Problematic and Objectives

Short fibre reinforced metals (SFRMs) are directly strengthened by means of an effective transfer of the applied load from the matrix to the rigid short fibres (SFs). In the case of aluminium-silicon (Al-Si) based SFRMs, the eutectic Si and the SFs form a three-dimensional hybrid network in which the SFs are connected by rigid Si bridges (see Fig. 1). The strength and stability of this structure is sensitive to the size and amount of the Si bridges. Furthermore, the morphology of the eutectic Si varies during high temperature exposure due to the diffusion driven spheroidisation of Si, resulting in a subsequent change of the interconnectivity of the hybrid Si-SFs structure. One of the reasons for the creep strengthening of short fibre reinforced Al-Si alloys during long term creep exposure is the Si diffusion increasing the interconnectivity of this Si-SF hybrid structure. The morphology and degree of interconnectivity of this hybrid structure can only be effectively revealed using three dimensional characterization methods such as X-Ray tomography. However, conventional X-ray tomography is not able to give the necessary contrast to distinguish Si from Al due to the practically identical X-ray absorption coefficienst of these two elements. On the other hand, synchrotron microtomography has proved to be a very powerful tool to obtain contrast of interfaces due to the high coherence of the X-ray beam (see Fig. 2). However, this phase contrast is still not enough to carry out an automatic segmentation of the Al and Si phases and a time consuming hand segmentation is the current methodology applied. The objective of the proposed diploma thesis is the development of an automatic segmentation tool that enables the segmentation between Si and Al in microtomographies obtained using synchrotron holotomography. The multidisciplinary nature of the work will involve a close collaboration with researchers from the institute of computer graphics and algorithms and from the institute of materials science and technology together with the research team of the European Synchrotron Radiation Facility in Grenoble in order to implement the necessary image analysis methods which shall help to understand the underlying physical phenomena observed experimentally. The developed methodology is expected to be generally applicable to the analysis of phase contrast tomographies of other multiphase materials.

Supervision

Institute of Computer Graphics and Algorithms: Prof. Eduard Gröller (groeller#cg.tuwien.ac.at)
Institute of Materials Science and Technology: Dr. Guillermo Requena
Institute of Materials Science and Technology: Prof. Hans Peter Degischer

Additional Information

Example-Video
Duration : 8 Months
Financial Support : Û2100

• Computer Graphics Group
• Courses
• Projekte
• Tomographic Segmentation of multiphase materials