Wave or particle? Heisenberg, take a hike!

The dual nature of light, electrons, and even atoms has long fascinated physicists. Quantum mechanical rules permit these elementary objects to act both as waves and particles, but scientists can witness only one of those states at a time. Why the curtain falls on half the show has been a matter of debate since the 1920s.

Physicists have long believed that Werner Heisenberg's famous uncertainty principle manages the stage (SN: 2/19/94, p. 118). The act of measuring the position or momentum of a quantum mechanical entity collapses the duality, they say, because the principle forbids both quantities to be simultaneously known with precision.

Now, German experimenters find that the mere existence of information about an entity's path causes its wave nature to disappear. In the Sept. 3 NATURE, they offer experimental evidence that something deeper than uncertainty yanks offstage the wave- or particle-half of the duet. That unseen hand is known as entanglement, or correlations.

The University of Konstanz team passed ultracold rubidium atoms across a pattern of laser light that in essence splits each atom's path, enabling the atoms, wavelike, to simultaneously take divergent routes through an experimental apparatus. As long as no attempt was made to determine which way the atoms went, they emerged as a pattern of light and dark bands, or interference fringes.

That fringe pattern vanished, however, when the researchers turned on microwaves that would orient the spins of the atoms upward if they took one intermediate path and downward if they took the other. The researchers say this spin-flipping so little affected momentum that it shouldn't squelch the atom's wave, according to the uncertainty principle.

Entangling spin with which-way data tilts the wave-particle scales without invoking uncertainty, notes Gerhard Rempe, who led the team. "To explain the wave-particle duality, we need entanglement and correlation," he says. "The Heisenberg uncertainty relation has nothing to do with wave-particle duality."

The finding also deepens the quantum world's mystery, he says, since merely requiring the atoms to carry which-way data was enough to affect their behavior. "There is information stored, but we don't read the information," he says. "This is very strange indeed."

"I think this is an experiment that will be included in future textbooks," says Berthold-Georg Englert of the Max Planck Institute for Quantum Optics in Garching, Germany. It appears to confirm predictions made in 1991 by Englert, Marian O. Scully of Texas A&M University in College Station, and other theorists.

Critics of Englert's analysis acknowledge a subtle effect but disagree that the new experiment sends the uncertainty principle packing. "Correlations are observations about relationships between quantities and do not cause physical processes to occur," asserts Sze Tan of the University of Auckland in New Zealand.

If "we think of the uncertainty principle as more general than simply the inability to simultaneously measure position and momentum," it still applies, he argues.