Interface Structures using Neural Network Potential*

Recently, the construction of interatomic potentials using first-principles calculation data and machine-learning technique has been widely tried because of higher reliability and low computational costs. In the present study, we have tried to construct the four-element neural network potential (NNP) [1,2] to investigate the Au(111)/Li₃PO₄ interface system, where the understanding of the interface structures and Li-ion distribution near the interface is of significance for the development of all-solid state Li-ion batteries and novel memory devices [3]. Using the constructed NNP, we then performed structure optimization with a large interface model of Au(111)/Li₃PO₄. In the meeting, we will discuss the calculated interface structures and the Li defect formation energies.

[1] J. Behler et al., Phys. Rev. Lett. 98, 146401 (2007).
[2] W. Li et al., J. Chem. Phys. 147, 214106 (2017).
[3] I. Sugiyama et al., APL Mater. 5, 046105 (2017).