A tau analysis using IDEA

A preliminary analysis has been conducted to examine the tau neutrino mass using τ → 5π ντ events from Delphi LEP 1 data. This has been performed using the recently developed Integrated DElphi Analysis framework (IDEA), the testing and evaluation of which was a motivation for this project. The physics of obtaining a limit on the tau neutrino mass is briefly described, with a discussion of the channels available for this. The IDEA project and its aims are summarised, and an overview of the code developed for this analysis given. The procedure used to select the required events is examined, and improvements considered. Finally, the analysis of the selected events, using ROOT, is described, including the fitting process. A very preliminary upper limit on the tau neutrino mass is obtained, of ντ . 135 MeV, with 90% confidence. work done as a summer student in the Delphi group at CERN, Geneva, from 26th June to 14th September 2001. Supervised by D. W. Reid. Full version. [email protected] 1 The tau neutrino mass, ντ The question of whether or not neutrinos have mass is an important one. The current Standard Model requires massless neutrinos, and the possibility that this may not be the case, even if their mass is very small, requires the development of new fundamental theories. It also has implications for cosmology, as massive neutrinos may account for some of the ‘dark matter’ observed in the universe. Assuming the mass hierarchy of the three families applies across all particle types, then if they have a mass, the tau neutrino is the heaviest of the neutrinos. This suggests that the tau may be the easiest neutrino for which to demonstrate a non-zero mass. Cosmological constraints [1, 5] allow an unstable ντ with mass of the order of 10-20 MeV, or a stable ντ with sub-MeV mass. An unstable ντ is a strong possibility, and so accelerator experiments, giving limits on the order of tens of MeV, are useful. In these accelerator experiments, weak decays of the τ lepton are examined, and the kinematical variables extracted analysed to give confidence intervals for the ντ mass. The τ may decay into lighter leptons (e and μ) or an odd number of charged hadrons plus neutrals. The channels with single charged products are difficult to identify in a detector and have backgrounds too large for them to be useful. The 3and 5-charged particle channels are far more identifiable. Decays to multiple leptons or more than 5 charged hadrons are very rare (decay channels and branching ratios are listed in the Review of Particle Physics [2]). Therefore, the channels available for analysis are decays to 3 or 5 charged hadrons plus neutrals. Most of the hadrons produced in τ decays are pions, with a small number of kaons and very few more exotic hadrons. The 3-prong decay has a reasonable branching ratio, but many of the events do not give significant information on the tau neutrino mass. Conversely, the 5-prong channel has mostly statistically useful events, but suffers from a very low branching ratio. 2 Integrated DElphi Analysis framework