Cold nuclear matter effect measured with high
This presentation is the property of its rightful owner.
Sponsored Links
1 / 35

Takao Sakaguchi Brookhaven National Laboratory For the PHENIX Collaboration PowerPoint PPT Presentation


  • 85 Views
  • Uploaded on
  • Presentation posted in: General

Cold nuclear matter effect measured with high p T hadrons and jets in 200GeV d+Au collisions in PHENIX. Takao Sakaguchi Brookhaven National Laboratory For the PHENIX Collaboration. Why are we interested in d+Au collisions?. PHENIX, Phys. Rev. Lett . 91, 072301 (2003) .

Download Presentation

Takao Sakaguchi Brookhaven National Laboratory For the PHENIX Collaboration

An Image/Link below is provided (as is) to download presentation

Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author.While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server.


- - - - - - - - - - - - - - - - - - - - - - - - - - E N D - - - - - - - - - - - - - - - - - - - - - - - - - -

Presentation Transcript


Takao sakaguchi brookhaven national laboratory for the phenix collaboration

Cold nuclear matter effect measured with high pT hadrons and jets in 200GeV d+Au collisions in PHENIX

Takao Sakaguchi

Brookhaven National Laboratory

For the PHENIX Collaboration


Why are we interested in d au collisions

Why are we interested in d+Au collisions?

PHENIX, Phys. Rev. Lett. 91, 072301 (2003)

p0 in Au+Au at 200GeV

cartoon

  • In order to confirm that the high pThadron suppression in Au+Au collisions is due to final state effect, and not cold nuclear matter (CNM) effect

    • Need system without additional effects from a hot medium.

  • CNM effect include:

    • kT-broadening (Cronin enhancement at moderate pT)

    • Shadowing of parton distributions

    • Cold nuclear matter energy loss

    • And possibly more…

  • d+Au is more favorable for RHIC operation

    • p+Au becomes feasible now (M.Bai)

T. Sakaguchi, [email protected] Valley


Nuclear pdfs are centrality dependent

Nuclear PDFs are centrality dependent?

Helenius, Eskola et.al. published centrality dependent nuclear PDFs (arXiv:1205.5359)

Compared to PHENIX p0RdA published in 2003

Theory curves are scaled up/down within systematics

New data can help better constraining nPDFs

T. Sakaguchi, [email protected] Valley


Phenix detector

PHENIX Detector

View From Beam

View From Side

PHENIX recorded d+Au events of 80 nb-1 in 2008 (2.74 nb-1 in 2003)

  • Photon measurement

    • EMCal(PbSc, PbGl): Energy measurement and identification of real photons

    • Tracking(DC, PC): Veto to Charged particles

  • Charged particle measurement

    • Tracking measure momentum

    • RICH: Identify electrons

    • EMCal add information on identifying electrons

  • Event triggered by a coincidence of BBC South and BBC North

    • Sitting in 3.1<|h|<3.9

    • Centrality defined by BBC south charge (Gold going direction)

T. Sakaguchi, [email protected] Valley


P 0 and h reconstruction

p0 and h reconstruction

T. Sakaguchi, [email protected] Valley


How we measure p 0 h

How we measure p0, h?

p0

PRC82, 011902(R) (2010)

p0

h

arXiv:1208.2254,

accepted for PRC

h

  • Reconstruct hadrons via 2g invariant mass in EMCal (example is in Au+Au)

  • Subtract Combinatorial background

    • Compute Mass using gs from different events. (mixed-event technique)

T. Sakaguchi, [email protected] Valley


Systematic errors

Systematic errors

p0 systematic errors

h systematic errors

Type A: point-by-point fluctuating errors

Type B: pT-correlated errors

Type C: overall normalization errors

T. Sakaguchi, [email protected] Valley


P 0 and h p t spectra in 200gev d au collisions

p0 and hpT spectra in 200GeV d+Au collisions

h

p0

T. Sakaguchi, [email protected] Valley


P 0 r da by centrality

p0 RdAby centrality

  • New p0RdA from Run8

  • Better statistics than Run 3

    • Extends pT reach by 5 GeV/c

    • Better constraint for nPDFs

  • Peripheral is most enhanced

  • Central consistent with no modification at pT > 2 GeV/c

  • How do we understand this?

  • Competing nuclear effects?

T. Sakaguchi, [email protected] Valley


H r da by centrality

hRdAby centrality

  • New hRdA from Run8

  • Better statistics than Run 3

    • Extends pT reach by 5 GeV/c

    • Better constraint for nPDFs

  • Peripheral is most enhanced

  • Central consistent with no modification at pT > 2 GeV/c

  • How do we understand this?

  • Competing nuclear effects?

T. Sakaguchi, [email protected] Valley


Jet reconstruction

Jet reconstruction

T. Sakaguchi, [email protected] Valley


Reconstruction of jets in p p

Reconstruction of jets in p+p

  • Gaussian Filter: Cone-like, but infrared and collinear safe

  • Shape of the filter:

    • Optimizes the signal-to-background by focusing on the core of the jet

    • Stabilizes the jet axis in the presence of background

  • Naturally handles isolated particles vs. collective background

T. Sakaguchi, [email protected] Valley


Reconstruction of jets in p p1

Reconstruction of jets in p+p

  • Gaussian Filter: Cone-like, but infrared and collinear safe

  • Shape of the filter:

    • Optimizes the signal-to-background by focusing on the core of the jet

    • Stabilizes the jet axis in the presence of background

  • Naturally handles isolated particles vs. collective background

T. Sakaguchi, [email protected] Valley


Systematic errors1

Systematic errors

Type B: pT-correlated errors

Type C: overall normalization errors

Bin-by-bin unfolding to correct for pT increase from mild d+Au underlying events (evaluated with embedding analysis)

Small residual fake rate (< 5%) above > 9 GeV/c

T. Sakaguchi, [email protected] Valley


Jet p t spectra in 200gev d au collisions

Jet pT spectra in 200GeV d+Au collisions

T. Sakaguchi, [email protected] Valley


Jets r da

Jets RdA

T. Sakaguchi, [email protected] Valley

  • Higher pT reach than p0s

  • Enhancement in peripheral,suppression in central.

    • Similar trend seen in p0 and h


Comparison to p 0 and h r da

Comparison to p0 and hRdA

p0 , h and jets of same pT sample slightly different parton scales, but let’s overlay them anyways…

Good agreement within uncertainties, and given the difference in observables.

T. Sakaguchi, [email protected] Valley


Comparison to p 0 and h r cp

Comparison to p0 and hRcp

p0 , h and jets of same pT sample slightly different parton scales, but let’s overlay them anyways…

Good agreement within uncertainties, and given the difference in observables.

T. Sakaguchi, [email protected] Valley


Comparison with models

Comparison with models

T. Sakaguchi, [email protected] Valley


Comparison to shadowing calculation i

Comparison to shadowing calculation (I)

p0RdA

  • (Very basic) shadowing calculation uses EPS09 PDF modification* + Glauber MC + PYTHIA (x,Q2) sampling for p0.

  • Shadowing effects match reasonably well within the global scale uncertainties in central events (where modification is weak), but is not compatible with the pT shape in peripheral.

  • *nPDF modification assumed to scale linearly with longitudinal nuclear thickness.

T. Sakaguchi, [email protected] Valley


Comparison to shadowing calculation ii

Comparison to shadowing calculation (II)

JetsRdA

  • (Very basic) shadowing calculation uses EPS09 PDF modification* + Glauber MC + PYTHIA (x,Q2).

  • Shadowing effects match reasonably well within the global scale uncertainties in central events (where modification is weak), but is not compatible with the pT shape in peripheral.

  • *nPDF modification assumed to scale linearly with longitudinal nuclear thickness.

~ 0.3

<x> ~ 0.2

T. Sakaguchi, [email protected] Valley


Interpretation of the results

Interpretation of the results

  • Strong centrality dependence of RdA

    • p0, h and jets RdA are in very nice agreement in spite of completely different systematics

    • nPDF may be strongly centrality dependent

  • Some thought

    • Before making the final conclusion, we would like to confirm some global things.

T. Sakaguchi, [email protected] Valley


How we define centralities

How we define centralities?

Charge distribution in BBC

(South, gold going direction)

60-88%

40-60%

20-40%

0-20%

  • Use Beam-Beam counter (BBC) installed in 3.1<|h|<3.9

    • Centrality defined by BBC south charge (Gold going direction)

    • Participant region

  • Compare with Monte Carlo simulations and determine TAB

    • Glauber calculation folded with negative binominal distributions (NBD)

T. Sakaguchi, [email protected] Valley


Possible dynamics in d au collisions

Possible dynamics in d+Au collisions

BBC

Mid-rapidity

Detector

  • We talk about peripheral collision case

  • Soft- and hard-dominated events may produce different hit distributions in BBC

  • In case that hardest jets are produced, less energy will be available for soft-production at high h

  • There could be pT dependent effect?

    • We use BBC for event triggering as well as centrality definition

T. Sakaguchi, [email protected] Valley


Pythia simulation

PYTHIA simulation

Number of hits in BBC vs

reconstructed jet pT

Fraction of reconstructed jets

when requiring BBC trigger

Errors are RMS’s of dists.

Ran PYTHIA and look reconstructed jets in mid-rapidity

PYTHIA sees a small anti-correlation effect

T. Sakaguchi, [email protected] Valley


Ampt p p simulation

AMPT p+p simulation

Number of hits in BBC vs

hadronpT in mid-rapidity

Fraction of hadrons

when requiring BBC trigger

Errors are RMS’s of dists

  • AMPT is a HIJING + hadron cascade event generator

  • Plotted as a function of hadronpT in mid-rapidity

    • Jets are not reconstructed.

    • AMPT also sees similar effect in single hadrons

T. Sakaguchi, [email protected] Valley


Conclusion

Conclusion

  • Very solid results from p0, h and jets on RdA that would help constraining nuclear PDF from d+Au collision data

    • Very consistent each other even though the systematics are very different

  • Huge enhancement in the yield in d+Au peripheral collisions compared to the one expected from p+p collisions

  • We are making several global checks in preparation for publishing the results.

T. Sakaguchi, [email protected] Valley


Backup

Backup

T. Sakaguchi, [email protected] Valley


Takao sakaguchi brookhaven national laboratory for the phenix collaboration

T. Sakaguchi, [email protected] Valley


R da of identified hadrons

RdA of Identified Hadrons

  • Mesons follow similar trend w/ pT in all centralities.

  • Cronin enhancement at moderate pT?

  • Or nPDF moving through antishadowing region into EMC region?

  • Proton enhancement still not explained by Cronin or shadowing models.

  • R. Hwa, et al. reproduced RCP using recombination of shower + thermal(?) partons (nucl-th/0404066).

T. Sakaguchi, [email protected] Valley


What is the merging effect

What is the merging effect?

Because of limited granularity of the detector, two g’s from p0 can not be resolved at very high pT (g’s merged. mass can not be reconstructed).

Opening angle: q ~ mass/pT

We corrected for the inefficiency due to merging, but also introduced a large systematic error.

Low pT

p0

High pT

p0

Probability of detecting two g’s from p0

PbGl

PbSc

pT of p0

T. Sakaguchi, [email protected] Valley


Takao sakaguchi brookhaven national laboratory for the phenix collaboration

60-88%

PHENIX Preliminary

0-20%

T. Sakaguchi, [email protected] Valley


P 0 r da

p0 RdA

T. Sakaguchi, [email protected] Valley


Cronin enhancement

Cronin Enhancement

p+

  • Enhancement of hadron production in heavy ion collisions

  • Usually modeled as multiple scattering of the incoming parton on the nucleus.

  • Most models don’t have any PID dependence…

  • However, measured enhancement is larger for protons than pions/kaons.

  • Originally thought to be due to steeper pT spectrum of protons and that it would go away at higher energies.

  • But proton enhancement is still much larger at RHIC energies!

K+

p

T. Sakaguchi, [email protected] Valley


Systematic errors2

Systematic errors

Type A: point-by-point fluctuating errors

Type B: pT-correlated errors

Type C: overall normalization errors

T. Sakaguchi, [email protected] Valley


  • Login