Artificial nanofluidic memristors and Hodgkin-Huxley dynamics in two-dimensional graphene slits

New energy-efficient architectures inspired by the brain have been growing as an alternative to traditional von Neumann computing. Yet, existing hardware implementations use electrons as charge carriers, while neurons rely on transport of ions to carry out computations. We predict neuromorphic behaviour in a recently demonstrated two-dimensional electrolyte confined between graphite surfaces. We show that ions in the monolayer form tightly bound Bjerrum pairs that assemble into micelle-like clusters when an electric field is applied. Our model can be extended to the time-dependent case, where the slow dynamics of ionic assemblies induce memory effects in the system’s conductivity. These artificial ‘memristors’ can then be assembled to implement the Hodgkin-Huxley neuron model in a nanofluidic device capable of emitting voltage spikes trains. Our findings reveal a minimal, experimentally-accessible neuron architecture and pave the way for the development of ion-based computing and prototype ionic machines.

Reference:
P. Robin, N. Kavokine, L. Bocquet, “Artificial nanofluidic memristors and Hodgkin-Huxley dynamics in two-dimensional graphene slits”, submitted (2020).