Introduction Analysis strategy Results Additional Higgs bosons ? Conclusion
Observation and properties of a Higgs boson andsearch for additional Higgs bosons in theTwo-Photon Decay Channel with CMS
Benoit Courbon
IPN Lyon/Universite Claude Bernard Lyon 1, on behalf of the CMS collaboration
Higgs Hunting, 30/07/2015
B.Courbon CMS H→ γγ 1 / 24
Introduction Analysis strategy Results Additional Higgs bosons ? Conclusion
Overview of the Standard Model H → γγ analysis
B.Courbon CMS H→ γγ 2 / 24
Diphoton channel : Small branching fraction(∼ 2.10−3 at mH = 125 GeV) but cleansignature : 2 high-pT isolated photons
Analysis strategy : Search for a narrow peak on asmoothly decreasing background in the diphotonmass spectra
Large diphoton background : Reducible (jet-jetor γ-jet with jet faking photon) and irreducible(γ-γ)
Mass resolution is crucial (energy correctionsand vertex identification)
Categorization of diphoton events to gain insensitivity and to tag production modes
We use the full run 1 dataset : 5.1 fb−1 (2011, 7 TeV) + 19.6 fb−1 (2012, 8TeV)
Trigger : pT ,γ > 26(18) or pT ,γ > 36(22) on the leading (trailing) photon +loose isolation, shower shape criteria and mass cut → 99.5 % efficiency
Challenging Pile-Up conditions : <PU>=21, up to 40
Eur. Phys. J. C 74(2014) 3076 ;Hgg Public TWiki
Introduction Analysis strategy Results Additional Higgs bosons ? Conclusion
Mass resolution
B.Courbon CMS H→ γγ 3 / 24
mγγ =√
2E1E2 cosα1,2
Energy measurment :
ECAL performance (crystal intercalibration,response changes correction...)
High-level correction : Photon energyregression method. Multivariate approachaiming to take in account showering,gap/crack effects, PU ...
Final energy scale extraction from Z → eeevents (cross-check with Z → µµγ)
Vertex identification :
The angular term is negligible if δz < 1 cm
CMS has no intrisic pointing capability
We use Boosted Decision Trees to identify the primary vertex, based on thekinematics of the recoiling tracks + the tracks of identified conversions
More than 80% average vertexing efficiency
Energy resolution vs η
arXiv :1502.02702
Introduction Analysis strategy Results Additional Higgs bosons ? Conclusion
Selection and categorization
B.Courbon CMS H→ γγ 4 / 24
We apply a first BDT classifier (photon ID)to discrimate prompt photons from fakes(neutral hadrons), based on shower shape andisolation
Then we categorize diphoton events :We build exclusive classes (1% of theevents) in order to tag specificproduction modes, by requiringadditional objects (forward jets,leptons, MET...) in the event.We apply another classifier (diphotonBDT) to categorize the remainingdiphoton events (99%) based on theirmass resolution, their kinematics andthe photon ID.5 (4) untagged categories + 9 (7)exclusive categories for 2012 (2011)dataset
Introduction Analysis strategy Results Additional Higgs bosons ? Conclusion
Signal and background modelling
B.Courbon CMS H→ γγ 5 / 24
Signal model :We fit MC signal samples with a sumof gaussians, with granularity of 5GeV, for each processThen the signal model is interpolatedbetween these points.
Background model :It is data driven : We choose thefunctional form with the best fit todata, among a class of differentfunctions (exponentials, power laws,polynomials...).The choice of the bkg model istreated as a discrete nuisanceparameter.We make sure that the potential biason the signal strength is negligible.
Signal model at mH = 125 GeV
Introduction Analysis strategy Results Additional Higgs bosons ? Conclusion
Results
B.Courbon CMS H→ γγ 6 / 24
Local significance of 5.7 σ
mH = 124.70± 0.34[±0.31(stat)± 0.15(syst)] GeV
µ(mH = 124.7GeV ) = 1.14+0.26−0.23
[±0.21(stat)+0.09−0.05(syst)+0.13
−0.09(theo)]
Introduction Analysis strategy Results Additional Higgs bosons ? Conclusion
Low-mass search
B.Courbon CMS H→ γγ 7 / 24
CMS-HIG-14-037
Light scalar searches in the diphoton channel areperformed in the range [80,110] GeVMotivations of this search : NMSSM, 2HDM ...
The Z peak contamination, with 2 electronsmisidentified as photons, needs to be added to thediphoton continuum background
4 categories (using the diphoton classifier fromstandard H→ γγ analysis)
No evidence for new particle production has beenobserved
(GeV)Hm80 85 90 95 100 105 110
(pb)
95%
CL
)γ γ →
Br(
H× σ
0
0.05
0.1
0.15
0.2
0.25Observed
σ 1±Expected
σ 2±Expected
CMS Preliminary -1=8 TeV L=19.7fbs
(GeV)γγm75 80 85 90 95 100 105 110 115 120
Eve
nts
/ 1.0
0GeV
0
100
200
300
400
500
600
700
800
900 DataBkg Fit
σ1±σ2±
-1 = 8 TeV, L = 19.7 fbsCMS Preliminary
Class 0
(GeV)Hm80 85 90 95 100 105 110
Loca
l p-v
alue
-310
-210
-110
σ1
σ2
σ3
Expected for SM-like Higgs Boson
8TeV Observed
CMS Preliminary -1=8 TeV L=19.7fbs
Introduction Analysis strategy Results Additional Higgs bosons ? Conclusion
High-mass search
B.Courbon CMS H→ γγ 8 / 24
arXiv :1506.02301
Heavy scalar searches in the diphoton channel are performed in the range[150,850] GeVThese searches are motivated by several BSM scenarios, such as 2HDM
Not only mass scan but also width scan (from 0.1 GeV to 10% of the scalar mass)
4 categories (barrel/endcap, converted/unconverted)
No evidence for new particle production has been observed
Γ = 0.1GeV Γ = 10%mX
Introduction Analysis strategy Results Additional Higgs bosons ? Conclusion
Conclusion
H→ γγ is one of the most sensitive channels and allows precise measurementof the Higgs sector :
Observation with 5.7σ significance (single-channel discovery)mH = 124.7± 0.34 GeVµ = 1.14± 0.24
All the measured properties are compatible with the SM Higgs bosonAdditional Higgs bosons searches, motivated by BSM scenarios, have beenperformed at low and high mass, showing no evidence of a new scalar
B.Courbon CMS H→ γγ 9 / 24
Introduction Analysis strategy Results Additional Higgs bosons ? Conclusion
BACK-UP
B.Courbon CMS H→ γγ 10 / 24
Introduction Analysis strategy Results Additional Higgs bosons ? Conclusion
H → γγ vertexing efficiency
B.Courbon CMS H→ γγ 11 / 24
Introduction Analysis strategy Results Additional Higgs bosons ? Conclusion
H → γγ categories
B.Courbon CMS H→ γγ 12 / 24
Introduction Analysis strategy Results Additional Higgs bosons ? Conclusion
H → γγ significance plot
B.Courbon CMS H→ γγ 13 / 24
Introduction Analysis strategy Results Additional Higgs bosons ? Conclusion
H → γγ mass measurment
B.Courbon CMS H→ γγ 14 / 24
Introduction Analysis strategy Results Additional Higgs bosons ? Conclusion
H → γγ reduced couplings
B.Courbon CMS H→ γγ 15 / 24
Introduction Analysis strategy Results Additional Higgs bosons ? Conclusion
H → γγ width
B.Courbon CMS H→ γγ 16 / 24
Introduction Analysis strategy Results Additional Higgs bosons ? Conclusion
H → γγ spin
B.Courbon CMS H→ γγ 17 / 24
Introduction Analysis strategy Results Additional Higgs bosons ? Conclusion
Low-mass Data/MC comparison
(GeV)γγm80 100 120 140 160 180
Eve
nts
/ 2 G
eV
0
5000
10000
15000
20000
25000
30000
35000
40000 Data
γ2 prompt
γ 1 fake γ1 prompt
γ2 fake
Drell-Yan
5× (95 GeV) γγ→H
-1 = 8 TeV L = 19.7 fbsCMS preliminary
B.Courbon CMS H→ γγ 18 / 24
Introduction Analysis strategy Results Additional Higgs bosons ? Conclusion
Low-mass exclusion limits (ggh+tth)
(GeV)Hm80 85 90 95 100 105 110
(pb)
95%
CL
)γ γ →
Br(
H× σ
0
0.05
0.1
0.15
0.2
0.25Observed
σ 1±Expected
σ 2±Expected
CMS Preliminary -1=8 TeV L=19.7fbs
B.Courbon CMS H→ γγ 19 / 24
Introduction Analysis strategy Results Additional Higgs bosons ? Conclusion
Low-mass significance plot (ggh+tth)
(GeV)Hm80 85 90 95 100 105 110
Loca
l p-v
alue
-310
-210
-110
σ1
σ2
σ3
Expected for SM-like Higgs Boson
8TeV Observed
CMS Preliminary -1=8 TeV L=19.7fbs
B.Courbon CMS H→ γγ 20 / 24
Introduction Analysis strategy Results Additional Higgs bosons ? Conclusion
Low-mass exclusion limits (vbf+vh)
(GeV)Hm80 85 90 95 100 105 110
(pb)
95%
CL
)γ γ →
Br(
H× σ
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14 Observed
σ 1±Expected
σ 2±Expected
CMS Preliminary -1=8 TeV L=19.7fbs
B.Courbon CMS H→ γγ 21 / 24
Introduction Analysis strategy Results Additional Higgs bosons ? Conclusion
Low-mass significance plot (vbf+vh)
(GeV)Hm80 85 90 95 100 105 110
Loca
l p-v
alue
-310
-210
-110
σ1
σ2
σ3
Expected for SM-like Higgs Boson
8TeV Observed
CMS Preliminary -1=8 TeV L=19.7fbs
B.Courbon CMS H→ γγ 22 / 24
Introduction Analysis strategy Results Additional Higgs bosons ? Conclusion
High-mass Data/MC comparison
B.Courbon CMS H→ γγ 23 / 24
Introduction Analysis strategy Results Additional Higgs bosons ? Conclusion
High-mass exclusion limits in spin 2 interpretation
B.Courbon CMS H→ γγ 24 / 24
Top Related