Ionization cross section data of nitrogen, methane, and propane for light ions and electrons and their suitability for use in track structure simulations

Track structure Monte Carlo simulations are frequently applied in micro- and nanodosimetry to calculate the radiation transport in detail. The use of a well-validated set of cross section data in such simulation codes ensures accurate calculations of transport parameters, such as ionization yields. These cross section data are, however, scarce and often discrepant when measured by different groups. This work surveys literature data on ionization and charge-transfer cross sections of nitrogen, methane, and propane for electrons, protons, and helium particles, focusing on the energy range between 100 keV and 20 MeV. Based on the evaluated data, different models for the parametrization of the cross section data are implemented in the code PTRA, developed for simulating proton and alpha particle transport in an ion-counting nanodosimeter. The suitability of the cross section data is investigated by comparing the calculated mean ionization cluster size and energy loss with experimental results in either nitrogen or propane. For protons, generally good agreement between measured and simulated data is found when the Rudd model is used in PTRA. For alpha particles, however, a considerable influence of different parametrizations of cross sections for ionization and charge transfer is observed. The PTRA code using the charge-transfer data is, nevertheless, successfully benchmarked by the experimental data for the calculation of nanodosimetric quantities, but remaining discrepancies still have to be further investigated (up to 13% lower energy loss and 19% lower mean ionization cluster size than in the experiment). A continuation of this work should investigate data for the energy loss per interaction as well as differential cross section data of nitrogen and propane. Interpolation models for ionization and charge-transfer data are proposed. The Barkas model, frequently used for a determination of the effective charge in the ionization cross section, significantly underestimates both the energy loss (by up to 19%) and the mean ionization cluster size (up to 65%) for alpha particles. It is, therefore, not recommended for particle-track simulations.