Search for MSSM and NMSSM Higgs bosons with the CMS detector

(1)

Christian Veelken

for the CMS Collaboration

Search for MSSM and NMSSM Higgs bosons

with the CMS detector

ICHEP Conference, July 4

^th

2014

(2)

http://www.phdcomics.com/higgs

A SM-‐like Higgs boson has been discovered at the LHC

Christian Veelken Search for MSSM and NMSSM Higgs bosons 2

(3)

A SM-‐like Higgs boson has been discovered at the LHC

It looks like the SM Higgs boson, sounds like the SM Higgs boson and smells like the SM Higgs boson… but: Is it really the SM Higgs boson ?

– or just one of mul<ple Higgs bosons in a supersymmetric world ?

(4)

A SM-‐like Higgs boson has been discovered at the LHC

It looks like the SM Higgs boson, sounds like the SM Higgs boson and smells like the SM Higgs boson… but: Is it really the SM Higgs boson ?

– or just one of mul<ple Higgs bosons in a supersymmetric world ?

Two (complementary) ways to ﬁnd-‐out:

•  Precision measurements of BRs

(good sensi<vity already with 7+8 TeV LHC data è see next BEH session)

•  Find more Higgs bosons!

• 

è Subject of this Presenta<on: Results of CMS searches for Higgs bosons in the context of MSSM and NMSSM

(5)

NB.: m_h = 125 GeV is right where SUSY predicted the Higgs boson to be!

MSSM NMSSM

2 Higgs doublets ^[*] 2 Higgs doublets ^[*]

+ complex Higgs singlet

2 scalars: h, H 1 pseudo-‐scalar: A

2 charged Higgs bosons: H⁺, H^-‐

3 scalars: h₁, h₂, h₃ 2 pseudo-‐scalars: a₁, a₂ 2 charged Higgs bosons: H⁺, H^-‐

Higgs mass > m_Z due to radia<ve correc<ons: large stop mass and/or large stop mixing required to reach m_h = 125 GeV

m_h = 125 GeV for Higgs with SM-‐like couplings can be achieve by mixing between doublet and singlet ﬁelds;

much less constraints on radia<ve correc<ons

Simplicity Naturalness

[*] TypeII 2HDM

(6)

Search for light NMSSM

Higgs bosons

(7)

pp ! h ! 2a

₁

! 4 μ

Model independent search for events containing 2 muon pairs of same mass

Event selecOon

Exactly 2 muon pairs of :

•  zero charge

•  mass m_μμ < 5 GeV

•  same mass (within 5 <mes detector resolu<on)

(8)

Background Estimation

Main backgrounds:

•  decays to muons

•  Direct di-‐ produc<on

b ¯ b

J/

8 Events observed in sideband

è Background es<mate in signal region = 3.8 ± 2.1

m

_µµ₁

6 = m

_µµ₂

esOmated from sideband in data

Signal region (blinded)

(9)

Results

1 event observed in signal region, compa<ble with background

predic<on of 3.8 ± 2.1

No evidence for a signal

è Set limit on cross-‐sec<on x BR

(10)

Model independent Limit

Eﬀect of the 1 event observed in signal region

(11)

Searches for charged MSSM

Higgs bosons

(12)

H

⁺

! cs (m

_H⁺

< m

_top

)

Event selecOon

≥ 2 b-‐tagged jets of P_T > 30 GeV & |η| < 2.4

H⁺

(13)

H

⁺

! cs (m

_H⁺

< m

_top

)

Event selecOon

≥ 2 further jets of P_T > 30 GeV & |η| < 2.4

H⁺

(14)

H

⁺

! cs (m

_H⁺

< m

_top

)

Event selecOon

≥ 2 further jets of P_T > 30 GeV & |η| < 2.4

E_T^miss > 20 GeV

1 isolated muon of P_T > 25 GeV & |η| < 2.1

H⁺

(15)

Separation of H

⁺

Signal from Backgrounds

H⁺ H⁺ ! cs signal

SM Q background

è Dis<nguish signal from background by looking at mass of jets without b-‐tag Main background: SM q

Minor background: single top

Diﬀerence between signal and background: t ! H⁺b instead of t ! Wb decay

other backgrounds heavily suppressed by requirement of b-‐tagged jets

(16)

[GeV]

Mjj

0 20 40 60 80 100 120 140 160 180 200

Normalized to Unity

0 0.02 0.04 0.06 0.08 0.1 0.12 0.14

t t

s

! c H+

= 8 TeV s

CMS Simulation,

= 120 GeV

H+

M

Dijet Mass Reconstruction

Dijet mass reconstructed by constrained kinemaOc ﬁt,

Choose ﬁt solu<on with lowest χ²:

•  For deciding which of the 2 b-‐tagged jets to pair with the non-‐b-‐tagged jets

•  In case there are > 4 jets in the event

using measured momenta +

•  muon

•  b-‐tagged and non-‐b-‐tagged jets

•  E_T^miss

plus their respec<ve resolu<ons

Shape analysis of dijet mass distribuOon Analysis strategy:

constraints

•  m(b + qq’) = m_t

•  m(b + μν) = m_t

•  m(μ + v) = m_W

(17)

(GeV) Mjj

0 20 40 60 80 100 120 140 160 180 200

Events

500 1000 1500 2000

2500 ^Observed

Diboson l-

l+

!

"

Z/

SingleTOP l# W!

t t Total Unc.

s

! c H+

= 120 GeV

H+

M

b) = 0.2 H+

! Br(t

= 8 TeV, 19.7 fb-1

s CMS Preliminary,

(GeV) Mjj

0 20 40 60 80 100 120 140 160 180 200

BkgData-Bkg

-0.5 -0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4 0.5

Signal Extraction

In the case of a t ! H⁺b signal:

•  a deﬁcit of events at m_W ^[*]

•  an excess of events at m_H⁺

is expected in the dijet mass distribu<on.

The yields of the t à H⁺b signal and of the backgrounds (SM q and other) is determined via a ﬁt of the dijet mass spectrum observed in data.

No evidence for a signal

è Set limits on BR(t à H⁺b)

[*] due to reduc<on of the number of t à Wb à qq’b decays (total number of q events does not change)

(18)

Limits

Christian Veelken Search for MSSM and NMSSM Higgs bosons 18 (GeV)

H+

90 100 110 120 130 140 150m 160 )± bH!95% CL limit on BR(t

0 0.02 0.04 0.06 0.08 0.1 0.12

0.14 Observed

Expected

"

± 1 Expected limit

"

± 2 Expected limit

s c

+!

±b, H H

! t

) = 1 s

! c BR(H+

+ jets final state µ

= 8 TeV, 19.7 fb-1

s CMS Preliminary,

Limit on t à H⁺b computed

assuming all charged Higgs bosons to decay into dijets

(19)

Searches for neutral MSSM

Higgs bosons

(20)

Events are analyzed in 2 Categories,

targe<ng diﬀerent MSSM neutral Higgs produc<on mechanisms:

1. No-‐B-‐tag: Events without b-‐tagged jets

2. B-‐tag: Events containing ≥ 1 b-‐tagged jet of P_T > 20 GeV & |η| < 2.4 gluon-‐gluon Fusion

dominates cross-‐sec<on in case tanβ is small

pp ! ϕ

b-‐associated ProducOon dominates cross-‐sec<on in case tanβ is large

pp ! ϕb

ϕ ! ττ

generic label for any one of the neutral MSSM Higgs bosons h, H or A

(21)

Events are analyzed in 2 Categories,

targe<ng diﬀerent MSSM neutral Higgs produc<on mechanisms:

1. No-‐B-‐tag: Events without b-‐tagged jets

2. B-‐tag: Events containing ≥ 1 b-‐tagged jet of P_T > 20 GeV & |η| < 2.4 gluon-‐gluon Fusion

dominates cross-‐sec<on in case tanβ is small

pp ! ϕ

b-‐associated ProducOon dominates cross-‐sec<on in case tanβ is large

pp ! ϕb

ϕ ! ττ

generic label for any one of the neutral MSSM Higgs bosons h, H or A

Presence of b-‐jet helps to reduce background (one b-‐jet usually outside acceptance)

(22)

Signal events selected in 5 out of 6 channels:

BR

•  τ_hτ_h 42.0%

•  eτ_h 23.1%

•  μτ_h 22.6%

•  eμ 6.2%

•  μμ 3.0%

96.9% of possible signal included in analysis

"

(not yet included in analysis: ee, BR = 3.1%)

ττ Decay Channels

(23)

[GeV]

τ

mτ

0 50 100 150 200 250

[1/GeV]ττ1/dm

0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16

= 125 GeV mH

τ τ

→ H

τ τ Z → = 8 TeV s

CMS Simulation µτ_h

Higgs Mass Reconstruction

Reconstruc<on of ditau mass based on Likelihood method, using as input:

•  Measured e, μ, τ_h-‐jet momenta

•  Reconstructed E_T^miss and event-‐by-‐event es<mate of E_T^miss resolu<on (E_T^miss reconstructed by a mul<variate regression technique)

Search for peak in ditau mass distribuOon Analysis strategy:

Resolu<on on m_ττ is O(20%), improves separa<on of signal from backgrounds

J.Phys.Conf.Ser. 513 (2014) 022035

(24)

No evidence for a signal beyond 125 GeV

è Set upper limits on cross-‐sec<on x BR

for the 2 produc<on processes pp à ϕ and pp à ϕb

Compute region in m_A-‐tanβ parameter space excluded by non-‐observa<on of a signal

Results

(25)

m

_ττ

Distributions for all Channels

(26)

Model independent Limits on ϕ

pp ! ϕb x BR(ϕ!ττ)

m_ττ spectra observed in B-‐tag and No-‐B-‐tag category are ﬁqed simultaneously, with shape templates for ϕ à ττ signal and for background processes.

pp à ϕb (pp à ϕ) cross-‐sec<on treated as nuisance parameter when compu<ng limit on pp à ϕ (pp à ϕb)

red line: limit expected in case no SM Higgs produc<on is present in the data blue line: limit expected in the presence of Higgs à ττ with SM cross-‐sec<on x BR

pp ! ϕ x BR(ϕ!ττ)

(27)

Exclusion contour in m

_A

-tanβ plane

InterpretaOon of results on new MSSM benchmark scenarios in preparaOon

red line: limit expected in case no SM Higgs produc<on is present in the data grey line: limit expected in the presence of Higgs à ττ with SM cross-‐sec<on x BR

Fit for sum of A+H+h,

taking dependency of cross-‐sec<on, BR and mass of each Higgs on

m_A and tanβ into account excluded

excluded

excluded

(28)

Summary & Outlook

(29)

Summary

CMS has searched for:

No evidence for a supersymmetric Higgs signal observed in CMS data so far

è Stringent limits on cross-‐sec<on x BR have been set

•  Light NMSSM Higgs bosons

•  Charged MSSM Higgs bosons

•  Neutral MSSM Higgs bosons

(30)

Outlook

è  Greatly enhanced sensiOvity,

in parOcular for signals of high mass

# Stay tuned!

CMS is about to restart taking data in 2015, with:

•  Higher center-‐of-‐mass energy TeV

•  Higher luminosity

•  New, more powerful, experimental techniques

p s = 13

(31)

CMS Searches for SUSY Higgs bosons

σ x BR

m

_H

[GeV]

10 125 173

100pb

1pb

10}

H⁺ à τν

ϕb à 3b

H⁺ ! cs

Presented at ICHEP 2014

Presented at Conferences previously

Phys.LeQ. B722 (2013) 207 CMS PAS HIG-‐12-‐052

CMS PAS HIG-‐13-‐035

CMS PAS HIG-‐13-‐024 Phys.Rev.LeQ. 109 (2012) 121801

a₁ à 2μ

h ! 2a₁ ! 4μ

CMS PAS HIG-‐13-‐010

CMS PAS HIG-‐13-‐021

ϕ à μμ

ϕ ! ττ

(32)

Backup

(33)

The CMS Detector

https://cms-docdb.cern.ch/cgi-bin/PublicDocDB/ShowDocument?docid=11520

ECAL

HCAL

Muon system

Silicon tracking detector

(34)

MSSM neutral Higgs cross-section

10 100 1000 10000

1 10 100 1000

gg Σqq

WJS2012

ratios of LHC parton luminosities:

14 TeV / 8 TeV and 33 TeV / 8 TeV

luminosity ratio

MX (GeV)

MSTW2008NLO

_

Parton luminosity increases by O(10) in run 2 compared to run 1

Signal cross-‐sec<on dropping rather steeply with Higgs mass

http://www.hep.ph.ic.ac.uk/~wstirlin/plots/plots.html

(35)

Width of MSSM neutral Higgs boson

≤ 3% of Higgs mass

Width of MSSM neutral Higgs boson

Higgs width relevant for ϕ à μμ analysis only

(experimental resolu<on on reconstructed Higgs mass O(15%) in ϕ à bb and O(20%) in ϕ à ττ decays)

(36)

m_h^max

m_h^mod+

Iight stop

τ-‐phobic Iight stau

m_h^max scenario disfavored by observa<on of SM-‐like Higgs boson of mass 125 ± 3 GeV

New MSSM benchmark scenarios compa<ble with 125 GeV in most of parameter space

MSSM Benchmark scenarios

m_h^mod-‐

Eur.Phys.J. C73 (2013) 2552

(37)

pp ! a

₁

! 2 μ

No evidence for signal

è Set limits

Search for peak in dimuon mass distribuOon Phys.Rev.LeQ. 109 (2012) 121801

(38)

H

⁺

! τν (m

_H⁺

< m

_top

)

Shape analysis of R = p^lead.track/E^τ

Data in agreement with background expecta<on

è Set limits

Frac<on R of energy carried by highest P_T (“leading”) track of hadronic tau is sensi<ve to τ-‐lepton polariza<on, which is opposite for taus origina<ng from t à Wb à τνb and taus produced in H⁺ à τν decays.

CMS PAS HIG-‐12-‐052

(39)

ϕb ! 3b

Search for peak in dijet mass distribuOon

Sizeable signal, but also large backgrounds

No evidence for signal

è Set limits

Phys.LeQ. B722 (2013) 207

(40)

ϕ ! μμ

No evidence for signal peak

in dimuon mass spectrum observed in data

è Set Limit

h

H+A

Dimuon mass Resolu<on ~3%

Search for peak in dimuon mass distribuOon CMS PAS HIG-‐12-‐011

(41)

Backgrounds to ϕ ! ττ

Z/γ*→ττ

Obtained using Embedding technique:

Z à μμ events selected in data,

reconstructed muons replaced by simulated taus

Electroweak (W+jets, diboson, Z ! ee/μμ) Shapes from MC simula<on

Normaliza<on of W+jets obtained by

extrapola<on from high m_T control region in data, others from MC simula<on

QCD

Fully data driven: Normaliza<on (shape) obtained from same-‐sign events with isolated (non-‐isolated) muons

Q, single top

Shape from MC simula<on, normaliza<on from data

Events

0 2000 4000 6000 8000 10000 12000 14000 16000 18000

at 8 TeV CMS Preliminary, 19.7 fb-1

τh

µ

Selection Baseline

Control Region High mT

observed bkg. uncertainty

τ τ

→ Z

µ µ Z→ electroweak

t t QCD

[GeV]

mT

0 20 40 60 80 100 120 140 160

Obs/Bkg 0.6

0.8 1.0 1.2 1.4

(42)

Hadronic tau Reconstruction

~35%

~57%

Mass m_τ= 1.78 GeV Life<me c  τ = 87 μm

è Taus typically decay before reaching ﬁrst detector layer

Charged and neutral pions reconstructed by ParOcle Flow algorithm

(43)

Tau Decay Mode reconstruction

single Hadron

Hadron + Strip(s)

Three Hadrons

Approach works well up to high pile-‐up

(44)

[GeV]

!

pT

Efficiency

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8

3-hit Loose MC

"

HPS #

3-hit Loose Data

"

HPS #

3-hit Medium MC

"

HPS #

3-hit Medium Data

"

HPS #

3-hit Tight MC

"

HPS #

3-hit Tight Data

"

HPS #

[GeV]

!

pT

20 40 60 80

SimulationData

0.85 0.90 0.95 1.00 1.05 1.10 1.15

= 8 TeV, L = 19.7 fb-1

s CMS Preliminary 2012,

50 100 150 200 250 300 350 400 450 500

Fake-rate

10-4

10-3

10-2

10-1

3-hit Loose Data

!

"

HPS

3-hit Loose MC

!

"

HPS

3-hit Medium Data

!

"

HPS

3-hit Medium MC

!

"

HPS

3-hit Tight Data

!

"

HPS

3-hit Tight MC

!

"

HPS = 8 TeV, L = 19.7 fb-1

s CMS Preliminary 2012,

[GeV/c]

jet

PT

50 100 150 200 250 300 350 400 450 500

SimulationData - Simulation -0.4

-0.2 0 0.2 0.4

Tau ID Performance

CMS-‐DP-‐2014-‐015

See also poster by Rosamaria Vendir CMS Tau ID performed well during run 1

Data well modeled by Monte Carlo simula<on

ID = τ_h Decay Mode reconstruc<on & Isola<on

Search for MSSM and NMSSM Higgs bosons with the CMS detector

Christian Veelken

A SM-­‐like Higgs boson has been discovered at the LHC

A SM-­‐like Higgs boson has been discovered at the LHC

MSSM NMSSM

Search for light NMSSM

Model independent search for events containing 2 muon pairs of same mass

Event selecOon

Main backgrounds:

Results

Searches for charged MSSM

Event selecOon

Dijet Mass Reconstruction

Signal Extraction

Limits

Searches for neutral MSSM

generic label for any one of the neutral MSSM Higgs bosons h, H or A

generic label for any one of the neutral MSSM Higgs bosons h, H or A

ττ Decay Channels

Results

Exclusion contour in m

excluded

Summary & Outlook

• Charged MSSM Higgs bosons

• Higher luminosity

Backup

Muon system

MSSM neutral Higgs cross-section

Width of MSSM neutral Higgs boson

MSSM Benchmark scenarios

Backgrounds to ϕ ! ττ

Hadronic tau Reconstruction

Tau ID Performance

A SM-‐like Higgs boson has been discovered at the LHC

A SM-‐like Higgs boson has been discovered at the LHC

•  Charged MSSM Higgs bosons

•  Higher luminosity