1. Measurements of associated W/Z plus heavy flavor production in pp collisions at LHC
CMS Collaboration
IFT seminar – Physics at LHC
9th of May, 2017
Juan Pablo Fernández Ramos(CIEMAT).
CMS-PAS-SMP (

2. Introduction Measurements of σ(pp⇾Z/W+c/b)
Tests pQCD predictions. Results sensitive to hard scattering process & associated soft QCD radiation (input to theory refinement & validation of MC techniques)
Allows better understanding of proton structure.
Vector boson+HV jets tests PDF for s,c,b
Background to some SM processes and in searches
We use Z/W leptonic decays modes. Clean signatures with low level of backgrounds and sufficient amounts of data (selection efficiency ~ 10 %). 2
Intrinsic Charm-quark component inside the proton enhances σ(Z+c) @ ↑pt(Z)

3. Samples and Strategy : How are the measurements done from the experimental point of view ?

4. Reconstruction of the event
From the experimental
point of view ...

5. Reconstruction of the event
What type of collision(s) could have produced such a final state?

6. Charged particles deposit part of their energy in the material

7. LHC collisions are more like these
(multiple)

8. Samples 8 DATA: 2011 7 TeV ( 5 fb-1 ) or 2012 8 TeV ( 20 fb-1 )
MC: W/DY+jets generated w. MADGRAPH + PYTHIA
Cross section normalized to σ(pp⇾Z/W+X) at NNLO calculated with FEWZ
Primary signal contribution:g-g & q-q processes (<5% from MPI)
Main backgrounds: -
Contributions from ttbar, diboson, Z+light processes (from simulations, ttbar can be taken from data)
Clasification: according to the flavour of the gen. level jet (gen jet)
Gen jet is b-flavoured if b-hadron
c-flavoured if c-hadron and no b-hadron
(light otherwise)

9. Analysis strategy: Identify heavy hadrons in final states by taking
Z → l+l−, W+→l+ν with pT(l)>20 GeV &|η(l)|<2.1
SV-mass SMP
(Z+b)
Dilepton invariant mass: [70,110] GeV
Identify heavy hadrons in final states by taking
advantage of long life of heavy hadrons:
Displaced tracks or tracks forming a SV
Lepton in jet (charm and bottom analyses)
Exclusive decays in jet (D*+/- and D+/- in charm analyses)
pTjet >25 GeV & |jet|<2.5 9
Charm hadrons decay weakly (with decay lenghts > 100μm)
LO diagram for Z+c(c) production

10. b/c/l separation (discriminants)
SV-mass SMP
(Z+b)
Invariant mass of all charged particles attached
to the reconstructed secondary vertex
JP : likelihood estimate of prob.
of jet constituents to come from primary vertex
The larger the IP of a track the more incompatible w.r.t. PV
JP discriminant from BTV
Template modeling:
- sometimes is data driven
- others from MC but corrected with data
- or from MC 10
Z+b calibrated with data sample enriched in top (eμ) prior to signal extraction.
The calibration strategy validates the shape and vertex efficiencies in simulation.

11. Modeling from data (Selection of W+c)
For Z+c, templates from c-jets taken from W+c production.
W e plus jets with similar identification of heavy flavor
OS–SS subtraction to remove symmetric backgrounds
After OS-SS subtraction the purity in W+c of the resulting sample is > 90%
JHEP (W+c) 11

12. More template modeling from data-
For Z+c,Z+b,W+b,W+c templates from b-jets taken from tt production.
Using a sample where the two W bosons from the top(antitop) quark decay leptonically into leptons of different flavour.
Identification of heavy flavor: muon inside the jet 12

13. Cross section determination
Signal extraction & cross section
Total # of observed Z/W+c/b extracted from a ML or χ2 fit of the Z/W+c/b templates to the experimental distributions of vertex mass, JP or any other used discriminants. E.g. Z+c:
ni = Number of events in data
NiZ+c, NiZ+b = Number of Z+c, Z+b
Z+light, ttbar, Dibosons subtracted
Parameters to fit: μz+c & μz+b
μz+c & μz+b in the range
Cross section determination
Z ⇾ ee
In W+c there is no fit, Nfitted is Nsel - Nbkg 13

14. Results : Comparison with predictions

15. W+bb Primary contribution is g splitting
Physics Letters B 735: (2014) at √s = 7 TeV
Primary contribution is g splitting
W+bb production is a background in H+Z/W (H⇾ bb)
The pT of the leading jet chosen as the fit variable because of its discrimination power against top
The European Physical Journal C 77:92 (2017)
Use W ⇾ μν
μ(pT)>25 GeV & |η|<2.1, and 2 b-tagged jets (jets originated from b quarks) jet(pT)>25 GeV &|η|<2.4.
Data is in agreement with
MC and theory
ATLAS W+b differs from predictions (difference larger in the production of events with a collinear bb pair reconstructed as 1 jet )
JHEP 06 (2013) 084
topology afflicted by significant theoretical uncertainties.
7 TeV To cross-check these results, an independent study is performed with looser b-tagging criteria,
corresponding to an efficiency of 70% for selecting a jet containing b-flavored hadrons, while
the misidentification probability for light-quark and gluon jets is 1% and for c-quark jets is
11%.
SM Higgs boson production in association with an electroweak gauge boson and subsequent decay to bb
8 TeV In this
analysis, CSV b-tagged jets are required to pass a threshold with an efficiency of 40% in the signal phase space and a misidentification probability of 0.1% (1%) for light charm jets
Fit to mass of “top” ( lepton-W , b, bar “object” )

16. Z+bb Primary contribution:
JHEP 06 (2014) at √s = 7 TeV
Primary contribution:
Use Z ⇾ ll (ee or μμ)
μ(pT)>20 GeV & |η|<2.4, 76 < Mll < 106 GeV and 1 or 2 b-tagged jets jet(pT)>25 GeV &|η|<2.1 & ΔR(l, j) > 0.5
Testing two different flavour schemes for the choice of initial-state partons.
Cross sec. Measured MADGRAPH MCFM MADGRAPH
(pb) (5F) (5F) (parton level) (4F) (4F)
sZ+1b ± ±
sZ+2b ± ± ± ±
sZ+b ± ±
sZb/Zj(%) ± ±
Z+1b data favors five-flavour scheme(b massless).Z+2b favors 4FS .
~2σ data vs MCFM (parton-level) cross section difference specific to the modeling of the Z+b-jets final state.
Cross section ratio points to source of difference MCFM data : modelling of Z+b jets final state
CSV: The selection applied to the discriminating Background estimationvariable gives a b-tagging efficiency of about 50% and a misidentification probability of 0.1%for light jets and 1% for c jets

17. Z+bb
To be published in EPJC at √s = 8 TeV
1 or 2 b-tagged jets
Testing two different flavour schemes for the choice of initial-state partons.
Testing two different flavour schemes for the choice of initial-state partons.
Agreement with 5FS-based theoretical calculations.4FS pt

18. W+c W en or mn Identification of heavy flavor: μ in jet or D-hadron
The MCFM predictions for this process do not include contributions from gluon splitting into a cc pair
Data and predictions agree->validation of strange PDFs
First evidence for an asymmetry in the W++c and W−+c production.
Validation of PDF : the fitted strange quark & anitquark parton distribution functions
NNPDF2.3collNNLO set [50], which is based on high energy collider data only,and thus does not rely on the neutrino DIS charm information. In particular, it includes W andZ production data from ATLAS, CMS, and LHCb, and leads to a larger strangeness contentof the proton than that of global PDF sets. These four sets span a wide range of values for thestrange-quark PDF, and the strangeness content from other PDF analyses falls within this inter-val. NNPDF2.3 has the smallest strangeness, and NNPDF2.3collthe largest one
The MCFM predictions for this process do not include contributions from gluon splitting into a cc pair, but only contributions where the strange (or the down) quark couples to the W boson

19. c participates in the hard interaction
 (Z+c)
Z+c
c participates in the hard interaction
μ,e(pT)>25 GeV & |η|<2.1, jet(pT)>25 GeV &|η|<2.5
Cross section ratio 19

20. Conclusions
Evaluated W/Z + heavy hadron associated production, inclusive and differential (more in back-up)
Agreement with predictions from MadGraph5 and Madgraph renormalized to a FEWZ calculation.
There has been a lot of improvement in the last decades and there is more to come from both , theoretical and experimental results 20
The differential (Z + c) cross section measurement at high pZ could be slightly enhanced due to the existence of a charm component in the proton of nonperturbative origin. The most sensitive region, 60 GeV < pZ < 200 GeV, can be explained using PDFs that include only charm generated perturbatively. Future measurements with the new 13 TeV dataset at higher pZ > 200 GeV, where this enhancement is expected to be larger, should be able to exclude/confirm the existence of a intrinsic charm component inside the proton.

21. Back up

22. Differential cross sections vs pTZ(left) pTjet(right)
Intrinsic Charm-quark component inside the proton enhances σ(Z+c) @ ↑pt(Z)
Bins pTZ[GeV] : 0-30, and
Bins pTjet[GeV] : 25-40, and 22

23. Modeling for Z+c
Templates after subtraction of remaining (little) background
Agreement in general distributions (pTjet , NSV )
SV-mass W+c MC and Z+c MC agree
SV-mass W+c MC and W+c data do not agree 23

24. Systematic uncertainties
PU : uncertainties in the evaluation of the pileup
profile (assuming a different inelastic cross section )
PDFs (PDFNNPDF & PDFCT10) : reweighted DY MC
according to the new PDF's : NNPDF23_nnlo & PDFCT10
Jet Energy scale and Resolution (JES & JER) :
Change scale and resolution correction factors
by their uncertainties
Branching ratios of D⇾Xμ hadrons and fragmentation c⇾D (BRFRAG): We reweight the simulation to match the PDG on D-hadron channels. The syst comes from uncertainty on those reweighting factors For semileptonic channel the systematic comes from the difference between the BR from inclusive or exclusive individual contributions
Missing-et : Misestimations on the Missing transverse energy: modify the missing E_t by 10% of the unclustered missing E_t.
Cgluon-splitting, misestimation of the contribution to our background from gluon-splitting from charm : increase a factor 50 % the weight on events with 2c with
ΔR (jet,c) < 0.5
Bgluon-splitting, increase 50% the weight on events with 2b w. ΔR(jet,b) < 0.5
c-(b-) tagging efficiency. Repit analysis with efficiency increased by its error
Shape (semileptonic mode) : change Z+b template correction factor by its error
Lepton efficiencies :change efficiencies by their errors
Luminosity : 2.6 %

25. Preselection Preselection of Z ee and Z  plus jets Z ee: Zμμ :
Dielectron trigger thres-hold of 17 and 8 GeV
Electron Id Medium WP
pT(e)> 20 GeV, |(e)| < 2.1
Isolated: Icomb/pT <0.15
Energy scale corrected
MC corrected for differences in Electron Id and Iso efficiencies
Zμμ :
DiMuon trigger thres-hold of 17 and 8 GeV
Tight selection
pT(e)> 20 GeV, |(e)| < 2.1
Isolated: Icomb/pT <0.2
Momentum scale and resolution corrected (Rochester correction)
MC corrected for differences in Muon Trigger, Id and Iso
Jets:
Jets anti-kT, R=0.5
JEC corrected, both in data and MC
pT(jet) > 25 GeV, |(jet)| < 2.5

26. Previous measurements at hadron colliders:
- Z+c+X)/σ(ppbar-Z+b+X) and σ(ppbar-Z+c+X)/σ(ppbar-Z+jets+X), pT(jet) > 20 GeV
2) pT(D*) > 3 GeV [σ(ppbar-W+D*+X)/σ(ppbar-W+X) as well]
Evidence and σ( pp-Z+D0+X) andσ( pp-Z+D+X) in the forward region