1. Charm measurements at SPS to RHIC Y. Akiba (KEK) March 14, 2003 Strangeness in Quark Matter 2003

2. Outline Physics –Physics with open charm pp, pA, AA –Physics with charmonium pp, pA, AA Open charm measurements at fixed target –Nuclear dependence: cross section,, A dependence –Enhancement in Pb+Pb (?) (NA50 IMR) Charmonium measurements at SPS and FNAL –A dependence: normal nuclear absorption, –Anomalous suppression in Pb+Pb (NA50) Open Charm measurements at RHIC –Charm measurement via single lepton –Run-1 (130 GeV) data (PHENIX) –Run-2 (200 GeV) prelimnary data (PHENIX) Charmonium measurement at RHIC –First J/  data at RHIC (PHENIX)

3. Physics with open charm production pp and hadron-hadron collisions –Comparison with pQCD calculation –Measurement of gluon density G(x) –Intrinstic –Base line for charm physics in pA and AA pA –Gluon shadowing –Energy loss of gluons in cold nuclear matter –Parton multiple scattering and –Base line (normal nuclear effect) for charm physics in AA AA –Gluon shadowing –Energy loss of charm in high density matter –Thermal production of charm in high temperature QGP

4. Physics with Charmonium production pp: study of production mechanism –Color evapolation model, Color singlet model, Color Octet model –Production cross section and its pt and s 1/2 dependence –Polarization –Production of J/ ,  (2S),  c, states –Base line for pA and AA pA: study of “normal nucleus effect” –Nuclear dependence of  (J/  ) ….A  or  abs (nuclear absorption) –Nuclear dependence of …multiple scattering effect? –Base line for AA AA : study of “medium effect” in high density matter –J/  suppression : signature of QGP (Matsui/Satz) –J/  formation by c quark coalescence at RHIC/LHC ?

5. Measurement of charm and lepton/lepton pairs J/  has a large branching ratio (6%) to lepton pairs, and it is almost exclusively measured by lepton pair decay. Charmed mesons has a large leptonic branching ratio (D 0 : 7 %, D + :17%). Charm production can be measured indirectly by single lepton in 0.5< pt<3 GeV/c (RHIC/PHENIX) and lepton pairs in 1<M<3 GeV (SPS/NA50). –More direct measurement of D- meson reconstruction is difficult without a precision vertex detector J/    e + e -,  +  -  S)  e + e -,  +  -  c  J/  +  Charmonium: lepton pairs Open charm: single lepton and lepton pairs Charm signal: single lepton lepton pairs

6. Open charm at fixed target (cross section) Charm cross section by fixed target experiments are reasonably reproduced by LO pQCD event generator (PYTHIA) with large K-factor, or by NLO pQCD calculation (HVQMNR). Note that pQCD may or may not be applicable to charm production because charm mass is small (~1.5GeV) In the left figure, world pi+N data and p+N data are compared with PYTHIA calculation. The s 1/2 dependence of the calculation mainly reflects the underlying PDF. s 1/2 (GeV)  cc in pN,  N

7. Open charm at fixed target ( ) FNAL E791 and CERN WA92 measured hight statistics Pt distribution of D-mesons in  +N collision. (No high statistics data in p+A is available) In the left figure, E791 data and WA92 data are compared with PYTHIA calculation with different values. ~ 1.5 GeV/c seems to be required to reproduces the data. Comparison with NLO pQCD calculation also suggests that observed pt distribution is harder than predicted, and large (>1GeV/c) intrinsic is needed to reproduce the data. No published data on nuclear dependence in but it can be similar to that of J/  p T 2 (GeV 2 /c 4 ) E791 d  /dp T 2 p T 2 (GeV 2 /c 4 ) WA92 d  /dp T 2

8. Nuclear dependence of Open charm cross section Open charm cross section has little or no nuclear dependence  = 1.00± 0.05 (E769 250GeV  +A)  = 0.92 ±0.06(WA82 340GeV  +A)  = 1.02 ±0.03 ±0.02(E789 800GeV p+A) This is consistent with that there is little nuclear shadowing at the x region probed by the fixed target charm experiments. Significant nuclear suppression is reported by large x f region (WA78,  =0.81 ± 0.05). This can be due to nuclear shadowing in small x, but this is not well understood. At RHIC, we can prove x region much smaller than in fixed target. We may observe nuclear shadowing effect in charm production. E769 250 GeV  ± PRL 70,722 (1993)  = 1.00 ± 0.05

9. Open Charm enhancement (?) at SPS Pb+Pb NA50 measured di-muon distribution in 1.5<M<2.5 GeV. The mass distribution is well reproduced by DD pairs + Drell-Yan components in p+A (@450GeV) with charm cross section  cc /A = 36.2±9.1  b, consistent with charm cross sections of other experiments In Pb+Pb, an excess of dimuon signal is observed. The excess can be explained by enhancement of charm production by factor of ~3 in central Pb+Pb. The other explanation is possible: e.g. thermal dimuon pair produciton. NA50

10. J/  at fixed target and ISR Many experimental data in the last 30 years near threshold to s 1/2 = 63 GeV Rapid increase of production cross section to s 1/2 ~ 20 GeV followed by a gradual increase in higher energy. The s 1/2 dependence is similar to that of charm production cross section, suggesting that J/  and charm ratio is almost constant with energy Detailed, high statistics study of A dependence by E772/789/866 at 38 GeV and p+A and A+A measurement by NA38/50

11. JPSI at fixed target: A dependence Significant nuclear dependence of cross section is observed Power law parameterization  = A   = 0.92(E772. PRL66(1991) 133) (limited pt acceptance bias?)  = 0.919 ± 0.015 (NA38. PLB444(1998)516)  = 0.954 ± 0.003(E866 @ x F =0. PRL84(2000),3258 )  = 0.934 ± 0.014(NA50, QM2001) Absorption model parameterization  = 6.2 mb (NA38/50/51) to 4.4 mb (NA50, QM2002) Small difference in  between J/  and  (2S) (E866)  (J/  ) –  (  S  ) ~ 0.02-0.03 @ x F = 0

12. JPSI at fixed target: of J/PSI increases with beam energy – increase with energy ? –Gluon radiation effect? (  pQCD ) of J/  increases with L (path length in target nuclear matter) or A 1/3 –Consistent with multiple scattering of incoming partion (  Cronin effect) vs L vs A vs s 1/2

13. J/  suppression at SPS NA50 observed anomalous suppression of J/Y in Pb+Pb collisions at 158 GeV Deviation from the nuclear absorption model (with  ~4 to 6 mb) is seen in E T > 40 GeV or  >2.5 GeV/fm 3 If the suppression is due to J/  break-up in high energy density matter, this is a strong evidence for QGP formation If J/  suppression is due to QGP formation, almost all of initially produced J/  should be suppressed at RHIC energy NA50

14. Charm and single electron at RHIC At RHIC, it is expected that charm decay can be the dominant component of single electron in pt > 1.5 GeV/c –Large production cross section of charm ( 300-600 ub) –Production of the high pt pions is strongly suppressed relative to binary scaling –Production of charm quark roughly scale with binary collisions. PHENIX observed “ excess ” in single electron yield over expectation from light meson decays and photon conversions  Observation of charm signal at RHIC Simulation before RHICPHENIX data (PRL88)

15. PHENIX single electron data PHENIX observed excess of single electron yield over the contribution from light meson decays and photon conversoins Spectra of single electron signal is compared with the calculated charm contribution. Charm contribution calculated as EdN e /dp 3 = T AA Ed  /dp 3 –T AA : nuclear overlap integral –Ed  /dp 3 : electron spectrum from charm decay calculated using PYTHIA The agreement is reasonably good. Assuming that all single electron signal is from charm decay and the binary scaling, charm cross section at 130 GeV is obtained as PHENIX PRL88 192303

16. Comparison with other experiments PHENIX single electron cross section is compared with the ISR data single electron data Charm cross section derived from the electron data is compared with fixed target charm data Single electron cross sections and charm cross sections are compared with –Solid curves: PYTHIA –Shaded band: NLO QCD Assuming binary scaling, PHENIX data are consistent with  s systematics o (within large uncertainties)!

17. Single electron measurement by PHENIX in RUN2 photon converter method Single electron spectra : data with the converter data w/o the converter If all electrons are from photonic source (  0 Dalitz,  conversion, etc), the ratio is constant. But the data shows that electron yield approach at high p T each other. It is an evidence for non-photonic electrons (i.e. charm/bottom decay) NeNe 00 1.1% 1.7% Dalitz : 0.8% X 0 equivalent 0 With converter Conversion in converter W/O converter Conversion from pipe and MVD 0.8% Non-photonic

18. RUN2 single electron result The yield of non-photonic electron at 200 GeV is higher than 130 GeV The increase is consistent with PYTHIA charm calculation (  cc (130GeV)=330  b   cc (200GeV)=650  b) Large systematic uncertainty due to material thickness without converter. The error will be reduced in the final result.

19. Centrality Dependence PHENIX data is consistent with the PYTHIA charm spectrum scaled by number of binary collisions in all centrality bins!

20. Observation from PHENIX single electron data NA50 has inferred a factor of ~3 charm enhancement at SPS. PHENIX does not see such a large effect at RHIC. PHENIX observes a factor of ~3-5 suppression in high p T  0 relative to binary scaling. PHENIX does not see such a large suppression in the single electrons. Initial state high pt suppression excluded? (  CGC, strong shadowing) smaller energy loss for heavy quark ? (dead cone effect) NA50 - Eur. Phys. Jour. C14, 443 (2000). N part Enhancement of Open Charm Yield Binary Scaling PHENIX Preliminary

21. J/  at RHIC J/  production in N+N –J/  production mechanism –Spin dependence (gluon polarization  G(x)) J/  production in Au+Au –Suppression by QGP formation? –Enhancement by recombination of charm and anti-charm in QGP? PHENIX can measure J/   ee in |y|<0.35 and J/PSI   in 1.2<|y|<2.4 –Wide kinematics coverage –Independent measurement in two channels PHENIX can measure of both open charm and J/  –VERY strong constraint on the models of J/  production

22. J/   ee and J/    in pp @ 200 GeV |y| <0.35 N J/  = 24 + 6 + 4 (sys) Bd  /dy = 52 + 13 (stat) + 18 (sys) nb 1.2 < y < 1.7 N J/  = 26 + 6 + 2.6 (sys) B d  /dy = 49 + 22% + 29% (sys) nb 1.7 < y < 2.2 N J/  = 10 + 4 + 1.0 (sys) B d  /dy = 23 + 37% + 29% (sys) nb e+e- |y|<0.35    - 1.2 < y < 2.2

23. 1.2<y<2.2 shape of p T distribution is consistent with a PYTHIA calculation average p T – y=1.7 = 1.66 ± 0.18 (stat.) ± 0.09 (syst.) GeV/c –slightly larger than measured at lower energies –consistent with a PYTHIA extrapolation to RHIC energy pp  J/  at RHIC ( )

24. µ+µ-µ+µ- e+e-e+e- Br (J/  l + l - )  (p+p  J/  X) = 226  36 (stat.)  79 (sys.) nb  (p+p  J/  X) = 3.8  0.6 (stat.)  1.3 (sys.) µ b PHENIX data agrees with the color evaporation model prediction at  s=200 GeV pp  J/  at RHIC

25. J   e + e - in Au+Au collisions at  s NN = 200 GeV e + e - invariant mass analysis very limited statistics N=10.8  3.2 (stat)  3.8 (sys) split minimum bias sample into 3 centrality classes

26. Centrality dependence of J/  yield J/  B-dN/dy per binary collision compared to different models for J/  absorption patterns –J/  scale with the number of binary collisions –J/  follow normal nuclear absorption with given absorption cross sections –J/  follows same absorption pattern as observed by NA50 (Phys. Lett. B521(2002)195) =791 p+p =297 =45 Attention: all curves are normalized to the p+p data point! present accuracy NO discrimination power between different scenarios Much more statistics is needed!

27. Coming soon --- J/  measurement in d-Au at RHIC PHENIX has seen J/  signal in d-Au collisions and expects to get a few thousands J/  at the end of the run. Observed signal online Crude calibration Nuclear shadowing effects in J/Y production soon will be studied by PHENIX. Comparison of J/  yield in central arm and two muon arm is useful to distinguish different shadowing models J.C.Peng PHENIX μ PHENIX e E866 (mid-rapidity) NA50

28. Summary and outlook SPS and fixed target –Open charm production Charm production scale with A in p+A collision (no nuclear effect) NA50 IMR data can be interpreted as a large enhancement of charm in Pb+Pb –J/  production Extensive data on normal nuclear effect in J/  production in p+A by E866 and NA38/50 Suppression of J/  production in Pb+Pb beyond normal nuclear absorption is observed by NA50 RHIC –Charm production measured by single electron is consistent with the binary scaling (with relatively large error). Little gluon shadowing at RHIC? Little energy loss effect in charm quark? –The first J/  data in pp and AuAu at RHIC is obtained Much more statistics is needed to measure J/  suppression at RHIC