Image: NIST physicist Krister Shalm with the photon source used in the Bell test that strongly supported a key prediction of quantum mechanics: There are in fact spooky actions at a distance. Credit: J. Burrus/NIST
Physicists have proven that it’s possible for particles to influence each other instantaneously at a distance, ruling out Albert Einstein’s view that objects can only be influenced by their immediate surroundings.
Krister Shalm, an alumnus of CIFAR’s Global Scholars program, led the project at the National Institute of Standards and Technology with support from collaborators including CIFAR Fellow Thomas Jennewein (University of Waterloo) group.
Their results are the best evidence yet that the description of reality given by quantum mechanics is true. The experiment relied on two principles of quantum physics. First, a property of a particle is inherently probabilistic, and can’t be known until it is measured. And second, despite this, two particles can be “entangled” so that a physical property of one is correlated with a property of another.
Taken together, these principles suggest that the particles can somehow influence each other instantaneously, even at great distances. (Despite this instantaneous influence, it is impossible to use such entanglement to send messages or communicate faster than the speed of light).
Many experiments have shown this influence exists. But the results are so counter-intuitive that people have suggested there must be loopholes. Maybe the particles are somehow communicating with one another. Maybe the measurement isn’t being done randomly. Maybe particles aren’t being sampled fairly for measurement.
The experiment seems to close all of those loopholes. The experimenters began by stimulating a crystal to produce many pairs of photons, and then sending each one to a different detector, located more than 100 meters away from the source and from one another. While the photons were in flight, each detector randomly–and independently–chose how to measure the polarization of the photons.
Over many repetitions, scientists found that both polarizers fired together in a way that is consistent with what quantum mechanics predicts for entangled particles. However, it is extremely unlikely that this would happen if each photon was operating independently, with no influence on its partner — about a 1 in 170 million chance.
Because of the speed of the measurements and the distance between devices, no signal traveling at the speed of light could have passed between the photon pairs. And because the measurements were made at random, there is no way that the measuring device was influenced.
After months of hard work building the experiment, the collaborators were thrilled to see it was working, even before they had analyzed the data.
“I was in the main lab and I got a call and they said ‘We think it’s working,’ and I took off down the hall sprinting as fast as I could and careened into the command centre,” Shalm says, “We were so excited.”
With the results and through further testing, the researchers eliminated the possibility that a photon was being controlled by something within its local environment instead of its distant pair. The new finding is consistent with the theory of quantum mechanics.
Albert Einstein famously doubted quantum mechanics and argued physics needed better models, saying that “physics should represent a reality in time and space, free from spooky actions at a distance.” This experiment did not prove that quantum entanglement exists, but it showed that there can be spooky actions at a distance. These could be caused by entanglement, but that hasn’t been proven yet.
“Quantum mechanics predicts that there are these actions at a distance, and in our experiment we were able to rule out the types of models that Einstein was advocating for,” Shalm says.
“I am excited to see this new experiment finally providing some closure to this question” said Jennewein, who over the past 20 years was involved in several experiments addressing some of these loopholes. “I am convinced that 2015 will be remembered as the year of the loophole-free Bell test”.
The researchers have submitted the paper to Physical Review Letters.
Shalm says the NIST setup is particularly exciting for applications that require random numbers, such as in secure communication technology.
“You can generate random numbers that are certifiable,” Shalm says. “This sounds strange, because how hard is it to come up with something random, but it is an outstanding problem. So many of our security applications rely on good sources of randomness.”
Phase two of the project is to create a type of ‘randomness beacon’ at NIST, that can provide service to those in need of reliably random numbers.