Search for neutron-antineutron oscillations at the Sudbury Neutrino Observatory
Marc Bergevin for the SNO Collaboration LLNL, Rare Event Detection Group May, 2017
Simulation of a neutron-antineutron oscillation event in SNO
M. Bergevin, LLNL 1 The Sudbury Neutrino Observatory (SNO)
2 SNO was a heavy water ( H2O or D2O) Cherenkov imaging detector that was in operation from November 2, 1999 to November 28, 2006
The main focus of SNO was the neutrino oscillation observation, for which it was awarded the Nobel prize in 2015
For better physics characterization, the experiment was divided into three operational phases:
• Phase I : D2O • Phase II : D2O+NaCl • Phase III: D2O+Proportional counter arrays
M. Bergevin, LLNL 2 Exposure to neutron-antineutron oscillations for the operational phases of SNO
Phase I : D2O (50% blinded)
350.43 ± 0.01 days; (6.021±0.007)×1031 deuterons
Phase II : D2O+NaCl (85% blinded) 499.42 ± 0.01 days; (6.021±0.007)×1031 deuterons
Phase III: D2O+Proportional counter arrays (80% blinded) 392.56 ± 0.01 days; (6.015±0.007)×1031 deuterons
Total exposure
M. Bergevin, LLNL 3 Neutron-antineutron oscillations – an experimentalist’s perspective
The oscillation probability of an neutron to an antineutron can be evaluate in analogous ways to neutrino oscillation probability.
A small δm perturbation allows the neutron and antineutron to oscillate between each state
Paris potential is used for deuteron to evaluate suppression factor R [(2.48±0.08)×1022 s−1]. Optical potential used for heavier nuclei; newer calculations for Heavier nuclei showed a decrease of the suppression factor by a factor of 2
M. Bergevin, LLNL 4 Signature consists of multiple particles that will create rings within the detector
In 16O, ~23% of pions are absorbed by surrounding nuclear media. Deuteron does not suffer from this and is the focus of the analysis
In D2O, π- and π+ may be re-absorbed at the same rate
Mean free path for pion interaction is 3 times
lower in D2O compared to H2O
D Vertex reconstruction, invariant mass and momentum reconstruction is challenging
Analysis concentrates on identifying ring behavior of events for nnbar in the deuteron M. Bergevin, LLNL 5 Visible energy of nnbar events and decay modes under consideration
0.045 R − I + • Momentum regime I (at rest) 0.040 π f’2 0 π+f 0.035 A study of channels of an antiproton colliding π+X’0 0 0.030 π+X with a neutron near rest showed a majority of 0 - π A2 0.025 +A0 2-body intermediate states*. These π 2 + 0.020 π ω intermediate states can then decay into ρ+ρ0 0.015 + 0
Event Distribution π ρ channels including multiple pions. π+X(2π+2π-) 0.010
0.005
• Momentum regime II (∼ 250 MeV) : 0.000 0 5000 10000 15000 Alternative interaction channels [**] for n ̄p Photoelectrons
annihilation have also been modeled using 0.045 R − II + 0 beam data of p ̄n collisions at momenta 0.040 π π π+2π0 comparable to the 16O 0.035 π+3π0 2π+π-π0 0.030 2π+π-2π0 2 + - 0.025 π π ω Both models are investigated and give similar 3π+2π-π0 results. Weighted average of both is used in 0.020 0.015 Event Distribution
analysis 0.010
0.005 *R. Bridges et al., Phys. Rev. Lett. 56, 215 (1986) **K. Abe et al., Phys. Rev. D 91, 072006 (2015) 0 5000 10000 15000 Photoelectrons M. Bergevin, LLNL 6 Backgrounds to neutron-antineutron oscillations - atmospheric neutrinos
PHYSICAL REVIEW D 80, 012001 (2009)
Nuance models and Bartol prediction are used. Bartol prediction are corrected by scaling correction (1.22 ± 0.09) based on prior atmospheric neutrino measurement at SNO
M. Bergevin, LLNL 7 Ring counting is automated via a multiple ring finding algorithm
Part 2 Part 3 Separation of candidates Validation of ring Part 1 into multiple phase candidates Populate phase- space 0.12 µ-like ξ space with ring exp e-like ξexp candidates 0.10
0.08 Bin Expectation (PDF) 0.06
0.04
0.02
0.000 10 20 30 40 50 60 opening angle (degree)
M. Bergevin, LLNL 8 Complexity of charged pion reconstruction (x’,θ) = (0,0)
Reconstruction of π+ simulated at the origin of the 180
detector using a non-showering and showering 160 Primary Ring (degree) rec expectation, highlighting the complexity of single θ 140 non-showering pion reconstruction. 120 ring selection criterion 100 Visual validation shows good reconstructed ring for 80 60
events below x’rec < 350 cm (ring size requirement) 40
20
0 Due to greater ring finding efficiency, only -800 -600 -400 -200 0 200 400 600 800 showering expectations are used to identify rings x’rec (cm)
180 T=206.9ϒ P=-59.5ϒ 160 Primary Ring (degree) rec
θ 140
120 ring selection criterion 100 80 showering 60
40
20
0 -800 -600 -400 -200 0 200 400 600 800 Run: 1 GTID: 2 x’rec (cm)
M. Bergevin, LLNL 9 Ring result after box opening is consistent with background expectation
Ring distribution of contained events
102 event distribution 10
1 data
νatm MC + syst
n−n MC + syst
10-1 012345 # rings / event
M. Bergevin, LLNL 10 Isotropy-photoelectron phase space is consistent with background expectation
Isotropy cut Λ
1.6 background expectation Λ 1.6 signal expectation 1.4 data 1.4 data Λ = m p.e. + b 1.2 m m 1.2 Λ = mm p.e. + b m
1.0 1.0
0.8 0.8 0.6 0.6 0.4 0.4 0.2 0.2 0.0 5000 10000 15000 0.0 Number of photoelectrons (p.e.) 5000 10000 15000 Number of photoelectrons (p.e.)
M. Bergevin, LLNL 11 Results after Multi-Ring and Isotropy-Cut criteria
Ring distribution of contained events Isotropy cut Λ Λ 1.6 1.6 background expectation signal expectation
data 1.4 data 102 1.4
Λ = mm p.e. + bm = m p.e. + b 1.2 1.2 Λ m m
event distribution 1.0 1.0 10 0.8 0.8
0.6 0.6 1 data MC + syst νatm 0.4 0.4 n−n MC + syst 0.2 0.2 10-1 012345 0.0 0.0 # rings / event 5000 10000 15000 5000 10000 15000 Number of photoelectrons (p.e.) Number of photoelectrons (p.e.)
M. Bergevin, LLNL 12 Statistical map of Λ-photoelectron phase space shows data consistent with statistical fluctuations Isotropy cut Λ 1.6 background expectation
1.4 data
Λ = m p.e. + b 1.2 m m
1.0
0.8 Data sets extracted from random sampling of νatmo and nn MC
0.6 1.0 0.4 data mean
0.2 Λ ν -only (trial data sets)
P/E mean atm 0.8 0.0 5000 10000 15000 nn-only (trial data sets) Number of photoelectrons (p.e.)
Λ 0.6 1.6 signal expectation
1.4 data
1.2 Λ = mm p.e. + b m 0.4
1.0 isotropy cut 0.8 0.2 0.6
0.4 2000 4000 6000 8000 10000 12000 14000 16000 18000 0.2 photoelectron mean
0.0 5000 10000 15000 Number of photoelectrons (p.e.) M. Bergevin, LLNL 13 11.7% systematics uncertainty on detection efficiency and overall background systematics of 24.5%
M. Bergevin, LLNL 14 Results for neutron-antineutron oscillation search in deuteron for SNO using the profile likelihood method
Likelihood includes systematic uncertainties
Profile likelihood method
Intranuclear to free neutron-antineutron relation
M. Bergevin, LLNL 15 Conclusion
31 SNO finds a limit of Tintranuclear > 1.48 x10 years at 90% CL, corresponding 8 to τfree > 1.37x10 s at 90% unbounded
Paper has been submitted PRD and can be found at arXiv:1705.00696
First limit using deuteron as a target
Part of this work was performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under contract DE-AC52-07NA27344. Lawrence Livermore National Security, LLC. Release number LLNL-PRES-731181