NuFact15 - Rio de Janeiro, Brazil - August 2015

Baksan Experiment on Sterile Transitions (BEST)∗

Yury Malyshkin,1, 2, † Bruce Cleveland,3 Sergei Danshin,1 Vladimir Gavrin,1

Valery Gorbachev,1 Tatiana Ibragimova,1 Anatoly Kalikhov,1

Tatiana Knodel,1 Yulia Kozlova,1 Victor Matveev,1 Ilya Mirmov,1

Jeff Nico,4 Alexander Shikhin,1 Evgeny Veretenkin,1 Steven Elliott,5

Hamish Robertson,6 David Sinclair,7 and John Wilkerson8

1Institute for Nuclear Research RAS, Russia2Pontifical Catholic University of Chile

3SNOLAB, Canada4National Institute of Standards and Technology, USA

5Los Alamos National Laboratory, USA6University of Washington, USA

7SNO Lab, Canada8University of North Carolina, USA

(Dated: December 17, 2015)

100

NuFact15 - Rio de Janeiro, Brazil - August 2015

AbstractA discrepancy between the results and their theoretical predictions is observed in radiochemical

calibration experiments SAGE and GALLEX in measurements with artificial radioactive sources.

This discrepancy can be explained by neutrino transitions to a hypothetical sterile state on a short

baseline, corresponding to the squared mass difference of the order of 1 eV2. A new underground

experiment BEST to search for this type of neutrino transition is planned to be carried out at

the Baksan Neutrino Observatory on the Gallium-Germanium Neutrino Telescope, which has been

used in the solar neutrino experiment SAGE. The idea is to observe the neutrino capture rate at

two distances from the source. A Ga target is divided in two concentric zones in a way that the

neutrino path lengths in each zone are equal. A statistically significant difference of the neutrino

capture rates in these zones, as well as a considerable deficiency of the average rate in both zones

in comparison with the expected rate, will indicate the existence of the neutrino oscillation on a

short baseline. The key features of this experiment are the intense compact neutrino source, which

provides a high flux of monochromatic neutrinos, low backgrounds (including solar neutrinos), and

a well established during decades technique of neutrino detection. The experiment allows to put

constraints on squared mass difference and mixing angle corresponding to the oscillations to the

hypothetical sterile neutrinos.

INTRODUCTION

In addition to three known neutrino species (νe, νµ and ντ ) there can exist one or moreneutrino flavor eigenstates. However, only three neutrino flavors with mass less than theZ0 boson mass can participate in the electro-weak interactions [1]. This gives the name‘sterile’ to other hypothetical neutrinos.

Nowadays the possibility of existence of sterile neutrinos is one the most attractive ques-tions. Up to date there is neither clear evidence of sterile neutrino observation, nor anyevidence of their non-existence. A set of unexplained experimental results (νe appearancein a pure νµ beam [2, 3], reactor anomaly [4], gallium anomaly [5, 6], and some ohters)can be interpreted as an indication of sterile neutrinos with ∆m2 ∼ 1eV2. However, thereare other experimental results making constraints on sterile neutrino parameters, e.g. [7–9].The significance of data in favor of and in contradiction to the hypothesis of sterile neutrinoexistence is more than 2.5σ [10]. This situation requires a clarification through dedicatedresearches.

GENERAL IDEA

The idea of Baksan Experiment on Sterile Neutrino Transitions (BEST) is to observetransitions of active electron neutrino from an intensive 51Cr source to sterile states on theirpassage through a Ga metal target. The probability of non-disappearance can be written inthe following form:

Pee → 1− sin2 2θ sin

(1.27

∆m2(eV2)L(m)

Eν(MeV)

),

where θ is sterile neutrino mixing angle, ∆m2 is mass difference, L is distance from thesource, and Eν is neutrino energy.

101

NuFact15 - Rio de Janeiro, Brazil - August 2015

0 1 2 3 4 5 6 7 8 9 100.650.700.750.800.850.900.951.001.051.101.151.201.25

R2/R1

R1

R2

m2(eV2)

FIG. 1: Ratios of measured ν capture rates to

expected ones in the absence of sterile neutrino

transitions in the inner zone (R1), in the outer

zone (R2), and their ratio (R2/R1) as a function

of ∆m2 for mixing angle θ = 0.3.FIG. 2: Scheme of BEST experimental setup.

The Ga detector is sensitive to only active electron flavors, therefore neutrinos oscillatedto a sterile flavor are invisible for it. This leads to a decrease of neutrino detection rate. Sinceneutrino transition to sterile state has a sinusoidal character, the detection rate depends onthe distance from the source.

The source will be placed at the center of a two-zone Ga target. The zones are completelyindependent and have different base-lines. This configuration makes it possible to observetransitions to sterile neutrino in two ways:

• comparing measured and expected rates under assumption of no sterile neutrino tran-sitions,

• comparing detection rates in two zones.

The first method requires an accurate knowledge of the source activity and neutrino reactioncross section. The second does not rely on this information, however it is not conclusive forsome ∆m2 values when the rates in both zones are close to each other. Fig. 1 illustratesthis, showing the expected capture rates R1 and R2 depending on ∆m2 (here and below therates R1,2 are normalized to ones without sterile neutrino transitions). One can see thatR1/R2 = 1 does not necessarily mean the absence of oscillations. Indeed,R1 = R2 for some∆m2, however the rates R1 and R2 are lower than it is expected in the absence of transitionsto sterile neutrino.

Finally, it is worth to mention, that the mixing angle θ defines the amplitudes of theoscillations. Fig. 1 shows the case when θ = 0.3.

DETECTION SYSTEM

The scheme of the BEST experimental setup is shown in Fig. 2. A compact high intensity3 MCi 51Cr neutrino source is planned to be placed at the center of a 50-t target of liquid

102

NuFact15 - Rio de Janeiro, Brazil - August 2015

Ga metal that is divided into independent inner and outer zones, and neutrino capture ratesin each zone are measured simultaneously. The zones are specially constructed so that theaverage neutrino path lengths (55 cm) in each zone are nearly the same. This means thatthe observed neutrino capture rates in both volumes are the same if there are no transitionsto a sterile state. As it has been already mentioned, the transition of electron neutrinoto sterile states will lead to decrease of the detected capture rates and, for certain ∆m2

values, to their difference. The zones’ baselines are optimized for the highest sensitivity at∆m2 ∼ 1eV2.

The detection technique to be used in BEST is exactly the same as was used in theSAGE experiments for decades [11]. Neutrinos are detected via decay of 71Ge produced incharge-current reaction νe+71 Ga→71 Ge+e− in the 50 t Ga target. The method has shownits reliability and we will not go into details here.

The biggest challenges for the BEST experiment are a production of a highly radioactiveneutrino source and a precise measurement of its activity. The source is planned to beproduced in research nuclear reactor SM-3 (RIAR) by exposing by neutrons of 3.5 kg of50Cr enriched to 97%. Produced by this way 51Cr nuclei will serve as a source emittingalmost monochromatic neutrinos: 90% of neutrinos have energy of 750 keV and other 10%– 430 keV. Their capture fractions are 95% and 5% respectively. Relatively small size of thesource (Ø8.6 × 9.5 cm) will allow to get a practically pure oscillation sinusoid, which givesa better sensitivity and makes the results interpretation more straightforward.

The source activity is planned to be measured by two independent methods and theiruncertainties are expected to be within 1%. The first method is based on precise measure-ments of thermal heat released from the source due to 51Cr decay [12]. A 3 MCi sourcewill emit about 650 W of heat. The second method is based on measuring with PHGe thecontinuous spectrum of gammas emitted by the source [13].

EXPECTED RESULTS

The source with initial activity of 3 MCi placed at the center of the concentric tanks willproduce a mean of 65 atoms of 71Ge per day in each zone at the beginning of measurements.The background is mostly from the Sun (0.16 and 0.83 atoms/day in the inner and theouter zones respectively) and is negligible in comparison with the signal from the source.Expected statistical and systematic uncertainties are 3.7% and 2.6% respectively, giving thetotal uncertainty of 4.5%.

Sterile neutrino transitions can be observed via certain combinations of measured topredicted capture rates under assumption of absence of oscillations in two zones R1 and R2.Possible positive outcomes are:

• R1 and R2 differ significantly from each other,

• R1 ' R2 but the average rate in both zones is considerably below the expected rate.

The improvement of the SAGE+GALLEX sensitivity by adding BEST results is shown inFig. 3. Region of sin 2θ > 0.1 can be excluded with a 3σ significance for 1 eV < ∆m2 <10 eV.

103

NuFact15 - Rio de Janeiro, Brazil - August 2015

0.01 0.1 10.1

1

10

5

SAGE+GALLEX

1.5

SAGE+GALLEX+2-zone

3

5

m2 (e

V2 )

sin22

FIG. 3: Sensitivity of the 2-zone experiment

SUMMARY

A new experiment BEST searching for an evidence of sterile neutrino transitions is underpreparation at the Baksan Neutrino Observatory. The baseline is ∼ 1 m and the experi-ment is most sensitive to ∆m2 ∼ 1eV2. The well established technique used in the SAGEexperiment will be used for neutrino detection. High source activity (3 MCi) will providerich statistic and high signal-to-backgrounds ratio. Almost monochromatic neutrino energyand the compactness of the source will allow to observe a pure sinusoidal oscillation and tosimplify the results interpretation. The experiment will start as soon as the source will beready and the measurements will take about a year.

∗ Presented at NuFact15, 10-15 Aug 2015, Rio de Janeiro, Brazil [C15-08-10.2]† malyshkin@inr.ru; Speaker

[1] D. E. Groom et al. (Particle Data Group), Eur. Phys. Jour. C15, 1 (2000).

[2] A. Aguilar-Arevalo et al. (LSND), Phys. Rev. D 64, 112007 (2001), hep-ex/0104049.

[3] A. Aguilar-Arevalo et al. (MiniBooNE Collaboration), Phys. Rev. Lett. 110, 161801 (2013).

[4] G. Mention, M. Fechner, T. Lasserre, T. A. Mueller, D. Lhuillier, M. Cribier, and A. Le-

tourneau, Phys. Rev. D 83, 073006 (2011).

[5] C. Giunti and M. Laveder, Phys. Rev. D 82, 053005 (2010).

[6] V. N. Gavrin, V. V. Gorbachev, E. P. Veretenkin, and B. T. Cleveland (2010), arXiv:1006.2103.

[7] A. Aguilar-Arevalo et al. (MiniBooNE Collaboration), Phys. Rev. Lett. 102, 101802 (2009).

[8] B. Armbruster et al. (KARMEN), Phys. Rev. D 65, 112001 (2002), hep-ex/0203021.

[9] P. Astier et al. (NOMAD), Nucl. Phys. B611, 3 (2001), hep-ex/0106102.

[10] K. N. Abazajian et al. (2012), arXiv:1204.5379.

[11] J. N. Abdurashitov et al. (SAGE), Phys. Rev. C 80, 015807 (2009).

[12] E. P. Veretenkin, V. N. Gavrin, S. N. Danshin, T. V. Ibragimova, Y. P. Kozlova, and I. N.

Mirmov, Phys. Atom. Nucl. 77, 1 (2015).

[13] V. V. Gorbachev and Y. M. Malyshkin, Instrum. Exp. Tech. 58, 418 (2015).

104