Calibration of parallel-plate chambers: Resolution of several problems by using Monte Carlo calculations

Publisher Website
Google Scholar
Abstract
For five commonly used parallel-plate ion chambers, Monte Carlo calculations are presented of the wall attenuation and scatter correction factors (Kwall = Awall-1 or katt-1) and the correction for nonhomogeneous composition of the chamber (Kcomp or related to km). The chambers are assumed to have 0.5 g/cm2 buildup caps made of the predominant material in each chamber. These correction factors are needed if air-kerma calibration factors are used to deduce the chamber's cavity-gas calibration factor, Ngas. The scatter from the material around the cavity more than compensates for the attenuation in the front wall and hence Kwall values are less than unity. The corrections for the nonhomogeneous composition can be significant. Thin collectors or insulators behind the collecting volume can have a major effect because electron backscattering depends strongly on material. The calculations agree well with experimental data and explain previous results that had not been understood. Calculations for different buildup caps predict trends in chamber in-air responses, which go in the opposite direction to the km values for homogeneous chambers of the same materials. It is argued that Pwall, the wall correction factor for photon-beam measurements in-phantom is, to first order, the same correction factor as Kcomp, calculated for a buildup cap of the same material as the phantom material and the initial assumption that the wall material is made of the same material. Hence, Pwall can also be significantly different from unity. Recommended values of kmkatt or the equivalent Ngas/Nx, and Pwall (in 60Co beams) are presented for the 5 chambers. These values treat the Exradin P11, Markus and Holt parallel-plate chambers as homogeneous in construction whereas the values for the NACP and Capintec PS-033 chambers demonstrate that the chambers are distinctly nonhomogeneous.

This publication has 0 references indexed in Scilit: