Multi-scenario application of non-volatile smart invisible textiles based on copper-ion modulated MoS2

Abstract

Flexible lappable memristors are ideal building blocks for designing smart fabric information processing devices, which empowers versatile integration of smart fabrics. The volatile type of memristors have certain limitations in applications and most of them suffer from random formation of conductive channels and lack certain stability. In this paper, such small-area contacts were fabricated on carbon fibers using direct solution deposition, where each interwoven fiber contact can be used as a memristor, and excellent resistive switching performance was obtained by copper atom modification, with switching ratios of up to 10⁶, setup voltages as low as 0.8 V, and hold-up times of more than 10⁴ s. Our systematic study demonstrates that heat treatment temperatures and copper modification can achieve a volatile-to-non-volatile transition. The textile-based memory elements successfully demonstrate synaptic plasticity under electrical stimulation, accurately emulating fundamental neural network operations. The resulting textile-type memristor arrays have high inter-device homogeneity and are capable of high-recognition accuracy image recognition and processing of complex physiological data, such as brainwave signals. Textiles can switch back and forth between wave-transparent and wave-blocking at low frequencies, effectively preventing interference between devices. This dual function of combining neuromorphic computing with filterable electromagnetic protection offers great potential for next-generation smart textile applications.