About the project
This project will use device simulations combined with transient electronic and optoelectronic characterisation techniques to unambiguously identify the resistive switching mechanisms underlying memristors based on a wide variety of materials systems including perovskites and titanium-based compounds.
A memristor is a fundamental two-terminal passive electronic device, the resistive state of which depends on its history. Interconnecting an array of memristors produces a hardware neural network capable of in-memory computing with the potential to realise neuromorphic chips with improved energy efficiency, real-time learning and greater fault tolerance as compared to the current silicon-based Artificial Intelligence accelerators. Despite this, many challenges remain for the research field, including understanding the various resistive switching mechanisms, device stability, and state retention.
In this project, you will use device simulations combined with transient electronic and optoelectronic characterisation techniques to identify the resistive switching mechanisms underlying memristive devices, based on a wide variety of materials systems including hybrid perovskites, silicon-carbide, and titanium-based compounds. You will be encouraged to take ownership of the project while ensuring that the research remains within its defined scope.
An existing mixed ionic-electronic device simulation tool (Driftfusion) will be applied and adapted to account for a variety of memristor resistive switching mechanisms including filament formation, electronic charge trapping, phase change, and Joule heating. The simulations will facilitate the design of new transient electronic and optoelectronic characterisation protocols, which will be tested on real-world devices. The overall aim of the project is to develop and apply a set of measurement protocols that can unambiguously determine the resistive switching mechanism of any given memristive device.
