Please use this identifier to cite or link to this item:
Title: Halide perovskite memristors as artificial synapses
Authors: Neoh, Eng Tiong
Keywords: Engineering::Materials::Microelectronics and semiconductor materials
Issue Date: 2022
Publisher: Nanyang Technological University
Source: Neoh, E. T. (2022). Halide perovskite memristors as artificial synapses. Final Year Project (FYP), Nanyang Technological University, Singapore.
Abstract: With recent advancements in computer science, and booming interest in the development of artificial intelligence, there is an increasing demand for more efficient computing systems. The von-Neumann architecture (embodiment of modern digital processing) is inherently inefficient, especially during the handling of data-intensive tasks. Brain-inspired analog computing elements have re-emerged as we approach the limitations in digital processing. Neuromorphic computing and specifically physical neural networks had been touted as a potential alternative for the modern computer architecture and has garnered significant attention in recent years. In order to realize such hardware neural networks, there is a need to develop analog memory devices known as artificial synapses. These devices are inspired by the synaptic connections extensively interconnecting the neuron cells in the brain. The strengthening and weakening of such connections are what govern the memory and processing in the brain. The memristor is a nanoelectronic device with a variable conductance, allowing it to store information as conductance states. This conductance modulation is analogous (conductive state represents strong connection, resistive state represents weak connection) to the biological synapse. The memristor device is hence studied as a candidate of artificial synapses to emulate the synaptic connections in the brain. While recent development in memristor technology has made significant advancement in the field, most memristor designs encounter issues regarding volatility in data storage and physical device stability. The retention of data is a critical parameter that enables long-term memory storage and capability to learn. The loss of data retention can be attributed to the back-diffusion of ions and the lack of stable redox changes in the system. In this study, we propose a novel design for memristors based on the tunability of charge transfer properties of bipyridinium compounds using viologen-templated perovskites. As opposed to conventional memristors, the reduction of viologen moieties is expected to produce high conductivity changes, stable conductance states and interestingly optically active changes. Here, we assessed the electrochemical properties of the viologen-templated perovskite using voltammetry methods for the first time. We then leveraged the quasi-reversible redox changes in the viologen-templated perovskite to fabricate a redox-stabilized memristor device. Non-volatile conductance changes are then demonstrated with direct current I-V and pulsed electrical measurement. In order to understand the modulation in electronic structure of the viologen-templated perovskite, optical characterization methods such as UV-Vis and Photoluminescence spectroscopy were carried out. Lastly, in order to develop a better understanding of the dynamic redox processes in the device, we have developed a lateral device design to facilitate in-situ testing in the future. By stabilizing the conductance changes via the reduced moiety, the proposed redox-stabilized memristor using viologen-templated perovskites offers significant advantages in memristor performance.
Schools: School of Materials Science and Engineering 
Fulltext Permission: restricted
Fulltext Availability: With Fulltext
Appears in Collections:MSE Student Reports (FYP/IA/PA/PI)

Files in This Item:
File Description SizeFormat 
FYP Draft vFinal_Neoh Eng Tiong.pdf
  Restricted Access
1.51 MBAdobe PDFView/Open

Page view(s)

Updated on May 30, 2023

Google ScholarTM


Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.