Framework-Constrained Materials Generation

Maya M. Martirossyan, Hillary Pan, Philipp Höllmer, Stefano Martiniani

Keywords: generative models, materials design, inorganic crystals, symmetry-constrained models

TL;DR: A new type of generative model for inorganic crystals in which a structural framework can be used to constrain predictions of new materials.

Abstract: Theoretical predictions of novel materials rely on a variety of techniques, from data-mining-based ion substitution methods to fully _de novo_ generative models. Each approach imposes specific constraints and encounters unique challenges when extrapolating beyond the distribution of known materials to discover new ones. The foremost goal of materials generation is to produce stable, unique, and novel (SUN) structures. Predicted structures that meet these criteria, however, typically result from novel chemical compositions mapped onto known structural prototypes or frameworks. We leverage this heuristic insight to present a new generative approach for the _de novo_ prediction of inorganic crystalline materials constrained by their _structural framework_—_i.e._, the unit cell shapes and respective particle positions. Our method, termed OMatG-FC (Open Materials Framework-Constrained Generation), learns _via_ a combination of discrete flow matching on the atomic species and stochastic interpolants for the volume scaling of the unit cell lattice, while fixing in-place the fractional coordinates and the unit-volume lattice vectors, dramatically reducing the dimensionality of the problem compared to unconstrained (framework-free) generation. Our method is more flexible than existing structure-constrained methods, as it allows for symmetry-breaking or chemical disorder on sites of equivalent crystalline symmetry. We also investigate the number of unique frameworks in the $MP$-20 dataset, demonstrating the propensity of duplicate frameworks within the dataset. We benchmark the performance of our model using the LeMat-GenBench suite using three ML interatomic potentials and achieve state-of-the-art SUN rate, eschewing the need to learn atomic packings for materials discovery.

Submission Track: Full Paper

Submission Category: AI-Guided Design

Submission Number: 57