Rapidity gap cross sections in $pp$ interactions at $\sqrt{s} = 7 \ {\rm TeV}$

Pseudorapidity gap distributions are studied using a data sample of integrated luminosity $7.08 \pm 0.24 \ {\rm \mu b^{-1}}$ from the first stable physics run provided by the Large Hadron Collider at $\sqrt{s} = 7 \ {\rm TeV}$. Events were collected using a minimum-bias trigger and are subjected to a search for large gaps in the pseudorapidity distribution of final state particles, based both on energy deposits in the ATLAS calorimeters and on tracks reconstructed in the inner detector. The data are used to tune the fraction of the inelastic cross section which is due to diffractive processes in Monte Carlo generators, yielding results close to $30\%$ for each of the PYTHIA8, PYTHIA6 and PHOJET models. Forward pseudorapidity gaps of size $\Delta \eta^F < 7$, extending to the limit of the detector acceptance at $\eta = \pm 4.9$, are used to measure cross sections differentially in gap size. At large $\Delta \eta^F$, these cross sections are dominated by contributions from the single diffractive dissociation process ($pp \rightarrow Xp$) and double dissociation ($pp \rightarrow XY$), where the invariant mass of the lighter of the two dissociation systems satisfies $M_Y \stackrel{<}{_{\sim}} 7 \ {\rm GeV}$. The result is ${\rm d} \sigma / {\rm d} \Delta \eta^F \approx 1 \ {\rm mb}$ over a wide range of gap size with $\Delta \eta^F \stackrel{>}{_{\sim}} 3$.