Drilling through Earth
With their contempt for matter, neutrinos can easily drill through the 13 000 kilometres of the Earth and fly out into space on the far side. Just as a champion motor-racing driver is not immune to being run over when crossing the road, even after surviving transit through the Earth there is always a probability that neutrinos will interact in the next obstacle they reach. Thus a few neutrinos coming up through the Earth are finally absorbed in the SuperKamiokande detector.
Even more intriguing than the deficit of muon neutrinos coming down from the sky is the marked difference between the signals due to downward neutrinos reaching detectors from the atmosphere, and upward neutrinos having passed through the Earth.
The interpretation is that the neutrinos "oscillate". Although they slip right through the Earth, they do not emerge entirely unscathed, and have their electron, muon, or tau allegiance changed en route.
Results from other neutrino experiments under very different conditions (January, page 5) appear to rule out electron-to-muon oscillations, leaving muontau as the most likely oscillation culprit neutrinos entering the Earth as muon-type leave it as tau-type.
To pin down the details of this effect needs continued systematic study of tau neutrinos. At 1777 MeV, the tau particle is much heavier than the muon (105 MeV) and the electron (0.5 MeV). So investigating tau neutrinos and tau particles is helped by having a neutrino beam from a particle accelerator with enough energy for its particles to interact and produce taus.
Neutrinos have traditionally been considered to be massless, spinning left-handedly (anticlockwise around their direction of motion) through the cosmos at the speed of light. However, the possibility of neutrino oscillations suggests this might not be 100% accurate. To oscillate, neutrinos must have some mass.
The oscillation probability of neutrinos depends on the quantity Dm2 L/E, where Dm2 is the squared mass difference between the oscillating neutrinos, L is the "baseline" (the distance between the site of neutrino production and detection), and E is the energy. Dm2 can be very small the new results suggest in the region of 103 eV2, corresponding to mass differences of a few hundredths of an electronvolt.
For each type of oscillation, a large range of oscillation probabilities has to be explored. For the muontau channel, the pioneer Chorus and Nomad tau neutrino experiments at CERN, with a baseline of about 1 kilometre, see no evidence for oscillation. Taking the new results from SuperKamiokande and other detectors at face value, this is hardly surprising to reveal such small mass differences needs long baselines.
The high-energy neutrino beams from CERN and from Fermilab can explore this unexplored oscillation territory. Both, strangely enough, have underground neutrino detectors 730 kilometres away: for Fermilab the Soudan mine in Minnesota, for CERN the Italian Gran Sasso laboratory.
In Japan, the K2K project to fire a 1.4 GeV neutrino beam from the KEK laboratory towards the SuperKamiokande underground detector 250 kilometres is nearing completion. This could produce indirect evidence for oscillations via the disappearance of muon neutrinos.
Fermilab and Soudan are the scenes of a major effort for the MINOS (Main Injector Neutrino Oscillation Search September 1996, page 20) using 10 GeV range neutrinos, while CERN, whose neutrinos currently point north-west, is putting the finishing touches to a plan to point 20 GeV range neutrinos in roughly the reverse direction and shoot them towards Gran Sasso. (In his original 1979 Gran Sasso proposal,
Antonino Zichichi had already pointed out such a possibility, and the experimental halls are deliberately oriented towards CERN.)
For the detailed project plan now drawn up, a proton beam from the SPS synchrotron with an energy of up to 450 GeV would be focused on a target to produce pions and kaons, which would then be magnetically focused using a horn and reflector to point the pions and kaons in the required direction once produced, the electrically neutral neutrinos cannot be steered. After about 1000 metres, many of these pions and kaons will have decayed, producing the required neutrinos, and the remaining hadrons absorbed by a beam stop (which the neutrinos easily penetrate).
Gran Sasso is already the home of the ICARUS detector using a liquid-argon time projection chamber and originally foreseen as a solar neutrino detector.
To investigate a detector scenario at Gran Sasso for CERN neutrinos, a joint meeting of CERN's SPS Experiments Committee and the Gran Sasso Scientific Committee was organized for early November at CERN (October, page 5).