Accelerating high-throughput catalyst screening by direct generation of equilibrium adsorption structures
The adsorption energy serves as a crucial descriptor for the large-scale screening of catalysts. Nevertheless, the limited distribution of training data for the extensively utilised machine learning interatomic potential, predominantly sourced from near-equilibrium structures, results in unreliable adsorption structures and consequent adsorption energy predictions. In this context, we present DBCata, a deep generative model that integrates a periodic Brownian-bridge framework with an equivariant graph neural network to establish a low-dimensional transition manifold between unrelaxed and DFT-relaxed structures, without requiring explicit energy or force information. Upon training, DBCata effectively generates high-fidelity adsorption geometries, achieving an interatomic distance mean absolute error of 0.035 Å on the dataset from Catalysis-Hub, which is approximately 70% lower than that of the current state-of-the-art machine learning potential models. Moreover, the corresponding DFT accuracy can be improved within 0.1 eV in 94% of instances by identifying and refining anomalous predictions through a hybrid chemical-heuristic and self-supervised outlier detection approach. We demonstrate that DBCata enables accelerated high-throughput computational screening for alloy catalysts in the oxygen reduction reaction, highlighting the potential of DBCata as a powerful tool for catalyst design and optimisation.
Institutions
- Shanghai Jiao Tong University (CN)
Publication Details
- Journal
- Nature Communications
- Published
- 2026-11-21
- DOI
- https://doi.org/10.1038/s41467-026-79243-9
- Citations
- 1
- Primary Topic
- Machine Learning in Materials Science
- Type
- article
- Field-Weighted Citation Impact
- 0.83
Funders
- National Key Research and Development Program of China