[PPT] - Observation of the Higgs particle in events and search for the Higgs PowerPoint Presentation

SLIDE 1

Observation of the Higgs particle in γγ events and search for the Higgs particle in Zγ events at ATLAS

LIU Kun

1Laboratoire de Physique Nucl´

eaire et de Hautes Energies (LPNHE)

2University of Science and Technology of China (USTC)

June 24th, 2014

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 1 / 72

SLIDE 2

Outline

The Standard Model Higgs Boson Thesis motivation Experimental Setup Photon Performance Search for a Higgs Boson in H → Zγ → ℓℓγ (ℓ = µ, e) Observation of the Higgs Boson in γγ Events Summary

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 2 / 72

SLIDE 3

The Standard Model Higgs Boson Success of the Standard Model (SM)

Success of the Standard Model (SM), but...

Spin- 1

2 fermions, grains of matter.

Spin-1 bosons for interactions. Most of the SM predictions have been confirmed in experiments. Spontaneous Symmetry Breaking through Higgs mechanism:

rigin of particle masses.

The spin-0 Higgs boson is pre- dicted by the Higgs mechanism. The Higgs boson had not been detected before July 2011. The Higgs boson mass is a free parameter.

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 3 / 72

SLIDE 4

The Standard Model Higgs Boson Constraints on the SM Higgs Boson Mass

Constraints on the SM Higgs boson mass in July 2011

Theoretical bounds: the existence of the Higgs boson leads to the cancellation of the ultraviolet divergence in the scattering amplitude

f W +

L W − L → W + L W − L , provided the Higgs boson mass is not too

heavy: mH <

4π

√ 2 3GF ∼ 700 GeV Experimental direct limits (95 % CL), in July 2011, from LEP, Tevatron. Exclusion region: mH < 114.4 GeV mH within [156, 177] GeV.

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 4 / 72

SLIDE 5

The Standard Model Higgs Boson Constraints on the SM Higgs Boson Mass

Constraints on the SM Higgs boson mass in July 2011

A global fit to electroweak data (from Hfitter group): Input parameters: W , Z, top masses and widths, cross sections Best fitted mH = 91+30

−23 GeV (without direct limits on mH as inputs)

This constraint favours a low mass value for the SM Higgs boson

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 5 / 72

SLIDE 6

The Standard Model Higgs Boson The Higgs Boson Production and Decay

The Higgs boson production and decay in the SM

Five main production processes (ggH, VBF, WH, ZH, t¯ tH):

q q q q H V V

H t, b g g

u/d d/u H W W± q q H Z Z H g g t t

(a) (b) (c) (e) (d)

[GeV]

H

m 80 100 200 300 400 1000 H+X) [pb] → (pp σ

2

10

1

10 1 10

2

10 = 8 TeV s

LHC HIGGS XS WG 2012 H (NNLO+NNLL QCD + NLO EW) → pp q q H ( N N L O Q C D + N L O E W ) → p p WH (NNLO QCD + NLO EW) → pp ZH (NNLO QCD +NLO EW) → pp ttH (NLO QCD) → pp

For a light Higgs boson in the SM

H → γγ and H → ZZ → 4ℓ (ℓ = µ, e) clean final states and large sensitivity H → Zγ → ℓℓγ (ℓ = µ, e) and H → µµ: clean final states but small BR

[GeV]

H

m

80 100 120 140 160 180 200

Higgs BR + Total Uncert

4

10

3

10

2

10

1

10 1

LHC HIGGS XS WG 2013

b b τ τ µ µ c c gg γ γ γ Z WW ZZ

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 6 / 72

SLIDE 7

Motivation The H → γγ and the H → Zγ → ℓℓγ Decays

The H → γγ and the H → Zγ → ℓℓγ decays in the SM

They are loop-induced decays, dominated by W loop. Two kinds of backgrounds: irreducible and reducible background

γγ (75%), γ+jet(jet+γ), jet+jet (for H → γγ) ℓℓ + γ (82%), ℓℓ+jet (for H → Zγ → ℓℓγ)

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 7 / 72

SLIDE 8

Motivation The H → γγ and the H → Zγ → ℓℓγ Decays

The H → γγ and the H → Zγ → ℓℓγ decays in the SM

At mH = 125 GeV, σH × BRH→γγ = 40 (50) fb at √s = 7 (8) TeV, while σH × BRH→Zγ→ℓℓγ (ℓ=µ,e) = 1.8 (2.3) fb. H → γγ and H → Zγ → ℓℓγ decays

clean final states excellent photon and lepton energy response resolution loop-induced decays: sensitive probe for new physics

Reminders on the photon performance

high photon selection efficiency and jet rejection needed to improve S/ √ B accurate measurements of the photon trigger and identification efficiencies are needed for precise measurements of cross sections and Higgs boson properties.

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 8 / 72

SLIDE 9

Experimental Setup The Large Hadron Collider (LHC)

The Large Hadron Collider (LHC)

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 9 / 72

SLIDE 10

Experimental Setup The ATLAS Detector

Running parameters in 2011 (2012)

ATLAS recorded 5.1 (21.3) fb−1 of pp collisions at √s = 7 (8) TeV (left plot) Mean number of interactions per crossing was 9.1 (20.7) in 2011 (2012) (right plot)

Month in Year J a n A p r J u l O c t J a n A p r J u l O c t

1

fb Total Integrated Luminosity 5 10 15 20 25 30 ATLAS Preliminary

= 7 TeV s 2011, = 8 TeV s 2012, LHC Delivered ATLAS Recorded

1

fb Delivered: 5.46

1

fb Recorded: 5.08

1

fb Delivered: 22.8

1

fb Recorded: 21.3

Mean Number of Interactions per Crossing 5 10 15 20 25 30 35 40 45 /0.1]

1

Recorded Luminosity [pb 20 40 60 80 100 120 140 160 180 Online Luminosity ATLAS

> = 20.7 µ , <

1

Ldt = 21.7 fb

∫

= 8 TeV, s > = 9.1 µ , <

1

Ldt = 5.2 fb

∫

= 7 TeV, s

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 10 / 72

SLIDE 11

Experimental Setup The ATLAS Detector

The ATLAS detector components

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 11 / 72

SLIDE 12

Experimental Setup The ATLAS Detector

Excellent performance of the ATLAS tracker system

The precision tracking detectors

the Inner detector (ID) : vertex and track reconstruction, precision track momentum measurement the Muon Spectrometer (MS) : muon track reconstruction and precision momentum measurement

Very high vertex and track reconstruction efficiencies

µ 5 10 15 20 25 30 35 40 45 Vertex Reconstruction Efficiency

0.98 0.985 0.99 0.995 1 Simulation t t µ µ → Z ee → Z

ATLAS Preliminary LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 12 / 72

SLIDE 13

Experimental Setup The ATLAS Detector

The Electromagnetic Calorimeter (EMC)

A Pb-LAr sampling detector with accordion-shaped kapton electrodes. 3 layers to reconstruct γ direction. 1st layer segmentation to separate prompt γ from photon pairs from π0, η... A thin LAr presampler (|η| <1.8) to estimate the energy lost before the accordion calorimeter.

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 13 / 72

SLIDE 14

Photon performance Photon Reconstruction and Identification

Photon reconstruction

Unconverted photons: clusters not matched to tracks. Converted photons: clusters matched to at least one track (and ambiguity resolution w.r.t. electrons). About half of the photons convert before the EMC. Photon reconstruction efficiencies are very high

[GeV]

T

p 20 40 60 80 100 120 Reconstruction efficiency 0.8 0.82 0.84 0.86 0.88 0.9 0.92 0.94 0.96 0.98 1 ATLAS Preliminary Simulation

All photons Unconverted photons Converted photons

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 14 / 72

SLIDE 15

Photon performance Photon Reconstruction and Identification

Photon/electron isolation

Discriminating γ/e from fake and non-direct γ/e coming from jets. Calorimeter isolation: collecting energy deposited in a cone ∆R = 0.4 around the γ/e object

ETcone40: from all cells E Topo

Tcone40 : from cells belonging to topological clusters

Track isolation: in a cone of ∆R = 0.2 around the γ/e object

pTcone20: scalar sum of pT of tracks nucone20: number of tracks

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 15 / 72

SLIDE 16

Photon performance Photon Reconstruction and Identification

Photon performance

[ATLAS-CONF-2012-123] Photon identification efficiency measurements, Photon trigger

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 16 / 72

SLIDE 17

Photon performance Photon Reconstruction and Identification

Photon identification

9 discriminating variables As example Eratio, ∆E

ratio

E 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Entries/0.02 1 10

2

10

3

10

4

10

5

10

data γ ll → Z corrected MC γ ll → Z ll)+jet corrected MC → Z(

1

Ldt=20.3 fb

∫

=8 TeV, s γ Converted

ATLAS Preliminary E [MeV] ∆ 1000 2000 3000 4000 5000 6000 7000 8000 Entries/160 MeV 1 10

2

10

3

10

4

10

data γ ll → Z corrected MC γ ll → Z ll)+jet corrected MC → Z(

1

Ldt=20.3 fb

∫

=8 TeV, s γ Unconverted

ATLAS Preliminary LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 17 / 72

SLIDE 18

Photon performance Photon Identification Efficiency Measurements

Photon identification efficiency measurements

Two identification algorithms

cut-based: loose, tight ID neural-network based ID (only used in H → γγ at √s =7 TeV)

Three data-driven methods:

using Final State Radiative (FSR) photons from Z → ℓℓγ decays matrix method electron extrapolation

[GeV]

T

E 10 20 30 100 200 1000 2000 0.2 0.4 0.6 0.8 1

data-driven MC-based

Data-MC correction on signal MC Matrix method Extrapolation ee → from Z Z radiative decays LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 18 / 72

SLIDE 19

Photon performance Photon Identification Efficiency Measurements

Photons from radiative Z decays

FSR photons from Z → ℓℓγ (ℓ = µ, e) decays, in low ET [10, 100] GeV. Event selection

requirements on lepton quality mℓℓ within [40, 83] GeV mℓℓγ within [80, 96] GeV

120k photon probes in √s =8 TeV data Photon purity from mℓℓγ fit

∼90% for E γ

T in [10,15] GeV

≥ 98% for E γ

T > 15 GeV

Residual backgrounds

subtracted in 10 < E γ

T < 15 GeV

as systematic uncertainty for higher E γ

T photons(< 2%).

[GeV]

γ ll

m 20 40 60 80 100 120 140 160 180 200 [GeV]

ll

m 20 40 60 80 100 120 140 160 Events 100 200 300 400 500 600 700 800 900

1

L dt = 20.3 fb

∫

= 8 TeV s , e) µ (l= γ ll → Z ATLAS work in progress [GeV]

γ µ µ

m 60 65 70 75 80 85 90 95 100 105 110 ]

1

[GeV

γ µ µ

dN/dm 10

2

10

3

10

4

10

data fit signal background signal region

ATLAS work in progress

1

L dt = 20.3 fb

∫

Unconverted photon < 15 GeV

T γ

10 < E 0.2)% ± P = (93.4

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 19 / 72

SLIDE 20

Photon performance Photon Identification Efficiency Measurements

The matrix method

Photon yield (and purity) before/after photon ID can be estimated using track isolation to discriminate photons from fakes.

NT

pass = NS pass + NB pass

NIso

pass

= εs

p × NS pass + εb p × NB pass

NT

fail = NS fail + NB fail

= ⇒ NIso

fail

= εs

f × NS fail + εb f × NB fail

nce track isolation efficiencies εs

p, εs f , εb p and εb f are known.

Distributions of εs

p and εb p (left), εs f and εb f (right):

Passing tight-ID criteria Failing tight-ID criteria

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 20 / 72

SLIDE 21

Photon performance Photon Identification Efficiency Measurements

The matrix method

Photon track isolation efficiencies (εs

p, εs f )

from simulated prompt photon events data/MC difference in Z→ee (1%) included in systematic uncertainty

Fake photon track isolation efficiencies (εb

p, εb f )

measured in a fake photon enriched data sample selected by reversing part of ID requirements (Fside, w3, ∆E, Eratio) signal leakage in the fake photon control region is subtracted non-closure in di-jet simulation included in systematic uncertainties

Sources of systematic uncertainties:

n the photon track isolation efficiencies (< 1%)
n the fake track isolation efficiencies (1%–10%)

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 21 / 72

SLIDE 22

Photon performance Photon Identification Efficiency Measurements

Efficiency comparison and combination

Three data-driven results are in good agreement within uncertainties. Relative systematic uncertainty of combined efficiency

in 2011: 4% to < 1% from 20 to 200 GeV in 2012: 4% to < 1% from 10 to 500 GeV

Pile-up dependence is well modeled in MC.

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 22 / 72

SLIDE 23

Photon performance Photon Trigger

Photon medium trigger in 2012 data-taking

Reduce rate (to half for γγ trigger) and pile-up dependence

ptimizing criteria on variables used in loose trigger

adding medium cut on extra variable Eratio

Trigger efficiency vs number of primary vertex: loose trigger and medium trigger

Nvtx 5 10 15 20 25 30 35 40 45 50 s ∈ 0.8 0.82 0.84 0.86 0.88 0.9 0.92 0.94 0.96 0.98 1

EF_g20_loose Weta2(tighten) + Rhad(loosen) + Reta(loosen) Weta2(tighten) + Rhad(loosen) + Reta(loosen) + Eratio(loose) Weta2(tighten) + Rhad(loosen) + Reta(loosen) + Eratio(medium) Weta2(tighten) + Rhad(loosen) + Reta(loosen) + Eratio(medium++) Weta2(tighten) + Rhad(loosen) + Reta(loosen) + Eratio(tight)

The default trigger for H → γγ analysis in Run-II

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 23 / 72

SLIDE 24

Photon performance Photon Trigger

Photon trigger efficiency measurement

Using photons from radiative Z decays. W.R.T. off-line photon selection criteria. Di-photon trigger efficiencies are the products of the efficiencies for the two single trigger objects. Trigger efficiency uncertainty is negligible (< 0.5%) compared to

thers in the H → γγ analysis.

[GeV]

T γ

E 10 20 30 40 50 60 70 80 90 100 Efficiency/2GeV 0.2 0.4 0.6 0.8 1

EF_g20_medium EF_g20_loose

ATLAS work in progress

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 24 / 72

SLIDE 25

Physics The H → γ and H → Zγ → ℓℓγ Decays

We were ready to search for the Higgs boson

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 25 / 72

SLIDE 26

Search for a Higgs Boson in H → Zγ → ℓℓγ decays

Search for a Higgs Boson in H → Zγ → ℓℓγ (ℓ = µ, e) decays

[ATLAS-CONF-2013-009], [Phys. Lett. B 732(2014)8-27] Using 4.5 (20.3) fb−1 of data at √s = 7 (8) TeV

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 26 / 72

SLIDE 27

Search for a Higgs Boson in H → Zγ → ℓℓγ decays Event Selection

Event selection (I)

Using lepton triggers At least one primary vertex Muon selection:

pT > 10 GeV, |η| < 2.7 muon track quality impact parameter: |d0| <1 mm, |z0| < 10 mm

Electron selection

ET > 10 GeV, |η| < 2.47 identification based on shower shape and ID information |z0| <10 mm

verlap removal: remove the electron if its track is within a ∆R < 0.2

around a muon track

[GeV]

T

p 20 40 60 80 100 120 140 [1/GeV]

T

dN/dp 200 400 600 800 1000 1200 1400 1600

γ

1

l

2

l µ µ → , Z γ Z → H = 8 TeV s = 125 GeV,

H

m process = gg fusion

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 27 / 72

SLIDE 28

Search for a Higgs Boson in H → Zγ → ℓℓγ decays Event Selection

Event selection (II)

Photon selection

ET > 15 GeV, |η| <2.37 ∆Rℓγ > 0.3 and passing tight ID criteria E Topo

Tcone40 < 4 GeV

Z → ℓℓ reconstruction and selection

two same flavor and opposite sign leptons the lepton pair whose mℓℓ is closest to mZ the reconstructed leptons are matched to trigger objects mℓℓ > 81.18 GeV

H → Zγ reconstruction and selection

lepton isolation requirements muon: pTcone20/pT < 0.15, ETcone20/ET < 0.3 electron: pTcone20/ET < 0.15, ETcone20/ET < 0.3 |d0|/σd0 < 3.5 (6.5) for muons (electrons)

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 28 / 72

SLIDE 29

Search for a Higgs Boson in H → Zγ → ℓℓγ decays Event Categorization

Event categorization

Classify events exploiting the different kinematic properties of signal and background to improve analysis sensitivity Two variables are used:

|∆ηZγ|, the |η| difference between the photon and Z boson pTt, the component of the pH

T that is orthogonal to

pγ

T −

pZ

T

Optimized values of 30 GeV on pTt and of 2.0 on |∆ηZγ| are chosen

[GeV]

Tt

p 20 40 60 80 100 120 140 / 5 GeV

Tt

1/N dN/dp

4

10

3

10

2

10

1

10

gg fusion VBF WH/ZH ttH Data MC γ Z ee → , Z γ Z → H = 8 TeV s = 125 GeV,

H

m

ATLAS |

γ Z

η ∆ | 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 |

γ Z

η ∆ 1/N dN/d| 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16 0.18

gg fusion VBF WH/ZH ttH Data MC γ Z ee → , Z γ Z → H = 8 TeV s = 125 GeV,

H

m

ATLAS

10 categories based on lepton flavour, center-of-mass energy, and previous variables (in table in later slide).

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 29 / 72

SLIDE 30

Search for a Higgs Boson in H → Zγ → ℓℓγ decays Background Composition

Background composition

Expecting 15 signal events with huge backgrounds after selection. The mℓℓγ distributions: ∼82% (∼ 17%) of Z + γ (Z+jets) events

[GeV]

γ ll

m 100 150 200 250 300 Events/4 GeV 200 400 600 800 1000 1200

γ Z Z+jets + WZ t t bkg uncertainty Data

ATLAS

µ µ → , Z γ Z → H

1

Ldt = 20.3 fb

∫

= 8 TeV, s LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 30 / 72

SLIDE 31

Search for a Higgs Boson in H → Zγ → ℓℓγ decays Exclusion Limits and p−values

95% C.L. limit (CLs) and p-value

Likelihood-based statistical test :

95% C.L. limit on cross section × BRH→Zγ p0-values : compatibility of data with background only hypothesis

Unbinned likelihood :

discriminating variable x = mℓℓγ profile likelihood ratio is used (µ = Nsignal

NSM

signal ): λ(µ) = L(µ,ˆ

ˆ θ(µ)) L(ˆ µ, ˆ θ)

in each category: Lc(µ, θc|xc) = e−N′

cN′

c Nc Nc k=1 Lc(xk|µ, θc)

for each event Lc(x|µ, θc) =

Nsignal,c(µ,θnorm

c

) Nsignal,c+Nbkg,c fsignal,c(x|θshape c

) +

Nbkg,c Nsignal,c+Nbkg,c fbkg,c(x|θbkg c

)

The nuisance parameters (θ) are constrained either with a Log-normal distribution (affecting signal yields) or a Gaussian distribution.

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 31 / 72

SLIDE 32

Search for a Higgs Boson in H → Zγ → ℓℓγ decays Signal Parameterization

Signal parameterization

Expected signal yield from: theoretical production cross section and branching ratio, selection efficiency from simulation, and measured luminosity Signal model: Crystal Ball + Gaussian function fit from simulation

[GeV]

γ ll

m 105 110 115 120 125 130 135 140 145 ]

1

[GeV

γ ll

dN/dm 100 200 300 400 500 600

µ µ → , Z γ Z → H = 125 GeV

H

m process = gg fusion FWHM = 3.77 GeV = 1.56 GeV

CB

σ = 1.331

CB

α = -394 MeV

CB

µ ∆ = 90.46 %

CB

f = 3.69 GeV

G

σ

ATLAS work in progress

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 32 / 72

SLIDE 33

Search for a Higgs Boson in H → Zγ → ℓℓγ decays Background Properties

Background properties

Choice of fit functions and fit range so that significance is maximum while potential bias is < 20 % of fitted signal uncertainty Fit mℓℓγ distribution in [115, 170] GeV on data

[GeV]

γ ll

m 120 130 140 150 160 170 Events/GeV 100 200 300 400 500 600

Data 50) ×

SM

σ =125 GeV,

H

(m γ Z → H

=7 TeV s ,

1

Ldt = 4.5 fb

∫

=8 TeV s ,

1

Ldt = 20.3 fb

∫

ATLAS

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 33 / 72

SLIDE 34

Search for a Higgs Boson in H → Zγ → ℓℓγ decays Parametrization

Expected signal and background yields

In [-5, +5] GeV mass window around mH = 125 GeV: background number (NB) is extracted in background-only data fit.

√s ℓ Category Model NS NB ND

NS

√

NB

FWHM [TeV] [GeV] 8 µ high pTt

Exp. of 2nd order pol.

2.3 310 324 0.13 3.8 8 µ low pTt, low |∆η| 5th order Chebyshev pol. 3.7 1600 1587 0.09 3.8 8 µ low pTt, high |∆η| 4th order Chebyshev pol. 0.8 600 602 0.03 4.1 8 e high pTt

Exp. of 2nd order pol.

1.9 260 270 0.12 3.9 8 e low pTt, low |∆η| 5th order Chebyshev pol. 2.9 1300 1304 0.08 4.2 8 e low pTt, high |∆η| 4th order Chebyshev pol. 0.6 430 421 0.03 4.5 7 µ high pTt Exponential 0.4 40 40 0.06 3.9 7 µ low pTt 4th order Chebyshev pol. 0.6 340 335 0.03 3.9 7 e high pTt Exponential 0.3 25 21 0.06 3.9 7 e low pTt 4th order Chebyshev pol. 0.5 240 234 0.03 4.0 Inclusive 14 5145 5138

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 34 / 72

SLIDE 35

Search for a Higgs Boson in H → Zγ → ℓℓγ decays Systematic Uncertainties

Systematic uncertainties at mH =125 GeV

Theory uncertainties

cross section: scale uncertainty(∼8% for ggF) and PDFs+αs uncertainty (gg − (qq−)initiated:8(4)%). branching ratio: 9.4%. An extra 5% uncertainty for the missing contributions from H → ℓℓ∗ → ℓℓγ and for the interference with H → Zγ → ℓℓγ.

Main experimental uncertainties in 2012 (2011) analysis

luminosity: 2.8 (1.8) % γ ID efficiency : ∼ 3 % e reconstruction+ID efficiency: ∼ 1.5⊕2.5 (2.5⊕1) % e/γ energy resolution: 10.6 (10.0) % for electron channel 2.7 (3.3) % for muon channel µ momentum resolution: 0.5 (1.5) % signal migration in categories: ∼ 3 %

All systematic uncertainties are taken correlated between 7 TeV and 8 TeV categories, except the one of luminosity.

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 35 / 72

SLIDE 36

Search for a Higgs Boson in H → Zγ → ℓℓγ decays Results

p0

The compatibility between data and the background-only hypothesis

[GeV]

H

m 120 125 130 135 140 145 150 Local p

2

10

1

10 1 10

bserved p

γ Z → H expected p γ Z → SM H = 8 TeV s Data 2012,

1

Ldt = 20.3 fb

∫

= 7 TeV s Data 2011,

1

Ldt = 4.5 fb

∫

ATLAS σ = 1.6

max

Z = 142.0 GeV

H

m σ 1 σ 2

No significant excess with respect to the background is observed.

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 36 / 72

SLIDE 37

Search for a Higgs Boson in H → Zγ → ℓℓγ decays Results

Exclusion limit (CLs)

95% limit on σ(H → Zγ)/σSM(H → Zγ) at mH = 125.5 GeV

expected w/o Higgs boson: 9 expected w/ Higgs boson: 10

bserved : 11

[GeV]

H

m 120 125 130 135 140 145 150 ) γ Z → (H

SM

σ )/ γ Z → (H σ 95% CL limit on 5 10 15 20 25 30 Observed Expected σ 1 ± σ 2 ±

= 7 TeV s ,

1

Ldt = 4.5 fb

∫

= 8 TeV s ,

1

Ldt = 20.3 fb

∫

ATLAS

Expected significance with 300 (3000) fb−1 of data at √s = 14 TeV: 2.3 (3.9) standard deviations

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 37 / 72

SLIDE 38

Observation of the Higgs Boson in γγ Events

Observation of the Higgs boson in γγ events

[Phys. Lett. B716(2012)1-29], [ATLAS-CONF-2013-012] Using 4.8 (20.7) fb−1 of data at √s = 7 (8) TeV Results shown at Moriond EW2013

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 38 / 72

SLIDE 39

Observation of the Higgs Boson in γγ Events Event Selection and Categorization

Event selection and categorization at √s = 7 TeV

Di-photon selection

diphoton trigger at least one primary vertex two photons E (sub)leading

T

> (30)40 GeV within |η| < 1.37 or 1.56 < |η| < 2.37 tight ID criteria isolated : pTcone20 < 2.6 GeV and E Topo

Tcone40 < 6 GeV

VBF category: two forward jets

|ηjet| < 4.5, pjet

T > 25 GeV

jet-vertex-fraction (JVF) > 0.75 ∆ηjj > 2.8 mjj > 400 GeV ∆φjj,γγ > 2.6

The remaining events are classified into 9 categories depending on photon conversion status, pseudorapidity and the di-photon pTt.

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 39 / 72

SLIDE 40

Observation of the Higgs Boson in γγ Events Event Selection and Categorization

Event categorization at √s = 8 TeV

One-lepton category

at least one good lepton
veto events with 84< mγe <94 GeV

E miss

T

Two-dimensional template fit method

The data isolation distribution is fitted with the sum of 2D templates of backgrounds (γγ, γj, jγ, jj). The 2D template of γγ, γj and jγ are 1×1 templates extracted from data

jet template: in jet control sample by reversing tight criteria γ template: events passing tight criteria after subtracting jets

The jet-jet 2D template is extracted in data by reversing tight criteria on both leading and subleading objects. 1D projection on leading object isolation of the 2d fit (left) and the estimated background components mγγ distribution (right).

[GeV]

iso T,1

E

4
2

2 4 6 8 Events / ( 0.25 GeV ) 1000 2000 3000 4000 5000 6000 7000 8000 9000

γ γ j γ +jj γ j +jj γ j+j γ + γ γ Data

1

Ldt = 20.7 fb

∫

= 8 TeV, s Data 2012, > 30 GeV

,2 γ T

> 40 GeV, E

,1 γ T

E

[GeV]

γ γ

m 100 110 120 130 140 150 160 ]

1

[GeV

γ γ

dN/dm 500 1000 1500 2000 2500 3000 3500 4000

γ γ γ j + j γ j j Fit statistical error Total Error Data

Data 2012

1

Ldt = 20.7 fb

∫

= 8 TeV, s

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 41 / 72

SLIDE 42

Observation of the Higgs Boson in γγ Events Signal and Background Modeling

Signal and background modeling

Studies are similar to those of the H → Zγ analysis. In a mass window around mH = 126.5 GeV containing 90% of the expected signal for √s = 8 TeV categories.

Category σCB(GeV ) Observed NS NB NS/√NB Background Model

Unconv. central, low pTt

1.50 911 46.6 881 1.56 Expo-pol 2

Unconv. central, high pTt

1.40 49 7.1 44 1.07 Exponential

Unconv. rest, low pTt

1.74 4611 97.1 4347 1.47 4th order pol.

Unconv. rest, high pTt

1.69 292 14.4 247 0.92 Exponential

Conv. central, low pTt

1.68 722 29.8 687 1.14 Expo-pol 2

Conv. central, high pTt

1.54 39 4.6 31 0.83 Exponential

Conv. rest, low pTt

2.01 4865 88.0 4657 1.29 4th order pol.

Conv. rest, high pTt

1.87 276 12.9 266 0.79 Exponential

Conv. transition

2.52 2554 36.1 2499 0.72 Expo-pol 2 Loose High-mass two-jet 1.71 40 4.8 28 0.91 Exponential Tight High-mass two-jet 1.64 24 7.3 13 2.02 Exponential Low-mass two-jet 1.62 21 3.0 21 0.65 Exponential E miss

T

significance 1.74 8 1.1 4 0.55 Exponential One-lepton 1.75 19 2.6 12 0.75 Exponential Inclusive 1.77 14025 355.5 13280 3.08 4th order pol.

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 42 / 72

SLIDE 43

Observation of the Higgs Boson in γγ Events Signal and Background Modeling

Inclusive diphoton invariant mass mγγ distribution

Inclusive invariant mass distributions of di-photon candidates for the combined √s = 7 TeV and √s = 8 TeV data sample

Events / 2 GeV

2000 4000 6000 8000 10000

ATLAS Preliminary γ γ → H

1

Ldt = 4.8 fb

∫

= 7 TeV, s

1

Ldt = 20.7 fb

∫

= 8 TeV, s Selected diphoton sample Data 2011+2012 =126.8 GeV)

H

Sig+Bkg Fit (m Bkg (4th order polynomial) [GeV]

γ γ

m

100 110 120 130 140 150 160

Events - Fitted bkg

200
100

100 200 300 400 500

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 43 / 72

SLIDE 44

Observation of the Higgs Boson in γγ Events Systematic Uncertainties in 8 (7) TeV categories

Systematic uncertainties in 8 (7) TeV

Uncertainties on the signal yield (for mH = 126.5 GeV)

theory uncertainties on cross section (∼ 8 ⊕ 8%) and BR (4.8%) luminosity : 3.6 (1.8) % γ ID⊕isolation : 2.4⊕1 (8.4⊕1) % γ trigger : 0.5 %

Uncertainties on the mass resolution

EMC energy resolution ⊕ [e → γ extrapolation response]: 14 → 23% pile-up effect: 1.5%

Event migration between categories

material mis-modeling : ∼4% Higgs pT mis-modeling: 1 → 10% jet energy scale and resolution underlying event modeling: w/o multi-parton interaction modeling of ∆φγγ,jj and η∗ variables E miss

T

uncertainty

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 44 / 72

SLIDE 45

Observation of the Higgs Boson in γγ Events Results in H → γγ Channel

Results in H → γγ channel

Expected and observed p0 at mH =126.5 GeV:

expected : 4.1 σ

bserved : 7.4 σ

Best-fit mass: mH = 126.6 ± 0.2(stat) ± 0.7(syst) GeV Inclusive signal strength: µ = 1.65+0.24

−0.24(stat)+0.25 −0.18(syst). The excess

w.r.t. µ = 1 is at level of 2.3 σ.

[GeV]

H

m 110 115 120 125 130 135 140 145 150 Local p

14

10

12

10

10

10

8

10

6

10

4

10

2

10 1

2

10

4

10

5

10 σ 1σ 2 σ 3 σ 4 σ 5 σ 6 σ 7

(category) Observed p (category) Expected p (inclusive) Observed p (inclusive) Expected p

= 7 TeV s Data 2011,

1

Ldt = 4.8 fb

∫

= 8 TeV s Data 2012,

1

Ldt = 20.7 fb

∫

ATLAS Preliminary

γ γ → H

Signal strength 1 2 3 4 5 6

Preliminary ATLAS 2011-2012 = 126.8 GeV

H

m γ γ → H = 7 TeV s ,

1

Ldt = 4.8 fb

∫

= 8 TeV s ,

1

Ldt = 20.7 fb

∫

Total Stat. Syst. µ

ggH+ttH

µ

VBF

µ

VH

µ

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 45 / 72

SLIDE 46

Observation of the Higgs Boson in γγ Events Measurements in Combined Channels

Higgs mass measurement

Channels: H → γγ and H → ZZ ∗ → 4ℓ mγγ

H = 125.98 ± 0.42(stat) ± 0.28(syst) GeV

mZZ→4ℓ

H

= 124.51 ± 0.52 ± 0.04(stat)(syst) GeV Compatibility between the two meaurements: ∆mH = 1.47 ± 0.67(stat) ± 0.28(syst) GeV, corresponding to 2 standard deviations with ∆mH = 0. Combined Higgs boson mass : mH = 125.36 ± 0.37(stat) ± 0.18(syst) GeV

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 46 / 72

SLIDE 47

Observation of the Higgs Boson in γγ Events Measurements in Combined Channels

Signal strength (µ) and spin measurements

At mH = 125.5 GeV, the combined signal strength is µ = 1.30 ± 0.12(stat)+0.14

−0.11(sys).

The compatibility between this measurement and the SM prediction is about 7%. The data favours the SM Higgs boson JP = 0+ w.r.t. 0−, 1±, 2+

m. ) µ Signal strength (

0.5

0.5 1 1.5 2

ATLAS Prelim.

1

Ldt = 4.6-4.8 fb

∫

= 7 TeV s

1

Ldt = 20.3 fb

∫

= 8 TeV s

= 125.5 GeV

H

m

0.28

0.33

+

= 1.57 µ γ γ → H

0.12

0.17

+ 0.18

0.24

+ 0.22

0.23

+

0.35

0.40

+

= 1.44 µ 4l → ZZ* → H

0.10

0.17

+ 0.13

0.20

+ 0.32

0.35

+

0.29

0.32

+

= 1.00 µ ν l ν l → WW* → H

0.08

0.16

+ 0.19

0.24

+ 0.21

0.21

+

0.20

0.21

+

= 1.35 µ

, ZZ*, WW* γ γ → H Combined

0.11

0.13

+ 0.14

0.16

+ 0.14

0.14

+

0.6

0.7

+

= 0.2 µ b b → W,Z H

<0.1 0.4 ± 0.5 ± 0.4

0.5

+

= 1.4 µ

(8 TeV data only)

τ τ → H

0.1

0.2

+ 0.3

0.4

+ 0.3

0.3

+

0.32

0.36

+

= 1.09 µ

τ τ , b b → H Combined

0.04

0.08

+ 0.21

0.27

+ 0.24

0.24

+

0.17

0.18

+

= 1.30 µ

Combined

0.08

0.10

+ 0.11

0.14

+ 0.12

0.12

+

Total uncertainty µ

n

σ 1 ±

(stat.) σ

)

theory sys inc.

(

σ (theory) σ

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 47 / 72

SLIDE 48

Summary Summary

Summary

Photon trigger optimization and efficiency measurement

criteria optimization to reduce trigger rate and pile-up dependence the trigger efficiency uncertainty in the H → γγ analysis is reduced to be less than 0.5%.

Photon identification efficiency measurement

pure photon probes from radiative Z decays matrix method: measure efficiency for photons in wide ET range provides accurate efficiency value, reducing uncertainty from 10%→ 1%

Search for a Higgs boson in H → Zγ → ℓℓγ

no significant excess with respect to the background is observed 95% C.L. limit on the σH→Zγ is 11 times the SM expectation

Observation of the Higgs Boson in γγ Events

7.4 σ observation of the Higgs particle mH = 126.6 ± 0.2(stat) ± 0.7(syst) GeV inclusive signal strength: µ = 1.65+0.24

−0.24(stat)+0.25 −0.18(syst) (corresponding

to 2.3 standard deviations from 1)

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 48 / 72

SLIDE 49

Summary Summary

Publications

Measurements of the photon identification efficiency with the ATLAS detector using 4.9 fb−1 of pp collision data collected in 2011. ATLAS-CONF-2012-123. Observation of a new particle in the search for the Standard Model Higgs boson with the ATLAS detector at the LHC. Phys. Lett. B 716 (2012) 1-29. Measurements of the properties of the Higgs-like boson in the two photon decay channel with the ATLAS detector using 25 fb−1 of proton-proton collision data. ATLAS-CONF-2013-012. Measurement of isolated-photon pair production in pp collisions at √s = 7TeV with the ATLAS detector. JHEP01(2013)086. Search for the Standard Model Higgs boson in the H → Zγ decay mode with pp collisions √s = 7 and 8 TeV. ATLAS-CONF-2013-009. Search for Higgs boson decays to a photon and a Z boson in pp collisions at √s= 7 and 8 TeV with the ATLAS detector. Phys. Lett. B 732 (2014) 8-27.

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 49 / 72

SLIDE 50

Backup

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 50 / 72

SLIDE 51

Backup Spontaneous Symmetry Breaking

Spontaneous symmetry breaking of a global symmetry

The Lagrangian obeys the global symmetry (φ = (φ1 + iφ2)/ √ 2):

L = ∂µφ∗∂µφ − V (φ) V (φ) = µ2φ∗φ + λ(φ∗φ)2

Ground state: minimizing the potential

µ2 > 0, λ > 0 : φ = φ∗ = 0, µ2 < 0, λ > 0: |φ|2 = − µ2

2λ ≡ υ

The Lagrangian under small oscillation around a chose vacuum state (φ1 = υ, φ2 = 0):

η = φ1 − υ, ξ = φ2 L = 1 2 (∂µη)2 − (λυ2)η2

massive scalar particle η

+ 1 2 (∂µξ)2 + 0 ∗ ξ2

massless scalar particle ξ

+ · · ·

higher order terms

One massive particle (η) and one massless particle (ξ) are generated.

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 51 / 72

SLIDE 52

Backup Spontaneous Symmetry Breaking

Spontaneous symmetry breaking of a local symmetry

To keep the Lagrangian invariant under local U(1) phase transformation:

∂µ → Dµ = ∂µ + iqAµ Aµ(x) → A′

µ(x) = Aµ(x) − 1

q ∂µθ(x)

The Lagrangian near the vacuum state (h = φ1 − υ, φ2 = 0):

φ = 1 √ 2 (υ + h) L = 1 2 (∂µh)2 − λυ2h2

massive scalar particle h

+ 1 2 e2υ2A2

µ

gauge field with mass

+ e2υA2

µh + 1

2 e2A2

µh2

Higgs and gauge fields interaction

− λυh3 − 1 4 λh4

Higgs self-interactions

This mechanism gives rise to a mass for the gauge boson Aµ and introduces a new real scalar field (h) with mass, interaction terms between the gauge boson and h, and the self-interaction terms of the h field.

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 52 / 72

SLIDE 53

Backup The Higgs Mechanism

The Higgs mechanism - gauge bosons masses

A doublet of complex scalar fields: φ =

φ†

φ0

=

1 √ 2

φ1 + iφ2

φ3 + iφ4

Symmetry group SU(2)L ⊗ U(1)Y and the transformation laws:

U(1)Y : φ → φ′ = e−iI.θ(x)φ, SU(2)L : φ → φ′ = e−i ˆ

τ.ˆ θ(x)/2φ

∂µ → Dµ = ∂µ + ig 2 ˆ τ ˆ Wµ + ig′ 2 ˆ Bµ

The kinetic density (Dµφ)†Dµφ = υ2

8 [g2(W +)2 + g2(W −)2 + (g2 + g′2)Z 2

µ + 0 ∗ A2 µ]

leads

W +

µ =

1 √ 2 (W 1

µ − iW 2 µ)

W −

µ =

1 √ 2 (W 1

µ + iW 2 µ)

Zµ = 1

g2 + g′2 (gW 3

µ − g′Bµ)

Aµ = 1

g2 + g′2 (g′W 3

µ + gBµ)

MW + = MW − = 1 2 υg MZ = υ 2

g2 + g′2

Mγ = 0 MW MZ = g′

g2 + g′2 ≡ cos(θw)

mh = √ 2λυ2

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 53 / 72

SLIDE 54

Backup The Higgs Mechanism

The Higgs mechanism - fermions masses

Similarly, after spontaneous SU(2)L ⊗ U(1)Y symmetry breaking, the Lagrangian of fermion mass term (taken electron e as an example):

L = −λe 1 √ 2 [

¯

ψνe,L, ¯ ψe,L

L
υ + h
ψe,R + ¯

ψe,R 0, υ + h ψνe,L ψe,L

]

= −λe(υ + h) √ 2 ( ¯ ψe,Lψe,R + ¯ ψe,Rψe,L) = − λeυ √ 2 ¯ ψeψe

electron mass term

− λe √ 2 h ¯ ψeψe

electron-higgs coupling term

The electron mass : me = λeυ

√ 2

The electron coupling to the Higgs boson, λe

√ 2 = me υ , is proportional

to its mass.

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 54 / 72

SLIDE 55

Backup Constraints on the SM Higgs Boson Mass

Constraints on the SM Higgs boson mass

Theoretical bounds: the existence of the Higgs boson leads to the cancellation of the ultraviolet divergence in the scattering amplitude

f W +

L W − L → W + L W − L , provided the Higgs boson mass is not too

heavy: mH <

4π

√ 2 3GF ∼ 700 GeV Experimental direct limits (95 % CL), til June of 2012, experiments from LEP, Tevatron and LHC. Exclusion region: mH < 117.5 GeV, mH > 128 GeV, mH within [118.5, 122.5] GeV.

1 10 100 110 120 130 140 150 160 170 180 190 200 1 10 mH (GeV/c2) 95% CL Limit/SM

Tevatron Run II Preliminary, L ≤ 10.0 fb-1

Observed Expected w/o Higgs ±1 s.d. Expected ±2 s.d. Expected LEP Exclusion

Tevatron +ATLAS+CMS Exclusion

SM=1 Tevatron + LEP Exclusion

CMS Exclusion ATLAS Exclusion ATLAS Exclusion LEP+ATLAS Exclusion

ATLAS+CMS Exclusion ATLAS+CMS Exclusion

June 2012

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 55 / 72

SLIDE 56

Backup Constraints on the SM Higgs Boson Mass

Constraints on the SM Higgs boson mass

Besides the direct limits, Hfitter group provides constraints from a global fit on the electroweak parameters. Input parameters : W , Z, top masses and widths, cross sections Best fitted mH = 94+25

−22 GeV [1.3 σ deviation from the ATLAS and

CMS measurements]. Using the measured mH as input, the fitted MW and mt are in good agreements with their experimental measurements.

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 56 / 72

SLIDE 57

Backup Photon Performance

Photon identification (9 discriminating variables)

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 57 / 72

SLIDE 58

Backup Photon Performance

FSR photons from radiative Z decays

ET spectra of FSR photon candidates selected from Z → ℓℓγ decays in data at √s = 8 TeV.

[GeV]

γ T

E 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 [1/GeV]

T

/dE

γ

dN

1

10 1 10

2

10

3

10

4

10

γ Data 2012 Unconverted = 8 TeV s ,

1

Ldt = 20.7 fb

∫

γ ll → Z γ µ µ → Z γ ee → Z

ATLAS Preliminary

[GeV]

γ T

E 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 [1/GeV]

T

/dE

γ

dN

1

10 1 10

2

10

3

10

4

10

γ Data 2012 Converted = 8 TeV s ,

1

Ldt = 20.7 fb

∫

γ ll → Z γ µ µ → Z γ ee → Z

ATLAS Preliminary

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 58 / 72

SLIDE 59

Backup Photon Performance

Electron extrapolation

Exploiting the similarity of EM showers of electrons and photons. Mapping electron shower shapes to photon ones using Smirnov transformation from MC samples. Electrons from Z → ee (9000k). Systematic uncertainties:

difference of e/γ EM shower correlations and kinematics: < 2%, distorted material: < 4%, residual background in electron sample < 1 %.

φ

R 0.4 0.5 0.6 0.7 0.8 0.9 1 pdf

4

10

3

10

2

10

1

10

0.6 ≤ | η 30 GeV and 0.1 < | ≤ T 25 GeV < E electrons photons

ATLAS Preliminary Simulation

φ

R 0.4 0.5 0.6 0.7 0.8 0.9 1 CDF 0.2 0.4 0.6 0.8 1 ATLAS Preliminary Simulation

electrons photons

electrons 0.4 0.5 0.6 0.7 0.8 0.9 1 photons 0.4 0.5 0.6 0.7 0.8 0.9 1

Smirnov Transform

ATLAS Preliminary Simulation

φ

R 0.4 0.5 0.6 0.7 0.8 0.9 1 pdf

4

10

3

10

2

10

1

10

Transformed electrons photons

ATLAS Preliminary Simulation

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 59 / 72

SLIDE 60

Backup Photon Performance

Pile up dependence of photon identification efficiency

Using photons from radiaitive Z decays Photon tight-ID efficiency vs Number of primary vertex

5 10 15 20 25 ID

ε

0.5 0.55 0.6 0.65 0.7 0.75 0.8

γ Unconverted = 8 TeV s ,

1

Ldt = 20.7 fb

∫

data-driven γ Z->ll simulation γ Z->ll

ATLAS Internal

PV

N 2 4 6 8 10 12 14 16 18 20 22 24

Data

ε /

MC

ε

0.96 0.98 1 1.02 1.04 5 10 15 20 25 ID

ε

0.5 0.55 0.6 0.65 0.7 0.75 0.8

= 8 TeV s ,

1

Ldt = 20.7 fb

∫

γ Converted data-driven γ Z->ll simulation γ Z->ll

ATLAS Internal

PV

N 2 4 6 8 10 12 14 16 18 20 22 24

Data

ε /

MC

ε

0.96 0.98 1 1.02 1.04

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 60 / 72

SLIDE 61

Backup Search for a Higgs Boson in H → Zγ → ℓℓγ (ℓ = µ, e)

Improvements from corrections

Improvements on signal resolution: together with the identified primary vertex, collinear FSR photons correction and Z mass constraints

[GeV]

γ µ µ

m 105 110 115 120 125 130 135 140 145 ]

1

[GeV

γ µ µ

1/N dN/dm 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16 0.18 0.2 0.22 0.24 ATLAS Simulation

µ µ → , Z γ Z → H → gg =8 TeV s =125 GeV,

H

m

γ µ µ

raw m FWHM = 4.82 GeV

γ µ µ

corrected m FWHM = 3.80 GeV

[GeV]

γ e e

m 105 110 115 120 125 130 135 140 145 ]

1

[GeV

γ e e

1/N dN/dm 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16 0.18 0.2 ATLAS Simulation

ee → , Z γ Z → H → gg =8 TeV s =125 GeV,

H

m

γ ee

raw m FWHM = 5.47 GeV

γ ee

corrected m FWHM = 4.10 GeV

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 61 / 72

SLIDE 62

Backup Search for a Higgs Boson in H → Zγ → ℓℓγ (ℓ = µ, e)

Decomposition of the expected signal in H → Zγ → ℓℓγ

Decomposition of the expected signal from various production processes at mH = 125.5 GeV

signal composition (%) 10 20 30 40 50 60 70 80 90 100

γ Z

η ∆ high

T t

Low p

γ Z

η ∆ low

T t

Low p

T t

High p Inclusive

ggF VBF WH ZH ttH ATLAS Simulation µ µ → , Z γ Z → H = 8 TeV s = 125 GeV

H

m signal composition (%) 10 20 30 40 50 60 70 80 90 100

γ Z

η ∆ high

T t

Low p

γ Z

η ∆ low

T t

Low p

T t

High p Inclusive

ggF VBF WH ZH ttH ATLAS Simulation ee → , Z γ Z → H = 8 TeV s = 125 GeV

H

m signal composition (%) 10 20 30 40 50 60 70 80 90 100

T t

Low p

T t

High p Inclusive

ggF VBF WH ZH ttH ATLAS Simulation µ µ → , Z γ Z → H = 7 TeV s = 125 GeV

H

m signal composition (%) 10 20 30 40 50 60 70 80 90 100

T t

Low p

T t

High p Inclusive

ggF VBF WH ZH ttH ATLAS Simulation ee → , Z γ Z → H = 7 TeV s = 125 GeV

H

m

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 62 / 72

SLIDE 63

Backup Search for a Higgs Boson in H → Zγ → ℓℓγ (ℓ = µ, e)

Exclusion limit (CLs)

95% CL limits on the production cross section of a Higgs boson decaying to Zγ

[GeV]

H

m 120 125 130 135 140 145 150 ) [pb] γ Z → x BR(H σ 95% CL limit on 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2

Observed Expected σ 1 ± σ 2 ± SM cross section 25 x SM cross section

= 7 TeV s ,

1

Ldt = 4.5 fb

∫

ATLAS [GeV]

H

m 120 125 130 135 140 145 150 ) [pb] γ Z → x BR(H σ 95% CL limit on 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

Observed Expected σ 1 ± σ 2 ± SM cross section 10 x SM cross section

= 8 TeV s ,

1

Ldt = 20.3 fb

∫

ATLAS

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 63 / 72

SLIDE 64

Backup Search for a Higgs Boson in H → Zγ → ℓℓγ (ℓ = µ, e)

CMS H → Zγ → ℓℓγ (ℓ = µ, e) analysis

5.0 (19.6) fb−1 data at √s = 7 (8) TeV. Events are classified into 11 categories according to the ηℓ1, ηℓ2, ηγ, photon conversions and requirements for VBF events. The observed and expected limits for mℓℓγ = 125 GeV are within one

rder of magnitude of the SM prediction.

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 64 / 72

SLIDE 65

Backup Search for a Higgs Boson in H → Zγ → ℓℓγ (ℓ = µ, e)

Interplay between H → Zγ and H → γγ channels

In the inert Higgs Doublet Model (IDM), decay rates of H → γγ (Rγγ) and H → Zγ (RZγ) normalized by their respective SM expectations:

The continuous line represents the rates are enhanced or suppressed by the same parameters associated with charged scalar The straight line below (1,1) is for the case that rates are damped by a big common constant from invisible decay channels where the Higgs boson decays to inert scalars (dark scalars or D-scalars).

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 65 / 72

SLIDE 66

Backup Observation of the Higgs Boson in γγ Events

Decomposition of the expected signal in H → γγ

Decomposition of the expected signal from various production processes at mH = 126.5 GeV for √s = 8 TeV

signal composition (%) 10 20 30 40 50 60 70 80 90 100

One-lepton significance

miss T

E Low-mass two-jet Tight high-mass two-jet Loose high-mass two-jet

Conv. transition

Tt

Conv. rest high p

Tt

Conv. rest low p

Tt

Conv. central high p

Tt

Conv. central low p

Tt

Unconv. rest high p

Tt

Unconv. rest low p

Tt

Unconv. central high p

Tt

Unconv. central low p

Inclusive

ggF VBF WH ZH ttH ATLAS γ γ → Preliminary (simulation) H

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 66 / 72

SLIDE 67

Backup Observation of the Higgs Boson in γγ Events

Systematic uncertainties in 8 (7) TeV categories

Uncertainties on the mass measurement (only for mass measurement)

electron energy scale: 0.3% material effect (e → γ): 0.3% presampler energy scale: 0.1% non-linearities of the EM calorimeter electronics: 0.15% corrections for later energy leakage: 0.1% fraction of photon conversion: 0.13% relative calibration of the first and second sampling of the EMC: 0.2% photon direction: 0.03% background modeling: 0.1%

In total: 0.45%⊕0.32%, corresponding to 0.6 GeV⊕0.4 GeV.

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 67 / 72

SLIDE 68

Backup Observation of the Higgs Boson in γγ Events

Changes in mass measurement

Previous combined results (July 2013): mH = 125.5 ± 0.2+0.5

−0.6 GeV

(2.4 σ tension: ∆m = 2.3+0.6

−0.7 ± 0.6 GeV)

Changes in new measurements

new MVA-based e/γ calibration

10% improvement on H → γγ mass resolution mass shift expectation: −0.45 ± 0.35 GeV

improved e, γ and µ energy scale uncertainties

larger clean control samples: 5.5M Z → ee, 4.3M Z → µµ, 34M W → eν, 0.2M Jψ → ee and 0.2M Z → ℓℓγ 17% to 55% (depending on category) improvement for H → γγ channel 10% improvement for H → ZZ channel

H → γγ: optimized event categorization for mass measurement (20% improvement on σmH w.r.t inclusive analysis) H → ZZ: using 2D likelihood fit instead of using 1D fit(8% improvement on σmH)

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 68 / 72

SLIDE 69

Backup Observation of the Higgs Boson in γγ Events

Limit on the Higgs width in ATLAS

Observed (expected) limits from H → γγ and H → ZZ ∗ analysis are < 2.5 (6.2) and 5.0 (6.2) GeV. Higgs width from off-shell Higgs decay (ΓH/ΓSM

H ) using H → 4ℓ and

H → ℓℓνν

combined observed (expected) limit: µoff −shell < 6.7 (7.3) at 95 % C.L. using the H → 4ℓ µon−shell : ΓH/ΓSM

H

< 4.1 at 95% C.L.

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 69 / 72

SLIDE 70

Backup Observation of the Higgs Boson in γγ Events

The spin and parity quantum numbers

The JP = 0+ hypothesis of the Higgs boson has been compared to various hypotheses in three di-boson decay channels

H → γγ: JP = 0+, 2+ H → ZZ → 4ℓ: JP = 0±, 1±, 2+ H → WW → eνµν/µνeν: JP = 0±, 2+

The data favours the JP = 0+

JP = 0− is rejected at 97.8% C.L. JP = 1± is rejected at 99.7% C.L. JP = 2+ is rejected at 99.9% C.L.

= 0

P

J

+

= 1

P

J

= 1

P

J

m +

= 2

P

J

ATLAS

4l → ZZ* → H

1

Ldt = 4.6 fb

∫

= 7 TeV s

1

Ldt = 20.7 fb

∫

= 8 TeV s

γ γ → H

1

Ldt = 20.7 fb

∫

= 8 TeV s

ν e ν µ / ν µ ν e → WW* → H

1

Ldt = 20.7 fb

∫

= 8 TeV s

σ 1 σ 2 σ 3 σ 4

Data expected

s

CL

+

= 0

P

assuming J

σ 1 ±

)

alt P

( J

s

CL

6

10

5

10

4

10

3

10

2

10

1

10 1 LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 70 / 72

SLIDE 71

Backup Observation of the Higgs Boson in γγ Events

Latest results from other channels at mH = 125 GeV

H → τ +τ −(3 channels: lep-lep, lep-had, had-had): the observed (expected) significance corresponds to 4.1 (3.2) σ, µ = 1.4+0.5

−0.4.

H → µµ(only 8 TeV data): the observed (expected) limit is 9.8 (8.2). t¯ tH(→ b¯ b): the observed (expected) limit is 4.1 (2.6). VH → b¯ b: the observed (expected in the absence of signal) is 1.4 (1.3).

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 71 / 72

SLIDE 72

Backup Observation of the Higgs Boson in γγ Events

Expectation in Run-II

Relative uncertainty on the total signal strength µ for mH = 125 GeV

µ / µ ∆ 0.2 0.4

(comb.) (incl.) (comb.) (comb.) (VBF-like) (comb.)

ATLAS Simulation Preliminary

= 14 TeV: s

1

Ldt=300 fb

∫

;

1

Ldt=3000 fb

∫

µ µ → H τ τ → H ZZ → H WW → H γ Z → H γ γ → H µ µ → H τ τ → H ZZ → H WW → H γ Z → H γ γ → H

1.5 →

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 72 / 72

SLIDE 73

Backup Observation of the Higgs Boson in γγ Events

Crystal Ball function (from WIKIPEDIA)

LIU Kun (LPNHE&USTC) Thesis defense June 24th, 2014 73 / 72