Abstract

Two-dimensional (2D) materials are promising candidates for the next generation of spintronic devices as they provide flat interfaces that embed many interesting physical effects. The properties of the magnetic tunnel junction (MTJ) such as the magnetism and tunnel magnetoresistance (TMR) effect are not only determined by the materials but also by the interfaces. For example, room temperature magnetism is required for devices based on 2D magnets. Due to the Mermin-Wagner theorem which requires magnetic anisotropy to sustain the magnetism for 2D spin-lattice, only a dozen 2D magnets are experimentally realized down to the monolayer limit, in which few magnets can sustain magnetism at room temperature. Therefore, finding a suitable substrate to contact the magnet, thereby modulating the magnetism is a promising way to realize high transition temperature. Here, we study both the conventional substrate and the 2D substrate. For example, Fe4GeTe2 is a quasi-2D ferromagnet with an intrinsic Curie temperature (TC) approaching 300K. We show that by contacting with sapphire 001 surface, the Curie temperature can rise to 530K. We identify the substrate recovered for the ferromagnetism. Apart from TC, we also show that the type of interface, such as physisorbed and chemisorbed interface, significantly impacts the spin transport. We investigate the TMR effect of h-BN/Co MTJ . TMR of MTJ with physisorbed interfaces is 1000 times higher than that of chemisorbed interfaces. Finally, we consider all van der Waals MTJ Fe4GeTe2/TMD/ Fe4GeTe2, where TMD stands for transition metal dichalcogenides. Without interfacial bonds altering the spin injection, we found the TMR can be known a priori via Brillouin zone filtering in hexagonal lattices, where the MTJ can be designed to achieve ultrahigh TMR .