Laser-based stroboscopic Kerr microscopy
The laser-based setup of our time-resolved wide-field Kerr microscope uses a pulsed laser to illuminate the sample in a stroboscopic way. We are using a mode-locked solid-state Nd:YVO4 laser that produces 12 ps wide laser pulses at a repetition rate of 23 MHz for this purpose. These laser pulses are sent through a rotating glass disc and focussed into a multi-mode glass fibre (as shown in Fig. 1). The other end of the glass fibre replaces the standard Xe arc lamp, which is used for quasi-static and camera-based time-resolved Kerr microscopy. The nearly point-like light source is focussed into the back focal plane of the objective, to insure a parallel incidence of light onto the sample. The coherency of the laser light causes disturbing interference and speckle patterns on the sample. This effect is minimized by the rotating glass disc, which moves the speckle pattern during the integration time of the CCD camera. This results in a more homogeneously illuminated image.
Fig. 1: Setup of the laser-based stroboscopic Kerr microscopy
The excitation of the samples can not be achieved by the usage of coils, as the induction in the windings of the coils reduces the rise time of the applied fields. Therefore we use coplanar waveguides that are capable to transmit high-frequency voltage and current pulses. The current pulse travelling along the waveguide produces a pulsed magnetic field around the center conductor, which can be used to excite high-frequency magnetization processes. The samples are deposited on top of the coplanar. To ensure an excitation of the samples that is in-phase with the illuminating laser pulses a photodiode picks up a small portion of the laser pulses and delivers the trigger signal for the pulse generator. A single shot imaging is not possible because of two reasons: (i) the signal-to-noise ratio of an image acquired with one single laser pulse is too low and (ii) the single shot images have to be produced at a rate of more than 1 GHz to reproduce the magnetization process and this is orders of magnitude too fast for any kind of optical camera.
Like in the experiment with the camera-based stroboscope only those processes are visible that are reproducible within the repetition rate of the experiment (in this experiment determined by the laser frequency of 23 MHz). An example image series acquired with the laser-based setup can be found here.