Multiplicity Structure in DIS and DDIS at HERA

1
Multiplicity Structure in DIS and DDIS at HERA

• Eddi A. De Wolf
• University of Antwerpen, Belgium
• On behalf of H1 collaboration

• Outline
• Charged particle multiplicity distributions
• Kinematic dependences of ltngt in DDIS
• Comparison DIS/DDIS
• Summary

2
Motivation

• H1 analysis on 94 DDIS data
• Dependence of ltngt on MX
• DDIS W,x,Q2, ß, xIP ,t, MX
• Which ones
• are relevant for multiplicity?
• H1 analysis on 2000 DDIS data
• Large statistics allows more differential study
• W, Q2, ? dependences at fixed MX
• Compare DIS and DDIS

3
Charged particle multiplicity of the hadronic
final state

• The multiplicity distribution P(n)
• Independent emission of single particles Poisson
  distribution
• Deviations from Poisson reveal correlations and
  dynamics
• Mean multiplicity ltngt of charged particles
• Particle density in Rapidity Space
• Koba-Nielsen-Olesen (KNO) scaling ?(z)
• energy scaling of the multiplicity distribution
• ?(z) ltngt Pn vs z n/ltngt

4
Diffraction ep -gt eXY
10 of DIS events have a rapidity gap
MX inv. mass of X
5
Models for diffraction
Combine QCD Regge theory resolved IP model

• Proton infinite momentum frame
• Colourless IP is built up of quarks/gluons
• Based on QCD and Regge factorization naïve,
  probably incorrect but works!
• Needs subleading IR component

6
Models for diffraction contd

• In proton rest frame ? splits into q-qbar
  dipole
• Model the dipole cross section, for example
• Saturation model Golec-Biernat and Wusthoff (GBW)

Colour dipole approach
7
Data selection DIS and DDIS

• 2000 nominal vertex data 46.65 pb-1
• Data corrected via Bayesian unfolding procedure
• DIS MC DJANGOH 1.3, proton pdf CTEQ5L
• DDIS MC RAPGAP resolved pomeron

• DIS selection
• Good reconstruction of scattered electron
• Kinematic cuts
• 0.05 lt yav lt 0.65
• 5 lt Q2 lt 100 GeV2
• 80 lt W lt 220 GeV

• DDIS selection
• Rapidity gap
• No activity in the forward detectors
• ?max lt 3.3
• Kinematic cuts
• 4 lt MX lt 36 GeV
• xIP lt 0.05

8
Track selection

• Primary vertex fitted tracks
• 15 lt ? lt 165o and pT gt 150 MeV
• Boost to hadronic ?p CMS

use tracks with ? gt 1
acceptance
resolution


9
Multiplicity results
Charged particles with ?gt1 in ?p CMS frame

• Multiplicity distribution and moments
• Q2 dependence of ltngt in DIS/DDIS at fixed W
• ? dependence of ltngt in DDIS at fixed MX
• W dependence of ltngt in DDIS at fixed MX
• Comparison of DIS and DDIS
• Density in Rapidity
• KNO scaling

10
Kinematics Bjorken Plot
DIS W2 Q2/x ymaxln (W/mp)
DDIS
11
Q2 dep. of ltngt in DIS/DDIS at fixed W

• ltngt vs Q2
• DIS/DDIS data (at fixed MX)
• No stat. signif. dependence on Q2
• Weak W- dependence
• Fit ltngt to
• ltngt a b log(Q2)

12
Q2 dep. of dn/d(y-ymax) in DIS at fixed W

• (1/N)dn/d(y-ymax)
• ymax ln(W/m?)
• Weak Q2 dependence
• QCD scaling violations of fragmentation function

DIS ltngt predominantly function of W, not Q2
13
Q2 dep. of dn/d(y-ymax) in DDIS at fixed W

• dn/d(y-ymax)
• ymax ln(W/m?)
• Weak Q2 dependence
• Weak W dependence
• Effect of the gap at lowest W!

DDIS ltngt predominantly function of MX, not Q2
14
? dependence of ltngt

• Intuitive picture expect no ? dependence at
  fixed MX (since no Q2 dependence observed)

15
? dependence of ltngt
low ltngt
triplet/anti-triplet
octet/octet
high ltngt

• dipole models
• relative fraction of q and g fragmentation
  depends strongly on ?
• expect ltngt depends on ?

16
? dep. of ltngt at fixed MX in DDIS

• MX kept fixed
• No obvious ? dependence

DDIS ltngt predominantly function of MX, not Q2 or
? separately
17
W dependence of ltngt?

• Changing W? changing the gap width
• Gap ln(1/xIP)
• Investigate ltngt dependence on xIP

18
W dep. of ltngt at fixed MX in DDIS
All Q2

• At fixed MX change W means change xIP
• Regge factorisation
• diffractive pdfs independent of xIP
• Breaking of Regge factorisation
• In resolved IP model
• pomeron reggeon

19
W dep. of ltngt at fixed MX in DDIS

• Fit ltngt
• ltngt a b log W2
• Regge factorisation breaking expected eg. in
  multiple scattering models
• Effects predicted to diminish with increasing Q2
  ( shorter interaction time) not seen here

Factorisation breaking not dependent on Q2
within errors
20
Particle Density in y DIS ? DDIS

• (1/N) dn/d(y-ymax)
• Central region
• particle density similar for DIS and high MX DDIS

DDIS DIS particle density not much different
although MXltlt W
21
Comparison DIS/DDIS KNO scaling

• Negative particles
• ?(z) ltngt Pn
• vs z n/ltngt
• Approximate KNO scaling for DIS and DDIS
• Shape of KNO distribution similar for DIS and
  DDIS
• Means Correlations very similar

22
Summary
Charged particle multiplicity

• studied for DIS and DDIS in ep at HERA
• ltngt in DIS main dependence only on W, not Q2 or x
  separately
• ltngt in DDIS main dependence on MX (and a bit on
  W), not on Q2 and ? separately
• Lack of ? dependence is surprising (q-qbar
  q-qbar g)
• DIS and DDIS density in rapidity similar at
  highest MX
• DIS and DDIS approximate KNO scaling same
  shape.

Kinematic dependences
Comparison DIS and DDIS