1. A review on the long baseline neutrino experiments Pasquale Migliozzi INFN, Sezione Napoli, Italy

2. The PMNS matrix e   is suppressed due to small  m 12 2  m 23 2 and  23 dominate  CP is the CP violation phase Leading oscillations in vacuum

3. Neutrino Oscillation Masses Three different types of experiments  three different mass ranges! Mass range accessible to long baseline experiments

4. Super-Kamiokande atmospheric results 1289 days

5. The atmospheric neutrino deficit There is an apparent deficit of atmospheric  's seen in both Cerenkov detectors and calorimeters While the atmospheric flux has large uncertainties, the L/E dependence implies oscillations If it is oscillations,  m 2 ~3x10 -3 eV 2 @sin 2 2  =1 It is likely to be  -  (  - s oscillations are excluded at 99% CL)

6. Why long baseline experiments? Check atmospheric neutrino results with a controllable beam See  appearance Measure the product |  m 2 23 |x  23 with ~10% precision Measure   e and  13 Constrain or measure   s

7. K2K 250 km Close Detectors (CD) Far Detector

8. Sensitivity of the experiment Sensitive to  m 2 above 2x10 -3 eV 2 !

9. The neutrino beam Pion monitor Target/Horn system Al alloy 250 kA current Proton intensity ~5.5x10 12 /spill Decay pipe 200 m long and filled with Helium @1atm Average energy ~ 1.3 GeV

10. K2K close detector Features Good vertex resolution Good angular resolution Good momentum resolution for high energy muons Aim Measure energy Measure profile Features Similar systematics as Super-Kamiokande Aim Measure rate Measure  0 production Fine grained calorimeter 1kton detector (MRD)  to SK MRD consists Drift tubes + 12 layers of Fe (10cm x4 + 20cm x8) Fiducial mass 312 ton 200k evts/2.1x10 19 pot

11. Profile and spectrum stability monitored with the MRD Stability of the profileStability of the spectrum To SK 1mrad beam direction is stable within 1 mrad

12. interaction @ different locations interaction @ different locations

13. Event selection at SK Using GPS timing require 28 evts fully contained in the fiducial volume have been observed 10 -3 background events up to now mainly from atmospheric events Trigger similar to the one used for atmospheric neutrino induced events

14. Expected # evts at SK (1) #evts expected at SK assuming null oscillations #evts observed in the near detector Live time correction using proton on target Event ratio R from MC calculation but  SK and  CD confirmed by pion monitor ( E > 1 GeV)

15. Expected # evts at SK (2) The 1kton water Cerenkov detector is used as reference CD detector same neutrino interaction target (cross section uncertainty cancels out) same detection energy threshold same event reconstruction scheme  small systematic error The MRD and Scifi/Water detectors are used to check the consistency of the 1kton detector results

16. Expected # evts at SK (3) Main source of sys. Error Sys. error from 1kton meas. (mainly due to fid. vol.) 5% Sys. error from SK meas. 3% Sys. Error from extrapolation Far/Close 7% Statistical error 1% From June 1999 to June 2000 (2.3x10 19 pot) Results consistent with a deficit @90% C.L. K2K is running now, energy spectrum distortion is under study

17. The CNGS neutrino beam ICARUS The beam will start in may 2005

18. The CNGS beam Shared SPS operation 200 days/year 4.5x10 19 pot / year Nominal beam  ( m -2 / pot) 7.78x10 -9  CC / pot / kton 5.85x10 -17 ( GeV ) 17 ( e + e ) /  0.87 %  /  2.1 %  prompt negligible

19. ICARUS liquid argon imaging The ICARUS technique is based on the fact that ionization electrons can drift over large distances (meters) in a volume of purified liquid Argon under a strong electric field. If a proper readout system is realized (i.e. a set of fine pitch wire grids) it is possible to realize a massive "electronic bubble chamber", with superb 3-D imaging. C.R. shower from 3 ton prototype

20. The ICARUS T600 module 3

21. Atmospheric neutrinos Large event statistics with Detection down to production thresholds Complete event final state reconstruction Identification all neutrino flavors Identification of neutral currents Excellent resolution on L/E reconstruction Direct  appearance search Neutrinos from CERN Search for µ   Search for µ  e Solar neutrinos Energy threshold: 5 MeV Large statistics, precision measurements “Smoking gun”: CC & NC ICARUS 5kton physics reach (I)  m 2 32,  23  m 2 12  m 2 32  23  13  m 2 12,  12 extract

22. Proton decay Large variety of decay modes accessible  study branching ratios free of systematics Background free searches  linear gain in sensitivity with exposure Neutrino “factory” Precise measurement oscillation Matter effects, sign of  m 2 23 First observation of e   CP violation ICARUS 5kton physics reach (II)  m 2 32  23  13  m 2 32 >0 or  m 2 32 <0 ? Unitarity of mixing matrix  0? extract

23.    oscillations (I)    oscillations (I) Analysis of the electron sample Exploit the small intrinsic e contamination of the beam (0.8% of  CC) Exploit the unique e/π 0 separation Statistical excess visible before cuts  this is the main reason for performing this experiment at long baseline ! At  m 2 =3.5x10 -3 eV 2 110  e events are expected Main background from charged current interactions of e in the beam 470 events are expected

24.    oscillations (II)    oscillations (II) Reconstructed energy Reconstructed visible energy spectrum of electron events clearly evidences excess from oscillations into tau neutrino

25.    oscillations (III)    oscillations (III) Transverse missing P T Kinematical selection in order to enhance S/B ratio Can be tuned “a posteriori” depending on the actual  m 2 For example, with cuts listed below, reduction of background by factor 100 for a signal efficiency 33%

26.  m 2 32 =3.5x10 –3 eV 2 ; sin 2 2  23 = 1 Search for  13 ≠0 (I) 4 years @ CNGS

27.  m 2 32 =3.5x10 –3 eV 2 ; sin 2 2  23 = 1 ; sin 2 2  13 = 0.05 Transverse missing P T Total visible energy Search for  13 ≠0 (II)

28. Sensitivity to  13 in three family- mixing Estimated sensitivity to   e oscillations in presence of    (three family mixing) Factor 5 improvement on sin 2 2  13 at  m 2 = 3x10 –3 eV 2 Almost two-orders of magnitude improvement over existing limit at high  m 2 4 years @ CNGS

29. The OPERA experimental technique Emulsion Cloud Chamber (ECC) Emulsions for tracking, passive material as target Basic technique works charmed “X-particle” first observed in cosmic rays (1971) DONUT/FNAL beam-dump experiment:  events observed  m 2 = (1.6 - 4) x10 -3 eV 2 ( SuperK)  M target ~2 kton of ECC large detector  sensitivity, complexity modular structure (“bricks”): basic performance is preserved Ongoing developments in the emulsion technique, required by the large vertex detector mass: industrially produced emulsion films automatic scanning microscopes with ultra high-speed Pb Emulsion layers  1 mm Experience with emulsions and/or  searches : E531, CHORUS, NOMAD and DONUT

30. ~ 10 m The detector at Gran Sasso (modular structure, configuration with three “supermodules”)  spectrometer Magnetised Iron Dipoles Drift tubes and RPCs target and  decay detector Each “supermodule” is a sequence of 24 “modules” consisting of - a “wall” of Pb/emulsion “bricks” - planes of orthogonal scintillator strips scintillator strips brick wall module brick (56 Pb/Em. “cells”) 8 cm (10X 0 ) supermodule Detector ready by may 2005

31. Sensitivity to    oscillations 5 years 3 years Summary of  detection efficiencies (in % and including BR) Expected background (5 years data taking)  m 2 = 1.2x10 -3 eV 2 at full mixing sin 2 (2  ) = 6.0x10 -3 at large  m 2 After 5 years data taking

32. Full mixing 5 years with shared SPS operation (2.25x10 20 pot) Average target mass = 1.8 kton (accounting for mass reduction with time, due to brick removal for analysis) Uncertainties on background and efficiencies accounted for in the following Expected  events

33. Statistical significance  6 events 4   discovery  Events observed Significance (equivalent  ) N4N4N4N4 Poisson distribution of the expected background #evts observed NnNn Probability that the b.g. fakes the signal: < P n  if #observed evts  N n  P 4  = 6.3x10 -5 P 3  = 2.7x10 -3

34. Probability of 4  significance Schematic view of the SK allowed region sin 2 2   m 2 (eV 2 ) 68% 22% 9% Simulate a large number of experiments with oscillation parameters generated according to the SuperK probability distribution N 4  events required for a discovery at 4  Evaluate fraction P 4  of experiments observing  N 4  events

35. 90 % CL limits *  m 2 ( 10 -3 eV 2 ) 1.6 2.5 4.0 Upper limit 2.2 3.1 4.6 Lower limit 0.9 1.8 3.4 (U - L) / (2xTrue) 41 % 26 % 15 % OPERA 90 % CL in 5 years * assuming the observation of a number of events corresponding to those expected for the given  m 2 Determination of  m 2 (mixing constrained by SuperK)

36. The MINOS Experiment Two Detector Neutrino Oscillation Experiment (Start 2004) Near detector: 980 tons Far detector: 5400 tons Iron/Scintillator Sampling calorimeter 1 cm x 4 cm plastic scintillator + 2.5 cm iron plates

37. The neutrino beam Target and horn 2 are moveable the beam energy can be changed 1st oscillation maximum Need the low energy beam with = 7.6 GeV to see 1st oscillation maximum which occurs at  2 GeV SK best fit Oscillation Probability

38. Physics Measurements Obtain firm evidence for oscillations Charge current (CC) interaction rate and energy distribution NC/(CC+NC) ratio (T-test) Measurement of oscillation parameters,  m 2, sin 2 2  CC energy distribution Determination of the oscillation mode(s)  or s from NC and CC energy distributions   e limits or observation by identification of electrons

39. Limit from the T-test N CC-like  events with identified muon N NC-like  events with no muon 10 kton-yr exposure 2% overall flux uncertainty 2% CC effciency uncertainty 2% NC trigger efficiency uncertainty

40. Limit from the CC Energy Spectrum 10 kton-yr exposure 2% overall flux uncertainty 2% bin-to-bin flux uncertainty 2% CC efficiency uncertainty

41. CC Energy Spectrum for various  m 2

42.  m 2, sin 2 2  sensitivity 10 kton-yr exposure 2% overall flux uncertainty 2% bin-to-bin flux uncertainty 2% CC efficiency uncertainty For  m 2 = 0.0035 eV 2 should be able to achieve better than 10% error at 68% C.L on both  m 2 and sin 2 2 

43. Limit on   e MINOS 10 kt-yr 90% C.L. limit Chooz 1999 m 3 > m 2 Matter effects included  m 2 solar = 3  10 -5 eV 2  12 =  23 = 45 degrees  = 0 10% systematic error on background U e3 2 3  10 -3 eV 2 sin 2 2  13 3  10 -3 eV 2 Already close to systematics limited with 10% error on background

44. JHF-Kamioka neutrino experiment Approved in December 2000 Construction 2001-2006 50 GeV PS machine Super-Kamiokande as a far detector Baseline 295 km Low energy neutrino beam tuned at the oscillation maximum

45. Physics measurements Factor 10 improvement in  disappearance  (sin 2 2  23 )~0.01  (  m 2 23 )~2x10 -4 eV 2 Search for   e appearance with a sensitivity 20 times better than CHOOZ limit sin 2 2  e  0.5 x sin 2 2  13 > 0.003 Search for a small admixture of sterile neutrinos

46. Layout of JHF and the beam Narrow Band BeamOff Axis beams Thin solid line shows the WBB A large variety of beams is available to tune the energy at the oscillation maximum Neutrino beam energy scan possible Energy peak around 1 GeV Electron neutrino contamination well below 1%

47.  disappearance  disappearance energy reconstruction for QE (red) and non-QE interactions  m 2 = 0.003 eV 2 sin 2 2  23 = 1 non-QE events All events w/o oscillations with oscillations The error bar is from the statistics of 5 years

48.  disappearance sensitivity  disappearance sensitivity Ratio of measured spectrum with oscillations to the expected one after subtraction of non-QE events Final sensitivity to oscillation parameters: Off Axis 2  beam NBB 1.5 GeV  NBB 3 GeV 

49.   e appearance search   e appearance search Signal: e (Far)/  (Near) expected to appear at the  disappearance dip Backgrounds   misidentification: negligible e contamination ~0.2-03%  0 (neutral current) background ~0.3% Sensitivity to sin 2 2   e > 0.003 A factor 20 better than the CHOOZ limit

50. ~ 3x10 20 e  yr ~ 3x10 20   yr The CERN present scenario Neutrinos from a muon storage ring A very complex acceleration and storage system

51. Optimal baseline is around 3000 km for CP violation + matter effects. Search for long-baseline detector laboratories

52. Physics from   e e  with a long baseline program at a Neutrino Factory  disappearance  (  m 2 ) ~ 5 x 10 -5  (sin 2 2  23 ) ~ 5 x 10 -3  – e appearance  sensitivity down to sin 2 2   e ~ 10 -3 - 10 -4 Matter effects  sign of  m 13 CP violation High energy e essential and unique Neutrino interaction rates x 10 or more w/r to present beams Very large detectors needed

53. Conclusion The existing experiment is running smootly and could give soon conclusive results (yes or no to oscillations) Experiments foreseen in the next years will probe mixing angles deeply (  23 and  13 ) Future accelerators will be able to pin down with high precision the oscillation parameters and study CP violation (  in the PMNS matrix) There is a lot of work to do!