Standard Model Higgs Search at CMS
Aaron DominguezWichita State, September 7, 2011
Known Force Particles
Graviton?
Known Matter Particles
Something’s Missing...
• Generations of matter different only in mass of the particles, otherwise basically duplicates
• Why do they have different masses?• Why is the weak force “weak?” Because mW,
mZ > 0.
• OK, so why are mW, mZ > 0?• One “solution” is a new particle: Higgs boson
Higgs Field• According to this model, the Higgs field fills the
universe
• It doesn’t disturb gravity, strong or the EM force• It does disturb the weak force and makes it short
ranged
• In simplest model, same particle could also be giving mass to all other particles
gravity
electric force
weak force
Accelerators: currently our best tools
!"#$%&'(
)*+,
-.
-.)/0 /1
234%".5'6
Accelerators: currently our best tools
!"#$%&'(
)*+,
-.
-.)/0 /1
234%".5'6
Interesting physics hard to find
Lothar A T Bauerdick Fermilab UW Madison February 2, 2004 6
Large Hadron Collider
Particle
Proton- Proton 2835 bunch/beam
Protons/bunch 1011
Beam energy 7 TeV (7x1012 eV)
Luminosity 1034 cm-2 s-1
Crossing rate 40 MHz
Collision rate ~109 Hz
Parton(quark, gluon)
Proton
Event Selection:1 in 10,000,000,000,000
l l
jetjet
Bunch
SUSY.....
Higgs
Zo
Zoe+
e+
e-
e-
New physics rate ~ 0.00001 Hz
LHC • Proton-Proton 1418 bunches/beam• Protons/bunch ~1011
• Beam energy 3.5 TeV• Inst. Luminosity 2.5E34 cm-2s-1
• Crossing rate 20 MHz• Collision rate ~109 Hz• New physics rate ~0.00001 Hz• Event selection 1 in ~10 trillion!
!"#$%&'()#'*%+,-*+.,"%"(+/,#$00*&*$%&
!"#$%'()*+",%*-++)."
%/#'.%)0("'#*()-01%234536*$%78#'9,1%+"&(-0,1%0":%'()*+",%."*#;
/#.'-0)%:/#(A,%+">(%%B%*$%@@%34%$
!"#"!"$"%"""#
(Wolfram Erdmann)
Have to be able to “see” the collisions
(Open in FlashPlayer)
How Do We “See?”
http://cms-project-cmsinfo.web.cern.ch/cms-project-cmsinfo/Resources/Website/Media/Videos/Animations/files/CMS_Slice.swfhttp://cms-project-cmsinfo.web.cern.ch/cms-project-cmsinfo/Resources/Website/Media/Videos/Animations/files/CMS_Slice.swf
Higgs Production in 7 TeV pp Collsions
gg! H is dominant production mechanism
2
Irreducible backgrounds in H ! WW, ZZ, !! are from qq annihilation; S/B better than at Tevatron except in VH
-
(V. Sharma)
Search Strategy• SM Higgs mass is unknown parameter• It would be produced in a few ways at the LHC, and in
association with certain particles
• It would couple to particles in proportion to their mass• It’s neutral, and interaction would conserve energy and
other QM quantities
• So, the decay signature would look different depending on mH
• Do a series of searches, in different mass ranges, for most important, kinematically allowed, decay modes and combine them all together in one big search!
3
SM Higgs Decay Modes Vs Mass
[GeV]HM100 200 300 400 500
BR
[pb]
!
-410
-310
-210
-110
1
10
LHC
HIG
GS
XS W
G 2
011
SM = 7TeVs
"l = e, ,",e =
q = udscbbb# lWH
bb-l+ lZH
-+ VBF H
-+ H
qq# lWW
-l+ lWW
qq-l+ lZZ -l+ lZZ
-l+l-l+ lZZ
Mode Mass Range Data Used (fb-1) CMS Document H ! !! 110-150 1.7 HIG-11-021
H ! bb 110-135 1.1 HIG-11-012 (NEW)
H ! "" 110-140 1.1 HIG-11-009
H !WW !2l 2# 110-600 1.5 HIG-11-014
H ! ZZ !4l 110-600 1.7 HIG-11-015
H ! ZZ !2l2" 180-600 1.1 HIG-11-013 (NEW)
H ! ZZ !2l2j 226-600 1.6 HIG-11-017
H ! ZZ !2l2# 250-600 1.5 HIG-11-016 (V. Sharma)
4
Cross Sections for Key SM Background Processes
W + jets
Z + jets Z! !+!"
W ! !"
tt
tt
t + X
t + X(s-chan)
(t-chan)
tW W +W !
WZZZ
165 pb NNLO
43 pb
18 pb5.9 pb
28000 pb NLO
2800 pb NLO
63 pb NLO
10.6 pb
4.6 pb
! (pb) s = 7 TeV
(V. Sharma)
Data set usedfor latest Higgs
search 1.1 to 1.7 fb-1
Data set usedfor latest Higgs
search 1.1 to 1.7 fb-1
Since April, the inst. lumi has
increased by x10!!!
Data set usedfor latest Higgs
search 1.1 to 1.7 fb-1
Data set usedfor latest Higgs
search 1.1 to 1.7 fb-1
In dataset usedhave ~6 simultaneousinteractions (“pileup”)
Data set usedfor latest Higgs
search 1.1 to 1.7 fb-1
• Would be a narrow peak in a falling diphoton mass spectrum
• Experimental resolutions important• Trigger on diphotons. Photons required to be
isolated and energetic (>40 GeV and >30 GeV)
• Background is mainly from large, irreducible QCD (measured from sidebands in data)
• Split selected events into different 8 categories based on the purity of the sample: both in barrel, both pass conversion veto,
H → γγ
1
pγγT > 40 GeV
1
Low Mass Higgs Search : H! !!
7
!1= 86 GeV
!2=56 GeV
Signature: 2 energetic, isolated !, narrow mass peak Background: Large & partly irreducible QCD. Measured from M!! sidebands in data
• Data divided in 8 categories depending on !! mass resolution & their PT
• Background shape fitted by 2nd order polynomial constrained to be positive
• Signal shape : Sum of 3 Gaussians, parameters determined from Z ! ee data
Simulated data not used in analysis
Unconverted !!, both in ECAL barrel
1 converted !, both in ECAL barrel
1 converted !, one in ECAL endcap
Unconverted !!, one in ECAL endcap
Low Mass Higgs Search : H! !!
8
LEE!Prob. to observe max. excess as large as seen in data = 0.05 (1.7!)
CMS Prelim
2- 4"SM
(V. Sharma)
• Look for three final states of tau pair decay (leptonic, hadronic):
• Build visible mass of event from electrons, muons, hadronic taus (doesn’t try to account for neutrinos)
• Look for excess of events in visible mass spectrum
• Divide into two sets of categories based on VBF-likeness of jets:
• Dominant background from
H → τ+τ−
1
e + τhad, µ + τhad, e + µ
1
pT > 30 GeV, mjj > 350 GeV, ∆ηjj > 3.5, η1 · η2 < 0
1
Z → τ+τ−
1
[GeV]vism0 100 200 300 400 500
Even
ts
0
1
2
3
4
5
6
7
8CMS Preliminary
=7 TeVs -11.1 fbh!e!
!!"(10x) HObserved
!!"Ztt
ElectroweakQCD
=120Hm
[GeV]vism0 100 200 300 400 500
Even
ts
0123456789
10CMS Preliminary
=7 TeVs -11.1 fbh!µ!
!!"(10x) HObserved
!!"Ztt
ElectroweakQCD
=120Hm
[GeV]vism0 100 200 300 400 500
Even
ts
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5CMS Preliminary
=7 TeVs -11.1 fbµ!e!
!!"(10x) HObserved
!!"Ztt
ElectroweakFakes
=120Hm
VBF VBF VBF
Low Mass Higgs Search : H! !!
9
µ PT =20 GeV
Jet2 ET =46 GeV
Visible Mass(!!) = 75 GeV
Mass ( jj ) = 580 GeV
!" (jj) = 3.5
Missing ET = 97 GeV
Jet1 ET = 177 GeV
! # "+"0 #$! PTvis = 70 GeV
Search H! !! in two categories: • VBF: 2 jets (PT>30 GeV), "#jj>3.5 #1"#2 350 GeV • Non-VBF: $ 1jet , or 2j failing VBF • In !e + !h , !µ + !h , !e +!µ final states • Background: top, EWK, Z ! !! (irriducible)
VBF modes are cleanest, most sensitive
Important mode, sensitivity improves ~ linearly with data and sophisticated methods
CMS Preliminary
Low Mass Higgs Search : H! b b • gg! H! bb and VBF are dominant
production modes but overwhelmed by enormous QCD di-jet background
• Best option: qq! VH; H ! bb – Major backgrounds are V+jets, VV, ttbar
• Use – VH topology : !"(V,H) > 3 – PT(V)> 100-160 GeV (boosted W/Z) – Tight b-tagging & MET quality – Backgrounds estimated from control data
10
_
Cut
CMS Prelim
(V. Sharma)
µ PT 32 GeV e PT
34 GeV
MET 47 GeV
Intermediate Mass 130 < MH
pp ! WW Is Major Irreducible Background
14
too large !"ll
2010 Data
too large !"ll
Limits From H! WW ! 2l 2!
18
SM Higgs boson with mass 147 < MH < 194 GeV ruled out at 95% CL. SM Higgs boson expected sensitivity 136 < MH < 200 GeV
PS: This update featured only Cut-based analysis. MVA based result coming soon !
CMS Prelim
H ! ZZ ! 4e, 4µ, 2e 2µ : The Golden Channels • Signal: 4 isolated lepton from common vertex • Fully reconstructed, Mass resolution~ 2-4 GeV • Reducible Backgrounds:
– ttbar ! 2l2!2b ; Z+bb – Removed by Isolation & Impact parameter
requirements
• Irreducible background: pp ! ZZ Continuum • Event Selection: Same Flavor, opposite charge
– Z1 : PT(min) > 10, PT(max)>20 GeV, 60 4
• Reducible background contribution from data • ZZ Continuum:
– Shape known at NLO, corrected for gg ! ZZ !4l evaluated with MCFM
– Rate obtained from Z yield in data & theoretical prediction for ratio of ZZ to Z cross sections
19
PT 43 GeV
PT 26 GeV
PT 20 GeV PT
48 GeV
M4µ = 201 GeV
(V. Sharma)
H ! ZZ ! 4l : Observed Yields & Limits
20
Three pairs of events at M4l = 122, 142 & 165 GeV Only M4l =142 GeV consistent with SM Higgs expectation
CMS PRELIM
(V. Sharma)
More High Mass Channels
• Also have looked in ZZ channels with: 2 leptons + 2 taus; 2 leptons + 2 neutrinos; 2 leptons + 2 quarks (with/without b-tag)
• Neutrino channel most sensitive because of large missing energy and higher branching fraction for Z
• Also have no evidence of Higgs like other channels
Combine All Searches
6 3 Search channels used in the combination
3 Search channels used in the combinationThe combination presented in this note is based on eight major channels classified by the finalHiggs decay chain signature as shown in Table 1. The mass search regions are specific to eachanalysis. The analyzed integrated luminosity varies from channel to channel in the range from1.1-1.7 fb−1.
From Table 1, one can also see that different analysis strategies are employed in differentsearches. They include three basic types: cut-and-count analyses, analyses of binned distri-butions, and unbinned analyses tracking individual events and using parametric models ofsignal and background shapes.
Each of the major analyses is a combination in itself with 2 to 8 independent sub-channels.In the overall combination there are 27 independent sub-channels at low mass Higgs bosonsearches (mH < 135 GeV/c2) and 24 sub-channels in the high mass region (mH > 250 GeV/c2).
The last column in Table 1 shows the number of nuisance parameters (systematic uncertainties)in each analysis. The total number of independent nuisance parameters in the current combi-nation is 267, of which 241 are used in the combination in the low mass range and 146 in thehigh mass range. There are 25 correlated sources of uncertainties appearing in more than onemajor search. The remaining ones are specific to individual analyses 1.
Table 2 shows the full list of uncertainties correlated across more than one major analysis. Thetop block in the table is a subset of the list prepared by the LHC Higgs Combination Group [20].The bottom block are correlated errors that are correlated within CMS only. Quantities af-fected by the uncertainties listed in Table 2 are all positive definite and, hence, modeled aslog-normals.
In the following subsections, we give a brief description of search strategies for the eight chan-nels used in this combination. Detailed information can be found in references provided withineach sub-section.
Table 1: Summary information on the analyses included in the combination. The first numberin the last column gives the number of nuisance parameters correlated across two or moreanalyses. The second number refers to the number of nuisance parameters specific to oneanalysis only.
channel mass range luminosity number of type number of(GeV/c2) (fb−1) sub-channels of analysis nuisances
H → γγ 110-150 1.7 8 mass shape (unbinned) 3+40=43H → ττ 110-140 1.1 6 mass shape (binned) 10+25=35H → bb 110-135 1.1 5 cut&count 10+59 = 69
H → WW → 2�2ν 110-600 1.5 5 cut&count 15 +79 =94H → ZZ → 4� 110-600 1.7 3 mass shape (unbinned) 14+20=34
H → ZZ → 2�2τ 180-600 1.1 8 mass shape (unbinned) 13+10=23H → ZZ → 2�2ν 250-600 1.6 2 cut&count 14+4=18H → ZZ → 2�2q 226-600 1.6 6 mass shape (unbinned) 12+15=27
TOTAL (8) 110-600 1.1-1.7 27 for low mH 241 for low mH24 for high mH 146 for high mH
1The majority of them are actually also correlated between different sub-channels within an analysis.
)2Higgs boson mass (GeV/c100 200 300 400 500 600
SM!/
!95
% C
L lim
it on
1
10
210 = 7 TeVsCMS Preliminary, Combined observedCombined expected
)-1 bb (1.1 fb"H )-1 (1.1 fb## "H )-1 (1.7 fb$$ "H )-1 WW (1.5 fb"H )-1 4l (1.7 fb" ZZ "H )-1 (1.1 fb# 2l 2" ZZ "H )-1 2l 2q (1.6 fb" ZZ "H )-1 (1.6 fb% 2l 2" ZZ "H
Combined observedCombined expected
)-1 bb (1.1 fb"H )-1 (1.1 fb## "H )-1 (1.7 fb$$ "H )-1 WW (1.5 fb"H )-1 4l (1.7 fb" ZZ "H )-1 (1.1 fb# 2l 2" ZZ "H )-1 2l 2q (1.6 fb" ZZ "H )-1 (1.6 fb% 2l 2" ZZ "H
Combined observedCombined expected
)-1 bb (1.1 fb"H )-1 (1.1 fb## "H )-1 (1.7 fb$$ "H )-1 WW (1.5 fb"H )-1 4l (1.7 fb" ZZ "H )-1 (1.1 fb# 2l 2" ZZ "H )-1 2l 2q (1.6 fb" ZZ "H )-1 (1.6 fb% 2l 2" ZZ "H
)2Higgs boson mass (GeV/c100 200 300 400 500 600
of S
M H
iggs
hyp
othe
sisS
CL
-510
-410
-310
-210
-110
1
90%95%
99%
-1 = 1.1-1.7 fbintCombined, L = 7 TeVsCMS Preliminary,
Observed! 1±Expected ! 2±Expected
Observed! 1±Expected ! 2±Expected
At 95% CL exclude 145-216, 226-228, 310-400 GeV
In Near Future
G. Tonelli, CERN/INFN/UNIPI MB150 August 29 2011 23
Higgs Sensitivity : 1, 2, 5 and 10 fb-1 @ 7 TeV
(NB: study a little dated)
Conclusions
• LHC and CMS (ATLAS too ;) performing very well• We are really now, finally!, closing in on the SM
Higgs
• If we get something like >5 fb-1 at 7 TeV, it looks like we should either be able to make an observation of the SM Higgs, or exclude it
• Either case would be a major result• Thanks for having me down to Wichita!