BSM Higgs Searches in ATLAS and CMS (part 2) · BSM Higgs Searches in ATLAS and CMS (part 2) On...

Livia Soffi

“Sapienza” Università di Roma - INFN Sez. di Roma

BSM Higgs Searches in ATLAS and CMS

(part 2)

On Behalf of the ATLAS and CMS Collaborations

lunedì 21 luglio 2014

Livia Soffi BSM Searches - Higgs Hunting 2014 2

Overview• Vital question: Is it in fact the discovered Higgs boson from the SM or part of an extended sector?

• Space to probe any non-SM property

• Additional Higgs bosons still a possibility

?

• Indirect searches from observed Higgs couplings measurement (not in this talk)



Outline

• Rare Higgs Decays

• Invisible Higgs Decays

• Higgs Decays to Long-Lived

• Additional Higgs in multilepton and photons channels

• MultiHiggs in cascade

• Di-Higgs production in diphoton and di b-jets channels

• New diphoton resonances

•Talk focuses on results from ATLAS and CMS

NEW

NEW

NEW

Non-SM property

Additional Higgs bosons

Many new results from ICHEP and after:

• Lepton Flavour Violation NEW



Rare decays

CMS PAS HIG-14-003

ATLAS: arXiv:1402.3051

CMS: Phys. Lett. B 726(2013) 587



• Rare Higgs decays as probes of new couplings and SM extensions• Loop and tree level processes contribute to μμγ final state

Search for

mμμ< 100 GeV Higher mμμ

H → γ∗γ → µµγ

• First CMS search for Dalitz decays with γ* internal conversion in μμ

(GeV)Hm120 125 130 135 140 145 150

SM!/

!95

% C

L lim

it on

0

5

10

15

20

25

30CMS Preliminary

"µµ#"*"#H-1 = 19.7 fbintL = 8 Tev; s

ObservedExpected

! 1±Expected ! 2±Expected

@125 GeV CMS ATLAS

Z γ 9.5X SM 11X SM

γ* γ 10X SM

H

Z/γ

γ γ

γ

Z/γ

μ

μ

μ

μ

μ

μ

H H

• mμμ < 20 GeV to separate γ* γ and Z γ



CMS PAPER HIG-13-030

Invisible decays

ATLAS-CONF-2014-011



Higgs decay to invisible particles

•Search in the VBF and ZH (Z →ll; Z →bbar) modes

Large cross-section Large SM background reduced by VBF jet topology

Lower cross-section Clear topologySensitivity increase by the leptons an bs

• One possibility: Higgs portal of DM interaction mDM< mH/2

• What if Higgs couples to something invisible?

•Higgs mediator between SM and DM particles

HDM

DM



Results

• No evidence for signal observed in any of the three searches

[GeV]!MDM Mass 10 210 310

[pb]

SI -N!

"DM

-nuc

leon

cro

ss s

ectio

n

-1310

-1210

-1110

-1010

-910

-810

-710

-610

-510

-410

-310

-210

-110

!CRESST 1!CRESST 2

XENON100(2012)XENON10(2011)DAMA/LIBRACoGeNT(2013)/90%CLCoGeNT(2013)/99%CLCDMS(2013)/95%CLCOUPP(2012)LUX(90%CL)

CMS invisible"ZH, H Combination of VBF and

= 125 GeVHm inv) < 0.51 @ 90% CL#B(H

(VBF+ZH)-1 = 8.0 TeV, L = 18.9-19.7 fbs (ZH)-1 = 7.0 TeV, L = 4.9 fbs

vector

scalar

fermion

MinLatticeMax

• Upper limit on BR(H → inv) : constrain the DM mass and its elastic cross section on nucleons

• 95% CL Upper Limits on σ · B(H → inv)

@125 GeV CMS ATLASZ(→ll)H 0.75XSM 0.75X SM

Z(→bb)H 1.82X SM -VBF 0.69X SM -

COMB 0.58X SM -•CMS combination paper just accepted for publication.

•Results interpreted in the Higgs portal of DM interaction model



Decays to long-lived particles

ATLAS-CONF-2014-041 NEW



Scalar boson decay to long-lived particles

• Hidden Valley Benchmark Model: coupling via a heavy scalar particle, ΦHS

• What if Higgs couples to Long-Lived particles?

CMS: search for displaced vertex using tracks

ATLAS:Decays in HCal/ECal NEW

ΦHS

πv

πv

ffbar

ffbar

JET

JET

• Lifetime of the πv is free parameter

Both πv decay in the hadronic calorimeter or near the outer edge of the electromagnetic calorimeter

πv decays result as displaced vertex

• πv neutral and long-lived



Analysis Strategy

• Dedicated trigger:

At least one narrow jet with no charged tracks associated

Requirement on the EH/EEM

Average probability to fire the trigger ~ 20% in EB and 6% in EE

Timing of the jet used to discard out-of-time background•Non collision background:

• 95% CL Upper Limits on cross-section times BR vs πv proper decay length

r [m]1 1.5 2 2.5 3 3.5 4 4.5 5

v!

prob

abilit

y

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35ATLASSimulation Preliminary

10 GeVv!

126 GeV - m"

m

25 GeVv!

126 GeV - m"

m

40 GeVv!

140 GeV - m"

mbarrel

proper decay length [m]v

!

-110 1 10

SM" / "

95%

CL

Upp

er L

imit

on

-110

1

10 GeVv!

126 GeV - m#

m

25 GeVv!

126 GeV - m#

m

40 GeVv!

126 GeV - m#

m

10 GeVv!

126 GeV - m#

m

25 GeVv!

126 GeV - m#

m

40 GeVv!

126 GeV - m#

m

10 GeVv!

126 GeV - m#

m

25 GeVv!

126 GeV - m#

m

40 GeVv!

126 GeV - m#

m

ATLAS Preliminary

BR 30%

BR 10%

-1 L dt = 20.3 fb$

= 8 TeVs

BSM Searches - Higgs Hunting 2014



Lepton Flavor Violation

NEWCMS PAS HIG-14-005



LFV at LHC


• What if we observe an unexpected decay of the new boson?

• LFV decays occur naturally in 2HDM, composite Higgs, models with flavor symmetries and Randall-Sundrum

• Constraints from indirect searches: B(H → μτ) < O(0.1), B(H → eτ) < O(0.1)

• First dedicated search for H → μτe and H → μτhad at LHC

•W.r. t. H → τμτhad and H → τμτe:

[GeV]T

pµ0 20 40 60 80 100 120 140 160 180 200

x B

r [pb

] / 5

GeV

!

-510

-410

-310

-210

-110 had" µ

-"+"#hµ " #h

= 8 TeVsCMS simulation preliminary

)!" (e,# $0 0.5 1 1.5 2 2.5 3

x B

r [pb

] / 0

.7%

-310

-210

-110

e& µ -&+&'hµ & 'h

= 8 TeVsCMS simulation preliminary

Harder muon pT

spectrum

MET collinear with visible τ decay

H → μτ



Results


• expected upper limit: B(H → μτ) < (0.75 ± 0.38)% • observed upper limit: B(H → μτ) < 1.57%

), %!µ"95% CL Limit on Br(h0 2 4 6 8 10

1.57% (obs.)0.75% (exp.)

!µ"h3.84% (obs.)3.77% (exp.)

, 2 Jetse!µ

2.38% (obs.)1.66% (exp.)

, 1 Jete!µ

2.04% (obs.)1.32% (exp.)

, 0 Jetse!µ

3.29% (obs.)1.95% (exp.)

, 2 Jetshad!µ

2.11% (obs.)2.10% (exp.)

, 1 Jethad!µ

2.94% (obs.)2.35% (exp.)

, 0 Jetshad!µ

Observed

Expected

# 1±Expected

# 2±Expected

= 8 TeVs, -119.7 fbCMS preliminary

| !µ

|Y-410 -310 -210 -110 1

|

µ!|Y

-410

-310

-210

-110

1 = 8 TeVs, -119.7 fbCMS preliminary

BR

<0.1%

BR

<1%

BR

<10%

BR

<50%

!!"LHC h

observed

expected!µ"h

µ 3"!

# µ "!

2/v!mµ

|=mµ!

Y!µ|Y

slight excess of observed number of events

Best fit: B(H → μτ) = (0.89+0.40)%

Constraint on B(H → μτ) interpreted in terms of LFV Higgs Yukawa couplings



Summary of Higgs decay modes

= preliminary= published

@125 GeV CMSCMSCMS ATLASATLASATLAS

Dataset Status Results Dataset Status ResultsFavoured decay modesFavoured decay modesFavoured decay modesFavoured decay modesFavoured decay modesFavoured decay modesFavoured decay modes

H(ZZ)

5.1 + 19.7 fb-1

HIG-14-009

μ= 1.00 +/- 0.09 (stat)

+0.08-0.07 (theo)+/-0.07(syst)

4.8 + 20.3 fb-1 ATLAS-CONF-2014-009μ= 1.30

+/- 0.12 (stat)+0.14-0.11(syst)

H(WW)

5.1 + 19.7 fb-1

HIG-14-009

μ= 1.00 +/- 0.09 (stat)

+0.08-0.07 (theo)+/-0.07(syst)

4.8 + 20.3 fb-1 ATLAS-CONF-2014-009μ= 1.30

+/- 0.12 (stat)+0.14-0.11(syst)H(γ γ) 5.1 + 19.7

fb-1HIG-14-009

μ= 1.00 +/- 0.09 (stat)

+0.08-0.07 (theo)+/-0.07(syst)

4.8 + 20.3 fb-1 ATLAS-CONF-2014-009μ= 1.30

+/- 0.12 (stat)+0.14-0.11(syst)

H(tau tau)

5.1 + 19.7 fb-1

HIG-14-009

μ= 1.00 +/- 0.09 (stat)

+0.08-0.07 (theo)+/-0.07(syst)

4.8 + 20.3 fb-1 ATLAS-CONF-2014-009μ= 1.30

+/- 0.12 (stat)+0.14-0.11(syst)

V-H(bb)

5.1 + 19.7 fb-1

HIG-14-009

μ= 1.00 +/- 0.09 (stat)

+0.08-0.07 (theo)+/-0.07(syst)

4.8 + 20.3 fb-1 ATLAS-CONF-2014-009μ= 1.30

+/- 0.12 (stat)+0.14-0.11(syst)

ttH(bb)

5.1 + 19.7 fb-1

HIG-14-009

μ= 1.00 +/- 0.09 (stat)

+0.08-0.07 (theo)+/-0.07(syst)

20.3 fb-1 ATLAS-CONF-2014-011 μ< 4.1(3.4)

VBF-H(bb) 19.0 fb-1 HIG-13-011 μ< 3.6(3.0) 4.7 + 13.0 fb-1 ATLAS-CONF-2012-161 μ< 1.8(1.9)

Rare decay modesRare decay modesRare decay modesRare decay modesRare decay modesRare decay modesRare decay modesH(mu mu ) 5.0+ 19.7 fb-1 HIG-13-007 μ< 7.4 24.8 fb-1 arXiv:1406.7663 μ< 7.2

H(Z γ) 5.0+ 19.6 fb-1 arXiv:1307.5515μ< 10

4.5 + 20.3 fb-1 arXiv:1402.3051 μ< 10

H(γ* γ) 19.7 fb-1 HIG-14-003μ< 10

- - -

Invisible decay modesInvisible decay modesInvisible decay modesInvisible decay modesInvisible decay modesInvisible decay modesInvisible decay modesZ(ll)-H(inv)

4.9 + 19.7 fb-1 arXiv:1404.1344 BR< 58(46)%

4.5 + 20.3 fb-1 arXiv:1402.3244 BR< 75(62)%

Z(bb)-H(inv) 4.9 + 19.7 fb-1 arXiv:1404.1344 BR< 58(46)% - - -

VBF-H(inv)

4.9 + 19.7 fb-1 arXiv:1404.1344 BR< 58(46)%

- - -Exotic decay modesExotic decay modesExotic decay modesExotic decay modesExotic decay modesExotic decay modesExotic decay modes

H(tau mu) 19.7 fb-1 HIG-14-005 BR<1.57% - - -H(long-lived) - - - 20.3 fb-1 ATLAS-CONF-2014-041 UL vs ctau


http://arxiv.org/abs/1406.7663


http://arxiv.org/pdf/1307.5515.pdf









Extended Higgs Sector IntroductionCMS ATLAS

EWK Singlet ModelEWK Singlet Model Future V

2HDM

MSSM H(tau tau) V V

2HDM

MSSM H(bb) V -

2HDM

H+(tau nu) V V

2HDMH+(tau jet) - V

2HDM H+(csbar) V -2HDMH(hh),A(Zh) V V

2HDM

H(multi γ) - Future

2HDM

H(ttbar) Future -

2HDM

Heavy H cascade - V

NSSM

h1->a1->2mu V Future

NSSM

h1->a1a1->4mu V -NSSM h2->h1h1->4tau Future -NSSM

h2->h1h1->2tau2mu - FutureNSSM

h2->h1h1->2tau2b Future -

NSSM

H+(Wa1) - Future

Resonant searches

Low Mass H(γ γ) - V

Resonant searches

HighMass H(γ γ) V VResonant searches

HighMass H(γ γbb) V VResonant searches HighMass H(bbbb) - VResonant searches

HighMass WW V V

Resonant searches

HighMass ZZ V V



Heavy Higgs decays to h

CMS PAS HIG-13-025

ATLAS: CERN-PH-EP-2013-173



2HDM Overview

• Five physical Higgs sector particles survive EWSB with masses < TeV and accessible at LHC (h, H,A, H+-)

Multilepton signature with unusual kinematics characteristics

Resonant decay of the SM-Higgs h to two photons

• If mH and mA> 2mh H → hh and A → Zh promising avenues for discovery even when the couplings of the light Higgs within a few percent of SM predictions.

H → hh A → Zh

• h has a nominal mass of 126 GeV and Brs to WW, ZZ, ττ, bb and γγ channels appropriate to SM



Interpretation in 2HDM

• Search procedure in exclusive channels depending upon the number of flavor of leptons, hadronic taus, photons, jet flavors and missing energy.

• Observed and expected limits in the 2HDM for masses in the range [260-360] GeV

• α and β determine cross-section and BRs for H and A production and decays

)!-"cos(-0.6 -0.4 -0.2 0 0.2 0.4 0.6

"ta

n

-110

1

10

210CMS Preliminary -1 = 19.5 fbt dL# = 8 TeV, s

hh$TYPE I 2HDM H

= 300 GeVHm95% C.L. CLs Limits

ObservedNLO expected

%1±NLO expected %2±NLO expected

)!-"cos(-0.6 -0.4 -0.2 0 0.2 0.4 0.6

"ta

n

-110

1

10

210CMS Preliminary -1 = 19.5 fbt dL# = 8 TeV, s

Zh$TYPE I 2HDM A

= 300 GeVAm

Zh$TYPE I 2HDM A

= 300 GeVAm95% C.L. CLs Limits

ObservedNLO expected

%1±NLO expected %2±NLO expected

EXCLUDED

EXCLUDED

EXCLUDED

EXCLUDED EXCLUDED

cos(β -α ) = 0 : Decoupling limit: h behaves exactly like in SM



ATLAS search for multi-Higgs sector

• Single heavy neutral H decays to charged H+-

and a W. H+-

decaying to W and h and h to bbar pair.

• One W assumed to decay hadronically and the other leptonically

• Signal topology:

1. One lepton + MET2. At least 4 jets - 2 b-tagged

• Main background contributions:

1. ttbar (~90%) Estimated from simulation and validated in control regions2. V+ jets

3. Multi-jets Small. Estimated from data

• MultiHiggs Cascade relevant for mH > 800 GeV



Results

• Observed and expected 95% CL model independent upper limits on cross-section

• Analysis performed for any combination of mH and mH+-

• Gain sensitivity for Very High Masses

= 1025, 225 GeV±H, m0H

Trained for m BDT output

-1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8

Even

ts

1

10

210

310

410

510 -1 L = 20.3 fb!Data

ttW+jets

OtherBDT threshold

= 225 GeV±Hm = 1025 GeV0Hm

Signal (1.00 pb)

ATLAS = 8 TeVs

= 1025, 225 GeV±H, m0H

Trained for m BDT output

-1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8

(Dat

a - S

M) /

SM

-0.6-0.4-0.2

00.20.40.6

Uncertainty

•Counting experiment with events passing the BDT output threshold

•Multivariate analysis to discriminate signal and main bkg ttbar

Mass [GeV]0H400 500 600 700 800 900 1000

Mas

s [G

eV]

±H

300

400

500

600

700

800

900

95%

C.L

. Upp

er L

imits

[pb]

-110

1

10-1Ldt = 20.3 fb!

ATLAS = 8 TeVs

±W±Wbb"±W±W0h"±H±W"0H



X->hhCMS PAS HIG-13-032

ATLAS: CERN-PH-EP-2014-113

NEW

ATLAS-CONF-2014-005 NEW



Overview

CMS ATLASγ γ bb V Vγ γ bb

non resonant Vbbbb V• Non Resonant SM Higgs pair production not

expected to be observable at LHC 8 TeV

(GeV)!! m100 110 120 130 140 150 160 170 180

Fra

ctio

n of

eve

nts

/ 2 G

eV

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7 Signal shapes = 300 GeVXM = 500 GeVXM = 700 GeVXM = 1000 GeVXM

standard model Higgs

bb!! " HH "X

= 8 TeVsCMS Preliminary Simulation

• γ γ bb: Large BR of the H → bbar and the low background and good resolution of the H → γγ channel

• Model independent analyses: Results interpreted in terms of Graviscalars or Radion production

• bbbb: More sensitive at high mass

• Heavy H resonant search performed in channels which allow full reconstruction of the decay chain

X → hh → ?



γ γ bb Analysis Strategy•CMS[260-1100]GeV: Analysis performed in two ranges: mX<400 GeV and mX > 400 GeV

• Low mass: signal extracted from a fit to the m γγ data spectrum

• High mass: similar procedure with a fit to the m γγjj

(GeV)!!m100 110 120 130 140 150 160 170 180

Even

ts /

( 1 G

eV )

1

2

3

4

5

6

= 8 TeV s -1 CMS Preliminary L = 19.7 fb

= 300 GeVX mData Fit

" 2 ±Fit " 1 ±Fit

High puritybb!! # HH #X

[GeV]!!m110 120 130 140 150 160

Even

ts /

2.5

GeV

020406080

100< 2 b-Tag Control Region

[GeV]!!m

Even

ts /

2.5

GeV

0

2

4

6

8

10

Signal RegionDataFitted Signal + BkdsSingle Higgs Boson + BkdContinuum Background

ATLAS

" = 8 TeVs, -1Ldt = 20 fb

• Non resonant search:

Fit to the unbinned m γγ spectrum

• Resonant production search:Counting experiment due to small number of expected events

•ATLAS [260-550]GeV:


Not possible to fit a bump in the m γγjj below 400 GeV m γγjj has kinematic peak ~300 GeV



bbbb Analysis Strategy

• 4-tag selection: 4 b-tagged high energy jets

• Kinematics requirement on dijet system to veto ttbar events

• Elliptical cut in the plane of the leading and the subleading dijet invariant mass

• >90% background in the signal region from multijet events + 10% ttbar

Multijet: m4j shape and normalization from datattbar: m4j shape from MC and normalization from data

XHH =

��m1

jj− m̃1

jj

σm1jj

�2

+

�m2

jj− m̃2

jj

σm2jj

�2



Results• 95% CL Upper Limits set on the cross-section times BR of the process

• Resonant Searches: Results interpreted in terms of KK-graviton, radion and 2HDM models

• Non Resonant Search assuming SM BR(hh): Exp (Obs) 1.0 (2.2) pb

[GeV]G*m600 800 1000 1200 1400

) [fb

]bbb

b→

HH

→

G*)

x B

R(G

* →

(pp

σ

1

10

210

Expected Limit (95% CL)σ 1±Expected σ 2±Expected

Observed Limit (95% CL) = 1.0PlanckMRS Graviton, k/

ATLAS Preliminary -1 Ldt = 19.5 fb∫ = 8 TeV: s

[GeV]Xm300 350 400 450 500

hh) [

pb]

! B

R(X

" X#

1

1.5

2

2.5

3

3.5

4

4.5ATLAS$ = 8 TeVs at -1Ldt = 20 fb

Observed 95% CL Limit#1±Expected Limit #2±Expected Limit

Type I 2HDM:)=-0.05%-&=1, cos(& tan

(GeV)X m300 400 500 600 700 800 900 1000 1100

HH

) (fb

)!

BR

(X

" X

) !

(pp

#

210

310

410

510 WED: kl = 35, k/Mpl = 0.2, elementary top, no r/H mixing = 3 TeV)R$radion ( = 1 TeV)R$radion (

RS1 KK-graviton Bulk KK-graviton

Observed 95% upper limitExpected 95% upper limit

# 1 ±Expected limit # 2 ±Expected limit

= 8 TeVs -1CMS (Unpublished) L = 19.7 fb

bb%% ! HH !X

4 b-jets

γ γ bb


γ γ bb



X->γγ

ATLAS-CONF-2014-031CMS PAS HIG-14-006 NEW



Overview

CMS ATLASMass Range

[GeV] 150-850 65-600

Width Range[GeV] 0-85 0

Spin 0,2 0

• Method developed for the SM H → γγ channel extended to search for diphoton resonances in a wider mass range

• Model Independent search for local excesses in the diphoton spectrum exploiting a fit technique

200 400 600 800 1000

Even

ts/1

0.00

-110

1

10

210

310

410All Categories Combined

Data + jet!! + !

Bkg Err

/NDF: 2.0642"

(8 TeV)-119.7 fb

CMSPreliminary

[GeV]! !m200 400 600 800 1000D

ata/

Bkg.

0.0001

1.0001

2

0 8.4 16.8 25.2 33.6

pull

-4

-2

0

2

4

0.092±RMS: 1.197 0.131 ±Mean: -0.395

H → γγ interesting in 2HDM in the decoupling limit


Livia Soffi BSM Searches - Higgs Hunting 2014 29 [GeV]! !m300 400 500 600

Even

ts/(6

.7)

1

10

210

310

Category 0

Data

Fit Model

(8 TeV)-119.7 fb

CMSPreliminary

Fit chi square/dof = 0.856Chi square Prob = 0.774

[GeV]! !m100 120 140 160 180 200

Even

ts /

( 1 )

-210

-110

1

10

210Simulation

BW

Resolution"BW

Cat: 0 (8 TeV)-119.7 fb

CMSSimulation

CMS Analysis

• High Mass analysis performed in four classes according to the two photons kinematics properties to increase the search sensitivity

• Parametrized signal model through analytic function with two free parameters: mX and ΓX

• Background estimated fitting directly data assuming negligible signal

Bias Study to validate the fit technique

Sliding window fit range




ATLAS Analysis

60 70 80 90 100 110 120

Even

ts /

2 G

eV

5000

10000

15000

20000

UU categoryDataContinuum+DY fitContinuum component of the fit

60 70 80 90 100 110 120

5000

10000

15000

CU categoryDataContinuum+DY fitContinuum component of the fit

[GeV]γγm60 70 80 90 100 110 120

1000

2000

3000

4000

CC categoryDataContinuum+DY fitContinuum component of the fit

ATLAS Preliminary-1 L dt = 20.3 fb∫ = 8 TeV s

[GeV]γγm100 200 300 400 500 600 700

]-1 [G

eVγγ

dN/d

m-110

1

10

210

310

data = 125 GeV)

XContinuum+H fit (m

= 250 GeV)X

Continuum+H fit (m = 500 GeV)

XContinuum+H fit (m

= 8 TeVs-1 L dt = 20.3 fb∫

ATLAS Preliminary

115 120 125 130 135800

100012001400160018002000

data = 125 GeV)

XContinuum+H fit (mContinuum part of the fit

• Search split into a categorized low-mass [65-110] GeV and a inclusive high mass [110-600] GeV analysis

• Low-mass: Main background from Drell-Yan production estimated from data.

Events categorized according to the number of converted photons

• High-mass: Sliding window fit range using analytic function

SM-Higgs production as background

• Parametrized narrow signal model with mX parameter



200 400 600 800

10

20

30

40

50

60

70

80

-410

-310

-210

All Categories Combined - Observed

UL = 0.002 [pb]

UL = 0.003 [pb]

UL = 0.004 [pb]

UL = 0.010 [pb]

UL = 0.020 [pb]

(pb)

95%

CL

)! ! "

BR

(X

# $

[GeV]Xm

[GeV

]X

%

(8 TeV)-119.7 fbCMS Preliminary

[GeV]Xm100 200 300 400 500 600

BR

[fb]

⋅ fidσ

95%

CL

limit

on

-110

1

10

210

310ATLAS Preliminary

= 8 TeVs -1 L dt = 20.3 fb∫

ObservedExpected

σ 1 ±σ 2 ±

60 80 100 120 140 160

50

100

Results

• No excess observed over the full mass range. 95% CL limits set on the fiducial cross-section times BR

[GeV]Xm200 300 400 500 600 700 800

(pb)

95%

CL) ! !

" B

R(X

# $

-410

-310

-210

-110

(8 TeV)-119.7 fb

CMSPreliminary Width: 0.1 [GeV]

All Categories Combined

Observed

$ 1±Expected

$ 2±Expected



Conclusions

• Search for BMS physics in the Higgs sector:

Directly from decays of neutral and charged Higgses

Indirectly by interpreting measured properties of the light Higgs

• Many analyses completed at ATLAS + CMS on full 8 and 7 TeV data:

No significant excess observed and various cross-section limits and exclusion regions for the parameter space of several models have been provided.

BSM Higgses might be just around the corner...

• 2015 and sqrt(s)=13 TeV will greatly enhance our sensitivity



BACKUP



Invisible:Search in VBF and ZH channels

1. Two final state quarks separated by a large rapidity gap and with high invariant mass

• VBF Signal topology:

2. Large missing energy

1.Z(ll)H(inv): Pair of isolated leptons and High MET - Limited jet activity

• ZH Signal topology:

2.Z(bbar)H(inv): B-tagged jet pair and High MET (same as Z(νν)H(bbar))

Even

ts /G

eV

-310

-210

-110

110

210

310

410

510

610

710 CMS-1 = 8 TeV, L = 19.7 fbs

Z(ll) H(inv)

Observed

inv)=100%!B(H =125GeV),

HZH(m

DY(ll)+jets

,tW,WW,W+jetstt

ZZ

WZ

[GeV]missTE

0 100 200 300 400 500

MC

Dat

a

0

1

2

/10)

radi

ans

!Ev

ents

/ (

-1101

10210310410510610710810910

10101110121013101410 CMS

-1 = 8 TeV, L = 19.7 fbsZ(ll) H(inv)

Observed

inv)=100%"B(H =125GeV),

HZH(m

DY(ll)+jets

,tW,WW,W+jetstt

ZZ

WZ

) [radians]missT

(ll,E!#0 1 2 3

MC

Dat

a

0.9

1

1.1

Angular separation MET-Z system



INVISIBLE:Search in vector boson fusion channel

1. Two final state quarks separated by a large rapidity gap and with high invariant mass


2. Large missing energy

• Main background from V+jets estimated from control regions in data

[GeV]missTE

150 200 250 300 350 400 450 500

Even

ts /

20 G

eV

-110

1

10

210

310

410 CMS-1 = 8 TeV, L = 19.5 fbs

VBF H(inv)

Observed

inv) = 100%!B(H = 125 GeV,HVBF m

V+jets

, tW, DY(ll)+jets, VVtt

[GeV]jjM1500 2000 2500 3000 3500

Even

ts /

100

GeV

-210

-110

1

10

210

310

410 CMS-1 = 8 TeV, L = 19.5 fbs

VBF H(inv)

Observed

inv) = 100%!B(H = 125 GeV,HVBF m

V+jets

, tW, DY(ll)+jets, VVtt

• Signal region defined: MET > 130 GeV && mjj > 1100 GeV && Δη>4.2CMS: Search for invisible decays of Higgs bosons in the vector boson fusion and associated ZH production modes 9

Table 2. Summary of the estimated number of background and signal events, together with the observed yield, in the VBFsearch signal region. The signal yield is given for mH = 125GeV and B(H → inv) = 100%.

Process Event yieldsZ(νν)+jets 99± 29 (stat.)± 25 (syst.)W(µν)+jets 67± 5 (stat.)± 16 (syst.)W(eν)+jets 63± 9 (stat.)± 18 (syst.)W(τhν)+jets 53± 18 (stat.)± 18 (syst.)QCD multijet 31± 2 (stat.)± 23 (syst.)Sum (tt, single top quark, V V , DY) 20.0± 8.2 (syst.)Total background 332± 36 (stat.)± 46 (syst.)VBF H(inv.) 210± 30 (syst.)ggF H(inv.) 14± 11 (syst.)Observed data 390S/B (%) 70

selection for the Z(��)H(inv) signal at mH = 125GeV is5.6%, estimated from MC simulation.

6.3 Background estimation

After the full selection, the dominant backgrounds arisefrom WZ and ZZ processes, which are modeled using MCsimulation. The pre-fit normalization of these backgroundsis obtained from their respective NLO cross sections com-puted with mcfm.

The DY(��)+jets background is modeled from an or-thogonal control sample of events with a single isolatedphoton produced in association with jets (γ + jets). Thischoice has the advantage of providing a large statisticssample, which resembles Z boson production in all im-portant aspects: production mechanism, underlying eventconditions, pileup scenario, and hadronic recoil. The kine-matic distributions and overall normalization of the γ +jets events are matched to Z(��) + jets in data throughevent weights, determined as a function of the Z boson pTmeasured from data. This procedure takes into accountthe dependence of the Emiss

Ton the associated hadronic

activity.Further discrepancies can arise due to differences in

the pileup distribution of the γ + jets sample due to thefact that photon data was collected with triggers whoseprescales varied as a function of photon threshold anddata-taking period. These are taken into account by fur-ther weighting events in the control sample, according tothe distribution of number of reconstructed vertices in thesignal sample. The electroweak backgrounds to the controlsample, involving photons and neutrinos, are subtractedusing predictions from MC simulation.

This procedure yields an accurate model of the Emiss

T

distribution in DY(��)+jets events, as shown in Fig. 4(left), which compares the Emiss

Tdistribution of the weighted

γ+jets events, summed with other backgrounds, to theEmiss

Tdistribution of the dilepton events in data. Figure 4

also compares the distributions of (center) ∆φ(��, Emiss

T)

and (right) |Emiss

T−p��

T|/p��

Tobtained from this background

model to the same distributions in the dilepton sample.Good agreement is found between data and background

predictions. The uncertainties in the electroweak back-ground to the photon control sample yield a 100% un-certainty in the normalization of the residual DY(��)+jetsbackground. However, since the Drell–Yan background af-ter the full selection is very small, the large uncertaintyhas negligible impact on the final results.

The remaining background processes do not involveZ boson production, and are referred to as non-resonantbackgrounds. Such backgrounds arise mainly from lep-tonic W boson decays in tt, tW decays and WW events.Also included in the estimate of non-resonant backgroundsare small contributions from single-top-quark events pro-duced from s- and t-channel processes, W+jets produc-tion, and Z → ττ events in which τ leptons produce elec-trons/muons and Emiss

T.

We estimate these backgrounds using a control samplein data, consisting of events with opposite-charge different-flavor dilepton pairs (e±µ∓) that otherwise pass the fullselection. The backgrounds in the e+e− and µ+µ− finalstates are then estimated by applying scale factors (αee,αµµ) to the number of events in the control sample, Neµ:

Nee = αee ×Neµ, Nµµ = αµµ ×Neµ. (5)

We compute the two factors αee and αµµ in the side-bands (SB) of the Z peak (40 < m�� < 70GeV and 110 <m�� < 200GeV) by using the following relations:

αee =NSB

ee

NSBeµ

, αµµ =NSB

µµ

NSBeµ

, (6)

where NSB

ee, NSB

µµ , and NSB

eµ are the number of eventsin the Z sidebands counted in a top-quark-enriched sam-ple of e+e−, µ+µ−, and e±µ∓ final states, respectively.The requirements for this sample are Emiss

T> 65GeV,

p��T

> 50GeV, 0.4 < Emiss

T/p��

T< 1.8, and a b-tagged jet.

The kinematic requirements are looser than in the sig-nal region, in order to reduce the statistical uncertaintiesin the scale factors. The measured values of these fac-tors with the corresponding statistical uncertainties areα7TeV

ee= 0.42 ± 0.04, α7TeV

µµ = 0.64 ± 0.06 and α8TeV

ee=

0.43±0.02, α8TeV

µµ = 0.69±0.03. The validity of the proce-dure for computing the scale factor is checked by closure



1.Z(ll)H(inv): Pair of isolated leptons and High MET - Limited jet activity


2.Z(bbar)H(inv): B-tagged jet pair and High MET

• Dominant background from boson and diboson production w/o jets estimated from control regions.

• Signal region 1: MET > 120 GeV, Δɸ(ll, MET)>2.7, |MET-pT,ll|/pT,ll

Signal extracted with a 2-dimensional fit of Δɸ and mT

of the dilepton-MET system

(l,l) [radians]!!0 0.5 1 1.5 2 2.5 3

Even

ts

0

20

40

60

80

100CMS

-1 = 8 TeV, L = 19.7 fbs

Z(ll) H(inv)

Observed

inv)=100%"B(H =125GeV),

HZH(m

ZZ

WZ

DY(ll)+jets

,tW,WW,W+jetstt

[GeV]Tm200 300 400 500 600 700 800 900 1000

Even

ts

0

10

20

30

40

50

60CMS

-1 = 8 TeV, L = 19.7 fbs

Z(ll) H(inv)

Observed

inv)=100%!B(H =125GeV),

HZH(m

ZZ

WZ

DY(ll)+jets

,tW,WW,W+jetstt

•For the Z(bb)H(inv) a BDT technique is used to select the signal.

INVISIBLE:Search in associated channel



LFV: Analysis strategy


• Mcollinear between de decay products as estimator of the Higgs mass

• Events divided into categories according to the number of jets in the event

Even

ts /

10 G

eV0

1000

2000

3000

4000

5000

6000

e!µData,

Bckg UncertaintySM Higgs

(embedded)!!Z+-l+Z+l

Single top quarktt

VV*" / W"W

Fakes (leptons)LFV GG Higgs (Br=100%)LFV VBF Higgs (Br=100%)


) [GeV]e!µcollinear M(

0 100 200 300

Bckg

Dat

a-Bc

kg

-0.50

0.5

MH = Mcollinear =Mvis√xτvis

• Z → ττ and misidentified leptons from W+jets and QCD multi-jet from data

Exploit the kinematics of the boosted τ from H decay



LFV: Mass spectra


Even

ts /

10 G

eV

0

1000

2000

3000

4000

5000

6000

e!µData,



Single top quarktt

VV*" / W"W




0 100 200 300

Bckg

Dat

a-Bc

kg

-0.50

0.5

Even

ts /

10 G

eV

0

200

400

600

800

1000

1200

1400

1600

1800

2000

e!µData,



Single top quarktt

VV*" / W"W




0 100 200 300Bc

kgD

ata-

Bckg

-0.50

0.5

Even

ts /

20 G

eV

0

500

1000

1500

2000

2500

3000

3500had!µData,



ttSingle top quarktt

VV)!"Fakes (jet

LFV GG Higgs (Br=100%)LFV VBF Higgs (Br=100%)


) [GeV]had!µcollinear M(

0 100 200 300

Bckg

Dat

a-Bc

kg

-0.50

0.5Ev

ents

/ 10

GeV

0

100

200

300

400

500

600

700e!µData,



Single top quarktt

VV*" / W"W




0 100 200 300

Bckg

Dat

a-Bc

kg

-0.50

0.5

Even

ts /

10 G

eV

200

400

600

800

1000

1200

1400

1600

1800

2000

2200had!µData,




VV)!"Fakes (jet




0 100 200 300

Bckg

Dat

a-Bc

kg

-0.50

0.5

Even

ts /

10 G

eV

0

1000

2000

3000

4000

5000

6000

7000

8000had!µData,




VV)!"Fakes (jet




0 100 200 300

Bckg

Dat

a-Bc

kg

-0.50

0.5



2HDM: Background estimation

• Main reducible contributions:

• Multilepton searches allow probing regions of parameter space inaccessible to hadronic searches.

1. Z+jets, W+jets with bosons decay leptonically and additional fake lepton

Estimated from data

2. ttbar with W’s leptonically decaysEvaluated in control regions

• Irreducible contributions:

1.VV+jets with >=3 real leptons

2. Drell-Yan processes with internal asymmetric conversions

• Diphoton plus lepton searches:

Main background reduced by the diphoton mass cut around the SM-Higgs observed value

Estimated with sidebands



2HDM:Results

(GeV)missTE

0-50 50-100 100-150 150-200 >200

Even

ts

10

210

310

410Observed

hh 300 GeV!HBkg UncertaintiesData-driventtWZZZWttZtt

SM Higgs

CMS Preliminary -1 = 19.5 fbint = 8 TeV, Ls

3-leptons+OSSF1+low-Z+no taus+no 1 b-jets

(GeV)missTE

0-50 50-100 100-150 150-200 >200

Even

ts

10

210

310

410

(GeV)missTE

0-30 30-50 50-100 >100

Even

ts

1

10

ObservedZh 300 GeV!A

Bkg UncertaintiesData-driven

ttWZZZ

WttZtt

SM Higgs

CMS Preliminary -1 = 19.5 fbint = 8 TeV, Ls

4-leptons+OSSF2+(TwoZ)+on-Z+no taus+at least 1 b-jet

(GeV)missTE

0-30 30-50 50-100 >100

Even

ts

1

10

• Most sensitive search channels for a Heavy Higgs search

[GeV]Am260 280 300 320 340 360

[pb]

!"Br

0

1

2

3

4

5

6

7CMS Preliminary -1 = 8 TeV, L = 19.5 fbs

Zh# A#gg

95% C.L. CLs LimitsObservedexpected

!1±expected !2±expected

[GeV]Hm260 280 300 320 340 360

[pb]

!"Br

0

2

4

6

8

10

12

14

16

18CMS Preliminary -1 = 8 TeV, L = 19.5 fbs

hh# H#gg95% C.L. CLs Limits

Observedexpected

!1±expected !2±expected

• 95% CL limits on cross-section times Br



[GeV]bbm0 20 40 60 80 100 120 140 160 180 200

Frac

tion

of e

vent

s/20

GeV

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4 ATLAS = 8 TeVs

Simulation = 125 GeV0hm = 325 GeV±Hm = 525 GeV0Hm

= 125 GeV0hm = 525 GeV±Hm = 825 GeV0Hm

= 125 GeV0hm = 725 GeV±Hm = 1025 GeV0Hm

[GeV]Wbbm200 300 400 500 600 700 800 900 1000

Frac

tion

of e

vent

s/40

GeV

0

0.05

0.1

0.15

0.2

0.25 ATLAS = 8 TeVs

Simulation

[GeV]WWbbm400 600 800 1000 1200 1400

Frac

tion

of e

vent

s/60

GeV

0

0.05

0.1

0.15

0.2

0.25

ATLAS = 8 TeVs

Simulation

MultiHiggs: Event Reconstruction

• Identification of the leptonically decaying W

• h candidate reconstructed with the two b-tagged jets

•Hadronically decaying W reconstructed from the remaining jets.

•H+- reconstructed from h and the W which gives the highest mass

•H formed with the bbarWW system reconstructed



(GeV)kinjj!! m

200 400 600 800 1000 1200

Fra

ctio

n of

eve

nts

/ 40

GeV

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7 Signal shapes = 300 GeVXM = 500 GeVXM = 700 GeVXM = 1000 GeVXM

standard model Higgs

bb!! " HH "X

= 8 TeVsCMS Preliminary Simulation

• Di-Higgs system reconstructed from a pair of photons and a pair of jets originating from b-quarks

bbgg: Signal Reconstruction

• Resonant signal topology:

1. Peak around mH (125 GeV) in diphoton and dijet spectra

2. Peak around mX (unknown) in the 4-body spectrum

• mX mass range: ATLAS [260-500] GeVCMS [260-1100] GeV

• Narrow resonance signal hypothesis

SM-Higgs considered as resonant background • Dominant Background: non resonant production of photons and jets (QCD)



Summary on decay modesCMSCMSCMS ATLASATLASATLAS

Dataset Status Results Dataset Status ResultsFavoured decay modesFavoured decay modesFavoured decay modesFavoured decay modesFavoured decay modesFavoured decay modesFavoured decay modes

H(ZZ) 5.1 + 19.6 fb-1 arXiv:1312.5353 μ= 0.93 4.8 + 20.3 fb-1 ATLAS-CONF-2014-009 μ= 1.44

H(WW) 5.1 + 19.4 fb-1 arXiv:1312.1129 μ= 0.72 4.8 + 20.3 fb-1 ATLAS-CONF-2014-009 μ= 1.00

H(γ γ)

5.1 + 19.4 fb-1 arXiv:1312.1129 μ= 0.72 4.8 + 20.3 fb-1 ATLAS-CONF-2014-009 μ= 1.00

H(γ γ) 5.1 + 19.7 fb-1 arXiv:1407.0558 μ= 1.14 4.8 + 20.3 fb-1 ATLAS-CONF-2014-009 μ= 1.57

H(tau tau)

5.1 + 19.7 fb-1 arXiv:1407.0558 μ= 1.14 4.8 + 20.3 fb-1 ATLAS-CONF-2014-009 μ= 1.57

H(tau tau) 4.9 + 19.7 fb-1 arXiv:1401.5041 μ= 0.78 4.8 + 20.3 fb-1 ATLAS-CONF-2014-009 μ= 1.44

V-H(bb) 5.1 + 18.9 fb-1 arXiv:1310.3687 μ< 1.89(0.95) 4.7 + 20.3 fb-1 ATLAS-CONF-2013-079 μ< 1.4(1.3)VBF-H(bb) 19.0 fb-1 HIG-13-011 μ< 3.6(3.0) 4.7 + 13.0 fb-1 ATLAS-CONF-2012-161 μ< 1.8(1.9)

ttH(bb) 19.5 fb-1 HIG-14-010 μ< 2.9(3.3) 20.3 fb-1 ATLAS-CONF-2014-011 μ< 4.1(3.4)

Rare decay modesRare decay modesRare decay modesRare decay modesRare decay modesRare decay modesRare decay modesH(mu mu ) 5.0+ 19.7 fb-1 HIG-13-007 μ< 7.4 24.8 fb-1 arXiv:1406.7663 μ< 7.2

H(Z γ) 5.0+ 19.6 fb-1 arXiv:1307.5515μ< 10

4.5 + 20.3 fb-1 arXiv:1402.3051 μ< 10

H(γ* γ) 19.7 fb-1 HIG-14-003μ< 10

- - -Invisible decay modesInvisible decay modesInvisible decay modesInvisible decay modesInvisible decay modesInvisible decay modesInvisible decay modes

Z(ll)-H(inv)

4.9 + 19.7 fb-1 arXiv:1404.1344 BR< 58(46)%

4.5 + 20.3 fb-1 arXiv:1402.3244 BR< 75(62)%

Z(bb)-H(inv) 4.9 + 19.7 fb-1 arXiv:1404.1344 BR< 58(46)% - - -VBF-H(inv)

4.9 + 19.7 fb-1 arXiv:1404.1344 BR< 58(46)%

- - -Exotic decay modesExotic decay modesExotic decay modesExotic decay modesExotic decay modesExotic decay modesExotic decay modes

H(tau mu) 19.7 fb-1 HIG-14-005 BR<1.57% - - -H(long-lived) - - - 20.3 fb-1 ATLAS-CONF-2014-041 UL vs proper decay

length

= preliminary= published






















Date post:	22-May-2020
Category:	Documents
Upload:	others
View:	7 times
Download:	0 times

BSM Higgs Searches in ATLAS and CMS (part 2) · BSM Higgs Searches in ATLAS and CMS (part 2) On...

Documents