Low-energy-threshold analysis of the Phase I and Phase II data sets of the Sudbury Neutrino Observatory

B. Aharmim, S. N. Ahmed, A. E. Anthony, N. Barros, E. W. Beier, A. Bellerive, B. Beltrán, M. Bergevin, S. D. Biller, K. Boudjemline, M. G. Boulay, T. H. Burritt, B. Cai, Y-d. Chan, D. Chauhan, M. Chen, B. T. Cleveland, G. A. Cox, X. Dai, Huiqiu Deng, J. A. Detwiler, M. DiMarco, P. J. Doe, G. Doucas, P.‐l. Drouin, C. A. Duba, F. A. Duncan, M. Dunford, E. D. Earle, S. R. Elliott, H. C. Evans, G. T. Ewan, J. Farine, H. Fergani, F. Fleurot, R. J. Ford, J. A. Formaggio, N. Gagnon, J. Tm. Goon, K. Graham, E. Guillian, S. Habib, R. L. Hahn, A. L. Hallin, E. D. Hallman, Paul J. Harvey, R. Hazama, W. J. Heintzelman, J. Heise, R. L. Helmer, A. Hime, Christian Howard, M. A. Howe, M. Huang, B. Jamieson, N. A. Jelley, K. Keeter, J. R. Klein, L. L. Kormos, M. Kos, C. Kraus, C. B. Krauss, T. Kutter, Christopher C. M. Kyba, J. Law, I. Lawson, K. T. Lesko, J. R. Leslie, I. Levine, J. C. Loach, R. MacLellan, S. Majerus, H.‐b. Mak, J. Maneira, R. D. Martín, N. McCauley, A.b. McDonald, S. McGee, M. L. Miller, B. Monreal, J. Monroe, B. Morissette, B. G. Nickel, A. J. Noble, H. M. O’Keeffe, N. S. Oblath, G. D. Orebi Gann, S. M. Oser, Richard A. Ott, S. J. M. Peeters, A. W. P. Poon, G. Prior, S. D. Reitzner, K. Rielage, B.c. Robertson, R. G. H. Robertson, M. H. Schwendener, J. A. Secrest, S. R. Seibert, O. SimardPhysical 2010 · DOI · arXiv

Abstract

Results are reported from a joint analysis of Phase I and Phase II data from the Sudbury Neutrino Observatory. The effective electron kinetic energy threshold used is ${T}_{\mathrm{eff}}=3.5$ MeV, the lowest analysis threshold yet achieved with water Cherenkov detector data. In units of ${10}^{6}$ cm${}^{\ensuremath{-}2}$ ${\mathrm{s}}^{\ensuremath{-}1}$, the total flux of active-flavor neutrinos from $^{8}\mathrm{B}$ decay in the Sun measured using the neutral current (NC) reaction of neutrinos on deuterons, with no constraint on the $^{8}\mathrm{B}$ neutrino energy spectrum, is found to be ${\ensuremath{\Phi}}_{\mathrm{NC}}=5.{140}_{\ensuremath{-}0.158}^{+0.160}{\text{(stat)}}_{\ensuremath{-}0.117}^{+0.132}\text{(syst)}.$ These uncertainties are more than a factor of 2 smaller than previously published results. Also presented are the spectra of recoil electrons from the charged current reaction of neutrinos on deuterons and the elastic scattering of electrons. A fit to the Sudbury Neutrino Observatory data in which the free parameters directly describe the total $^{8}\mathrm{B}$ neutrino flux and the energy-dependent ${\ensuremath{\nu}}_{e}$ survival probability provides a measure of the total $^{8}\mathrm{B}$ neutrino flux ${\ensuremath{\Phi}}_{{}^{8}\mathrm{B}}=5.{046}_{\ensuremath{-}0.152}^{+0.159}{\text{(stat)}}_{\ensuremath{-}0.123}^{+0.107}\text{(syst)}$. Combining these new results with results of all other solar experiments and the KamLAND reactor experiment yields best-fit values of the mixing parameters of ${\ensuremath{\theta}}_{12}=34.{06}_{\ensuremath{-}0.84}^{+1.16}$ degrees and $\ensuremath{\Delta}{m}_{21}^{2}=7.{59}_{\ensuremath{-}0.21}^{+0.20}\ifmmode\times\else\texttimes\fi{}{10}^{\ensuremath{-}5}$ eV${}^{2}$. The global value of ${\ensuremath{\Phi}}_{{}^{8}\mathrm{B}}$ is extracted to a precision of ${}_{\ensuremath{-}2.95}^{+2.38}%$. In a three-flavor analysis the best fit value of $\mathrm{sin}{}^{2}{\ensuremath{\theta}}_{13}$ is $2.{00}_{\ensuremath{-}1.63}^{+2.09}\ifmmode\times\else\texttimes\fi{}{10}^{\ensuremath{-}2}$. This implies an upper bound of $\mathrm{sin}{}^{2}{\ensuremath{\theta}}_{13}<0.057$ ($95%$ C.L.).

Introduction

The "Solar Neutrino Problem" stood as a central mystery in particle physics for decades: why did terrestrial detectors observe significantly fewer neutrinos from the Sun than predicted by the Standard Solar Model? The Sudbury Neutrino Observatory (SNO) provided the definitive solution by demonstrating that neutrinos possess mass and undergo flavor oscillation during their journey to Earth. While SNO's initial results were groundbreaking, the precision of neutrino oscillation parameters and the total solar neutrino flux remained constrained by the energy thresholds of the detectors.

Historically, analyses of water Cherenkov data were limited to energies above 5.0 MeV to avoid the overwhelming "wall" of low-energy radioactive backgrounds. This paper presents a landmark joint analysis (LETA - Low Energy Threshold Analysis) of SNO's Phase I and Phase II data, successfully pushing the effective electron kinetic energy threshold down to 3.5 MeV. This lower threshold is critical for probing the transition region of the Mikheyev-Smirnov-Wolfenstein (MSW) effect, where neutrino oscillations shift from matter-dominated to vacuum-dominated regimes.

Methodology and Signal Extraction

The reduction of the analysis threshold to 3.5 MeV was made possible through several technical innovations. A primary advancement was the "Full-Time-Kinetic" (FTK) energy estimator. Unlike previous algorithms, FTK incorporates scattered and reflected light—components previously discarded as noise—resulting in a 6% improvement in energy resolution.

Signal extraction was performed using an extended maximum likelihood fit across four observable dimensions: effective kinetic energy ($T_{eff}$), reconstructed radial position ($R^3$), direction relative to the Sun ($\cos \theta_\odot$), and an isotropy parameter ($\beta_{14}$). The $\beta_{14}$ parameter proved particularly vital during the "Salt Phase" (Phase II), as it allowed for the statistical separation of single-electron events (from Charged Current interactions) and the multi-gamma cascades resulting from neutron capture on Chlorine (Neutral Current interactions).

Distributions of the four observable parameters used in the signal extraction for Phase I and Phase II signals.

Operating at such a low threshold required meticulous modeling of radioactive backgrounds. The team accounted for contributions from the acrylic vessel, photomultiplier tubes (PMTs), and trace radioactivity within the heavy water itself.

Datasets and Calibration

The analysis integrates the complete datasets from the first two operational phases of SNO:

  • Phase I (Pure $D_2O$): 277.4 live days from November 1999 to May 2001.
  • Phase II (Salt Phase): 391.4 live days from July 2001 to August 2003, with $NaCl$ added to enhance the detection efficiency of the Neutral Current reaction.

Rigorous calibration was essential for characterizing the detector response at low energies. The team utilized a variety of sources, including $^{16}$N for 6.13 MeV gammas, $^{8}$Li for high-energy betas, and a "laserball" for optical characterization across multiple wavelengths.

Summary of calibration sources used across SNO Phase I and Phase II.

Key Results and Discussion

Lowering the threshold to 3.5 MeV significantly enhanced the statistical power of the dataset. Statistics for Charged Current (CC) and Elastic Scattering (ES) events increased by approximately 30%, while Neutral Current (NC) event statistics grew by nearly 70%.

The total flux of active-flavor neutrinos from $^8$B decay was measured via the NC reaction as $\Phi_{NC} = 5.140 \times 10^6 \text{ cm}^{-2}\text{s}^{-1}$. This measurement, performed without assuming a specific energy spectrum shape, achieved a total uncertainty of approximately 4%—a twofold improvement in precision over previous SNO publications.

By combining these results with data from other solar neutrino experiments and the KamLAND reactor experiment, the team derived highly precise neutrino mixing parameters: $\theta_{12} = 34.06^{+1.16}_{-0.84}$ degrees and $\Delta m_{21}^2 = 7.59^{+0.20}_{-0.21} \times 10^{-5} \text{ eV}^2$. Furthermore, a three-flavor oscillation analysis yielded a competitive upper bound on the third mixing angle: $\sin^2 \theta_{13} < 0.057$ (95% C.L.).

Conclusion

The LETA analysis represents a significant achievement in water Cherenkov detector physics, demonstrating that sophisticated reconstruction and background modeling can push energy thresholds into previously inaccessible regimes. While the exponential rise of radioactive backgrounds remains a challenge for further threshold reductions, this work provides the most precise measurement of the $^8$B solar neutrino flux to date. These results solidify the LMA-MSW oscillation framework and set the stage for future experiments, such as SNO+, to explore even lower-energy solar neutrinos and search for neutrinoless double-beta decay.

Summary updated

Abstract

Results are reported from a joint analysis of Phase I and Phase II data from the Sudbury Neutrino Observatory. The effective electron kinetic energy threshold used is ${T}_{\mathrm{eff}}=3.5$ MeV, the lowest analysis threshold yet achieved with water Cherenkov detector data. In units of ${10}^{6}$ cm${}^{\ensuremath{-}2}$ ${\mathrm{s}}^{\ensuremath{-}1}$, the total flux of active-flavor neutrinos from $^{8}\mathrm{B}$ decay in the Sun measured using the neutral current (NC) reaction of neutrinos on deuterons, with no constraint on the $^{8}\mathrm{B}$ neutrino energy spectrum, is found to be ${\ensuremath{\Phi}}_{\mathrm{NC}}=5.{140}_{\ensuremath{-}0.158}^{+0.160}{\text{(stat)}}_{\ensuremath{-}0.117}^{+0.132}\text{(syst)}.$ These uncertainties are more than a factor of 2 smaller than previously published results. Also presented are the spectra of recoil electrons from the charged current reaction of neutrinos on deuterons and the elastic scattering of electrons. A fit to the Sudbury Neutrino Observatory data in which the free parameters directly describe the total $^{8}\mathrm{B}$ neutrino flux and the energy-dependent ${\ensuremath{\nu}}_{e}$ survival probability provides a measure of the total $^{8}\mathrm{B}$ neutrino flux ${\ensuremath{\Phi}}_{{}^{8}\mathrm{B}}=5.{046}_{\ensuremath{-}0.152}^{+0.159}{\text{(stat)}}_{\ensuremath{-}0.123}^{+0.107}\text{(syst)}$. Combining these new results with results of all other solar experiments and the KamLAND reactor experiment yields best-fit values of the mixing parameters of ${\ensuremath{\theta}}_{12}=34.{06}_{\ensuremath{-}0.84}^{+1.16}$ degrees and $\ensuremath{\Delta}{m}_{21}^{2}=7.{59}_{\ensuremath{-}0.21}^{+0.20}\ifmmode\times\else\texttimes\fi{}{10}^{\ensuremath{-}5}$ eV${}^{2}$. The global value of ${\ensuremath{\Phi}}_{{}^{8}\mathrm{B}}$ is extracted to a precision of ${}_{\ensuremath{-}2.95}^{+2.38}%$. In a three-flavor analysis the best fit value of $\mathrm{sin}{}^{2}{\ensuremath{\theta}}_{13}$ is $2.{00}_{\ensuremath{-}1.63}^{+2.09}\ifmmode\times\else\texttimes\fi{}{10}^{\ensuremath{-}2}$. This implies an upper bound of $\mathrm{sin}{}^{2}{\ensuremath{\theta}}_{13}&lt;0.057$ ($95%$ C.L.).
BibTeX
@inproceedings{aharmim2010low,
  author = {B. Aharmim and S. N. Ahmed and A. E. Anthony and N. Barros and E. W. Beier and A. Bellerive and B. Beltrán and M. Bergevin and S. D. Biller and K. Boudjemline and M. G. Boulay and T. H. Burritt and B. Cai and Y-D. Chan and D. Chauhan and M. Chen and B. T. Cleveland and G. A. Cox and X. Dai and Huiqiu Deng and J. A. Detwiler and M. DiMarco and P. J. Doe and G. Doucas and P.‐L. Drouin and C. A. Duba and F. A. Duncan and M. Dunford and E. D. Earle and S. R. Elliott and H. C. Evans and G. T. Ewan and J. Farine and H. Fergani and F. Fleurot and R. J. Ford and J. A. Formaggio and N. Gagnon and J. TM. Goon and K. Graham and E. Guillian and S. Habib and R. L. Hahn and A. L. Hallin and E. D. Hallman and Paul J. Harvey and R. Hazama and W. J. Heintzelman and J. Heise and R. L. Helmer and A. Hime and Christian Howard and M. A. Howe and M. Huang and B. Jamieson and N. A. Jelley and K. Keeter and J. R. Klein and L. L. Kormos and M. Kos and C. Kraus and C. B. Krauss and T. Kutter and Christopher C. M. Kyba and J. Law and I. Lawson and K. T. Lesko and J. R. Leslie and I. Levine and J. C. Loach and R. MacLellan and S. Majerus and H.‐B. Mak and J. Maneira and R. D. Martín and N. McCauley and A.B. McDonald and S. McGee and M. L. Miller and B. Monreal and J. Monroe and B. Morissette and B. G. Nickel and A. J. Noble and H. M. O’Keeffe and N. S. Oblath and G. D. Orebi Gann and S. M. Oser and Richard A. Ott and S. J. M. Peeters and A. W. P. Poon and G. Prior and S. D. Reitzner and K. Rielage and B.C. Robertson and R. G. H. Robertson and M. H. Schwendener and J. A. Secrest and S. R. Seibert and O. Simard},
  title = {{Low-energy-threshold analysis of the Phase I and Phase II data sets of the Sudbury Neutrino Observatory}},
  booktitle = {Physical},
  year = {2010},
  doi = {10.1103/physrevc.81.055504},
}
Figure