M.NakahataKamioka observatory, ICRR,IPMU, Univ. of Tokyo

Professor Yoji Totsuka(1942-2008)

Kamiokande spokesman: 1987 April ---- end

Super-Kamiokande spokesman: beginning ---- 2002

Kamiokande detector (1983 – 1996)

16 m high, 15.6 m diameterInner counter: 948 20-inch PMTs

Anti-counter123 20-inch PMTs

neutrino

e

3000 ton water tank

Photo-sensitive: 2140 t

Fiducial volume: 680 t

(for solar neutrino)

Photocoverage: 20 %

50000 t water tank (42m high, 40m diameter)

32000 t photo-sensitive volume

22000 t fiducial volume

11146 20-inch PMTs

Photocoverage: 40%

1000m underground in Kamioka mine

X 30 fiducial volume than Kamiokande

Super-Kamiokande detector (1996 – )

History of Super-Kamiokande detector

11146 ID PMTs(40% coverage)

5182 ID PMTs(19% coverage)

11129 ID PMTs(40% coverage)

EnergyThreshold(total electron energy)

1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009

SK-I SK-II SK-III SK-IV

Acrylic (front)+ FRP (back)

ElectronicsUpgrade

SK-I SK-II SK-III SK-IV

5.0 MeV 7.0 MeV 4.5 MeVwork in progress

< 4.0 MeVtarget

Original purpose of KamiokandeOriginal purpose of KamiokandeSearch for proton decay

p→e+0 Monte Carlo simulation

e+

0→

High resolution detector for measuring the branching ratio of proton decay.

It should be useful to pin down the true GUT model.

3500

Low energy neutrino detectionLow energy neutrino detectionIt was found that the large photo-collection efficiency is useful also for detecting low energy neutrino.

Reconstructed energy = 19.8 MeV

An event at Kamiokande

Advantage of Kamiokande as a “Advantage of Kamiokande as a “telescopetelescope””

DirectionalityImaging Cherenkov detector has excellent directionality.

Energy informationThe number of observed Cherenkov photon is proportional to energy of particle.

Real time detectionReal time counter experiment.

neutrino

electron

+ e + e

Advantage of Kamiokande as a “Advantage of Kamiokande as a “telescopetelescope””

Another advantage of Kamiokande: Particle identification(PID)

electron

muon

Mis-identification is less than 1%.

Evis=540 - 1200 MeV

Evis=270 - 540 MeV

Evis=130 - 270 MeV

Evis=80 - 130 MeV

Evis=30 - 80 MeV

PID was very important for the atmospheric neutrino analysis.

First solar neutrino plot at Kamiokande

Jan,1987 --- May, 1988 (450 days)

Observed number of solar neutrino events was ~50.

Confirmed the “solar neutrino problem”.

Solar model prediction

K.S.Hirata et al., Phys. Rev. Lett. 63(1989) 16

Solar neutrinos ( Ｓｕｐｅｒ－Ｋａｍｉｏｋａｎｄｅ )May 31, 1996 – July 13, 2001 (1496 days )

e-

sun

COSsun

Ee = 5.0 - 20 MeV

22400 solar events

(14.5 events/day)

8B flux : 2.35 0.02 0.08 [x 106 /cm2/sec]

(BP2004: 5.79 x 106 /cm2/sec)= 0.406 +0.014-0.013

0.004

DataSSM(BP2004)

Combined analysis of SK, SNO CC and NCCombined analysis of SK, SNO CC and NC

SK ESSNO ES

SNO NC

SNO CC

SSM prediction (1)

8B solar neutrino e flux and (+) flux

Evidence for neutrino oscillation

Solar neutrino energy spectrumKamiokande II and III (2079 days )

Based on ~600 solar events

Super-Kamiokande (1496 days )

Based on ~22400 solar events

5

Excluded region by energy spectrum and day/nightSuper-Kamiokande 1496 days

S.Fukuda et al., Phys. Lett. B 539 (2002) 179

Solar Neutrino future prospects in SK

pp7Be

8B

Aim to reduce background in SK

P(

e

e)

Vacuum osc. dominant

Transition from vacuum to matter osc.Upturn is expected in 8B spectrum.

matter dominant

e survival probability(at best fit parameter)

~70% reduction below 5.5MeV and lower threshold to 4MeV

Expected spectrum distortion with 5 years low BG SK data

,IV

Supernova at LMC (February 23, 1987)Supernova at LMC (February 23, 1987)

BeforeAfter

TimeJT: 1987 Feb 23 16:35:35 (±1min)UT: 7:35:35

Background level

SN1987A signal by KamiokandeSN1987A signal by Kamiokande

sec

11 events in 13 sec.

It was when the Kamiokande detector was almost ready for solar neutrino detection.

Vis

ible

en

erg

y (M

eV

)

SN1987A: supernova at LMC(50kpc)Kamiokande-II IMB-3 BAKSAN

Kam-II (11 evts.)IMB-3 (8 evts.)Baksan (5 evts.)

Tot

al B

indi

ng E

nerg

y

95 % CLContours

Spectral e Temperature__

TheoryTheory

from G.Raffelt

Feb.23, 1987 at 7:35UT

24 events total

Super-K: Expected number of eventsSuper-K: Expected number of events

~7,300 e+p events~300 +e events~360 16O NC events ~100 16O CC events (with 5MeV thr.)

for 10 kpc supernova

Neutrino flux and energy spectrum from Livermore simulation (T.Totani, K.Sato, H.E.Dalhed and J.R.Wilson, ApJ.496,216(1998))

Super-K: Time variation measurement by e+pAssuming a supernova at 10kpc.

Time variation of event rate Time variation of mean energy

Enough statistics to discriminate models

ep e+n events give direct energy information (Ee = E – 1.3MeV).

Simulation of a SN at 10kpc

e+p

e+p

e+p e+p

+e +e

+e +e

Super-K: Expected angular distributionSuper-K: Expected angular distribution

Spectrum of +e events can be statistically extracted using the angular distributions.

Direction of supernova can be determined with an accuracy of ~5 degree.

Neutrino flux and spectrum from Livermore simulation

S.Ando, NNN05

Supernova Relic NeutrinosSupernova Relic NeutrinosS.Ando, Astrophys.J.607:20-31,2004.

10-6

10-5

10-4

10-3

10-2

10-1

11010 210 310 410 510 610 7

0 10 20 30 40 50 60 70 80Neutrino Energy (MeV)

Neut

rino

Flux

(/cm

2 /s

ec /M

eV)

Supernova Relic NeutrinosSupernova Relic Neutrinos

Constant SN rate (Totani et al., 1996)Totani et al., 1997Hartmann, Woosley, 1997Malaney, 1997Kaplinghat et al., 2000 Ando et al., 2005Lunardini, 2006Fukugita, Kawasaki, 2003(dashed)

Solar 8B e

Solar hep e

Expected number SRN events0.8 -5.0 events/year/22.5kton(10-30MeV)

(0.3 -1.9 events/year/22.5kton for 18-30MeV)

Large target mass like SK and high background reduction are necessary.

Large target mass like SK and high background reduction are necessary.

SRN predictions(e fluxes)

Reactor e

Atmospheric e

0

0.5

1

1.5

2

2.5

3

3.5

4

Constant SN rate

(Totani et al. 1996)

Totani et al. 1997

Malaney et al. 1997)

Hartmann et al. 1997)

Kaplinghat et al. 2004

Ando et al. 2005

Fukugita et al. 2003

Lunardini et al. 2006

SK-II limit = 3.68 /cm2/sec

SK-I limit = 1.25 /cm2/sec Combined limit = 1.08 /cm2/sec

preliminary(E>18MeV)

Super-K results so farFlux limit VS predicted flux

Energy spectrum of SK-IEnergy spectrum of SK-I and SK-IIand SK-II (>18MeV)(>18MeV)

Atmospheric → invisible → decay e

Atmospheric e

90% CL limit of SRN Total

background

Energy (MeV)Atmospheric e

Atmospheric → invisible → decay e

Spallation background

SK-I (1496days) SK-II(791 days)

Eve

nts

/4M

eV

Observed spectrum is consistent with estimated background.Search is limited by the invisible muon background.

e can be identified by delayed coincidence.

Neutron tagging in water Neutron tagging in water Cherenkov detectorCherenkov detector

e

e+

pn

Positron and gamma ray vertices are within ~50cm.

Gd

n+Gd →~8MeV T = ~30 sec

Neutron capture gamma

(J.Beacom and M.Vagins) Phys.Rev.Lett.93:171101,2004

Add 0.2% GdCl3 in water

Possibility of SRN detectionRelic model: S.Ando, K.Sato, and T.Totani, Astropart.Phys.18, 307(2003) with flux revise in NNN05.

If invisible muon background can be reduced by neutron tagging

Assuming invisible muon B.G. can be reduced by a factor of 5 by neutron tagging.

By 10 yrs SK data,Signal: 33, B.G. 27(Evis =10-30 MeV)

SK10 years (=67%)

Assuming 67% detection efficiency.

0123456789

10

10 15 20 25 30 35 40 45 50

relic+B.G.(inv.mu 1/5)

B.G. inv.mu(1/5)

atmsph.–

e

Visible energy (MeV)

even

ts/1

0yea

rs/2

MeV

We are studying feasibility of introducing gadolinium. (effect on water transparency, corrosion, cable connectors and etc.)

Atmospheric neutrino anomaly in KamiokandeAtmospheric neutrino anomaly in Kamiokande

Initial hint

→e decay ratio

Data from 1983 to1985Small →e decay ratio

e

Momentum of single ring events

Paper in 1988

-like/e-like ratio is 60% of expectation.

EXPERIMENTAL STUDY OF THE ATMOSPHERIC NEUTRINO FLUX.KAMIOKANDE-II Collaboration (K.S. Hirata et al.), Phys.Lett.B205:416,1988

Atomospheric anomaly in KamiokandeAtomospheric anomaly in KamiokandeZenith angle distribution of multi-GeV events (1994)

Y.Fukuda et al., Phys. Lett. B 335 (1994) 237.

downwardupward

Zenith Angle distribution of SK

SK-I dataMonte Carlo (no oscillations)Monte Carlo (best fit oscillations)

\\

cos θzenith cos θzenith

cos θzenith

Zenith Angle Analysis: SK-I + SK-II

Best fit:Δm2 = 2.1 x 10-3 eV2

sin2 2θ = 1.02χ2 = 830.1 / 745 d.o.f.

L/E Analysis: SK-I + SK-II

χ2 fit to 43 bins of log10(L/E) with 29 systematic error terms

DatasetsSK-I FC/PC μ-like: 1489 daysSK-II FC/PC μ-like: 799 days

Use only event categories with good L/E resolution:

Partially-contained muons Fully-contained muons

Compare against:Neutrino decoherence (5.0σ)Neutrino decay (4.1σ)

3 flavor analysis: SK-I + SK-IINormal Hierarchy Inverted Hierarchy

preliminary

Note: one mass scale dominance method(m212 is set to 0)

Full 3-flavor analysis is being prepared.

SK-IV electronics: New front-end electronics, QBEE

QTC TDC FPGA

Network Interface Card

PMTsignal

Ethernet Readout

60MHz ClockTDC Trigger

QTC-Based Electronics with Ethernet(QBEE)

24 channel input QTC (custom ASIC)

3 gain stages Wide dynamic range(>2000pC) factor 5 larger than old electronics

Pipe line processing multi-hit TDC (AMT3) FPGA

Ethernet Readout 60MHz common system clock Internal calibration pulser Low power consumption ( <

1W/ch )

Calibration Pulser

Difference in readout system

FormerElectronics

(ATM)

Readout (backplane, SCH, SMP)

Trigger (1.3sec x 3kHz)

HITSUMTrigger

logic

NewElectronics

(QBEE)

Readout (Ethernet)

Periodic trigger(17sec x 60kHz)

Clock

Hardware Triggerusing number of hit

(HITSUM)

1.3secevent window

Variableevent window

by software trigger

No hardware trigger. All hits are readout. Apply software trigger.No hardware trigger. All hits are readout. Apply software trigger.

12PMTsignals

permodule

24PMTsignals

permodule

Collect ALL hits every 17sec time window. The 60kHz clock synchronize time of hit information.

Former readout system

New readout system

Performance of new electronics for supernova burst

Distance to SN vs. number of events

Dead time free in the new system

Performance for high rate

100% efficiency up to 130kHz for each channel.

It corresponds to ~1000 x supernova at galactic center.(100 times better than previous system.)

# of

hits

burst hit rate (kHz) 130kHz

input

output

Dead time free even for a supernova as close as 0.3kpc

Previous system

Conclusion• Neutrino astronomy was born in Kamiokande. And it was evolved in

Super-Kamiokande.– KAM observed deficit of solar neutrinos, and SK contributed to the evidence for the

solar neutrino oscillation and parameter determination.– Neutrinos from SN1987A by KAM, and a large statistical observation of galactic

supernova is expected in SK.– Atmospheric neutrino anomaly in KAM, and evidence for atmospheric neutrino

oscillation in SK. Detailed analysis is going on in SK.– The flux upper limit of supernova relic neutrinos is close to the theoretical expectation.

SK is studying possibility of neutron tagging by gadolinium.

• New electronics and online system was installed in September 2008 at SK, and SK-IV is running.

• T2K will start soon (from April 2009).

• More physics outputs are expected at SK.