Observation of νμ → ντ oscillations in the CNGS beam with the OPERA experiment

M. De Serio University of Bari & INFN Bari

On behalf of the OPERA Collaboration

3rd International Conference of New Frontiers in Physics - Kolymbari, Crete

2M. De Serio - ICNFP 2014

Observation of νμ → ντ oscillations in the CNGS beam with the OPERA experiment

The OPERA experiment

Physics motivation detection principle Detector Analysis chain Status of data analysis

νμ → ντ oscillation results

Other physics results

The OPERA Collaboration

M. De Serio - ICNFP 2014 3

About 140 physicists from 28 institutions

http://operaweb.lngs.infn.it

BelgiumIIHE-ULB Brussels

CroatiaIRB Zagreb

FranceLAPP Annecy

IPHC Strasbourg

GermanyHamburg

IsraelTechnion Haifa

ItalyBari

BolognaFrascatiL’Aquila

LNGSNaplesPadova

RomeSalerno

JapanAichiTohoKobe

NagoyaUtsunomiya

RussiaINR RAS MoscowLPI RAS Moscow

ITEP MoscowSINP MSU Moscow

JINR Dubna

SwitzerlandBern

TurkeyMETU Ankara

KoreaJinju

M. De Serio - ICNFP 2014 4

Neutrino oscillation appearance

Neutrino oscillation in the atmospheric sector established in 1998 in disappearance mode (SuperKamiokande, MACRO, …)

Oscillation mechanism later confirmed by accelerator experiments (K2K, MINOS) as well as solar neutrino experiments (KamLAND, SNO, Borexino, …)

Oscillated neutrino appearance: a missing tile in the picture.

OPERA designed to observe appearance in a pure beam.

Requirements:

High neutrino energy High beam intensity Large target mass Long baseline Detector capability to identify short-lived

lepton

5M. De Serio - ICNFP 2014

Oscillation Project with Emulsion tRacking Apparatus

L = 732 kmCERNLNGS

Cern Neutrinos to Gran Sasso beam

High-energy beam (<E 17 GeV>) optimised to maximise interactions at LNGS

Gran Sasso National Laboratory:1400 m rock shield

(cosmic flux reduction ~ 10-6)

M. De Serio - ICNFP 2014 6

Decay kink

t-nm oscillation

-, h-, e-

(+ 3h)nt

n

nm

-

ντ detection signature

Detection of ’s from oscillation in a sea of interactions through the observation of decays into muon, electron and hadron(s):

• Target mass O(kt) (small interaction cross-section)• High granularity detector ( decay detection, background rejection)

appearance on event-by-event basis

)4/(sincos)2(sin 223

213

423

2 ELmP

7M. De Serio - ICNFP 2014

OPERA detector

SUPERMODULE 1 SUPERMODULE 2

Hall C Laboratori Nazionali Gran Sasso

CNGS Beam

Muon spectrometerdipolar magnet instrumented

with RPCs and drift tubes

Target sectionwalls of lead – nuclear

emulsion bricks interfaced with planes of plastic

scintillator strips (Target Tracker, TT)

Veto

Muon ID, momentum and charge measurement Event trigger and brick

finding

8M. De Serio - ICNFP 2014

OPERA detector

SUPERMODULE 1 SUPERMODULE 2

Hall C Laboratori Nazionali Gran Sasso

CNGS Beam

Muon spectrometerdipolar magnet instrumented

with RPCs and drift tubes

Target sectionwalls of lead – nuclear

emulsion bricks interfaced with planes of plastic

scintillator strips (Target Trackers, TT)

Veto

id efficiency 95% (with TT)

p/p < 20% , p < 50 GeV/c

misidentified charge prob. : 1.2%

● Trigger efficiency > 99% ● Brick finding : 6080%

9M. De Serio - ICNFP 2014

OPERA detector

6 m

150,000 bricks (total target mass ~ 1.25 kt)

arranged in 62 walls Changeable Sheets (CS):

2 nuclear emulsion films in a removable

box glued on the downstream face of

the brickacting as an interface between TT and brick

Brick:57 nuclear emulsion films interleaved with 56 lead

plates (1mm thick)

12.5cm

10.2cm

10 mm

25 mrad

TT - CS

60 m

10 mrad

CS - brick

10M. De Serio - ICNFP 2014

From ED data analysis, a brick probability map is built.

The brick with the highest probability is extracted from the target.

First interaction reconstructed in OPERA target (2007)

Event location strategy

11M. De Serio - ICNFP 2014

Event location strategy• Analysis of CS confirmation of interaction, predictions for brick analysis

• Track following in the brick film by film up to disappearance point

• High-energy cosmic rays used for film inter-calibration

Micrometric resolution achieved in track and vertex reconstruction

JHEP 11 (2013) 036

Location efficiency vs event energyImpact parameter of tracks w.r.t.

reconstructed interaction vertices

12M. De Serio - ICNFP 2014

Decay search strategy

Decay confirmation benefits of :

• Momentum measurement by Multiple Coulomb Scattering

• E.m. shower detection and energy measurement

• Detection of highly-ionizing nuclear fragments

(discrimination between interactions and decays)

Impact parameter crucial to detect decay daughters

Details will be given in A. Pastore’s talk this afternoon

Impact parameter vs. longitudinal distance from reconstructed interaction vertex

13M. De Serio - ICNFP 2014

Search for short-lived particle decays: charmed hadrons as control sample

Charmed hadrons have masses and lifetimessimilar to those of the lepton:

powerful tool to directly test detection efficiency

50 events observed in the sample 2008 2010 54 4 events expected

arXiv1404.4357 Accepted for publication in EPJC

Details will be given in A. Pastore’s talk this afternoon

14M. De Serio - ICNFP 2014

Run p.o.t.(1019)

SPSefficienc

y

Beamdays

ν interaction

s

2008

1.7 61% 123 1931

2009

3.53 73% 155 4005

2010

4.09 80% 187 4515

2011

4.75 79% 243 5131

2012

3.86 82% 257 3923

Total

17.97 77% 965 19505

Status of the analysis

1.18 kt on average

80% of the expected intensity

Analysis strategy Runs 2008 – 2009:

1st and 2nd most probable bricks analysed Runs 2010 2012:

1st brick only analysed so far. Extension to less probable bricks in progress.

15M. De Serio - ICNFP 2014

νμ → ντ oscillation analysis

Data sample: 4685 located and fully analysed interactions

Expected events: 2.1 0.4 (m2 = 2.32 x 10-3 eV2, 23 = /4 )

Observed events: 4

16M. De Serio - ICNFP 2014

The first candidate event

Physics Letters B 691 (2010) 138

First direct detection of oscillations in appearance mode

Variable Value

Kink angle (mrad)

41 ± 2

Decay length (µm)

1335 ± 35

P daughter (GeV/c)

12 +6-3

Pt (MeV/c)470 +230

-

120

Missing Pt at 1ry vertex (MeV/c)

570 +320-

170

ϕ (deg) 173 ± 2

τ−→ρ− ντ

ρ−→π0 π- π0 → γ γ

17M. De Serio - ICNFP 2014

2000 m

The second candidate event

Journal of High Energy Physics 11 (2013) 036

- h- h- h+

candidate decay in plastic baseFlight length: 1.54 mm

18M. De Serio - ICNFP 2014

The third candidate event

𝑩→

First measurement of the lepton charge

in appearance mode

Phys. Rev. D 89 (2014) 051102(R)

- -

Muon momentum: by range (electronic detectors) 2.8 0.2 GeV/c

by MCS in emulsion 3.1 [2.6, 4.0] GeV/c

19M. De Serio - ICNFP 2014

The fourth candidate event

arXiv 1407.3513 - Submitted to PTEP

The fourth candidate event

M. De Serio - ICNFP 2014 20

All tracks attached to the reconstructed interaction have been followed in downstream bricks in order to check the compatibility

with the muon hypothesis.

Track 2 crosses 9 wall and stops in the spectrometer;not compatible with hypothesis based on

momentum – range correlation

Track 3: visible interaction in downstream brick #2

Track 4 crosses 1 wall; identified as a proton based on its ionization

21M. De Serio - ICNFP 2014

The fourth candidate event

22M. De Serio - ICNFP 2014

Expected events after completion of 2nd bricks for 2010 2012 runs: 3.2 events

νμ → ντ oscillation analysis summary

Not yet accounted for: foreseen extension of the analysis

to 3rd and 4th bricks in the probability map (~ 20% effect )

Visible energy in emulsion

1st candidate3rd candidate

4th candidate

2nd candidate

23M. De Serio - ICNFP 2014

Expected events after completion of 2nd bricks for 2010 2012 runs: 3.2 events

νμ → ντ oscillation analysis summary

Not yet accounted for: foreseen extension of the analysis

to 3rd and 4th bricks in the probability map (~ 20% effect )

Decay channel

Expected signal

Δm232 = 2.32

meV2

Total backgroun

d

Observed

h 0.41 ± 0.08 0.033 ± 0.006

2

3h 0.57 ± 0.11 0.155 ± 0.030

1

0.52 ± 0.10 0.018 ± 0.007

1

e 0.62 ± 0.12 0.027 ± 0.005

0

Total 2.11 ± 0.42 0.233 ± 0.041 4

Two statistical methods Fisher combination of single channel p-values Likelihood ratio

4.2 significance of null observationObservation of appearance

24M. De Serio - ICNFP 2014

νμ → ντ oscillation analysis

OPERA Preliminary (tau appearance)

ANTARES (atm. neutrino)

MINOS (pure atm. anti-nu)

MINOS (anti-nu beam)

MINOS (pure atm. nu)

MINOS (atmospheric)

T2K

MINOS (2ν, maximal mixing)

First measurement of m223

with appearance

90% CL intervals on Δm232

assuming sin2(2θ23) = 1

Feldman & Cousin [1.8 – 5] x 10-3 eV2

Bayesian [1.9 – 5] x 10-3 eV2

OPERA Off-peak L/<E> ~ 43 Km/GeV

(L/<E>)peak ~ 500 Km/GeV

25M. De Serio - ICNFP 2014

νμ → νe oscillation results

νe events searched for in a sample of 505 neutrino interactions without muon in the final state (runs 2008-

2009) .Extension to full statistics in progress.

19 (4) events observed, 19.8 2.8 (4.6) events expected (E<20GeV)

JHEP 1307 (2013) 004

Non standard oscillation limit: sin2(2θnew) < 7.2 x 10-3 for large m2

Details in V. Tioukov’s talk

26M. De Serio - ICNFP 2014

Cosmic ray physics results

OPERA: a deep underground detector with charge and momentum reconstruction and excellent timing capabilities (~10 ns)Measurement of the atmospheric muon charge ratio Rμ = Nμ+ / Nμ- in the TeV region

Complete data-sample analysed:~2.22 M μ events during 20082011 runs, 625.0 days of livetime in Standard Magnet Polarity ~823 k μ events during 2012 run, 234.8 days of livetime in Inverted Magnet Polarity Misalignment systematics reduced at the level of 0.1 mrad

Compatible with the expectation from a simple π-K model; no significant contribution

of the prompt component up to Eμ cos θ* ∼ 10 TeV

arXiv 1403.0244 - accepted for publication in EPJC

Rμ as a function of the vertical surface muon energy

Details in L. Consiglio’s poster

27M. De Serio - ICNFP 2014

Conclusions and Outlooks

Successful data taking from 2008 to 2012, 17.97 x 1019 p.o.t. integrated

Oscillation analysis results: 4685 located and fully analysed interactions4 candidate events observed 2.1 0.4 expected events

Observation of νμ ντ oscillation in appearance mode (significance 4.2 )

Other physics results:• Measurement of the atmospheric muon charge ratio in the TeV region • e oscillation

ντ search now being extended to other bricks of the probability map

Improvements in the statistical analysis under study

M. De Serio - ICNFP 2014 28

Thank you

Image taken using OPERA emulsion film with pinhole hand made cameraCourtesy by D.onato Di Ferdinando

M. De Serio - ICNFP 2014 29

νμ → ντ oscillation: background