The Dynamic Orchestration of Self-Driving Laboratories

Abstract

Self-Driving Labs (SDLs) combine automated hardware with computational experiment planning tools to reduce the time between experiments and liberate chemists from routine work, allowing them to focus on bigger, more conceptual problems. SDLs thus have the potential to acceleration of chemical research. However, SDLs are often difficult to implement for existing labs. Barriers include financial cost, lack of accessible Application Programming Interfaces (APIs) for chemical hardware, and modularity. Another one is orchestration: the administration of the plethora of toolsavailable to the laboratory users. Unfortunately, existing frameworks lack one of the following key elements: modularity, data collection strategies, and a comprehensive real-life implementation. To address these concerns, this research presents a framework for orchestrating chemical labs, and the automation of laboratory instruments. In addition, implementations are presented to demonstrate the implementation of this framework: A campaign for the synthesis of organic laser molecules, as well as an electrochemical optimization experiment.