New interrogation method makes an ion clock far less sensitive to motion-related frequency errors
Researchers report a new way to run an optical ion clock so that frequency errors caused by the ion’s motion largely cancel themselves out. They demonstrate the method on a new ytterbium clock that probes the 467 nm electric octupole (E3) transition and cut the combined uncertainty from motion-induced shifts from 1.3×10⁻¹⁸ to 0.3×10⁻¹⁸ in fractional frequency. The team also measured the E3/E2 (electric quadrupole at 435 nm) frequency ratio as 0.932829404530965340(39) and used that, with earlier data, to set a limit on a possible time change of the fine-structure constant of 2.4(2.7)×10⁻¹⁹ per year, in line with existing bounds.
Why motion matters: trapped ions are not perfectly still. Their residual motion changes the measured clock frequency in two main ways. One is a second-order Doppler shift (a small time-dilation effect when the ion moves). The other is a quadratic DC Stark shift (an energy shift caused by the trap’s electric field interacting with the atom). Both of these shifts grow when the ion picks up motional energy during the long coherent interrogations that give the best clock stability. Controlling or correcting those motion-induced shifts is a major challenge for pushing clock accuracy further.
What the researchers did: instead of relying on a particular trap setting that makes the two shifts cancel (a solution that only works for some atomic transitions), they engineered a third, controllable frequency shift. They do this by changing the standard Ramsey interrogation sequence — a method that uses two light pulses separated by a dark period to measure frequency — so that the drive laser is intentionally detuned during the pulses. This engineered shift depends on the ion’s motional energy in a way that can be tuned to cancel the combined Doppler and Stark effects over a wide range of heating rates.
How it works at a high level: the strength of the atom–light interaction during the pulses (often called the Rabi frequency) depends on the ion’s motional quantum state. By using specially shaped and phase-managed pulses — including a composite first pulse and an adjusted second pulse — the team makes the spectroscopy signal acquire a motion-dependent bias. That engineered bias is chosen to oppose the natural motion-induced shifts. Crucially, the protocol suppresses the net shift to first order even when the ion’s motional energy increases at unknown or varying rates during the interrogation.