Heavy Flavored Jet Modification at CMS

Heavy Flavored Jet Modification at CMS

Available online at www.sciencedirect.com Nuclear Physics A 956 (2016) 874–877 www.elsevier.com/locate/nuclphysa Heavy Flavored Jet Modification at C...

185KB Sizes 0 Downloads 28 Views

Available online at www.sciencedirect.com

Nuclear Physics A 956 (2016) 874–877 www.elsevier.com/locate/nuclphysa

Heavy Flavored Jet Modification at CMS Kurt Jung (for the CMS Collaboration)1 Department of Physics and Astronomy, Purdue University, 525 Northwestern Ave., West Lafayette, IN, USA

Abstract The energy loss of jets in heavy-ion collisions is expected to depend on the flavor of the fragmenting parton. Thus, measurements of jet quenching as a function of flavor place powerful constraints on the thermodynamical and transport properties of the hot and dense medium. Specifically, pPb measurements provide crucial insights into the behavior of the cold nuclear matter effect, which is required to fully understand the hot and dense medium effects on jets in PbPb collisions. In these proceedings, we present the heavy flavor jet spectra and measurements of the nuclear modification √ factors of b jets in both PbPb and pPb as a function of transverse momentum and pseudorapidity, at sNN = 2.76 and 5.02 TeV, respectively. In addition, we present the first ever measurement of charm-tagged jets in a heavy-ion environment, including cross-sections and comparisons to PYTHIA in both pPb and pp. Keywords: QGP, b jets, c jets, heavy flavor, energy-loss

1. Introduction In relativistic collisions of heavy nuclei, experiments at both the RHIC BNL and CERN LHC have observed the phenomenon known as “jet quenching” [1, 2]. This suppression of inclusive high-pT hadrons and jets is thought to be due to collisional and radiative energy loss. When observed as a function of jet parton flavor, however, the energy loss mechanisms are less clear. Due to the additional color factors for gluon emission in the heavy quark sector and the large gluon-jet contribution to an inclusive heavy flavored jet measurement, heavy flavored jets may quench in different ways than light flavored jets [3, 4]. While measurements of jet quenching in heavy collisions can quantify the jet-medium interaction strength, measurements of jet energy loss in proton-lead collisions allow for a factorization of initial-state or “cold nuclear matter” effects from the suppression effects seen in heavy collisions. Energy loss measurements as a function of collision species will allow for a visualization of the quenching power directly from the hot and dense medium, since it is expected that the QGP is not produced in proton-lead collisions. These proceedings also present the first measurement of a charm jet cross-section [5]. In conjunction with the measurements of nuclear modification factor in pPb for inclusive jets [6] and b-jets [7], a comprehensive picture of flavor-dependent energy loss can be obtained, from which mass-dependent effects may be observed. 1A

list of members of the CMS Collaboration and acknowledgements can be found at the end of this issue.

http://dx.doi.org/10.1016/j.nuclphysa.2016.04.008 0375-9474/© 2016 Elsevier B.V. All rights reserved.

K. Jung / Nuclear Physics A 956 (2016) 874–877

875

35 nb-1 (5.02 TeV)

CMS

5

10

-2.5 < η

< 1.5

104

lab

10

90 < p < 110 GeV/c T

SSV > 2.0 jets

103 102

pPb data b c udsg

|η |<2.0

5

Number of jets

Number of jets

CM

20.9 nb-1 (5.02 TeV)

CMSPreliminary

pPb data b c udsg

c tagged sample (SSVHP>1.68)

80 < p < 120 GeV/c T

4

10

103 102 10

10 0

1

2

3

4

5

Secondary vertex mass (GeV/c2)

6

1

0

1

2

3

4

5

6

Corrected secondary vertex mass (GeV/c2)

Fig. 1. An example of a template fit of secondary vertex mass (left) and corrected secondary vertex mass (right) used for b-jet and c-jet purity calculations, respectively. Distributions of light, charm, and bottom jets from simulation (blue, green, and red filled histograms, respectively) are fit to data (filled black points) after selecting jets identified by the secondary vertex tagger.

2. Tagging Methodology The procedure at CMS [8] for identifying b-jets makes use of the long lifetime of the b-jet and applies selections on displaced secondary vertices. This algorithm is known as the “simple secondary vertex” (SSV) tagger and applies a discriminator based on the three-dimensional vertex displacement divided by its uncertainty. Selections on this quantity enhance the relative b-jet fraction by roughly a factor of ten. After applying a selection on the SSV tagger, templates of secondary vertex mass are created in simulation, where the relative yields of light, charm, and bottom jets are allowed to float such that the fit to the distributions in data is maximized. An example of this fit is shown in Fig. 1 (left). From this procedure, the tagging purity is obtained. Then, tagging efficiency is found via simulation, but is cross-checked using an alternate tagging algorithm independent of secondary vertex mass [9]. The efficiency is calculated by taking the ratio of the purity of tagged jets to the purity of untagged jets, multiplied by the tagged jet fraction. This is done in each jet pT bin, after which the full pT spectra are unfolded to correct for detector resolution effects. Charm jets are defined in these proceedings as any jet containing a charm quark within the jet cone, but ignoring charm quarks that stem from the b→c cascade. In principle, charm jets can be identified from thier displacement from the primary vertex, but in practice, the cτ of these displacements is only on the order of 100 μm, so this tagging is difficult. At CMS, the presence of a silicon tracker close to the interaction point allows for the possibility to resolve features at these length scales and is crucial for this analysis. The c-tagging strategy used here is similar to that used to tag b-jets, but with some important differences. Charm jets are again tagged using displaced vertices, but a requirement of at least three associated tracks is placed on the vertex. The inclusion of the third associated track reduces the light-jet contamination to the sample by an additional factor of three while still maintaining the charm-jet and b-jet tagging efficiencies. With a reduced light-jet component, the charm jets begin to dominate the sample in small regions of kinematic phase space which is crucial to constrain the template fits used to extract the relative fractions of bottom, charm, and light jets. The second distinction is the use of the so-called “corrected” secondary vertex mass (Mcorr ), defined in Eq. (1), where M is the standard secondary vertex mass, p is the jet momentum, and θ is the angle between the secondary vertex displacement vector and the vector sum of the particles associated with the vertex.  Mcorr = M 2 + p2 sin2 θ + p sin θ (1) This tool was developed by the experiments at the LEP collider [10] and is currently used by the LHCb Collaboration [11]. Essentially, the variable corrects for any anglar discrepancy between the final state particle momenta and the displacement vector of the secondary vertex due to neutral or invisible particles.

K. Jung / Nuclear Physics A 956 (2016) 874–877

RPYTHIA pA

876

pPb 35 nb-1 (5.02 TeV)

2.5

1.2

CMS

b jet RPYTHIA, -2.5 < η pA

CM

0.8 b-jet R

1

PbPb, 150 μb-1 pp, 5.3 pb-1

0.6 0.4

0.5

pPb luminosity uncert.

pQCD: PLB 726 (2013) 251-256 gmed = 1.8

0.2

gmed = 2.0

pp reference uncert.

0 0

|η| < 2

AA

1.5

sNN = 2.76 TeV

1

PLB 726 (2013) 251-256

2

CMS 0-100%

< 1.5

50 100 150 200 250 300 350 400

jet p [GeV/c] T

gmed = 2.2

0

100

150 200 b-jet p (GeV/c)

250

T

HIA (left) and b-jet R Fig. 2. Nuclear modification factor comparison for b-jet RPYT AA (right). The RAA values shown are for 0-10% pA centrality, while the R pA values are for inclusive-centrality. Both figures show a comparison to theoretical predictions assuming a radiative and collisional energy loss model, discussed in Refs. [12] and [13].

With the presence of a decay neutrino, for example, the direction of the vector sum of all reconstructed particles in the vertex may not match the expectation based on the displacement vector of the secondary vertex. In this case, one can assume that there must be missing energy associated to the vertex, and can add additional mass-energy to correct for this effect. Templates of Mcorr are built in simulation, where, as in b-jet tagging, the relative contributions of each jet flavor are allowed to float independently such that the fit to the Mcorr distribution in data is optimized. An example template fit is shown in Fig. 1 (right). A visual comparison of the templates of secondary vertex mass and corrected secondary vertex mass shows a clear difference in b-tagged jet template shape, leading to a more robust calculation of charm-jet purity. As in the b-jet methodology, the efficiency is found via simulation and is cross-checked using a tagger independent of secondary vertex reconstruction. 3. Results and Conclusions HIA (in pPb, relative to The b-jet nuclear modification factors RAA (in PbPb, relative to pp data) and RPYT pA a pythia simulation using the Z2 tune) are shown in Fig. 2. While the PbPb results show clear suppression in central events, the pPb cross-section is consistent with that from a pythia simulation using the Z2 tune. In addition, both the pPb and PbPb results are consistent with suppression factors of inclusive jets [7, 9], indicating that mass-dependent effects are small, at least at high-pT . Both panels in the figure also show a comparison to a theoretical model, described in Refs. [12] and [13], based on primarily collisional and radiative energy loss using perturbative QCD. While the PbPb suppression is predicted very well when a stronger gluon-to-medium coupling parameter is used, the relative enhancement seen in pPb is not anticipated by the model. However, this is expected due to the fact that initial state effects are not included in the models, so the observation in data that initial state enhancement dominates any energy loss effects in pPb is compatible with the theoretical interpretation. The c-jet spectrum in pPb and pp is compared to pythia simulations at 5.02 and 2.76 TeV, respectively. After scaling the pPb results by the Glauber nuclear overlap factor, we observe relative consistency with the pythia simulations. Charm jet spectra from data and pythia are shown in Fig. 3 in pPb (left) and pp collisions (right). Fitting a constant to the data/pythia ratios gives 1.00 ± 0.19 (stat.+syst.) in pPb and 1.15 ± 0.27 (stat.+syst.) in pp. These values indicate that charm jet energy is not significantly modified in pPb or

K. Jung / Nuclear Physics A 956 (2016) 874–877

−6

CMS Preliminary pp data, -2 < η < 2

10− 5

PYTHIA Z2 (2.76 TeV)

T

1 dσ TpA dp

10

4.8 pb-1 (2.76 TeV)

10− 4

dσ dp

mb GeV/c

pPb data, -2 < ηlab < 2 PYTHIA Z2 (5.02 TeV)

−5

T

10

CMS Preliminary

mb GeV/c

35 nb-1 (5.02 TeV)

10− 4

877

−6

10 10− 7

10− 7 10

−8

10− 8 10

−9

0

100

200

300

400

c jet p [GeV/c] T

Data/PYTHIA

Data/PYTHIA

10− 9 2 1.5 1 0.5 0

2 1.5 1 0.5 0

0

50

100

150

200

250

c jet p [GeV/c] T

HIA (left) and b-jet R Fig. 3. Nuclear modification factor comparison for b-jet RPYT AA (right). The RAA values shown are for 0-10% pA centrality, while the R pA values are for inclusive-centrality. Note the similarity between both the R pA and RAA curves between the b-jet and inclusive-jet data.

pp relative to pythia within the present systematic uncertainties. In summary, we present the first measurement of charm-tagged jets in a heavy ion environment, along with studies of b-tagged jets, in proton-proton, proton-lead and lead-lead collisions. Significant modification of b-jets is observed in PbPb collisions, while Glauber scaled b-jet cross sections in data are consistent with pythia simulations in pPb collisions, both for charm and bottom jets. The proven viability of the charm tagging methodology described in these proceedings is indicative that a measurement of charm jet energy modification in PbPb may be possible using the data from upcoming LHC run periods. References [1] CMS Collaboration, Observation and studies of jet quenching in PbPb collisions at nucleon-nucleon center-of-mass energy =2.76 TeV, Phys. Rev. C 84 (2011) 024906. arXiv:1102.1957, doi:10.1103/PhysRevC.84.024906. √ [2] ATLAS Collaboration, Jet size dependence of single jet suppression in lead-lead collisions at sNN = 2.76 TeV with the ATLAS detector at the LHC, Phys. Lett. B 719 (2013) 220–241. arXiv:1208.1967, doi:10.1016/j.physletb.2013.01.024. [3] M. Djordjevic and M. Djordjevic, Heavy flavor puzzle at RHIC: analysis of the underlying effects, Phys. Rev. C 90 (2014) 034910. arXiv:1407.3670, doi:10.1103/PhysRevC.90.034910. [4] A. Buzzatti and M. Gyulassy, Jet Flavor Tomography of Quark Gluon Plasmas at RHIC and LHC, Phys. Rev. Lett. 108 (2012) 022301. arXiv:1106.3061, doi:10.1103/PhysRevLett.108.022301. [5] CMS Collaboration, Charm-tagged jet production in pPb collisions at 5.02 TeV and pp collisions at 2.76 TeV, CMS Physics Analysis Summary HIN-15-012, CMS (2015). √ [6] CMS Collaboration, Measurement of inclusive jet nuclear modification factor in pPb collisions at sNN = 5.02 TeV with CMS, CMS Physics Analysis Summary HIN-14-001, CMS (2014). √ [7] CMS Collaboration, Transverse momentum spectra of inclusive b jets in pPb collisions at sNN =5.02 TeVSubmitted to Phys. Lett. B. arXiv:1510.03373. [8] CMS Collaboration, The CMS experiment at the CERN LHC, JINST S08004. √ [9] CMS Collaboration, Evidence of b-jet quenching in pbpb collisions at sNN = 2.76 tev, Phys. Rev. Lett. 113 (2014) 132301. arXiv:1312.4198, doi:10.1103/PhysRevLett.113.132301. [10] G. J. Barker, b-quark physics with the LEP collider, Springer Tracts Mod. Phys. 236 (2010) 170. doi:10.1007/978-3-642-05279-8. [11] LHCb Collaboration, Identification of beauty and charm quark jets at LHCb, JINST 10 (2015) P06013. arXiv:1504.07670, doi:10.1088/1748-0221/10/06/P06013. [12] Jinrui Huang, Zhong-Bo Kang, and Ivan Vitev, Inclusive b-jet production in heavy ion collisions at the LHC, Phys. Lett. B 726 (2013) 251–256. arXiv:1306.0909, doi:10.1016/j.physletb.2013.08.009. [13] Rishi Sharma and Ivan Vitev, High transverse momentum quarkonium production and dissociation in heavy ion collisions, Phys. Rev. C 87 (2012) 044905. arXiv:1203.0329, doi:10.1103/PhysRevC.87.044905.