Measurements of strange and non strange beauty production in PbPb collisions at 5.02 TeV with the CMS detector Ta-Wei Wang (MIT) on behalf of the CMS collaboration Quark Matter, Venezia, Italy 15, May, 2018
Measurements of strange and non strange beauty production in PbPb collisions at
5.02 TeV with the CMS detector
Ta-Wei Wang (MIT)on behalf of the CMS collaboration
Quark Matter, Venezia, Italy15, May, 2018
Ta-Wei Wang (MIT), Quark Matter 2018 (Venezia, Italy 14-19, May, 2018)
Flavor dependence parton energy loss
�2
• Heavy flavor are predominantly produced in the early hard scattering, timescale < QGP formation ➜ full information of QGP evolution history
• Heavy ➜ negligible thermal production and annihilation rate
• Parton mass dependence of energy loss
• Medium induced energy loss (Eloss): 1. Collisional 2. Radiative
• Kinematics: “Dead cone effect” [1]: gluon radiation is suppressed at angles < quark mass/energy
• Eloss in light quarks > Eloss in heavy quarks
• Suppression of induced radiation at low pT and the disappearance of this effect at high pT
• A measure to quantify this: nuclear modification factor (RAA)
[1] Y.L. Dokshitzer, D. E. Kharzeev, Phys. Lett. B 519 (2001) 199.
Collisional Radiative
Ta-Wei Wang (MIT), Quark Matter 2018 (Venezia, Italy 14-19, May, 2018)
Strangeness enhancement & recombination mechanism
�3
• Proposed an enhancement of strangeness in the QGP state comparing to hadron gas [1]
• Verified by RHIC in strange baryon production measurements [2]
• Suggested that heavy quarks could hadronise via a recombination with other quarks in the medium in addition to fragmentation
[1] J. Rafelski et. al, Phys. Rev. Lett. 48, 1066[2] STAR Collaboration, Phys. Rev. C 77, 044908
[3] ALICE Collaboration, arXiv:1804.09083
- Beauty quark: even cleaner probe
- First time probing combination of beauty and strange
• ALICE measurement of Ds production show indications of higher Ds meson RAA w.r.t other D meson species [3]
Ta-Wei Wang (MIT), Quark Matter 2018 (Venezia, Italy 14-19, May, 2018) �4
• B quark decay kinematic:
• Comparable long decay length (exploit this)
• Precise vertexing & tracking
• High statistics dataset
• No-hadronic PID utilized
b quark ➞ B+, B0, Bs
Primary vertex
Secondary vertex
Hadronize
Decay
Charged tracks
cτ O(500) μm
QGP
Measuring b hadron
Ta-Wei Wang (MIT), Quark Matter 2018 (Venezia, Italy 14-19, May, 2018)
Measuring b hadron
�5
• Strategy 1, exclusive: full reconstruction of the B hadron decay chain• Exclusive decay channels:
• B+ ➞ J/ψ(μμ) + K (branch ratio ~ 0.06%)
• Bs ➞ J/ψ(μμ) + φ(KK) (branch ratio ~ 0.03%)
b quark ➞ B+, B0, Bs
Primary vertex
Secondary vertex
Hadronize
Decay
cτ O(500) μm
QGP
K
µ
µ
Non-prompt J/ψ
B+
Ta-Wei Wang (MIT), Quark Matter 2018 (Venezia, Italy 14-19, May, 2018) �6
• Strategy 1, exclusive: full reconstruction of the B hadron decay chain
• Pros: Access to the original B hadron kinematics Distinguish B hadrons, e.g. B+ v.s. Bs
• Cons: Low Statistic Combinatorial background
b quark ➞ B+, B0, Bs
Primary vertex
Secondary vertex
Hadronize
cτ O(500) μm
QGP
K
µ
µ
K
Non-prompt J/ψ
Φ
Bs
Decay
Measuring b hadron
Ta-Wei Wang (MIT), Quark Matter 2018 (Venezia, Italy 14-19, May, 2018) �7
• Strategy 2, inclusive: reconstruction some of the daughter resonance of b hadrons
• Inclusive decay channels:
• b hadrons ➞ J/ψ(μμ)
• b hadrons ➞ D0(πK)
b quark ➞ B+, B0, Bs
Primary vertex
Secondary vertex
Hadronize
cτ O(500) μm
QGP µ
µ
Non-prompt J/ψ
Decay
Measuring b hadron
Ta-Wei Wang (MIT), Quark Matter 2018 (Venezia, Italy 14-19, May, 2018) �8
• Strategy 2, inclusive: reconstruction some of the daughter resonance of b hadrons
• Pros: High statistics
High reconstruction efficiency
• Cons:
Convolution of b mesons and b baryons decay
b quark ➞ B+, B0, Bs
Primary vertex
Secondary vertex
Hadronize
cτ O(500) μm
QGP Decay
D0
π
K
Measuring b hadron
Ta-Wei Wang (MIT), Quark Matter 2018 (Venezia, Italy 14-19, May, 2018)
Full reconstruction: B+
�9
• Signal channel: B+ ➜ J/ψ K+
• Charged tracks are assigned a kaon mass
• Muon pair + track ➜ common vertex fitting ➜ fitting for yield extraction ➜ efficiency correction ➜ RAA
• Cut optimization is crucial to beat down the massive combinatorial background
Tracks
J/ψ + track common vertex fit
B candidates
B
µ+
Primary vertex
Secondary vertex
µ-
B decay
B meson
J/ψ candidates
MVA cut optimizationBoosted Decision Tree (BDT):
• Track kinematics (pT, rapidity…)• Vertex fitting probability• Opening angle• Decay length
Non-prompt J/ψ Muon pairs
K+
Ta-Wei Wang (MIT), Quark Matter 2018 (Venezia, Italy 14-19, May, 2018)
Full reconstruction: Bs
�10
B
µ+
Primary vertex
Secondary vertex
µ-
B decay
B mesonK+
Φ
Non-prompt J/ψ
K-
• Signal channel: Bs ➜ J/ψ Φ
• Charged tracks are assigned a kaon mass
• Muon pair + track pairs ➜ common vertex fitting ➜ fitting for yield extraction ➜ efficiency correction ➜ RAA
• Cut optimization is crucial to beat down the massive combinatorial background
Track pairs
J/ψ + Φ common vertex fit
B candidates
J/ψ candidates
MVA cut optimizationBoosted Decision Tree (BDT):
• Track kinematics (pT, rapidity…)• Vertex fitting probability• Opening angle• Decay length
Muon pairs
Φ candidates
Ta-Wei Wang (MIT), Quark Matter 2018 (Venezia, Italy 14-19, May, 2018)
Signal extraction: B+
�11
Phys. Rev. Lett. 119, 152301
Signal: Double Gaussian with same mean
Peaking BG: error function + two sided Gaussian
• Signals extracted: fit on invariant mass spectra (maximum likelihood)
• Peaking background: B hadrons other than the signal decay channel
- Examples: B+ → J/ψ + K*+ or B0 → J/ψ + K*0, K0…etc
pp PbPb
Ta-Wei Wang (MIT), Quark Matter 2018 (Venezia, Italy 14-19, May, 2018)
)2 (GeV/c(KK)φ)µµ(ψJ/m5 5.2 5.4 5.6 5.8 6
)2Ev
ents
/ (2
0 M
eV/c
020406080
100120140160180
DataFitSignalCombinatorial
CMSPreliminary
s0B
(pp 5.02 TeV)-128.0 pb
/nDOF: 65/45 = 1.42χSignificance = 12
Yield = 170
< 50 GeV/cT
7 < p|y| < 2.4
)2 (GeV/c(KK)φ)µµ(ψJ/m5 5.2 5.4 5.6 5.8 6
)2Ev
ents
/ (2
0 M
eV/c
0
5
10
15
20
25 DataFitSignalCombinatorial
CMSPreliminary
s0B
(PbPb 5.02 TeV)-1bµ351
/nDOF: 47/45 = 1.12χSignificance = 4.0
Yield = 20
< 50 GeV/cT
7 < p|y| < 2.4
Signal extraction: Bs
�12
Signal: Double Gaussian with same mean
• Narrow natural width of Φ meson ➜ no peaking background components
• First fully reconstructed Bs meson measurement in heavy-ion collision by CMS
HIN-17-008
pp PbPb
Ta-Wei Wang (MIT), Quark Matter 2018 (Venezia, Italy 14-19, May, 2018)
)2 (GeV/c-µ+µm2.6 2.7 2.8 2.9 3 3.1 3.2 3.3 3.4 3.5
)2Ev
ents
/ ( 0
.025
GeV
/c
0
200
400
600
800
1000
1200
1400 < 5.5 GeV/cµµ
T p≤4.5
| < 2.4µµ |y≤1.8
Cent. 0-100%
DataTotal fit
ψPrompt J/ from b hadronsψJ/
Background
(5.02 TeV)-1bµPbPb 368
CMS
�13
Partial reconstruction: non-prompt J/ψ
1. Muon pair fit to a common vertex ➞ J/ψ candidates
2. 2D Fit on invariant mass and decay length spectra
3. Extracted yields corrected using a data-driven approach (tag & probe)
All J/ψprompt J/ψ:
direct production + feed down (ex ψ’)
non-prompt J/ψ: from B decays (ex B
➜ J/ψ X)
• Distinguish non-prompt component from prompt component
• Utilize the fact that b hadrons has a long decay length
b
μ+
μ-
Non-prompt J/ψ
(mm)ψJ/l3− 2− 1− 0 1 2 3 4
Even
ts /
( 0.1
mm
)
1−10
1
10
210
310
410
510
610
< 5.5 GeV/cµµT
p≤4.5
| < 2.4µµ |y≤1.8
Cent. 0-100%DataTotal fit
ψPrompt J/ from b hadronsψJ/
Background
(5.02 TeV)-1bµPbPb 368
CMS
arXiv:1712.08959 Submitted to Eur. Phys. J. C
Ta-Wei Wang (MIT), Quark Matter 2018 (Venezia, Italy 14-19, May, 2018) �14
Partial reconstruction: non-prompt D
b
πK
Non-prompt D
• Distinguish non-prompt component from prompt component
• Distance of closest approach (DCA)
1. Track pair fit to a common vertex ➞ D candidates
2. Signal extraction in DCA interval ➞ DCA distribution
3. Template fit (from simulation) on DCA distribution to extract non-prompt DHIN-16-016
W. Xie’s poster
Ta-Wei Wang (MIT), Quark Matter 2018 (Venezia, Italy 14-19, May, 2018) �15
Non-prompt J/ψ & D @ 5.02 TeV
Ta-Wei Wang (MIT), Quark Matter 2018 (Venezia, Italy 14-19, May, 2018)
Non-prompt J/ψ RAA v.s. Npart and rapidity
�16
• Strong suppression of non-prompt J/ψ production
• Increased suppression with event activity in both collision energy
• No significant difference in RAA between the two collision energies
• Interplay between: 1). initial momentum spectrum. 2). increase energy loss ??
arXiv:1712.08959 (5.02 TeV), EPJC 77 (2017) 252 (2.76 TeV)
|y|0 0.4 0.8 1.2 1.6 2 2.4
AAR
0
0.2
0.4
0.6
0.8
1
1.2
1.4
= 5.02 TeVNNs
= 2.76 TeVNNs
< 50 GeV/cT
6.5 < pCent. 0-100%
from b hadronsψJ/ (5.02 TeV)-1, pp 28.0 pb-1bµPbPb 368
CMS
partN0 50 100 150 200 250 300 350 400
AAR
0
0.2
0.4
0.6
0.8
1
1.2
1.4
= 5.02 TeVNNs
= 2.76 TeVNNs
|y| < 2.4 < 50 GeV/c
T6.5 < p
from b hadronsψJ/ (5.02 TeV)-1, pp 28.0 pb-1bµPbPb 368 (<30%) / 464 (>30%)
CMS
Ta-Wei Wang (MIT), Quark Matter 2018 (Venezia, Italy 14-19, May, 2018) �17
• At 2.76 TeV, some mild indication of rapidity dependance was seem
• No y-dependence at 5 TeV, and the results are compatible with the 2.76 TeV results
|y|0 0.4 0.8 1.2 1.6 2 2.4
AAR
0
0.2
0.4
0.6
0.8
1
1.2
1.4
= 5.02 TeVNNs
= 2.76 TeVNNs
< 50 GeV/cT
6.5 < pCent. 0-100%
from b hadronsψJ/ (5.02 TeV)-1, pp 28.0 pb-1bµPbPb 368
CMS
arXiv:1712.08959 (5.02 TeV), EPJC 77 (2017) 252 (2.76 TeV)
partN0 50 100 150 200 250 300 350 400
AAR
0
0.2
0.4
0.6
0.8
1
1.2
1.4
|y| < 0.6
1.8 < |y| < 2.4
< 50 GeV/cT
6.5 < p from b hadronsψJ/
(5.02 TeV)-1, pp 28.0 pb-1bµPbPb 368 (<30%) / 464 (>30%)
CMS
Non-prompt J/ψ RAA v.s. Npart and rapidity
Ta-Wei Wang (MIT), Quark Matter 2018 (Venezia, Italy 14-19, May, 2018) �18
Non-prompt J/ψ RAA v.s. pT
• Measurement down to pT 3 GeV at forward rapidity
• RAA dependence for low pT J/ψ
(GeV/c)T
p0 5 10 15 20 25 30 35 40 45 50
AAR
0
0.2
0.4
0.6
0.8
1
1.2
1.4
= 5.02 TeVNNs
= 2.76 TeVNNs
|y| < 2.4Cent. 0-100%
from b hadronsψJ/ (5.02 TeV)-1, pp 28.0 pb-1bµPbPb 368
CMS
arXiv:1712.08959 (5.02 TeV), EPJC 77 (2017) 252 (2.76 TeV)
(GeV/c)T
p0 5 10 15 20 25 30
AAR
0
0.2
0.4
0.6
0.8
1
1.2
1.4
|y| < 0.6
1.8 < |y| < 2.4
Cent. 0-100% from b hadronsψJ/
(5.02 TeV)-1, pp 28.0 pb-1bµPbPb 368
CMS
Ta-Wei Wang (MIT), Quark Matter 2018 (Venezia, Italy 14-19, May, 2018)
(GeV/c)T
p1 10 210
AAR
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
5.02 TeV pp + PbPbCMS Preliminary
Cent. 0-100%
from b hadrons0D|y|<1
globaluncertainty
CUJETEPOS2+MC@sHQTAMUPHSD
�19
• Suppression of non-prompt D production in PbPb collisions
X. Jiechen et al., Journal of High Energy Physics 2 (2016) 169 P. B. Gossiaux et al, Nucl. Phys., A931 (2014) 581M. He et al., Physics Letters B 735 (2014) 445 – 450 T. Song et al., hys. Rev. C 92 (2015)
HIN-16-016
Non-prompt D0 RAA v.s. pT
• Compatible with theory prediction that includes both collisional and radiative energy loss (e.g., CUJET)
• The model including only collisional energy loss (PHSD), seems to predict a different behavior compared with other models and data at high-pT
W. Xie’s poster
Ta-Wei Wang (MIT), Quark Matter 2018 (Venezia, Italy 14-19, May, 2018) �20
B+ & Bs @ 5.02 TeV
Ta-Wei Wang (MIT), Quark Matter 2018 (Venezia, Italy 14-19, May, 2018) �21
• Suppression of B+ meson production in PbPb collisions
• B+ meson RAA ~ 0.3 to 0.6 with no obvious trend within uncertainty
• Compatible with theory prediction within uncertainty for pT 10-50 GeV/c
• Necessity for high pT measurement : distinguishing pQCD vs AdS/CFT base models
M. He et al., Physics Letters B 735 (2014) 445 – 450 M. Djordjevic, Phys. Rev. C 94 (Oct, 2016) 044908 X. Jiechen et al., Journal of High Energy Physics 2 (2016) 169 W. A. Horowitz, Phys. Rev. D 91 (2015) 085019P. B. Gossiaux et al, Nucl. Phys., A931 (2014) 581
Phys. Rev. Lett. 119, 152301
B+ nuclear modification factor
Ta-Wei Wang (MIT), Quark Matter 2018 (Venezia, Italy 14-19, May, 2018)
(GeV/c)T
p1 10 210
AAR
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
5.02 TeV pp + PbPbCMS Preliminary
Cent. 0-100%
from b hadrons:ψJ/
from b hadrons |y|<10D
|y|<10prompt D|<1ηcharged hadrons |
|y| < 2.4±B
1.8 < |y| < 2.4|y| < 2.4global
uncertainty
�22
HIN-16-016
Phys. Rev. Lett. 119, 152301
• Compatible results from three beauty RAA measurements (non-prompt D, B+, and non-prompt J/ψ)
• Beauty seems to separate from charm and light flavor at ~ 20 GeV
• RAA between prompt D, charged particle, B+, non-prompt J/ψ and non-prompt D merging above ~ 20 GeV
• Note:• Rapidity range• Inclusive vs exclusive• B to J/ψ or D kinematic
RAA zoo: B v.s. D v.s. light
arXiv:1708.04962
JHEP 04 (2017) 039
Ta-Wei Wang (MIT), Quark Matter 2018 (Venezia, Italy 14-19, May, 2018)
(GeV/c)T
p10 20 30 40 50
AAR
0
0.5
1
1.5
2
2.5 (PbPb) 5.02 TeV-1bµ (pp) + 351 -128 pb
|y| < 2.4
CMSPreliminary AA Rs
0BAA R+B
TAMUCUJET
�23
Bs nuclear modification factor
HIN-17-008
• Results derived in two Bs pT bins (7~15 and 15~50 GeV)
• Indication of less suppression of Bs
comparing to B+
• Substantial statistical and systematic uncertainty
⁉
M. He et al., Physics Letters B 735 (2014) 445 – 450 X. Jiechen et al., Journal of High Energy Physics 2 (2016) 169
Ta-Wei Wang (MIT), Quark Matter 2018 (Venezia, Italy 14-19, May, 2018) �24
Bs nuclear modification factor
HIN-17-008
• Bs / B+ RAA ratio
• Correlated systematic uncertainties sources (e.g., muon acceptance) are cancelled
• CMS’s capability to perform fully reconstructed Bs measurement
• Incoming LHC run ➜ more precise measurement
(GeV/c)T
p0 10 20 30 40 50 60
AA R+
/ B
AA R s0 B
0
1
2
3
4
5
6
7
(PbPb) 5.02 TeV-1bµ (pp) + 351 -128 pb
|y| < 2.4
CMSPreliminary ratioAA R+/Bs
0BTAMUCUJET
M. He et al., Physics Letters B 735 (2014) 445 – 450 X. Jiechen et al., Journal of High Energy Physics 2 (2016) 169
Ta-Wei Wang (MIT), Quark Matter 2018 (Venezia, Italy 14-19, May, 2018)
(GeV/c)T
p0 10 20 30 40 50 60
AA R+
/ B
AA R s0 B
0
1
2
3
4
5
6
7
(PbPb) 5.02 TeV-1bµ (pp) + 351 -128 pb
|y| < 2.4
CMSPreliminary ratioAA R+/Bs
0BTAMUCUJET
(GeV/c)T
p1 10 210
AAR
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
5.02 TeV pp + PbPbCMS Preliminary
Cent. 0-100%
from b hadrons:ψJ/
from b hadrons |y|<10D
|y|<10prompt D|<1ηcharged hadrons |
|y| < 2.4±B
1.8 < |y| < 2.4|y| < 2.4global
uncertainty
�25
• CMS beauty quark measurements (exclusive and inclusive), results are consistent
• Suppression for B+, non-prompt J/ψ, and non-prompt D
• Beauty seems to separate from charm and light flavor at ~ 20 GeV
• First Bs measurement in heavy ion collision (albeit with a substantial uncertainty)
• Comparison between B+ and Bs at low pT unveils information on flavor recombination.
• Future HL-LHC data with more precise measurements
Summary & outlook
- The MIT group's work was supported by US DOE-NP
Ta-Wei Wang (MIT), Quark Matter 2018 (Venezia, Italy 14-19, May, 2018) �26
Backups
Ta-Wei Wang (MIT), Quark Matter 2018 (Venezia, Italy 14-19, May, 2018) �27
Signal extraction: non-prompt D
Signal: Double Gaussian
with same mean
Combinatorial BG: 3rd order polynomial
Swapped BG: Wrong track mass
assumption
• Signal region DCA
— Sideband region DCA (X 0.5)
HIN-16-016
Ta-Wei Wang (MIT), Quark Matter 2018 (Venezia, Italy 14-19, May, 2018) �28
HIN-16-016
pp PbPb
• Two components template fit (from simulation) on data DCA distribution
• Corrected for potential difference in resolution between data and simulation
Signal extraction: non-prompt D
Ta-Wei Wang (MIT), Quark Matter 2018 (Venezia, Italy 14-19, May, 2018)
Non-prompt cross sections
�29
• Cross section measured between pT 6.5~50
GeV, |y|<2.4 (J/ψ) and 2~100 GeV, |y|<1.0(D0)
arXiv:1712.08959 HIN-16-016
(GeV/c)T
p0 5 10 15 20 25 30 35 40 45 50
)c (n
b / G
eV/
T) d
N/d
pM
B N
AA (1
/T×
Β o
r T
/dp
σ d×
Β
4−10
3−10
2−10
1−10
1
10
210
pp
PbPb, Cent. 0-100%
pp global unc. 2.3 %
3.9 %−+3.4 %PbPb global unc.
from b hadronsψJ/-µ+µ → ψJ/
|y| < 2.4
(5.02 TeV)-1, pp 28.0 pb-1bµPbPb 368
CMS
(GeV/c)T
p10 210
c)-1
(pb
GeV
T /
dpσd
110
210
310
410
510
610
710
810 CMS2
)0D+0(D→B
pp 5.02 TeV|y|<1
global sys: 8.4%
pp dataFONLL
(GeV/c)T
p10 210
c)-1
(GeV
TdN
/ dp
9−10
8−10
7−10
6−10
5−10
4−10
3−10
2−10
1−10 PbPb 5.02 TeV|y|<1
global sys: 8.3%
PbPb dataAA T×FONLL
(GeV/c)T
p0D10 210
)AAT
×FO
NLL
(da
ta
0
0.5
1
1.5
pp PbPb FONLL uncertainty
|y|0 0.4 0.8 1.2 1.6 2 2.4
(nb)
y) d
N/d
MB
NAA
(1/T
× Β
or
y/dσ
d× Β
0.2
0.4
0.6
0.8
1
1.2
pp
PbPb, Cent. 0-100%
pp global unc. 2.3 %
3.9 %−+3.4 %PbPb global unc.
from b hadronsψJ/-µ+µ → ψJ/
< 50 GeV/cT
6.5 < p
(5.02 TeV)-1, pp 28.0 pb-1bµPbPb 368
CMS
Ta-Wei Wang (MIT), Quark Matter 2018 (Venezia, Italy 14-19, May, 2018) �30
Non-prompt J/ψ fraction
• Proportion of measured J/ψ mesons coming from b-hadron decays
A. PbPb > pp ➜ indication of RAA non-prompt > prompt ??
B. Strong pT dependence, non-prompt fraction increases with pT (20% to 60%)
C. No significant dependence on rapidity is observed
arXiv:1712.08959. Submitted to Eur. Phys. J. C
(GeV/c)T
p0 5 10 15 20 25 30 35 40 45 50
frac
tion
ψN
onpr
ompt
J/
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
pp
PbPb, Cent. 0-100%
|y| < 2.4
(5.02 TeV)-1, pp 28.0 pb-1bµPbPb 368
CMS
|y|0 0.4 0.8 1.2 1.6 2 2.4
frac
tion
ψN
onpr
ompt
J/
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
pp
PbPb, Cent. 0-100%
< 50 GeV/cT
6.5 < p
(5.02 TeV)-1, pp 28.0 pb-1bµPbPb 368
CMS
Ta-Wei Wang (MIT), Quark Matter 2018 (Venezia, Italy 14-19, May, 2018) �31
(GeV/c)T
p0 5 10 15 20 25 30
AAR
0
0.2
0.4
0.6
0.8
1
1.2
1.4
Cent. 0-10%Cent. 10-30%Cent. 30-100%
|y| < 2.4 from b hadronsψJ/
(5.02 TeV)-1, pp 28.0 pb-1bµPbPb 368
CMS
partN0 50 100 150 200 250 300 350 400
AAR
0
0.2
0.4
0.6
0.8
1
1.2
1.4
< 6.5 GeV/cT
3 < p < 50 GeV/c
T6.5 < p
1.8 < |y| < 2.4 from b hadronsψJ/
(5.02 TeV)-1, pp 28.0 pb-1bµPbPb 368 (<30%) / 464 (>30%)
CMS
Differential RAA v.s. Npart and pT
arXiv:1712.08959
• Similar information, more suppressed for higher pT and in central collision
Ta-Wei Wang (MIT), Quark Matter 2018 (Venezia, Italy 14-19, May, 2018)
MVA optimization
�32
To reach high signal to background significance (low production rate of b) → Multivariate analysis (MVA) for cut value optimization
Variables employed in MVA• track kinematic: track pT and pseudorapidity• Probability of the vertex fit (χ2): B secondary vertex fitting probability• Normalized d0: normalized distance between primary vertex and B secondary vertex• Opening angle(θ): angle between B meson displacement vector and B meson momentum
Primaryvertex
Secondaryvertex
B+
pB θMaximize Figure of merit: S∕√(S+B)• S: signals from simulation (normalized to FONLL prediction)• B: background from real data (sidebands of the B mass
spectrum)• Optimization conducted independently for pp and PbPb
Ta-Wei Wang (MIT), Quark Matter 2018 (Venezia, Italy 14-19, May, 2018)
Signal extraction: Bs
�33
HIN-17-008
)2 (GeV/c(KK)φ)µµ(ψJ/m5 5.2 5.4 5.6 5.8 6
)2Ev
ents
/ (2
0 M
eV/c
0
20
40
60
80
100DataFitSignalCombinatorial
CMSPreliminary
s0B
(pp 5.02 TeV)-128.0 pb
/nDOF: 81/45 = 1.82χSignificance = 8.6
Yield = 93
< 15 GeV/cT
7 < p|y| < 2.4
)2 (GeV/c(KK)φ)µµ(ψJ/m5 5.2 5.4 5.6 5.8 6
)2Ev
ents
/ (2
0 M
eV/c
0
2
4
6
8
10
12DataFitSignalCombinatorial
CMSPreliminary
s0B
(PbPb 5.02 TeV)-1bµ351
/nDOF: 39/45 = 0.862χSignificance = 2.6
Yield = 9.0
< 15 GeV/cT
7 < p|y| < 2.4
)2 (GeV/c(KK)φ)µµ(ψJ/m5 5.2 5.4 5.6 5.8 6
)2Ev
ents
/ (2
0 M
eV/c
01020304050607080
DataFitSignalCombinatorial
CMSPreliminary
s0B
(pp 5.02 TeV)-128.0 pb
/nDOF: 44/45 = 0.972χSignificance = 7.9
Yield = 79
< 50 GeV/cT
15 < p|y| < 2.4
)2 (GeV/c(KK)φ)µµ(ψJ/m5 5.2 5.4 5.6 5.8 6
)2Ev
ents
/ (2
0 M
eV/c
02468
1012141618
DataFitSignalCombinatorial
CMSPreliminary
s0B
(PbPb 5.02 TeV)-1bµ351
/nDOF: 43/45 = 0.952χSignificance = 3.0
Yield = 11
< 50 GeV/cT
15 < p|y| < 2.4
Ta-Wei Wang (MIT), Quark Matter 2018 (Venezia, Italy 14-19, May, 2018)
0 20 40 60 80 100
b/G
eV]
µ [ TX)
/dp
+ B
→(p
p σd
3−10
2−10
1−10
1
10|<1.45)BData (13 TeV, |y|<2.1)BData (13 TeV, |y
FONLL (13 TeV)PYTHIA (13 TeV)
|<1.45)BData (7 TeV, scaled to |y|<2.1)BData (7 TeV, scaled to |y
FONLL (7 TeV)PYTHIA (7 TeV)
CMS (13 TeV)-148.1 pb
(a)
[GeV]BT
p0 20 40 60 80 100R
atio
s to
FO
NLL
0.51
1.52
�34
Compatible with pp 13 and 7 TeV results: upper edge of FONLL @ low pT
(GeV/c)T
p10 20 30 40 50
c)-1
( pb
GeV
Tdpσd
410
510
610
710CMS
(pp 5.02 TeV)-128.0 pb
|y| < 2.4
±B
Global uncert. 3.8%
DataFONLL
(GeV/c)T
p10 20 30 40 50
Dat
a / F
ON
LL
0.5
1
1.5
Phys. Lett. B 771 (2017) 435 Phys. Rev. Lett. 119, 152301
Cross section in pp
Ta-Wei Wang (MIT), Quark Matter 2018 (Venezia, Italy 14-19, May, 2018)
(GeV/c)T
p10 20 30 40 50
c)-1 (
pb G
eVT
dpσd 410
510
610
710 CMSPreliminary
(pp 5.02 TeV)-128.0 pb
|y| < 2.4
s0BData
FONLL
Global uncertainty:7.9%±pp:
(GeV/c)T
p10 20 30 40 50
Dat
a/FO
NLL
0.51
1.5
Cross section in pp
�35
Phys. Rev. Lett. 119, 152301 HIN-17-008
- Result derived in 5(3) B+(Bs) pT bins, from 7 to 50 GeV in |y| < 2.4- Consistent with FONLL predictions [1]
[1] M. Cacciari, M. Greco, P. Nason, “The pT Spectrum in Heavy-Flavour Hadroproduction”, JHEP 007, 9805 (1998)
Ta-Wei Wang (MIT), Quark Matter 2018 (Venezia, Italy 14-19, May, 2018)
(GeV/c)T
p10 20 30 40 50
c)-1 (
pb G
eVT
dpdN AAT1
310
410
510
610
710 CMSPreliminary
(PbPb) 5.02 TeV-1bµ (pp) + 351 -128 pb
|y| < 2.4
s0BData pp
Data PbPb
Global uncertainty:7.9%±pp:
8.5%−8.3, +PbPb:
Cross section in PbPb
�36
Phys. Rev. Lett. 119, 152301 HIN-17-008
Ta-Wei Wang (MIT), Quark Matter 2018 (Venezia, Italy 14-19, May, 2018) �37
Cold nuclear matter effect?
(GeV/c)T
p0 10 20 30 40 50 60
pPb
R
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
| < 1.93CM
: |yψNonprompt J/ < 1.93
CM: -2.86 < y+B
Open beauty
(5.02 TeV)-1, pp 28.0 pb-1pPb 34.6 nb
CMS
• Suppression observed in both non-prompt J/ψ and B meson in PbPb collisions• Consequence of CNM effect? e.g.
Modification of nPDF and potential
nuclear absorption
• CMS measurements of both non-prompt J/ψ[1] and B meson[2] in pPb collisions• No evidence of suppression observed
in pPb collisions• The suppression seen in PbPb is likely
a final state effect.
[1] EPJC 77 (2017) 269
[2] Phys. Rev. Lett. 116, 032301
Ta-Wei Wang (MIT), Quark Matter 2018 (Venezia, Italy 14-19, May, 2018) �38
CMy
3− 2− 1− 0 1 2
pPb
R
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
< 30 GeV/cT
: 10 < pψNonprompt J/ < 60 GeV/c
T: 10 < p+B
Open beauty
(5.02 TeV)-1, pp 28.0 pb-1pPb 34.6 nb
CMS
(GeV/c)T
p0 10 20 30 40 50 60
pPb
R
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
| < 1.93CM
: |yψNonprompt J/ < 1.93
CM: -2.86 < y+B
Open beauty
(5.02 TeV)-1, pp 28.0 pb-1pPb 34.6 nb
CMS
[1] EPJC 77 (2017) 269
[2] Phys. Rev. Lett. 116, 032301
Cold nuclear matter effect?
Ta-Wei Wang (MIT), Quark Matter 2018 (Venezia, Italy 14-19, May, 2018)
B to J/ψ kinematics difference
�39
• The difference in pT between parent B meson and J/ψ
[GeV/c]T
p0 5 10 15 20 25 30
Bin
num
ber
Tp
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
PYTHIA
T pψJ/
Tparent B p
Ta-Wei Wang (MIT), Quark Matter 2018 (Venezia, Italy 14-19, May, 2018)
Future
�40
(GeV)T
p1 10 210
AAR
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6 CMSProjection
= 5.02 TeVNNs pp + PbPb
Centrality 0-100%
Charged hadrons-1 < 50 GeV), 0.2 nbT(p
-1 > 50 GeV), 10 nbT(p-1 < 20 GeV), 0.2 nbT (p0D
-1 > 20 GeV), 10 nbT (p0D-1, 10 nb+B
-1, 10 nbψNon-prompt J/
B meson projection central value based on Djordjevic (PRC 94 (2016) 044908)
• Future prospect: HL-LHC• 25 times more statistic expected
(factor of 5 reduction in uncertainty)
• Comparison @ low pT ➜ more definite conclusion on flavor
dependence energy loss