Search for Neutral Higgs Bosons which Decay to tau Pairs

Search for Neutral Higgs Bosons which Decay to tau Pairs"

9-Jan-13! 1!Sridhara Dasu (Wisconsin)!

Higgsteria!"

The Best of Times! for particle physics!!!Is the new boson ! the Higgs of SM?!!Does the new boson! couple to taus, as it

should if it is SM H?!!We have x2 data now,

and good future J!9-Jan-13! Sridhara Dasu (Wisconsin)! 2!

Standard Model Higgs Production"

9-Jan-13! Sridhara Dasu (Wisconsin)! 3!

[GeV] HM100 200 300 400 500 1000

H+X

) [pb

]

�(p

p �

-210

-110

1

10= 7 TeVs

LHC

HIG

GS

XS W

G 2

010

H (NNLO+NNLL QCD + NLO EW)

�pp

qqH (NNLO QCD + NLO EW)

�pp

WH (NNLO QCD + NLO EW)

�pp

ZH (NNLO QCD +NLO EW)

�pp

ttH (NLO QCD)

�pp

Standard Model Higgs Decay"


We search for several Higgs decay channels.

H → bb

H → ττ

H → γγ

H → WW

H → ZZ

Cleaner S

ignal

Better M

ass Resolution

Decays to ZZ to 4-light leptons"

Golden mode – Zoom of loss 4L mass region!


Clearly observe production of Z

decay to 4 leptons

Cluster of events rising above the

background consistent at

3-sigma level with SM Higgs signal M

Z =

91 G

eV

125.

8 ±

0.4

± 0.

4 G

eV

CMS HIG-12-041

Use Angular Information"


Study additional properties of these events!•  Angles shown carry information of scalar (SM H),

pseudo-scalar vs spin-2 decay versus ZZ production!

Strong Evidence in decays to ZZ"


Boost using angular information from 3 to 4.5 sigma!•  Reduce background, while keeping signal-like events!

CMS HIG-12-041

Is the new boson a scalar?"

Not yet significant, but favors 0+ over 0– by ~2sigma!


CMS HIG-12-041

-2ln(L /L )-15 -10 -5 0 5 10 15

Gen

erat

ed e

xper

imen

ts

0

1000

2000

3000

4000

5000 0+2+Observed

CMS -1 = 7 (8) TeV, L = 5.1 (12.2) fbs

b)

+0+2

SM Higgs?"

Signal strength consistent with SM Higgs!Excesses seen in γγ, ZZ and WW in both experiments!


Statistical and systematic errors

included

Does it couple to fermions? Answers, beginning to emerge

Self-coupling?

Must wait for several hundred fb–1

CMS HIG-12-045

Higgs Decay to tau Pairs"

9-Jan-13! 10!Sridhara Dasu (Wisconsin)!

Forward Jet 1

Forward Jet 2

Tau Muon

MET

Taus with high branching fraction can probe in all production modes: W and Z boson fusion; W, Z, top associated production and Gluon fusion

Tau Decays"

Weak decays!•  Not fully reconstructed – MET from neutrino!•  Leptonic electron/muon + 2 neutrinos (35%)!•  Hadrons + 1 neutrino (65%)!

•  Single charged hadron (~50%)!•  ρ → π+π0!•  a1 → π+π–π+ , π+π0π0!

Hadronic tau identification is important!•  Reconstruct intermediate light mesons!•  Take advantage of good charged tracking!•  Take advantage of highly segmented ECAL!•  Suppress large background from jets!


Decay Mode Reconstruction"The good!

–  Single prong: well measured isolated charged hadron!–  Three prong: softer well measured and vertexed charged hadrons!

The bad!–  n π0s leading to 2n γs -- π0 not always reconstructed fully!–  EM energy clusters spread out in φ-strips!

The ugly!–  Missing neutrino: fit the missing momentum vector

and secondary vertex of visible τ-pair components!

9-Jan-13 Sridhara Dasu (Wisconsin)! 12

Single Prong Three Prong Single Prong + Strip

Not “tau-jets”

τ lepton reconstruction"Bachtis (Wisconsin), Friis (Davis)!

•  Instead of jets of narrow cone, shrinking cone … we truly reconstruct taus!•  In single and three prong modes, adding neutral pizeros!

•  Decay mode algorithms HPS and TaNC!•  Established HPS algorithm and analysis techniques for τ physics !

JINST 7 (2012) P01001

9-Jan-13!13 Sridhara Dasu (Wisconsin)

τ reconstruction performance"


Swanson (Wisconsin), Rebecca (Imperial)!

τ reco pileup dependence"

Efficiency is ~flat at > 60% independent of pileup level!

!Fake rate 1-3%!!Measured using real data with tag

and probe technique to 7%!


Chan (MIT)!Swanson (Wisconsin), Veelken (LLR)!

τ energy scale"


Bachtis (Wisconsin), Veelken (LLR)!

Z to ττ : 2010 Data "Bachtis (Wisconsin), Friis (Davis), Swanson (Wisconsin), Cutajar (Imperial)!

•  Measure Z to τ τ cross section and τ identification efficiency!•  Necessary precursor to searches for higgs decays in τ modes!

J. High Energy Phys. 08 (2011) 117

Clean Z signals in τ modes enabled cross section measurement + ��simultaneously extract τ ID efficiency.

9-Jan-13!17 Sridhara Dasu (Wisconsin)

Visible Mass [GeV]0 50 100 150 200

Even

ts /

(10

GeV

)

0

50

100

150

dataττ → Z

t EWK+t QCD

yields from fit

CMS

= 7 TeVs at -136 pb

hadτµτ → ττ →Z

Visible Mass [GeV]0 50 100 150 200

Even

ts /

(10

GeV

)

0

50

100

150

dataττ → Z

t EWK+t QCD

yields from fit

CMS

= 7 TeVs at -136 pb

hadτeτ → ττ →Z

-ID Efficiency (data/sim)hadτ0.6 0.8 1 1.2 1.4

) [n

b]ττ

→ B

(Z

× Z

X)

→(p

p σ

0.7

0.8

0.9

1

1.1

1.2

1.3

68% CL

95% CL

central value

σ 1 ±

CMS

= 7 TeVs at -136 pb

Z to ττ Cross Section"

Sridhara Dasu (Wisconsin)! 18 9-Jan-13

Taus

are

legi

t !!

MET Improvement"

Particle flow MET corrected using Z recoil in μμ data events!Further improved using multi-variate analysis!


ICHEP ICHEP

HCP HCP

Big change compared to ICHEP

Harris (CERN)!

MT Improvement"


Harris (CERN)!

pZeta Improvement"

Kinematic variable in eμ channel with larger MET!


Harris (CERN), Dutta (MIT) !

Di-tau Mass Reconstruction"


Visible Mass

SVFit Mass

Veelken (LLR), Lorenzo (LLR), Friis (Wisconsin) !

Decay Channels Covered"


Backgrounds in Inclusive Analysis"


Event Characterization"


0 Jet

High Background Constrains fit

Used to fix

normaliza;on & tau efficiency

1 Jet, Low pT Enhancement from Jet

Requirement

1 Jet, High pT Enhancement from pT and Jet requirement

VBF

2 Jets Rapidity Gap Veto

Δη > 3.5

MJJ > 500 GeV

Jets pT > 30 GeV

pT τh (τµ)

Signal is negligible Gluon Fusion Enhanced VBF Enhanced

Simplified compared to ICHEP

All categories are fit simultaneously!

μμ Channel"

Overwhelming Z to μμ background results in poor sensitivity!!MET and Mττ are used in 2D fit to extract signal for this channel!


CMS HIG-12-043

Bethani (DESY), Raspareza (DESY) !

ττ Channel"

Trigger required additional jet, so only high boost and VBF!Data driven background estimates!Larger than desired thresholds and QCD BG make this less sensitive!!!!!!!!!!Mττ shape fit is used to extract signal!


CMS HIG-12-043

Hinzmann (CERN)!

eμ Channel"

Channel with least background, but also small signal!Mττ shape fit is used to extract signal!!


CMS HIG-12-043

Dutta (MIT)!

eτ Channel"

Second most sensitive channel!Mττ shape fit is used to extract signal!!!


CMS HIG-12-043

Swanson (Wisconsin), Gilbert (Imperial)!

μτ Channel"

Most sensitive channel!Mττ shape fit is used to extract signal!!!


CMS HIG-12-043

Bianchini (LLR), Swanson (Wisconsin), Gilbert (Imperial)!

By Channel Expected Significance"

Combined significance approaches SM sensitivity!


CMS HIG-12-043

Wolf (MIT), Friis (Wisconsin), Harris (CERN)!

Expected Significance By Category"

Boosted 1-jet (gluon fusion) and 2-jet (VBF) are ~equal!


CMS HIG-12-043


Z Associated Production"

Reconstructed eight final states with llτhτh, lllτh!Z1 in light leptons with 60 < Mll < 120 GeV!ZZ BG from simulation!Non-prompt BG using data!!


CMS HIG-12-050

Mohammadi (Belgium)!

W Associated Production"

Reconstructed eμτh and μμτh (eeτh, eτhτh and μτhτh in to do list)!Same charge light leptons to reject Z+jet background!Kill fakes with cut on scalar sum of leptons, LT > 80 GeV!Extra lepton and b-jet veto!WZ and ZZ from simulation!Non-prompt BG (Z+jet, top, etc)

estimated from data (fake rate)!!Higgs visible mass shape used for

limit setting!


CMS HIG-12-050

Friis (Wisconsin)!

VH Limits"

By themselves neither are sensitive at SM level!However, they provide important VH sensitivity!


CMS HIG-12-050

Friis (Wisconsin), Mohammadi (Belgium)!

Limit on SM Higgs"Everything combined (5 channels x 3 categories) + VH!Small excess, but within systematic uncertainty bands!


CMS HIG-12-043


Observed vs SM H(125 GeV) Injected"

Everything combined (5 channels x 3 categories) + VH!Consistent with SM H signal – large systematic band!


CMS HIG-12-043


Change from ICHEP?"

MVA MET vs PF MET; Simplified VBF; Updated Reco!


Results obtained from the same data

set have changed L


Overlap of Events"

Small overlap L -- however, reduced BGs in HCP!


2012

2011 HCP Selected

Worry: Re-reconstruction seems to have affected the results

Swanson (Wisconsin), Harris (CERN)!

SM Higgs Signal Strength"Consistent with SM H and also zero L!~50% more statistics for final result from 2012!Need ~50 fb–1 at 14 TeV for 3-sigma observation!!


CMS HIG-12-043

CMS HIG-12-045


Beyond the Standard Model"

There are many possibilities that change the precise predictions of the minimal higgs sector of the Standard Model"–  Fourth (heavy) generation of fermions modify H couplings"

•  Enhances SM4 higgs cross section over SM"•  Already ruled out in entire parameter space with 2011 data"

–  Fermiophobic – fermion mass of different origin than higgs"•  Changes low mass higgs production & decays dramatically"•  Also ruled out for 126 GeV object"

–  Beyond minimal higgs doublet field"•  Two higgs doublet model (2HDM)"

–  Multiple higgs bosons: 3 neutral and 2 charged"–  Minimal Supersymmetric Model (MSSM) requires 2HDM"

•  NMSSM, triplets … have even more higgses"–  Very light pseudoscalar higgs, Doubly charged …"

–  This talk focuses on these non-standard higgs bosons"

9-Jan-13 Sridhara Dasu (Wisconsin) 41

MSSM Higgs"

Higgs sector in SUSY theory is more complicated!•  Need 2 higgs doublets each with 4 degrees of freedom!

•  Results in the Standard Model like Higgs (h0)!•  Plus, two neutral higgs (A0, H0) and charged (H±)!•  However, only 2 parameters (MA, tanβ – ratio of the two doublets)!•  Masses of higgs and Z related!•  Search in (MA, tanβ) plane!

Neutral Higgs!–  Look for ϕ=(h0, A0, H0) in decays to tau-leptons!

Charged Higgs!–  Look for H± in top decays!


MSSM ϕ(h, H, A)"Use MHMAX Scenario!Enhanced coupling to b-quarks and τ-leptons!

•  Production rate enhanced × tan2β!•  Gluon fusion with b,t loops + associated b quark production!

•  Decays to b-quark and τ-lepton pairs enhanced at all masses!


[GeV]AM100 200 300 400 500 600 700 800 9001000

)!!"

BR(A

0

0.05

0.1

0.15

0.2

0.25= 5#tan= 10#tan= 15#tan= 20#tan= 25#tan= 30#tan= 35#tan= 40#tan= 45#tan= 50#tan

= 7 TeVsmhmax scenario

[GeV]AM100 200 300 400 500 600 700 800 9001000

)!!"

BR(A

0

0.05

0.1

0.15

0.2

0.25

Implications of SM Like H126on MHMAX scenario"

Sridhara Dasu (Wisconsin)! 44! 9-Jan-13!

Mass of ττ : 17 fb–1 Data "CMS HIG-12-051

Tau pairs reconstructed in decays to leptons (e or μ) + hadrons (1 or 3 prong) or two leptons (eμ) Kinematic fit to obtain tau pair mass – used to search for H to ττ contribution

Two categories: non-b-tagged and b-tagged to enhance bbϕ

9-Jan-13 45 Sridhara Dasu (Wisconsin)

[GeV]ττm0 200 400 600 800 1000

[1/G

eV]

ττdN

/dm

-210

-110

1

10

210

310

410

510

610

710CMS -1 = 7-8 TeV, L = 17 fbsPreliminary, µτµτ

=8β, tanττ→(160 GeV)φ×10observed

ττ→Zµµ→Z

electroweaktt

QCDbkg. uncertainty

[GeV]ττm0 100 200 300 400 500

[1/G

eV]

ττdN

/dm

-110

1

10

210

CMS -1 = 7-8 TeV, L = 17 fbsPreliminary, µτµτ


ττ→Zµµ→Z

electroweaktt

QCDbkg. uncertainty

Bethani (DESY), Raspareza (DESY) !





[GeV]ττm0 200 400 600 800 1000

[1/G

eV]

ττdN

/dm

-110

1

10

210

310

CMS -1 = 7-8 TeV, L = 17 fbsPreliminary, µτeτ


ττ→Zelectroweaktt

QCDbkg. uncertainty

[GeV]ττm0 100 200 300 400 500

[1/G

eV]

ττdN

/dm

-110

1

10

CMS -1 = 7-8 TeV, L = 17 fbsPreliminary, µτeτ



QCDbkg. uncertainty

Dutta (MIT) !





[GeV]ττm0 200 400 600 800 1000

[1/G

eV]

ττdN

/dm

-210

-110

1

10

210

310

CMS -1 = 7-8 TeV, L = 17 fbsPreliminary, hτeτ


ττ→Z ee→Z

electroweakttQCDbkg. uncertainty

[GeV]ττm0 100 200 300 400 500

[1/G

eV]

ττdN

/dm

-210

-110

1

10

CMS -1 = 7-8 TeV, L = 17 fbsPreliminary, hτeτ


ττ→Z ee→Z

electroweakttQCDbkg. uncertainty

Swanson (Wisconsin), Gilbert (Imperial)!





[GeV]ττm0 200 400 600 800 1000

[1/G

eV]

ττdN

/dm

-210

-110

1

10

210

310

410CMS -1 = 7-8 TeV, L = 17 fbsPreliminary, hτµτ



QCDbkg. uncertainty

[GeV]ττm0 100 200 300 400 500

[1/G

eV]

ττdN

/dm

-210

-110

1

10

210CMS -1 = 7-8 TeV, L = 17 fbsPreliminary, hτµτ



QCDbkg. uncertainty

Bianchini (LLR), Swanson (Wisconsin), Gilbert (Imperial)!

Limit in mA-tanβ Plane"

Sridhara Dasu (Wisconsin) 49 9-Jan-13

CMS HIG-12-051

Wolf (MIT), Friis (Wisconsin),Vazquez Acosta (Imperial)!

MSSM Higgs Summary"


CMS PAS HIG-11-019

ATLAS CONF 2012-11

CMS PAS HIG-12-050

CMS PAS HIG-12-026 CMS PAS HIG-12-027

Friis (Wisconsin)!

Sridhara Dasu (Wisconsin)! 51! 9-Jan-13!

Summary"LHC discovered a new particle!

–  Both ATLAS & CMS see the same thing at the same mass!–  Updated evidence in Z-pair decay mode is excellent!

Is it the Standard Model Higgs Boson?!–  It is a boson – because it decays to bosons!–  Its mass is in the right window – consistency with SM!–  However, its properties are yet to be determined to confirm that it is

the SM Higgs Boson, 0+ slightly favored!–  Not confirmed yet if the new boson couples to taus L!–  Fermion channels are consistent with it being SM Higgs!

LHC has performed well in 2012 but will be down 2013-2014!•  Both experiments collected > 20 fb–1 in 2012!•  Should be able to update with 50% more data in Spring 2013!•  It will take ~50 fb–1 14 TeV to confirm SM Higgs to taus, 2015-2016?!•  If it is not SM like Higgs, it will be even more exciting!!

Nagging Issues Statistics, Biases, …"

Appears that some 2-sigmas (read agreeing with SM) are sometimes easier to get out of the collaboration than others!

Difficult to get out if they are “difficult” to explain to the public and theorists J!


ZZ M4L ZZ M4L & Mass Errors Fit Default

??

ZZ M4L Default

Latest ATLAS Update (after HCP)"ATLAS sees larger signals than CMS!

•  ZZ (MZZ = 123.5GeV)!•  Gamma,Gamma (Mγγ = 126.5 GeV)!

!


ATLAS Mass and Signal Strength"


CMS : γγ Mass 7+8 TeV 10 fb–1"

Divide data into four categories by mass resolution!

Separate reduced background di-jet category!

2011 (7 TeV)!2012 (8 TeV)!!!


CMS HIG-12-015

Documents

Search for Neutral Higgs Bosons which Decay to tau Pairs