Supernova Neutrino Detection with IceCube

Transcription

1 Supernova Neutrino Detection with IceCube Reina Maruyama Supernova Physics and DUSEL: UCSD/UCLA workshop September 16-17, 2009

2 The IceCube Neutrino Observatory 1km 3 instrumental volume 59 Strings deployed in 5 yrs 86 Strings by strings ~ 125 m apart 60 DOMs/string at 17 m vertical spacing 6 special strings, 62 m apart, 7 m vertical spacing (high QE PMTs) Deep Core: 6 high-qe + 7 nearest standard strings 50 m 1450 m 2450 m 2820 m IceTop 160 Tanks,2 DOMs per tank 2009: 59 strings in operation 2011: Project completion with 86 strings IceCube In-Ice 86 Strings 60 DOMs per string AMANDA-II: (Precursor to IceCube, shut down in 2009) Deep Core 6 strings (+7 surrounding strings) Optimized for low-energies

3 125 m

4 The IceCube collaboration USA: Bartol Research Institute, Delaware University of California, Berkeley University of California, Irvine Pennsylvania State University Clark-Atlanta University Ohio State University Georgia Institute of Technology University of Maryland University of Alabama, Tuscaloosa University of Wisconsin-Madison University of Wisconsin-River Falls Lawrence Berkeley National Lab. University of Kansas Southern University and A&M College, Baton Rouge University of Alaska, Anchorage Sweden: Uppsala Universitet Stockholm Universitet UK: Oxford University Netherlands: Utrecht University Switzerland: EPFL Germany: DESY-Zeuthen Universität Bonn Universität Mainz Universität Dortmund Universität Wuppertal Humboldt Universität MPI Heidelberg RWTH Aachen Ruhr-Universität Bochum Belgium: Université Libre de Bruxelles Vrije Universiteit Brussel Universiteit Gent Université de Mons-Hainaut Japan: Chiba University 34 institutions, approx. 250 members New Zealand: University of Canterbury

5 Standard Event Topologies Topologies

6 Moon shadow 5 months of IC40 Moon max. altitude at the South Pole (2008): 28 Median primary cosmic ray energy: 30TeV Deficit: 5 σ (~900 events of ~28000) - consistent with expectation. Verification of angular resolution and absolute pointing. More statistics will allow study of angular response function Preliminary

7 Cosmic ray large scale anisotropy Data: IC22 4.3x10 9 events Median angular resolution: 3 o Median primary cosmic ray energy: 12 TeV. IceCube skymap is consistent with northern hemisphere observations from Tibet Array and Milagro

8 Search for point sources 40-string(6month) all-sky results days livetime Preliminary events: 6796 up-going down-going Hottest location in the all-sky search: r.a.=114.95, dec.=15.35 Pre-trial -log 10 (p-value) = 4.43 Best-fit # of source events = 7.1 Best-fit spectral index = 2.1 No excess found! all-sky p-value is 61%, not significant

9 WIMP Searches IceCube + Deep Core Solar WIMP search will probe large region of allowed phase space in the SD scattering cross section Abbasi et al., Phys. Rev. Lett. 102, (2009) arxiv: limit for IceCube + Deep Core (prelim) These are models with low SI cross sections and difficult for direct detection experiments SI cross section well constrained by direct detection experiments Requires models of solar dark matter distributions, annihilation model hard W + W -, soft bb

10 IceCube performance Low noise rates: ~280Hz (SPE/sec) Rate with correlated pulses ~500Hz Supernova detection High duty cycle: >96% Event rates (59 strings) Muons: ~1.5 khz Neutrinos: ~160/day Strings Year Livetime μ rate ν rate IC days 80 Hz 1.7 / day IC days 550 Hz 28 / day IC ~365 days 1000 Hz 110 / day IC ~365 days 1500 Hz 160 / day IC86* 2011 ~365 days 1650 Hz 220 / day

11 IceCube as MeV ν detector first proposed by Halzen, Jacobsen & Zas, astro-ph/ DOM If there is a supernova nearby: ice will be uniformly illuminated by large number of neutrinos > 500,000 hits /15 sec for SN1987A-like event at 10 kpc detect correlated rate increase on top of PMT noise Disadvantage: no pointing no event-by-event detection Advantage: high statistics (0.25% stat. error) low dark noise rate (~280 Hz)

12 Detecting MeV neutrinos in IceCube Main detection is inverse beta decay: ν ep ne + Other channels: νee, ν ee, νxe, ν xe, νeo, ν eo For every neutrino interacting in the ice, at most 1 photon is detected by a DOM Single photon hits counted for each DOM Perform online analysis Store all DOM rates in 0.5 s resolution continuously Search for collective rate fluctuation in 0.5, 4, & 10 sec Store Supernova candidate in 2 ms resolution Send alarms to SNEWS

13 Digital Optical Module

14 Digital Optical Module, a.k.a DOM Hamamatsu 10 inch PMT Flasher Board PMT base Main board 2x 300MHz waveform digitizers 1x 40 MHz FADC digitizer Trigger in coincidence w/ neighbor DOMs Transmits data to surface on request Data sent over 3.3 km twisted pair copper cable Knows the time to within 3 ns to all other DOMs in the ice Power comsumption: 3W Low Noise: ~ 280 Hz Large dynamic range 1000 pe/10 ns 10,000 pe/1 us Clock stability: nsec / sec Synchronized periodically to precision of O(2 nsec) 33 cm Benthosphere 14

15 Capturing Waveforms with the DOMs ATWD x16 x2 x MHz 10-bit flash ADC for slow high energy events 2 parallel Analog Transient Waveform Digitizer (ATWD) chips with 10-bit resolution and sampling speeds programmable from 250 MHz to 1 GHz Each ATWD contains 3 gain paths: x16, x2, x0.25 looking at the PMT input and giving an effective 14-bits of resolution to span the PMT dynamic range Reprogrammable from surface Reina Maruyama SN at DUSEL September 16-17, 82009

16 DOMs in Production DOM testing freezer Modular Dark Freezer Lab DOM Assembly DOM sealing station

17 SN Data Handling Each DOM stores hits in ms scaler bins, the scaler values are sent to surface ~every 1 sec (600+ bins). Counts from all DOMs are reordered and rebinned into global 2 ms bins. Near-real time search for collective rate fluctuations in 0.5 s, 4 s, and 10 s Limited data transfer rate: each supernova candidate event is stored in 2 ms bins, the rest in 0.5 s DOM clocks synchronized to < 2 ns, absolute time given by GPS with utc time stamp given to bins. Data sent over satellite for later analysis Uptime in 2008: 94% (better in 2009) pdaq: 97%, SNi3DAQ: 97%

18 Noise Rate of DOMs Correlated noise from scintillation in glass caused by K/U/Th in the glass Dark Noise Rate in Ice for 6.4 usec deadtime Standard DOMs High QE DOMs 250 μs artificial deadtime ~ 280 Hz

19 Effective Volume and Optical Sensitivity M. Ackermann et al.(2006) J. Geophys. Res., 111, D13203 Preliminary Effective volume depends on absorption length in ice absorption length for 350 nm: ~ m average eff. volume per photon per DOM: ~ 0.19 m DOMs Deep Core DOMs IceCube is a 2.5 Mton detector for e + for 15 MeV ν e

20 IceCube as MeV ν detector /0.5 s accretion phase Preliminary Helmholtz cooling phase Supernova at 7.5 kpc Average rate increase/dom : 13 Hz (0.7σ) Average noise rate/dom: 280 Hz Noise rate for 4800 DOMs: 1.3 x 10 6 ± 1.8 x 10 2 Hz Collective rate increase: 6100 Hz (34σ) Simulation based on a numerical Livermore model, normalized to SN1987A at 7.5 kpc Totani, Sato, Dalhed & Wilson, ApJ 496 (1998) 216 See also: Dighe, Keil & Raffelt, hep-ph/

21 Looking for core collapse SN IceCube AMANDA center of Milky Way 5σ signal for SN of 1987A strength

22 Sources of Background Anything that raises the collective noise rate can trigger our detector e.g. seismic activity, muon rates, magnetic field? Require that the entire detector is illuminated to eliminate spurious effects, but we still have fluctuations: muons are the biggest source AMANDA 2002 Background simulation with simple noise rate fluctuations including fluctuations from muons

23 Neutronization Burst, a Standard Candle? Peak rather independent of mass of star neutrino transport nuclear equation of state 25m 11m Peak strongly dependent on MSW effect in supernova θ 13 neutrino mass hierarchy Else, if neutrino mixing matrix known: excellent method to measure distance with <10% accuracy Kachelriess,Tomàs, Buras,Janka, Marek & Rampp,astro-ph/

24 Can IceCube see the deleptonization peak? Consider the two possible hierarchies (sin 2 Θ13 > 10-3 ): Preliminary Normal Hierarchy Inverted hierarchy expected signal without oscillations Supernova at 7.5 kpc Supernova at 7.5 kpc Only MSW included Detection may depend on details of the onset of the accretion phase & oscillations

25 High Quantum Efficiency DOMs More than 4000 sensors with standard 10 PMT (R ) integrated and tested in IceCube 400+ high quantum efficiency PMT (10 ) tested with IceCube standard production test program. Result: Quantum efficiency ~38% higher (405 nm, -40C), ~ 30% higher noise rate in ice. No problems found Low temperature (-40C) noise behavior scales with quantum efficiency as expected. 100 deployed last season, ~320 will be deployed this season. Optical efficiency from freezer tests Standard DOMs High QE DOMs Dark Noise Rate in Ice for 6.4 usec deadtime Standard DOMs High QE DOMs

26 High QE DOMs: Noise vs temperature Noise rate of the high-qe DOMs were measured in our test freezer facility shows qualitative agreement with expected behavior. A few caveats... Measurements have been made of the noise rate of the DOM, not the PMT. Preliminary The resulting data are upper limits and at low temperatures they are dominated by noise from the glass pressure housing. in freezer Will provide a pretty good estimate on the PMT noise dependence on temperature once the glass noise is subtracted. Freezer temperature shown. PMT is a few degrees warmer Study on-going

27 Summary Construction of IceCube in full gear (59 / 86 deployed) highly reliable, low noise modules (280 Hz with 254 μs dead time) IceCube is sending alerts to SNEWS Sensitivity of IceCube to MeV neutrinos unabiguous detection from galactive supernovae with 2 ms time resolution 5 sigma sensitivity at 50 kpc detection capability of neutronization peak will depend on the details of the accretion phase and neutrino self-scattering Series of high QE DOMs were tested for noise rate and optical characteristics. Next: Addition of the Deep Core high QE DOMs to the IceCube s supernova sensitivity studies. Earth-matter effects with other detectors? self-scattering?