Dissociation of virtual photons in events with a leading proton at HERA

TL;DR

This study measures diffractive dissociation in γ^* p interactions with a leading proton at HERA, covering Q^2 ranges from 0.03–0.60 GeV^2 and 2–100 GeV^2 with M_X>1.5 GeV. By tagging the outgoing proton with the ZEUS LPS, the analysis obtains differential cross sections in t and Φ, and extracts the diffractive structure function F_2^{D(3)}(β,Q^2,x_{Iar{P}}), enabling a QCD-based interpretation via diffractive PDFs and color-dipole models. The results show a t-slope b ≈ 7.9 GeV^{-2} at small x_{Iar{P}}, a near-zero Φ-azimuthal asymmetry, and strong Q^2 evolution of F_2^{D(3)} driven by gluon-dominated diffractive PDFs, with α_{Iar{P}}(0) ≈ 1.16. BEKW and BGK saturation models describe several features, while a NLO QCD fit confirms a gluon-rich diffractive structure and yields a diffractive gluon momentum fraction around 82% at Q^2=2 GeV^2. Overall, the work provides quantitative tests of Regge factorization and diffractive QCD dynamics, informing the understanding of diffraction in ep scattering and its partonic content.

Abstract

The ZEUS detector has been used to study dissociation of virtual photons in events with a leading proton, gamma^* p -> X p, in e^+p collisions at HERA. The data cover photon virtualities in two ranges, 0.03<Q^2<0.60 GeV^2 and 2<Q^2<100 GeV^2, with M_X>1.5 GeV, where M_X is the mass of the hadronic final state, X. Events were required to have a leading proton, detected in the ZEUS leading proton spectrometer, carrying at least 90% of the incoming proton energy. The cross section is presented as a function of t, the squared four-momentum transfer at the proton vertex, Phi, the azimuthal angle between the positron scattering plane and the proton scattering plane, and Q^2. The data are presented in terms of the diffractive structure function, F_2^D(3). A next-to-leading-order QCD fit to the higher-Q^2 data set and to previously published diffractive charm production data is presented.

Figures (18)

• Figure 1: Schematic diagram of the reaction $ep \rightarrow eXp$.
• Figure 2: Comparison of the measured (points) and Monte-Carlo simulated (histograms) distributions for $x_L$, $|t|$, $Q^2$, $W$, $M_X$ and $x_{I\!\!P}$ in the low-$Q^2$ analysis. The $Q^2$ and $W$ distributions were obtained without the LPS requirement (see text).
• Figure 3: Comparison of the measured (points) and Monte-Carlo simulated (histograms) distributions for $x_L$, $|t|$, $Q^2$, $W$, $M_X$ and $x_{I\!\!P}$ in the high-$Q^2$ analysis.
• Figure 4: Distribution of $E+p_Z$ for the high-$Q^2$ events. The hatched histogram represents the beam-halo sample obtained as discussed in the text. The empty histogram is the sum of the RAPGAP Monte Carlo and the beam-halo contribution. The vertical dashed line is at $E+p_Z=1655$GeV, the value of the cut used to suppress beam-halo events.
• Figure 5: (a) The differential cross-section $d\sigma^{ep\rightarrow eXp}/dt$ in the region $x_{I\!\!P}<0.01$, $2<Q^2<100$GeV$^2$ and $M_X>1.5$GeV. The inner error bars show the statistical uncertainties and the full bars indicate the statistical and the systematic uncertainties added in quadrature. The overall normalisation uncertainty of $\pm 10$% is not shown. The line shows the result of the fit described in the text. (b) The value of the slope parameter $b$ of the differential cross-section $d\sigma^{ep \rightarrow eXp}/dt$ as a function of $Q^2$. (c) The value of the slope parameter $b$ of the differential cross-section $d\sigma^{ep \rightarrow eXp}/dt$ as a function of $x_{I\!\!P}$. The mean value of $\beta$ in each bin is also given. (d) The value of the slope parameter $b_{p_T^2}$ of the differential cross-section $d\sigma^{ep\rightarrow eXp}/dp_T^2$ as a function of $x_{I\!\!P}$. The symbols labelled ZEUS 97 indicate the present results. Earlier ZEUS results are also shown: ZEUS 95 low-xl, ZEUS 94 lps94b ($Q^2=0$) and ZEUS 94 lps94a ($Q^2>0$).