<br/>The development of high-throughput experimentation over the past three decades allows the screening of large compositional and operational spaces and create large datasets under reproducible conditions. Machine learning is an avenue to extract knowledge and unravel multidimensional relationships present in such large experimental datasets.<br/>A framework will be described that incorporates machine learning algorithms with experimental high-throughput catalytic data, spectroscopic data, and literature data to discover new materials. The framework allows us to rank features and predict catalyst performance for previously unknown compositions. Application of this framework for low-temperature ammonia decomposition and ammonia synthesis under mild conditions predicted several novel Ru-based catalyst compositions with multiple promoters and different supports. These compositions were experimentally validated against state-of-the-art ammonia catalysts and were found to have exceptional performance at substantially lower weight loadings of Ru. Features having a strong influence on catalyst performance can also be ranked by the approach and can contribute to fundamental understanding.

High-throughput experimentation and spectroscopy meet artificial intelligence: Catalyst discovery for enabling ammonia as a carbon-free fuel

Where thousands of chemistry professionals meet to share ideas and advance scientific and technical knowledge. Spring 2023 abstracts, posters, and presentations will launch at the end of March.

ACS Spring 2023

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.