FET100 Short Course
FET Technology and Its Diverse Application
The concept of the Field Effect Transistor (FET) was first documented in a patent filed by Julius Lilienfeld on October 22, 1925. Since then, FETs have been a fundamental component in almost all electronic devices. To celebrate its 100th anniversary, IEEE Electron Devices Society (EDS) is organizing a series of activities under the theme of “the 100 Years of the FET” (FET100), including the short course “FET Technology and Its Diverse Applications”, which will be held on the first day of DTDA 2025.
In this short course, six distinguished speakers will give lectures on various applications of FETs, technological applications using new materials, and innovations based on new principles. Participants will gain a deeper understanding of FET technology and its applications, gaining valuable insights for their own research and development fields.
Future of MedTech Opened Up by CMOS: Wearable and Implantable Biomedical Devices
Graduate School of Biomedical Engineering, Tohoku University
Abstract
CMOS transistor technology has excellent characteristics such as high speed, multi-function integration, and low power consumption. CMOS is extremely important for MedTech for inspection, diagnosis, treatment, and vital sign monitoring.
CMOS transistor technology has made medical devices smaller and more power-efficient. It has also enabled high portability and wearability, as well as long-term implantation using batteries, contributing to improved quality of life for users.
In particular, the recent practical application of CMOS 3DICs using the TSV-based stacking technology has greatly expanded the possibilities of MedTech.
Among several types of 3DIC technologies, multiple chips-to-wafer (MCtW) stacking can stack known-good chips fabricated with different technologies and sizes, leading to a true 3D heterogeneous system.
A fully implantable retinal prosthesis with a 3D-stacked artificial retina (AR) chip has been developed at Tohoku University using MCtW stacking. In the 3D-stacked AR chip, the photoreceptor chip and stimulus current generator with image processing circuits are vertically stacked and electrically connected by many TSVs. The 3D-stacked AR chip has a layered structure similar to the human retina, so the AR chip of approximately 1,500 pixels can be fabricated. By implanting the 3D-stacked AR chip into the eyeball, the patients can employ their lens and cornea and shift the gaze point by moving the eyeball. The 3D-stacked AR chip leads to a small chip size, lightweight, large fill factor, high resolution, and a high quality of life for the patients.
Furthermore, a non-contact interface device (Nail Conductor) will be introduced as a next-generation wearable biomedical device. The nail conductor senses blood flow at the fingertips via the nails and allows users to operate several devices without touching them.
CMOS will continue to revolutionize MedTech in the future.
Steep Slope “PN-Body Tied SOI-FET” for RF-Energy Harvesting and Neuromorphic Applications
Abstract
Ultra-low power (ULP) and high-performance applications are the frontiers for FET technology, realizing “Trillion Sensors Universe” and “Artificial General Intelligence (AGI)”. R&D of the steep subthreshold slope (SS) FET, such as tunnel FETs and negative capacitance FETs for both applications, has still been the challenging topics of the future FET technology. Moreover, power reduction is one of the big issues on R&D of AGI because the current GPGPU approach consumes huge power. We have proposed “PN-Body Tied SOI-FET(PNBT-FET)” as a steep slope FET. We confirmed SS of PNBT NMOS and PMOS below 1mV/dec over 6 orders of the drain current with Vdd=0.1V. We applied it for the ultra-low power rectification diode for RF energy harvesting (EH). We also started applying our PNBT-FET for morphing neuron functions with a single device, aiming as a basic device candidate for spiking neural network system which meets ultra-low power by working with spike and event-driven, not clock system. I will introduce our PNBT-FET and status of applying it for RF-EH and Nuro-morphic applications.
Emerging semiconductors meet new applications: security and neuromorphics
Abstract
To overcome these obstacles, we introduce a unified photonic framework that exploits the inherent randomness of light-matter interactions for both probabilistic computing and secure entropy generation. Here, we introduce two focuses: (1) photonics-based entropy engine that utilizes random light-material interactions to generate high-entropy random numbers and (2) Gaussian transistors to enable Bayesian neural networks capable of real-time uncertainty estimation.
Key implementations include: (i) true random number generation via photodetectors (ii) real-time demonstration of photoinduced security devices (iii) Gaussian transistors and their applications to Gaussian mixture model, activated functions, Gaussian Naive Bayes.
Hafnia-based ferroelectric transistors: Insight into device physics and emerging applications
Abstract
HfZrOₓ-based ferroelectric field-effect transistors (FeFETs) are emerging innovative devices that demonstrate their potential in non-volatile memory and computing-in-memory applications, driven by their CMOS compatibility, scalability, and low power consumption. This talk will offer a comprehensive overview of FeFET technology, delving into the latest advancements and emerging trends, exploring the vast untapped potential of these devices, and addressing the critical challenges that must be overcome to unlock their full capabilities in both memory and computing applications. An in-depth exploration of the device physics behind FeFETs will be provided, with a focus on critical phenomena such as charge trapping, which presents both benefits and concerns regarding device operation and reliability. The talk will also briefly touch upon emerging applications, including multiply-accumulate accelerators and reservoir computing, where the unique characteristics of FeFETs offer a promising path forward for future computing technologies.
Solid-state electrochemical thermal transistors with enhanced switching performance using earth-abundant cerium oxide
Abstract
Thermal transistors (or thermal switches) are devices that can electrically switch “heat flow” on and off, like a semiconductor field effect transistor that switches “electric current” on and off. We can reuse waste heat exhausted to the environment using devices composed of thermal transistors such as thermal displays. Although several thermal transistors have been demonstrated thus far, the use of liquid electrolytes (or ionic liquids or ion gels) may limit the application from the viewpoint of reliability or liquid leakage. Recently, we developed oxide-based solid-state electrochemical thermal transistors that were fabricated on single crystal YSZ plate used as a solid electrolyte. In this talk, we will mainly explain cerium oxide-based solid-state electrochemical thermal transistors [A. Jeong, M. Yoshimura, H. Ohta et al., Science Advances 2025].
Vibronics: Concept, topics, and prospect
Abstract
Vibronics encompasses energy transport mechanisms in disordered systems, meta-dimensional structures, polymers, and interfaces where conventional phonon descriptions become insufficient. By developing unified theoretical frameworks and experimental methodologies, Vibronics enables cohesive descriptions of energy transport phenomena in solids regardless of structural complexity or dimensionality. The field integrates perspectives from physics, chemistry, materials science, and device engineering to expand traditional phonon engineering concepts. This convergence of disciplines creates powerful new approaches for controlling vibrational energy at multiple length scales. Such capabilities are increasingly critical for thermal management in quantum technologies, advanced semiconductor devices, and thermal functional materials.
This presentation will outline the fundamental concepts of Vibronics, highlight current research directions, and discuss future prospects for both fundamental understanding and technological applications. By bridging theoretical insights with practical engineering challenges, Vibronics aims to establish a foundation for next-generation energy transport solutions in complex material systems, potentially revolutionizing thermal management strategies across multiple technological domains.