Optical lattice clocks have recently opened the most promising route towards more accurate and stable frequency standards. But to improve the present accuracy of a few 10-16 of the best atomic clocks, it was necessary to lift the most serious limitation. These clocks consist in measuring the optical frequency corresponding to the transition between two atomic levels of 87Sr. The objective is to be able to control the transition frequency of 500 THz (5 1014 Hz) to within one millihertz (10-3 Hz), or 10-18 in fractional units. To reach this level, the motions of the atoms must be totally under control.
This is done by tightly confining the atoms in an optical lattice which freezes the atomic motion. Such a trap however can significantly shift the transition frequency and thus severely affect the clock accuracy. Previous studies have shown that this frequency shift can be cancelled to first order by adjusting the wavelength of the trapping laser. Given the required level of control however, effects of higher order must be considered and so far nobody had been able to conclude regarding these effects.
The experimental results published by A. Brusch and collaborators are the first study of these higher orders effects, which was achieved by operating the clock at an unprecedented laser intensity. This study shows that these effects are not a limitation for a future accuracy at the 10-18 level.