Transport in resistive switching devices

Abstract

Resistive switching (RS) devices are a critical component for the realisation of computational systems that mimic biological neural systems. In biological systems, connection weights between the neurons is dependent on the time lapse between the neuron s action potentials. RS devices (memristors) with multiple conductance states can mimic this biological Spike Time Dependent Plasticity (STDP) behaviour, with multiple conductance states being analogous to different synaptic weights. Additionally, memristors are one of the contenders as storage elements in non-volatile memories. The change in conductance in memristors can be, broadly, attributed to vacancy migration inside the switching layer (Valence Change Memory (VCM)), or to the formation of metallic filaments inside the switching layer (Electrochemical Metallization (ECM)). newlineElectrochemical Metallization (ECM)-based memristors with vertical transport and newlinesmall inter-electrode distances have been reported recently. Their device characteristics exhibit multiple conductance states with relatively low switching voltages, which make them well-suited for low power neuromorphic applications. Our work models the transport in these memristors, with a focus on explaining and capturing their CurrentVoltage (I-V) characteristics. The physics-based model also captures the dynamics of switching (with an emphasis on the estimation of switching energies and delays), and explains the experimentally observed STDP behaviour in these devices. We have proposed models for filament growth and dissolution along one dimension (axial), and newlinealong two dimensions (axial and radial). The simulation results obtained using our model (and implemented in Verilog-A) have been validated with experimental data from multiple sources. Our work demonstrates the flexibility of including different newlinetransport mechanisms (such as, tunneling, space charge limited conduction) in a unified newlineframework.