1 / 23

Particle Correlation Measurements from

Particle Correlation Measurements from. Stephen C. Johnson – Lawrence Livermore National Lab for the PHENIX Collaboration. HBT results presented here: submitted to PRL: 16 Jan `02 nucl-ex/0201008. R. 2. r. 1. q. The Physics of Hanbury-Brown Twiss. r 1 , p 1. S(r).

wachtel
Télécharger la présentation

Particle Correlation Measurements from

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. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Particle Correlation Measurements from Stephen C. Johnson – Lawrence Livermore National Lab for the PHENIX Collaboration HBT results presented here: submitted to PRL: 16 Jan `02 nucl-ex/0201008

  2. R 2 r 1 q The Physics of Hanbury-Brown Twiss r1, p1 S(r) • Consider a source S(r) of identical bosons (g or p) whose wave functions can be described as plane waves. • Assume: • Production amplitudes independent of momentum • Mutually incoherent x r2, p2 y Amplitude for this diagram: Corresponding normalized probability:

  3. S(x,p) Some details: Pairs from same event In principle: In practice: • One dimension works fine when measuring stars: • Static • Isotropic emission (no position momentum correlations) • Heavy Ion Collisions are anything but • Consider a more complicated source • S() – effective single particle Wigner phase space density • Often replaced by a classical phase-space density in practical calculations • Note: due to mass shell constraints, the function is non-invertible (model assumptions). Pairs from ‘mixed’ events

  4. k q qlong qout Kinematic Variables • Currently the field tends to consider projections of the momentum difference into: • qlong, qside, qout • qlong Rlong ~longitudinal extent • qside/qout Rside/Rout ~transverse extent qside qlong qout

  5. Rside Rout HBT and the QGP • “Naïve” picture: • Rout2=Rside2+(bpairt)2 • Concrete predictions are few: • Pratt PRD 1314 (`86): fireball and EOS  t ~ 90 fm/c • Bertsch NPA 173 (89) QGP + cascade  t ~ 12 fm/c • Hydro calculation of Rischke & Gyulassy expects Rout/Rside ~ 2->4 @ kt = 350 MeV. • Result robust to Tfreeze, dQ/dH, 1st order vs. rapid cross-over. • Response: can hadronic rescattering mask this prediction?

  6. One step closer • Soff, Bass, Dumitru (PRL 86) • Couple microscopic transport to hydro with phase transition • Still expect Rout/Rside>1  measurements at high kt are very interesting. • Note: • Hydro: Ro/Rs(200)<Ro/Rs(160) • Longer time at phase transition • Transport: Ro/Rs(200)>Ro/Rs(160) • Longer time rescattering

  7. PHENIX – Year 1 Configuration • Both arms provide hadron PID (contrary to popular belief) • East: • DC + TOF (~100 ps) • p/K separation to 2 GeV/c • West: • DC + PbSc (~600 ps) • p/K separation to 1 GeV/c • B-field + geometry limits lower kt bound to 200 MeV. I use ‘electron’ acceptance for this hadron analysis

  8. Centrality definition and sample • This data sample uses the 30% most central collisions • <centpairs> = 10% • 493k events • After all analysis cuts: • 3.1M p+ pairs • 3.3M p- pairs BBC ZDC

  9. Pair acceptance and corrections • p definition  <1.5s from p peak && >2.5s from K peak • Require pairs from mixed events to have reconstructed vertices within 1cm • Acceptance varies as a function of vertex position • Remove pairs within 2cm in drift chamber • Ghosting • Remove tracks with EMC clusters within 12cm of each other in both real and mixed sample • Shower + tower size in EMC • Correct for two track inefficiencies at low relative f in the drift chamber. • Full Coulomb Correction modified for momentum smearing • Partial correction changes radii results marginally • Residual correlations in event mixed background  <2% error • Momentum smearing correction to correlation function Systematic Errors: Rlong & Rside: 8% Rout: 4.5%

  10. Results: Assume T = 125 MeV, bf = .69/hf=.85 From fits to singles spectra in centrality region 5-15% (J. Burward-Hoy) • Theoretical hydro-inspired fits. • Chapman, Nix, Heinz PRC 52, 2694 • Wiedemann, Scotto, Heinz PRC 53, 918 • However, hydro calculations predict Ro than data, Rs, smaller. • Much larger than comparable 1D RMS Au radius of 3.07fm • kt dependence suggest larger bf/T (hf/T) than fits to singles Systematic Errors: Rlong & Rside: 8% Rout: 4.5%

  11. Compared to STAR • Well described by hydro model if consider datasets separately. • But they indicate a much higher flow/temp ratio when taken together • Need to be careful about systematics between different measures. • Both experiments should be able to sort this out in the next dataset. • Inconsistent with models of QGP that include an hadronic rescattering phase. Soff, Bass, Dumitru (PRL 86)

  12. Energy dependence … M. Lisa et al., PRL 84, 2798 (2000) R. Soltz et al., to be sub PRC C. Adler et al., PRL 87, 082301 I.G. Bearden et al., EJP C18, 317 (2000) • Rout and Rside are energy independent within error bars. • Smooth energy dependence in Rlong • No immediate indication of very different physics • Fit Rlong to: • AGS: A = 2.19 +/- .05 • SPS: A = 2.90 +/- .10 • RHIC: A = 3.32 +/- .03 A = t0T in 1st order T/mT calculation t0 = average freeze-out proper time

  13. Even an experimentalist can’t predict this • “A prognostication” • Local PHENIX HBT expert keeps his predictions on the web. • Zajc at Nucleus-Nucleus 97 showed this slide • Extrapolation to RHIC multiplicities? • (dN/dy)1/3 => R ~ 9 fm • (dN/dy) => R ~20 fm • Neither. • What is happening here? • ** ** To be fair: Bill’s radii predictions are way off. His guess at the charged particle multiplicity in the extrapolations…bang on

  14. R.A. Soltz E859 Si+Al/Si+Au/Au+Au Q & A • Q: Why are Rout and Rside ~identical over an order of magnitude of beam energy? • There is ample evidence at AGS and SPS of dependence of radii on # target+projectile nucleons • Even though the larger nuclei have larger flow the radii follow a simple scaling • Why no multiplicity/energy dependence? • Q: Flow higher at RHIC leads to smaller radii? • HBT results suggest high flow but spectra imply flow comparable to SPS ** • kt dependence of radii is similar for different energies (competing mechanisms that create similar bf/T?) • Q: Higher opacity at RHIC energies? • Why would opacity be higher at RHIC than at AGS? • pn->D->pn  pp->r->pp • And why would size and opacity effects identically cancel? • Red herring? – You can’t create opaque source with smaller lifetime. • Q: Why is it only changing in the longitudinal direction? • Some surprise among theorists that the difference is so small. • Is there a quantitative expectation of the Rlong dependence? ** To my knowledge no one has fit both SPS and RHIC results to same hydro model

  15. A taste of the future • For year-1 we wrote ~5 million min. bias events. • 1/2 million events in this analysis after all cuts • In the past run we wrote ~90 million min. bias events (+14 million rare event triggers) • Therefore, for the year-2 pion correlations: • easily 10 high statistics bins in kt from .2 to 1.0 GeV + a few bins from 1.0 to 2.0 GeV • Capability to exclude detailed theories • Systematic errors start to become the real problem • Beyond pions: • 1D proton and kaon correlations • Non-identical correlations (pK, pp, etc.) • p0 • maybe possible in year 2 … year 3? • Probe of very high kt. • anti-neutron correlations • Better than proton correlations  lack of coulomb. Ro,s,l kt

  16. The scientists that make me look good* *or at least better than I normally do

  17. Supporting slides:

  18. PHENIX Preliminary PHENIX Preliminary PHENIX Preliminary 118-126 MeV 0.71-0.73 PHENIX Preliminary Fitting the Single Particle Spectra:Jane Burward-Hoy (LLNL) 1/mt dN/dmt = A  f()  d mT K1( mT /Tfo cosh  ) I0( pT /Tfo sinh  ) • Linear velocity profile • Gaussian spatial distribution • Exclude pion resonance region • Exclude mt-m0>1.0 GeV

  19. Switching frames • Problem: radii extracted depend on the frame in which you measure • Historically we have used: • Qinv (1D pair rest frame) • Collision Center of Mass • AGS • LCMS (longitudinally co-moving frame) • Longitudinally boost invariant sources • SPS(?), RHIC • Jet frame • p-p collisions. • PCMS (pair center-of-mass – 3D version of Qinv) • … • What is the ‘correct’ frame?? (= the frame in which the source is not moving) • Can we bound it experimentally?

  20. Our source?? 2p HBT measurement is simultaneous in the PCMS frame p1 bpair p2 bsource Source Detector Pair rest frame != source rest frame What is the bsource? We don’t know, but we can bound it: >0 <bpair

  21. ‘side’ ‘out’ p1 p’1 bpair p2 p’2 unphysical LCMS vs. PCMS • LCMS • A Lorentz contracted measurement of the frame • PCMS • A measurement of the length of the train in the train frame True frame unphysical

  22. Fits to entire dataset (p-) LCMS PCMS* *PCMS = Pair Center of Mass System

  23. PCMS results • Rside and Rlong unchanged • And not plotted • RoutPCMS differs by <gpair> from RoutLCMS • Be careful about deducing a lifetime from Rout2=Rside2+(bpairt)2 • bpair=0

More Related