Optical methods for exploring ultrafast magnetization reversal and spin-wave dynamics

The work presented in this thesis concerns the investigation of magnetization dynamics on the sub-picosecond timescale using optical methods. We explore two research areas: all-optical switching, which is the reversal of the magnetization by an ultrashort optical excitation, and propagating-spin-wave spectroscopy, a technique that can probe spin waves - periodic wave-like excitations in magnetic materials. In both areas we introduce new methodologies, and find that the underlying physics is governed by a competition between exchange interactions on the one hand and effective fields, such as dipolar fields, on the other. The field of ultrafast magnetization dynamics emerged with the discovery of ultrafast demagnetization, in which an ultrashort (fs) laser pulse can give rise to a (partial) loss of the magnetic order on a sub-picosecond timescale. Later it was found that it is not only possible to temporarily alter the magnetic state, but that in certain materials the magnetization can be completely switched using a single laser pulse, an effect dubbed all-optical switching (AOS). Magnetic materials have long been used for digital data storage, and it was quickly recognized that these findings could pave the way for a new generation of ultrafast, energy-efficient data storage devices. Further research uncovered that there are two different types of AOS, helicity-independent and helicity-dependent AOS. In the former a single pulse is enough to switch the magnetization, which has a toggle switching behavior where every subsequent pulse reverses the magnetization. It is generally found in ferrimagnetic alloys and bilayers, which have two sublattices that point in opposite directions. The process is governed by the exchange interaction and the transfer of spin angular momentum between these two sublattices. For helicity-dependent AOS, multiple pulses are needed to achieve switching, and the direction of the switch is determined by the helicity of the light. It is found in a wider variety of materials, including ferromagnetic thin films and granular media. %The mechanism behind it has long been poorly understood, with two competing explanations being given---one relying on IFE, a field generated in the magnetic layer that causes switching, the other relying on MCD, a difference in absorption between regions where the magnetization is up or down, that indirectly causes switching. Other work, more phenomenological, has found a criterion that links the stable domain size in a material to the ability to achieve HD-AOS. In chapter 4 we investigate helicity-independent AOS in Co/Gd bilayers, which are known to exhibit toggle switching. We first introduce a novel addition to the general TR-MOKE setup, which allows us to measure the switching dynamics of Co/Gd bilayers without the need for an applied magnetic field. We use this method to show that the switching in Co/Gd happens at the same ultrafast timescale as in GdFeCo alloys. Furthermore, we examine the influence of applied fields on the switching behavior, resolving an interesting competition between exchange-driven longitudinal switching and field-driven precessional effects. We find that fields of several hundreds of mT influence the dynamics within ~50 ps. We look into three different mechanisms that can explain this phenomenon, and conclude that precessional switching is the most important contribution. In chapter 5 we focus on the second type of switching, helicity-dependent AOS. We investigate the domain patterns that arise after sweeping a train of laser pulses over a Pt/Co/Pt sample. When using linearly polarized light, we find two characteristic domain sizes in our sweeps, both resulting from a competition between exchange and dipolar contributions: small domains, in a state of lowest magnetostatic energy, in the end of the sweep and larger domains, in a metastable state, in the rest of the sweep. We then move to circularly polarized light, and examine the influence of repetition rate, fluence and Co thickness. Our results are generally in line with models used in literature to explain helicity-dependent AOS, but we provide a minor clarification regarding a criterion presented in literature that links the stable domain size and laser spot size to the appearance of helicity-dependent AOS. We then turn our focus towards propagating spin waves. First predicted in the early twentieth century, spin waves have long been of interest for both their fundamental properties and applications. Due to their distinct advantages—frequencies up to the THz range, low power consumption due to the lack of Joule heating and access to wave-based computing—they are nowadays seen as a candidate for data carriers in a new generation of computing devices. While magnonic logic gates are already realized experimentally, a major challenge concerns the miniaturization of these devices, specifically regarding the generation and detection of spin waves with very small wavelengths, at the transition between being dominated by the dipolar coupling and the exchange interaction. In chapter 6 we propose a new type of spin wave detection scheme, using ultrafast laser pulses, in which it is possible to optically detect spin waves with a wavelength smaller than a diffraction-limited laser spot size. Using a metallic grating we can selectively detect the magnetization, allowing for small-wavelength, k-selective, spatially resolved spin wave detection. We also present preliminary measurements in which this method is used to determine the strength of the interfacial Dzyaloshinskii–Moriya interaction, an antisymmetric exchange interaction that, among other things, can stabilize skyrmions. The work presented in this thesis provides new insights in the fields of all-optical switching and spin waves, which bring a new generation of logic and storage devices one step closer to reality.