[Halld-physics] Eta Proposal

[Matthew Shepherd]

Hi Liping and Sascha,

I've had a chance to read through your eta proposal and have some questions.

I'm looking through plots to try to find some demonstration of the expected background levels in photoproduction.  Have you done this exercise (e.g., with the Pythia generator tuned for GlueX)?  

The proposal seems to assume that backgrounds in other experiments are an indicator of what is expected in GlueX.  This may be true for cross feed from other eta decay modes, but doesn't, in general, apply to other sources.  Photoproduction in GlueX is different all the other production mechanisms you list, with the exception of Crystal Ball at Mainz which has the following curious statement in your text:

"Using a bremsstrahlung photon beam from a 1.5 GeV electron beam, they are continuing a rare eta decay program using the same reaction as us, gamma p to eta p. Their eta production rates are reportedly over an order of magnitude higher than that expected in JEF, but so are their backgrounds."

Does this imply that we haven't done a good enough job studying our backgrounds?

You list three "priority" channels:

1) pi^0 gamma gamma:  measured before, precision measurement needed
2) pi^0 pi^0:  CP violating
3) eta -> pi0 gamma or 3gamma

The second process has a limit in the charged channel set at < 1.3 x 10^-5.  This implies that the neutral channel is limited to something like < 7 x 10^-6 (assuming isospin conservation).  Are there predictions for CP violation or other physics that would show up in the neutral channel but not the charged channel?  If not, would your limits be competitive with the implied 7 x 10^-6 limit?

It seems like the first process is a background for the second process.  How will you separate 2 from 1 (or 2 from other backgrounds)?  I can imagine some multi-dimensional fitting methods, but all of these will require some statistics.  The text says you make cuts around pi0 mass.  What fraction of the eta gamma gamma background passes these cuts?  Do you have some plots to illustrate the analysis for those bits of text -- I'm having a hard time following the arguments in the proposal.  This is around pages 46-47.  (How do you estimate the candidates remaining?  Candidates for what?)

I think that channel 3 will have similar problems that John encountered with the e+e-.  You say that the eta cross section is 70 nb.  The rho cross section is 20 microbarns.  The rho decay to pi0 gamma is about 10-3.  The eta to pi0 gamma limit is at 10-4.  Assuming the true eta -> pi0 gamma BF is at the limit you have:

( sigma_rho / sigma_eta ) * ( rho -> pi gamma / eta -> pi0 gamma ) ~ 3000

So you will produce one eta -> pi0 gamma event for every 3000 rho -> pi0 gamma events if the BF is at the current limit.  Of course the rho events will peak around the rho mass, but the tail extends well into the eta region.  How does this affect your sensitivity?  More importantly:  what other large photoproduction cross sections may contribute to the background?  I'm sure that Pythia has all of this stuff in it, but I haven't seen a detailed analysis of the Pythia backgrounds yet.

It seems like for such a rare decay proposal, where you know both exactly what you are looking for and the rate at which you want to be sensitive, it would be helpful to see what types of backgrounds Pythia generates that may leak into the signal region.  Then you can check if those photoproduction reactions have been studied (and compare measured rates with generated rates in Pythia to check).  There can always be the unknown background, but the minimum goal should be to try to probe all possible known backgrounds.

Cheers,

Matt