Dijet production at the LHC at next-to-leading order

Abstract

This thesis presents an analysis of NLO inclusive dijets produced form proton-proton collisions at center of mass energies √s = 13 TeV. These dijets were simulated using Powheg+Pythia8 parton shower Monte Carlo event generators. The jets were reconstructed using FastJet and the anti-k_t jet clustering algorithm. The purpose of this analysis is to contribute in the calculation of the ratio of jet-gap-jet events divided by the number of inclusive dijet events, a ratio highly sensitive to the effects that have been predicted by the Balitsky–Fadin–Kuraev–Lipatov (BFKL) evolution equation. These effects have proven difficult to separate from other effects predicted by perturbative QCD. The cuts applied for the jet-gap-jet selection process and the observables chosen for study were in accordance with a recent preliminary analysis preformed by the CMS Collaboration at √s= 13 TeV. The main observables were the difference in pseudorapidity of the two leading jets, ∆η j j, the azimuthal angle separation of the two leading jets, ∆φ j j, the momentum of the subleading jet, p T 2, and the number of charged particles produced in the "gap" region of jet-gap-jet events. The results of our analysis found several interesting features. The first was that ∆φ j j was found to deviate from the back-to-back topology found in events produced at LO accuracies. We believe this is caused by NLO corrections that allow for the generation of three partons at the parton level instead of the maximum two partons found in LO calculations. We also found that our Monte Carlo event generators, Powheg+Pythia8, were producing an overabundance of low-energy charged particles. It was not until we reached a minimum transverse momentum for the charged particles of 1.0 GeV that the distribution of the number of charged particles began to behave as expect. We conclude that further analysis needs to be done to deduce the cause of the surplus of low-energy charged particles, and more NLO inclusive dijets events should be produced to reduce statistical uncertainty and reveal subtle features in the distributions of the obsrevables.