Harnessing Ion Tuning Mechanisms for Neuromorphic Computing: From Artificial Synapses to Dynamically Reconfigurable Architectures

Non-von Neumann computing using neuromorphic systems based on analog synaptic and neuronal elements has emerged as a powerful approach, but its full potential has not been realized due to the lack of materials and devices with the appropriate attributes. Recently, solid state electrochemical ion-insertion, also known as electrochemical random access memory (ECRAM) has emerged as a promising approach realize the needed device characteristics. Unlike memristors which require large write currents to drive phase transformations or filament growth, every electron transferred through the external circuit in an ion tunable element corresponds to the migration of ∼1 ion, which is used to store analogue information. Like static dopants in traditional semiconductors, electrochemically inserted ions increase or decrease the conductivity by locally perturbing a host’s electronic structure but do so in a dynamic and reversible manner. The resulting change in conductance can span orders of magnitude, from gradual increments needed for artificial synapses, to large, abrupt changes required for artificial neurons. In my presentation, I will briefly cover the history of ECRAM, the recent progress in devices spanning various types of materials, circuits, architectures, and discuss the rich portfolio of challenging, fundamental questions and how we can harness ECRAM to realize a new paradigm for low power neuromorphic computing.