Resistive switching in emerging materials and their characteristics for neuromorphic computing

Abstract

Resistive random access memory would be an important component of microelectronics in the era of big data storage due to its efficient characteristics such as low cost, fast operating speed, low power consumption, and high performance in respect of endurance and retention. In this present review, we have focused on surveying three physical mechanisms which lead to resistive switching: electrochemical metallization, valence change mechanism, and ferroelectric polarization. A detailed discussion has been carried out on how these physical mechanisms work in various materials used for resistive switching based on nonvolatile random access memory elements such as oxides, ferroelectric, chalcogenides, polymers, graphene-based resistive switching, etc. The desirable electrical and optical properties for the representation of analog resistive switching in neuromorphic computing have also been discussed. An extensive report has examined the device requirement of different materials for artificial memristors.