NNuclear and Particle Physics Proceedings 00 (2020) 1–7
Nuclear andParticle PhysicsProceedings
Measurements of heavy-flavor jets with ALICE at the LHC ∗ Ashik Ikbal Sheikh (for the ALICE Collaboration)
Variable Energy Cyclotron Centre, Kolkata - 700064, IndiaHomi Bhabha National Institute, Mumbai - 400094, India
Abstract
Heavy quarks created in ultra-relativistic heavy-ion collisions are mostly produced in hard QCD processes duringthe early stages of the reaction. They interact with the hot and cold nuclear matter throughout the evolution of themedium via semi-hard and soft processes such as energy loss via gluon radiations and collisions. Nuclear modificationof heavy flavors in p-A systems provides insight into cold nuclear matter e ff ects such as (anti)shadowing and k T -broadening, and serves as a baseline for A-A studies. In addition to that the fully reconstructed heavy-flavor jetsprovide additional information on the flavor (or mass) dependence of fragmentation, color charge e ff ects as wellas insight into the contribution of late gluon splitting. In this contribution, we present the measurements of b -jetproduction in p–Pb collisions at √ s NN = .
02 TeV and c -jet production in pp, p–Pb and Pb–Pb collisions measuredby the ALICE experiment at the LHC. The measurements of the nuclear modification factors for c -jet in p–Pb andPb–Pb collisions are also presented. The experimental measurements are compared with the results from Monte Carloevent generators (PYTHIA 6, PYTHIA 8 and Herwig 7) and the NLO pQCD calculations (POWHEG + PYTHIA6).We find good agreement of the measurements with the results from Monte Carlo event generators and from NLOpQCD calculations.
Keywords:
Heavy quarks, Jets, Quark-Gluon Plasma
1. Introduction
The relativistic heavy-ion collision programs at theRelativistic Heavy Ion Collider (RHIC) at BNL and theLarge Hadron Collider (LHC) at CERN aim to producea hot and dense deconfined state of QCD matter, calledquark-gluon plasma (QGP). Many experimental resultsindicate that this new deconfined state of matter hasbeen formed during relativistic heavy-ion collisions atthe RHIC [1] and the LHC [2]. One of the features ofthis deconfined plasma is the energy loss of hard partonsleading to reduced yield of open-charm and open beauty ∗ Talk given at 22nd High Energy Physics International Conferencein Quantum Chromodynamics (QCD 19), 2 - 5 July 2019, Montpellier- FR
Email address: [email protected] (Ashik Ikbal Sheikh (forthe ALICE Collaboration)) mesons in Pb-Pb collisions as compared to the yieldin pp collisions scaled with the number of collisions,a phenomenon known as jet quenching. The proton-nucleus (p-A) collisions are essential to understand thee ff ects that take place in the cold nuclear matter (CNM),which serve as a baseline for the measurements done inA-A collisions. The influence of the CNM e ff ects canbe studied by measuring the nuclear modification factorin p-A collisions as: R pA = A d σ pA / dp T d σ pp / dp T (1)where, d σ pA / dp T and d σ pp / dp T are the p T -di ff erentialproduction cross sections of a given particle species inp-A and pp collisions, respectively, and A is the numberof nucleons in the nucleus. a r X i v : . [ nu c l - e x ] S e p Nuclear and Particle Physics Proceedings 00 (2020) 1–7 The usual measurements of heavy flavor mesons (Dand B) reconstructed either directly or via their decayelectrons (HFe) provide important information aboutthe jet quenching and collective motion of the c and b quarks within the medium. Reconstruction of jetscontaining a heavy quark also provides the informationabout the flavor dependence of the fragmentation mech-anism. The partonic energy loss is expected to be massdependent and the corresponding e ff ects should be morefor partons with low transverse momentum ( p T ).Heavy quarks (charm ( c ) and bottom ( b )) are mostlyproduced in primordial stage of the heavy-ion collisionsfrom the initial fusion of partons. Like light quarks orgluons, heavy quarks fragment into jets, called heavy-flavor jets ( c -jet and b -jet). The loss of energy in thedense medium due to heavy quarks is di ff erent fromthose due to light quarks and gluons, particularly inthe low and intermediate transverse momentum region.Therefore, the jet quenching depends on the flavor ofthe fragmenting parton as discussed in Refs. [3–5].Experimentally, the heavy quark content of a jet canbe identified by looking for the presence of heavy-flavorhadrons inside the jet. The hadrons containing heavyquarks have su ffi cient lifetimes ( ∼ − s), so theytravel some distances ( ∼ few mm ) before decaying. Theproperties of their decay vertices allow us to identifyheavy-flavor tagged jets. The CMS [6, 7] and AT-LAS [8] collaborations at the LHC have measured theheavy-flavor jet production and suppression in heavy-ion collisions.In this paper, we show the first ALICE measurementsof p T -di ff erential production cross section for b -jets inp–Pb collisions at √ s NN = .
02 TeV, and for D -mesontagged jets in pp collisions at √ s = . , √ s = .
02 TeV.The measurements of fractional jet momentum carriedby D-meson along the jet axis direction ( z ch || ) are shownfor D -meson tagged jet. The z ch || is defined as: z ch || = (cid:126) p jet .(cid:126) p D (cid:126) p jet .(cid:126) p jet (2)where, (cid:126) p jet and (cid:126) p D are the jet momentum and D-mesonmomentum respectively. We also report the measure-ments of nuclear modification factor ( R pPb , R AA ) for D-meson and HFe tagged jet in p–Pb and Pb–Pb colli-sions at √ s NN = .
02 TeV. For further details on mea-surements of D -meson tagged jets, we refer to [9].The experimental measurements are compared with re-sults from Monte Carlo event generators (PYTHIA 6,PYTHIA 8 and Herwig 7) and the NLO pQCD calcu-lations (POWHEG + PYTHIA6). The ALICE heavy fla- vor jet measurements are performed down to very low p T region, particularly for D-meson and HFe tagged jetsthe measured p T range is 5 < p chT , jet <
30 GeV / c and for b -jet the range is 10 < p chT , jet <
100 GeV / c . The excel-lent particle tracking capabilities of the ALICE detectormake the low momentum heavy flavor jet measurementpossible as discussed in Sec. 2.The paper is organized as follows: In the next sec-tion we outline very briefly the ALICE detector setupand data sample used in the measurements. In Sec. 3,we describe the analysis procedure for heavy flavor jetmeasurements. We discuss the experimental results inSec. 4. Section 5 contains the summary and conclu-sions.
2. ALICE Detector and Data Sample
The ALICE detector [10, 11] setup has excellent par-ticle identification, low- p T track reconstruction and ver-texing capabilities. The ALICE detector system is im-mersed in a longitudinal magnetic field B = . π ). The iden-tification of electron at high p T is performed with theEMCal, while TPC and TOF are used in the low p T re-gion.The measurements presented here are carried out us-ing data recorded by the ALICE detector setup. Forthe Monte Carlo (MC) simulations, we have used thePYTHIA6 [12]. The generated particles are transportedthrough the ALICE apparatus using the GEANT3 trans-port model [13]. Nuclear and Particle Physics Proceedings 00 (2020) 1–7
3. Analysis Procedure
The identifications of heavy flavor jets rely on theidea of finding heavy flavor content within the jets. Theheavy flavor hadrons within the jets decay after travelingsome distances from the primary vertex. The identifica-tion procedure uses the information on displacement ofthe decay vertex from the interaction vertex.Firstly, the jets are reconstructed from the selectedcharged tracks using the infrared and collinear safe anti- k T algorithm from the FastJet package [14]. The tracksare required to have | η | < . p T > .
15 GeV / c, at least70 associated TPC space points (out of a maximum of159), χ / nd f < b-Jet Identification The b -jet candidate is identified by means of recon-struction of a displaced secondary vertex (SV) withinthe jet. The SV is reconstructed from jet constituentsand is required to have 3 tracks. When there aremany candidates for SV, we choose the one whichis maximally displaced. Discriminating variables ex-ploit properties of beauty-hadron decays determinedby their long lifetimes and large masses. The maindiscriminating variables are: (i) S L xy = L xy /σ L xy > cut on signi f icance , where L xy is the projection of thedistance of the reconstructed SV from the primary ver-tex on the (x, y) plane and σ L xy is the resolution of L xy .(ii) σ SV < cut on S V resolution , where the SV reso-lution is calculated as, σ SV = (cid:118)(cid:117)(cid:116) (cid:88) i = d , where d i are the closest approaches of the tracks (usedto reconstruct SV) to the SV in 3D. The resolution, σ SV has a discrimination power only if a cut on S L xy is ap-plied.In Fig. 1, we display the jet flavor tagging e ffi ciency(obtained from EPOS + PYTHIA6 simulation) as a func-tion of jet p T for S L xy > σ S V < .
03 cm. Thetagging e ffi ciency is defined as the number of true fla-vor jets ( b -jet, c -jet and light flavor jet) after taggingw.r.t total number jets ( b -jet, c -jet and light flavor jet)before tagging.The primarily identified all b -jet candidates (usingdiscriminating variables S L xy and σ S V ) N allb , are cor-rected for the tagging e ffi ciency ( (cid:15) b ) and purity ( P b ) as: ) c (GeV/ ch, recoT, jet p
10 20 30 40 50 60 70 80 90 100 SV t agg i ng e ff i c i en cy − − − ALICE Preliminary = 5.02 TeV (EPOS+PYTHIA6 simulation) NN s Pb − p | < 0.5 labjet η = 0.4, | R , T k charged jets, anti < 0.03 cm SV σ > 7 xy L σ / xy L b jets c jets light flavor jets Figure 1: The b -jet, c -jet and light flavor (lf)-jet tagginge ffi ciency as a function of jet p T for S L xy > σ SV < .
03 cm. N b = N allb P b /(cid:15) b . The e ffi ciency and purity corrected b -jet samples are further corrected for the detector e ff ects(unfolding) and systematic uncertainties. c-Jet Identification The main concept of c -jet tagging is the D -mesonsidentification within the jets. The D -mesons are recon-structed via their hadronic decay channel, D → K − π + (branching ratio ∼ ± -meson candidates and theirdecay vertices are constructed from pairs of tracks withopposite charge. The reconstructed D -meson candi-dates are corrected for the reconstruction e ffi ciency. TheD -mesons selection criteria have been well-establishedby the ALICE Collaboration as discussed in Ref. [15,16]. The b-hadron feed-down corrections are furtherdone to the reconstructed D -meson candidates withinthe jets. Finally, the jets containing the D -meson iscorrected for the detector e ff ects (unfolding) and sys-tematic uncertainties to extract the D -meson tagged jetyields. More details can be seen in [17]. HFe Tagged Jet Identification
The specific energy loss dE / dx in the TPC volumeis used for electron identification over the momentumrange 0 . < p T <
12 GeV / c . However, the electrondE / dx band intersects with the hadron band below 2.5GeV / c and merges with the hadron band above 6 GeV / c .The TOF and EMCal are used in the momentum range0 . < p T < . / c and 6 < p T <
12 GeV / c re-spectively to resolve this issue. Once the electrons areidentified, the next step is to subtract the non-HFe con-tribution where the main sources are the photonic con-version and Dalitz decay of neutral mesons. Finally, the Nuclear and Particle Physics Proceedings 00 (2020) 1–7 HFe yields are obtained after performing the electronreconstruction, selection e ffi ciency correction and cor-rection for detector geometry. The details of the HFeidentification procedure can be found in [18].
4. Results and Discussions
The p T -di ff erential production cross section for b -jetswith resolution parameter R = .
4, reconstructed fromcharged particles in minimum bias p–Pb collisions at √ s NN = .
02 TeV is shown in Fig. 2. We have com-pared the measured b -jet cross section with NLO pQCDcalculations (POWHEG + PYTHIA). The measured b -jet cross-section is in agreement with the NLO pQCDcalculations within the experimental and theoretical un-certainties as seen in the ratio (data over calculation)plot in the lower panel of Fig. 2.
10 20 30 40 50 60 70 80 90 100 / G e V ) c ) ( m b η d c h T , j e t p / ( d σ d − − − − − ALICE Preliminary = 5.02 TeV NN s Pb − p = 0.4 R , T k charged b jets, anti ) c (GeV/ chT,jet p
10 20 30 40 50 60 70 80 90 100 R a t i o t o da t a Datasystematic uncertaintyPOWHEG HVQPOWHEG systematic uncertainty
Figure 2: Upper panel: The measured b -jet corss-section asa function of charged jet p T . Lower panel: The ratio of themeasured b -jet spectra with the NLO pQCD calculations(POWHEG). Figures. 3, 4 and 5 show the p T -di ff erential cross sec-tion of charm jets containing a D -meson in pp colli-sions at √ s = . , -mesons used to tag the jets have a minimum trans-verse momentum 3 GeV / c for √ s = . , / c for √ s =
13 TeV. The measurements are com-pared with NLO pQCD calculations obtained with thePOWHEG-BOX V2 framework [19–21], matched withPYTHIA 6 (Perugia-2011 tune) for the generation ofthe parton shower and of the non-perturbative aspects ofthe simulation, such as hadronization of colored partonsand generation of the underlying event. The theoreti-cal uncertainties are estimated by varying the renormal-ization and factorization scales (0 . µ ≤ µ F , R ≤ . µ with 0 . ≤ µ R /µ F ≤ . m c = . , . / c with m c , = . / c ) and theparton distribution function (central points: CT10nlo;variation: MSTW2008nlo68cl [22]). Two process im-plementations of the POWHEG framework are em-ployed: the heavy-quark [23] and the di-jet implemen-tation [24]. A good agreement is found within the theo-retical and experimental uncertainties between the mea-sured p T -di ff erential cross section and the cross sec-tion obtained with the POWHEG heavy-quark imple-mentation as shown in Figs. 3, 4 and 5. However, thePOWHEG di-jet implementation systematically overes-timates the production yield at √ s = ≈ . ALI-PREL-309045
Figure 3: Charm jet (tagged with D -meson) p T -di ff erential cross section in pp collisions at √ s = . + PYTHIA6 NLO pQCDcalculations.
In Fig 6, we show the z ch || -di ff erential cross section ofD -meson tagged jets for 5 < p chT , jet <
15 GeV / c (up-per plot) and for 15 < p chT , jet <
30 GeV / c (lower plot).The measurements are compared with simulations ob-tained with the POWHEG heavy-quark implementationand the Herwig 7 MEPP2QQ process. The simulated re-sults agree well with the experimental data. The D -mesons used to tag the jets have a minimum transversemomentum p T , D > / c for 5 < p chT , jet <
15 GeV / c and p T , D > / c for 15 < p chT , jet <
30 GeV / c . Thesekinematic cuts allow one to fully access the z ch || distribu-tion in 0 . < z ch || < . p chT , jet interval, a pronounced peak at z ch || ≈ -meson is the only constituent. Whereas, in case ofthe higher p chT , jet interval single-constituent jets are much Nuclear and Particle Physics Proceedings 00 (2020) 1–7 - - - - ] - ) c [ m b ( G e V / j e t h d c h T , j e t p d s d = 7 TeV s ALICE, pp, = 0.4 R , T k Charged Jets, Anti-| < 0.5 jet h | c > 3 GeV/ T,D p , with D POWHEG hvq + PYTHIA 6POWHEG dijet + PYTHIA 6Data ) c (GeV/ chT,jet p M C / D a t a Figure 4: Same as Fig. 3 but in pp collisions at √ s = − − − − − ) c m b ( G e V / η d T p d σ d DataSyst. unc. (data)POWHEG+PYTHIA6Syst. unc. (theory)ALICE Preliminary = 13 TeV s pp, | < 0.5 jetlab η = 0.4, | R , T k charged jets, anti c < 36 GeV/ T,D p , 2 < with D ) c (GeV/ T,ch jet p da t a / t heo r y Figure 5: Same as Fig. 3 but in pp collisions at √ s =
13 TeV. rarer and the peak at z ch || ≈ p chT , jet increases, the fragmentation becomes softer, afeature that has been observed also for inclusive jet mea-surements by ATLAS Collaboration [25].The Fig. 7 depicts the p T -di ff erential cross sectionof heavy flavor jet in pp collisions at √ s = .
02 TeV.The identification is performed by HFe where the mo-mentum range of HFe is 4 < p T , e <
18 GeV / c . Thedata are compared with POWHEG calculations and thecomparison shows the best agreement between data andPOWHEG calculations within their uncertainties.The nuclear modification factor R pPb of HFe taggedjets as a function of p T in p–Pb collisions at √ s NN = .
02 TeV is shown in Fig. 8. The R pPb is consistent withunity within the uncertainties over the entire p T range ofthe measurements. The production of heavy flavor jet is ] - ) c [ m b ( G e V / j e t h d c h T , j e t p d c h || z d s d = 7 TeV s ALICE, pp, = 0.4 R , T k Charged Jets, Anti- | < 0.5 jet h , | c < 15 GeV/ chT,jet p c > 2 GeV/ T,D p , with D POWHEG hvq + PYTHIA 6Herwig 7 MEPP2QQData ch|| z M C / D a t a - · ] - ) c [ m b ( G e V / j e t h d c h T , j e t p d c h || z d s d = 7 TeV s ALICE, pp, Charged Jets = 0.4 R , T k Anti- c < 30 GeV/ chT,jet p
15 < | < 0.5 jet h | with D c > 6 GeV/ T,D p POWHEG hvq + PYTHIA 6Herwig 7 MEPP2QQData ch|| z M C / D a t a Figure 6: The z ch || -di ff erential cross section ofCharm jet (tagged with D -meson) in pp collisionsat √ s = < p chT , jet <
15 GeV / c (up-per plot) and 15 < p chT , jet <
30 GeV / c (lower plot).The measurements are compared with the results ofPOWHEG + PYTHIA6 and Herwig 7 MEPP2QQ. thus consistent with binary collision scaling of the ref-erence spectrum for pp collisions at the same centre-of-mass energy. The measurements immediately suggestthat there is a negligible initial state e ff ect (CNM e ff ect)present in case of heavy flavor jets. The suppression ofthe heavy flavor jet yield in Pb–Pb collisions is thus thefinal state e ff ect induced by the produced hot mediumas shown in the Fig. 9.The nuclear modification factor in Pb–Pb collisionsis defined as: R AA = A d σ AA / dp T d σ pp / dp T (3)where, d σ AA / dp T and d σ pp / dp T are the p T -di ff erential Nuclear and Particle Physics Proceedings 00 (2020) 1–7
10 15 20 25 30 35 40 45 50 55 60 ) ) c ( m b ( G e V / η d T p d σ d − − − − − DataSyst. Unc. (data)POWHEG+PYTHIA8Syst. Unc. (theory)
ALICE Preliminary = 5.02 TeV NN s pp, |<0.5 jet η = 0.4, | R , T k Charged Jets, Anti |<0.6 e y , | c < 18 GeV/ T,e p e, 4 < → with c,b ) c (GeV/ T,ch jet p
10 15 20 25 30 35 40 45 50 55 60 D a t a / T heo r y Figure 7: The p T -di ff erential cross section of heavy flavorjet (tagged with HFe) in pp collisions at √ s = .
02 TeV,compared with the results of POWHEG + PYTHIA8. ) c (GeV/ T,ch jet p p P b R ALICE Preliminary = 5.02 TeV NN s p Pb, T k Charged Jets, Anti c < 18 GeV/ T,e p e, 4 < → with c,b |<0.6 e y |<0.6, | jet η = 0.3, | R |<0.6 e y |<0.5, | jet η = 0.4, | R |<0.6 e y |<0.3, | jet η = 0.6, | R Figure 8: Heavy flavor jet (tagged with HFe) nuclear modifica-tion factor ( R pPb ) in p–Pb collisions at √ s NN = .
02 TeV. production cross sections of a given particle species inA-A and pp collisions, respectively, and A is the num-ber of nucleons in the nucleus. In Fig. 9, we display the R AA of charm jet (tagged with D -meson) in Pb–Pb col-lisions at √ s NN = .
02 TeV for the centrality range 0-20%. We also show here the R pPb for D -meson taggedjets as reference. The measurements show the strongsuppression of the D -meson tagged jets in Pb–Pb col-lisions which ensures the in-medium energy loss of thecharm jets in the produced hot medium.
5. Summary
We presented the measurements of p T -di ff erentialproduction cross section of b -jet in p–Pb collisions ALI-PREL-309073
Figure 9: The nuclear modification factor( R pPb , R AA ) of charm jet (tagged with D -meson) inp–Pb and Pb–Pb collisions at √ s NN = .
02 TeV. at √ s NN = .
02 TeV and that of c -jet in pp colli-sions at √ s = . , + PYTHIA6) and we find a good agreementbetween them. Along with that we reported the z ch || -di ff erential cross section of D -meson tagged jets forthe jet momentum range, 5 < p chT , jet <
15 GeV / c and15 < p chT , jet <
30 GeV / c . The measurements agreewell with the results from POWHEG heavy-quark im-plementation and the Herwig 7 MEPP2QQ process. The z ch || -di ff erential measurement suggests softer heavy fla-vor jet fragmentation as p chT , jet increases, similar to in-clusive jet [25] measurement. The measured cross sec-tion for heavy-flavor jet production in pp collisions at √ s = .
02 TeV agrees with POWHEG calculationswithin uncertainties. Besides that, we also presented themeasurements of nuclear modification factor of heavyflavor jets in p–Pb and Pb–Pb collisions at √ s NN = . R pPb is consistent with unity implies insignif-icant initial state e ff ect (CNM e ff ect) for heavy flavorjets. The measured R AA of charm jet in Pb–Pb colli-sions is less than unity, showing a strong suppression ofcharm jet in the produced hot medium. The suppressionis caused due to the energy loss of charm quark jets inthe medium. References [1] M. Gyulassy and L. McLerran, Nucl. Phys. A750, 3063 (2005).[2] J.W Harris and B. Muller, Ann. Rev. Nucl. Part. Sci. 46, 71-107(1996).[3] J. Casalderrey-Solana and C.A. Salgado, Acta Phys. Pol. B 383731 (2007).
Nuclear and Particle Physics Proceedings 00 (2020) 1–7 (2013) P04013; Phys. Rev. Lett. (2014) 132301; Phys. Lett. B (2016) 59.[7] CMS Collaboration, Phys. Lett. B (2017) 306.[8] ATLAS Collaboration, Eur. Phys. J. C
295 (2012); ATLAS-CONF-2010-042.[9] ALICE Collaboration, arXiv:1905.02510 [nucl-ex].[10] ALICE Collaboration, JINST , (2008) S08002.[11] ALICE Collaboration, Int. J. Mod. Phys. A , (2014)1430044.[12] T. Sjstrand, S. Mrenna and P. Skands, J. High Energy Phys. (2011) 136.[13] R. Brun et. al., CERN-DD-78-2.[14] M. Cacciari, G. P. Salam and G. Soyez, Eur. Phys. J. C (2012) 1896.[15] ALICE Collaboration, J. High Energy Phys. , (2012) 128.[16] ALICE Collaboration, Phys. Rev. C , (2016) 054908.[17] ALICE Collaboration, arXiv:1905.02510 [nucl-ex].[18] ALICE Collaboration, Phys. Lett. B 754 (2016) 81.[19] P. Nason, J. High Energy Phys. , (2004) 040.[20] S. Frixione, P. Nason and C. Oleari, J. High Energy Phys. ,(2007) 070.[21] S. Alioli, P. Nason, C. Oleari and E. Re, J. High Energy Phys. , (2010) 043.[22] A. D. Martin, W. J. Stirling, R. S. Thorne, and G. Watt, Eur.Phys. J. C (2009) 189.[23] S. Frixione, G. Ridolfi, and P. Nason, J. High Energy Phys. 09(2007) 126.[24] S. Alioli, K. Hamilton, P. Nason, C. Oleari, and E. Re, J. HighEnergy Phys. (2011) 081.[25] ATLAS Collaboration, Eur. Phys. J. C71