the stereo experiment · 10th october - the stereo experiment - h. almazan reactor antineutrino...

THE STEREO EXPERIMENTHELENA ALMAZÁN

ON BEHALF OF THE STEREO COLLABORATION MAX-PLANCK-INSTITUT FÜR KERNPHYSIK

HEIDELBERG

10TH OCTOBER - THE STEREO EXPERIMENT - H. ALMAZAN

REACTOR ANTINEUTRINO ANOMALY

2

L [m]

R=N

expN

cal

10 102 1030.70

0.80

0.90

1.00

1.10

1.20

R = 0.934 ± 0.024

Bugey−3Bugey−4Chooz

Daya BayDouble ChoozGosgen

ILLKrasnoyarskNucifer

Palo VerdeRENORovno88

Rovno91SRPRAA:

• deficit at 2.8σ in flux measured by several experiments at different distancies from reactors

Possible explanations

arXiv: 1703.00860, S. Gariazzo et al.

→ sterile neutrino hypothesis?→ wrong prediction? Discrepancy between observed and predicted IBD yields from 235U by Daya Bay. Suggests 235U could be the primary contributor of the RAA

∼ 1 eV sterile neutrino new oscillation angle

and mass splitting need dedicated measurement at short

distances

P⌫̄e!⌫̄e(E⌫̄e , L) = 1� sin2(2✓new)sin2

✓1.27

�m2new[eV

2]L[km]

E⌫̄e [MeV ]

◆

⌫̄e

arXiv: 1704.01082, Daya Bay Collaboration

https://arxiv.org/abs/1703.00860




STEREO EXPERIMENT

3

→ spectral shape studies Measurement of a pure 235U energy spectrum → Help to understand Daya Bay results

→ sterile neutrino hypothesis to probe the RAA region by measuring relative distortions of the energy spectrum as a function of the distance [9-11m]

arXi

v: 1

512.

0665

6, C

. Buc

k et

al.

↑rea

ctor↑

segmented detector (6 cells)

∼ 10 m

Cell

1

Cell

2

Cell

3

Cell

4

Cell

5

Cell

6

Independent from predicted energy spectrum

⌫̄e

⌫̄e



EXPERIMENTAL SITE - ILL

4

ILL research reactor

Grenoble (France)• nominal power 58.3 MWtherm 1019. s-1

• compact fuel element 40cm ∅

• HEU fuel 93% 235U

Experimental

site• short baseline experiment

8.9m < Lcore < 11.1m

• ground-level experiment

• gamma and neutron

background from neighbour

experiments

⌫̄e


STEREO DETECTOR

5

3m

1.5m

PMTs

Gamma-Catcher----------

outer crown+

liquid scintillator(without Gd)

Mineral oil

Acrylic wall2mm thick

Reflective foilVM2000Nylon net

(air gap)

Neutrino Target • 2.2 x 1.5 x 0.9 m3

• Gd-β-diketonate liquid scintillator → good response

Gamma catcher • Outer crown volume

• filled Gd-free liquid scintillator

Buffer • 48 PMTs (8inch) • acrylic buffers • mineral oil (optical coupling)

STEREO detector


DETECTION PRINCIPLE

6

p

n

e+ e-

⌫e + p ! e+ + n

Interaction via Inverse Beta Decay IBD

2. Delayed Event neutron capture

• Gd nuclei X

i

E�,i ⇠ 8MeV

E� ⇠ 2.2MeV• H nuclei

1. Prompt Event positron annihilation

Evis,e+ ' E⌫e � 0.8MeV

STEREO detector with shielding


DATA TAKING - TIME LINE

7

NovOct Nov Dec Jan Feb Mar Oct2016

Nov Dec Jan Feb Mar2017

Commissioning ON OFF ON OFF ON

2017Apr May Jun Jul Aug Sep Oct

OFF

2018

OFFON ON

Data Taking - Phase I Data Taking - Phase II

Phase I: • Loss of optical coupling between PMTs

and target for one target and on GC cell • Evolving light cross-talks between cells → repaired during summer 2017

Phase II: • Stable conditions ∼ 95% of data taking

time


DETECTOR RESPONSE: LIGHT CROSS-TALKS

8

Date [MM/DD/YY]10/31/16 12/31/16 03/02/17

[%]

ij L

0

5

10

15

20 Cell3→Cell2

GC Back→Cell6

GC D19→Cell3 Cell6→Cell5

Date [MM/DD/YY]10/31/17 12/31/17 03/02/18

[%]

ij L

0

5

10

15

20 Cell3→Cell2

GC Back→Cell6 GC D19→Cell3

Cell6→Cell5

liquid slowly infiltrating into acrylic walls detector stability after walls reparation

Evolution of the light cross-talk between cells estimated by muon events in neutrino runs

arXiv

: 180

4.09

052,

N. A

llem

ando

u et

al.

Phase I Phase II




0 200 400 600 800 1000 1200 1400Charge [PE]

3−

2−

1−

0

1

2

3

Non

-line

arity

[%]

Charge deposited in cell 6 [PE]50 100 150 200 250

Even

ts /

4 PE

02468

10121416182022

310×

DATA

MC

DETECTOR RESPONSE: CALIBRATION

9

LED system, used to study the single photoelectron and the

PMT-DAQ linearity in the detector target

extensive circulation of radioactive sources Ge68 , Sb124 , Cs137 , Mn54 , Zn65 , Na24 , AmBe

along 3 different calibration systems

arXi

v: 1

804.

0905

2, N

. Alle

man

dou

et a

l.

Cell response for Mn source internal deployment in Cell 6

Deviation from linearity for all the Target PMTs



DETECTOR RESPONSE: ENERGY RECONSTRUCTION

10

~Ereco = M�1 · ~Q

arXi

v: 1

804.

0905

2, N

. Alle

man

dou

et a

l.

Charge and energy connection by inverse M matrix

collected photons/MeV from calibration runs in cell i

cross-talk cells j->i measured online

Mij = CjLji

described by

Date [MM/DD/YY]10/31/17 12/31/17 03/02/18

[%]

ij L

0

5

10

15

20 Cell3→Cell2

GC Back→Cell6 GC D19→Cell3

Cell6→Cell5

Charge deposited in cell 6 [PE]50 100 150 200 250

Even

ts /

4 PE

02468

10121416182022

310×

DATA

MC



DETECTOR RESPONSE: ENERGY RECONSTRUCTION

11

~Ereco = M�1 · ~Q

arXi

v: 1

804.

0905

2, N

. Alle

man

dou

et a

l.

Charge and energy connection by inverse M matrix

Mij = CjLji

described by

1.00%0.35%

n-H peak n-Gd peak Time Stability Cell to cell deviation

Stability of the reconstructed n-H & n-Gd peaks – probing whole target volume



BACKGROUND IN STEREO

12

Reactor Induced Environmental Radioactivity Muon induced

• neutrons • gamma radiation

from n-capture

• Thorium/Uranium (concrete)

• Radon/Argon (air)

• spallation neutrons (in shielding)

• stopping muons

Reactor OFFReactor ON


passive shielding

SHIELDING IN STEREO

13



from n-capture




• stopping muons


SHIELDING IN STEREO

14



from n-capture




• stopping muons

water pool, active veto


ANTINEUTRINO SELECTION AND CORRELATED BACKGROUND

15

⌫e + p ! e+ + nIBD coincidence selection Prompt: • 1.6 < Eprompt < 7.1 MeV • Prompt contained in a target cell

and its 4 neighboring cells

Delayed: • 4.5 < Edelayed < 10 MeV

Cosmic rejection: • Muons: Veto 100µs after each

detected muon • Fast neutrons/multiple neutron

captures: Isolation cut of 100µs after and before prompt

• Stopping muons: Cut high asymmetry of the light collection in the vertex cell

e+n capture

Accidental pair: estimated with shifted time windows for the delayed search

Pulse shape discrimination for prompt signal: background estimation for neutron induced reactions


PULSE SHAPE DISCRIMINATION

16

Pulse shape discrimination (PSD)

allow us to distinguish background events (neutron induced reactions) with

real positron interactions

But PSD follows temperature changes • seasons and when reactor going on or

off (lasting for several weeks) • A PSD cut does not permit to have at

the same time: a constant neutrino acceptance and a constant background rejection


ANTINEUTRINO SIGNAL EXTRACTION: FROM PSD DISTRIBUTION

17

→ Solution: fit PSD distributions to extract neutrino rates

Reactor OFF Reactor ON

Self-consistent method to estimate background under component: • Adapt to PSD variations (temperature sensitivity) • Local rescaling to global norm (pressure sensitivity)

Multi-Gaussian fit for each cell / energy / time bin Aγ/Ap ratio constrained by the OFF model


ANTINEUTRINO SIGNAL EXTRACTION: FROM PSD DISTRIBUTION

18

Phase-II: Updated background model with increased statistics (large time binning)

A second component (Ap) for p-recoils, anchored relatively to the first one (Ap) → Possible physical origin: multiple proton recoils (under study)


RELATIVE COMPARISON OF NEUTRINO ENERGY DISTRIBUTION

19

Oscillation test using ratio of energy distributions - cell 1 taken as reference • Reduced systematics • Insensitive to absolute flux normalization and predicted spectrum shape

MC takes into account cells differences,

detection efficiencies etc.

• Vi is the covariance matrix of the 5 ratios (common reference for each cell) for the energy bin i • {α} are nuisance parameters to take into account estimated systematics

Pull terms Cell-to-cell correlated Uncorrelated

Energy scale 0.35% 1.00% from energy scale

Normalization - 1.70% from neutron efficiencies arXi

v:18

06.0

2096

, STE

REO

Col

labo

ratio

n



NO OSCILLATION HYPOTHESIS TEST

20

• Minimized pull terms stay within ± 1 σ • Non-oscillation hypothesis (H0) can not be rejected: p-value = 34 % for phase-I

Comparison between cells for data

ar

Xiv:

1806

.020

96, S

TERE

O C

olla

bora

tion




EXCLUSION CONTOURS

21

• Phase I and II combined results (66 + 47 data days reactor ON)

Considered as two independent measurements:

• Raster scan approach (Δχ2 law simulated in each Δm2 bin)

• Best-fit value of the RAA rejected at 98% C.L.

�2 = �2I(~↵I) + �2

II(~↵II)

★: RAA oscillation best fit ∆m2RAA = 2.3 eV2 - sin2 (2θRAA) = 0.14

arXiv:1806.02096, STEREO Collaboration Talk at Neutrino 2018


https://zenodo.org/record/1286998#.W7odxS-B0Wo


CONCLUSION

22

• Stereo is now running under stable conditions

• Data taking will continue until end 2019, reaching 300 days of reactor-ON data

• The correlated background understanding improves using reactor-OFF periods

• Exclusion contour obtained using the robust ratio method, rejects the original RAA best fit value is at 98% CL using a ratio comparison insensitive to bias in spectra prediction arXiv:1806.02096, STEREO Collaboration (submitted and accepted at PRL)

• Improved results are coming soon, with a pure 235U spectrum

thanks for your atte

ntion



THE STEREO COLLABORATION

BACK-UP


BACKGROUND SPECTRUM

25

Reactor OFF prompt energy spectrum

both prompt and delayed are neutron captures

H(n,γ)

12C(n,n’γ)12C


BACKGROUND SPECTRUM

26


BACKGROUND SPECTRUM

27


SHIELDING

28


SHIELDING

29


FIT

30


SENSITIVITY CONTOUR

31


NEUTRON EFFICIENCY STUDIES

32deployment position

norm

alize

d Da

ta/M

C ra

tio

0.92

0.94

0.96

0.98

1.00

1.02

1.04

1.06

Homogeneity of the Cut Efficiency Cell 1Cell 4Cell 6

1.08

-1%

+1%

bottom10 cm

top80 cm

center45 cm

n-Hn-Gd

reconstructed energy [MeV]0 1 2 3 4 5 6 7 8 9 10

entri

es/0

.05

MeV

0

100

200

300

400

500

600

700

800captures in Gd = (88.9 ± 0.2) %

AmBe neutron capture energy spectra


SYSTEMATIC UNCERTAINTIES

33

the stereo experiment · 10th october - the stereo experiment - h. almazan reactor antineutrino...

Documents