A structure translation model for crystal compounds

High-throughput virtual screening for crystals aims to discover new materials by evaluating the property of every virtual candidate in the database exhaustively. During this process, the major computational bottleneck is the costly structural relaxation of each hypothetical material on the large-scale dataset using density functional theory (DFT) calculations. Here, we present a generative domain translation framework that maps the unrelaxed structural domains to the relaxed domains, enabling data-driven structural translations. The model predicts the materials formation energy with a small mean absolute error without DFT relaxations, and furthermore can produce the atomic coordinates consistent with the DFT relaxed structures. The utility of the proposed concept is not restricted to the structural domains, and we expect that it can be extended to translate the domain of easy-to-compute properties into the domain of more difficult properties.

Automated summary

High-throughput virtual screening for crystals discovers new materials by evaluating the property of every virtual candidate exhaustively, with the computational bottleneck being the costly structural relaxation using DFT calculations. A generative domain translation framework is proposed to map the unrelaxed structural domains to the relaxed domains, enabling data-driven structural translations. This framework accurately predicts the materials formation energy without DFT relaxations and produces atomic coordinates consistent with DFT relaxed structures.