Atomic control of ionic processes in resistive memory devices

Regents Of The University Of Michigan - Ann Arbor, Ann Arbor MI

Investigators

Abstract

Large amounts of data are being generated constantly in every aspect of today's society, from manufacturing to social networks, from large organizations to individual consumers, and this trend is continuing at an accelerated pace. Better and faster memory technologies are needed to store and analyze the data quickly and efficiently. At the same time, traditional memory technologies are facing fundamental scaling limits and major practical constraints that make it ever more challenging to keep up with the demand. This project aims to develop a fundamentally new memory technology that can store data by changing the material's internal configuration at the atomic scale, leading to significantly improved device performance. The proposed program will significantly advance fundamental scientific research and enhance the understanding of interface control, ionic and electronic transport and hybrid device integration at the nanoscale. Results obtained from this NSF-funded program will also likely lead to successful technology transfers that can result in industry-leading memory products, and have a transformative impact on the crucial semiconductor industry sector. The tools, methods and techniques developed in the program can be applied to a wide range of nanoscale devices and systems to stimulate research in a broad range of areas. The project will in turn provide interdisciplinary training of graduate and undergraduate students, and draw broad participation of students of different levels and backgrounds in collaborative research and education. The proposed device is based on the concept of resistive random-access memory (RRAM), which has shown excellent scalability and several other performance metrics. However, the technology still faces fundamental challenges, including large device variations, high programming current and small on/off ratio, which prevent the device from practical applications. The proposed project aims to achieve atomic-level control of the fundamental ionic processes underlying the resistance changes in RRAM, and produce devices with significant performance improvements and new device functions. Specifically, the proposed device exploits atomically-thin graphene as an ion-blocking layer to regulate oxygen vacancy (VO) generation and migration through nanoscale openings in the graphene film. The filament formation will further be confined inside the switching layer to localized channels that are favorable for VO transport and storage through controlled doping. The experimental studies will be supported by theoretical calculations and modeling, including first-principles calculations that predict the VO formation energy and O-O distance, as well as numerical modeling that predicts the VO distribution and dynamic migration processes. The proposed approaches will eliminate excessive VOs in the switching layer and will significantly improve the on/off ratio, programming current, and device variability of RRAM devices. In the meantime, by leveraging unique properties of emerging materials with emerging devices, the proposed project advances the frontier of nanoscale device research and naturally promotes a tight integration with physics, materials research and device engineering.