Open Catalyst Project - Oxide Catalyst Dataset Development

Education and Research

Technology

AI models developed by Meta and Carnegie Mellon to simulate chemical reactions and predict catalyst performance. These models use the oxide catalyst dataset to identify cheaper alternatives to precious metals.

A critical chemical process in hydrogen production that requires expensive precious metal oxides like ruthenium and iridium. The OER is a key step in electrolysis-based hydrogen production.

A type of catalyst, specifically developed by Adzic, that is more effective than conventional platinum catalysts and helps reduce the cost of fuel cell production without sacrificing performance.

Pennsylvania

A collaborative research initiative between Meta's AI division and Carnegie Mellon University's Department of Chemical Engineering to develop machine learning models for simulating chemical reactions and identifying cheaper catalyst materials for hydrogen production.

An open-source dataset containing approximately 8 million data points from 40,000 unique simulations across oxide materials spanning 52 elements. Designed to support AI research in identifying cheaper catalyst alternatives for hydrogen production.