Machine-learned force fields have transformed the atomistic modeling of materials by enabling simulations of ab initio quality on unprecedented time and length scales. However, they are currently limited by: (i) the significant computational and human effort that must go into development and validation of potentials for each particular system of interest; and (ii) a general lack of transferability from one chemical system to the next. Here, using the state-of-the-art MACE architecture we introduce a single general-purpose ML model, trained on a public database of 150k inorganic crystals, that is capable of running stable molecular dynamics on molecules and materials.
The concrete external anchor is Yohannes et al.’s 2023 nitride CO₂RR screening study: ~800 nitrides were screened with DFT, while adsorption-energy regressors used active-motif representations. That separation of structural representation, adsorption chemistry, and validation is exactly the methodological question worth preserving in a public workflow.
Would you welcome a possible future introduction to Prof. Siahrostami’s group around a small, clearly bounded representation-comparison reference case? Silence is completely fine and will be treated as no consent, not as disinterest.