Optimizing Filter Thickness in MCNP5 for Phosphorus 32 Bremsstrahlung Reduction

[Addali sabah]

I am working on the optimization of the thickness of filters in order to reduce the continuous Bremsstrahlung spectrum emitted by phosphorus 32 measured by the Ge(HP) detector using some MCNP5 code but unfortunately I couldn't find the exact form of declaration of bremsstrahlung in the MCNP input file.
and this is my input file can anyone find the mistakes in it
input file (The modeled geometry is composed of the detector( Ge HP) and a parallelepiped source)

1 1 -5.323 (2 :3 )1 -4 -5
2 2 -1.83 6 -7 8 -9 10 -11
3 0 -12 #1 #2
4 0 12

1 pz 0
2 pz 5.1
3 cz 0.6
4 pz 6.1
5 cz 3.025
6 px -0.5
7 px 0.5
8 py -0.5
9 py 0.5
10 pz 11
11 pz 11.1
12 so 15

mode p e
m1 32000. 1 $Ge
m2 15000. 1 $P
imp:p 1 2r 0 $ 1, 4
imp:e 1 2r 0 $ 1, 4
sdef par=3 erg=1.7 x=d1 y=d2 z=d3 vec= 0 0 -1 dir=1
si1 -0.5 0.5
si2 -0.5 0.5
si3 11 11.1
sp1 0 1
sp2 0 1
sp3 0 1
f8:p 1
e8 0 1e-04 120i 1.8
cut:p,e j 0.001
nps 1000000
phys:e 2 0 0 0 0 1 1 1 1 0

 

[Alex A]

"x=d1 y=d2 z=d3", I don't understand what this is intended to be, but I'd set it to be inside, or impinging on, your source.

15000 is better and will produce bremsstrahlung, but probably not right. You won't literally have a lump of phosphorous smoking in the air I assume. Have a think what your source is actually made of. All the bits, is it on foil?, is it on a carrier?. You may want to relax the vectors on your source, the real thing would be omnidirectional and the interactions with the rest of the material of the source will contribute to the photon background.

 

[Addali sabah]

x=d1 y=d2 z=d3", because I have parallelepiped source

 

[Alex A]

Ohhh. I think I see. I don't know source biasing.

Why are the SP lines "0 1"?
Isn't that the same as "D 0 1", ie 2 bins?
Instead of "-21 0" which ought to be uniform sampling?

It might also help to have a cel=2 on the sdef line.

 

[Addali sabah]

thank you sir ,
for the sdef card we cannot use both definitions at the same time ( cell= and x, y, z) because x=d1 y=d2 z=d3 these are the extensions of the geometry of the source (this is the definition of a parallelepiped source in the mcnp5 code).
in this case the sp is the probability of taking the values of si for this si= -0.5 0.5 sp= 0 1 means either it takes the value 0 for a si -0.5 or a value 1 for si= -0.5 and the same for si= 0.5.
i don't understand this "Instead of "-21 0" which ought to be uniform sampling?".

 

[Alex A]

Putting a cel= in the sdef line will cause source particles generated that are not within that cell to be skipped.

-21 is a built in distribution function x^a, with a=0 this results in a flat probability spectrum. This is a method for a volume source.

As it is, you seem to have a 1cm x 1cm x 1mm strip of elemental phosphorus. It emits electrons in one direction, just from the corner points. This is intentional?

 

[Alex A]

I realised too late to edit, small correction. Corner point not points. This is a point source at 0.5,0.5,11.1 at the corner of the strip. All the other corners generated by the SI cards have zero probability because each of the SP probability entries correspond to an SI bin when this format is used.