Data from a massive neutrino experiment show that the elusive subatomic particles must literally be of two mutually exclusive types at once—poking a hole in our intuitive sense of reality. The result is bedrock quantum mechanics. But it's the sort of thing typically shown with highly controlled quantum optics experiments and not with nearly undetectable neutrinos.
"If you had told me 10 years ago that we would use neutrinos to study quantum foundations, I would have said that you'd been smoking something very exciting," says Andrew White, a physicist at the University of Queensland, St. Lucia, in Brisbane, Australia, who was not involved in the work. "The result is utterly unsurprising and yet utterly attractive because it tells us that there's a new system for testing quantum foundations."
According to quantum theory, minuscule things behave nothing like everyday objects. Unlike an apple, a subatomic particle can be in two places or of two different types at once. Those two-way "superposition" states are fragile, however. Measure, say, a particle of light or photon that is simultaneously polarized both horizontally and vertically and it will randomly "collapse" one way or the other.
Still, according to quantum theory, the photon's polarization doesn’t exist until it's measured. Albert Einstein disdained that idea, arguing that a physical property of an object has to be "an element of reality" that exists independently of measurement. To salvage "realism," some physicists argued that the result of such a measurement is predetermined by some "hidden variable" within the photon.
In 1964, U.K. theorist John Bell devised a way to test that notion. In quantum theory, two photons in two-way states can be "entangled" so that a measurement on one instantly determines not only its polarization but that of the other photon as well, even if it's light-years away. That quantum connection produces correlations between the particles that are stronger than hidden variables allow, Bell showed. Last year, physicists in the Netherlands and the United States performed the best demonstrations yet of those correlations, nixing such hidden variables.
The test with neutrinos involves correlations between measurements separated not in space, but in time. In 1985, theorists Anupam Garg, now at Northwestern University, Evanston, in Illinois, and Anthony Leggett of the University of Illinois, Urbana-Champaign, considered repeated measurements of a single quantum system: a ring of superconductor in which an unquenchable current flows one way or the other. The ring mimics a coin, which can be heads or tails, except that current can also flow both ways at once.
According to quantum theory, the current will oscillate between the two directions. So a measurement will reveal it flowing, say, clockwise, with a probability that depends on the time. Leggett and Garg found that certain correlations among three or more measurements would be stronger than classical physics allows—if the current has no direction until it's measured.
Experimenters have approximated the Leggett and Garg test. In 2011, White and colleagues demonstrated the extrastrong correlations in quantum optics, although in an average way and not with a single photon. Now, Joseph Formaggio, a neutrino physicist at the Massachusetts Institute of Technology in Cambridge, and colleagues provide a demonstration using data from the Main Injector Neutrino Oscillation Search (MINOS) experiment at Fermi National Accelerator Laboratory (Fermilab) in Batavia, Illinois, which fires neutrinos at near-light-speed 735 kilometers to a 5.4-kiloton detector in the Soudan Mine in Minnesota.
Neutrinos come in three flavors that morph into one another. Those fired from Fermilab start as so-called muon neutrinos and "oscillate" mainly to electron neutrinos in a process that resembles the one analyzed by Leggett and Garg. MINOS experimenters didn't repeatedly measure individual neutrinos, as detecting a neutrino destroys it. However, each neutrino starts in the same state whose evolution depends only on the time since it left Fermilab. So measuring many neutrinos was equivalent to measuring the same one repeatedly.
The MINOS physicists also didn't measure the neutrinos at different distances from Fermilab, so Formaggio and colleagues couldn't directly compare measurements made after different flight times. However, the rate at which neutrinos oscillate varies with their energy, with the clock ticking faster for more energetic neutrinos. So instead of looking for correlations between neutrinos measured at different times, Formaggio and colleagues looked for equivalent correlations in the number of muon neutrinos arriving in Minnesota with different energies.
The researchers observed the strong correlations predicted by Leggett and Garg, as they report in a paper in press at Physical Review Letters. "As we expected, it's a very obvious effect," Formaggio says. The data underscore that the neutrino has no flavor until it's actually measured, he says.
The result is not surprising, Garg says, as neutrino oscillations are inherently quantum mechanical. Still, he says, it "probes the conflict between the quantum and classical worlds in a new regime."
Next would be to see whether neutrinos can test quantum mechanics in other ways, Formaggio and White say. Garg says he still hopes somebody will push the test he and Leggett devised as originally intended: to test whether realism holds for a truly macroscopic object. If it fails, our sense of reality really would go out the window.