[Gate-users] Abslength issue

[D.J. van der Laan]

Hi Sean,

You are right. This is a bug. There was an error in the code that reads
in the unit used in the property vectors: the unit is not read in and
the default unit of Geant4 is always used (for lengths this is mm). This
bug has been fixed in Gate v3.1.0 and newer. So you should either switch
to a newer version, or always specify the lengths in mm's.

Since I don't know how the values of the absorption and attenuation
length were measured by the manufacturer, it is difficult to say what
they exactly mean. I would expect that the attenuation length includes
the contributions of absorption and scatter (and ...?), and will
therefore be shorter than the absorption length. When the attenuation
length includes only absorption and scatter, the scatterlength can be
calculated using (see also [1]):

1/l_atten = 1/l_absor + 1/l_scatt

In Gate v3.1.0/1 it is now also possible to specify the (Rayleigh)
scatter length using the RAYLEIGH property vector (as for absorption).
This is not yet in the documentation.

I hope this helps.

Regards,
Jan


[1] Moisan, C.; Vozza, D. & Loope, M. Simulating the performances of an
LSO based position encoding detector for PET, IEEE Nuclear Science
Symposium and Medical Imaging Conference Record, IEEE, 1996 , 2, 1211-1215.

Sean Jones wrote:
> I am running a simulation of a phoswich detector.  In order to
> determine light capture efficiency of the system, I am looking to
> determine from the ASCII file output, the number of photons created
> vs. the number generated.
> 
> I have spent much time and attention on the simulation.  However, I
> seem to have run across and interesting observation that I believe may
> be a bug.  Unfortunately, this is a large speed bump I must overcome
> in order to move ahead with this research.
> 
> In running simulations I have input a value of 250 (units of cm) for
> the property abslength in the materials.xml file.  When doing so I can
> count optical photons that strike the photocathode.  When I run the
> same simulation, and switch the value from 250 (with units of cm) to
> 2.5 (units of m), ALL of the optical photons generated are absorbed
> before hitting the photocathode.  Each simulation generates a similar
> number of optical photons, and in theory the two values should be
> identical and yield similar results.  I've switched back and forth
> various times, and it seems that the input depends on the magnitude of
> the number input and disregards the unit used.  As such, I believe
> I've encountered a bug in the GATE code and how it implements the unit
> value for abslength property, or perhaps I've missed something
> important between the two manuals.
> 
> Furthermore, I am uncertain of the definition of the abslength value
> that must be used for optical materials.  I am aware that these are
> empirical values:
> Bulk absorption (discussed in the GEANT4 manual), and Bulk attenuation
> (I believe reduces light by a factor of e^-1).  I have these two
> values for only one of my materials from our detector manufacturer.
> However, if you have any input as to what those values actually mean,
> which one must be implemented in the GATE framework (specifically what
> units the input seems to be stuck to, if it is somehow unchageable as
> a bug), and where
> one could find a list of this empirical data for various optical and
> scintillation materials, I would be most appreciative.
> 
> Thank you for your time.
>