Optimizing the performance of organic electrochemical transistors (OECTs) through high-throughput electrochemistry guided by machine learning

Organic electrochemical transistors (OECTs) have recently gained attention for a diverse range of applications, including bioelectronics, neuromorphic computing, and biosensing. Their performance is strongly influenced by the properties of the active layer, which typically consists of a conjugated polymer organic mixed ionic-electronic conductor (OMIEC). Factors such as polymer chemical structure, processing conditions, and electrolyte selection all play critical roles in determining device performance. Because of the large number of variables, finding optimal OECT configurations can be challenging. To address this, we have developed an automated robotic system—guided by machine learning—that fabricates and characterizes OMIEC thin films in a high-throughput and reproducible manner. By systematically investigating how molecular ion identity affects OECT performance, we are able to predict and discover new high-performance ions. Overall, our work reveals which chemical properties of electrolytes are most important for realizing OECTs with superior performance.