Memristive synapses with high reproducibility for flexible neuromorphic networks based on biological nanocomposites

Ge, Juan; Li, Dongyuan; Huang, Changqiao; Zhao, Xuanbo; Qin, Jieli; Liu, Huanyu; Ye, Weiyong; Xu, Wenchao; Liu, Zhiyu; Pan, Shusheng

doi:10.1039/c9nr08001e

Cited by 54 publications

(42 citation statements)

References 61 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…[ 61,62 ] When the amplitude and number of pulses are increased, the synaptic weight is enhanced (Figure S8c, Supporting Information). [ 63 ] XPS analysis at 100, 200, 500 nm depth of MXene confirms the occurrence of electrochemical doping (Figure S8d, Supporting Information).…”

Section: Resultsmentioning

confidence: 62%

Redox MXene Artificial Synapse with Bidirectional Plasticity and Hypersensitive Responsibility

Wei

Gong

et al. 2020

Adv Funct Materials

View full text Add to dashboard Cite

Artificial synapses are key elements for the nervous system which is an emulation of sensory and motor neuron signal transmission. Here, the design and fabrication of redox‐behavior the metal carbide nanosheets, termed MXene artificial synapse, which uses a highly‐conductive MXene electrode, are reported. Benefiting from the special working mechanism of ion migration with adsorption and insertion, the device achieves world‐record power consumption (460 fW) of two‐terminal synaptic devices, and so far, the bidirectionally functioned synaptic device could effectively respond to ultra‐small stimuli at an amplitude of ±80 mV, even exceeding that of a biological synapse. Potential applications have also been demonstrated, such as dendritic integration and memory enhancement. The special strategy and superior electrical characteristics of the bidirectionally functioned electronic device pave the way to high‐power‐efficiency brain‐inspired electronics and artificial peripheral systems.

show abstract

Section: Resultsmentioning

confidence: 62%

Redox MXene Artificial Synapse with Bidirectional Plasticity and Hypersensitive Responsibility

Wei

Gong

et al. 2020

Adv Funct Materials

View full text Add to dashboard Cite

show abstract

“…This conductance modulation can be used to emulate the synaptic weight update characteristics. Due to simple structure and conductance modulation characteristics, various two-terminal devices such as memristors, PCMs, and FTJs have been researched as artificial synapses ( Ge et al., 2020 ; Boybat et al., 2018 ; Li et al., 2020a ; Chen et al., 2018a ). This section reviews the device characteristics and switching mechanism of two-terminal artificial synapses for DNNs.…”

Section: Artificial Synapses Based On Emerging Materialsmentioning

confidence: 99%

“…These conductance change behaviors of memristors depend on the types of materials, operation mechanism, and device structures. A memristor composed of carboxymethyl iota-carrageenan (CιC) with Ag nanoclusters could implement synaptic weight update characteristics of DNNs ( Ge et al., 2020 ). A CιC:Ag layer was deposited using spin-coating, and nanopores were formed in the CιC:Ag layer by adding acetic acid in the solution.…”

Section: Artificial Synapses Based On Emerging Materialsmentioning

confidence: 99%

See 1 more Smart Citation

Emerging Materials for Neuromorphic Devices and Systems

et al. 2020

View full text Add to dashboard Cite

“…[7][8][9][10] Therefore, the imitation of synapses as building blocks for neural processing are of crucial importance for constructing an e cient arti cial neuromorphic system. [11][12][13] For emulating synaptic behaviors, a variety of structures such as metal/insulator/metal (MIM)-stack switches, [14][15][16][17][18] electrolyte/semiconductor hetero-junctions, 19,20 and multi-terminal transistors [21][22][23][24][25] have been employed to realize the signal transmission. Among them, the con gurations featuring ion migrations can opportunely facilitate the neuromorphic synapses to simulate the biological sensory and motor neurons in bioinspired ionotronic systems.…”

Section: Introductionmentioning

confidence: 99%

Graphdiyne-Based Ultra-Responsive Artificial Synapse with Multi-Ion Diffusive Dynamics for Mimicking Efferent Nerves

Wei¹,

Shi²,

Sun³

et al. 2020

Preprint

View full text Add to dashboard Cite

Somatosensory nerves require synapses to respond efficiently and in parallel for receving and transmiting biological signals. The gap between biological systems and conventional electronics needs ionotronics to bridge. However, the exploration of new materials and the systematic construction of ionotronics still pose challenges. Graphdiyne, a highly π-extended two-dimensional (2D) carbon allotrope, has demonstrated potential applications in ionic peripheral systems for its inherent network holes that can be used for rapid and selective transmission of diverse ions. Here, a graphdiyne-based artificial synapse (GAS), exhibiting intrinsic short-term plasticity, has been proposed to mimic the biological signal transmission behaviors. An record-breaking impulse responsiveness (±5 mV) that is an order of magnitude exceeding biological level has been realized for ultra-sensitive and power-efficient brain-inspired applications, with the lowest femtowatt-level consumption (~16.7 fW). Most importantly, GAS is capable of parallelly processing signals transmitted from multiple preneurons and therefore realizing dynamic logics and spatiotemporal rules. In a proof-of-concept demonstration, our artificial efferent nerve, connecting GAS with artificial muscles, completes the information integration of preneurons and the information output of motor neurons, which is advantageous for coalescing multiple sensory feedbacks (e.g., visual and tactile) and reacting to these events. Our synaptic element has potential applications in bioinspired peripheral nervous systems of soft electronics and neurorobotics.

show abstract

Memristive synapses with high reproducibility for flexible neuromorphic networks based on biological nanocomposites

Abstract: A memristive synapse based on novel biomaterial nanocomposites is proposed and simulations including the non-ideal factors prove an online learning accuracy of 94.3%.

Cited by 54 publications

References 61 publications

Redox MXene Artificial Synapse with Bidirectional Plasticity and Hypersensitive Responsibility

Redox MXene Artificial Synapse with Bidirectional Plasticity and Hypersensitive Responsibility

Emerging Materials for Neuromorphic Devices and Systems

Graphdiyne-Based Ultra-Responsive Artificial Synapse with Multi-Ion Diffusive Dynamics for Mimicking Efferent Nerves

Contact Info

Product

Resources

About