A5040367300- Advanced Memory and Neural Computing
- Neural Networks and Reservoir Computing
- Ferroelectric and Negative Capacitance Devices
- Semiconductor materials and devices
- Neural dynamics and brain function
- Machine Learning and ELM
- ZnO doping and properties
- Neural Networks and Applications
- Transition Metal Oxide Nanomaterials
- Ferroelectric and Piezoelectric Materials
- Machine Learning in Materials Science
University of Sheffield
2018-2025
An electrochemical device capable of manifesting reversible charge storage at the interface an active layer offers formidable advantages, such as low switching energy and long retention time, in realizing synaptic behavior for ultralow power neuromorphic systems. Contrary to a supercapacitor-based field-effect that is prone memory due fast discharge, solid electrolyte-gated ZnO thin-film exhibiting battery-controlled mechanism via mobile charges its with tantalum oxide demonstrated. Analysis...
Physical dynamic reservoirs are well-suited for edge systems, as they can efficiently process temporal input at a low training cost by utilizing the short-term memory of device in-memory computation. However, two-terminal memristor-based limits duration inputs, resulting in more reservoir outputs per sample classification. Additionally, forecasting requires multiple devices (20-25) prediction single time step, and long-term reintroduction forecasted data new input, increasing system...
Negative capacitance transistors are a unique class of switches capable operation beyond the Boltzmann limit to realize subthermionic switching. To date, negative effect has been predominantly attributed devices employing an unstable insulator with ferroelectric properties, exhibiting two-well energy landscape, in accordance Landau theory. The theory and solid electrolyte field transistor (SE-FET) subthreshold swing less than 60 mV/dec absence gate dielectric demonstrated this work. Unlike...
The processing of sequential and temporal data is essential to computer vision speech recognition, two the most common applications artificial intelligence (AI). Reservoir computing (RC) a branch AI that offers highly efficient framework for inputs at low training cost compared conventional Recurrent Neural Networks (RNNs). However, despite extensive effort, two-terminal memristor-based reservoirs have, until now, been implemented process by reading their conductance states only once, end...
With the vast amounts of real-time data today, efficient storage and processing are critical to temporal applications for making informed decisions. Physical memory devices provide effective solutions to...
We demonstrate a novel concept for low power compute-in-memory applications in room temperature fabricated ZnO/Ta 2 O 5 thin film transistor.By writing during the off-state, device consumption is reduced to nW despite large L/W ratio.A thinner gate insulator thickness gives higher on/off ratio, nevertheless this reduces retention time.The ratio thicker oxide can be improved by asymmetric voltage pulses of magnitude off state without affecting consumption.Benchmarked against other ReRAM...
The duration and interval of the input signal in two terminal memristor based reservoir systems is constrained by decay time a conventional memristor. Here, we demonstrate that third Solid Electrolyte <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\text{ZnO}/\text{Ta}_{2}\mathrm{O}_{5}$</tex> Thin-film Transistor (SE-FET) can be used to control using variable read voltage, without any additional circuit elements. Using this approach, have...
Abstract We establish that the phenomenon of transient negative capacitance, conventionally linked to delay in response a domain switching, ferroelectric material, and modelled by non-linear capacitor, can fact be considered more generally applicable any represented an RC-equivalent circuit. demonstrate conditions for sub-60 mV/dec switching RC-FET, even if R C were constant along both forward backward sweeps. For semiconductor charge Q ch , we show necessary condition (d )/(d Ψ s ) = ( q /(...
Implementation of accurate neural network models in edge applications such as wearables is limited by the hardware platform due to constraints power/area. We highlight novel concepts reservoir computing that rely on a volatile three terminal solid electrolyte thin film synaptic transistor, whose conductance can be controlled gate and drain voltages enhance richness operate off-state. The proposed approach achieves an accuracy 94% image processing, significantly higher than equivalent based...
In this work, we demonstrated a room temperature fabricated ZnO/Ta <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> O xmlns:xlink="http://www.w3.org/1999/xlink">5</sub> transistor for low power compute-in-memory application. By writing during the off-state, device programmed shows consumption in nW. using variable pulse amplitudes SET/RESET allows control of on/off ratio resistance states without affecting consumption. Benchmarked against...
The asynchronous supervised-learning ability of a Ta <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> O xmlns:xlink="http://www.w3.org/1999/xlink">5</sub> /ZnO synaptic thin film transistor, capable write operation in the off-state, is demonstrated. Poisson drift-diffusion model device enhanced to reproduce Faradaic type charge storage mechanism its gate current characteristics. Training as well implementation OR and AND operations 2x2...