Bulk neutrinos as an alternative cause of the gallium and reactor anti-neutrino anomalies

We consider an alternative explanation for the deficit of ${\nu }_e$ in Ga solar neutrino calibration experiments and of the ${\overline{\nu }}_e$ in short-baseline reactor experiments by a model where neutrinos can oscillate into sterile Kaluza-Klein modes that can propagate in compactified submicrometer flat extra dimensions. We have analyzed the results of the gallium radioactive source experiments and 19 reactor experiments with baseline shorter than 100 m, and showed that these data can be fit into this scenario. The values of the lightest neutrino mass and of the size of the largest extra dimension that are compatible with these experiments are mostly not excluded by other neutrino oscillation experiments.

Figure 1

Survival probability as a function of the distance from the ${\nu }_e$ (${\overline{\nu }}_e$) source averaged over the detector position (reactor energy spectrum). To illustrate how LED, in principle, can explain the short-baseline anomalies we show this probability for (a) standard oscillation with ${sin}^{2}2{\theta }_{13}=0.0$ and ${sin}^{2}2{\theta }_{13}=0.1$ and (b) LED for IH with ${\theta }_{13}=0$ and some values of the LED parameters: $a=0.3\mu \mathrm{m}$, $m_0=0$, $a=1.0\mu \mathrm{m}$, $m_0=0.2\mathrm{eV}$, $a=0.08\mu \mathrm{m}$, $m_0=0.1\mathrm{eV}$. We also show the average experimental deficits for the source test experiments GALLEX and SAGE, as well as for the reactor experiments ILL, Bugey-3, Bugey-3/4, ROVNO88-1S/3S, ROVNO91, ROVNO88-1I/2I, ROVNO88-2S, SRP-I, SRP-II, Gösgen-I, Gösgen-II, Gösgen-III, Krasnoyarsky-I, Krasnoyarsky-II, Krasnoyarsky-III, Palo Verde, and CHOOZ where the reactor data were taken from the Table II of Ref. 16.

Figure 2

Regions in the plane $m_0$ versus $a$ at 68%, 95%, and 99.73% CL (that is 1, 2, and 3 $\sigma$) allowed by GALLEX and SAGE source calibration experiments (left panels), by short-baseline reactor data (middle panels) and by the combined case of these two data set (left panels). For each case, the upper (lower) panel correspond to the normal (inverted) hierarchy. The hatched areas correspond to the 95% CL limits from terrestrial oscillation experiments derived in Ref. 30.