CBE project selected for ARPA-E effort to automate and speed up design of new industrial catalysts

The U.S. Department of Energy’s Advanced Research Projects Agency-Energy (ARPA-E) has selected a project led by faculty in the University of Wisconsin-Madison’s Department of Chemical and Biological Engineering for its Catalytic Application Testing for Accelerated Learning Chemistries via High-throughput Experimentation and Modeling Efficiently (CATALCHEM-E) program.

The team includes Richard L. Antoine Professor George Huber; Ernest Micek Distinguished Chair, James A. Dumesic Professor, and Vilas Distinguished Achievement Professor Manos Mavrikakis; Baldovin-DaPra Professor Victor Zavala; Duane H. and Dorothy M. Bluemke Assistant Professor Siddarth Krishna; Faculty Associate Eric Codner; honorary research fellow and VP of catalysis at Pyran, Inc. Daniel McClelland; as well at two of CBE’s new faculty members, Paul A. Elfers Professor Jeff Greeley and Gerald and Louise Battist Associate Professor Joel Paulson. BASF, one of the world’s largest chemical companies, is serving as industrial partner on the project, represented by Amit Gokhole, a 2005 CBE PhD graduate.

The $2.84 million award will fund a project titled AI-Driven Design of Industrial Fixed Bed Reactor Catalysts for Production of Renewable Fuels and Chemicals from Oxygenated Feedstocks (AI-FIXCAT). The goal is to develop catalysts to convert ethanol into alcohol for fuels and specialty chemicals by combining AI models with automated lab tools and industrial scale catalyst testing. The team will also explore integrating the technology into standard laboratory information management systems to make the design workflow widely accessible to a range of commercial entities. 

The UW-Madison project is one of a dozen projects selected for CATALCHEM-E at universities, national labs and private research organizations across the United States. The goal of the $34 million program is to pair artificial intelligence with self-driving laboratories to dramatically accelerate industrial catalyst development for fuels and chemicals production.

CATALCHEM-E aims to compress catalyst development timelines from roughly a decade to about one year. Projects will seek to integrate machine learning, AI-guided design, and high-throughput experimentation platforms into continuous, automated discovery workflows. The program targets 10x faster progress in designing and validating industrially relevant catalysts such as those that convert oil and gas feedstocks into widely used fuels and commodity chemicals. 

“America is upgrading its industrial base to reclaim our global leadership, and commodities like industrial catalysts play a foundational role in this effort,” says ARPA-E director Conner Prochaska. “CATALCHEM-E’s goal is to harness the power of AI paired with self-driving labs to slash the development timeline for these critical building blocks from a decade to a year. This will empower American refineries, factories, and industrial plants to strengthen manufacturing, energy independence, and national security.”

A full list of selected CATALCHEM-E projects is available here.