Electron backscatter corrections for parallel-plate chambers.

The wall of an ionization chamber is commonly assumed to have a negligible effect on chamber response in electron beams. For cylindrical chambers with thin walls this assumption is valid. However, parallel-plate chambers commonly possess large mechanical supports which may affect chamber response in a manner not accounted for in current dosimetry protocols. This is due to changes with energy in the relative backscattered electron fluence between chamber support and phantom materials. To investigate this effect, electron backscatter from low atomic number materials has been measured with electrons from 6 to 20 MeV. The effect of the diameter and thickness of the backscattering material has also been studied. Based on these data, Lucite and polystyrene chambers in water phantoms are expected to underrespond by 1% and 2% at 6 MeV. The expected underresponse decreases to 0.8% and 0.4% for polystyrene and Lucite at 12 MeV and is insignificant above 16 MeV. Two commercially available parallel-plate chambers were compared with a cylindrical chamber in electron beams from 6 to 20 MeV. Using the 20-MeV intercomparison, the expected chamber responses at the lower energies were calculated and compared with measurements. Both parallel-plate chambers underresponded by approximately 1% at 6 MeV and 0.5% at 9 MeV which is qualitatively consistent with the electron backscatter data. Recommendations for minimizing electron backscatter effects through chamber design are discussed.