14th International Conference of Medical Physics
ICMP2005Nuremberg, Germany, Sept. 14-17, 2005
Estimation of Neutron Dose Conversion Coefficients for a Voxel Phantom with mcnpx Valery Taranenko and Maria Zankl, GSF—Research Center for Environment and Health, Neuherberg, Germany
This work describes the estimation of the fluence-to-absorbed dose conversion coefficients for the icrp Reference Man adjusted male voxel phantom Godwin for the external monoenergetic neutrons of various source configurations using the latest version of the mcnpx general purpose Monte Carlo code. The results are provided for an anterior-posterior parallel source for a few organs. The comparison of the results against analogous datasets provides insight into the degree of difference due to various voxel phantoms used as well as modeling parameterizations employed. Taking into account the variability between the datasets, the results comply with other evaluations.
1 Introduction In radiological protection against external neutrons fluenceto-absorbed dose conversion coefficients play an importance role because the other protection quantities are based upon them [1]. In this study we have calculated for the first time the conversion coefficients for a modern male voxel phantom. This phantom has been recently adjusted in accordance with icrp Reference Man organ mass data [2], providing an improved degree of realism. To simulate the 3d transport of radiation in the phantom we used the recent version 2.5.f of the mcnpx general purpose Monte Carlo program [3]. mcnpx is capable to follow neutrons in the range of interest for the present study (below 20 MeV) on the bases of continuous-energy nuclear data libraries. Secondary photons and electrons are followed down to 1 keV.
2 mcnpx setup The geometry setup in mcnpx was made via the repeated structures feature as described in [4]. The phantom consists of about 7.5 million voxels of which about 2 million represent 84 different tissues and organs. Explicit isotopical composition was set for 27 materials in use. Isotopes and libraries used (from dlc-0205/004 library package): 1H, natural C, 14N, 16O, 31P, natural Ca, 56Fe from la150n (endf/b-vi.6); 23Na, natural Mg, S, Cl and K, 127I from endf60 (endf/b-vi); natural Ar from rmccsa (endl 85). The S(α,β) thermal scattering treatment for hydrogen in light water at 300 K was based on table lwtr.01t from the library tmccs. This explicit S(α,β) capability takes into account the effects of chemical binding for incident neutron energies below about 4 eV. The S(α,β) thermal treatment was improved in mcnpx compared with older versions of mcnp/x codes. In addition, photoatomic data based on epdl 97 and standard electron library el03 were employed. We used a coupled neutron-photon mode for production runs, since the explicit transport of electrons doesn’t noticeably affect the organ absorbed doses.
The default cutoff energies were used, i.e. neutrons 0 MeV, photons 1 keV. The scoring of energy deposition was accomplished via the +f6 tally which is a track-length estimate of flux folded with heating number. These heating numbers are taken from the cross-section libraries and represent the amount of deposited energy per unit track-length. Hence, it estimates the kerma. Several external source beams were simulated. 10 million of primary neutrons were assigned for each run to yield a statistical uncertainty about 1%.
3 Results Results for anterior-posterior (ap) broad unidirectional monoenergetic neutron beams for several critical organs are shown in Fig 1. In the figure, the results are compared with the results of similar evaluations: firstly, with icrp 74 data [1], with gsf data for the “mathematical” male phantom Adam (V. Mares, private communication), and with the data for vip-Man [5]. The latter are the only results (available to the authors) that are based on neutron transport simulations in a voxel geometry. On a 3-GHz Intel Pentium 4 cpu, the calculation speed at 1 meV and 20 MeV is about 105 and 8×104 showers per minute, respectively. The typical difference between the mcnpx results for absorbed dose compared with the vip-Man model and the icrp data is about 10–20%, increasing up to 50–70% at energies from 100 keV to 1–3 MeV. These discrepancies are attributable to differences in the anatomical models used as well as to a lesser degree due to different cross-section libraries. Taking into account the overall discrepancy between the four datasets depicted in Fig. 1, we conclude that on the basis of the male phantom Godwin it is possible to provide consistent results for neutron transport simulations with mcnpx.
A bladder
�T / Φ, pGy cm²
�odwin (this work) �an �dam
³
¹
³
⁵
B colon
⁷ �n₀, eV �T / Φ, pGy cm²
³
¹
³
⁵
⁷ �n₀, eV
C skin
�T / Φ, pGy cm²
³
¹
³
⁵
⁷ �n₀, eV �T / Φ, pGy cm²
D stomach
³
¹
³
⁵
⁷ �n₀, eV
Fig. 1 Comparison of different evaluations of organ absorbed doses as a function of source energy for an ap unidirectional neutron beam
4 Acknowledgments This work has been financially supported by the German Federal Office of Radiation Protection under contract no. StSch 4417.
5 Literature [1] International Commission on Radiological Protection: Conversion Coefficients for use in Radiological Protection against External Radiation, Annals of icrp, Publication 74. 1997 [2] Zankl M, Becker J, Fill U, Petoussi-Henss N, and Eckerman K F: GSF male and female adult voxel models representing icrp reference man. In Monte Carlo 2005 Topical Meeting, Chattanooga, tn, usa [3] Hendricks J S et al.: mcnpx extensions, version 2.5.0, LAUR-04-0570, Los Alamos, nm, usa, http://mcnpx.lanl. gov, 2004 [4] Taranenko V, Zankl M, and Schlattl H: Voxel phantom setup in mcnpx. In Monte Carlo 2005 Topical Meeting, Chattanooga, tn, usa [5] Bozkurt A, Chao T C, and Xu X G: Fluence-to-dose conversion coefficients from monoenergetic neutrons below 20 MeV based on the vip-Man anatomical model. Phys. Med. Biol. 45, pp. 3059–3079. 2000