.png){kind=icon}
.png){kind=icon}
Return
2D Material-Based Bioinspired Devices for Neuromorphic Computing
.png){kind=icon}
PRE
AI
DOI：10.1002/smll.202504118.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
The rapid advancement of artificial intelligence has underscored the limitations of traditional von Neumann architecture, particularly their inherent “memory wall” and “power wall” bottlenecks, which hinder efficient computing. Neuromorphic computing, mimicking the brain's parallel computing paradigm, offers a promising solution by enabling high energy efficiency and fast computing speed, making it a key approach for overcoming these computing bottlenecks in the post-Moore era. Two-dimensional (2D) materials have emerged as exceptional candidates for next-generation neuromorphic devices, owing to their atomic-scale thickness, tunable physical properties, and superior integration compatibility. A range of 2D materials have successfully emulated biological synaptic plasticity and neuronal dynamics through diverse device architectures, such as two-terminal memristors, three-terminal field-effect transistors (FETs), and optoelectronic transistors. These advancements have demonstrated significant potential for edge computing and autonomous control systems, enabling multi-modal sensory integration across vision, hearing, touch, smell, and taste. Despite these breakthroughs, critical challenges remain, particularly in wafer-scale material synthesis, device uniformity, and large-scale integration. This review systematically summarizes the latest progress in structural design, performance optimization, and practical applications of 2D neuromorphic devices. Furthermore, it explores the current challenges and emerging opportunities in 2D neuromorphic computing while providing insights into future development in electronics and heterogeneous integration architectures.
Keywords:
2D materials
neuromorphic computing
neuromorphic devices