Abstract

The advances in computing and mobile devices have realized massive generation, collection, and processing of data. The concept "Internet of things" and "Big data" have necessiated large-scale parallel processing, as computing of collected data is highly dependent on matrix multiplication process, which in inherently parallel.In this context, conventional machines are exhibiting their limits as they are unsuited for parallel processing, resulting in low performance and high power consumption [1]. The need for advances in parallel processing hardware is emerging.In this regard, memristors have gathered attention for their high potentials as parallel processing units. There have been reports of memristors with high on/off ratio, fine retention, high switching enderance, and fast switching spped, etc [2-4]. Memristor is a non-volatile memory which stores data as internal resistance, modulated by applied voltage. Even though memristors contain high capabilities for parallel processing, scaling of devices still remain as a challenge. Conventional materials have their limit for both lateral and vertical scaling.To achieve a major breakthrough, we have fabricated memristors based on 2D materials with atomic-scale thickness of ~ 7 nm [5]. As surface-to-volume ratio is high in 2D materials, they are chemically active, leading to susceptible properties to external environment and instability in fabricated devices with 2D materials. To compensate the drawback, we applied fluoropolymer passivation layer and obtained stable switching endurance for 100 potentiation & depression cycles with 25 states.For device fabrication, we used direct current sputtering & wet etching to pattern bottom Au/Cr (50/5 nm) electrodes. Molybdenum disulfide flakes were mechanically exfoliated from bulk mineral and dry-transferred onto bottom electrodes by PDMS (Polydimethylsiloxane) stamps. Top electrodes were patterned by photolithography & evaporation of Ni/Au (5/100 nm).With the introduction of 2D materials to next-generation electronics, memristors can be even more revolutionized towards extreme scaling of devices. By compensating susceptibility in the material itself and studying degradation mechanism, parallel computing would be realized away from power-plugged environment, and accelerate artificial intelligence in our everyday lives. Figure Caption Fig 1. Electrical characterization of the resistive switching MoS2 memristor with non-volatile memory behavior. (a) Semi-log plots of IV curve, resistive switching triggered by voltage sweeping. Inset: False-color SEM image of a fabricated device; scale bar: 4μm. (b) Scheme for pulse voltages (top), potentiation and depression of conductance states (bottom). (c) Encapsulation-enhanced switching stability of MoS2 memristors. Detailed illustration of 20 PD cycles in CYTOP-encapsulated (top) and bare devices (bottom).

Full Text
Paper version not known

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call

Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.