An experimental study of recombination and polarity effect in a set of customized plane parallel ionization chambers

Plane parallel ionization chambers are an important tool for dosimetry and absolute calibration of electron beams used for radiotherapy. Most dosimetric protocols require corrections for recombination and polarity effects, which are to be determined experimentally as they depend on chamber design and radiation quality. Both effects were investigated in electron beams from a linear accelerator (Varian 21CD) for a set of four tissue equivalent plane parallel ionization chambers customized for the present research by Standard Imaging (Madison WI). All four chambers share the same design and air cavity dimensions, differing only in the diameter of their collecting electrode and the corresponding width of the guard ring. The diameters of the collecting electrodes were 2 mm, 4 mm, 10 mm and 20 mm. Measurements were taken using electron beams of nominal energy 6 to 20 MeV in a 10 cm x 10 cm field size with a SSD of 100 cm at various depths in a Solid Water slab phantom. No significant variation of recombination effect was found with radiation quality, depth of measurement or chamber design. However, the polarity effect exceeded 5% for the chambers with small collecting electrode for an effective electron energy below 4 MeV at the point of measurement. The magnitude of the effect increased with decreasing electron energy in the phantom. The polarity correction factor calculated following AAPM protocol TG51 ranged from approximately 1.00 for the 20.0 mm chamber to less than 0.95 for the 2 mm chamber at 4.1 cm depth in a electron beam of nominally 12 MeV. By inverting the chamber it could be shown that the polarity effect did not depend on the polarity of the electrode first traversed by the electron beam. Similarly, the introduction of an air gap between the overlying phantom layer and the chambers demonstrated that the angular distribution of the electrons at the point of measurement had a lesser effect on the polarity correction than the electron energy itself. The magnitude of the absolute difference between charge collected at positive and negative polarity was found to correlate with the area of the collecting electrode which is consistent with the explanation that differences in thickness of the collecting electrodes and the number of electrons stopped in them contribute significantly to the polarity effect. Overall, the polarity effects found in the present study would have a negligible effect on electron beam calibration at a measurement depth recommended by most calibration protocols. However, the present work tested the corrections under extreme conditions thereby aiming at greater understanding of the mechanism underlying the correction factors for these chambers. This may lead to better chamber design for absolute dosimetry and electron beam characterization with less reliance on empirical corrections.