Reactor neutrino experimentsZeyuan Yu, IHEP, CAS

yuzy@ihep.ac.cn

June 25, SJTU

1950s 2000s 2010s 2020s

Nuclear reactor as antineutrino source

• Nuclear reactors produce pure

νe from beta decays of fission

daughters

• 6 νe per fission

• 2*1020 νe per second per GWth

• Commercial reactor: ~ 3 GWth

• Free, huge flux

• Research reactor: ~ MWth

2

Beta decay in 1920s

• Why the beta decay spectrum is

continuous?

• Break of energy conservation law?

3

ZX -> β + Z+1X

Pauli

• Why the beta decay spectrum is

continuous?

4

ZX -> β + Z+1X + ν

Fermi

• In 1932, Chadwick discovered neutron, then Fermi proposed to change

Pauli’s ‘neutron’ to ‘neutrino’ – a minor neutron

• In 1933, Fermi proposed β-decay results from some sort of interaction

between the nucleons, the electron and the neutrino

• This interaction is different from all other forces and will be called the weak

interaction

• But the paper was rejected by Nature

• “Because it contained speculations too

remote from reality to be of interest

to the reader”

5

How to detect neutrino

• ν + 37Cl 37Ar + β

• Used by Raymond Davis

6

• ν + p n + e+

• Used by Reines and Cowan

Since 1953

Hanford reactor

0.3m3 liquid scintillator

90 2” PMTs

Since 1948

CCl4 detector

BNL reactor

Neutrino discovery

• Reines and Cowan moved the experiment to Savannah River reactor plants

• In 1956, the neutrino was observed

• Nobel Prize of physics, 1995

7

Solar neutrino

• Raymond Davis moved the CCl4 detector to Homestake in 1960s

• ν + 37Cl 37Ar + β

• ν + 37Cl 37Ar + β

• The observation of solar neutrinos, Nobel Prize of 2002

8

What we learned from Reines and Cowan

• Detection principles

• Inverse beta decay in CdCl3 water solution coincidence of prompt and delayed signal

• Liquid scintillator + PMTs

• Underground

• Modern experiments are still quite similar, except

• Loading Gd into liquid scintillator

• Larger, better detector

• Deeper underground, better shielding

9

CHOOZ and Paolo Verde

10

• They were built around 1997 with ~ 1 km reactor to detector distance

• Aimed to search for the neutrino oscillation with Δm2 ~ 10-3 eV2

1998-1999, US

11.6 GWth

Segmented detector

12 ton 0.1% Gd-LS

Shallow overburden

32 mwe

1997-1998, France

8.5 GWth

300 mwe

5 ton 0.1% Gd-LS

Bad Gd-LS

R=1.012.8%(stat) 2.7%(syst), sin2213<0.17

KamLAND

11

2002-, Japan, 53 reactors, 80 GWth

1000 ton LS, 2700 mwe

Radioactivity fiducial cut, Energy threshold

Baseline 180 km

KamLAND

12

The first observation of reactor antineutrino disappearance

Confirmed antineutrino disappearance at 99.998% CL

Excluded neutrino decay at 99.7% CL

Excluded decoherence at 94% CL

R=0.6580.044(stat) 0.047(syst)

Neutrino oscillation @ 2003

13

1 2 3

1 2 3

1 2 3

1

2

3

e e ee U U U

U U U

U U U

1 13 13

13 13

23 23

23 23

2 12

12 12

1 0 0

0 0

0

0 0

0 c s

0 s c 0

c s 0

s c 0

0 0 1

c 0 s

0 0

s 0 c 1

i

iiU ee

e

23 ~ 45

Atmospheric

Accelerator

12 ~ 34

Solar

Reactor

0

13 = ?

Reactor

Accelerator

In a 3- framework

θ13 = ?

• Reactor neutrino experiments use ν disappearance

• Clean in physics, only related to 13

14

2

4 2 2

1

2

3 12

2

13 31

2

21

1 sin / 4

cos si

sin

n 2 sin / 4

2

e e

m

m

P L E

L E

“New generation” θ13 experiments

15

Parameter Error Near-far

Reaction cross section 1.9 % 0

Energy released per fission 0.6 % 0

Reactor power 0.7 % ~0.1%

Number of protons 0.8 % < 0.3%

Detection efficiency 1.5 % 0.2~0.6%

CHOOZ Combined 2.7 % < 0.6%

Major sources of uncertainties:

• Reactor related ~2%

• Detector related ~2%

• Background 1~3%

Lessons from past experience:

CHOOZ: Good Gd-LS

Palo Verde: Better shielding

KamLAND: No fiducial cut

Near-far relative measurement

Mikaelyan and Sinev, hep-ex/9908047

16

Angra, Brazil

Diablo Canyon, USA

Braidwood, USA

Double Chooz, France

Krasnoyarsk, Russia

KASKA, Japan

Daya Bay, China

RENO, Korea

8 proposals, most in 2003 (3 on-going)• Fundamental parameter• Gateway to -CPV and Mass Hierachy measurements• Less expensive

Reactor antineutrino detection

17

Capture on H

Capture on Gd

Daya Bay experiment

• 6 reactor cores, 17.4 GWth

• Relative measurement

– 2 near sites, 1 far site

• Multiple detector modules

• Good cosmic shielding

– 250 m.w.e @ near sites

– 860 m.w.e @ far site

18

Three on-going experiment @ 2009

19

ExperimentPower

(GW)

Detector(t)

Near/Far

Overburden (m.w.e.)

Near/Far

Sensitivity

(3y,90%CL)

Daya Bay 17.4 40 / 80 250 / 860 ~ 0.008

Double Chooz 8.5 8 / 8 120 / 300 ~ 0.03

RENO 16.5 16 / 16 120 / 450 ~ 0.02

Huber et al. JHEP 0911:044, 2009

Daya Bay

20

Three on-going experiment @ 2018

• Daya Bay: running to Dec. 2020, sin22θ13 precision better than 3%

• RENO: running to 2020

• Double Chooz: data taking stopped in Dec. 201721

3.4%

2.8%

Neutrino flux measurements

• The neutrino oscillation study is a near and far relative measurement

• The absolute neutrino flux can also be measured

22

Reactor neutrino predictions

• Summation method: 10% uncertainty• Sum over the fission products’ νe spectra

from the nuclear database

• 235U, 239Pu, 241Pu: conversion method, ~2.7% uncertainty• Convert ILL’s measured beta spectra to νe

ones with virtual beta-decay branches

• ILL + Vogel model since 1980s• Predicted flux was consistent with Bugey-3

and other short baseline experiments

• Huber + Mueller Model• In 2011, two conversion re-analyses increased

the predicted flux by ~5%

23

Flux measurements @ 2011

• Measured flux is 6% higher than the Huber-Mueller model prediction

• eV scale sterile neutrino?

• A lot of short baseline experiments were proposed

24G. Mention et al.

Phys.Rev. D83 (2011) 073006

fit with sterile 𝜈Δ𝑚2 ≈ 1 𝑒𝑉2

Daya Bay measurement

• The 6% flux deficit is confirmed

• 5 sigma discrepancies are found in the

neutrino spectra

25

1904.07812

1808.10836

Fuel evolution study

• With nuclear fuel burning, larger 239Pu fission fraction smaller

neutrino yield

26

Fuel evolution study

• With nuclear fuel burning, larger 239Pu fission fraction smaller

neutrino yield

27PRL118, 251801 (2017)

Fuel evolution study

• Combined fit for major fission isotopes 235U and 239Pu

• σ235 is 7.8% lower than Huber-Mueller model (2.7% meas. uncertainty)

• σ239 is consistent with the prediction (6% meas. uncertainty)

• 2.8σ disfavor equal deficit (H-M model & sterile hypothesis)

28PRL118, 251801 (2017)

Isotropic neutrino spectra

• Daya Bay extracted the neutrino

spectrum of 235U and 239Pu fissions

• The first 235U spectrum at commercial

reactors

• The first 239Pu spectrum measurement

29

1904.07812

30

31

PROSPECT STEREO

Best-fit value of reactor antineutrino anomaly is rejected at 99% level

32

NEOS PROSPECT STEREO

The global neutrino spectrum analysis is imminent.

Daya Bay and PROSPECT have started the joint analysis.

IAEA technical meeting

• Partrick Huber: the ~3% uncertainty of model prediction seems too aggressive

33

Neutrino oscillation @ 2012

34

1 2 3

1 2 3

1 2 3

1

2

3

e e ee U U U

U U U

U U U

1 13 13

13 13

23 23

23 23

2 12

12 12

1 0 0

0 0

0

0 0

0 c s

0 s c 0

c s 0

s c 0

0 0 1

c 0 s

0 0

s 0 c 1

i

iiU ee

e

23 ~ 45

Atmospheric

Accelerator

12 ~ 34

Solar

Reactor

0

13 ~ 8o

Reactor

Accelerator

In a 3- framework

JUNO: mass hierarchy

35

Yangjiang NPP

Taishan NPP

Daya Bay NPP

Huizhou

NPP

Lufeng

NPP

53 km

53 km

Hong Kong

Macau

Guang Zhou

Shen Zhen

Zhu Hai

2.5 h drive

Kaiping,Jiang Men city,Guangdong Province

Overburden ~ 700 m

by 2020: 26.6 GWDaya Bay ~60 km JUNO

JUNO

36

• 20 kton LS detector

• 3% energy resolution

• 700 m underground

• Rich physics possibilities• Reactor neutrino

for Mass hierarchy and precision measurement of oscillation parameters

• Supernovae neutrino

• Geoneutrino

• Solar neutrino

• Atmospheric neutrino

• Proton decay

• Exotic searches

JUNO

• Precision energy spectrum measurement interference between P31 and P32

37

Summary

• Nuclear reactors have played a crucial role in studying neutrino

properties and neutrino oscillations

• There are many on-going and future reactor neutrino experiments

covering a wide range of physics opportunities

• Neutrino oscillations: JUNO

• Sterile neutrino search: world-wide program

• Understanding reactor models: world-wide program

• Neutrino-electron scattering: TEXONO and possibly others

• Coherent Elastic Neutrino-nucleus Scattering: world-wide program

• BSM physics: many opportunities

38

39

Searching for sterile neutrino• The existence of sterile neutrino would introduce an additional spectral distortion

• Daya Bay, RENO and NEOS set limits to sin22θ14 at different |Δm241| region

• A combined analysis between DYB, MINOS and Bugey-3 excluded the MiniBooNE and LSND allowed parameter space at Δm2

41<0.8 eV2

40Phys. Rev. Lett. 118, 121802 (2017)

41

Oscillation results

sin22θ13 = [8.41±0.27(stat.)±0.19(syst.)]× 10-2

|Δm2ee| = [2.50±0.06(stat.)±0.06(syst.)]× 10-3 eV2

χ2/NDF = 234.7/263

Phys. Rev. D 95, 072006 (2017)

• Rate analysis: sin22θ13 = 0.071±0.011 χ2/NDF = 6.3/6

• Consistent results with those of the n-Gd analysis

• Spectrum distortion consistent with the oscillation hypothesis

42

sin22θ13 through n-H

Phys. Rev. D 93, 072011 (2016)

43

Spectrum evolution

• The evolution slopes are different at different energy ranges

• Neutrino spectrum do change with 239Pu fission fraction, in agreement with most models’ predictions

• No strange behavior at 4 to 6 MeV region

• Larger statistics and better detection efficiency estimates would improve the fuel evolution results

44

Muon flux modulation

• The underground muon flux was known to be positively correlated with the atmospheric temperature

• Daya Bay measures the correlation coefficient which is consistent to model prediction

• The only experiment measuring the coefficient with functionally identical detectors at different overburdens

arXiv:1708.01265

Muon rate

Temperature

Coefficient

Prospect detector

45

46