StudyoftheTensorStructureofHiggsBoson CouplingswiththeH ZZ ... · StudyoftheTensorStructureofHiggsBoson CouplingswiththeH! ZZ! 4ℓ DecayChannelwiththeATLASDetector IMPRSYoungScientistWorkshopatRingbergCastle

Study of the Tensor Structure of Higgs BosonCouplings with the H → ZZ∗ → 4ℓDecay Channel with the ATLAS DetectorIMPRS Young Scientist Workshop at Ringberg Castle

Verena WalbrechtMax Planck Institute for Physics(Werner-Heisenberg-Institut)

July 17th, 2017

Introduction Study of the Tensor Structure of Higgs Boson Couplings Summary

The Standard Model is not the End of the Story...

Open questions:– General relativity– Neutrino oscillations– Nature of dark matter and dark energy– Matter - antimatter asymmetry

Theories beyond the Standard Model:– Supersymmetry, string theory, M-theory and extra dimensions

Experimental search for physics beyond the Standard Model in pp collisions:

1. Direct searches: search for new particles (heavy resonances, SUSYparticles, dark matter ...)

2. Indirect searches: precise measurements of known particles of the SM -search for deviations from the prediction of the SM⇒ Search for anomalous Higgs boson couplings

06/12/2017 V. Walbrecht - Study of the Tensor Structure of Higgs Boson Couplings 2/14

mailto:[email protected]


The Higgs Boson in the Standard Model2012: discovery of the Higgs boson

SM prediction: scalar CP-evenparticle (JP=0+)

Conjugation scalar

Spin: J 0Charge: C +1Parity: P +1

JP 0+

SM Higgs boson Spin 0+ hypotheses preferred bythe Run I data

But: small admixtures of 0− state to 0+ are still possible (Beyond SM):

|HBSM⟩ = cos(α)|0+⟩+ sin(α)|0−⟩

CP-even

[GeV]4lm100 150 200 250

Events

/5 G

eV

0

5

10

15

20

25

1Ldt = 4.8 fb∫ = 7 TeV: s

1Ldt = 5.8 fb∫ = 8 TeV: s

4l→(*)ZZ→H

Data(*)

Background ZZ

tBackground Z+jets, t

=125 GeV)H

Signal (m

Syst.Unc.

ATLAS

Question

: Isit a

SMHigg

s boson

?




Measurement of the Higgs Boson CP Properties

1. Decay Channel for CP properties measurement:

2. Observable to study CP properties:

)1θcos(1− 0.5− 0 0.5 1

Ent

ries

/ 0.2

50

2

4

6

8

10

12

14

16

18 ATLAS

4l→ZZ*→H-1 = 7 TeV, 4.5 fbs

-1 = 8 TeV, 20.3 fbs

Data

Background ZZ*


SM+ = 0PJ- = 0PJ

Data

Background ZZ*


SM+ = 0PJ- = 0PJ

3. Approach to account for BSM contributions: effective field theory

⇒ Statistics very low: using event rates

Z∗

Z

H

`+

`−

`+

`−




Measurement of Anomalous CouplingsEffective field theory assumption:

– Physics beyond the SM appears at energy scale Λ≫ E⇒ Point-like interaction can be assumed

LV0 =

{κSM cosα

[12gHZZZµZ

µ + gHWWW+µW

−µ]

− 1

4

[κHgg cosα gHggG

aµνG

a,µν + κAgg sinα gAggGaµνG

a,µν]

− 141Λ

[κHZZ cosα ZµνZ

µν + κAZZ sinα ZµνZµν]

− 121Λ

[κHWW cosαW+

µνW−µν + κAWW sinαW+

µνW−µν

]}X0

X0

Z⇤

Z

g

g

X0

Z∗

Z

κHgg, κAgg κSM, κAZZ, κHZZ

1. non-SM coupling to gluons 2. non-SM coupling to vector bosons

SM CP-evenanomalous (BSM) CP-evenanomalous (BSM) CP-odd






LV0 =

{κSM cosα

[12gHZZZµZ

µ + gHWWW+µW

−µ]

− 1

4

[κHgg cosα gHggG

aµνG


a,µν]

− 141Λ

[κHZZ cosα ZµνZ


− 121Λ

[κHWW cosαW+


µνW−µν

]}X0

X0

Z⇤

Z

g

g

X0

Z∗

Z









LV0 =

{κSM cosα

[12gHZZZµZ

µ + gHWWW+µW

−µ]

− 1

4

[κHgg cosα gHggG

aµνG


a,µν]

− 141Λ

[κHZZ cosα ZµνZ


− 121Λ

[κHWW cosαW+


µνW−µν

]}X0

X0

Z⇤

Z

g

g

X0

Z∗

Z







Analysis Strategy

Too low statistics to use the information from angular distributions⇒ Event rates

Approach: effective field theory

Too low statistics to measure all predicted couplings simultaneously⇒ Test two different models with different assumptions:

All other couplings are set to the SM value

X0

Z⇤

Z

g

g

X0

Z∗

Z






1. Measurement of Hgg Tensor Coupling

t/b

g

g

H

W/Z

W/Z

q′

q

q′

q

H

Z∗

Z

H

`+

`−

`+

`−

κ2Hgg

κ2Agg

κ2SM

κ2SM

ggF

VBF

H → ZZ∗ → 4ℓ

Production:

Decay:σggF ∝ κ2

Agg

Dependence:

σVBF =const

Production rate information sensitive to BSM contributions

Production and decay rates are dependent on the anomalous couplings:




1. Measurement of Hgg Tensor Coupling

t/b

g

g

H

W/Z

W/Z

q′

q

q′

q

H

Z∗

Z

H

`+

`−

`+

`−

κ2Hgg

κ2Agg

κ2SM

κ2SM

ggF

VBF

H → ZZ∗ → 4ℓ

Production:

Decay:σggF ∝ κ2

Agg

Dependence:

σVBF =const



)αsin(⋅Aggκ1− 0.5− 0 0.5 1

at L

O [p

b]σ

0

5

10

15

20

25

30

35

= 13TeVsMadGraph5_aMC@NLO at LO Standalone,

=16.5 pbggF,SMσ

=4.4 pbVBF,SMσ

ggF

σVBF,BSM=

VBF




2. Measurement of the HZZ Tensor Coupling

t/b

g

g

H

W/Z

W/Z

q′

q

q′

q

H

Z∗

Z

H

`+

`−

`+

`−

κ2Hgg

Effectivecoupling

κ2SM, κ2HZZ, κ

2AZZ

κ2SM, κ2HZZ, κ

2AZZ

ggF

VBF

H → ZZ∗ → 4ℓ

Production:

Decay:σggF ∝ κ2

XZZ

Dependence:

σVBF ∝ κ4XZZ






2. Measurement of the HZZ Tensor Coupling

t/b

g

g

H

W/Z

W/Z

q′

q

q′

q

H

Z∗

Z

H

`+

`−

`+

`−

κ2Hgg

Effectivecoupling

κ2SM, κ2HZZ, κ

2AZZ

κ2SM, κ2HZZ, κ

2AZZ

ggF

VBF

H → ZZ∗ → 4ℓ

Production:

Decay:σggF ∝ κ2

XZZ

Dependence:

σVBF ∝ κ4XZZ



)α sin(⋅ AZZκ30− 20− 10− 0 10 20 30

at L

O [p

b]σ

0

5

10

15

20

25

30

35

= 13TeVsMadGraph5_aMC@NLO at LO Standalone,

=16.5 pbggF,SMσ

=4.4 pbVBF,SMσ

ggFVBF




Event CategorisationProduction mode splitting for the SM Higgs bosonfor mH =125 GeV:

0 20 40 60 80 100

ttH enriched

VH-leptonic

VH-hadronic

VBF enriched

mixed

ggF enriched

t/b

g

g

H

W/Z

W/Z

q′

q

q′

q

H

W/Z

q

q

W/Z

H

g

g

t/b

t/b

H

ggF87.4%

VBF6.8%

VH4.0%

ttH1.8%

ggF VBF WH ZH ttH

0 jet

1 jet

≥ 2 jet, mjj>120 GeV

≥ 2 jet, mjj<120 GeV

≥ 1 additional ℓ

≥ 1 nb,jet and≥ 4 njet




Signal ModellingContinuous signal model to describe signal expectation in dependence onBSM couplings (κAgg, κHZZ, κAZZ)

Predicts number of expected events at every parameter point based on adiscrete set of simulated input samples

Nout(κout) =∑Ninput

i=1 wi(κout; κi) · Ni(κi)

1. Hgg Tensor Coupling: κ = (κHgg, κAgg)

2. HZZ Tensor Coupling: κ = (κSM, κHZZ, κAZZ)

Output distribution weight Input distributionκ1

κ2

κout




Results of the Tensor Coupling MeasurementMeasuring BSM coupling parameter:Comparison of observed number of events in each category with the onepredicted by signal model

118 GeV<m4ℓ<129 GeV (for SM Signal):

Event ggF mixed VH- VBF VH- ttH-Category enriched hadronic enriched leptonic enriched

Signal 26.80± 2.50 15.67± 1.33 3.54± 0.51 06.87± 0.81 0.32± 0.02 0.39± 0.04ZZ∗ 13.70± 1.00 04.12± 0.42 0.66± 0.16 01.17± 0.32 0.05± 0.01 0.01± 0.01Z+jets,tt 02.23± 0.31 00.96± 0.42 0.02± 0.02 00.45± 0.04 0.01± 0.00 0.07± 0.04Expected 42.70± 2.70 20.85± 1.41 4.39± 0.51 08.42± 0.91 0.38± 0.02 0.47± 0.05Observed 49.00± 0.00 24.00± 0.00 3.00± 0.00 19.00± 0.00 0.00± 0.00 0.00± 0.00

Test statistic:

q = −2 lnL(κ)L(κ)

= −2ln(λ)

L(κ): is the maximum of the likelihood




Results of the Tensor Coupling MeasurementMeasuring BSM coupling parameter:Comparison of observed number of events in each category with the onepredicted by signal model

118 GeV<m4ℓ<129 GeV (for SM Signal):

Event ggF mixed VH- VBF VH- ttH-Category enriched hadronic enriched leptonic enriched

Signal 26.80± 2.50 15.67± 1.33 3.54± 0.51 06.87± 0.81 0.32± 0.02 0.39± 0.04ZZ∗ 13.70± 1.00 04.12± 0.42 0.66± 0.16 01.17± 0.32 0.05± 0.01 0.01± 0.01Z+jets,tt 02.23± 0.31 00.96± 0.42 0.02± 0.02 00.45± 0.04 0.01± 0.00 0.07± 0.04Expected 42.70± 2.70 20.85± 1.41 4.39± 0.51 08.42± 0.91 0.38± 0.02 0.47± 0.05Observed 49.00± 0.00 24.00± 0.00 3.00± 0.00 19.00± 0.00 0.00± 0.00 0.00± 0.00

Test statistic:

q = −2 lnL(κ)L(κ)

= −2ln(λ)

L(κ): is the maximum of the likelihood




Results of the Hgg Tensor Coupling Measurement

Aggκ

1− 0.5− 0 0.5 1

)λ2

ln (

0

5

10

15

20

25

68% CL

95% CL

Observed

SM expected

ATLAS Preliminary

4l→ ZZ* →H 113 TeV, 36.1 fb

= 1SMκ = 1, Hggκ

| = 0.43AggκObserved: |

= 0.00AggκExpected:

BSM couplingparameter

Allowed range at 95% confidence level (CL)Best fit expected for SM observed

κAgg ±0.43 [-0.47,0.47] [-0.68,0.68]⇒ Compatible with the SM prediction within 1.8 standard deviations

κSM · cosα=1

κHgg · cosα=1Regionsexcludedat 95% CL

Expected and observed distributions of the test statistic q for fits of κAgg:




Results of the HZZ Tensor Coupling Measurement

Avvκ

8− 6− 4− 2− 0 2 4 6 8

)λ2

ln (

0

5

10

15

20

25

30

68% CL

95% CL

Observed

SM expected

ATLAS Preliminary

4l→ ZZ* →H 113 TeV, 36.1 fb

= 1SMκ = 1, Hggκ

| = 2.9AvvκObserved: |

= 0.0AvvκExpected:

Hvvκ

6− 4− 2− 0 2 4 6

)λ2

ln (

0

5

10

15

20

25

30

68% CL

95% CL

Observed

SM expected

ATLAS Preliminary

4l→ ZZ* →H 113 TeV, 36.1 fb

= 1SMκ = 1, Hggκ

= 2.9HvvκObserved:

= 0.0HvvκExpected:

BSM couplingparameter

Allowed range at 95% confidence level (CL)Best fit expected for SM observed

κAZZ ±2.9 [-3.5,3.5] [-5.2,5.2]κHZZ 2.9 [-2.9,3.2] [0.8,4.5]

⇒ Compatible with the SM prediction within 1.4 and 2.3 standard deviations

Expected and observed distributions of the test statistic q for fits of...... κAZZ (κHZZ=0) ... κHZZ (κAZZ=0) κSM · cosα=1

Regions

excluded

at 95% CL




Summary

Measurement of the anomalous couplings :

– More events observed in each category then expected

– Too low statistic to observe a significant deviation from the SM predication

– Excluded regions at 95% confidence level:

κAgg < -0.68 and κAgg > 0.68κAZZ < -5.2 and κAZZ > 5.2κHZZ < 0.8 and κHZZ > 4.5

Outlook: Combine event yield and angular distribution information




BACKUP

06/12/2017 V. Walbrecht - Study of the Tensor Structure of Higgs Boson Couplings


Full Categorization

ttH-like

118 GeV < m4l < 129 GeV

≥ 1 extra leptons

≥ 2 Jets

= 0 Jet

= 1 Jet

mJJ > 120 GeV

mJJ < 120 GeV

60 GeV < pT 4l < 120 GeV

pT 4l > 120 GeV

pT 4l < 60 GeV

ggF-enrichedVBF-enrichedVH-enrichedttH-enriched

PT 4l > 150 GeV

PT 4l < 150 GeV

PT j1 > 200 GeV

PT j1 < 200 GeV

06/12/2017 V. Walbrecht - Study of the Tensor Structure of Higgs Boson Couplings B/U 1


κSM free

Avvκ8− 6− 4− 2− 0 2 4 6 8

)λ-2

ln (

0

2

4

6

8

10

12

14

16

18

20

68% CL

95% CL

Observed

SM expected

ATLAS Internal 4l→ ZZ* →H

-113 TeV, 36.1 fb

freeSMκ = 1, Hggκ

= 1.2SMκ| = 1.5, AvvκObserved: | = 1.0SMκ = 0.0, AvvκExpected:

Hvvκ6− 4− 2− 0 2 4 6

)λ-2

ln (

0

5

10

15

20

25

30

68% CL

95% CL

Observed

SM expected


-113 TeV, 36.1 fb

freeSMκ = 1, Hggκ

= 1.2SMκ = 2.2, HvvκObserved: = 1.0SMκ = 0.0, HvvκExpected:



2D scans

Avvκ6− 4− 2− 0 2 4 6

SMκ

0.6

0.8

1

1.2

1.4

1.6

1.8

2

2.2

2.4

Best FitObserved 95% CL

SM

SM expected 95% CL


-113 TeV, 36.1 fb

= 0Hvvκ = 1, Hggκ

= 1.2SMκ| = 1.5, AvvκObserved: | = 1.0SMκ = 0.0, AvvκExpected:

Hvvκ6− 4− 2− 0 2 4 6

SMκ

0.6

0.8

1

1.2

1.4

1.6

1.8

2

2.2


SM

SM expected 95% CL


-113 TeV, 36.1 fb

= 0Avvκ = 1, Hggκ

= 1.2SMκ = 2.1, HvvκObserved: = 1.0SMκ = 0.0, HvvκExpected:



2D scans

Avvκ6− 4− 2− 0 2 4 6

Hvv

κ

4−

2−

0

2

4

6

8

10

12

14


SM

SM expected 95% CL


-113 TeV, 36.1 fb

= 1SMκ = 1, Hggκ

= 2.9Hvvκ| = 0.5, AvvκObserved: | = 0.0Hvvκ = 0.0, AvvκExpected:

Avvκ6− 4− 2− 0 2 4 6

Hvv

κ

4−

2−

0

2

4

6

8

10

12

14


SM

SM expected 95% CL


-113 TeV, 36.1 fb

freeSMκ = 1, Hggκ

= 1.7SMκ = 2.1, Hvvκ| = 0.3, AvvκObserved: | = 1.0SMκ = 0.0, Hvvκ = 0.0, AvvκExpected:



SM Higgs Boson Production Cross Section

[TeV] s6 7 8 9 10 11 12 13 14 15

H+

X)

[pb]

→(p

p σ

2−10

1−10

1

10

210 M(H)= 125 GeV

LH

C H

IGG

S X

S W

G 2

016

H (N3LO QCD + NLO EW)

→pp

qqH (NNLO QCD + NLO EW)

→pp

WH (NNLO QCD + NLO EW)

→pp

ZH (NNLO QCD + NLO EW)

→pp

ttH (NLO QCD + NLO EW)

→pp

bbH (NNLO QCD in 5FS, NLO QCD in 4FS)

→pp

tH (NLO QCD, t-ch + s-ch)

→pp



CP-Violation in the 2HDM

Additional SU(2) doublet→ two neutral scalars (h,H), a pseudoscalar (A), two charged (H±)

2HDM potential:

V =1

2λ1(ϕ†1ϕ1

)2+

1

2λ2(ϕ†2ϕ2

)2+ λ3

(ϕ†1ϕ1

)(ϕ†2ϕ2

)+ λ4

(ϕ†1ϕ2

)(ϕ†2ϕ1

)+

1

2

[λ5(ϕ†1ϕ2

)2+ h.c.

]+

{[λ6(ϕ†1ϕ1

)2+ λ7

(ϕ†2ϕ2

)2] (ϕ†1ϕ2

)+ h.c.

}−{m211

(ϕ†1ϕ1

)+[m212

(ϕ†1ϕ2

)+ h.c.

]+ m2

22

(ϕ†2ϕ2

)}

no CP violation in the Higgs sector and no FCNC if:λ6 = λ7 = m2

12 = 0 (Z2 Symmetry)



CP-Violation in the 2HDM

Simplest case of CP violation in the Higgs sector:λ6 = λ7 = 0 and m2

12 = 0Parametrization of the minimum of the potential:

ϕ1 =

(0

1√2v1

), ϕ2 =

(0

1√2v2eiξ

)Relation: Im

(m2

12eiξ)=Im(λ5e2iξ

)v1v2

rephasing invariance⇒ ξ = 0

Mass squared matrix of the neutral sector:

M2 =

M211 M2

12 − 12 Imλ5v2 sinβ

M212 M2

22 − 12 Imλ5v2 cosβ

− 12 Imλ5v2 sinβ − 1

2 Imλ5v2 cosβ M233

λ5 = 0: three neutral Higgs state mix⇒ CP-violation



Backup

H → ZZ∗ → 4µ

µ µ

µ µ



H → ZZ∗ → 4ℓ Event Selection

2 pairs opposite electric charge,same flavourCut on invariant masses:

– Leading lepton pair:50 GeV < m12 < 106 GeV

– Off-shell Z: mthreshold < m34 < 115 GeV

Muon (electron) isolation:Trackisolation:Itrackµ(e) =

(∑ptrackT

)/pℓT(E

ℓT) < 0.15 (0.15)

Calorimeterisolation:Icaloµ(e) =

(∑EclusterT

)/pℓT(E

ℓT) < 0.30 (0.20)

d0-significance:Muons: |d0/σd0 | < 3.0Electrons: |d0/σd0 | < 5.0

ℓ ℓ

ℓ ℓ

Higgs-Signal,ZZ∗



H → ZZ∗ → 4ℓ Event Selection

2 pairs opposite electric charge,same flavourCut on invariant masses:

– Leading lepton pair:50 GeV < m12 < 106 GeV

– Off-shell Z: mthreshold < m34 < 115 GeV

Muon (electron) isolation:– Track-based isolation:

Itrackµ(e) =(∑

ptrackT

)/pℓT(E

ℓT) < 0.15 (0.15)

– Calorimeter-based isolation:Icaloµ(e) =

(∑EclusterT

)/pℓT(E

ℓT) < 0.30 (0.20)

d0-significance:– Muons: |d0/σd0 | < 3.0– Electrons: |d0/σd0 | < 5.0

ℓ ℓ

ℓ ℓ

Higgs-Signal,ZZ∗

ℓℓ

ℓ ℓ

Z+ bbb

b

y

x

d0

trackPrimaryVertex



Branching Ratios of the SM Higgs Boson

[GeV]HM120 122 124 126 128 130

Bra

nchi

ng R

atio

-410

-310

-210

-110

1

LH

C H

IGG

S X

S W

G 2

016

bb

ττ

µµ

cc

gg

γγ

ZZ

WW

γZ



pp Collision Data Recorded with the ATLAS Detector

Number of expected pp collisions:

Nexp =L · σ =

∫dt L · σ

L ∝ nb · N2b · frev

Run I (2011+2012):–

√s = 7 TeV: L= 4.6 fb−1

–√s = 8 TeV: L= 20.7 fb−1

Run II (since 2015):–

√s = 13 TeV: L= 36.1 fb−1

Delivered luminosity:

Month in YearJan Apr Jul

Oct

]1

De

live

red

Lu

min

osity [

fb

0

5

10

15

20

25

30

35

40

45

ATLAS Online Luminosity

= 7 TeVs2011 pp

= 8 TeVs2012 pp

= 13 TeVs2015 pp

= 13 TeVs2016 pp

2/1

7 c

alib

ratio

n

This study:Run II pp collision data L= 36.1 fb−1

2016

2012

20112015



Production and Decay of the SM Higgs Boson at the LHC

Production of the SM Higgs boson with a mass of 125 GeV at√s=13 TeV:

t/b

g

g

H

W/Z

W/Z

q′

q

q′

q

HW/Z

q

q

W/Z

H

g

g

t/b

t/b

H

Decay of the SM Higgs boson:

H

f

f

H

Z∗/W ∗

Z/W

t/b/W+

t/b/W

t/b/W−

H

γ

γ

t

t

t

H

g

g

ggF VBF VH ttHbbH

H → VV

87.4% 6.8% 4.0% 1.8%

66% 24%

0.0124%

0.2% 9%

H → ff H →γγ H → gg

H → ZZ∗ → 4ℓ



Full Event Selection

Table:Event selection criteria applied in the H → ZZ∗ → 4ℓ analysis [?].

Lepton Requirements

ElectronsET > 7 GeV and |η| < 2.47z0 · sin(θ) < 5 mm

Muons

pT > 5 GeV and |η| < 2.7pT > 15 GeV and |η| < 0.1 (CT muons)z0 · sin(θ) < 5 mm|d0| < 1 mm

Event selection

Higgs Boson Candidate

Two lepton pairs of same-flavour and opposite-chargepT > 20, 15, 10 GeV for the three highest-pT leptons∆R(ℓ, ℓ′) > 0.10 (0.20) for same- (different-) flavour leptonsmℓℓ > 5 GeV for same-flavour opposite-charge di-lepton pairs50 GeV< m12 < 106 GeVmthreshold < m12 < 115 GeV

Lepton Isolation

Muon track isolation (∆R ≤ 0.3): Itrackµ < 0.15

Muon calorimeter isolation (∆R = 0.2): Icaloµ < 0.30

Electron track isolation (∆R ≤ 0.2): Itracke < 0.15

Electron calorimeter isolation (∆R = 0.2): Icaloe < 0.20|d0/σd0 | < 3(5) for muons (electrons)

Vertex Selectionχ2/Ndof < 6 for 4µ candidatesχ2/Ndof < 9 for 2e2µ, 2µ2e, 4e candidates



The H → ZZ → 4ℓ Decay ChannelSignal:

Z∗

Z

H

`+

`−

`+

`−

Irreducible background:

Z/γ

Z/γ

`+

`−

`+

`−

q′

q

Reducible background: at least one lepton originates from a gluon or jet

Z

g

`+

`−

q

qq′

q

b

g

t

t

W−

bW+

`−

ν`

`+

ν`

g

g

ℓ± = e±,µ±

ZZ∗

Estimated with Monte Carlo Simulation

Data driven estimation

tt

Z+ jets



Results of the Event Selection

Invariant mass spectrum after event selection:

Good agreement between dataand simulation

Observed and expected numberof events in118 GeV < m4ℓ < 129 GeV:

Signal 54.0 ± 4.0ZZ∗ 19.7 ± 1.5Z+jets, tt 3.9 ± 0.5

Total Expected 77.0 ± 4.0Total Observed 95.0 [GeV]Constrained

4lm

80 100 120 140 160

Events

/2.5

GeV

0

10

20

30

40

50

60 Data

= 125.09 GeV)H

(mHiggs

ZZ*

+V, VVV tt

tZ+jets, t

Uncertainty

ATLAS Preliminary

4l→ ZZ* →H 113 TeV, 36.1 fb



CP measurements in the H → ZZ∗ → 4ℓ channelEffective field theory (EFT) implemented in the so called Higgscharacterisation model (arXiv:1306.6464)

LV0 =

{cακSM

[1

2gHZZZµZµ + gHWWW+

µW−µ

]−

1

4

[cακHγγgHγγAµνAµν + sακAγγgAγγAµν Aµν

]−

1

2

[cακHZγgHZγZµνAµν + sακAZγgAZγZµν Aµν

]−

1

4

[cακHgggHggGa

µνGa,µν + sακAgggAggGa

µν Ga,µν]

−1

4

1

Λ

[cακHZZZµνZµν + sακAZZZµν Zµν

]−

1

2

1

Λ

[cακHWWW+

µνW−µν + sακAWWW+

µνW−µν

]−

1

Λcα[κH∂γZν∂µA

µν + κH∂ZZν∂µZµν + κH∂W(W

+ν ∂µW

−µν + h.c.)] }

X0

Lf0 =−

∑f=t,b,τ

ψf (cακHffgHff + i sin (α)κAffgAffγ5)ψfX0

CP violation: Mixture of CP even and CP odd

Bosons

Fermions

Done in Run 1SM CP-even,tree-level

BSM CP-evenBSM CP-oddα = CP mixing angle

κ = HC coupling parameter

g = coupling strength SM or MSSM

Λ = cut-off energy

cα = cos (α)

sα = sin (α)



Measurement of the HVV Tensor Coupling

CP-sensitive kinematic discriminants in the decay system :

z′

z

Φ1

Φ

p X

Z2

Z1

p

µ+

µ−

θ1

θ∗

θ2

e+

e−

m12, m34, cos(θ1), cos(θ2), ϕ, ϕ1, θ∗ )1θcos(1− 0.5− 0 0.5 1

Ent

ries

/ 0.2

50

2

4

6

8

10

12

14

16

18 ATLAS

4l→ZZ*→H-1 = 7 TeV, 4.5 fbs

-1 = 8 TeV, 20.3 fbs

Data

Background ZZ*


SM+ = 0PJ- = 0PJ

Data

Background ZZ*


SM+ = 0PJ- = 0PJ

Information from kinematicdistributions (used in Run-1)

∢(Z1, ℓ−)

Eur. Phys. J. C75 (2015) 476



Signal Modelling via Morphing Method

Number of input samples Ti is dependent on the number of couplings onewants to model

N =np (np + 1)

2· nd (nd + 1)

2+

(4 + ns − 1

4

)+

(np · ns +

ns (ns + 1)

2

)· nd (nd + 1)

2+

(nd · ns +

ns (ns + 1)

2

)· np (np + 1)

2

+ns (ns + 1)

2· np · nd + (np + nd)

(3 + ns − 1

3

)



Relevant couplings in ggF/VBF production in H4ℓ channel

Production: ggF,VBF

t/b

g

g

H

W/Z

W/Z

q′

q

q′

q

H

Decay: H → 4ℓ

Z∗

Z

H

`+

`−

`+

`−κHgg

κAgg

κSM,κHzz, κHww, κHzγ ,κHγγ , κH∂z,

κH∂w, κH∂γ

κAzz, κAww,

κAγγ , κAzγ ,

κSM

κAzz, κAγγ , κAzγ ,

κHzz, κHγγ , κHzγ ,κH∂z, κH∂γ



Signal Modelling via Morphing MethodStatistical uncertainty of output distribution:

∆Tbinout =

√√√√∑i

wi (κout; κi)NbinMC,i (κi) ·

(σi(κi)LNMC,i(κi)

)2

SMκ

BS

Mκ

Validation Point

B1 B2

B3

A1

A2 A3

Observable sensitive to anomalous coupling

Num

ber

of e

xpec

ted

even

ts

Validation sampleMorphing output set AMorphing output set B

Different choice of input samples⇒ different statistical uncertaintiesIn addition, require that predicted value of the observable agrees withvalidation value

fixed κ

Set of input samples:◦ A1, ◦ A2,◦ A3• B1, • B2,• B3



Set of Input Samples for Modelling VBF and VH productionVBF and VH production have the same coupling structure:

W/Z

W/Z

q′

q

q′

q

HW/Z

q

q

W/Z

H

⇒ VBF and VH production are combinedAssumption: κXZZ = κXWW, where X = H,ASeparate model for κAZZ and κHZZ

VBF VHκHZZ, κAZZκSM

κHZZ, κAZZ κSM



Set of Input Samples for Modelling VBF and VH production

W/Z

W/Z

q′

q

q′

q

H

Z∗

Z

H

`+

`−

`+

`−

Production mode np nd ns N

VBF+VH 0 0 2 5

Azzκ15− 10− 5− 0 5 10 15

Hzz

κ

10−

5−

0

5

10

V_kHVV5_kSM1

V_kHVV2p5_kSM1

V_kHVVm5_kSM1

V_kHVVm2p5_kSM1

V_kAVV5_kSM1

V_kAVV2p5_kSM1

V_kAvvm5_kSM1

V_kAVVm2p5_kSM1

V_kSM1

V_kAVV15_kSM0

V_kHVV10_kSM0

=1SM

input samples, g

=0SM

input samples, g

=1SM

validation samples, g

κSMκHZZ or κAZZ

κSMκHZZ or κAZZ

H → ZZ∗ → 4ℓ

VBF

Input samples in κAZZ-κHZZ plane:




W/Z

W/Z

q′

q

q′

q

H

Z∗

Z

H

`+

`−

`+

`−


VBF+VH 0 0 2 5

12m

50 55 60 65 70 75 80 85 90 95 100

cros

s se

ctio

n in

[pb]

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

V_kAVVm2p5_kSM1, validation sample

V_kAVVm2p5_kSM1, signal model

= 13 TeVs4l, →ZZ→MadGraph5_aMC@NLO, VBF: H

κSMκHZZ or κAZZ

κSMκHZZ or κAZZ

H → ZZ∗ → 4ℓ

VBF

Validation of the input sample set: κAZZ

Good agreement between predic-tion and validation point




W/Z

W/Z

q′

q

q′

q

H

Z∗

Z

H

`+

`−

`+

`−


VBF+VH 0 0 2 5

12m

50 55 60 65 70 75 80 85 90 95 100

cros

s se

ctio

n in

[pb]

0.5

1

1.5

2

2.5

3

3.5

4

4.5

5

V_HVVm2p5_kSM1, validation sample

V_HVVm2p5_kSM1, signal model

= 13 TeVs4l, →ZZ→MadGraph5_aMC@NLO, VBF: H

κSMκHZZ or κAZZ

κSMκHZZ or κAZZ

H → ZZ∗ → 4ℓ

VBF

Validation of the input sample set: κHZZ

Good agreement between predic-tion and validation point



CP sensitive discriminants in the decay system

Eur. Phys. J. C75 (2015) 476

[GeV]12m50 60 70 80 90 100 110

Ent

ries

/ 5 G

eV

0

5

10

15

20

25ATLAS

4l→ZZ*→H-1 = 7 TeV, 4.5 fbs

-1 = 8 TeV, 20.3 fbs

Data

Background ZZ*


SM+ = 0PJ- = 0PJ

Data

Background ZZ*


SM+ = 0PJ- = 0PJ

[GeV]34m20 30 40 50 60 70

Ent

ries

/ 5 G

eV

0

5

10

15

20

25ATLAS

4l→ZZ*→H-1 = 7 TeV, 4.5 fbs

-1 = 8 TeV, 20.3 fbs

Data

Background ZZ*


SM+ = 0PJ- = 0PJ

Data

Background ZZ*


SM+ = 0PJ- = 0PJ



CP sensitive discriminants in the decay system

Eur. Phys. J. C75 (2015) 476

)2θcos(1− 0.5− 0 0.5 1

Ent

ries

/ 0.2

5

0

5

10

15

20

25 ATLAS

4l→ZZ*→H-1 = 7 TeV, 4.5 fbs

-1 = 8 TeV, 20.3 fbs

Data

Background ZZ*


SM+ = 0PJ- = 0PJ

Data

Background ZZ*


SM+ = 0PJ- = 0PJ

φ3− 2− 1− 0 1 2 3

/4)

πE

ntrie

s / (

0

5

10

15

20

25 ATLAS

4l→ZZ*→H-1 = 7 TeV, 4.5 fbs

-1 = 8 TeV, 20.3 fbs

Data

Background ZZ*


SM+ = 0PJ- = 0PJ

Data

Background ZZ*


SM+ = 0PJ- = 0PJ



Signal Modelling via Morphing MethodContinuous signal model to describe signal expectation in dependence onBSM couplingsPredicts kinematic distributions and cross-sections at every parameter point

Tout(κout) =Ninput∑i=1

wi(κout; κi) · Tin(κi) κ = (κSM, κ1BSM, . . . , κ

nBSM)

Assumption:

T (κ) ∝ |M(κ)|2 =(∑

α∈p,s

καO(κα)

)2

︸︷︷︸production

·(∑

α∈d,s

καO(κα)

)2

︸︷︷︸decay

Challenge: Find the set of input samples which gives the lowest statisticaluncertainty

p: productiond: decays: shared in both

Output distribution weight Input distribution



Morphing: Calculation of weights functions

Tout(gout) =Ninput∑i=1

wi(gout; gi) · Tin(gi) e.g. T = σ · BR, T = cos θ1

Ansatz for morphing weights:

wi = (ai1gSM2 + ai2gBSM

2 + ai3gSMgBSM)

Requirement for calculation of constants:

wi = 1 and wj =i = 0 if gout = gi

⇒ Linear system of equations, solveable through matrix inversion



Morphing: Specific example

SM

Mix

BSM

Interferenceg2SM

g2BSM

+1

−1

−1

gSM · gBSM

(gSM,gBSM)

(1,0)

(1,1)

(0,1)

Morphing function for this specific example:Tout(gSM, gBSM) = (gSM2 − gSMgBSM)︸︷︷︸

=w1

Tin(1, 0)+(gBSM2 − gSMgBSM)︸︷︷︸=w2

Tin(0, 1)+gSMgBSM︸︷︷︸=w3

Tin(1, 1)

⇒ Set of input samples can be arbitrarily chosen as long as linear system ofequations can be solved



Documents

StudyoftheTensorStructureofHiggsBoson CouplingswiththeH ZZ ... · StudyoftheTensorStructureofHiggsBoson CouplingswiththeH! ZZ! 4ℓ DecayChannelwiththeATLASDetector IMPRSYoungScientistWorkshopatRingbergCastle