New award to WPI catalchem-e control number: 3505-1551 title: autonomous high throughput experimentation to accelerate chemical manufacturing (a-team) the objective of the a-team project is to disrupt…

The Department of Energy Advanced Research Projects Agency awarded Worcester Polytechnic Institute a $1.375 million Cooperative Agreement on August 4, 2026, to develop autonomous, high-throughput experimentation workflows that compress traditional catalyst development cycles from 10–15 years into 18-month phases under the Advanced Research Projects Agency - Energy program (CFDA 81.135).

WPI will establish an integrated human-AI-robot collaboration ecosystem for accelerated thermal catalytic transformation development. The work combines forward machine-learning models and inverse design diffusion models to rapidly explore catalytic design spaces, while robotic platforms execute synthesis and screening with minimal human intervention. High-velocity proxy reactions and measurement provide rapid feedback to refine models, and multiscale validation—from atomistic simulations to pilot-scale reactor testing—ensures industrial relevance. By Phase 2, the workflow will deliver high-performing heterogeneous hydroformylation catalysts that eliminate precious metal loss and excessive energy consumption to strengthen domestic manufacturing competitiveness.

Performance occurs in Worcester, Massachusetts, with a period of performance running through August 3, 2029. The funding integrates model and workflow leadership, industrial expertise in chemical and catalyst manufacturing, and hardware innovation and data infrastructure capabilities to create a self-correcting design-build-test-learn ecosystem capable of navigating the inherent uncertainties and multiscale complexities of heterogeneous catalyst discovery.

New award to WPI catalchem-e control number: 3505-1551 title: autonomous high throughput experimentation to accelerate chemical manufacturing (a-team) the objective of the a-team project is to disrupt traditional industrial catalyst development cycles-typically spanning 10-15 years-by establishing an autonomous, agentic workflow that compresses these cycles into 18-month phases. The overarching goal is to accelerate the development and deployment of thermal catalytic transformations for chemical manufacturing through an integrated human-AI-robot collaboration. The integration of model/workflow leadership, industrial expertise in chemical/catalyst manufacturing, and a hardware innovation/data infrastructure provider provides a self-correcting design-build-test learn ecosystem capable of navigating the inherent uncertainties and multiscale complexities of heterogeneous catalyst discovery. Forward machine-learning models and inverse design diffusion models rapidly explore vast catalytic design spaces, while robotic platforms execute synthesis and screening with minimal human intervention. High-velocity proxy reactions and measurement provide rapid feedback to refine models, and multiscale validation-from atomistic simulations to pilot-scale reactor testing-ensures industrial relevance. By phase 2, the workflow will deliver high-performing heterogeneous hydroformylation catalysts that eliminate precious metal loss and excessive energy consumption, directly bolstering domestic manufacturing competitiveness.