Co-reporter:John Mark P. Martirez and Emily A. Carter
Journal of the American Chemical Society March 29, 2017 Volume 139(Issue 12) pp:4390-4390
Publication Date(Web):March 7, 2017
DOI:10.1021/jacs.6b12301
Localized surface plasmon resonances (LSPRs) offer the possibility of light-activated chemical catalysis on surfaces of strongly plasmonic metal nanoparticles. This technology relies on lower-barrier bond formation and/or dissociation routes made available through energy transfer following the eventual decay of LSPRs. The coupling between these decay processes and a chemical trajectory (nuclear motion, charge-transfer, intersystem crossing, etc.) dictates the availability of these alternative (possibly lower barrier) excited-state channels. The Haber–Bosch method of NH3 synthesis from N2 and H2 is notoriously energy intensive. This is due to the difficulty of N2 dissociation despite the overall reaction being thermodynamically favorable at ambient temperatures and pressures. LSPRs may provide means to improve the kinetics of N2 dissociation via induced resonance electronic excitation. In this work, we calculate, via embedded n-electron valence second-order perturbation theory within the density functional embedding theory, the excited-state potential energy surfaces for dissociation of N2 on an Fe-doped Au(111) surface. This metal alloy may take advantage simultaneously of the strong LSPR of Au and the catalytic activity of Fe toward N2 dissociation. We find the ground-state dissociation activation energy to be 4.74 eV/N2, with Fe as the active site on the surface. Consecutive resonance energy transfers (RETs) may be accessed due to the availability of many electronically excited states with intermediate energies arising from the metal surface that may couple to states induced by the Fe-dopant and the adsorbate molecule, and crossing between excited states may effectively lower the dissociation barrier to 1.33 eV. Our work illustrates that large energetic barriers, prohibitive toward chemical reaction, may be overcome through multiple RETs facilitating an otherwise difficult chemical process.
Co-reporter:Thomas P. Senftle and Emily A. Carter
Accounts of Chemical Research March 21, 2017 Volume 50(Issue 3) pp:472-472
Publication Date(Web):March 21, 2017
DOI:10.1021/acs.accounts.6b00479
Technologies for reducing the concentration of CO2 in our atmosphere are essential for mitigating the risks of climate change, and novel chemistry is required for such technologies to work at scale. Here, we highlight challenges that chemists must overcome to realize the Holy Grail of an economically viable strategy for CO2 capture, utilization, and storage.
Co-reporter:Donald F. Johnson
The Journal of Physical Chemistry C March 18, 2010 Volume 114(Issue 10) pp:4436-4444
Publication Date(Web):Publication Date (Web): February 24, 2010
DOI:10.1021/jp907883h
We assess two alloys, FeAl and Fe3Si, in terms of their ability to inhibit ingress of carbon into steels using periodic density functional theory to model alloy thin films. Absorption of carbon into Fe3Si via its (100) and (110) surfaces is predicted to be considerably more endothermic compared to absorption into FeAl or pure Fe. Predicted energy barriers for surface to subsurface diffusion are ∼1 eV or larger for all four surfaces studied. A very endothermic dissolution enthalpy (1.65 eV) and large diffusion activation energies (e.g., 1.29 eV) suggest Fe3Si should inhibit carbon uptake into bulk steel and slow bulk diffusion. Combined with the results of other recent work [Johnson, D. F.; Carter, E. A. Acta Mater. 2010, 58, 638], we expect that a protective coating of Fe3Si should be effective at preventing steel degradation by hydrogen and carbon containing gases.
Co-reporter:Jin Cheng, Florian Libisch, Kuang Yu, Mohan Chen, Johannes M. Dieterich, and Emily A. Carter
Journal of Chemical Theory and Computation March 14, 2017 Volume 13(Issue 3) pp:1067-1067
Publication Date(Web):January 26, 2017
DOI:10.1021/acs.jctc.6b01010
Embedding theories offer an elegant solution to overcome intrinsic algorithmic scaling and accuracy limitations of simulation methods. These theories also promise to achieve the accuracy of high-level electronic structure techniques at near the computational cost of much less accurate levels of theory by exploiting positive traits of multiple methods. Of crucial importance to fulfilling this promise is the ability to combine diverse theories in an embedding simulation. However, these methods may utilize different basis set and electron–ion potential representations. In this first part of a two-part account of implementing potential functional embedding theory (PFET) at a correlated wave function level, we discuss remedies to basis set and electron–ion potential discrepancies and assess the performance of the PFET scheme with mixed basis sets.
Co-reporter:Jin Cheng, Kuang Yu, Florian Libisch, Johannes M. Dieterich, and Emily A. Carter
Journal of Chemical Theory and Computation March 14, 2017 Volume 13(Issue 3) pp:1081-1081
Publication Date(Web):January 26, 2017
DOI:10.1021/acs.jctc.6b01011
Quantum mechanical embedding theories partition a complex system into multiple spatial regions that can use different electronic structure methods within each, to optimize trade-offs between accuracy and cost. The present work incorporates accurate but expensive correlated wave function (CW) methods for a subsystem containing the phenomenon or feature of greatest interest, while self-consistently capturing quantum effects of the surroundings using fast but less accurate density functional theory (DFT) approximations. We recently proposed two embedding methods [for a review, see: Acc. Chem. Res. 2014, 47, 2768]: density functional embedding theory (DFET) and potential functional embedding theory (PFET). DFET provides a fast but non-self-consistent density-based embedding scheme, whereas PFET offers a more rigorous theoretical framework to perform fully self-consistent, variational CW/DFT calculations [as defined in part 1, CW/DFT means subsystem 1(2) is treated with CW(DFT) methods]. When originally presented, PFET was only tested at the DFT/DFT level of theory as a proof of principle within a planewave (PW) basis. Part 1 of this two-part series demonstrated that PFET can be made to work well with mixed Gaussian type orbital (GTO)/PW bases, as long as optimized GTO bases and consistent electron–ion potentials are employed throughout. Here in part 2 we conduct the first PFET calculations at the CW/DFT level and compare them to DFET and full CW benchmarks. We test the performance of PFET at the CW/DFT level for a variety of types of interactions (hydrogen bonding, metallic, and ionic). By introducing an intermediate CW/DFT embedding scheme denoted DFET/PFET, we show how PFET remedies different types of errors in DFET, serving as a more robust type of embedding theory.
Co-reporter:Linan Zhou, Chao Zhang, Michael J. McClain, Alejandro Manjavacas, Caroline M. Krauter, Shu Tian, Felix Berg, Henry O. Everitt, Emily A. Carter, Peter Nordlander, and Naomi J. Halas
Nano Letters 2016 Volume 16(Issue 2) pp:1478-1484
Publication Date(Web):January 22, 2016
DOI:10.1021/acs.nanolett.5b05149
Hydrogen dissociation is a critical step in many hydrogenation reactions central to industrial chemical production and pollutant removal. This step typically utilizes the favorable band structure of precious metal catalysts like platinum and palladium to achieve high efficiency under mild conditions. Here we demonstrate that aluminum nanocrystals (Al NCs), when illuminated, can be used as a photocatalyst for hydrogen dissociation at room temperature and atmospheric pressure, despite the high activation barrier toward hydrogen adsorption and dissociation. We show that hot electron transfer from Al NCs to the antibonding orbitals of hydrogen molecules facilitates their dissociation. Hot electrons generated from surface plasmon decay and from direct photoexcitation of the interband transitions of Al both contribute to this process. Our results pave the way for the use of aluminum, an earth-abundant, nonprecious metal, for photocatalysis.
Co-reporter:Kuang Yu and Emily A. Carter
Chemistry of Materials 2016 Volume 28(Issue 3) pp:864
Publication Date(Web):January 21, 2016
DOI:10.1021/acs.chemmater.5b04351
Although a promising photovoltaic material that is inexpensive and easy to manufacture, Cu2ZnSnS4 (CZTS) suffers from a low open circuit voltage thought to be due to local potential fluctuations caused by a disordered Cu/Zn sublattice. The disordered character of CZTS is difficult to study experimentally and has been universally neglected in computational studies. Here, we develop a cluster expansion model that enables simulation of the order–disorder phase transition in CZTS for the first time. With a proper atomic structure of the disordered phase in hand, we investigate the temperature-dependent voltage loss in CZTS, illustrating intrinsic limitations of existing synthesis methods and suggesting an optimal annealing temperature. We offer one explanation why Cu-poor CZTS is optimally efficient, as Cu vacancies increase the band gap via interactions between free carriers and the disordered nature of as-grown CZTS. Accordingly, increasing carrier concentrations may be an effective strategy to flatten the fluctuating potentials.
Co-reporter:Kuang Yu and Emily A. Carter
Chemistry of Materials 2016 Volume 28(Issue 12) pp:4415
Publication Date(Web):June 3, 2016
DOI:10.1021/acs.chemmater.6b01612
Cu2ZnSnS4 (CZTS) is an environmentally friendly photovoltaic material with promising applications in thin-film solar cells. Although CZTS’s efficiency is currently too low, stoichiometry/defect engineering presents a strategy for further improvement. As-grown CZTS is typically disordered and therefore prone to form secondary phases, making the final product stoichiometry difficult to determine and even harder to control. We use first-principles quantum mechanics in combination with Monte Carlo simulations to determine CZTS stoichiometry under various experimental conditions. We develop an approach to predicting the optimal CZTS stoichiometry, explaining the physical origin of Zn-enrichment observed in experiments. We further propose practical ways to introduce more free carriers into CZTS in order to screen observed local potential fluctuations, increase conductivity, and ultimately improve the efficiency of CZTS.
Co-reporter:Thomas P. Senftle, Martina Lessio, and Emily A. Carter
Chemistry of Materials 2016 Volume 28(Issue 16) pp:5799
Publication Date(Web):July 22, 2016
DOI:10.1021/acs.chemmater.6b02084
Pyridine (Py) is an effective cocatalyst during the photoelectrochemical reduction of CO2 to methanol over GaP, CdTe, and CuInS2 semiconductor surfaces. Identifying the role Py plays in the catalytic reduction mechanism is essential for optimizing the design of such photocatalytic processes. The Py-enhanced mechanism, however, is under considerable debate. Recent studies suggest that the semiconductor surface itself participates in a heterogeneous mechanism, and for this reason a detailed understanding of the interaction between Py and the surface is required. Additionally, surface reconstructions occurring during operation alter the nature of adsorption sites available for interaction with the solution, therefore impacting the performance of the electrode. To address this issue, we report a density functional theory investigation of the stability of GaP(111) and CdTe(111) surface reconstructions, as well as adsorption trends of intermediate species across sites created by such reconstructions. We also determine band edge positions of the solvated, reconstructed surfaces, which we compare to calculated reduction potentials involved in proposed elementary steps of the overall CO2 reduction mechanism. This allows us to determine which reduction steps are thermodynamically feasible based on the energy of a photoexcited electron in the conduction band of the semiconductor. Given the band edge alignment of the GaP(111) surface, we determine that the 1e– reduction of the solvated pyridinium cation most favorably results in the formation of adsorbed Py* + H* species and that the formation of a 1-pyridinyl radical in solution is unlikely. Furthermore, we find that it is thermodynamically feasible to form a newly proposed adsorbed 2-pyridinyl intermediate adsorbed on the surface, which may act as a powerful hydride donor. On the CdTe(111) surface, we find that no 1e– reductions are thermodynamically feasible, leaving only 2e– reductions leading to the formation of dihydropyridine (DHP) as possible reduction steps. These results identify stable intermediate species along the CO2 reaction path over reconstructed surfaces, thus lending insight into the Py-catalyzed reaction mechanism.
Co-reporter:Mohan Chen, Xiang-Wei Jiang, Houlong Zhuang, Lin-Wang Wang, and Emily A. Carter
Journal of Chemical Theory and Computation 2016 Volume 12(Issue 6) pp:2950-2963
Publication Date(Web):May 4, 2016
DOI:10.1021/acs.jctc.6b00326
Orbital-free density functional theory (OFDFT) is a quantum-mechanics-based method that utilizes electron density as its sole variable. The main computational cost in OFDFT is the ubiquitous use of the fast Fourier transform (FFT), which is mainly adopted to evaluate the kinetic energy density functional (KEDF) and electron–electron Coulomb interaction terms. We design and implement a small-box FFT (SBFFT) algorithm to overcome the parallelization limitations of conventional FFT algorithms. We also propose real-space truncation of the nonlocal Wang–Teter KEDF kernel. The scalability of the SBFFT is demonstrated by efficiently simulating one full optimization step (electron density, energies, forces, and stresses) of 1,024,000 lithium (Li) atoms on up to 65,536 cores. We perform other tests using Li as a test material, including calculations of physical properties of different phases of bulk Li, geometry optimizations of nanocrystalline Li, and molecular dynamics simulations of liquid Li. All of the tests yield excellent agreement with the original OFDFT results, suggesting that the OFDFT-SBFFT algorithm opens the door to efficient first-principles simulations of materials containing millions of atoms.
Co-reporter:Andrew M. Ritzmann, Johannes M. Dieterich and Emily A. Carter
Physical Chemistry Chemical Physics 2016 vol. 18(Issue 17) pp:12260-12269
Publication Date(Web):06 Apr 2016
DOI:10.1039/C6CP01720G
Reducing operating temperatures is a key step in making solid oxide fuel cell (SOFC) technology viable. A promising strategy for accomplishing this goal is employing mixed ion–electron conducting (MIEC) cathodes. La1−xSrxCo1−yFeyO3−δ (LSCF) is the most widely employed MIEC cathode material; however, rational optimization of the composition of LSCF requires fundamental insight linking its electronic structure to its defect chemistry. To provide the necessary insight, density functional theory plus U (DFT+U) calculations are used to investigate the electronic structure of LSCF (xSr = 0.50, yCo = 0.25). The DFT+U calculations show that LSCF has a significantly different electronic structure than La1−xSrxFeO3 because of the addition of cobalt, but that minimal electronic structure differences exist between La0.5Sr0.5Co0.25Fe0.75O3 and La0.5Sr0.5Co0.5Fe0.5O3. The oxygen vacancy formation energy (ΔEf,vac) is calculated for residing in different local environments within La0.5Sr0.5Co0.25Fe0.75O3. These results show that configurations have the highest ΔEf,vac, while have the lowest ΔEf,vac and may act as traps for . We conclude that compositions with more Fe than Co are preferred because the additional sites would lead to higher overall ΔEf,vac (and lower concentrations), while the trapping strength of the sites is relatively weak (∼0.3 eV).
Co-reporter:Ting Tan, Xueliang Yang, Yiguang Ju and Emily A. Carter
Physical Chemistry Chemical Physics 2016 vol. 18(Issue 6) pp:4594-4607
Publication Date(Web):14 Jan 2016
DOI:10.1039/C5CP07282D
The kinetics of hydrogen abstraction by five radicals (H, CH3, O(3P), OH, and HO2) from a biodiesel surrogate, methyl propanoate (MP), is theoretically investigated. We employ high-level ab initio quantum chemistry methods, coupled-cluster singles and doubles with perturbative triples correction (CCSD(T)) and multi-reference singles and doubles configuration interaction (MRSDCI) with the Davidson-Silver (DS) correction, and obtain chemically accurate reaction energetics. Overall, MRSDCI + DS predicts comparable energetics to CCSD(T) for MP + H/CH3/O/OH. The rate constants are computed using transition state theory (TST-Rice–Ramsperger–Kassel–Marcus theory) in conjunction with the separable-hindered-rotor approximation, variable reaction coordinate TST, and the multi-structure all-structure (MS-AS) approach. A simplified method, semi-multi-structure, is also employed for MP + OH/HO2, and the rate coefficients with this less expensive method are in good agreement with the results obtained with the MS-AS method. The fitted modified Arrhenius expressions are provided over a temperature range of 250 to 2000 K. The predicted rate coefficients for MP + OH agree remarkably well with experimental data over a wide temperature range. Branching ratio analysis of all the studied reactions shows that abstractions of the secondary H atoms within MP are expected to dominate the consumption of fuel at low temperatures, and the contributions of abstractions from the two methyl groups increase with temperature for all abstracting radicals.
Co-reporter:Ting Tan, Xueliang Yang, Yiguang Ju, and Emily A. Carter
The Journal of Physical Chemistry B 2016 Volume 120(Issue 8) pp:1590-1600
Publication Date(Web):September 28, 2015
DOI:10.1021/acs.jpcb.5b07959
The dissociation and isomerization kinetics of the methyl ester combustion intermediates methoxycarbonyl radical (CH3OĊ(═O)) and (formyloxy)methyl radical (ĊH2OC(═O)H) are investigated theoretically using high-level ab initio methods and Rice–Ramsperger–Kassel–Marcus (RRKM)/master equation (ME) theory. Geometries obtained at the hybrid density functional theory (DFT) and coupled cluster singles and doubles with perturbative triples correction (CCSD(T)) levels of theory are found to be similar. We employ high-level ab initio wave function methods to refine the potential energy surface: CCSD(T), multireference singles and doubles configuration interaction (MRSDCI) with the Davidson–Silver (DS) correction, and multireference averaged coupled-pair functional (MRACPF2) theory. MRSDCI+DS and MRACPF2 capture the multiconfigurational character of transition states (TSs) and predict lower barrier heights than CCSD(T). The temperature- and pressure-dependent rate coefficients are computed using RRKM/ME theory in the temperature range 300–2500 K and a pressure range of 0.01 atm to the high-pressure limit, which are then fitted to modified Arrhenius expressions. Dissociation of CH3OĊ(═O) to ĊH3 and CO2 is predicted to be much faster than dissociating to CH3Ȯ and CO, consistent with its greater exothermicity. Isomerization between CH3OĊ(═O) and ĊH2OC(═O)H is predicted to be the slowest among the studied reactions and rarely happens even at high temperature and high pressure, suggesting the decomposition pathways of the two radicals are not strongly coupled. The predicted rate coefficients and branching fractions at finite pressures differ significantly from the corresponding high-pressure-limit results, especially at relatively high temperatures. Finally, because it is one of the most important CH3Ȯ removal mechanisms under atmospheric conditions, the reaction kinetics of CH3Ȯ + CO was also studied along the PES of CH3OĊ(═O); the resulting kinetics predictions are in remarkable agreement with experiments.
Co-reporter:Mohan Chen, John Roszell, Emanuel V. Scoullos, Christoph Riplinger, Bruce E. Koel, and Emily A. Carter
The Journal of Physical Chemistry B 2016 Volume 120(Issue 26) pp:6110-6119
Publication Date(Web):March 30, 2016
DOI:10.1021/acs.jpcb.6b02092
Determining the strength of Li binding to Mo is critical to assessing the survivability of Li as a potential first wall material in fusion reactors. We present the results of a joint experimental and theoretical investigation into how Li desorbs from Mo(110) surfaces, based on what can be deduced from temperature-programmed desorption measurements and density functional theory (DFT). Li desorption peaks measured at temperatures ranging from 711 K (1 monolayer, ML) to 1030 K (0.04 ML), with corresponding desorption onsets from 489 to 878 K, follow a trend similar to predicted Gibbs free energies for Li adsorption. Bader charge analysis of DFT densities reveals that repulsive forces between neighboring positively charged Li atoms increase with coverage and thus reduce the bond strength between Mo and Li, thereby lowering the desorption temperature as the coverage increases. Additionally, DFT predicts that Li desorbs at higher temperatures from a surface with vacancies than from a perfect surface, offering an explanation for the anomalously high desorption temperatures for the last Li to desorb from Mo(110). Analysis of simulated local densities of states indicates that the stronger binding to the defective surface is correlated with enhanced interaction between Li and Mo, involving the Li 2s electrons and not only the Mo 4d electrons as in the case of the pristine surface, but also the Mo 5s electrons in the case with surface vacancies. We suggest that steps and kinks present on the Mo(110) surface behave similarly and contribute to the high desorption temperatures. These findings imply that roughened Mo surfaces may strengthen Li film adhesion at higher temperatures.
Co-reporter:Houlong Zhuang, Alexander J. Tkalych, and Emily A. Carter
The Journal of Physical Chemistry C 2016 Volume 120(Issue 41) pp:23698-23706
Publication Date(Web):September 27, 2016
DOI:10.1021/acs.jpcc.6b09687
Computational searches for catalysts of the hydrogen evolution reaction commonly use the hydrogen binding energy (HBE) as a predictor of catalytic activity. Accurate evaluation of the HBE, however, can involve large periodic supercell slab models that render high-throughput screening relatively expensive. In contrast, calculations of other relevant surface properties, such as the surface energy, work function, and potential of zero charge (PZC), require only small surface unit cells and are hence less expensive to compute. Correlations between catalytic activity and these surface properties warrant exploration because of this reduced computational cost. Here, we use density functional theory in conjunction with three different exchange-correlation functionals—the local density approximation (LDA), the Perdew–Burke–Ernzerhof (PBE) generalized gradient approximation, and the PBEsol functional (a reparameterization of the PBE functional)—to calculate the lattice constants, surface energy, cohesive energy, and work function of six common catalysts: three metals (Au, Pd, and Pt) and three transition-metal carbides (TMCs; WC, W2C, and Mo2C). The three exchange-correlation functionals produce identical trends, and PBEsol yields results between those calculated using LDA and PBE and most often closer to experiment. We therefore use PBEsol to obtain the surface energy, work function, and PZC of nine novel hybrid catalysts, each containing a metal monolayer on a TMC substrate. Importantly, a volcano-shaped correlation between the experimental exchange current density and the theoretical surface energies emerges. We also investigate solvation effects on the surface energy and work function using a polarizable continuum model within the framework of joint density functional theory. For these particular materials, the surface energies in vacuum are nearly unchanged upon exposure to an aqueous solution, prior to any reaction with water. The volcano-shaped correlation observed between the exchange current densities and the surface energies is not observed for the work function or PZC. Our work thus reveals opportunities for more rapid computational screening of reduced Pt-loading catalysts using the surface energy as a computationally efficient catalytic descriptor.
Co-reporter:Martina Lessio, Thomas P. Senftle, and Emily A. Carter
ACS Energy Letters 2016 Volume 1(Issue 2) pp:464
Publication Date(Web):July 30, 2016
DOI:10.1021/acsenergylett.6b00233
The role of the photoelectrode surface during pyridine-catalyzed CO2 reduction on p-GaP photocathodes is currently under debate. Understanding the extent of the photoelectrode’s direct participation in the catalytic CO2 reduction mechanism is essential to improving the design of such photoelectrochemical systems. Here, we present new theoretical results demonstrating that the previously proposed pyridinyl radical intermediate is unlikely to form and that reduction of pyridinium to adsorbed pyridine and H species remains the most favorable reduction pathway, even when accounting for the aqueous environment. Furthermore, we conclude, based on recently reported experimental evidence and our new computational results reported herein, that the mechanism of CO2 reduction operating in this system is likely heterogeneous. We also introduce a new heterogeneous mechanism involving a recently proposed radical species, which we predict will be stable on the electrode surface and that may serve as the active catalytic species in this system.
Co-reporter:John Mark P. Martirez and Emily A. Carter
ACS Nano 2016 Volume 10(Issue 2) pp:2940
Publication Date(Web):February 1, 2016
DOI:10.1021/acsnano.6b00085
The Haber-Bosch process for NH3 synthesis is arguably one of the greatest inventions of the 20th century, with a massive footprint in agriculture and, historically, warfare. Current catalysts for this reaction use Fe for N2 activation, conducted at high temperatures and pressures to improve conversion rate and efficiency. A recent finding shows that plasmonic metal nanoparticles can either generate highly reactive electrons and holes or induce resonant surface excitations through plasmonic decay, which catalyze dissociation and redox reactions under mild conditions. It is therefore appealing to consider AuM (M = Fe, Co, Ni, and Mo) alloys to combine the strongly plasmonic nature of Au and the catalytic nature of M metals toward N2 dissociation, which together might facilitate ammonia production. To this end, through density functional theory, we (i) explore the feasibility of forming these surface alloys, (ii) find a pathway that may stabilize/deactivate surface M substituents during fabrication, and (iii) define a complementary route to reactivate them under operational conditions. Finally, we evaluate their reactivity toward N2, as well as their ability to support a pathway for N2 dissociation with a low thermodynamic barrier. We find that AuFe possesses similar appealing qualities, including relative stability with respect to phase separation, reversibility of Fe oxidation and reduction, and reactivity toward N2. While AuMo achieves the best affinity toward N2, its strong propensity toward oxidation could greatly limit its use.Keywords: ammonia synthesis; Au alloys; Haber-Bosch; heterogeneous catalysis; nitrogen dissociation; surface plasmon; transition metal catalysis;
Co-reporter:Martina Lessio
Journal of the American Chemical Society 2015 Volume 137(Issue 41) pp:13248-13251
Publication Date(Web):September 29, 2015
DOI:10.1021/jacs.5b08639
Experimental evidence suggests that pyridinium plays an important role in photocatalytic CO2 reduction on p-GaP photoelectrodes. Pyridinium reduction to pyridinyl has been previously proposed as an essential mechanistic step for this reaction. However, theoretical calculations suggest that this step is not feasible in solution. Here, cluster models and accurate periodic boundary condition calculations are used to determine whether such a reduction step could occur by transfer of photoexcited electrons from the p-GaP photocathode and whether this transfer could be catalyzed by pyridinium adsorption on the p-GaP surface. It is found that both the transfer of photoexcited electrons to pyridinium and pyridinium adsorption are not energetically favored, thus making very unlikely pyridinium reduction to the pyridinyl radical and the proposed mechanisms requiring this reduction step. Given this conclusion, an alternative and energetically viable pathway for pyridinium reduction on p-GaP photoelectrodes is proposed. This pathway leads to the formation of adsorbed species that could react to form adsorbed dihydropyridine, which was proposed previously to play the role of the active catalyst in this system.
Co-reporter:Christoph Riplinger and Emily A. Carter
ACS Catalysis 2015 Volume 5(Issue 2) pp:900
Publication Date(Web):December 22, 2014
DOI:10.1021/cs501687n
[Re(bpy)(CO)3]− and [Mn(bpy)(CO)3]− are homogeneous electrocatalysts for the reduction of CO2 to CO. Their turnover frequencies depend on the type of Brønsted acid used, with the Mn catalyst exhibiting no catalytic turnover without added Brønsted acid. In this work, we use density functional theory together with continuum solvation and microkinetics simulations to understand these differences. The computed turnover frequencies reproduce the experimental trends. In absolute numbers, the computed turnover frequencies differ from the experimental ones by about an order of magnitude. We find that some of the experimentally used acids are too weak to protonate CO2 or to stabilize CO2 binding. Catalysis with these acids requires more negative applied potentials or higher acid concentrations compared to catalysis with stronger acids. This trend is more pronounced for the Mn catalyst than for the Re catalyst, the latter working at maximum turnover with acids that produce submaximum turnover with the Mn catalyst. In the absence of Brønsted acids, the first catalytic steps are driven by the solvent acetonitrile, which can act as proton donor for protonation of CO2 in the case of the Re catalyst. For the Mn catalyst, the endergonic CO2 binding free energy prevents protonation by acetonitrile. C–O bond cleavage, however, cannot be assisted by acetonitrile for either catalyst. Electrolyte-assisted C–O bond cleavage via Hofmann degradation is also predicted to be strongly disfavored kinetically. Water produced during catalysis might be responsible for completing the reaction cycle.Keywords: Brønsted acid; CO2 reduction; electrocatalysis; homogeneous catalysis; microkinetics simulation; turnover frequencies
Co-reporter:Kuang Yu and Emily A. Carter
Chemistry of Materials 2015 Volume 27(Issue 8) pp:2920
Publication Date(Web):April 3, 2015
DOI:10.1021/acs.chemmater.5b00172
Cu2ZnSnS4–xSex (CZTS) is an important semiconductor with significant potential for applications in the next generation of solar cells. CZTS has an optimal band gap (∼1.5 eV) and contains no expensive or toxic elements. However, CZTS-based solar cells suffer from low efficiency because of poor crystal quality, which is partly caused by secondary phase formation during synthesis. We use density functional theory+U calculations to systematically investigate the stabilities of three CZTS phases: kesterite, stannite, and wurtzite. In agreement with previous experiment and theory, we confirm that these three phases have very similar formation energies. This finding is consistent with the known difficulties in synthesizing pure kesterite CZTS, the phase that is desirable for photovoltaic applications. To overcome this problem, we characterize surfaces and interfaces of CZTS and are able to identify certain “beneficial surfaces” that could be exploited to potentially provide extra stability for the kesterite phase. We propose the zinc blende ZnS (001) surface as a substrate to induce formation of these beneficial surfaces and to stabilize the kesterite phase, thereby serving as an effective crystallization template for the fabrication of high-performance CZTS solar cells.
Co-reporter:David B. Krisiloff, Caroline M. Krauter, Francis J. Ricci, and Emily A. Carter
Journal of Chemical Theory and Computation 2015 Volume 11(Issue 11) pp:5242-5251
Publication Date(Web):October 9, 2015
DOI:10.1021/acs.jctc.5b00762
To treat large molecules with accurate ab initio quantum chemistry, reduced scaling correlated wave function methods are now commonly employed. Optimization of these wave functions in practice requires some approximation of the two-electron integrals. Both Cholesky decomposition (CD) and density fitting (DF) are widely used approaches to approximate these integrals. Here, we compare CD and DF for use in local multireference singles and doubles configuration interaction (LMRSDCI). DF-LMRSDCI provides less accurate total energies than CD-LMRSDCI, but both methods are accurate for energy differences. However, DF-LMRSDCI is significantly less computationally expensive than CD-LMRSDCI on the molecules tested, suggesting that DF-LMRSDCI is an efficient, often sufficiently accurate alternative to our previously reported CD-LMRSDCI method.
Co-reporter:Ting Tan, Xueliang Yang, Yiguang Ju and Emily A. Carter
Physical Chemistry Chemical Physics 2015 vol. 17(Issue 46) pp:31061-31072
Publication Date(Web):28 Oct 2015
DOI:10.1039/C5CP06004D
The unimolecular dissociation and isomerization kinetics of the three methyl propanoate (MP) radicals, CH3CH2C(O)OĊH2 (MP-m), CH3ĊHC(O)OCH3 (MP-α), and ĊH2CH2C(O)OCH3 (MP-β), are theoretically investigated using high-level ab initio methods and the Rice–Ramsperger–Kassel–Marcus (RRKM)/master equation (ME) theory. Stationary-point energies are obtained using the coupled cluster singles and doubles with perturbative triples correction (CCSD(T)), multi-reference singles and doubles configuration interaction (MRSDCI) with the Davidson-Silver (DS) correction, and multi-reference averaged coupled pair functional (MRACPF2) theories. The isomerization barriers between the three radicals are predicted to be generally lower than the corresponding bond dissociation channels, leading to a strongly coupled reaction system in subsequent kinetics studies. The phenomenological temperature- and pressure-dependent rate coefficients are computed using the RRKM/ME theory over a temperature range of 500 to 2000 K and at a pressure range of 0.01 atm to the high-pressure limit, which are then fitted to modified Arrhenius expressions. The β-scission rate coefficients of MP-α to CH3CHC(O) and CH3Ȯ are predicted to be the smallest because of its highest activation energy among all studied unimolecular reactions channels. Analysis of branching fractions shows that both MP-m and MP-α radicals mainly decompose to the bimolecular products CH3CH2Ċ(O) and H2CO, whereas the MP-β radical primarily decomposes via cleavage of a C–C bond to form C2H4 and CH3OĊ(O). The isomerization channels dominate at low temperatures, the branching fractions of which decrease with increasing temperature and become very minor at about 2000 K. Our accurate rate coefficients and branching fractions help to illuminate the unique combustion properties of MP.
Co-reporter:Nima Alidoust and Emily A. Carter
Physical Chemistry Chemical Physics 2015 vol. 17(Issue 27) pp:18098-18110
Publication Date(Web):18 Jun 2015
DOI:10.1039/C5CP03429A
Alloying nickel oxide (NiO) with lithium oxide (Li2O) at high Li concentrations may reduce NiO's band gap and expand its use as a light absorber in photocatalytic and tandem dye-sensitized solar cell technologies. In this work, we evaluate the viability of this alloy as a p-type hole transport material. We use embedded cluster models, along with unrestricted Hartree–Fock and complete active space self-consistent field theories, to study the impact of alloying on polaronic transport of holes. Our calculated energy barrier for hole transfer in undoped NiO is in excellent agreement with the experimental value of ∼0.1 eV. We predict that hole transport in NiO is anisotropic and mostly confined parallel to the (111) ferromagnetic planes. Applying the same model to Li-doped NiO indicates that isolated Li ions do not introduce free holes into NiO samples. However, free holes can be created in the homogeneous Li0.125Ni0.875O alloy, in which the Li concentration is very high. Our kinetic Monte Carlo calculations show that hole mobility in this alloy is lower than in undoped NiO. However, the additional free holes and the predicted lower band gap of Li0.125Ni0.875O should increase hole conductivity compared to NiO upon alloy formation. Therefore, Li0.125Ni0.875O alloys have potential for use as a hole transporter, as well as a sunlight absorber, in a variety of solar energy applications.
Co-reporter:Johannes M. Dieterich, Emily A. Carter
Computational and Theoretical Chemistry 2015 Volume 1051() pp:47-56
Publication Date(Web):1 January 2015
DOI:10.1016/j.comptc.2014.10.030
•Avoiding the serial I/O bottleneck: simple, direct integral reassembly on-the-fly.•Higher parallel efficiency through integral-direct computations.•Local correlation approximations and Cauchy–Schwarz prescreening drastically reduce the number of significant integrals.We present different integral-direct implementations in a local Cholesky-decomposed multi-reference configuration interaction framework. We discuss their performance, parallel efficiency and scaling properties for a range of alkyne test systems. As we are able to introduce accuracy-preserving integral truncations within a direct algorithm, we observe superior performance and, through the drastically reduced I/O operations, better parallel efficiency for the truncated integral-direct kernels compared to their conventional counterparts.
Co-reporter:Victor B. Oyeyemi, Johannes M. Dieterich, David B. Krisiloff, Ting Tan, and Emily A. Carter
The Journal of Physical Chemistry A 2015 Volume 119(Issue 14) pp:3429-3439
Publication Date(Web):March 16, 2015
DOI:10.1021/jp512974k
We previously developed a fast, local, reduced scaling Cholesky-decomposed multireference averaged-coupled pair functional (CD-LMRACPF2) method, which takes advantage of the locality of dynamic correlation and numerical approximations such as Cholesky decomposition and integral screening. Motivated by the desire to study large biodiesel methyl ester molecules, here we validate CD-LMRACPF2 for the computation of bond dissociation energies (BDEs) in a suite of oxygenated molecules, and show that the low-cost method is very accurate compared to the conventional variant. We then demonstrate the power of CD-LMRACPF2 for fast and accurate computation of energies of molecules containing up to 13 second-row atoms within a polarized triple-ζ (cc-pVTZ) basis set. We use biodiesel methyl esters as a chemically interesting model system and furnish BDEs of C10 and C18 methyl esters, with the latter performed within a cc-pVDZ basis set. We describe trends in the BDEs and explain how structural (isomeric) differences affect BDEs, as well as discuss implications of BDE trends for biodiesel physical and chemical properties.
Co-reporter:Ting Tan, Xueliang Yang, Caroline M. Krauter, Yiguang Ju, and Emily A. Carter
The Journal of Physical Chemistry A 2015 Volume 119(Issue 24) pp:6377-6390
Publication Date(Web):May 14, 2015
DOI:10.1021/acs.jpca.5b03506
The kinetics of hydrogen abstraction by five radicals (H, O(3P), OH, CH3, and HO2) from methyl acetate (MA) is investigated theoretically in order to gain further understanding of certain aspects of the combustion chemistry of biodiesels, such as the effect of the ester moiety. We employ ab initio quantum chemistry methods, coupled cluster singles and doubles with perturbative triples correction (CCSD(T)) and multireference averaged coupled pair functional theory (MRACPF2), to predict chemically accurate reaction energetics. Overall, MRACPF2 predicts slightly higher barrier heights than CCSD(T) for MA + H/CH3/O/OH, but slightly lower barrier heights for hydrogen abstraction by HO2. Based on the obtained reaction energies, we also report high-pressure-limit rate constants using transition state theory (TST) in conjunction with the separable-hindered-rotor approximation, the variable reaction coordinate TST, and the multi-structure all-structure approach. The fitted modified Arrhenius expressions are provided over a temperature range of 250 to 2000 K. The predictions are in good agreement with available experimental results. Abstractions from both of the methyl groups in MA are expected to contribute to consumption of the fuel as they exhibit similar rate coefficients. The reactions involving the OH radical are predicted to have the highest rates among the five abstracting radicals, while those initiated by HO2 are expected to be the lowest.
Co-reporter:Jin Cheng; Florian Libisch
The Journal of Physical Chemistry Letters 2015 Volume 6(Issue 9) pp:1661-1665
Publication Date(Web):April 14, 2015
DOI:10.1021/acs.jpclett.5b00597
The interaction between O2 molecules and Al surfaces has long been poorly understood despite its importance in diverse chemical phenomena. Early experimental investigations of adsorption dynamics indicated that abstraction of a single O atom by the surface, instead of dissociative chemisorption, dominates at low O2 incident kinetic energies. Abstraction of the closer O atom suggests low barrier heights at perpendicular incidence. However, recent measurements suggest that parallel O2 orientations dominate sticking at low energies. We resolve this apparent contradiction by a systematic ab initio embedded correlated wavefunction study of the stereochemistry of O2 reacting with Al(111). We identify two important new details: (i) initially, roughly parallel oxygen molecules tend to tilt upright while approaching the surface, suggesting that the abstraction channel does dominate at low energies and (ii) the reaction channel with the lowest barrier indeed corresponds to a parallel orientation, which ultimately evolves either into dissociative chemisorption or toward abstraction.
Co-reporter:Ting Tan, Xueliang Yang, Yiguang Ju, and Emily A. Carter
The Journal of Physical Chemistry A 2015 Volume 119(Issue 42) pp:10553-10562
Publication Date(Web):October 5, 2015
DOI:10.1021/acs.jpca.5b08331
The unimolecular dissociation and isomerization kinetics of the methyl ester combustion intermediates methoxycarbonylmethyl (CH2C(═O)OCH3) and acetyloxylmethyl (CH3C(═O)OCH2) are theoretically investigated using high-level ab initio methods and the Rice–Ramsperger–Kassel–Marcus (RRKM)/master equation (ME) theory. Potential energy surfaces (PESs) are obtained using coupled cluster singles and doubles with perturbative triples correction (CCSD(T)), multireference singles and doubles configuration interaction (MRSDCI) with the Davidson–Silver (DS) correction, and multireference averaged coupled pair functional (MRACPF2) theory. The transition states exhibit high T1 diagnostics in coupled cluster calculations, suggesting the need for a multireference correlated wave function treatment. MRSDCI+DS and MRACPF2 capture their multiconfigurational character well, yielding lower barrier heights than CCSD(T) for these reactions. The rate coefficients are computed using the RRKM/ME theory over a 500–2500 K temperature range and at a pressure range of 0.01 atm to the high-pressure limit. The temperature- and pressure-dependent rate coefficients are given in modified Arrhenius expressions. The β-scission of CH2C(═O)OCH3 is predicted to have a much higher barrier than the corresponding isomerization reaction and the β-scission of CH3C(═O)OCH2. Consequently, the rate coefficients for β-scission of CH2C(═O)OCH3 are the smallest among the three reactions and the isomerization followed by decomposition to CH3C(═O) and HCHO is the dominant reaction pathway for CH2C(═O)OCH3. Both radicals CH2C(═O)OCH3 and CH3C(═O)OCH2 are predicted to mainly decompose to CH3C(═O) + HCHO rather than to the bimolecular product CH2C(═O) + CH3O. A newly developed MA combustion mechanism, using our theoretical rate coefficients for the MA-related reactions, predicts combustion properties in good agreement with available experimental data.
Co-reporter:Christoph Riplinger
The Journal of Physical Chemistry C 2015 Volume 119(Issue 17) pp:9311-9323
Publication Date(Web):April 6, 2015
DOI:10.1021/acs.jpcc.5b00383
Water adsorption on solid surfaces plays a part in a variety of processes, including renewable energy applications. Water adsorption can occur either dissociatively, monomolecularly, or as clusters. In contrast to metal surfaces, the compositional and structural complexity of metal oxide surfaces has inhibited atomic-scale understanding of their interactions with water. Cu2O is a promising photocatalyst and (photo)electrochemical catalyst. Here, we investigate water adsorption on its (111) surface, using density functional theory + U with dispersion corrections. A number of monomolecular and dissociated adsorbate geometries are considered on the two most stable surface terminations. H2O is found to adsorb most strongly when datively coordinated to an unsaturated surface Cu cation; dissociative adsorption is not as favorable as this dative bonding mode of molecular chemisorption. If these Cu cations are not present, H2O binds via hydrogen bonding and electrostatic interactions in surface cavities. We also examine a large variety of mixed modes of coadsorption. Mixtures of datively bonded and hydrogen-bonded water molecules adsorb most strongly, exhibiting a strong lateral interaction. The resulting water clusters can adapt to the underlying adsorption site template and maintain significant water–surface and water–water interactions at the same time. This is possible through the proximity of the unsaturated cationic and anionic adsorption sites. The combination of dative and hydrogen bonding to the surface enables water clustering even at low temperatures, probably due to rapid surface diffusion of the more weakly bonded monomers. The strong dative and lateral interactions keep water clusters adsorbed up to unusually high temperatures. The water hexamer can still be observed at room temperature under ultrahigh vacuum conditions; we predict that datively bound monomers, hexamers, and other similarly constructed clusters will remain bound to the stoichiometric surface up to quite high temperatures under conditions of high relative humidity. We suggest that metal oxides with similar surface compositions should show similar properties. Finally, we predict vibrational frequencies for the adsorbed water molecules and distinguish between water–water and water–surface vibrations for comparison with future experimental studies.
Co-reporter:Alexander J. Tkalych
The Journal of Physical Chemistry C 2015 Volume 119(Issue 43) pp:24315-24322
Publication Date(Web):October 7, 2015
DOI:10.1021/acs.jpcc.5b08481
NiOx, long studied for its use in nickel-based secondary batteries, has been the subject of much recent interest due to its efficacy as an oxygen evolution catalyst. Despite extensive study over more than a century, however, many outstanding questions remain surrounding both the structure and the activity of NiOx. Further compounding this ambiguity is the recent finding that much of the previous experimental work on NiOx may have been influenced by incidental doping. Here, we report a computational study of the two simplest members of the NiOx family: β-Ni(OH)2 and β-NiOOH. Using DFT+U calculations, we first identify a β-NiOOH structure with a staggered arrangement of intercalated protons that is more consistent with experimental crystal structures of β-NiOOH than previously proposed geometries. Next, by conducting a thorough study of various initial spin configurations of this β-NiOOH structure, we found that a low-spin d7 Ni3+ configuration is always favored, which suggests a Jahn–Teller distortion, rather than disproportionation, explains the different Ni–O bond distances found in experiment. G0W0 calculations performed on β-Ni(OH)2 and β-NiOOH reveal electronic structures consistent with previous experimental results. Lastly, calculations of various low-index surface energies of both β-Ni(OH)2 and β-NiOOH demonstrate that the (001) surface is the most thermodynamically stable surface, in keeping with numerous experimental results but in contrast to recent computational models.
Co-reporter:John A. Keith;Ana B. Muñoz-García;Martina Lessio
Topics in Catalysis 2015 Volume 58( Issue 1) pp:46-56
Publication Date(Web):2015 February
DOI:10.1007/s11244-014-0341-1
Sunlight-powered CO2-photoelectroreduction is a promising and potentially sustainable route to recycle CO2 byproducts back into energy-dense liquid fuels. One of the most intriguing processes known to date is the pyridinium-catalyzed CO2 reduction on p-type GaP photoelectrodes, where conversion to methanol has reported faradaic efficiencies nearing 100 %. Modeling this reactive environment requires understanding energetics of differently charged species at semiconductor electrodes, so we develop a cluster model and benchmark binding energies from it to those from Kohn–Sham density functional theory calculations that employ periodic boundary conditions. We then use this cluster model to theoretically predict structures and binding energies for charged and neutral adsorbates on the GaP(110) surface with and without the presence of van der Waals interactions and implicit solvation. We discuss the relative magnitudes of binding energy contributions for different adsorbates considered relevant in this CO2 reduction process and provide details showing pitfalls when using cluster models.
Co-reporter:Florian Libisch, Chen Huang, and Emily A. Carter
Accounts of Chemical Research 2014 Volume 47(Issue 9) pp:2768
Publication Date(Web):May 30, 2014
DOI:10.1021/ar500086h
Ab initio modeling of matter has become a pillar of chemical research: with ever-increasing computational power, simulations can be used to accurately predict, for example, chemical reaction rates, electronic and mechanical properties of materials, and dynamical properties of liquids. Many competing quantum mechanical methods have been developed over the years that vary in computational cost, accuracy, and scalability: density functional theory (DFT), the workhorse of solid-state electronic structure calculations, features a good compromise between accuracy and speed. However, approximate exchange–correlation functionals limit DFT’s ability to treat certain phenomena or states of matter, such as charge-transfer processes or strongly correlated materials. Furthermore, conventional DFT is purely a ground-state theory: electronic excitations are beyond its scope. Excitations in molecules are routinely calculated using time-dependent DFT linear response; however applications to condensed matter are still limited.By contrast, many-electron wavefunction methods aim for a very accurate treatment of electronic exchange and correlation. Unfortunately, the associated computational cost renders treatment of more than a handful of heavy atoms challenging. On the other side of the accuracy spectrum, parametrized approaches like tight-binding can treat millions of atoms. In view of the different (dis-)advantages of each method, the simulation of complex systems seems to force a compromise: one is limited to the most accurate method that can still handle the problem size. For many interesting problems, however, compromise proves insufficient. A possible solution is to break up the system into manageable subsystems that may be treated by different computational methods. The interaction between subsystems may be handled by an embedding formalism.In this Account, we review embedded correlated wavefunction (CW) approaches and some applications. We first discuss our density functional embedding theory, which is formally exact. We show how to determine the embedding potential, which replaces the interaction between subsystems, at the DFT level. CW calculations are performed using a fixed embedding potential, that is, a non-self-consistent embedding scheme. We demonstrate this embedding theory for two challenging electron transfer phenomena: (1) initial oxidation of an aluminum surface and (2) hot-electron-mediated dissociation of hydrogen molecules on a gold surface. In both cases, the interaction between gas molecules and metal surfaces were treated by sophisticated CW techniques, with the remainder of the extended metal surface being treated by DFT. Our embedding approach overcomes the limitations of conventional Kohn–Sham DFT in describing charge transfer, multiconfigurational character, and excited states. From these embedding simulations, we gained important insights into fundamental processes that are crucial aspects of fuel cell catalysis (i.e., O2 reduction at metal surfaces) and plasmon-mediated photocatalysis by metal nanoparticles. Moreover, our findings agree very well with experimental observations, while offering new views into the chemistry. We finally discuss our recently formulated potential-functional embedding theory that provides a seamless, first-principles way to include back-action onto the environment from the embedded region.
Co-reporter:Christoph Riplinger ; Matthew D. Sampson ; Andrew M. Ritzmann ; Clifford P. Kubiak
Journal of the American Chemical Society 2014 Volume 136(Issue 46) pp:16285-16298
Publication Date(Web):October 17, 2014
DOI:10.1021/ja508192y
[Re(bpy)(CO)3]− is a well-established homogeneous electrocatalyst for the reduction of CO2 to CO. Recently, substitution of the more abundant transition metal Mn for Re yielded a similarly active electrocatalyst, [Mn(bpy)(CO)3]−. Compared to the Re catalyst, this Mn catalyst operates at a lower applied reduction potential but requires the presence of a weak acid in the solution for catalytic activity. In this study, we employ quantum chemistry combined with continuum solvation and microkinetics to examine the mechanism of CO2 reduction by each catalyst. We use cyclic voltammetry experiments to determine the turnover frequencies of the Mn catalyst with phenol as the added weak acid. The computed turnover frequencies for both catalysts agree to within one order of magnitude of the experimental ones. The different operating potentials for these catalysts indicate that different reduction pathways may be favored during catalysis. We model two different pathways for both catalysts and find that, at their respective operating potentials, the Mn catalyst indeed is predicted to take a different reaction route than the Re catalyst. The Mn catalyst can access both catalytic pathways, depending on the applied potential, while the Re catalyst does not show this flexibility. Our microkinetics analysis predicts which intermediates should be observable during catalysis. These intermediates for the two catalyzed reactions have qualitatively different electronic configurations, depending on the applied potential. The observable intermediate at higher applied potentials possesses an unpaired electron and therefore should be EPR-active; however, the observable intermediate at lower applied potentials, accessible only for the Mn catalyst, is diamagnetic and therefore should be EPR-silent. The differences between both catalysts are rationalized on the basis of their electronic structure and different ligand binding affinities.
Co-reporter:Andrew M. Ritzmann, Michele Pavone, Ana B. Muñoz-García, John A. Keith and Emily A. Carter
Journal of Materials Chemistry A 2014 vol. 2(Issue 21) pp:8060-8074
Publication Date(Web):08 Apr 2014
DOI:10.1039/C4TA00801D
Although solid oxide fuel cells (SOFCs) provide clean and efficient electricity generation, high operating temperatures (T > 800 °C) limit their widespread use. Lowering operating temperatures (600 °C < T < 800 °C) requires developing next-generation mixed ion-electron conducting (MIEC) cathodes that permit facile oxygen transport. One promising MIEC material, La1−xSrxCo1−yFeyO3 (LSCF), can operate at intermediate temperatures, has a longer cell lifetime, and permits less expensive interconnect materials. However, the road to optimization of LSCF compositions for SOFC applications would benefit from fundamental, atomic-scale insight into how local chemical changes affect its oxygen ion conductivity. We provide this insight using ab initio density functional theory plus U (DFT+U) calculations to analyze the factors governing oxygen transport in the LSCF parent material LaCoO3. We show that oxygen diffusion in LaCoO3 depends strongly on the spin state of the Co3+ ions: in particular, low spin Co3+ promotes higher oxygen vacancy concentrations than other spin states. We also predict that different spin states of Co3+ significantly affect the oxygen ion migration barrier. Through electronic structure analysis, we uncover the fundamental details which govern oxygen diffusivity in LaCoO3.
Co-reporter:Junchao Xia, Emily A. Carter
Journal of Power Sources 2014 Volume 254() pp:62-72
Publication Date(Web):15 May 2014
DOI:10.1016/j.jpowsour.2013.12.097
•Orbital-free density functional theory (OFDFT) is used to study Li–Si alloys.•Accurate cell lattice vectors, bulk moduli, and densities are predicted by OFDFT.•OFDFT elastic constants and alloy formation energies agree well with KSDFT.•OFDFT Li atom adsorption energies on a Si(100) surface are close to KSDFT values.•Linear-scaling OFDFT appears quite promising for large-scale Li–Si simulations.Li–Si interactions are of great interest currently due to the potential use of silicon anodes in Li-ion batteries. As a first step toward eventual nanoscale characterization of lithiation of silicon, here we study the crystalline Li–Si alloys LiSi, Li12Si7, Li7Si3, Li13Si4, Li15Si4, and Li22Si5 using orbital-free density functional theory (OFDFT). The recently proposed Wang–Govind–Carter decomposition (WGCD) and Huang–Carter (HC) kinetic energy density functionals (KEDFs) are used to evaluate the electron kinetic energy. Both KEDFs predict accurate cell lattice vectors, equilibrium volumes, bulk moduli, and ground-state densities when compared to Kohn–Sham density functional theory (KSDFT) benchmarks. Elastic constants and alloy formation energies calculated with the WGCD KEDF also agree reasonably well with KSDFT. Finally, Li atom adsorption energies on the Si(100) − 2 × 1 surface are calculated as a simple initial test of the Li–Si mixing process during lithiation of silicon. The OFDFT adsorption energies again are fairly close to KSDFT values. The results in this work demonstrate the accuracy of the WGCD and HC KEDFs for materials with mixed covalent-metallic character and their considerable transferability under different chemical environments. Because of its quasilinear scaling, coupled with the level of accuracy shown here, OFDFT appears quite promising for large-scale simulation of such materials phenomena.
Co-reporter:Ilgyou Shin, Emily A. Carter
Acta Materialia 2014 Volume 64() pp:198-207
Publication Date(Web):February 2014
DOI:10.1016/j.actamat.2013.10.030
Abstract
First-principles quantum mechanics is an increasingly important tool for predicting material properties when designing novel alloys with optimized mechanical properties. In this study, we employ first-principles orbital-free density functional theory (OFDFT) to study plastic properties of body-centered-cubic (bcc) Mg–Li alloys as potential lightweight metals for use in, e.g., vehicle applications. The accuracy of the method as a predictive tool is benchmarked against the more accurate Kohn–Sham DFT (KSDFT). With a new analytic local electron–ion pseudopotential, OFDFT is shown to be comparable in accuracy to KSDFT with the conventional non-local pseudopotential for many properties of Mg–Li alloys, including lattice parameters and energy differences between phases. After this validation, we calculate generalized stacking fault energies (SFEs) of a perfect lattice and Peierls stresses (σp’s) for dislocation motion in various bcc Mg–Li alloys. Such predictions have not been made previously with any level of theory. Based on analysis of SFE barriers, we propose that alloys with 31–50 at.% Li will exhibit the greatest strength. Their σp’s are predicted to be 0.18–0.31 GPa. The Li concentration in this range (31–50 at.%) has little impact on plastic properties of bcc Mg–Li alloys, while atomic-level disorder may decrease the σp. This range of σp is similar to the industrial goal for potential lightweight Mg alloys.
Co-reporter:Johannes M. Dieterich, David B. Krisiloff, Alexander Gaenko, Florian Libisch, Theresa L. Windus, Mark S. Gordon, Emily A. Carter
Computer Physics Communications 2014 Volume 185(Issue 12) pp:3175-3188
Publication Date(Web):December 2014
DOI:10.1016/j.cpc.2014.08.016
We present a shared-memory parallelization of our open-source, local correlation multi-reference framework, TigerCI. Benchmarks of the total parallel speedup show a reasonable scaling for typical modern computing system setups. The efficient use of available computing resources will extend simulations on this high level of theory into a new size regime. We demonstrate our framework using local-correlation multireference computations of alkyl-substituted dioxirane and solvated methyl nitrene as examples.
Co-reporter:Dr. Dalal K. Kanan;Dr. John A. Keith;Dr. Emily A. Carter
ChemElectroChem 2014 Volume 1( Issue 2) pp:407-415
Publication Date(Web):
DOI:10.1002/celc.201300089
Abstract
The demand for renewable hydrogen derived from CO2-neutral water-splitting processes spurs efforts to develop new catalysts, including those inspired by nature. A first-principles quantum mechanics (Kohn–Sham density functional theory + U) approach has been used to model electrocatalytic water oxidation on the visible-light-absorbing transition-metal oxide alloy, MnO:ZnO; a material that can be considered a heterogeneous analogue to the photosystem II photocatalyst. Ab-initio-derived U values were used to correct self-interaction errors in the highly correlated material. It has been confirmed that previously established scaling relationships between the binding energies of reaction intermediates are valid. The predicted electrochemical overpotential for water oxidation under experimentally relevant conditions (0.82 V versus the standard hydrogen electrode) is slightly higher than those values reported for manganese oxides and comparable to those previously calculated values for hematite photoanodes.
Co-reporter:Victor B. Oyeyemi, John A. Keith, and Emily A. Carter
The Journal of Physical Chemistry A 2014 Volume 118(Issue 17) pp:3039-3050
Publication Date(Web):April 7, 2014
DOI:10.1021/jp501636r
As part of our ongoing investigation of the combustion chemistry of oxygenated molecules using multireference correlated wave function methods, we report bond dissociation energies (BDEs) in C1–C4 alcohols (from methanol to the four isomers of butanol) and C1–C4 aldehydes (from methanal to butanal). The BDEs are calculated with a multireference averaged coupled-pair functional-based scheme. We compare these multireference BDEs with those derived from experiment and single-reference methods. Trends in BDEs for the alcohols and aldehydes are rationalized by considering geometry relaxations of dissociated radical fragments, resonance stabilization, and hyperconjugation. Lastly, we discuss the conjectured association between bond strengths and rates of hydrogen abstraction by hydroxyl radicals. In general, abstraction reaction rates are higher at sites where the C–H bond energies are lower (and vice versa). However, comparison with available rate data shows this inverse relationship between bond strengths and abstraction rates does not hold at all temperatures.
Co-reporter:Victor B. Oyeyemi, John A. Keith, and Emily A. Carter
The Journal of Physical Chemistry A 2014 Volume 118(Issue 35) pp:7392-7403
Publication Date(Web):March 13, 2014
DOI:10.1021/jp412727w
Accurate bond dissociation energies (BDEs) are important for characterizing combustion chemistry, particularly the initial stages of pyrolysis. Here we contribute to evaluating the thermochemistry of biodiesel methyl ester molecules using ab initio BDEs derived from a multireference averaged coupled-pair functional (MRACPF2)-based scheme. Having previously validated this approach for hydrocarbons and a variety of oxygenates, herein we provide further validation for bonds within carboxylic acids and methyl esters, finding our scheme predicts BDEs within chemical accuracy (i.e., within 1 kcal/mol) for these molecules. Insights into BDE trends with ester size are then analyzed for methyl formate through methyl crotonate. We find that the carbonyl group in the ester moiety has only a local effect on BDEs. C═C double bonds in ester alkyl chains are found to increase the strengths of bonds adjacent to the double bond. An important exception are bonds beta to C═C or C═O bonds, which produce allylic-like radicals upon dissociation. The observed trends arise from different degrees of geometric relaxation and resonance stabilization in the radicals produced. We also compute BDEs in various small alkanes and alkenes as models for the long hydrocarbon chain of actual biodiesel methyl esters. We again show that allylic bonds in the alkenes are much weaker than those in the small methyl esters, indicating that hydrogen abstractions are more likely at the allylic site and even more likely at bis-allylic sites of alkyl chains due to more electrons involved in π-resonance in the latter. Lastly, we use the BDEs in small surrogates to estimate heretofore unknown BDEs in large methyl esters of biodiesel fuels.
Co-reporter:Nima Alidoust, Maytal Caspary Toroker, and Emily A. Carter
The Journal of Physical Chemistry B 2014 Volume 118(Issue 28) pp:7963-7971
Publication Date(Web):April 1, 2014
DOI:10.1021/jp500878s
We use two different ab initio quantum mechanics methods, complete active space self-consistent field theory applied to electrostatically embedded clusters and periodic many-body G0W0 calculations, to reanalyze the states formed in nickel(II) oxide upon electron addition and ionization. In agreement with interpretations of earlier measurements, we find that the valence and conduction band edges consist of oxygen and nickel states, respectively. However, contrary to conventional wisdom, we find that the oxygen states of the valence band edge are localized whereas the nickel states at the conduction band edge are delocalized. We argue that these characteristics may lead to low electron–hole recombination and relatively efficient electron transport, which, coupled with band gap engineering, could produce higher solar energy conversion efficiency compared to that of other transition-metal oxides. Both methods find a photoemission/inverse-photoemission gap of 3.6–3.9 eV, in good agreement with the experimental range, lending credence to our analysis of the electronic structure of NiO.
Co-reporter:Nima Alidoust;Dr. Maytal Caspary Toroker;Dr. John A. Keith;Dr. Emily A. Carter
ChemSusChem 2014 Volume 7( Issue 1) pp:195-201
Publication Date(Web):
DOI:10.1002/cssc.201300595
Abstract
Long-term sustainable solar energy conversion relies on identifying economical and versatile semiconductor materials with appropriate band structures for photovoltaic and photocatalytic applications (e.g., band gaps of ∼1.5–2.0 eV). Nickel oxide (NiO) is an inexpensive yet highly promising candidate. Its charge-transfer character may lead to longer carrier lifetimes needed for higher efficiencies, and its conduction band edge is suitable for driving hydrogen evolution via water-splitting. However, NiO’s large band gap (∼4 eV) severely limits its use in practical applications. Our first-principles quantum mechanics calculations show band gaps dramatically decrease to ∼2.0 eV when NiO is alloyed with Li2O. We show that LixNi1−xO alloys (with x=0.125 and 0.25) are p-type semiconductors, contain states with no impurity levels in the gap and maintain NiO’s desirable charge-transfer character. Lastly, we show that the alloys have potential for photoelectrochemical applications, with band edges well-placed for photocatalytic hydrogen production and CO2 reduction, as well as in tandem dye-sensitized solar cells as a photocathode.
Co-reporter:Michele Pavone ; Ana B. Muñoz-García ; Andrew M. Ritzmann
The Journal of Physical Chemistry C 2014 Volume 118(Issue 25) pp:13346-13356
Publication Date(Web):June 3, 2014
DOI:10.1021/jp500352h
We characterize the structural, electronic, and defect behavior of La1–xSrxMnO3 (LSM) (xSr = 0.0, 0.25, and 0.5) by means of density functional theory + U (DFT+U) and hybrid DFT methods. Aliovalent substitution of Sr2+ for La3+ induces formation of holes in the LSM electronic structure. These holes affect electron and oxide ion transport, two key processes occurring within LSM when used as a solid oxide fuel cell (SOFC) cathode. To improve fundamental understanding of these processes, we investigated the atomic-scale effects of increasing Sr content and two different Mn magnetic moment alignments. In agreement with low-temperature experiments, we find a metallic, ferromagnetic (FM) electronic ground state with holes delocalized across the Mn and O sublattices. We also employ an antiferromagnetic (AFM) arrangement of Mn ions to model LSM’s high-temperature paramagnetic state. In contrast to FM LSM, the holes in AFM LSM localize to form Mn4+ ions, consistent with the observed high-temperature polaronic transport in LSM. The formation of oxygen vacancies governs oxide ion transport in bulk LSM. We find that the ease with which oxygen vacancies form is strongly influenced by the Sr content and the overall magnetic arrangement of Mn ions. These atomic-scale insights enable us to propose new guidelines for enhanced nanoscale LSM-based SOFC cathodes.
Co-reporter:Peilin Liao and Emily A. Carter
Chemical Society Reviews 2013 vol. 42(Issue 6) pp:2401-2422
Publication Date(Web):30 Oct 2012
DOI:10.1039/C2CS35267B
Photocatalytic production of transportation fuels should be among our long term strategies to achieve energy and environmental sustainability for the planet, but the technology is hampered by a lack of sufficiently efficient catalysts. Although efficiency is ultimately determined by laboratory measurements, theory and computation have become powerful tools for examining underlying mechanisms and guiding avenues of inquiry. In this review, we focus on first principles calculations of transition metal oxide semiconductor photocatalysts. We discuss how theory can be applied to investigate various aspects of a photocatalytic cycle: light absorption, electron/hole transport, band edge alignments of semiconductors, and surface chemistry. Emphasis is placed on identifying accurate models for specific properties and theoretical insights into improving photocatalytic performance.
Co-reporter:John A. Keith ; Kyle A. Grice ; Clifford P. Kubiak
Journal of the American Chemical Society 2013 Volume 135(Issue 42) pp:15823-15829
Publication Date(Web):September 20, 2013
DOI:10.1021/ja406456g
A complete mechanism for the proton-dependent electrocatalytic reduction of CO2 to CO by fac-Re(bpy)(CO)3Cl that is consistent with experimental observations has been developed using first principles quantum chemistry. Calculated one-electron reduction potentials, nonaqueous pKa’s, reaction free energies, and reaction barrier heights provide deep insight into the complex mechanism for CO2 reduction as well as the origin of selectivity for this catalyst. Protonation and then reduction of a metastable Re–CO2 intermediate anion precedes Brønsted-acid-catalyzed C–O cleavage and then rapid release of CO at negative applied potentials. Conceptually understanding the mechanism of this rapid catalytic process provides a useful blueprint for future work in artificial photosynthesis.
Co-reporter:John A. Keith and Emily A. Carter
Chemical Science 2013 vol. 4(Issue 4) pp:1490-1496
Publication Date(Web):14 Feb 2013
DOI:10.1039/C3SC22296A
One of the most promising CO2 reduction processes presently known suffers from a lack of fundamental understanding of its reaction mechanism. Using first principles quantum chemistry, we report thermodynamical energies of various pyridine-derived intermediates as well as barrier heights for key homogeneous reaction mechanisms. From this work, we predict that the actual form of the co-catalyst involved in pyridinium-based CO2 reduction is not the long-proposed pyridinyl radical in solution, but is more probably a surface-bound dihydropyridine species.
Co-reporter:Leah Y. Isseroff and Emily A. Carter
Chemistry of Materials 2013 Volume 25(Issue 3) pp:253
Publication Date(Web):January 9, 2013
DOI:10.1021/cm3040278
Cuprous oxide (Cu2O) is an attractive material for solar energy applications, but its photoconductivity is limited by minority carrier recombination caused by native defect trap states. We examine the creation of trap states by cation vacancies, using first principles calculations based on density functional theory (DFT) to analyze the electronic structure and calculate formation energies. With several DFT-based methods, a simple vacancy is predicted to be consistently more stable than a split vacancy by 0.21 ± 0.03 eV. Hybrid DFT is used to analyze the density of states and charge density distribution, predicting a delocalized hole for the simple vacancy and a localized hole for the split vacancy, in contrast to previously reported results. The differing character of the two defects indicates that they contribute to conduction via different mechanisms, with the split vacancy as the origin of the acceptor states that trap minority carriers. We explore methods of improving photoconductivity by doping Cu2O with Li, Mg, Mn, and Zn, analyzing their impact on vacancy formation energies and electronic structures. Results suggest that the Li dopant has the greatest potential to improve the photoconductivity of the oxide by inhibiting the creation of trap states.Keywords: Cu2O; density functional theory; doping; photoconductivity; solar energy; trap states; vacancy formation energy;
Co-reporter:Andrew M. Ritzmann, Ana B. Muñoz-García, Michele Pavone, John A. Keith, and Emily A. Carter
Chemistry of Materials 2013 Volume 25(Issue 15) pp:3011
Publication Date(Web):June 27, 2013
DOI:10.1021/cm401052w
Incorporating mixed oxygen-ion-electron conducting (MIEC) cathode materials is a promising strategy to make intermediate-temperature solid oxide fuel cells (IT-SOFCs) viable; however, a lack of fundamental understanding of oxygen transport in these materials limits their development. Density functional theory plus U (DFT+U) calculations are used to investigate how the Sr concentration affects the processes that govern oxygen ion transport in La1-xSrxFeO3-δ (LSF, x = 0, 0.25, and 0.50). Specifically, we show that oxygen vacancies compensate holes introduced by Sr and that this compensation facilitates oxygen vacancy formation in LSF. We also find that oxygen migration in LaFeO3 is accompanied by electron transfer in the opposite direction. Our results explicitly identify and clarify the role of electron-deficient substitutions in promoting oxygen diffusion in LSF. This atomic level insight is important for enabling rational design of iron-based SOFC cathode materials.Keywords: La1-xSrxFeO3; oxygen migration; oxygen vacancies; solid-oxide fuel cells;
Co-reporter:Maytal Caspary Toroker and Emily A. Carter
Journal of Materials Chemistry A 2013 vol. 1(Issue 7) pp:2474-2484
Publication Date(Web):17 Dec 2012
DOI:10.1039/C2TA00816E
First-row transition metal oxides (TMOs) are inexpensive potential alternative materials for solar energy conversion devices. However, some TMOs, such as manganese(II) oxide, have band gaps that are too large for efficiently absorbing solar energy. Other TMOs, such as iron(II) oxide, have conduction and valence band edges with the same orbital character that may lead to unfavorably high electron–hole recombination rates. Another limitation of iron(II) oxide is that the calculated valence band edge is not positioned well for oxidizing water. We predict that key properties, including band gaps, band edge positions, and possibly electron–hole recombination rates, may be improved by alloying TMOs that have different band alignments. A new metric, the band gap center offset, is introduced for simple screening of potential parent materials. The concept is illustrated by calculating the electronic structure of binary oxide alloys that contain manganese, nickel, iron, zinc, and/or magnesium, within density functional theory (DFT)+U and hybrid DFT theories. We conclude that alloys of iron(II) oxide are worth evaluating further as solar energy conversion materials.
Co-reporter:Dalal K. Kanan and Emily A. Carter
Journal of Materials Chemistry A 2013 vol. 1(Issue 32) pp:9246-9256
Publication Date(Web):07 Jun 2013
DOI:10.1039/C3TA11265A
Rationally engineering photocatalytic devices that power water splitting or CO2 reduction reactions requires identifying economical materials that efficiently absorb sunlight and have suitable band edge placements. Recent theoretical investigations have predicted that a 1:1 alloy of MnO and ZnO meets these criteria. However, poor hole conductivity in undoped MnO:ZnO alloys (with up to 10% ZnO) severely limits this material's utility in electronic devices, and its electron conductivity has not yet been characterized. Here we investigate carrier transport in pure and doped MnO and MnO:ZnO with ab initio quantum chemistry calculations. Electrostatically embedded clusters are used to compute and compare relative electron/hole transfer barriers within the small polaron model. We assess the effects of Al, Ga, In, Sc, Y, Ti, Sb, Gd, F (n-type dopants) and Li (a p-type dopant) to determine which may enhance conductivity in MnO:ZnO. Our findings indicate that Ga, Sc, Ti, F, and Sb dopants create deep traps whereas In forms shallower traps that merit further investigation. Y, Al, Gd, and Li dopants should increase the carrier concentration while maintaining favorable electron and hole transport pathways. The latter are recommended for increasing the conductivity of MnO:ZnO and its effectiveness for solar energy conversion.
Co-reporter:Ana B. Muñoz-García, Michele Pavone, Andrew M. Ritzmann and Emily A. Carter
Physical Chemistry Chemical Physics 2013 vol. 15(Issue 17) pp:6250-6259
Publication Date(Web):07 Mar 2013
DOI:10.1039/C3CP50995H
We use ab initio density functional theory + U calculations to characterize the oxide ion diffusion process in bulk Sr2Fe1.5Mo0.5O6−δ (SFMO) by analyzing the formation and migration of oxygen vacancies. We show that SFMO's remarkable ionic conductivity arises from its intrinsic content of oxygen vacancies and a predicted very low migration barrier of such vacancies. Theoretical analysis of the electronic structure reveals a crucial role played by strongly hybridized Fe 3d/O 2p states to achieve the attendant mixed ion-electron conductor character so important for intermediate temperature fuel cell operation. We predict a next-nearest-neighbor-type migration pathway for the O2− ion should dominate. The low energy barrier of this pathway is mainly related to electrostatic interactions with homogeneously distributed Mo in the SFMO sublattice. We identify the reasons why Fe-rich perovskites, with the key addition of a certain concentration of Mo, produce excellent electronic and ionic transport properties so crucial for efficient operation of intermediate temperature solid oxide fuel cells.
Co-reporter:Leah Isseroff Bendavid, Emily A. Carter
Surface Science 2013 Volume 618() pp:62-71
Publication Date(Web):December 2013
DOI:10.1016/j.susc.2013.07.027
•Calculated adhesion energies of Cu2O(111)/ZnO(101¯0) show it is a weakly stable interface•Adhesion occurs via localized chemical interactions between substrate and coating•ZnO modifies the electronic structure of the Cu2O coating, reducing its band gapDensity functional theory (DFT)-based methods are used to understand atomic level interactions and calculate adhesion energies of the Cu2O(111)/ZnO101¯0 interface with varying Cu2O coating thickness. We first establish an accurate model of the ZnO substrate, validating DFT + U against the more accurate hybrid-DFT to calculate properties of bulk wurtzite ZnO and the ZnO101¯0 surface. DFT + U is then used to analyze the structure of the Cu2O(111) surface, characterizing the formation of surface copper dimers. The Cu2O(111)/ZnO101¯0 interface is found to be only weakly interacting, with a DFT + U–derived adhesion energy of 0.85 ± 0.07 J/m2. Charge density analysis reveals that some interface stabilization occurs because of local ZnO and CuO bonding interactions at the interface. We find that the overall impact of the ZnO substrate on the electronic structure of the Cu2O overlayer is to reduce the Cu2O band gap.
Co-reporter:Dalal K. Kanan, John A. Keith, Emily A. Carter
Surface Science 2013 Volume 617() pp:218-224
Publication Date(Web):November 2013
DOI:10.1016/j.susc.2013.07.023
•DFT + U study of water–electrode interface provides water/MnO:ZnO(001) phase diagram.•MnO:ZnO(001) is hydrophilic: ~ 50% water dissociates under various water conditions.•MnO:ZnO work function is sensitive to water orientation and adsorbed hydroxyls.•Photocatalytic activity of MnO:ZnO could be enhanced by H2O dissociative adsorption.Improving photochemical water oxidation processes on sunlight absorbing materials requires understanding the photoelectrode-solution interface. We use ab initio density functional theory (DFT) + U to investigate the structure and energetics of water adsorbed on MnO:ZnO(001), a potential photoanode material we previously identified as having suitable band gaps and band edge placements for visible light induced water splitting. Our calculations show that there is a preference for molecular adsorption at water coverages of less than half a monolayer (ML). At higher coverages, cooperative water–water interactions facilitate water dissociation at the interface. We find that the work function is very sensitive to water dipole orientation and/or presence of hydroxyls on the surface. The computed phase diagram reveals the surface to be fairly hydrophilic with a preference for the first water ML to be 33% dissociated at 0.75 ML, 50% dissociated at 1 ML, and 50% dissociated at 2 ML water coverage under various conditions away from water-poor conditions.
Co-reporter:Leah Isseroff Bendavid and Emily A. Carter
The Journal of Physical Chemistry C 2013 Volume 117(Issue 49) pp:26048-26059
Publication Date(Web):November 19, 2013
DOI:10.1021/jp407468t
Adsorption of CO2 on the Cu2O(111) surface is investigated using density functional theory + U with and without dispersion corrections. A number of adsorbate geometries are considered on four different surface terminations that include the bulk-terminated surface and surfaces with oxygen and/or copper vacancies. CO2 is found to adsorb most strongly as a tilted linear molecule coordinated to an unsaturated surface cation. Surface vacancies allow for bent adsorbate configurations to be accessed but they are all less stable than the linear adsorbate. Bader analysis confirms that adsorption of bent CO2 is accompanied by charge transfer from the surface to the molecule, whereas minimal charge transfer occurs in linear physisorption. We show that surface oxygen vacancies have a small impact on adsorption free energies, while surface copper vacancies result in a significant reduction of CO2 adsorption. Including dispersion corrections increases the stability of adsorbed CO2, but adsorption is mostly endoergic at 298.15 K and 1 atm. Our findings reveal that adsorption of CO2 on cuprous oxide is contingent on the presence of copper cations at the surface. Thus, this surface’s chemistry is dominated by Lewis acidity, in contrast to other oxide surfaces where oxygen anions may act as Lewis bases to form carbonates. The suppression of carbonate formation is important, as it confirms that this surface may be useful for CO2 reduction to fuels.
Co-reporter:John A. Keith and Emily A. Carter
The Journal of Physical Chemistry Letters 2013 Volume 4(Issue 23) pp:4058-4063
Publication Date(Web):November 7, 2013
DOI:10.1021/jz4021519
We propose a novel and general reaction mechanism to explain the unique performance of nitrogen-heterocycle-promoted (photo)electrochemical CO2 reduction reactions. This mechanism is based on observations from recent computational and experimental studies of pyridinium-catalyzed CO2 reduction. Herein we report pKas and standard reduction potentials of species adsorbed on GaP photoelectrodes derived from first-principles quantum chemistry computations. We show that on GaP surfaces, proton reduction or pyridinium reduction is energetically unfavorable even at very negative electrode potentials. However, it is thermodynamically favorable to convert a surface-bound pyridine into a 2e– reduced species such as dihydropyridine at less negative applied potentials. Intriguingly, these transient 2e– reduced species share a similar chemical moiety as some biological redox catalysts (e.g., NADH), and their reduction potentials are similar to the thermodynamic redox potentials that would convert CO2 to a variety of products.Keywords: CO2 conversion; density functional theory; electrocatalysis; reduction potentials; renewable energy;
Co-reporter:Dalal K. Kanan and Emily A. Carter
The Journal of Physical Chemistry C 2013 Volume 117(Issue 27) pp:13816-13826
Publication Date(Web):June 13, 2013
DOI:10.1021/jp4024475
MnO:ZnO was recently proposed as a novel, potentially visible-light absorbing catalyst with many advantages for water splitting. However, little is known about its optical absorption properties and whether they are suited for solar applications. We address this issue by examining excitations in MnO:ZnO and pure MnO using electrostatically embedded cluster models treated with complete active space second order perturbation (CASPT2) theory. The lowest-lying excitations in MnO are single Mn d → d ligand field excitations (∼2.6 eV, ∼108 s lifetime), followed by double d → d excitations (∼5.2 eV, ∼106 s lifetime), Mn 3d → 4s excitations (∼6.3 eV, ∼10–3 s lifetime), and higher-lying O 2p → Mn 3d ligand-to-metal charge-transfer (LMCT) excitations (∼10.1 eV, ∼10–4 s lifetime). The longer-lived transitions should exhibit better electron–hole pair separation and enhance photoconductivity depending on ease of carrier transport. Alloying MnO with ZnO notably lowers the LMCT transition to ∼8.3 eV with the metal orbital comprised mainly of Zn 4s states, leaving most other absorption properties relatively unchanged. We find near degeneracies among spin-allowed and spin-forbidden LMCT states that could facilitate intersystem crossing resulting in longer lifetimes. We suggest seeking other materials that exhibit similar LMCT excitations but that are visible-light activated as a design strategy for further enhancing photon conversion efficiencies.
Co-reporter:Leah Isseroff Bendavid and Emily A. Carter
The Journal of Physical Chemistry B 2013 Volume 117(Issue 49) pp:15750-15760
Publication Date(Web):September 27, 2013
DOI:10.1021/jp406454c
For a photocatalytic reaction to be thermodynamically allowed, a semiconductor’s band edges need to be placed appropriately relative to the reaction redox potentials. We apply a recently developed scheme for calculating band edges with density functional theory (DFT)-based methods to Cu2O, evaluating its available thermodynamic overpotential for redox reactions such as water splitting and conversion of CO2 to methanol. Because these calculations are surface dependent, we first study the low-index surfaces of Cu2O using periodic DFT+U theory to characterize and identify the most stable surface, which will be the most catalytically relevant. We employ various techniques to calculate the surface energy, including the method of “ab initio atomistic thermodynamics”. The Cu2O(111) surface with (1 × 1) periodicity and surface copper vacancies is identified as the most stable at all oxygen partial pressures, although the ideal stoichiometric Cu2O(111) surface is relatively close in energy under oxygen-poor conditions. These surfaces are then used to calculate the band edges. Comparison of the band edges to redox potentials reveals that Cu2O is thermodynamically capable of photocatalytic reduction of CO2 to methanol and the reduction and oxidation of water.
Co-reporter:Ana B. Muñoz-García
Journal of the American Chemical Society 2012 Volume 134(Issue 33) pp:13600-13603
Publication Date(Web):August 8, 2012
DOI:10.1021/ja3063106
The structural and electronic properties of the GaP(110)/H2O interface have been investigated by first-principles density functional theory calculations. Our results suggest that hydride-like H atoms are present on the surface as a consequence of the dissociation of water in contact with the GaP surface. This feature opens up a new feasible reduction pathway for CO2 where the GaP(110) surface is the electrochemically active entity.
Co-reporter:Ana B. Muñoz-García ; Daniel E. Bugaris ; Michele Pavone ; Jason P. Hodges ; Ashfia Huq ; Fanglin Chen ; Hans-Conrad zur Loye
Journal of the American Chemical Society 2012 Volume 134(Issue 15) pp:6826-6833
Publication Date(Web):March 26, 2012
DOI:10.1021/ja300831k
We characterize experimentally and theoretically the promising new solid oxide fuel cell electrode material Sr2Fe1.5Mo0.5O6−δ (SFMO). Rietveld refinement of powder neutron diffraction data has determined that the crystal structure of this material is distorted from the ideal cubic simple perovskite, instead belonging to the orthorhombic space group Pnma. The refinement revealed the presence of oxygen vacancies in the as-synthesized material, resulting in a composition of Sr2Fe1.5Mo0.5O5.90(2) (δ = 0.10(2)). DFT+U theory predicts essentially the same concentration of oxygen vacancies. Theoretical analysis of the electronic structure allows us to elucidate the origin of this nonstoichiometry and the attendant mixed ion–electron conductor character so important for intermediate temperature fuel cell operation. The ease with which SFMO forms oxygen vacancies and allows for facile bulk oxide ion diffusivity is directly related to a strong hybridization of the Fe d and O p states, which is also responsible for its impressive electronic conductivity.
Co-reporter:John A. Keith
Journal of the American Chemical Society 2012 Volume 134(Issue 18) pp:7580-7583
Publication Date(Web):April 23, 2012
DOI:10.1021/ja300128e
The role of pyridinium cations in electrochemistry has been believed known for decades, and their radical forms have been proposed as key intermediates in modern photoelectrocatalytic CO2 reduction processes. Using first-principles density functional theory and continuum solvation models, we have calculated acidity constants for pyridinium cations and their corresponding pyridinyl radicals, as well as their electrochemical redox potentials. Contrary to previous assumptions, our results show that these species can be ruled out as active participants in homogeneous electrochemistry. A comparison of calculated acidities and redox potentials indicates that pyridinium cations behave differently than previously thought, and that the electrode surface plays a critical (but still unknown) role in pyridinium reduction. This work substantially alters the mechanistic view of pyridinium-catalyzed photoelectrochemical CO2 reduction.
Co-reporter:Peilin Liao, John A. Keith, and Emily A. Carter
Journal of the American Chemical Society 2012 Volume 134(Issue 32) pp:13296-13309
Publication Date(Web):July 12, 2012
DOI:10.1021/ja301567f
In photoelectrochemical cells, sunlight may be converted into chemical energy by splitting water into hydrogen and oxygen molecules. Hematite (α-Fe2O3) is a promising photoanode material for the water oxidation component of this process. Numerous research groups have attempted to improve hematite’s photocatalytic efficiency despite a lack of foundational knowledge regarding its surface reaction kinetics. To elucidate detailed reaction mechanisms and energetics, we performed periodic density functional theory + U calculations for the water oxidation reaction on the fully hydroxylated hematite (0001) surface. We investigate two different concentrations of surface reactive sites. Our best model involves calculating water oxidation mechanisms on a pure (1×1) hydroxylated hematite slab (corresponding to 1/3 ML of reactive sites) with an additional overlayer of water molecules to model solvation effects. This yields an overpotential of 0.77 V, a value only slightly above the 0.5–0.6 V experimental range. To explore whether doped hematite can exhibit an even lower overpotential, we consider cation doping by substitution of Fe by Ti, Mn, Co, Ni, or Si and F anion doping by replacing O on the fully hydroxylated surface. The reaction energetics on pure or doped hematite surfaces are described using a volcano plot. The relative stabilities of holes on the active O anions are identified as the underlying cause for trends in energetics predicted for different dopants. We show that moderately charged O anions give rise to smaller overpotentials. Co- or Ni-doped hematite surfaces give the most thermodynamically favored reaction pathway (lowest minimum overpotential) among all dopants considered. Very recent measurements (Electrochim. Acta2012, 59, 121–127) reported improved reactivity with Ni doping, further validating our predictions.
Co-reporter:John A. Keith and Emily A. Carter
Journal of Chemical Theory and Computation 2012 Volume 8(Issue 9) pp:3187-3206
Publication Date(Web):August 7, 2012
DOI:10.1021/ct300295g
Sensibly modeling (photo)electrocatalytic reactions involving proton and electron transfer with computational quantum chemistry requires accurate descriptions of protonated, deprotonated, and radical species in solution. Procedures to do this are generally nontrivial, especially in cases that involve radical anions that are unstable in the gas phase. Recently, pyridinium and the corresponding reduced neutral radical have been postulated as key catalysts in the reduction of CO2 to methanol. To assess practical methodologies to describe the acid/base chemistry of these species, we employed density functional theory (DFT) in tandem with implicit solvation models to calculate acidity constants for 22 substituted pyridinium cations and their corresponding pyridinyl radicals in water solvent. We first benchmarked our calculations against experimental pyridinium deprotonation energies in both gas and aqueous phases. DFT with hybrid exchange-correlation functionals provide chemical accuracy for gas-phase data and allow absolute prediction of experimental pKas with unsigned errors under 1 pKa unit. The accuracy of this economical pKa calculation approach was further verified by benchmarking against highly accurate (but very expensive) CCSD(T)-F12 calculations. We compare the relative importance and sensitivity of these energies to selection of solvation model, solvation energy definitions, implicit solvation cavity definition, basis sets, electron densities, model geometries, and mixed implicit/explicit models. After determining the most accurate model to reproduce experimentally-known pKas from first principles, we apply the same approach to predict pKas for radical pyridinyl species that have been proposed relevant under electrochemical conditions. This work provides considerable insight into the pitfalls using continuum solvation models, particularly when used for radical species.
Co-reporter:Victor B. Oyeyemi, John A. Keith, Michele Pavone, and Emily A. Carter
The Journal of Physical Chemistry Letters 2012 Volume 3(Issue 3) pp:289-293
Publication Date(Web):January 6, 2012
DOI:10.1021/jz201564g
Density functional theory (DFT) is often used to determine the electronic and geometric structures of molecules. While studying alkynyl radicals, we discovered that DFT exchange-correlation (XC) functionals containing less than ∼22% Hartree–Fock (HF) exchange led to qualitatively different structures than those predicted from ab initio HF and post-HF calculations or DFT XCs containing 25% or more HF exchange. We attribute this discrepancy to rehybridization at the radical center due to electron delocalization across the triple bonds of the alkynyl groups, which itself is an artifact of self-interaction and delocalization errors. Inclusion of sufficient exact exchange reduces these errors and suppresses this erroneous delocalization; we find that a threshold amount is needed for accurate structure determinations. Below this threshold, significant errors in predicted alkyne thermochemistry emerge as a consequence.Keywords: ab initio calculations; alkynyl, alkylethynyl radicals; delocalization error; density functional theory; exchange-correlation functionals; self-interaction error; thermochemistry;
Co-reporter:Dalal K. Kanan and Emily A. Carter
The Journal of Physical Chemistry C 2012 Volume 116(Issue 18) pp:9876-9887
Publication Date(Web):April 16, 2012
DOI:10.1021/jp300590d
Practical implementation of solar-powered water splitting and CO2 reduction to fuels requires the discovery of efficient and inexpensive photocatalytic (PC) materials. One possible materials design strategy aims to tune properties of relatively inexpensive transition metal oxide catalysts to increase sunlight absorption while preserving potential redox reactivity. Here we consider MnO for PC use by alloying it with ZnO in solid solutions. A combined density functional theory and GW scheme is used to study the band gap and band edge placements as a function of composition. We predict that alloying MnO with ZnO in varying amounts reduces MnO’s band gap for more efficient light absorption while maintaining advantageous band edge placements. The 1:1 alloy of MnO and ZnO is identified as a new (2.6 eV band gap) visible-light-absorbing material with band edges suitably placed with respect to both water-oxidation and CO2-reduction reactions, making it a compelling candidate for solar PC chemistry.
Co-reporter:Maytal Caspary Toroker and Emily A. Carter
The Journal of Physical Chemistry C 2012 Volume 116(Issue 33) pp:17403-17413
Publication Date(Web):July 23, 2012
DOI:10.1021/jp3047664
We propose ways to enhance the conductivity of wüstite (i.e., naturally p-type FeO), a visible-light-absorbing, inexpensive, abundant, and nontoxic potential alternative material for solar energy conversion devices. Unfortunately, the conversion efficiency of FeO is inhibited by its low hole conductivity. Increasing the iron vacancy concentration or adding p-type dopants improves FeO conductivity by increasing the number of holes; however, it is not known which strategy introduces larger energy traps that would hinder hole conductivity. Here we employ the small polaron model along with ab initio calculations on electrostatically embedded clusters to analyze the local trapping effects of iron vacancies and several substitutional p-type dopants that are soluble in FeO, including copper, nitrogen, lithium, and sodium, and also hydrogen as an interstitial dopant for comparison. We find that vacancies create stronger traps than dopants and that copper and nitrogen dopants form deeper traps than lithium, sodium, or hydrogen. Furthermore, hydrogen repels the hole and substantially decreases the trap formed by an iron vacancy. Because of the shallower traps formed compared to vacancies, lithium-, sodium-, or hydrogen-doped, nanostructured or alloyed FeO may be worth evaluating as a p-type semiconductor for solar energy conversion applications.
Co-reporter:Ting Tan, Michele Pavone, David B. Krisiloff, and Emily A. Carter
The Journal of Physical Chemistry A 2012 Volume 116(Issue 33) pp:8431-8443
Publication Date(Web):July 25, 2012
DOI:10.1021/jp304811z
Combustion of renewable biofuels, including energy-dense biodiesel, is expected to contribute significantly toward meeting future energy demands in the transportation sector. Elucidating detailed reaction mechanisms will be crucial to understanding biodiesel combustion, and hydrogen abstraction reactions are expected to dominate biodiesel combustion during ignition. In this work, we investigate hydrogen abstraction by the radicals H·, CH3·, O·, HO2·, and OH· from methyl formate, the simplest surrogate for complex biodiesels. We evaluate the H abstraction barrier heights and reaction enthalpies, using multireference correlated wave function methods including size-extensivity corrections and extrapolation to the complete basis set limit. The barrier heights predicted for abstraction by H·, CH3·, and O· are in excellent agreement with derived experimental values, with errors ≤1 kcal/mol. We also predict the reaction energetics for forming reactant complexes, transition states, and product complexes for reactions involving HO2· and OH·. High-pressure-limit rate constants are computed using transition state theory within the separable-hindered-rotor approximation for torsions and the harmonic oscillator approximation for other vibrational modes. The predicted rate constants differ significantly from those appearing in the latest combustion kinetics models of these reactions.
Co-reporter:Dalal K. Kanan, Sahar Sharifzadeh, Emily A. Carter
Chemical Physics Letters 2012 s 519–520() pp: 18-24
Publication Date(Web):
DOI:10.1016/j.cplett.2011.11.003
Co-reporter:Michele Pavone, Andrew M. Ritzmann and Emily A. Carter
Energy & Environmental Science 2011 vol. 4(Issue 12) pp:4933-4937
Publication Date(Web):17 Oct 2011
DOI:10.1039/C1EE02377B
Low oxide ion conductivity in perovskite-type transition metal oxides is one of the major problems with solid oxide fuel cells (SOFCs). Here, simple quantum mechanical analyses of LaMO3 (M = Cr, Mn, Fe, Co) materials provide new insights into what drives the relative ease of formation of oxygen vacancies, which is a prerequisite for and predictor of oxide ion bulk diffusion. From our results, we derive design principles based on easily measurable or computable properties to improve SOFC cathode materials.
Co-reporter:Peilin Liao, Maytal Caspary Toroker, and Emily A. Carter
Nano Letters 2011 Volume 11(Issue 4) pp:1775-1781
Publication Date(Web):March 22, 2011
DOI:10.1021/nl200356n
Hematite (α-Fe2O3) is a promising candidate for photoelectrochemical splitting of water. However, its intrinsically poor conductivity is a major drawback. Doping hematite to make it either p-type or n-type enhances its measured conductivity. We use quantum mechanics to understand how titanium, zirconium, silicon, or germanium n-type doping affects the electron transport mechanism in hematite. Our results suggest that zirconium, silicon, or germanium doping is superior to titanium doping because the former dopants do not act as electron trapping sites due to the higher instability of Zr(III) compared to Ti(III) and the more covalent interactions between silicon (germanium) and oxygen. This suggests that use of n-type dopants that easily ionize completely or promote covalent bonds to oxygen can provide more charge carriers while not inhibiting transport.
Co-reporter:Ivan Milas, Berit Hinnemann and Emily A. Carter
Journal of Materials Chemistry A 2011 vol. 21(Issue 5) pp:1447-1456
Publication Date(Web):26 Oct 2010
DOI:10.1039/C0JM02212H
We use density functional theory with periodic boundary conditions to investigate diffusion of Al, O, Pt, Hf, and Y on the α-Al2O3(0001) surface as a simple model to obtain insight into possible diffusion mechanisms occurring at alumina grain boundaries (GBs) in thermal barrier coatings (TBCs). We calculate diffusion pathways, activation energies, and diffusion constants. For the elements involved in alumina growth, we find that Al diffusion is facile and involves a simple hop from one most stable adsorption site to the next without local minima in between whereas O diffusion has a much higher barrier and has local minima along the minimum energy pathway. This trend is consistent with relative rates of diffusion at alumina GBs. Regarding common dopants in the metal (bond coat) alloy substrate from which alumina grows, we predict that Pt diffusion is facile but thermodynamically disfavored relative to Pt remaining in the bond coat alloy, while Hf and Y diffusion involve much higher barriers than Al diffusion. We use these results to rationalize some observations regarding the role of reactive elements Hf and Y as well as Pt in TBCs.
Co-reporter:Tsz S. Chwee and Emily A. Carter
Journal of Chemical Theory and Computation 2011 Volume 7(Issue 1) pp:103-111
Publication Date(Web):December 3, 2010
DOI:10.1021/ct100486q
We demonstrate that valence excited states in large molecules can be treated using local multireference singles and doubles configuration interaction (LMRSDCI). The interior eigenvalues corresponding to the excited states of interest are transformed and shifted to the extrema of the spectrum by way of oblique projections and a matrix shift within a modified Davidson diagonalization scheme. In this way, the approximate wave function associated with the excited state of interest can be isolated independently of the lower lying roots, and residual minimization is used for final convergence to the target eigenstate. We find that vertical excitation energies calculated using LMRSDCI are mostly within 0.2 eV of nonlocal MRSDCI values.
Co-reporter:Maytal Caspary Toroker, Dalal K. Kanan, Nima Alidoust, Leah Y. Isseroff, Peilin Liao and Emily A. Carter
Physical Chemistry Chemical Physics 2011 vol. 13(Issue 37) pp:16644-16654
Publication Date(Web):19 Aug 2011
DOI:10.1039/C1CP22128K
The positions of electronic band edges are one important metric for determining a material's capability to function in a solar energy conversion device that produces fuels from sunlight. In particular, the position of the valence band maximum (conduction band minimum) must lie lower (higher) in energy than the oxidation (reduction) reaction free energy in order for these reactions to be thermodynamically favorable. We present first principles quantum mechanics calculations of the band edge positions in five transition metal oxides and discuss the feasibility of using these materials in photoelectrochemical cells that produce fuels, including hydrogen, methane, methanol, and formic acid. The band gap center is determined within the framework of DFT+U theory. The valence band maximum (conduction band minimum) is found by subtracting (adding) half of the quasiparticle gap obtained from a non-self-consistent GW calculation. The calculations are validated against experimental data where possible; results for several materials including manganese(II) oxide, iron(II) oxide, iron(III) oxide, copper(I) oxide and nickel(II) oxide are presented.
Co-reporter:Kristen A. Marino;Berit Hinnemann
PNAS 2011 Volume 108 (Issue 14 ) pp:5480-5487
Publication Date(Web):2011-04-05
DOI:10.1073/pnas.1102426108
To maximize energy efficiency, gas turbine engines used in airplanes and for power generation operate at very high temperatures,
even above the melting point of the metal alloys from which they are comprised. This feat is accomplished in part via the
deposition of a multilayer, multicomponent thermal barrier coating (TBC), which lasts up to approximately 40,000 h before
failing. Understanding failure mechanisms can aid in designing circumvention strategies. We review results of quantum mechanics
calculations used to test hypotheses about impurities that harm TBCs and transition metal (TM) additives that render TBCs
more robust. In particular, we discovered a number of roles that Pt and early TMs such as Hf and Y additives play in extending
the lifetime of TBCs. Fundamental insight into the nature of the bonding created by such additives and its effect on high-temperature
evolution of the TBCs led to design principles that can be used to create materials for even more efficient engines.
Co-reporter:Linda Hung
The Journal of Physical Chemistry C 2011 Volume 115(Issue 14) pp:6269-6276
Publication Date(Web):March 16, 2011
DOI:10.1021/jp112196t
Elastic properties and mechanisms for tensile yielding of fcc Al nanowires are explored using orbital-free density functional theory. We quasistatically load ultrathin nanowires that have circular cross sections, diameters 1−8 nm, lengths up to ∼20 nm, and axes along the [111] direction. We find that Young’s modulus is roughly consistent for the nanowires and bulk Al but that the equilibrium interlayer spacing, elastic limit, and yield strength are diameter dependent. Plasticity is nucleated by 3-fold symmetric axial displacements of surface atoms for all nanowires examined. However, the nanowire with 4 nm diameter yields via partial slip before achieving the theoretical strength of bulk Al, while the 1 and 2 nm nanowires yield via amorphous mechanisms, above or near the theoretical strength. These results give new insight into the plastic yielding mechanisms of ultrathin fcc nanowires with edge-free cross sections and [111] orientation.
Co-reporter:Peilin Liao
The Journal of Physical Chemistry C 2011 Volume 115(Issue 42) pp:20795-20805
Publication Date(Web):September 14, 2011
DOI:10.1021/jp206991v
Hematite is a candidate for use as a photoanode in water-splitting reactions. Part of the efficiency depends on the capability of hematite to absorb sunlight and convert solar energy to electron energy. While there have been optical spectroscopy measurements as well as molecular orbital studies on the optical spectra of hematite, accurate characterizations of excited states from theory are missing. To fill this gap, we study excited states of electrostatically embedded hematite clusters using complete active space self-consistent field theory and complete active space with second-order perturbation theory. Overall, we found that the lowest lying excitations within hematite are Fe d–d ligand field transitions (starting at ∼2.5 eV), which are highly localized around Fe centers. The O 2p to Fe 3d ligand to metal charge transfer excitations are higher lying excited states (∼6 eV). These excitation energies are used to verify some of the earlier peak assignments for the optical spectra of hematite. In addition, we demonstrate that density functional theory energy differences between different spin states of Fe in the embedded FeO69– cluster are significantly biased by the choice of exchange-correlation functional. Therefore, any conclusions derived from those types of calculations should be viewed with caution.
Co-reporter:Peilin Liao and Emily A. Carter
Journal of Materials Chemistry A 2010 vol. 20(Issue 32) pp:6703-6719
Publication Date(Web):05 Jul 2010
DOI:10.1039/C0JM01199A
Stress-corrosion cracking of steel is a ubiquitous phenomenon in which steel is chemically corroded, followed by fracture induced by applied loads. The products of steel corrosion in moist air at high temperature (>570 °C) are α-Fe2O3, FeO, and Fe3O4. Here we employ an ab initio density functional theory + U method to predict the tensile properties of these oxides, to gain insight into failure mechanisms. The universal binding energy relationship of Hayes et al. is employed to extrapolate atomic scale data to macroscopic sample sizes. The extrapolated macroscopic predictions are consistent with experimental measurements. The ordering of tensile strengths is shown to be FeO < Fe3O4 < α-Fe2O3, which correlates with increasing ionicity (as given by the formal charge on the Fe cations) in the three oxides. The direction that has weakest tensile strength is predicted to be along [01(3/2)(a/c)22] for α-Fe2O3 and along [001] for both FeO and Fe3O4. The direction dependence of tensile properties for these three iron oxides can be understood via a local bond strain analysis. We also predict that loading FeO along [110] and Fe3O4 along [001] or [111] produces a plastic response prior to brittle fracture at high temperature.
Co-reporter:Kristen A. Marino, Emily A. Carter
Acta Materialia 2010 Volume 58(Issue 7) pp:2726-2737
Publication Date(Web):April 2010
DOI:10.1016/j.actamat.2010.01.008
Abstract
First-principles density functional theory calculations are used to study Al diffusion in β-NiAl. The activation energy and diffusion constant pre-exponential factors are calculated for five previously postulated Al diffusion mechanisms: next-nearest-neighbor Al jumps, the triple defect mechanism and three variants of the six-jump cycle mechanism beginning with an Al vacancy. We predict that the triple defect mechanism has the lowest activation energy and is the mechanism by which Al diffusion occurs in NiAl. In order to elucidate why Pt has a beneficial effect on thermal barrier coating lifetime, the effect of Pt on each of these mechanisms is also examined. In all cases, Pt decreases the diffusion activation energy, which should enhance Al diffusion in the coatings.
Co-reporter:Donald F. Johnson, Emily A. Carter
Acta Materialia 2010 Volume 58(Issue 2) pp:638-648
Publication Date(Web):January 2010
DOI:10.1016/j.actamat.2009.09.042
Abstract
We characterize the ability of two potential surface alloys, FeAl and Fe3Si, to prevent H incorporation into steel, with a view toward inhibiting steel embrittlement. Periodic density functional theory calculations within the generalized gradient approximation are used to evaluate H dissolution energetics and the kinetics of H diffusion into and through FeAl and Fe3Si. We predict increased dissolution endothermicities and diffusion barriers in both alloys compared to bulk Fe. Fe3Si is predicted to be the most effective at inhibiting H incorporation, with a 1.91 eV [0.97 eV] surface-to-subsurface diffusion barrier on the (1 1 0) surface [(1 0 0 surface)] and a 0.79 eV endothermicity to bulk dissolution, compared to a 1.02 eV [0.38 eV] barrier and 0.20 eV dissolution energy in pure Fe [37]. We therefore propose that a thin layer of Fe3Si may provide protection against H embrittlement of the underlying steel.
Co-reporter:Kristen A. Marino, Emily A. Carter
Intermetallics 2010 Volume 18(Issue 8) pp:1470-1479
Publication Date(Web):August 2010
DOI:10.1016/j.intermet.2010.03.044
First-principles density functional theory calculations are used to examine long-range Ni and short-range Al diffusion in Ni-rich NiAl. Since Ni diffusion in Ni-rich NiAl has been observed to have a lower activation energy than Ni diffusion in stoichiometric NiAl, another diffusion pathway, the antistructure bridge (ASB) mechanism, is believed to be activated once a threshold concentration of Ni antisite atoms has been reached. Our calculated activation energies for Ni diffusion via the ASB mechanism agree well with experimental data, confirming the ASB mechanism is likely operating in the Ni-rich region of the phase diagram, providing a means of long-range Ni diffusion. Since the ASB mechanism does not involve Al atoms, no long-range transport of Al occurs via this mechanism. Instead we looked at short-range Al transport via next-nearest-neighbor Al jumps into Al vacancies created during the ASB mechanism. The effect of Pt on these mechanisms was also examined; we predict that the presence of Pt enhances the rate of the ASB mechanism, which has implications for the lifetime of jet engine thermal barrier coatings.
Co-reporter:Nicholas J. Mosey, Emily A. Carter
Acta Materialia 2009 Volume 57(Issue 10) pp:2933-2943
Publication Date(Web):June 2009
DOI:10.1016/j.actamat.2009.03.001
Abstract
A method for predicting the shear strength of materials over multiple length scales is developed and tested. The method is based on renormalizing the energies and shear displacements obtained through electronic structure calculations of nanoscale models of the material of interest. All material- and size-dependent quantities are incorporated into the renormalization factors, yielding a universal model that can be applied to many materials and length scales. The model is used to predict the shear strength of Cr2O3 along three relevant slip planes and slip directions. The results demonstrate that the shear strengths of the nanoscale systems used in the calculations range from 19.4 to 29.4 GPa. These data are then renormalized to predict the shear strength of a grain that is 10 μm thick, yielding shear strengths ranging from 189 to 342 MPa. The large decrease in the shear strength with increasing grain size is consistent with the behavior of many materials. The ability to capture this change using electronic structure calculations that do not require experimental input may be useful in developing cohesive laws of novel materials for use in large-scale mechanical engineering simulations of materials failure.
Co-reporter:Linda Hung, Emily A. Carter
Chemical Physics Letters 2009 Volume 475(4–6) pp:163-170
Publication Date(Web):25 June 2009
DOI:10.1016/j.cplett.2009.04.059
We present a fully linear scaling (at most O(N · log(N))) and parallel algorithm for orbital-free density functional theory (OFDFT), for the first time exhibiting linear scaling in all terms (electronic and ionic). OFDFT solves directly for the electron density; consequently, the electron kinetic energy is determined using density functionals, which must be nonlocal to provide sufficient accuracy. The systematic elimination of bottlenecks within OFDFT renders the entire algorithm quasilinear scaling for all system sizes (no crossover point). Now an unprecedented number of atoms (∼1 million) can be treated explicitly quantum mechanically within OFDFT with a modest number of processors, opening up the door to treatment of ever more complex features in materials (precipitates, dislocations, etc.) without introducing empirical assumptions.Standard orbital-free density functional theory (OFDFT) scales quadratically with system size beyond a few thousand atoms due to ionic energy terms, but with cardinal B-spline approximations, OFDFT can be made to scale fully linearly for all system sizes. To the right is shown the total time required to optimize the electron density and to then calculate the total energy, forces, and stress tensor for fixed ion positions and cell shapes within OFDFT, for systems of bulk aluminum on a single processor.
Co-reporter:Donald F. Johnson, Emily A. Carter
Surface Science 2009 Volume 603(Issue 9) pp:1276-1283
Publication Date(Web):1 May 2009
DOI:10.1016/j.susc.2009.03.018
We use density functional theory to evaluate the stability of molybdenum disilicide coatings on a nickel substrate, as a possible bond coat alloy for high temperature coating applications. We consider the MoSi2(0 0 1)/Ni(1 1 1), MoSi2(1 0 0)/Ni(1 1 1), and MoSi2(1 1 0)/Ni(1 1 1) interfaces and predict quite strong (3.5–3.8 J/m2) adhesion of this metal–silicide ceramic to nickel. The origin of this strong adhesion is elucidated by examining the geometric and electronic structure of the interfaces. We predict that Mo and Si atoms at the interface primarily occupy Ni 3-fold hollow sites, the typical adsorption site on Ni(1 1 1). Projected local densities of states and electron density difference plots reveal a mixture of localized, covalent Si–Ni bonds and more delocalized metallic Mo–Ni bonding, as the origin of the strong interfacial bonding. As emphasized in our earlier work, creation of strong covalent bonds at interfaces results in very strong adhesion. Such strong adhesion makes MoSi2 a potential candidate for use in thermal barrier applications, in conjunction with a yttria-stabilized zirconia topcoat.
Co-reporter:Ivan Milas
Journal of Materials Science 2009 Volume 44( Issue 7) pp:
Publication Date(Web):2009 April
DOI:10.1007/s10853-008-3191-z
We investigate by means of periodic density functional theory the mechanism of grain boundary sliding along the α-alumina Σ11 tilt grain boundary. We identify minimum and maximum energy structures along a preferential sliding pathway for the pure grain boundary, as well as for grain boundaries doped with a series of early transition metals, as well as barium, gadolinium, and neodymium. We predict that the segregation of those dopants results in a considerable increase in the grain boundary sliding barrier. Grain boundary sliding occurs by a series of bond breaking and forming across the grain boundary. Our results suggest that the presence of large cations inhibits the regeneration of bonds during sliding, which results in a decrease in total number of bonds across the grain boundary interface, thereby raising the barrier to sliding. Trends in predicted grain boundary sliding energies are in good agreement with recently measured creep activation energies in polycrystalline alumina, lending further credence to the notion that grain boundary sliding plays a dominant role in alumina creep.
Co-reporter:Sahar Sharifzadeh, Patrick Huang, Emily A. Carter
Chemical Physics Letters 2009 470(4–6) pp: 347-352
Publication Date(Web):
DOI:10.1016/j.cplett.2009.01.072
Co-reporter:Donald F. Johnson and Emily A. Carter
The Journal of Physical Chemistry A 2009 Volume 113(Issue 16) pp:4367-4373
Publication Date(Web):March 13, 2009
DOI:10.1021/jp8110259
Ceramics such as SiC have the potential to act as protective coatings, primarily due to their high melting points and wear resistance. We use periodic density functional theory (DFT) within the generalized gradient approximation (GGA) to calculate the adhesion strength between SiC and Fe, for Si- and C-terminations of SiC(100) and two surfaces of Fe: (100) and (110). We predict a maximum ideal work of adhesion of 6.51 J/m2 at the SiC(100)/Fe(110) interface for C−Fe interfacial bonding, which is stronger than the traditional chrome coating’s adherence to Fe. We characterize the interfacial bonding via local densities of states and electron density difference analysis and find strong covalent bonding and some evidence of metallic bonding between Si (C) and Fe. Our results suggest that SiC might prove useful as a thin adhesion layer in a multilayer protective coating for steel.
Co-reporter:Kristen A. Marino
ChemPhysChem 2009 Volume 10( Issue 1) pp:226-235
Publication Date(Web):
DOI:10.1002/cphc.200800528
Abstract
Platinum is added to thermal barrier coatings (TBCs) as it is observed empirically to extend their lifetime, but the mechanism by which Pt acts is unknown. Since Pt has been proposed to alter diffusivities in NiAl, a key component of TBCs, we use first-principles quantum mechanics calculations to investigate atomic level diffusion mechanisms. Here, we examine the effect of Pt on five previously proposed mechanisms for Ni diffusion in NiAl: next-nearest-neighbor jumps, the triple defect mechanism, and three variants of the six jump cycle. We predict that Pt increases the rate of Ni diffusion by stabilizing point defects and defect clusters that are diffusion intermediates. Previously, we predicted the triple defect mechanism to be a dominant Ni diffusion mechanism; it simultaneously results in long-range Al diffusion in the opposite direction. Since Pt increases the rate of Ni diffusion, it also increases Al diffusion in NiAl, which may be key to extending the coating lifetime.
Co-reporter:Kristen A. Marino
ChemPhysChem 2009 Volume 10( Issue 14) pp:
Publication Date(Web):
DOI:10.1002/cphc.200990058
No abstract is available for this article.
Co-reporter:Patrick Huang and Emily A. Carter
Nano Letters 2008 Volume 8(Issue 4) pp:1265-1269
Publication Date(Web):March 22, 2008
DOI:10.1021/nl0804203
We report an ab initio study of the Kondo states formed from a Co adatom on Cu(111) and Cu(100). The model consists of a CoCun cluster (n = 5–19) embedded in (111) and (100) Cu slabs. An embedding potential derived from density functional theory treats the interaction between the periodic crystal surroundings and the CoCun cluster, while strong electron correlations within CoCun are explicitly accounted for via configuration interaction (CI) methods. Analysis of the embedded CI wave function provides insight into the nature of the Kondo state, specifically into the influence of the crystal host on the Co d-electronic structure. We predict that different d-orbitals are preferentially singly occupied in Co on Cu(111) versus Cu(100) as a result of the different crystalline environment. We propose that these variations in the local d-electronic structure on Co, not accounted for in previous theories, are responsible for the drastically different Kondo resonance line shapes observed in scanning tunneling microscopy experiments on these two surfaces.
Co-reporter:K.A. Marino, E.A. Carter
Acta Materialia 2008 Volume 56(Issue 14) pp:3502-3510
Publication Date(Web):August 2008
DOI:10.1016/j.actamat.2008.03.029
Abstract
First-principles density functional theory is used to examine the effect of Pt on point defects and defect clusters in NiAl. It is found that Pt promotes the formation of Ni and Al vacancies and Ni and Al antisite atoms. Defect clusters that are minima in postulated Ni diffusion mechanisms in NiAl are also found to be stabilized by the presence of Pt. By decreasing defect formation energies, Pt may decrease the overall activation barrier to the diffusion of Ni and Al in NiAl. The results provide clues as to how Pt enhances thermal barrier coating lifetime.
Co-reporter:Gregory S. Ho, Vincent L. Lignères, Emily A. Carter
Computer Physics Communications 2008 Volume 179(Issue 11) pp:839-854
Publication Date(Web):1 December 2008
DOI:10.1016/j.cpc.2008.07.002
We present PROFESS (PRinceton Orbital-Free Electronic Structure Software), a new software package that performs orbital-free density functional theory (OF-DFT) calculations. OF-DFT is a first principles quantum mechanics method primarily for condensed matter that can be made to scale linearly with system size. We describe the implementation of energy, force, and stress functionals and the methods used to optimize the electron density under periodic boundary conditions. All electronic energy and potential terms scale linearly while terms involving the ions exhibit quadratic scaling in our code. Despite the latter scaling, the program can treat tens of thousands of atoms with quantum mechanics on a single processor, as we demonstrate here. Limitations of the method are also outlined, the most serious of which is the accuracy of state-of-the-art kinetic energy functionals, which limits the applicability of the method to main group elements at present.Program summaryProgram title: PROFESSCatalogue identifier: AEBN_v1_0Program summary URL:http://cpc.cs.qub.ac.uk/summaries/AEBN_v1_0.htmlProgram obtainable from: CPC Program Library, Queen's University, Belfast, N. IrelandLicensing provisions: Standard CPC licence, http://cpc.cs.qub.ac.uk/licence/licence.htmlNo. of lines in distributed program, including test data, etc.: 35 933No. of bytes in distributed program, including test data, etc.: 329 924Distribution format: tar.gzProgramming language: Fortran 90Computer: Intel with ifort; AMD Opteron with pathf90Operating system: LinuxRAM: Problem dependent, but 2 GB is sufficient for up to 10,000 ionsClassification: 7.3External routines: FFTW (http://www.fftw.org), MINPACK-2Nature of problem: Given a set of coordinates describing the initial ion positions under periodic boundary conditions, recovers the ground state energy, electron density, ion positions, and cell lattice vectors predicted by orbital-free density functional theory. Except for computation of the ion–ion and ion–electron terms, all other terms are effectively linear scaling. Up to ∼10,000 ions may be included in the calculation on just a single processor.Solution method: Computes energies as described in text; minimizes this energy with respect to the electron density, ion positions, and cell lattice vectors.Restrictions: PROFESS cannot use nonlocal (such as ultrasoft) pseudopotentials. Local pseudopotential files for aluminum, magnesium, silver, and silicon are available upon request. Also, due to the current state of the kinetic energy functionals, PROFESS is only reliable for main group metals and some properties of semiconductors.Running time: Problem dependent: the test example provided with the code takes less than a second to run. Timing results for large scale problems are given in the paper.References:[1] Y.A. Wang, N. Govind, E.A. Carter, Phys. Rev. B 58 (1998) 13465; Y.A. Wang, N. Govind, E.A. Carter, Phys. Rev. B 64 (2001) 129901 (erratum).[2] S.C. Watson, E.A. Carter, Comput. Phys. Comm. 128 (2000) 67.
Co-reporter:Chen Huang and Emily A. Carter
Physical Chemistry Chemical Physics 2008 vol. 10(Issue 47) pp:7109-7120
Publication Date(Web):20 Oct 2008
DOI:10.1039/B810407G
One obstacle in orbital-free density functional theory (OF-DFT) is the lack of accurate and transferable local pseudopotentials (LPSs). In this work, we build high quality LPSs by inverting Kohn–Sham (KS) equations on bulk valence electron densities to obtain an atom-centered local pseudopotential. With this approach, we build LPSs for Mg, Al, and Si, and then test them in KS DFT calculations of static bulk properties for several Mg, Al, and Si bulk structures as well as β″-Al3Mg. Our Mg, Al, and Si LPSs produce correct ground state properties and phase orderings. These LPSs are then tested in KS-DFT calculations of surface energies for several low-index Mg and Al surfaces, point defect properties in hexagonal-close-packed (hcp) Mg, face-centered cubic (fcc) Al, and diamond Si, and stacking fault energies in fcc Al. All of these LPS results agree quantitatively with the results from nonlocal pseudopotentials with errors less than or equal to 40 meV per atom. Finally, we perform OF-DFT calculations for various Mg and Al structures, employing the Wang–Govind–Carter (WGC) nonlocal kinetic energy density functional (KEDF). The OF-DFT results generally agree well with the corresponding KS-DFT results. With our new Mg and Al LPSs and the WGC KEDF, OF-DFT now provides a practical method for accurate, large-scale first principles simulations of main group metals and their alloys.
Co-reporter:Karin M. Carling, Emily A. Carter
Acta Materialia 2007 Volume 55(Issue 8) pp:2791-2803
Publication Date(Web):May 2007
DOI:10.1016/j.actamat.2006.12.020
Abstract
We investigate with first-principles density functional theory (DFT) the adhesion of the Al2O3(0 0 0 1)/NiAl(1 1 0) interface as a model for the thermally grown oxide/bond coat alloy interface in thermal barrier coatings. We find that the clean interface has an ideal work of adhesion of 0.66 J m−2. We predict that S impurities reduce interfacial adhesion significantly, due to a reduction in cross-interface bonds. The presence of Pt alters the interface adhesion only slightly, while Hf dopants dramatically increase adhesion via formation of strong Hf–O bonds, as expected from Hf’s open-shell character. We discuss the implications of these predictions, which are consistent with experimental observations of the effects of S, Pt, and Hf on the lifetime of thermal barrier coatings.
Co-reporter:Donald F. Johnson, D.E. Jiang, Emily A. Carter
Surface Science 2007 Volume 601(Issue 3) pp:699-705
Publication Date(Web):1 February 2007
DOI:10.1016/j.susc.2006.10.034
Properties of the Cr(1 0 0)/Fe(1 0 0) and Cr(1 1 0)/Fe(1 1 0) interfaces are investigated with spin-polarized density functional theory within the generalized gradient approximation (DFT-GGA) for electron exchange and correlation. Contrary to earlier predictions for a monolayer of Cr on bulk Fe, we find intermixing of Cr and Fe at the interface of thick films to be endothermic; hence here we focus on characterizing abrupt, unalloyed interfaces. The ideal work of adhesion for both the (1 0 0) and (1 1 0) abrupt interfaces is predicted to be ∼5.4 J/m2. We propose that this anomalously strong adhesion between heterogeneous interfaces is derived from significant spin correlations and d–d bonding at the interface.
Co-reporter:Keith Niedfeldt, Peter Nordlander, Emily A. Carter
Surface Science 2007 Volume 601(Issue 5) pp:L29-L33
Publication Date(Web):1 March 2007
DOI:10.1016/j.susc.2006.12.085
We investigate the broadening of the 2s energy level of a Li atom outside a Si(0 0 1) surface using a first principles approach. The covalent nature of the Si surface produces large variations in Li energy level widths as a function of lateral position across the surface. The widths above symmetric Si dimers are predicted to be much larger than above buckled Si dimers, suggesting that charge transfer will occur primarily above symmetric dimers. We discuss the ramifications of our results on the controversy surrounding the relative abundance of the buckled vs. symmetric dimers on the Si surface.
Co-reporter:Gregory S. Ho, Chen Huang, Emily A. Carter
Current Opinion in Solid State and Materials Science 2007 Volume 11(5–6) pp:57-61
Publication Date(Web):October–December 2007
DOI:10.1016/j.cossms.2008.06.005
Orbital-free density functional theory (OF-DFT) can be made to scale linearly with sample size, allowing thousands of atoms to be treated explicitly with quantum mechanics. State-of-the-art kinetic energy density functionals and ion–electron pseudopotentials are used to obtain accurate structural property predictions for nanoparticles, nanowires, extended surfaces, and nanoindentation of simple metals.
Co-reporter:Keith Niedfeldt, Emily A. Carter, Peter Nordlander
Surface Science 2006 Volume 600(Issue 21) pp:L291-L295
Publication Date(Web):1 November 2006
DOI:10.1016/j.susc.2006.08.005
Using a first principles electronic structure method, we investigate how a band gap in the surface projected density of states of a noble metal influences the resonant tunneling lifetime of an atomic level near the surface. Comparing Li atoms outside Cu(1 1 1), where the Li 2s level is within the gap, with Cu(0 0 1), where the Li 2s level is below the gap, reveals that the lifetime of the ionization level of Li is longer on Cu(1 1 1). This difference is caused both by the decreased final state phase space for the tunneling electron on Cu(1 1 1) and by a reduction in intra-atomic hybridization in Li outside a surface whose band gap encompasses the ionization level. Our results are consistent with previous experiments and theory that predict a similar trend for the ionization level of Cs above Cu(1 1 1) and Cu(0 0 1).
Co-reporter:Karin M. Carling, Will Glover, Hakan Gunaydin, Tracy A. Mitchell, Emily A. Carter
Surface Science 2006 Volume 600(Issue 10) pp:2079-2090
Publication Date(Web):15 May 2006
DOI:10.1016/j.susc.2006.02.047
First-principles periodic slab density-functional theory (DFT) calculations with a plane-wave basis are used to predict the properties of S, Pt, and Hf adsorption on NiAl(1 1 0). Stable adsorption sites are identified, and adsorbate binding energies and structures are predicted. We find that while S adsorbs in a threefold site, the metals prefer to adsorb in the Ni–Ni twofold bridge site. The latter finding is consistent with scanning tunneling microscopy experiments for adsorption of various transition metals on NiAl(1 1 0) by Ho and coworkers. S is predicted to easily diffuse between threefold sites. We find that Pt and Hf both induce significant changes in the local surface structure, changing twofold bridge sites into fourfold coordination sites by drawing next-nearest-neighbor atoms nearly equidistant with the nearest-neighbor atoms. We find Pt favors interaction with Al slightly more than Ni, while Hf shows a particularly strong affinity for Ni compared to Al. We also predict that Hf may diffuse one-dimensionally along Ni rows with a barrier of ≈0.6 eV.
Co-reporter:D.E. Jiang, Emily A. Carter
Acta Materialia 2005 Volume 53(Issue 17) pp:4489-4496
Publication Date(Web):October 2005
DOI:10.1016/j.actamat.2005.06.001
Abstract
We report a spin-polarized periodic density functional theory investigation of the atomic structure, bonding, and ideal work of adhesion of the MoSi2/Fe interface, in order to explore the potential of MoSi2 as a protective coating for steel. We find that MoSi2 strongly adheres to Fe, with an ideal work of adhesion of ∼3.85 J/m2 for two low-index, low-strain interfaces. This value will be a lower bound to measured adhesion energies, since the latter will be larger due to plasticity. This ideal adhesion energy for a ceramic coating to Fe is much stronger than predicted previously for ceramic coating materials such as ZrC and TiC. We attribute this stronger adhesion to increased covalent interfacial bonding for MoSi2/Fe compared to metal carbide/Fe interfaces (where metallic bonding plays a larger role), as evidenced by the rearrangement of electron density and the character of the local density of states upon formation of the interface.
Co-reporter:D.E. Jiang, Emily A. Carter
Surface Science 2005 Volume 583(Issue 1) pp:60-68
Publication Date(Web):20 May 2005
DOI:10.1016/j.susc.2005.03.023
We report first principles density functional theory (DFT) results of H2S and HS adsorption and dissociation on the Fe(1 1 0) surface. We investigate the site preference of H2S, HS, and S on Fe(1 1 0). H2S is found to weakly adsorb on either the short bridge (SB) or long bridge (LB) site of Fe(1 1 0), with a binding energy of no more than 0.50 eV. The diffusion barrier from the LB site to the SB site is found to be small (∼0.10 eV). By contrast to H2S, HS is predicted to be strongly chemisorbed on Fe(1 1 0), with the S atom in the LB site and the HS bond oriented perpendicular to the surface. Isolated S atoms also are predicted to bind strongly to the LB sites of Fe(1 1 0), where the SB is found to be a transition state for S surface hopping between neighboring LB sites. The minimum energy paths for H2S and HS dehydrogenation involve rotating an H atom towards a nearby surface Fe atom, with the S–H bonds breaking on the top of one Fe atom. The barrier to break the first S–H bond in H2S is low at 0.10 eV, and breaking the second S–H bond is barrierless, suggesting deposition of S on Fe(1 1 0) via H2S is kinetically and thermodynamically facile.
Co-reporter:Kyle J. Caspersen
PNAS 2005 102 (19 ) pp:6738-6743
Publication Date(Web):2005-05-10
DOI:10.1073/pnas.0408127102
We present a method to identify transition states and minimum energy paths for martensitic solid–solid phase transformations,
thereby allowing quantification of the activation energies of such transformations. Our approach is a generalization of a
previous method for identifying transition states for chemical reactions, namely the climbing image-nudged elastic band algorithm,
where here the global deformation of the crystalline lattice (volume and shape fluctuations) becomes the reaction coordinate
instead of atomic motion. We also introduce an analogue to the Born–Oppenheimer approximation that allows a decoupling of
nuclear motion and lattice deformation, where the nuclear positions along the path are determined variationally according
to current deformation state. We then apply this technique to characterize the energetics of elemental lithium phase transformations
as a function of applied pressure, where we see a validation of the Born–Oppenheimer-like approximation, small energy barriers
(expected for martensitic transformations), and a pronounced pressure dependence of various properties characterizing the
phase transitions.
Co-reporter:Ilgyou Shin, Emily A. Carter
International Journal of Plasticity (September 2014) Volume 60() pp:58-70
Publication Date(Web):1 September 2014
DOI:10.1016/j.ijplas.2014.04.002
•We investigate dislocation mobility in HCP Mg using the linear scaling OFDFT method.•Mg Peierls stresses are extracted for the first time from quantum simulations.•OFDFT Peierls stresses are in excellent agreement with experiments.•Anisotropic plasticity between prismatic vs. basal slip is well-reproduced.•This anisotropy is strongly correlated with the number of atoms moving collectively.The strength and ductility of metals are governed by the motion of dislocations, which is quantified by the Peierls stress (σp). We use orbital-free density functional theory (OFDFT) to characterize the motion of 13〈112¯0〉 dislocations on the basal {0001} and prismatic {11¯00} planes in hexagonal-close-packed magnesium (Mg) in order to understand its deformation mechanisms. We predict σp values of edge dislocations on the basal and prismatic planes to be 0.6 and 35.4 MPa, respectively. The presence of stable stacking faults only on the basal plane produces partial dislocation splitting, which significantly lowers σp for basal dislocations. Our atomic scale simulations reveal that dislocation mobility is strongly correlated with the number of core atoms moving collectively. OFDFT σp results are in excellent agreement with experiments (∼0.5 and 39.2 MPa), further validating OFDFT as an independent and predictive tool for simulating plastic behavior in main group metals at the mesoscale with first principles’ accuracy.
Co-reporter:Nicholas J. Mosey, Emily A. Carter
Journal of the Mechanics and Physics of Solids (February 2009) Volume 57(Issue 2) pp:287-304
Publication Date(Web):1 February 2009
DOI:10.1016/j.jmps.2008.10.009
Periodic density functional theory (DFT) and DFT+U calculations are used to evaluate various mechanical properties associated with the fracture of chromia (Cr2O3) along the [0 0 0 1] and [011¯(3/2)(a/c)22] directions. The properties investigated include the tensile strength, elastic constants, and surface energies. The tensile strengths are evaluated using an ideal tensile test, which provides the theoretical tensile strength, and by fitting the calculated data to universal binding energy relationships (UBER), which permit the extrapolation of the calculated results to arbitrary length scales. The results demonstrate the ability of the UBER to yield a realistic estimate of the tensile strength of a 10-μm-thick sample of Cr2O3 using data obtained through calculations on nanoscopic systems. We predict that Cr2O3 will fracture most easily in the [011¯(3/2)(a/c)22] direction, with a best estimate for the tensile strength of 386 MPa for a 10 μm grain, consistent with flexural strength measurements for chromia. The grain becomes considerably stronger at the nanoscale, where we predict a tensile strength along the same direction of 32.1 GPa for 1.45 nm crystallite. The results also provide insight into the origin of the direction dependence of the mechanical properties of Cr2O3, with the differences in the behavior along different directions being related to the number of Cr–O bonds supporting the applied tensile load. Additionally, the results shed light on various practical aspects of modeling the mechanical properties of materials with DFT+U calculations and in using UBERs to estimate the mechanical properties of materials across disparate length scales.
Co-reporter:Maytal Caspary Toroker, Dalal K. Kanan, Nima Alidoust, Leah Y. Isseroff, Peilin Liao and Emily A. Carter
Physical Chemistry Chemical Physics 2011 - vol. 13(Issue 37) pp:NaN16654-16654
Publication Date(Web):2011/08/19
DOI:10.1039/C1CP22128K
The positions of electronic band edges are one important metric for determining a material's capability to function in a solar energy conversion device that produces fuels from sunlight. In particular, the position of the valence band maximum (conduction band minimum) must lie lower (higher) in energy than the oxidation (reduction) reaction free energy in order for these reactions to be thermodynamically favorable. We present first principles quantum mechanics calculations of the band edge positions in five transition metal oxides and discuss the feasibility of using these materials in photoelectrochemical cells that produce fuels, including hydrogen, methane, methanol, and formic acid. The band gap center is determined within the framework of DFT+U theory. The valence band maximum (conduction band minimum) is found by subtracting (adding) half of the quasiparticle gap obtained from a non-self-consistent GW calculation. The calculations are validated against experimental data where possible; results for several materials including manganese(II) oxide, iron(II) oxide, iron(III) oxide, copper(I) oxide and nickel(II) oxide are presented.
Co-reporter:Ting Tan, Xueliang Yang, Yiguang Ju and Emily A. Carter
Physical Chemistry Chemical Physics 2015 - vol. 17(Issue 46) pp:NaN31072-31072
Publication Date(Web):2015/10/28
DOI:10.1039/C5CP06004D
The unimolecular dissociation and isomerization kinetics of the three methyl propanoate (MP) radicals, CH3CH2C(O)OĊH2 (MP-m), CH3ĊHC(O)OCH3 (MP-α), and ĊH2CH2C(O)OCH3 (MP-β), are theoretically investigated using high-level ab initio methods and the Rice–Ramsperger–Kassel–Marcus (RRKM)/master equation (ME) theory. Stationary-point energies are obtained using the coupled cluster singles and doubles with perturbative triples correction (CCSD(T)), multi-reference singles and doubles configuration interaction (MRSDCI) with the Davidson-Silver (DS) correction, and multi-reference averaged coupled pair functional (MRACPF2) theories. The isomerization barriers between the three radicals are predicted to be generally lower than the corresponding bond dissociation channels, leading to a strongly coupled reaction system in subsequent kinetics studies. The phenomenological temperature- and pressure-dependent rate coefficients are computed using the RRKM/ME theory over a temperature range of 500 to 2000 K and at a pressure range of 0.01 atm to the high-pressure limit, which are then fitted to modified Arrhenius expressions. The β-scission rate coefficients of MP-α to CH3CHC(O) and CH3Ȯ are predicted to be the smallest because of its highest activation energy among all studied unimolecular reactions channels. Analysis of branching fractions shows that both MP-m and MP-α radicals mainly decompose to the bimolecular products CH3CH2Ċ(O) and H2CO, whereas the MP-β radical primarily decomposes via cleavage of a C–C bond to form C2H4 and CH3OĊ(O). The isomerization channels dominate at low temperatures, the branching fractions of which decrease with increasing temperature and become very minor at about 2000 K. Our accurate rate coefficients and branching fractions help to illuminate the unique combustion properties of MP.
Co-reporter:Nima Alidoust and Emily A. Carter
Physical Chemistry Chemical Physics 2015 - vol. 17(Issue 27) pp:NaN18110-18110
Publication Date(Web):2015/06/18
DOI:10.1039/C5CP03429A
Alloying nickel oxide (NiO) with lithium oxide (Li2O) at high Li concentrations may reduce NiO's band gap and expand its use as a light absorber in photocatalytic and tandem dye-sensitized solar cell technologies. In this work, we evaluate the viability of this alloy as a p-type hole transport material. We use embedded cluster models, along with unrestricted Hartree–Fock and complete active space self-consistent field theories, to study the impact of alloying on polaronic transport of holes. Our calculated energy barrier for hole transfer in undoped NiO is in excellent agreement with the experimental value of ∼0.1 eV. We predict that hole transport in NiO is anisotropic and mostly confined parallel to the (111) ferromagnetic planes. Applying the same model to Li-doped NiO indicates that isolated Li ions do not introduce free holes into NiO samples. However, free holes can be created in the homogeneous Li0.125Ni0.875O alloy, in which the Li concentration is very high. Our kinetic Monte Carlo calculations show that hole mobility in this alloy is lower than in undoped NiO. However, the additional free holes and the predicted lower band gap of Li0.125Ni0.875O should increase hole conductivity compared to NiO upon alloy formation. Therefore, Li0.125Ni0.875O alloys have potential for use as a hole transporter, as well as a sunlight absorber, in a variety of solar energy applications.
Co-reporter:Andrew M. Ritzmann, Johannes M. Dieterich and Emily A. Carter
Physical Chemistry Chemical Physics 2016 - vol. 18(Issue 17) pp:NaN12269-12269
Publication Date(Web):2016/04/06
DOI:10.1039/C6CP01720G
Reducing operating temperatures is a key step in making solid oxide fuel cell (SOFC) technology viable. A promising strategy for accomplishing this goal is employing mixed ion–electron conducting (MIEC) cathodes. La1−xSrxCo1−yFeyO3−δ (LSCF) is the most widely employed MIEC cathode material; however, rational optimization of the composition of LSCF requires fundamental insight linking its electronic structure to its defect chemistry. To provide the necessary insight, density functional theory plus U (DFT+U) calculations are used to investigate the electronic structure of LSCF (xSr = 0.50, yCo = 0.25). The DFT+U calculations show that LSCF has a significantly different electronic structure than La1−xSrxFeO3 because of the addition of cobalt, but that minimal electronic structure differences exist between La0.5Sr0.5Co0.25Fe0.75O3 and La0.5Sr0.5Co0.5Fe0.5O3. The oxygen vacancy formation energy (ΔEf,vac) is calculated for residing in different local environments within La0.5Sr0.5Co0.25Fe0.75O3. These results show that configurations have the highest ΔEf,vac, while have the lowest ΔEf,vac and may act as traps for . We conclude that compositions with more Fe than Co are preferred because the additional sites would lead to higher overall ΔEf,vac (and lower concentrations), while the trapping strength of the sites is relatively weak (∼0.3 eV).
Co-reporter:Peilin Liao and Emily A. Carter
Journal of Materials Chemistry A 2010 - vol. 20(Issue 32) pp:NaN6719-6719
Publication Date(Web):2010/07/05
DOI:10.1039/C0JM01199A
Stress-corrosion cracking of steel is a ubiquitous phenomenon in which steel is chemically corroded, followed by fracture induced by applied loads. The products of steel corrosion in moist air at high temperature (>570 °C) are α-Fe2O3, FeO, and Fe3O4. Here we employ an ab initio density functional theory + U method to predict the tensile properties of these oxides, to gain insight into failure mechanisms. The universal binding energy relationship of Hayes et al. is employed to extrapolate atomic scale data to macroscopic sample sizes. The extrapolated macroscopic predictions are consistent with experimental measurements. The ordering of tensile strengths is shown to be FeO < Fe3O4 < α-Fe2O3, which correlates with increasing ionicity (as given by the formal charge on the Fe cations) in the three oxides. The direction that has weakest tensile strength is predicted to be along [01(3/2)(a/c)22] for α-Fe2O3 and along [001] for both FeO and Fe3O4. The direction dependence of tensile properties for these three iron oxides can be understood via a local bond strain analysis. We also predict that loading FeO along [110] and Fe3O4 along [001] or [111] produces a plastic response prior to brittle fracture at high temperature.
Co-reporter:John A. Keith and Emily A. Carter
Chemical Science (2010-Present) 2013 - vol. 4(Issue 4) pp:NaN1496-1496
Publication Date(Web):2013/02/14
DOI:10.1039/C3SC22296A
One of the most promising CO2 reduction processes presently known suffers from a lack of fundamental understanding of its reaction mechanism. Using first principles quantum chemistry, we report thermodynamical energies of various pyridine-derived intermediates as well as barrier heights for key homogeneous reaction mechanisms. From this work, we predict that the actual form of the co-catalyst involved in pyridinium-based CO2 reduction is not the long-proposed pyridinyl radical in solution, but is more probably a surface-bound dihydropyridine species.
Co-reporter:Ivan Milas, Berit Hinnemann and Emily A. Carter
Journal of Materials Chemistry A 2011 - vol. 21(Issue 5) pp:NaN1456-1456
Publication Date(Web):2010/10/26
DOI:10.1039/C0JM02212H
We use density functional theory with periodic boundary conditions to investigate diffusion of Al, O, Pt, Hf, and Y on the α-Al2O3(0001) surface as a simple model to obtain insight into possible diffusion mechanisms occurring at alumina grain boundaries (GBs) in thermal barrier coatings (TBCs). We calculate diffusion pathways, activation energies, and diffusion constants. For the elements involved in alumina growth, we find that Al diffusion is facile and involves a simple hop from one most stable adsorption site to the next without local minima in between whereas O diffusion has a much higher barrier and has local minima along the minimum energy pathway. This trend is consistent with relative rates of diffusion at alumina GBs. Regarding common dopants in the metal (bond coat) alloy substrate from which alumina grows, we predict that Pt diffusion is facile but thermodynamically disfavored relative to Pt remaining in the bond coat alloy, while Hf and Y diffusion involve much higher barriers than Al diffusion. We use these results to rationalize some observations regarding the role of reactive elements Hf and Y as well as Pt in TBCs.
Co-reporter:Ting Tan, Xueliang Yang, Yiguang Ju and Emily A. Carter
Physical Chemistry Chemical Physics 2016 - vol. 18(Issue 6) pp:NaN4607-4607
Publication Date(Web):2016/01/14
DOI:10.1039/C5CP07282D
The kinetics of hydrogen abstraction by five radicals (H, CH3, O(3P), OH, and HO2) from a biodiesel surrogate, methyl propanoate (MP), is theoretically investigated. We employ high-level ab initio quantum chemistry methods, coupled-cluster singles and doubles with perturbative triples correction (CCSD(T)) and multi-reference singles and doubles configuration interaction (MRSDCI) with the Davidson-Silver (DS) correction, and obtain chemically accurate reaction energetics. Overall, MRSDCI + DS predicts comparable energetics to CCSD(T) for MP + H/CH3/O/OH. The rate constants are computed using transition state theory (TST-Rice–Ramsperger–Kassel–Marcus theory) in conjunction with the separable-hindered-rotor approximation, variable reaction coordinate TST, and the multi-structure all-structure (MS-AS) approach. A simplified method, semi-multi-structure, is also employed for MP + OH/HO2, and the rate coefficients with this less expensive method are in good agreement with the results obtained with the MS-AS method. The fitted modified Arrhenius expressions are provided over a temperature range of 250 to 2000 K. The predicted rate coefficients for MP + OH agree remarkably well with experimental data over a wide temperature range. Branching ratio analysis of all the studied reactions shows that abstractions of the secondary H atoms within MP are expected to dominate the consumption of fuel at low temperatures, and the contributions of abstractions from the two methyl groups increase with temperature for all abstracting radicals.
Co-reporter:Martina Lessio, Christoph Riplinger and Emily A. Carter
Physical Chemistry Chemical Physics 2016 - vol. 18(Issue 38) pp:NaN26443-26443
Publication Date(Web):2016/09/09
DOI:10.1039/C6CP04272D
Adsorbed protons that develop hydride character have been proposed to play a role in the mechanism of CO2 reduction catalyzed by pyridine on GaP photoelectrodes. Investigating their stability represents an important step towards vetting this mechanism. In this contribution, the relative stability of the adsorbed protons is determined using cluster models with dispersion-corrected density functional theory and continuum solvation. Proton acidity constants computed under typical experimental conditions are compared to the acidity constants of other relevant species. The adsorbed protons are predicted to be very stable, suggesting that they will be present on the surface and available to be reduced to surface hydrides that could possibly react with adsorbed pyridine to form adsorbed dihydropyridine, a previously proposed co-catalyst. However, the high stability of such protons also suggests that the surface does not represent a significant proton source; as a consequence, protons required in the proposed mechanism must be provided by a different source such as the acidified aqueous solution in contact with the electrode surface.
Co-reporter:Chen Huang and Emily A. Carter
Physical Chemistry Chemical Physics 2008 - vol. 10(Issue 47) pp:NaN7120-7120
Publication Date(Web):2008/10/20
DOI:10.1039/B810407G
One obstacle in orbital-free density functional theory (OF-DFT) is the lack of accurate and transferable local pseudopotentials (LPSs). In this work, we build high quality LPSs by inverting Kohn–Sham (KS) equations on bulk valence electron densities to obtain an atom-centered local pseudopotential. With this approach, we build LPSs for Mg, Al, and Si, and then test them in KS DFT calculations of static bulk properties for several Mg, Al, and Si bulk structures as well as β″-Al3Mg. Our Mg, Al, and Si LPSs produce correct ground state properties and phase orderings. These LPSs are then tested in KS-DFT calculations of surface energies for several low-index Mg and Al surfaces, point defect properties in hexagonal-close-packed (hcp) Mg, face-centered cubic (fcc) Al, and diamond Si, and stacking fault energies in fcc Al. All of these LPS results agree quantitatively with the results from nonlocal pseudopotentials with errors less than or equal to 40 meV per atom. Finally, we perform OF-DFT calculations for various Mg and Al structures, employing the Wang–Govind–Carter (WGC) nonlocal kinetic energy density functional (KEDF). The OF-DFT results generally agree well with the corresponding KS-DFT results. With our new Mg and Al LPSs and the WGC KEDF, OF-DFT now provides a practical method for accurate, large-scale first principles simulations of main group metals and their alloys.
Co-reporter:Ana B. Muñoz-García, Michele Pavone, Andrew M. Ritzmann and Emily A. Carter
Physical Chemistry Chemical Physics 2013 - vol. 15(Issue 17) pp:NaN6259-6259
Publication Date(Web):2013/03/07
DOI:10.1039/C3CP50995H
We use ab initio density functional theory + U calculations to characterize the oxide ion diffusion process in bulk Sr2Fe1.5Mo0.5O6−δ (SFMO) by analyzing the formation and migration of oxygen vacancies. We show that SFMO's remarkable ionic conductivity arises from its intrinsic content of oxygen vacancies and a predicted very low migration barrier of such vacancies. Theoretical analysis of the electronic structure reveals a crucial role played by strongly hybridized Fe 3d/O 2p states to achieve the attendant mixed ion-electron conductor character so important for intermediate temperature fuel cell operation. We predict a next-nearest-neighbor-type migration pathway for the O2− ion should dominate. The low energy barrier of this pathway is mainly related to electrostatic interactions with homogeneously distributed Mo in the SFMO sublattice. We identify the reasons why Fe-rich perovskites, with the key addition of a certain concentration of Mo, produce excellent electronic and ionic transport properties so crucial for efficient operation of intermediate temperature solid oxide fuel cells.
Co-reporter:Maytal Caspary Toroker and Emily A. Carter
Journal of Materials Chemistry A 2013 - vol. 1(Issue 7) pp:NaN2484-2484
Publication Date(Web):2012/12/17
DOI:10.1039/C2TA00816E
First-row transition metal oxides (TMOs) are inexpensive potential alternative materials for solar energy conversion devices. However, some TMOs, such as manganese(II) oxide, have band gaps that are too large for efficiently absorbing solar energy. Other TMOs, such as iron(II) oxide, have conduction and valence band edges with the same orbital character that may lead to unfavorably high electron–hole recombination rates. Another limitation of iron(II) oxide is that the calculated valence band edge is not positioned well for oxidizing water. We predict that key properties, including band gaps, band edge positions, and possibly electron–hole recombination rates, may be improved by alloying TMOs that have different band alignments. A new metric, the band gap center offset, is introduced for simple screening of potential parent materials. The concept is illustrated by calculating the electronic structure of binary oxide alloys that contain manganese, nickel, iron, zinc, and/or magnesium, within density functional theory (DFT)+U and hybrid DFT theories. We conclude that alloys of iron(II) oxide are worth evaluating further as solar energy conversion materials.
Co-reporter:Dalal K. Kanan and Emily A. Carter
Journal of Materials Chemistry A 2013 - vol. 1(Issue 32) pp:NaN9256-9256
Publication Date(Web):2013/06/07
DOI:10.1039/C3TA11265A
Rationally engineering photocatalytic devices that power water splitting or CO2 reduction reactions requires identifying economical materials that efficiently absorb sunlight and have suitable band edge placements. Recent theoretical investigations have predicted that a 1:1 alloy of MnO and ZnO meets these criteria. However, poor hole conductivity in undoped MnO:ZnO alloys (with up to 10% ZnO) severely limits this material's utility in electronic devices, and its electron conductivity has not yet been characterized. Here we investigate carrier transport in pure and doped MnO and MnO:ZnO with ab initio quantum chemistry calculations. Electrostatically embedded clusters are used to compute and compare relative electron/hole transfer barriers within the small polaron model. We assess the effects of Al, Ga, In, Sc, Y, Ti, Sb, Gd, F (n-type dopants) and Li (a p-type dopant) to determine which may enhance conductivity in MnO:ZnO. Our findings indicate that Ga, Sc, Ti, F, and Sb dopants create deep traps whereas In forms shallower traps that merit further investigation. Y, Al, Gd, and Li dopants should increase the carrier concentration while maintaining favorable electron and hole transport pathways. The latter are recommended for increasing the conductivity of MnO:ZnO and its effectiveness for solar energy conversion.
Co-reporter:Andrew M. Ritzmann, Michele Pavone, Ana B. Muñoz-García, John A. Keith and Emily A. Carter
Journal of Materials Chemistry A 2014 - vol. 2(Issue 21) pp:NaN8074-8074
Publication Date(Web):2014/04/08
DOI:10.1039/C4TA00801D
Although solid oxide fuel cells (SOFCs) provide clean and efficient electricity generation, high operating temperatures (T > 800 °C) limit their widespread use. Lowering operating temperatures (600 °C < T < 800 °C) requires developing next-generation mixed ion-electron conducting (MIEC) cathodes that permit facile oxygen transport. One promising MIEC material, La1−xSrxCo1−yFeyO3 (LSCF), can operate at intermediate temperatures, has a longer cell lifetime, and permits less expensive interconnect materials. However, the road to optimization of LSCF compositions for SOFC applications would benefit from fundamental, atomic-scale insight into how local chemical changes affect its oxygen ion conductivity. We provide this insight using ab initio density functional theory plus U (DFT+U) calculations to analyze the factors governing oxygen transport in the LSCF parent material LaCoO3. We show that oxygen diffusion in LaCoO3 depends strongly on the spin state of the Co3+ ions: in particular, low spin Co3+ promotes higher oxygen vacancy concentrations than other spin states. We also predict that different spin states of Co3+ significantly affect the oxygen ion migration barrier. Through electronic structure analysis, we uncover the fundamental details which govern oxygen diffusivity in LaCoO3.
Co-reporter:Peilin Liao and Emily A. Carter
Chemical Society Reviews 2013 - vol. 42(Issue 6) pp:NaN2422-2422
Publication Date(Web):2012/10/30
DOI:10.1039/C2CS35267B
Photocatalytic production of transportation fuels should be among our long term strategies to achieve energy and environmental sustainability for the planet, but the technology is hampered by a lack of sufficiently efficient catalysts. Although efficiency is ultimately determined by laboratory measurements, theory and computation have become powerful tools for examining underlying mechanisms and guiding avenues of inquiry. In this review, we focus on first principles calculations of transition metal oxide semiconductor photocatalysts. We discuss how theory can be applied to investigate various aspects of a photocatalytic cycle: light absorption, electron/hole transport, band edge alignments of semiconductors, and surface chemistry. Emphasis is placed on identifying accurate models for specific properties and theoretical insights into improving photocatalytic performance.
