Strong radial electric field scaling near nanoscale conductive filaments and the ReRAM resistive switching mechanism

TL;DR

The paper addresses the unresolved reset mechanism in bipolar filamentary ReRAM by proposing a nanoscale surface-charge–driven radial electric-field effect that scales inversely with filament radius. It derives analytical expressions for radial and axial fields in a cylindrically symmetric filament and validates them with finite-element simulations, showing that radial fields can reach the order of $10^5$ to $10^6$ V/cm at modest bias for sub-10 nm filaments. The results reconcile long-standing experimental discrepancies by demonstrating that radial transport and filament rupture can be driven by radial electrostatics rather than solely diffusion, offering a potentially universal reset mechanism across materials. This size-dependent effect has implications for scaling, retention at elevated temperatures, and the design of reliable filamentary ReRAM devices.

Abstract

The physics underlying reset in bipolar resistive memory has been the subject of decades of controversy and has been identified as the primary barrier to resistive memory technology development. This manuscript introduces a nanoscale effect in current carrying conductors, whereby surface charge induced radial electric fields are found to be inversely proportional to the radius of the conductive path. This nanoscale effect is then applied to explain the negative resistance switching (reset) mechanism in filamentary metal oxide resistive switching memory devices (memristors). Previous explanations for the negative resistive switching mechanism state that diffusion constitutes the radial driving mechanism for oxygen ions, and drift under electric fields is restricted to the direction parallel to current flow. This explanation conflicts with retention and microscopy data collected in a subset of devices presented in literature. We demonstrate that the electric field's dependency on the on the radius of a nanoscale conductive path can result in radial fields on the order of 10^5 to 10^6 V/cm at only -1 V bias, sufficient to govern the negative resistance switching mechanism in filamentary metal oxides. By accounting for this nanoscale size effect, long standing anomalous experimental data about the negative (reset) resistance switching mechanism in bipolar filamentary resistive memory devices is finally reconciled. Wide understanding of surface charges and associated electric fields in nanoscale conductive paths could prove important for further scaling of integrated circuits and aid in elucidating many nanoscale phenomena.

Figures (5)

• Figure 1: $TaO_x$ bipolar Valence Change Mechanism (VCM) ReRAM electrical characteristics. [Top] A typical $TaO_x$ device is composed of an active Ta electrode Top Electrode (TE), a reduced $TaO_x$ layer, and an inert TiN bottom electrode (BE). The device can be treated as being cylindrically symmetric around the filament. [Bottom] set-reset characteristics of the ReRAM device; Positive bias applied to the top electrode (set operation) brings the device to the Low Resistance State (LRS), negative bias applied to the top electrode (reset operation) brings the device to the high resistance state. The set operation is well understood, but the reset operation is still the subject of much controversy.
• Figure 2: [Top] Three diagrams of a $TaO_x$ ReRAM cell during reset with the direction of oxygen ion drift/diffusion indicated according to currently established theory. The center of the filament is located on the left-hand side of each image and the figure is radially symmetric around the filament. SLAC X-ray experiments have shown a ring of excess oxygen which indicated a radial reset switching mechanism invalidating earlier electric field-only switching mechanisms. Currently, the only drift/diffusion vector with the correct direction to explain the reset mechanism is thought to be Fick diffusion. (Bottom) Three diagrams of a $TaO_x$ ReRAM cell during reset with the direction of oxygen ion drift/diffusion indicated according the theory presented in this manuscript. Recent experiments have shown that in $TaO_x-TaO_y$ the direction of the chemical potential is reversed from the concentration gradient, so there is no diffusive mechanism that can explain reset in this subset of devices. TEM experiments have shown that the filament consistently breaks near the Bottom Electrode (BE), but there is not a simple explanation for this behavior in established theory.
• Figure 3: The boundary conditions for a simplified resistive memory cell with a cylindrical filament in a metal insulator metal (MIM) capacitor structure with negative bias applied to the top electrode and ground on the bottom electrode is shown above. The ground path is assumed to be a V=0, at a distance b from the filament in the r direction, which is $>>$ the radius of the top and bottom electrodes. The radius of the filament is R and is aligned with the R direction. This system is fully cylindrically symmetric in $\theta$.
• Figure 4: Analytical solution for the radial component of the electric field near a 1 nm diameter (left) and 5nm diameter (right) conductive path, with a distance to the return path b=500 nm, 1 mA current on the filament, and a filament height of 10 nm. The radial component of the electric field is significantly stronger for thinner filaments.
• Figure 5: Finite element model of radial electric field for a cylindrical tantalum filament of 1 nm and 5 nm respectively under application of -1V bias to the top of the cylinder and ground applied to the bottom of the cylinder. Black arrows show the electric field direction and their length is proportional to magnitude. Radial electric fields reach 5.49 MV/cm near the 1 nm filament surface which is stronger than the dielectric strength of $Ta_2O_5$, and radial fields reach 4.05 MV/cm for the 5 nm filament. These fields may be enhanced above the analytical expressions due to the sharp corners in the model. The vector direction of the radial field is critical, as negatively charged oxygen ions move towards the bottom of the filament where it has been shown to rupture in experiment.