Explainable GNN framework guided by local chemical features to predict binding energies in bimetallic alloys

Abstract

Adsorption energies are key catalytic descriptors that reveal adsorbate-site interactions on heterogeneous catalysts. However, their computation via DFT is time-consuming, limiting high-throughput screening. This work presents a machine learning (ML) methodology based on graph representations of local adsorption sites, using a Graph Neural Network (GNN) with per-atom local descriptors derived from accessible physicochemical properties. The approach is evaluated on two bimetallic datasets. The first includes AB-type bimetallic flat surfaces with varying A:B ratios, predicting binding energies for small monodentate adsorbates (C, N, O, S, H) with MSEs of 0.073/0.181 eV2 (train/test). The second dataset comprises reaction energies of key intermediates for CO2 hydrogenation on Ni-Ga-based surfaces. The GNN model achieves an impressive performance (MSE: 0.001/0.002 (train/test) eV2) on complex atomic configurations, even bidentate ones. Beyond predictive performance, clustering analysis provides an explainable framework, showing how structural and electronic descriptors can rationally guide catalyst design and deepen understanding of adsorbate-metal interactions.