Artificial Synapses Research Articles

Fiber-Shaped Cu-Ion Diffusive Memristor for Neuromorphic Computing

Fiber-shaped memristors have attracted enormous attention as potential wearable electronics. Here, a Cu-ion diffusive memristor with fiber shape was proposed for artificial synapse and neuromorphic computing. The fiber-shaped diffusive memristor exhibits gradual conductance modulation characteristics under consecutive voltage sweeps. Typical synaptic plasticity including EPSC, PPF, PPD, LTP/LTD and learning behaviors were all successfully achieved by the memristor. The active Cu <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2+</sup> of diffusive memristor was similar as Ca <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2+</sup> diffusion in biological synapse, which is the basis of realizing the functions of synaptic plasticity. The fiber-shaped Cu <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2+</sup> diffusive memristor acting as artificial synapse paves the way for next-generation wearable neuromorphic computing system.

TiW/SiOX:Al/TiW memristor with negative differential resistance effect for neuromorphic computing

Memristors, acting as artificial synapses, are proposed to be a promising candidate for neuromorphic computing applications. In this work, the CMOS process-compatible TiW/SiOX:Al/TiW memristor with negative differential resistance (NDR) effect is explored for this application. Nonpolar switching with a 340 on/off ratio, data retention beyond 106 s, and endurance of 106 cycles are realized. The device shows excellent analog behavior with nonlinearities of 1.69 and 0.65 of long-term potentiation and depression, respectively, under identical pulse stimuli. The synaptic features such as long-term potentiation (LTP), long-term depression (LTD), spike-timing-dependent plasticity (STDP), and paired-pulse facilitation (PPF) are mimicked. Moreover, on the basis of the symmetry and linearity of the conductance of TiW/SiOX:Al/TiW memristor, the neural network simulation for supervised learning presents successful pattern recognition, with an accuracy of 93.11% achieved after 20 iterations. It is proposed that the nonpolar NDR switching originates from the discontinuous Al metal nanoparticles that form deeply localized states in the energy band and result in the trap/de-trap of electronic carriers. Overall, this memristor with the NDR effect presents a unique way to simulate artificial synapse behavior for neuromorphic computing.

Realize low-power artificial photonic synapse based on (Al,Ga)N nanowire/graphene heterojunction for neuromorphic computing

The fast development of brain-inspired neuromorphic computing systems has stimulated urgent requirements for artificial synapses with low-power consumption. In this work, a photonic synaptic device based on (Al,Ga)N nanowire/graphene heterojunction has been proposed and demonstrated successfully. In the artificial synaptic device, the incident light, the nanowire/graphene heterojunction, and the light-generated carriers play the roles of action potential, pre-synaptic/post-synaptic membrane, and neurotransmitter in a biological synapse, respectively. As a key synaptic function, the paired pulse facilitation index of the photonic synapse can reach 202%, which can be modulated by the interval time between two adjacent light pulses. It is found that the graphene defects, the surface band bending, and the Al vacancies on the surface of (Al,Ga)N nanowires can be the key reasons contributing to the synaptic characteristics of artificial photonic devices. Hence, the dynamic “learning–forgetting” performance of the artificial synaptic device can resemble the “learning–forgetting” behavior of the human brain. Furthermore, the hand-written digits are set up to mimic a typical characteristic of human perceptual learning. After only three training epochs, the simulated network can achieve a high recognition rate of over 90% based on the experimental conductance for long-term potentiation and long-term depression. In supervised learning processes, such few training times are beneficial to reduce energy consumption significantly. Therefore, in the area of neuromorphic computing technology and artificial intelligence systems requiring low-power consumption, this work paves a potential way to develop the optoelectronic synapse based on semiconductor nanowires.

Demonstration of Synaptic Characteristics in VRRAM with TiN Nanocrystals for Neuromorphic System

AbstractTo efficiently develop an extremely intensive storage memory, the resistive random‐access memory (RRAM), which operates by producing and rupturing conductive filaments, is essential. However, due to the stochastic nature of filament production, this filamentary type resistive switching has an inherent limitation, which entails the unpredictability of the driving voltage and resistance states. Several strategies such as doping, research into multilayer stacks, and interface engineering, are suggested to tackle this challenge. This work fabricates a CMOS‐compatible TiN/HfOx/TiN‐NCs (nanocrystals)/HfOx/TiN RRAM to implement analog resistive switching and advance the development of the synaptic device. Specifically, atomic force microscopy (AFM), scanning electron microscopy (SEM), and transmission electron microscopy (TEM) are utilized to observe the formation of TiN nanocrystals, which play a crucial role in the enhancement of resistive switching. By comparing HfOx–based RRAM devices with and without NCs, the DC I–V curves, retention, endurance, and switching speed are properly examined. Interestingly, it is found that the TiN/HfOx/TiN‐NCs/HfOx/TiN device is more appropriately utilized as an artificial synapse in neuromorphic systems mainly due to its stable and reliable resistive switching properties. Finally, this work demonstrates well‐controlled resistive switching 3D vertical RRAM with TiN‐NCs, which is particularly suitable for high‐density memory.

Neuron-Inspired Sticky Artificial Spider Silk for Signal Transmission.

Neurons exhibit excellent signal transmission capacity, which inspire artificial neuron materials for applications in the field of wearable electronics and soft robotics. In addition, the neuron fibers exhibit good mechanical robustness by sticking to the organs, which currently has rarely been studied. Here, a sticky artificial spider silk was developed by employing a proton donor-acceptor (PrDA) hydrogel fiber for application as artificial neuron fibers. Tuning the molecular electrostatic interactions by modulating the sequences of proton donors and acceptors, enables combination of excellent mechanical properties, stickiness, and ion conductivity. In addition, the PrDA hydrogel exhibits high spinning capacity for a wide range of donor-acceptor combinations. The PrDA artificial spider silk would shed light on the design of new generation of artificial neuron materials, bio-electrodes, and artificial synapses. This article is protected by copyright. All rights reserved.

Recent developments of block copolymers in nonvolatile memory and artificial synapses

AbstractThe rapid development of artificial intelligence has significantly accelerated computational calculation and enlarged the demands in information storage. Therefore, high‐performance memory devices based on new architectures and functionalities, such as resistive memory, transistor memory, and phase change memory, along with their applications in neuromorphic computing and artificial synapse have arisen extensive interest among researchers. In order to improve the memory performance, block copolymer electrets with diversified self‐assembled structures have been extensively developed, and their structure–performance properties have been investigated. Therefore, in this focused review, a variety of block copolymers combining incompatible polymers, such as hydrophilic/hydrophobic, polar/nonpolar, and conjugated/insulating polymers, are introduced in this focused review. The design concepts and recent advances of block copolymers in nonvolatile memory and artificial synapses were addressed. With these divergent structure designs, block copolymers are regarded as potential candidates to provide high performance in memory device applications. This review sheds light on the great potential and importance of block copolymers for optoelectronic device applications.

An Aqueous Electrolyte Gated Artificial Synapse with Synaptic Plasticity Selectively Mediated by Biomolecules

AbstractThe emulation of functions and behaviors of biological synapses using electronic devices has inspired the development of artificial neural networks (ANNs) in biomedical interfaces. Despite the achievements, artificial synapses that can be selectively responsive to non‐electroactive biomolecules and directly operate in biological environments are still lacking. Herein, we report an artificial synapse based on organic electrochemical transistors and investigate the selective modulation of its synaptic plasticity by glucose. The enzymatic reaction between glucose and glucose oxidase results in long‐term modulation of the channel conductance, mimicking selective binding of biomolecules to their receptors and consequent long‐term modulation of the synaptic weight. Moreover, the device shows enhanced synaptic behaviors in the blood serum at a higher glucose concentration, which suggests its potential application in vivo as artificial neurons. This work provides a step towards the fabrication of ANNs with synaptic plasticity selectively mediated by biomolecules for neuro‐prosthetics and human‐machine interfaces.

An Aqueous Electrolyte Gated Artificial Synapse with Synaptic Plasticity Selectively Mediated by Biomolecules.

The emulation of functions and behaviors of biological synapses using electronic devices has inspired the development of artificial neural networks (ANNs) in biomedical interfaces. Despite the achievements, artificial synapses that can be selectively responsive to non-electroactive biomolecules and directly operate in biological environments are still lacking. Herein, we report an artificial synapse based on organic electrochemical transistors and investigate the selective modulation of its synaptic plasticity by glucose. The enzymatic reaction between glucose and glucose oxidase results in long-term modulation of the channel conductance, mimicking selective binding of biomolecules to their receptors and consequent long-term modulation of the synaptic weight. Moreover, the device shows enhanced synaptic behaviors in the blood serum at a higher glucose concentration, which suggests its potential application in vivo as artificial neurons. This work provides a step towards the fabrication of ANNs with synaptic plasticity selectively mediated by biomolecules for neuro-prosthetics and human-machine interfaces.

Black Phosphorus/Ferroelectric P(VDF-TrFE) Field-Effect Transistors with High Mobility for Energy-Efficient Artificial Synapse in High-Accuracy Neuromorphic Computing.

The neuromorphic system is an attractive platform for next-generation computing with low power and fast speed to emulate knowledge-based learning. Here, we design ferroelectric-tuned synaptic transistors by integrating 2D black phosphorus (BP) with a flexible ferroelectric copolymer poly(vinylidene fluoride-trifluoroethylene) (P(VDF-TrFE)). Through nonvolatile ferroelectric polarization, the P(VDF-TrFE)/BP synaptic transistors show a high mobility value of 900 cm2 V-1 s-1 with a 103 on/off current ratio and can operate with low energy consumption down to the femtojoule level (∼40 fJ). Reliable and programmable synaptic behaviors have been demonstrated, including paired-pulse facilitation, long-term depression, and potentiation. The biological memory consolidation process is emulated through ferroelectric gate-sensitive neuromorphic behaviors. Inspiringly, the artificial neural network is simulated for handwritten digit recognition, achieving a high recognition accuracy of 93.6%. These findings highlight the prospects of 2D ferroelectric field-effect transistors as ideal building blocks for high-performance neuromorphic networks.

Memristive artificial synapses based on Au–TiO2 composite thin film for neuromorphic computing

Neuromorphic computing architecture is considered to be a highly desirable next-generation computing architecture as it simulates the way the brain processes information. The basic device supporting such an architecture is called an artificial synapse, which possesses synapse-like functionalities. Here in this work, an Au–TiO2 composite thin film (Au nanoparticles embedding into TiO2 matrix) based memristive artificial synapse has been fabricated with excellent interface-type resistive switching (RS) characteristics. The conductivity of the device can be continuously tuned by applying different sequences of pulses, which could be analogous to the weight change of synapses. Various synaptic behaviors have been emulated, such as long-term potentiation/depression, short-term/long-term memory, learning-forgetting process, and paired-pulse facilitation. Finally, an artificial neural network for hand-written digits recognition has been constructed with an accuracy level as high as ∼90%. The excellent performance of the Au–TiO2 based device demonstrates the availability of incorporating the second phase to tune RS properties and shows its potential in a memristor for artificial synapses and neuromorphic computing with enhanced performance.

수직구조 3단자 인공 시냅스 소자

Recently, synaptic devices have the advantages of being able to process information in parallel. However, nondestructive weight control is limited in 2-terminal synaptic devices because reading and writing are conducted in a common electrode. Hence, 3-terminal synaptic devices with separately reading and writing processes are currently emerging for not only nondestructive weight control without data loss but also a very short vertical channel length. Since the length of the vertical channel is determined as the thickness of the active layer, nanometer range channel length can be achieved. According to the nano-scaled channel length, vertical 3-terminal artificial device can be operated with low voltage and energy consumption. In this review paper, vertical 3-terminal artificial synaptic devices were classified by electric field transmission or ion migration into organic semiconductor. According to a recent study, vertical 3-terminal artificial synapse was able to simulate a biological synapse even with a low driving voltage of up to 10 μV and a current density of MAcm−2 level. This review article gives an overview of vertical 3-terminal artificial synaptic devices, and suggest a practical strategy to achieve the massive data processing with high speed and low power consumption.

A halide perovskite thin film diode with modulated depletion layers for artificial synapse

Artificial synapse is basic electronic components used to prepare neuromorphic hardware with integrated storage and computing, high speed and low energy consumption. This work reports an artificial synapse based on perovskite diodes with learning, memory and computing functions. The artificial synapse demonstrates a mechanism for the dual roles of depletion layer width modulation and ion conduction channel, enabling it to complete the calculation task of neural network and memorize data. Artificial synapses are modeled with various neural networks, including single-layer neural networks, deep neural networks, and convolutional neural networks. The results show that artificial synapses show good recognition accuracy for the MNIST dataset. Especially on the convolutional neural network, only relying on more than 3000 parameters has achieved an accuracy rate of 92.4 %.

Multi-neuron connection using multi-terminal floating–gate memristor for unsupervised learning

Multi-terminal memristor and memtransistor (MT-MEMs) has successfully performed complex functions of heterosynaptic plasticity in synapse. However, theses MT-MEMs lack the ability to emulate membrane potential of neuron in multiple neuronal connections. Here, we demonstrate multi-neuron connection using a multi-terminal floating-gate memristor (MT-FGMEM). The variable Fermi level (EF) in graphene allows charging and discharging of MT-FGMEM using horizontally distant multiple electrodes. Our MT-FGMEM demonstrates high on/off ratio over 105 at 1000 s retention about ~10,000 times higher than other MT-MEMs. The linear behavior between current (ID) and floating gate potential (VFG) in triode region of MT-FGMEM allows for accurate spike integration at the neuron membrane. The MT-FGMEM fully mimics the temporal and spatial summation of multi-neuron connections based on leaky-integrate-and-fire (LIF) functionality. Our artificial neuron (150 pJ) significantly reduces the energy consumption by 100,000 times compared to conventional neurons based on silicon integrated circuits (11.7 μJ). By integrating neurons and synapses using MT-FGMEMs, a spiking neurosynaptic training and classification of directional lines functioned in visual area one (V1) is successfully emulated based on neuron’s LIF and synapse’s spike-timing-dependent plasticity (STDP) functions. Simulation of unsupervised learning based on our artificial neuron and synapse achieves a learning accuracy of 83.08% on the unlabeled MNIST handwritten dataset.

Biocompatible AlOOH-based memristor with biomimicking synaptic functions for artificial nociceptor applications

Electronic devices with biomimetic synaptic characteristics are one of the most critical components in health monitoring and artificial intelligence. The design of neuromorphic responsive devices, however, yet remains a ground challenge. Herein, a biocompatible and electrically responsive boehmite material (γ-AlOOH) is utilized to fabricate artificial memristors to realize mimicking biological synapsefunctions. The obtained AlOOH memristor is flexible, highly transparent, and biocompatible, exhibiting nonvolatile resistive switching behaviours with low threshold voltages, high ON-OFF ratios (>103), good switching durability (>104 cycles), and long retention time (>104 s). Specifically, the device can mimic the biosynaptic functions, such as paired-pulse facilitation (PPF), long-term potentiation/depression (LTP/LTD), amplitude-dependent plasticity (SADP), and spike time dependent plasticity (STDP), and the typical sensitization, desensitization, hyperalgesia, and allodynia behaviours of the biological nociceptors. This study reveals that the AlOOH memristor with excellent artificial synapses is resulted by the high proton mobility along with the OH bond length variation of AlOOH subjected to applied stimuli, such as voltages, pulse width, and pulse intervals. Therefore, the biocompatible AlOOH-based memristors with artificial synapse behaviour have promising potential for the applications in high-performance neuromorphic devices and future artificial intelligence.

Biologically plausible information propagation in a complementary metal-oxide semiconductor integrate-and-fire artificial neuron circuit with memristive synapses

Neuromorphic circuits based on spikes are currently envisioned as a viable option to achieve brain-like computation capabilities in specific electronic implementations while limiting power dissipation given their ability to mimic energy-efficient bioinspired mechanisms. While several network architectures have been developed to embed in hardware the bioinspired learning rules found in the biological brain, such as spike timing-dependent plasticity, it is still unclear if hardware spiking neural network architectures can handle and transfer information akin to biological networks. In this work, we investigate the analogies between an artificial neuron combining memristor synapses and rate-based learning rule with biological neuron response in terms of information propagation from a theoretical perspective. Bioinspired experiments have been reproduced by linking the biological probability of release with the artificial synapse conductance. Mutual information and surprise have been chosen as metrics to evidence how, for different values of synaptic weights, an artificial neuron allows to develop a reliable and biological resembling neural network in terms of information propagation and analysis.

High sensitivity and wide response range artificial synapse based on polyimide with embedded graphene quantum dots

Artificial electronic synapses are commonly used to simulate biological synapses to realize various learning functions, regarded as one of the key technologies in the next generation of neurological computation. This work used a simple spin coating technique to fabricate polyimide (PI):graphene quantum dots(GQDs) memristor structure. As a result, the devices exhibit remarkably stable exponentially decaying postsynaptic suppression current over time, as interpreted in the spike-timing-dependent plasticity phenomenon. Furthermore, with the increase of the applied electrical signal over time, the conductance of the electrical synapse gradually changes, and the electronic synapse also shows plasticity dependence on the amplitude and frequency of the pulse applied. In particular, the devices with the structure of Ag/PI:GQDs/ITO prepared in this study can produce a stable response to the stimulation of electrical signals between millivolt to volt, showing not only high sensitivity but also a wide range of “feelings”, which makes the electronic synapses take a step forwards to emulate biological synapses. Meanwhile, the electronic conduction mechanisms of the device are also studied and expounded in detail. The findings in this work lay a foundation for developing brain-like neuromorphic modeling in artificial intelligence.

Multiple-State Nonvolatile Memory Based on Ultrathin Indium Oxide Film via Liquid Metal Printing.

In this work, the ultrathin two-dimensional (2D) indium oxide (InOx) with a large area of more than 100 μm2 and a high degree of uniformity was automatically peeled off from indium by the liquid-metal printing technique. Raman and optical measurements revealed that 2D-InOx has a polycrystalline cubic structure. By altering the printing temperature which affects the crystallinity of 2D-InOx, the mechanism of the existence and disappearance of memristive characteristics was established. The tunable characteristics of the 2D-InOx memristor with reproducible one-order switching was manifest from the electrical measurements. Further adjustable multistate characteristics of the 2D-InOx memristor and its resistance switching mechanism were evaluated. A detailed examination of the memristive process demonstrated the Ca2+ mimic dynamic in 2D-InOx memristors as well as the fundamental principles underlying biological and artificial synapses. These surveys allow us to comprehend a 2D-InOx memristor using the liquid-metal printing technique and could be applied to future neuromorphic applications and in the field of revolutionary 2D material exploration.

Organic Memristor-Based Flexible Neural Networks with Bio-Realistic Synaptic Plasticity for Complex Combinatorial Optimization.

Hardware neural networks with mechanical flexibility are promising next-generation computing systems for smart wearable electronics. Several studies have been conducted on flexible neural networks for practical applications; however, developing systems with complete synaptic plasticity for combinatorial optimization remains challenging. In this study, the metal-ion injection density is explored as a diffusive parameter of the conductive filament in organic memristors. Additionally, a flexible artificial synapse with bio-realistic synaptic plasticity is developed using organic memristors that have systematically engineered metal-ion injections, for the first time. In the proposed artificial synapse, short-term plasticity (STP), long-term plasticity, and homeostatic plasticity are independently achieved and are analogous to their biological counterparts. The time windows of the STP and homeostatic plasticity are controlled by the ion-injection density and electric-signal conditions, respectively. Moreover, stable capabilities for complex combinatorial optimization in the developed synapse arrays are demonstrated under spike-dependent operations. This effective concept for realizing flexible neuromorphic systems for complex combinatorial optimization is an essential building block for achieving a new paradigm of wearable smart electronics associated with artificial intelligent systems.

Organic All-Photonic Artificial Synapses Enabled by Anti-Stokes Photoluminescence.

All-photonic synaptic devices with the merits of visible signals and high spatiotemporal resolution are promising to break the Von Neumann bottleneck. Although organic synapses outperform their inorganic counterpart for easy molecular modulation and lower energy consumption, the organic all-photonic artificial synapse has never been reported. Here, all-photonic synaptic characteristics were unprecedentedly observed in an organic semiconductor, (3,6-dimethyl-9H-carbazol-9-yl)(thiophen-2-yl) methanone (S2OC), with anti-Stokes photoluminescence. Impressively, the intensity of fluorescence from the higher excited state (S3) exhibited synaptic performance, which constantly increased with irradiation time through a channel composed of intersystem crossing, triplet-triplet annihilation, and energy transfer. More importantly, the relationship between the molecular structure and synaptic performance was established. Based on the synaptic photoplasticity property, noncontacted multilevel anticounterfeiting and imaging recognition were realized in all-photonic synapse arrays. This work provides a universal strategy for tuning the performances of organic synapses upon regulating the molecular structures, which paves the way for the application of organic semiconductors in artificial intelligence.

Periodic Ferroelectric Stripe Domains in α-In2Se3 Nanoflakes Grown via Reverse-Flow Chemical Vapor Deposition.

The two-dimensional (2D) layered semiconductor α-In2Se3 has aroused great interest in atomic-scale ferroelectric transistors, artificial synapses, and nonvolatile memory devices due to its distinguished 2D ferroelectric properties. We have synthesized α-In2Se3 nanosheets with rare in-plane ferroelectric stripe domains at room temperature on mica substrates using a reverse flow chemical vapor deposition (RFCVD) method and optimized growth parameters. This stripe domain contrast is found to be strongly correlated to the stacking of layers, and the interrelated out-of-plane (OOP) and in-plane (IP) polarization can be manipulated by mapping the artificial domain structure. The acquisition of amplitude and phase hysteresis loops confirms the OOP polarization ferroelectric property. The emergence of striped domains enriches the variety of the ferroelectric structure types and novel properties of 2D In2Se3. This work paves a new way for the controllable growth of van der Waals ferroelectrics and facilitates the development of novel ferroelectric memory device applications.