Abstract
If the sample and the calibration standard are identical in geometry and in density, many laboratories of radioactivity measurement using gamma spectrometry, do not proceed to the correction of the effect of the possible difference in chemical composition between them. Some laboratories proceed to the correction of this effect using the usual simple ratio between the self absorption factor of the sample and the calibration standard based on the linear attenuation coefficients obtained using transmission experiments. In this work we studied this question by conducting Monte Carlo simulation of several typical environmental samples with different chemical compositions. We found that at low energy (E < 100 keV), chemical composition effect must be corrected. We found also that its correction using the usual simple ratio between the self absorption coefficient of the sample and the calibration standard is not useful for that purpose.