Chemists and engineers at the University of British Columbia and pharmaceutical company Pfizer have collaborated to develop an AI-powered system for automating chemical workup processes. Published in the journal Chemical Science, their research outlines a chemical processing system that combines computer vision with real-time machine learning monitoring.
Chemical workup processes involve isolating a pure product through selective separation from other components, a task that can be tedious and prone to errors. In an attempt to automate this process, the team has integrated computer vision, real-time monitoring techniques, machine learning, and computer processing to carry out workup processes without human intervention.
The system, known as Heinsight2.0, builds on its predecessor’s knowledge and includes components such as a webcam, reaction vessel, dosing unit, temperature probe, and overhead stirrer. It also features a secondary device that displays iControl, real-time reaction trends, EasyMax, and CV model output.
The system functions by monitoring and controlling the workup process. It responds to sensory cues like a human chemist, initiating follow-up actions based on observed changes. It is capable of handling various scenarios involving solid-liquid mixing, crystallizations, exchange distillations, and liquid-to-liquid extraction.
The team has made the program script publicly available, enabling other chemists to build their own units and utilize the code for running their systems. Furthermore, the researchers plan to enhance the system’s capabilities through further development.
The development of an AI-powered system for automating chemical workup processes holds promise for improving efficiency and reducing errors in chemistry labs. By combining computer vision, real-time monitoring, and machine learning, the system can accurately monitor multiple sensory cues and respond accordingly. Additionally, the availability of the program script allows chemists to replicate and adapt the system for their own use.
While the publication of the research highlights the successful implementation of Heinsight2.0, the team recognizes that there is still room for improvement in terms of expanding the system’s capabilities. By continuing their work and making further advancements, they aim to advance automation in chemical processes.
Overall, the collaboration between chemists, engineers, and pharmaceutical experts in developing an AI-powered system for automating chemical workup processes marks a significant advancement in the field. The integration of computer vision, real-time monitoring, and machine learning has the potential to revolutionize the way chemical workup processes are conducted, increasing efficiency and reducing human error. As the system evolves, it holds great promise for the future of chemical research and development.
