Studies on quark compositeness in ATLAS

1

• Studies on quark compositeness in ATLAS

2
Studies on quark compositeness at ATLAS

• What is quark compositeness and why do we care?
• What is ATLAS experiment and how can one measure
  the compositeness?
• Use of Combined testbeam 2004 data for assumption
  of ATLAS measurement results.

3
Are quarks composite?

• One can describe effects of quark compositeness
  by analogy of Fermi theory.
• Izoscalar left-left effective lagrangian
  describes contact four-fermion interaction
• ?1 interference sign, g24?, ? compositeness
  scale.
• We consider simple low energy approximation and
  study differences to standard QCD, for instance
• Inclusive jet production cross-section d?/dpT.
• Jet angular distribution.
• Are quarks composite?
• Tevatron up to ? ? 2 TeV no quark structure
  seen.
• LHC, ATLAS with total integral luminosity of
  300 fb-1 ability to spot even ? ? 40 TeV.

4

• ATLAS (A Toroidal LHC ApparatuS)
• A versatile particle detector for Large Hadron
  Collider at CERN
• pp 77 TeV, 40 MHz, 25 ns
• Sub-detectors and magnetic field system
• Inner detector (tracker)
• Central solenoid magnetic field 2T
• Liquid Argonne (LAr) em calorimeter and hadron
  end-caps
• Hadron calorimeter Tilecal
• Muon spectrometer
• Barrel toroid (8x 26 m long coil) a 2 x End-Cap
  toroid field 4T
• Coordinates
• Pseudorapidity

y
LHC
x
?
0
Z
-?
?
5
Invariant amplitudes in Pythia

• Physical data versus assumed simulated detector
  response

Particle generators (Pythia, Herwig,)
Detector response simulation (full GEANT4, fast
ATLFAST)
Data analysis (ROOT, Athena)

• What is done
• Calculation of cross-sections of all processes at
  parton tree level.
• Detailed comparison of theoretical angular
  distributions to Pythia results (Two references
  for M2 were in contradiction)
• Fast simulation of detector response (ATLFAST)
• Found bug in Pythia and acquired its correction

6
Invariant amplitudes check in Pythia

• Pythia 6.2 Pyt quark structure described by
  isoscalar model (Chi, Eich). Effective
  lagrangian see page 3.
• Two formulas of M2 referenced in Pythia are in
  contradiction
• qiqj-gtqiqj

Chia
Eich

• qiqi-gtqiqi
• Calculations by Tomá Davídek in agreement with
  Chia. But which formula is used in Pythia?

7
Invariant amplitudes check in Pythia

• Jet angular distributions for four different
  processes
• Pythia simulation with ?1 TeV
• Angular distribution fit according to Chia
  agrees in high precision.
• Eich formula does not describe such spectra.
• -gt Pythia includes correct M2 .

Chia
Eich
8
Fast simulation in Athena (6.0.3) Atlfast

• Particles generated by Pythia become input data
  for ATLAS detector simulation. Output of such a
  simulation are already reconstructed jets,
  isolated e, ?,?, ETmiss etc.
• Atlfast simulation
• Both negative and positive interferences
• ?5, 10, 20 a 40 TeV (106 events for each set)
• All quarks composite
• Cone jet algorithm R1.0
• Low luminosity
• With ? ? ? quark structure effect dissappears
  (QCD only).

9
Positive interference
! Spectra of ?20 TeV and ?40 TeV lie below QCD
spectrum.!
10
Negative interference
The same problem spectra of ?20 TeV and ?40
TeV lie below QCD spectrum!
11
Detailed bug analysis at Pythia level

• Problem spectra of ?20 TeV and ?40 TeV lie
  below QCD spectrum, opposite should be observed
  (ATL)!
• Analysis of all individual 4-fermion contact
  interactions revealed, that lower values of
  problematic spectra are caused by wrong
  performance of
• qi qi ?qi qi

Number of events of this process decreases with
growing ?, eventually for ?gt5 TeV not an event is
found in a 10 mil. ntuple. (In older Pythia
5.720 such a case havent appeared, but its
compositeness model did not included interference
terms. It was more simple.)
12
Pythia 6.2 correction

• After reporting this problem to Pythia author
  T.Sjostrand we received a remedy.
• Correct spectrum with Atlfast (R0.4, corrected
  Pythia, but old Atlfast Fortran version due to
  Athena obstacles) -gt
• In Pythia 6.223 this bug is already corrected.

13
Dijet angular distribution

• Studying dijet angular distribution is a useful
  tool because of its
• smaller sensitivity to Parton Distribution
  Functions
• smaller sensitivity to calorimeter non-linearity
• Atlfast cone jet algorithm with R0.4
• Every event contains several jets (mean 7)
• Two jets with the highest pT were chosen (?1,?2)

• For case of 2-gt2 parton scattering it is related
  to CMS scattering angle T as follows

• Several invariant mass bins were studied.

14
Dijet angular distribution
15
Dijet angular distribution

• ?20 and 40 TeV does not differ much from
  standard QCD. Higher statistics needed (here only
  106 events).

16
Conclusions and Plans

• Corrected Pythia must be implemented in next
  Athena release.
• Plans
• Fast simulation to obtain better statistics
• Convert to full simulation
• Inquire influence of different jet algorithms,
  PDFs, etc.
• Evaluate influence of calorimeter calibration
  and linearity
• Jet construction using physical data obtained at
  Combined testbeam (2004) and their comparison
  with MC further development of a method already
  used by Tomá Davídek for CTB data 1996
  (ATL-TILECAL-2000-010).

17
Combined Testbeam 2004
Motivation Construct the jet response using
signals from individual particles from the
combined beam test gt studies based on TB data
(instead of MC). Real subdetectors performance.
Investigation of many specific physics
processes (especially quark compositeness in our
case). Preparation for ATLAS data-taking and
analysis.
18
Constructing the response to jets

• Jets generated with Pythia
• Every particle then replaced by appropriate
  testbeam data event
• Available data sets will be e, ?, p, ? at 1 ?
  180 (300) GeV
• Particle substitution
• Muons replaced with ?, energy loss scaled
• using Bethe-Bloch formula.
• Mesons replaced with ?
• Baryons replaced with p (if ?/p separation
• in Cerenkov possible), otherwise ? used
• Electrons, gammas replaced with e
• Special treatment of K and very low energy
• particles
• Generated particle replaced with a TB event at
  closest available energy, signal scaled by
  Ebeam/Eparticle (except of muons)

19
Constructing the response to jets

• Building the calorimeter signal in a cone around
  jet axis
• Reconstruct the jet axis as a ET-weighted centre
  of the jet
• Signal evaluation in each calorimeter radial
  samplings separately
• Calculate ??cell(TB) , ??cell(TB) distance of
  the cell from the beam axis at TB for each TB
  readout cell and then assign real cell
  coordinates for the given particle direction
  (?part , ?part)
• ?cell ?part ??cell(TB)
• Sum-up all TB cells obeying
• (?cell ? ?jet)2 (?cell ? ?jet)2 ? R2
• Noise cuts applied to avoid accumulating noise
  when summing several single particle TB events
  into one jet.

20
Energy reconstruction

• Both calorimeters calibrated to the elmg. scale
  (ELAr , ETile)
• e/? - method for hadron energy reconstruction
  used in Tilecal
• cryostat correction (same as used in the
  combined run 1996 paper)
• Ecryo ccryo ? (? ? ELAr,3 ? (e/?)Tile ?
  ETile,1)0.5
• Reconstructed energy
• Erec ?(E,R) ? (? ? ELAr (e/ ?)Tile ? ETile
  Ecryo )
• function ? plays a role of weighted share of
  elmg. and hadronic particles, depends on total
  energy as well as ELAr . Since LAr is
  non-compensated, ?gt1.
• function ? corrects for several effects very
  low-energy particles caught by magnetic field
  within solenoid, finite size of the cone R. Both
  functions are parametrised, actual values of
  parameters obtained from fits (done in 1996).

21
Plans

• Jet response construction using experimental
  testbeam data provides a tool for studying
  detector performance and/or physics processes and
  was succesfully used for 1996 data.
• Forthcoming combined testbeam with final detector
  design geometry even more attractive, since
  (compared to 1996)
• Parts of all sub-detectors in beam-line (ID,
  calorimetry, muon system)
• Larger range of beam energies should be
  available (especially VLE)
• Beams at various pseudorapidities
• Plans
• Construct the jet response from CTB data in the
  similar way.
• Study the combined calorimetry resolution
  linearity with various energy reconstruction
  methods.
• Use the above described method to simulate
  compositeness effects.

22
References

• Atl ATLAS Detector and Physics Performance TDR,
  CERN/LHCC/99-15, p933.
• Pyt www.thep.lu.se/torbjorn/Pythia.html,
  Pythia 6.2 manual, p151.
• Eich E.Eichten, I.Hincliffe, K. Lane and C.
  Quigg, Rev. Mod. Phys 56 (1984), 579 Rev. Mod.
  Phys. 58 (1985), 1065.
• Chia P. Chiappetta and M. Perrottet, in Large
  Hadron Collider Workshop, eds. G. Jarlskog and
  D. Rein, CERN 90-10 (Geneva 1990), Vol. II, p806.