— Detection of UHE tau neutrinos with a

- Detection of UHE tau neutrinos with a surface detector array Saclay 25/01/06 M. Iori , Roma 1

OUTLINE Ø Ø • Neutrino tau Detection method → an alternative method From theory we know… Neutrino conversion efficiency Results of tests of a prototype at high altitude (3600 m) Saclay 25/01/06 M. Iori , Roma 1 2

Neutrino tau • Standard acceleration processes in astrophysics rarely produce τ. • e : μ : τ =1: 2: 0 • In case of oscillation full mixing, the flux ratios evolves towards • e : μ : τ =1: 1: 1 for a very large range in Δm 2 ~ 50%Eντ ντ →CC → τ + h → shower(1) + shower (2) ντ →NC → ντ + h → τ → shower (1) CC 1 st shower: decays into hadrons (0. 64), this shower carries visible Eτ ~ 0. 5 Eν 2 nd shower: neutrino-nucleon scattering, this shower carries 0. 2 Eν Eτ=(1 -y)Eν 0. 2=0. 25 and Eτ=0. 75 Eν Motivation: Ø μ ↔ τ oscillation of atmospheric neutrinos well established ØNo direct evidence for astrophysics τ appearance observed yet Saclay 25/01/06 M. Iori , Roma 1 3

Could the flavour ratios differ from 1: 1: 1? • High Energy neutrinos are belived to be produced through the decays π+ →μ+ +νμ → e+ + νμ + νe + νμ produced in pγ, pp, pn interactions. The ratio of the fluxes of neutrinos is expected to be 1: 2: 0 • if energy losses is considered (radiation) for µ from high energy pion decay → suppression of contribution of Neutrinos from muon decay π and μ decays Ø Then we expect a ratio at source 0: 1: 0 Ø Reactor results give each mass eigenvalues contains equal fraction of νμ –ντ then at Earth the ratio is 1: 2: 2 Ref. J. . F. Beacom N. F. Bell, D. Hooper, S. Pakvasa, T. J. Weiler Phys. Rev. D 68 093005 2003 Saclay 25/01/06 M. Iori , Roma 1 νμ νμ νe π+ 1 1 1 π- 1 1 νe 1 4

Other scenarios if not 1: 1: 1 • • Neutrino decay (Beacom, Bell, Hooper, Pakvasa& Weiler) CPT violation (Barenboim& Quigg) Oscillation to steriles with very tiny delta δm 2 (Crocker et al; Berezinskyet al. ) Pseudo-Dirac mixing (Beacom, Bell, Hooper, Learned, Pakvasa& Weiler) 3+1 or 2+2 models with sterile neutrinos (Dutta, Reno and Sarcevic) Magnetic moment transitions (Enqvist, Keränen, Maalampi) Varying mass neutrinos (Fardon, Nelson & Weiner; Hung & Pas) Saclay 25/01/06 M. Iori , Roma 1 5

Detection methods UHEC neutrinos E>108 Ge. V are difficult to be detected: Ø The atmosphere is transparent → low density ØThe Earth opaque to them, ν interaction length ≤ 2000 km 1. q Ø • • • Detection by Large Volume detectors → Cerenkov: Cerenkov rate Φshower = NAv ρ∫(dΦν /d. E) d. Eσν A ντ channel is quite different from νμ τ will decay and generate a cascade decay vertex and showers are separated by several ten meters at Pe. V → double bang event. This signature is constrained by effective volume Similar to detection techniques used in low energy experiments (Super-K) Technique with a proven capacity to detect neutrinos (atmospheric) Cascade detection: e, τ Muon tracking: Effective volume >> instrumented Volume Excellent pointing, Poor energy resolution Saclay 25/01/06 Effective volume = instrumented volume Poor pointing accuracy, Good energy resolution M. Iori , Roma 1 6

ντ detection technique in a large Volume detectors Cascade detection Lτ =49 Eτ m/Pe. V τ ντ ντ km 3 Saclay 25/01/06 M. Iori , Roma 1 7

Detection methods cont’d 2. Detection by Large Surface detector: • • • ρair =10 -3 g/cm 3 hence large acceptance → 104 km 3 sr → Auger Horizontal neutrino shower rate Φshower =NAv ρair ∫ (dΦν /d. E) d. Eσν A If we increase ρ of a factor ~2500 (i. e. rock) we can reduce the acceptance τ → Saclay 25/01/06 M. Iori , Roma 1 alternative method → 8

An alternative method → Earth skimmed neutrinos * Earth skimming process implies: ü neutrino propagation in the Earth during which interacts and regenerates (~10% →see Beacom et al. ) ü tau propagation in the Earth during which looses energy and decays in the atmosphere after exit from Earth surface. * → upward going air shower emerging from the ground may be detected using the fluorescence light (low duty cycle ~10%) or a directional detector. J. F. Beacom, D. Fargion, J. L. Feng, E. W. Kolb and E. Zas Saclay 25/01/06 M. Iori , Roma 1 9

Detection strategy: aim at the horizon • strategy to detect τ is aim at the horizon To do that we must have: • • direction detection (i. e time of flight) good geometric acceptance Saclay 25/01/06 M. Iori , Roma 1 10

Different models evaluate the UHE neutrino flux AGN: galaxies with extremely Energetic processes in the nuclei. They are the most likely source of cosmic rays above 1015 e. V. AGNs are thought to be fueled by matter accreting onto black holes. Z-burst: A very high energy (>1021 e. V) neutrino interacts with a relic neutrino left over from the Big Bang. This produces showers of particles, some of which are neutrinos. GZK: Any high energy hadron will interact with the cosmic microwave background (CMB). This will eventually lead to the production of neutrinos. The rate at which hadrons interact with the CMB is controlled by the injection spectrum of the hadrons and the characteristics of the universe. The uncertainties in this rate can be as high as 300%. Saclay 25/01/06 M. Iori , Roma 1 topological defects (TDs) which are relics of symmetry breaking just after the Big Bang excluded by Auger exp Pune Conf. 2005 11

AGN and GZK Excluded by A-B 10 Astr. Phys 22 339 05 AGN integrated flux ~104 km-2 yr-1 sr-1 @ Eτ 108 -1011 Ge. V GZK is a factor 102 -103 less AGN GZK mechanism: ⌐ p+ e- + νe p+γcmb→Δ+ →n +π+ ∟μ++ν ∟e++νe +νμ Saclay 25/01/06 M. Iori , Roma 1 AGN GZK hep-ph/0011176 JCAP 0404 2004 Cosmological neutrino flux and sensitivity for planned projects 12

Search for ντ up-going Tau events: • τ shower: high multiplicity 10 particles/m 2 • 64% hadronic decays n(π)ντ • Curved shower wave front (detected by timing) unlike the plane wavefront of atmospheric showers at large angles • Shower emerging from ground Θ>900 • Decay vertex above the ground Backgrounds: • Horizontal atmospheric shower, p interaction → low multiplicity 0. 01 particles /m 2 • DIS with pions. . muon bundles Saclay 25/01/06 M. Iori , Roma 1 13

Time structure of tau shower tau decay : shower with particles spreaded in time → curvature radius The shower Max occurs after 3 -4 km the decay point μ density in a time window of 500 ns Ordinary hadrons or nuclei (p, Fe. . ) interact at the top of the atmosphere. Electromagnetic component reduced, muons → plane wavefront Saclay 25/01/06 M. Iori , Roma 1 14

A comment on acceptance for neutrino showers Geometric Acceptance is function of Array Surface and shower radius mainly hence → q Effective area, A = (a sin(α-β)+2 r)(b+2 r)dΏ where Ø Surface of Array, S=axb Ø r=shower radius of shower at first interaction point Ø Emerging angle Β =Θzenith-90 ~ 50 • • Horizontal array α~0, sinβ~10 -2 hence → no emerging showers can be detected, only horizontal → Auger surface detector Inclined array sin(α-β)~0. 6 q Detection and trigger efficiency good → no particular event topology required Saclay 25/01/06 M. Iori , Roma 1 15

Probability to get an emerging Tau ν interaction length in rock neutrino cross section Τau Range in air The tau flux is determined by ratio tau range to CC neutrino N interaction length Ratio @ 1017 -18 e. V ~10 -3 hence ν flux at energy>1017 e. V Rdecay = E/mτ cτ Saclay 25/01/06 M. Iori , Roma 1 Τau Range in rock. behaviour including photonuclear pair and bremsstrahlung 16

Charged particles as function of the depth of the shower Gaisser_Hillas distribution total γ μ e π τ → πππντ ü At large depth muons constant and electrons drop; at 3 km electrons/muons ~100 Saclay 25/01/06 Laundau-Pomeranchuck-Migdal effect retards the development of em component M. T. Dova astro-ph/0505583 M. Iori , Roma 1 17

The neutrino conversion efficiency, ε in the Earth crust we computed the probability for τ of energy E to survive for a slant depth x inside the crust of Earth and converting to τ within [x, x+dx] e(-x/λν) dx/λν (int. length, λν=1/NAσρ) The probability for a tau lepton of energy E’ to survive for a slant depth L-x inside the Earth crust and exit the crust with the energy Eτ including the energy loss • The neutrino conversion efficiency : ε=∫ e(-x/λν) e(L-x/λτ) dx/λν is convoluted with decay length and longitudinal shower development and neutrino flux 1/E 2 gives : Saclay 25/01/06 M. Iori , Roma 1 18

The τ sample versus θ and ε for Eν = 108 -1011 Ge. V Toy Monte. Carlo L ντ L= 250 500 km 92. 50 95. 00 L in rock (km) Saclay 25/01/06 events with at least 10 part/m 2 on the array and Tau decay length from ground L= 200 -300 km Detector at 8 km ε ~ 1. 0 – 1. 5% To detect τ shower requiring at least 10 part/m 2 it must decay at 4 -5 km from the ground level Colors: nu interactions where tau reaches the ground with shower max M. Iori , Roma 1 19

ε as function of detector distance and L for Eν = 108 -1011 Ge. V 92. 50 950 ± 200 conversion efficiency ε Detector aperture versus L at different A=∫Ω Seffcos(θ-α)dΩ= Seff 0. 15 km 2 sr detector distances L bin=100 km Saclay 25/01/06 M. Iori , Roma 1 20

Rate calculation Assuming a neutrino energy spectrum following a power law Φ(E) ~ 10 -6/E 2 (Gev-1 cm-2 s-1 sr-1) ∫Φ(E)d. E ~ 3 x 103 (km-2 yr-1 sr-1) integrated between 108 – 1011 Ge. V N τ shower = ∫Φ(E)Pdet τau(E)d. Edtd. AdΏBr= 50 x 0. 15 x 0. 64 ~ 4. 8 (km-2 yr-1) or a upper flux limit for a spectrum Φ(E) ~ 10 -7/E 2 *(single tower → 1. 6) **No tau regeneration taken into account Saclay 25/01/06 M. Iori , Roma 1 21

Predictions for Auger and future experiments: Ice-Cube and Nemo Auger fluorescence Bugaev et al. astro-ph/0311086 Guzzo et al hep-ph/0312119 Saclay 25/01/06 M. Iori , Roma 1 22

Results from Corsika 6. 204 tau shower 5 x 1018 e. V particle μ, e γ tau shower density and momentum at the detector level Saclay 25/01/06 M. Iori , Roma 1 23

Determination of Mountain slope β=Θ-90 The slope of the mountain slope, θ angle and L Saclay 25/01/06 M. Iori , Roma 1 24

Performances studies of the array The unit of the array § How many tower? § Angular resolution: 0. 50 (core inside the array) § Duty cycle 100% Ø We can improve the acceptance if we put 2 modules close Ø separate em component by lead layer under study 60 cm lead Saclay 25/01/06 Number hits versus number towers >500 tower no large improvements in density M. Iori , Roma 1 25

From Décor-Nevod at large zenith angle 10 -9 10 -4 Albedo muon intensity Muon intensity O. Saavedra talk at Cosmic Ray Physics Workshop Moscow 17 Oct 2005 10 -5 Bundles ~ 10 -5 m-2 s-1 sr-1 Saclay 25/01/06 M. Iori , Roma 1 26

Trigger 1 To define a trigger we have to remind Ø Electrons do not escape at that energies Ø Muons from νµ do not produce visible signal Ø Neutrino related showers can be easily distinguished from those induced by hadrons in high atmosphere by multiplicity Ø Muon bundles high multiplicity small area ~10 -5 m-2 s-1 sr-1 1 st Level record data bank (TDC charge flag etc. . ) 2 nd Level select up-ward tracks 3 rd Level multiplicity logic with a coincidence gate (~10µs) Saclay 25/01/06 M. Iori , Roma 1 27

Trigger 2 detector Event trigger logic: ØSub-array → µ bundles flag ØEvent by multiple sub-array o Rate Level 1: 10 -2 Hz each tower zenith angle 930 o Rate Level 2: 10 -3 Hz each tower Cut on charge to select up-ward and down ward tracks -> 10 -6 Hz only up-ward tracks o Rate Level 3: NCMRate. L 2 x(Rate. L 2 xgate)M = 10 -24 Hz Sub-array Sinlge tower at 930 measures a flux of ~10 -7 cm-2 s-1 sr-1 Saclay 25/01/06 M. Iori , Roma 1 28

Electronics board Saclay 25/01/06 M. Iori , Roma 1 29

Test performed with a prototype at Jungfraujoch Station (3600 m) Switzerland Field of view Up-going tracks Time of flight Saclay 25/01/06 M. Iori , Roma 1 30

Preliminary results (April 05) from 2 towers Tower 1 Tower 2 950 3 cm Lead in front tile 2 of Tower 2 (yellow plot) ADC tile 1 vs tile 2 Tower 1 Tower 2 1000 Saclay 25/01/06 M. Iori , Roma 1 31

Time resolution central region, -3 ns to +3 ns Signal region, ± 3 ns tile C 1 tile C 2 time resolution Saclay 25/01/06 M. Iori , Roma 1 32

The tiles Saclay 25/01/06 M. Iori , Roma 1 33

Preliminary test results at Jungfraujoch Station Flux measured at Jungfraujoch Station 3600 m P>30 Me. V/c • • Saclay 25/01/06 M. Iori , Roma 1 Decor experiment P>7 Ge. V/c at sea level 34

西部高山后面的宇 活动星系核、 暴、 GZK, TD, Z-burst等 宙CRTNT proposal Z. Cao. M. A. Huang, P. Sokolsky Y. Hu 超高能中微子天 Astro-ph/0411677 电子中微子 μ中微子 文学 中 微 子 振 荡 τ中微子 效应 荧光/契伦科夫光 空气簇射 x 16 AGN event rate: 8~10 event/yr using 16 telescopes Saclay 25/01/06 τ中微子大气荧 光/C光 成像望远镜 3 m M. Iori , Roma 1 2. 5 m 35

Conclusions Conventional horizontal arrays are not optimally adapted to Earth skimming events → a new strategy to be considered is: ü to aim at the horizon and detect the direction of the shower by time of flight Saclay 25/01/06 M. Iori , Roma 1 36

R&D 1 - development of the electronic board with high resolution TDC (ACAM) and MATACQ chip to sample the signal al High Frequency ( 1 GHz sampling) (see http: //www. caen. it/nuclear/product. php? mod=V 1729) 2 - buy 20 -30 PM's and 10 -15 solar Power supplies 3 - design the mechanics to move within 15 deg in zenith each tower and the pointing system 4 - define material to be used to built the detector (alluminium, PVC and possible stress due to gradient of the temperature) 5 - study an other PM reading (fibers? ) to improve the separation of up-ward down-ward tracks 6 - study possible part id (by lead + 5 mm scintillator plate read by fibers+PM 4 ch) and the signal shape to separate PID 8 - Study Trigger by MC, how store the data and its managment 7 - built 10 -15 towers and move them at the final destination to test the system wireless device (GPS-GSM Motorola 12+) and Daq. test the longitudinal as well as transversal time resolution Saclay 25/01/06 M. Iori , Roma 1 37

setup Saclay 25/01/06 M. Iori , Roma 1 38

Cost estimate Scintillating tiles ---- 1000 PMT Hamamatsu low voltage $275 k mechanics $70 k Frame detector, cables $60 k Electronics board $300 k Power supply $100 k GPS-GSM Motorola 12+ $25 k DAQ $100 k installation $100 k total $1030 k Saclay 25/01/06 M. Iori , Roma 1 39

Neutrino cross section R 2 Making measurements at different n θ angle we can estimate σν e u t r Neutrino cross section i n o R 1 • R= Φτ (E)/Φν (E)=∫ 0 L dz∫Nρd. Eν (dσcc /d. Eτ ) e-z. Nρσ • R 1/R 2 ≈ (1 - e-L 1 Nρσ )/ (1 - e-L 2 Nρσ ) • <σ>=ln(R 1 -R 2 )/Nρ(L 2 – L 1 ) Saclay 25/01/06 M. Iori , Roma 1 1017 ≤Eν ≤ 1020 e. V 40

• MC Trigger studies and array • Improve e/mu ID (i. e. preshower) and define final version • Front-end configuration/trigger • Mechanics • Test PMT as function of temperature • Test Solar power system Hip-55172 Sanyo 55 Watts 17 V/3. 3 A • Test on wireless lan devices (ADlink) • Test prototype on site Saclay 25/01/06 M. Iori , Roma 1 41