The coincidence analysis of proton-proton (pp) scattering events from a MeV proton microprobe is an unique method for 3D hydrogen microscopy, where sub-ppm atomic sensitivity is achieved in light materials with μm depth and lateral resolution using 17 MeV protons. Up to 25 MeV protons can be used with a new detector system at the Munich microprobe SNAKE. Hence, the analysis of unsupported samples with high mass density or large thickness up to e.g. 50 μm tungsten, 150 μm silicon or 250 μm plastics becomes possible. At these energies, new challenges arise and are solved in this thesis for the dedicated setup at the Munich Tandem accelerator: (1) The optimum beam brightness and stability is required in particular for μm lateral resolution in large depth as well as for meeting reasonable measurement times. Therefore, a new multicusp ion source has been installed at the 14-MV Tandem accelerator with the brightness of up to B = 27 μAmm-2mrad-2MeV-1 at the space charge limit of 30 kV extraction potential. Beam transport calculations are performed and reveal the limits due to the Tandem accelerator stripper foil in conjunction with the intrinsic astigmatism and possible parasitic in uences on Bmax = 23.5 μAmm-2mrad-2MeV-1. It brings up the injection parameters and improvements for a future injection system to get an optimized phase space volume for an optimum brightness Bexp at the experiment. With the existing system, even Bexp = 2 μAmm-2mrad-2MeV-1 has been achieved. This is 20 times more than with the previously used proton source and equals the originally suggested design value for the microprobe SNAKE. Additionally, it was required to construct and install a new water-cooled beam micro slit system as an object aperture for SNAKE. It includes a possibility of tandem stability control as well as combined heating and cooling pipe system that minimizes vibrations and temperature dilatation to less than 1μm even up to 250W blockedbeam power. (2) At the data analysis, the efficiency of the angular and coincidence filtering has been calibrated using a special Monte-Carlo simulation code. This enables an accuracy better than 5% for the quantification of the hydrogen content independent of the thickness of the sample, the atomic number/composition or density. The simulation has been verified on multilayered sandwich targets, demonstrating a depth calibrating from the energy signal of the protons with an accuracy of better than 1% of the sample thickness. As a basic requirement, the energy dependent scattering cross section data have been evaluated with a precision of 0.2%. Furthermore, detectors energy resolution of better than 30 keV for 17MeV protons has been achieved with a new calibration method using elastic and inelastic scattering signals of thin foils. This allows correction of energy loss effects in the pixels of the detector and finally optimizes the sensitivity to the sub-ppm level. The system has been successfully applied for measuring the hydrogen content in buckled niobium hydrogen films showing enrichment in the buckels and confirming theory of hydrogen release with relevance to hydrogen storage devices.    «

The coincidence analysis of proton-proton (pp) scattering events from a MeV proton microprobe is an unique method for 3D hydrogen microscopy, where sub-ppm atomic sensitivity is achieved in light materials with μm depth and lateral resolution using 17 MeV protons. Up to 25 MeV protons can be used with a new detector system at the Munich microprobe SNAKE. Hence, the analysis of unsupported samples with high mass density or large thickness up to e.g. 50 μm tungsten, 150 μm silicon or 250 μm plasti...    »