1 Introduction

Searches for sterile neutrinos using the T2K off-axis near detector

Debra Dewhurst on behalf of the T2K Collaboration,
Department of Physics
University of Oxford
OX1 3RH Oxford, UK

To appear in the proceedings of the Prospects in Neutrino Physics Conference, 15 – 17 December, 2014, held at Queen Mary University of London, UK.

1 Introduction

The results from a number of short baseline (SBL) neutrino experiments [1, 2] and the reanalysis of previous reactor experiments with updated antineutrino fluxes [3] suggest some incompatibility with the standard three-neutrino model (the gallium and reactor anomalies). A possible solution is the existence of sterile neutrinos [4]: right-handed particles that do not interact via the weak interaction. Their existence can be studied through their mixing with the three active Standard Model neutrinos. The 3+1 model assumes there is only one sterile neutrino whose mixing is described by a unitary matrix. If the mass squared difference is much larger than the other mass differences () the mixing can lead to SBL oscillations. The survival probability is given by

where and E are the flight path and energy of the neutrino respectively.

The existence of sterile neutrinos can be probed with the T2K Experiment [5]: a long baseline neutrino oscillation experiment in Japan. Protons impinge on a graphite target, producing a beam of , , and . Neutrinos are detected at a near detector complex 280 m from the target, and at the Super-Kamiokande far detector 295 km from the target. The near detector complex consists of two detectors, one situated on-axis (INGRID) and one situated off-axis (ND280). At ND280 the component of the beam is peaked at 600 MeV/c and the dominant interaction is charged current (CC) quasi-elastic (QE) scattering (). At higher energies pions are produced in CC resonant single pion production (CCRES), coherent pion production (CCCoh) and multi pion production due to deep inelastic scattering (CCDIS).

Here we present a search for disappearance using ND280. The data analysed corresponds to an exposure of protons on target (POT). We also present an introduction to a new analysis looking for disappearance with ND280, which promises to have interesting results in 2015.

2 Search for short baseline oscillations using ND280

To search for SBL oscillations, binned templates are built using the Monte Carlo (MC) reconstructed energy distribution assuming CCQE interactions. The templates can be weighted event-by-event with the oscillation probability to determine the dependence on the oscillation parameters of a given model. The oscillation probability affects the signal events based on the true energy and flight path of the neutrino. These templates are then compared to data in a binned likelihood ratio fit with systematic errors included as nuisance parameters with gaussian constraints, similar to other T2K analyses [6].

2.1 Search for short baseline disappearance

At ND280 the 3+1 model is investigated with in order to investigate the gallium and reactor anomalies. A sample of events is selected with a purity (efficiency) of (26[7]. The largest background comes from CCDIS or neutral current interactions where a is produced (). A control sample is used to measure this background, predominantly consisting of photon conversions from in neutral current and CCDIS interactions, with a purity (efficiency) of ().

A measurement of CC interactions at ND280 is used to reduce the flux and the correlated cross-section uncertainties, as described in [6]. The sample of CC interactions is subdivided into events without charged pions (CC0), events with one positive pion (CC) and other interactions that produce pions (CCOth). This provides sensitivity to the rate of CCQE, CCRES and CCDIS interactions.

Figure 1: Reconstructed energy distributions of the (left) and control (right) samples broken down by interactions (signal), backgrounds inside and outside the fiducial volume due to (In-FV and OOFV respectively), and all other sources of background ( other). The ratio of data to MC in the null oscillation hypothesis is shown. The red error band corresponds to the fractional systematic uncertainty. Black dots represent the data with statistical uncertainty.

The results from this analysis are based on data taken from January 2010-May 2013 (corresponding to POT). Figure 1 shows the reconstructed energy distributions of the signal and control samples. From the likelihood fit to data the best fit oscillation parameters are and  [8]. The 2D confidence intervals in the - parameter space can be seen in Figure 2. The p-value of the null hypothesis is 0.085.

Figure 2: The T2K exclusion region for disappearance at CL compared with other experimental results: allowed regions of gallium and reactor anomalies and excluded regions by - carbon interaction data and solar neutrino data [9]. The T2K best fit is marked by a green star, and those of other experiments by filled circles of the same colour as the corresponding limits.

2.2 Search for short baseline disappearance

The signal sample for this analysis is the same as the CC sample described in Section 2.1, binned in terms of the reconstructed neutrino energy assuming CCQE interactions and input into the likelihood fit described above.

Figure 3 shows the expected sensitivity for a 3+1 analysis, at CL, when flux and cross-section systematics are evaluated, compared to other experimental results. These preliminary results are promising. Once the detector systematics and final state interaction systematics have been included the data will be analysed.

Figure 3: The expected sensitivity for disappearance, at CL, based on POT of MC scaled to POT with flux and cross-section systematics included. The red and the yellow lines show the CL when the CC0, CC and CCOth samples are combined into a single CC inclusive sample. The dashed purple line shows the CL when the three samples are kept separate. The shaded region indicates the 90 CL limits from the CCFR [10] and CDHS [11] experiments. The black line represents the CL limits from MiniBooNE/SciBooNE measurements [12].

3 Summary

A search for disappearance caused by SBL oscillations has been performed with the T2K off-axis near detector. The exclusion region at CL is approximately given by and . The p-value of the null oscillation hypothesis is 0.085. These results exclude parts of the gallium anomaly and a small part of the rector anomaly allowed regions. The analysis is limited by statistical uncertainties and therefore further data from T2K will help to improve the analysis.

Searches for disappearance due to SBL oscillations are being constructed. A full MC sensitivity study with systematic uncertainties is almost complete for the 3+1 model and promises to have interesting results in 2015.


  • [1] F. Kaether, W. Hampel, G. Heusser, J. Kiko and T. Kirsten, Phys. Lett. B 685 (2010) 47.
  • [2] G. Cheng et al. [MiniBooNE and SciBooNE Collaborations], Phys. Rev. D 86 (2012) 052009.
  • [3] T. A. Mueller, D. Lhuillier, M. Fallot, A. Letourneau, S. Cormon, M. Fechner, L. Giot and T. Lasserre et al., Phys. Rev. C 83 (2011) 054615.
  • [4] M. A. Acero, C. Giunti and M. Laveder, Phys. Rev. D 78 (2008) 073009.
  • [5] K. Abe et al. [T2K Collaboration], Nucl. Instrum. Meth. A 659 (2011) 106.
  • [6] K. Abe et al. [T2K Collaboration], Phys. Rev. Lett. 112 (2014) 061802.
  • [7] K. Abe et al. [T2K Collaboration], Phys. Rev. D 89 (2014) 9, 092003.
  • [8] K. Abe et al. [T2K Collaboration], Phys. Rev. D 91 (2015) 5, 051102.
  • [9] C. Giunti, M. Laveder, Y. F. Li, Q. Y. Liu and H. W. Long, Phys. Rev. D 86 (2012) 113014 [arXiv:1210.5715 [hep-ph]].
  • [10] I. E. Stockdale, A. Bodek, F. Borcherding, N. Giokaris, K. Lang, D. Garfinkle, F. S. Merritt and M. Oreglia et al., Phys. Rev. Lett. 52 (1984) 1384.
  • [11] F. Dydak, G. J. Feldman, C. Guyot, J. P. Merlo, H. J. Meyer, J. Rothberg, J. Steinberger and H. Taureg et al., Phys. Lett. B 134 (1984) 281.
  • [12] K. B. M. Mahn et al. [SciBooNE and MiniBooNE Collaborations], Phys. Rev. D 85 (2012) 032007.
Comments 0
Request Comment
You are adding the first comment!
How to quickly get a good reply:
  • Give credit where it’s due by listing out the positive aspects of a paper before getting into which changes should be made.
  • Be specific in your critique, and provide supporting evidence with appropriate references to substantiate general statements.
  • Your comment should inspire ideas to flow and help the author improves the paper.

The better we are at sharing our knowledge with each other, the faster we move forward.
The feedback must be of minimum 40 characters and the title a minimum of 5 characters
Add comment
Loading ...
This is a comment super asjknd jkasnjk adsnkj
The feedback must be of minumum 40 characters
The feedback must be of minumum 40 characters

You are asking your first question!
How to quickly get a good answer:
  • Keep your question short and to the point
  • Check for grammar or spelling errors.
  • Phrase it like a question
Test description