Materials discovery acceleration by using condition generative methodology

Abstract

With the rapid advancement of AI technologies, generative models have been increasingly employed in the exploration of novel materials. By integrating traditional computational approaches such as density functional theory (DFT) and molecular dynamics (MD), existing generative models, including diffusion models and autoregressive models, have demonstrated remarkable potential in the discovery of novel materials. However, their efficiency in goal-directed materials design remains suboptimal. In this work we developed a highly transferable, efficient and robust conditional generation framework, PODGen, by integrating a general generative model with multiple property prediction models. Based on PODGen, we designed a workflow for the high-throughput crystals conditional generation which is used to search new topological insulators (TIs). Our results show that the success rate of generating TIs using our framework is 5.3 times higher than that of the unconstrained approach. More importantly, while general methods rarely produce gapped TIs, our framework succeeds consistently, highlighting an effectively $\infty$ improvement. This demonstrates that conditional generation significantly enhances the efficiency of targeted material discovery. Using this method, we generated tens of thousands of new topological materials and conducted further first-principles calculations on those with promising application potential. Furthermore, we identified promising, synthesizable topological (crystalline) insulators such as CsHgSb, NaLaB${}_{12}$, Bi${}_4$Sb${}_2$Se${}_3$, Be${}_3$Ta${}_2$Si and Be${}_2$W.

References in corpus (28)

• VASPKIT: A User-friendly Interface Facilitating High-throughput Computing and Analysis Using VASP Code
• Topological Insulators in Three Dimensions
• Topological Insulators with Inversion Symmetry
• Maximally localized Wannier functions: Theory and applications
• Crystal Graph Convolutional Neural Networks for an Accurate and Interpretable Prediction of Material Properties
• WannierTools: An open-source software package for novel topological materials
• Maximally-localized Wannier functions for entangled energy bands
• Topological quantum chemistry
• Graph Networks as a Universal Machine Learning Framework for Molecules and Crystals
• Implementation strategies in phonopy and phono3py
• The (High Quality) Topological Materials In The World
• A Universal Graph Deep Learning Interatomic Potential for the Periodic Table
• Catalogue of Topological Electronic Materials
• An equivalent expression of Z2 Topological Invariant for band insulators using Non-Abelian Berry's connection
• Atomistic Line Graph Neural Network for Improved Materials Property Predictions
• All Topological Bands of All Nonmagnetic Stoichiometric Materials
• Data-driven discovery of novel 2D materials by deep generative models
• MatterSim: A Deep Learning Atomistic Model Across Elements, Temperatures and Pressures
• Open Materials 2024 (OMat24) Inorganic Materials Dataset and Models
• Transferable E(3) equivariant parameterization for Hamiltonian of molecules and solids
• Universal Machine Learning Kohn-Sham Hamiltonian for Materials
• Con-CDVAE: A method for the conditional generation of crystal structures
• Language models can generate molecules, materials, and protein binding sites directly in three dimensions as XYZ, CIF, and PDB files
• CrystalFlow: A Flow-Based Generative Model for Crystalline Materials
• MatterGPT: A Generative Transformer for Multi-Property Inverse Design of Solid-State Materials
• DPmoire: A tool for constructing accurate machine learning force fields in moiré systems
• Large-scale machine-learning-assisted exploration of the whole materials space
• DeepH-2: Enhancing deep-learning electronic structure via an equivariant local-coordinate transformer