Soft & Neuromorphic Electronics

Next-generation flexible systems have attracted much attention where vast amounts of data such as visual, auditory, behavioral, and emotional information are being collected in real-time. Silicon-based semiconductors have played significant roles of signal processing, nerve stimulation, memory storage, and wireless communication in implantable electronics. However, the rigid and bulky LSI chips have limited its uses in in vivo and flexible devices.

Human electronics systems realized by flexible memory devices enable data extraction, processing and analysis with low power consumption. We have developed silicon-based flexible large-scale integrated circuits (LSI) for bio-medical applications. We fabricated flexible LSI interconnected with thousand nano-transistors on silicon wafer by state-of-the-art 0.18 CMOS process, and then the entire bottom substrate except top 100 nm active circuit layer was removed by wet chemical etching. This work could provide an approach to flexible LSI for an ideal artificial retina system and other bio-medical devices. Also, the result represents an exciting technology with the strong potential to realize fully flexible consumer electronics such as application processor (AP) for mobile operating system, high-capacity memory, wireless communication in the near future.
A memristor can comprehensively emulate the neural components rather than imitating a single characteristic superficially due to its analog and hysteretic resistive switching. Bio-plausible mimicry aims to emulate biological working mechanisms to implement the complicated functional characteristics of a neural network for artificial intelligence (AI). Bio-plausible neuromorphic device using memristor is a direct and efficient approach for the emulation of biological systems, contributing to the realization of brain-like intelligence beyond limited AI applications.
Our research goal is to realize brain-computer interface by using artificial neural components. Synaptic plasticity and non-synaptic intrinsic plasticity work concurrently in the cognitive system of the human brain. We recently reported a neurosynaptic threshold switch-phase change memory (TS-PCM) device that emulated the intrinsic plasticity coupled with synaptic plasticity. Volatile TS and nonvolatile PCM layers were stacked in a single cell to implement neuron and synapse simultaneously. This work is based on our previous research about hybrid PCM architecture of TiW/GST/NiO/Ni to control the crystallinity of GST with self-structured Ni filaments. These remarkable achievements and advancements of bio-plausible BNN will open a new era of the brain-like intelligence of AI systems.

[Related References]

"ACF Packaged Flexible NAND Flash Memory" IEDM, Washington DC, 19.3, 1, 2015

"Simultaneous Roll Transfer and Interconnection of Flexible silicon NAND Flash Memory" Adv. Mater., 28, 8371, 2016

"Flexible Crossbar-Structured Phase Change Memory Array via Mo-based Interfacial Physical Lift-Off: Adv. Mater. 29, 1806338, 2019

"Simultaneous emulation of synaptic and intrinsic plasticity using a memristive synapse" Nat. Commun, 13, 2811, 2022