UvA-DARE is a service provided by the library of the University of Amsterdam (https://dare.uva.nl)
UvA-DARE (Digital Academic Repository)
Static and dynamic X-ray resonant magnetic scattering studies on magnetic
domains
Soriano, J.M.
Publication date
2005
Link to publication
Citation for published version (APA):
Soriano, J. M. (2005). Static and dynamic X-ray resonant magnetic scattering studies on
magnetic domains.
General rights
It is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons).
Disclaimer/Complaints regulations
If you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, stating your reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Ask the Library: https://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam, The Netherlands. You will be contacted as soon as possible.
7 7
C O N C L U S I O NN AND O U T L O O K
Magneto-opticall techniques have been extremely important in magnetic research,, especially in magnetic domain studies and ultrafast dynamical stud-ies.. This is mainly due to their versatility: they can easily be combined with highh magnetic fields and a wide range of temperatures. The discovery of the strongg magneto-optical effects in the X-ray range greatly extended the possi-bilities,, at first through the discovery of magneto-optical sum rules and more recentlyy through the exploitation of the short wavelengths which offer in prin-ciplee much higher spatial resolution.
Thiss thesis explores the novel technique of X-ray Resonant Magnetic Scatteringg in transmission mode in studies of the static domain structures in thinn films as well as their dynamical evolution. Before concluding this thesis withh a summary of the results, we think it is worthwhile to briefly assess the experimentall technique, in particular in comparison with the most important otherr optical techniques: transmission X-ray microscopy as regards spatial res-olution,, and the magneto-optical Kerr effect as regards magnetization measure-ments. .
Onee of the most important technical aspects of XRMS in transmission geometryy is that it is, in principle, a very straightforward technique. In the pi-oneeringg phase in which the experiments described here were performed, this simplicityy was often offset by the burden of the detailed alignment of the setup onn the beamline at the start of each experiment. The beamline on which these experimentss were performed is a fantastic spectroscopy beamline with high flux andd perfect control over the polarization, but it was not designed with scattering
114 4
inn mind. With a stable soft X-ray small angle scattering station on a purpose-buildd beamline, with well characterized collimators, beam stops, detectors, etc., itt is possible to investigate large volumes of samples relatively easily [63, 68], Wee will briefly venture to access the merits of such a system in comparison with aa transmission X-ray microscopy beamline.
Thee transmission geometry renders both techniques sensitive to the mag-netizationn profiles integrated over the film thickness. Transmission limits the thicknesss of the films to about hundred nanometer and requires the use of ul-trathinn S13N4 membranes as substrate, which clearly limits the number of sys-temss that can be studied, a shortcoming that is shared with Lorentz transmis-sionn electron microscopy and transmission X-ray microscopy. Although XRMS iss also used in reflectivity geometry, the detailed interpretation of the results is correspondinglyy more complicated in that case.
Thee strongest resonances (transition-metal L23 and rare-earth M4 5 edges) occurr in the soft X-ray range at wavelengths between 1.5 and 0.8 nm. In theory, XRM55 therefore has an excellent spatial resolution, given by the diffraction limit off the wavelength of the resonant X rays. Indeed, this resolution has been at-tainedd in XRMS studies of electronic superstructures in correlated systems such ass multilayers [111] and layered manganites [188]. However, as in any scatter-ingg experiment, in order to exploit that resolution, the scattered intensity has too be measured up to large scattering angles. More importantly, in scattering experimentss the extractable information strongly depends on the order of the sample.. In magnetic domain studies, the best resolution is obtained in highly orderedd parallel stripe domain lattices [13, 18, 67, 72, 92]. In this case it is pos-siblee to obtain information on the structure of domain walls with a resolution off 10 to 25 nm which can be compared with micromagnetic calculations. In the casee of the disordered domain structures that are discussed in this thesis, the resolutionn is determined by the disorder, and the average structure can be re-solvedd with some 50 nm resolution at best, which is more than two times worse thann the resolution of a transmission X-ray microscope.
TXMM currently has a resolution of some 25 nm, still very far from the diffractionn limit, and this was the main stimulus for our use of XRMS. However, thee development of the Fresnel zone plate lenses used in them is still continu-ing.. More importantly, these lenses produce a direct image and one does not
havee to take recourse to modelling the measured intensities as is necessary in XRMS. .
Anotherr much quoted advantage of X-ray magneto-optical techniques iss their chemical sensitivity, which allows one to measure sublattice magneti-zationss with great accuracy [189, 125]. This should be qualified to some ex-tentt since, for instance, most alloyed and multilayer systems have fairly well definedd magneto-optical spectral features in the infrared to ultraviolet range, whichh can also be used to separate the magnetic contributions of different sub-latticess in multicomponent systems to some extent. For static spatially-resolved X-rayy techniques like resonant scattering or microscopy, the chemical sensitivity iss not very important, as the different sublattices are linked by exchange inter-actionss and the information obtained at different resonances is the same.
Wee conclude that the unique strength of XRMS in comparison to MOKE andd TXM is in dynamic nucleation studies in homogeneous systems, whe-e the nucleationn can appear randomly, which means that imaging techniques would nott observe any structure under stroboscopic illumination. The reverse side off this maximum disorder situation is that the amount of information that can bee extracted is low. In more ordered samples, such as patterned thin films or nanostructures,, the dynamics is more reproducible, and transmission X-ray mi-croscopyy may be more useful.
7.1.. Outlook
Thee main goal of this thesis project was to study domain nucleation dy-namicss in homogeneous systems using time-resolved XRMS. Overall this goal hass been achieved. We have shown that even with the used tiny beams and single-bunchh operation of the synchrotron, the magnetization and the total scat-teredd intensity can be followed with a resolution of 100 ps. Preliminary data on thee magnetic correlation lengths relate mainly to the interdomain distances. An importantt objective was to follow the form factor of the nucleating domains. Thiss aim could not yet be achieved due to lack of flux and poor detector sensi-tivity.. However, vast improvements in data quality are still possible by using a moree optimized setup.
evi-116 6 C H A P T E RR 7
dencee for spin reorientation transitions of the subnetworks in the ferrimagnetic GdFee film, superimposed on the reversal dynamics. This is an interesting re-searchh area in itself and, to our knowledge, no prior results on spin reorienta-tionn transitions on these timescales are available. Ironically, insufficient time meantt we could not provide the direct proof of this feature in this thesis and we hadd to use magneto-optical Kerr effect data to obtain the information on the Fe sublattice.. Nonetheless, we think we do have proven that the potential for spin reorientationn transitions studies using the very intense pulses of microcoils is huge. .
Withh the aim of increasing the amount of order in the nucleation land-scape,, we patterned the thin films with very low doses of Ga+ ions using a focused-ion-beamm system. Although dynamic experiments could not yet be performed,, the influence of these nucleation centers on the diffraction pattern inn static fields is reported. Such anisotropy-engineered systems present many interestingg phenomena in their own right, which are the subject of a follow-up studyy using field-dependent MFM and transmission X-ray microscopy, the lat-terr ultimately also in a pump-probe mode.
Finally,, an experiment on perpendicular exchange-bias systems shows thatt even very thin samples can be studied and that, even in transmission ge-ometry,, the sensitivity is sufficient to measure uncompensated interfacial spins whichh play a decisive role in the exchange bias process. Here the transmission geometryy has the advantage that the magnetic contributions to the magnetic momentt of the uncompensated spins can in principle be measured in an abso-lutee way.
Time-resolvedd dynamical studies using visible light have a tradition of somee 30 years, and clearly pulsed lasers outperform the third-generation syn-chrotronss in terms of flux per pulse and pulse duration. This means that time-resolvedd XRMS is not able to access the forefront of magnetization dynamics att present, which addresses coherent precessional switching. The development off X-ray Free Electron Lasers (XFEL) over the coming decade will completely changee this situation. It can be expected that there it will be possible to com-binee 100 fs time resolution with nanometer spatial resolution. For those days too come, it may become possible to access spin-magnon interactions during do-mainn nucleation.
