Current* Neutrino Oscillation Experiments

PPAP MeetingBirmingham15 July 2009

Elisabeth FalkUniversity of Sussex

Def. current: running or under construction

E. Falk, U. of Sussex 2

Outline

• Neutrino-oscillation experiments:what to measure, and how

• Accelerator-based experiments• Reactor experiments• Conclusions and outlook

15/7/09

E. Falk, U. of Sussex 3

Neutrino Mixing

15/7/09

3

2

1

321

321

321

UUU

UUU

UUU eeee

diagMaj

2323

2323

1313

1313

1212

1212

MNSP

cossin0

sincos0

001

cos0sin

010

sin0cos

100

0cossin

0sincos

CP

CP

U

e

e

Ui

i

MNSPU

Solar n + KamLAND: q12 = (34 ± 3)o

Dm212 = 7.6 x 10-5 eV2

Atmospheric n + MINOS, K2K: q23 = (45 ± 7)o

Dm232 = 2.5 x 10-3 eV2

Unknown CP-violating phase

Reactor n (CHOOZ):sin2 2q13 < 0.11 @ 90% CL(or q13 <~ 10o)Dm2

31 ~ Dm232

Dm212

n2

n1

n3

Dm232

nent n

m

What to measure, and how

E. Falk, U. of Sussex 4

Accelerator-Based vs. Reactor Experiments

Reactor experiments:q13 only:• Look for disappearance (ne nn e) as a fnc of

L and E• Near detector to measure unoscillated flux• P (ne ne) independent of d; matter effects

small

15/7/09

ND FD

LBL accelerator experiments:sin2 q12, Dm2

32:• Look for disappearance (n m nn m) as a fnc of L

and E• Near detector to measure unoscillated spectrumq13, δCP, sign(Dm2

32):• Look for appearance (n m nn e) in nm beam vs. L

and E• Near detector to measure background nes (beam

+ mis-id)• P (nm ne) = f (d, sign(Dm31

2))Accelerator-based and reactor experiments complementary

MINOS, T2K, NOnA Double Chooz, Daya Bay, RENO

What to measure, and how

E. Falk, U. of Sussex 5

Experimental Approach• (nm) disappearance measurements:

– Measure unoscillated nm spectrum at Near Detector• Extrapolate using MC

– Compare to measured spectrum at Far Detector

15/7/09

spectrum ratio

Monte Carlo

Unoscillated

Oscillated

Monte Carlo

nm spectrum

What to measure, and how

• ne appearance measurements:– Measure ne spectrum at Far Detector– Near Detector to measure

background nes (beam + mis-id)• Predict Far Detector background

using MC

~ sin2 2q23

~ Dm232

E. Falk, U. of Sussex 6

Neutrino Beam

• Protons strike graphite target• Magnetic horns focus secondary

p/K– Decay of p/K produces

(mostly) muon neutrinos

15/7/09

Accelerator-based experiments

NuMI

NuMI

E. Falk, U. of Sussex 7

Players: MINOS, T2K, NOnA

MINOS:• NuMI (Fermilab) to Soudan

mine, MN– 735 km baseline

• Magnetised sci/steel calorimeters

• Taken beam data since 2005– Two more years of data-taking

• Large UK involvement– All ana/working groups have

UK co-convener(s)– Made significant h/w & DAQ

contributions15/7/09

Accelerator-based experiments

T2K:• J-PARC (Tokai) to Kamioka

mine– 295 km baseline, off-axis– New beam

• FD: Super-K (water Cerenkov)

• Start data-taking in 2010• Large UK involvement

– ND280 off-axis detector(P0D, ECAL, DAQ, ….)

NOnA:• NuMI (Fermilab) to Ash

River, MN– 810 km baseline, off

axis– Upgrade beam power

• “Totally active” tracking liquid sci calorimeters

• Start data-taking in 2013• No UK involvement

E. Falk, U. of Sussex 8

MINOSMain physics results to date:• nm disappearance:

|Δm232|=(2.43 ± 0.13) x 10-3 eV2

at 68% C.L.

sin2(2θ23) > 0.90 at 90% C.L.– Most precise result for |Δm2

32| to date

• ne appearance:– 1.5s excess (within realms statistical

fluctuation):– Normal hierarchy (dCP = 0):

sin2(2θ13) < 0.29 (90% C.L.)

– Inverted hierarchy (dCP = 0):sin2(2θ13) < 0.42 (90% C.L.)

• nm disappearance:– Study 7% nm component – See 1.9s fewer events than CPT-conserving

• Neutral-current interaction rate:– Limit on sterile neutrinos:

fs < 0.68 (90% C.L.)15/7/09

Accelerator-based experiments nm disappearance

ne appearance

E. Falk, U. of Sussexß 9

MINOS

Future plans:• Update all analyses with more

than double the data set– Data in the can

(3.2 7E20 POT)• Muon antineutrino running

– Switch beam magnetic horns to focus p-

• Start in September this year

– Make the first direct measurement• Reduce uncertainty on Dm2

32 by an order of magnitude

• nm mode for 2E20 POT, or until July 2010

15/7/09

Accelerator-based experiments

E. Falk, U. of Sussex 10

T2K

• Physics programme:– Phase I:

• Discovery/measurement of q13

• D(sin2 2q23) ~ 0.01 (1%)D|Dm2

32| < 1 x 10-4 eV (a few %) (90% CL)

– Potential phase 2:• Search for CP violation

• Milestones:– ND suite operational by end of 2009– Beam power ramped over next few

years(2009: 30 kW; 2010: 100 kW; 2011: 300 kW; + “a few years”: 750 kW

– Phase 1: total of 5E21 POT

15/7/09

Accelerator-based experiments

E. Falk, U. of Sussex 11

NOnA

• Significant injection of funds via US stimulus package revived NOnA

• Physics programme:– Optimised for ne appearance

• An order of magnitude more sensitive than MINOS

– Improve sin2 2q23 and |Dm232| by an order of

magnitude (compared to MINOS)– Mass hierarchy (sign (Dm2

32))• Long baseline: only experiment sensitive to

this

• Milestones:– 1 May 2009: FD groundbreaking– Autumn 2011: Start of beam upgrades

(400 kW 700 kW)– 2013: Data-taking with full FD

15/7/09

Accelerator-based experiments

E. Falk, U. of Sussex 12

Dominant source of systematic error in CHOOZ:Reactor neutrino spectrum~ Cancels out with two detectors

Measuring q13: ne Disappearance

15/7/09

Reactor experiments

Nuclear power station Far Detectord = 1 - 1.8 km

ne ne

Present limit from CHOOZ:sin2(213) < 0.15 (90% C.L.) atDm2

31 = 2.5 x 10-3 eV2

Near Detectord = 300 - 400 m

(nm,t

)

E. Falk, U. of Sussex 13

The Detectors

15/7/09

Prompt e+ signalEprompt = En – (Mn – Mp) + me

Delayed n capture on Gd(t ~ 30 ms) with emissionof g cascade ( E ~ 8 MeV)

Outer Veto:Plastic scintillator panels

n Target:10 m3 liquid scintillatordoped with 0.1% Gd

g Catcher:23 m3 liquid scintillator

Buffer:114 m3 mineral oil with~400 PMTs

Inner Veto:90 m3 liquid scintillatorwith 80 PMTs

Shielding:15 cm steel

Reactor experiments

Example: Double Chooz

Inverse Beta Decay

E. Falk, U. of Sussex 14

Players: Double Chooz, RENO, Daya Bay

15/7/09

Reactor experiments

Double Chooz, FranceExpected sin22n13~0.03

85 ton-GWth

Small UK interest

(Sussex, no longer funded)

Daya Bay, ChinaExpected sin22n13~0.01

1400 ton-GWth

RENO, KoreaExpected sin22n13~0.03

250 ton-GWth

Main differences:• Reactor power/no of cores• Configuration cores vs. detectors; no. of detectors• Detector target mass

E. Falk, U. of Sussex 15

Status and Expected Milestones*

15/7/09

Reactor experiments

2009

2010

2011

2012

2013

2014

2009

2010

2011

2012

2013

2014

2009

2010

2011

2012

2013

2014

Double Chooz RENO Daya Bay

* DC schedule includes my estimated delays – not official, so don’t quote it! RENO, DB official schedules, but at least DB will likely be delayed by a few months

ND and FD readyfor data-taking

N and F tunnelscompleted

ND and FD commissioning

sin2 2q13 ~ 0.03 ???

First detector complete; start dry run

Near Hall ready for data-taking

Far Hall ready for data-taking;Ling Ao Hall ready abit earlier

Near Hall occupancy

sin2 2q13 ~ 0.01

FD ready for data-taking

sin2 2q13 ~ 0.06ND ready fordata-taking

sin2 2q13 ~ 0.03

ND hall + tunnelconstruction begins

E. Falk, U. of Sussex 16

Comparison of q13 Sensitivities

15/7/09

M. Mezzetto, Venice, March 2009

E. Falk, U. of Sussex 1715/7/09

Comparison of q13 SensitivitiesM. Mezzetto, Venice, March 2009

E. Falk, U. of Sussex 18

Conclusions and Outlook

• Past and present experiments have pinned down q13, Dm2

12, q23, Dm232

• MINOS recently announced 1.5s result for q13

• T2K and reactor experiments come on-line within ~a year from now

• Next 5 years will see sensitivity to sin2 2q13 to ~0.01, plus order of magnitude improvement on q23 and Dm2

32

• Need combination of different baselines + reactor to resolve q13, δCP, sign (Dm2

32)

• Would be difficult to search for δCP if sin2 2q13 < 0.01

15/7/09

E. Falk, U. Sussex 19

Reactor Experiment Howto: Improve on Chooz

• Statistics– More powerful reactor (multi-core)– Larger detection volume– Longer exposure

• Experimental error: n flux and cross-section uncertainty– Multi-detector– Identical detectors

• Reduce inter-detector systematics (normalisation, calibration...)

• Background– Improve detector design– Increase overburden– Improve knowledge of background by direct measurement

Larger S/B

CHOOZ : Rosc = 1.01 ± 2.8% (stat) ± 2.7% (syst)

Reach ~1% precision

E. Falk, U. Sussex 20

Double Chooz Systematic Uncertainties

n energy, Dt, (distance e+ - n)

E. Falk, U. of Sussex 21

Baselines and Reactor Power

15/7/09

Reactor experiments

Double Chooz RENODaya Bay

P = 16.1 GWth/6 coresL ~ 1.4 km

P = 8.2 GWth/2 coresL = 1.05 km

P = 11.6 GWth/4 cores ~2011: 17.4 GWth/6 coresL ~ 1.8 km

Daya Bay and Ling Ao Nuclear Power Plant

Daya Bay NPP 2.9GWn2 LingAo NPP 2.9GWn2

LingAo II NPP 2.9GWn2Under construction (2010)

E. Falk, U. of Sussex 23

Daya Bay Run Plan

• Istall first two detectors in DB Near Hall within a year from now

• Take data (engineering run) while Ling Ao and Far Halls completed

• Additional detectors deployed in pairs, one in Ling Ao and one in Far Hall, as they become ready – except last pair

• One detector to be moved from Near to Far Hall• Last pair deployed in Near + Far Halls• Run for three years• Publish

15/7/09

Reactor experiments

DYB (40 t)

LA (40 t)

Far (80 t)

Daya Bay: Redundancy

• Measuring sin22n13 to 0.01 need to control systematic errors very well.• We believe that the relative (near/far) detector systematic error could be

lowered to 0.38%, with near/far cancellation and improved detector design.

• Side-by-side calibration: Event rates and spectra in two detectors at the same near site can be compared How IDENTICAL our detectors are?

• Detector swapping: Daya Bay antineutrino detectors are designed to be MOVABLE. All detectors are assembled and filled with liquids at the same place. Detectors at the near sites and the far site can be swapped, although not necessary to reach our designed sensitivity, to cross check the sensitivity and further reduce the systematic errors.

DYB (40 t)

LA (40 t)

Far (80 t)

E. Falk, U. of Sussex 2515/7/09

Reactor experiments

2010 2011 201420132012 2015

sin2 2

q13

lim

it

Far Detectoronly

Far+Near Detectorsssys= 2.6%

ssys= 0.6%

Comparison of q13 SensitivitiesDaya Bay

Double Chooz

E. Falk, U. of Sussex 26

Comparison of q13 Sensitivities

15/7/09

M. Mezzetto, Venice March 2009

Assumptions:

• DC to start end of 2009; ND 1.5 yrs later• More likely 2nd quarter of

2010 (my estimate)

• DB to start mid 2011• From C. White: Schedule

slipping by a few months

• T2K: 0.1 MW in 2010; 0.45 MW in 2011; 0.75 MW thereafter• From D. Wark:

0.1 MW in 2010; 0.3 MW in 2011; ramp to 0.75 MW over a few years

• MINOS, OPERA: sensitivity as per proposal, scaled by POT

E. Falk, U. of Sussex 27

Overview• Long baseline

Fermilab 735 km Soudan• Near detector at Fermilab

– Measure beam composition, energy spectrum

• Far detector in Minnesota– Search for and study oscillations

• Test the nm nx oscillation hypothesis– Measure precisely |Dm2

32| and sin2 (2q23)

• Search for sub-dominant nm ne oscillations– Sensitive to q13

• Other MINOS physics:– Search for sterile neutrinos, CPT/Lorentz

violation– Compare nm, nm oscillations– Studies of cosmic rays and atmospheric

neutrinos– Neutrino interaction studies in the

Near Detector15/7/09

MINOS

E. Falk, U. of Sussex 28

Muon-Neutrino Disappearance Analysis

• Strong energy-dependent spectrum distortion

• Spectrum fit with two-flavour oscillation hypothesis

• Fit constrained to physical region + includes three largest systematics

• Results:

|Δm232|=(2.43 ± 0.13) x 10-3 eV2

at 68% C.L.

sin2(2θ23) > 0.90

at 90% C.L.

15/7/09

PRL 101 131802 (2008)

Most precise measurement of |nm232| to date

MINOS

E. Falk, U. of Sussex 29

Antimuon-Neutrino Disappearance Analysis

• Measure |Δm232|, sin2(2θ23)

• Test CPT• FNAL Wine & Cheese three

weeks ago• nm CC events are 7% of beam

– Mis-ID muon and NC backgrounds relatively larger

– As CC analysis, but with extra cuts• Track-length likelihood, charge-sign

significance

15/7/09

MINOS

16th-22nd July 2009 30Justin Evans

The FutureAll these analyses have used 3x1020 of the 7x1020 protons-on-target that have been recorded

– As of this June’s summer shutdown

Over the next year, the analyses will be updated with the increased dataset– Using the blind analysis policy on the new data

Graphs below show the θ13 sensitivity for 7x1020 PoT

If the excess persists If the excess goes away

E. Falk, U. of Sussex 31

Electron-Neutrino Appearance Analysis

• Sub-dominant neutrino oscillations

P(νμ→νe) ≈ sin2θ23 sin22θ13 sin2(1.27Δm231L/E)

• Also CPv and matter effects: not shown here but included in fit

• ANN algorithm to select ne events

• Background measured in ND– NC events, high-y nm CC, beam ne, oscillated nt

at FD– Data-driven technique: compare horn

on/off data

• Predict FD background from measured ND background + MC

15/7/09

nnCC Event

NC Event

neCC Event

MINOS

E. Falk, U. of Sussex 32

Electron-Neutrino Appearance Analysis

• Far Detector spectrum obtained after blind analysis

• Expected background:27 ± 5(stat) ± 2(syst)

• Contours from Feldman-Cousins method– Fit to number of events

• Observed events:37

• Results:– Normal hierarchy (dCP = 0):

sin2(2θ13) < 0.29 (90% C.L.)

– Inverted hierarchy (dCP = 0):sin2(2θ13) < 0.42 (90% C.L.)

• 1.5s excess– Well within realms of statistical

fluctuation15/7/09

MINOS

E. Falk, U. of Sussex 33

Neutral Current Analysis

• General NC analysis overview:– All active neutrino flavours

participate in NC interaction– Mixing to a sterile ν will cause a

deficit of NC events in Far Detector– Assume one sterile neutrino and

that mixing between νμ, νs and ντ occurs at a single Δm2

• Survival and sterile oscillation probabilities become:

(αμ,s = mixing fractions)

15/7/09

Simultaneous fit to CC and NC energy spectra yields the fraction of νμ that oscillate to νs:

MINOS

E. Falk, U. of Sussex 34

Future Plans

• Update all analyses with more than double the data set

• Muon antineutrino possibilities– Switch beam magnetic

horns to focus p-

– MINOS can make the first direct measurement• Reduce uncertainty on

Dm232 by an order of

magnitude

15/7/09

MINOS

E. Falk, U. of Sussex 35

Main Measurements: sin2q13

• Use CCQE: ne + n e- + p• Main backgrounds:

– Beam ne contamination– NC p0 events

15/7/09

T2K

Simulated SK spectrum

E. Falk, U. of Sussex 36

Main Measurements: sin2q23, Dm232

• Use CC quasi-elastic events:nm + n m- + p– Background from non-CCQE

interactions

15/7/09

T2K

Dm322

sin22q23

nm disappearance

• Precision prospect: D (sin22q23) ~ 0.01 (1%) D (Dm2

32) < 1 x 10-4 eV2 (a few %) at 90% C.L.

• @ Dm2 ~ 2.5 x 10-3 eV2

E. Falk, U. of Sussex 37

J-PARC Accelerator Complex

15/7/09

T2K

Commissioned,in use for material & life science

Protons to 30 GeV on 23/12/08Commissioning is on time

Commissioned,in use for material & life science

n beam-line componentsCommissioning in progress

Neutrino Beam Line

E. Falk, U. of Sussex 38

Near Detectors at 280 m• On- and off-axis detectors

280 m downstream of target– Understand the neutrino

beam before oscillations occur

• T2K timeline– Apr 09: Beam-line

commissioning 10 days– Summer-autumn 09: Install

INGRID & ND280– Oct 09: Restart beam

commissioning– Dec 09-June 10: Physics run

• Aim to improve CHOOZ limit

15/7/09

On-axis: INGRIDScintillator/ironmodules

ND 280: Off-axis detectorsinside UA1 magnet

T2K

E. Falk, U. of Sussex 39

Beam Power Projections

15/7/09

April 1, 2009

T2K

E. Falk, U. of Sussex 40

Discovery Potential

15/7/09

T2K