CMW long paper [up-to-date status: submitted to PRC]
***********************************************************************************************************
* Search for the Chiral Magnetic Wave with anisotropic flow of identified particles at RHIC
* PAs: Qiye Shou, Gang Wang, Hongwei Ke, Maria Sergeeva, Aihong Tang, Haojie Xu
* Target Journal: PRC
***********************************************************************************************************
* Corresponding short paper (Observation of charge asymmetry dependency of pion elliptic flow and possible chiral magnetic wave in heavy ion collisions)
* Analysis note: https://drupal.star.bnl.gov/STAR/starnotes/private/psn0603
* STAR Collaboration, Phys. Rev. Lett. 114, 252302 (2015).
***********************************************************************************************************
_____________________________________________________________________________________________________________________________________________________________________________
Analysis Note
https://drupal.star.bnl.gov/STAR/starnotes/private/psn0700
_____________________________________________________________________________________________________________________________________________________________________________
Code
$CVSROOT/offline/paper/psn0700/
_____________________________________________________________________________________________________________________________________________________________________________
Paper Draft
paper_version1
paper_version2
paper_version3
paper_version4
paper_version5
paper_version6
See file attachments below for the latest version in 2022
_____________________________________________________________________________________________________________________________________________________________________________
GPC
[256] Sergei Voloshin (Chair), Shusu Shi, Richard Seto (English QA), Niseem Abdelrahman (Code QA), Jiangyong Jia (PWG Rep), Qi-Ye Shou (PA Rep), Gang Wang (PA Rep)
1st round
The 2nd round comments include a few minor modification of expression (a bit softened conclusion) and language improvement.
_____________________________________________________________________________________________________________________________________________________________________________
Institutional Review
Comments and replies
_____________________________________________________________________________________________________________________________________________________________________________
Paper Proposal
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
First round PWG: Aug 31, 2016
Slides: cmw_paper_proposal_bulk.pdf
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
First round PWGC: Oct 7, 2016
Minutes: http://www.star.bnl.gov/HyperNews-star/protected/get/starpwgc/4246.html
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Second round PWG: Nov 8, 2017
Slides: cmw_paper_proposal_bulk_2.pdf
Minutes:
Q: For fig. 2 and 3, what about directly presenting delta(pT)/<pT> and delta(v2)/<v2>?
A: The relative variation of <pT> and v2 is more meaningful than absolute variation. Therefore, we have already included the number of relative variation in the text, “… because over the same Ach range, the relative variation of ⟨pT ⟩(∼1‰) is typically smaller than the relative variation of v2 (∼1%) by an order of magnitude.” For the figures, the present way of illustrating (as functions of Ach) could provide a direct impression to the reader. We think the combination of the text and figure is clear enough to make our point.
Q: For fig. 5 and 7, what about putting pi,k,p linear dependences together in the same figure? For fig. 6, using pt<1GeV/c for pi?
A: It’s ok to put the plots of delta(v2) vs Ach for pi,k,p together, so we’ve made a backup version for further decision. However, the plot becomes less clear if putting v2 vs Ach for k and p together due to the difference of the y-axis range. In fig.6, the pion results are directly from the published paper. In addition, the previous section has proven that the results of pt<0.5 and <1 are consistent with each other, so we still prefer to use <0.5 for pions.
Q: Why larger systematic error bars for proton?
A: The larger systematic errors for proton mainly come from the source of PID. We apply both TPC and TOF for the central value, while only TPC for the systematic estimation. More uncertainties are introduced without TOF.
Q: What about presenting v3 linear plots for all centralities to see how slopes change from neg to pos?
A: We think it’s feasible to show the slopes in all centralities so we’ve made an alternative version for further decision. We prefer to directly present v3 than normalized v3 here.
Q: What about adding the curve/band of theoretical prediction? Why most peripheral (and most central) of A+A negative? What about using high multiplicity for p+A?
A: The theoretical prediction would basically be zero for p(d)+Au collisions. We’ve talked with a few theorists and will think about add a curve/band here. Results of high multiplicity (most central) p(d)+A are still consistent with zero and are in the note. For the paper, we don't have strong intention to particularly include it.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
CME focus group: Feb 27, 2018
Slides: cmw_paper_proposal_cmefg.pdf
Minutes:
Figure 3, the offset of dv2 depends on pT, but not the slope. This is due to the inclusive +/- particles don't have the same v2, and +/- yield ratio may also depends on pT.
Figure 4, what is the interpretation of the differences in very peripheral events? can this be local-charge conservation? (fig 3 only shows mid-central)
Figure 6, why the Kaon and Pion have similar slopes? If CMW picture is correct, the heavy strange quark mass must lead to a smaller signal for kaon? Suggest to make a statement in the text to point out this puzzling results. It probably should also be mentioned in the conclusion.
Figure 10, suggest to change the legends to be dv2/v2 and dv3/v3 (or n=2 and 3). Lots of discussions on similarities between dv2 and dv3 at mid-central 30-50%. The conclusion stated that for a wider centrality range, 20-60%, the difference is 2.9 sigma. But this really depends on which centrality range to use. The difference between 30-50% centrality would suggest there is no difference. Request to clarify this point in the text, the conclusion on this (2nd paragraph on page 16) need to be softened and need to mention both the signal and background interpretation. e.g. just say clear difference in central but difference become much smaller in mid-central. This trend is insistent with pure LCC scenario and pure CMW scenario. Maybe also mention the behaviour is different from LHC.
Figrue 11, the U+U dv2 slope is larger than Au+Au slope for the same Npart. The expected magnetic field is also stronger for same Npart from Glauber model calculation.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Second round PWGC: Mar 9, 2018
Minutes:
The conveners have reviewed the BulkCorr paper proposal for the second time. The first review was conducted on 10/7/2016. Prior to this second preview, the PAs have completed the steps described in http://www.star.bnl.gov/HyperNews-star/protected/get/bulkcorr/3644.html
The panel finds that the analysis is mature, the targeted journal is appropriate and the paper proposal should move forward. The following comments were made during the discussion.
The conclusion that the CMW picture stays a viable interpretation should be softened, and include in the language both signal and background, e.g. neither CMW nor LCC alone can describe data.
A: The conclusion is the most important part of the paper. Due to the fact that several different comments have been proposed during the QM preview meeting, we think it would better to collaborate with GPC to reach a consensus.
Add X-axis labels to figures with multiple panels (e.g. Figure 2)
Making the marker styles/colors consistently representing the same quantity for the same particle species across the paper
A: Done.
Figure 2: The figure shows little dependence of mean pT on Ach and difference between pi+ and pi- for the lowest pT region (0.15significant when the pT range is increased to <1 or <2 GeV/c. The main motivation for this figure is to explain why we would like to focus on the lowest pT region, in contrast to LHC, where much wider pT ranges were used. It was pointed out that this figure without a good explanation for the change of the behavior
for different pT regions may lead to questions on detector effects and consequently our physics results. It was suggested that PAs should try to understand these observed dependences.
Figure 3: only the statistical uncertainties are plotted for this figure. It was suggested to explain in the paper text why no systematic uncertainty is necessary for this figure.
Figure 4: PAs are still trying to understand the different results in the peripheral bin with different pT and eta cuts
A: We believe that there could be some additional reasons concerning this result. PAs are performing an independent model study to try to understand it.
Figure 7 (or the alternative version Figure 5 on slide 12): use $p$ and $\bar{p}$ instead of p+ and p- in the legend
Figure 10: add pion to the legend
A: Done.
Table I: both the statistical and systematic uncertainties need to be considered to quantify the significance level for the difference between the normalized v2 and v3 slopes. When comparing the normalized v2 and v3 slopes for several centrality bins together, the uncorrelated and correlated uncertainties should be properly calculated and combined.
A: Have been updated. And now we're thinking about if it's really necessary to have this table. If so, we prefer to do in centrality 10% by 10%, instead of the previous way of combining. We'll work with GPC to find the optimal solution.
Tracking efficiency correction is applied to Ach and mean pT, but not to the pT-integrated v2. Based on the known pT dependence of v2 and tracking efficiency, the effect should be negligible. The PAs were asked to include sufficient info to justify the conclusion in the AN.
A: Such effect has been tested, and we confirm that it's negligible. We've included the info in the note.
_____________________________________________________________________________________________________________________________________________________________________________
Abstract
In this more comprehensive study, we present the STAR measurements of elliptic flow and triangular flow ($v_{3}$) of charged pions, along with the $v_{2}$ of charged kaons and protons, as functions of $A_{\rm ch}$ in Au+Au collisions at $\sqrt{s_{\rm NN}}$ = 27, 39, 62.4 and 200 GeV. The slope parameters of $\Delta v_{2}$($A_{\rm ch}$) or $\Delta v_{3}$($A_{\rm ch}$) for different particle species are reported and compared in different centrality intervals. The $\Delta v_{2}$($A_{\rm ch}$) slopes for charged pions in small systems (p+Au and d+Au at $\sqrt{s_{\rm NN}}$ = 200 GeV) are also presented and compared with those in large systems (Au+Au at $\sqrt{s_{\rm NN}}$ = 200 GeV and U+U at 193 GeV). Our results provide new insights for the CMW picture, and further constrain the background contributions, such as local charge conservation, in heavy-ion collisions at RHIC energies.
Conclusions
The previous experimental evidence of the CMW, the $\Delta v_{2} (A_{\rm ch})$ slope for pions, has been challenged by the LCC effect and the isospin effect. In this work, we present the $\Delta v_{2} (A_{\rm ch})$ slopes for low-$p_{T}$ kaons in Au+Au collisions at $\sqrt{s_{\rm NN}}$ = 27, 39, 62, 200 GeV. The similarity between pion and kaon slopes suggests that the isospin effect is not the dominant contribution to the pion or kaon slopes. The isospin effect, however, remains a potential contributor to the proton slopes in Au+Au collisions at $\sqrt{s_{\rm NN}} = 200$ GeV.
The LCC background has been investigated via the pion $\Delta v_{3} (A_{\rm ch})$ slopes in Au+Au collisions at $\sqrt{s_{\rm NN}} = 200$ GeV. The measured $\Delta v_{3}$ slopes are typically negative or consistent with zero, in contrast to the positive values predicted by the LCC effect. With data combined over wide centrality intervals, we normalize the $\Delta v_{2}$ and $\Delta v_{3}$ slopes. The former is above the latter in all cases. The difference between the normalized $\Delta v_{2}$ and $\Delta v_{3}$ slopes for pions with $0.15<p_T<0.5$ GeV/$c$ is a $2.9\sigma$ effect for $20-60\%$ collisions, where the CMW signal is expected to be most prominent. Nevertheless, the similarity between the $\Delta v_{2}$ and $\Delta v_{3}$ slopes in the $30-40\%$ centrality as well as the negative $\Delta v_{3}$ slopes at most central and peripheral collisions call for further theoretical inputs. In short, the CMW picture stays a viable interpretation of the pion $\Delta v_{2} (A_{\rm ch})$ slopes, but further investigation of the LCC effect is warranted in future.
The pion $\Delta v_{2} (A_{\rm ch})$ slopes have also been reported for p+Au and d+Au collisions at $\sqrt{s_{\rm NN}}$ = 200 GeV and U+U collisions at $\sqrt{s_{\rm NN}}$ = 193 GeV. The CMW signals in the small systems are expected to disappear owing to the orientation decoupling between the magnetic field and the 2nd-order event plane. The measured slopes are consistent with zero. The difference in the pion $\Delta v_{2} (A_{\rm ch})$ slope between Au+Au and U+U qualitatively resembles the expectation from the CMW picture. Another important systematic, the $\langle p_{T} \rangle$ effect on $A_{\rm ch}$ (and further on $\Delta v_{2}$) has also been discussed in length. In general, the $\langle p_{T} \rangle$ effect is not noticeable in the slope measurements, except for the difference between normalized $\Delta v_{2}$ and $\Delta v_{3}$ slopes.
_____________________________________________________________________________________________________________________________________________________________________________
Supporting Material
Preliminary v3 and K results presented in QM2014
Updated version of preliminary K results (Qiye's talk at Chirality Workshop 2016)
Do pions from lambda affect the linear relationship?
https://www.star.bnl.gov/protected/bulkcorr/rexwg/flow/Parity/CMW/pAu_dAu/
_____________________________________________________________________________________________________________________________________________________________________________
- qyshou's blog
- Login or register to post comments