Jets at the Highest Energies (Jets at D

1
Jets at the Highest Energies(Jets at DØ CDF)

• Don Lincoln
• Fermilab
• (for the DØ CDF collaborations)

2
The Richness of Jet Physics
3
Subjects for Jet Study

• Mechanics of QCD
• Calculations with approximations (LO, NLO, NNLO,
  etc.)
• Scale dependences (integration cutoffs)
• Algorithm choices (cone, k?, etc.)
• Constrain parton distribution functions (PDFs)
• Showering/fragmentation models
• New physics
• Many collisions at high CM energy
• Background studies
• Jet (s mb or mb)
• Higgs top (s pb) ? 106 - 108!

4
Fermilab Tevatron
Run II Ecm 1.96 TeV Instantaneous
Luminosity 1032 cm-2 s-1 New Integrated
Luminosity Milestone!
2 _at_ 400 GeV 5 _at_ 600 GeV
NEW
5
Tevatron Milestone1 fb-1
!
NEW
6
DETECTORS

• New Silicon Detector
• New Central Drift Chamber
• New End Plug Calorimetry
• Extended muon coverage
• Faster DAQ

• New Silicon Detector and Central
• Fiber Tracker in a 2T solenoid
• Substantially upgraded muon system
• Faster DAQ

7
Triggering/Jet Algorithms

• Both experiments rely on multi-tiered triggers
• start with calorimeter towers
• followed by ever more sophisticated jet
  algorithms
• Jet Reconstruction
• Cone algorithm DR(y,f ) R ? (0.5, 0.7, 1.0)
• Iterative, start with calorimeter seeds
• Remove duplicates
• Check the midpoint between jets (IR safety)
• k? Algorithm (more later)

8
(No Transcript)
9
Inclusive Jet Cross Section

• Run II Cone Algorithm
• R 0.7
• L 380 pb-1
• Two rapidity bins
• Highest Pt jet is 630 GeVEcm of 1.2 TeV
• Note the 9 orders of magnitude

10
Inclusive Jet Cross Section

• Jet energy uncertainty is dominant error (JES)

11
Di-jet Mass

• L 143 pb-1
• Run II cone algorithm
• R 0.7
• yjet lt 0.5
• JETRAD comparison
• mR mf 0.5 Ptmax
• Rsep 1.3
• Update expected this winter.

12
Inclusive Jet Cross Section
CDF RunII Preliminary L 177 pb-1 0.1 lt hDet lt
0.7 JetClu Cone R 0.7

• Run II Cone Algorithm
• Radius of 0.7
• L 177 pb-1
• Central rapidity

13
L 177 pb-1 0.1 lt hDet lt 0.7 JetClu Cone R
0.7
Comparison of data and theory
L 177 pb-1 0.1 lt hDet lt 0.7 JetClu Cone R
0.7
Ratio of cross section
14
KT Algorithm

• Cluster together calorimeter towers by their kT
  proximity.
• Infrared and collinear safe at all orders of
  pQCD.
• No splitting and merging.
• No ad hoc Rsep parameter necessary to compare
  with parton level.
• Every parton, particle, or tower is assigned to a
  jet.
• No biases from seed towers.
• Favored algorithm in ee- annihilations!

KT Algorithm
Raw Jet ET 533 GeV
Raw Jet ET 618 GeV
CDF Run 2
Only towers with ET gt 0.5 GeV are shown
15
KT Jet Cross-Section
0.1 lt yjet lt 0.7, L 385 pb-1
D 0.5
D 0.7
D 1.0
16
m -Tagged Jets

• Jets containing heavy flavor can contain muons
• e.g. b ? c W ? m n
• Jets also can be m - tagged from p and K decay (
  others)
• All jets are eventually m tagged
• Define a jet as m tagged only if
• muon collinear with jet
• muon created in a cylindrical volume of radius 10
  cm

Before r 10 cm correction

• Run II Cone Algorithm
• Radius of 0.5
• Requires muon in 0.5
• L 300 pb-1
• yjet lt 0.5
• P?(m) gt 5 GeV/c

17
m -Tagged Jets

• Cant find vertex of muon-creation most of the
  time
• p ? m n is low-kink most of the time
• some pion decay inside calorimeter volume
• Pythia-derived corrections

Dominant error is fraction of pion decay inside
calorimeter
18
m -Tagged Jets
NLO Calculation of m - Tagged Jets

• Calculate NLO inclusive jet cross section
  (NLOJet)
• Multiply cross section by Pythia-determined ratio

DØ Preliminary
DØ Preliminary
After corrections
19
The b-Jet Inclusive Cross-Section
Construct the invariant mass of particles
pointing back to the secondary vertex!
98 lt pT(jet) lt 106 GeV/c
Monte-Carlo Templates
MidPoint Jets Rcone 0.7 Fmerge 0.75 yjet lt
0.7
20
The b-Jet Inclusive Cross-Section

• The data are compared with PYTHIA (tune A)
  Data/PYTHIA 1.4
• Comparison with MC_at_NLO coming soon!

21
The b-diJet Mass Cross-Section

• Etjet 1 gt 30 GeV, Etjet 2 gt 20 GeV, hjet lt 1.2

Preliminary CDF Result
sbb 34.5 ? 1.8 ? 10.5 nb
QCD Monte-Carlo Predictions
PYTHIA Tune A CTEQ5L 38.71 0.62nb
HERWIG CTEQ5L 21.53 0.66nb
MC_at_NLO 28.49 0.58nb
Differential Cross Section as a function of the
b-bbar DiJet invariant mass!
Predominately Flavor Creation!

• Large Systematic Uncertainty
• Jet Energy Scale (20).
• b-tagging Efficiency (8)

22
The b-Jet Inclusive Cross-Section

• Etjet 1 gt 30 GeV, Etjet 2 gt 20 GeV, hjet lt 1.2

Preliminary CDF Result
sbb 34.5 ? 1.8 ? 10.5 nb
QCD Monte-Carlo Predictions
PYTHIA Tune A CTEQ5L 38.7 0.6 nb
HERWIG CTEQ5L 21.5 0.7 nb
MC_at_NLO 28.5 0.6 nb
MC_at_NLO Jimmy 35.7 2.0 nb
Differential Cross Section as a function of the
b-bbar DiJet invariant mass!
JIMMY Runs with HERWIG and adds multiple parton
interactions!
JIMMY MPI J. M. Butterworth J. R. Forshaw M. H.
Seymour
Adding multiple parton interactions (i.e. Jimmy)
to enhance the underlying event increases the
b-bbar jet cross section!
23
b-bbar DiJet Correlations
Tune A!
Differential Cross Section as a function of Df of
the two b-jets!

• The two b-jets are predominately back-to-back
  (i.e. flavor creation)!
• Pythia Tune A agrees fairly well with the Df
  correlation!

Not an accident!
24
b-bbar DiJet Correlations

• The two b-jets are predominately back-to-back
  (i.e. flavor creation)

• Pythia Tune A agrees fairly well with the Df
  correlation

Agrees very well with MC_at_NLO HERWIG JIMMY!
25
Azimuthal Decorrelation

• Standard DØ (y,f ) cone algorithm
• R 0.7
• 150 pb-1
• Published
• Phys. Rev. Lett. 94 221801 (2005)

LO fails in expected way NLO fails at
large Df, where soft effects matter
At high PT, events are more back-to-back
26
Azimuthal Decorrelation
Data/Herwig agreement good Data/Pythia agreement
improved by addition of more ISR PARP(67) 1.0
? 4.0 Tune A-ish
Data/NLO agreement good
27
Direct Photons

• Sensitive to gluon PDF and hard scatter dynamics
• Separating photons from jet backgrounds is hard
• Use neural net (NN).
• Track isolation, and calorimeter shower shape
  variables

28
Direct Photons

• L 330 pb-1
• y lt 0.9
• Highest Pt(g) is 442 GeV/c
• 3 events above 300 GeV/c not displayed
• Very clean probe of hard scatter dynamics

29
Direct Photons

• No JES error
• purity dominates (added in displayed error)
• Errors 20
• Need NNLO for sensitivity to PDF effects?

30
The Transverse Regionsas defined by the
Leading Jet

• Look at charged particle correlations in the
  azimuthal angle Df relative to the leading
  calorimeter jet (JetClu R 0.7, h lt 2).
• Define Regions
• Toward
• Df lt 60o
• Transverse
• 60o lt -Df lt 120o and
• 60o lt Df lt 120o
• Away
• Df gt 120o

Charged Particle Df Correlations pT gt 0.5
GeV/c h lt 1
31
Charged Particle DensityDf Dependence Run 2

• Df dependence of the charged particle density,
  dNchg/dhdf
• Particle cuts
• pT gt 0.5 GeV/c
• h lt 1
• Df plotted as compared to jet 1
• rotated to 270o
• 30 lt ET(jet 1) lt 70 GeV

Leading Jet h lt 2.0 JetClu R 0.7
Back to Back h1,2 lt 2.0 JetClu R 0.7 Df12 gt
150o ET(jet 2)/ET(jet 1) gt 0.8 ET(jet 3) lt 15 GeV
32
Charged Particle DensityDf Dependence Run 2
Leading Jet
Back to Back
33
Charged PTsum DensityPYTHIA Tune A vs HERWIG
PYTHIA Tune A Works well (Unsurprising,
with emphasis on Tune)
HERWIG Insufficient charged particle flow
in Transverse Region
34
Tuned JIMMY vs PYTHIA Tune A
Charged PTsum Density dPT/dhdf
100.0
Charged Particles
Leading Jet
30 lt ET(jet 1) lt 70 GeV
Compare PYTHIA to JIMMY (include ISR, FSR and
multiple parton interactions)
h
(
lt1.0, PTgt0.5 GeV/c)
PY Tune A
10.0
Charged PTsum Density (GeV/c)
1.0
CDF Preliminary
Jet1
"Transverse"
data uncorrected
Region
theory CDFSIM
0.1
0
30
60
90
120
150
180
210
240
270
300
330
360
Df
(degrees)
Leading Jet, Data vs. PYTHIA (Tune A)
JIMMY MPI J. M. Butterworth J. R. Forshaw M. H.
Seymour
35
Summary
KT Algorithm
36
www-d0.fnal.gov/lucifer/PowerPoint/PIC2005.ppt