Exchange scattering as the driving force for ultrafast all-optical and bias-controlled reversal in ferrimagnetic metallic structures

Abstract

Experimentally observed ultrafast all-optical magnetization reversal in ferrimagnetic metals and heterostructures based on antiferromagnetically coupled ferromagnetic d− and f −metallic layers relies on intricate energy and angular momentum flow between electrons, phonons and spins. Here we treat the problem of angular momentum transfer in the course of ultrafast laser-induced dynamics in a ferrimagnetic metallic system using microscopical approach based on the system of rate equations. We show that the magnetization reversal is supported by a coupling of d− and f − subsystems to delocalized s− or p− electrons. The latter can transfer spin between the two subsystems in an incoherent way owing to the (s; p) − (d; f ) exchange scattering. Since the effect of the external excitation in this process is reduced to the transient heating of the mobile electron subsystem, we also discuss possibility to trigger the magnetization reversal by applying a voltage bias pulse to antiferromagnetically coupled metallic ferromagnetic layers embedded in point contact or tunneling structures. We argue that such devices allow controlling reversal with high accuracy. We also suggest to use the anomalous Hall effect to register the reversal, thus playing a role of reading probes.

Role of electron and phonon temperatures in the helicity-independent all-optical switching of GdFeCo

“…However, in these models it is often clearly claimed 5,8,19 that initially T e needs to quickly overcome T C in order to decouple the sublatices and allow a faster demagnetization of the Fe sublattice. Other microscopic models 23,24 that treat the energy and angular momentum exchange through scattering processes, also reach similar conclusions, and state the necessity of short and intense pulses for the initial demagnetization of both sublattices to happen at different rates.…”

“…6, for ∆t = 12.5 ps pulses, T e will only be heated to ∼ 530 K. We are not able to exactly determine whether T e needs to reach T C or not, due to the uncertainty (∼ 20%) in the critical fluence. Despite this open question, our result raises questions on the proposed scenario where very high electron temperatures (1000 − 2000 K) are necessary for AOS 5,8,[19][20][21][22][23][24] . We posit that helicity independent switching is a three step process, where there is no need for high electron temperatures.…”

“…For example, in HAMR, the lattice temperature T p of the system is heated close to the critical Curie temperature T C to reduce the anisotropy before an applied field favors a particular direction for the magnetization upon cooling 1 . In contrast, AOS models for ferrimagnets 5,8,[19][20][21][22][23][24] do not require the lattice temperature of the film to approach the Curie temperature. Instead, these models rely on transient electron temperatures that are out of equilibrium with the lattice to induce the dynamics of the Gd and Fe magnetic sublattices 5 .…”

“…Crossing T C would imply a loss of memory of the magnetic order. The final magnetization would then be determined by the cooling conditions, analogous to HAMR, which is in contrast with what most AOS models assume 5,8,[19][20][21][22][23][24] . In this work, we carefully measure F C for the helicityindependent AOS of GdFeCo films, through single shot switching and stroboscopic pump-probe experiments.…”

Electric bias-controlled switching of magnetization of ferrimagnetically coupled Mn delta-layers in a GaAs-AlGaAs quantum well

“…In order to gain insight into microscopical nature of the all-optical switching, dissipationless energy and angular momentum exchange between TM and RE sublattices mediated by 5d-4f coupling in RE ions has been explored in [20], and the exchange electron-electron scattering as the driving mechanism of the magnetization reversal was discussed in [21]. In [22] a general microscopic approach based on the rate equations was suggested for addressing the problem of the angular momentum exchange between two nonequivalent magnetic sublattices in a metal. In the latter work, the exchange scattering was found to be the driving mechanism of the switching.…”