Overview of Reactor Neutrino Measurements

1
Overview of Reactor Neutrino Measurements

• Michael Shaevitz
• Columbia University
• NOW2004
• With the recent confirmation by Kamland and
  isolation of the Dmsolar2 in the LMA region, the
  field is turning to
• Measuring the last mixing angle q13
• Obtaining better precision on q23, q12, Dmsolar2
  and Dmatm2 (along with checking LSND)
• ? Road to measuring n-mass hierarchy and
  n-CP violation
• Reactor oscillation measurements unique for
• Measuring q13
• Constraining q23

2
Outline

• Reactor neutrino measurements
• Physics capabilities
• Experimental method
• How to measure a small disappearance signal
• Possible experimental sites
• Sensitivity studies
• Comparisons and combinations with offaxis
  experiments

3
Brief Aside on??e Elastic Scattering

• Use Osc. Experiment near detector for
  measurement
• Use far detector to measure background

• A reactor-based weak mixing angle (sin2?W)
  measurement
• Probe new physics in neutrino sector at low Q2
• With errors comparable to APV, Moller Scattering,
  and NuTeV
• Also, sensitivity to ? magnetic moment (x3
  better than current).

4
Weak Mixing Angle Measurement

• Measure the elastic scattering rate using the
  Near Detector
• ??e e ? ??e e
• Measure the neutrino flux using Inverse Beta
  Decay (IBD) events in the same fiducial volume
  with same deadtime.
• ??e p ? e n
• Reduction of backgrounds
• Only use events in the 3 lt Evisible lt 5 MeV
  window
• Veto events with delayed coincident neutrons
• Veto events with various types of associated
  cosmic muons
• Use the Far Detector to monitor the non-elastic
  background rate.

• Experiment needs a close detector (lt 200m) with
  gt 300 mwe shielding
• Detector requirements very similar to oscillation
  experiment ? Low cost addition to osc. Program
• ? Need to obtain reduced backgnds

? d(sin2qW) 0.0019(compare to
NuTeV 0.0017)
(See J.Conrad, J.Link, and M.S., hep-ex/0403048)
5
Reactor Measurements of ?13

• Nuclear reactors are a very intense sources of??e
  with a well understood spectrum
• 3 GW 21021 MeV/s ? 61020?ne/s
• Reactor spectrum peaks at 3.7 MeV
• Oscillation Max. for Dm22.5?10-3 eV2 at L 1.8
  km
• Osc prob (sin2q131.0) 0.81 for 1.8 km
  0.55 for 1.05 km

• Disappearance Measurement Look for small rate
  deviation from 1/r2 measured at a near and
  far baselines
• Counting Experiment
• Compare events in near and far detector
• Energy Shape Experiment
• Compare energy spectrum in near and far detector

6
Reactor Neutrino Event Signature

• The reaction process is inverse ß-decay followed
  by neutron capture
• Two part coincidence signal is crucial for
  background reduction.
• Positron energy spectrum implies the neutrino
  spectrum
• In undoped scintillator the neutron will capture
  on hydrogen
• More likely the scintillator will be doped with
  gadolinium to enhance capture

E? Evis 1.8 MeV 2me
n H ? D g (2.2 MeV)
n mGd ? m1Gd gs (8 MeV)
7
Physics of ?13 at Reactors

• The reactor experiment can make an unambiguous
  measurement of the mixing parameter sin22?13.

• Small dependence on
• mass hierarchy
• CP violation phase, ?CP
• sin2?23 and ?23 degeneracy
• sin2?12 and Dmsolar2

8
Current Status of ?13 Knowledge

• Recent global fits including solar KamLAND
  CHOOZ
• Kamland spectral distortion fixes Dm2Solar ?
  Compare to solar gives q13
• Gives reduced 90 CL upper limits especially at
  smaller Dm132
• sin22?13 lt0.09 _at_ Dm1322.5?10-3 eV2
• sin22?13 lt0.12 _at_ Dm1321.3?10-3 eV2

• CHOOZ and Palo Verde Experiments
• Neither experiments found evidence for?ne
  oscillation
• sin22q13lt 0.14 at 90 CL (at Dm22.510-3 eV2)

CHOOZ Only
Maltoni et al.hep-ph/0405172
9
Physics of ?13 at Reactors

• The reactor experiment can make an unambiguous
  measurement of the mixing parameter sin22?13.
• Sensitivity should reach sin22?13 0.01 to 0.03
  at 90 CL in three years of running. Better
  sensitivity is also possible.
• In conjunction with the off-axis (beam
  experiments) the reactor experiment breaks the
  degeneracy of ?23 lt45? or gt45?.

10
The ?23 Degeneracy Problem
Disappearance neutrino measurements are sensitive
to sin22?23 But the leading order term in
offaxis ?µ??e oscillations is
Super-K / Minos / T2K Measures
Offaxis q13 Measures
Example Measurement of sin22 ?23 0.95
? ?23 38? or 52? Prediction
for appearance rate ? sin2?23 ?
sin2?23 0.38 or 0.62 (x1.6 uncertainty)
11
Physics of ?13 at Reactors

• The reactor experiment can make an unambiguous
  measurement of the mixing parameter sin22?13.
• Sensitivity should reach sin22?13 0.01 to 0.03
  at 90 CL in three years of running. Better
  sensitivity is possible.
• In conjunction with the off-axis (beam
  experiments) the reactor experiment breaks the
  degeneracy of ?23 lt45? or gt45?
• Direct knowledge of the mixing angles is
  important in its own right! Could be crucial to
  constructing a theory of flavor.
• The reactor measurement determines the
  feasibility of CP violation and mass hierarchy
  studies in off-axis.

12
How Do You Measure a Small Disappearance?

• Use identical near and far detectors to cancel
  many sources of systematics.

13
Sin22?13 Reactor Experiment Basics
Well understood, isotropic source of electron
anti-neutrinos
Oscillations observed as a deficit of ?e
E? 8 MeV
1.0
Unoscillated flux observed here
Probability ?e
Survival Probability
Distance
1200 to 1800 meters
14
How Do You Measure a Small Disappearance?

• Use identical near and far detectors to cancel
  many sources of systematics.
• Design detectors to eliminate the need for
  analysis cuts that may introduce systematic error.

15
Detector Design Basics

• Homogenous Volume
• Viewed by PMTs
• Coverage of 20 or better
• Gadolinium Loaded, Liquid Scintillator Target
• Enhances neutron capture
• Unloaded Scintillator Region
• To capture energy from gamma rays. Eliminates
  need for fiducial volume cut.
• Pure Mineral Oil Buffer
• To shield the scintillator from radioactivity in
  the PMT glass. Allows you to set an energy cut
  well below the 1 MeV ee- annihilation energy.

16
How Do You Measure a Small Disappearance?

• Use identical near and far detectors to cancel
  many sources of systematics.
• Design detectors to eliminate the need for
  analysis cuts that may introduce systematic
  error.
• Detector cross calibration may be used to
  further reduce the near/far normalization
  systematic error.
• ? Use common sources to cross calibrate
• ? Move far detectors to near site for cross
  calibration

17
How Do You Measure a Small Disappearance?

• Use identical near and far detectors to cancel
  many sources of systematics.
• Design detectors to eliminate the need for
  analysis cuts that may introduce systematic
  error.
• Detector cross calibration may be used to
  further reduce the near/far normalization
  systematic error.
• Reduce background rate and uncertainty

18
Backgrounds

• There are two types of background
• Uncorrelated - Two random events that occur
  close together in space and time and mimic the
  parts of the coincidence.
• This BG rate can be estimated by measuring
  the singles rates, or by switching the order of
  the coincidence events.
• Correlated - One event that mimics both parts of
  the coincidence signal.
• These may be caused by fast neutrons (from
  cosmic ms) that strike a proton in the
  scintillator. The recoiling proton mimics the e
  and the neutron captures.
• Or they may be cause by muon produced isotopes
  like 9Li and 8He which sometimes decay to ßn.
• Estimating the correlated rate is much more
  difficult!

19
How Do You Measure a Small Disappearance?

• Use identical near and far detectors to cancel
  many sources of systematics.
• Design detectors to eliminate the need for
  analysis cuts that may introduce systematic
  error.
• Detector cross calibration may be used to
  further reduce the near/far normalization
  systematic error.
• Reduce background rate and uncertainty
• Go as deep as you can
• Veto

20
Veto Background Events
Fast neutrons Veto ms and shield neutrons
9Li and 8He
E 10.6 MeV t½ 0.18 to 0.12 s 0.075
produced/ton/day (450 mwe) 50 to 16 correlated
ßn
Shielding
KamLAND Data
A ½ second veto after every muon that deposits
more that 2 GeV in the detector should eliminate
70 to 80 of all correlated decays.
6 meters
21
How Do You Measure a Small Disappearance?

• Use identical near and far detectors to cancel
  many sources of systematics.
• Design detectors to eliminate the need for
  analysis cuts that may introduce systematic
  error.
• Detector cross calibration may be used to
  further reduce the near/far normalization
  systematic error.
• Reduce background rate and uncertainty
• Go as deep as you can
• Veto
• Use vetoed events to measure the background
• Redundant measurements to give convincing signal
• Multiple detectors at each site
• See osc. signal in both rate and spectral
  distortion

22
Scales of Experiments and Sensitivities

• small sin22q130.03
• Goal fast experiment to explore region x3-4
  below the Chooz limit.
• Sensitivity through rate mainly
• Example Double-Chooz experiment (300
  GW-ton-yrs)
• medium sin22q130.01
• Make a discovery of q13 in region of interest for
  the next 10-20 year program
• Sensitivity enough to augment the physics of
  offaxis measurements
• Sensitivity both to rate and energy shape
• Example Braidwood, Daya Bay (3000 GW-ton-yrs)
• large sin22q130.002-0.004??
• Measurement capability comparable to second
  generation offaxis experiments
• Sensitivity mainly through energy shape
  distortions
• MiniBooNE/Kamland sized detector (20,000
  GW-ton-yrs)

23
Counting (Rate) vs Energy Shape Measurement
Small detectors give a rate comparison near to
farLarge detectors can show a energy shape
distortion near to far
medium
sin22?13 Sensitivity
Fit uses spectral shape only
Statistical error only
From Huber et al.hep-ph/0303232
large
small
Exposure (GWtonyears)
The location of the transition from rate to shape
depends on the level of systematic error ? snorm
relative normalization of the two detectors
24
Proposed Reactor Oscillation Experiments
25
Proposed Sites Around the World
Site Power (GWthermal) Baseline Near/Far (m) Shielding Near/Far (mwe) Sensitivity 90 CL
Krasnoyarsk, Russia 1.6 115/1000 600/600 0.03
Kashiwazaki, Japan 24 300/1300 150/250 0.02
Double Chooz, France 8.4 150/1050 30/300 0.03
Diablo Canyon, CA 6.7 400/1700 50/700 0.01
Angra, Brazil 5.9 500/1350 50/500 0.02
Braidwood, IL 7.2 200/1500 450/450 0.01
Daya Bay, China 11.5 250/2100 250/1100 0.01
Stop
On Hold
Go
Many Sites have been investigated as potential
hosts to a reactor neutrino experiment. This is
appropriate since getting the cooperation of the
reactor company and host country is a main
challenge.
See Talks this afternoon by G. Mention and M.
Goodman
26
Proposed Sites Around the World
Site Power (GWthermal) Baseline Near/Far (m) Shielding Near/Far (mwe) Sensitivity 90 CL
Krasnoyarsk, Russia 1.6 115/1000 600/600 0.03
Kashiwazaki, Japan 24 300/1300 150/250 0.02
Double Chooz, France 8.4 150/1050 30/300 0.03
Diablo Canyon, CA 6.7 400/1700 50/700 0.01
Angra, Brazil 5.9 500/1350 50/500 0.02
Braidwood, IL 7.2 200/1500 450/450 0.01
Daya Bay, China 11.5 250/2100 250/1100 0.01
Kashiwazaki, Japan 24 300/1300 150/250 0.02
Status
27
Kashiwazaki-Kariwa, Japan (
)
Minakata, Sugiyama, Yasuda, Inoue, and Suekane
hep-ph/0211111

• 7 Reactors, 24 GWth
• Three 8.5 ton detectors
• Two near detectors at baselines of 400 m
• One far detector at 1.3 1.8 km
• 21 different baselines but not a problem
• Sensitivity of sin22?130.025 in 2 years
• Fast! Reaches systematics limit quickly
• Currently working its way through the
• Japanese system
• (2 yr RD budget approved)

far
near
near
28
Proposed Sites Around the World
Site Power (GWthermal) Baseline Near/Far (m) Shielding Near/Far (mwe) Sensitivity 90 CL
Krasnoyarsk, Russia 1.6 115/1000 600/600 0.03
Kashiwazaki, Japan 24 300/1300 150/250 0.02
Double Chooz, France 8.4 150/1050 30/300 0.03
Diablo Canyon, CA 6.7 400/1700 50/700 0.01
Angra, Brazil 5.9 500/1350 50/500 0.02
Braidwood, IL 7.2 200/1500 450/450 0.01
Daya Bay, China 11.5 250/2100 250/1100 0.01
Double Chooz, France 8.4 150/1050 30/300 0.03
Status
29
Double CHOOZ , France

• Use old far detector hall at 1050 meters
• Near detector at 125-250 meters (50 mwe)
• 11 ton Gd loaded detectors.
• Sensitivity of sin22?130.03 in 3 years
• Fast and Inexpensive
• Has scientific approval
• Released an LOI (hep-ex/0405032)

30
Double CHOOZ Experiment

• Cost estimate
• Detectors 7M
• Civil Construction 10M
• Schedule
• Now Securing approvals from agencies and
  company
• Feb. 05 If approved, complete design and put out
  bids
• May 07 Complete far detector
• Early 08 Complete near detector

• Experimental Goals
• Statistical error 0.4
• Background error 1
• Relative detector error 0.6
• Advantages
• Infrastructure exists for far site
• Reactor company would build near site if approved
• Disadvantages
• Shallow near detector site (50-100 mwe)
• Deadtime 500 ms / muon? 30 - 60 Deadtime
• Far baseline only 1km which limits sensitivity
  especially for low Dm2

31
Proposed Sites Around the World
Site Power (GWthermal) Baseline Near/Far (m) Shielding Near/Far (mwe) Sensitivity 90 CL
Krasnoyarsk, Russia 1.6 115/1000 600/600 0.03
Kashiwazaki, Japan 24 300/1300 150/250 0.02
Double Chooz, France 8.4 150/1050 30/300 0.03
Diablo Canyon, CA 6.7 400/1700 50/700 0.01
Angra, Brazil 5.9 500/1350 50/500 0.02
Braidwood, IL 7.2 200/1700 450/450 0.01
Daya Bay, China 11.5 250/2100 250/1100 0.01
Braidwood, IL 7.2 200/1500 450/450 0.01
Status
32
Braidwood, Illinois

• Four 65 ton detectors
• Near detectors at 200 meters 450 mwe
• Two far detectors located at 1500 meters, 450
  mwe
• Sensitivity of sin22?130.01 in 3 years
• High level of cooperation with utility

Braidwood
33
Braidwood Experiment

• Cost estimate
• Civil 35M
• Detectors 15M
• Schedule
• 2005 RD proposal submission.
• 2006 Full proposal submission
• 2007 Project approval start const.
• 2009 Start data collection

• Experimental Goals
• Statistical error 0.15
• Background error 0.4
• Relative detector error 0.6
• Advantages
• Deep near site allows other reactor physics
  measurements
• Favorable geology and low bkgnd
• Redundancy and cross checks
• Both rate and spectral distortion analysis
• Surface transport of detectors for cross
  calibration
• Multiple near and far detectors
• Disadvantages
• Infrastructure costs high due to new undeveloped
  site

Braidwood3 yrs
34
Proposed Sites Around the World
Site Power (GWthermal) Baseline Near/Far (m) Shielding Near/Far (mwe) Sensitivity 90 CL
Krasnoyarsk, Russia 1.6 115/1000 600/600 0.03
Kashiwazaki, Japan 24 300/1300 150/250 0.02
Double Chooz, France 8.4 150/1050 30/300 0.03
Diablo Canyon, CA 6.7 400/1700 50/700 0.01
Angra, Brazil 5.9 500/1350 50/500 0.02
Braidwood, IL 7.2 200/1500 450/450 0.01
Daya Bay, China 11.5 250/2100 250/1100 0.01
Daya Bay, China 11.5 250/2100 250/1100 0.01
Status
35
Daya Bay, China

• 4 Reactors, 11.5 GWth
• Several 8 ton detectors
• Near detectors at baseline of 300 and 400
  meters, 200 to 250 mwe
• Far detectors at baselines of 1800 and 2400
  meters, 1100 mwe
• Sensitivity of sin22?130.01 in 3 years

• Utility/government approval is likely
• China would support civil construction, but
  foreign support is needed for detectors

36
Daya Bay Experiment

• Cost estimate
• 25M
• Schedule
• 2005 RD, engineering design, and secure funding
• 2006-2007 Construction
• 2008 Start data collection

• Experimental Goals
• Statistical error 0.2
• Background error 0.3
• Relative detector error 0.4
• Advantages
• Horizontal access tunnel approach
• Large overburden reduces bkgnd
• Flexibility to change baseline
• Easy to service detector
• Cross calibration by moving detectors to near
  site
• Disadvantages
• Uncertainties associated with the Chinese
  approval system

37
Reactor Sensitivity Studies(Comparing and
Combining with Offaxis Measurements)
(K. McConnel and M. Shaevitz hep-ex/0409028)

• Try to do estimates including all effects
• ?m232 2.5?10-3 eV2 and allowed to vary with ?
  0.1?10-3 eV2
• Include both mass hierarchies ?m232 gt 0 and
  ?m232 lt 0
• sin22?23 0.95 and allowed to vary with ?
  0.01
• Include ambiguity of ?23 lt 45? or gt 45?
• ?m122 and ?12 fixed at current values
• dCP allowed to vary between 0? to 360?

• Reactor experimental inputs
• Small scale d(sin22q13) 0.03
• Medium scale d(sin22q13) 0.01
• Large scale d(sin22q13) 0.005(90 CL upper
  limit sensitivities at ?m2 2.5?10-3 eV2)

• Offaxis experimental inputs
• JPARC to SuperK (T2K) exp.
• Rates for n and?n from LOI
• with / without upgrade x5 rate
• Offaxis NuMI (Nova) exp.
• Rates for n and?n new LOI appendix
• with / without proton driver upgrade (x5 rate)

38
Comparisons of Reactor and Off-axis Sensitivities
large medium small reactor
5 beamrate
90 CL upper limits for underlying null sin22?13
( 0) A medium scale reactor experiment, gives
a more stringent limit on sin22?13 in several
regions than off axis, even with proton driver
like statistics (5 beam rate). After a medium
scale reactor limit, only a small window of
opportunity exists for an observation of ?µ to ?e
with an off-axis.
Green Offaxis exp. (5yr ? only)Blue Combined
Medium Reactor plus OffaxisWhite Offaxis Only
(x5 rate)
combinewith med.reactor
combinewith med.reactor
5 yrs ?-only
39
Comparison of Reactor to Off-axis Signal Meas.
Chooz-like, small scale Braidwood-like medium
scale
90 CL regions for sin22?13 0.05, dCP0 and
?m2 2.510-3 eV2 A medium scale reactor
experiment makes a significantly better
measurement of sin2?13 than the offaxis
experiments.
Green Offaxis exp. OnlyBlue Combined Medium
Reactor plus OffaxisRed Combined Small Reactor
plus Offaxis
5 yrs ?-only
40
Resolving the ?23 Degeneracy
Example sin22?23 0.95 ? ?23 38? or ?23
90?-38 ? 52?
(?23TRUE 38?)

• For sin2?13 0.05, combining with off-axis ?
  and?? running
• Medium scale reactor breaks the ?23 degeneracy.
• Small reactor is not sufficient to break the
  degeneracy.

(3 yrs ? 3 yrs?? )
41
Regions where ?23 Degeneracy Resolved at 2?(Is
?23 lt 450 or gt 450 ?)
3 yrs ? 3 yrs??
3 yrs ? 3 yrs??
42
Constraining the CP Violation Parameter, ?CPand
Determining the Mass Hierarchy

• Oscillation probability vs dCP for offaxis
  exp(?m2 2.5x10-3 eV2 , sin22?13 0.05)

• Use Medium Reactor (d(sin22q13) ?0.01) can
  predict the neutrino prob.

P(????e)
sin22?130.1
Neutrino, normal hierarchy
Neutrino, inverted hierarchy
?CP
43
Reactor Contribution to CP Violation

• For large sin2?13, can restrict ?CP with reactor
  and ?-only offaxis data.
• For smaller sin2?13, need offaxis?? data to
  restrict ?CP.
• In all cases, precision on determining sin2?13
  comes from reactor data.

(3 yrs)
3 yrs ? 3 yrs??
5 yrs ?-only
44
CP Violation 3? Discovery Regions
Reactor measurement does not contribute much to
measuring CP violation But a null reactor
measurement, even at the Double Chooz
sensitivity, can mostly rule out accessible
regions for T2K / Nova.
with proton driver
with x5 upgrade
To the right of the curve, ?CP0 or p is excluded
by at least three standard deviations
?m2 2.510-3 eV2 sin22?23 0.95 ? 0.01
with x5 upgrades
3 yrs ? 3 yrs??
45
Determining Reach in Mass Hierarchy
To the right of the curve, mass hierarchy is
resolved by at least two sigma
Reactor measurement does not contribute much to
resolving the mass hierarchy Again, a null
reactor measurement, even at the Double Chooz
sensitivity, can mostly rule out accessible
regions for T2K and Nova.
?m2 2.510-3 eV2 sin22?23 0.95 ? 0.01
with proton driver
with x5 upgrades
3 yrs ? 3 yrs??
46
Far Future ? Reactor plus Hyper-K Nova(x10)

• Hyper-K alone makes poor determination of CP
  violation phase, ?CP
• ?23 degeneracy limits accuracy
• Mass hierarchy introduces wide range
• ? Medium scale reactor data can lift ?23
  degeneracy (black dashed curves)
• ? Combination with Nova can determine
  hierarchy
• Bottom Line Combination of reactor plus Hyper-K
  plus Nova can make precise determination(black
  dashed curves)

47
Conclusions

• A reactor experiment is the prime and only
  unambiguous measurement of q13
• q13 is a important physics parameter
• Needed to constrain the models of lepton mixing
  matrix
• If very small, probably indicates a new symmetry
• q13 is key for planning future long-baseline
  experiments to measure CP violation and the mass
  hierarchy
• If sin22q13 is gt 0.02, T2K and Nova make a nice
  program
• If sin22q13 is lt 0.01, need other techniques to
  access the physics (1st,2nd max. measurements
  Superbeam exps, Neutrino Factory.)
• Reactor measurements are important for sorting
  out the q23 ambiguity (q23 vs 900- q23)
• Again this is an important, fundamental physics
  parameter (like q13)
• May be important for CP violation and mass
  hierarchy measurements
• Reactor experiments are being pursued at many
  sites and it is likely that several will be
  approved and go forward.