pandas

.NET

Advancing Zn-Ion Battery Electrolytes with Robotic Experimentation and Machine Intelligence

Cyclic-Voltammetry testing is an intensive experiment requiring significant time from a human operator. In this project, it is used to determine the performance of aqueous electrolyte solutions, by repeatedly simulating the charge and discharge cycles of a battery. The system includes 3 electrodes. A cathode, anode, and reference, all connected to an Autolab electrochemical testing station.

Our platform is an XY stage operating over a space of 40 in x 20 in. This operating space is segmented into slots in which different modules can be placed. A key constraint is that the two working electrodes have to be swapped out between each test, while all three electrodes must reestablish a connection to the autolab station to conduct testing. This was solved in two parts. First, we designed exchangeble electrode holders for each set of working electrodes. Second we designed a custom end effector that can attach and detach these electrode holders, while maintaining a connection to the Autolab. This project also required integration with existing lab workflows which was achieved using our existing sample id labeling system, as well as an API for protocol generation.

Each test can take up to 2 hours to perform, and up to 54 samples can be tested per batch. This platform can be run 24/7 completely unsupervised, substantially increasing the experimental throughput of our lab.