A foundational postulate of quantum physics has been directly verified by an experiment for the first time. Formulated by Richard Feynman nearly 80 years ago, it has long formed the backbone of much of modern physics.
Quantum particles are notoriously difficult to analyse, as interacting with them affects their properties. But researchers have developed powerful mathematical tools for predicting what a particle is likely to do under particular conditions.
One such tool is the “Feynman path integral”, which posits that the most likely path a quantum particle will take between two points can be predicted by performing a calculation that sums all the possible paths between those points in a special way. Since Feynman proposed it in 1948, his path integral had made its way into textbooks and is now invaluable for many scientists, including physicists who study particles.
Advertisement
Now, for the first time, Feynman’s idea has been tested directly in an experiment.
at South China Normal University and his colleagues achieved this by building on their past work, in which they measured another quantum object that pertains to particles of light, or photons.
This object, known as the propagator, predicts how a photon’s quantum state changes as it travels from one part of the experiment to another – in this case, as it travels through a maze of tiny mirrors, lenses and crystals.
Previously, the researchers had worked out how to measure different properties of photons, such as polarisation, then reconstruct their propagators. Now, they realised that a path integral for a given photon could be constructed by measuring a succession of propagators, then multiplying them together.
Specifically, they measured five propagators for a given photon across the experiment. Because there were so many different trajectories that those propagators could correspond to as the photon moved through the maze, the researchers ended up measuring a total of 1,419,857 distinct paths. They then plugged all of those paths into Feynman’s path integral formula to find out what it predicted about the photon’s behaviour – and found it agreed with what the photon actually did.
“Given the extraordinary success of Feynman’s formulation over nearly eight decades, the result itself was not unexpected. But seeing it work … was still astonishing,” says Zhu. “The familiar phrase ‘sum over all paths’ was no longer just a symbolic instruction in a textbook, we could see its consequences emerge directly from experimental data.”
He says that the experiment was very challenging because the sheer number of tested paths opened the door for errors and experimental noise, which could easily have accumulated and made the final result unintelligible.
“We had to improve almost every aspect of the experiment simultaneously,” says Zhu. The researchers had to significantly improve the precision of their past experiment because, had they used it as is, after the five propagator multiplications, the data would have looked nearly completely random, he says.
at the University of Glasgow in the UK says that the precision of the experiment is a real scientific advance. Given that a photon’s path cannot be traced directly as it moves, the experiment showcases sophisticated quantum measurement and control methods. And it confirms that the foundations of quantum physics are just as researchers have so far taken them to be.
If the team’s measurement had turned up something else, it would have been very surprising and the whole framework of using the path integral for quantum calculations would have crumbled, says Götte.
The new experiment could lead to more complicated tests of the path integral – for example, directly measuring how the different paths add up when photons travel through materials rather than air, he says. “We also hope that researchers from different backgrounds can adapt our experimental approach to their own systems,” says Zhu.
Science Advances
