Thermal mapping of current density in filamentary switching devices

Carnegie Mellon University, Pittsburgh PA

Investigators

Abstract

Nontechnical description: Storing the information in semiconductor memory devices and transferring it to and from the processing unit are the two major roadblocks in continued increase of information processing speed and efficiency. One of the most promising ways to address these roadblocks is to employ memristors i.e., resistors with memory. It is generally agreed that resistance change is caused by redistribution of ions in the device material. This project is focused on answering fundamental questions concerning the driving forces of ion motion, the speed of motion, and the temperature necessary for it. A secondary goal of the project is the development of experimental technique to measure the temperature distribution in nanoscale devices under bias. The project provides opportunities for training of undergraduate and graduate students in areas of processing of thin films and electronic devices and contributes to education of the workforce in an important area of the national economy. Outreach activities target high school students to encourage them to pursue careers in semiconductor technology. Technical description: The project targets identification of the operating mechanisms in two types of resistive switching structures: threshold and memory switches. Both device types form a small diameter conducting filament that is either volatile or non-volatile. Two models have been proposed for formation of the filament in resistive switches (i) based on lateral motion induced by thermophoresis and (ii) vertical motion and exchange of oxygen ions with the anode. The proposed mechanisms for threshold switches include carrier injection, local heating of the material, creating of hot carriers, and nucleation of a secondary metallic phase. The Carnegie Mellon team is going to use Scanning Thermal Microscopy and Scanning Joule Expansion Microscopy to measure the temperature distribution in structures with applied voltage, extract the filament size, and its changes with biasing conditions and map the local changes of composition by High Angle Annular Dark Field and X-Ray Energy Dispersive Spectroscopy. Since the temperature distribution is the distinguishing characteristics of different models, the results will identify the operating mechanism. The final goal of the project is building a predictive and quantitative model of processes involved the resistance change. Results are expected to form fundamental underpinnings of the important new electronic device technology that potentially can revolutionize the field of non-volatile high-density memories. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.