3d Reconstructions of Micro-systems Using X-ray Tomographic Methods

During the last years tomographic X-ray investigation methods have been transferred from research centers to industrial labs and have been used for research and development as well as quality control. At the moment 3-dimensional tomographic methods are introduced into production for process survey and in-line control. Furthermore there is a trend towards highly accurate tomographic systems with resolutions down to several microns or even several hundreds of nanometers resulting in new demands for those systems. Region of interest reconstruction methods have to be applied to objects with large expansion in two dimensions in order to achieve a high resolution within a small volume. For this purpose problem optimized scanning and reconstruction algorithms are necessary. In this paper we present the state of the art of high-resolution industrial X-ray CT and laminography systems. As an example a new cost effective, small and fast tabletop mini 3D-CT system is shown. It allows the usage of μ-CT in many trades as e.g. seed production, or plastics, Al-casting or electronic industry. In combination with laminographic methods as tomosynthesis or swing-laminography this system can be used as well for the inspection of flat components. First results will be presented. Introduction: High demands on the quality of materials and technical components require nondestructive testing methods for the optimisation of production processes. The wish to determine, characterise and measure voluminous defects requires fast non-destructive testing methods with a high geometrical and contrast resolution. X-ray irradiation is well known as a non-destructive testing method for technical components. However using a simple irradiation technique there is no possibility to get any information about the depth of the imaged structures from a single projection. As early as twenty-two years after the discovery of X-rays it was mathematically shown that it is possible to calculate the density distribution of an object from its X-ray projections [1]. This led to the development of computed tomography (CT) in the medical field in the early 70. Since the 80 CT is also used for the examination of technical objects. Two dimensional CT (2D-CT) with a line detector reconstructs with one measurement – a complete rotation of the object – one slice of the object. An examination of the whole volume of an object with two-dimensional CT (2D-CT) is very time consuming due to the fact that a large number of object slices have to be measured and reconstructed one after each other. For each measurement the object has to be completely rotated. This difficulty can be overcome by the three-dimensional CT (3D-CT). Instead of a line detector a two-dimensional flat X-ray detector is used and only one rotation is necessary to obtain the information for the reconstruction of the whole object volume. So significant shorter inspection times can be reached compared with the situation in 2D-CT. The three-dimensional distribution of the linear absorption coefficient can be reconstructed from the projections using mathematical reconstruction algorithms. Optimisation and parallelisation of such algorithms can lead to fast reconstruction methods, which are even suitable for the examination of random samples in industrial applications. Figure 1 shows the principle of a 3D-CT measurement geometry. Volume z Rotation axis