The information computation through synapse networks in the brain plays a vital role for cognitive behaviors such as image/video recognition, self-learning, and decision-making. Achieving proper synaptic networks by conventional semiconductor and memristive devices has encountered critical issues such as the spatial density requiring a number of transistors for one synapse, reliable filament formation in memristors, or emulating diverse excitatory and inhibitory synaptic plasticity with two-terminal device geometry. Here, we report selective growth of variously doped MoS 2 with controllable conductance plasticity, which can be used for emulating diverse synaptic junctions. The conductance plasticity in the monolayer MoS 2 was found to originate from resistive-heating near the junctions with electrodes in the two-terminal device geometry and the carrier concentration-dependent metal-insulator transition in the MoS 2 channel. A spatiotemporal synaptic summation is demonstrated where the firing of a proper postsynaptic membrane potential can be designed for cognitive processes. Compared with previously reported three terminal synaptic devices with atomically thin materials, our two-terminal devices with flexible synaptic strengths have advantages for integrating three-dimensional neuronal networks. This provides a new insight on two-dimensional materials as a promising arena for integrated synaptic functionalities in artificial neural networks.
No takes yet. Share an insight, caveat, or question.
Sun et al. (2018) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:
More from 2D Materials