Accelerating the Discovery of Metastable IrO2 for the Oxygen Evolution Reaction by the Self-Learning-Input Graph Neural Network

The discovery of active and stable catalysts for the oxygen evolution reaction (OER) is vital to improve water electrolysis. To date, rutile iridium dioxide IrO2 is the only known OER catalyst in the acidic solution, while its poor activity restricts its practical viability. Herein, we propose a universal graph neural network, namely, CrystalGNN, and introduce a dynamic embedding layer to self-update atomic inputs during the training process. Based on this framework, we train a model to accurately predict the formation energies of 10,500 IrO2 configurations and discover 8 unreported metastable phases, among which C2/m-IrO2 and P62-IrO2 are identified as excellent electrocatalysts to reach the theoretical OER overpotential limit at their most stable surfaces. Our self-learning-input CrystalGNN framework exhibits reliable accuracy, generalization, and transferring ability and successfully accelerates the bottom-up catalyst design of novel metastable IrO2 to boost the OER activity.

Automated summary

By introducing a dynamic embedding layer, we propose a universal graph neural network called CrystalGNN, which can automatically update atomic inputs during the training process. We train a model based on this framework to accurately predict the formation energies of 10,500 IrO2 configurations and discover 8 unreported metastable phases. Among them, C2/m-IrO2 and P62-IrO2 are identified as excellent electrocatalysts that can reach the theoretical OER overpotential limit at their most stable surfaces. Our self-learning-input CrystalGNN framework exhibits reliable accuracy, generalization, and transferring ability and successfully accelerates the bottom-up catalyst design to boost the OER activity.