A new optical clock that is insensitive to an important source of noise has been developed by physicists at the National Institute of Standards and Technology (NIST) in the US. The researchers believe the new design, which allows the clock to reach its peak precision much more quickly than before, could provide a step towards allowing optical clocks to be used in a wider range of applications than is possible today.

Optical lattice clocks trap atoms in a standing-wave potential created by two counter-propagating laser beams. A third laser is used to repeatedly excite and de-excite a specific atomic transition, which gives the "ticks" of the clock. Such a clock's principal advantage over a similar timekeeper based on trapped ions is that technical difficulties currently prevent more than one ion being used at a time. This makes ion clocks prone to the inherent quantum randomness in the way the ion behaves when excited by the laser – called the quantum projection noise. In contrast, thousands of neutral atoms can be used in the same trap at once and this greatly reduces the quantum projection noise.

In 2013, Andrew Ludlow and colleagues at NIST in Boulder, Colorado, demonstrated two optical lattice clocks that are stable to within a record-breaking half a second in the age of the universe. Physicists have proposed that, if such clocks could be made robust enough to be taken out of the laboratory, the precise measurements of time dilation taken at various points on Earth could give important insights into the internal composition of our planet. Taking such clocks to space could allow physicists to look for deviations from Einstein's general theory of relativity and quantum effects in gravity.

Billiard balls

"Dick noise" is an important effect in optical clocks and it arises because the atoms cannot be monitored continuously. "They'll typically stay around [in the trap] for a few seconds before molecules of background gas bump into them and knock them out like billiard balls," explains Ludlow, "and so we have to get some more." During the "dead time" while they do this, the laser frequency can vary slightly. The effect of these random variations can be averaged down by measuring for many hours, but this is experimentally cumbersome.

Researchers have tried to minimize the problem using ultra-stable clock lasers. "The clock laser is becoming the most difficult part of the experiment," explains team member Marco Schioppo, who is now at Heinrich Heine University of Düsseldorf in Germany. "The laser cavity must be as isolated as possible from the environment, both thermally and vibrationally. The clock laser is definitely the one piece of equipment it is extremely difficult to move anywhere."

Schioppo, Ludlow and colleagues have now produced a clock containing two trapped atomic ensembles – essentially a timekeeper comprising two optical clocks. While one trap is being refilled and its atomic state prepared and measured, the laser is locked to the other trap. This has been proposed before, says Ludlow, but the researchers are the first to implement it successfully in an optical lattice clock: "To be able to do it with just two clocks, you need to make sure that the amount of time you can coherently interact with the atoms is at least the same as the amount of dead time," he explains. "For a long time, the dead time would be much larger than the spectroscopy time."

Simple, robust laser

The new clock reaches extreme stability 10 times faster than the team's 2013 clock. "As soon as you improve the instability, you decrease the timescale for your measurement, and then you're really able to pin down systematic effects more effectively," explains Schioppo. The researchers suggest that, as the laser is permanently locked to one cavity or another, a simpler, more robust laser system could also be used.

"I think it's a big step forward," says optical-clock specialist Helen Margolis of the National Physical Laboratory in Teddington, UK. Quantum metrologist Piet Schmidt of Germany's Leibniz University Hannover agrees, although he adds that the numerous difficulties the researchers had to overcome leave him wondering if the work provides a plausible route towards a simpler or more portable clock: "You need to have your two clocks synchronized extremely well to not lose a cycle of your clock laser. If you can come up with a way of producing a more stable laser source you could possibly have the same gain for less effort, but that remains to be seen."

The research is described in Nature Photonics.