Graphene Nanoribbon-based Synapses with Versatile Plasticity

Abstract

Designing and implementing artificial systems that can be interfaced with the human brain or that can provide computational ability akin to brain's processing information efficient style is crucial for understanding human brain fundamental operating principles and to unleashing the full potential of brain-inspired computing. As basic neural network components, responsible for information transfer between neurons, artificial synapses able to emulate analog biological synaptic behaviour are of particular interest. State of the art CMOS and memristorbased synapses suffer from scalability drawbacks (large energy consumption and area footprint), variability-induced instability, and are not bio-compatible. In this paper, we propose a generic Graphene Nanoribbon (GNR) based synapse structure and demonstrate that by changing GNR geometry and external bias voltages it can emulate different synaptic plasticity behaviours, i.e., Spike Timing Dependent Plasticity and Long-Term Depression and Potentiation, and that both excitatory and inhibitory synaptic behavior can be obtained with the same GNR geometry. To demonstrate biologically plausible operation, we make use of low voltage bias, i.e., 0.1 V, 0.2 V, and consider inputs consistent with measured brain synapses data, i.e., −50 mV to 50 mV pre-and post-synaptic spikes voltage range, and −60 ms to 60 ms time range. The simulations indicate that by changing the GNR shape we can enrich the plasticity behaviour (potentially beyond the considered cases) and the plasticity change of 100% provided by natural synapses can be achieved. Our investigation clearly suggests that the proposed GNR synapse structure is a promising candidate for large-scale neuromorphic systems integration, which might potentially bring novel insight on brain neurophysiology, as it requires a small footprint, is energy effective, biocompatible, and versatile from the synaptic behaviour point of view.

Compact Graphene-Based Spiking Neural Network With Unsupervised Learning Capabilities

“…The distance between two neighbor carbon atoms is denoted as a = 0.142 nm. We note that the capability of such graphenebased device to provide a rich set of complex functionalities has been demonstrated by the utilization of specially tailored (in terms of topology and dimensions) versions of it for the implementation of Boolean gates and individual synapses and neurons [17], [19], [21].…”

“…Depending on the applied top-gate voltage V g value charges are trapped or released by the graphene oxide interface, which causes an equivalent top-gate voltage shift and affects the topgate conductance modulation ability. This phenomenon makes the GNR conductance dependent on the cumulated device history activities and as such makes GNR devices suitable for the emulation of synaptic plasticity [21] and neuron membrane potential dynamics [19].…”

“…Graphene has emerged as a promising material for nanoelectronics, as it exhibits outstanding properties, e.g., ballistic transport, flexibility, ultimate thinness, and biocompatibility [15], [16]. Due to these attractive properties, graphene-based Boolean logic gate [17], [18] and spiking neuron and synapse [19], [20], [21] implementations have been reported. However, previous work concentrated on individual synapse and neuron designs while disregarding input-output compatibility aspects, which preclude their direct utilization for the implementation of graphene-based Spiking Neural Networks.…”

A Reconfigurable Graphene-Based Spiking Neural Network Architecture

“…Moreover, when compared with other emerging technologies, graphene is a biocompatible material, which makes it favourable for graphene-based neuromorphic biointerfaces. Previous work demonstrated graphene's suitability for artificial synapse [20], [21], neuron [22], and SNN unit implementations [23], thus a versatile, generic graphene-based SNN architecture that can be reconfigured for various practical tasks, would facilitate the exploration of grahene-based neuromorphic computing capabilities.…”

Ultra-Compact, Entirely Graphene-Based Nonlinear Leaky Integrate-and-Fire Spiking Neuron

“…Due to its properties graphene transistor-based logic, which follows the traditional CMOS design style has been proposed in [9] [10], while alternative approaches towards gate realizations departing from the switch-based mainstream have been introduced in, e.g., [11], [12]. Moreover, as graphene is biocompatible and can model complex functionality within a single Graphene Nanoribbon (GNR), GNR-based synapses have been proposed in [13].…”