The best atomic clocks approach an accuracy of about a part in , but that’s not good enough for researchers looking for possible tiny drifts in fundamental constants. A newly refined proposal based on nuclear excitations in a single ion could, in principle, do almost times better.
The energy to create an excited nuclear state should be much less sensitive to stray external fields than are the electronic excitations used in the best existing atomic clocks. In particular, in 2003, researchers proposed exploiting a relatively low-energy transition in thorium- nuclei, which could be excited with ultraviolet lasers. In Physical Review Letters, Corey Campbell at the Georgia Institute of Technology, Atlanta, and co-workers note that a different transition in the same nucleus should be even less sensitive to external fields.
The team proposes a clock based on a single trapped thorium ion, and analyzes a dozen different effects that could limit its accuracy. The two biggest potential errors are stray electric fields that nudge the ion away from the sweet spot of the trap, and uncertainty about the height of the trap. The gravitational effect of even a millimeter height error would cause a significant frequency shift, according to general relativity. Combining all of the errors, the researchers estimate a timing accuracy of parts in , corresponding to a -millisecond error over the entire age of the universe. – Don Monroe