Accelerator searches for oscillations Roumen Tzenov CERN and

Accelerator searches for oscillations Roumen Tzenov CERN and University of Sofia Ioannina, 20 -23 April 2000

Legend • Introduction to neutrino oscillations • Short baseline accelerator searches for – CHORUS – NOMAD • Future long baseline accelerator searches

Neutrino mixing Associate neutrino flavour with the charged lepton flavour as seen in charged-current interactions: For massive neutrinos: flavour eigenstate need not be a mass eigenstate but can be a coherent superposition: Mixing matrix U is unitary The propagation of different mass eigenstates leads to flavour oscillation in vacuum: Simplification for 2 mixing flavours with mixing angle q (phase d): Interactions are now nondiagonal with the mass eigenstates!

Neutrino oscillations The probability that a neutrino oscillates (changes flavour): With definition: To have a large effect: Maximum at 1/4 oscillation length

Dm 2 (e. V 2) Two parametric oscillation plot Cosmologically relevant Solar n + seesaw + DM ? 100 10 1 10 -2 10 -3 10 -4 10 -5 10 -6 10 -7 10 -8 10 -9 10 -10 10 -11 10 -12 LSND Atmospheric n Kamiokande Super. Kamiokande Solar n (MSW) CHOR US and NOMA D Solar n (vacuum oscillation) 10 -5 10 -4 10 -3 sin 2 10 -2 2 Q 10 -1 1. e e e

The Collaboration Belgium (Brussels, Louvain-la-Neuve), CERN, Germany (Berlin, Münster), Israel (Haifa), Italy (Bari, Cagliari, Ferrara, Naples, Rome, Salerno), Japan (Toho, Kinki, Aichi, Kobe, Nagoya, Osaka, Utsunomiya) , Korea (Gyeongsang), The Netherlands (Amsterdam), Russia (Moscow), Turkey (Adana, Ankara, Istanbul)

CHORUS Main objective • nt appearance in the SPS WBB n beam via oscillation • P(n ® nt ) down to 1 • 10 -4 for m 2 ~10 e. V 2 • nt direct detection in 770 kg nuclear emulsion target Tag: visible 1 - and 3 - prongs decay of primary -lepton (decay path ~1. 5 mm) - h- n o e- e BR 18 % + - - n o 50 % 18 % 14 % “Kink”

CERN West Area Neutrino Facility 450 Ge. V SPS Beryllium protons target horn reflector vacuum tunnel earth/iron shielding ~0. 6 km; L(rms)/L~0. 2 124 m • WBB, = 26. 6 Ge. V 290 m • ~5 • 10 19 protons on target • ~840 K CC in CHORUS CC / CC ~ 3. 10 -6 • (~0. 1 background event) 408 m CHORUS NOMAD

WANF West Area Neutrino Facility The “horn”

SPS and WANF (n ) neutrino beam

CHORUS detector Nucl. Instr. Meth A 401 (1997) 7 - T=5° Calorimeter h 770 kg emulsion target and scintillating fibre tracker Muon spectrometer Air core spectrometer and emulsion tracker Veto plane

Scintillating fibre trackers Nucl. Instr. Meth A 412 (1998) 19 ~ 2 mrad, xy~150 mm

External electronic detectors: • sign and momentum of pions • Hadronic and e-m shower energy and direction • Muon momentum and id Event pre-selection and post -scanning analysis P(h±)<20 Ge. V/c p/p ~25% 3

Neutrino data-taking collection efficiency 1994 -1997 N. B. Longest/Largest emulsion exposure ever done

e tanq Predictions and Scanback

Nuclear emulsion yesterday u 1947, first nuclear emulsions. Lattes et al. , Brown et al. : Discovery of e

CHORUS emulsion plate Target = 4 stacks (1. 4 m 2) 1 stack = 36 plates MIP : 30 40 grains / 100 m 1/4 plate 100 m emulsion 350 mkm base 90 mkm – Grain size ~ 0. 3 mm 80 m – Angular resolution 1. 5 mrad

CHORUS automatic microscopes CCD and XYZ stage New Track Selector Host CPU Network data storage

CHORUS automatic microscopes Megapixel CCD and XYZ stage DSPs High Performance optics 1 m Processing Cluster

Inside a “vertex plate” n beam -54 mm -36 mm View size: 120 x 150 m 2 Focal depth : ~3 m Red frame: ~30 x 40 m 2 -21 mm 0 mm

Decay search

t- kink detection (parent search) Principle: Principle Parent track ( ) can be detected by wider view and general angle scanning at the vertex plate Offline selection small impact parameter between parent and daughter n kink point is in the vertex plate n scan-back track Impact parameter or h general scanning area

Off-line video-image analysis CHORUS Emulsion Display

Manual scanning on special events: Diffractive D*s production, double leptonic decay Phys. Lett. B 435(1998) 458 -464.

Status of Phase I scanning * 0 decay search not finished yet (1996 -1997), not included in current results

t Det efficiency: Ratio of Acceptances S=Nt if P t=1 A=detector acceptance N 1 =normalization h=Kink finding efficiency In the same way, it is applied to the 0 sample Located Vertexes

Background • 1 sample ( - -) – charm production from antineutrino CC (with primary lepton (e+ or +) unidentified ~10 -6 / - nt contamination of the beam N 1 ~10 -7 / N 1 • 0 sample ( - h-) – charm production from antineutrino CC ~2 • 10 -6 / N 1 – 1 -prong nuclear interaction without visible recoil or nuclear break-up (White kinks) ~2 • 10 -5 / N 1

Current Result • No n candidates CHORUS current limit found sin 2 2 < 8 • S = 6003 ± 17% (syst) • P < 2. 38 / 6003 = (@ 90% C. L. ) m 2/e. V 2 • n n CC (expected) = P • S, 4 • 10 -4 Includes also 17% systematic error (NIM A 320 (1993) 331) sin 2 2 • 10 4

Outlook: • Phase I scanning: Going to finish this year Expected gain in sensitivity: • ~1. 2 from 1 (short decays, statistics) • ~1. 2 from 0 (3 prongs, 0 96+97) • Phase II scanning and analysis: years 2000 -2001 • New generation of automatic systems • Upgraded predictions • 3 prongs dedicated search • e ? (electron id by MS in emulsion) • Full vertex analysis (NETSCAN, General tracking) charm physics: |Vcd|2, cc, D+/D 0 P < 1. 0 • 10 -4 (in absence of -candidates)

What to dofurther with accelerator beams?

Long baseline experiments

Long base line beams compared to WANF

MINOS detector

ICANOE detector

OPERA detector 200 ton iron/emulsion sandwiches + muon identificator

OPERA module

Long baseline experiments’ claims. . .

Conclusion u Current short baseline accelerator searches for oscillations have almost done their job; u No oscillations seen (so far) for large m 2 and small mixing angles; u Atmospheric neutrino data suggest, on the opposite, small mass difference and large mixing angle; u Several long baseline accelerator experiments are on the start scratch to clarify the issue. . .