Message: Fano cavity simulations in a magnetic field?  Not Logged In (login) 
Hello all,
I am performing fano cavity simulations (similar to extended example fano cavity 2) for my work to investigate charged particle transport in the presence of a magnetic field. I am following the methods of Malkov and Rogers (MedPhys 43 44474458 (2016)) for this work. Essentially, three geometries are considered for these simulations, two of which consist of cylindrical slabs with the thickness of the gas volume equal to 0.2 cm or 2 cm and the third geometry is a simplified ion chamber. The particle source for all geometries are monoenergetic electrons of various starting energies (0.01, 0.1, 1, and 10 MeV). For the slab geometries, a line source in the center of the slabs are considered whereas the source for the ion chamber geometry is uniform per unit mass throughout the geometry. The electron emission is isotropic for all geometries. I originally started using Geant4 10.02, but switched to geant4 v10.04.p01 as it was clear that significant improvements were made to the GoudsmitSaunderson MSC model and the UseSafetyPlus step limit type. The parameters used are shown in the attached table. Agreement within 0.16% was observed for all energies and geometries considered (see attached plot). With these parameters I started performing simulations with a 1.5 tesla magnetic field present. I followed extended example field01 and wrote my own magnetic field class to control all of the parameters related to magnetic field transport. I've run into an interesting problem and I was hoping someone could give me some ideas to help with data interpretation. Attached are the simulated dose/energy fluence in the two slab geometries for all the electron energies considered. The transport parameters related to the magnetic field are also attached. As you can see, there is a significant drop in the dose/energy fluence at particular energygeometry combinations, specifically 0.1 MeV in the slab 0.2 cm geometry and 1 MeV in the slab 2 cm geometry. Furthermore, when the dose/energy fluence is simulated as a function of magnetic field strength for these two energygeometry combinations (attached), the results are very puzzling. I am having trouble understanding why the simulated dose/energy fluence decreases with a magnetic field present and why the dose/energy fluence from some energygeometry combinations are more sensitive to the presence of the magnetic field. Any ideas to help explain this trend would be greatly appreciated. Thanks, Eric
Attachment: http://hypernews.slac.stanford.edu/HyperNews/geant4/get/AUX/2018/04/05/13.4474485fano_cavity_0T.PNG http://hypernews.slac.stanford.edu/HyperNews/geant4/get/AUX/2018/04/05/13.4450221_transport_parameters.PNG http://hypernews.slac.stanford.edu/HyperNews/geant4/get/AUX/2018/04/05/13.4423224ial_1,5T_slab_results.PNG http://hypernews.slac.stanford.edu/HyperNews/geant4/get/AUX/2018/04/05/13.4430325_transport_parameters.PNG http://hypernews.slac.stanford.edu/HyperNews/geant4/get/AUX/2018/04/05/13.4430137gnetic_field_strength.PNG

to: 