Recent advancements in metal oxide-based hybrid nanocomposite resistive random-access memories for artificial intelligence

Abstract

Artificial intelligence (AI) advancements are driving the need for highly parallel and energy-efficient computing analogous to the human brain and visual system. Inspired by the human brain, resistive random-access memories (ReRAMs) have recently emerged as an essential component of the intelligent circuitry architecture for developing high-performance neuromorphic computing systems. This occurs due to their fast switching with ultralow power consumption, high ON/OFF ratio, excellent data retention, good endurance, and even great possibilities for altering resistance analogous to their biological counterparts for neuromorphic computing applications. Additionally, with the advantages of photoelectric dual modulation of resistive switching, ReRAMs allow optically inspired artificial neural networks and reconfigurable logic operations, promoting innovative in-memory computing technology for neuromorphic computing and image recognition tasks. Optoelectronic neuromorphic computing architectured ReRAMs can simulate neural functionalities, such as light-triggered long-term/short-term plasticity. They can be used in intelligent robotics and bionic neurological optoelectronic systems. Metal oxide (MOx)–polymer hybrid nanocomposites can be beneficial as an active layer of the bistable metal–insulator–metal ReRAM devices, which hold promise for developing high-performance memory technology. This review explores the state of the art for developing memory storage, advancement in materials, and switching mechanisms for selecting the appropriate materials as active layers of ReRAMs to boost the ON/OFF ratio, flexibility, and memory density while lowering programming voltage. Furthermore, material design cum-synthesis strategies that greatly influence the overall performance of MOx–polymer hybrid nanocomposite ReRAMs and their performances are highlighted. Additionally, the recent progress of multifunctional optoelectronic MOx–polymer hybrid composites-based ReRAMs are explored as artificial synapses for neural networks to emulate neuromorphic visualization and memorize information. Finally, the challenges, limitations, and future outlooks of the fabrication of MOx–polymer hybrid composite ReRAMs over the conventional von Neumann computing systems are discussed.