2D MoTe₂ memristors for energy-efficient artificial synapses and neuromorphic applications

Area of Science:

• Materials Science
• Nanotechnology
• Computer Engineering

Background:

• Two-dimensional (2D) transition metal dichalcogenides (TMDs) are gaining interest for advanced electrical and low-energy neuromorphic applications.
• Molybdenum telluride (MoTe2) exhibits unique properties suitable for artificial synapses.

Purpose of the Study:

• This review summarizes the creation, characteristics, and applications of MoTe2-based memristive devices.
• Emphasis is placed on their potential for neuromorphic computing as artificial synapses.

Main Methods:

• Review of existing literature on MoTe2 memristive devices.
• Analysis of MoTe2 properties like resistance switching, synaptic linearity, and phase states.
• Examination of experimental performances including decimal arithmetic and synaptic plasticity simulations.

Main Results:

• MoTe2 memristors demonstrate resistance switching, linear synaptic potentiation, and customizable phase states.
• They enable basic computational functions (decimal arithmetic, addition, multiplication) with low energy consumption.
• MoTe2 memristor arrays achieve over 90% accuracy in handwritten digit recognition, showing scalability and stability.

Conclusions:

• MoTe2 memristors are highly promising for artificial synapses in neuromorphic computing due to their unique properties and demonstrated performance.
• Overcoming challenges like environmental robustness, phase transition sensitivity, and thermal stability is crucial for practical applications.
• Future research should focus on scalable synthesis, hybrid materials, doping, structural modification, and comprehensive material property analysis.