1

Search for effects related to Chiral Magnetic Wave at STAR

Gang Wang (UCLA)for STAR Collaboration

2

Motivation

Peak magnetic field ~ 1015 Tesla !

(Kharzeev et al. NPA 803 (2008) 227)

CSE + CME Chiral Magnetic Wave: • gapless collective excitation• signature of Chiral Symmetry Restoration

3

Observables

Formation of electric quadrupole: ,

where charge asymmetry is defined as .

Then π- v2 should have a positive slope, and π+ v2 should have a negative slope with the same magnitude.The integrated v2 of π- is not necessarily bigger than π+: (other physics)only the A± dependency matters for CMW testing.

Y. Burnier, D. E. Kharzeev, J. Liao and H-U Yee, Phys. Rev. Lett. 107, 052303 (2011)

4

Observables CMW + Parity-odd domain,

=> charge separation across RP

Kharzeev, PLB633:260 (2006)Kharzeev, McLerran, Warringa, NPA803:227 (2008)

RPad

dN

sin21

This charge separation effect needs to be beyond statistical fluctuation and conventional physics background.

5

STAR

6

STAR: excellent PID and tracking

The correlation measurements at STAR are accurate to relative 0.1%.

7

Observed charge asymmetry

• N+ (N-) is the number of positive (negative) particles within |η|<1.

• The distribution was divided into 5 bins, with roughly equal counts.

• Tracking efficiency was corrected later.

8

Charge asymmetry dependency

• v2 was measured with the Q-cumulant method.

• Clear A± dependency

• v2(A±) slopes for π±:• opposite sign• similar magnitude

• v2 difference vs A± may have a non-zero intercept: other physics?

9

Correction for tracking efficiency

• Fit with a straight line to extract the slope .

• Do the same for all centralities

10

Slope vs centrality

• Statistical errors only

• Smooth trend

• Positive and significant for most centrality bins

• Systematic uncertainty:• weak decays (Λ et al.)• tracking efficiency• <A±> bin center effect• different v2 methods

11

Theoretical calculations

• Very similar trend between data and theoretical calculations

• To compare the magnitude, the acceptance effects need investigation.

Y. Burnier, D. E. Kharzeev, J. Liao and H-U Yee,

private communication.

12

UrQMD

The slopes from UrQMD are consistent with 0 using the same approach.

Hongwei Ke

13

Outlook

• Further studies of the weak decay contributions• DCA (Distance of Closest Approach) cut on particles• simulation such as UrQMD and AMPT

• Beam energy scan (39 GeV, 27 GeV and 19.6 GeV)• different charge asymmetry distribution• different magnetic field?

• Kaon v2

• opposite v2 ordering than pion (v2(K+) > v2(K-))• what about A± dependency?

• Handle on the magnetic field• trigger on spectators with Zero Degree Calorimeters• is the signal still there when we turn off B?

14

CMW + Local Parity Violation

RPad

dN

sin21

A direct measurement of the P-odd quantity “a” should yield zero.

S. Voloshin, PRC 70 (2004) 057901

Directed flow: vanishesif measured in a symmetric rapidity range

Non-flow/non-parity effects:largely cancel out P-even quantity:

still sensitive to charge separation

15

Results with different EPsSTAR Preliminary

The correlators using TPC/ZDC event planes are consistent with each other.

Lost in the medium?

16

Dilution effect

STAR Preliminary

The factor Npart is used to compensate for dilution effect.

Weaker B field

Non-zero Radial flow?

Momentum conservation?

outin BB

17

(OS - SS)*Npart

If Nspec is a measure for B, the signal (OS-SS)*Npart is roughly proportional to the magnetic field in central/mid-central collisions.Experimentally, we may trigger on Nspec to control B.

18

Possible physics background

-+ ΨRP+ -

charge conservation/cluster + v2 Physics background for LPV:

Qualitatively similar trend! Need to trigger on very small Nspec.

Phys. Rev. C72 (2005) 014904

19

Beam energy scan

As we lower the beam energy, changes start to show from the peripheral collisions.27 GeV and 19.6 GeV coming soon...

20

If we consider OS-SS to be signal...

The signal seems to be disappearing at 7.7 GeV, but the statistical errors are large.

21

• Further studies of identified particles• Kπ correlation• pπ, Λπ...• ΛΛ correlation: vorticity

• Beam energy scan (27 GeV and 19.6 GeV)• Is there a smooth transition from 200 GeV to 7.7 GeV? • different magnetic field?

• Handle on the magnetic field• trigger on spectators with ZDCs• is there still signal when we turn off B?

• U+U collisions• different v2 trend from Au+Au?

Outlook

See also Kent Riley's talk!

Phys.Rev.Lett. 105 (2010) 172301

22

Backup slides

23

Multi-component Coalescence (MCC) + Quark Transport

John Campbell & Mike Lisa, preliminary study, publication in preparation.

24

Dilution effect

What do we know about the position Rn after n steps? Rn follows a Gaussian distribution: mean = 0, and rms =

Our measurement of PV is like Rn2, expected to be n.

Compared with going in one fixed direction, where Rn2 = n2,

the "random-walk" measurement is diluted by a factor ~ n ~ Nch.

In the quark-gluon medium, there could be multiple P-odd domains.

The net effect is like a random walk, but one-dimensional.

n

25

Possible physics backgroundcharge conservation/cluster + v2

Qualitatively similar trend! Need to trigger on very small Nspec.

Phys. Rev. C72 (2005) 014904

Phys.Rev.C83:014913,2011

26Balance function

Phys.Rev.C83:014913,2011