[Clascomment] OPT-IN: Photoproduction of the f1(1285) Meson

[Reinhard Schumacher]

Dear Moskov,

In the revised version of the paper, which should appear very shortly, 
we have taken another look at the comparison of the eta' cross 
sections.  Thanks to your concern, we remade Figure 7 of the paper to 
show the comparison of the eta' cross section in the present analysis 
with two previous CLAS published results.

1) The first comparison is with the Williams et al. paper.  The analysis 
method was a partial wave analysis based on event-by-event maximum 
likelihood fitting using events in which the proton pi+ and pi- were 
detected.  The Q value method was used to identify the reaction of 
interest in which an eta was missing.  This work used the g11 data set.

2) The second comparison is with the Dugger et al. paper.  The analysis 
method was to use single-arm detection of the proton and to reconstruct 
the eta' from the missing mass.  The proton was identified using the 
simplest imaginable PID cuts, and the acceptance was computed the 
simplest possible way using GSIM. Background was subtracted bin by bin 
using a polynomial function. This work used the g1c data set.

3) The analysis we have done for the f1 is under collaboration review.  
The analysis method was to use kinematic fitting of 3-track events 
containing proton pi+ pi- to identify events with the eta missing.  
Background was subtracted bin-by-bin with a polynomial function.   The 
acceptance was computed with GSIM and used all the corrections developed 
for the very well-studied g11 data set. The analysis did not use the 
Q-value method.

The three-way comparison among these analyses, that use three different 
approaches, shows fair to good agreement among the three. The bins I 
care about the most are for W=2.35 GeV and above, since that is where 
the f1 cross section is extracted.  This energy is above the energy 
published in the Dugger paper.  But at 2.15 and 2.25 GeV the 3-way 
agreement is good.   Thus we have three independent analyses arriving at 
the same result.  I see for the first time that the agreement between 
the three analyses is not so good at the lowest energy bin of 2.05 GeV.  
We have not pursued what is going on in this bin since it is not in the 
range of energies where the results of this paper are being 
presented.    We can just note that the lowest-energy bin is just 100 
MeV above threshold, where complications due to energy loss and 
acceptance may be worst - not an issue facing us in the present paper in 
the bins we care about.

We also understand where your preliminary analysis of the g12 data is 
having problems. Specifically, you mentioned the decay mode of the eta' 
to eta pi0 pi0, followed by the eta decaying to pi+ pi- pi0.  If you 
detect one or two of those charged pions then you don't reconstruct the 
eta correctly in the missing mass off the proton pi+ and pi-.  As 
Cathrina showed in her talk at the last working group meeting, the 
rejected events have bumps in missing mass off the proton at the 
location of the eta' and the f1.  You were worried about background that 
peaks at the location of the desired particles, so that, depending on 
how you chose to do your background subtractions and acceptances, you 
could get a wrong result.

Consider this.  Yes, there are eta decays that create charged pions that 
may be mistaken for the pions you want to detect from the decay of the 
eta' to eta pi+ pi-.  And yes, those events end up in the "background" 
outside of the selected eta band.  In Cathrina's talk this is 
illustrated on slides 8 and 9.  On slide 8 one sees the Q-value 
selection of the eta.  Events in the "red" group are rejected.  The 
eta-identification of a genuine eta was corrupted by the mistaken 
selection of the "wrong" pions while testing the identity of the missing 
particle.  However, just because the eta was (wrongly) rejected in the 
missing mass off p pi+ pi-, it still follows that the eta' or the f1 
should can show up in the missing mass off the proton.  This is shown on 
slide 9.  It is totally natural that events in the "red" group, events 
that were rejected because the eta was not successfully reconstructed, 
should nevertheless remain uncorrupted as far as the missing mass off 
the proton in concerned.  The proton track was not disturbed by the fact 
the that a wrong pion was selected.

Thus, it would be a mistake to use the rejected "red group" events for 
any further work toward computing the cross section.  The next step in 
the calculation is then calculation of the acceptance.  This acceptance 
calculation must include the issue of pion combinatorics and possibly 
picking the wrong pion, as outlined above.  The proper procedure is to 
simulate the full decay chain of the event, with proper branching 
fractions, etc., and submit those simulated events to the same analysis 
chain.  That is only natural, and I am sure you do it too.  Thus, the 
events that get "lost" due to eta's decaying in an awkward way are 
statistically compensated-for by the acceptance calculation that 
recovers their number.  This is procedure we used when extracting the 
cross section for f1 photoproduction and also for the eta' 
photoproduction channels.  The cross section that comes out will be the 
correct one.

Another place to look into this for clarification is Figure 3.2, and the 
discussion around it, in Ryan Dickson's thesis.  It shows the 2D plot of 
MM(g, p pi+ pi-) vs. MM(g, p) and various projections.  You can see by 
eye the events we are talking about.

I'm not sure how you want to solve the problem within your analysis:  
it's not my task here, but it should be straightforward. When you get to 
the stage of a technical review of a complete analysis I would be happy 
to contribute.

Best Regards,
Reinhard

____________________________________________________________________
Reinhard Schumacher         Department of Physics, 5000 Forbes Ave.
Carnegie Mellon University, Pittsburgh, PA 15213, U.S.A.
phone: 412-268-5177         web: www-meg.phys.cmu.edu/~schumach
____________________________________________________________________