Assessment of different MCNP Monte Carlo codes in electron absorbed dose
Background: MCNP is a general-purpose Monte Carlo code for simulation of neutrons, photons and electrons or coupled neutron/photon/electron transport. This code is based on ETRAN/ITS codes. There are different versions of this code. Aim: This work aims to compare the more recently released MCNP codes with the earlier version in terms of the central axis absorbed dose (CADD), the energy spectrum and the computational efficiency. MCNP codes 4A, 4B, 4C, X and 5 were compared for a 10MeV electron beam in water. Materials/Methods: The energy spectra of electrons were scored on the phantom surface and planes 3 and 5cm deep using F2 tally subdivided into 0.1MeV energy bins. This tally also was repeated for 4A, 4B, 4C and X with smaller energy bins (0.05MeV). The simulated geometry and other input parameters were kept the same. Both the default and ITS energy indexing algorithms (EIA) were used in 4B, 4C and X, while only the default EIA was employed in 4A and 5. Results: With default indexing, X and 5 showed no difference in CADD compared to 4B and 4C and were within 3% of 4A. We found no differences in CADD between codes when 4B, 4C and X were used with ITS indexing. The ITS algorithm improved computational efficiency. For the energy spectrum at the phantom surface, all codes except X show very similar results (within 2%). However, changing the energy indexing to ITS as well as using a 0.05MeV bin removed this discrepancy at the surface for X code. Conclusions: While, under the examined conditions, versions 4B and later behaved similarly in terms of the resulting CADD, the ITS indexing should be used due to its agreement with measurements and computational efficiency.