Room: Exhibit Hall
Purpose: To calculate radiation yields of electrons and positrons accurately using Monte Carlo techniques instead of the CSDA method of integrating the ratio of the radiative to total stopping powers as done in ICRU Report 37.
Methods: The EGSnrc Monte Carlo code and a modified version of the distributed application `g' was used for electron and positron initial kinetic energies between 10 keV and 30 MeV to calculate the radiation yield, g, defined as the fraction of a charged particle’s kinetic energy which is lost via radiative processes while slowing to rest. The fraction lost to bremsstrahlung, fluorescence and for positrons, annihilation in flight, were scored separately.
Results: For electrons in water, g(total)=g(brems) but both are between 4.4%(20 MeV) and 3.1%(1 MeV) greater that the ICRU37 values. The differences are due to the more accurate effects of energy-loss straggling included in the Monte Carlo results. For positrons on water, g(brems) is similarly close to the ICRU37 values for electrons which are close to those for positrons except at low energies. However, the g(total) is substantially greater than the ICRU37 values due to a significant contribution from annihilation in flight: by 64%, 100% and a factor 6.4 at 20, 12 and 5 MeV respectively. For electrons in aluminum, g(fluor) starts to have a significant effect, especially at lower energies: 18% and 47% at 1 and 0.1 MeV respectively.
Conclusion: The radiative losses from non-brems processes are significant, as are the effects of energy-loss straggling. While Seltzer explicitly included all these effects when calculating mass energy absorption coefficients, it is important to account for these processes in other calculations using radiation yields. EGSnrc uses electron brems cross sections for positrons and this leads to some significant errors in g values for low-energy positrons.
Funding Support, Disclosures, and Conflict of Interest: Supported by the Natural Science and Engineering Research Council of Canada.
Not Applicable / None Entered.