Li‐Ion Synaptic Transistor for Low Power Analog Computing

Abstract

Nonvolatile redox transistors (NVRTs) based upon Li-ion battery materials are demonstrated as memory elements for neuromorphic computer architectures with multi-level analog states, "write" linearity, low-voltage switching, and low power dissipation. Simulations of backpropagation using the device properties reach ideal classification accuracy. Physics-based simulations predict energy costs per "write" operation of <10 aJ when scaled to 200 nm × 200 nm.

Bridging Biological and Artificial Neural Networks with Emerging Neuromorphic Devices: Fundamentals, Progress, and Challenges

“…[108][109][110][111] In addition, electrochemical random-access memory (ECRAM) based on ion intercalation has recently been reported as a promising synaptic cell, showing multi-states and incremental switching with near-ideal switching symmetry and linearity. [29,[112][113][114][115][116][117][118][119] The electrochemically driven ion intercalation process is more controllable than filament-related ion movements in RRAM; therefore, ECRAM also exhibits a much smaller stochasticity. In addition, by borrowing the battery concept, those devices successfully decouple the read and write operations and thus realize low programming energy and long retention time simultaneously.…”

The future of electronics based on memristive systems

“…This kind of chemical 'gating' effect can be obtained by using ions with low energy barrier (for example, Li ions) to drive the charge-discharge redox reactions in the conduction channel in a battery-like fashion. In this case, switching can occur at a very low voltage (for example, 5 mV), resulting in excellent power efficiency 71 . Beyond synaptic behaviours, memristive systems can be used to implement neuronal elements that ultimately receive, process and transmit information in bioinspired computing systems.…”

Hybrid Silicon Nanowire Devices and Their Functional Diversity

“…[270] The brain, a natural computing machine, has a great ability to predict, analyze, and infer links among the events, which are difficult tasks for nowadays computers. Although memristive memory cells are mainly used for artificial synapses, many synaptic or neuron-like transistors using either memristive channel [13,[272][273][274][275][276][277] or dielectric materials [278][279][280][281][282][283][284] have been reported. Recently, the several inspiring developments related to artificial neurons and synapses have been published.…”