Atomistic Insights on the Full Operation Cycle of a HfO₂-Based Resistive Random Access Memory Cell from Molecular Dynamics

Abstract

We characterize the atomic processes that underlie forming, reset, and set in HfO 2 -based resistive random access memory (RRAM) cells through molecular dynamics (MD) simulations, using an extended charge equilibration method to describe external electric fields. By tracking the migration of oxygen ions and the change in coordination of Hf atoms in the dielectric, we characterize the formation and dissolution of conductive filaments (CFs) during the operation of the device with atomic detail. Simulations of the forming process show that the CFs form through an oxygen exchange mechanism, induced by a cascade of oxygen displacements from the oxide to the active electrode, as opposed to aggregation of pre-existing oxygen vacancies. However, the filament breakup is dominated by lateral, rather than vertical (along the filament), motion of vacancies. In addition, depending on the temperature of the system, the reset can be achieved through a redox effect (bipolar switch), where oxygen diffusion is governed by the applied bias, or by a thermochemical process (unipolar switch), where the diffusion is driven by temperature. Unlike forming and similar to reset, the set process involves lateral oxygen atoms as well. This is driven by field localization associated with conductive paths.

Thermodynamic and Transport Properties of LiF and FLiBe Molten Salts with Deep Learning Potentials

“…34 Regardless, we obtain agreeable results for the thermal and electronic conductivity in the systems studied here, possibly due to the strongly ionic, well-defined characteristic of the ions in the melt. Future works on MD simulations of molten salts including fixed charges, dynamically altered partial charges, such as those found in the modeling of HfO 2 RRAM cells using the ReaxFF potential, 40 or MLP trained with partial charges from charge equilibration method 34 combined with long-ranged solvers for Coulombic interactions should be considered. 36 MD with the RIM, 33 fitted line from MD with polarized ion model, 24 experimental results, 41,42 and by a predictive model.…”

An Atomistic Model of Field-Induced Resistive Switching in Valence Change Memory

“…50 The melt-and-quench step is performed using the LAMMPS MD code 51 with a ReaxFF force field for HfO 2 . 26 A subsequent relaxation step is performed at the DFT level (further details are in the Methods section). It serves to remove any coordination defects and ensure that the atomic structure is consistent with the DFT calculations later on.…”

“…They identify a dependence of activation energies on the local defect environment, electric field, grain boundaries, and interface environment, , as well as the energy signature of different defects . Simulations at the structural level explore possible switching dynamics using molecular dynamics (MD) − or kinetic Monte Carlo (KMC) methods. − For many oxide stacks, they generally recover the anticipated switching mechanism in which ion exchanges at an interface coupled with the generation of a percolating path of oxygen vacancies lead to an increase in electrical conductance. Finally, transport simulations have revealed a high sensitivity of the current to the thickness of the conductive filament, down to a single line of atoms .…”

Revealing Resistive Switching Mechanism in CaFeOx Perovskite System with Electroforming‐Free and Reset Voltage‐Controlled Multilevel Resistance Characteristics

“…According to a previous study, the Ti electrode could serve as an oxygen reservoir layer, [40,41] creating a substoichiometric region at the interface that improves the performance of the RRAM device. Therefore, after we characterized the as-deposited CFO thin film, a Ti top electrode was deposited on the CFO film via e-gun evaporation, and the diameter of the Ti electrode was ≈80 µm.…”