[Clascomment] OPT-IN:Measurement of the generalized form factors near threshold via gamma*p --> n pi+ at high Q^2

[Volker Burkert]

Dear lead authors:

You have invested a big effort in the analysis that is the basis of this paper, and it is an interesting topic. However, having looked at the paper for the first time in some detail I have some problems with the current paper and recommend not to push publication on it until the questions below are addressed. Perhaps these are pretty obvious questions and have been asked and resolved during the reviews in the committees already. In that case I apologize for asking possibly repetitive questions.  


1) This paper attempts to extract generalized form factors near pi+ threshold. The lowest W bin is W= 1.10-1.12 GeV with the bin starting just 20 MeV above threshold. The Q^2 range is from 1.9-4.5 GeV. At the lowest Q2 and W value the kinematics is nearly elastic and excellent knowledge of the electron acceptance at small scattering angles is needed. In addition, extracting cross sections this close to threshold requires excellent knowledge of the beam energy. I wonder if this has been studied, as nothing is written about it in the paper?  Also, the neutron miss mass resolution as shown in Fig. 8 as sigma~19MeV, i.e. the total resolution width is ~45 MeV. Also from previous analyses of the same date we know that the width of the elastic peak is about 50MeV, i.e. much larger than the 20MeV distance above threshold and the bin size of 20MeV. 

2) The beam energy is assumed to be 5.754GeV, and there is no discussion on this. What happens if the energy is 20MewV higher or lower? In several analyses of the e1-6 data the beam energy has been a subject of discussion and I have seen uncertainties in the beam energy of up to 20 - 30MeV.  While for deep inelastic processes the beam energy is less important, it is crucial to know the beam energy at a few MeV level for physics close to threshold. There is no discussion on this in the paper.

3) One result where threshold effect maybe visible and possibly generated what looks to me like unphysical results, is seen in Fig,13. For the lowest W=1.11 bin and the lowest Q^2=2.05 bin the cross section is smaller than the cross section for the higher Q^2=2.44 bin. If one compares with the MAID curves, they fit well the higher Q^2 data but are a factor 2 or more above the data at the Q^2-2.05 point. I think this is an unphysical result. 

4) Another problem with the data is the discrepancy between the extraction of the Legendre moments and the extraction of multipoles. The Q^2=2.05  D_0^(L+T} moment is =0 in Fig.16, while they are large at all higher Q^2 values. From the definition of D_0^(L+T) in eq. (5) one would conclude that both form factor G1^(npi)^2 and GMn^2  must be =0 if the moment is zero. However, we do know from CLAS measurements that G_Mn > 0 and large. Extracting E_(0+) multipole from its correspondence to G1^(npi) would then yield an unphysical value. This is not discussed. However, the results for E_(0+) extracted form the multipole analysis shown in Fig. 17-19 give E_(0+) large and positive. If one takes the dipole form factor normalization out the Q^2=2.05 point is larger than the higher Q^2 points. This is, of course, reasonable, but it seems to contradict what is seen in Fig. 13 and Fig. 16. 

5) The paper concludes in section VIII. summary that the E_(0+) multipole has been extracted for W=1.11-1.15 GeV. The final graphs show only results for the W=1.11 GeV bin. Since this lowest W bin is the one most vulnerable to calibration and acceptance uncertainties I wonder why the higher W bins are not presented. I think it is essential to show the E_(0+) multipole as extracted from the higher W values as well to show the consistency with the lowest W. Since this is a full research paper and no page limit, one should certainly expand on this. 

6) In general, many of the cross section points in Fig. 13-15 appear to fluctuate more than what would be allowed by statistics, even including systematic errors. This should give us pause when publishing this paper in the form it is in.

Volker Burkert