MCNP 6.2 simulations of energy deposition in low-density volumes corresponding to unit-density volumes on the nanometre level

Abstract

The MCNP 6.2 event-by-event transport of electrons has been used for the first time to simulate energy imparted in low-density objects corresponding to nanometric unit-density objects. It is shown that the isolated atom assumption and the use of an average excitation energy based on stopping power, significantly alters the value of energy imparted from excitations. This limits to some extent the use of MCNP 6.2 in micro- and nanodosimetric applications, at least when excitations play an important role. Simulated single-event distributions show pronounced and anomalous excitation peaks related to the atomic composition of the target volume. In the case of air the dominating peak is related to excitations in nitrogen, while for tissue-equivalent gas the two dominating peaks are related to excitations in hydrogen and carbon. The excitation peaks in a nanometric unit-density water volume are also shown to be shifted compared to simulations with Geant4-DNA that take molecular excitation levels into account.Despite these limitations in addition to large cross-section uncertainties at low energies, the MCNP 6.2 simulated dose-average lineal energies are shown to agree well with experimental values in 60Co and 137Cs gamma beams as well as a 100 kVP x-ray beam. Simulated and measured values for an air-filled ionisation chamber agree within 10% for object diameters from 136 μm down to 20 nm. For smaller objects the experimental values based on ionisations and a conventional We-value are as expected significantly higher than the simulated values. Only a few percent of this difference is related directly to the excitation peaks in the MCNP 6.2 simulations.