Interrogating Electrochemical Doping in Organic Mixed Conductors Through a Robotic Platform Guided by Machine Learning

Conjugated polymer organic mixed ionic-electronic conductors (OMIECs) emerged recently as promising materials for bioelectronics, neuromorphic computing, and energy storage. OMIECs function in an electrochemical cell through a process called electrochemical doping. In electrochemical doping, an underlying electrode injects electronic charge carriers into the polymer and ions from solution enter the polymer to compensate for these electronic charges, making the polymer more conductive. The process of electrochemical doping depends on a large parameter space that includes the polymer chemistry, polymer processing conditions, polymer post-processing, and the properties of the electrolyte. To explore this parameter space, we developed a robotic system that fabricates, processes, and characterizes OMIEC thin films. Initially, we investigate electrochemical doping in poly(3-[2-[2-(2-methoxyethoxy)ethoxy]ethyl]thiophene) (P3MEEET) as a function of electrolyte properties. By measuring over 200 electrolytes and developing a model using supervised learning, we uncover new structure-property relationships that connect the chemical properties of the ion to the efficiency of electrochemical doping.