UniversityofChicagoNovember2-3,2018

ComplementarityofShort-BaselineNeutrinoOscillationSearcheswithCEvNS

JoelW.WalkerSamHoustonStateUniversity

withDutta,Gao,Kubik,Mahapatra,Mirabolfathi,Strigari - Phys.Rev.D94,093002+ Dutta,Denttoappear

&RepresentingtheMI𝜈ERCollaboration

Pleaseseetheexcellentreview1803.10661byDentler,Hernandez-Cabezudo,Kopp,Machado,Maltoni,Martinez-Soler,Schwetz

2

Outline

• SummaryofStatusofAnomalies:ThereareSEVERAL,generallyconsistentwithin“types”,butnotobviouslyconsistentwitheachotherglobally.SeetalkbyDannyMarfatia foronepossibleapproach–asterilesectorWITHNon-StandardInteractions

• ReviewofPhysicsandAnalysis:ProsandConsofvariousexperimentalapproachestocharacterizingshort-baselinesteriles withCEvNS

• ProjectionofSensitivitywithCEvNS:Complementarityofbeamandreactorsearches

3

Formalism

• Transitiontoselfandtransitiontoalternateflavor

• Atthematrixelementlevel:Sumoverintermediatestatesandsquaretheamplitude

4

REACTORandGALLIUMANOMALIES• Nuclear reactors produce �̅�# flavor states; effect of steriles is *disappearance*

• The “REACTOR ANOMALY”: There is a global ~ 3𝜎 flux deficit relative to thetheoretical expectation. This is amplified by recent reevaluation of the theory(Huber / Mueller et. al 1101.2663 & 1106.0687). Observed/Expected is ~ 94%

• Radiactive source experiments with Gallium (GALLEX and SAGE – 0711.4222 &1006.3244) likewise show a flux deficit.

• There is an observed “bump” in the reactor spectrum near 5 MeV (1610.04326)

• Daya Bay (1704.02276) has used time evolution of the fuel composition to breakdown flux contributions. There is a suggestion that the anomaly is associatedwith 235U, while 239Pu is consistent. This would disfavor a sterile interpretation.However, there is some disagreement on methodolgoy (1510.08948)

• Dentler et. al (1709.04294) find goodness of fit 73% with free flux normalizationsvs. 18% with fixed flux plus sterile Δ𝑚(~eV(.

• However, DANSS and NEOS prefer sterile to flux rescaling. This weakens theglobal preference. Including time-dependence of decay chains and neutroncapture on fission productsr reduces Daya Bay’s preference below 2𝜎 – P. Huber

5

Dentler, Hernandez-Cabezudo, Kopp, Machado, Maltoni, Schwetz,’18

Without reactors, a larger |Ue4 |2~ 5 - 6 x 10-2 is ok

MiniBooNe:

FromDutta

6

Daya BayDANSSandNEOS

• Newer(1607.01174,1610.0534,1606.02896)reactoranalysestakeRATIOSofobservationsatdifferentbaselinesinordertoREMOVEdependenceuponthefluxnormalizationandintrinsicspectralshape.

• Inclusionofasterileimprovesthefitatthelevelof3𝜎 (1803.10661)

7

LSNDandMiniBooNE

• At MiniBooNE, 8 GeV protons from FNAL Booster strike a Be target. Magneticallyfocused charged pions produce 𝜈- or �̅�- beams. Detector is 818 tons of mineral oil at~ 540 m baseline. Detection is flavor-sensitive CCQE off electrons. Neutrino energiesare around 500 MeV. (1805.12028)

• Around 10(0 protons on target

• There is 4.8𝜎 evidence of an excess of electron neutrino appearance.

• Two neutrino mu to e oscillation has goodness of fit 20.1%. Background onlyhypothesis is 5×1067 relative to best fit with 𝐿/𝐸; ≈ 1[m/MeV].

• This is MUCH too short for standard neutrino oscillation to be responsible. BUT – thetransition could occur *through* a sterile.

• In combination with results form the prior similar LSND experiments at Los Alamos(which is compatible) the significance is 6.1𝜎

8

MiniBooNE Results• 1805.12028Left:NeutrinoModeandRight:CombinedwithAnti-Neutrino• Bestfit“dot”shouldnotbestronglypreferredoverregionsincontours

3-10 2-10 1-10 1q22sin

2-10

1-10

1

10

210)2 (e

V2

mD

90% CLKARMEN2

90% CLOPERA

LSND 90% CL

LSND 99% CL

68% CL90% CL95% CL99% CL

CLs3 CLs4

3-10 2-10 1-10 1q22sin

2-10

1-10

1

10

210)2 (e

V2

mD

90% CLKARMEN2

90% CLOPERA

LSND 90% CL

LSND 99% CL

68% CL90% CL95% CL99% CL

CLs3 CLs4

9

Neutrino4

• Hosted at a megawatt research reactor in Russia. 95% 235U. 480 live days.

• Baseline is 6-12 meters. Core is compact and detector is segmented.

• Gadolineum-doped liquid scintillator with 1.8 m3 detects neutrinos via inverse betadecay (�̅�# + 𝑝 → 𝑒F + n).

• Analysis uses RATIOS of events and plots in 𝐿/𝐸; to extract oscillation withoutdependence upon normalization of flux.

• Claim 3𝜎 preference for oscillation. NOTE: this is a DELTA 𝜒(. The no-oscillationhypothesis is a reasonably good fit. This is NOT a 3𝜎exclusion of the SM.

• The IBD detection FULLY RECONSTRUCTS the neutrino energy – this allows for“coherency” of the oscillation over many cycles, with deep cuts as a function of∆𝑚(. It is also flavor sensitive.

• But, the cross-section is very low compared to coherent scattering

10

Neutrino4

11

Neutrino4

• Yellow,Green,andBlueareincreasinglyfavored

12

Dentler, Harnadex-Cabezudo, Kopp, Machado, Maltoni, Schwetz,’18

|Uµ4|2≤0.01

FromDutta

SearchingforNewPhysicswithCEvNS

• Largestatisticsallowprecisiondiscrimination• Cansearchfornewneutralcurrents,e.g.Z’,NSI– thiscreatesamodificationtotheRATEonly

• SensitivityisBESTtomodelsthatimpactalsotheexpectedeventdistributionSHAPE:

• Lightmediators,magneticmoment,sterile

SM&BSMEventRates

• Hugeeventratesof~1/kg/hourarepossibleintheSM• Thesignalregionstandsoutb/cofnarrowbandwidthandcoherencyenhancement• ForBSMphysicslooktodistinguishrate,shape,andSiVs.Ge

SM Coherent Nuclear Scattering

μν = 10-10 × μBohr CEνNS

MZ' = 1 TeV (E6 χ Model) CEνNS

SM Elastic Electron Scattering

μν = 10-10 × μBohr Electron Scattering

1 10 100 1000 10410

50

100

500

1000

5000

104

Detector Recoil Energy Threshold [eV]

IntegratedEventRate[kg-year]-1

Elastic Scattering of Reactor Anti-Neutrinos at 1 Meter

Solid: 72Ge, Dashed: 28Si

OscillationtoSterile4th FlavorNeutrino

• Probabilityforoscillationdependsonmixing(amplitude)andmassgap(phase)• Fortheregionofinterest,anexperimentalbaselineontheorderofmetersisrelevant• Dimensionlessscale-invariantbasisfunctionsencapsulateallaspectsoftheory

DepletionviaOscillation

10-20

20-35

35-55

55-80

80-110

110-155

155-220

220-350

350-1000

Unbinned

0.1 0.5 1 5 10 50 1000.0

0.2

0.4

0.6

0.8

1.0

Δm142 [eV]2 × L [m]

γi≡

(1-NOsc/NExp)

÷sin22θ14

Sterile Neutrino Oscillation in Reactor CEνNS with 72Ge

Nuclear Kinetic Recoil Binning [eV]

• Largervaluesintheverticalcorrespondtogreaterdepletionviaoscillation• Universalcurvebasesarerescaled(vert.)bymixingamplitudeand(horiz.)massgap• Binsareselectedforapproximatelyequivalentpopulationeventrates• Evenwithafixedlengthscale,multipleenergysamplesgivesensitivitytooscillation• Oscillationdecoheres overmultiplecycles&withmixingintheneutrinoenergy

17

COHERENTattheSNS• Stopped Positive Pion produces isotropic muon

neutrino 𝜈-of fixed energy ~ 30 MeV

• This is ~ 20X the mean energy of a reactor neutrino

• Subsequently the delayed decay of the 𝜇F to 𝑒F𝜈#�̅�-yields calculable SPECTRA with endpoint energy 𝑚-(1804.09459). The 𝜈-: 𝜈# ∶ �̅�- flavors are producedin equal proportion. BUT, for a NR threshold ~ 5 keV,the coherent scattering rates are around 0.2 : 0.3 : 0.5due to rate enhancement at higher energy.

• INTEGRATED cross section is 202 = 400X larger andrecoils are similarly more energetic – this is why lowthreshold is less critical for COHERENT. In principle, italso allows for much more massive detectors.

• Timing information helps with backgroundsuppression.

• BUT flux is ~ 105 times lower than a reactor.

18

CoherentScatteringataReactor

• Flux is high, (1012 – 1013 per cm2 per second) and backgrounds are challenging

• The reactor spectrum is (reasonably) well known.

• Because of the neutral current coherent, scattering detection never resolves flavor.

• Because of the differential cross-section, a given neutrino can produce manydifferent recoils, and the map is NOT INVERTABLE. BUT harder neutrinos will tend toproduce harder recoils, so binning in energy is essential.

• On an event-by-event basis one never knows what the neutrino energy was(directional detection would resolve this)

ReactorAnti-NeutrinoSource

• 235Uyieldsathermalenergyof202MeVperfission• Neutrinoyieldincascadeis6.14with1.5MeVmeanenergy• Ifreactorpowerisknown,thentheneutrinofluxisknown• Spectrumisexperimental(Schreckenbach etal.)above2MeV• Belowinverse𝛽 threshold,spectrumistheoretical(Kopeiken)• Coherencyofscatteringisnaturallywell-maintained• MWreactordeliversfluxof1.5×100(/cm2/sec@1m(vs.Solar~5×10P/cm2/sec)

0 2 4 6 8 1010-6

10-5

10-4

0.001

0.010

0.100

1

Anti-Neutrino Energy [MeV]

Anti-NeutrinosperMeV

Normalized Anti-Neutrino Fission Spectrum

IntegratedEventRate

• Integrateinthephysicalregionoverrecoilsandoverthenormalized𝐸; spectrum• Resultisproportionaltoflux,time,andmass,andinverselysotodistance-square• Formfactor𝐹((𝑞() issuppressed(assumedequaltounity)• ForMeVorderneutrinos,anultra-lowdetectionthresholdisvital• Note“area”isfromtheinteractioncrosssection– NOTthephysicaldetectordimension

21

Formalism

• CEvNS Neutralcurrenttouchesallflavors– useunitarity atreactors

• AndattheSNSbeamline.Ifweidealizepromptanddelayedasseparateexperimentswecansolvethesystem.

22

SNSDelayed

22

SNSDelayed

1 5 10 50 100 500 10000.0

0.2

0.4

0.6

0.8

1.0

Δm142 [eV]2 × L [m]

γi≡

(1-NOsc/NExp)

÷sin

22θ14

Sterile Neutrino Oscillation in Reactor CEνNS with CsI

23

SNSPrompt

1 5 10 50 100 500 10000.0

0.2

0.4

0.6

0.8

1.0

Δm142 [eV]2 × L [m]

γi≡

(1-NOsc/NExp)

÷sin

22θ14

Sterile Neutrino Oscillation in Reactor CEνNS with CsI

24

Reactor

0.1 1 10 100 10000.0

0.2

0.4

0.6

0.8

1.0

Δm142 [eV]2 × L [m]

γi≡

(1-NOsc/NExp)

÷sin

22θ14

Sterile Neutrino Oscillation in Reactor CEνNS with Ge

25

SNSDelayed

10-3 10-2 10-1 10010-2

10-1

100

101

sin2 2θxx

Δm142[eV]2

SNS Delayed CsI, 100 [kg·d/m2], 20 [m], ER > 5 [keV], 0 BG, 1% Sys q0A

1

2

3

4

5

26

SNSPrompt

10-3 10-2 10-1 10010-2

10-1

100

101

sin2 2θμμ

Δm142[eV]2

SNS Prompt CsI, 100 [kg·d/m2], 20 [m], ER > 5 [keV], 0 BG, 1% Sys q0A

1

2

3

4

5

27

Reactor

10-3 10-2 10-1 10010-2

10-1

100

101

sin2 2θ14

Δm142

[eV]2

Ge χ2 Significance, 10 [GW·kg·d/m2], ER > 100 [eV], 1 dru,1% Sys q0A

1

2

3

4

5

28

Reactor

10-3 10-2 10-1 10010-2

10-1

100

101

sin2 2θ14

Δm142

[eV]2

q0A = 3, Ge 10 [GW·kg·d/m2], ER > 40 [eV], 1 dru,1% Sys

1 [m]3 [m]8 [m]20 [m]50 [m]

29

SNSDelayed

10-2 10-1 10010-1

100

101

sin2 2θxx

Δm142[eV]2

q0A = 3, SNS Delayed CsI, 100 [kg·d/m2], ER > 5 [keV], 0 BG, 1% Sys

10 [m]15 [m]20 [m]30 [m]50 [m]

30

ReactorThreshold• Lowthresholdisessentialforadditionalchannels

10 50 100 500 10000

2

4

6

8

Nuclear Recoil Threshold [eV]

DiscoverySignificance

*PRELIMINARY* Sterile Oscillation, 1 DRU, sin22θ = 0.1, Δm2= 1 eV2, 10,000 kg-Days, MInER 3 [m]

31

ReactorBinning• Onemustbininordertoseparatecorrelatedeffects

5 10 200.0

0.5

1.0

1.5

2.0

Number of Bins

Exposurefor3

σ/90

%Discovery

[GW·kg·d/m2 ]

*PRELIMINARY* Sterile Oscillation at Reactor, sin22θ = 0.1, 1 DRU, 5% Systematics

32

ReactorSystematics• Largesystematicsrequirelowthresholds

1 5 10 50 100

0.5

1

5

10

Nuclear Recoil Threshold [eV]

Exposurefor3

σ/90

%Discovery

[GW·kg·d/m2 ]

*PRELIMINARY* Sterile Oscillation at Reactor, sin22θ = 0.1, 1 DRU

1% Systematics5% Systematics