A Review of MXene Memristive Networks: Atomic-Scale Engineering to Neuromorphic System Integration

Abstract

With the ever-growing demands of artificial intelligence and big data, the advancement of the conventional von Neumann framework is increasingly hindered by limitations in memory and power consumption. The human brain's energy-efficient neural mechanisms (e.g., synaptic plasticity) have driven innovations in brain-inspired computing architectures. Inspired by this, memristors, especially those containing MXenes, can efficiently simulate low-power, high-performance synaptic behaviors. MXenes are known for their tunable surface chemistry and excellent electrical conductivity, enabling the fabrication of superior neuromorphic components under ambient conditions. This study elucidates the effectiveness of MXene memristors in simulating synaptic plasticity and adaptive learning, thoroughly examines the challenges in advancing neuromorphic systems, and outlines future directions, thereby providing new possibilities for revolutionizing artificial intelligence and computing technologies.