<html>
<head>
</head>
<body text="#000000" bgcolor="#ffffff">
Hovannes,<br>
<br>
I have now got this code working, just this morning. Here are some
results for you to check out against known references. If you want,
I can just answer specific questions you have, or else I can give
you the output from the program for you to play with yourself. <br>
<br>
E(gamma) = 9.0 GeV, linear polarization 100%<br>
pair plane asymmetry = 13.0 +/- 0.3 % without cuts on
electron/positron energy<br>
pair plane asymmetry = 24.5 +/- 0.5 % with cuts 4.0 < E(electron)
< 5.0 GeV<br>
<br>
<a href="http://zeus.phys.uconn.edu/halld/pairspectrometer/xsect-1-2011/allpairs.png">http://zeus.phys.uconn.edu/halld/pairspectrometer/xsect-1-2011/allpairs.png</a><br>
<a href="http://zeus.phys.uconn.edu/halld/pairspectrometer/xsect-1-2011/allpairs.png">http://zeus.phys.uconn.edu/halld/pairspectrometer/xsect-1-2011/sympairs.png</a><br>
<br>
Plots are found at the above links. The code itself only computes
the basic Feynman diagram, which is differential in 5 variables. To
get useful results you have to integrate over many of these, which I
do using Monte Carlo integration. This is why there are error bars
on the cross section plots: they are the statistical errors coming
from the Monte Carlo integration. I have 10 million Monte Carlo
events, saved with 5 kinematic variables E(electron), phi(electron),
M(e+,e- pair mass), Q2 (recoil momentum squared), and phi(recoil).
Saved together with these 5 generated variables in a root Tree is
the differential cross section (diffXS), the Monte Carlo weight
factor (weight), and their product (weightedXS) that is most useful
for making plots. Let me know if you want access to these data.<br>
<br>
-Richard J.<br>
<br>
<br>
<br>
<br>
On 1/19/2011 11:53 AM, Hovanes Egiyan wrote:
<blockquote type="cite" cite="mid:4D37171C.40809@gmail.com">
<pre wrap="">Hi Richard,
during last Hall D beamline meeting you said that you could setup
a calculator for me to evaluate the azimuthal asymmetries for
pair production with linearly polarized photons. What you needed
was an atomic formfactor of my choice. I would suggest to use the
form factor for the case of complete screening :
F(q) = 1/ ( 1 + ( 111*q*Z^(-1/3) )^2 )
I did find some parameterizations of the atomic formfactors but those would
need to be coded, and I think it would be useful to first check what we
get with
this formfactor.
thanks in advance,
Hovanes.
</pre>
</blockquote>
<br>
</body>
</html>