Status of the KATRIN neutrino mass experiment

TL;DR

The KATRIN experiment aims to measure the neutrino mass with high precision using tritium beta decay, and also explores potential Lorentz violation effects within the Standard-Model Extension.

Contribution

This paper reports on the current status and future outlook of the KATRIN experiment, including its sensitivity to neutrino mass and Lorentz-violating operators.

Findings

Initial tritium run in 2018 completed

Data collection in 2019 underway for first neutrino mass results

Potential to detect Lorentz violation effects in beta decay spectrum

Abstract

The KArlsruhe TRItium Neutrino (KATRIN) experiment is designed to measure tritium $\beta$-decay spectrum with enough precision to be sensitive to neutrino mass down to 0.2eV at 90$\%$ Confidence Level. After an initial first tritium run in the summer of 2018, KATRIN is taking tritium data in 2019 that should lead to a first neutrino mass result. The $\beta$ spectral shape of the tritium decay is also sensitive to four countershaded Lorentz Violating (LV), oscillation-free operators within the Standard-Model Extension that may be quite large. The status and outlook of KATRIN to produce physics results, including in the LV sector, are discussed.

Equations

$$\frac{d\Gamma}{dT}\simeq B+C\Big{[}\big{(}\Delta T+k(T_{\oplus})\big{)}^{2}-\tfrac{1}{2}m_{\nu}^{2}\Big{]},$$

$$k(T_{\oplus })$$

Full text

\bodymatter

1 Introduction

The massive nature of neutrinos remains one of the major open questions in physics that cannot be explained by the minimal Standard Model. Among the various ways for directly probing the absolute neutrino mass, $\beta$ decay is the least model dependent unlike cosmology (cosmological models)[1] or neutrinoless double beta decay (nuclear models and quenching).[2] Kinematics alone defines the $\beta$-decay spectrum, from which the electron-antineutrino mass, in the form of the effective mass $m^{2}_{\mathrm{eff}}(\nu_{\mathrm{e}})=\Sigma_{\mathrm{i}}|U^{2}_{\mathrm{ei}}|m^{2}_{\mathrm{i}}$, will determine the spectral-shape modification very near the high-energy endpoint.

A good candidate for very precise $\beta$ spectral measurements is tritium (T). Tritium (3H) decays into 3He with the emissions of an electron and an antineutrino. The relatively low $Q$ value of $18.3\,$keV means that the fractional change to the spectrum by the neutrino mass would be large. A high-luminosity tritium source, necessary for high statistics since only a small fraction of the decays are near the endpoint energy, can be designed due to the tritium half-life being 12.3 years.

The previous generation of tritium beta-decay experiments, located at Troitsk and at Mainz, set a limit on the antineutrino mass at $\sim\!2\,$eV at 90$\%$ c.l.[4, 3] KATRIN is a next-generation tritium experiment that has recently started taking data. The ultimate sensitivity of KATRIN is expected to be $\sim\!0.2\,$eV at 90$\%$ c.l., an improvement by an order of magnitude.

2 Experiment Description

The KATRIN experiment can achieve its desired sensitivity due to a high-activity, gaseous molecular-tritum (T2) source and a high-resolution spectrometer. A $\beta$ electron will be adiabatically guided by magnetic fields to travel down the entire 70-meter beamline to the detector. After the decay takes place in the windowless, gaseous T2 source, a two-stage transport section eliminates the non-$\beta$-electron components (ions and neutral atoms) ahead of the main spectrometer. The main spectrometer, with the Magnetic Adiabatic Collimation combined with an Electrostatic filter design, will select with fine energy resolution ($0.93\,$eV at the endpoint energy of $18.6\,$keV) only $\beta$ electrons above a specific energy to pass. This filter adiabatically transforms the $\beta$ electrons’ momenta to the longitudinal direction, the same direction as the precisely set spectrometer voltage, allowing the endpoint silicon PIN diode detector to count essentially just the number of electrons with decay energies above an energy threshold. A more detailed description of the KATRIN experimental setup can be found in Ref. \refciteKATRIN_JINST. In normal KATRIN operation, an integrated spectrum of $\beta$ decay is measured.

The measurement-time distribution, i.e., where and how long to scan the $\beta$ spectrum via the main spectrometer potential setting, is optimized for the best neutrino mass sensitivity. Reference \refciteKATRIN lists the theoretical modification to the $\beta$ spectrum that KATRIN will need to consider. From the integrated spectrum with those effects considered, the neutrino mass will be one of four fundamental parameters to be fitted, alongside the background, the amplitude, and the endpoint energy, which is also important for Lorentz-violation (LV) searches.

3 Prospects for LV searches with KATRIN

The Standard-Model Extension (SME) provides a framework for small, testable Lorentz and CPT breakdown.[7] In the neutrino sector of the SME, the majority of LV coefficients are expected to be heavily suppressed by the tiny ratio of the electroweak and Planck scales ($m_{w}$/$m_{p}$) of about $10^{-17}$. However, four oscillation-free operators may be examples of the rare countershaded LV; these effects may be large but difficult to detect, akin to a shark’s different color on its belly and its back. These four operators are renormalizable, flavor blind, and of mass dimension three; other than breaking Lorentz and CPT symmetry, all other physics is conventional.[8]

The corresponding countershaded coefficients to these operators are one timelike $(a^{(3)}_{\mathrm{of}})_{00}$ and three spacelike $(a^{(3)}_{\mathrm{of}})_{1m}$ with $m=0,\pm 1$. While the $00$ term is independent of apparatus location, the spacelike terms depend on $\chi$, $\xi$, and $\theta_{0}$ representing the coordinate change to the standard choice of SME frame; for terrestrial experiments like KATRIN, the leading contribution to this transformation arises from the rotation of the Earth around its axis.

The paper by Díaz et al., on Lorentz violation and beta decay[8] both analyzed the published Troitsk and Mainz results[4, 3] to get the current best limits on two of the oscillation-free operators and provided the road map to figure out the other two operators, which requires binning the data based on sidereal time. The rest of this section will mostly summarize that paper.

For the tritium-decay energy range $\Delta T$ near the endpoint energy $T_{0}$, i.e., $\Delta T=T_{0}-T$ is small, the decay rate can be written as

[TABLE]

where $B$ is the experimental background rate and $C$ is approximately constant. The SME coefficients contribute to the function $k(T_{\oplus})$, which depends on the sidereal time $T_{\oplus}$:

[TABLE]

where ${\mathrm{Re}}(a^{(3)}_{\mathrm{of}})_{11}=(a^{(3)}_{\mathrm{of}})_{11}$ and ${\mathrm{Im}}(a^{(3)}_{\mathrm{of}})_{11}=(a^{(3)}_{\mathrm{of}})_{1-1}\equiv-(a^{(3)}_{\mathrm{of}})_{11}^{*}$.

Equation 2 shows that $k(T_{\oplus})$ will only shift the endpoint energy of the decay spectrum without changing the shape and is independent of the neutrino mass. Without knowing the sidereal time, the harmonic oscillation effects of the $1-1$ and $11$ terms on the spectrum will average out. Limits on $({a}^{(3)}_{\rm of})_{00}$ and $({a}^{(3)}_{\rm of})_{10}$ can then be obtained from the limit on the potential endpoint energy shift.

By conservatively taking $<5\,$eV to be the Troitsk and Mainz sensitivity, $|(a^{(3)}_{\mathrm{of}})_{00}|<2\times 10^{-8}\,$GeV and $|(a^{(3)}_{\mathrm{of}})_{10}|<5\times 10^{-8}\,$GeV limits were set. For comparison, the neutrinoless double beta-decay experiment EXO-200 also searched for $(a^{(3)}_{\mathrm{of}})_{00}$ from the two-neutrino double beta-decay spectrum shape;[9] their limit of $-9.39\times 10^{-5}\,$GeV$<(a^{(3)}_{\mathrm{of}})_{00}<2.69\times 10^{-5}\,$GeV is significantly worse than what a single $\beta$-decay experiment can do.

Reference \refciteLV_betadecay predicts that 30 days of nominal KATRIN run can improve the limit on $(a^{(3)}_{\mathrm{of}})_{00}$ by two orders of magnitude. A KATRIN analysis that considers sidereal time can set the first limits on $(a^{(3)}_{\mathrm{of}})_{11}$ and $(a^{(3)}_{\mathrm{of}})_{1-1}$.

4 Outlook

An initial tritium injection June 5th–18th, 2018 resulted in a successful “First Tritium” run of 81 hours worth of tritium data. The data taken from March until May of 2019 with a higher concentration of tritium, named “KNM1,” will be the first KATRIN run to be used for a neutrino mass result. With some off-season adjustments, KATRIN plans to operate at the nominal settings starting in 2020. A three-year run spanning five calendar years should allow KATRIN to reach the design sensitivity of $0.2\,$eV.

The timeline for a KATRIN LV result will depend on the prior release of neutrino mass results. Even if the blinding scheme for the neutrino mass analysis does not mask the endpoint energy result that is needed for the LV analysis, a comprehensive understanding of the endpoint energy systematics will not be available until after the completion of the main neutrino mass analysis. With some strong theoretical motivations to search for LV with the $\beta$-spectrum method, KATRIN will likely produce LV results following each neutrino mass result release in the coming years.

Acknowledgments

This work is supported by the DOE Office of Science under award no. $\#$DE-SC0019304.