Sterile n search with the ICARUS T 600

Sterile n search with the ICARUS T 600 in the CNGS beam 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

Outline 1. Sterile n and LSND/Mini. Boo. NE anomaly 2. ICARUS T 600 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 (ar. Xiv: 1209. 0122) § e signal selection § Results 4. Conclusions Neutrino Telescopes, 11 -15 March 2013 2

The new frontier § 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. Neutrino Telescopes, 11 -15 March 2013 3

Sterile neutrinos ? § 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 Dm 2≈10 -2 -1 e. V 2 from e observation in m beam accelerator experiment, „LSND anomaly” - disappearance may have been observed in nuclear reactors and very intense (mega. Curie) e-conversion sources with maybe comparable mass differences. Neutrino Telescopes, 11 -15 March 2013 4

Overall evidence Combined evidence s ≈ 3. 8 Zhang, Qian, Vogel: „Reactor (…) with known θ 13” (ar. Xiv: 1303. 0900) → s ≈ 1. 4 ? Combined evidence for some possible anomaly: many standard deviations. Neutrino Telescopes, 11 -15 March 2013 5

Neutrino related anomalies Allowed LSND and Mini. Boo. NE signals ICARUS T 600 search for the LSND like nm→ne signal from the CNGS nm beam at 730 km and 10 ≤ En ≤ 30 Ge. V is presented. Neutrino Telescopes, 11 -15 March 2013 6

ICARUS Collaboration M. Antonelloa, P. Aprilia, B. Baibussinovb, M. Baldo Ceolinb, , P. Benettic, E. Calligarichc, N. Cancia, S. Centrob, A. Cesanaf, K. Cieslikg, D. B. Clineh, A. G. Coccod, A. Dabrowskag, D. Dequalb, A. Dermenevi, R. Dolfinic, C. Farneseb, A. Favab, A. Ferrarij, G. Fiorillod, D. Gibinb, A. Gigli Berzolaric, , S. Gninenkoi, A. Guglielmib, M. Haranczykg, J. Holeczekl, A. Ivashkini, J. Kisiell, I. Kochanekl, J. Lagodam, S. Manial, G. Mannocchin, A. Menegollic, G. Mengb, C. Montanaric, S. Otwinowskih, L. Perialen, A. Piazzolic, P. Picchin, F. Pietropaolob, P. Plonskio, A. Rappoldic, G. L. Rasellic, M. Rossellac, C. Rubbiaa, j, P. Salaf, E. Scantamburloe, A. Scaramellif, E. Segretoa, F. Sergiampietrip, D. Stefana, J. Stepaniakm, R. Sulejm, a, M. Szarskag, M. Terranif, F. Varaninib, S. Venturab, C. Vignolia, H. Wangh, X. Yangh, A. Zalewskag, A. Zanic, K. Zarembao. a b c d e f g h i j k l m n o p Laboratori Nazionali del Gran Sasso dell'INFN, Assergi (AQ), Italy Dipartimento di Fisica e INFN, Università di Padova, Via Marzolo 8, I-35131 Padova, Italy Dipartimento di Fisica Nucleare e Teorica e INFN, Università di Pavia, Via Bassi 6, I-27100 Pavia, Italy Dipartimento di Scienze Fisiche, INFN e Università Federico II, Napoli, Italy Dipartimento di Fisica, Università di L'Aquila, via Vetoio Località Coppito, I-67100 L'Aquila, Italy INFN, Sezione di Milano e Politecnico, Via Celoria 16, I-20133 Milano, Italy Henryk Niewodniczanski Institute of Nuclear Physics, Polish Academy of Science, Krakow, Poland Department of Physics and Astronomy, University of California, Los Angeles, USA INR RAS, prospekt 60 -letiya Oktyabrya 7 a, Moscow 117312, Russia CERN, CH-1211 Geneve 23, Switzerland Institute of Theoretical Physics, Wroclaw University, Wroclaw, Poland Institute of Physics, University of Silesia, 4 Uniwersytecka st. , 40 -007 Katowice, Poland National Centre for Nuclear Research, A. Soltana 7, 05 -400 Otwock/Swierk, Poland Laboratori Nazionali di Frascati (INFN), Via Fermi 40, I-00044 Frascati, Italy Institute of Radioelectronics, Warsaw University of Technology, Nowowiejska, 00665 Warsaw, Poland INFN, Sezione di Pisa. Largo B. Pontecorvo, 3, I-56127 Pisa, Italy Neutrino Telescopes, 11 -15 March 2013 7

ICARUS @ LNGS: the first LARGE LAr-TPC Hall B cathode 1. 5 m E cryogenics (behind) LN 2 vessels readout electronics T 300 E T 300 readout wire arrays T 300 module: two TPCs with the common cathode q Two identical T 300 modules q 3 readout wire planes at 0°, ± 60°, 3 mm plane (2 TPC chambers per module) spacing (for each TPC chamber): § 53000 wires, 3 mm pitch q LAr active mass 476 t: § 2 Induction planes, 1 Collection § (17. 9 x 3. 1 x 1. 5 for each TPC) m 3; § drift length = 1. 5 m; q PMT for scintillation light (128 nm): § Edrift = 0. 5 k. V/cm; vdrift = 1. 6 mm/ms § (20+54) PMTs § trigger and t 0 § Total energy reconstruction of events from charge integration. § Full sampling, homogeneous calorimeter; excellent accuracy for contained events. Neutrino Telescopes, 11 -15 March 2013 8

ICARUS LAr-TPC detection technique § 2 D projection for each of 3 wire planes per TPC § 3 D spatial reconstruction from stereoscopic 2 D projections § charge measurement from Collection plane signals Collection (top view) Induction 2 (top view) Induction 1 (frontal view) CNGS m charge current interaction, one of TPC’s shown Neutrino Telescopes, 11 -15 March 2013 9

II MODULE max drift I MODULE 60 ppt O 2 equiv. Key feature: LAr purity Impurities from electro-negative molecules: O 2, H 2 O, CO 2 q. Ultra High Vacuum techniques; highly efficient filters based on Oxysorb/Hydrosorb; q. Continuous purification by recirculation in liquid & gas phases; q. During data taking: te > 5 ms ≈ 60 ppt O 2 equivalent impurities max. 17% signal attenuation after 1. 5 m for the longest e– drift length, Edrift = 0. 5 k. V/cm. Neutrino Telescopes, 11 -15 March 2013 10

CNGS data taking: 2010 -2012 detector live-time > 93% CNGS data taking (Oct. 2010 – Dec. 2012) total 8. 6 x 1019 pot collected • large sample of n interactions • superluminal n searches 1. Cherenkov-like e+e– emissions: P. L. B 711 (2012) 270 2. timing measurement: P. L. B 713 (2012), 17 3. precision measurement: JHEP 11 (2012) 049 • n oscillations 2010: Oct. 1 – Nov. 22 m → t t - → e– e t nm → ne ”LSND/Mini. Boo. NE” anomaly 5. 8 1018 pot in press: Eur. Phys. J. C, ar. Xiv: 1209. 0122 2011: Mar. 19 – Nov. 14 4. 44 1019 pot Neutrino Telescopes, 11 -15 March 2013 2012: Mar. 23 – Dec. 3 3. 54 1019 pot 11

LAr TPC performance Total energy reconstr. from charge integration § Full sampling, homogeneous calorimeter with excellent accuracy for contained events Tracking device m. CC energy deposit § Precise 3 D topology and accurate ionization § Muon momentum via multiple scattering Measurement of local energy deposition d. E/dx § e/g remarkable separation (0. 02 X 0 samples) § Particle identification by d. E/dx vs range Low energy electrons: σ(E)/E = 11% / √ E(Me. V)+2% Electromagn. showers: 1 m. i. p. data d. E/dx vs residual range compared to Bethe-Bloch curves 2 m. i. p. σ(E)/E = 3% / √ E(Ge. V) Hadron showers: σ(E)/E ≈ 30% / √ E(Ge. V) Neutrino Telescopes, 11 -15 March 2013 12

Precise particle tracking and identification CNGS beam primary vertex kaon cathode New 2 D => 3 D approach for LAr TPC (in press: AHEP, ar. Xiv: 1210. 5089) 3 D object driven by optimization of its 2 D projections (no need for drift matching) Induction Collection 3 D reconstruction p K d. E/dx based PID μ Neutrino Telescopes, 11 -15 March 2013 13

d. E/dx of muon tracks in CNGS Ek = 2. 14 Ge. V, length = 10. 0 m Collection long muon from m CC interaction Induction 2 MC: m=2. 37 Me. V/cm; RMS=1. 37 data: μ=2. 32 Me. V/cm; RMS=1. 31 Induction 1 (a) muon track: • reconstructed with Collection and Induction 2 • projected to Induction 1 (b) d. E/dx in 3 mm track segments (3 D), after removing d rays and e. m. cascades MC – data agreement on the level of 2% signal/noise from Landau + gaussian ≈ 10 14 Neutrino Telescopes, 11 -15 March 2013

e/g separation: d. E/dx in cascade initial part Ek = 102 ± 10 Me. V 0 reconstruction: θ • MC: single electrons (Compton) • MC: e+ e– pairs (g conversions) • data: EM cascades (from 0 decays) 2 m. i. p. Ek = 685 ± 25 Me. V Collection • ~0. 02 X 0 sampling; g’s separated from vtx; • g rejection increases with energy; • e selection efficiency const with energy; 1 m. i. p. 2 m. i. p. 1 m. i. p. MC Neutrino Telescopes, 11 -15 March 2013 15

A search for LSND effects with ICARUS at CNGS m→ e signal from the CNGS m beam at: L = 730 km, 10 ≤ En ≤ 30 Ge. V § Differences w. r. t. the LSND experiment: • L/E ≈ 1 m/Me. V at LSND, but L/E ≈ 36. 5 m/Me. V at CNGS • LSND-like short distance oscillation signal averages to: sin 2(1. 27 Dm 2 new L /E) ≈ ½ and: <P>nm→ne ≈ 1/2 sin 2(2 qnew) § When compared to other long baseline results (MINOS and T 2 K) 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 ne events with high efficiency. Neutrino Telescopes, 11 -15 March 2013 16

Data sample • CNGS neutrinos: almost pure nm peaked in the range 10 ≤ En ≤ 30 Ge. V – 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 1019 pot (out of total 8. 6 x 1019 pot collected) – m CC events identified by L>250 cm track from primary vtx, no hadr interaction – the signature of the nm→ne 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 Ge. V – 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 Neutrino Telescopes, 11 -15 March 2013 17

Signal selection Visibility cuts for nm→ne event identification: 1. Charged track from the vertex, m. i. p. over 8 wires (d. E/dx < 3. 1 Me. V/cm excluding d-rays), developing into EM-shower 2. Spatial separation (150 mrad) from other tracks at the vertex, at least in one transverse (± 60°) view Expected conventional ne– like events: (cuts: vtx in fiducial volume; visible energy ≤ 30 Ge. V) 3. 0 ± 0. 4 ev. due to the intrinsic ne beam contamination 1. 3 ± 0. 3 ev. due to q 13 oscillations, sin 2(q 13) = 0. 0242 ± 0. 0026 0. 7 ± 0. 05 ev. of nm→nt 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 ne events: h =0. 74±. 05 in a good approximation, h independent from shape of energy spectrum 3. 7 ± 0. 6 expected background events after cuts. Neutrino Telescopes, 11 -15 March 2013 18

Signal selection efficiency in MC simulation • e events generated according to m 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 d. E/dx cut on the initial part of cascade; • no e-like events selected among NC simulated sample of 800 events. Typical MC event E = 11 Ge. V, pt= 1. 0 Ge. V/c. (only the vertex region is shown) Neutrino Telescopes, 11 -15 March 2013 19

Signal selection efficiency check in MC simulation • automatic cuts mimicking data selection, large sample of MC C 1: inside fiducial volume and Edep < 30 Ge. V; C 2: no identified muon, at least one shower; C 3: one shower: initial point (or g conversion point) < 1 cm from vtx, separated from other tracks; C 4: 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. Neutrino Telescopes, 11 -15 March 2013 20

2 events observed in data a b (a) Etot = 11. 5 ± 1. 8 Ge. V, pt = 1. 8 ± 0. 4 Ge. V/c (b) vis Etot = 17 Ge. V, pt = 1. 3 ± 0. 18 Ge. V/c In both events: single electron shower in the transverse plane clearly opposite to hadronic component Neutrino Telescopes, 11 -15 March 2013 21

The observation is presently compatible with the absence of a LSND anomaly • 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: Pνμ→νe ≤ 5. 4 x 10 -3 (90% CL) Pνμ→νe ≤ 1. 1 x 10 -2 (99% CL) • The exclusion area is shown for the plot Dm 2 vs sin 2(2 q). At small Dm 2 ICARUS strongly enhances the probability with respect to short baseline experiments. with (Dm 2, sin 2(2 q)) = (0. 11 e. V 2, 0. 10) as many as 30 events should have been seen. Neutrino Telescopes, 11 -15 March 2013 22

Result combined with the low energy excess 1: 1 -5: 6: 6 -9: Mini. Boo. NE best fit in the combined 3+1 model excluded by present ICARUS result (=> excessive osc. P @ large L/E ) best value including ICARUS result compatible with present ICARUS Neutrino Telescopes, 11 -15 March 2013 23

LSND anomaly surviving area allowed Mini. Boo. NE allowed LSND 90% allowed LSND 99% limit of KARMEN present ICARUS exlusion area ICARUS result strongly limits the window of possible parameters for LSND anomaly indicating a narrow region (Dm 2, sin 22 q) = (0. 5 e. V 2, 0. 005) where there is an overall agreement (90% CL): • the present ICARUS limit • the limits of KARMEN • the positive signals of LSND and Mini. Boo. NE Collaborations Neutrino Telescopes, 11 -15 March 2013 24

Conclusions • Successful operation of ICARUS T 600 on CNGS beam. • LAr TPC allows unambiguous identification of individual e events with high efficiency. • ICARUS result strongly limits the window of possible parameters for LSND anomaly indicating a narrow region: (Dm 2, sin 22 q) = (0. 5 e. V 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 Mini. Boo. NE • Definitive answer to sterile search: ICARUS T 600 + anew 150 t detectors exposed to new 2 Ge. V beam at CERN SPS at 1. 6 km and 460 m baseline. Neutrino Telescopes, 11 -15 March 2013 25

Thank you! Neutrino Telescopes, 11 -15 March 2013 26

ICARUS at CERN New CERN SPS 2 Ge. V neutrino facility in North Area 100 Ge. V primary proton beam fast extracted from SPS in North Area: C-target station next to TCC 2 + two magnetic horns, 100 m decay pipe, 15 m of Fe/graphite dump, followed by m stations. Interchangeable n and n focussing. Far position: 1600 m ICARUS-T 600 detector + magnetic spectrometer Near position: 330 m 150 t LAr-TPC detector to be build + magnetic spectrometer Neutrino Telescopes, 11 -15 March 2013 27

ICARUS at CERN ● Clarify the surviving LSND/Mini. Boo. NE (Dm -sin 2 q) region 2 2 § far: ICARUS-T 600 + near: T 150 at CERN SPS § shorter distances: 1600 m and 330 m from target § lower neutrino energies (~ 2 Ge. V) than in CNGS beam. → increased event rate, reduced overall event multiplicity, enlarge the angular range improving substantially e selection efficiency proton recoil minimum ionizing ● In absence of oscillations: n spectra at different distances equal § independent of specific experimental event signatures § without any MC comparisons Neutrino Telescopes, 11 -15 March 2013 28

ICARUS at CERN: mass and mixing angle ● Dual method, unlike LNSD and Mini. Boo. NE, 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 Dm 2 = 0. 4 e. V 2, sin 2(2 q) = 0. 01 ● For d=1. 6 km, E < 5 Ge. V and 4. 5 1019 pot (1 y) a e oscillation signal of ≈1200 events is expected above 5000 backgr. events T 150 near detector Neutrino Telescopes, 11 -15 March 2013 T 600 far detector 29

ICARUS at CERN: expected sensitivities e-appearance: 1 year μ beam (left) 2 year anti- μ beam (right) for 4. 5 1019 pot/year, 3% syst. uncertainty on energy spectrum μ→ e LSND allowed region fully explored in both cases e-disappearance: 1 year μ beam 1 year μ + 2 years anti- μ beams (right) e→ e combined “anomalies”: from reactor s, Gallex and Sage experiments. Neutrino Telescopes, 11 -15 March 2013 30

L/En nm ICARUS window Neutrino Telescopes, 11 -15 March 2013 31

Precision study, e. g. : m CC Quasi-elastic interactions CNGS data Collection p m proton: §visible part Ek = 121± 24 Me. V §inelastic interaction at: Ek = 45± 9 Me. V §init direction w. r. t. beam axis: cos: (-0. 601, -0. 779, -0. 179) ± 3 o Induction 2 precision study possible thanks to: § particle identification § energy, momentum reconstruction § vertex region analysis QE: very clean events • single visible particle identified as a stopping/interacting proton • events with high energy and higher particle multiplicity and reinteractions Neutrino Telescopes, 11 -15 March 2013 32