Development of a Robotic System for Automatic Organic Chemistry Synthesis

Abstract

Automated chemical synthesis has great promises of safety, efficiency, and reproducibility for both research and industry laboratories. Current approaches are based on specifically designed automation systems, which present two major drawbacks: 1) existing apparatus must be modified to be integrated into the automation systems and 2) such systems are not flexible and would require substantial redesign to handle new reactions or procedures. In this article, we propose a system based on a robot arm that mimics motions of human chemists, performs complex chemical reactions with no modifications to the existing setup used by humans, and thus removes human interventions. The automated system is capable of precise liquid handling, mixing, and filtering and is flexible; new skills and procedures could be added with minimum effort. The production sequence is customizable by chaining tasks together. We show that the robot is able to perform a Michael reaction, reaching a yield of 34%, which is comparable to that obtained by a junior chemist (undergraduate student in Chemistry). <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Note to Practitioners</i> —This article explored methods to reduce the need of additional modifications on structured environments to implement automation. Existing approaches in chemical synthesis automation require substantial modifications on apparatus to incorporate automation, and hence, they are inflexible to changes in the procedures or reactions. This article suggested a new system that performed chemical reactions in an existing setup used by human chemist and was flexible in the production sequence or addition of tasks. We showed how the robot arm was capable of automating a Michael reaction and repeating the experiment again, which could be further extended multiple times. The yield obtained by the robot was comparable to junior chemist and consistent but lower than a senior chemist. Hence, we could improve the yield by understanding skills of a senior chemist and transferring them to the robot.