DOCTORAL CANDIDATE
Anjana Thomas

SUPERVISORS
Bertrand Vilquin, École Centrale de Lyon (EC Lyon)
Prof. Sharath Sriram, RMIT University (RMIT)

Neuromorphics, Thin Films, Renewable Energy, Material Science, Catalysis

Modern computing systems, based on the Von Neumann architecture, are facing significant limitations due to the separation of processing and memory, leading to the “Memory Bottleneck.” To overcome these challenges, this project aims to develop neuromorphic networks inspired by the human brain, where information storage (synapses) and processing units (neurons) are closely integrated for parallel and energy-efficient processing.

The focus of this project is to create artificial synapses using ferroelectric hafnium zirconium oxide (HfZrO2), a material compatible with CMOS technology that shows promise for use in neuromorphic devices. The project seeks to optimise the deposition conditions of HfZrO2, including factors like thickness, stoichiometry, and interface properties, all within a low thermal budget (<450°C) to meet the requirements of semiconductor manufacturing.

Key goals include developing CMOS-compatible artificial synapses based on HfZrO2, identifying optimal deposition parameters to enhance ferroelectric properties, and fabricating integrated synaptic matrices of the “crossbar” type. These advancements are crucial for implementing dense networks of artificial synapses that can emulate the brain’s synaptic plasticity, essential for learning and memory functions.

This project will be realised in the framework of common labs between INL and CEA-LETI and STMicroelectronics. This project aims to pave the way for the next generation of efficient, brain-inspired computing systems by successfully developing these ferroelectric synapses.