Publications

Abstract not provided.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Abstract not provided.

This report discusses aspects of neuromorphic computing and how it is used to model microsystems.

A major class of resistive memory devices is based on transition metal oxides, where mobile oxygen vacancies allow these devices to exhibit multiple resistance states. Ta2O5 based devices in particular have recently demonstrated impressive endurance and forming-free results. Deposition of substoichiometric Ta2Ox (x < 5) films is a critical process in order to produce the required oxygen vacancies in these devices. This paper describes a physical vapor deposition (PVD) reactive sputtering process to deposit substoichiometric Ta2Ox films. The desired film stoichiometry is achieved by feedback control of the oxygen partial pressure in the PVD chamber. A calibration procedure based on Rutherford backscattering spectroscopy is described for locating the optimum oxygen partial pressure.

Resistive random access memory (ReRAM), or memristors, may be capable of significantly improve the efficiency of neuromorphic computing, when used as a central component of an analog hardware accelerator. However, the significant electrical variation within a device and between devices degrades the maximum efficiency and accuracy which can be achieved by a ReRAMbased neuromorphic accelerator. In this report, the electrical variability is characterized, with a particular focus on that which is due to fundamental, intrinsic factors. Analytical and ab initio models are presented which offer some insight into the factors responsible for this variability.

Abstract not provided.