TL;DR
This paper investigates the "reactor antineutrino anomaly" by analyzing data from the Daya Bay experiment and other reactor neutrino experiments. They find that while the Daya Bay data alone slightly disfavors sterile neutrino oscillations in favor of variations in the 235U fuel component, a combined analysis with other experiments shows a preference for sterile neutrino oscillations, particularly when combined with an enhancement of the 239Pu fuel component.
What Problem Does This Paper Address?
For decades, experiments have observed a slight deficit in the number of electron antineutrinos detected from nuclear reactors compared to theoretical predictions. This "reactor antineutrino anomaly" has led to various explanations, including the possibility of electron antineutrinos oscillating into a new, "sterile" neutrino state that doesn't interact via the weak force. However, the precise nature of this anomaly remains unclear, with different experiments and analyses yielding slightly different conclusions. This paper aims to reconcile these observations by performing a detailed statistical analysis of the Daya Bay experiment's fuel evolution data, considering various theoretical models, and combining it with previously published data from other reactor neutrino experiments. They specifically focus on disentangling the effects of potential sterile neutrino oscillations from uncertainties in the antineutrino flux originating from different nuclear fuel isotopes (primarily 235U and 239Pu).
Methods
The authors employ a standard least-squares fitting technique to analyze the experimental data. They define several theoretical models to explain the observed antineutrino rates:
- Fuel Variation Models: These models allow for independent variations in the cross-sections per fission for specific isotopes, primarily 235U (denoted as "235") and 239Pu (implicitly included in "235+239"). This accounts for uncertainties in the theoretical calculations of antineutrino production from these fuels.
- Oscillation Model (OSC): This model assumes active-sterile neutrino oscillations, where electron antineutrinos transform into sterile neutrinos with a survival probability ($P_{ee}$) that is independent of the fuel composition.
- Composite Models (e.g., 235+OSC, 239+OSC): These models combine fuel variations with active-sterile neutrino oscillations, allowing for a more comprehensive explanation of the data.
Crucially, the analysis explicitly incorporates the theoretical uncertainties associated with the antineutrino fluxes from the four main fissionable isotopes (235U, 238U, 239Pu, 241Pu) using covariance matrices. They also use Monte Carlo simulations to perform statistical comparisons of non-nested models and calculate p-values, providing a robust assessment of the significance of their findings.
Datasets
The paper utilizes two primary datasets:
- Daya Bay Evolution Data: This dataset, presented in Ref. [1], tracks the antineutrino detection rate as a function of the changing fuel composition (specifically, the 239Pu fission fraction) in the Daya Bay reactors.
- Global Reactor Rate Data: This dataset comprises previously published results from a collection of short-baseline reactor antineutrino experiments (e.g., Bugey, Rovno, Gosgen, ILL, Krasnoyarsk, SRP, Nucifer, Chooz, Palo Verde, RENO, Double Chooz) available before the Daya Bay fuel evolution data release.
Key Insights and Ideas
- Conflicting Signals: The Daya Bay evolution data alone shows a slight preference for explaining the observed anomaly through variations in the 235U fuel component rather than sterile neutrino oscillations. This is a key finding that contrasts with some earlier interpretations.
- Importance of Theoretical Uncertainties: Explicitly including the theoretical uncertainties in the antineutrino fluxes from different isotopes significantly impacts the statistical significance of the results, particularly when comparing oscillation models to fuel variation models.
- Synergy of Global Data: When combined with other reactor neutrino experiments, the sterile neutrino oscillation hypothesis (OSC) becomes more favored, and the suppression of the 235U flux is disfavored.
- Composite Models Shine: The most compelling fits are achieved by composite models that simultaneously consider both fuel variations (especially an enhancement of the 239Pu flux) and sterile neutrino oscillations. This suggests that the anomaly might be a combination of effects.
- 239Pu Enhancement: The best-fit model for the combined data involves an enhancement of the 239Pu flux, which is an interesting deviation from simply assuming suppression or standard theoretical values.
Key Results
- Daya Bay Data Alone:
- Oscillations are disfavored at 2.6σ compared to a 235U flux suppression.
- Oscillations are disfavored at 2.5σ compared to variations in both 235U and 239Pu fluxes.
- Global Reactor Rate Data Alone:
- Sterile neutrino oscillations are favored, disfavoring 235U flux suppression at 3.1σ.
- Variations in 235U and 239Pu fluxes are disfavored at 2.8σ.
- Combined Analysis:
- Shows a slight preference for oscillations over fuel variations.
- The best-fit model is the 239+OSC model, featuring an enhancement of the 239Pu flux and relatively large sterile neutrino oscillations.
Limitations and Future Work
- Statistical Significance: While the combined analysis shows a preference for certain models, the authors note that the significant uncertainties in reactor rate measurements and the good fits achieved by models both with and without sterile neutrinos mean the search for the anomaly's explanation is still open.
- Future Experiments: The paper highlights the importance of upcoming short-baseline reactor neutrino experiments (PROSPECT, SoLid, STEREO, DANSS, Neutrino-4) that will utilize reactors with different fuel compositions. These experiments are expected to provide crucial data to definitively resolve the reactor antineutrino anomaly.
Summary updated