JART ECM v1 varhttps://doi.org/10.26165/JUELICH-DATA/WHSSZAAhmad, Rana WaliedMenzel, StephanJülich DATA2025-08-152025-08-15T11:27:36ZA purely physics-based variability-aware compact model of electrochemical metallization memory (ECM) cells is presented. Since this extension consists of several different features allowing for a realistic variability-aware fit, it depicts a unique model comprising physics-based, stochastically and experimentally originating variabilities and reproduces them well. It is based on the deterministic ECM model JART ECM v1. The variability-aware model introduces device-to-device variability by choosing the model parameters from a physically reasonable value range. The cycle-to-cycle variability can be introduced by updating these parameters according to a random walk algorithm after a certain time step. Moreover, a stochastic feature is added to the gap evolution within the model’s main dynamics-determining differential equation. The model is validated by experimental data of Cu/SiO2/W ECM cells. This model can be used in higher-level circuit simulators like Spectre to design variability-aware application circuits. [1] shows (a) experimentally measured and (b) simulatively verified device-to-device variability for SET kinetics analysis. [2] shows experimentally measured I–V sweeps in red, simulated I–V sweeps in blue: (a) experimentally recorded I–V sweep, (b) simulated I-V sweep characteristics with all simulation modifications, (c) - (f) simulated I–V sweep characteristics showing individual types of simulation modifications.Computer and Information ScienceEngineeringPhysicsOthermemristive devices, variability-aware modeling, ReRAM, ECM, CBRAM, SPICE level, compact modelR. W. Ahmad et al. “Variability-Aware Modeling of Electrochemical Metallization Memory Cells”. In: Neuromorphic Computing and Engineering 4.3 (2024), doi, 10.1088/2634-4386/ad57e7, https://iopscience.iop.org/article/10.1088/2634-4386/ad57e7Ahmad, Rana Walied2025-07-27CC0CC0 Waiver