Tritium ions in KATRIN: blocking, removal and detection

Abstract

The KArlsruhe TRItium Neutrino (KATRIN) experiment aims at the model-independent determination of the absolute neutrino mass scale. For this purpose, KATRIN investigates the imprint of the neutrino mass on the tritium beta spectrum near its kinematic end point. KATRIN employs a high-luminosity Windowless Gaseous Tritium Source (WGTS), which creates $10^{11}$ beta electrons per second. The beta electrons are guided by a strong magnetic field up to 6 T to the Main spectrometer, which acts as a high-pass MAC-E filter. Inadvertently, tritium ions are created from the source gas by tritium beta decay and inelastic scattering of beta electrons. Under nominal conditions, $2\cdot 10^{11}$ positive ions/s leave the WGTS along the magnetic guiding field towards the spectrometers. There, they could create background due to the radioactivity of the tritium and via the ionisation of residual gas. For this reason, the ion flux into the Pre-spectrometer (PS) has to be restricted to $1\cdot10^4$ ions/s. Ring and dipole electrodes were therefore installed in the KATRIN beamline in order to block the positive tritium ions with electrostatic potentials and to remove the stored ions via $E\times B$ drift in negative dipole potentials. A proof-of-principle for the ion blocking and removal was achieved in the First Light measurement campaign in 2016 with a pencil beam of non-radioactive deuterium ions. Nevertheless, the ring electrode potentials could be neutralised during tritium operation due to the trapping of secondary electrons and negative ions in the positive potential wells. For this reason, a constant monitoring of the ion flux into the PS was installed and used during the First Tritium measurements in May and June 2018 with about 0.5% of the nominal tritium concentration. With regard to ion safety, an upper limit on the ion flux into the PS at the aspired level of $1\cdot 10^{4}$ ions/s could be derived via the ionisation of residual gas in the PS. During the first scans of the tritium beta spectrum with KATRIN, a current measurement in the PS monitored the ion flux into the PS continuously. In the meantime, a current measurement at the one dipole electrode closest to the source measured only about 50% of the expected ion flux, while the other dipole electrodes observed no ions at all. Such inadvertent ion blocking had already been observed during First Light and is attributed to work function differences of different beam tube materials. Whether the inadvertent ion blocking also occurs during measurements with 100% tritium and how the trapped ions might affect the neutrino mass measurements needs to be subject of future investigations.