No abstract

An electron antineutrino mass has been measured in tritium β-decay in the Troitsk ν-mass experiment. The setup consists of a windowless gaseous tritium source and an electrostatic electron spectrometer. The whole data set acquired from 1994 to 2004 was reanalyzed. A thorough selection of data with the reliable experimental conditions has been performed. We checked every known systematic effect and obtained the following experimental estimate for neutrino mass squared m 2 ν = −0.67 ± 2.53 eV 2 . This gives an experimental upper sensitivity limit of mν < 2.2 eV , 95% C . L. and upper limit estimates mν < 2.12 eV , 95% C .L. for Bayesian statistics and mν < 2.05 eV , 95% C .L. for the Feldman and Cousins approach.

We performed a search for any sign of an additional neutrino mass state in β-electron spectrum based on data reanalysis of direct electron antineutrino mass measurements in Tritium beta-decay in the Troitsk nu-mass experiment. The existing data set allows us to search for such a state in the mass range up to 100 eV . The lowest value at a 95% C.L. upper limit for the contribution of a heavy eigenstate into electron neutrino is around or less than 1% for masses above 20 eV .Throughout the last couple decades it has become clear that the Standard Model (SM) of elementary particles cannot explain some of the observed phenomena in particle physics, astrophysics and cosmology. These are baryon asymmetry, dark matter, neutrino oscillations and others. Neutrino oscillations from short baseline experiments favor the existence of an additional neutrino mass state to the three active neutrinos in the SM. Astrophysical observations and cosmology also point to the fact that the effective number of neutrinos is greater than 3 [1]. This can be interpreted as a possible existence of at least one sterile neutrino. Sterile neutrino is a natural consequence of the non-zero neutrino mass and appears in many theories beyond the SM. From this point of view, addition of a. Would a sterile neutrino be found it will be the first particle beyond the Standard Model. While there is a number of the results which disfavor or even are in contradiction with the hypothesis of sterile neutrino, many experiments are undergoing or are planning to search for them. For details we refer to the "Light sterile neutrinos: a white paper" [2].It becomes important to check all possible experimental data to prove, disprove or set an upper limit for the sterile neutrino hypothesis. Results of the reanalysis of our data on the direct electron antineutrino mass measurements in Tritium β-decay in the Troitsk nu-mass experiment [4] are presented in this paper. The group led by V. M. Lobashev was obtaining these data in the period of 1997-2004. We used the same file set and analysis framework as for the electron antineutrino mass. We performed a search for any sign of an additional neutrino mass state in the β-electron spectrum. Such a state with a finite mass would exhibit itself as a kink in the spectrum. Recently a similar analysis was published based on the Mainz data [3].The Troitsk experiment has two major parts: an integrating electrostatic spectrometer with adiabatic magnetic collimation and a windowless gaseous tritium source as a volume for β-decays. The spectrometer resolution was about 4 eV. We measured an integrated electron spectrum at the region of the last 200-300 eV from the spectrum endpoint (18575 eV) by varying the electrostatic potential V on the spectrometer electrode. All details on experimental setup, data taking, analysis, corrections and estimation of systematic error are published in Ref. [4]. For electron antineutrino mass squared we published the value m 2 ν = −0.67 ± 2.53 eV 2 . In accordance with Ref.[4], the spectrum of electrons ...

In this paper we present the details of our previously published results for a search for an additional neutrino mass state in β-electron spectrum from the Troitsk nu-mass experiment. Here we present steps of the analysis, show a set of likelihood functions obtained for each additional heavy mass value. We demonstrate how systematic errors were estimated. We also compare our results with those published recently for a similar analysis for Mainz data and try to explain why there is a factor of 2-3 difference in the sensitivity for an additional heavy mass. I. INTRODUCTIONRecently we have published a short paper on upper limits on an additional neutrino mass eigenstate in 2 to 100 eV region [1] obtained after reanalysis of our data on the direct electron antineutrino mass measurements in Tritium β-decay in the Troitsk nu-mass experiment [2]. The group led by V. M. Lobashev took these data in the period of 1997-2004. The same file set and the same analysis framework as for the electron anti neutrino mass were used. A search was performed for any sign of an additional neutrino mass state in the β-electron spectrum. Such a state with a finite mass would exhibit itself as a kink in the spectrum. The analysis drew much attention to the fact how data were analyzed, how systematic errors were estimated and why our limits appeared to be by a factor of 2-3 better than those from the similar analysis for Mainz data [3]. Here we present all details of the analysis and make a direct comparison with Mainz results.

The KATRIN experiment will probe the neutrino mass by measuring the β-electron energy spectrum near the endpoint of tritium β-decay. An integral energy analysis will be performed by an electro-static spectrometer ("Main Spectrometer"), an ultra-high vacuum vessel with a length of 23.2 m, a volume of 1240 m 3 , and a complex inner electrode system with about 120 000 individual parts. The strong magnetic field that guides the β-electrons is provided by super-conducting solenoids at both ends of the spectrometer. Its influence on turbo-molecular pumps and vacuum gauges had to be considered. A system consisting of 6 turbo-molecular pumps and 3 km of non-evaporable getter strips has been deployed and was tested during the commissioning of the spectrometer. In this paper the configuration, the commissioning with bake-out at 300 • C, and the performance of this system are presented in detail. The vacuum system has to maintain a pressure in the 10 −11 mbar range. It is demonstrated that the performance of the system is already close to these stringent functional requirements for the KATRIN experiment, which will start at the end of 2016.