Inclusive Jet Production using the k T Algorithm at CDF

Régis LefèvreAnalysis done with Mario Martínez and Olga Norniella

IFAE Barcelona

Inclusive Jet Production using the kT Algorithm at CDF

IMFP2006XXXIV International Meeting

on Fundamental Physics

April 2nd-7th 2006, Madrid, Spain

HRPN-CT-2002-00292

E.U. ResearchTraining Network

Probe for New Physics

Régis Lefèvre, IMFP 2006, April 6th 2006, El Escorial, Madrid , Spain 2

• Proton-antiproton collisions

• s = 1.96 TeV

• 36 bunches: crossing time = 396 ns

• Peak luminosity ~ 1.2 1032 cm-2 s-1

The Tevatron in Run II

• Collecting ~ 20 pb-1 / week

• About 1.6 fb-1 delivered


CDF • Highly upgraded for Run II– New silicon tracking

– New drift chamber

– Upgraded muon chambers

– New plug calorimeters

– New TOF

• Data taking efficiency ~ 85 %

• About 1.3 fb-1 on tape – New results based on 1 fb-1


Motivations• Legacy from Run I

– Great interest on apparent excess at high ET

– SM explanation•Gluon PDF increased at high x•New PDFs from global fit include CDF and

D0 jet data from Run I (CTEQ6, MRST2001)

• Stringent test of pQCD– Over ~ 8 order of magnitudes

• Tail sensitive to New Physics– Probing distances ~ 10-19 m

• PDFs at high Q2 & high x

• Production enhanced at high pT

thanks to new s

Run I CDF Inclusive Jet Data(Statistical Errors Only)JetClu RCONE=0.7 0.1<||<0.7R=F=ET /2 RSEP=1.3

CTEQ4M PDFsCTEQ4HJ PDFs


Cone Jet Algorithms and pQCD

• Infrared and Collinear Safety– Fixed order pQCD contains not fully cancelled

infrared divergences• Inclusive jet cross section affected at NNLO

– Run I Cone Algorithm: JetClu• Neither infrared nor collinear safe

– Run II Cone Algorithm: Midpoint• Uses midpoints between pairs of proto-jets

as additional seeds

Infrared and collinear safety restored

• Merging/Splitting– NLO pQCD uses larger cone radius R’ = R RSEP

to emulate experimental merging/splitting• Arbitrary parameter RSEP: prescription RSEP = 1.3

(based on parton level approximate arguments)

below threshold(no jets)

above threshold(1 jet)


The kT Algorithm

• Inclusive kT algorithm– Merging pairs of nearby particles

in order of increasing relative pT

•

•

– D parameter controls merging termination and characterizes size of resulting jets

• pT classification inspired by pQCD gluon emissions

– Infrared and Collinear safeto all orders in pQCD

– No merging/splitting•No RSEP issue comparing to pQCD

2

222

D

ÄR)p,(pmind jT,iT,ij =

2iT,ii pd =

– Successfully used at LEP and HERA

– Relatively new in hadron-hadron collider• More difficult environment Underlying Event

Multiple Interactions per crossing (MI)

Results from ZEUS / D0 Run ID0 Run I

Disagreement at low pT

Suggests Underlying Event not properly accounted for


Framework / Related Topics

• Look first at central jets: 0.1 < |y| < 0.7– Where calorimeter simulation is best

– Use D = 0.5, 0.7 and 1.0•To make sure that Underlying Event and MI

contributions are well under control

• Data fully corrected to particle level– Requires a good simulation of the detector

– Monte-Carlo generator should be able to reproduce the Jet Shapes

•Jet fragmentation and parton cascades

• NLO pQCD corrected to hadron level– Parton level pQCD calculation corrected

for the Underlying Event and Hadronization •Requires a Monte-Carlo generator able to

reproduce the Underlying Event


Underlying Event• Everything but the hard scattering process

– Initial state soft radiations

– Beam-beam remnants

– Multiple Parton Interactions (MPI)

• Studied in the transverse region– Leading jet sample

– Back-to-back sample


Energy Flow Inside Jets

(1-)

∑=jets T

T

jets R)(0,P

r)(0,P

N

1Ø(r)

Jet shapes governed by multi-gluon emission from primary parton

– Test of parton shower models

– Sensitive to underlying event structure

– Sensitive to quark and gluon mixture in the final state Phys. Rev. D

71 112002 (2005)

37 < pT < 380 GeV/c


Calorimeter Response to Jets• (First set electromagnetic scale using Z e+e-)• Absolute jet energy scale

– E/p of isolated tracks used to tune the showering simulation (G-Flash)•Residual discrepancies taken as systematic errors

– Induced uncertainty on jet energy scale between 1 and 3%•Reasonable simulation of the pT spectrum of the particles within a jet by PYTHIA

and HERWIG fragmentation models (fundamental as non-compensated calorimeters) – Induced difference on jet energy scale < 1%

– Photon-jet balance•Data and Simulation agree

at 1% to 2% level

• Non uniformity versus – Dijet balance

•Relative response known to 0.5% level

• Resolution– Bisector method

•Jet energy resolutions known within relative uncertainties of few %


Theoretical Predictions

• NLO pQCD: JETRAD

– Scale: R = F = max (PTJET) / 2

– PDFs: CTEQ6.1M package

– Main uncertainty comes from PDFs

•Gluon PDF at high x

• CHAD = parton-to-hadron correction factor– Accounts for non perturbative contributions

•Underlying Event (U.E.)

•Hadronization

– PYTHIA-Tune A used as nominal

•HERWIG used for uncertainty

(parton level no U.E.)

(hadron level with U.E.)CHAD =

PDF uncertainty

CHAD

D=0.7 and 0.1 < |y| < 0.7


Published Results

Phys. Rev. Lett. 96 122001 (2006)

D = 0.70.1 < |y| < 0.7


kT Jets vs. D (0.1 < |y| < 0.7)

D = 0.5 D = 1.0


Forward Jets

• Essentials to pin down PDFs vs. eventual New Physicsat higher Q2 in central region– DGLAP gives Q2 evolution

• Expend x range toward low x

High-x Low-x

“Rutherford type” parton backscattering


New Results

• D = 0.7

• 5 rapidity regions up to |y| = 2.1– |y| < 0.1

– 0.1 < |y| < 0.7

– 0.7 < |y| < 1.1

– 1.1 < |y| < 1.6

– 1.6 < |y| < 2.1

L ~ 1 fb-1


Data / Theory


Conclusion

• Good agreement with NLO

– Stringent test of pQCD over ~ 8 orders of magnitude

•pT reach extended by ~ 150 GeV/c with respect to Run I

• Careful treatment of non perturbative effects

– Underlying Event well under control

• To be used in future PDF global fits in orderto better constrain the gluon PDF at high x

– Forward jets essentials

• Prospect

– cos * vs. dijet invariant mass

•Limit on contact interactions

Backup Slides


kT Algorithm Step-by-Step

jet

jet

jet

jet jet

jet jet

jet

2

222

D

ÄR)p,(pmind jT,iT,ij =

2iT,ii pd =

Longitudinally invariant kT algorithm (Ellis-Soper inclusive mode)


PT//

PTPERP

PERP axis

// axis (bisector)

Transverse Plan

PTJET2

PTJET1

Bisector Method

// ISR

PERP ISR Detector Resolution

• Assuming ISR democratic in D = (2

PERP - 2//) / 2 Detector Resolution

0.1 < |y| < 0.7

Inclusive Jet Production using the k T Algorithm at CDF

Documents