The Open Catalyst 2022 (OC22) dataset and challenges for oxide electrocatalysts
Published in ACS Catalysis, 2023
Citation: Richard Tran, Janice Lan, Muhammed Shuaibi, Brandon M Wood, Siddharth Goyal, Abhishek Das, Javier Heras-Domingo, Adeesh Kolluru, Ammar Rizvi, Nima Shoghi, Anuroop Sriram, Félix Therrien, Jehad Abed, Oleksandr Voznyy, Edward H Sargent, Zachary Ulissi, C Lawrence Zitnick, ACS Catalysis 13 (5), 3066-3084, 2023 https://pubs.acs.org/doi/abs/10.1021/acscatal.2c05426
The Open Catalyst 2022 (OC22) dataset aims to accelerate the development of machine learning models for oxide electrocatalysts, which are crucial for the Oxygen Evolution Reaction (OER). The dataset consists of 62,331 Density Functional Theory (DFT) relaxations across various oxide materials, coverages, and adsorbates, enabling property prediction beyond adsorption energies. The authors establish clear benchmarks for future efforts by defining generalized total energy tasks, testing baseline performance of graph neural networks, and providing predefined dataset splits.