[Gate-users] Potential bug in material definition?

[David Leibold]

Dear Gaters,

thanks to a hint from my colleague Stefan van der Sar the issue got solved. I took the values from the NIST database values for the mixture instead of compound, while using the formula of the compound. Equivalently, the Gate database fractions have to be given in fractions of weights.

I’m sorry for any confusion I might have caused.

Kind regards,
David



On 8 Mar2021, at 17:05, David Leibold <D.Leibold at tudelft.nl<mailto:D.Leibold at tudelft.nl>> wrote:

Dear Gaters,

(Note: I’m using Gate on the vGate machine 9.0, including GATE version 9.0 and geant4-10-06-patch-02.)

I checked the physics in Gate by simulating the attenuation of a monoenergetic beam through phantoms of different compositions. So far I have checked water, LuAP, LYSO, aluminium, silicon, calcium and iodine. For all materials (except LuAP), the relative error to the NIST reference data is below 2 % in the energy range 10 – 150 keV.

In the case of LuAP, the results were initially off by about a factor of about 2 (not a constant factor, though).

This is the spectra I got:

<attachment.png>



The graph labelled with "n=..." is obtained by defining LuAP in Gate’s material database with the following definition, using integer ratios:

    LuAP: d=8.34 g/cm3; n=3 ; state=Solid
        +el: name=Lutetium ; n=1
        +el: name=Aluminium; n=1
        +el: name=Oxygen; n=3

Please note that this is the definition as shipped with the Gate examples.

The graph labelled with "f=..." is obtained by defining LuAP in Gate’s material database with the following definition, using float ratios of 1:

    LuAP: d=8.34 g/cm3; n=3 ; state=Solid
        +el: name=Lutetium ; f=0.2
        +el: name=Aluminium; f=0.2
        +el: name=Oxygen; f=0.6

The graph labelled "NIST" denotes the attenuation coefficient data as obtained via NIST, which perfectly aligns with the graph labelled “f=…”, i.e. defining the material as fractions of 1.

Except for how the LuAP material was defined in the Gate database, the Gate scripts and the evaluation code are the same.

As a sanity check and to see what material composition Gate is using, I decided to fit the individual attenuation coefficients for lutetium, aluminum and oxygen as obtained by NIST to the output data "n=...", i.e. the data that is off. Here's the fit result:


<attachment.png>

The fit coefficients are as follows:

    0.69914713*NISTdata_Lu +
    0.13694515*NISTdata_Al +
    0.1097257 *NISTdata_O

As can be seen, the composition as used by Gate has little in common with the composition as defined in the database…

So is this a bug or am I doing something wrong?

Thanks in advance for your help!

Kind regards,
David Leibold
_______________________________________________
Gate-users mailing list
Gate-users at lists.opengatecollaboration.org<mailto:Gate-users at lists.opengatecollaboration.org>
https://urldefense.proofpoint.com/v2/url?u=http-3A__lists.opengatecollaboration.org_mailman_listinfo_gate-2Dusers&d=DwIGaQ&c=XYzUhXBD2cD-CornpT4QE19xOJBbRy-TBPLK0X9U2o8&r=lfXR69GFfS7NT-Wp5HZqbtBDrbYoilDdmNEG7fKW7aM&m=w821jisJJCJyX6ocrZbdl1YyYs3LWTPzVldTNKwWeMM&s=tJ6_rP-1Gnf9RGj_MsyoADH3EbIXtcFCPcbKcilAx7Q&e=

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.opengatecollaboration.org/pipermail/gate-users/attachments/20210308/dae87885/attachment.html>