TiSe2 is a narrow-gap insulator with a rich array of unique properties. In addition to being a superconductor under certain modifications, it is commonly thought to be a rare realisation of an excitonic insulator. Below 200 K, TiSe2 undergoes a transition from a high-symmetry (P3m1\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P\bar{3}m1$$\end{document}) phase to a low-symmetry (P3c1\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P\bar{3}c1$$\end{document}) charge density wave (CDW). Here we establish that it is indeed an insulator in both P3m1\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P\bar{3}m1$$\end{document} and P3c1\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P\bar{3}c1$$\end{document} phases. However, the insulating state is driven not by excitonic effects but by symmetry-breaking. In the CDW phase it is static. At high temperature, thermally driven instantaneous deviations from P3m1\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P\bar{3}m1$$\end{document} break the symmetry on the characteristic time scale of a phonon. Even though the time-averaged lattice structure assumes P3m1\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P\bar{3}m1$$\end{document} symmetry, the time-averaged energy band structure is closer to the CDW phase - a rare instance of a metal-insulator transition induced by dynamical symmetry breaking. We establish these conclusions from quasiparticle self-consistent GW (QSGW) and many-body calculations (QSGW\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$G\widehat{W}$$\end{document}), in combination with molecular dynamics simulations to capture the effects of thermal disorder. The many-body theory includes explicitly ladder diagrams in the polarizability, which incorporates excitonic effects in an ab initio manner. We find that the excitonic modification to the potential is weak, ruling out the possibility that TiSe2 is an excitonic insulator.

The localization of electrons caused by atomic disorder is a well-known phenomenon. However, what circumstances allow electrons to remain delocalized and retain band-like characteristics even when the crystal structure is completely absent, as found in certain amorphous solids, is less well understood. To probe this phenomenon, we developed a fully first-principles description of the electronic structure and charge transport in amorphous solids by combining a novel representation of the amorphous state with the state-of-the-art many-body (QSGW) electronic structure theory. Using amorphous In2O3 as an example, we demonstrate the accuracy of our approach in reproducing the band-like nature of the conduction electrons as well as their disorder-limited mobility. Our approach reveals the physical origins responsible for the electron delocalization and the survival of the band dispersions despite the absence of long-range order.

The interactions between substituted arylvinyl phosphonic acid (AVPA) ligands within a Eu-AVPA complex are shown to influence the outcomes of excited state evolution after photoexcitation. Compared with unfunctionalized AVPAs, pairs of ligands functionalized with CF3 in the para position preassociate in the ground state of complexes with Eu3+ according to calculated geometry optimizations. The CF3-substituted AVPA complexes show evidence of red-shifted optical absorption and undergo more efficient excimer formation, as revealed by transient absorption spectroscopy. We rationalize this behavior through simulations of excited-state geometry optimizations that reveal evolution toward interligand phenyl-phenyl planarity for specific excited states. Emission from complexed Eu3+ after energy transfer from the ligand is found to be weaker with CF3 substitution, which we hypothesize is due to intracomplex, interligand aggregates with excimer-promoting geometries. These observations point to the need to consider ground-state geometries as well as dynamic excited-state processes to understand the flow of energy in rare earth coordination complexes.

TiSe$_2$ is thought to be an insulator with a bandgap of ~0.1eV. It has attracted a much interest because, among of a rich array of unique properties, many have thought TiSe$_2$ is a rare realisation of an excitonic insulator. Below 200 K, TiSe$_2$ undergoes a transition from a high-symmetry ({P-3m1}) phase to a low-symmetry ({P-3c1}) phase. Here we establish that TiSe$_2$ is indeed an insulator in both {P-3m1} and {P-3c1} phases. However, the insulating state is driven not by excitonic effects but by symmetry-breaking of the {P-3m1} phase. In the CDW phase the symmetry breaking is static. At high temperature, thermally driven instantaneous deviations from {P-3m1} break the symmetry on the characteristic time scale of a phonon. Even while the time-averaged \emph{lattice} structure assumes {P-3m1} symmetry, the time-averaged \emph{energy band} structure is closer to the CDW phase -- a rare instance of a metal-insulator transition induced by dynamical symmetry breaking. We establish these conclusions from a high-fidelity, self-consistent form of many body perturbation theory, in combination with molecular dynamics simulations to capture the effects of thermal disorder. The many-body theory includes explicitly ladder diagrams in the polarizability, which incorporates excitonic effects in an \emph{ab initio} manner. The excitonic modification to the potential is slight, ruling out the possibility that TiSe$_2$ is an excitonic insulator. Charge self-consistency is essential distinguish the metallic from insulating state.

Anion exchange membrane electrolysis enables the use of earth abundant metals for the catalyst and other components of the membrane electrode assembly. However, earth abundant metal catalysts may be sensitive to poisoning and degradation due to the extreme pH of the electrolyte or interactions with the ionomer. Plane-wave density functional theory calculations probed the binding strength of different functional groups from popular ionomers (Nafion, Sustainion, tetramethylammonium-based ETFE/Gen 2/Georgia Tech, Versogen) on model catalysts such as Platinum Group Metal IrO2 and earth abundant NiO.1 We evaluated the stability of these various ionomers on these model catalysts, whether they remain intact or degrade, as well as their catalytic effects on the oxygen evolution reaction (OER) e.g. in the presence of OH*. These calculations allowed us to understand the poisoning of active sites: 1) if the functional group bound more strongly than key reaction intermediates, then the ionomer effectively blocked activity by displacing the reaction intermediate; 2) if the functional group introduced competing reactions, then the ionomer was consuming the OxHy* intermediates necessary for OER. Ideally, ionomers remain stable in contact with the catalyst and shuttle hydroxide ions from the cathode to the anode. 1. S. Ghoshal, B. S. Pivovar and S. M. Alia, J. Power Sources, 2021, 488, 229433. Figure 1

TiSe$_2$ is thought to be an insulator with a bandgap of ~0.1eV. It has attracted a much interest because, among of a rich array of unique properties, many have thought TiSe$_2$ is a rare realisation of an excitonic insulator. Below 200 K, TiSe$_2$ undergoes a transition from a high-symmetry ({P-3m1}) phase to a low-symmetry ({P-3c1}) phase. Here we establish that TiSe$_2$ is indeed an insulator in both {P-3m1} and {P-3c1} phases. However, the insulating state is driven not by excitonic effects but by symmetry-breaking of the {P-3m1} phase. In the CDW phase the symmetry breaking is static. At high temperature, thermally driven instantaneous deviations from {P-3m1} break the symmetry on the characteristic time scale of a phonon. Even while the time-averaged \emph{lattice} structure assumes {P-3m1} symmetry, the time-averaged \emph{energy band} structure is closer to the CDW phase -- a rare instance of a metal-insulator transition induced by dynamical symmetry breaking. We establish these conclusions from a high-fidelity, self-consistent form of many body perturbation theory, in combination with molecular dynamics simulations to capture the effects of thermal disorder. The many-body theory includes explicitly ladder diagrams in the polarizability, which incorporates excitonic effects in an \emph{ab initio} manner. The excitonic modification to the potential is slight, ruling out the possibility that TiSe$_2$ is an excitonic insulator. Charge self-consistency is essential distinguish the metallic from insulating state.

Manifesting chemical differences in individual rare earth (RE) element complexes is challenging due to the similar sizes of the tripositive cations and the corelike 4f shell. We disclose a new strategy for differentiating between similarly sized Dy3+ and Y3+ ions through a tailored photochemical reaction of their isostructural complexes in which the f-electron states of Dy3+ act as an energy sink. Complexes RE(hfac)3(NMMO)2 (RE = Dy (2-Dy) and Y (2-Y), hfac = hexafluoroacetylacetonate, and NMMO = N-methylmorpholine-N-oxide) showed variable rates of oxygen atom transfer (OAT) to triphenylphosphine under ultraviolet (UV) irradiation, as monitored by 1H and 19F NMR spectroscopies. Ultrafast transient absorption spectroscopy (TAS) identified the excited state(s) responsible for the photochemical OAT reaction or lack thereof. Competing sensitization pathways leading to excited-state deactivation in 2-Dy through energy transfer to the 4f electron manifold ultimately slows the OAT reaction at this metal cation. The measured rate differences between the open-shell Dy3+ and closed-shell Y3+ complexes demonstrate that using established principles of 4f ion sensitization may deliver new, selective modalities for differentiating the RE elements that do not depend on cation size.

Ligands that can discriminate between individual rare earth elements are important for production of these critical elements. A set of aryl-vinyl phosphonic acid ligands for extracting rare earth elements were designed and synthesized under the hypothesis that the strength of the rare earth-ligand interactions could be tuned by changing the dipole moment of the ligand. The ligands were synthesized via a two-step reaction procedure using a Heck coupling reaction to functionalize vinyl phosphonic acid, followed by Steglich esterification to obtain high-purity styryl phosphonic acid monoesters with varying dipole moments along the P-C bond. The metal binding strength and composition of the rare earth complexes formed with these styryl phosphonic acid monoesters were experimentally studied by liquid-liquid extraction techniques, while DFT calculations were performed to determine the dipole moments of the free and complexed ligands and the electronic structure of the complexes formed. All three prepared ligands were much stronger extracting agents for europium(III) than the dialkylphosphonic acids usually used for this separation. However, the order of increasing extraction strength was found to match the order of the decreasing calculated dipole moment along the P-C bond of the three styryl-based ligands, rather than correlating with increasing ligand basicity, as reflected by the pK a of the ligands. These findings suggest that this approach can be used to systematically alter the extraction strength of aromatic phosphonic monoesters for rare earth element purification.

Density functional theory calculations in conjunction with statistical mechanical arguments are performed on the rutile IrO2 (110) facet in order to characterize multiple reaction pathways on the surface at the highest active limit (the stoichiometric surface with all metal sites available) and at the lowest active limit (the oxygen-terminated surface). Alternative pathways to the oxygen evolution reaction (OER) are found, with multiple pathways determined at each step of the four proton-coupled electron transfer reaction. Of particular interest is the detailed characterization of a co-adsorption pathway utilizing neighboring, adsorbed O, OH species in order to evolve oxygen; activation energies of this pathway are <0.5 eV and therefore easily surmountable at the high operating potentials of OER. We also determined that surface Ir atoms can potentially participate in deprotonating an OOH* intermediate; the activation energy to this is 0.67 eV on the oxygen-terminated surface. These theoretical findings explain in part the high activity present in iridium oxide catalysts and also provide insight into the mechanistic pathways available on metal oxide catalysts, which may require the concerted interaction of nearest neighbor co-adsorbates to produce chemicals of interest.

Although electricity feedstock currently dominates hydrogen production costs in commercial electrolysis, capital cost will become a significant factor as electrochemical water splitting is directly coupled with low-cost power sources. [1,2] To minimize device costs and reach cost targets, reducing catalyst loading in proton exchange membrane-based (PEM) electrolysis will be necessary and efforts to evaluate and improve upon catalyst performance and durability will become critical. Beyond PEM-based systems, anion exchange membrane-based (AEM) electrolyzers can also reduce capital cost, with the high pH enabling non-platinum group metal (PGM) catalysts and improving the durability of system components. In these areas, efforts at the National Renewable Energy Laboratory have included developing and evaluating electrolyzer catalysts, establishing baseline performance and durability, and linking ex- and in-situ testing for commercial nanoparticles and novel materials. In PEM electrolysis, ink compositions and coating methodologies were evaluated in rotating disk electrodes, establishing best practices for screening catalysts in the oxygen evolution reaction. [3] Test factors were considered, including working electrode dissolution and catalyst delamination, and electrochemical surface area measurements adapted to differentiate between site quantity and quality. While half- and single-cell performances and durabilities do not match, higher activities translated to kinetic improvements in membrane electrode assemblies and half-cell testing was a reasonable tool for assessing relative differences between material sets. Projecting differences at the device level, however, required segregating materials based on surface composition and oxide content, and some caution is needed when using rotating disk electrodes to assess catalysts for electrolysis applications. In membrane electrode assemblies, baseline durability was evaluated when accounting for low catalyst loading and intermittent load profiles. How catalyst layers are incorporated has a significant impact on device performance and durability, and efforts have been made to translate improvements from spray coating to roll to roll coating and manufacturing appropriate processes. Various catalyst and membrane combinations have been used to assess that ability of component development and system control to limit performance loss with extended operation. In AEM electrolysis, baselines and best practices have been established in rotating disk electrodes for catalysts in the oxygen and hydrogen evolution reactions. Test factors, including electrolyte purity, conditioning protocols, and counter electrode choices were found to significantly impact measurements, and some care is needed to avoid under- or over-estimating kinetic improvements. Novel catalysts were developed for the hydrogen and oxygen evolution reactions, including low- and non-PGMs, sulfides, and metal organic frameworks, where activity improvements ex-situ generally translated to membrane electrode assemblies. [4] In in-situ testing, ionomers were varied with standard catalysts to assess the role of catalyst-ionomer interactions and supporting electrolytes in device performance. [1] H2 at Scale: Deeply Decarbonizing our Energy System. Presented at Annual Merit Review, U.S. Department of Energy; Washington, DC, June 6−10, 2016. https://www.hydrogen.energy.gov/pdfs/review16/2016_amr_h2_at_scale.pdf. [2] Denholm, P.; O’Connell, M.; Brinkman, G.; Jorgenson, J. Overgeneration from Solar Energy in California: A Field Guide to the Duck Chart; Vol. NREL/TP-6A20-65023; National Renewable Energy Laboratory: Golden, CO, 2015. http://www.nrel.gov/docs/fy16osti/65023.pdf. [3] S. M. Alia and G. C. Anderson, J. Electrochem. Soc., 166, F282 (2019). [4] S. Ghoshal, S. Zaccarine, G. C. Anderson, M. B. Martinez, K. E. Hurst, S. Pylypenko, B. S. Pivovar and S. M. Alia, ACS Applied Energy Materials, 2, 5568 (2019).

H2 remains an important chemical precursor in processes such as the Haber-Bosch process to synthesize ammonia and hydrocracking for refining of hydrocarbons and also exists as an alternative fuel for transportation. Currently, 95% of H2 is produced by steam methane reforming, whose byproduct is pollutant CO. Electrolysis is a clean, efficient method run at room temperature to split water into H2and O2. However, electrolysis is not cost-competitive with steam methane reforming, utilizing large amounts of platinum group materials. Computational chemistry can give atomic and electronic levels of understanding of a material by identifying highly active facets of a known material or predicting the catalytic potential of unknown materials. We will model catalysts used in the oxygen evolution reaction, the half-cell reaction requiring the most catalyst and occurring many orders of magnitude slower than the hydrogen evolution reaction. We will focus on the effect of the ensemble: modeling different facets of iridium oxide, characterizing all possible configurations of key reaction intermediates, and incorporating explicit solvation to fully understand the nuances of optimizing a premier catalyst such as IrO2. In particular, we will focus our calculations on two facets: (110), the more thermodynamically stable facet and (001), a potentially more active facet due to the presence of undercoordinated, surface Ir sites. We will compare the reaction profile of the ensemble of key reaction intermediates on the (110) versus the (001) facet and in vacuum versus with solvation. The ensemble of intermediates allows us to explore multiple reaction pathways on the surface that have not been characterized. Figure 1. A schematic summarizing the different isomers found at each reaction step (TOP) and the reaction profile of the oxygen evolution reaction on the (110) facet of iridium oxide (BOTTOM). Figure 1

Block cracking is commonly observed in form of a series of interconnected cracks that divide the surface of multilayered materials into approximately rectangular or square pieces. A three-dimensional (3D) elastic fracture model is developed to study the block cracking in thin film/substrate structures which consist of a thin protective metal oxide coating fully bonded to a polymer substrate. Under a temperature change, fracture occurs in the coating due to the thermal stress caused by the material mismatch. Using the plane assumption and non-shearing assumptions, the displacement and stress fields in this substrate-coating system under thermal loading are explicitly solved and verified with the finite element (FE) simulation results. Therefore, the energy release rate (ERR) can be calculated from the work done by the stress on the crack opening before crack and has been used to study the fracture initiation, infilling and saturation. Additionally, the theoretical fracture analysis results are verified by FE simulation using the cohesive zone model (CZM) and experimental data from literature. The results show that the fracture model presented in this study is able to capture the displacement and stress distributions in the thin hard oxide film fully bonded to a polymer substrate accurately and predicts the fracture initiation, infilling and saturation successfully.

This paper combines density functional theory calculations and electrochemical testing to study activity differences among iridium (Ir) surfaces in the oxygen evolution reaction. Ir metal/hydroxide is significantly more active than Ir oxide, which may be due to oxide skins at the surface weakening O-binding relative to pure metal or oxide surfaces. Here we report a disparity in activity between Ir and Ir oxide in half-cells not observed in single-cells. Extended operation at elevated temperature and potential were found to result in oxide growth, limiting how surface differences affect electrolyzer performance. Comparisons of half- and single-cell testing were used to assess how well rotating disk electrode testing predicts membrane electrode assembly performance and durability. Although oxygen evolution activities in half-cells can translate to single-cells, standard rotating disk electrode test procedures can exaggerate the activity benefit of a metal/hydroxide surface relative to membrane electrode assembly performance under typical operating conditions; it also appears that a half-cell test cannot reasonably accelerate activity loss from continual operation. While a variety of novel catalyst approaches, including alloying, faceting, morphology, and supports can improve oxygen evolution kinetics, these results suggest that Ir surfaces at different oxide states may struggle to improve performance at the device level. (C) The Author(s) 2019. Published by ECS.

Machine learning methods have shown promise in predicting molecular properties, and given sufficient training data, machine learning approaches can enable rapid high-throughput virtual screening of large libraries of compounds. Graph-based neural network architectures have emerged in recent years as the most successful approach for predictions based on molecular structure and have consistently achieved the best performance on benchmark quantum chemical datasets. However, these models have typically required optimized 3D structural information for the molecule to achieve the highest accuracy. These 3D geometries are costly to compute for high levels of theory, limiting the applicability and practicality of machine learning methods in high-throughput screening applications. In this study, we present a new database of candidate molecules for organic photovoltaic applications, comprising approximately 91 000 unique chemical structures. Compared to existing datasets, this dataset contains substantially larger molecules (up to 200 atoms) as well as extrapolated properties for long polymer chains. We show that message-passing neural networks trained with and without 3D structural information for these molecules achieve similar accuracy, comparable to state-of-the-art methods on existing benchmark datasets. These results therefore emphasize that for larger molecules with practical applications, near-optimal prediction results can be obtained without using optimized 3D geometry as an input. We further show that learned molecular representations can be leveraged to reduce the training data required to transfer predictions to a new density functional theory functional.

Temperature change after formation commonly results in thermal residual stress in multilayered structures due to the different thermal and mechanical properties of each layer. In this paper, a three-dimensional (3D) elastic model is developed to study the residual stress and opening-mode fractures (OMFs) in a multilayered structure consisting of arbitrary number of layers under temperature change. The general solution of displacement field in the multilayered structure is derived by solving the elastic boundary value problem. In order to verify the proposed model, the elastic field in the advanced polymeric solar reflectors that consist of four layers is solved by applying the present model and compared with the finite-element (FE) simulation. In addition, parametric studies are conducted to investigate the effect of the thickness ratio between each layer on the accuracy of the developed model. Based on the obtained elastic field, the fracture energy release rate (ERR) in the surface layer of the advanced polymeric reflector is obtained and used to study the fracture initiation, infilling, and saturation successfully.

Degradation pathways of small molecule donors for organic photovoltaics are shown to be dependent on chemical traits and not just redox properties.

In commercial electrolysis today, the cost of electricity input drives the cost of hydrogen production. Electrochemical water splitting, therefore, typically uses high catalyst loading and constant power input at high capacity. Catalyst development under these operating conditions, for cost reduction or durability improvement, are less immediate a focus. To reach hydrogen production targets, however, electrolyzers will need to be coupled directly with low-cost power sources to reduce feedstock costs. At that point, catalyst loading reductions are needed to reduce capital cost and catalyst efforts to evaluate material performance and durability become critical.[1-3] We will present efforts comparing half- and single-cells and evaluating to what extent rotating disk electrodes can be used to project membrane electrode assembly performance and durability. In rotating disk electrodes, coating technique and test parameters have a clear influence on baseline activity and the observed properties of materials. Under specific conditions, changes to these variables can shift the activity of standard catalysts by orders of magnitude. Similarly, how membrane electrode assemblies are coated, including spray temperature, ink composition, and ionomer content, has an impact on their initial performance and long-term operation. Although kinetic activities in half- and single-cell tests do not match, half-cell testing appears to be a reasonable indicator of cell performance provided that robust baselines are used. Performances between the two techniques were correlated for a variety of material types, including supported catalysts, alloys, and multicomponent materials with consistent activity trends observed. In certain cases, however, particularly for different surface types (metals, hydroxides, oxides) a disconnect was found between rotating disk electrodes and membrane electrode assemblies. Testing methodologies were explored to resolve these discrepancies and single-cell conditioning was found to incorporate aspects of half-cell durability testing. Modeling has been used to address performance differences between these surfaces, including: how surface and near-surface oxidation influences activity; and how that activity changes over time. By evaluating different catalysts and test parameters, we are looking to establish performance and durability guidelines for catalyst development efforts, and to share our perspective on how catalyst development and system controls factor into electrolysis at low loading and with intermittent operation. These tests have significant implications as electrolysis shifts toward low-cost hydrogen production coupled with renewable power inputs. [1] Alia, S. M., H2@Scale: Experimental Characterization of Durability of Advanced Electrolyzer Concepts in Dynamic Loading. Department of Energy, U. S., Ed. https://www.hydrogen.energy.gov/pdfs/review18/tv146_alia_2018_p.pdf, 2018. [2] Denholm, P.; O’Connell, M.; Brinkman, G.; Jorgenson, J. Overgeneration from Solar Energy in California: A Field Guide to the Duck Chart; Vol. NREL/TP-6A20-65023; National Renewable Energy Laboratory: Golden, CO, 2015. Available at the following: http://www.nrel.gov/docs/fy16osti/65023.pdf. [3] Alia, S. M.; Rasimick, B.; Ngo, C.; Neyerlin, K. C.; Kocha, S. S.; Pylypenko, S.; Xu, H.; Pivovar, B. S. J. Electrochem. Soc. 2016, 163, F3105−F3112.

Thin metal and metal oxide films deposited on glass or polymer substrates have been widely applied to manufacture front surfaced solar reflectors such as those used in the concentrated solar power (CSP) systems due to the facts that they are highly transparent, resistant to ultraviolet (UV) light and abrasion, and chemically inert. However, the reliable operation over long periods of outdoor exposure of such solar reflectors depends on the adequate adhesion between each layer to prevent delamination and subsequent failure. We report results from measurements of the delamination fracture energy (DFE) of the weakest interface in both as-deposited and aged reflectors, where the reflector construction consists of a glass/copper/silver/oxide stack. In this study, DFE was measured using a width-tapered beam method and the test results showed that the weakest interface of this solar reflector is the copper-glass interface, with a delamination fracture energy of approximately 4.4 J/m2. In order to verify the test results, finite element (FE) simulations were run using the commercial software package ABAQUS 6.14 based on the cohesive zone model (CZM). In the simulations, the DFE derived from the delamination measurement was treated as an input and the delamination forces from the FE simulations were compared to the experimental results. The good agreement between the tested and simulated delamination force indicated that this width-tapered beam method is accurate even for very small DFE, making it appropriate to test the DFE of front surfaced glass solar reflectors. Additionally, in order to study the aging effect on the DFE, accelerated aging tests were conducted using samples with different protective coatings, one group of samples with alumina (Al2O3) protective coating and the other group with titania (TiO2) protective coating. The DFE after 50 and 100 hours of accelerated aging were measured. The test results showed that the copper-glass interface would still be the weakest interface for all samples after aging, and the DFE for samples with Al2O3 protective coating was unchanged whereas the copper-glass interface was found to undergo statistically significant change after both 50 and 100 hours of aging for samples with TiO2 protective coating. We found that DFE at the copper/glass interface decreased after 50 hours of aging then increased after 100 hours of aging. The origin of this non-monotonic response of TiO2 coated reflectors under accelerated aging conditions is unknown.