An Autonomous Electrochemical Test stand for Machine Learning Informed Electrolyte Optimization
Preprints are manuscripts made publicly available before they have been submitted for formal peer review and publication. They might contain new research findings or data. Preprints can be a draft or final version of an author's research but must not have been accepted for publication at the time of submission.
A fully automated, computer-controlled test stand capable of rapidly creating and electrochemically characterizing any arbitrary liquid electrolyte solution is described. Hundreds of different electrolytes were studied, and the results were used to verify the precision and accuracy of the system. To test the functionality of the approach, several 2-dimensional co-solvated electrolyte solutions containing blends of aqueous sulfates and nitrates were rapidly created and examined automatically. The test stand took less than a day to conduct these searches, while conventional manual methods would have taken much longer. The demonstrated standard error of the test-stand was 0.5 mS/cm on conductivity and 0.02 V for voltage stability window measurements, and several of the combinations studied revealing surprisingly high voltage stability and conductivity values. The demonstrated success of the test-stand in a 2-dimensional search spaces shows the promise of conducting high speed co-optimization studies of liquid electrolytes in particular when used in concert with a machine learning-based real time/in-loop data assessment computational package.
Read the published paper
in Journal of The Electrochemical Society