Mixed Conduction in Polymeric Materials: Imparting Plasticity to Electrochemical Devices for Brain-Like Computing

Organic semiconductors have been traditionally developed for making low-cost and flexible transistors, solar cells and light-emitting diodes. In the last few years, emerging applications in health case and bioelectronics have been proposed. An interesting class of materials in this application area takes advantage of mixed ionic and electronic conduction in certain semiconducting polymers. In particular, mixed conductors can be used to emulate the compute-in-memory operation of the brain that is the key to its energy efficiency. This principle has been invoked as one of the keys to neuromorphic, or brain-like, computing. Indeed, the brain can perform massively parallel information processing while consuming only ~1- 100 fJ per synaptic event. I will describe a novel electrochemical neuromorphic device that switches at record-low energy (<0.1 fJ projected, <100 fJ measured), displays >500 distinct, non-volatile conductance states within a ~1 V operating range. The tunable resistance of the device behaves very linearly, allowing low write-noise and blind updates in a neural network when operated with the proper access device. Organic electrochemical synapses also display outstanding endurance achieving over 109 switching events with very little degradation. I will describe our recent efforts at scaling and materials selection, allowing us to reach 20 ns write pulses and operation at high temperature (up to 120°C) in an all-solid-state device.