ICHEP Moscow July 06

1
Results on Higgs Searches _at_ the Tevatron

• ICHEP Moscow July 06
• Gregorio Bernardi, LPNHE-Paris
• for the CDF and DØ Collaboration

Run IIb upgrade relevant for Higgs -
Layer Ø for DØ SMT
Layer 0 now inserted and fully readout
Luminosity Introduction to Higgs SM Higgs
Searches at High mass Combined CDF, D0 and
CDFD0 results. Sensitivity Prospects
Excellent noise performance S/N18!!
2
Tevatron Performance
85 Average Efficiency
Currently in shutdown Silicon and Trigger
upgrades
4-8 fb-1 expected by end of 2009
This talk 260- 1000 pb-1
Run I Lumi
3
SM Higgs Production and Decays
Decays
Excluded
mH (GeV/c2)
Production cross section (mH 115-180) ? in the
0.8-0.2 pb range for gg ? H ? in the 0.2-0.03
pb range for WH associated vector boson
production
Dominant Decays ? bb for MH for MH 135 GeV
Search strategy MH production WH and ZH with H?bb decay
Backgrounds Wbb, Zbb, top, WZ,QCD
complement with WWW WW MH 135 GeV gg ?H
production with decay to WW Backgrounds
WW, DY, W/ZZ, tt, tW, ?? complement with WWW
4
SM Heavy Higgs H ? WW ? lnln
Search strategy ? 2 high Pt leptons and
missing Et ? WW comes from spin 0
Higgs leptons prefer to point in the same
direction. But Higgs mass peak cannot be
reconstructed due to the presence of 2 n? look
for an excess CDF and DØ already published on
0.3-0.4 fb-1
H
4th generation?
CDF
5
H ? WW ? lnln

• Selection Strategy
• Presection lepton
  ID, trigger, opposite charge leptons
• Remove QCD and Z?ll- ET 20 GeV
• Reject events with mismeasured jet energy
• Higgs Mass Dependent Cuts Invariant Mass
  (Mll-) Min. Transverse Mass
  Sum of lepton pTl and ET (S pTl ET)
• Anti tt(bar) cut HT S
  PTjet
• Spin correlation in WW pair Df(l,l)

DATA CORRESPOND TO LUMI L 950 pb-1 in H ? WW
? ee and em
L 930 pb-1 in H
? WW ? mm
Major backgrounds
WW production
Now measured at the Tevatron by both expts. in
agreement with NLO calculation12-13.5 pb
6
H ? WW ? ee / em / mm (950-930 pb-1)
Preselection Trigger, ID, leptons of opposite
charge pTl1 15 GeV pTl2 10 GeV
m m
e e
e m
Missing Transverse Energy 20 GeV Cut (to
suppress Z/g ? ll- background)
e m
e e
m m
7
H?WW final selection
mm
em
ee
selections
MH160GeV (x10)
8
Results for H?WW
SM only a factor 4 away We exclude 4th
generation models, for which mH150-185
GeV
9
Search for WH?WWW (2 like-sign leptons)
Low BR, but channel important in the
intermediate region 125-145 GeV
10
WH?WWW (360-380 pb-1)

• Event preselection
• Same charge di-lepton from W (one from H?WW,
  the other from prompt W)
• Suppress SM backgrounds pT (lepton) 15 GeV,
  h

11
Final Selection for WWW
Topological Likelihood discriminant, built on 3
variables per channel
12
Exclusion limit for WWW
Events after Topological Likelihood discriminant
Effect on global combination significant in the
intermediate region (125-145 GeV)
13
Combining Higgs boson Searches
Combination Procedure
CDF uses a Bayesian approach
DØ uses the CLs (LEP) Method the CLS
confidence interval is a normalization of CLSB
CLSB signal bkgd
hypothesis, CLB bkgd only hypothesis
CLS CLSB/CLB CLSB CLB are defined
using a test statistic Test statistic used
is the Log-Likelihood Ratio (LLR-2 ln Q)
generated via Poisson statistics
(Qe-(sb)(sb)d/e-bbd) s,b,dsig.,bkd,data)
Tevatron Higgs combination is done with both
methods ? they give results compatible within 10.
14
Systematic Uncertainties _at_ CDF
15
Combined Higgs boson Search _at_ CDF
All low mass channels analyses use 1 fb-1 of data
! WH (ln bb) ZH (nn bb) ZH (ll- bb) For mH
115 GeV, the 95CL Limit/SM is 9 (expected) and
13(observed)
16
Combined Higgs boson Search _at_
Combination Procedure

• 16 independent channels have been combined,
  taking into account correlated uncorr.
  systematics
• WH (e) ST, DT
• WH (m) ST,DT
• WH ( l ) ST DT
• ZH (nn) ST,DT
• ZH (ll-) DT
• WWW (ee,em,mm)
• WW (ee,em,mm) ? 1 fb-1, others 260-390
  pb-1)

WH? enbb
17

With low statistics (0.3 fb-1) Data/SM 15 at
115 GeV 9 at 160 GeV With
higs statistics (1 fb-1) Data/SM 5 at 160
GeV
Based on our first experiences with Data, can
cdfdØ reach exclusion for a 115 GeV Higgs with
2 fb-1 as early simulation studies were
expecting ?
18
Examples of Developments

• New b-tagging tool
• Combines various variables from the track based
  b-tagging tools in a Neural Network, trained on
  Monte Carlo
• Performance measured on data
• Substantial improvement in performance over
  constituent input b-taggers
• Increase of 33 in efficiency for a fixed fake
  rate of 0.5

• Jet energy resolution (recalibration track-jet
  algorithm)
• Subtract expected energy deposition in cal
• Add the track momentum
• Add the energy of out-of-cone tracks
• Improve the jet energy resolution by 20
• Use improved calorimeter calibration
• Z? bb to calibrate b-jet response
• L1-cal Trigger (DØ Upgrade)
• Important for difficult channels ? efficiency
  improvement (ZH?nnbb, hbb)
• Addition of SMT Layer 0 (Upgrade)
• r_at_L1 2.7 cm ? r_at_L0 1.6 cm
  
• better impact parameter resolution
• More redundancy in pattern
  
• Recognition for higher luminosity reduce fake
  track
• Keep functionality in case of degradation of
  Layer 1 due to radiation damage

19
Sensitivity Prospects extrapolated from 0.3 fb-1
results
On the Horizon

• Will 6X more lumi take our x-section factors of
  15/9 to 1?? Maybe.. with analyses improvements
  under work in CDF and DØ ? Close !!!!!

Equiv Lumi gain (_at_115)
Xsec Factor Xsec Factor
Ingredients (DØ)
mH115 GeV mH 160 GeV
Today with 330 fb-1
-
15 9
6.0
6.1 3.7
Lumi 2.0 fb-1
NN b-Tagger/L0
3.0
3.5
NN analysis selections
1.7
2.7 2.8
Dijet-mass resolution
1.5
2.2
Increased Acceptance
1.2
2.0 2.5
New channels
1.2
1.9 2.1
1.7
Reduced Systematics
1.2
1.2 1.5
Combine DØ and CDF
2.0

• At 160 GeV needs 4.5 fb-1
• (2 fb-1 x1.52)

• At 115 GeV needs 2.8 fb-1

? 95 CL exclusion for mH 115-180 GeV with 6
fb-1
20
Tevatron SM Higgs Combination
21
SM Higgs Summary
Conclusions

• First time with essentially complete result
• Full impact of systematics uncertainties is
  included
• Analyses have been steadily improving due to
  optimization, already close to the Prospective
  reports
• Combined limit looks very promising
• High mass region benefits a lot from H?WW type
  analyses (H and WH production), but low mass as
  well, as low as 120 GeV.
• Our outlook for the future looks very interesting
• LHC experiments will have to work hard to get the
  signal, if the Higgs is light (
  Tevatron has insight also if it is close to 160
  GeV. Barring accidents, the Tevatron could have
  evidence by 2009, if its there.
• If the Higgs is light we might end-up combining
  with LHC!

22

• Conclusions

Will we soon understand better whats going on in
the Universe?
if we get somewhere around here the
answer is YES! So well try hard !!!
23
H ? WW ? ee / em 950 pb-1
Preselection Trigger, ID, leptons of opposite
charge pTl1 15 GeV pTl2 10 GeV
e e
e m
PRESELECTION (L 950 pb-1)
e m
e e
24
H ? WW ? mm 930 pb-1
m m
m m

• 2 muons of opposite charge
• Leading muon Pt 15 GeV
• Nex- to- leading muon
• Pt 10 GeV
• Cuts optimized for a Higgs mass of 160GeV

m m
m m