Conduction mechanisms and thermally assisted multilevel switching in MOCVD-grown MoS2 resistive random-access memory devices
Year: 2026
Type: article
Abstract: Nonvolatile resistive random-access memory based on memristive switching has emerged as promising technology for next-generation computing and artificial intelligence hardware. Two-dimensional (2D) materials offer unique advantages due to their layered structure, uniform thickness, and defect-tunable properties, making them strong candidates for high-performance memory devices. Metalorganic chemical vapor deposition (MOCVD) provides large-area, wafer-compatible growth of 2D materials, offering a scalable path beyond small, exfoliated flakes. Here, we report multilevel resistive switching in MOCVD-grown multilayer MoS2 memory devices. The devices employ a metal–insulator–metal (MIM) configuration, where an 8 nm-thick MoS2 film is sandwiched between Ti top and Au bottom electrodes. Stable intermediate resistance states are obtained through controlled partial modulation of the conductive pathway during the RESET process. Temperature-dependent DC sweeps are used to analyze the underlying conduction mechanism. In addition, RF transient pulse measurements capture the nanosecond dynamics of the RESET transition and shed light on the mechanism of dissolution of the conductive filament. Statistical cycle-to-cycle analysis further confirms the reproducibility and uniformity of the multilevel switching behavior. These results highlight the promise of MOCVD-grown MoS2 as a platform for high-density multilevel memory storage applications.
Source: Journal of Applied Physics
Authors: Siyu Wu, Pranav Rama, Luke Sloan, Johnny Shen, Jatin V. Singh, Keunhwi Cho, Jaehyun Ahn, Dong-Won Kim, Sanjay K. Banerjee
Language: English
Subfield: Electrical and Electronic Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: hybrid
Estimated APC paid: $3,500
Funder: Samsung
OpenAlex ID: W7160180185