Re: polarimetry (fwd)

[Jim Stewart <jstewart@jlab.org>]

---------- Forwarded message ----------
Date: Wed, 14 May 2008 10:25:41 -0400 (EDT)
From: Franz Klein <fklein@jlab.org>
To: Jim Stewart <jstewart@jlab.org>
Cc: Richard Jones <jonesrt@uconnvm.uconn.edu>,
     Elke-Caroline Aschenauer <elke@jlab.org>
Subject: Re: polarimetry


Jim,

in general, we will analyze the bremsstrahlung spectrum (using Richard's
code and if you want the anb code from Tuebingen, which is used in Mainz,
Lund, Bonn, Hall B. There exists another code from Tokyo U. but we found
it inferior to the Tuebingen code). In Hall B we are analyzing the
spectrum in steps of 2 seconds using scalers (fed into datastream) with
cross check by TDC data. The main reason for variations has been beam
trips (followed by position offsets/divergence during recovering
~10-30sec).
We estimate the accuracy of such analysis at about 3-5% (based on beam
asymmetry for piN and accuracy of describing the spectrum ... using some
fudge factors for beam emittance+divergence (somehow the CASA calculation
usually don't agree with our results .. it's actually similar in Mainz).
Richard and Hradchya (NIM A554(05),75) claim that they achieve an absolute
accuracy of 0.01-0.02.

The pair polarimeter requires dedicated runs: (a) since the area between
active target (thin scintillator, 50um) and mustrip planes is field-free,
the opening angles are only of order mrad - thus the mustrip planes are in
the beamline, (b) in our tests at LEGS,Spring8,Hall B, the rate was only
~10-50Hz due to the multiplexer (old stuff) and hit ambiguities - thus low
intensity beam to start with (smaller pitch and new readout electronics
can get to 1kHz ... still low intensity!).
Method and tests described in NIM A515(03),605; Eur.Phys.J.A19(04),275);
clas-note 2006-01.
Why it's not further used in Hall B? Main reason: after the successful
test at Spring8 (the paper has a long list of collaborators who wanted to
visit Japan) the South Carolina group and George Wash.U. groups weren't
interested any more; mustrip detectors were malfunctioning; light coupling
of 50um target appeared to be more tricky than expected; motion of the
mustrip detector in vacuum was not accurate ... most of the things could
have been prevented if the 2 groups weren't simply playing around.

Other option for pair polarimeter: dipole between target and mustrips or
fibers, which are then out of the beamline, i.e. the device can be used
during regular data taking. Problems: to calculate the trajectory, the
hit position on the farget and the field have to be well known. The TAGX
setup didn't deliver accurate results (with low analyzing power), the
YerPhi group meanwhile claims that they can achieve abs.0.02-0.03 (and 25%
analyzing power) (NIM A579(07),973).

Finally, there has been always the idea of using triplet production for
polarimetry - what I've seen till now is very questionable (several poor
experiments - I got from geant and the Tokyo code some options that may
allow for ~7-8% analyzing power but not more!).

Jim,
as I understood the reaction to polarimetry in earlier years, the
collaboration wasn't interested at all ... that changed! Nonetheless,
a polarimeter is not part of the basic Hall-D setup (simply have the space
available).

Greetings

Franz

On Mon, 12 May 2008, Jim Stewart wrote:

> Dear Franz and Richard
>
> I am trying to understand the debate around polarization measurements.
>
> I understand that with the pair spectrometer option a silicon
> telescope is placed in front of the magnet but after a 1.5 meter drift
> space after the radiator. You measure a phi dependence in the yield
> of pairs produced at 90 degrees in the CM frame (symmetric). From the
> opening angle you know the gamma energy and then QCD gives you the
> analyzing power.
>
> The other way it is proposed to measure the polarization is to measure the
> shape of the coherent bremstrahlung peak. Richard claimed in the meeting he
> could get a 1% measurement rather quickly from this. Is there a note which
> talks about this somewhere? I was wondering what systematic effects could
> enter this determination. To me a changing beam emittance, beam size, and
> a crystal mis-orientation would effect the shape in a complicated way.
>
> Thanks Jim
>
>
>

===============================================================
                  Franz J. Klein, Associate Professor
                  CUA, Department of Physics
                  Washington, DC 20064
  office: Hannan Hall 206          phone: 202-319-6190
  or: Jefferson Lab,CC F-243       phone: 757-269-6671
---------------------------------------------------------------