Re: LV12066 Measurements of the electric form factor of the neutron up to Q 2=3.4 GeV 2 using the reaction 3 vec He( vec e,e prime n)pp by S. Riordan, S. Abrahamyan, B. Craver, et al. Dear Dr. Wojtsekhowski, The above manuscript has been reviewed by our referees. We ask you to consider the enclosed comments from the reports. While we cannot make a definite commitment, the probable course of action if you choose to resubmit is indicated below. ( ) Acceptance, if the editors can judge that all or most of the criticism has been met. (x) Return to the previous referee(s) for review if available. ( ) Submittal to new referee(s) for review. With any resubmittal, please include a summary of changes made and a brief response to all recommendations and criticisms. Yours sincerely, Abhishek Agarwal Assistant Editor Physical Review Letters Email: prl@ridge.aps.org Fax: 631-591-4141 http://prl.aps.org/ Physics - spotlighting exceptional research: http://physics.aps.org/ ---------------------------------------------------------------------- Report of Referee A -- LV12066/Riordan ---------------------------------------------------------------------- The authors report a measurement of the neutron electric form factor up to higher momentum transfer than is now available. The technique is sound and the researchers are known to me as leaders in this field, with excellent track records for quality experimental work. The result is important and timely, and is well within the constraints for publication in Physical Review Letters. Indeed, I urge that it be published soon, although I strongly suggest that something be added, namely a better discussion of systematic error. The two figures show what appears to be a significant inconsistency with the result of Ref [6] Plaster, et al, which includes many of the present authors. That is, the highest momentum transfer point in [6] is nearly the same as the lowest in this paper, yet the measured GEn values differ well outside the quoted error bars. The two techniques are different enough, so that the systematic errors might be quite different, and the ~2 standard deviation difference is hard to attribute to the slightly different momentum transfers. (I am somewhat bewildered that these authors neglected to take a data point that overlaps in momentum transfer.) That is not to say that either of these experiments has some flaw. It nevertheless behooves the authors to address the potential systematic error so that experts can evaluate one experiment against the other. Indeed, Table III shows that a ~10% systematic error dominates the uncertainty in the first data point. However, the only discussion that I see of systematic error is in the second paragraph in the first column of page four ("The final steps...") which concludes with the statement that "The systematic uncertainty was obtained by combining in quadrature the contributions of individual effects (the largest of which were described above)." All I see there are 2% and 3% uncertainties, a far cry from 10% added in quadrature. Presumably other effects, which are discussed but without numerical details, give rise to the quoted, total systematic error. So, I suggest a table be included with lists the systematic error contributions, along with their quadrature sum, for each of the three momentum transfer points. I suspect that such a table should be easy for the researchers to include. I realize that there are page limit constraints for Phys Rev Letters, but, in my opinion, this is much more important than including, for example, three different ways to plot the data, equations (1) or (2), or even the detailed numbers given in Tables I and II. ---------------------------------------------------------------------- Report of Referee B -- LV12066/Riordan ---------------------------------------------------------------------- This manuscript reports on new measurements of the electric form factor of the neutron. This is one of the most fundamental properties of the nucleon, and it is not fully understood. The new data add most significantly to the rather small set of reliable and accurate data on this quantity. The momentum range over which the form factor is measured here doubles the existing range. As such, these data deserve to be published in PRL. The paper is well written, and contains the essential information on the experiment and analysis. It has a very good introduction showing the importance of the new data to non-experts in the field of nucleon structure. The conclusions are valid and clear. Since this paper addresses an important puzzle in nucleon structure, and adds significantly to our knowledge base in this area, I recommend publication in Physical Review Letters. ---------------------------------------------------------------------- Report of Referee C -- LV12066/Riordan ---------------------------------------------------------------------- This letter reports new high-impact measurements of the neutron electric form factor, GEn, obtained from measurements of the double-polarization asymmetry in polarized-electron, polarized-3He quasielastic scattering. As summarized nicely by the authors, measurements of both the proton and neutron electromagnetic form factors are important for an understanding of nucleon structure; however, until now, the Q^2 range for GEn was severely restricted relative to those of the other three form factors. Data on the neutron form factors, especially GEn, are needed in the Q^2 = 1 - 6 GeV^2 range, where unexpected results for the proton form factors emerged from experiments utilizing spin degrees of freedom. Indeed, as the authors show, while many theoretical models of nucleon structure yield agreement with the proton form factor data, the results of these model calculations for the neutron form factors tend to diverge at Q^2 values > 1.5 GeV^2 (i.e., beyond the existing data on GEn). The important new results reported in this letter have more than doubled the existing Q^2 range of data on GEn (out to Q^2 = 3.4 GeV^2), thereby providing important constraints on competing theoretical models (and, ultimately, as benchmark tests of lattice QCD calculations). Indeed, any successful model or calculation must be able to reproduce data on both the proton and neutron form factors. The results of this experiment have been long anticipated and should merit publication in Physical Review Letters, as they are of broad interest to those working in hadron/QCD structure. However, I have several comments which the authors will need to address in a satisfactory manner prior to my recommending this paper for publication in PRL. Primary Comments: ---------------- (1) The results for the form factor ratio g_n quoted in Table III list relative systematic uncertainties of 10.77%, 8.56%, and 8.94% at the three Q^2 points. There appears to be no accounting within the letter for the origin of these quoted systematic uncertainties. The authors state that "The systematic uncertainty was obtained by combining in quadrature the contributions of individual effects (the largest of which were described above)." However, the only statements I could find of any systematic uncertainties were: (a) A statement that the beam polarization was determined with a "relative accuracy" of 3%; and (b) A statement that the target polarization was measured every six hours to a "relative accuracy" of 4%; and (c) A statement that the estimated accuracy of the GEA calculations was 2%. These certainly do not add in quadrature to ~10%. I see no mention anywhere else in the text as to what the additional sources of the quoted ~10% systematic uncertainty may be. I would strongly suggest that the authors list the uncertainties in all of the parameters reported in Table II (for all those parameters following the list of analysis cuts). It is typical for letters to report their systematic error budget, and I believe it is the authors' responsibility to report theirs. In particular, I note that the D_p/n charge-exchange dilution factor is quite large, ~20%, and it would be quite interesting to know what the uncertainty in this factor was. (2) The dilution from inelastic pion-production events, D_in, is relatively large (~15%) at the highest Q^2 point. It is then stated that the correction for these events is small, "thanks to" the fact that the observed asymmetry appeared to be quite close to the inelastic asymmetry. It is not clear to me how this comparison was made. Presumably the inelastic asymmetry could be measured at kinematics away from the quasielastic "peak". However, it is not clear if the experimental acceptance permitted such a measurement. The questions then become: What exactly is the inelastic asymmetry that is reported in Table II: a measurement or a calculation? If it is a measurement, how was the asymmetry extrapolated into the quasielastic peak? Was the asymmetry proven to not be a function of the invariant mass? If it is a calculation, how were the calculations bench marked for these particular kinematics? I also note that the A_in is significantly larger (-0.254) at the middle Q^2 point? (This also does not agree with the authors' statement that the observed asymmetry was quite close to the inelastic asymmetry.) Other Comments: -------------- (3) It is stated that the scraping of the target cell was "regularly investigated", but no statement is given as to how small/large the effect was found to be. (4) How uniform was the holding field over the length of the 40-cm target? Alternatively, it is stated in the text that the field direction was measured to an accuracy of 1 mrad. However, if the field is not perfectly uniform and varies slightly over the target cell length, the spin angles theta* and phi* would presumably vary according to the location of the event vertex along the cell, and it would then need to be known to what accuracy the event vertex was reconstructed. (5) The vetos were stated to be 97% efficient. Does this number refer to all protons incident on the detector, or to only those protons incident on the vetos? In other words, to what degree did the vetos form a "hermetic surrounding seal" for the tagging of all possible p-n charge-exchange events occurring anywhere in the experimental hall? (6) It would be useful to state explicitly what the Geant4 simulation was used for, other than that it was "found to be in good agreement with the detector characteristics". For example, was the Geant4 simulation (with its presumably full detector geometry) employed for the calculations of the dilution factors D_in and D_p/n, and in the averaging of the GEA calculations over the kinematic acceptance? (7) I assume the statement "[a]n appropriate average of A^p and A^a" simply means a statistically-weighted average of the two datasets. (8) It is stated in the text that corrections for nuclear re-scattering effects were found to be small (within 3% of the plane-wave values), but it is not clear from the text if the reported values for g_n were actually corrected for these effects, or if the ~3% factors were absorbed into the systematic error budget. In either case, it would be useful to report the values for all three of the Q^2 points.