Exploiting the switching dynamics of HfO2-based ReRAM devices for reliable analog memristive behavior

Abstract

The utilization of bipolar-type memristive devices for the realization of synaptic connectivity in neural networks strongly depends on the ability of the devices for analog conductance modulation under application of electrical stimuli in the form of identical voltage pulses. Typically, filamentary valence change mechanism (VCM)-type devices show an abrupt SET and a gradual RESET switching behavior. Thus, it is challenging to achieve an analog conductance modulation during SET and RESET. Here, we show that analog as well as binary conductance modulation can be achieved in a Pt/HfO2/TiOx/Ti VCM cell by varying the operation conditions. By analyzing the switching dynamics over many orders of magnitude and comparing to a fully dynamic switching model, the origin of the two different switching modes is revealed. SET and RESET transition show a two-step switching process: a fast conductance change succeeds a slow conductance change. While the time for the fast conductance change, the transition time, turns out to be state-independent for a specific voltage, the time for the slow conductance change, the delay time, is highly state-dependent. Analog switching can be achieved if the pulse time is a fraction of the transition time. If the pulse time is larger than the transition time, the switching becomes probabilistic and binary. Considering the effect of the device state on the delay time in addition, a procedure is proposed to find the ideal operation conditions for analog switching.

Memristive State Equation for Bipolar Resistive Switching Devices Based on a Dynamic Balance Model and its Equivalent Circuit Representation

“…which coincides with the original expression (2) corresponding to the voltage-acceleration law for metal ions/oxygen vacancies displacement [19][20][21][22][23]. and V0=1/S [V] in (4) are constants of the model and can be found experimentally [24].…”

Variability-Aware Modeling of Filamentary Oxide-Based Bipolar Resistive Switching Cells Using SPICE Level Compact Models

“…Simulations using the deterministic version of the JART VCM v1b model were shown in [32]. Since the exact changes were only indicated in [32], this work is used to describe the deterministic version from [32] as well as the variability version. The JART VCM v1b model therefore exists as a deterministic model and as a variability model.…”

“…If the deterministic version is to be used, all parameters get the index 'det' instead of the index 'var' in section III-A. Simulations with the deterministic parameter set can be found in [32]. The higher level of complexity in the switching modeling is represented by a variability model.…”

“…In [32] we showed a deterministic model, the Jülich Aachen Resistive Switching Tools (JART) VCM v1b model that describes the SET and RESET kinetic behavior with a special focus on the initial states before the SET and the RESET. For [32] the equations have been updated and modified and the parameters have been adjusted to better match the dynamics of SET and RESET processes as well as the I-V characteristic with the high resistive state (HRS) and the low resistive state (LRS), compared to JART VCM v1, which was presented in [33].…”

Transport mechanism of copper sulfide embedded carbon nitride thin films: a formation free memristor

“…Nonvolatile resistive memory has attracted significant attention for next-generation data-storage applications 1,2 due to faster device performance, data nonvolatility and low power consumption. 3,4 A resistive memory device utilizes the bistable resistances of the high-resistance state and low-resistance state of materials. 5 The simple architectural design also makes a resistive memory device more attractive over other memory candidates, such as ferroelectric and magnetoresistive memory devices.…”