Soft QCD Phenomena in High-E T Jet Events at CDF

Andrey Korytov, University of Florida EPS 2003 July 17-23, 2003, Aachen

Soft QCD Phenomena in High-ET Jet Events at CDF

Abstracts covered in this talk• A356 Fragmentation Differences of Quark and Gluon Jets at CDF• A362 Measurement of Jet Shapes and Energy Flows in Dijet Production at the Tevatron• A353 Jet Evolution and the Underlying Event in Run 2 at CDF

Official Title Studies of Jet Shapes and Fragmentation Differences of Quark and Gluon Jets with the CDF Detector(note: Underlying Event fell out from the title)

Better TitleStudies of Soft QCD Phenomena in High-ET Jet Events with the CDF Detector

Andrey Korytov

(for the CDF Collaboration)


Soft QCD Phenomena in High-ET Jet Events

Anatomy of events with high ET jets: hard scattered partons final state radiation initial state radiation multi-parton interactions proton/antiproton remnants Note: separation between sub-processes is not cleandue to entangled color connections…

Soft QCD Phenomena: Jet fragmentation is largely driven by soft QCD So is the UE physics

Tools available: re-summed pQCD approximations

(analytic, but only for limited number of observables) Monte Carlo generators (generic, but have many tunable ad-hoc

knobs)

JETS

UNDERLYINGEVENT


Why bother?

Jet Fragmentation: Jet development physics:

• Parton shower stage—challenge for pQCD calculations at the very soft limit (kT~QCD)

• Hadronization stage—still remains a mystery

Many high-ET physics analyses depend on good understanding of jet properties

Underlying Event: UE physics is poorly understood:

• MC Generators implement UE differently and often with many (too many?) parameters

• Even when tuned to match the accessible data, MC predictions for LHC vary wildly

UE pollutes many analyses source of systematic errors

LHC MinBias

kT=1 GeV/c


Results presented in this talk

Jets: Momentum distribution of charged particles in jets vs.

NLLA Multiplicities of charged particles in g- and q-jets vs.

NLLA Energy flow in jets (jet shapes) vs. MC

Underlying Event: Energy flow away from jets vs. MC Charged particle multiplicity flow away from jets vs. MC Momentum distribution of away-from-jet charged particles

vs. MC


Jets: doing fragmentation analytically

Jet fragmentation: parton shower development:

MLLA’ Modified Leading Log Approximation

with one kT-cutoff parameter Qeff=Qcutoff=QCD

hadronization: LPHDHypothesis of Local Parton Hadron Duality with one parameter KLPHD=Nhadrons/Npartons

MLLA+LPHD:cannot describe all details…but all analytical…and works surprisingly well…

Momenta of charged particles in jets: Qeff = 230 40 MeV

KLPHD() = 0.56 0.10

CDFCharged particle momentum spectra (cone=0.47) and MLLA+LPHD fit

particle

jet

px

E


Jets: gluon vs quark jet differences

Ratio r = Nhadrons(gluon jet) / Nhadrons(quark jet) recent calculations (for partons): extensions of NLLA, r=1.4-1.7

(Q=10-100 GeV) lots of results from LEP, not all self-consistent: r = 1 to 1.5

jet-ID biased, model-dependent, few “unbiased/model-independent”


Jets: gluon vs quark jets at CDF

di-jet events (~60% gluon jets) and -jet events (~80% quark jets) di-jet or -jet center of mass frame: Ejet = ½Mjj or ½Mj

Nch multiplicity in cones with opening angle from ~0.3 to ~0.5 rad Energy scale Q=Ejet Results

• Ratio: r=1.60.2, almost energy scale independent• multiplicities in quark and gluon jets: see plot


Jets: Energy flow inside jets (vs MC)

Run IICDF preliminary

Jet shape: fractional energy flow (r) = ET(0:r) / ET(0:R), where R=1

In central region, do it with• Calorimeter towers (●)

• Charged tracks (○)

Either way—shapes are nearly identical

Herwig and Pythia practically coincide and agree with data


Run II CDF preliminary

Jets: Energy flow inside jets (vs MC)

In forward region, do it with• Calorimeter towers only

Data vs MC discrepancy:

the higher jet and

the smaller jet’s ET,

the larger the disagreement

Is it real or do we have simulation problems with the new plug calorimeter?• expand tracker analysis

to higher region…• any D0 data?

?


UE studies with charged tracks

Event sample: min-bias, jet events (central jets) Measure:

• Average Number of particles in direction transverse to leading jet: n=d2N / dd• PT spectrum of particles in direction transverse to leading jet: dn/dPT

• Average Energy summed over charged particles in transverse direction: d2ET/dd

Confront data and MC:• Identify importance of various MC knobs/parameters

Leading Jet

“Transverse” “Transverse”

“transverse” particlesas a probe of the underlying event

ET(jet)

Charged tracks:

• d2N/dd• d3N/dddPT

• d2ET/dd


UE: data vs. default Pythia and Herwig

Default Pythia and Herwig fail to reproduce data one way or another, e.g.:

Pythia 6.206 underestimates number of tracks in transverse direction…

Herwig 6.4 gives too soft spectrum for particles in transverse direction, especially in events with small ET jets (missing MPI now have been added)

"Transverse" Charged Particle Density: dN/dd

0.00

0.25

0.50

0.75

1.00

0 5 10 15 20 25 30 35 40 45 50

PT(charged jet#1) (GeV/c)

"Tra

ns

ve

rse

" C

ha

rge

d D

en

sit

y

CTEQ3L CTEQ4L CTEQ5L CDF Min-Bias CDF JET20

1.8 TeV ||<1.0 PT>0.5 GeV

Pythia 6.206 (default)MSTP(82)=1

PARP(81) = 1.9 GeV/c

CDF Datadata uncorrectedtheory corrected

Charged Particle Density

1.0E-06

1.0E-05

1.0E-04

1.0E-03

1.0E-02

1.0E-01

1.0E+00

1.0E+01

0 2 4 6 8 10 12 14

PT (GeV/c)

Ch

arg

ed D

ensi

ty d

N/d

d d

PT

(1/

GeV

/c)

||<1CDF Preliminary

Min-Bias Data, 1.8 TeV

HW "Soft" Min-Biasat 0.63, 1.8, and 14 TeV


UE: tune Pythia to match CDF data

PYTHIA 6.206 and CDF Tune A (CTEQ5L)Parameter Default Tune Description

PARP(67) 1.0 4.0 Scale factor that governs the amount of initial-state radiation.

MSTP(81) 1 1 Turns on multiple parton interactions (MPI).

MSTP(82) 1 4 Double Gaussian matter distribution.

PARP(82) 1.9 2.0 Cut-off for multiple parton interactions, PT0.

PARP(83) 0.5 0.5 Warm Core: 50% of matter in radius 0.4.

PARP(84) 0.2 0.4 Warm Core: 50% of matter in radius 0.4.

PARP(85) 0.33 0.9 Probability that the MPI produces two gluons with color connections to the "nearest neighbors".

PARP(86) 0.66 0.95Probability that the MPI produces two gluons either as described by PARP(85) or as a closed gluon loop. The remaining fraction consists of quark-antiquark pairs.

PARP(89) 1,000.0 1,800.0 Determines the reference energy E0.

PARP(90) 0.16 0.25Determines the energy dependence of the cut-off PT0 as follows

PT0(Ecm) = PT0(Ecm/E0)PARP(90).

Pythia: CDF Tune A vs. Default 6.206• Enhanced initial state radiation• Smoothed out probability of Multi-Parton Interactions (vs. impact)• MPIs are more likely to produce gluons than quark-antiquark pairs

and MPI gluons are more likely to have color connection to p-pbar remnants

• …


UE: Pythia Tune A describes Data

"Transverse" Charged Particle Density: dN/dd

0.00

0.25

0.50

0.75

1.00

1.25

0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150


"Tra

nsvers

e"

Ch

arg

ed

Den

sit

y CDF Run 1 Min-Bias

CDF Run 1 JET20

CDF Run 2

PY Tune A

||<1.0 PT>0.5 GeV/c

CDF Preliminarydata uncorrectedtheory corrected

"Transverse" Charged PTsum Density: dPTsum/dd

0.00

0.25

0.50

0.75

1.00

1.25

0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150


"Tra

nsv

erse

" P

Tsu

m D

ensi

ty (

GeV

)

CDF Run 1 Min-Bias

CDF Run 1 JET20

CDF Run 2

PY Tune A

CDF Preliminarydata uncorrectedtheory corrected

||<1.0 PT>0.5 GeV/c

"Transverse" Charged Particle Density

1.0E-05

1.0E-04

1.0E-03

1.0E-02

1.0E-01

1.0E+00

0 2 4 6 8 10 12 14 16 18 20

PT(charged) (GeV/c)

Ch

arg

ed D

ens

ity

d3 N

/d d

d

PT (

1/G

eV

/c)

CDF Preliminarydata uncorrected

theory corrected

PYTHIA Tune A 1.96 TeV

30 < PT(chgjet#1) < 70 GeV/c

70 < PT(chgjet#1) < 95 GeV/c


Summary

Jet fragmentation

Momenta of charged particles in jets are well described by NLLA pQCD:• MLLA kT-cutoff Qeff=230 40 MeV• LPHD Nhadrons/Npartons = KKLPHD() = 0.56 0.10

Multiplicities in gluon and quark jets and their ratio r = 1.6 02 agree with extended NLLA pQCD and recent LEP data

Jet shapes mostly agree with PYTHIA and HERWIG, but more studies/tuning are needed for high- jets

Underlying event

Run II and Run I data agree, Run II analysis is being expanded

MC generators with default parameters do not quite work, but can be tuned to match data… insights into UE physics?


A few backup slides follow this page…


Tevatron upgrade: Run II vs. Run I

Original Plan: Run I Run II

CM energy: 1.8 1.96 TeVBx spacing: 3.5 0.4 s Max luminosity: 2x1031 50x1031 cm-2 s-1

Integrated luminosity: 0.1 15 fb-1 (before LHC turn-on)

As of May 2003:

Peak luminosity so far: 4.5x1031 cm-2 s-1

Total delivered (incl. commissioning): 0.24 fb-1

On tape (CDF, incl. commissioning): 0.18 fb-1 Good for physics (CDF) >0.13 fb-1

CDF current data taking efficiency ~90%

Long-term plan (by end of 2008):

Base goal: 6 fb-1

Stretch goal: 11 fb-1


CDF upgrade: what is new

Retained from CDF I: Solenoid Central Calorimeters Central Muon System

Brand new in CDF II: 5-layer 3D vertex Si

detector Intermediate Si layers Central Drift Tracker Plug Calorimeter Mini-plug calorimeter Time of Flight System Expanded Muon Coverage TRIGGER, now includes:

• displaced vertex

• track pT>1.5 GeV/c

Faster Front End Electronics

• 3d vertex coverage: <2• Tracking coverage: <2• Calorimeter coverage: <3.6• Mini-plug calorimeter: 3.6<<5.1• Muon coverage: <1.5


Jets: Gluon vs Quark jets at CDF

Momentum distributions of charged particles in gluon and quark jets Ratio reaches max and flattens for soft part of spectrum at ~1.80.2 Same pattern was observed at LEP

Documents

Soft QCD Phenomena in High-E T Jet Events at CDF