Role of ferromagnetic spin structure in magnetization reversal and exchange bias phenomena

Abstract

Recently, the dependence of exchange bias (EB) on ferromagnetic layer thickness (t FM ) and temperature has become a matter of controversy, triggering renewed research efforts to decipher the key aspects of these intriguing phenomena. To demonstrate these anomalous dependences linked to the ferromagnetic spin structures, a modified Monte Carlo method is used on models of NiFe/antiferromagnet and Fe/antiferromagnet bilayers where a twisted configuration is favored along the film depth in the ferromagnetic layer during magnetization reversals. It is found that the EB field deviates from the reciprocal relation with t FM towards a lower value as t FM increases, and the maximum value of the EB field with temperature appears at 97 K for NiFe and 123 K for Fe, not the lowest temperature. The decrease in the EB field with t FM may also depend on the thickness of the antiferromagnetic layer and the twisted angle. The twisted configuration formed in a thick ferromagnetic layer encourages the propagation of domain walls through the ferromagnetic film to decrease the external switching field, leading to a suppressed EB. On the other hand, this twisted configuration simultaneously widens the domain walls and removes the unstable pinning points at relatively high temperatures. As a result, EB is recovered.

Spin‐Glass Irreversibility Temperature and Magnetic Stabilization in Ferromagnet/Spin‐Glass Bilayers

“…The hysteresis loops are recorded by cycling field between −3000 and 3000 Oe after field‐cooling from 1000 K under H FC = 3000 Oe. The time is measured by Monte‐Carlo step (MCS), where spin energies are calculated exactly to judge their evolved state, based on a modified Metropolis algorithm . For each temperature and field variation, 12 000 MCS are performed; the first 10 000 MCS are used to equilibrate system and then discarded, followed by 2000 MCS to average magnetization and magnetic energy quantities.…”

Prediction of reentering and switching ferromagnet/antiferromagnet exchange bias by antiferromagnetic proximity effect

“…For obtaining EB training, the hysteresis loop is repeatedly measured by a cycle index of n=20. The Monte Carlo Metropolis algorithm has been modified by explicitly calculating the spin energy in the framework of the Stoner-Wohlfarth (SW) model to determine the spin flipping probability, which can effectively prevent the non-physical spin flip by tunneling across the energy barrier [27][28][29][30][31][32][33]. A sufficient simulation time measured by Monte Carlo step is used to guarantee the thermal quasiequilibrium, that is, the 10 5 Monte Carlo steps are initially performed and then discarded to equilibrate the system, followed by another 10 5 Monte Carlo steps to calculate the magnetization and magnetic energy.…”

Spin glass properties mapped by coercivity in ferromagnet/spin glass bilayers

“…Then the magnetic hysteresis (m-h) loop is recorded by cycling H between −1 and 1 kOe in steps of |ΔH|=10 Oe. To update the spin state undergoing the processes of field cooling and isothermal magnetizing, a modified Monte Carlo Metropolis method is used [5,18,32,33,44]. In the standard Monte Carlo method, the flipping probability of a single moment was only determined by the energy difference between the trial and original states.…”

“…Subsequently, Jalil [30] promoted Xu et al's method at low temperature and conceived five subsidiary states around the minimum energy points as fluctuation states and Du et al [31] extended the fluctuation states to a continuous area as they should be. Finally, Hu et al [32,33] applied this method to interacting systems and have proven that the simulation results obtained are informative for interpreting the hysteresis loop behaviors of systems with finite anisotropies. However, to the best of our knowledge, the K SG in an SG and its role as well as potential application have scarcely been studied experimentally and theoretically.…”