Electron absorbed dose comparison between MCNP5 and Penelope Monte Carlo code for microdosimetry

The objective of the present work was to compare electron absorbed dose results between two widespread used codes in international scientific community: MCNP5 and Penelope-2003. Individual water spheres with masses between 10-9 g up to 10-3 g immersed in an infinite water medium (density of 1g/cm3) and monoenergetic electron sources with energy from 0.002 MeV to 0.1 MeV have been considered. The absorbed dose in the spheres was evaluated by both codes and the relative differences have been quantified. The results shown that Penelope gives, in general, higher results that, in some cases saturate or reach a maximum point and then rapidly drops. Particularly, for the 40 keV electron source we have done additional tests in three different scenarios: more points in the region of lower masses to a better definition of the curve behavior; MCNP used 200 substeps and Penelope was set to a full detail history methodology, and almost same parameters of case B but with the density of exterior medium increased to 10 g/cm3. The three cases show the influence of the backscattering that contribute with an important fraction of absorbed dose, finally we can infer a range of reliability to use the codes in this kind of simulations: both codes can calculate close results for up to 10-4 g.Even though MCNP5 uses the condensed history method, if simulation parameters are chosen carefully it can reproduce results very close to those obtained using detailed history mode. In some cases, the use of higher number of electron substeps causes significant differences in the result. (author)