It’s one of the strangest concepts in the already strange field of quantum physics: Measuring the condition or state of a quantum particle like an electron can instantly change the state of another electron—even if it’s light-years away. That idea irked the likes of Albert Einstein, as it suggests that something can travel faster than light and that reality is somehow determined by the measurements we make. But now, a team of experimenters says it has clinched the case for this concept, sealing up loopholes in previous demonstrations.

"This is an absolute landmark paper in quantum physics," says Howard Wiseman, a physicist at Griffith University, Nathan, in Australia, who was not involved in the work. "There can no longer be any reasonable doubt that the physical world is profoundly different from our everyday intuitions." Christopher Ferrie, a physicist at the University of Sydney in Australia, notes that for many physicists, the issue was settled long ago. "Poll any physicists of my generation or later and they will be completely unfazed by [it]," he says. The real advance, he says, is in opening the way for ultrasecure quantum communications technologies.

The experiment was performed by Ronald Hanson, a physicist at Delft University of Technology in the Netherlands, and colleagues. Hanson declined to discuss the paper, which is posted on the arXiv preprint server, as it's under review at an undisclosed journal.

The experiment involves a concept called entanglement. Consider an electron. Like a top, it can spin in one direction (up) or the other (down). Bizarrely, quantum theory says that the electron can also spin equally both ways at once—although if you measure it, the quantum state will "collapse" so that you'll find the electron spinning either up or down with equal probability. How such a measurement is made is important. According to quantum theory, you can’t simply read the spin directly; you have to use an analyzer that can be set, a bit like a dial, to a particular orientation to see whether the electron is spinning that way or the opposite way. In the case of the both-ways spin, setting the analyzer vertically leads the electron to collapse into the 50-50 result.

Even weirder, two electrons can be entangled so the spin of each electron is completely uncertain, but the two spins are completely locked together and correlated. Suppose then that Alice and Bob share two entangled electrons and each has an analyzer set vertically. If Alice measures her electron and finds it spinning up, she knows instantly that Bob's is spinning down, even if he's a galaxy away. That "spooky action at a distance" bothered Einstein, as it suggests the quantum wave describing the electrons collapses at faster-than-light speed. It also suggests that the "reality" of an electron’s spin state—what is knowable about it—isn't determined until the electron is measured and the quantum wave collapses.

Einstein found this idea unpalatable. He argued instead that quantum mechanics was incomplete—essentially, that "hidden variables" encoded in each electron but outside the scope of the theory determine the results of Bob's measurements. That concept obviates faster-than-light collapse, because the determining factor travels along with Bob's electron. It also jibes with the notion that measurements reveal some aspect of the world that exists independently of them—just as we assume the color of a tennis ball exists before we look at it.

However, in 1964, British theorist John Bell found a way to test the difference between collapsing quantum waves and hidden variables. According to quantum theory, if Alice and Bob tilt their analyzers to different angles, they should no longer see perfect correlations in their measurements. For example, suppose Alice keeps her analyzer vertical and Bob tilts his by 45°. Then, if Alice finds her electron spinning up, the chance that Bob will find his electron spinning down—defined in his new orientation—is only 71%. Bell imagined that Alice and Bob repeatedly varied the orientations of their analyzers. He proved mathematically that hidden variables would produce correlations weaker than a certain limit—spelled out in a formula called Bell's inequality. Collapsing quantum waves could yield stronger correlations. The formula offered a litmus test for determining whether the hidden variables were really there.

Bell also explained that the faster-than-light collapse of the waves wouldn't necessarily violate relativity's prohibition on faster-than-light travel. Because Alice cannot control the results of her measurements, she cannot use them to send Bob information faster than light. She and Bob can merely confirm the correlations after the fact. That is now the standard interpretation of relativity.

In the 1970s, experimenters began taking measurements designed to see whether Bell’s inequality holds. They consistently found correlations stronger than hidden variables allow. Those results generally convinced physicists that Einstein was wrong. Either quantum waves must indeed collapse faster than light, or the results of measurements could not be predestined by hidden variables: Until an electron spinning both ways is measured, it literally spins both ways.

However, performing an airtight test of Bell's theorem is tricky, and in recent years physicists have fretted over "loopholes" that would allow some effect other than the instantaneous collapse of quantum waves to skew the results. Now, Hanson and 18 colleagues claim to have done the first loophole-free test of Bell's theorem.

To test Bell's idea, physicists must make sure that no influence other than that of the measurements can travel between the electrons in the time it takes to perform the measurements. That's a tall order, as light travels 299,792 kilometers per second. Hanson and colleagues separated the two stations with their electrons by 1.28 kilometers on the Delft campus. That gave them 4.27 microseconds to perform both measurements before a light-speed signal from one station could reach the other.

The researchers still had to entangle the distant electrons. To do that, they first entangled each spinning electron with the state of a photon that they then sent down an optical fiber to a third station between the other two. Only if the two photons arrived simultaneously and interfered with each other in just the right way would the electrons become entangled, through a process called entanglement swapping. Fewer than one out of 150 million photon pairs registered the right interference signal. Still, the researchers could start the measurements on the electrons before the photons met and go through the data afterward to find the trials that worked. In the preprint, they report 245 successful trials in 22 hours of data-taking.

Finally, the physicists have to close the loophole that opens if they can't reliably read the electrons' state. Such a failed measurement could obscure the true correlations between the electrons' spins. To overcome that, Hanson's team used individual electrons trapped in atomic-size defects in diamonds cooled to near absolute zero. In the defects the electrons easily maintain their delicate spin states and can be manipulated with microwaves and light. The physicists measured the spin of each electron with greater than 95% efficiency.

With both loopholes nailed shut, the researchers see a clear violation of Bell's inequality—torpedoing Einstein's hidden variable and vindicating collapsing quantum waves. "The only significant concern one could have about this paper is the small data set, which means the result is not as surely established as one would ideally like," Wiseman says. "But I am sure this will be rectified soon."

It's always possible to dream up even wilder loopholes, Ferrie says. But the experiment closes the ones that might be used to attack certain developing quantum technologies, such as schemes to use entangled particles to securely distribute the keys for encoding secret messages in so-called "device independent quantum key distributions." "This is a huge technical milestone," Ferrie says, "and prerequisite for many future quantum technologies, which are sure to enable the probing and eventual understanding of new physics."

***Correction, 30 August, 9:47 a.m,: **The story has been changed to correct a mathematical mistake in the explanation of Bell's argument.