NSF PCL-test bed: speed: programmable cloud laboratories for autonomous chemistry and materials discovery

The National Science Foundation awarded North Carolina State University $5 million on August 1, 2026, under the NSF Technology, Innovation, and Partnerships program (CFDA 47.084) to establish SPEED, a programmable cloud laboratory for autonomous chemistry and materials discovery.

SPEED will operate as a shared, remotely accessible research facility combining robotics, artificial intelligence, and real-time measurement tools to enable researchers to design, run, and optimize experiments in chemistry and materials science. The facility will focus on three research areas: ligand-controlled homogeneous catalysis, solution-processable semiconductor materials, and photocatalysis as a cross-cutting domain linking the first two. By integrating autonomous experimentation, real-time analysis, and machine-learning-guided decision-making, SPEED will accelerate discovery of catalysts for efficient chemical synthesis and development of semiconductor materials for energy, sensing, and electronics applications.

The laboratory will serve academic and industrial users nationwide, including researchers at institutions lacking autonomous science capabilities or major research infrastructure. The award supports workforce development through training environments, onboarding, and educational activities preparing students and researchers to work across science, data, robotics, and artificial intelligence. Place of performance is Raleigh, North Carolina, with a period of performance extending through July 31, 2030. The cooperative agreement structure enables collaborative research partnership between NSF and the recipient institution.

Abstract

NSF PCL-test bed: speed: programmable cloud laboratories for autonomous chemistry and materials discovery this project will establish speed, a programmable cloud laboratory that will allow researchers to design, run, and improve experiments through a shared, remotely accessible facility for chemistry and materials science. By combining robotics, artificial intelligence (AI), and real-time measurement tools, speed will accelerate the discovery of catalysts that make valuable chemicals more efficiently and the development of solution-processed semiconductor materials for emerging energy, sensing, and electronics technologies. Speed will also advance light-driven chemical systems that connect these two fields. This project will strengthen U.S. Leadership in autonomous science, expand access to advanced research infrastructure, and support innovation relevant to manufacturing, semiconductors, and the chemical enterprise. Speed will serve academic and industrial users nationwide, including researchers at institutions without autonomous science capabilities or major research infrastructure. Through training environments, onboarding, and workforce development activities, speed will prepare students and researchers to work at the intersection of science, data, robotics, and AI. Speed will establish a programmable cloud laboratory node for solution-phase chemistry and materials science by combining autonomous experimentation, real-time analysis, and machine-learning-guided decision-making in a shared user facility. The project will focus on ligand-controlled homogeneous catalysis and solution-processable semiconductor materials, with photocatalysis serving as a cross-cutting area that links the two. These areas present closely related scientific and engineering challenges where performance depends strongly on ligand identity, reaction conditions, and processing history, yet discovery remains slow because synthesis, testing, and optimization are still carried out in separate, largely manual workflows. Speed will address these gaps by creating a closed-loop CO-design environment that connects ligand synthesis, homogeneous catalysis, materials synthesis, and functional testing. Building on NC state?s existing self-driving laboratory infrastructure, speed will unify batch and flow synthesis platforms, multimodal characterization tools, digital twins, and a common data framework that records the full history of each experiment. This environment will allow AI models to propose new experiments, update predictions from incoming data, and test hypotheses under realistic laboratory conditions. By generating portable workflows, richly annotated datasets, interoperable ontologies, and predictive models that connect synthesis history to structure, function, and performance, speed will provide an experimental foundation for autonomous science and a national test bed for verifiable and transferable cloud-based experimentation. The resulting workflows and datasets will provide a rigorous foundation for training, validating, and transferring predictive models across instruments, users, and sites nationwide. Speed will also establish community-accessible benchmarks for evaluating AI under realistic experimental conditions. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the foundation's intellectual merit and broader impacts review criteria.- subawards are planned for this award.