qcd in the heat bath: results from phenix thomas k hemmick for the phenix collaboration university...
DESCRIPTION
PHENIXThomas K Hemmick The “Calibrated” Plasma Probe Hard scattering processes and products: Occur at short time scales. Are “calculable” (even by experimentalists) in simple models (e.g. Pythia) with appropriate fudging: Intrinsic k T K scaling factor. Find themselves enveloped by the medium Are “visible” at high p T despite the medium Promise to be our laser shining (or not) through the dense medium created at RHIC.TRANSCRIPT
QCD in the Heat Bath:Results from PHENIX
Thomas K Hemmickfor the PHENIX Collaboration
University at Stony Brook
PHENIX Thomas K Hemmick
Overheard at APS ~1992:During a plenary RHI talk at APS, I wound up seated among “real” plasma physicists who made numerous comments about us:
“These guys are stupid…”Always a possibility.
“…why don’t they just shoot a laser through it and then they’d know if its plasma for sure!”
Visible light laser…bad idea.Calibrated probe through QGP…good idea……but not new. (Wang, Gyulassy, …)
PHENIX Thomas K Hemmick
The “Calibrated” Plasma ProbeHard scattering processes and products:
Occur at short time scales.Are “calculable” (even by experimentalists) in simple models (e.g. Pythia) with appropriate fudging:
Intrinsic kTK scaling factor.
Find themselves enveloped by the mediumAre “visible” at high pT despite the mediumPromise to be our laser shining (or not) through the dense medium created at RHIC.
PHENIX Thomas K Hemmick
QCD and the Heat BathI will make the following artificial divisions:
QCDCharged Hadron + Neutral pion spectra at high pT.Heavy quark production measured via electrons at high pT.J/, direct, +- coming in the future…
Heat BathMultiplicity, ET, …Identified hadrons
Singles, pairsFlow, Hadro-chemistry
Caveat:Short mean free path means that the view of the heat bath is heavily influenced by the latest stage prior to breakup.Nonetheless, initial state information (azimuthal anisotropy) is apparent.
Controlling parameterNature’s variation of the “centrality” controls the heat bath.Size/Energy variation is better, but not yet available.
PHENIX Thomas K Hemmick
Pioneering High Energy Nuclear Interaction eXperiment
11 Countries51 Institutions
400+ Collaborators
PHENIX Thomas K Hemmick
PHENIX from Above
PHENIX Thomas K Hemmick
West Armtracking:
DC,PC1, PC2, PC3electron ID:
RICH, EMCal
PhotonsEMCal
East Armtracking:
DC, PC1, TEC, PC3hadron& electron ID:
RICH,TEC, TOF, EMCphotons:
EMCal
PHENIX Setup During Year 2001 Run
full heavy ion program and first spin physics including electron & muon pairs
South Armtracking:
MuTrmuon ID:
MuID
Other DetectorsVertex
& centrality:ZDC, BBC,MVD
accumulated ~200 106 Au-Au
•The main 2001 data production pass has begun.
•Physics results still from 2000 data set.
PHENIX Thomas K Hemmick
Centrality DeterminationZDC measures neutral energy (principally free spectator neutrons) along collision axis.BBC measures produced charged particles within +/-(3.0-3.9).Their joint response is divided into centrality slices :
Each colored band is 5%.Central is at lower right.
The PHENIX trigger accepts 92% of minimum bias events.A Glauber calculation was used to deduce the numbers of participants (Npart) and binary collisions (Nbinary) for each centrality.
0-5%5-10%
cross sectionfraction
Npart Nbinary
0-5 348 10 1009 915-15 271 9 712 6515-30 180 7 406 4230-60 79 5 131 2260-92 14 3 14 5
PHENIX Thomas K Hemmick
Transverse Energy Production
Transverse Energy production yields Bjorken Estimate for Energy Density as:
For =1.0 fm/c, this yields ~4.6 GeV/fm3! well above estimated critical value c~0.6-1.2 GeV/fm3
central 2%
PHENIX
ddE
cRT
Bj11
2
~ 1.5X higher than at CERN
PHENIX Thomas K Hemmick
Multiplicity rise with Npart is well reproduced by “Saturation” model.This model produces particles via parton-parton scattering.Production is affected by the initial-state gluon distribution which “saturates” as low x gluons fill available space.
dNch
/d/
(0.5
Np)
Kharzeev & Levin, nucl-th/0108006Schaffner-Bielich et al, nucl-th/0108048
Multiplicity and ET vs. Npart
Multiplicity and ET both rise faster than linearly with the number of participants.This has been parameterized as:
The “Binary Scaling” contribution varies , ~25% in peripheral collisions~50% in central collisions<ET>/<Nch> ~ constant with centrality
12.034.0;28.088.0
BA
NBNAd
dNbinarypart
ch
soft hard
PHENIX Thomas K Hemmick
Multiplicity and ET vs. s
Nch and ET evolution very well described by ln(s)Saturation model also quite reasonable here (and especially with the PHOBOS result…).
with quenching
no jet quenchingPHENIX
saturation
PHENIX Thomas K Hemmick
Composition of the Heat Bath: PID
Charged hadron ID is achieved via high-res TOF measurement.
True photon pairs show a peak that is absent from mixed sample.Subtracted spectrum can be fitted for 0 yield.
K+
p
d
d
Au+Au @ 200 GeV
PHENIX Thomas K Hemmick
Yield vs. Centrality
Yield per participant for all species rises with Npart.Not surprising since the total Nch rises faster than Npart.Integral of , K, p yields in excellent agreement with dNch/d result.
x1.2
x1.7
Surprising to me:• Yields of K+, K-, p, p-
bar rise faster with Npart than pion yields.
• The “faster than Npart” character of Nch production also involves a change in relative species abundances favoring K and nucleon production!
PHENIX Thomas K Hemmick
Net and Total Baryon DensityTypically we have measure and quote “net baryon” density (b—bbar).Some physical processes will more naturally depend upon the “total baryon” density:
e.g. medium modifications to mesons are predicted to be same for baryonic and anti-baryonic dense matter (Rapp).
Here we calculate both densities:
10288Total baryon density
20.180.4
520
8.621.4
2768
p – p Participating nucleons (p – p ) A/Z dN( p ) / dy Produced baryons (p, p, n, n )
RHIC(Au-Au)
SPS(Pb-Pb) Caveat:
We don’t measure neutrons!Assume: --(n+p)part=A/Z--nprod = pprod
PHENIX Thomas K Hemmick
<pT>: Energy Distribution in Bath
<pT> variation with centrality roughly the same for all species:
~20% rise. and K close to pp production for peripheral collisions.Proton differs from pp significantly at most peripheral bin.
PHENIX Thomas K Hemmick
Hadron SpectraHadron spectra stiffen with increasing hadron mass.Well known consequence of collective motion in the final state.Spectra well fitted by “flowing thermal Ansatz:
emperatureFreezeOutTTocitySurfaceVel
Rr
dT
pIT
mKfmAdmm
dN
FO
s
ns
FO
T
FO
TT
TT
)(tanh;
)sinh()cosh()(
1
01
PHENIX Thomas K Hemmick
Thermal Fit Error ContoursFor each TFO and , a best fit and 2 are found.2
min/dof = 34/40.Contours show 1, 2, 3…Similar T to CERN, higher flow velocity.
118-126 MeV
0.71-0.73
PHENIX Preliminary
Although the result of such a technique is only as good as the assumed functional form, the differences between experiments can be quite instructive.
PHENIX Thomas K Hemmick
HBTHBT correlations also reflect underlying collective flow velocities via the kT dependence of the radius parameter.World systematics of HBT nearly universal trend of Rout, Rside???Rout~Rside???Only Rlong variation with s???A real dilemma.
PHENIX Thomas K Hemmick
What’s the composition of h+
+h-?PID for charged hadrons run out prior to the pT required for hard processes.Plots w/o renormalization show the species content of hadrons leading into our high pT probe particles.
The stiff slopes (and increased yield) of nucleons cause the high pT spectrum to be more than 50% nucleons!
PHENIX Thomas K Hemmick
Where does hard scattering begin?
Initial state Hard scattering should produce a p/ ratio which falls with pT.mT scaling (flow) will producing a rising p/.ISR: Onset of hard processes noted above 2 GeV/c.RHIC: No clear onset of hard scattering is found in the data up to pT = 3 GeV/c !!
pT
mT scaling
pQCDp/
rat
io
soft/hard transition
measured
PHENIX Thomas K Hemmick
Charged Hadron Spectra to High pT
Charged hadron spectra in bins of centrality show a gradual loss in concavity as they evolve from peripheral to central collisions.The loss in concavity is opposite to the effect one might expect comes from the known stiff proton contribution.The protons have a significantly higher slope than pions and become the majority contributor in the range 2-3 GeV/c.
To quantify the shape evolution we can use either pp data or peripheral collisions.
PHENIX Thomas K Hemmick
Unfortunately, no pp data at s=130 GeV exist.We choose to interpolate using ISR, UA1, and CDF data.The data establish the parameters of the “Power Law Fit” as a function of s.
These are smoothly interpolated to produce the solid curve in the figure.
Interpolation of pp Data
nTTT
ppA
dpdd
p
0
2
121
PHENIX Thomas K Hemmick
Spectral comparisonGood fit to peri-collisions using interpolated power law.Power Law overpredicts the central collisions
0
The discrepancy between the neutral pions and the interpolated power law is more severe than for the charged hadrons.
PHENIX Thomas K Hemmick
Difference between h and 0
We subtract the charged nucleon contribution from the all-charged hadron distribution.The result is in reasonable agreement with the neutral pion spectrum demonstrating that the charged part. contamination is principally nucleons.
PHENIX Thomas K Hemmick
RaaWe define the nuclear modification factor as:
By definition, processes that scale with Nbinary will produce RAA=1.
ddpdNddp
NdNpR
T
NN
NNinel
binary
T
AA
evtTAA
2
21
)(
RAA is below 1 for both charged hadrons and neutral pions.
The neutral pions fall below the charged hadrons since they do not suffer proton contamination
PHENIX Thomas K Hemmick
RAA--PeripheralWe can also reference RAA using measured peripheral collisions to avoid the systematic error in the pp interpolation.
Again, the central data drops below 1.0 and shown no Cronin!
PHENIX Thomas K Hemmick
RAA—0 at RHIC and CERN
WA98 measurements of RAA for 0 show a dramatically different behavior than at RHIC.
PHENIX Thomas K Hemmick
RAA influences from “ordinary” effects.
x~2pT/s~0.02
X.N.Wang
soft/hardtransition?
CERN - SPS
Cronin effect: Multiple scatters of the projectile broaden the incoming pT
RAA exceeds 1.0 at high pT e.g. at CERN.
Shadowing?…No.
PHENIX Thomas K Hemmick
HIJING prediction for 0
One explanation for the RAA spectrum comes from multiple scattering from a colored medium.The HIJING model using 0.25 GeV/fm energy loss well reproduces the measurement.
PHENIX Thomas K Hemmick
Electron IdentificationThe Electromagnetic Calorimeter and RICH were used to select single electrons.Neither device has sufficient identification working alone.The combination shows a clear peak of signal electrons that both fire the RICH and deposit energy equal to their momentum in the EMcal.
0.8 GeV/c < p< 0.9 GeV/c
All charged
With RICH ring
Random background
E/p ratio
PHENIX Thomas K Hemmick
Electron SinglesIdentified electron spectra are shown on the left.Low pT single electrons dominantly come from 0 Dalitz decay:0->e+e-
and other neutral meson decays.Our measurements of the 0, +, - spectra determine this background level.Other backgrounds can also be modeled and subtracted as an “electron cocktail”.
PHENIX Thomas K Hemmick
Tuning Cocktail SpectraThe pion spectra are fit to an exponential + power law.Dalitz decay is an internal photon conversion.External photons conversions are normalized using a GEANT simulation and modeled as an upscaling to the Dalitz.Heavier meson spectra are assumed to follow mT scaling.Heavier meson yields are normalized by the pp measured ratio at high Pt (consistent w/ thermal).
mT scaling OK
FitData/Fit
PHENIX Thomas K Hemmick
Decomposition of Electrons
Green band indicates error (statistical+systematic) in background.Data Shows excess over Cocktail at high pT.
PHENIX Thomas K Hemmick
Pythia (after tuning…)
Akiba will show detail about the tuning of Pythia to describe open charm production in pp.The addition of open charm decays ala’ Pythia well describes the electron excess.
PHENIX Thomas K Hemmick
Fit the excess as charmThe subtracted spectrum can be used to normalize the charm yield and thereby the cross section.The result is comparable within large error bars to the Pythia calculation.
0 10%cc 380 60 200 b and
PHENIX Thomas K Hemmick
Summary-1The RHIC medium is certainly intriguing and exciting:
Particle production scales faster than Npart possibly signaling a hard Nbinary component.Kaon and Nucleon contributions grow in percentage with centrality.Low net baryon density but high total baryon density.Strong radial flow implied by relative slope constants of hadrons.Nucleons begin to be dominant contributors to pT spectra above 2 GeV/c.HBT Rout~Rside???
Probes of the medium yield exciting resultsRAA below 1. Especially dramatic for identified 0 which have no proton background.Charm = Pythia within huge errors. Why?
PHENIX Thomas K Hemmick
Summary-2PHENIX has produced a wealth of data and has much more in analysis.The RHIC era opens the hard scattering or QCD toolbox (Pandora box?) in RHI physics.
My wife usually criticises my use of new tools:LMH: “Don’t you ever read the manual?”TKH: “Honey, these tools don’t come with manuals!”
Maybe this week we’ll begin writing the manual for QCD in the RHIC era.
PHENIX Thomas K Hemmick
END OF ALL SLIDES
SLIDES BEYOND THIS POINT WILL NOT BE SHOWN
PHENIX Thomas K Hemmick
Singles SpectraMost Central Mid-Central Peripheral
PHENIX Thomas K Hemmick
Singles==all
PHENIX Thomas K Hemmick
Singles mt
PHENIX Thomas K Hemmick
Yields
PHENIX Thomas K Hemmick
<pt>
PHENIX Thomas K Hemmick
Ratios
PHENIX Thomas K Hemmick
Ratios2
PHENIX Thomas K Hemmick
ID vs dN/dn
PHENIX Thomas K Hemmick
Acceptance
PHENIX Thomas K Hemmick
HBT Results
PHENIX Thomas K Hemmick
HBT spectra
PHENIX Thomas K Hemmick
Charged Particle PID
PHENIX Thomas K Hemmick
Pi0 PID
PHENIX Thomas K Hemmick
Flow
PHENIX Thomas K Hemmick
Hard Spectra
PHENIX Thomas K Hemmick
Hard Spectra 2
PHENIX Thomas K Hemmick
Hard pi0 spectrum
PHENIX Thomas K Hemmick
SQRT(s)with quenching
no jet quenchingPHENIX
saturation
PHOBOS
RHIC
with quenching
PHENIX Thomas K Hemmick
SQRT(s)-2
PHENIX Thomas K Hemmick
Raa
PHENIX Thomas K Hemmick
Raa2
PHENIX Thomas K Hemmick
Raa3
PHENIX Thomas K Hemmick
Hard pp
PHENIX Thomas K Hemmick
Suppression—Pt
PHENIX Thomas K Hemmick
dE/dx-theory
PHENIX Thomas K Hemmick
LambdaPHENIX Preliminary
½ x
PHENIX Thomas K Hemmick
Electron PID
PHENIX RUN 12280 SEQ 0014 EVENT 850
View from North Side
South Side
East Arm West Arm
6 PMT RICH ring2.55 GeV/c track2.5 GeV EMCal hitelectron candidate
EMCal
RICHPC1
DC
EMCal
RICHPC1
DC
TOF
TECPC3
PHENIX Thomas K Hemmick
Cocktail
PHENIX Thomas K Hemmick
Electron Spectra
PHENIX Thomas K Hemmick
Parameters
PHENIX Thomas K Hemmick
Pythia
PHENIX Thomas K Hemmick
Flow—v2
PHENIX Thomas K Hemmick
Thermal
t()
(No QGP phase transition)
2 fm/c20 fm/c
5 fm/c
PHENIX Thomas K Hemmick
Ralf
PHENIX Thomas K Hemmick
Ralf2
PHENIX Thomas K Hemmick
Ralf3