Resistance noise spectroscopy and autocorrelation analysis to study resistive switching mechanism

Resistive switching (RS) holds significant potential for next-generation memory devices. Despite considerable progress, these devices remain far from mass production due to the stochastic nature of switching. Both volatile and non-volatile RS exhibit inconsistent behavior, attributed to complex and poorly understood internal processes in materials, leading to challenges in achieving reliable device performance. In this work, we employ low-frequency resistance noise spectroscopy to gain insights into physical processes in material systems exhibiting resistive switching behavior. In NbO₂, dual RS states are observed, and time-domain autocorrelation analysis reveals two distinct mechanisms: the first RS is attributed to inhomogeneous conduction. In contrast, the second RS is linked to an insulator-to-metal transition. Similarly, in ZIF-8, noise spectroscopy characterizes multi-state behavior, supporting its application in multi-bit memory storage. In vanadium oxide bronzes, a multi-order power spectral density (PSD) change is ascribed to a polaron hopping mechanism. Time-domain autocorrelation plots and PSD emerge as powerful, non-destructive tools for elucidating the mechanisms behind RS in various materials, including transition metal oxides and metal-organic frameworks, offering deeper insights critical for advancing memory technologies.