Low-energy sterile neutrinos: Theory
Antonio Palazzo
Max-Planck-Institut f¨ur Physik (Werner Heisenberg Institut), F¨ohringer Ring 6, 80805 M¨unchen, Germany
Abstract
Several experimental anomalies seem to point towards the existence of light sterile neutrinos. We focus on the low-energy anomalous results (the so-called gallium and reactor anomalies), which indicate a non-zero admixture Ue4of the electron neutrino with a fourth (mostly) sterile mass eigenstateν4. We point out that solar sector data, in combination with the precision measurement ofθ13, provide the constraint|Ue4|2<0.041 (90% C.L.), independent of the reactor flux determinations.
Keywords: Neutrino oscillations, sterile neutrinos
1. Introduction
The emergence of new anomalies in very-short- baseline (VSBL) neutrino oscillation measurements and in cosmological data analyses has given new impetus to the investigation of light sterile neutrinos — Standard Model gauge singlets with mass in the eV range.
From a theoretically standpoint, it seems natural to expect the existence of new gauge singlets as they ap- pear in many extensions of the Standard Model. Al- though the most popular mechanisms of neutrino mass- generation, the so-called see-saw mechanisms, involve much heavier sterile neutrinos,a priorithere is no theo- retical constraint on the mass of these particles. In fact, several models have been proposed in which light ster- ile neutrinos easily arise (see the overview given in [1]).
Essentially, the theory only tells us that sterile neutrinos may exist, without giving any certain information on their number and their mass-mixing properties, which ultimately have to be determined by the experiments.
From a phenomenological perspective, it must be en- sured that the putative new neutrino species do not spoil the basic success of the standard 3-flavor paradigm.
In the simplest and currently favored realization (the so-called 3+1 scheme), only one sterile neutrino νs
is introduced. In this scenario, the four flavor eigen- states (νe, νμ, ντ, νs) mix through the matrix elements
(Ue4,Uμ4,Uτ4,Us4) with the fourth mass eigenstateν4, assumed to be mostly sterile (|Us4| ∼1). In addition, the mass eigenstateν4is supposed to be separated by a large mass-squared splitting (Δm214 Δm224 Δm234 ∼1 eV2) from the three standard mass eigenstates (ν1, ν2, ν3). In this way the 3+1 scheme realizes a genuine perturbation of the standard 3νscenario.
2. New and old anomalies
New and more refined calculations of the reactor an- tineutrino spectra [2, 3] indicate fluxes which are ∼ 3.5% higher than previous estimates and have raised the so-called reactor anomaly [4]. In fact, adopting the new fluxes, all the VSBL (L100 m) reactor measurements show a clear deficit (a∼ 3σeffect) with respect to the theoretical expectations.
An apparently unrelated deficit (again at the ∼ 3σ level) has been evidenced in the calibration campaign conducted at the solar neutrino experiments GALLEX and SAGE [5, 6], employing radioactive sources.
In addition, the latest data released by the Mini- BooNE accelerator experiment [7], sensitive to νμ → νe transitions, now (differently from the past) seem to lend support to the longstanding LSND anomalous re- sults [8]. As a matter of fact, both experiments evidence an excess of electron neutrino events a the∼3.8σlevel.
Available online at www.sciencedirect.com
Nuclear Physics B (Proc. Suppl.) 237–238 (2013) 121–123
0920-5632/$ – see front matter © 2013 Elsevier B.V. All rights reserved.
www.elsevier.com/locate/npbps
http://dx.doi.org/10.1016/j.nuclphysbps.2013.04.073
Figure 1: Left panel: regions allowed by the solar sector data (diagonal bands) and by the dual-baseline reactor (Daya Bay and RENO) experiments (vertical bands). Middle panel: regions allowed by their combination. Right panel: combination of the constraints in the middle panel with those coming from the reactor anomaly taken from [6]. The contours refer toΔχ2=1 (dotted line) andΔχ2=4 (solid line).
A direct test of the LSND and MiniBooNE anomalies has been recently performed at the long-baseline accel- erator experiment ICARUS [9], which however is not sensitive enough to rule out all the relevant mass-mixing region.
Finally, the latest cosmological data analyses [10, 11], show a weak (∼ 2σ) but stable trend towards an extra radiation content. This hint is not in contrast with primordial nucleosynthesis [12], which however leaves less space to sterile neutrinos, allowing for no more than one additional light species.
3. An independent test of the reactor anomaly As we have shown in [13], where we have presented the analytical treatment of the solar MSW transitions in a 3+1 scheme (see also [14]), the solar sector data (Solar and KamLAND) offer a sensitive probe of the admix- ture of the electron neutrino with new sterile species. In a subsequent paper [15], we have shown how the first evidence for non-zero θ13 further improved the sensi- tivity of the solar sector to Ue4. In these proceedings we present an updated version of the analysis performed in [15] by incorporating the latest (strongest) constraints onθ13, which is now determined with far better preci- sion.
In the left panel of Fig. 1, the diagonal bands indi- cate the region allowed by the combined solar and Kam- LAND data in the plane [sin2θ13,sin2θ14].1 We stress that the KamLAND analysis has been performed using only the spectral shape information so as to render its
1In the parametrization adopted in [13, 15]Ue42 ≡sin2θ14.
results independent of the reactor antineutrino flux nor- malization.
In the same panel, the vertical bands identify the range allowed for θ13 by the combination of the dual-baseline reactor experiments Daya Bay [16] and RENO [17]. In order to understand this behavior, it should be observed that at distances of a few hundreds meters, typical of the near and far detectors of the dual- baseline reactor experiments, the oscillations driven by the new mass-squared splitting get completely aver- aged forΔm214 in the region of current interest (around
∼ 1 eV2) and their effect (an energy independent sup- pression) is identical at the near and far sites. Therefore, the estimate ofθ13 (based on a near/far comparison) is insensitive toθ14(as well as to the reactor flux normal- ization).2
The superposition of the regions allowed by the two datasets (solar+KamLAND and Daya Bay+RENO) ev- idences their complementarity in constraining the two mixing angles. Their combination, shown in the middle panel of Fig. 1, leads to the strong upper bound
sin2θ14 ≡ |Ue42|<0.041 (90% C.L.). (1) As already stressed, this bound does not depend on the normalization of the reactor fluxes and thus represents an independent constraint on|Ue4|2. Therefore, it makes sense to combine such a bound with the information coming from the reactor anomaly.3 The result of such
2The same conclusion is not true for Double Chooz [18], currently working only with the far detector (see the discussion presented in [19]).
3We do not combine the bound in Eq. (1) with the constraints com- A. Palazzo / Nuclear Physics B (Proc. Suppl.) 237–238 (2013) 121–123
122
an exercise is shown in the right panel of Fig. 1, where we have taken the reactor anomaly likelihood from [6].
The effect of the bound in Eq. (1) is a downshift of the 2σrange allowed for|Ue4|2 from [0.011,0.054] to [0.005,0.050], and a sensitive reduction of the overall statistical significance (from∼3σto∼2.5σ) of the in- dication ofθ14 >0.
4. Conclusions
We have discussed the status of light sterile neutrino searches focusing on the low-energy anomalies com- ing from reactor and gallium calibration measurements, which point towards a non-zero admixture of the elec- tron neutrino with a fourth sterile species. We have shown that it is possible to obtain an independent limit on such a mixing from the solar neutrino sector. New experiments will be needed to improve on our bound.
Acknowledgments
We acknowledge support from the European Com- munity through a Marie Curie IntraEuropean Fellow- ship, grant agreement no. PIEF-GA-2011-299582, “On the Trails of New Neutrino Properties”.
References
[1] K. N. Abazajianet al., arXiv:1204.5379 [hep-ph].
[2] T. A. Muelleret al., Phys. Rev. C83, 054615 (2011).
[3] P. Huber, Phys. Rev. C84, 024617 (2011) [Erratum-ibid. C85, 029901 (2012)].
[4] G. Mentionet al., Phys. Rev. D83, 073006 (2011).
[5] J. N. Abdurashitovet al., Phys. Rev. C73, 045805 (2006).
[6] C. Giuntiet al., arXiv:1210.5715 [hep-ph].
[7] A. A. Aguilar-Arevaloet al., arXiv:1207.4809 [hep-ex].
[8] A. Aguilaret al., Phys. Rev. D64, 112007 (2001).
[9] M. Antonelloet al., arXiv:1209.0122 [hep-ex].
[10] J. Hamannet al., Phys. Rev. Lett.105, 181301 (2010).
[11] M. Archidiaconoet al., Phys. Rev. D86, 065028 (2012).
[12] G. Mangano and P. D. Serpico, Phys. Lett. B701, 296 (2011).
[13] A. Palazzo, Phys. Rev. D83, 113013 (2011).
[14] C. Giunti and Y. F. Li, Phys. Rev. D80, 113007 (2009).
[15] A. Palazzo, Phys. Rev. D85, 077301 (2012).
[16] F. P. Anet al., arXiv:1210.6327 [hep-ex].
[17] J. K. Ahnet al., Phys. Rev. Lett.108, 191802 (2012).
[18] Y. Abeet al., Phys. Rev. D86, 052008 (2012).
[19] C. Giunti and M. Laveder, Phys. Rev. D85, 031301 (2012).
ing from the gallium anomaly since they are not independent. Indeed, a shift in the theoretical estimate of the cross-sectionνe+71Ga→71 Ge+e−, which is a critical issue for the gallium anomaly, would also modify the solar bound onθ14(andθ13).
A. Palazzo / Nuclear Physics B (Proc. Suppl.) 237–238 (2013) 121–123 123
