Results of the OPERA experiment

Transcription

1 Results of the OPERA experiment Giovanni De Lellis University Federico II and INFN Napoli On behalf of the OPERA Collaboration For the XLVII Meeting of the LNGSC Committee On behalf of the OPERA Collabora2on 02/04/17 Giovanni De Lellis, XLVII LNGSC Mee2ng 1

2 The CNGS beam along its five years of operation Year Beam days P.O.T. (10 19 ) Total p.o.t. (10 18 ) date 02/04/17 Giovanni De Lellis, XLVII LNGSC Mee2ng 2

3 DATA ANALYSIS COMPLETED located interactions 6785 decay search 02/04/17 Giovanni De Lellis, XLVII LNGSC Mee2ng 3

4 NON-OSCILLATION PHYSICS 02/04/17 Giovanni De Lellis, XLVII LNGSC Mee2ng 4

5 COSMIC-RAY PHYSICS on surface R µ OPERA L3+C Utah MINOS CMS πk model Surface energy Measurement of TeV atmospheric muon charge ratio Eur. Phys. J. C74 (2014) on surface ε µ (GeV) R µ OPERA L3+C Utah MINOS CMS R µ N µ +/N µ, egative charged m Vertical surface energy πk model πk + RQPM model πk + QGSM model πk + VFGS model /04/17 Giovanni De Lellis, XLVII LNGSC Mee2ng cos θ* (GeV) 5 ε µ 5

6 Cosmic-muon rate and temperature dependence Gran Sasso underground ~ 3800 m w.e. à Minimum muon energy ~ 1.8 TeV Atmospheric temperature increase à density decrease à increase the pion decay rate à muon rate increase R 1 0 R 1 0 I µ (t) =Iµ 0 + T (x)w (x)dx T eff = W (x)dx High W in high atmosphere à high energy muons apple 2 I µ = Iµ 0 + I µ cos T (t t 0) I µ I 0 µ = T T e T e 02/04/17 Giovanni De Lellis, XLVII LNGSC Mee2ng 6

7 Annual modulation of cosmic-muon rate I µ I 0 µ =(1.54 ± 0.07)% t 0 = (176±4) days, June 26 th Temperature data by the European Center for Medium-range Weather Forecasts (ECMWF) 02/04/17 Giovanni De Lellis, XLVII LNGSC Mee2ng 7

8 Muon rate vs temperature variations I µ I 0 µ = T T e T e α T = 0.94±0.04 OPERA 02/04/17 8

9 MULTIPLICITY STUDIES IN NEUTRINO LEAD SCATTERING 02/04/17 Giovanni De Lellis, XLVII LNGSC Mee2ng 9

10 Track multiplicity distributions reflecting the dynamics of interactions Shower tracks Grey tracks Nuclear evaporation and break-up tracks 02/04/17 10

11 Multiplicity features KNO Scaling 02/04/17 Giovanni De Lellis, XLVII LNGSC Mee2ng 11

12 OSCILLATION PHYSICS ν µ ν e ANALYSIS ν µ ν τ ANALYSIS 02/04/17 Giovanni De Lellis, XLVII LNGSC Mee2ng 12

13 Electron neutrinos one event with a π 0 as seen in the brick electron Interface films γ showers 2 mm 10 mm ECC CS 02/04/17 35 candidates in the full data sample Figure 5: Display of the reconstructed emulsion tracks of one of the ν e candidate events. The reconstructed neutrino energy is 32.5 GeV. Two tracks are observed at the neutrino Giovanni De Lellis, interaction XLVII LNGSC Mee2ng vertex. One of the two generates 13 an electromagnetic shower and is identified as an electron. In addition, two

14 Electron neutrino energy Energy spectrum of ν e candidates ( data) Number of events / 10 GeV overflow from 3-flavour oscillations ν e, ν e from the beam contamination ν e, ν e ν τ from 3-flavour oscillations (τ -> e decay channel) 2 π 0 background [GeV] E rec candidates found ν e 02/04/17 Giovanni De Lellis, XLVII LNGSC Mee2ng 14

15 STERILE NEUTRINO SEARCH 3+1 model: bounds from ν e appearance with profile Likelihood method P µ! e Interference term ~standard oscillation Exotic oscillation C =2 U µ3 U e3 ij = 1.27 E m2 ij L µe = Arg(U µ3 U e3u µ4u e4 ) sin 2 2# µe =4 U µ4 2 U e4 2 Energy distribution to constrain the parameter space: shape analysis Sensitivity curves vs energy cut 02/04/17 15

16 CONSTRAINING STERILE NEUTRINOS WITH A 3+1 MODEL First sterile neutrino search in a long baseline with ν µ à ν e and a 3+1 model 2 flavour approx. invalid at CNGS baselines sin 2 2# µe =4 U µ4 2 U e4 2 02/04/17 Giovanni De Lellis, XLVII LNGSC Mee2ng 16

17 ν µ ν τ ANALYSIS 02/04/17 Giovanni De Lellis, XLVII LNGSC Mee2ng 17

18 ν µ ν τ ANALYSIS STRATEGY runs No kinematical selection: get confidence on the detector performances before applying any kinematical cut Slower analysis speed (signal/noise not optimal) Kinematical selection applied for the candidate selection, coherently for all runs Good data/mc agreement shown runs Pµ < 15 GeV/c, to suppress charm background Prioritise the analysis of the most probable brick in the probability map: optimal ratio between efficiency and analysis time Analyse the other bricks in the probability map 02/04/17 Giovanni De Lellis, XLVII LNGSC Mee2ng 18

19 THE FIRST ν τ CANDIDATE in the brick 02/04/17 τ ρ ν τ ρ π 0 π - π 0 γ γ Physics Giovanni Letters De Lellis, XLVII B691 LNGSC (2010) Mee2ng

20 THE SECOND ν τ CANDIDATE in the brick τ π - π + π - ν τ 02/04/17 Journal Giovanni of High De Lellis, Energy XLVII LNGSC Physics Mee2ng 11 (2013)

21 THE THIRD ν τ CANDIDATE in the brick τ µ - ν µ ν τ Phys. Rev. D 89 (2014) (R) 02/04/17 Giovanni De Lellis, XLVII LNGSC Mee2ng 21

22 THE FORTH ν τ CANDIDATE in the brick τ h - ν τ PTEP 2014 (2014) 101C01 02/04/17 Giovanni De Lellis, XLVII LNGSC Mee2ng 22

23 OUTSTANDING PAPER AWARD BY THE PHYSICAL SOCIETY OF JAPAN 02/04/17 Giovanni De Lellis, XLVII LNGSC Mee2ng 23

24 THE FIFTH ν τ CANDIDATE in the brick τ h - ν τ PRL 115 (2015) /04/17 Giovanni De Lellis, XLVII LNGSC Mee2ng 24

25 02/04/17 Giovanni De Lellis, XLVII LNGSC Mee2ng 25

26 PRL 115 (2015) /04/17 Giovanni De Lellis, XLVII LNGSC Mee2ng 26

27 NEW EVENT ANALYSIS Widen selection cuts to increase the statistics Candidate identification mostly topological with looser kinematical cuts Statistical gain to reduce uncertainties Use likelihood approach, simulation fully validated with data in all kinematical corners 02/04/17 Giovanni De Lellis, XLVII LNGSC Mee2ng 27

28 NEW SELECTION 02/04/17 Giovanni De Lellis, XLVII LNGSC Mee2ng 28

29 NEW TAU NEUTRINO CANDIDATES τ à 3h Ev Ev τ à 3h τ à h Ev Ev τ à h τ à h Ev /04/17 29

30 A CLOSER LOOK AT ONE OF THESE EVENTS 02/04/17 Giovanni De Lellis, XLVII LNGSC Mee2ng 30

31 AN EVENT WITH THREE VERTICES WITHOUT ANY MUON IN THE FINAL STATE 02/04/17 Giovanni De Lellis, XLVII LNGSC Mee2ng 31

32 KINK TOPOLOGY Flight Length = 1160 µm With γ attached IP = 8±8 µm # kink = 90 mrad! P? = 240 MeV not passing the standard cuts to be a tau candidate 02/04/17 32

33 Track segments showing a double vertex topology in the same lead plate 2 prong vertex Flight Length = 103 µm Invariant mass = 1.8±0.5 GeV 02/04/17 Giovanni De Lellis, XLVII LNGSC Mee2ng 33

34 Leading Feynman diagrams 02/04/17 Giovanni De Lellis, XLVII LNGSC Mee2ng 34

35 Expected yield and multivariate analysis Sample Muon misidentified Expected events (10 3 ) CC + charm 45 µ CC + charm + h int yes 21 µ NC + c c 13 CC + h int 9 µ CC + 2h int yes 4 µ NC + 2h int 4 Total /04/17 35

36 Neutral network output 02/04/17 Giovanni De Lellis, XLVII LNGSC Mee2ng 36

37 Observation of a tau neutrino interaction with a charmed hadron production x PDF from ANN output n i = yield of i-th process Background only à µ = 0 CL =(2.6 ± 0.2) 10 4! /04/17 Giovanni De Lellis, XLVII LNGSC Mee2ng 37

38 ANALYSIS OF THE EXTENDED SAMPLE 02/04/17 Giovanni De Lellis, XLVII LNGSC Mee2ng 38

39 VISIBLE ENERGY OF ALL CANDIDATES Sum of the momenta of charged particles and γ s measured in emulsion P sum entries ν τ bkg data P sum (GeV/c) 02/04/17 Giovanni De Lellis, XLVII LNGSC Mee2ng 39

40 Δm 2 measurement N / P ( µ! ) ν τ cross-section = 0 EK(E) SM value 0 = cm 2 GeV 1 40

41 Significance of the tau neutrino appearance using 8 channels Test statistic: Profile likelihood ratio one sided p value = significance 90%CL interval on signal strength µ: [0.51, 2.6] 02/04/17 41

42 Input variables for BDT analysis the τ à h channel 02/04/17 42

43 BDT output the τ à h channel 02/04/17 Giovanni De Lellis, XLVII LNGSC Mee2ng 43

44 Likelihood analysis with BDT discrimination using 4 channels p value = significance 02/04/17 Giovanni De Lellis, XLVII LNGSC Mee2ng 44

45 Publications being issued Cosmic-ray annual modulation Study of charged particle multiplicity in high-energy neutrino-lead interactions Search for sterile neutrinos in the muon to electron channel Observation of a tau neutrino candidate with charmed hadron production Extended ν µ à ν τ search and Δm 2 measurement in appearance mode 02/04/17 Giovanni De Lellis, XLVII LNGSC Mee2ng 45

46 Forthcoming publications Search for sterile neutrinos in the muon to tau channel with the full data sample Combination of electron and tau appearance with muon disappearance to constraint the oscillation parameters 02/04/17 Giovanni De Lellis, XLVII LNGSC Mee2ng 46

47 OPERA DATA PRESERVATION CERN willing to host OPERA data for its preservation OPERA data sample < 100 TB (~10 tapes) equivalent to one of the LEP experiments LEP support granted until 2030 Agreement with data preservation office at CERN a year ago First data sets transferred Contacts with LNGS Computing Service for a similar action 02/04/17 Giovanni De Lellis, XLVII LNGSC Mee2ng 47

48 OPERA OPEN DATA AT CERN CERN has a program of open data access mostly for educational purposes (CMS, LHCb, ) CERN willing to include OPERA data among the open data OPERA agreed to provide neutrino interactions reconstructed in the bricks: both data & event display (effective for educational purposes) 02/04/17 Giovanni De Lellis, XLVII LNGSC Mee2ng 48

49 TECHNOLOGICAL DEVELOPMENTS OPERA system Running JINST 8 (2013) P01023 JINST 10 (2015) P11006 JINST 11 (2016) P /04/ Scanning 0 speed: ~190 cm 2 /h ~10 times faster R&D Giovanni De Lellis, XLVII LNGSC Mee2ng 2-3 years 49

50 02/04/17 Giovanni De Lellis, XLVII LNGSC Mee2ng 50

51 ACKNOWLEDGEMENTS The OPERA Collaboration is very much indebted to: The INFN LNGS hosting laboratory CERN and the CNGS beam team Funding agencies All industrial partners and suppliers Our technical collaborators, engineers and undergraduate students 02/04/17 Giovanni De Lellis, XLVII LNGSC Mee2ng 51