Fishing for new physics with massive neutral dibosons: measurements of ZZ production cross section and the search for invisible Higgs boson decays beyond the Standard Model with the CMS detector at the LHC

The Standard Model of particle physics is a theory describing the fundamental interactions and properties of subatomic particles. A key feature is its ability to explain particle mass through the Higgs mechanism, and a by-product of this mechanism is the Higgs boson. The discovery of the Higgs boson, in 2012 at CERN, completed the Standard Model particle zoo, but observed phenomena, like dark matter, remain unexplained. The analyses presented explore proton-proton collison events resulting in a Z boson plus missing transverse energy (MET). The motivation for this is to investigate two processes: Standard Model (SM) ZZ production, and beyond Standard Model (BSM) ZH production, in particular the ZZ → �+�−νν¯ and ZH → �+�− + Hinv channels. The place-holder Hinv is for all Higgs boson decay modes resulting in undetected “invisible” particles, which may branch to new physics, like dark matter particles. The data used are from Run 1 (2011–2012) of CMS, where proton-proton collisions at 7 TeV and 8 TeV were delivered by the LHC. The Compact Muon Solenoid (CMS) is a generalpurpose detector located along the Large Hadron Collider (LHC), which is a particle accelerator at CERN in Geneva, Switzerland. To extract these signals containing real MET from background containing fake mismeasured MET, a new “reduced MET” variable is constructed and optimized. This assists in the measurement of the ZZ production cross section. The results of the exclusive ZZ → �+�−νν¯ cross section measurement are 201+82 −69 fb and 264+81 −64 fb from the 7 and 8 TeV portions of Run 1 data, respectively. Bayesian unfolding is used to measure a cross section of 224+68 −70 fb from the 8 TeV data. These results both agree with next-to-leading order predictions from the Standard Model. The differential cross section as a function of transverse momentum of the Z boson is also measured from unfolding, for the purpose of providing a way to compare data to new theories. To distinguish ZH → �+�−+Hinv from ZZ → �+�−νν¯ a machine learning algorithm is used with physical variables as the input. A shape analysis is performed on the resulting distribution, and an upper limit is placed at 95% C.L. on the invisible branching fraction of the Higgs boson. For a Higgs boson with a Standard Model cross section and mass of 125 GeV, the observed limit on the branching fraction is 52% and the expected is 49%. Considering a mass spectrum of 115-200 GeV, a fully invisible Higgs is excluded for masses below 163 GeV.