Einstein was wrong. So, in essence, begins many a screed by some poor crackpot who aspires to be a physicist. Still, many legitimate physicists are willing to entertain the possibility that, in formulating his theories of relativity, Einstein didn't get it exactly right. Some are looking for slight deviations from the prediction of his theory that might point to some deeper understanding of the universe. Alas, they'll have to search even harder than before: A new experiment tests the stretching of time predicted by Einstein's special theory of relativity and finds it spot on the money to 1 part in 10 million.
Special relativity joins space and time into a single weird thing called spacetime that appears different to observers moving relative to each other. Imagine you hold a firecracker in each of your outstretched hands and you see the two go off simultaneously. (Ouch!) Bizarrely, another person zooming by at near-light speed will see the firecrackers explode at different times. (She'll also claim your arms are shorter than you say they are.) In much the same way, a clock flying at near-light speed would tick more slowly than the watch on your wrist, and a person who travels through space in a high-speed rocket would appear to age unusually slowly to someone stuck on Earth.
Such "time dilation" seems preposterous. But in 1907, Einstein proposed an experiment to test the idea. Atoms and ions give off light of specific colors. Light is a kind of rippling wave, and the rate of rippling can be viewed essentially as the ticking of a clock. Einstein reasoned that if the ion were accelerated to near-light speed, time would slow for it and, hence, so would the rippling of the emitted light waves. That is, the speeding ions would emit light of a lower frequency.
Now, Gerald Gwinner, a physicist at the University of Manitoba in Winnipeg, Canada, and colleagues have tested time dilation in just this way. Using a ring-shaped accelerator at the Max Planck Institute for Nuclear Physics in Heidelberg, Germany, they accelerated lithium ions around a ring-shaped accelerator to 6% of the speed of light and used lasers to tickle them and make them radiate. The researchers then measured the frequency of the emitted light and found that, to within a precision of 1 part in 10 million, it had slowed just as the special theory of relativity predicts, the researchers report this month in Nature Physics. By using faster ion beams and sophisticated techniques for observing emitted light, the researchers verified relativity to 100,000 times greater precision than the first experiments in 1938 and 10 times better precision than any earlier result. The new experiment doesn't prove Einstein was exactly right--that kind of proof isn't possible--but it puts a very tight limit on how wrong special relativity might be, Gwinner says. "We can only push the limit."
The experiment provides a particularly useful new datum, says Alan Kostelecky, a physicist at Indiana University, Bloomington, who developed the theory that aims to incorporate all possible violations of special relativity. Known as the Standard-Model Extension, the theory contains 19 parameters, or "coefficients," that allow for wiggle room. "They have placed an improved bound on a coefficient that is particularly difficult to measure," Kostelecky says. Any theories that move beyond special relativity would have to agree with special relativity to within this very tight margin. So, to improve their credibility, the crackpots should start their manuscripts with, "Einstein was not quite right."