Abstract:Scalable synaptic transistor memory from solution‐processed carbon nanotubes, exhibiting reconfigurable weights and long‐term synaptic plasticity, enables parallel neuromorphic data processing with a speed of 1 M bit per second per input channel. Neural networks as a core information processing technology in machine learning and artificial intelligence demand substantial computational resources to deal with the extensive multiply‐accumulate operations. Neuromorphic computing is an emergent solution to address this problem, allowing the computation performed in memory arrays in parallel with high efficiencies conforming to the neural networks. Here, scalable synaptic transistor memories are developed from solution‐sorted carbon nanotubes. The transistors exhibit a large switching ratio of over 105, a significant memory window of ≈12 V arising from charge trapping, and low response delays down to tens of nanoseconds. These device characteristics endow highly stabilized reconfigurable conductance states, successful emulation of synaptic functions, and a high data processing speed. Importantly, the devices exhibit uniform characteristic metrics, e.g., with a 1.8% variation in the memory window, suggesting an industrial‐scale manufacturing capability of the fabrication. Using the memories, a hardware convolution kernel is designed and parallel image processing is demonstrated at a speed of 1 M bit per second per input channel. Given the efficacy of the convolution kernel, a promising prospect of the memories in implementing neuromorphic computing is envisaged. To explore the potential, large‐scale convolution kernels are simulated and high‐speed video processing is realized for autonomous driving.

Carbon nanotube neurotransistors with ambipolar memory and learning functions

Carbon Nanotube-Based Flexible Ferroelectric Synaptic Transistors for Neuromorphic Computing

Carbon Nanotube-Based Programmable Devices for Adaptive Architectures

Carbon Nanotube Transistor with Short-Term Memory

Charge trap-based carbon nanotube transistor for synaptic function mimicking

Bio-inspired artificial heterosynapse based on carbon nanotubes memtransistor with dynamically tunable analog switching behavior

Scalable Synaptic Transistor Memory from Solution‐Processed Carbon Nanotubes for High‐Speed Neuromorphic Data Processing

Neuromorphic Function Learning with Carbon Nanotube Based Synapses.

Aqueous-solution-processed Proton-Conducting Carbon Nitride/polyvinylpyrrolidone Composite Electrolytes for Low-Power Synaptic Transistors with Learning and Memory Functions

Design and Modeling of a Neuro-Inspired Learning Circuit Using Nanotube-Based Memory Devices

Neuromorphic Computing Using a Floating Gate Modulated Photoelectric Synaptic Transistor

A Low-Power Vertical Dual-Gate Neurotransistor with Short-Term Memory for High Energy-Efficient Neuromorphic Computing

Long-ionic-gated graphene synaptic transistor with enhanced memory, learning function and humidity perception

An Evolvable Organic Electrochemical Transistor for Neuromorphic Applications

A memristive nanoparticle/organic hybrid synapstor for neuro-inspired computing

Tactile tribotronic reconfigurable p-n junctions for artificial synapses

Hydrophobically gated memristive nanopores for neuromorphic applications

Optically Modulated Nanofluidic Ionic Transistor for Neuromorphic Functions

Nitrogen-doped titanium dioxide nanorod array memristors with synaptic features and tunable memory lifetime for neuromorphic computing

Optically and electrically modulated printed carbon nanotube synaptic transistors with a single input terminal and multi-functional output characteristics

Emulation of Synaptic Plasticity on a Cobalt-Based Synaptic Transistor for Neuromorphic Computing