Gallium Anomaly: critical view from the global picture of νe and $$ {\overline{\nu}}_e $$ disappearance

C. Giunti, Yufeng Li, Christoph A. Ternes, O. Tyagi, Zhao XinJournal of High Energy Physics 2022 · DOI

TL;DR

This paper critically re-examines the "Gallium Anomaly" – a persistent deficit in observed electron neutrino events in experiments designed to detect solar neutrinos. By comparing Gallium Anomaly data with a comprehensive set of other neutrino experiments (reactor, tritium, and solar), the authors find a significant tension, strongly suggesting that the anomaly is not explained by the currently favored "3+1" sterile neutrino oscillation model. Instead, they propose that the Gallium Anomaly likely arises from other, as yet unidentified, causes.

What Problem Does This Paper Address?

For decades, experiments like GALLEX and SAGE have observed fewer electron neutrino events than predicted when using radioactive sources to test solar neutrino detection. This discrepancy is known as the "Gallium Anomaly." While initially attributed to uncertainties in theoretical calculations of neutrino detection cross-sections, the recent BEST experiment results have reinforced this anomaly. A leading theoretical explanation for such anomalies is the existence of "sterile neutrinos" – hypothetical particles that don't interact via the weak force – which would cause electron neutrinos to oscillate into these sterile counterparts. The "3+1" model, which posits one sterile neutrino alongside the three known "active" neutrinos, has been a popular candidate. This paper aims to rigorously test this 3+1 model by comparing the Gallium Anomaly data with a broad spectrum of other experimental constraints on electron neutrino disappearance. The core question is: does the 3+1 sterile neutrino model consistently explain all available data, or is the Gallium Anomaly a sign of something else entirely?

Methods

The authors employ a statistical analysis framework to compare different experimental results within the context of the 3+1 active-sterile neutrino mixing model. This involves:

  • Gallium Anomaly Analysis: Quantifying the anomaly's significance using various theoretical models for the neutrino detection cross-section.
  • Reactor Neutrino Analysis: Examining data from reactor neutrino experiments, both for total event rates and spectral ratios (the distribution of neutrino energies at different distances from the reactor). This is crucial because reactor antineutrinos are the antiparticles of electron neutrinos and their disappearance would also be affected by sterile neutrinos.
  • Tritium Beta Decay Analysis: Utilizing data from experiments like KATRIN, Mainz, and Troitsk, which search for deviations in the beta decay spectrum of tritium, sensitive to the mass of sterile neutrinos.
  • Solar Neutrino Analysis: Incorporating constraints from solar neutrino experiments, which probe neutrino oscillations over astronomical distances.
  • Global Fits: Combining data from all these sources to constrain the parameters of the 3+1 model, specifically the mixing angle ($sin^2 2\theta_{ee}$) and the mass-squared difference ($Am_{41}$) associated with the sterile neutrino.
  • Tension Quantification: Using a "parameter goodness of fit" (GoFPG) metric to quantify the statistical tension between different datasets when interpreted within the 3+1 model. A low GoFPG indicates a significant incompatibility.

Datasets

The paper utilizes publicly available data from a wide range of experiments:

  • Gallium Source Experiments: BEST, GALLEX, and SAGE.
  • Reactor Neutrino Experiments:
    • Rate Data: Various experiments contributing to the "Reactor Antineutrino Anomaly" discussion.
  • Spectral Ratio Data: DANSS, PROSPECT, STEREO, NEOS, and Bugey-3.
  • Tritium Beta Decay Experiments: KATRIN, Mainz, and Troitsk.
  • Solar Neutrino Experiments: Data from charged current, neutral current, and electron scattering interactions.

The authors also consider different theoretical models for reactor neutrino fluxes (HM, HKSS, EF, KI) and Gallium detection cross-sections, which are critical for interpreting the experimental results.

Key Insights and Ideas

  • Gallium Anomaly is Robust: The anomaly persists even with updated measurements and various theoretical cross-section models, indicating it's a significant experimental observation.
  • 3+1 Model Under Strain: The 3+1 sterile neutrino model, while a popular explanation for anomalies, faces severe challenges when confronted with the combined evidence from reactor, tritium, and solar neutrino data.
  • Reactor Data is Key: Modern reactor experiments, especially those measuring spectral ratios at different distances, provide strong constraints that are largely incompatible with the parameters favored by the Gallium Anomaly.
  • Solar Neutrinos Agree with Reactors, Not Gallium: Solar neutrino data, when combined with reactor constraints, further disfavors the sterile neutrino parameters needed to explain the Gallium Anomaly.
  • Tension is Severe: The statistical tension (quantified by GoFPG) between the Gallium Anomaly and other experimental datasets within the 3+1 model is often below 0.1%, indicating a strong incompatibility.

Key Results

  • Gallium Anomaly Explanation Requires Large Mixing: Explaining the Gallium Anomaly within the 3+1 model necessitates relatively large mixing angles between electron neutrinos and the sterile neutrino, implying the sterile neutrino is not "almost sterile."
  • Reactor Rate Data: Reactor rate data, even with updated flux models, imposes upper bounds on sterile neutrino mixing that are in tension with Gallium Anomaly explanations.
  • Reactor Spectral Ratio Data: This data shows a preference for short-baseline oscillations (2.6-3.3 sigma), but with mixing angles smaller than those required by the Gallium Anomaly, leading to a significant tension.
  • Combined Global Fit: When all data (reactor, tritium, solar) are combined excluding the Gallium data (due to the severe tension), a preference for 3+1 oscillations is found at the 2.6-3.3 sigma level. However, this preference is not compatible with the Gallium Anomaly data.
  • Conclusion: The authors conclude that the Gallium Anomaly is very likely not due to 3+1 active-sterile neutrino oscillations and that other explanations must be sought.

Limitations and Future Work

  • Monte Carlo Simulations: The authors note that a more reliable estimate of statistical significance for some indications (e.g., from spectral ratio data) would require computationally intensive Monte Carlo simulations, which were not performed.
  • Future Experiments: The paper implicitly calls for future experiments that can either resolve the Gallium Anomaly directly or provide even more precise measurements of neutrino oscillations to further clarify the situation.
  • Alternative Explanations: The paper explicitly states that other solutions beyond short-baseline oscillations should be pursued for the Gallium Anomaly. This could include CPT violation, new physics in neutrino detection, or unknown systematic errors.

Summary updated

Abstract

A bstract The significance of the Gallium Anomaly, from the BEST, GALLEX, and SAGE radioactive source experiments, is quantified using different theoretical calculations of the neutrino detection cross section, and its explanation due to neutrino oscillations is compared with the bounds from the analyses of reactor rate and spectral ratio data, β -decay data, and solar neutrino data. In the 3+1 active-sterile neutrino mixing scheme, the Gallium Anomaly is in strong tension with the individual and combined bounds of these data sets. In the combined scenario with all available data, the parameter goodness of fit is below 0.042%, corresponding to a severe tension of 4–5 σ , or stronger. Therefore, we conclude that one should pursue other possible solutions beyond short-baseline oscillations for the Gallium Anomaly. We also present a new global fit of ν e and $$ {\overline{\nu}}_e $$ <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:msub> <mml:mover> <mml:mi>ν</mml:mi> <mml:mo>¯</mml:mo> </mml:mover> <mml:mi>e</mml:mi> </mml:msub> </mml:math> disappearance data, showing that there is a 2.6–3.3 σ preference in favor of short-baseline oscillations, which is driven by an updated analysis of reactor spectral ratio data.
BibTeX
@article{giunti2022gallium,
  author = {C. Giunti and Yufeng Li and Christoph A. Ternes and O. Tyagi and Zhao Xin},
  title = {{Gallium Anomaly: critical view from the global picture of νe and $$ \{\textbackslash\{\}overline\{\textbackslash\{\}nu\}\}\_e $$ disappearance}},
  journal = {Journal of High Energy Physics},
  year = {2022},
  doi = {10.1007/jhep10(2022)164},
}
Figure