IceCube cherche à s’étendre

L’expérience IceCube, au pôle Sud, a été une pionnière dans le domaine de l’astronomie des neutrinos. Elle a transformé un kilomètre cube de glace de l’Antarctique en un immense détecteur Tchérenkov, qui détecte la lumière des particules chargées produites lorsque des neutrinos à haute énergie venus du cosmos interagissent avec des noyaux situés à l’intérieur du détecteur. IceCube a détecté les neutrinos les plus énergétiques jamais observés, mais leur origine reste un mystère. Pour cette raison, la collaboration IceCube élabore des plans pour un détecteur plus grand, " Gen2 ", d’un volume instrumenté environ dix fois plus grand que celui d’IceCube, ce qui accélérera considérablement la collecte des données.

The IceCube experiment at the South Pole has been one of the pioneers of the field of neutrino astronomy. During a seven-year-long construction campaign that ended in 2010, the 325 strong IceCube collaboration transformed a cubic kilometre of ultra-transparent Antarctic ice into a giant Cherenkov detector. Today, 5160 optical sensors are suspended beneath the ice to detect Cherenkov light from charged particles produced when high-energy neutrinos from the cosmos interact with nuclei in the detector. So far, IceCube has detected neutrinos with energies in the range 1011–1016 eV, which include the most energetic neutrinos ever recorded (see image of the proposed Gen2 array). However, we do not yet know where these neutrinos come from. For this reason, the IceCube collaboration is developing designs for an expanded "Gen2" detector.

IceCube observes astrophysical neutrinos in two ways. The first approach selects upgoing events by using the Earth to filter out the large flux of cosmic-ray muons. At low energies (below 100 TeV), the measured flux of muon neutrinos is consistent with an atmospheric origin, whereas at higher energies, a clear excess of events with a significance of 5.6σ is observed. The second approach selects neutrinos that interact inside the detector. A total of 54 cosmic-neutrino events with energies ranging from 30–2000 TeV were detected during four years of operation, excluding a purely atmospheric explanation at the level of 6.5σ. Although there is some tension between the results from the two approaches, a combined analysis finds that the data are consistent with an at-Earth flux equally shared between three neutrino flavours, as is expected for neutrinos originating in cosmic sources.

Towards a new detector

Despite multiple searches for the locations of these sources, however, the IceCube team has yet to find any statistically significant associations. Searches for neutrinos from gamma-ray bursts and some classes of galaxies have also come up empty. Although these observations have disfavoured many promising models of the origin of cosmic rays, the ultimate goal of neutrino astronomy is to detect multiple neutrinos from a single source. This requires many hundreds of events, which would take an array of the scale of IceCube at least 20 years to detect.

To speed up data collection, an expanded IceCube collaboration is planning a greatly enhanced instrument (see image of the proposed Gen2 array) with multiple elements: an enlarged array to search for high-energy astrophysical neutrinos; a dense infill array to determine the neutrino properties (PINGU); a larger surface air-shower array to veto downgoing atmospheric neutrinos; and possibly an array of radio detectors targeting neutrinos with energies above 1017 eV. Most importantly, thanks to the clarity of the Antarctic ice, we would be able to increase the instrumented volume of this next-generation array by a factor of 10 without a corresponding increase in the number of deployed sensors – or in the cost. The Gen2 proposal would therefore see an instrumented volume of approximately 10 km3 comprising strings of optical modules, but with improved hardware and deployment methods compared with IceCube.

For the in-ice component PINGU (Precision IceCube Next Generation Upgrade), the Gen2 collaboration is exploring a number of optimised designs for the optical modules, as well as longer strings deployed with improved drilling methods. Photomultipliers (PMTs) with higher quantum efficiency will be used, as is already the case for DeepCore in IceCube, and pressure spheres with improved glass and optical gel will improve sensitivity by transmitting more ultraviolet Cherenkov light. Some designs include more than one phototube per optical module (see image), while more radical concepts envision the addition of long cylindrical wavelength shifters to improve information about the photon arrival direction. Many-PMT designs were pioneered by the KM3NeT collaboration, which is proposing to build a cubic-kilometre-sized European neutrino Cherenkov telescope in the Mediterranean Sea, but are also attractive to IceCube.

The increased complexity of these approaches would be offset by new electronics, and increased computing power will allow the use of more sophisticated software algorithms that better account for the positional dependence of the optical properties of the ice and the stochastic nature of muon energy loss. This will result in improved pointing and energy resolution of both tracks and showers and better identification of tau neutrinos. IceCube has produced a white paper for the Gen2 proposal (arXiv:1412.5106) that fits well with the US National Science Foundation’s recent identification of multi-wavelength astronomy as one of six future priorities, and a formal proposal will be completed in the next few years.

Physics in order

PINGU will build on the success of DeepCore in measuring atmospheric neutrino-oscillation parameters. It consists of a dense infill array in the centre of DeepCore with a threshold of a few GeV, allowing the ordering of the neutrino masses to be determined by matter-induced oscillations of the atmospheric neutrino flux. By precisely measuring the oscillation probability as a function of neutrino energy and zenith angle, PINGU will be able to determine which neutrino is lightest.

Like the present IceTop (a surface air-shower array that covers IceCube’s surface), an expanded surface array will tag and veto downgoing atmospheric neutrinos that are accompanied by cosmic-ray air showers. Current Gen2 designs envision a 75 km2 surface array that would allow IceCube to collect a clean sample of astrophysical neutrinos over a much larger solid angle, including the galactic centre. It will also result in much improved cosmic-ray studies and more sensitive searches for PeV photons from galactic sources. To study the highest-energy (above typically 1017 eV) neutrinos, Gen2 may also include an array of radio detectors to observe the coherent radio Cherenkov emission from neutrino-induced showers. Radio detection is now pursued by the ARA (the Askaryan Radio Array at the South Pole) and ARIANNA (located on Antarctica’s Ross Ice Shelf) experiments, but coincident observations with IceCube Gen2 would be preferable.

Of course, IceCube is not the only neutrino telescope in town. ANTARES has been taking data in the Mediterranean Sea since 2008 and will be followed by KM3NeT (CERN Courier March 2016 p12), while the Gigaton Volume Detector (Baikal-GVD) is currently being built in Lake Baikal, Russia (CERN Courier July/August 2015 p23). Seawater, lake water and Antarctic ice present different challenges and advantages to cosmic-neutrino observatories, and sites in the Northern Hemisphere benefit because the galactic centre is below the horizon. While we all benefit from friendly competition and from sharing R&D resources, size has undeniable advantages. IceCube-Gen2, should the project go ahead, will be larger than any of the proposed alternatives, and is therefore well placed to write the next chapter in neutrino astronomy.