Nanoscale memristor devices: materials, fabrication, and artificial intelligence

Abstract

With the advent of the big data era and the Internet of Things, promising hardware techniques for data storage and energy-efficient data processing are in great demand. This is especially more urgent with the emergence of the generated artificial intelligence (AI) applications, such as ChatGPT, which are changing and will continuously impact every aspect of our daily lives. Due to their capability to implement the memory and computational functions in a single device, memristor/memristive devices hold great potential to replace non-von Neumann architecture for high-performance and low-power consumption computing for future information technology. Herein, the recent progress in nanoscale memristor devices from the aspects of materials, fabrication, and artificial intelligence applications is reviewed. First, the memristor devices based on nanoscale materials of different dimensions, such as zero-dimensional, one-dimensional, and two-dimensional materials, and polymer materials were summarized. Moreover, memristor device fabrication methods, including conventional thin film, innovative nanojoining, and 2D/3D printing techniques, were systematically discussed. Finally, the applications of memristor devices in artificial intelligence, such as synapse and neuron emulation and neuromorphic computing, and the latest progress in molecular memristors were demonstrated. This review ends with the conclusion and outlook on memristor devices from the perspectives of materials, fabrication, and applications.