aa r X i v : . [ nu c l - e x ] J a n Recent Heavy-Flavor results at STAR
Zhangbu Xu (for the STAR Collaboration)
Physics Department, Brookhaven National Laboratory, Upton, NY 11973, USAE-mail: [email protected]
Abstract.
We present the recent results on non-photonic electron (NPE) yields fromRHIC run8 p+p collisions. The e/π ratio as a function of p T in run8 with a factorof 10 reduction of the inner detector material at STAR is found to be consistent withthose results from run3 taking into account the NPE from charm leptonic decay andthe difference of photonic electron yield from photon conversion in detector material. J/ Ψ spectra in p + p and Cu+Cu collisions at √ s NN = 200 GeV with high sampledluminosity J/ Ψ spectrum at high- p T follows x T scaling, but the scaling is violated atlow p T . J/ψ -hadron correlations in p + p collisions are studied to understand the J/ Ψ production mechanism at high p T . We observed an absence of charged hadronsaccompanying J/ Ψ on the near-side, in contrast to the strong correlation peak in the di-hadron correlations. This constrains the B -meson contribution and jet fragmentationto inclusive J/ Ψ to be < ∼ p + p collisions scaled by the underlying binary nucleon-nucleon collisionsin the measured p T range. Other measurements and future projects related to heavy-flavors are discussed.PACS numbers: 12.38.Mh, 14.40.Gx, 25.75.Dw, 25.75.Nq ecent Heavy-Flavor results at STAR p T can serve as aprobe of the degree of a QGP thermalization analogy to classic Brownian Motion. Thedissociation of J/ Ψ and Υ due to color-screening in a Quark-Gluon Plasma (QGP)created in relativistic heavy-ion collisions [2] is a classic signature of de-confinementof the fundemental theory of Quantum Chromodynamics (QCD). Recently, techniquesbased on the AdS/CFT duality have been utilized to study the dissociation of quark-antiquark pairs with high velocities relative to the QGP. Calculations in this frameworkshow that bound states of heavy fermion pairs (an analog of quarkonium in QCD) havean effective dissociation temperature that decreases with p T (or velocity) as 1 / √ γ [3]. Totest this conjecture, measurements of J/ Ψ R AA to p T > J/ Ψ dissociation temperature is expected to be below the temperature reachedat RHIC collisions ( ∼ T c ). J/ Ψ in hadron-hadron collisions can be produced fromthe following processes: (i) gluon and heavy-quark fragmentation, (ii) decay feed-downfrom B mesons and Ξ c states, and (iii) direct production either through charm quarkand anti-quark pair in a color-octet or color-singlet state. Therefore it is important toidentify the J/ Ψ production before J/ Ψ can be used as a probe of the color dissociationin QGP.We report the recent results on non-photonic electron (NPE) yields from RHIC run8p+p collisions and the J/ Ψ spectra at high transverse momentum (5 < p T <
14 GeV/c)from EMC triggered events and at low transverse momentum from minbias events in p + p and Cu+Cu collisions at √ s NN = 200 GeV measured by the STAR experimentat RHIC/BNL. The e/π ratio as a function of p T in run8 with a factor of 10 reducedinner detector material at STAR is used to compare step-by-step with those results fromrun3 [6, 7, 8] to assess if large amount of the electrons from the photon-conversion indetector material in run3 has produced an artifically high NPE yield. The electron and π identification is provided by a combination of dE/dx in the STAR Time ProjectionCharmber (TPC) [9] and velocity from Time-of-Flight (TOF). This technique resultsin a small systematic error ( < ∼ e/π ratio since most of the detector acceptanceand efficiency cancels. The large acceptance of TPC and the Barrel ElectroMagneticCalorimeter (BEMC) [10] with | η | < J/ψ -hadron correlations in p + p collisions to understand the J/ Ψ productionmechanism at high p T .In run8, STAR has removed the inner silicon tracker (SVT and SSD). This reducesthe detector material close to the beam pipe by a factor of 10. STAR has also installedone sector (out of 24 in total) of new TPC electronics, which increases the TPC DataAcquisition rate by a factor of 10 and at the same time provides a buffered readoutscheme to reduce the deadtime [11] to few percent at 1KHz readout rate. The old TPCelectronics provides a maximum of 100Hz readout at 100% deadtime. Within the samesector, five TOF trays with final detector configuration have been installed as well. This ecent Heavy-Flavor results at STAR (GeV/c) T p0 0.5 1 1.5 2 2.5 3 3.5 π e / −4 −3 −2 −1 π Dalitz/ π STAR Simulation run8 p+p π STAR inclusive/ PRL96 π PHENIX inclusive/ PRL98 π STAR EMC NPE/ PRL94 π STAR TOF NPE/ PRL94 π / bin STAR d+Au NPE/N predicted π STAR inclusive/ in run8 p+p π e/ Figure 1. (Color online). Inclusive e/π ratio (red circles) as a function of p T from run8p+p collisions. The dotted dash lines are e/π ratio with electrons from π → γe + e − and the dotted line is the η Dalitz decay. The solid line is the NPE in d+Au minbiascollisions scaled by the binary collisions over the pion yield in p+p collisions as in other e/π ratios. The Dashed line is the inclusive e/π ratio from the sum of all the electronyields. special sector took data at 200Hz with a L0 trigger requiring at least a hit in the TOF.We refer the detailed analysis to Ref. [12].Figure. 1 shows the e/π ratio as a function of p T from run8. In the samefigure, the e π → γe + e − /π ratio from π → γe + e − (Dalitz decay) and a similar curvefor e/π ratio from the η Dalitz decay are presented. The pion spectrum was takenfrom an average of the π ± spectra measured by STAR in non-singly diffractive p+pcollisions. The spectrum was well described by a Levy function with fit functionas: dN/dy ( n − n − / (2 πnT ( nT + m ( n − / (1 + ( q p T + m − m ) /nT ) n wheredN/dy=1.38, n=9.7 and T=0.131. The red line is the ratio of NPE over the pionspectrum where NPE is the non-photonic electron yields obtained from a fit to thecombined results of NPE and D0 yields in d+Au minbias collisions scaled by its binarycollisions [6]. The inclusive electron yields consist of photonic electrons from π and η Dalitz decays, photon conversions at the detector material and non-photonic electronsfrom heavy-flavor semileptonic decays. Other sources ( φ → e + e − , direct photon) are atfew percent level and have very similar spectrum shape as those of photonic sourcesfrom Dalitz and photon conversions. To match the low- p T ( < ∼ . π Dalitz decay.We denote this detector dependent electron background as e run bg /π . This means that e run bg /π = 0 . × e π → γe + e − /π . Since the Branching Ratio of π Dalitz decay is 1.2%,the equivalent detector material for a photon conversion from π → γγ or η → γγ decays at the 90% of the e/π ratio from π Dalitz decay is 0 . × . / . . × / .
69% radiation length ( X ). The total sum ecent Heavy-Flavor results at STAR e/π ratios fromrun3 and an early PHENIX result (the only published inclusive electron yields) [13]are presented for comparison. To reproduce run3 data [6], we need a factor of 10more conversion background (10 × e run bg ), consistent with the different amounts ofmaterial existing in run3 and run8. PHENIX inclusive electron spectrum is similar toour current inclusive electron spectrum. This provides an opportunity to compare theinclusive electron yields, the background subtraction and NPE step-by-step between twoexperiments [13, 14, 8].Besides the NPE measurements, STAR excels in other heavy-flavor relatedmeasurements: minbias D measurements and D* in a jet without secondary vertex,e-h, e-D0 correlations. These results are presented in Ref. [15]. We have measured thefraction of B/ ( B + D ) from e-h correlation in p+p collisions and NPE R AA . This meansthat we can infer the B and D R AA in a model dependent analysis. Reference [15] showsour preliminary result of B R AA vs D R AA . It suggests that the bottom hadrons are assuppressed as charm hadrons. To directly reconstruct bottom and charm hadrons, thecurrent STAR upgrade plans include Time-of-Flight for particle identification, Heavy-Flavor Tracker for secondary vertex, DAQ1000 faster readout rate and future muontelescope detector.Both the TPC and the BEMC at STAR can provide electron identification [16, 17].BEMC has been used as a fast online trigger to enrich the data sample with high- p T electrons. The combination of shower energy deposit in BEMC towers and showershape from Shower-Maximum Detector (SMD) provides powerful hadron rejection.At moderate p T , the TPC can identify electrons efficiently with reasonable hadronrejection. This allows a study of J/ Ψ at high- p T . In this analysis, the high p T J/ Ψwas reconstructed through the dielectron decay channel with a decay branch ratio of5.9%. The electron at high p T was identified by combining the energy and shower shapemeausred in the BEMC and ionization energy loss ( dE/dx ) measured by the TPC; theother electron at lower p T was identified by the dE/dx only with better efficiency butlower purity. The data were from p + p and Cu+Cu runs in 2005 and p + p run in2006 at RHIC. An online BEMC trigger that requires only the transverse energy ( E T )deposit in one BEMC tower to be above certain threshold [18]. In addition, this triggerwas in coincidence with a minimum bias trigger requiring a coincidence between the twoZero Degree Calorimeters (ZDCs). The integrated luminosity is ∼ pb − for p + p collisions collected in year 2005 (2006) with E T > ∼ µb − for Cu+Cu collisions with E T > | η | < . p + p collisions have been presented in Ref. [18, 27] and found to follow the x T scalinglaw [28, 29, 30]: E d σdp = g ( x T ) √ s n , where x T = 2 p T / √ s . The value of the power n depends ecent Heavy-Flavor results at STAR s/ T =2p T x -3 -2 -1 ] [ nb / ( G e V / c ) / dp σ E d × B n / G e V ) s ( UA2 540STAR 200ISR 63 (n=6.6) ×π proton (n=6.6) STAR 200ISR 53FNAL 27.4 (n=5.6) ψ J/ CDF 1960CDF 1800UA1 630 ) T STAR 200 (low pSTAR 200 @2005STAR 200 @2006ISR 63ISR 53ISR 30
STAR Preliminary π p ψ J/ Figure 2. (Color online). (a): J/ Ψ invariant cross section times the dielectronbranching ratio as a function of p T in p + p collisions from year 2005 data (stars) andyear 2006 data (circles) and in Cu+Cu collisions (squares) at √ s NN = 200 GeV. TheCu+Cu results are scaled by 1/100 for clarity. (b): x T scaling of pions, protons and J/ψ s. The pion and proton results are from Ref. [19, 20, 21, 22]. The J/ Ψ results fromother measurements are from the following references, CDFII [23], CDF [24], UA1 [25],and ISR [26]. on the quantum exchanged in the hard scattering and is related to the number of point-like constituents taking an active role in parton model. It reaches 8 in the case of adiquark scattering and reaches 4 in more basic scattering processes (as in QED). Figure2 shows the x T scaling of J/ψ , pion and proton in p + p collisions. The J/ Ψ data[23, 24, 25, 31, 26] covers the √ s range from 30 GeV (ISR) to 1960 GeV (CDFII).The high p T J/ Ψ cross section at these various center-of-mass energies also follow the x T scaling law. These data are fitted simultaneously at the high p T region using thefunction (1 − x T ) m /p nT , The power n is found to be 5 . ± . J/ψ , which is lower thanthat for pion and proton (6 . ± . p T J/ Ψ productionmechanism is likely to originate from a 2 → p T J/ Ψ shows clear deviation from the x T scaling, very similar tothe behavior of the pion and proton yields at p T < p T J/ Ψ must originate from a hard process, the subsequent soft process could determinethe J/ Ψ formation and yields. In this regard, there is no reason to believe/assume theinitial J/ Ψ production at low p T should follow a binary scaling in a nucleus-nucleusor nucleon-nucleus collision. In fact, there is no experimental evidence that the binaryscaling is followed, although the effect is often attributed to the cold nuclear absorption.This may explain why the J/ Ψ suppression in Au+Au collisions at RHIC is strongerat forward rapidity than at midrapidity. This observation may strengthen the recenttheoretical development on J/ Ψ production mechanisms [32, 33].The nuclear modification factor R AA is the ratio of the p T spectra in Cu+Cuand p + p collisions scaled by the number of the underlying binary nucleon-nucleon ecent Heavy-Flavor results at STAR φ∆ -1 0 1 2 3 4 T r g / N φ ∆ / d c h d N +X, 17% ψ J/ → B > 5 GeV/c ψ J/T p > 0.5 GeV/c associateT p STAR Preliminary (GeV/c)p ) ψ ) / ( i n c l u s i ve J / ψ J / → ( B (pQCD)/(STAR Spectra), p+p 200GeV-hadron correlation, p+p 200GeV ψ STAR, J/ 630GeVp-hadron correlation, p+ ψ UA1, J/ 1.8TeVp impact parameter, p+ µ D0, 1.96TeVpCDF, B life time, p+
STAR Preliminary
Figure 3. (Color online). (a)
J/ψ -hadron azimuthal correlations after backgroundsubtraction. (b) Fraction of B → J/ Ψ + X over the inclusive J/ Ψ from differentmeasurements at UA1, STAR, D0 and CDF. collisions [18]. The systematic uncertainty is similar to that on the invariant spectrawith the contribution from efficiency partly cancelling out in the ratio. The R AA tendsto increase from low to high p T , although the error bars currently do not allow todraw strong conclusions. If we assume the systematic and statistical errors of PHENIXCu+Cu data points at high p T are correct and are independent of those from STAR,we can obtain more high p T data points by combining PHENIX Cu+Cu results withSTAR p + p results, the average R AA at p T > . ± . stat. ) ± . syst. ).These results are consistent with unity and two standard-deviation higher than that atlow p T ( R AA ∼ .
6) measured by PHENIX [34]. This result is also in contrast to theexpectation from AdS/CFT-based model (dotted-dashed curve) [3, 35] and from theTwo-Component-Approach model (dashed curve) [36], which predict a decreasing R AA with increasing p T . Similar result was also observed by NA60 Collaboration in In + In collisions at √ s NN = 17 . R AA reaches unity at much smaller p T than at RHIC and most likely of a different physics origin. These results could indicatethat other J/ Ψ production mechanisms such as recombination or formation time [38, 39]may play an important role at high p T .The large S/B ratio of the J/ Ψ in p + p collisions allows the study of J/ψ -hadroncorrelations to understand the J/ Ψ production mechanism at high p T . Figure 3. a showsthe azimuthal angle correlations between a high p T J/ Ψ ( p T > p T > . c in the same event. No significant near side correlationswere observed, which is in contrast to the dihadron correlation measurements [40] wherethe height of the near-side correlation at zero degree is no less than that of the away-sidecorrelation at 180 degree. Since the Monte Carlo simulations show a strong near sidecorrelation if the J/ Ψ is produced from B -meson decay [25, 41], these results can beused to constrain the B -meson contribution to J/ Ψ production. The contribution tothe
J/ψ -hadron correlation from B -meson decay was simulated with the same kinematicacceptance in PYTHIA events. If we attribute all the near-side excess to the B -mesonfeed-down as was done in UA1 [25, 41], we conclude that B -meson feed-down contributes ecent Heavy-Flavor results at STAR ) ) (GeV/c - e + (e Inv M ) C oun t s / ( M e V / c - +e + e →ψ (a) J/ d+Au @ 200 GeV > 2.5 GeV/c T p STAR Preliminary unlike-signlike-sign ) ) (GeV/c - e + (e Inv M ) C oun t s / ( M e V / c - +e + e →Υ (b) d+Au @ 200 GeV STAR Preliminary unlike-signlike-sign
Figure 4. (Color online) (a) High- p T J/ψ in d+Au collisions from run8 (b)Υ rawyields in d+Au collisions from run8. < ∼ ±
3% to the inclusive J/ Ψ yields at p T > B -mesonproduction based on pQCD [42] with the B → J/ Ψ + X decay form factor from CLEOmeasurments [43] shows that the fraction of J/ Ψ from B -meson feed-down at high p T is sensitive to the B -meson cross section and should contribute to the J/ Ψ yields at thelevel of 20-40%. Apart from a conculsion on the B -meson contribution of < ∼ J/ Ψ are produced alone most of the time ( > ∼ J/ Ψ production mechanism.Further correlation measurements of J/ Ψ - γ with high statistics will provide the fractionof J/ Ψ from Ξ c decay. Future measurements of J/ Ψ R AA from Au+Au and p+pcollisions with RHIC luminosity and detector upgrades are anticipated to provide aprecision test on the p T dependence of J/ Ψ suppression [18, 44].The STAR experiment already reported on the first RHIC measurement of theΥ(1S+2S+3S) cross section at mid-rapidity in p+p collisions at √ s = 200 GeV [45].The first ever measurements of Υ mesons in Au+Au collisions at √ s NN = 200 GeV areunderway. We observe a stable signal, that will allow us to get first information on thenuclear modication factor of the Υ. This will be complemented by measurements ind+Au collisions taken in 2008. A clean signal with negligible background from d+Aucollisions in run8 is shown in Fig. 4 together with the J/ Ψ invariant mass distributionfrom the same run.In summary, we reported the preliminary non-photonic electron results from run8p+p collisions taken from a new TOF and TPC readout sector with low inner detectormaterial budget at STAR. We also reported measurements of J/ Ψ spectra in p + p andminimum bias Cu+Cu collisions from low p T to high p T at RHIC mid-rapidity throughthe dielectron channel. The high p T J/ Ψ production was found to follow the x T scalingwith a beam energy dependent factor ∼ √ s NN 5 . ± . while the low p T J/ Ψ fails the x T scaling test. The average of J/ Ψ nuclear modification factor R AA at p T > . ± . stat. ) ± . syst. ) and is 0 . ± . ± .
13 when combined from all RHIC data.This is consistent with no J/ Ψ suppression, and is about 2 σ above the values at low p T ecent Heavy-Flavor results at STAR p T J/ Ψ on the near side. The fraction of J/ Ψ from B -meson decay is found to beless than 17 ±
3% at p T > References [1] Yuri L. Dokshitzer and D. E. Kharzeev.
Phys. Lett. , B519:199–206, 2001.[2] T. Matsui and H. Satz.
Phys. Lett. , B178:416, 1986.[3] H. Liu, K. Rajagopal, and U.A.Wiedemann.
Phys. Rev. Lett. , 98:182301, 2007.[4] Hong Liu, Krishna Rajagopal, and Urs Achim Wiedemann.
Phys. Rev. Lett. , 97:182301, 2006.[5] Elena Caceres and Alberto Guijosa.
JHEP , 11:077, 2006.[6] John Adams et al.
Phys. Rev. Lett. , 94:062301, 2005.[7] B. I. Abelev et al. arXiv: , nucl-ex::0805.0364, 2008.[8] B. I. Abelev et al.
Phys. Rev. Lett. , 98:192301, 2007.[9] M. Anderson et al.
Nucl. Instrum. Meth. , A499:659–678, 2003.[10] M. Beddo et al.
Nucl. Instrum. Meth. , A499:725–739, 2003.[11] Esteve Bosch et al.
IEEE Trans. Nucl. Sci. , 50:2460–2469, 2003.[12] Jin Fu (for the STAR Collaboration). Strangeness in Quark Matter 2008.[13] Stephen Scott Adler et al.
Phys. Rev. Lett. , 96:032001, 2006.[14] A. Adare et al.
Phys. Rev. Lett. , 97:252002, 2006.[15] Shingo Sakai (for the STAR Collaboration). Strangeness in Quark Matter 2008.[16] B. I. Abelev et al.
Phys. Rev. Lett. , 98:192301, 2007.[17] John Adams et al.
Phys. Rev. Lett. , 94:062301, 2005.[18] Zebo Tang.
J. Phys. , G35:104135, 2008.[19] M. Banner et al.
Phys. Lett. , B115:59, 1982.[20] John Adams et al.
Phys. Lett. , B637:161–169, 2006.[21] B. Alper et al.
Nucl. Phys. , B100:237, 1975.[22] D. Antreasyan et al.
Phys. Rev. , D19:764, 1979.[23] Darin E. Acosta et al.
Phys. Rev. , D71:032001, 2005.[24] F. Abe et al.
Phys. Rev. Lett. , 79:572–577, 1997.[25] C. Albajar et al.
Phys. Lett. , B256:112–120, 1991.[26] C. Kourkoumelis et al.
Phys. Lett. , B91:481, 1980.[27] Mauro R. Cosentino. nucl-ex:0806.0353, 2008.[28] A. G. Clark et al.
Phys. Lett. , B74:267, 1978.[29] A. L. S. Angelis et al.
Phys. Lett. , B79:505–510, 1978.[30] Stephen Scott Adler et al.
Phys. Rev. , C69:034910, 2004.[31] A. Adare et al.
Phys. Rev. Lett. , 98:232301, 2007.[32] Dmitri Kharzeev, Eugene Levin, Marzia Nardi, and Kirill Tuchin. arXiv: , hep-ph::0808.2954, 2008.[33] M. Mishra, C. P. Singh, and V. J. Menon. arXiv , hep-ph:0711.3359, 2007.[34] A. Adare et al.
Phys. Rev. Lett. , 101:122301, 2008.[35] T. Gunji.
J. Phys.G: Nucl. Part. Phys. , 35:104137, 2008.[36] Xingbo Zhao and Ralf Rapp.
Phys. Lett. B , 664:253–257, 2008.[37] R. Arnaldi.
J. Phys.G: Nucl. Part. Phys. , 35:104133, 2008.[38] F. Karsch and R. Petronzio.
Phys. Lett. , B193:105, 1987.[39] J. P. Blaizot and Jean-Yves Ollitrault.
Phys. Lett. , B199:499–503, 1987.[40] John Adams et al.
Phys. Rev. Lett. , 95:152301, 2005.[41] C. Albajar et al.
Phys. Lett. , B200:380, 1988.[42] Matteo Cacciari, Paolo Nason, and Ramona Vogt.
Phys. Rev. Lett. , 95:122001, 2005.[43] S. Anderson et al.
Phys. Rev. Lett. , 89:282001, 2002.[44] Zhangbu Xu and Thomas Ullrich. arXiv: , nucl-ex::0809.2288, 2008.[45] Debasish Das.