[Gate-users] Influence of the system shape?

[Kiessling, Nicole]

Dear Gate-Users,

I found some strange results for the interaction of gamma radiation depending on the scanner system and I cannot find an explanation for it. 
This is what I did:
I want to analyse the cross-talk behaviour of LSO crystals. That’s why I have modeled 7 crystals side by side and focused a caesium gamma radiation source only on the crystal in the middle. I have summed up the deposited energy in a special range from a neighbor crystal and set it in ratio with the corresponding energy from the crystal in the middle.
At first I used the sytem “scanner” with the shape “cylinder” defined by Rmax and Rmin. Later I tried to use a box instead of the cylinder. But I get different results using the different shapes and when I change the size of the shapes - even if I use the same random numbers and even if the rest of the scannersystem is defined as vacuum.
I would have expected nearly the same results because I thought that if I use vacuum there would not be any noticeable interaction (before I used air but there were a lot of interactions I did not want to get so I changed it into vacuum).

What can be the reason for this effect?
Which information is predetermined with the predefined system? 
Can anybody help me with this?

I would be very glad about every help.
Thanks in advance.

Kind regards,
Nicole