Low-Mass Dielectron Production in pp, p-Pb and Pb-Pb Collisions with ALICE
NNuclear Physics A 00 (2018) 1–4
NuclearPhysics A / locate / procedia Low-Mass Dielectron Production in pp, p–Pb and Pb–PbCollisions with ALICE
Patrick Reichelt (for the ALICE Collaboration)
Institut f¨ur Kernphysik, Goethe-Universit¨at Frankfurt am Main, Germany
Abstract
The ALICE Collaboration measures the production of low-mass dielectrons in pp, p–Pb and Pb–Pb collisions at theLHC. The main detectors used in the analyses are the Inner Tracking System, Time Projection Chamber and Time-Of-Flight detector, all located at mid-rapidity. The dielectron yield in p–Pb collisions shows an overall agreement withthe hadronic cocktail. The pair transverse momentum distributions are sensitive to the contributions from open heavy-flavours. In Pb–Pb collisions, uncorrected background-subtracted yields have been extracted in two centrality classes.In pp collisions the production of virtual photons relative to the inclusive yield is determined by analyzing the dielectronexcess with respect to the expected hadronic sources. The direct photon cross section is then calculated and found to bein agreement with NLO pQCD calculations. A feasibility study for LHC Run 3 after the ALICE upgrade indicates thepossibility for a future measurement of the early e ff ective temperature. Keywords: dielectron, electron, heavy-flavour, virtual photon
1. Introduction
The measurement of electron-positron pairs in the low invariant mass region allows studying the vacuumand in-medium properties of light vector mesons. Additionally, dielectrons from semileptonic decays ofcorrelated heavy-quark mesons carry information on the heavy-flavour production in the di ff erent collisionsystems. Low-mass dielectrons are also produced by internal conversion of virtual direct photons. Toquantify modifications of the dielectron production in heavy-ion collisions, measurements in pp collisionsserve as a reference, while the analysis of p–A collisions allows disentangling cold from hot nuclear mattere ff ects. In ALICE [1] at the LHC, dielectron measurements are performed using the central barrel detectorsat mid-rapidity. In this proceedings we present the invariant mass analyses in p–Pb and Pb–Pb collisions,a mass-di ff erential study in p–Pb, and a virtual direct photon measurement in pp collisions. Prospects of afuture measurement in Pb–Pb collisions after the ALICE upgrade for LHC Run 3 are also discussed.
2. Low-mass dielectrons in ALICE: analysis and results
The analyses presented here are based on 3 · minimum-bias pp collisions at √ s = minimum-bias p–Pb collisions at √ s NN = .
02 TeV, as well as 1 . · and 1 . · events in central(0–10%) and semi-central (20–50%) Pb–Pb collisions, respectively. The electron selection in the pp and a r X i v : . [ h e p - e x ] D ec / Nuclear Physics A 00 (2018) 1–4 p–Pb analyses share the same fiducial cuts on transverse momentum ( p T > . / c ) and pseudorapidity( | η | < . . < p T < . / c ensures exclusion of charged pions and reducesthe contribution of soft electrons from conversions and π -Dalitz decays to the combinatorial background.A conversion rejection cut on the pair level is done in all collision systems. In pp collisions, a clean electronsample is achieved by applying cuts on the time-of-flight (TOF) and on the specific energy loss d E / d x inthe Time Projection Chamber (TPC). In p–Pb and Pb–Pb, a TOF signal is not required in order to increasethe electron e ffi ciency, yet used if available to purify the electron sample. Kaons and protons are rejectedby an electron inclusion cut on the d E / d x measured by the Inner Tracking System (ITS). Dielectron spectraare created for unlike-sign (ULS) and like-sign (LS) combinations of selected particles. The signal yield isobtained as N sameULS − N sameLS · R , where R accounts for any acceptance di ff erence between unlike- and like-signpairs due to detector e ff ects. It is built via a mixed-event technique as R = N mixULS / N mixLS and stays within 5%around unity for the full invariant mass range considered in these analyses. ) - ) c (( G e V / ee m / d N d N S D e v t N / -5 -4 -3 -2 -1 Cocktail sum with uncertaintiesee γ → π ee γ → η ee π → ω ee and → ω ee η → φ ee and → φ ee γ → ' η ee → ρ = 6.9mb) cc σ > x pp PYTHIA MNR, pPbcoll (
Fig. 1. Dielectron mass spectrum in p–Pb collisions in compar-ison to the hadronic cocktail. ) c (GeV/ ee m ) - ) c r a w y i e l d (( G e V /
10 0-10%20-50% ALICE Preliminary = 2.76 TeV NN s Pb, − Pb | < 0.76 e η , | c < 3.5 GeV/ eT p c ≤ pairT p ≤ ALI-DER-90991
Fig. 2. Dielectron raw subtracted yield in central and semi-central Pb–Pb collisions after background subtraction.
The corrected dielectron yield in p–Pb collisions, integrated over pair- p T ( p eeT ), is shown in Fig. 1.The data points and systematic uncertainties are extracted from the mean values and the spread of resultsobtained with 22 di ff erent combinations of analysis cut settings (inspired by [2]). The data are comparedto the hadronic cocktail, which uses the charged pion measurement [3] as π input and m T -scaling forthe other light mesons. Open heavy-flavour and J /ψ contributions are calculated from PYTHIA, tunedto independent ALICE measurements in pp and p–Pb collisions [4, 5]. Uncertainties coming from theinput of the cocktail sources are shown as a grey band in Fig. 1. Data and cocktail are in reasonableagreement within their uncertainties over the full mass range. Figure 2 shows, for Pb–Pb collisions in bothcentralities, the background-subtracted uncorrected yields in the low-mass region for 1 < p eeT < / c .The corresponding signal-to-background ratios within 0 . < m ee < . / c are 0 .
01 for 20–50% and0 .
003 for 0–10% centrality. The e ffi ciency correction and cocktail comparison are in progress.The kinematic region p eeT (cid:29) m ee is useful to study the production of virtual direct photons. Figure 3 showsthe di ff erential dielectron cross-section in pp collisions at low m ee for one pair- p T region, compared to ahadronic cocktail and its components. Also shown is the expected mass distribution of dielectrons comingfrom virtual direct photons, after having applied the single-electron fiducial cuts on p T and η . A fit to the datais performed to determine the virtual photon fraction r by using the function f combined = (1 − r ) · f cocktail + r · f γ, dir in the range 0 . < m ee < . / c . This fit is done in four pair- p T regions and the extracted virtual photonfraction is then multiplied by the inclusive photon cross section, also measured by ALICE via the photonconversion method (PCM) [6]. In Figure 4 the resulting direct photon cross section is compared to NLO Nuclear Physics A 00 (2018) 1–4 ) c (GeV/ ee m ) c ( m b / G e V / T p d ee m d σ d data ,dir γ f ,dir γ *f r + cocktail )*f r (1 cocktail sum c < 3.2 GeV/ Tee p ALICE preliminary =7 TeVspp, c >0.2 GeV/ Te p |<0.8 e η | π η ’ η ωφ c ALI−PREL−69064
Fig. 3. Low-mass region of the dielectron cross section in pp collisions for2 . < p eeT < . / c compared to the hadronic cocktail, as well as to afit to extract the virtual direct photon fraction r . ) c (GeV/ T p ) c ( pb G e V p / d σ d E (W.V.) NLOdirect γ =0.5 µ =1.0 µ =2.0 µ incl γ × r = direct γ (PCM) incl γ
95 % C.L.
ALICE preliminary =7 TeVspp,
ALI−PREL−69076
Fig. 4. Direct photon cross section compared to NLOpQCD calculations, and the inclusive photon crosssection extracted from PCM. pQCD calculations. Reasonable agreement is found, within uncertainties, between data and model. Detailson this analysis can be found in [7].The dielectron yield in p–Pb collisions is analyzed as a function of pair- p T in di ff erent mass regions to gainadditional sensitivity in the hadronic cocktail comparison. At LHC energies, heavy-flavour contributionsalready become relevant for 0 . < m ee < .
75 GeV / c , as shown in Fig. 5, while they completely dominatethe spectrum for 1 . < m ee < . / c , as shown in Fig. 6. The analysis is being extended to highermomenta to evaluate di ff erent model calculations and extract heavy-flavour cross sections. ) - ) c (( G e V / ee T p / d N d N S D e v t N / -5 -4 -3 -2 Cocktail sum with uncertaintiesee γ → η ee π → ω ee and → ω ee γ → ' η ee → ρ = 6.9mb) cc σ > x pp PYTHIA MNR, pPbcoll (
Fig. 5. Dielectron pair- p T spectrum in p–Pb collisions comparedto the hadronic cocktail, for 0 . < m ee < .
75 GeV / c . ) - ) c (( G e V / ee T p / d N d N S D e v t N / -6 -5 -4 -3 -2 Cocktail sum with uncertainties = 6.9mb) cc σ > x pp PYTHIA MNR, pPbcoll (
Fig. 6. Same as Fig. 5, but for 1 . < m ee < . / c , whichis most sensitive to c ¯ c and b ¯ b production.
3. Upgrade Study for future dielectron measurements in Pb–Pb collisions
A fine-binning di ff erential low-mass dielectron measurement in Pb–Pb collisions is a major physics casefor the ALICE upgrade for LHC Run 3 [8]. We evaluated to which precision a possible thermal excess / Nuclear Physics A 00 (2018) 1–4 yield [9] could be reconstructed in Pb–Pb collisions at full LHC energy ( √ s NN = c ¯ c contribution. Figure 8 shows the expectedspectrum after subtraction of the hadronic sources, where a thermal fit can be applied. The ALICE upgradeis expected to allow for a temperature measurement in the intermediate mass region within 10% statistical(based on 2 . · events) and 20% systematic uncertainty [10]. ) (GeV/c ee M0 0.5 1 1.5 2 2.5 ) - d y ( G e V ee d N / d M -6 -5 -4 -3 -2 -1 SumRapp in-medium SFRapp QGP 10%) – ( r cocktail w/o 20%) – ee ( fi cc2.5E9 'measured' 0.25%) – Syst. err. bkg. ( = 5.5 TeV NN sPbPb @ 0 - 10%, 2.5E9 events| < 0.84 e |y > 0.2 GeV/c eT p < 3.0 t,ee Fig. 7. Predicted dielectron signal composition for √ s NN = . ) (GeV/c ee M0 0.5 1 1.5 2 2.5 ) - d y ( G e V ee d N / d M -6 -5 -4 -3 -2 -1 Rapp SumRapp in-medium SFRapp QGP - cockt.c2.5E9 'meas.' - cSyst. err. bkg. + cocktailcSyst. err. c = 5.5 TeV NN sPbPb @ 0 - 10%, 2.5E9 events| < 0.84 e |y > 0.2 GeV/c eT p < 3.0 t,ee Fig. 8. Excess dielectron yield after hadronic cocktail subtrac-tion and propagating expected uncertainties. An exponential fitfor m ee > . / c gives the early e ff ective temperature. [10]
4. Conclusions
The dielectron yield measured in p–Pb collisions is in agreement with the expected hadronic sources,and the p eeT -distributions are sensitive to heavy-flavour production. From Pb–Pb collisions, raw subtractedyields at low mass have been extracted in two centrality classes. In the related kinematic region p eeT (cid:29) m ee ,virtual direct photons are observed in pp collisions. The extracted direct photon cross section agrees withNLO pQCD calculations within uncertainties. A feasibility study indicates that the ALICE upgrade for LHCRun 3 will facilitate a measurement of the e ff ective temperature at the early times of a Pb–Pb collision. Acknowledgements.
This work is supported by the German BMBF and the Helmholtz Association.
References [1] K. Aamodt et al. (ALICE Collaboration), JINST 3 (2008) S08002.[2] R. Barlow, arXiv:hep-ex / //