Exchange bias interactions in polycrystalline/amorphous bilayers

For technologically relevant systems of polycrystalline antiferromagnetic layer coupled to an amorphous ferromagnetic layer, quantitative models and micromagnetic simulations are challenging due to inherent structural differences. We present a numerical study, performed using a surface integral technique with finite element micromagnetic simulations, that can incorporate structural and magnetic parameters such as grain crystallography, mixed spin-interface coupling and granular stability, arising from grain volume and anisotropy. We show that this model is in good agreement with experimental results for exchange bias and coercive fields as well as the training effect.