Lepton Identification at Hadron Colliders

1
Lepton Identification at Hadron Colliders

• c. mills

2
Introduction Leptons in Physics

• At hadron colliders, QCD processes prevail
• Higher cross-section than electroweak
• Leptons only produced by electroweak processes
• Flag for these rarer processes
• Used in triggers and offline selection
• Look for W, Z, top (strong production, weak
  decay), and ?
• Start with general idea, then move to actual
  implementation

3
Leptons in a Generic Detector

• Nature 3 leptons
• e (stable)
• m (2.2 x 10-6 s)
• Even a 10 GeV muon has a 99.99 chance of
  escaping the detector (5 m radius) without
  decaying
• t (2.9 x 10-13 s)
• Even a 1 TeV tau has an immeasurably small (1
  part in 1045) chance to escape the detector
• Jargon lepton e or m

Decays inside detector, usually hadronically,
into a jet of particles
4
A Generic Detector
muon

• Electrons
• Track
• Stop (shower) in EM calorimeter
• Muons
• Track
• Passes through calorimeter
• track in muon detector

Muon detectors
Hadronic cal.
EM cal

tracking
electron
5
Electron Backgrounds

• Jet Catch-all term for fakes of hadronic origin
• Tracks energy in calorimeter
• Nasty case pp0 gives one track EM energy
• Photon
• Need to pick up a track
• Conversion g ? e e
• Muon
• Yes, really Energetic muons can emit
  bremstrahlung photon in EM cal track from muon
  (rare)
• Heavy-flavor decay
• Real electrons but treated as background tricky

6
Muon Backgrounds

• Less background than electrons in general
• Jet Catch-all term for fakes of hadronic origin
• Tracks energy in calorimeter
• Nasty case punch-throughs, K decay-in-flight
• Cosmic rays
• Real muons
• Heavy-flavor decay
• Real muons but treated as background tricky

7
CDF A Real Detector

• Forward-backward and azimuthally symmetric
• From the beamline outward
• Silicon vertex detector
• Drift chamber tracker
• Solenoid
• Electromagnetic calorimeter (with shower maximum)
• Hadronic calorimeter
• Shielding
• Muon chambers and scintillator

Cutaway view of the CDF II detector
Protons go in here
Interaction point
8
CDF Tracking

• Silicon strip tracking (Solid state)
• Charged particle creates electron-hole pairs,
  apply HV to collect charge
• Good resolution, radiation tolerance (close to
  IP)
• R-phi, stereo, and Z type layers (7-8 layers,
  some double-sided)

• Drift chamber tracking
• Metal wires in closed chamber full of gas
• Charged particle ionizes gas
• Alternating R-phi and stereo layers (4 of each)
• Algorithms reconstruct tracks from hits
• Group wires/strips with signal above threshold
  into clusters hits
• Momentum from curvature in 1.4 T field
• Use track quality, number of tracks

9
CDF Tracking
Apparently this is also a CDF tracker The
Grumman S-2T Turbine Tracker
10
CDF Calorimetry

• High-mass particle interacts with matter, stops
  ( transfers all its momentum)
• CDF alternating layers of scintillator, heavy
  material
• Shower develops in heavy material
• Collect photons from scintillator
• Electromagnetic calorimeter stops
  electrons/photons first (ideally)
• Lead-scintillator
• Hadronic calorimeter stops hadrons
• Iron-scintillator
• Designed to measure particle energy
• Very coarse granularity in eta, phi
• Projective geometry
• Towers point back at interaction point

scintillator
iron
scintillator
lead
shower maximum detector
one tower
central
forward
interaction point
11
CDF Small Tracking

• Shower maximum detectors electrons
• Small, shallow tracking at depth where EM shower
  peaks
• Wire chamber in central, scintillator strips in
  plug
• Better spatial resolution than calorimetry
• Run clustering algorithms, like central tracker
• h, j location of shower centroid
• Shower profile (collimated/ spread out?)
• Muon chambers
• Shallow wire tracker outside of calorimetry,
  shielding
• Short tracks, called stubs, indicate muons

12
Kinematic vs. ID selection

• Kinematic whats usable
• ET or pT cuts
• Fiducial (in volume where detector can measure
  reliably)
• Fraction of signal events passing these cuts
  determined by physics process (Acceptance)
• Identification (ID) cuts assume you have the
  above, aim is to reject backgrounds
• Probability for real lepton to pass is
  Efficiency
• Probability for something else to pass is the
  Fake Rate

13
Electron Identification

• Jet rejection
• Calorimeter Isolation Ratio of energy in a cone
  around the electron to the electron energy. Jets
  are wider objects
• Track Isolation Require electron track to be
  much higher pT than any other track around it

• Had/Em Ratio of energy in the hadronic
  calorimeter to energy in EM calorimeter. Jets
  typically deposit most of their energy in the
  hadronic calorimeter

14
Electron Identification

• Jet rejection (continued)
• Shower profile should be narrow (related to
  isolation)
• Track-shower max matching track should point at
  cluster centroid (particularly good for rejecting
  sneaky pp0 s
• Most of these (especially isolation-type
  variables, track-centroid matching) are also very
  good at rejecting real electrons from
  heavy-flavor decay, but not as powerful against
  that

15
Electron Identification
e

• Photons
• Correct EM signature
• Requiring a track gets rid of prompt photons
• Conversions Algorithm looks for opposite-sign
  tracks originating from the same, displaced point
• Muons
• Rare, but it happens
• Reject some with track-centroid matching
• Get rid of the rest by requiring that the
  electron not be pointing right at missing energy

e-
g
An exaggerated conversion
m
radiated photon showers in EM detector, just
like an electron
g
m
muon track points right at the cluster
16
Muon Identification

• Jet rejection similar to electrons
• Calorimeter, Track Isolation
• MIP signature Require there to be almost
  nothing (few GeV) in the calorimeters
• Muon stub Very few hadronic particles make it
  out of the calorimetry
• Impact parameter, track quality
• Kaon decays-in-flight have two low-pT tracks
  strung together to make one lousy high-pT track
• Smaller fake rates, still worry about real muons
  from heavy flavor decays

17
Muon Identification

• Cosmic rays
• Impact parameter unlikely to have crossed
  detector at exactly the interaction point
• Cosmic tagging algorithm looks at track timing
  information consistent with beam crossing?

18
Use in Analysis

• Ideally, apply all selection criteria to a Monte
  Carlo of the physics process of interest
• In practice, detector modeling is rarely perfect
• Trust MC for your acceptance, but not efficiency

• Quantify data/MC discrepancy by measuring the
  efficiency in both
• Pure sample of leptons? At CDF, use Z bosons
  (mass window opposite charge), background 2 or
  less)
• Compare to Z MC
• Take scale factor ratio of e (data)/ e (MC)
  eff, multiply MC Ae by this correction factor

19
Moving to CMS _at_ the LHC
20
Moving to CMS _at_ the LHC
? A physicists-eye view
21
CMS Tracking
All silicon, all the time
Almost 10 M readout channels

• Pixels lower occupancy close to interaction
  point
• Strips are faster to readout and easier to track
  with (less combinations)
• Endcap structures as well as radial
• Stronger field (4 T) will provide better momentum
  resolution for higher pT particles

22
CMS EM Calorimetry

• Instead of alternating dense material and
  scintillator, a very dense scintillator
• Crystals of lead tungstate (PbWO4, 98 metal by
  mass but completely transparent)
• Finer h-j resolution
• Crystals are 1 Moliere radius ( typical width
  of EM shower 22 mm) wide
• No shower max detector
• Instead, pre-radiator
• Two layers of lead (to start shower) followed by
  silicon layers (to measure position)

one crystal
23
CMS Hadron Calorimetry

• Sampling calorimeters, like CDF
• Central copper-scintillator sandwich
• Forward steel-quartz sandwich
• Robust for higher radiation evironment uses
  Cerenkov light instead of scintillation.
• Spatial resolution (central) 0.87 x 0.87 in h-j
  (compare to CDF at 0.11 x 0.26)
• All the calorimetry is inside the magnet
• Less material in front of calorimetry (except
  the tracker)
• Additional scintillator outside of magnet to get
  up to 11 absorption lengths

24
CMS muon detectors

• 4 muon stations interleaved with iron absorber/
  flux return
• Each station is layers of wire chambers
• Right outside the solenoid

• Enough lever arm for independent tracking

25
Signal, Background at 14 TeV

• From pp at 2 TeV to pp at 14 TeV
• More energetic leptons
• More bremstrahlung
• Adds tracks, confuses calorimeter information
• A use for the better tracking
• More noise in the event from underlying, softer
  interactions
• Need to re-think isolation variables?

26
Electron ID at CMS

• Much finer segmentation in calorimetry
• More detailed isolation and shower shape
  variables
• Instead of just an isolation ratio, look at shape
  of energy distrubution (electrons should be
  confined to one crystal)
• Important as events are very busy and occupancy
  is high
• With preradiator, may be able to discriminate
  against pp0
• look for indications of two particles, better
  resolution for track/cluster mismatch
• More material in tracker
• Conversions will be more of a problem, but
  perhaps it will be easier to catch them?

27
Muon ID at CMS

• All silicon tracking
• More stringent track quality requirements
• Forward muons more practical (coverage)
• Pointing at vertex in Z as well as j
• d0 resolution?
• Must understand tracking to do muon ID well
• Matching silicon track to muon chamber tracks
• More material, more energetic muons
• Challenge muons may radiate
• Too much acceptance loss from requiring MIP
  signature in ECAL?
• Use ECAL, preradiatior, accept muons that appear
  to be paired with a photon
• Still require MIP in HCAL

28
Summary

• Electrons and muons can be identified with good
  efficiency/ high purity
• Use to identify interesting physics
• Use all parts of detector to discriminate against
  backgrounds
• CMS brings new challenges but new tools to use as
  well