Self driving labs for the optimization of photochemical processes

The integration of automation, artificial intelligence (AI), and robotics into synthetic chemistry has ushered in a new era of self-driving laboratories (SDLs) capable of accelerating reaction optimization and chemical discovery. This thesis presents RoboChem, an advanced SDL platform designed for the optimization of photochemical processes. By leveraging Bayesian optimization and automated experimentation, RoboChem efficiently explores chemical space, reducing the number of required experiments while improving reproducibility and scalability.
Chapter 1 introduces the field of SDLs, emphasizing their potential to overcome inefficiencies in traditional organic synthesis. Chapter 2 details the development of the RoboChem platform, including its hardware and software architecture, fluidic automation, and AI-driven decision-making. Chapter 3 applies RoboChem to synthetic photochemistry, demonstrating its capability for optimizing reaction conditions, light intensity, and catalyst loadings. Chapter 4 expands the platform’s utility to automated substrate selection and functional group tolerance testing, enabling high-throughput dataset generation for machine learning applications.
By automating photochemical optimization, RoboChem enables rapid discovery of efficient reaction conditions, minimizes human intervention, and enhances safety. The findings in this thesis highlight the transformative role of SDLs in modern chemistry, paving the way for broader applications in reaction discovery, materials science, and beyond.