1. R. Sulej National Centre for Nuclear Research, Otwock/Swierk Poland for the ICARUS Collaboration XV International Workshop on Neutrino Telescope Venice 11-15 March 2013 Sterile search with the ICARUS T600 in the CNGS beam

2. 1. Sterile and LSND/MiniBooNE anomaly 2. ICARUS T600 LAr TPC at LNGS  Detector operation and performance 3. Sterile neutrinos search in the CNGS beam Experimental search for the „LSND anomaly” with the ICARUS detector in the CNGS neutrino beam, Eur. Phys. J. C (arXiv:1209.0122)  e signal selection  Results 4. Conclusions Neutrino Telescopes, 11-15 March 20132 Outline

3. Neutrino Telescopes, 11-15 March 20133  The discovery of a Higgs boson at CERN/LHC has crowned the successful Standard Model (SM) and will call for a verification of the Higgs couplings to the gauge bosons and to the fermions.  Neutrino masses and oscillations represent today a main experimental evidence of physics beyond the Standard Model.  Being the only elementary fermions whose basic properties are still largely unknown, neutrinos must naturally be one of the main priorities to complete our knowledge of the SM.  Albeit still unknown precisely, the incredible smallness of the neutrino rest masses, compared to those of other elementary fermions points to some specific scenario, awaiting to be elucidated.  The astrophysical importance of neutrinos is immense, so is their cosmic evolution. The new frontier

4. Neutrino Telescopes, 11-15 March 20134  Sterile neutrinos: hypothesized in a seminal paper by B. Pontecorvo in 1957, as particles not interacting via any of fundamental interactions of SM except gravity.  Since per se they may not interact directly, they are extremely difficult to detect. If they are heavy enough, they may also contribute to cold dark matter or warm dark matter.  Sterile ’s may mix with ordinary neutrinos via a mass term. Evidence may be building up by “anomalies” observed by several experiments: - sterile (s) with  m 2 ≈10 -2 -1 eV 2 from e observation in  beam accelerator experiment, „LSND anomaly” - disappearance may have been observed in nuclear reactors and very intense (megaCurie) e-conversion  sources with maybe comparable mass differences. Sterile neutrinos ?

5. Neutrino Telescopes, 11-15 March 20135 Zhang, Qian, Vogel: „Reactor (…) with known θ 13 ” (arXiv:1303.0900) →  ≈ 1.4 ? Combined evidence  ≈ 3.8 Combined evidence for some possible anomaly: many standard deviations. Overall evidence

6. Neutrino Telescopes, 11-15 March 20136 Allowed LSND and MiniBooNE signals ICARUS T600 search for the LSND like  → e signal from the CNGS  beam at 730 km and 10 ≤ E  ≤ 30 GeV is presented. Neutrino related anomalies

7. aLaboratori Nazionali del Gran Sasso dell'INFN, Assergi (AQ), Italy bDipartimento di Fisica e INFN, Università di Padova, Via Marzolo 8, I-35131 Padova, Italy cDipartimento di Fisica Nucleare e Teorica e INFN, Università di Pavia, Via Bassi 6, I-27100 Pavia, Italy dDipartimento di Scienze Fisiche, INFN e Università Federico II, Napoli, Italy eDipartimento di Fisica, Università di L'Aquila, via Vetoio Località Coppito, I-67100 L'Aquila, Italy fINFN, Sezione di Milano e Politecnico, Via Celoria 16, I-20133 Milano, Italy gHenryk Niewodniczanski Institute of Nuclear Physics, Polish Academy of Science, Krakow, Poland hDepartment of Physics and Astronomy, University of California, Los Angeles, USA iINR RAS, prospekt 60-letiya Oktyabrya 7a, Moscow 117312, Russia jCERN, CH-1211 Geneve 23, Switzerland kInstitute of Theoretical Physics, Wroclaw University, Wroclaw, Poland lInstitute of Physics, University of Silesia, 4 Uniwersytecka st., 40-007 Katowice, Poland mNational Centre for Nuclear Research, A. Soltana 7, 05-400 Otwock/Swierk, Poland nLaboratori Nazionali di Frascati (INFN), Via Fermi 40, I-00044 Frascati, Italy oInstitute of Radioelectronics, Warsaw University of Technology, Nowowiejska, 00665 Warsaw, Poland pINFN, Sezione di Pisa. Largo B. Pontecorvo, 3, I-56127 Pisa, Italy M. Antonello a, P. Aprili a, B. Baibussinov b, M. Baldo Ceolin b,, P. Benetti c, E. Calligarich c, N. Canci a, S. Centro b, A. Cesana f, K. Cieslik g, D. B. Cline h, A.G. Cocco d, A. Dabrowska g, D. Dequal b, A. Dermenev i, R. Dolfini c, C. Farnese b, A. Fava b, A. Ferrari j, G. Fiorillo d, D. Gibin b, A. Gigli Berzolari c,, S. Gninenko i, A. Guglielmi b, M. Haranczyk g, J. Holeczek l, A. Ivashkin i, J. Kisiel l, I. Kochanek l, J. Lagoda m, S. Mania l, G. Mannocchi n, A. Menegolli c, G. Meng b, C. Montanari c, S. Otwinowski h, L. Periale n, A. Piazzoli c, P. Picchi n, F. Pietropaolo b, P. Plonski o, A. Rappoldi c, G.L. Raselli c, M. Rossella c, C. Rubbia a,j, P. Sala f, E. Scantamburlo e, A. Scaramelli f, E. Segreto a, F. Sergiampietri p, D. Stefan a, J. Stepaniak m, R. Sulej m,a, M. Szarska g, M. Terrani f, F. Varanini b, S. Ventura b, C. Vignoli a, H. Wang h, X. Yang h, A. Zalewska g, A. Zani c, K. Zaremba o. ICARUS Collaboration Neutrino Telescopes, 11-15 March 2013 7

8. 8  Two identical T300 modules (2 TPC chambers per module)  LAr active mass 476 t:  (17.9 x 3.1 x 1.5 for each TPC) m 3 ;  drift length = 1.5 m;  E drift = 0.5 kV/cm; v drift = 1.6 mm/  s  3 readout wire planes at 0°, ±60°, 3mm plane spacing (for each TPC chamber):   53000 wires, 3 mm pitch  2 Induction planes, 1 Collection  PMT for scintillation light (128 nm):  (20+54) PMTs  trigger and t 0 Hall B LN 2 vessels readout electronics T300 cryogenics (behind) T300 module: two TPCs with the common cathode cathode readout wire arrays EE 1.5m  Total energy reconstruction of events from charge integration.  Full sampling, homogeneous calorimeter; excellent accuracy for contained events. ICARUS @ LNGS: the first LARGE LAr-TPC

9. Neutrino Telescopes, 11-15 March 20139 CNGS   charge current interaction, one of TPC’s shown Collection (top view) Induction 2 (top view) Induction 1 (frontal view)  2D projection for each of 3 wire planes per TPC  3D spatial reconstruction from stereoscopic 2D projections  charge measurement from Collection plane signals ICARUS LAr-TPC detection technique

10. Neutrino Telescopes, 11-15 March 201310 I MODULE II MODULE 60 ppt O 2 equiv. Impurities from electro-negative molecules: O 2, H 2 O, CO 2  Ultra High Vacuum techniques; highly efficient filters based on Oxysorb/Hydrosorb;  Continuous purification by recirculation in liquid & gas phases;  During data taking:  e > 5ms ≈ 60 ppt O 2 equivalent impurities  max. 17% signal attenuation after 1.5m for the longest e – drift length, E drift = 0.5kV/cm. max drift Key feature: LAr purity

11. Neutrino Telescopes, 11-15 March 201311 2012: Mar. 23 – Dec.3 3.54 10 19 pot 2011: Mar. 19 – Nov. 14 4.44 10 19 pot 2010: Oct. 1 – Nov. 22 5.8 10 18 pot CNGS data taking (Oct. 2010 – Dec. 2012) large sample of interactions superluminal searches 1. Cherenkov-like e + e – emissions: P. L. B711 (2012) 270 2. timing measurement: P. L. B713 (2012), 17 3. precision measurement: JHEP 11 (2012) 049  oscillations  →    → e – e   → e ”LSND/MiniBooNE” anomaly in press: Eur. Phys. J. C, arXiv:1209.0122 detector live-time > 93% total 8.6 x 10 19 pot collected CNGS data taking: 2010-2012

12. Neutrino Telescopes, 11-15 March 201312  CC energy deposit 1 m.i.p. 2 m.i.p. Total energy reconstr. from charge integration  Full sampling, homogeneous calorimeter with excellent accuracy for contained events Tracking device  Precise 3D topology and accurate ionization  Muon momentum via multiple scattering Measurement of local energy deposition dE/dx  e/  remarkable separation (0.02 X 0 samples)  Particle identification by dE/dx vs range data dE/dx vs residual range compared to Bethe-Bloch curves LAr TPC performance Low energy electrons: σ(E)/E = 11% / √ E(MeV)+2% Electromagn. showers: σ(E)/E = 3% / √ E(GeV) Hadron showers: σ(E)/E ≈ 30% / √ E(GeV)

13. Neutrino Telescopes, 11-15 March 201313 cathode CNGS beam primary vertex kaon Collection Induction 3D reconstruction dE/dx based PID p K  μ New 2D => 3D approach for LAr TPC (in press: AHEP, arXiv:1210.5089) 3D object driven by optimization of its 2D projections (no need for drift matching) Precise particle tracking and identification

14. Neutrino Telescopes, 11-15 March 201314 Collection Induction2 Induction1 (a) Induction1 (b) muon track: reconstructed with Collection and Induction2 projected to Induction1 E k = 2.14GeV, length = 10.0 m dE/dx in 3 mm track segments (3D), after removing  rays and e.m. cascades MC – data agreement on the level of 2% signal/noise from Landau + gaussian ≈ 10 dE/dx in 3 mm track segments (3D), after removing  rays and e.m. cascades MC – data agreement on the level of 2% signal/noise from Landau + gaussian ≈ 10 long muon from  CC interaction MC:  =2.37 MeV/cm; RMS=1.37 data: μ=2.32 MeV/cm; RMS=1.31 dE/dx of muon tracks in CNGS

15. Neutrino Telescopes, 11-15 March 201315 θ E k = 685 ± 25 MeV E k = 102 ± 10 MeV Collection m π o = 127 ± 19 MeV/c² θ = 28.0 ± 2.5º p π o = 912 ± 26 MeV/c MC: single electrons (Compton) MC: e + e – pairs (  conversions) data: EM cascades (from  0 decays) ~0.02 X 0 sampling;  ’s separated from vtx;  rejection increases with energy; e selection efficiency const with energy; 1 m.i.p. 2 m.i.p. 1 m.i.p. 2 m.i.p. MC  0 reconstruction: e/  separation: dE/dx in cascade initial part

16. Neutrino Telescopes, 11-15 March 201316  Differences w.r.t. the LSND experiment:  → e signal from the CNGS  beam at: L = 730 km, 10 ≤ E  ≤ 30 GeV  When compared to other long baseline results (MINOS and T2K) ICARUS operates in a L/E  region in which contributions from standard neutrino oscillations are not yet too relevant.  Unique detection properties of LAr- TPC allows to identify individual e events with high efficiency. L/E ≈ 1 m/MeV at LSND, but L/E  ≈ 36.5 m/MeV at CNGS LSND-like short distance oscillation signal averages to: sin 2 (1.27  m 2 new L /E) ≈ ½ and:  → e  ≈ 1/2 sin 2 (2  new ) A search for LSND effects with ICARUS at CNGS

17. Neutrino Telescopes, 11-15 March 201317 CNGS neutrinos: almost pure  peaked in the range 10 ≤ E  ≤ 30 GeV – beam associated e about ~1% – CNGS events triggered by PMT signal inside SPS proton extraction gate – empty events rejection, few interactions/day Present sample: 1091 neutrino events (within 6% with MC expectation) – data from 2010 and 2011, 3.3 x 10 19 pot (out of total 8.6 x 10 19 pot collected) –  CC events identified by L>250 cm track from primary vtx, no hadr interaction – the signature of the  → e signal is observed visually Fiducial volume: – primary vertex min. 5 cm from each side, min. 50 cm from downstream wall to allow shower identification Energy deposition cut: < 30 GeV – optimized signal/background ratio ( e contamination extends to higher E ), only 15% signal events rejected – excellent MC/data agreement for deposited energy  negligible syst. error on signal and bkg expectations – 10% syst. error due to e beam component Data sample

18. Neutrino Telescopes, 11-15 March 201318 Expected conventional e – like events: (cuts: vtx in fiducial volume; visible energy ≤ 30 GeV) 3.0 ± 0.4 ev. due to the intrinsic e beam contamination 1.3 ± 0.3 ev. due to  13 oscillations, sin 2 (  13 ) = 0.0242 ± 0.0026 0.7 ± 0.05 ev. of  →   oscillations with electron production, 3 mixing predictions. The total is therefore of 5.0 ± 0.6 expected events (uncertainty on the NC and CC contaminations included) Selection efficiency, validated on MC sample of e events:  =0.74±.05 in a good approximation,  independent from shape of energy spectrum 3.7 ± 0.6 expected background events after cuts. Visibility cuts for  → e event identification: 1. Charged track from the vertex, m.i.p. over 8 wires (dE/dx < 3.1 MeV/cm excluding  -rays), developing into EM-shower 2. Spatial separation (150 mrad) from other tracks at the vertex, at least in one transverse (±60°) view Signal selection

19. Neutrino Telescopes, 11-15 March 201319 Typical MC event E = 11 GeV, p t = 1.0 GeV/c. (only the vertex region is shown) e events generated according to  spectrum in order to reproduce oscillation behaviour; full physics and detector MC simulation in agreement with data 122 events over 171 simulated inside the detector, satisfy fiducial volume and energy cuts; visibility cuts: (3 independent scanners), leading to 0.74 ± 0.05 efficiency; < 1% systematic error from dE/dx cut on the initial part of cascade; no e -like events selected among NC simulated sample of 800 events. Signal selection efficiency in MC simulation

20. Neutrino Telescopes, 11-15 March 201320 automatic cuts mimicking data selection, large sample of MC C1:inside fiducial volume and E dep < 30 GeV; C2:no identified muon, at least one shower; C3:one shower: initial point (or  conversion point) < 1 cm from vtx, separated from other tracks; C4:ionisation signal from single mip in the first 8 wires. Signal selection efficiency (after the fiducial and energy cuts): 0.6/0.81 = 0.74, in agreement with the visual scanning method. Signal selection efficiency check in MC simulation

21. Neutrino Telescopes, 11-15 March 201321 b (a) E tot = 11.5 ± 1.8 GeV, p t = 1.8 ± 0.4 GeV/c (b) vis E tot = 17 GeV, p t = 1.3 ±0.18 GeV/c In both events: single electron shower in the transverse plane clearly opposite to hadronic component 2 events observed in data a

22. Neutrino Telescopes, 11-15 March 201322 The limits on no. of events due to the LSND anomaly are respectively 3.41 (90% CL) and 7.13 (99% CL) Given the present data sample of, the limits to the oscillation probability are: P νμ → νe ≤ 5.4 x 10 -3 (90% CL) P νμ → νe ≤ 1.1 x 10 -2 (99% CL) The exclusion area is shown for the plot  m 2 vs sin 2 (2  ). At small  m 2 ICARUS strongly enhances the probability with respect to short baseline experiments. with (  m 2, sin 2 (2  )) = (0.11 eV 2, 0.10) as many as 30 events should have been seen. The observation is presently compatible with the absence of a LSND anomaly

23. Neutrino Telescopes, 11-15 March 201323 1: MiniBooNE best fit in the combined 3+1 model 1-5: excluded by present ICARUS result (=> excessive osc. P @ large L/E ) 6: best value including ICARUS result 6-9: compatible with present ICARUS Result combined with the low energy excess

24. Neutrino Telescopes, 11-15 March 201324 the present ICARUS limit the limits of KARMEN the positive signals of LSND and MiniBooNE Collaborations allowed MiniBooNE allowed LSND 90% allowed LSND 99% present ICARUS exlusion area limit of KARMEN ICARUS result strongly limits the window of possible parameters for LSND anomaly indicating a narrow region (  m 2, sin 2 2  ) = (0.5 eV 2, 0.005) where there is an overall agreement (90% CL): LSND anomaly surviving area

25. Neutrino Telescopes, 11-15 March 201325 ICARUS result strongly limits the window of possible parameters for LSND anomaly indicating a narrow region: (  m 2, sin 2 2  ) = (0.5 eV 2, 0.005) where there is an overall agreement (90 % CL) between: – the present ICARUS limit – the limits of KARMEN – the positive signals of LSND and MiniBooNE Successful operation of ICARUS T600 on CNGS beam. LAr TPC allows unambiguous identification of individual e events with high efficiency. Definitive answer to sterile search: ICARUS T600 + anew 150 t detectors exposed to new 2GeV beam at CERN SPS at 1.6 km and 460 m baseline. Conclusions

26. Neutrino Telescopes, 11-15 March 201326 Thank you!

27. Neutrino Telescopes, 11-15 March 201327 100 GeV primary proton beam fast extracted from SPS in North Area: C-target station next to TCC2 + two magnetic horns, 100 m decay pipe, 15 m of Fe/graphite dump, followed by  stations. Interchangeable and  focussing. Near position: 330 m 150t LAr-TPC detector to be build + magnetic spectrometer Far position: 1600 m ICARUS-T600 detector + magnetic spectrometer New CERN SPS 2 GeV neutrino facility in North Area ICARUS at CERN

28. Neutrino Telescopes, 11-15 March 201328 → increased event rate, reduced overall event multiplicity, enlarge the angular range improving substantially e selection efficiency ● Clarify the surviving LSND/MiniBooNE (  m 2 -sin 2 2  ) region  far: ICARUS-T600 + near: T150 at CERN SPS  shorter distances: 1600 m and 330 m from target  lower neutrino energies (~ 2 GeV) than in CNGS beam. minimum ionizing proton recoil ● In absence of oscillations: spectra at different distances equal  independent of specific experimental event signatures  without any MC comparisons ICARUS at CERN

29. Neutrino Telescopes, 11-15 March 201329 ● Dual method, unlike LNSD and MiniBooNE, determines both the mass difference and the value of the mixing angle. ● Events rates shown both for [background] and [oscill.+background] at d=330 m and d=1600 m and the  m 2 = 0.4 eV 2, sin 2 (2  ) = 0.01 ● For d=1.6 km, E  < 5 GeV and 4.5 10 19 pot (1 y) a e oscillation signal of ≈1200 events is expected above 5000 backgr. events T150 near detectorT600 far detector ICARUS at CERN: mass and mixing angle

30. Neutrino Telescopes, 11-15 March 201330 e-appearance: 1 year μ beam (left) 2 year anti- μ beam (right) for 4.5 10 19 pot/year, 3% syst. uncertainty on energy spectrum μ → e  μ →  e e → e e-disappearance: 1 year μ beam 1 year μ + 2 years anti- μ beams (right) combined “anomalies”: from reactor s, Gallex and Sage experiments. LSND allowed region fully explored in both cases ICARUS at CERN: expected sensitivities

31. Neutrino Telescopes, 11-15 March 201331 L/E   ICARUS window

32. Neutrino Telescopes, 11-15 March 201332 CNGS data Collection Induction2 proton:  visible part E k = 121±24 MeV  inelastic interaction at: E k = 45±9 MeV  init direction w.r.t. beam axis: cos:(-0.601, -0.779, -0.179) ± 3 o precision study possible thanks to:  particle identification  energy, momentum reconstruction  vertex region analysis  p QE: very clean events single visible particle identified as a stopping/interacting proton events with high energy and higher particle multiplicity and reinteractions Precision study, e.g.:  CC Quasi-elastic interactions