5 December 2017 in The Journal of Chemical Physics
To clarify atomic diffusion in amorphous materials, which is important in novel information and energy devices, theoretical methods having both reliability and computational speed are eagerly anticipated. In the present study, we applied neural network (NN) potentials, a recently developed machine learning technique, to the study of atom diffusion in amorphous materials, using Li3PO4 as a benchmark material. The NN potential was used together with the nudged elastic band, kinetic Monte Carlo, and molecular dynamics methods to characterize Li vacancy diffusion behavior in the amorphous Li3PO4 model. By comparing these results with corresponding DFT calculations, we found that the average error of the NN potential is 0.048 eV in calculating energy barriers of diffusion paths, and 0.041 eV in diffusion activation energy. Moreover, the diffusion coefficients obtained from molecular dynamics are always consistent with those from ab initio molecular dynamics simulation, while the computation speed of the NN potential is 3–4 orders of magnitude faster than DFT. Lastly, the structure of amorphous Li3PO4 and the ion transport properties in it were studied with the NN potential using a large supercell model containing more than 1000 atoms. The formation of P2O7 units was observed, which is consistent with the experimental characterization. The Li diffusion activation energy was estimated to be 0.55 eV, which agrees well with the experimental measurements.
15 May 2023 in Journal of Materials Chemistry A
Seawater splitting remains far-fetched due to the interference of corrosive chlorine evolution reactions at the anode. The lowering of overpotential with the help of an effective electrocatalyst is one way to deal with it. However, to achieve hydrogen production under high current, a potential above 2.0 V needs to be applied practically, which is an electrochemical chloride oxidation–dominant region. In this study, we show that the presence of a trace amount of anionic inhibitors, especially carbonate, provides a much-needed anionic protective layer at the anode surface, selectively restricting chlorides but allowing hydroxides to reach the anode surface. The hypothesis was examined on a novel MnFeCr layered triple hydroxide as a bifunctional electrocatalyst. The developed electrocatalyst exhibits an overpotential of 242 mV for OER and 154 mV for HER at a benchmarking current density of 10 mA cm−2. Theoretical calculations were used to establish the possible OER pathways and explain the reason for a good electrocatalytic activity of the developed electrocatalyst. Having such an electrocatalytic activity, in alkaline real seawater, the electrocatalyst in the presence of carbonate inhibitors shows an impressive stability over 500 h maintaining a cell potential of 2.2–2.3 V without showing any sign of electrode degradation. The findings thereby provide a much-needed solution to make seawater splitting feasible for sustainable hydrogen production.
15 December 2020 in Journal of Physics: Energy
Owing to their simultaneous accuracy and computational efficiency, interatomic potentials machine-learned using first-principles calculation data are promising for investigating phenomena closely related to atomic motion in various energy materials. We have been working with one type of these potentials, high-dimensional (HD) neural network potentials (NNPs), and their applications, but we realized that our current understanding of HD NNPs, e.g. the meaning of the atomic energy mapping, remained insufficient, and that tuning their prediction performance for different target properties/phenomena often requires much trial and error. In this article, we illustrate the usefulness of NNPs through our studies on ion migration and thermal transport in energy and related materials. We also share our experiences with data sampling and training strategies and discuss the meaning of atomic energy mapping in HD NNPs.
12 August 2019 in Applied Physics Express
We demonstrate that a high-dimensional neural network potential (HDNNP) can predict the lattice thermal conductivity of semiconducting materials with an accuracy that is comparable to that of density functional theory (DFT) calculation. After a training procedure based on force, the root mean square error between the forces predicted by HDNNP and DFT is less than 40 meV Å−1. As typical examples, we present the results of Si and GaN bulk crystals. The deviation from the thermal conductivity calculated using DFT is within 1% at 200 to 500 K for Si and within 5.4% at 200 to 1000 K for GaN.
1 December 2021 in Chemical Engineering Journal
13 September 2007 in Physical Review Letters
The electronic structure near the Fermi level () of the -pyrochlore superconductor is studied using laser-excited ultrahigh-resolution photoemission spectroscopy. The superconducting gap clearly opens across the superconducting transition (), with the strong electron-phonon coupling value of . A fitting analysis identifies clear anomalies at in the temperature dependencies of the superconducting gap size and the quasiparticle relaxation lifetime. These anomalies and the fine spectral structures arising from phonons suggest that the existence of the rattling behavior of K ions significantly affects the superconductivity in .