AlphaFlow: autonomous discovery and optimization of multi-step chemistry using a self-driven fluidic lab guided by reinforcement learning

Closed-loop, autonomous experimentation enables accelerated and material-efficient exploration of large reaction spaces without the need for user intervention. However, autonomous exploration of advanced materials with complex, multi-step processes and data sparse environments remains a challenge. In this work, we present AlphaFlow, a self-driven fluidic lab capable of autonomous discovery of complex multi-step chemistries. AlphaFlow uses reinforcement learning integrated with a modular microdroplet reactor capable of performing reaction steps with variable sequence, phase separation, washing, and continuous in-situ spectral monitoring. To demonstrate the power of reinforcement learning toward high…

• NS
  National Science Foundation
• CA
  Camille and Henry Dreyfus Foundation

  Award: ML-21-064
• RO
  Research Opportunities Initiative, University of North Carolina
• DO
  Division of Chemical, Bioengineering, Environmental, and Transport Systems

  Award: 1940959
• DO
  Division of Civil, Mechanical and Manufacturing Innovation

  Award: 1902708