Recent results on jet physics from ALICE at the LHC

Transcription

1 Journal of Physics: Conference Series OPEN ACCESS Recent results on jet hysics from ALICE at the LHC o cite this article: Alexandre Shabetai and the ALICE Collaboration 23 J. Phys.: Conf. Ser View the article online for udates and enhancements. Related content - Probing the medium with jets in ALICE Pasquale Di Nezza and the Alice Collaboration - Measurement of D-meson roduction in, -Pb, and Pb-Pb collisions at the LHC with the ALICE detector Chitrasen Jena and the Alice Collaboration - Reconstructed Jet Results in +, d+au and Cu+Cu collisions at 2 GeV from PHENIX D V Pereelitsa and the Phenix Collaboration his content was downloaded from IP address on 28/4/28 at :54

2 29th Winter Worksho on Nuclear Dynamics (WWND23) Journal of Physics: Conference Series 458 (23) 225 doi:.88/ /458//225 Recent results on jet hysics from ALICE at the LHC Alexandre SHABEAI (for the ALICE Collaboration) SUBAECH - CNRS/IN2P3 - UMR Laboratoire de hysique subatomique et des technologies associées - 4 rue Alfred Kastler - BP Nantes cedex 3, France shabetai@in23.fr Abstract. An overview of recent results on jet hysics from the ALICE collaboration is resented. We mainly discuss azimuthal correlations and nuclear modification factors for both hadrons and jets. he very low tracking (> 5 MeV/c) caabilities of the ALICE detector associated to its electromagnetic calorimetry, allow to reconstruct both charged jets and full jets, with minimum bias on the jet fragmentation. A detailed measurement of the underlying event fluctuations will be discussed, as well as, jet sectra measurements, in and Pb-Pb collisions, using low constituent cuts and hadron-recoil jet correlations, which give a controlled way to subtract the combinatorial background.. Physics motivations Jets are roduced in hard scatterings of high energy quarks and gluons in the early stage of heavy-ion collisions. his allows to robe the hot and dense matter formed in these collisions. Jets can be used to study in medium energy loss and its ath length deendence thanks to their strong interaction with the medium, which leads to a modification of their structure and to a redistribution of their energy. Exerimentally, this can be seen as a marked reduction of their energy measured in a given reconstruction cone. heir fragmentation attern is also exected to be modified. his henomenon is called jet quenching []. Many exerimental observations such as the suression of back-to-back azimuthal correlations, the suression of inclusive hadron or even jet sectra can all be used in order to learn about in-medium energy loss. Comaring those to theoretical calculations rovides insight into the density of the system. Moreover, measurements of jets in roton-roton collisions rovide the baseline needed for heavy-ion studies, and at the same time allow to test QCD. In these roceedings, an overview of recent results on jet hysics from the ALICE collaboration is resented. We mainly discuss azimuthal correlations and nuclear modification factors, for both hadrons and jets. 2. Exerimental setu In ALICE, high articles or jets are reconstructed at mid-raidity using either two or three sub-systems: Charged articles are measured using the central tracking system [2] which consists of the ime Projection Chamber (PC) and of the Inner racking System (IS). Photons (mainly from π decay) and electrons are in turn measured by the ALICE Electromagnetic Calorimeter (EMCal) [3]. Content from this work may be used under the terms of the Creative Commons Attribution 3. licence. Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI. Published under licence by Ltd

3 29th Winter Worksho on Nuclear Dynamics (WWND23) Journal of Physics: Conference Series 458 (23) 225 doi:.88/ /458//225 he ALICE PC measures the momentum of charged article down to = 5 MeV/c over a seudo-raidity range of η <.9 and has a full azimuthal accetance ( < ϕ < 2π). he electromagnetic comonent of the jet energy (which is on average about /3 of the total energy of a given jet) is measured by the ALICE EMCal. his is a SHASHLIK [2, 3] (Pb-scintillator) samling electromagnetic calorimeter which is read-out by APDs. It covers η <.7 and has an energy resolution of %/ E +.7%. It has been fully installed in 2: its azimuthal accetance was increased to φ = 7 allowing it to be used for jet hysics. 3. Hadron R AA he first observable that we will consider is the nuclear modification factor of charged hadrons. his quantity is defined as the ratio of charged article yields in Pb-Pb (to study final state effects) or in -Pb (to study initial state effects) over the corresonding yield in collisions, normalize by /N coll calculated from the Glauber model [4]. In the absence of nuclear modifications, hard rocesses are exected to follow the N coll scaling. he corresonding nuclear modification factor would then be unity (R AA = ). R AA ALICE, Pb-Pb, K, y <.75 s = 2.76 ev, - 5 % Λ, y <.75 unidentified charged articles, η <.8 R AA ALICE, Pb-Pb, s NN = 2.76 ev charged articles, η <.8 norm. uncertainty ALI PREL 945 ALICE Preliminary ALICE (-5%) CMS (-5%) H (Chen et al.) lower density H (Chen et al.) higher density H (A.M.) ASW (.R.) YaJEM-D (.R.) elastic (.R.) large esc P elastic (.R.) small esc P WHDG (W.H.) π uer limit WHDG (W.H.) π lower limit Figure. Left: Comarison of the nuclear modification factor for different article secies. Right: he hadron R AA measured by ALICE and CMS. Initial state effects have been quantified in -A collisions: Using data at s NN = 5 ev taken during the LHC -A ilot run from Setember 22, ALICE made a first measurement of R A [5] (from.5 GeV/c < <2 GeV/c) which is found to be consistent with binary scaling, indicating that the Cronin effect [6] at LHC is very small (if not null) comared to the one measured at RHIC [7, 8, 9,, ]. In order to estimate final state effects in the medium, ALICE also measured the hadron R AA [2] in Pb-Pb collisions at s NN = 2.76 ev u to = 5 GeV/c. In central collisions ( 5% centrality) hadrons are losing a large art of their energy in the medium and are suressed by a factor 6 at 7 GeV/c. However, the amount of suression slowly decreases with increasing. Above 4 GeV/c, hadrons are suressed by a factor 2 and R AA.4.5. A comarison of the hadron nuclear modification factors (resented in the left anel of Fig. ) for different article secies (unidentified articles, ions, Kaons and Lambdas) has also been made at the same center of mass energy of 2.76 ev. hose measurements show a different behaviour at low (corresonding to the soft regime): the Λ R AA is close to unity at = 3 GeV/c, whereas the Kaon and ion R AA are smaller. he Λ/K ratio is found to be enhanced at intermediate but starting at 8 GeV/c and above an agreement is found for all articles: the magnitude of the suression is measured to be R AA = indeendent of the article tye. 2

4 29th Winter Worksho on Nuclear Dynamics (WWND23) Journal of Physics: Conference Series 458 (23) 225 doi:.88/ /458//225 he hadron R AA measured by ALICE and CMS (see right anel of Fig. ) [2, 3] are in good agreement and the trend of those measurements is reroduced by a large number of theoretical redictions. Many of the theory extraolations from RHIC, however, redict too strong suression at the LHC [4]. It has been suggested that taking into account the running of α s reduces the suression [5]. Overall, a full quantitative understanding of energy loss and the medium density requires further theoretical work [6]. 4. Di-hadron azimuthal correlations he second observable that we will consider are di-hadron azimuthal correlations. o build such correlations, a high article, called ger article is selected and associated with all other article in the event. One of the first exerimental measurement of jet quenching was erformed by the SAR collaboration at RHIC [9]. In Au-Au collisions, the disaearance of the away side eak of a di-hadron φ correlations was observed first by SAR. his indicated substantial interactions as the hard-scattered artons traverse the medium. he near side eak, on the other hand, was found to be very similar in and Au-Au collisions, one interretation of these observations is that there is a strong surface bias effect, i.e. that the ger article selects jets with a short in-medium ath length and/or little energy loss. he ALICE collaboration measured long range di-hadron η φ azimuthal correlations in Pb-Pb at s NN = 2.76 ev (see Fig. 2) [7]. A different behaviour is found for low and high regimes: At low (3 GeV/c < t < 4 GeV/c and 2 GeV/c < a < 2.5 GeV/c), a clear near side ridge structure is seen and the away side is broad. A Fourier decomosition was made and the measured distribution is well reroduced by summing v n harmonics with n = to n = 4. his suggests that in this range, the underlying hysics is well described by hydrodynamics. At higher (8 GeV/c < t < 5 GeV/c and 6 GeV/c < a < 8 GeV/c) a different scenario is observed: the near side ridge is relaced by a jet eak and in the away side a small recoiling jet distribution can be seen. In addition, the revious global fit fails indicating that the hysics is different in this higher regime. ALI-PUB-47 ALI-PUB-4 Figure 2. Di-hadron η φ azimuthal correlations in Pb-Pb for two different regime (see text). However, those measurements do not give access to the full jet energy: they are limited by the fact that a high leading article at LHC energies only carries about 5% of the energy of the corresonding jet. o overcome this limitation, we have to measure full jets. 5. Full Jets measurements A jet can be defined at several levels: A hard scattering rocess will lead into a artonic shower (which can be seen as a arton jet) which after hadronization will generate a hadronic shower 3

5 29th Winter Worksho on Nuclear Dynamics (WWND23) Journal of Physics: Conference Series 458 (23) 225 doi:.88/ /458//225 (corresonding to a hadron jet). he charged tracks(article jet) and the neutral energy of those hadrons will then be detected in the ALICE PC and EMCal (see section 2). Exerimentally, the jet energy has to be reconstructed. o do that, a simle algorithm consists in oening a cone, of a given radius R, around the jet axis and to sum the energy of all articles inside the cone. Sequential recombination algorithms, which have the advantage of collinear and infrared safety, canalsobeused. hek (forbackgroundclusters)andanti-k (forsignaljets)algorithms[8,9] from the FastJet ackage (with.2 R.4) were used in the following analyses. Analyseswithchargedjetsusingonlytrackinginformation(trackswith > 5MeV/c)and fully reconstructed jets including EMCal information (cluster energy with E > 5 MeV/c) have been erformed. Jet-by-jet, to avoid double counting, we correct for the energy contribution from charged articles to the energy measured with EMCal (hadronic correction). he contribution from the underlying event is also subtracted (see section 5.2). Background fluctuations have a large imact on the measured jet sectrum in Pb-Pb. A resonse matrix RM δ containing the smearing due to background fluctuations is constructed from the measured distribution. he detector effects (efficiency, detector resolution) are corrected using a second resonse matrix RM det. he two resonse matrices are combined to obtain the resonse matrix which will be needed for correcting the data using the unfolding rocedure in order to obtain the results that will follow. 5.. Inclusive differential jet cross-section in collisions dη (mb c/gev) σ/d 2 d NLO/data NLO/data anti-k, R =.4, η <.5 - ALICE s = 2.76 ev: L int = 3.6 nb Systematic uncertainty -6 NLO (N. Armesto) NLO (G. Soyez) NLO + Hadronization (G. Soyez) ,jet Figure 3. Uer anel: inclusive differential jet cross section for R =.4. he bands show the NLO QCD calculations. Lower anels: ratio of NLO QCD calculations to data. σ(r=.2)/σ(r=.4) anti-k, η <.5 ALICE s = 2.76 ev Systematic uncertainty LO (G. Soyez) NLO (G. Soyez) NLO + Hadronization (G. Soyez) ,jet Figure 4. Ratio of inclusive differential jet cross sections for R =.2 and R =.4, with QCD calculations. As a baseline (and as a reference for Pb-Pb analyses), in collisions at s NN = 2.76 ev, the inclusive differential jet cross-section was measured [2] at mid-raidity (with an integrated luminosity of 3.6 nb ). Jets are measured over the transverse momentum range 2 to 25 GeV/c and are corrected using a bin-by-bin technique. Calculations based on Next-to-Leading Order erturbative QCD and PYHIA 8 are in good agreement with the measurements (see Fig. 3). he ratio of inclusive jet cross-sections for jet radii R =.2 and R =.4 (see Fig. 4) is well reroduced by a Next-to-Leading Order erturbative QCD calculation when hadronization effects are included. 4

6 29th Winter Worksho on Nuclear Dynamics (WWND23) Journal of Physics: Conference Series 458 (23) 225 doi:.88/ /458// Background estimation and fluctuations in Pb-Pb One of the main exerimental difficulties of measuring jets in heavy-ion collisions is the estimation and the subtraction of the fluctuating background (contribution from the underlying event). In ALICE, the background energy density is estimated by clustering the whole event with the k -algorithm and calculating the density ρ = median( jet,i ) (where A is the area of a given jet A jet,i i) for every jet excet the two leading ones as roosed in [2]. he average backgroud density is then subtracted from the of signal jets (found using anti-k ) as sub = ρ A. ρ which was estimated for charged only (using 2 data [22]) and full (charged + neutral) jets (using 2 data [23]), scales with the event multilicity [22]: ρ N < >. In addition, oint-to-oint background fluctuations δ were estimated [22] by lacing random cones in the measured events, or by embedding a known (high ) robe in the event and then looking at the collection of jets found by the anti-k algorithm and matched to the embedded robe, in this event, the corresonding δ = jet ρ A robe. he δ distribution is then fitted with a Gaussian and the width (σ) of the distribution is extracted. he corresonding σ ch and σ ch+em were estimated for a given jet radius fromr =.2toR =.4. For smaller reconstructed jets radii, the total energy inside the cone is smaller and consequently, the corresonding background fluctuations are also reduced. σ ch GeV/c was measured for R =.4 comared to σ ch 4.5 GeV/c for R =.2. When neutral articles are included, the jet energy resolution increases, but the background fluctuations also become larger: σ ch+em > σ ch. For instance, for R =.3, σ ch+em 9 GeV/c and σ ch 7 GeV/c Jet sectrum in Pb-Pb at s NN 2.76 ev wo analyses have been erformed: he first one [24] uses data from 2 and is only using jets from charged articles. For this analysis jet radii of R =.2 and.3 were used. he minimum of the jet constituents is const =.5 GeV/c. All jets with a jet axis within η <.5 were considered. In order to reduce the number of combinatorial background jets, which hels to stabilize the unfolding rocedure, the transverse momentum of the leading track of a given jet was required to be leading > 5 GeV/c. he effect of this selection on the corrected sectra was found to be small in the case of this analysis. o quantify it, the inclusive sectrum was comared to the ones obtained using leading > 5 GeV/c and leading > GeV/c. he corrected sectra (using R =.3 jets) is shown in Fig. 5 (left anel). A centrality evolution (using 4 centrality classes from - to 5-8%) of the yield of jets, is observed. In addition, a ratio of the sectrum using R =.2 devided by the sectrum using R =.3 jets shows that there is no signifiquant jet broadening (from R =.2 to R =.3). A second analysis was made [25], following the same strategy but using fully reconstructed (including both charged and neutral articles) R =.2 jets from 2 Pb-Pb data. his full jet sectrum (shown Fig. 6 on the left anel) was obtained using the % most central events. As for the charged jet analysis (see above), the effect of the leading constituent track transverse momentum cut was studied using both leading > 5 GeV/c and leading > GeV/c. 6. Jet nuclear modification factors Using the charged jet sectrum measured in 4 centrality classes [24] from 2 Pb-Pb data, a jet R jet CP was obtained in 3 centrality classes (from % to 3 5%), using the sectrum measured in 5 8% centrality as a reference (see Fig. 5 central anel). A strong jet suression, similar to the hadron R AA, is observed for central events. For more eriheral events the suression decreases. his imlies that the full jet energy is not catured by jets with R =.2 and R =.3 in Pb-Pb collisions. 5

7 2 29th Winter Worksho on Nuclear Dynamics (WWND23) Journal of Physics: Conference Series 458 (23) 225 doi:.88/ /458//225 If we comare those measurements to other measurements from LHC (see Fig. 5 right anel), the jet R jet CP from ALICE, CMS and ALAS [26] are consistent within systematic errors. Using the full jet sectra [25] from 2 data, combined with the differential cross-section measured in collisions (see section 5.), a full jet nuclear modification factor R jet AA was built. A strong suression is observed (see Fig. 6 central anel) for R =.2 jets (in the % centrality class). At low jet this suression decreases with. At high the jet R AA = hese results are consistent with the CMS R =.2 results from the 5% most central events (see Fig. 6 right anel) [27], desite the differences in the detector configurations as well as in jet finding and background subtraction rocedures. As exlained in section 2, in ALICE, jets can be found using tracking and electromagnetic calorimetry, whereas CMS uses article-flow jets, which combine information from three detectors (tracker, electromagnetic and hadronic calorimeters). In order to fully quantify the measured jet quenching, additional measurements are ongoing in order to study the radius and the centrality deendence of R jet AA. - /dη N/d d /N coll /N evt Pb-Pb s NN =2.76 ev Centrality -% -3% 3-5% 5-8% Charged Jets Anti-k R =.3 track >.5 GeV/c R CP (-%)/(5-8%) (-3%)/(5-8%) (3-5%)/(5-8%) Charged Jets Anti-k R =.3 track >.5 GeV/c Pb-Pb s NN =2.76 ev correlated uncertainty shae uncertainty Anti-k algorithm R=.3 CMS Preliminary -%/5-9% Particle Flow Jet ALICE Preliminary -%/5-8% rack Jet ALAS -%/6-8% Calo Jet -9 - correlated uncertainty shae uncertainty - ALI PREL charged charged ALI PREL 6476 Figure 5. Left: Corrected jet sectrum and Rc jet (central anel) with charged tracks for jet radius R =.3. Right: Jet R jet CP from ALICE comared to CMS and ALAS results. ALI-PREL-4422 ALI-PREL-4426 Figure 6. Corrected sectrum (left anel) and R jet AA (central anel) of fully reconstructed jets (reconstructed using R =.2 and requiring a high leading track > 5 GeV/c) in the % centrality bin. Right: ALICE R jet AA comared to CMS R =.2 results from the 5% centrality bin. ALI-DER Hadron-jet correlations Both di-hadron and hadron-jet correlations are sensitive to the di-jet structure exected from hard scattering. Modelcalculationsshowthatahigh hadrongerinducesageometricalbias[28], towards jets generated close to the surface of the fireball. he jet oulation recoiling from such a ger is biased towards larger in-medium ath length. 6

8 29th Winter Worksho on Nuclear Dynamics (WWND23) Journal of Physics: Conference Series 458 (23) 225 doi:.88/ /458//225 - ch dn d N,jet /8/3 : -5 GeV : 5-2 GeV : 2-5 GeV Pythia / recoil Pb-Pb = recoil Pythia IAA h jet ϕ -ϕ -π <.6 recoil Signal ger hadron: 2 < < 5 GeV/c Reference ger hadron: 5 < < 2 GeV/c -4-5 PbPb anti-k = 2.76 ev -2% s NN const R =.4 A >.4 >.5 GeV ch,jet = reco -ρ A (G ev /c) Pb-Pb -2% = 2.76 ev s NN const anti-k, R=.4, >.5 GeV/c Diagonal element of covariance matrix Shae uncertainty Correlated uncertainty ch,jet Figure 7. Left: Recoil jet distributions in 2% central Pb-Pb collisions for R =.4 and const >.5 GeV/c. Right: IAA PYHIA. o exloit that effect, a jet-hadron measurement was made [29], using 2 Pb-Pb data, based on the semi-inclusive distribution of reconstructed charged article jets (using anti-k with R =.2 and.4) recoiling from a high ger hadron. he distribution of recoil jets is measured by counting the number of jets in the event within φ() φ(jet) < π.6 and normalized to the corresonding number of gers. Hadron gers are selected with > GeV/c in order to select a single hard rocess in the collision. he recoil jet distribution (measured in 2% central Pb-Pb collisions), lotted as function of the jet sub = reco ρ A, is shown Fig. 7 (left anel). For jet < 2 GeV/c the shae of the distribution is identical for all choices of ger. his region corresonds to combinatorial background jets. For jet > 2 GeV/c a clear evolution with the ger can be seen. his region, dominated by high Q 2 events, corresonds to the signal art of the recoil jet sectrum which deends strongly on the ger (as a consequence of the ger bias effect: the of selected artons increases while increasing the ger ). o get rid of the combinatorial jets (in a urely data driven way [3]), the difference (called recoil ) of two measured jet distributions (with hadron gers in different intervals) is used: with recoil ( ch,jet) = Y( ch,jet; min, max ) Y( ch,jet; min,ref, max,ref) () Y( ch,jet; min, max ) = N tr dn( ch,jet ; min,max ) d ch,jet In addition, this method does not imose any bias on the fragmentation of the recoil jets. he resulting recoil distribution (which is free of combinatorial jets) is then unfolded using both χ 2 minimization and the Bayes theorem. he difference between the two methods contributes to the anti-correlated shae uncertainty shown in Fig. 7 (right anel). he ratio I AA = Pb Pb recoil / PYHIA recoil of the measured recoil distribution over the same distribution from PYHIA is shown Fig. 7 (right anel). his measurement was made for R =.2 and R =.4 jets with const >.5 GeV/c and > 2 GeV/c. Overall, I AA.6.7 with no visible broadening of recoil jets from R =.2 to R =.4. In addition, the comarison of these distributions (relative to PYHIA) does not indicate a large energy redistribution towards lower constituents. const ALICE has also measured [3] the hadron I AA = Y PbPb /Y (where Y is the yield measured as the integral of a given φ correlation eak) from di-hadrons azimuthal correlations. On the away side a suression by a factor two is seen (redistribution of the energy of the high (2) 7

9 29th Winter Worksho on Nuclear Dynamics (WWND23) Journal of Physics: Conference Series 458 (23) 225 doi:.88/ /458//225 article in the medium) which can be comared to I jet AA from hadron-jet measurements: I jet AA hadron I AA. We can also note that on the near side a 2% yield enhancement was measured (effect of fragmentation after energy loss). 8. Summary he inclusive differential jet cross-section measured in collisions is in good agreement with NLO QCD calculations and rovides an imortant reference for Pb-Pb studies. In heavy-ion collisions, the background contribution and its fluctuations have been studied in details. Jets are strongly suressed in Pb-Pb collisions. his suggests significant out-of-cone radiation. Overall, hadron and jet observables tend to be in agreement (R jet AA hadron R AA and I jet AA hadron I AA ). Recoil jets are found to be significantly less suressed than inclusive jets (a similar behavior exists for single article i.e R AA vs I AA ). A qualitative agreement between exeriments exists. However, the measurements are not directly comarable as they use different jet reconstruction techniques, background estimation, etc. An ale to ale comarison would hel to build a coherent icture of jet quenching at LHC. Comaring jet measurements to theory redictions is essential to learn about the medium roerties but it is also a very challenging task, which requires close collaboration between theorists and exerimentalists in order to find good observables or even to define a common framework. In the near future, more differential measurements using larger cone radii as well as the recent -A run at LHC will rovide some additional constraints. References [] Wiedemann U A 29 (Prerint ) [2] Aamodt K et al. (ALICE Collaboration) 28 JINS 3 S82 [3] Abeysekara U et al. (ALICE EMCal Collaboration) 2 (Prerint 8.43) [4] Alver B, Baker M, Loizides C and Steinberg P 28 (Prerint 85.44) [5] Abelev B et al. (ALICE Collaboration) 22 (Prerint 2.452) [6] Cronin J, Frisch H J, Shochet M, Boymond J, Mermod R et al. 975 Phys.Rev. D 35 [7] Accardi A 22 (Prerint he-h/2248) [8] Adler S et al. (PHENIX Collaboration) 23 Phys.Rev.Lett (Prerint nucl-ex/362) [9] Adams J et al. (SAR Collaboration) 23 Phys.Rev.Lett (Prerint nucl-ex/3624) [] Arsene I et al. (BRAHMS Collaboration) 23 Phys.Rev.Lett (Prerint nucl-ex/373) [] Back B et al. (PHOBOS Collaboration) 24 Phys.Rev. C 7 69 (Prerint nucl-ex/467) [2] Abelev B et al. (ALICE Collaboration) 23 Phys.Lett. B (Prerint 28.27) [3] Chatrchyan S et al. (CMS Collaboration) 22 Eur.Phys.J. C (Prerint ) [4] Horowitz W and Gyulassy M 2 J.Phys. G (Prerint 7.236) [5] Buzzatti A and Gyulassy M 22 Nucl.Phys.A (Prerint 2.647) [6] Armesto N, Cole B, Gale C, Horowitz W A, Jacobs P et al. 22 Phys.Rev. C (Prerint 6.6) [7] Aamodt K et al. (ALICE Collaboration) 22 Phys.Lett. B (Prerint 9.25) [8] Cacciari M, Salam G P and Soyez G 28 JHEP (Prerint 82.89) [9] Cacciari M and Salam G P 26 Phys.Lett. B (Prerint he-h/522) [2] Abelev B et al. (ALICE Collaboration) 23 (Prerint ) [2] Cacciari M and Salam G P 28 Phys.Lett. B (Prerint ) [22] Abelev B et al. (ALICE Collaboration) 22 JHEP (Prerint ) [23] Aiola S (Collaboration for the ALICE) 23 (Prerint ) [24] Verweij M (ALICE Collaboration) 22 (Prerint ) [25] Reed R (ALICE collaboration) 23 (Prerint ) [26] ALAS-CONF-2-75 (ALAS Collaboration) 2 [27] CMS-PAS-HIN-2- (CMS Collaboration) 22 [28] Renk 22 Phys.Rev. C (Prerint ) [29] Cunqueiro L (ALICE Collaboration) 22 (Prerint 2.76) [3] de Barros G, Fenton-Olsen B, Jacobs P and Ploskon M 22 (Prerint 28.58) [3] Aamodt K et al. (ALICE Collaboration) 22 Phys.Rev.Lett (Prerint.2) 8