[Paper Review] Event shape analysis of deep inelastic scattering events with a large rapidity gap at HERA

Breitweg J, M. Derrick+422 more|arXiv (Cornell University)|Jan 1, 1997

Particle physics theoretical and experimental studiesPhysics and Astronomy54 references31 citations

TL;DR

This paper analyzes event shapes in deep inelastic scattering events with a large rapidity gap at HERA, focusing on the γ∗−pomeron rest frame. It finds that as the hadronic final state mass MX increases, events become more collimated and planar, requiring a significant gluon component in the pomeron to explain the data, which cannot be accounted for by models relying only on quark couplings or hard gluon bremsstrahlung.

ABSTRACT

A global event shape analysis of the multihadronic final states observed in neutral current deep inelastic scattering events with a large rapidity gap with respect to the proton direction is presented. The analysis is performed in the range $5 \leq Q^2 \leq 185\gev^2$ and $160 \leq W \leq 250\gev$, where $Q^2$ is the virtuality of the photon and $W$ is the virtual-photon proton centre of mass energy. Particular emphasis is placed on the dependence of the shape variables, measured in the $\gamma^*-$pomeron rest frame, on the mass of the hadronic final state, $M_X$. With increasing $M_X$ the multihadronic final state becomes more collimated and planar. The experimental results are compared with several models which attempt to describe diffractive events. The broadening effects exhibited by the data require in these models a significant gluon component of the pomeron.

Motivation & Objective

• To study the dependence of event shape variables on the mass MX of the hadronic final state in deep inelastic scattering with a large rapidity gap.
• To test whether existing Monte Carlo models can describe the observed event geometry and momentum flow in dijet-like final states.
• To determine the required partonic structure of the pomeron by comparing data with models that include either pointlike couplings (VBLY) or parton densities (RAPGAP).
• To assess the role of gluon radiation and pomeron structure in explaining the broadening and planarity of high-MX dijet events.
• To quantify the fraction of gluon-induced events needed to reproduce the data, especially in the context of non-perturbative pomeron structure.

Proposed method

• Event shape variables—sphericity, thrust, and transverse momentum components (in and out of the event plane)—were measured in the γ∗−pomeron center-of-mass frame.
• The analysis was performed in the kinematic range 5 ≤ Q² ≤ 185 GeV² and 160 ≤ W ≤ 250 GeV, with ηmax ≤ 1.8 to isolate large rapidity gap events.
• Data were compared with predictions from the POMPYT, VBLY, and RAPGAP Monte Carlo models, each with different assumptions about pomeron structure.
• The VBLY model used pointlike couplings to quark and gluon pairs with free parameters gIPqq and gIPgg, tuned to fit distributions including sphericity, thrust, β, and |cos θS|.
• The RAPGAP model included a pomeron with quark and gluon densities, tuned to H1 F₂^D structure function data without additional fitting in this analysis.
• Systematic uncertainties were evaluated by varying cuts on ηmax, mass reconstruction procedures, and threshold definitions; the dominant error source was systematic.

Experimental results

Research questions

• RQ1 How do event shape variables such as sphericity, thrust, and transverse momentum distributions evolve with increasing MX in large rapidity gap DIS events?
• RQ2 To what extent can standard QCD-based models like POMPYT, which rely on leading-log gluon bremsstrahlung, describe the observed planarity and collimation?
• RQ3 What is the required fraction of gluon-induced events in pomeron-exchange processes to reproduce the data, and how does this depend on the model used?
• RQ4 Does the observed broadening in high-MX events necessitate a non-pointlike, partonic structure of the pomeron, particularly a gluon component?
• RQ5 How do the results compare with e+e− annihilation data, and what does this imply about the dynamics of dijet final states in DIS?

Key findings

• With increasing MX, the mean sphericity ⟨S⟩ and ⟨1−T⟩ decrease, indicating a transition toward more collimated and planar event shapes.
• The mean transverse momentum component in the event plane increases with MX, while the out-of-plane component remains nearly constant, confirming the development of planarity.
• The POMPYT model, based on leading-log gluon bremsstrahlung, fails to reproduce the data, especially the high-MX broadening effects.
• The VBLY model requires a 50% fraction of gluon-induced events (via gIPgg coupling) to describe the data, while RAPGAP requires a 30% gluon component in the pomeron.
• RAPGAP with only a quark density fails to describe the data, and VBLY with only quark pair coupling cannot reproduce the sphericity, thrust, or β distributions.
• The observed broadening is stronger than in e+e− annihilation at comparable energies, indicating a more complex dynamics in DIS with large rapidity gaps.