Encapsulation-enhanced switching stability of MoS2 memristors

The advent of big data and machine learning has created an exploding demand for parallel processing. The lack of parallelism in von Neumann machines limits the processing of data in large quantities and with high dimensions. The memristor exhibits considerable potential for neuromorphic computing and has thus been studied as an alternative. Two-dimensional (2D) materials have been used to fabricate memristors with extreme density and ultralow power consumption. As 2D materials are environmentally susceptible, protection from the external environment is imperative to ensure reliable device operation. Herein, we apply an amorphous fluoropolymer barrier layer to molybdenum disulfide (MoS2)-based memristors, thereby enhancing the switching stability. The encapsulated devices exhibited an enhanced switching endurance of 100 cycles with analog switching behavior for 25 conductance states. The results of this study show that encapsulation can effectively suppress degradation and enhance the stability of MoS2 memristors from environment effects. (C) 2021 Elsevier B.V. All rights reserved.