More Like this

1. Combining localized orbital scaling correction and Bethe–Salpeter equation for accurate excitation energies

   https://doi.org/10.1063/5.0087498  

   Li, Jiachen ; Jin, Ye ; Su, Neil Qiang ; Yang, Weitao ( April 2022 , The Journal of Chemical Physics)

   We applied localized orbital scaling correction (LOSC) in Bethe–Salpeter equation (BSE) to predict accurate excitation energies for molecules. LOSC systematically eliminates the delocalization error in the density functional approximation and is capable of approximating quasiparticle (QP) energies with accuracy similar to or better than GW Green’s function approach and with much less computational cost. The QP energies from LOSC, instead of commonly used G 0 W 0 and ev GW, are directly used in BSE. We show that the BSE/LOSC approach greatly outperforms the commonly used BSE/ G 0 W 0 approach for predicting excitations with different characters. For the calculations of Truhlar–Gagliardi test set containing valence, charge transfer, and Rydberg excitations, BSE/LOSC with the Tamm–Dancoff approximation provides a comparable accuracy to time-dependent density functional theory (TDDFT) and BSE/ev GW. For the calculations of Stein CT test set and Rydberg excitations of atoms, BSE/LOSC considerably outperforms both BSE/ G 0 W 0 and TDDFT approaches with a reduced starting point dependence. BSE/LOSC is, thus, a promising and efficient approach to calculate excitation energies for molecular systems.
2. Guanine–adenine interactions in DNA tetranucleotide cation radicals revealed by UV/vis photodissociation action spectroscopy and theory

   https://doi.org/10.1039/d0cp02362k  

   Liu, Yue ; Huang, Shu R. ; Tureček, František ( July 2020 , Physical Chemistry Chemical Physics)

   Hydrogen-rich cation radicals (GATT + 2H) + ˙ and (AGTT + 2H) + ˙ represent oligonucleotide models of charged hydrogen atom adducts to DNA. These tetranucleotide cation radicals were generated in the gas phase by one-electron reduction of the respective (GATT + 2H) 2+ and (AGTT + 2H) 2+ dications in which the charging protons were placed on the guanine and adenine nucleobases. We used wavelength-dependent UV/Vis photodissociation in the valence-electron excitation region of 210–700 nm to produce action spectra of (GATT + 2H) + ˙ and (AGTT + 2H) + ˙ that showed radical-associated absorption bands in the near-UV (330 nm) and visible (400–440 nm) regions. Born–Oppenheimer molecular dynamics and density-functional theory calculations were used to obtain and rank by energy multiple (GATT + 2H) dication and cation-radical structures. Time-dependent density functional theory (TD-DFT) calculations of excited-state energies and electronic transitions in (GATT + 2H) + ˙ were augmented by vibronic spectra calculations at 310 K for selected low-energy cation radicals to provide a match with the action spectrum. The stable product of one-electron reduction was identified as having a 7,8-dihydroguanine cation radical moiety, formed by intramolecular hydrogen atom migration from adenine N-1–H. The hydrogen migration was calculated tomore »have a transition state with a low activation energy, E a = 96.5 kJ mol −1 , and positive activation entropy, Δ S ‡ = 75 J mol −1 K −1 . This allowed for a fast isomerization of the primary reduction products on the ion-trap time scale of 150 ms that was substantially accelerated by highly exothermic electron transfer.« less
3. Machine learning the Hohenberg-Kohn map for molecular excited states

   https://doi.org/10.1038/s41467-022-34436-w  

   Bai, Yuanming ; Vogt-Maranto, Leslie ; Tuckerman, Mark E. ; Glover, William J. ( November 2022 , Nature Communications)

   The Hohenberg-Kohn theorem of density-functional theory establishes the existence of a bijection between the ground-state electron density and the external potential of a many-body system. This guarantees a one-to-one map from the electron density to all observables of interest including electronic excited-state energies. Time-Dependent Density-Functional Theory (TDDFT) provides one framework to resolve this map; however, the approximations inherent in practical TDDFT calculations, together with their computational expense, motivate finding a cheaper, more direct map for electronic excitations. Here, we show that determining density and energy functionals via machine learning allows the equations of TDDFT to be bypassed. The framework we introduce is used to perform the first excited-state molecular dynamics simulations with a machine-learned functional on malonaldehyde and correctly capture the kinetics of its excited-state intramolecular proton transfer, allowing insight into how mechanical constraints can be used to control the proton transfer reaction in this molecule. This development opens the door to using machine-learned functionals for highly efficient excited-state dynamics simulations.
4. Chemical and Electronic Structures of Cobalt Oxynitride Films Deposited by NH3 vs. N2 Plasma: Theory vs. Experiment

   https://doi.org/10.1039/D0CP04168H  

   Osonkie, Adaeze ; Lee, Veronica ; Oyelade, Adeola ; Mrozek-McCourt, Maximillian ; Chukwunenye, Precious ; Golden, Teresa ; Cundari, Thomas R. ; Kelber, Jeffry A ( January 2020 , Physical Chemistry Chemical Physics)

   The chemical structures of Co oxynitrides – in particular, interactions among N and O atoms bonded to the same cobalt – are of great importance for an array of catalytic and materials applications. X-ray diffraction (XRD), core and valence band X-ray photoelectron spectroscopy (XPS) and plane wave density functional theory (DFT) calculations are used to probe chemical and electronic interactions of nitrogen-rich CoO1-xNx (x > 0.7) films deposited on Si(100) using NH3 or N2 plasma-based sputter deposition or surface nitridation. Total energy calculations indicate that the zincblende (ZB) structure is energetically favored over the rocksalt (RS) structure for x > ~ 0.2, with an energy minimum observed in the ZB structure for x ~ 0.8 - 0.9. This is in close agreement with XPS-derived film compositions when corrected for surface oxide/hydroxide layers. XRD data indicate that films deposited on Si (100) at room temperature display either a preferred (220) orientation or no diffraction pattern, and are consistent with either rocksalt (RS) or zincblende (ZB) structure. Comparison between experimental and calculated X-ray excited valence band densities of states – also similar for all films synthesized herein – demonstrates a close agreement with a ZB, but not an RS structure. Core levelmore »XPS spectra exhibit systematic differences between films deposited in NH3 vs N2 plasma environments. Films deposited by N2 plasma magnetron sputtering exhibit greater O content as evidenced by systematic shifts in N 1s binding energies. Excellent agreement with experiment for core level binding energies is obtained for DFT calculations based on the ZB structure, but not for the RS structure. The agreement between theory and experiment demonstrates that these N-rich Co oxynitride films exhibit the ZB structure, and forms the basis of a predictive model for understanding how N and O interactions impact the electronic, magnetic and catalytic properties of these materials.« less
5. Injecting Domain Knowledge from Empirical Interatomic Potentials to Neural Networks for Predicting Material Properties

   Shui, Zeren ; Karls, Daniel S. ; Wen, Mingjian ; Nikiforov, Ilia A. ; Tadmor, Ellad B. ; Karypis, George ( January 2022 , NeurIPS 2022)

   For decades, atomistic modeling has played a crucial role in predicting the behavior of materials in numerous fields ranging from nanotechnology to drug discovery. The most accurate methods in this domain are rooted in first-principles quantum mechanical calculations such as density functional theory (DFT). Because these methods have remained computationally prohibitive, practitioners have traditionally focused on defining physically motivated closed-form expressions known as empirical interatomic potentials (EIPs) that approximately model the interactions between atoms in materials. In recent years, neural network (NN)-based potentials trained on quantum mechanical (DFT-labeled) data have emerged as a more accurate alternative to conventional EIPs. However, the generalizability of these models relies heavily on the amount of labeled training data, which is often still insufficient to generate models suitable for general-purpose applications. In this paper, we propose two generic strategies that take advantage of unlabeled training instances to inject domain knowledge from conventional EIPs to NNs in order to increase their generalizability. The first strategy, based on weakly supervised learning, trains an auxiliary classifier on EIPs and selects the best-performing EIP to generate energies to supplement the ground-truth DFT energies in training the NN. The second strategy, based on transfer learning, first pretrains the NN onmore »a large set of easily obtainable EIP energies, and then fine-tunes it on ground-truth DFT energies. Experimental results on three benchmark datasets demonstrate that the first strategy improves baseline NN performance by 5% to 51% while the second improves baseline performance by up to 55%. Combining them further boosts performance.« less