Suche nach dem Higgs-Boson in hadronischen Endzuständen mit fehlender Energie am L3-Experiment bei LEP

This thesis presents searches for the Higgs boson, which is predicted by the Standard Model of Particle Physics and its extensions in order to endow fermions and bosons with their observed masses. While the Standard Model predicts only one Higgs boson "H", the Minimal Supersymmetric Extension contains five Higgs bosons. This analysis refers to the light neutral "h", which has nearly the same properties as the Standard Model Higgs boson. The analyses are based on data taken by the L3 experiment at the Large Electron Positron Collider (LEP) near Geneva during the years 1998 until 2000, which corresponds to an integrated luminosity of 629.7/pb and which were collected in a center-of-mass energy range from 189 GeV to 209 GeV. The Higgs boson searches presented here analyse the data with regard to hadronic events with missing energy, since the visible energy is reduced compared to the total energy given by the LEP collider. According to theoretical prediction these final states can be produced via the Higgs-strahlung process, e+e- --> Z* --> HZ oder e+e- --> Z* --> hZ, where a Higgs boson is radiated by a virtual Z boson that subsequently goes on-shell. The first part of the analysis is devoted to the search for the Standard Model Higgs boson in the reaction e+e- --> HZ --> q(anti-)q nu(anti-)nu, which is the second most frequent channel for Higgs production. The observed jets stem from the Higgs boson decaying into a quark- antiquark pair, being a b(anti-)b pair with about 75% probability in the mass range under consideration. With a probability of 20% the Z boson goes in neutrinos, so that the visible energy of the events in comparison to the center-of-mass energy shows a deficit according to the mass of the Z. The initial selection requires high multiplicity in order to separate hadronic and leptonic events and a minimum value of missing energy. Subsequently a neural network is used to search for Higgs decaying into b quarks, which in turn result in jets with B mesons after fragmentation. The selection is performed independently of the mass of the Higgs boson, only afterwards the neural network output is folded with the visible invariant mass in order to obtain a Higgs mass dependent discriminant for mass hypothesis in the range between 60 GeV and 120 GeV. The number of observed events agrees well with the expectation for the Standard Model background. A detailed statistical interpretation is carried out by means of a likelihood ratio. This procedure offers the possibility to prove the presence of a Higgs boson or to exclude its existence in the analysed mass range. The method yields an observed lower Higgs mass limit of m(H) > 95.6 GeV at 95% confidence level, which corresponds to a standard deviation of 1.4 sigma in comparison with the background prediction of the Standard Model. The lower mass bound stemming purely from background is 103.9 GeV, thus for example, if a Higgs boson with a mass of 100 GeV had existed, it would have been observed with a significance of 2.7 sigma. Using this method a candidate with a reconstructed mass of 114.4 GeV was observed by the L3 experiment. The combination of all event topologies in the LEP results shows a preference for a Higgs boson with a mass of 115.6 GeV. The main origin of this 2.1 sigma excess compared to the background expectation is the ALEPH four-jet channel. Because no significant excess was observed in the data of the other LEP experiments the combination yields a lower Higgs mass bound of 114.4 GeV with an expected value of 115.3 GeV from the background simulation. The second part of the analysis describes the search for the light Higgs boson h in the reaction e+e- --> HZ --> chi01 chi01 q(anti-)q. The paramter range of the Minimal Supersymmetric Standard Model (MSSM) under investigation allows for invisible decays of the Higgs boson h, those being decays to pairs of the lightest neutralino chi01, which in R-parity conserving models is a candidate for the Lightest Supersymmetric Particle (LSP). In this case two jets and missing energy are expected too, only this time the Z boson decays hadronically and the Higgs particle invisibly, vice versa to the Standard Model decay. A first preselection of events requires high multiplicity and missing energy as well as a visible mass consistent with the Z mass considering the properties of this signature. Subsquently the analysis is split for light and heavy Higgs bosons based on a cut in the relativistic velocity, since in the case of small Higgs masses the Z boson and its decay products become more energetic, so that the velocity of the jet-system is high. Then two selections are devised to get a sample of candidates, whereby mass hypothesis in the range between 50 GeV and 120 GeV were tested. Both selections use a Likelihood analysis, which exploits kinematic and event shape variables in order to obtain a Higgs mass dependent final discriminant. This function combines information about the visible mass of the Z boson and the recoil mass of a given Higgs boson. In both mass ranges the number of observed candidates is in good agreement with the expectation for the Standard Model background. Again a likelihood ratio is used to derive a mass bound for the Higgs boson. A lower limit of m(h) > 112.1 GeV is set at 95% confidence level for the hypothesis that the production cross section equals that of the Standard Model Higgs Boson. The expected boundary based on the presence of background only amounts to 111.4 GeV. A combination with the leptonic decay channels of the Z boson leads to a lower mass bound of 112.3 GeV in good agreement with the expected limit of 111.6 GeV. Relaxing the previous hypothesis, an upper limit on the production cross section is set to 0.07 pb due to the variation of the model dependent cross section. A combination of the results of all LEP experiments will follow. Presently a lower mass bound of 114.4 GeV is indicated using previously published data