Click here for free access to our latest coronavirus/COVID-19 research, commentary, and news.

Support nonprofit science journalism

Science’s extensive COVID-19 coverage is free to all readers. To support our nonprofit science journalism, please make a tax-deductible gift today.

Splitsville. The basic setup for the experiment in which a neutron follows one path and its magnetism another.

Splitsville. The basic setup for the experiment in which a neutron follows one path and its magnetism another.

Adapted from Ernecker/Creative Commons

Physicists Produce Quantum Version of the Cheshire Cat

In Lewis Carroll's famous children's novel Alice's Adventures in Wonderland, Alice meets the Cheshire Cat, which disappears and leaves only its grin behind. Now, physicists have created a quantum version of the feline by separating an object—a neutron—from its physical property—its magnetism. The experiment is the latest example of how quantum mechanics becomes even weirder using a technique called weak measurement and could provide researchers with an odd new experimental tool for performing precision measurements.

In quantum physics, tiny particles can be in opposite conditions or states at the same time, a property known as superposition. For instance, an electron can literally spin in opposite directions simultaneously. Try to measure the spin, however, and that state will "collapse" so that the electron is found spinning one way or the other. That's because quantum theory generally forbids you to measure a particle's state without altering it—at least ordinarily.

But in 1988, Yakir Aharonov, a theorist at Tel Aviv University in Israel, and colleagues dreamed up a way to measure delicate quantum states without disturbing them through so-called weak measurements. There's a price to pay, of course. A weak measurement can't reveal anything about an individual particle, but only the behavior of many particles all in the same state. And it requires not only putting the particles in just the right state to begin with, but also picking only those in a specific different state in the end, so the whole experiment has to be analyzed retrospectively. Nevertheless, weak measurements can probe phenomena that ordinary measurements can't, and last November Aharonov and colleagues described how they could be used to realize a quantum Cheshire Cat.

Here’s the idea. A beam of neutrons all magnetized in the same direction, say right, enters a device called a neutron interferometer (see diagram). The beam strikes a beam splitter, which splits not only the macroscopic beam but also the quantum wave describing each neutron. So after the beam splitter, each neutron is in the bizarre quantum state: in path 1, polarized right, and in path 2, polarized right. This is the "preselected" state. After taking different paths, the waves recombine at the second beam splitter and interfere with each other so that the neutrons all exit the interferometer through one of two "ports," the light port.

Now, here's where things get weird. Experimenters install a few gadgets before the second beam splitter that work like a filter so that if a neutron is in the state in path 1, polarized right and in path 2, polarized left—the "postselected state”—it will come out the dark port instead. That may sound superfluous, because each neutron is not in that state. However, the two states have a common part—in path 1, polarized right—and that overlap ensures that some neutrons emerge from the dark port, just by virtue of trying to filter out this postselected state.

If you look at only these postselected events, you can say for sure that the neutron went through path 1. That's because the only parts of the preselected and postselected states that overlap are the ones for path 1. On the other hand, if you try to measure the magnetism, you'll find that all the magnetism is in path 2. That's because to know the magnetism is there, you essentially have to apply a magnetic field that flips the neutron’s polarization. So after the measurement, the parts of the altered preselected state and postselected state that are identical are the ones for path 2.

The traditional interpretation is that the whole argument is moot. If you reach into path 1 with a neutron detector, then that measurement alters the original quantum state, making it pointless to speculate about what you would have seen if you'd measured magnetism in the path 2 instead, and vice versa. According to Aharonov’s theory though, the measurements could be done weakly, so that they would not alter the neutrons' state. And that's exactly what Yuji Hasegawa of the Vienna University of Technology and colleagues have done, as they report in a paper posted to the arXiv preprint server.

Using a neutron interferometer at the Institut Laue-Langevin in Grenoble, France, the researchers inserted an absorber that soaked up only a few percent of the neutrons—not enough to ruin the interference of the waves. When they put it in path 2, the rate of neutrons leaving the dark port remained the same. When they put it in path 1, the number decreased, proving that the neutrons in the postselected state go through path 1. Then, they applied a small magnetic field to slightly rotate the neutrons’ polarization and perturb the interference pattern. When the field was applied to path 1, it had no effect. But in path two, the number of neutrons exiting the dark port changed, proving the neutrons' magnetism was all in path 2. Thus the cat—the neutron—was separated from its grin—its magnetism.

The experiment will “surely help us understand better the counter-intuitive nature of quantum phenomena,” says Sandu Popescu, a theorist at the University of Bristol in the United Kingdom who was not involved in the experiment. The odd quantum phenomenon might even prove useful for making better precision measurements, he says. Some physicists have been testing whether Newton's law of gravity remains correct at distances shorter than a millimeter or so; the delicate experiments can be muddled by extraneous electromagnetic effects. But if researcher could split the mass of neutrons from their magnetism, then they might be able to study gravitational effects without being disturbed by electromagnetic ones, says Aephraim Steinberg, an experimenter at the University of Toronto in Canada.