Atomic clocks feel the heat

By Aussiegirl

If you thought that the second is the length of time it takes the car behind you in traffic to honk his horn once the light changes, you might be interested in the more precise definition: The second is defined as 9 192 631 770 periods of the radiation corresponding to a transition between two hyperfine energy levels in a caesium-133 atom. But I imagine it amounts to the same thing!

Atomic clocks feel the heat (August 2006) - News - PhysicsWeb

Atomic clocks feel the heat
Author, Belle Dumé
11 August 2006

Temperature is not something that most people take into account when trying to find out what time it is -- unless your watch has frozen or melted that is. But in the ultra-precise world of atomic timekeeping, which governs navigation technology such as the global positioning system, the temperature is vital.

Two teams of physicists in the US and Australia have now calculated the tiny shift in the atomic transition frequencies of a caesium atom, which is used to define the second, due to blackbody radiation. Although this shift has been worked out before, its value has varied by about 10% between different groups' estimates, introducing a sizeable uncertainty in the output of atomic clocks.

The second is defined as 9 192 631 770 periods of the radiation corresponding to a transition between two hyperfine energy levels in a caesium-133 atom. The current generation of caesium clocks boast accuracies of one part in 1015 -- equivalent to an error of less than one second in 30 million years. However, this accuracy could be improved by at least an order of magnitude if the tiny shift in the caesium levels due to thermal or black body radiation could be accurately determined. Although the shift could be completely removed by cooling the clock to absolute zero, this is impractical for most applications.

Kyle Beloy, Ulyana Safronova and Andrei Derevianko of the University of Nevada, and, independently, Elizabeth Angstmann, Vladimir Dzuba and Victor Flambaum at the University of New South Wales, have found that the discrepancies between previous estimates of the black-body shift are due to "intermediate continuum states" in the caesium atom, which were not taken into account.

Derevianko's team combined first-principle methods for calculating atomic structure with high-accuracy experimental data to achieved a fractional uncertainty of 6×10-17 for the black body radiation coefficient (Phys. Rev. Lett. 97 040801). This value improves the accuracy of clocks operated at room temperature by an order of magnitude, making them as accurate as those operated at 0K. Meanwhile, Flambaum's group reached similar conclusions using first-principle calculations alone, applying Planck's radiation law and perturbation theory to calculate the energy shift of each component of the hyperfine structure in caesium (Phys. Rev. Lett. 97 040802).

Together, the results pave the way for making atomic clocks more accurate, and also for more precise studies of how the fundamental constants of nature vary with time.