Experimental Investigation of Programmed State Stability in OxRAM Resistive Memories

TL;DR

This paper experimentally investigates the stability of programmed states in OxRAM resistive memories, revealing intrinsic instability mechanisms affecting data retention at low programming currents.

Contribution

It provides a detailed analysis of resistance drift and stochastic fluctuations in OxRAM, demonstrating their intrinsic nature and impact on memory stability.

Findings

Resistance evolution is driven by deterministic drift and stochastic fluctuations.

Intrinsic instability phenomena are independent of processing variations.

Verify algorithms are ineffective at low current due to inherent state instability.

Abstract

Oxide-based Random Access Memory (OxRAM), is part of the larger family of Resistive RAM (RRAM) memories. Generally OxRAM cells consist of a transition metal oxide (typically HfO2, Ta2O5, TiO2) sandwiched between two metal electrodes (typically TiN, TaN, W but also Pt, Ir). This thesis describes the experimental investigation performed on OxRAM memories during a 6-months internship project at imec research institute (Leuven, Belgium). From previous studies, HfO-based OxRAM memories showed good endurance properties (with more than 10e9 write cycles), low operating current (as low as 10 uA), and very fast switching (programming pulses of 100 ns). However, it has been observed that, under low programming current condition, data stability becomes challenging. Therefore, this programming state stability needed further investigation. This thesis addresses the problem of state stability, first, by presenting the equipment and algorithms used to study OxRAM state loss over short time intervals. Secondly, various experiments and tests are performed searching for a key parameter in order to control memory retention. Finally, this work demonstrates that resistance evolution over time is affected by both a determinist drift (logarithmically dependent on time) and a random stochastic fluctuation component. It also demonstrates that this behavior is intrinsic in the nature of the device itself and does not depend on processing issues or variation in the programming conditions. This experimental study proved the intrinsic nature of the program instability phenomena in OxRAM devices, which in turns leads to the ineffectiveness of verify algorithm for low current operation.