[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: How many photons in scintillator? CORRECTION (fwd)




---------- Forwarded message ----------
Date: Tue, 02 Oct 2007 13:14:10 -0600
From: George Lolos <George.Lolos@uregina.ca>
To: Matthew Shepherd <mrs43@cornell.edu>,
     Richard Jones <richard.t.jones@uconn.edu>,
     "elton@jlab.org" <elton@jlab.org>
Cc: Andrei Semenov <semenov@jlab.org>, Blake Leverington <leverinb@uregina.ca>,
     "zisis@uregina.ca" <zisis@uregina.ca>,
     Kathryn Janzen <janzen2k@uregina.ca>,
     "dzierba@indiana.edu" <dzierba@indiana.edu>
Subject: Re: How many photons in scintillator?  CORRECTION

Hi guys:

The demon of  "calculatius interruptus" hit me again!  The SciFi
acceptance of 9.3% is, of course, taking into consideration the "one
side" capture, so I should not have multiplied by a factor of two in the
end.  So, the number of photons/MeV is 2136 as calculated by X, not ~ 4200!

My apologies for such a mistake.  Incidentally, this number matched very
well claims in the literature as the "effective" photon yield in plastic
scintillators in general.

By the way, we have just started the cosmic ray measurements of Module 2
with green-sensitive PMT's (on loan from UofA) for the green section of
the module and we continue with the old arrangement for the blue
fibers.  Due to the small size of the green sensitive PMT's we now use
only one light guide-WC arrangement for them, thus exactly matching the
read-out proposed for the inner layers with SiPM's.  Next step down the
road is to use the Phase I arrays on the green fibers of Module 1.

Cheers,

George

George Lolos wrote:

> Hi guys:
>
> Time to revisit the much discussed question of how many scintillation
> photons do we really get in a scintillator that result in
> photo-electrons (pe's) in the PMT.  I will use the hard numbers
> obtained by the cosmic ray tests here in Regina using Module 2, the
> "hybrid green-blue", but the numbers are based on the blue SciFi's
> using standard BURLE 8575 2" PMT's.  The reason I base my calculations
> on the Regina rather than the JLab cosmic ray data is the very tight
> trigger we use here (we have even incorporated a Plexiglas Cherenkov
> counter to assure no showers are in the data) and the well simulated
> light guide -WC single piece devices.
> My assumptions are the following:
>
> 2 MeV/cm energy loss of muons in scintillator
> 4 cm of read-out corresponds to 2 cm of actual SciFi material, so we
> have 2cm of scintillator
> Actual scintillating content is 95% to account for the two claddings
> (dead) volume
> SciFi acceptance per side is 9.3%
> Length of SciFi is 2m
> Attenuation length is 2.4m (as extracted by Andrei and Alex)
> Light guide-WC transmission efficiency is 95% (from Blake's MC)
> I assume 90% transmission efficiency per optical grease coupling for a
> combined 81% for two surfaces
> A combined Q.E. and collection efficiency for the PMT of 10% pe/ph.
> Finally, we have approximately 22 to 25 pe's (depending on method) in
> excellent agreement with the Hall B results.
>
> So:
>
> X ph/MeV x 2MeV/cm x 2cm x 0.95 x exp(-2/2.4) x 0.95 x 0.8 x 0.1pe/ph
> = 25 pe's
>
> X = 2136 ph/MeV for one PMT (or with initial emission in half the
> SciFi volume)
>
> /Therefore, the number of photons/MeV in SciFi that "survived"
> absorption and propagated down the fiber is approximately/ *4,200*
> *photons/MeV*.
>
> This is interesting because my rough integration of the blue SciFi
> spectrum had indicated that out of 8000 ph/MeV only ~ 3200 survive
> initial absorption.    Now, some of the low wavelengths that have been
> absorbed have been re-emitted shifted up but I have no way of knowing
> how many. In addition, I had to make some assumptions on overall
> transmission efficiency and photon-to-pe conversion but I think the
> result is quite close to the actual number.
> To summarize, the measurements are indicative of an effective number
> of photons/MeV for MIP's of half the quoted value of 8000 ph/MeV and
> we should base our calculations from now on this number unless someone
> comes up with a more reliable one.
>
> George
>