A Leaky Integrate-and-Fire Neuron Based on Hexagonal Boron Nitride (h-BN) Monocrystalline Memristor

Abstract

As a competitive candidate for artificial neurons, memristors have become the focus of intense research owing to their intrinsic ion migration tunability, enabling an authentic implementation of biomimicry. However, they still suffer from variability issues due to 3-D uncontrollable filament dynamics in an amorphous medium and modeling of switching dynamics underlying filament growth and rupture is still under investigation. In this work, we present volatile memristors that exhibit desired characteristics for neuromorphic computing with low performance variations utilizing a hexagonal boron nitride (h-BN) monocrystalline as a switching medium. Theoretical investigations assisted by the Monte Carlo simulation combined with experimentally detected <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${I}$ </tex-math></inline-formula> – <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${V}$ </tex-math></inline-formula> characteristics described that the electric field dominates the set process, whereas the Gibbs–Thomson interfacial energy minimization and heat dissipation influence the relaxation process mostly. Additionally, h-BN memristors with high switching uniformity provide an ideal hardware platform for credible neuron emulation and software identification of digital images.