 kV to sustain spray. |
6746321af9980725cf0f27bc | 3 | All experiments were performed on a Waters SELECT Series Cyclic IMS (Q-IM-TOF) with a commercially available, stainless steel, thin SID device (3-mm along the ion path) located after the selection quadrupole in place of the DRE lens prior to the Trap stacked ring ion guide used for CID. The instrument was operated in ion mobility mode, with ion mobility separation traveling wave (TW) static height of 25 V and TW velocity of 375 m/s. All instrument settings were tuned to be as gentle as possible to minimize protein complex restructuring, unless collision-induced structural changes were intended. In that case, the cone voltage was adjusted in 10 V steps from 0 -160 V to vary the amount of isCA (named "cone isCA" by Waters for consistency with older instruments, although the voltage variation does not involve a source cone). Subsequently, data were acquired without further activation (cone isCA), or with additional, intentional activation by another method (cone isCAfixed SID or cone isCA -ECD). Data acquired via cone isCA were collected in triplicate on three separate days. Data processing was accomplished using CIUSuite 2. Cone isCA -fixed SID experiments were performed at a fixed SID potential, for the indicated quadrupole-selected charge states, while scanning the cone voltage in 10 V steps. All data were obtained in triplicate. Final plots are the subsequent averages and error bars represent one standard deviation from the mean. For SA 11+ and CRP 17+, the SID potential was 45 V (495 and 765 eV, respectively). For PhosB 21+ the SID potential was 190 V (3990 eV). For TRAP the SID potential was 70 V for the 15+ charge state (1050 eV) and 50 V for the 21+ charge state (1050 eV). SID data for SA, CRP, and TRAP were processed using DriftScope v3.0 to extract oligomer intensities to be sum normalized and plotted as a function of cone voltage. The same procedure was followed to obtain weighted average SID monomer charge plots. When the SID data for PhosB were visualized in DriftScope, the precursor dimer and product monomer lines overlapped in a way that prevented signal intensity extraction. Thus, an alternative strategy was devised. First, Waters Raw files were converted to mzML files using ProteoWizard MSconvert 44 (version 3.0.23244-bc8a3ad) with the Scan Summing filter applied with the following parameters: Precursor tolerance ยฑ0.05 m/z, Scan time tolerance ยฑ5 s, ion mobility tolerance ยฑ5 ms, and Sum MS1 scans. The mzML files were then imported to an HDF file and deconvolved using MetaUniDec, with the cone voltage as the input variable. MetaUniDec was then used to extract signal intensities and charge states for PhosB's monomer and dimer at each cone voltage. |
6746321af9980725cf0f27bc | 4 | Cone isCA -ECD experiments were conducted at two different cone settings, either 20 or 160 V, followed by quadrupole isolation of the same charge states as were selected for the isCA -SID experiments, then by IM separation, and finally by electron capture in the ExD device located in the pre-transfer guide after IM. Because ECD product ions tend to remain "stuck" (non-covalently bound) within native-like proteins/complexes due to non-covalent interactions, post-ECD activation (transfer collision energy (TCE)) was used after electron capture to liberate the fragments. Typical Transfer CE voltages ranged from 60 -130 V. All ECD spectra were processed using ExDViewer v4.5.8 (eMSion, an Agilent company), with the default deconvolution and peak picking settings and a mass tolerance of 20 ppm for product ion matching. |
6746321af9980725cf0f27bc | 5 | Experiment Type: Optimize Pre-IM Tuning With each new instrument generation, end-users must optimize instrument tuning to meet their experimental needs. For the nMS community, this often means tuning instrument voltages to minimize ion activation in the source and other pre-selection quadrupole areas to maintain native-like biomolecule conformations; and in the post-activation areas to minimize the risk of secondary fragmentation. While the Waters cIM features a unique IM region that can be complex to tune, the Zenobi lab produced a helpful body of work related to that topic. In terms of activation regions, the Waters cIM instrument features multiple places for deliberate activation (Figure ): the cone voltage between the StepWave and the source ion guide, the Trap cell, the IM pre-array, and the Transfer cell. To minimize activation prior to the selection quadrupole, the source region (ion source, StepWave, ion guide) should be carefully tuned to minimize unintentional activation and maintain nativelike protein conformations. Between the selection quadrupole and the cIM, the Trap gradient, post-Trap bias, and helium cell should be tuned to minimize pre-IM activation, or to minimize secondary fragmentation after SID or Trap CID. To tune our instrument, we used Ubq's measured CCS as a probe of pre-IM "hotness," and SID of CRP to fine-tune for native protein complexes. Figure . A -F) Collision cross-section distributions for Ubq 6+. He cell is the helium cell, pTb is the post-Trap bias, and BG and HG are the body gradient and head gradient, respectively, in the source StepWave. The underlined parameters highlight what is varied from one panel to the next. The flame icon indicates the "hottest" tuning profile, resulting in Ubq 6+ adopting extended conformations, while the snowflake indicates the "coldest" tuning profile, resulting in compact conformations. The red to orange to blue peak colors also indicate the transition from hot to cold tuning. |
6746321af9980725cf0f27bc | 6 | Depicted in Figure are calibrated collision cross-section (CCS) distributions for Ubq 6+ (CCS calibration was achieved using Agilent ESI Tuning Mix as the calibrants, with IMSCal software). Panel A features the hottest (most activating) tuning, and panel F is the coldest (least activating) tuning, based on matches with CCS in the literature for Ubq. The relevant instrument parameters are listed in the figure. As the helium cell entrance, bias, and exit were tuned to have lower voltage deltas, the CCS distribution shifted from extended conformations between 1250 -1750 ร
2 to slightly more compact conformations between 1000 -1500 ร
2 . As the post-Trap bias was lowered from 35 V to 20 V the CCS distribution shifted to favor the most compact conformational state, centered around 1100 ร
2 . When the StepWave body and head gradient deltas were lowered to 1 V apiece, a slight shift towards the most compact conformation is seen, though not nearly as dramatic of a shift as was seen in lowering the deltas and bias of the helium cell and post-Trap bias, respectively. It is possible that this slight shift could result from the larger, slower Ubq conformer being scattered in the He cell and not arriving at the detector. However, this could also affect the slower 5+ charge state as well, which is not observed when we compare relative intensities for 6+/5+ in the mass spectra under these different tuning conditions. In terms of absolute signal intensity, as the conditions are tuned from Figure to 2F, the mass spectra indicate an approximate 50% signal intensity loss as the instrument conditions become cooler. Despite that loss the spectra are still easily interpretable with appropriate S/N. With minimized deltas in the StepWave and the helium cell, and minimized post-Trap bias, the least activating tune parameters were found for Ubq. |
6746321af9980725cf0f27bc | 7 | With these minimally activating tune settings, CRP was used to ensure good ion transmission after SID. Figure depicts a series of SID bar charts for charge-reduced CRP (17+), in line with the charge state used for the data described later in the paper. With the least activating tune parameters for Ubq (Figure ), no ions were observed in the mass spectrum or mobiligrams (Figure ). Increasing the post-Trap bias from 20 to 35 V (panel B) resulted in evidence for trimer, tetramer, and pentamer, which suggests poor collection of complementary ions (monomer to accompany tetramer, dimer to accompany trimer). Additionally, the spectra and mobiligrams featured low ion intensities. Panels C -F were obtained by adjusting the indicated helium cell deltas and bias. Comparing these panels reveals that tuning the helium cell improperly can result in ion transmission bias, depending on the relationship between the helium cell entrance and exit deltas. Panel E appears to feature native-like SID for CRP, however, the high m/z region was missing from the mobiligram, which was concerning because complementary oligomers and charge states were not present when they reasonably should have been. These product ions appeared when the helium cell deltas were increased to the values in Panel F. |
6746321af9980725cf0f27bc | 8 | Panels G -I featured a pushes-per-bin value of 4, while panels A -F used a value of 3. The yellow star on panel I indicates the final tune settings we used for the SA, CRP, and TRAP experiments discussed below, except a pusher-per-bin value 0f 3 was used. Similar results could be obtained with the tune settings present in F -H. While it is important to consider different instrument configurations and detection strategies when comparing data acquired on different instruments, the Wysocki group has routinely shown that SID results are instrument agnostic, and that standard proteins should be used for benchmarking new SID installations. It is apparent when comparing the conditions in Figures and that what is optimal for ubiquitin (Figure ) is not ideal for CRP (Figure ). Because CRP is larger and features more non-covalent interactions than Ubq, the conditions that elongate Ubq do not have the same effect on CRP. This contrast highlights the importance of tuning an instrument for the analytes of interest and that no single tuning condition will preserve native-like structures for all proteins. Figure . A -I) SID bar chart for 765 eV SID of CRP 17+ pentamer. The letter acronyms are the same as in Figure , with the input value listed directly after the acronym. In the case of the helium cell (He) the entrance, cell bias, and exit are listed from left to right. The yellow star on panel I indicates the tuning we used for our experiments, though similar results could be obtained with conditions listed in panels F-H as well. Panel E also appears to give optimal SID results, but the high m/z portion was absent from spectra and the mobiligrams, which is not desirable. Panels F and G have the same tuning parameters, but the pushes-per-bin was increased from 3 for panels A -F, to 4 for panels G -I. The terms 1mer, 2mer, etc. were used in place of monomer, dimer, etc. due to space constraints within the figure panels. |
6746321af9980725cf0f27bc | 9 | To confirm the suitability of our final tuning, an SID energyresolved mass spectrum (ERMS) plot was obtained for CRP 18+, the charge state used by Harvey et al. The ERMS plot in Figure features the dimer/trimer pair as the initial dominant SID product pathway, followed by monomer/tetramer. These pairs rise together in normalized intensity until dimer/trimer deviate due to secondary fragmentation of trimers to form monomer/dimer pairs and potentially dimer to monomer and tetramer to dimer-dimer or monomer-trimer, which also causes the monomer/tetramer pair to no longer track together. As the SID energy increases, monomer formation becomes the dominant SID pathway. These results clearly depict native-like SID fragmentation for the CRP pentamer, confirming that our tuning parameters (Figure ) are suitable for native-like analyses of protein complexes. |
6746321af9980725cf0f27bc | 10 | As discussed above, and as will be illustrated below, SID is an excellent tool for probing the "nativeness" of protein complexes. It can clearly indicate when instrument tuning is too harsh. Thus, SID of a standard protein complexes was key to confirming that our instrument tuning was minimally activating prior to SID and cIM. |
6746321af9980725cf0f27bc | 11 | Experiment Types: isCA-IM and isCA-SID-IM Cone isCA collision-induced unfolding/collapse/restructuring (CIx) plots for all four charge-reduced protein complexes are found as overlays within panel A of Figures . The respective RMSD plots for the averaged triplicates are found in the SI (Figures ). The RMSD values are high, ranging from 8 -11. This is likely due to collecting the CIU data on separate days, which allows for pressure fluctuations that affect drift time. While CCS calibration could correct for such fluctuations, calibration would not affect the overall trends that are described hereafter. The plots obtained for SA (Figure ), and CRP (Figure ) reflect decreases in drift time as a function of increasing cone voltage. This suggests that the proteins' quaternary structures were rearranged to smaller volumes due to isCA, as previously reported. CRP shows the most significant decrease in drift time at the highest cone voltages. Its structural rearrangement is partially attributable to its ring-shaped quaternary structure (inset, Figure ) collapsing inward on itself, and possibly includes interactions between the elongating monomer chain with remaining complex (inferred from fragmentation patterns and charge states). Unlike monomeric proteins, protein complexes, especially when charge-reduced, are less likely to undergo major unfolding/expansion. The CIx plot for PhosB (Figure and S4) does not possess noteworthy features that suggest significant quaternary rearrangement. Please note, the CIx plot overlay in Figure covers isCA voltages from 0 -100 V to align with the x-axis of the isCA -SID experiment. The full CIx plot, up to 160 V, for PhosB is found in Figure . While PhosB's plot does not contain noteworthy features, the drift time does decrease linearly with increasing cone voltage. This appears to result from increasing amounts of collisional de-adducting as a function of isCA. Examining the mass spectra collected with cone settings of 20 V and 160 V (not shown) revealed a -14 m/z shift between the two cone settings, which corresponds to an ~300 Da mass loss for the 21+ charge state. Overall, while the isCA CIx plots (panel A in Figures ) can be helpful in diagnosing sourceinduced quaternary structure rearrangements, this paper focuses on SID's utility as a diagnostic tool, as described below in a series of experiments where the isCA was varied, as denoted by isCAโ, and the SID energy was held constant, as denoted by SIDC. Based on that focus, the isCAฮ-SIDC plots in panel A of Figures contrast with the overlaid isCA CIx plots to show how SIDC results show diagnostic differences that may not be reflected in a CIx plot. Note that the CIx plot involves activation by CID only (no SID). |
6746321af9980725cf0f27bc | 12 | Streptavidin isCAโ-SIDC. Streptavidin is a tetramer that features a dimer-of-dimers topology (inset structure in Figure ). Thus, dissociation to dimers should be the preferred fragmentation pathway for native-like SID (Figure ). This is the case for SA, as seen in Figure , which reflects dimer as the primary dissociation pathway at low cone voltages. This trend is stable up to a cone voltage of approximately 90 V, at which point the normalized dimer intensity decreases as the cone voltage continues to increase. This suggests that either secondary fragmentation begins to occur when the sum of the isCA and SID activation exceeds a threshold, or the initial quaternary structure rearranges to a non-native-like structure that becomes apparent by the change in SID fragmentation and/or the change in drift time and product charge state. The CIx plot overlay in Figure and S2, suggest that the drift time is stable until a cone voltage of 120 -130 V. Between that point and the max cone voltage of 160 V, the drift time shifts to lower values, reflective of a more compact, non-native-like structure. Clearly, SID, as a diagnostic tool, is sensitive enough to detect excess energy deposition in the source, or rearranged quaternary structure, before ion mobility hints at the possibility that something has changed. As seen in Figure , as the dimer's contribution to normalized intensity decreases, the contribution from monomer and trimer increase. The slope of the monomer increase tracks with dimer decrease and is steeper than trimer increase, suggesting secondary fragmentation. The increase in trimer (and the portion of monomer produced from tetramer, not dimer) is a strong indicator that the weak subunit interfaces between dimer subunits, which are the most likely to be cleaved by SID, have been altered in a way that results in non-native SID, the mobiligram for which can be seen in Figure . |
6746321af9980725cf0f27bc | 13 | In comparison with SA's CIx plot (Figure and S2), SID was a more sensitive diagnostic for isCA than IM. The SID results showed significant shifts, due to post-SID secondary fragmentation, at lower cone voltages than were reflected in the CIx plot. Thus, reliance on IM to determine how harsh the ion source is, and how well the experiment conserved a protein complex's native-like structure, could possibly lead to erroneous conclusions about the true nature of a complex with unknown topology. Additional evidence for quaternary structure rearrangement by isCA is found in Figure , which shows the monomer product's average charge as a function of cone voltage. The statistically significant (*, p < 0.05) increase in average charge state suggests structural rearrangement that allows the monomer to abscond with more charge per mass than is typical under SID conditions. As demonstrated recently, this can happen if the monomer becomes elongated, as typical for CID. Despite the CIx plot showing a decrease in drift time of the tetramer, indicating quaternary structure collapse, the monomer's increasing average charge state suggests that rearrangement involves monomeric expansion. This process is then measured and/or enhanced during SID, allowing for higher charged monomer to result. C-reactive Protein isCAโ -SIDC. We also examined the 17+ charge state of CRP under the same cone conditions. In contrast to SA, CRP is a cyclic pentamer with a ring-like topology (structure inset in Figure ). The cavity in the middle provides volume for structural rearrangement not available to SA. IsCA clearly induces structural rearrangement (collapse), as seen in the overlay of Figure and, where the drift time starts to decrease at about 100 V. This is 20 -30 V lower than when SA's drift time starts to decrease. Ostensibly, this suggests that SA may possess a more robust quaternary structure than CRP. However, CRP's charge state is higher than SA's, which would result in more energetic in-source ion-neutral collisions during isCA. CRP is also physically larger than SA, causing the pentameric protein to undergo more collisions than SA at the same source pressures. Regardless, isCA clearly induced structural rearrangement in CRP, as we've reported previously, and affected SID trends, as seen in Figure . |
6746321af9980725cf0f27bc | 14 | CRP has a pentamer-of-monomers topology with equal interfacial areas between all subunits. This suggests that possible SID products include complementary pairs of monomer/tetramer and dimer/trimer, depending on the collision energy. At an SID energy of 765 eV, at low levels of isCA, dimer/trimer formation was the primary dissociation pathway, with monomer/tetramer also present, in line with previous reports from the Wysocki lab. This strongly suggests that the cyclic quaternary structure is intact and native-like. |
6746321af9980725cf0f27bc | 15 | As isCA becomes more energetic, the observed dissociation pathways change to favor monomer formation. An interesting region of the plot is between 80 -110 V. In this region the monomer and dimer signals increase while the trimer and tetramer signals gradually decrease, and the residual pentamer signal dramatically decreases. These trends suggest post-SID secondary fragmentation, likely due to excess energy from isCA. As the ions undergo isCA, an amount of translational energy is converted to internal energy, leaving the ions internally hot. When these hot ions undergo SID, the sudden energy deposition results in native-like SID (dimer/trimer and monomer/tetramer formation), but the extra energy from isCA then drives the oligomeric product ions to dissociate further: trimer to monomer/dimer and tetramer to dimer/dimer and monomer/trimer. Since these secondary dissociation events have several pathways to monomer and dimer we see the commensurate rise in those respective signals. The trimer signal decreases overall since a portion of the initial trimer population dissociates to monomer/dimer while a new trimer population emerges since the initial tetramer product can form trimer as part of its secondary dissociation. This region shows how high amounts of isCA can have a dramatic effect on downstream tandem MS analyses, which would not have been apparent without carefully scanning isCA at a fixed SID potential. The end of this region coincides with the initial drift time shift in the overlayed CIx plot in Figure , which suggests that quaternary structure rearrangement has been minimal and that increased internal energy from isCA is driving the shift in SID trends. At isCA voltages above 110 V, the monomer product becomes dominant, which aligns with the dramatic drift time shift in the CIx plot which indicates a collapsing quaternary structure. |
6746321af9980725cf0f27bc | 16 | Figure tracks CRP's SID monomer product average charge state as a function of cone voltage. Evident from the plot is the statistically significant (*, p < 0.05) increase in monomer average charge state with increased isCA, which agrees with similar work from the Wysocki lab in 2015, performed on a Waters Synapt G2. The interpretation given at that time is that the cone isCA process, at 160 V, both rearranges the protein such that the measured CCS is smaller (shorter drift time in our data), while also causing a monomeric subunit to partially elongate but remain non-covalently bound to the remaining subunits. The SID process then allows that elongated monomer to abstract additional charge as it leaves the protein complex. This is also consistent with what we observed for SA. Returning to the interesting regions between 80 -110 V, the average monomer charge state increases at a faster rate than it did between 10 -80 V. Based on the discussion above, this could be due to the increase in secondary fragmentation from energy deposition during isCA followed by SID. As the CIx plot reflects quaternary structure collapse, the average monomer charge state appears to approach a rollover point to a constant value. Unfortunately, this rollover point is not completed due to the upper limit on the cone voltage. |
6746321af9980725cf0f27bc | 17 | Phosphorylase B isCAโ-SIDC. We also included PhosB, a homodimer whose CIx data are presented in Figure and S4, and SID data in Figure . In terms of CIx, the plot only shows a slight decrease in the drift time as a function of cone voltage, otherwise there are no distinct features that suggest quaternary structure rearrangement. This drift time decrease does suggest mass loss, which is confirmed when the deconvolved and centroided mass spectra collected with cone settings of 10 V and 160 V are compared for the 21+ charge state. The spectrum collected at a cone = 100 V features a deconvolved mass that is ~156 Da lower than that collected at cone = 10 V. At the highest cone voltage, 160 V, the deconvolved mass is ~300 Da lower than that collected at cone = 10 V. While structural rearrangement is not evident, the higher cone voltage clearly helped with in-source protein cleanup, i.e., loss of adducts such as (H2O), (NH4+) and (CH3COO-). |
6746321af9980725cf0f27bc | 18 | In terms of SID, the instrument has a total voltage limit of 300 V across all activating regions, including the cone voltage, SID/Trap, and Transfer regions. Thus, with an SID voltage of 190 V, the cone voltage could only be tuned up to ~100 V. Hence, the x-axis of Figure does not reach the same cone voltage as used in Figure . For SID of PhosB, monomer formation will be the favored dissociation pathway, if the SID energy is sufficiently high. Previous work in our lab found that the ratio of product monomer and surviving precursor dimer is about 50:50, even at the highest SID energies. Taking that into account, and factoring in the CIx results, we anticipated that increased isCA would not affect the SID trend since the only SID product is the monomer subunit. As seen in Figure , this is indeed the case. Regardless of the cone voltage, no significant changes occur in the SID trend. In striking comparison to the results obtained for SA, and CRP, this SID trend suggests no measurable structural rearrangement took place. Figure . A) 21+ PhosB isCAโ -SIDC (3990 eV) ERMS plot with overlayed CIx plot. The implied y-axis of the overlay is drift time. B) Monomer average charge state. Increased levels of isCA do not have a statistically significant effect on SID pathways nor on the monomer's subsequent charge state. The inset depicts the PhosB dimer (PDB 1GPB). |
6746321af9980725cf0f27bc | 19 | Additional evidence that suggests minimal to no quaternary structure rearrangement is found in the monomer product's average charge. As depicted in Figure , the average charge remains steady as a function of cone voltage. This was somewhat surprising, since isCA could have caused the monomeric subunits to elongate in a way that more charge could be placed on the monomers during SID, even if the SID trend was not disrupted. This apparently did not happen. To confirm this, the mobiligrams at each cone voltage were compared to each other and no difference in mobility arrival time was observed as a function of cone voltage. Given the fact that isCA was limited to low voltages when coupled to SID, it is unsurprising that no quaternary structure rearrangement was observed for PhosB. Figure . A) 15+ TRAP isCAโ -SIDC (1050 eV) plot for a few indicated SID products, the remainder of which are found in Figure ; the overlay is the TRAP 15+ CIx plot; B) SID monomer product average charge state, which shows no statistically significant trend. C) SID product comparison at 1050 eV of SID energy for the 15+ at two distinct cone voltages, and the 21+ at a low cone voltage. The left inset is the calibrated CCS for the indicated precursor charge states and their cone voltage settings. The CCS for the 15+ at high cone voltage and the 21+ are highly similar, suggesting they both feature a compact structure. The right inset structure depicts the TRAP bst 11mer (PDB 1GTN), with the tryptophan ligands removed. |
6746321af9980725cf0f27bc | 20 | TRAP isCAโ -SIDC. In 2005 Ruotolo, Robinson, and coworkers published a convincing report that native-like protein quaternary structure could be maintained in the gas phase after nESI from ammonium acetate. The report was founded on ion mobility measurements for TRAP from Bacillus subtilus. Careful CCS calibration and comparison with predicted CCS values for model structures indicated that the 21+ charge state likely featured a collapsed, globular, 11mer structure while the 19+ charge state was likely an intact cyclic ring. Since then, the Wysocki lab has published SID data comparing the 21+, 15+, and 16-charge states. The 21+ and 15+ had distinctly different SID fingerprints, with the latter showing native-like fragmentation for a ring-like quaternary structure. The 16-had similar features to the 15+. This distinct SID behavior persuaded us to include TRAP here as well, to illustrate the effect of overly energetic isCA on an undecameric protein complex . |
6746321af9980725cf0f27bc | 21 | However, the drift time decreases to a new stable feature at cone voltages above 120 V. This suggests that isCA causes the cyclic protein complex to collapse. The ultimate effect of isCA on SID is seen in Figure , which shows a few selected SID products. The full plot with all SID products can be seen in Figure . At cone voltages below 60 V and with a fixed SID energy of 1050 eV, the most prevalent SID products are the monomer, pentamer, and hexamer subunits, followed by residual 11mer precursor. As seen in Figure (blue bars), the other complementary oligomeric pairs (1/10, 2/9, 3/8, 4/7) seem to have complementary signal intensities, which should be due to native-like fragmentation. As isCA becomes more aggressive Figure shows a significant shift in SID behavior at a cone voltage of 50 -70 V, a trend that seems to level off by a cone voltage of 100 V, when the CIx plot barely starts to suggest the possibility of structural rearrangement. The monomer signal is this region, cone voltage 50 -100 V, appears similar to the monomer trend for SA (Figure ) and CRP (Figure ), which suggests that secondary fragmentation due to energy deposition from isCA and SID is at play here as well. As seen in Figure (red bars) at the highest cone voltage (most severe isCA), the SID product distribution shifts towards lower oligomeric states, more reflective of non-native fragmentation behavior as captured for the 21+ charge state at the same SID energy (Figure gold bars). |
6746321af9980725cf0f27bc | 22 | As shown in Figure , the SID monomer product's average charge shows limited variation with increasing isCA, as opposed to the increasing charge states with increasing isCA observed for SA, AV, and CRP, and previously reported by our lab. Figure compares SID product intensities of 15+ TRAP at two different cone voltages, and the 21+ charge state at low cone voltage. Based on the above mentioned papers, it was reasonable to expect that the 21+ charge state (purple bars) would feature non-native-like SID. This is clearly reflected in Figure , where the SID oligomeric product distribution is shifted towards lower oligomeric states, whereas the 15+ charge state at low cone voltage (dark blue bars) appears to have more balance across the oligomeric states, reflecting native-like partitioning when two interfaces are cleaved in an 11mer protein complex. The 15+ charge state at the highest cone voltage (light blue bars) is shifted to favor lower oligomeric states, similar to the 21+ charge state. This clearly shows that a high amount of isCA caused 15+ TRAP to adopt a collapsed quaternary structure, resulting in non-native-like SID. |
6746321af9980725cf0f27bc | 23 | Lastly, the inset of Figure is the CCS distribution for the same three precursor ion conditions (charge states 15+ at cone 10 and 160V, and charge state 21+ at cone 10V). Because we are comparing two different charge states, we felt it appropriate to calibrate TRAP's CCS. We did so using IMSCal, and a CCS database composed of protein complexes under native and charge-reducing conditions. The calibrants selected included: alcohol dehydrogenase, c-reactive protein, and streptavidin, which were prepared as described for other proteins complexes above. These calibrants were sprayed from 200 mM ammonium acetate and 160/40 mM ammonium acetate/TEAA. |
6746321af9980725cf0f27bc | 24 | As seen in the inset of Figure , there is a size difference between the 15+ and 21+ at a cone voltage of 10 V that suggests the 15+ is more expanded, perhaps cyclic, while the 21+ is indeed globular. When the 15+ complex undergoes isCA at a cone voltage of 160 V, the CCS decreases to a range that matches the CCS for the globular 21+. One possible reason for this match is that isCA caused the 15+ to collapse from a ring-like structure to a globular structure, similar to that possessed by the 21+. We note that in comparison to the CCS values published by the Robinson lab in 2005, our CCS values for TRAP are smaller by about 9%. This could be due to the advancement that traveling wave ion mobility has experienced since 2005, especially in terms of better understanding of its physical implementation, theoretical underpinnings, 72 application, and improved experimental CCS calibration. Experiment Type: isCA-IM-ECD To explain why we observed these SID trends as a function of isCA, we find it helpful to think about the subunit interfaces within the protein complexes. Previous work from our lab showed that SID will cause native-like protein complexes to dissociate along their weakest subunit interfaces. By this we mean the interfaces with the lowest interface area, fewest salt bridges, fewest hydrogen bonds, and hydrophobic surface area. When the protein's quaternary structure is disrupted by isCA, it is very likely that the subunit interfacial areas, salt bridges, and hydrogen bonds were disrupted along the way to low energy, non-native-like structures. These rearranged interfaces and interactions then influenced subsequent SID pathways. |
6746321af9980725cf0f27bc | 25 | To characterize how subunit interfaces were possibly rearranged by isCA, we employed top-down electron capture dissociation (TD ECD). ECD cleaves the peptide N-Cฮฑ bond, typically resulting in c/z product ion types. For native-like proteins, the released product ions can be indicative of protein secondary, tertiary, and even quaternary structure. Additionally, the obtained products can be sensitive to pre-ECD collisional activation, thus this technique can also diagnose rearranged protein structures Figures S9 and S10 show SA's ECD sequence coverage at cone voltages of 20 V and 160 V, respectively. We note that the depicted sequence includes an N-terminal methionine residue. The SA commercially available from Thermo Scientific does not have all N-terminal methionine residues cleaved, which is evident in the mass spectrum. Because the tetramer likely forms in a stochastic manner, there are mass spectral peaks that correspond to 0, 1, 2, 3, and 4 conserved methionine residues on the complex, with individual monomer subunits carrying 0 or 1 Met. Thus, we used both possible sequences, with and without the leading Met, but choose to display the ECD data featuring the leading Met. Cleavage sites were similar between the two protein sequences. |
6746321af9980725cf0f27bc | 26 | Using Proteins, Interfaces, Structures and Assemblies (PISA) (), an interface analysis was performed on PDB 1SWB, the SA tetramer. Identified interfacial residues are located within SA's anti-parallel ฮฒ-sheet. In Figures and these residues are highlighted in tan. The residues between the highlighted regions belong to the ฮฒ-sheet's hairpin turns and were not included as interfacial residues by PISA. Returning to Figures and, a cone voltage of 20 V resulted in 24% sequence coverage, not including internal fragments. The sequence coverage was 53% at a cone voltage of 160 V. Since the SID data indicate that isCA clearly rearranges the quaternary structure, possibly by causing monomer subunits to elongate, the uptick in sequence coverage was unsurprising. Next, we should note that the increased sequence coverage came from the N-terminus, and the interface between subunits, especially between residues 75 -120. This strongly suggests that isCA disrupted protein tertiary structure by elongating the N-terminal domain and rearranged the quaternary structure along the strong monomer-monomer interfacial areas. This evidence validates our conclusion that the SID results in Figure reflect interfacial changes between subunits, which was also suggested by the increase in average monomer charge state. |
6746321af9980725cf0f27bc | 27 | ECD analysis at cone voltages of 20 and 160 V was also attempted for 17+ CRP, 21+ PhosB, and 15+ TRAP, however, no fragments were obtained for these complexes. We did observe extensive gas phase electron capture charge reduction (ECCR) with the average charge state being 12+ for CRP, with a precursor charge state range of 8+ through 17+. For PhosB, the average charge was 14+ with a range encompassing 7+ through the 21+ precursor. The average charge for TRAP decreased to 12+, with a range of 6+ to 15+. We did attempt to tune the Transfer collision energy and Transfer cell pressure to release ECD fragments, but those efforts did not result in ECD fragments being released from the protein complexes. We also attempted to tune the transfer RF amplitude to help capture possible ECD fragments, but to no avail. As we checked PDB 1GNH for CRP we found that the N-terminus is solvent accessible in the native structure while the C-terminus is buried in an interfacial interaction. For PDB 1GPB for PhosB both N-and C-termini are solvent accessible, while 1GTN for TRAP features N-and Ctermini that are involved in interfacial interactions, some of which are present as hydrogen bonds. Because charge reduction without observed fragmentation was observed for CRP, PhosB, and TRAP, they can be classified as 'group II' following the Loo lab's designation scheme, which suggests that inter-and intra-subunit interactions limit the number of fragments that can be obtained after ECD. For CRP, whose N-terminus was solvent accessible, it is possible that ionization into the gas phase resulted in the N-terminus becoming protected by new non-covalent interactions during desolvation, thereby limiting c-type fragmentation. Additionally, we suppose that CRP's higher mass, compared to SA, could have contributed to low kinetic energy that Transfer CID could not overcome to liberate fragments. Thus, a heavier collision gas, such as argon or xenon, could increase the collision energy enough to free the ECD fragments from the complex or AI-ETD could be attempted. |
6746321af9980725cf0f27bc | 28 | This paper defines conditions to be used to keep protein complexes "native-like" in a cyclic IMS (q-IM-TOF) modified to include SID and ExD capabilities. Using Ubq 6+, with a goal of achieving a compact conformation that is considered in the literature to be native-like, we tuned the pre-IM voltage gradients to be as non-activating as possible. During this process we identified the post-Trap ion guide bias and the helium cell as the most activating regions, while adjusting the StepWave voltages resulted in minimal CCS changes. While such tuning was ideal to maintain Ubq's native-like structure, the shallow voltage gradients resulted in poor SID product transmission for the larger standard protein complex CRP. We then systematically increased the post-Trap ion guide bias and the helium cell entrance, exit, and bias voltages to obtain good SID product ion transmission for this larger, more structurally stable protein complex, noting that the new conditions did not cause non-native fragmentation behavior. Taking lessons from Ubq, we maintained the instrument's source, including the StepWave and cone voltage at low values as an added measure to preserve native-like structure for a range of standard protein complexes. We then systematically demonstrated that using the cone voltage setting for isCA to remove non-covalent adducts from protein complexes can result in both internal heating resulting in secondary fragmentation and, at higher energies, quaternary structure rearrangement. These effects were illustrated by using SID product ion distributions, CIx mobiligrams, and ECD. Secondary fragmentation, resulting from the additive nature of isCA and SID, altered fragmentation patterns and highlighted the importance of minimizing ion activation prior to the desired tandem MS steps even at isCA energies where drift time shift is not yet apparent. The approach reported here, that uses SID, CIx, and/or ECD of standard protein complexes, will be used to characterize native-like behavior in other instrument platforms in the future, to establish a standard "native" tuning protocol for new instrument acquisitions, released instrument modifications, or new practitioners of native MS. |
6746321af9980725cf0f27bc | 29 | The Supporting Information is available free of charge on the ACS Publications website. SID energy-resolved mass spectrometry plot for CRP 18+; averaged CIx plots for SA, CRP, PhosB, and TRAP and their respective RMSD plots, sample SID mobiligrams illustrating native and nonnative SID results, ECD sequence maps for SA at different cone voltages (docx) |
655d1092dbd7c8b54bbca93d | 0 | Hybrid organic-inorganic lead halide perovskites (i.e., methylammonium lead trihalide, CH3NH3PbX3 also known as MAPbX3, where MA = methylammonium or CH3NH3, and X = Cl, Br, or I) have garnered significant attention involving electronic, optoelectronic, laser, lightemitting diode, photodetector, and X-ray/ฮณ-ray detector device technologies. This is largely due to tunable process-dependent compositions, structures, and morphologies resulting in a wide range of material characteristics, such as changes in the bandgap, carrier mobility, and exciton binding energy. Additionally, the potential for adaptable low-cost manufacturing (compared to state-ofthe-art commercial technologies) make perovskites appealing for high-production through-put and scale-up manufacturing advancements. However, material and device development challenges impact the broader adoption of perovskite material integration at scale-including stability and durability limitations, material quality and property performance reproducibility, environmental or defect degradation, induced toxicity, and controlled synthetic scale-up methodologies. As such, it is common to observe a wide range of reported material properties that impact the electrical responses and optical characteristics of MAPbX3. Such structure-property-processing effects complicate electronic and optoelectronic device design and development optimization strategies for commercialization. The most prominent application focus for halide perovskites is the development of highperformance, low-cost solar cells. Hybrid organic-inorganic lead halide perovskite solar cells (PSCs) have been shown to achieve a power conversion efficiency (PCE) of over 25%, which rivals state-of-the-art high-performance silicon solar cells (โฅ25% PCE). Critical to the development of PSCs are the optical properties (refractive index, n, and extinction coefficient, k) of the halide perovskite active layer, which describe the way materials interact with light and subsequently function as a photovoltaic material. Proper derivation of perovskite optical constants enables the optimized design of efficient solar cells. While there are many considerations when designing an exemplary PSC (e.g., compositional bandgap tuning, surface nano or microstructured light trapping, multilayer tandem cell architectures, and enhanced charge carrier extraction), perhaps the most important device consideration is the perovskite layer absorptance within the given solar cell architecture. Perovskite optical absorptance in a solar cell device directly impacts the efficacy of solar radiation absorption for an engineered PSC and the resulting PCE. |
655d1092dbd7c8b54bbca93d | 1 | The n and k for MAPbX3 have been observed to vary between representative processing methodologies to date. If such process-induced optical dispersion variation is ignored at the design stage, PSCs can exhibit undesired parasitic optical absorptance by non-perovskite (non-active) layers resulting in variable optical transmission, reflection, or haze-like characteristics within solar radiation bands of interest (i.e., ultraviolet = 300-400 nm, visible = 400-700 nm, and infrared = 700-2500 nm). To our knowledge, no comprehensive experimentally appended optical dispersion data analysis to derive the optical constants of MAPbX3 has been performed to include notable anomalous optical dispersion characteristics including the sharp band edge transition (bandgap ~539 nm) and methylammonium spectral overtones below the bandedge (observed around 1050-2500 nm). As such, the optical property determination of exemplary MAPbX3 has been largely under approximated and, consequently, evokes uncertainty with respect to optical properties reported in prior work (discussed below). |
655d1092dbd7c8b54bbca93d | 2 | Fundamental derivation of MAPbX3 optical constants is paramount toward the development of high-performance electronic and optoelectronic perovskite devices. In this report, we present a rigorous optical dispersion data analysis of single crystal MAPbBr3 via variable angle spectroscopic ellipsometry appended with transmission intensity data. This results in a robust derivation of MAPbBr3 optical constants for normal and anomalous optical dispersion regimes. Using the derived optical constants for our single crystal MAPbBr3, exemplary modeled solar cell optical device designs are optimized for perovskite layer absorptance. Our optimized PSC designs demonstrate improved optical performance in comparison to designs prepared using illustrative MAPbBr3 literature optical constants reported to date. |
655d1092dbd7c8b54bbca93d | 3 | The MAPbBr3 crystals characterized in this work were grown following a modified inverse temperature crystallization (ITC) method reported in the literature. An illustration of the growth process is shown in Figure . A 1:1 molar ratio of PbBr2 and MABr was dissolved in DMF. The precursor solution was allowed to stir at room temperature for 24 hours to ensure that all starting materials dissolved. The precursor solution was then passed through a 0.2 ฮผm PTFE syringe filter and added directly to a glass crystallizing dish on a hot plate. Small seed crystals were grown in an initial crystal growth step. Once the seeds had grown to an appreciable size for ease of handling (~1ร1ร0.5 mm), they were harvested from the growth solution and transferred to a separate glass crystallizing dish containing fresh precursor solution for seeded growth propagation. Seeded crystal growth was performed on a glass substrate placed at the bottom of the crystallizing dish to achieve a flat crystal surface at the glass-crystal interface. The growth was allowed to propagate for approximately five days uninterrupted. Once the growth was complete, a ~2.7ร2.7ร0.4 cm crystal (Figure ) was removed from the solution and dried with a Kimwipe prior to further analysis. Figure shows an illustration of the cubic MAPbBr3 unit cell where the gray spheres at the center of the octahedra represent Pb 2+ , the green spheres on the corners represent the halide (X -), and the blue sphere at the center of the unit cell is the organic cation (A + ). |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.