Atomic Nb-doping of WS₂ for high-performance synaptic transistors in neuromorphic computing.

Area of Science:

• Materials Science
• Nanotechnology
• Device Physics

Background:

• Atomically thin two-dimensional (2D) transition-metal dichalcogenides (TMDCs) are increasingly explored for synaptic transistors due to their integration potential.
• Substitutional doping offers a method to precisely engineer the physical properties of 2D materials for synaptic plasticity.

Purpose of the Study:

• To enhance the performance of synaptic transistors by improving their switch ratio.
• To investigate the effects of niobium (Nb) substitutional doping in tungsten disulfide (WS2) channel materials.

Main Methods:

• Selective substitutional doping of WS2 with niobium (Nb) atoms at tungsten (W) sites.
• Fabrication and characterization of Nb-doped WS2 synaptic transistors.
• Evaluation of device performance, including switch ratio and learning capabilities.

Main Results:

• Nb-doped WS2 synaptic transistors achieved a switch ratio of 10^3, a 100-fold improvement over undoped WS2 devices.
• Niobium atoms effectively facilitated electron trapping and detrapping, crucial for synaptic function.
• The Nb-WS2 transistors demonstrated accurate simulation of synaptic potentiation, inhibition, and learning processes.
• Achieved 92.30% recognition accuracy on the MNIST handwritten digit dataset after 125 training iterations.

Conclusions:

• Substitutional doping of WS2 with Nb is a viable and accessible method to significantly enhance synaptic transistor performance.
• The engineered Nb-WS2 material shows promise for neuromorphic computing applications, particularly in handwritten digit recognition.
• This work provides insights into doping strategies for optimizing channel materials in next-generation synaptic devices.