Observing an invisible Higgs at the LHC via the ttH channel based on the full reconstruction of the

1
Observing an invisible Higgs at the LHC via the
ttH channel based on the full reconstruction of
the two top decays

• Teh Lee Cheng
• Supervisor Pedro Teixeira-Dias
• Royal Holloway University of London
• 4 September 2003

2
Outline

• Motivation
• Analysis
• Results
• Conclusions and outlook

3
Invisible Higgs search at the LHC
Motivation

• Extensions to standard model, e.g. supersymmetry.
• The minimal supersymmetric standard model (MSSM)
  predicts 5 Higgs bosons
• h, H (CP-even) ? A (CP-odd) ? H?
• h is the lightest Higgs.
• h? c10 c10 can be dominant (c10 lightest
  neutralino)
• In Majoron model (violate R parity)
• h? JJ can be dominant (J Majoron)

4
Search Channel

• Vector boson fusion (VBF) qq? qqH
• Associated production with a vector boson
  qq? VH, VW, Z
• Associated production with a tt pair gg, qq?ttH

5
Analysis
tt? bjjbln
t? bjj
6
Standard Analysis

• Selections and cuts
• 1 lepton e or m with separate thresholds and
  vetoes.
• 2 b-tagged jets, pT gt 30 GeV
• 2 or more light jets, pT gt 30 GeV
• All particles in central region, h lt 2.5
• Require 1 hadronic top
• Cut on pTmiss and mT(l, miss)

7
Standard Analysis (cont.)
Hadronic top reconstruction

• All pair combinations are sought, their invariant
  masses are evaluated and the following constraint
  imposed
• Jet pairs passing the constraint are calibrated
  such that
• Each combination paired with b-jets. Top mass
  constraint imposed to find hadronic top
  candidate

8
Full Reconstruction of the two top decays
tth signal

• Exact ONLY for the specific tt process
• tt ? bjj bln (only n is missing)

• Step 1
• Compute pz of neutrino (pzn) based on mln mW
• Two solutions, real (physical) or complex
  (unphysical).
• Physical pzn permit full rec -gt physical
  events.
• Unphysical pzn forbid full rec -gt unphysical
  events.
• Physical events are rejected by the standard
  analysis.
• Two analysis can be performed in parallel.

9
Full Reconstruction of the two top decays (cont.)

• Step 2
• 2 b-jets and 2 possible neutrino momenta
• ? fourfold ambiguity b1jjb2ln1(n2),
  b2jjb1ln1(n2)
• Resolve by relative distances of the inv.
  masses (mbjj / mbln) to mt 175 GeV.
• ? simple method the best hadronic top first,
  and then semileptonic top.
• Combination giving the closest inv. mass to
  mt is selected.
• ? c2 method reconstruct both simultaneously
  using a variable c2
• Combination giving minimal c2 is selected.
• Step 3
• Reconstruct a number of variables mT, pTmiss,
  Dfln, pT(top) etc.

10
Results
Results of standard analysis

• With optimised cuts (mT gt 110 GeV , pTmiss gt 170
  GeV), S/?B 3.62. (10 improvements)
• To reach 5s for discovery need 192 fb-1 (2 yrs
  LHC).

11
Results of full reconstruction

• After selecting physical events, S/?B 0.329
• Physical vs unphysical, 46 (signal) 82
  (background)

Signal
Background
12
Results of full reconstruction
Distributions of pTmiss and Dfln
13
Distributions of the reconstructed two top decays
Method 1 (simple)
Method 2 (c2)
Hadronic top
Semileptonic top
14
Conclusions and outlook

• Standard analysis reconstructs only the hadronic
  top and implements cuts on pTmiss and mT.
• Using optimised cuts, a discovery is possible in
  less than 2 years of LHC (high luminosity).
• Full reconstruction is possible for a minority of
  signal events and a majority of background
  events.
• Some but limited discriminating features between
  signal and background, e.g. Dfln.
• Statistical technique (e.g. Fisher discriminant)
  for maximum discrimination, in standard analysis
  and full reconstruction.
• Include other background e.g. ttZ and
  supersymmetric processes.