Three Experiments Contradict Reactor Anomaly Sterile Neutrino Hypothesis

Once again, experiments have ruled out a theory challenging the Standard Model of Particle Physics.

Several measurements of neutrino oscillation in connection with nuclear reactors suggested that there might be a fourth and/or fifth type of neutrino in addition to the three in the Standard Model (called the "Reactor Antineutrino Anomaly" in the paper below). 

The extra one or two neutrinos would oscillate with the three "active" neutrinos, but which are "sterile" in the sense that they like interactions via the weak force that active neutrinos are charged under. (All neutrinos lack electromagnetic charge and strong force color charge, and all particles interact via gravity, which is the fourth "Core Theory" force. It is unclear if neutrinos interact directly with the Higgs field as the other Standard Model particles with non-zero rest mass do.) As the introduction to the paper below explains:

The Reactor Antineutrino Anomaly (RAA) appeared in 2011, following a revision of the predicted neutrino fluxes for the main isotopes in nuclear fuel (235U, 238U, 239Pu, 241Pu). The upward reevaluation of predicted fluxes resulted in a data-to-prediction deficit of about 5%. 

One hypothesis to explain this deficit consists in oscillations to a sterile neutrino state, since sterile neutrinos are not observable in detectors. . . .

Based on data available at the time, the RAA best-fit parameters are of the order of sin^2(2θ(new)) ∼ 0.1 and ∆m(new)^2 ∼ 2 eV^2. 

Notably, this mass splitting translates into an oscillation length from 2 to 10 meters (depending on the neutrino energy), which is the typical length on which oscillations would develop. 

Therefore, a new generation of neutrino detectors were designed to study such oscillations, combining two key requirements: (i) a distance from detector to reactor core of about 10 m, to be able to probe the RAA hypothesis; (ii) a segmented detector covering several baselines over ∼ 2 m, so that oscillations could develop inside the detector and be seen by comparing spectra from the detector’s subparts. Experiments operating such detectors are, for instance, Stereo, PROSPECT and DANSS.

It is fair to say that a new round of experiments have ruled out the sterile neutrino hypothesis that the anomaly generated, consistent with the weak force carrier boson decay and cosmology evidence, both of which strongly disfavor this possibility. 

A conference paper from the Moriond 2021 conference reviewing the results of three experiments designed to test the sterile neutrino hypothesis shows that all three experiments strongly disfavor a sterile neutrino explanation of the Reactor Antineutrino Anomaly. Specifically, the paper notes that:

The best-fit point of the RAA is strongly excluded: at > 99% C.L. by Stereo, at > 95% C.L. by PROSPECT, and at > 5σ by DANSS. A large portion of the RAA 95% C.L. region is excluded as well. The only remaining region of the parameter space still unrejected corresponds to ∆m^2 > 5 eV^2 where experiments at O(10 m) have little sensitivity.

This is illustrated with the following trio of charts:

The conclusion of the paper explains that an alternative hypothesis, which is that the model used to predict the number of antineutrinos produced by commercial nuclear reactors with a mix of fuels was flawed because it overestimated the number of antineutrinos produced by the uranium-235 component of the fuel mix, is a more likely cause of the anomaly.

Since the emergence of the RAA, an intense experimental effort has developped around very short-baseline reactor neutrino detectors, in order to search for active-to-sterile oscillations. A decade later, the best-fit parameters and a large portion of the allowed region in parameter space are ruled out by Stereo, PROSPECT or DANSS. With these segmented detectors, model-independent analyses have been performed by comparing the antineutrino spectra from several baselines, and no sign of oscillations have been found. 

Another explanation is then required to understand the data-to-prediction deficit of about 5%, which first suggested the hypothesis of a sterile neutrino. 

The contributions of 235U and 239Pu to this deficit have been separated by the Daya Bay collaboration and favor that the deficit is mostly carried by 235U. The 235U deficit is measured at (7.8±2.7)%. The measurement by Stereo on a pure 235U flux yields a compatible deficit of (5.2±2.4)%, with a pure-235U world average now at (5.0±1.3)%. 

Finally, a recent repetition by Kopeikin et al. of the measurement of 235U and 239Pu β spectra, used as inputs for the Huber-Mueller model [ed. used to determine the predicted number of neutrino events], indicates that the ratio of 235U/239Pu fluxes may have been overestimated by 5.4%. The global picture . . . suggests that the 239Pu normalization may be correct, and the 235U normalization overestimated in the HM model by 5-6%.

The paper also discuses the "5 MeV bump" excess of events at that energy at experiments running at commercial reactors (Daya Bay, RENO, Double Chooz) that is observed. 

But this doesn't necessarily suggest fundamental new physics surrounding neutrinos as opposed to a lack of a full understanding of the details of a complex set of reactions in a mixed fuel commercial nuclear reactor similar to the one that produced the Reactor Antineutrino Anomaly. The data, collectively, concerning the 5 MeV bump is inconclusive at this time.

The paper and its abstract are as follows:

Reactor neutrinos have been an intense field of investigation for the last decade. Two anomalies are discussed in this document. First, a status of the sterile neutrino searches by STEREO, PROSPECT and DANSS is presented. The best-fit parameters of active-to-sterile oscillations from the Reactor Antineutrino Anomaly are strongly rejected by these experiments. 

Second, the analyses of the shape anomaly ("5 MeV bump") by STEREO and PROSPECT, both using a virtually pure-235U neutrino flux, are detailed. Results show a significant excess of events at 5-6 MeV and indicate that the bump observed at commercial reactors is not specific to a particular isotope but rather shared among U and Pu.

Matthieu Licciardi "Results of STEREO and PROSPECT, and status of sterile neutrino searches" arxiv.org (May 28, 2021) (Contribution to the 2021 EW session of the 55th Rencontres de Moriond).