Condensed-history Monte Carlo simulation of the dosimetric distribution of electron microbeam

Abstract

To evaluate the dose distributions of an electron microbeam and to help optimization of its design, the condensed-history (CH) Monte Carlo simulation algorithm implemented in the Geant4 toolkit was selected as an alternative to the conventionally used analog algorithm. Compared to the analog algorithm, the CH algorithm is cheaper and less limited by the lack of cross-sections. And, with a properly chosen production cut for secondaries, its accuracy for the problems of microdosimetry is satisfactory. In this work, calculations of the single-event (imparted energy epsilon) size distribution f (1)(epsilon) is described, for compartments in the Orlando electron micro beam. The results agree well with those obtained by use of the analog algorithm and reported in the literature. It is shown that substituting tissue with water in HeLa cells, and replacing Mylar with water of the same mass stopping power in the substrate, makes little difference. Additionally, the neighbor-to-target ratio of average event size R (NT) has been calculated as a function of the incident energy of the electrons. Comparison with analog results reported in the literature suggests that the average event size in neighbors, and hence the neighbor-to-target ratio, is sensitive to the selection of the energy threshold. Finally, the effect of finite beam radius on the event size distribution and the neighbor-to-target ratio has also been studied. All results presented suggest the condensed-history technique to provide an efficient and valuable alternative tool in the design of electron microbeams.