Accelerating High-Throughput Catalyst Screening by Direct Generation of Equilibrium Adsorption Structures
This addresses the bottleneck of limited training data distribution for catalyst screening researchers, though it appears to be an incremental improvement over existing generative approaches in computational chemistry.
The paper tackles the problem of unreliable adsorption energy predictions in catalyst screening by developing DBCata, a deep generative model that directly generates equilibrium adsorption structures without requiring explicit energy or force information. The model achieves an interatomic distance mean absolute error of 0.035 Å, nearly three times better than current state-of-the-art machine learning potential models, and improves DFT accuracy within 0.1 eV in 94% of cases.
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 (MLIP), 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 (DMAE) of 0.035 \textÅ on the Catalysis-Hub dataset, which is nearly three times superior to 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 the remarkable performance of DBCata facilitates accelerated high-throughput computational screening for efficient alloy catalysts in the oxygen reduction reaction, highlighting the potential of DBCata as a powerful tool for catalyst design and optimisation.