Hadron muoproduction at the COMPASS experiment

The COMPASS Collaboration has two main fields of interest: to improve our knowledge of the nucleon spin structure and to study hadrons through spectroscopy. These goals require a multipurpose universal spectrometer such as the COmmon Muon and Proton Apparatus for Structure and Spectroscopy, COMPASS. In its first years of data taking (2002-2007), the nucleon spin structure was studied with a polarized muon beam scattering off a polarized target. These studies resumed in 2010 and will continue until at least 2011. The years 2008 and 2009 were dedicated to hadron spectroscopy using hadron beams. In the case of the nucleon structure studies, it is crucial to detect with high precision the incoming beam muon (160 GeV), the scattered muon and the produced hadrons. The large amount of high quality data accumulated provides access to the unpolarized and polarized parton distributions of the nucleon and the hadronization process. Subtle differences (asymmetries) between polarized cross sections have been predicted for hadron production from polarized muon-nucleon interaction for COMPASS. It is based on these differences that the polarized parton distributions can be measured. In this context, it is important to first compare predictions with the gross features of the measured unpolarized semi-inclusive differential cross sections or the closely related differential multiplicities. In order to determine cross sections, the data has to be corrected for the acceptance of the spectrometer. In this thesis, a multidimensional acceptance correction method, based on Monte Carlo simulation, is developed and applied to the data measured in 2004. The method is first used to determine the inclusive muon-nucleon cross section which is compared with a global fit to world data. This serves as a test of the acceptance correction method and to verify if the results from previous experiments can be reproduced. Then, unpolarized differential multiplicities as a function of transverse momentum are presented for different kinematical intervals. These multiplicities can be used as benchmarks to verify the reliability of theoretical models. The subject of parton intrinsic transverse momentum is of growing interest to the spin structure community. The picture of partons moving collinear with the proton momentum is not sufficient to explain many spin features of the nucleons. Since a few years, transverse momentum dependent (TMD) distributions are integrated into theoretical models of nucleon interactions. Assumptions are made and it is to the experiment to test them. The Gaussian ansatz which assumes Gaussian behavior of the TMD distribution functions is applied and investigated. This model is very popular for its simplicity and ability to reproduce many experimental results. This analysis contributes to the verification of this model and suggests possible ameliorations. Based on this model, the intrinsic transverse momentum of the partons within the nucleon is extracted from the average transverse momenta of the measured hadrons. The extraction is carried out for different kinematical intervals to verify basic assumptions of the nucleon structure and fragmentation of partons into hadrons. Some insights are acquired about the flavor and kinematical dependence of the partons intrinsic transverse momenta. Finally, further studies and related analyses are proposed. xviii