Abstract. Most codes of practice for dosimetry of proton beams do not give a clear recommendation on the determination of recombination correction factors for ionization chambers. In this work, recombination corrections were measured in the low-energy clinical proton beam of the Clatterbridge Centre of Oncology (CCO) using data collected at different dose rates and different polarizing voltages. This approach allows the separation of contributions from initial and volume recombination and was compared with results from extrapolation and two-voltage methods. A modified formulation of the method is presented for a modulated beam in which the ionization current is time dependent. Using a set-up with two identical chambers placed face-to-face yielded highly accurate data for plane-parallel ionization chambers. This method may also be used for high-energy photon and electron beam dosimetry. At typical dose rates of 26 Gy min⁻¹ used clinically at the CCO, the recombination correction is 0.8% and thus is of importance for reference dosimetry. The proton beam should be treated as purely continuous given the high pulse repetition frequency of the cyclotron beam. The results show that the volume recombination parameter for protons is consistent with values measured for photon beams. Initial recombination was found to be independent of beam quality, except for a tendency to increase at the distal edge of the Bragg peak; this is only relevant for depth dose measurements. Using a general equation for recombination and generic values for the initial and volume recombination parameters (A = 0.25 V and m² = 3.97 × 10³ s cm⁻¹ nC⁻¹ V²), the experimental results are reproduced within 0.1% for all conditions met in this work. For the CCO beam and similar proton beams used for treating optical targets operating at high dose rates, the recombination correction factor can be overestimated by up to 2%, resulting in an overestimation of dose to water by the same amount, if the recommendation from IAEA TRS-398, which is only valid for pulsed beams, is followed without consideration.

Print publication: Issue 4 (21 February 2006)
Received 20 July 2005, in final form 6 December 2005
Published 1 February 2006