Synchrotron radiation microcomputer tomography (SRμCT) is an extension to conventional computer tomography. The monochromatic nature of the synchrotron light and its easy tunability (from "soft" to "hard" X-rays, e.g. 7-200 keV) lead to fewer artifacts than for the case of conventional X-ray tubes and allow an easy adjustment of the energy of the applied X-radiation to the absorption properties of the object under study. Objects with a low degree of mineralization (such as collagen sponges) have a low X-ray density and therefore require X-radiation with lower energy, and vice versa for highly mineralized objects (such as teeth). A high spatial resolution of about 1/1000 of the size of the sample is achieved, down to 1 μm or less. In the case of biominerals, this method is especially useful because the high X-ray absorption contrast between mineralized and nonmineralized tissue makes both easily distinguishable. Typical biominerals such as calcium carbonate (CaCO3) and calcium phosphate (like apatite, Ca5(PO4)3OH) show a high X-ray absorption which is very convenient to separate them from surrounding organic matter, such as collagen, chitin, etc. By its nature this method does not require a preparation of the sample, except for a mechanical fixation in the case of delicate or brittle objects. The latter is necessary to avoid a movement of the sample during the experiment. This fact makes it complementary to electron microscopy and histology, which both require mechanical cutting of the object and also preparation steps. These preparations may lead to artifacts which are difficult to describe and to predict. If the X-ray absorption data are quantitatively evaluated, it is possible to compute physical parameters such as linear X-ray attenuation coefficients. These are related to the composition of a given volume element, i.e. to the ratio of (highly absorbing) inorganic mineral and (low absorbing) organic matrix. The fact that each voxel of a given data set is associated with a specific X-ray density (i.e., a gray-value) allows a computation of the volume of a biomineralized structure with a given X-ray density within a sample. This allows us to derive the absolute volume of this hard tissue in this sample and to draw conclusions about dynamic phenomena (such as the volume growth of this tissue with age). In principle, this is also possible without quantitative calibration of the X-ray density, i.e. with relative values. Nevertheless, the quantitative evaluation of microcomputer tomographic data sets is far from being routine, due to the complex shapes of the objects and the large amount of data. Image analysis procedures have to be applied to obtain quantitative data, e.g. on pore size distribution, object size, and orientation, etc. We are convinced that this method will be of high future value to study the spatially different mineralization processes in mineralizing animals (and plants). This will be especially useful if dynamic effects are studied, like the initial formation of mineralized tissues or re- and demineralization processes which occur during the lifetime of an animal. For this, animals of different stages of development or of different age have to be investigated. Due to the necessary measuring time of a few hours at best, in situ monitoring of structural changes appears to be impossible at the moment, although this certainly would be of high value for rapidly developing organisms (such as mollusks). © 2008 American Chemical Society.
