Calculation of the replacement correction factors for ion chambers in megavoltage beams by Monte Carlo simulation

This article describes four methods of calculating the replacement correction factor, P(repl0 (or the product p(cav)P(dis) in the IAEA's notation), for a plane-parallel chamber in both electron and photon beams, and for a Farmer chamber in photon beams, by using the EGSnrc Monte Carlo code. The accuracy of underlying assumptions and relative merits of each technique are assessed. With careful selection of parameters it appears that all four methods give reasonable answers although the direct methods are more intellectually satisfying and more accurate in some cases. The direct methods are shown to have an accuracy of 0.11% when appropriate calculation parameters are selected. The depth dependence of P(repl) for the NACP02 plane-parallel chamber has been calculated in both 6 and 18 MeV electron beams. At the reference depth (0.6R50-0.1 cm) P(repl) is 0.9964 for the 6 MeV beam and 1.0005 for the 18 MeV beam for this well-guarded chamber; at the depth of maximum dose for the 18 MeV beam, P(repl) is 1.0010. P(repl) is also calculated for the NACP02 chamber and a Farmer chamber (diameter 6 mm) at a depth of 5 cm in a 60Co photon beam, giving values of 1.0063 and 0.9964, respectively. For the Farmer chamber, P(repl) is about half a percent higher than the value (0.992) recommended by the AAPM dosimetry protocol. It is found that the dosimetry protocols may have adopted an incorrect value of P(repl) for cylindrical chambers in photon beams. The nonunity values of P(repl) for plane-parallel chambers in lower energy electron beams imply a variety of values used in dosimetry protocols must be reassessed.