Data-driven discovery of novel 2D materials by deep generative models

TL;DR

This paper introduces a deep generative model called CDVAE that efficiently creates diverse and potentially synthesizable 2D materials, significantly expanding the known materials space for data-driven discovery.

Contribution

The study demonstrates that CDVAE can generate high-quality 2D materials with diverse structures and compositions, validated by DFT relaxations and systematic element substitution.

Findings

Generated 11630 new 2D materials with many predicted to be synthesizable.

CDVAE and lattice decoration methods are complementary, producing diverse structures.

The relaxed structures are available in the open C2DB database.

Abstract

Efficient algorithms to generate candidate crystal structures with good stability properties can play a key role in data-driven materials discovery. Here we show that a crystal diffusion variational autoencoder (CDVAE) is capable of generating two-dimensional (2D) materials of high chemical and structural diversity and formation energies mirroring the training structures. Specifically, we train the CDVAE on 2615 2D materials with energy above the convex hull $\Delta H_{\mathrm{hull}}< 0.3$ eV/atom, and generate 5003 materials that we relax using density functional theory (DFT). We also generate 14192 new crystals by systematic element substitution of the training structures. We find that the generative model and lattice decoration approach are complementary and yield materials with similar stability properties but very different crystal structures and chemical compositions. In total we find 11630 predicted new 2D materials, where 8599 of these have $\Delta H_{\mathrm{hull}}< 0.3$ eV/atom as the seed structures, while 2004 are within 50 meV of the convex hull and could potentially be synthesized. The relaxed atomic structures of all the materials are available in the open Computational 2D Materials Database (C2DB). Our work establishes the CDVAE as an efficient and reliable crystal generation machine, and significantly expands the space of 2D materials.