A sub-500 mV monolayer hexagonal boron nitride based memory device

The recent discovery of memristive devices based on two-dimensional materials have attracted much interest for emerging applications on flexible memory, neuromorphic computing, and so forth. Reducing the thickness to a single-layer level would prompt the scaling limit to sub-nanometer. However, mono...

Full description

Bibliographic Details
Main Authors: Jun Ge, Haiming Huang, Zelin Ma, Weilong Chen, Xucheng Cao, Huaheng Fang, Jianfeng Yan, Zhiyu Liu, Weiliang Wang, Shusheng Pan
Format: Article
Language:English
Published: Elsevier 2021-01-01
Series:Materials & Design
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S0264127520309023
Description
Summary:The recent discovery of memristive devices based on two-dimensional materials have attracted much interest for emerging applications on flexible memory, neuromorphic computing, and so forth. Reducing the thickness to a single-layer level would prompt the scaling limit to sub-nanometer. However, monolayer materials based vertical memristive devices generally suffer inferior performance with high operating voltage, large leakage currents, and poor reliability. In this study, an interfacial polymer layer is inserted between the monolayer hexagonal boron nitride (h-BN) and top electrodes, which not only helps to constrain the conducting filament size but also block the formation of excess filaments from the bottom Cu foil. Therefore, the device shows stable bipolar resistive switching behavior with low operating voltage (< 500 mV), large on/off ratio (up to 105), long retention time (> 105 s), and excellent flexibility. It is demonstrated that tunneling conduction is shown in off-state and on-state current conducts via metallic conducting filaments, which are formed by the substitute of metal ions for lattice vacancies in h-BN. This work presents a scalable interface engineering strategy to control the interactions between metal ions and defects in monolayer h-BN films and sheds light on their promising application for large-scale integrated ultrathin flexible memory.
ISSN:0264-1275