C. W. Oates

Time (or its inverse, frequency) is the most precisely measured physical quantity and therefore is exploited in many fundamental investigations of nature. The advent of atomic clocks based on optical transitions has led to 10-100 times improvement in our timekeeping capabilities, with some measurements of optical frequencies reaching the mHz level [1,2]. Timekeeping precision at 1 part in 1018 enables new timing applications in relativistic geodesy, enhanced Earth- and space-based navigation and telescopy, and new tests of physics beyond the standard model. Here, we describe the development of two optical lattice clocks, both using spin-polarized, ultracold atomic ytterbium, and we discuss their operation at 10-18 instability as well as measurements toward the goal of 10-18 uncertainty.

We have extended Ramsey spectroscopy by stepping the probe frequency during the two Ramsey excitation pulses to compensate frequency shifts induced by the excitation itself. This makes precision Ramsey spectroscopy applicable even for transitions that have Stark and Zeeman shifts comparable to the spectroscopic resolution. The method enables a new way to evaluate and compensate key frequency shifts, which benefits in particular, optical clocks based on magnetic field-induced, spectroscopy, two-photon transitions, or heavily forbidden transitions.

We demonstrate a method to suppress potentially troublesome frequency shifts in optical lattice atomic clocks based on alkaline-earth-like atoms with zero nuclear spin. Suppression of these shifts, which depend on the square of the lattice field intensity, would remove a potential barrier from the path to the next generation of atomic clocks.

We describe a new laser system we have developed to probe the ultra-narrow ¹S₀ harr ³P₀ clock transition at 578 nm in neutral ytterbium. The yellow light is produced by sum frequency generation in a periodically poled waveguide. With approximately 100 mW each from a fiber laser and Nd:YAG laser, we produce 10 mW of visible light. Stabilization of the laser to a resonance of a high finesse, environmentally isolated cavity has enabled resolution of spectroscopic features as narrow as 5 Hz.

We report an absolute frequency measurement of the optical clock transition at 657 nm in ⁴⁰Ca with a relative uncertainty of 7.5 × 10⁻¹⁵, one of the most accurate frequency measurements of a neutral atom optical transition to date. The frequency (455 986 240 494 135.8 ± 3.4) Hz was measured by stabilizing a diode laser system to a spectroscopic signal derived from an ensemble of 10⁶ atoms cooled in two stages to a temperature of 10 µK. The measurement used a femtosecond-laser-based frequency comb to compare the Ca transition frequency with that of the single-ion ¹⁹⁹Hg⁺ optical frequency standard at NIST. The Hg⁺ frequency was simultaneously calibrated relative to the NIST Cs fountain via the NIST time scale to yield an absolute value for the Ca transition frequency. The relative fractional instability between the two optical standards was 2 × 10⁻¹⁵ for 10 s of averaging time and 2 × 10⁻¹⁶ for 2000 s.

The paper describes two types of optical atomic clocks. The first is based on freely expanding calcium atoms and is optimized for experimental simplicity and high stability. The second is based on Yb atoms confined to an optical lattice that is designed to yield minimal shifts for the clock transition at 578 nm. Measurements of the effective beatnote between the clocks via a femtosecond-laser frequency comb show a fractional frequency instability of -16 @ 100 s averaging time