Premium
Monte Carlo dose calculations in homogeneous media and at interfaces: A comparison between GEPTS, EGSnrc, MCNP, and measurements
Author(s) -
Chibani Omar,
Li X. Allen
Publication year - 2002
Publication title -
medical physics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.473
H-Index - 180
eISSN - 2473-4209
pISSN - 0094-2405
DOI - 10.1118/1.1473134
Subject(s) - physics , monte carlo method , electron , positron , photon , imaging phantom , homogeneous , dosimetry , nuclear physics , computational physics , nuclear medicine , optics , statistical physics , medicine , statistics , mathematics
Three Monte Carlo photon/electron transport codes (GEPTS, EGSnrc, and MCNP) are benchmarked against dose measurements in homogeneous (both low‐ and high‐ Z ) media as well as at interfaces. A brief overview on physical models used by each code for photon and electron (positron) transport is given. Absolute calorimetric dose measurements for 0.5 and 1 MeV electron beams incident on homogeneous and multilayer media are compared with the predictions of the three codes. Comparison with dose measurements in two‐layer media exposed to a60 Co gamma source is also performed. In addition, comparisons between the codes (including the EGS4 code) are done for (a) 0.05 to 10 MeV electron beams and positron point sources in lead, (b) high‐energy photons (10 and 20 MeV) irradiating a multilayer phantom (water/steel/air), and (c) simulation of a90 Sr / 90 Y brachytherapy source. A good agreement is observed between the calorimetric electron dose measurements and predictions of GEPTS and EGSnrc in both homogeneous and multilayer media. MCNP outputs are found to be dependent on the energy‐indexing method (Default/ITS style). This dependence is significant in homogeneous media as well as at interfaces. MCNP ITSfits more closely the experimental data than MCNP DEF , except for the case of Be. At low energy (0.05 and 0.1 MeV), MCNP ITSdose distributions in lead show higher maximums in comparison with GEPTS and EGSnrc. EGS4 produces too penetrating electron‐dose distributions in high‐ Z media, especially at low energy (<0.1 MeV). For positrons, differences between GEPTS and EGSnrc are observed in lead because GEPTS distinguishes positrons from electrons for both elastic multiple scattering and bremsstrahlung emission models. For the60 Co source, a quite good agreement between calculations and measurements is observed with regards to the experimental uncertainty. For the other cases (10 and 20 MeV photon sources and the90 Sr / 90 Y beta source), a good agreement is found between the three codes. In conclusion, differences between GEPTS and EGSnrc results are found to be very small for almost all media and energies studied. MCNP results depend significantly on the electron energy‐indexing method.