Researchers from Kobe University have investigated how fabrication techniques influence the interface between graphene barriers and nickel-iron alloy electrodes in magnetic tunnel junctions (MTJs). These interfaces play a crucial role in determining the performance of spintronic devices, but their atomic structure and resulting electronic properties can vary significantly depending on how the materials are combined.
By comparing two main approaches - transferring graphene onto a nickel-iron substrate or depositing the alloy directly onto graphene - the team uncovered how the choice of process governs the stability of nickel-rich versus iron-rich surfaces, ultimately shaping the spin-dependent behavior of MTJs.
Using first-principles calculations supported by large-scale simulations on the Fugaku supercomputer, the Kobe researchers showed that for a bare nickel-iron substrate, nickel-rich surfaces are energetically more stable, regardless of composition. However, once graphene is adsorbed, the situation reverses: hybridization between the p orbitals of carbon atoms and the d orbitals of iron atoms stabilizes Fe-rich surfaces. This means that the graphene-transfer method tends to produce Ni-rich interfaces, while the alloy-deposition method favors Fe-rich ones.
This subtle but critical difference in surface composition directly impacts the electronic and magnetic properties of the junction. As carbon–iron interactions are stronger than carbon–nickel ones, the degree of orbital hybridization at the interface alters spin-dependent transport through the graphene barrier. As a result, the fabrication route itself provides a mechanism to tune device behavior, with direct implications for optimizing MTJs in both memory and sensing applications.
Importantly, the results extend beyond nickel-iron and graphene. By identifying a fundamental mechanism governing how ferromagnetic surfaces interact with two-dimensional materials, the study offers a principle that can be applied to other material systems as well. This establishes new opportunities for engineering high-performance spintronic devices, paving the way for faster, more energy-efficient electronics based on precisely tailored interfacial properties.