A Comprehensive Oxide-Based ReRAM TCAD Model with Experimental Verification

“©2021 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.”

Original IEEE publication: W. Goes et al., “A Comprehensive Oxide-Based ReRAM TCAD Model with Experimental Verification,” 2021 IEEE International Memory Workshop (IMW), 2021, pp. 1-4, doi: 10.1109/IMW51353.2021.9439623.

Abstract —During the last few years, oxide-based ReRAM technology has attracted intense industrial and scientific research interest. Therefore, we have performed an indepth computational study with a focus on data retention besides the resistive switching and the current run-away. Our newly developed comprehensive TCAD (Technology Computer Aided Design) model provides deep insight into the underlying microscopic processes and is validated against experimental data as an accurate and predictive simulation tool.

INTRODUCTION

Oxide-based ReRAM is an emerging memory technology that combines the advantages of both RAM and Flash technologies [1-4]. With the increasing demand for big data, artificial intelligence, and neuromorphic computing, ReRAM will play an important role in meeting future memory requirements. For optimizing this technology, predictive and physics-based TCAD simulations are vital as they can dramatically speed up the design, fabrication, and commercial use of new microelectronic technologies through the elimination of expensive and time-consuming experimental test wafers during technology adoption. So far, a few research groups have already undertaken interesting modeling attempts [4-8], which provided first microscopic insight behind the resistive switching. However, their fundamental works focused on the role of oxygen vacancies and interstitials during the SET, RESET, and forming operations alone [4-7]. In this work, we extend the ReRAM investigations to data retention while accounting for the frequently neglected impact of the current compliance on the device characteristics. For this purpose, we make use of the versatile applicability of the generic chemistry module in Silvaco’s Victory TCAD package [9] and validated a new ReRAM model against experimental measurements supplied by Weebit-nano.