A simulation-driven spectrometric method to determine α particle attenuation in air filters

International audienceThe analysis of radioactivity in air is often performed on filters in order to collect large amounts of radionuclides on a small surface and therefore, increase the measurement sensitivity. The same approach has been used for decades in the field of volcanology, where 210Po activities are commonly analyzed in volcanic aerosols collected on filters in order to better understand magma processes at depth. However, the attenuation of particles interacting with the filter matrix generally lowers the detection efficiency, which needs to be corrected in order to obtain accurate activities. In this contribution, we present a correction method performing a decomposition of the energy spectrum recorded by the detector over a base of simulated spectra corresponding to different depths of particle emission. These simulations are obtained by using the GATE code in order to describe both the detector environment and particles trajectories and interactions. The method based on the retrieval of the information conveyed by the spectra uses a regular linear inversion process. We validated this approach using natural radioactive aerosols of radon daughters ( 214Po) filtered on two different types of media (PTFE and cellulose acetate) and we found a good agreement between the activity corrected for attenuation using the spectrum and the total activity determined from simultaneous spectrometry of 214Bi, in radioactive equilibrium with its very short half-life daughter 214Po