Robust 2D MoS2 Artificial Synapse Device Based on a Lithium Silicate Solid Electrolyte for High-Precision Analogue Neuromorphic Computing

Sub-Millisecond and Energy-Efficient Electrochemical Synaptic Transistors with a Partially Reduced Graphene Oxide Channel

Designed artificial synaptic transistors, which emulate the functions of biological synapses, are intended to achieve information processing and computation, showcasing their promise in advancing artificial intelligence. Herein, we propose a synaptic transistor composed of a partially reduced graphene oxide (prGO) channel and a Nafion electrolyte, operating based on electrochemical reactions of the prGO channel, which are assisted by protons through the Nafion electrolyte. After electrical reduction of a pristine GO channel to the prGO channel by sweeping the drain voltage, the transistor exhibits over 200 distinct conductance states under applications of short gate voltage pulses down to 500 μs width, giving rise to a low energy consumption of 10-50 pJ per gate pulse. Using highly linear and symmetric long-term potentiation and depression characteristics, an image recognition accuracy using an artificial neural network based on a two-layer perceptron model is calculated to be 90%. If gate current pulses are used, the image recognition accuracy further increases to 94%, because of the improved linearity and symmetry of the conductance change. The transistor also exhibits short-term plasticity, such as paired-pulse facilitation and spike-timing-dependent plasticity, with time ranges of less than a few tens of milliseconds. These superior synaptic properties of the Nafion/prGO transistors will offer a remarkable paradigm for the development of neuromorphic computation architectures.

2D-SnS-Embedded Schottky Device with Neurotransmitter-Like Functionality Produced Using Proximity Vapor Transfer Method for Photonic Neurocomputing

Neuromorphic computing, which involves the creation of artificial synapses capable of mimicking biological brain activity, has intrigued researchers in the field of artificial intelligence (AI). To advance neuromorphic computing, a highly efficient 2D material-based artificial synapse capable of performing logical and arithmetic operations must be developed. However, fabricating large, uniform films or high-quality structures of 2D materials remains challenging because of their multistep and complex fabrication processes. In the present study, to produce large (Ø ≈ 3 in.), uniform, transparent neuromorphic devices, a novel single-step approach called proximity vapor transfer (PVT) that utilizes van der Waals (vdW) materials is employed. This single-step technique, which involves the fabrication of vdW materials on various substrates (glass, ITO, AZO, Mo, and Cu), allows control of the thickness and bandgap tunability. The Schottky device developed via the PVT method using vdW SnS with neurotransmitter (acetylcholine)-like functionality emulates biological synapses and exhibits photoelectronic synaptic behavior with wide-field-of-view synaptic plasticity. In addition, logic gate operations (NOT, OR, AND), reward-cascade neurotransmission, and imaging can be performed using 3 × 3 arrays of the device. This study represents a significant step toward the development of transparent and large-area synaptic devices, which are crucial for advancing AI applications.

Combined optical and electrical control of a low-power consuming (∼fJ) two-terminal organic artificial synapse for associative learning and neuromorphic applications

Optoelectronic synaptic devices outperform electrical synapses in speed, energy efficiency, and integration density. Recent progress in visual sensing and optogenetics has led to the integration of light-sensitive materials in these devices, promising unmatched speed, connectivity, and bandwidth. Here, we present a copper phthalocyanine (CuPc) based optoelectronic synaptic device boasting femto Joule power consumption stable at room temperature. The optoelectronic synapse can be operated with energy consumption as low as 430.4 fJ which is very attractive from the point of view of low-power neuromorphic devices. By modulating light pulses, the neuromorphic behavior can be emulated including excitatory post-synaptic current (EPSC), paired-pulse facilitation (PPF), transitioning from short-term plasticity (STP) to long-term plasticity (LTP), spike-rate dependent plasticity (SRDP) and spike-number dependent plasticity (SNDP), etc. Optical potentiation and electrical depression are observed with combined optical and electrical stimulation, proving the multi-functionality of the synapse. Furthermore, the device demonstrates classical associative learning behaviors like Pavlovian conditioning using optical and electrical stimuli. We have established the pain conditioning processes such as hyperalgesic response and pain extinction effects with varying optical pulse amplitudes. These results render the CuPc-based devices as multifunctional and highly versatile artificial synaptic devices for future computing applications, offering unprecedented efficiency and functionality in neuromorphic systems.

In2O3/ZnO heterojunction thin film transistor for high recognition accuracy neuromorphic computing and optoelectronic artificial synapses

Solid electrolyte-gated transistors exhibit improved chemical stability and can fulfill the requirements of microelectronic packaging. Typically, metal oxide semiconductors are employed as channel materials. However, the extrinsic electron transport properties of these oxides, which are often prone to defects, pose limitations on the overall electrical performance. Achieving excellent repeatability and stability of transistors through the solution process remains a challenging task. In this study, we propose the utilization of a solution-based method to fabricate an In2O3/ZnO heterojunction structure, enabling the development of efficient multifunctional optoelectronic devices. The heterojunction's upper and lower interfaces induce energy band bending, resulting in the accumulation of a large number of electrons and a significant enhancement in transistor mobility. To mimic synaptic plasticity responses to electrical and optical stimuli, we utilize Li+-doped high-k ZrOxthin films as a solid electrolyte in the device. Notably, the heterojunction transistor-based convolutional neural network achieves a high accuracy rate of 93% in recognizing handwritten digits. Moreover, our research involves the simulation of a typical sensory neuron, specifically a nociceptor, within our synaptic transistor. This research offers a novel avenue for the advancement of cost-effective three-terminal thin-film transistors tailored for neuromorphic applications.

Effects of Mg Doping to a LiCoO2 Channel on the Synaptic Plasticity of Li Ion-Gated Transistors

Artificial synapses with ideal functionalities are essential in hardware neural networks to allow for energy-efficient analog computing. However, the realization of linear and symmetric weight updates in real synaptic devices has proven challenging and ultimately limits the online training capabilities of neural network systems. Herein, we investigate the effect of Mg doping on a LiCoO2 (LCO) channel in a Li ion-gated synaptic transistor, so as to improve long-term and short-term plasticity. Two transistor structures, based on a lithium phosphorus oxynitride electrolyte, were examined by using undoped LCO and Mg-doped LCO as the channel material between the source and drain electrodes. It was found that Mg doping increased the initial channel conductance by 3 orders of magnitude, which is probably due to the substitution of Co3+ by Mg2+ and the compensation of hole creation. It was further found that the doped channel transistor showed good retention characteristics and better linearity of long-term potentiation and depression when voltage pulses were applied to the gate electrode. The improved retention and linearity are attributed to an extended range of the insulator-to-conductor transition by Mg doping and Li-ion extraction/insertion cooperated in the LCO channel. Using the obtained synaptic weight update, artificial neural network simulations demonstrated that the doped channel transistor shows an image recognition accuracy of ∼80% for handwritten digits, which is higher than ∼65% exhibited by the undoped channel transistor. Mg doping also improved short-term plasticity such as paired-pulse facilitation/depression and Hebbian spike timing-dependent plasticity. These results indicate that elemental doping to the channel of Li ion-gated synaptic transistors could be a useful procedure for realizing robust neuromorphic systems based on analog computing.

Ultrafast and Low-Power 2D Bi2O2Se Memristors for Neuromorphic Computing Applications

Memristors that emulate synaptic plasticity are building blocks for opening a new era of energy-efficient neuromorphic computing architecture, which will overcome the limitation of the von Neumann bottleneck. Layered two-dimensional (2D) Bi₂O₂Se, as an emerging material for next-generation electronics, is of great significance in improving the efficiency and performance of memristive devices. Herein, high-quality Bi₂O₂Se nanosheets are grown by configuring mica substrates face-down on the Bi₂O₂Se powder. Then, bipolar Bi₂O₂Se memristors are fabricated with excellent performance including ultrafast switching speed (<5 ns) and low-power consumption (<3.02 pJ). Moreover, synaptic plasticity, such as long-term potentiation/depression (LTP/LTD), paired-pulse facilitation (PPF), and spike-timing-dependent plasticity (STDP), are demonstrated in the Bi₂O₂Se memristor. Furthermore, MNIST recognition with simulated artificial neural networks (ANN) based on conductance modification could reach a high accuracy of 91%. Notably, the 2D Bi₂O₂Se enables the memristor to possess ultrafast and low-power attributes, showing great potential in neuromorphic computing applications.