Enhancing Resistive Switching Characteristics of MoS₂-Based Memristor Through O₂ Plasma Irradiation-Induced Defects

In this work, we report the tuning of resistive switching properties in MoS2-based memristors through varying durations of oxygen plasma exposure. Electrical transport measurements of untreated and plasma-treated memristors reveal that longer oxygen plasma treatment increases the low resistance state (LRS) current and memory window (MW) and reduces the SET/RESET voltages. Notably, graphene/MoS<inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${}_{\text {2-x}}$ </tex-math></inline-formula>Ox/graphene memristor with 18% oxygen atoms exhibits a MW <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$({{{\textrm {I}}_{\textrm {LRS}}}{\textrm {/}}{{\textrm {I}}_{\textrm {HRS}}}})$ </tex-math></inline-formula> of approximately <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$10^{2}$ </tex-math></inline-formula>, which is around <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$78.12\times $ </tex-math></inline-formula> higher than that of untreated memristors. Despite the switching performance benefits, the excessive oxidation of MoS2 via <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$O_{2}$ </tex-math></inline-formula> plasma leads to increased cycle-to-cycle variability and enhanced temperature dependency on the resistive switching characteristics. Furthermore, the near-linear modulation of conductance during long-term potentiation (LTP) and long-term depression (LTD) in MoS2-based memristors provides the excellent recognition accuracy of approximately 90% in a <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$784\times 100\times 10$ </tex-math></inline-formula> neural network for the handwritten MNIST dataset. The comprehensive performance evaluation of oxygen-irradiated MoS2-based memristors highlights their potential for large-scale crossbar memory and synaptic arrays. This study could also serves as a valuable reference for future research on controlled defect generation in memristive devices, paving the way for advancements in neuromorphic computing architectures.