Charged Particle Multiplicity in DIS

1
Charged Particle Multiplicity in DIS

• QCD Group Preliminary Request

M. Rosin, D. Kçira, and A. Savin University of
Wisconsin L. Shcheglova Moscow State University,
Institute of Nuclear Physics May 13, 2004
2
Outline

• Introduction and motivation
• Data selection simulation
• Correction methods
• Bin by bin (modified)
• Matrix
• ltnchgt vs. Meff
• with MC predictions
• In bins of x
• In bins of x and Q2
• Systematic LEPTO vs. ARIADNE for correcting in
  Q2 and x bins
• Comparison to second analysis
• Summary and plan

3
Motivation

• Mean charged multiplicity, ltnchgt , vs. Q
  shows logarithmic dependence for both ee-
  and pp on the effective energy going into
  hadronization, and dependence is universal
• ep points measured in current region of
  Breit frame (x2), shows nice agreement
• For pp and ee- hadronization is universal
  when you go to proper scale vs for ee-
  reactions, effective energy going into hadron
  production, vq2 for pp.
• Expect that ep (measured in lab frame) and
  pp (and ee-) are universal, using a proper
  scale.

4
Motivation

• Mean charged multiplicity, ltnchgt , vs. Q
  shows logarithmic dependence for both ee-
  and pp on the effective energy going into
  hadronization, and dependence is universal
• ep points measured in current region of
  Breit frame (x2), shows nice agreement
• For pp and ee- hadronization is universal
  when you go to proper scale vs for ee-
  reactions, effective energy going into hadron
  production, vq2 for pp.
• Expect that ep (measured in lab frame) and
  pp (and ee-) are universal, using a proper
  scale.

5
Motivation for the use of Meff as energy scale

• Analogous to the pp study, want to measure the
  dependence of ltnchgt of on its total invariant
  mass. (Energy available for hadronization)
• For ep in lab frame, measure visible part of
  ltnchgt with visible part of energy available for
  hadronization should show same universality as
  total ltnchgt and total energy
• Use Meff as scale for comparing to pp (and ee-)

Whad
Meff
Lab Frame
Meff HFS measured in the detector where the
tracking efficiency is maximized
6
1996-97 Data sample

• Event Selection
• Scattered positron found with E gt 12 GeV
• A reconstructed vertex with Zvtx lt 50 cm
• scattered positron position cut radius gt 25cm
• 40 GeV lt E-pz lt 60 GeV
• Diffractive contribution excluded by requiring
  ?maxgt 3.2
• Track Selection
• Tracks associated with primary vertex
• ? lt 1.75
• pT gt 150 MeV
• Physics and Kinematic Requirement
• Q2 da gt 25 GeV2
• y el lt 0.95
• y JB gt 0.04
• 70 GeV lt W lt 225 GeV ( W2 (q p)2 )

724,958 events after all cuts (38.58 pb-1)
7
Event simulation

• Ariadne 97 6v2.4
• Matrix elements at LO pQCD O(?s)
• Parton showers CDM
• Hadronization String Model
• Proton PDFs CTEQ-4D

(Simulates both 96 and 97 data no changes in
detector)
Luminosity of MC 36.5 pb-1
8
Correction to hadron level bin by bin
Part one correct to hadron level using only
hadrons generated with pT gt 0.15 GeV
Part two correct for hadrons with lower pT,
using ratio of ltgengt with pT cut to ltgengt no pT
cut in each bin.
9
Correction to hadron level bin by bin

Correction for detector effects
Correction of hadrons of pT gt 0.15 to all
hadrons
10
Correction to hadron level matrix
The matrix relates the observed to the generated
distributions in each bin of Meff by
11
Comparison of correction methods
Better than 3 agreement in all Meff bins
12
Results with MC predictions
ZEUS

• Good agreement with 1995 prelim. points,
  with smaller statistical, systematic errors
• Observe a significant difference between ep,
  ee- and pp.
• In general data agrees with the MC
  predictions, LEPTO predicts even higher
  multiplicities
• Check for ep points does ltnchgt dependence
  on Meff change depending on scale?

13
ltnchgt vs. Meff in x bins
Matrix method

• if difference is due to quark and gluon
  distributions, then maybe also x dependence
• x range split into similar bins as in previous
  multiplicity paper.
• weak x dependence in mc, in data dependence is
  stronger

14
ltnchgt vs. Meff in x bins
Bin-by-bin method

• if difference is due to quark and gluon
  distributions, then maybe also x dependence
• x range split into similar bins as in previous
  multiplicity paper.
• data agree better with mc than for matrix
  method, overall weak x dependence

15
x and Q2 bins

• No Q2 dependence observed
• for matrixData described by ARIADNE, except in
  high x bins
• for bin-by-bin well decribed by ARIADNE
• Bin-by-bin method is more stable against mc,
  matrix shows some mc dependence
• Need to investigate why matrix gives different
  results in high x bins

150 -1200
Matrix method
50 - 150
0.01 - 0.1
25 -50
0.0006 - 0.0012
0.0012 - 0.0024
0.0024 - 0.01
16
x and Q2 bins

• No Q2 dependence observed
• for matrixData described by ARIADNE, except in
  high x bins
• for bin-by-bin well decribed by ARIADNE
• Bin-by-bin method is more stable against mc,
  matrix shows some mc dependence
• Need to investigate why matrix gives different
  results in high x bins

17
LEPTO vs. ARIADNE for corrections

• As a systematic, used LEPTO for correction
• bin-by-bin method more stable
• data agrees with ARIADNE even when corrected
  using LEPTO

18
LEPTO vs. ARIADNEfor corrections

• As a systematic, used LEPTO for correction
• bin-by-bin method more stable
• data agrees with ARIADNE even when corrected
  using LEPTO

19
Comparison to 2nd analysis in Q2 and x bins

• Agreement between 1st and 2nd analysis within 1
  for matrix for x and Q2 bins

20
Comparison to 2nd analysis

• For full kinematic range, agreement between 1st
  and 2nd analysis is less than 1 for both
  correction methods.

21
Summary

• Shown 3 preliminary plots
• Measured in x bins, weak x dependence not enough
  to compensate for difference between ep and pp.
• Observed difference in dependence of ltnchgt on
  Meff between ep and pp (and ee-) is real
• Plan
• Look at Breit frame for consistency check
• Study diffractive events combine ARIADNE RAPGAP

22
Preliminary plots
23
Preliminary plots
24
Preliminary plots