CERN Accelerating science

Two-particle angular correlations are a useful tool to study the mechanisms of particle production by observing the angular separation $(\Delta \eta, \Delta \varphi)$ between pairs of particles in an event. Different structures in the $\Delta \eta - \Delta \varphi$ space are caused by various modes of particle production and interactions between particles shortly after production. Examining these structures can give us insight into the nature of these interactions. One of these structures is called ``the Ridge'' and its significance is that the best current explanation for its origin comes from interactions within the quark-gluon plasma (QGP). Therefore, the presence of the ridge could be an indicator of the formation of QGP in a particular system. It is however often overshadowed by other structures in the correlation. In this thesis, two-particle angular correlations from proton-proton collisions at $\sqrt{s}=7$~TeV are analysed using transverse sphericity and multiplicity to isolate and study different structures in the correlation function. Transverse sphericity ($S_T$) is a momentum space event shape variable giving a measure of how isotropically particles and their momenta are distributed within an event. This variable allows us to differentiate events containing jets produced in hard processes from events containing multiple soft, non-perturbative QCD processes. Differences in the shape of the correlation function as a function of transverse sphericity are presented. A drastic change in the shape of the correlation function is observed. There appears to be consistent shrinking of the jet peak together with an overall change in the size and shape of certain underlying long-range correlations. In order to quantify the data various projections are made which are subsequently fit. The width and yield of the jet peak are extracted from the fits for different $S_T$ bins showing a quasi-exponential decrease in the yield of the jet peak with increasing $S_T$, however the width remains constant. A couple of unexpected structures appear in the correlation function, including the presence of a long range correlation imitating the Ridge in a data sample where it shouldn't appear. This indicates that the Ridge may be obtained by introducing a mathematical bias into the data sample by means of $S_T$ and possibly other variables.