A study of QCD processes at low momentum transfer in hadron-hadron collisions

This PhD thesis concerns the study of QCD soft processes (i.e. involving low transferred momentum) at hadron colliders. QCD (Quantum Chromodynamics) is the theory that describes the strong interactions of quarks and gluons. Since protons consist of quarks and gluons, a collision between them can be explained in terms of a collision between their constituents (partons), which carry only a fraction of the energy of the original hadrons. Most of the times a hadron collision is soft, involving low transferred momentum, but sometimes partons can interact hardly, at large transferred momentum, and their interaction can be described by perturbative QCD. Usually the remnant partons can also interact, or partons can radiate gluons before they collide (initial state radiation). These contributions, which come from sources different from the partons responsible for the hard interaction, designate the underlying event and should be removed in order to test perturbative QCD. The processes involved in the underlying event usually happen at low transferred momentum, therefore perturbative calculation cannot be applied and they have to be described by models. At the moment the underlying event subtraction is the dominant source of systematic error for the inclusive CDF jet cross section at low transverse momentum. The actual assumption is that the underlying event energy is similar to the energy found in soft events (minimum bias events). We examined both jet and minimum bias events using data from the CDF detector at the Tevatron accelerator, in order to test this assumption. We performed investigations at two center of mass energies: 1800 GeV and 630 GeV with the goal of an extrapolation to LHC energies. The LHC (Large Hadron Collider) will collide protons on protons at a center of mass energy of 14 TeV. Data from both calorimeters and tracking chambers were analyzed and compared to simulations from two Monte Carlo programs. In order to study the underlying event in jet events, we considered the transverse energy and momentum of particles inside two cones in the detector far away from the leading jet. The cone with more energy between the two was called max cone, the one with less energy min cone. The max cone should be sensitive to both NLO perturbative corrections to 2 -> 2 hard scattering and underlying event, while the min cone is sensitive only to the underlying event contribution. In minimum bias events, we picked a random cone in the central rapidity region, the energy of which should be similar to the underlying event energy in jet events. We observed that for the same available energy the underlying event in a hard scattering is considerable more active than in a soft collision. Generally, the energy in a cone in minimum bias events is lower than the energy in the min cone in jet events. CDF data and the simulation exhibit a similar behavior for the transverse energy and momentum inside the max and the min cone; the energy in the min cone remains constant, while the energy in the max cone increases as a function of the transverse energy of the most energetic jet in the event. None of the examined Monte Carlo programs with their default parameters, reproduce the data from the Tevatron in every respect. Finally, results were extended to LHC energies, where we first performed a comparison of the fast and full simulation of the ATLAS detector, one of the detectors at LHC. This proved the reliability of the ATLAS fast detector simulation for an investigation of the underlying event in jet events and minimum bias events at LHC energies. We further employed the fast detector simulation and found that the momentum distribution of charged particles in minimum bias events at CDF center of mass energy (1800 GeV) and ATLAS center of mass energy (14 TeV) does not change significantly. From the Herwig simulation, we found about 70% more energy in the min cone in jet events than in a random cone in minimum bias events. In this case, the decision whether to assume the underlying event energy as the energy found in minimum bias events, or to derive it from jet events, would be significant. Monte Carlo programs are an absolutely essential tool to understand the physics at LHC. Indeed, a good understanding of soft dynamics will be of extreme importance, since a large number of soft interactions is expected to be superimposed on any hard event of interest