Calibration and background reduction of the KATRIN experiment

Neutrino masses, recently proven to be non-zero by neutrino oscillation experiments, are important parameters in particle and nuclear physics, as well as in cosmology and astrophysics, but they were not measured until now. There are several experiments trying to determine neutrino masses. The low-background KArlsruhe TRItium Neutrino experiment (KATRIN), as well as its predecessors in Mainz and Troitsk which were able to put independently an upper limit of ~2 eV for the neutrino mass, is based on a direct model-independent way to determine average electron neutrino mass from the kinematics of tritium $\beta$-decay. It considers $\beta$-spectrum in the endpoint region using MAC-E spectroscopy technique. KATRIN is aiming to reach the sensitivity of 0.2 eV (90% C.L.) in 3 years of pure measurement time; already from four weeks of data taking, it has pushed the experimental limit down to 1.1 eV. KATRIN measures the energy of $\beta$-decay electrons emitted in Windowless Gaseous Tritium Source (WGTS) via a tandem of so-called pre- and main spectrometers (MAC-E-filters). In the region between the two spectrometers, a Penning trap is created by the combination of retarding potentials of -18.3 kV in the pre-spectrometer and -18.6 kV in the main spectrometer together with a magnetic field of about 4.5 T (nominal value) produced by a common superconducting magnet. Even at the ultra-high vacuum conditions of KATRIN, electrons lose energy due to synchrotron radiation and (in)elastic scattering with residual gas. It leads to their accumulation in the trap causing background increase which raises the statistical uncertainty of the experiment. They could even produce discharges which may interrupt the data-taking process and damage parts of the spectrometer and detector section of KATRIN. As have been demonstrated in tests of the system at various pressure conditions, higher pressure leads to higher background level and less time for a discharge formation. As a countermeasure, three pneumatically-driven electron catchers (Inconel rods) were implemented in the beamline part between the two spectrometers. Trapped electrons are guaranteed to hit an inserted catcher within one magnetron turn (on a sub-ms scale). As is demonstrated in the present work, the electron catchers proved to have a good efficiency to quench discharges at up to ~10${}^{-9}$ mbar therefore being a perfect safety measure in case of possible discharge appearance. Keeping the spectrometer pressure at the nominal level of the order of 10${}^{-11}$ mbar is crucial to sufficiently suppress the Penning trap contribution to the KATRIN background. For the absolute energy calibration monitoring and precise determination of transmission functions of pre- and main spectrometers (vital to allow KATRIN to reach its sensitivity goal), among the other sources a Condensed Krypton Source (CKrS) is used. It has been developed at the University of Müunster and was installed at the KATRIN cryogenic pumping section (CPS) at the Karlsruhe Institute of Technology (KIT) in 2017. Its core component is a 4 cm${}^2$ HOPG (highly oriented pyrolytic graphite) substrate on which gaseous ${}^{83m}$Kr is condensed. This makes the CKrS a point-like source which in combination with an ability to be moved in vertical and horizontal directions across the flux tube allows per-pixel calibration of the KATRIN focal plane detector (FPD) with comparatively high rates. It can be exploited for frequent measurements due to its relative operational simplicity comparing to the gaseous ${}^{83m}$Kr source (GKrS) which KATRIN is using as another calibration source. The cleanliness of the substrate and quality of frozen radioactive films are crucial for the source characterization and control of the stability and reproducibility of the conversion electron spectrum. At the CKrS this is monitored by means of laser ellipsometry which allows to extract information about film thickness growth with a conservatively estimated uncertainty of 2.4 Å. In this thesis, analysis of the ellipsometry data and comparison with the spectroscopic data in terms of line stability from measurements with the CKrS in krypton campaigns of 2017 and 2018 are discussed.