Precisely measuring weakly radioactive samples by gamma-ray spectrometry requires optimizing the detection geometry and knowledge of the gamma-ray decay scheme. One can thus increase the counting rate and reduce the statistical uncertainty of the spectral peaks used to determine radioisotope activities. However, an increased sample volume requires a correction for the self-absorption of γ -rays in the sample itself, and approaching a sample to the detector gives rise to coincidence summing. MCNPX simulations permitted finding the separate influence of sample density and effective atomic number of the sample in the attenuation of photons with energies less than 100 keV. Peak-summing corrections were obtained with MCNPX, GESPCOR and ETNA. Thus a data base for 244 radionuclides could be established for SG50 and SG500 geometries in contact with a planar detector. In an application of the results to the health physics domain, construction materials were analyzed. Naturally-occurring Uranium-238, Thorium-232 and Potassium-40 activities were identified and corrected for the above-mentioned effects in order to evaluate the risk indexes, the absorbed dose and the annual effective dose received from different dimensions built of these materials. MCNPX simulations corroborated the model used to calculate the absorbed dose and gave its distribution in an enclosed space. The results obtained are within the recommended norms.
