ABSTRACT
One of the most exciting, yet puzzling, questions in materials science and condensed matter physics today is whether one can detect, understand, and control macroscopic spin orderings in highly nonequilibrium, nonthermal states at femtosecond time scales. Such processes are at least 1000 times faster than those of the traditional thermal magnetic processes (Zvezdin and Kotov, 1997; Sugano and Kojima, 2000; Zhang et al., 2002), which set the limit of the magnetic switching time to 100 ps-10 ns, as used in the modern magneto-optical (MO) recording industry (Zvezdin and Kotov, 1997). Recently, there is growing evidence that, at the femtosecond time scale, coherent modication of magnetism is feasible, which involves direct coupling between the light eld and spins. Compelling evidence for such coherent ultrafast magnetism has been obtained from photoinduced demagnetization in nickel lms (Bigot et al., 2009), spin precession in orthoferrite DyFeO3 (Kimel et al., 2005), and spin reorientation in the ferromagnetic semiconductor GaMnAs (Kapetanakis et al., 2009; Wang et al., 2009). However, exactly how laser pulses can substantially modify the collective spin ordering in the coherent regime or induce even a complete spin reversal or a magnetic phase transition at ultrafast time scales remains controversial. Some of the most challenging issues in this temporal regime are the highly nonequilibrium processes and photoexcited coherences involved, since the relevant characteristic time scales are comparable to or even shorter than one oscillation cycle of phonons and magnons, spin-dependent scattering, and dephasing times. These raise some fundamental questions deep into the cornerstones of our current understanding of magnetism and phase transitions, for example, microscopic origin of angular momentum conservation and the validity of the thermodynamic description of magnetism at these extremely short time scales.
