Arrayed neural network capable of artificially designing boundary based on two-dimensional material and preparation method and application thereof

Abstract

The invention discloses an arrayed neural network capable of artificially designing a boundary based on a two-dimensional material, and a preparation method and application thereof. The preparation method comprises the following steps of: 1) growing a single crystal transition metal compound thin film on a substrate through adopting a vapor deposition method; (2) transferring the thin film obtained in the step (1) to a substrate with a dielectric layer; (3) irradiating the thin film with the dielectric layer on the substrate in the step (2) by using laser according to a specified path to obtain a sample with a specified heterostructure; and 4) plating a source electrode and a drain electrode on the surface of the sample of the specified heterostructure by adopting an electron beam or thermal evaporation method to obtain the arrayed neural network capable of artificially designing the boundary based on the two-dimensional material. According to the arrayed neural network, and the preparation method and the application thereof, due to the sub-nano-scale thickness, the artificially modulatable boundary and the controllably generated surface defects, the Schottky barrier between the metal and the electrode is effectively reduced, so that the turn-on voltage is reduced, and the energy consumption is further reduced to the human brain level.

Description

technical field [0001] The invention relates to a two-dimensional material-based arrayed neural network capable of artificially designing boundaries, a preparation method and application thereof, and belongs to the field of nanotechnology. Background technique [0002] Neuromorphic computing aims to mimic biological brains to mimic the efficiency, versatility, and resilience of the human brain. Synapses are a key part of information transmission and processing in the brain. Synaptic plasticity refers to the ability of synapses to change the strength of their connections according to the strength of neural activity, and is considered the biological basis of learning and memory. Two-dimensional heterojunctions at the subnanometer scale are promising flexible and transparent electronic devices. In recent years, synaptic devices based on 2D layered materials have been found to have excellent performance in neuromorphic computing systems, such as electronic synapses and neurons...

AI Technical Summary

Problems solved by technology

In addition, in order to be used in conjunction with the visual system, the synaptic device also needs to exhibit a sensitive response to the light dose, and it is still difficult for existing devices to achieve photoelectric dual modulation.

Images