A Novel Neural-Based Design of Graphene and Molybdenum Disulfide Magnetic Tunnel Junction

Abstract

Magnetic tunneling junction (MTJ) research is currently growing dramatically due to the progress of spintronic devices. Due to the varying electrode and tunneling behaviors, it is critical to conduct a more accurate analysis of MTJ behavior also MTJ has limited behavior analysis research. To accurately analyze the behavior of MTJ a novel design of Graphene and Molybdenum Disulfide Magnetic Tunnel Junction is proposed. Initially, Graphene–MoS2–Graphene MTJ is designed, in this, graphene acts as the electrode and MoS_2 acts as the tunneling junction, which gains a better efficiency score, high TMR ratio, and current transmission range. Furthermore, to predict the magnetic and electrical properties of the designed MTJ, the intelligent behavior analysis strategy known as the radial basis magnetic behavior framework (RBMBF) is implemented, which makes the computations faster and provides better characterization. To analyze the implemented MTJ, it is developed in the MATLAB environment and trained on the RBMBF. Finally, the calculated metrics are compared to those of standard MTJs, and improvement scores are recorded. The intended model has a better transmission coefficient of 1.7 and a TMR ratio of 586% than the previous models.