Go Back

Modeling Questions

I had inquiry regarding a disk source vs cylinder source. The way I understood it, the disk source took no account of self-absorption, while activity is evenly distributed in a cylinder and self-absorption is taken into account. When I run a unit release of the Mo99/Tc99m scenario for 1cm2 source an dose area, there are cylinder heights for same area which get a greater dose rate than the disk source. I just wanted to confirm that this is correct. Thanks for your help.

What we are seeing here is the effect of backscatter. If you change all of the BSCF's in the .rad files to 1, you will see the reason for these results. I tested this with Mo-99 since its the main contributor to dose here. When you run the disk source, the dose from Mo-99 is 1.41E-03. When you run the 1 um thick cylinder this dose jumps to 1.62E-03. Now, when we take out the BSCF these doses change to 2.02E-03 and 2.01E-03, respectively. This means that the disk source is applying a BSCF of 0.7 and the cylinder is only applying a BSCF of 0.8 and therefore dose is not reduced as much. This is means that source self-absorption is fairly negligible as the dose hardly changes for the two geometries with backscatter. So why is the BSCF for the cylinder less? Its how backscatter is built into the integration process. As the source size is increased past that of a point or a disk, the amount of source material traversed in each integration step has an impact on the BSCF chosen for that point in the integration process. As a result, for a volume source, the BSCF applied will always be less than that of a point or a disc and will approach that of a point or a disk as the source size is decreased. As you continue to increase the source size from 1 um to 10 and so forth, the self absorption starts to have a greater impact than the increase in BSCF and this is why your dose starts to climb back down and eventually be less than the disk dose.