1

DDDiffractive Studies at the Tevatron

Gregory R. SnowUniversity of Nebraska

(For the D0 and CDF Collaborations)

• Introduction

• Hard Color Singlet Exchange (CSE) (forward jet - gap - forward jet)

• Hard Single Diffraction (forward gap - opposite dijets or W)

• Double Pomeron Exchange Signatures (forward gap - jets - forward gap)

• Improvements for Tevatron Run II

• Conclusions

ICHEP98Vancouver, BC, Canada

July 23 - 29, 1998

Workshop on Forward Physics and

Luminosity Determination at the LHC

Helsinki, Finland

November 2, 2000

International Europhysics ConferenceTampere, FinlandJuly 15 - 21, 1999

CMS Collaboration Meeting

March 8, 2001

2

DDD0 Run I Diffraction ResultsGregory R. Snow

University of Nebraska(For the D0 Collaboration)

• Introduction

• Hard Color Singlet Exchange (CSE) (forward jet - gap - forward jet)

• Hard Single Diffraction (forward gap - opposite dijets or W)

• Double Pomeron Exchange Signatures (forward gap - jets - forward gap)

• Diffractive W and Z Production (W/Z plus gap)

• Conclusions

Small-x WorkshopFermilab

September 17 - 20, 2003

New

New-ish

Nice result,need to publish

3

DDSo why am I here?

• Dino was my Ph.D. thesis advisor at RockU, somehow diffraction gets in your blood

• Oversaw some of this work as D0 Run I QCD co-convener 1998-2001 (with Schellman, Brandt, Elvira) and Editorial Board member/chair

• I have great appreciation for Andrew Brandt’s invention and leadership of the D0 diffraction effort

• Expanded D0’s Run I physics menu• Several unique Ph.D. theses and publications resulted• Run I pioneering work has led to Run II Forward Proton Detector – more comprehensive studies to come (see upcoming talks)

• I said “yes” to this talk

4

DDConclusion before the talk

So you’re at a cocktail party, and someone asks you:

“What fraction of (insert your favorite QCD dijet, W/Z, direct photon, heavy flavor process here) events areaccompanied by a striking rapidity gap in the calorimeter?”

Your answer: “About ONE PERCENT. Come here often?”

5

DDQCD Physics from DØ QCD publications represent 25% of the 125 papers published by D from its Run 1 data sample (100 pb-1)Jet physics“Transverse Energy Distributions within Jets at 1.8 TeV”“Studies of Topological Distributions of Inclusive Three- and Four-Jet Events at 1800 GeV”“Measurement of Dijet Angular Distributions and Search for Quark Compositeness”“Determination of the Absolute Jet Energy Scale in the D Calorimeters” “The Dijet Mass Spectrum and a Search for Quark Compositeness at 1.8 TeV”“The Inclusive Jet Cross Section in Collisions at 1.8 TeV”“Limits on Quark Compositeness from High Energy Jets in Collisions at 1800 GeV”“The ratio of Jet Cross Sections at 630 GeV and 1800 GeV”“Ratios of Multijet Cross Sections at 1800 GeV”“High-pT Jets at 630 and 1800 GeV”

Direct photon physics“The Isolated Photon Cross Section in the Central and Forward Rapidity Region at 1.8 TeV”“The Isolated Photon Cross Section at 1.8 TeV”QCD with W and Z“A Study of the Strong Coupling Constant Using W + Jets Processes”“Measurement of the inclusive differential cross section for Z bosons as a function of transverse momentum produced at 1.8 TeV”“Evidence of Color Coherence Effects in W+Jets Events at 1.8 TeV”“Differential production cross section of Z bosons as a function of transverse momentum at 1.8 TeV”“Differential Cross Section for W Boson Production as a Function of Transverse Momentum in Collisions at 1.8 TeVRapidity gaps, hard diffraction, BFKL dynamics“Rapidity Gaps between Jets in Collisions at 1.8 TeV”“The Azimuthal Decorrelation of Jets Widely Separated in Rapidity”“Jet Production via Strongly-Interacting Color-Singlet Exchange”“Color Coherent Radiation in Multijet Events from Collisions at 1.8 TeV”“Probing Hard Color-Singlet Exchange at 630 Gev and 1800 GeV”“Hard Single Diffraction in Collisions at 630 and 1800 GeV”“Probing BFKL Dynamics in Dijet Cross Section at Large Rapidity Intervals at 1800 and 630 GeV”“Observation of Diffractively Produced W and Z Bosons in Collisions at 1800 GeV”

6

DDD0 Hard Diffraction StudiesUnderstand the Pomeron via ……

Or W/Z

7

DDDØ Detector (Run I)

EM Calorimeter

L0 Detector

beam

(nL0 = # tiles in L0 detector with signal2.3 < || < 4.3)

Central Drift Chamber

(ntrk = # charged tracks with || < 1.0)

End Calorimeter

Central Calorimeter

(ncal = # cal towers with energy above threshold)

Hadronic Calorimeter

Central Gaps

EM Calorimeter

ET > 200 MeV

|| < 1.0

Forward Gaps

EM Calorimeter

E > 150 MeV

2.0 < || < 4.1

Had. Calorimeter

E > 500 MeV3.2 < || < 5.2

Calorimeter tower thresholdsused in rapidity gap analyses

Three different rapgap tags = 0.1 0.1

8

DDCentral Gaps between Dijets

• Two forward jets with | | > 1.9, > 4.0

and rapidity gap in | | < 1.0

• Signature for color singlet exchange

• Study gap fraction fS as function of

• Center of mass energy (1800, 630)

• Jet ET

• • Compare with Monte Carlo color singlet models

• Low energy run (630 GeV) very illuminating

No tracks orcal energy

jet

jet

| | < 1.0

9

DD

Nev

ents

ncal

DØ Central Gaps between Dijets

• 2-D multiplicity (n2-D multiplicity (ncalcal vs. n vs. ntrktrk) between two leading) between two leading

jets (Ejets (ETT > 12) for (a) 1800 GeV and (b) 630 GeV > 12) for (a) 1800 GeV and (b) 630 GeV

• Redundant detectors important to extractRedundant detectors important to extract rapidity gap signalrapidity gap signal

• Dijet events with Dijet events with one interactionone interaction per bunch crossing per bunch crossing selected using multiple interaction flagselected using multiple interaction flag

Negative binomial fit to “QCD” multiplicity

fS = (Ndata- Nfit)/Ntotal

Rapidity gap fraction

10

DDFraction of events with central gapfS

630 = 1.85 0.09 0.37 %fS

1800 = 0.54 0.06 0.16 % (stat) (sys)Sys. error dominated by backgroundfit uncertaintiesRatio (630/1800)fS

630/fS1800 = 3.4 1.2

DØ Central Gaps between Dijets

)1800(

)630()Model()Data( 630

18006301800 S

SRR

Gap fraction vs. second-highest jet ET

Example of comparison to differentcolor singlet models

s = 1800 GeV

Data consistent with a soft color rearrangement model preferring initial quark states, inconsistent with two-gluon, photon, or U(1) models

In the soft-color scenario, one canextract a ratio of gap survivalprobabilities

{

1.5 0.1

8.02.2)1800(

)630(

S

S

11

DDHard Single Diffraction

-4.0 -1.6 3.0 5.2

Measure minimum multiplicity here

-5.2 -3.0 -1. 1. 3.0 5.2

Two topologies

Measure multiplicity here

Central jetsForward jets

• Gap fractions (central and forward) at s = 630 and 1800 GeV

• Single diffractive distributions

• Comparisons with Monte Carlo to investigate Pomeron structure

• Physics Letters B531, 52 (2002)

Jet ET’s > 12, 15 GeV

12

DD

s = 1800 GeV

s = 630 GeV

Forward jets

Forward jets

Central jets

Central jets

L0 scintillatortiles used now

-4.0 -1.6 -1.0 1.0 3.0 5.2

orMeasure Multiplicity here

2-D Multiplicitydistributions

13

DD1800 Forward Jets

• Solid lines show show HSD candidate events• Dashed lines show non-diffractive events

Event Characteristics • Fewer jets in diffractive events• Jets are narrower and more back-to-back (Diffractive events have less overall radiation)• Gap fraction has little dependence on average jet ET

Hard Single Diffraction

14

DDHard Single Diffraction

SignalData

Background

Signal extraction• Forward 1800 GeV example• Signal: 2D falling exponential• Background: 4 parameter polynomial surface

Lessons

• Forward Jets Gap Fraction > Central Jets Gap Fraction

• 630 GeV Gap Fraction > 1800 GeV Gap Fraction

Measured Gap Data Set Fraction1800 Forward Jets (0.65 0.04)%1800 Central Jets (0.22 0.05)%630 Forward Jets (1.19 0.08)%630 Central Jets (0.90 0.06)%

Data Sample Ratio 630/1800 Forward Jets 1.8 0.2630/1800 Central Jets 4.1 1.01800 Fwd/Cent Jets 3.0 0.7630 Fwd/Cent Jets 1.3 0.1

Gap Fraction # diffractive Dijet Events / # All Dijets

Publication compares data with POMPYT M.C. using different quark and gluon structures.This data described well by a Pomeron composed dominantly of quarks, or a reduced fluxfactor convoluted with a gluonic Pomeron with both soft and hard components.

15

DD distributions using (0,0) bin

0.2 for s = 630 GeV

=p p

i

yT

s

eE i

i

s = 1800 GeV forward

central

s = 630 GeV forward

central

Hard Single Diffraction

Prescription based oncalorimeter energy deposition

used to extract distributions,fractional energy loss ofdiffracted beam particle

Shaded bands indicatevariance due to calorimeterenergy scale uncertainties

16

DDDouble Gaps at 1800 GeV|Jet | < 1.0, ET > 15 GeV

Demand gap on one side, Demand gap on one side, and measure multiplicity and measure multiplicity on opposite sideon opposite side

Gap Region 2.5<||<5.2

17

DDDouble Gaps at 630 GeV|Jet | < 1.0, ET > 12 GeV

Demand gap on one side, Demand gap on one side, and measure multiplicity and measure multiplicity on opposite sideon opposite side

Gap Region 2.5<||<5.2

18

DD

Publication (hep-ex/0308032) accepted by Phys. Lett. B

Diffractively Produced W and Z

Electron from W decay, with missing ET

May expect jets accompanying W or Z

Rapidity gap

Process probes quarkcontent of Pomeron

W eZ e+e-

consideredand

require singleinteraction to

preserve possiblerapidity gaps

(reduces availablestats considerably)

19

DD

Peaks in (0,0) bins indicate diffractive W

L0nL0 ncal

-1.1 0 1.1 3.0 5.2

Minimum side

Plot multiplicity in 3<||<5.2 (minimum side)

Diffractively Produced W’s

-2.5 -1.1 0 1.1 3.0 5.2

Minimum side

Central and forward electrons considered

L0ncal

nL0

68 of 8724 in (0,0)

23 of 3898 in (0,0)

20

DD

Peak in (0,0) bin indicates diffractive Z

-2.5 -1.1 0 1.1 3.0 5.2

Minimum side

ncalncal

L0nL0

Diffractively Produced Z’s

ncal

Plot multiplicity in 3<||<5.2 (minimum side)

9 of 811 in (0,0)

21

DD

MT=70.4

ET=36.9

ET=35.2

Standard W Events Diffractive W Candidates

ET=35.1

ET=37.1

MT=72.5

Compare diffractive W characteristics to all W’s

Electron ET

Missing ET

Transverse mass

Good agreementgiven lowerdiffractive statistics

22

DDFraction of diffractively produced W/Z

Sample Diffractive Probability Background All Fluctuates to Data Central W (1.08 + 0.19 - 0.17)% 1 x 10-14 7.7Forward W (0.64 + 0.18 - 0.16)% 6 x 10-8 5.3All W (0.89 + 0.19 – 0.17)% 3 x 10-14 7.5All Z (1.44 + 0.61 - 0.52)% 5 x 10-6 4.4

• Diffractive W/Z signals extracted from fits to the 2-D multiplicity distributions, similar to hard single diffraction dijet analysis

• Small correction to fitted signal from residual contamination from multiple interaction events NOT rejected by single interaction requirement (based on # vertices, L0 timing)

• Corrections due to jets misidentified as electrons and Z’s which fake W’s very small

{Opposite

trend comparedto hard diffractive

dijet case

23

DDW+Jet Rates

Jet ET Data Quark Hard Gluon

>8GeV (10 ± 3)% 14-20% 89 %>15GeV (9 ± 3)% 4-9 % 53 %>25GeV (8 ± 3)% 1-3 % 25 %

It is instructive to look at W+Jet rates for rapidity gap events compared to POMPYT Monte Carlo, since we expect

a high fraction of jet events if the Pomeron isdominated by the hard gluon NLO process.

The W+Jet rates are consistent with a quark dominatedpomeron and inconsistent with a hard gluon

dominated one.

24

DDRD = (WD ) / ( ZD ) = R*(WD/W)/ (ZD/Z)

where WD/W and ZD/Z are the measured gap fractions from this analysis

and

R=(W)/ (Z) = 10.43 ± 0.15(stat) ± 0.20(sys) ± 0.10(NLO)B. Abbott et al. (D0 Collaboration), Phys. Rev D 61, 072001 (2000)

Substituting in these values gives RD = 6.45 + 3.06 - 2.64

This value of RD is somewhat lower than, but consistent with, the non-diffractive ratio.

W/Z Cross Section RatioW/Z Cross Section Ratio

25

DDCalculate = p/p for W boson events using calorimeter energy deposition prescription:

Diffractive W Boson Diffractive W Boson

• Sum over all particles in event: those with largest ET and closest to gap given highest weight in sum (particles lost down beam pipe at – do not contribute

• Use only events with rapidity gap {(0,0) bin} to minimize non-diffractive background

• Correction factor 1.5 ± 0.3 derived from Monte Carlo used to calculate

=p p

i

yT

s

eE i

i

Most events < 10%, mean is 5%

26

DDRap gap fractions for different processes

0

0.5

1

1.5

2

2.5

0 1 2 3 4 5 6 7 8 9 10 11

Process Number

Rap

gap

frac

tion

(%)

Central gaps, opposite side dijets

Hard single diffraction, dijets

W boson

Z boson

OK, so it’s right – about 1%

1.0%

Central1800 GeV

Forward630 GeV Hard Single Diffraction

Central gap1800 GeV

Central gap630 GeV

Forward1800 GeV

Central630 GeV

All W

Forward W

Central W

All Z

Tomorrow, I’ll have a few words to say about augmenting this plot at LHC (CMS) energies