A5003630133- Advanced Memory and Neural Computing
- Ferroelectric and Negative Capacitance Devices
- Neural Networks and Reservoir Computing
- Neuroscience and Neural Engineering
- Photoreceptor and optogenetics research
- Advancements in Semiconductor Devices and Circuit Design
- Transition Metal Oxide Nanomaterials
- Semiconductor materials and devices
- Neural Networks and Applications
- ZnO doping and properties
- Semiconductor materials and interfaces
- 2D Materials and Applications
- Thin-Film Transistor Technologies
- Metal and Thin Film Mechanics
- Conducting polymers and applications
- CCD and CMOS Imaging Sensors
The University of Tokyo
2023-2025
Tokyo University of Information Sciences
2023-2024
Kwangwoon University
2020-2021
This study evaluated the memristive switching characteristics of a biomaterial solid polymer electrolyte (SPE) chitosan-based memristor and confirmed its artificial synaptic behavior with analog switching. Despite potential advantages organic memristors for high-end electronics, unstable multilevel states poor reliability devices must be overcome. The fabricated Ti/SPE-chitosan/Pt-structured has stable bipolar resistive (BRS) due to cation-based electrochemical reaction between polymeric...
This study proposes a hybrid electric double layer (EDL) with complementary metal-oxide semiconductor (CMOS) process compatibility by stacking chitosan electrolyte and Ta2O5 high-k dielectric thin film. Bio-inspired synaptic transistors excellent electrical stability were fabricated using the proposed EDL for gate layer. The high chemical resistance, thermal stability, mechanical strength enables CMOS-compatible patterning processes on biocompatible organic polymer electrolytes. technique...
In this study, we implemented a high-performance two-terminal memristor device with metal/insulator/metal (MIM) structure using solution-derived In-Ga-Zn-Oxide (IGZO)-based nanocomposite as resistive switching (RS) layer. order to secure stable memristive characteristics, IGZO:N nanocomposites were synthesized through the microwave-assisted nitridation of IGZO thin films, and resulting improvement in synaptic characteristics was systematically evaluated. The films clearly demonstrated by...
Abstract We experimentally demonstrate the anti-ferroelectric (AFE) behavior of a Hf 1-x Zr x O 2 (HZO)/Si FET and its potential for high-endurance nonvolatile memory operation. The AFE-HZO with content 75 % exhibits double polarization switching half-loop hysteresis under bipolar unipolar bias conditions, respectively. counterclockwise in transfer I d - V g characteristics is demonstrated sweep through AFeFET. steep subthreshold swing values were observed both forward backward sweeps curves...
We have experimentally demonstrated the physical reservoir computing by employing polarization switching current dynamics of an Hf1−xZrxO2 (HZO)-based metal/ferroelectric/metal capacitor with Zr content x = 0, 0.5, and 0.75. The spatial distribution crystalline phase HZO film reveals that tetragonal is a dominant crystal structure in [Zr] 75%, resulting anti-ferroelectric (AFE)-like double switching. Analyses using t-distributed stochastic neighbor embedding (t-SNE) find AFE-HZO effectively...
Abstract We have investigated how the parameters of an input gate voltage ( V g ) waveform and a drain d impact performance reservoir computing (RC) using Hf 0.5 Zr O 2 ferroelectric FET (FeFET). The RC is maximized by high swing amplitude most symmetrical polarization switching condition in triangular-shaped waveform, obtained center V, because enhanced FeFETs. Regarding dependence, amount current trade-off relationship. As result, moderate 1.0 becomes optimum terms difference responses...
In this study, we evaluated the improved memristive switching characteristics of hydrogen silsesquioxane (HSQ) nanocomposites embedded with a single-walled carbon nanotube (SWCNT) random network. A low-temperature solution process was implemented using flexible memristor device on polyethylene naphthalate (PEN) substrate. The difference in resistive (RS) behavior due to presence SWCNT network analyzed by current transport mechanism. Such not only improves RS operation but also facilitates...
In this study, we propose a polysilicon (poly-Si) channel thin-film synaptic transistor based on an organic-inorganic hybrid electric-double-layer (EDL) structured gate dielectric. To implement the EDLs, which play key role in artificial transistors, protonic mobile ion-based bio-inspired chitosan electrolyte and high-k 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> dielectric were...
Herein, a rare metal‐free oxide‐based semiconductor channel layer is constructed with excellent electrical properties and stabilities by fabricating zinc–tin oxide (ZTO) thin film transistors (TFTs) using tin (Sn) as the dopant element in ZnO. Furthermore, to improve performance stability of ZTO TFTs, microwave annealing (MWA) powered energy conversion from electromagnetic heat rather than transfer conducted post‐deposition (PDA) process layer. As result, TFTs improved MWA despite much...
In this study, we fabricated a resistive random access memory (ReRAM) of metal-insulator-metal structures using hydrogen silsesquioxane (HSQ) film that was deposited by low-cost solution process as resistance switching (RS) layer. For post-deposition annealing (PDA) to improve the performance HSQ-based ReRAMs, applied high energy-efficient microwave irradiation (MWI). comparison, ReRAMs with an as-deposited HSQ layer or conventional thermally annealed (CTA) were also prepared. The RS...
HfO <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf> -based ferroelectric-gate-insulator FETs (FeFETs) and metal/ferroelectrics/metal (MFM) capacitors are promising devices for AI computing applications. We have recently proposed experimentally demonstrated physical reservoir using dynamic polarization switching of Si FeFETs MFM capacitors. Here, history-dependent current waveforms utilized tasks. In this paper, we introduce the basic...
we have proposed and demonstrated reservoir computing (RC) by using HZO/Si FeFETs. In our FeFET reservoir, the time response of currents Si FeFETs is utilized as virtual nodes. We experimentally present fundamental properties RC performance address several methods to enhance performance. This scheme applied two typical AI tasks, prediction nonlinear time- series data speech recognition. show a classification accuracy 98.1 % in recognition task classify audio waveforms '0' '9' spoken digits.