Direct observation of ultrafast atomic motion using time-resolved X-ray diffraction
This thesis is dedicated to the study of the atomic motion in laser irradiated solids on a picosecond to subpicosecond time-scale using the time-resolved X-ray diffraction technique. In the second chapter, the laser system, the laser-plasma based X-ray source and the experimental setup for optical pump / X-ray probe measurements were presented. To perform the experiments described in this thesis, the construction of a new experimental setup for optical pump / X-ray probe measurements was required. The old experimental setup, due to its geometrical restrictions, did not allow the observation of some of the Bragg-reflections which are necessary for effective measurements of the transient Debye-Waller factor. In this chapter special attention is paid to the characterization of the used laser-plasma based X-ray source, because its parameters determine which applications the setup can be used for. The work on optimization of the laser system and the experimental setup allowed the experiments described in this thesis to be performed with high accuracy. The optical pump / X-ray probe experiments required in particular efficient recollection and focusing of the radiation of the laser-plasma based X-ray source because its radiation is emitted into the full solid angle. The X-ray focusing elements are therefore key elements in the experimental setup. Chapter 3 is devoted to the characterization and comparison of different types of X-ray optics. For each optic the amount and distribution of the X-rays in the focal plane was determined. It was found that all the optics tested can focus the X-ray radiation into a spot with the size (FWHM) of 100-200 μm. The measured fluxes of X-rays in the foci of all the optics were large enough to perform TRXD experiments. It was shown that for each type of experiment it is possible to choose a suitable kind of optic. In particular for the experiments where the measurements of the X-ray reflection profiles (rocking curves) are of interest, the toroidal mirrors are best suited to fulfill the role of the focusing optic due to their narrow bandwidth (thus high angular resolution) and large convergence angle of the focused X-rays. Multilayer mirrors and capillary optics focus typically the whole Ka radiation of the X-ray source including the Ka1 and Ka2 lines. Therefore, they are suitable if one is interested in the integral intensity of the diffracted signal from the sample under investigation but the profile of the rocking curve is not of interest. As multilayer mirrors and capillary optics have a relatively small convergence angle (due to their large magnification) they can be used in experiments where no large shifts of the rocking curves are expected. It turned out, that the Ge (400) toroidal mirror for Ti-Ka radiation, due to its relatively large bandwidth provides a good compromise when both the integrated intensity and the angular distribution of the diffracted signal are of interest. This mirror was used for the experiments discussed in this thesis. Chapter 4 presented the time-resolved X-ray diffraction experiments performed for this thesis. The first section of this chapter discusses the measurements of initially unexpected strain-induced transient changes of the integrated reflectivity of the X-ray probe beam. These changes should be taken into consideration in any type of experiment in which an X-ray beam probes a spatially inhomogeneously strained sample, otherwise they can mask other physical effects such as the Debye-Waller factor. In particular, this effect should be carefully considered in any optical pump / X-ray probe experiment where the size of the pump beam differs from the size of the probe beam by less than at least one order of magnitude. The elimination of the strain-induced transient changes of the integrated reflectivity described in the first section of chapter 4 represented an important prerequisite to perform the study of lattice heating in Germanium after femtosecond optical excitation by measuring the transient Debye-Waller effect. These measurements are described in the second section of chapter 4. It was found that the energy of the excited electrons is transferred to the lattice in just over one picosecond. The process of electron-to-lattice energy transfer had been investigated previously by the observation of changes in the electronic system of the material. The experiments discussed here demonstrate that TRXD enables us to investigate this process also “from the lattice point of view”, thus completing our understanding of the energy relaxation in solids after optical excitation. The third section describes the investigations of acoustic waves upon ultrafast optical excitation and discusses the two different pressure contributions driving them: the thermal and the electronic ones. The experiments performed here made it possible to estimate the relative strength of the electronic and thermal pressure contributions. The values of the strength and the decay time of the electronic pressure obtained from the experiments provided clarification on some seeming contradictions in the measurements on acoustic phonons in Germanium discussed in the literature.