Abstract
Computer Axial Tomography (CAT) maps the distribution of the linear attenuation coefficient (LAC) over a section or over an entire object. LAC represents the product between density and mass attenuation coefficient (MAC). These parameters are influenced in complete different manners by porosity provided that pore sizes are much smaller than voxels: density lineally decreases with porosity while MAC is almost independent of it. Consequently, in the case of single energy tomography, CAT images will be strongly influenced by the porosity. In the case of dual-energy tomography, when it results for the same section two CAT images, one depicting density and the other one the effective atomic number distributions, only the first one will reflect the sample porosity while the second one remains independent of the specimen porosity. This fact could be very useful in investigating complex samples such as rocks or soils.