3D kinetic Monte Carlo-based investigation of the influence of dopants on the reliability of VCM ReRAM

Abstract

Variability and retention are critical reliability challenges for redox-based resistive switching random access memory (ReRAM) devices operating via the valence change mechanism (VCM). Variability arises from stochastic oxygen vacancy dynamics at short timescales, while retention failure manifests itself by the long-term instability of the programmed states. Both challenges become more severe in large-scale memory arrays, requiring a better understanding of the mechanisms and effective optimization strategies. In this work, we employ an advanced 3D Kinetic Monte Carlo simulation model to systematically investigate the influence of dopants on these reliability issues. For variability, we demonstrate that dopants significantly enhance the signal-to-noise ratio by stabilizing oxygen vacancy dynamics through a trapping effect, which reduces the frequency of current fluctuations. For retention, we show that dopants improve the stability of current distributions over time, where they effectively suppress distribution broadening and slow mean current degradation. Our simulation results provide valuable insights into optimizing the reliability of VCM ReRAM via dopants for practical applications. While these findings align with existing experimental data, the scope of experimental studies on doped VCM ReRAM devices remains limited, particularly concerning large-scale statistics and short-term variability. Therefore, our simulations not only corroborate previous experimental observations but also highlight the need for further experimental investigations to enhance the performance of VCM ReRAM devices for next-generation memory applications.