NNuclear and Particle Physics Proceedings 00 (2021) 1–5
Nuclear andParticle PhysicsProceedings
Measurement of quarkonium production in ALICE ∗ Victor Feuillard, for the ALICE Collaboration
Physikalisches Institut, Heidelberg, Germany
Abstract
ALICE is designated to study the quark–gluon plasma (QGP), a state of matter where, due to high temperature anddensity, quarks and gluons are deconfined. One of the probes used to investigate this state of matter is quarkoniumstates, bound states of either a charm and anti-charm quark pair (charmonia) or a bottom and anti-bottom quark pair(bottomonia). The presence of the QGP is expected to modify the quarkonium production yields in a very specificway due to a balance between medium-induced suppression, and a recombination mechanism or a hadronizationmechanism. To understand the the properties of the QGP in nucleus-nucleus collisions, it is essential to measurethe quarkonium di ff erential yields in proton–proton collisions, as it provides a reference and allows the investigationof quarkonium production mechanisms, as well as in proton–nucleus collisions to understand the cold nuclear mattere ff ects that appear. In this contribution, the latest results for quarkonium production measured with the ALICE detectorin pp collisions at di ff erent collision energies are reported. The measurements of the nuclear modification factor andanisotropic flow in Pb–Pb collisions at √ s NN = .
02 TeV and in p–Pb at √ s NN = .
16 TeV at mid- and forwardrapidity are also reported. All measurements are compared to various theoretical predictions.
Keywords:
QGP, heavy-ion, quarkonium, J / ψ , Υ , elliptic flow, Cold Nuclear Matter, ALICE, LHC
1. Introduction
The quark–gluon plasma (QGP) is a state of mat-ter theoretically predicted by lattice quantum chromo-dynamics (QCD) where quark and gluons are decon-fined. There is a particular interest in studying the QGPsince in current cosmological models it was the firststate of matter in the early stages of the Universe (upto τ ≈ µ s). Experimentally it is possible to createa QGP using ultra-relativistic heavy-ion collisions, likeat RHIC [1] or the LHC [2], but only within a shortperiod of time and a very small volume ( ∼
10 fm / c and ∼ fm in Pb–Pb at √ s NN = .
76 TeV) [3]. Dueto this small time period and volume, it is impossibleto observe the QGP directly, therefore measurements of ∗ Talk given at 23rd International Conference in Quantum Chro-modynamics (QCD 20), 27 - 30 October 2020, Montpellier - FR
Email address: [email protected] (Victor Feuillard, for the ALICE Collaboration) the particles produced, such as quarkonia, are used asprobes. Because of their large mass, charm and bottomquarks are primarily produced at the very beginning ofthe collision and experience the entire medium evolu-tion. Quarkonium resonances, which are bound states ofa c¯c pair for charmonium or a b¯b pair for bottomonium,are therefore among the most direct signatures for theQGP formation. Firstly, theory predicts that quarkoniaare suppressed in a QGP due to the colour screening: thepresence of free colour charges in the medium, the bind-ing potential is screened [4]. This leads to a reduction ofthe number of quarkonium states produced. Secondly,a competing mechanism can occur, namely recombina-tion: if there are enough quark pairs produced, and theythermalize in the QGP, then quarkonia can be regener-ated by the recombination of these quark pairs, resultingin an increase of the quarkonium yields [5, 6].In the following, a selection of the latest results ob-tained by ALICE in pp, p–Pb, and Pb–Pb collisions are a r X i v : . [ nu c l - e x ] J a n Nuclear and Particle Physics Proceedings 00 (2021) 1–5 reported. A complete description of the ALICE detectorcan be found in [7, 8].
2. Results in pp collisions at √ s =
13 TeV
In addition to the measurements in heavy-ion colli-sions, it is also interesting to measure the charmoniumproduction in pp collisions. Indeed, it provides a ref-erence for the measurement in Pb–Pb collisions andalso allows to explore the QCD models, as the quarko-nium formation involves both hard-scale processes forheavy quark production, and soft-scale processes forhadronization. / ψ production cross section A new measurement of the J / ψ production cross sec-tion in pp collisions at √ s =
13 TeV at mid-rapidityas a function of transverse momentum ( p T ) is presentedin Fig. 1. It is compared with previous measurementsat mid-rapidity at collision energies of √ s = √ s = .
02 TeV [10]. With the increasing collisionenergy, a hardening of the p T is observed, similar to ob-servations at forward rapidity [11]. c (GeV/ T p −
10 110 )) c b / ( G e V / µ ( T p d y d σ d ± = 21.1 nb int L = 13 TeV, s
100 x (Preliminary) 4% ± = 5.6 nb int L = 7 TeV, s
10 x (Phys. Lett. B704 (2011) 442 455) 2.1% ± = 19.5 nb int L = 5.02 TeV, s ALICE pp |<0.9 y , | ψ Inclusive J/
ALI−PREL−319654
Figure 1: J / ψ cross section measured at mid-rapidity for di ff erent col-lision energies as a function of p T . / ψ yield as a function of charged-particle multiplicity A good test of QCD models is the measurement ofthe normalized J / ψ yield as a function of the normalizedcharged-particle multiplicity. The correlation betweenthe two quantities helps to constrain the interplay be-tween the soft and hard mechanisms in pp collisions.A new measurement of the normalized inclusiveJ / ψ yield at mid-rapidity as a function of normalized charged-particle pseudorapidity density at mid-rapidityin pp collisions at √ s =
13 TeV is presented in Fig. 2.The data are compared with several theoretical mod-els [12]. The normalized J / ψ yield exhibits a fasterthan linear increase with the normalized multiplicity.This behaviour is predicted by all the theoretical mod-els currently used. It is explained e ff ectively as the re-sult of a reduction of the charged-particle multiplicity athigh multiplicity, however each model attributes the ob-served behavior to di ff erent underlying processes (colorstring reconnection or percolation, gluon saturation, co-herent particle production, 3-gluon fusion in gluon lad-ders / Pomerons).
ALI-PUB-348246
Figure 2: Normalized inclusive p T -integrated J / ψ yield at mid-rapidityas a function of normalized charged-particle pseudorapidity density( | η | <
1) [12] compared with di ff erent models [13–17]. In order to better understand the mechanisms at play,more tests are required on the models. In addition, thepossibility to separate between prompt and non-promptJ / ψ could provide further information.
3. Results in p–Pb collisions at √ s NN = .
16 TeV
Measuring the nuclear modification in p–Pb colli-sions, meaning the ratio of production yields in p–Pbcollisions with respect to pp collisions, allows to un-derstand cold nuclear matter (CNM) e ff ects, such asnuclear shadowing of the partonic structure functions.It leads to a change in the probability for a quark orgluon to carry a fraction of the nucleon momentum ( x )and therefore a ff ects the production cross section of theheavy quark pair. This allows to distinguish in the Pb–Pb measurements the e ff ects that originate from the hot Nuclear and Particle Physics Proceedings 00 (2021) 1–5 medium from those caused by the presence of nuclearmatter. The asymmetry of the p–Pb collision allows usto probe di ff erent x regions of the lead nucleus: mea-suring in the p-going direction equates to investigatingthe low- x , whereas measuring in the Pb-going directioncorresponds to investigating the high- x region. A new measurement of the ψ (2S) nuclear modifi-cation factor ( R pPb ) in p–Pb collisions at √ s NN = .
16 TeV as a function of rapidity is presented in Fig. 3,compared with the J / ψ one [18]. We can observe thatat forward rapidity, corresponding to the p-going direc-tion, the J / ψ and the ψ (2S) are suppressed in p–Pb colli-sions. The values for the J / ψ and the ψ (2S) agree withinuncertainties. On the other hand, at backward rapidity,corresponding to the Pb-going direction, the ψ (2S) isalso suppressed whereas the J / ψ R pPb is compatible withunity. ALI-PUB-347553
ALI-PUB-347565
Figure 3: (Top) J / ψ and ψ (2S) R pPb as a function of rapidity com-pared with models including CNM only [18]. (Bottom) Same datacompared with models including final state e ff ects. In the top panel of Fig. 3, the data are compared withtheoretical models that include CNM e ff ects only [19–22], which are largely independent from the specific charmonium resonance and can therefore be comparedto both. The models show a good agreement with theJ / ψ values both at forward and backward rapidity. How-ever in the case of the ψ (2S), the models show a goodagreement with the result at forward rapidity but over-estimate the result at backward rapidity.In the bottom panel of Fig. 3, measurements are com-pared with theoretical models that include some finalstate e ff ects as well. These models show a good agree-ment with both resonance measurements, at both for-ward and backward rapidity, when available. This tendsto indicate that final state interactions, which can lead toa suppression of the charmonium resonance, may havea stronger e ff ect on the ψ (2 S ) due to its lower bindingenergy. In Fig. 4 the Υ (1S) R pPb in p–Pb collisions at √ s NN = .
16 TeV is presented as a function of rapidity [23]. It iscompared with the LHCb results [24] and with severaltheoretical models [20–22, 25, 26].
ALI-PUB-346423
Figure 4: Υ (1S) R pPb as a function of rapidity compared with modelsand LHCb results [23]. The results show a good agreement with the LHCbmeasurement in the two directions. A suppression ofthe Υ (1S), both at forward and backward rapidity, witha stronger suppression at forward rapidity is seen. Whencompared with the theoretical models, an overall agree-ment is observed considering the still significant exper-imental uncertainties.
4. Results in Pb–Pb collisions at √ s NN = .
02 TeV / ψ nuclear modification factor The J / ψ nuclear modification factor in Pb–Pb colli-sions at √ s NN = .
02 TeV, measured at mid-rapidity
Nuclear and Particle Physics Proceedings 00 (2021) 1–5 is presented in Fig. 5 as a function of the mean num-ber of participants [27]. A moderate suppression formid-central collisions, and an increase towards centralcollisions is seen. The measurement in the most pe-ripheral region is compatible with unity. These resultsstrengthen the hypothesis of regeneration, as a modelwith the color screening alone could not explain thelarge values nor the increase seen in central collisions. ALI-PUB-337728
Figure 5: R AA of the J / ψ at √ s NN = .
02 TeV [27] compared withtheoretical models [28–31].
The result is compared with several theoretical mod-els. The Statistical Hadronization Model (SHM) [28]shows a good agreement with the data, with the uncer-tainties entirely due to the c¯c cross section. The twoTransport Models (TM) [29, 30] and the Comover In-teraction Model (CIM) [31] tend to underestimate themeasurement in the most central collisions.When observed as a function of rapidity, togetherwith the measurement made in the dimuon decay chan-nel at forward rapidity, the J / ψ R AA exhibits a maxi-mum at mid-rapidity ( R y < | . | AA = . ± .
05 (stat . ) ± . . ) [27]) and decreases towards large rapidities( R . < y < . = . ± .
01 (stat . ) ± .
05 (syst . ) [32]). Thisresult was predicted for the regeneration scenario, dueto the larger value of the production cross section σ c¯c at mid-rapidity, leading to more quark pairs created atthe beginning of the collisions, and therefore, more re-combination of charmonium. For color screening, theopposite e ff ect is predicted. The azimuthal dependence of the particle production,the anisotropic flow, is a particularly interesting mea-surement. It can be expressed as a Fourier decomposi-tion: dNd ϕ ∝ + (cid:80) n v n cos[n( ϕ − Ψ n )], where ϕ is theazimuthal angle, Ψ n is the initial state symmetry plane for the n-th harmonic and v n is the n-th order Fourierspatial coe ffi cient. The initial anisotropy of the colli-sion is transformed into a momentum anisotropy of thefinal state particles, that can be quantified by the Fouriercoe ffi cients v n . In particular the v coe ffi cient, calledthe elliptic flow, is caused by the ellipsoidal shape ofthe overlap region in non-central collisions, and the v ,called the triangular flow, is understood to arise fromfluctuations in the initial energy-density profile.In Fig. 6 the elliptic flow of the J / ψ at forward ra-pidity is presented in the left panel, and the first mea-surement of the J / ψ triangular flow in ALICE in theright panel, as a function of p T [33]. It is comparedwith the mid-rapidity v and v of pions, D mesons andprotons [34, 35]. The J / ψ elliptic flows is positive andreaches a maximum at intermediate p T , before decreas-ing. In the low p T region ( p T < / c ), a mass hi-erarchy of the v can be observed: the heavier particle,in this case the J / ψ , has the smaller v , and the lighterthe particle, the larger the elliptic flow. In the high p T region, this ordering disappears and the elliptic flow isindependent from the particle species. ALI-DER-348523
Figure 6: J / ψ elliptic and triangular flow as a function of p T comparedwith other species. Regarding the triangular flow, a clear non-zero v is observed for the J / ψ , indicating that the initialstate energy-density fluctuations reflect also in theanisotropic flow of charm quarks. The same mass hi-erarchy as in the case of the v is observed at low p T .The J / ψ elliptic flow is compared with transportmodel predictions [36] in Fig. 7. It appears that themodel is in good agreement with the data for p T < / c , but strongly underestimates the data at high p T .In the bottom panel of Fig. 7, the elliptic flow of the Υ (1S) is presented as a function of p T , compared withthe J / ψ v and transport model predictions [36, 38]. The Υ (1S) v is compatible with zero, and is in agreement Nuclear and Particle Physics Proceedings 00 (2021) 1–5 ALI-PUB-347891 ) c (GeV/ T p v − ψ Inclusive J/(1S) ϒ (1S), TAMU model ϒ (1S), BBJS model ϒ = 5.02 TeV NN s Pb − ALICE Pb 60% − y ALI−PUB−325477
Figure 7: (Top) J / ψ elliptic flow as a function of p T [33] comparedwith transport model calculations [36]. (Bottom) Υ elliptic flow asa function of p T [37] compared with the J / ψ one and model calcula-tions [36, 38]. with the transport model predictions, which include lit-tle or no regeneration for the Υ (1S). However, the largeuncertainties on the measurement prevent from draw-ing any firm conclusion. Moreover, models where theb quark fully thermalize also predict a very small v forthe Υ (1S) in the measured p T region [39]. Therefore,more precise measurements are required to determinewether the beauty quark exhibits collective behaviour.
5. Conclusion
An overview of some of the latest results on quarko-nium production in ALICE has been presented. In ppcollisions at √ s =
13 TeV, a new measurement of theJ / ψ cross section has been performed as well as a newmeasurement of the J / ψ yield as a function of charged-particle multiplicity at mid-rapidity. It shows a fasterthan linear increase that is qualitatively reproduced bythe di ff erent models, but the exact mechanism has yet tobe understood. In p–Pb collisions at √ s NN = .
16 TeV, R pPb has been measured for the J / ψ , ψ (2S), and Υ (1S)quarkonium states. Models including final state e ff ectscan describe the two charmonium states at both forward and backwards rapidity. Finally, in Pb–Pb collisionsat √ s NN = .
02 TeV, the J / ψ R AA has been measuredat mid-rapidity. The experimental data require, beyondthe e ff ect of color screening and deconfinement, a dom-inant contribution from regeneration. Finally, the mea-surement of the J / ψ elliptic and triangular flow showsa positive v and v , and a mass hierarchy at low p T inboth cases. The results are in line with the regenerationscenario except at high p T , where the results exhibit alarge v . The Υ (1S) v has also been measured up to p T =