Technical Note: Effect of explicit M and N ‐shell atomic transitions on a low‐energy x‐ray source

Purpose: In EGSnrc, atomic transitions to and from the M and N ‐shells are treated in an average way by default. This approach is justified in which the energy difference between explicit and average M and N ‐shell binding energies is less than 1 keV, and for most applications can be considered negligible. However, for simulations of low energy x‐ray sources on thin, high‐Z targets, characteristic x‐rays can make up a significant portion of the source spectra. As of release V4‐2.4.0, EGSnrc has included an option to enable a more complete algorithm of all atomic transitions available in the EADL compilation. In this paper, the effect of M and N ‐shell averaging on the calculation of half‐value layer (HVL) and relative depth dose (RDD) curve of a 50 kVp intraoperative x‐ray tube with a thin gold target was investigated. Methods: A 50 kVp miniature x‐ray source with a gold target (The INTRABEAM System, Carl Zeiss, Germany) was modeled with the EGSnrc user code cavity , both with and without M and N ‐shell averaging. From photon fluence spectra simulations, the source HVLs were determined analytically. The same source model was then used with egs _ chamber to calculate RDD curves in water. Results: A 4% increase of HVL was reported when accounting for explicit M and N ‐shell transitions, and up to a 9% decrease in local relative dose for normalization at 3 mm depth in water. Conclusions: The EGSnrc default of using averaged M and N ‐shell binding energies has an observable effect on the HVL and RDD of a low energy x‐ray source with high‐Z target. For accurate modeling of this class of devices, explicit atomic transitions should be included.