The Pulse
DOE Taps SLAC to Test AI-Directed Catalyst Experiments
SLAC will lead a Department of Energy Genesis Mission project that uses an AI system to select experiments on working catalysts. The Phase I project will test whether a closed-loop approach can explain catalytic activity with fewer experime
AI.info Team · Written with AI ·
SLAC will lead one of nine new Department of Energy Genesis Mission projects it joins, with its own project asking an AI system to choose the next experiment on a working catalyst. Announced October 8, the Phase I award will test whether a closed-loop process can reveal how catalysts behave with fewer experiments than standard approaches.
At SSRL, each result will shape the next experiment
The project, called “Closed-Loop Autonomous Discovery of Mechanistic Activity-Selectivity-Stability Rules in Catalysis,” links AI reasoning to experiments at SLAC’s Stanford Synchrotron Radiation Lightsource (SSRL). The system will compare competing explanations for how a catalyst works, identify evidence that is missing, then select a measurement to test those explanations.
Researchers will combine real-time X-ray measurements with electrochemistry, computer models and published scientific knowledge. Each experimental result will return to the system and inform its next choice, creating a cycle of hypothesis, measurement and updated reasoning.
“The SI weighs the scientific evidence, decides which measurement would tell us the most, and that decision becomes a real experiment at SSRL,” said project principal investigator Dimosthenis Sokaras, director of the Chemistry & Catalysis Division at SSRL. “The result comes straight back into the reasoning process. That closed loop is a key step toward autonomous laboratories, where scientists can test ideas and learn much faster.”
The award tests a claim, not a finished capability
SLAC describes the project as a Phase I award. The team’s test is whether this method can reach scientific understanding with fewer experiments than standard approaches; the announcement does not report that it has already done so.
The distinction matters because the system is being built to guide research, not simply to run a predetermined series of measurements. SLAC says the project will evaluate whether integrating the AI system with scientific investigation can accelerate discovery, improve predictive capabilities, enhance experimentation or produce new scientific insights.
The work draws on SLAC’s existing combination of catalyst testing, theory and X-ray measurements of working catalysts. It also connects to the lab’s ISAAC AI Pathfinder, which brings together measurements from DOE facilities, calculations run on DOE supercomputers and scientific literature to weigh evidence and rank hypotheses.
Caltech and Berkeley Lab join the SLAC team
Sokaras is the project’s principal investigator. The collaboration includes the California Institute of Technology, led by engineering and applied science professor Harry Atwater, and Lawrence Berkeley National Laboratory, led by Molecular Biophysics and Integrated Bioimaging division director Junko Yano.
Researchers from the Liquid Sunlight Alliance and the SUNCAT Center for Interface Science and Catalysis are also involved. SUNCAT is a joint venture between SLAC and Stanford’s School of Engineering; the Liquid Sunlight Alliance is a multi-institution consortium led by Atwater that includes SLAC.
Faster insight into catalyst behavior could help researchers develop more efficient and durable processes for energy and chemical production, SLAC said. The Phase I work is intended to establish whether the closed-loop approach can support that broader goal.
SLAC also joins eight other Genesis projects
Beyond the catalysis effort, SLAC will partner on eight projects led by other laboratories, universities and a company. They include research on computational enzyme design, quantum magnets, scientific software, accelerator facilities, microelectronics, fusion energy and RNA structure.
Those collaborations include a fusion-energy digital-twin project led by Commonwealth Fusion Systems and an accelerator project led by Lawrence Berkeley National Laboratory. SLAC will also take part in work on enzyme design led by the University of Washington and RNA structure led by the University of California, San Diego.
The Department of Energy’s October 8 announcement separately named 12 Phase II Genesis awards totaling $159 million, alongside six new Phase I awards. SLAC identifies its catalysis project as Phase I; its announcement does not state a project-specific funding amount.