Physical Drivers of Adsorption Thermodynamics at 2D Interfaces for Resistive Switching in Atomristors

Two-dimensional (2D) materials with engineered defects are emerging as exciting candidates for next-generation memristors. In these "atomristors," resistive switching (RS) -- the ability to toggle between high and low resistance -- occurs through atomic-scale processes at the interface between the electrode and the 2D layer. Experiments suggest [1] that adsorption and desorption of electrode metal atoms at vacancies in the 2D lattice play a key role in this switching. In this work, we combine hybrid density functional theory with many-body dispersion corrections and AI (Sure Independence Screening and Sparsifying Operator [2]) to systematically study metal adsorption on transition metal dichalcogenides at gold interfaces. SISSO provides a predictive model for adsorption energies and uncovers the fundamental physical factors driving RS. Our sensitivity analysis shows how orbital hybridization and bonding interactions dominate the behavior, offering simple heuristics for understanding the atomic processes. We also create a comprehensive materials property map to guide the design of future atomristors -- and highlight regions that align with experimentally reported devices. Our approach bridges the gap between fundamental material properties and the emergent functionality of RS, providing a physically interpretable framework for designing the next generation of 2D memristive devices.

1) Ge et. al, Adv. Mat. 33 2007792 (2021)

2) R. Ouyang et. al, Phys. Rev. Materials 2, 083802 (2018)