Weighing Truth

1
Weighing Truth

• Erik Brubaker
• University of Chicago
• Cook/Brandenberger Symposium
• May 6, 2006

2
What is this talk about?
A Precision Measurement of the Top Quark Mass
Using the Template Method in the Lepton Jets
Channel at CDF II

• The Physics
• The Experiment
• The Measurement

3
Particle Physics

• Bearers of the faltering standard of
  reductionism.
• Most fundamental particles and forces.
• Standard Model of Particle Physics.

4
The Standard Model

• Particle content
• Interactions (forces) governed by a set of gauge
  symmetries
• SU(3)C ? SU(2)L ? U(1)EM ( gravity??)
• Massive gauge bosons ? symmetry broken
• Higgs mechanism

5
Outstanding questions in particle physics

• What is mass?
• Why do particles have the masses they do?
• How do particles acquire mass?
• How are the fundamental forces related?
• Strong, weak, e-m, gravity?
• Grand Unification Theory
• Theory of Everything

LU colloquium,September 1999!
6
The Top (Truth!) Quark

• Feels strong, electroweak, gravitational forces.
• Short-liveddoesnt hadronize (t510-25 s).
• Studied directly only at Fermilab.
• Especially interesting due to its unexpectedly
  large mass.
• Mtop related through QFT loop effects to other
  EW observables.
• Important for precision tests of SM.
• With mW, constrains mHiggs!

Mass (GeV/c2)
7
High-Energy Physics Experiments

• Massive in scale, cost, and personpower.
• Want to probe small distance scales
• Need high-energy interactions.
• Accelerate and collide beams of particles.
• Measure properties of everything coming out.
• Reconstruct the collision, including intermediate
  products.
• Infer parameters of a theory.

8
Fermilab Accelerator Complex

• Multiple stages accelerate protons from rest to
  980 GeV.
• Also produce and accelerate anti-protons!
• Final stage is the Tevatron
• Highest energy accelerator in the world
• Produces proton/antiproton collisions at 1.96 TeV.

2 km
9
Top Quarks in Tevatron Collisions

• Top quarks produced in pairs.
• Top decays to W boson and b quark.
• Quarks observed as jets of particles.
• W boson can decay to lepton/neutrino, or to two
  quarks.
• Our events contain one of each W decay type.

Final state e/m, n, 4 jets (2 from b)
10
Cross-sections, Luminosity, and Rates

• Cross-section
• The relative probability of a given process
  occurring.
• Literally, an area presented by a target particle
  to a beam particle.
• stt 7 pb.
• Luminosity
• Intensity-like property of the beam(s).
• Tevatron typical lumi L 1032 cm-2 s-1.
• Event rate
• Rate s x L
• Tevatron ttbar 0.7 mHz.
• Tevatron all events 1.7 MHz.

(CDF x 100)
Cross-section
pp center-of-mass energy
11
Collider Detector at Fermilab

• Onion-like set of subsystems.
• Each subdetector measures different particles or
  different properties.

12
An Easy Measurement Z Boson Mass

• Find Z boson decaying to two electrons.
• Electrons easy to identify.
• Low backgrounds.
• Electron energy momentum well measured.
• Combine 4-momenta and calculate mass of parent
  particle.
• Repeat several hundred thousand times. Done!

Z?ee Event Schematic
Calorimeter
E1
p1
p2
Tracker(B field)
E2
13
A Difficult Measurement

• Complicated events
• Unobserved neutrino
• Only 50 of signal events have leading 4 jets
  from tt decay.
• 12 ways to interpret 4 jets ? 4 quarks.

• Jet energy resolution effect andjet energy scale
  (JES) systematics
• Resolution 85/vET ? statistical uncertainty.
• Systematic 3 ? systematic uncertainty.
• Background contamination
• Well understood SB of 11101.
• Must be treated properly to avoid bias.

14
Top Mass at CDF

• Robust program of top mass measurements 12
  results
• Different final states
• Different methods
• Good statistics (360 events)? precision
  measurements
• Many analysis techniques with different
  sensitivities ? high confidence

15
Template Analysis (LJets) Overview

• Kinematic fit to reconstruct top quarkmass in
  each event

• Invariant mass of possible W decay jets

mtreco
Observables
mjj
DJES
Mtop
Parameters

• DJES deviation from nominalin units of
  external calib.

• Maximum likelihood methodusing 4 subsamples
  constrainDJES 0 1 sc

This method,318 pb-1 PRL 96, 022004PRD 72,
032003
16
Template (LJets) Results680 pb-1
2-tag
2-tag
1-tag(T)
1-tag(T)
1-tag(L)
0-tag
1-tag(L)
0-tag
mtreco in data w/ fit overlaid
mjj in data w/ fit overlaid
By simultaneously measuring the jet energy scale
using the W boson decay, a (fixed) systematic
uncertainty is converted to a statistical
uncertainty that goes like 1/sqrt(N) as we
accumulate more data!
17
Measurement Impact

• Is there a Standard Model Higgs?
• If so, what is its mass?
• A precise top quark mass measurement can help
  with the answers!

Summer 04 average
Summer 05 average
Winter 06 average
95 CL upper limit. But mH gt 114 GeV fromdirect
searches!
Summer 04 mHlt260GeV mH11469-45GeV
Summer 05 mHlt186GeV mH9145-32GeV
Winter 06 mHlt175GeV mH8942-30GeV
18
Conclusions

• Particle physics is an exciting field.
• Large Hadron Collider in Geneva will certainly
  discover something.
• Complicated experimental techniques are
  interesting and rely on other physics,
  statistics,
• Weve made an impressive and quite precise
  measurement of the top quark mass, further
  constraining our understanding of the Standard
  Model and what might lie beyond.
• Thank you Profs. Brandenberger and Cook!

19
Backup Slides
20
Particle Content of Fundamental Theory
21
Measure JES Using Dijet Mass

• Build templates using invariant mass mjj of
  allnon-tagged jet pairs.

• Rather than assuming JES and measuring MW...
• Assume MW and measure JES
• Parameterize P(mjjJES) same as P(mtrecoMtop)

22
Template (LJets) Results680 pb-1
Likelihood contours in Mtop-DJES plane
23
Combination of CDF Results

• Use BLUE (Best Linear Unbiased Estimator)
  technique.
• NIM A270 110, A500 391.
• Accounts for correlations in systematics.
• Stat correlations in progress.
• So far only combine measurements on independent
  datasets (incl run I).

Updated CDFD0 combined result coming
24
Keep an Eye on This

• See some discrepancy between Ljets, Dilepton
  channel Mtop measurements.
• Statistically consistent so far
• ME(dil) vsTempl(Ljets)c2 2.9/1, p0.09.
  (Accounts for correlated systematics)
• But what if it persists?
• Could there be a missing systematic?
• Would have to affects the channels differently
• Could our assumption of SM ttbar be incorrect?
• Will be interesting to see all-hadronic
  measurements.

Stay tuned
25
Conclusions

• CDF has surpassed our run IIa goal of 3 GeV/c2
  precision on Mtop.
• Goal assumed 2 fb-1!
• With in situ JES calibration, dominant
  systematic now scales as 1/sqrt(N).
• 1 uncertainty on Mtop is in sight as we
  concentrate on reducing remaining systematics

26
Systematics ISR/FSR/NLO

• Method in hand to use Drell-Yan events to
  understand and constrain extra jets from ISR.
• Constraint scales with luminosity.
• Easily extendible to FSR.
• MC_at_NLO sample shows no addl NLO uncertainty is
  needed.

Df(t-tbar)
27
The Higgs mechanism

• Widely accepted theory standard model status
• Particles interact with Higgs field, acquire mass
• Predicts existence of Higgs boson
• If its a particle, we can find it

28
Unification of Forces

• Is there an energy range where the basic forces
  have the same strength?
• Problem curves dont meet
• Supersymmetry is one possibility
• MSSM predicts new particles

29
More than Just Particles and Fields

• High energy physics experiments are huge
• Staffing
• Budget
• Complexity
• Data volume
• Sheer size
• Lines of code
• Number of components
• Number of coffee breaks