We experimentally investigate an optical clock based on ;{171}Yb (I = 1/2) atoms confined in an optical lattice. We have evaluated all known frequency shifts to the clock transition, including a density-dependent collision shift, with a fractional uncertainty of 3.4 x 10;{-16}, limited principally by uncertainty in the blackbody radiation Stark shift. We measured the absolute clock transition frequency relative to the NIST-F1 Cs fountain clock and find the frequency to be 518 295 836 590 865.2(0.7) Hz.

For the past 50 years, atomic standards based on the frequency of the cesium ground-state hyperfine transition have been the most accurate time pieces in the world. We now report a comparison between the cesium fountain standard NIST-F1, which has been evaluated with an inaccuracy of about 4 x 10(-16), and an optical frequency standard based on an ultraviolet transition in a single, laser-cooled mercury ion for which the fractional systematic frequency uncertainty was below 7.2 x 10(-17). The absolute frequency of the transition was measured versus cesium to be 1,064,721,609,899,144.94 (97) Hz, with a statistically limited total fractional uncertainty of 9.1 x 10(-16) the most accurate absolute measurement of an optical frequency to date.

We report the first accuracy evaluation of NIST-F2, a second-generation laser-cooled caesium fountain primary standard developed at the National Institute of Standards and Technology (NIST) with a cryogenic (liquid nitrogen) microwave cavity and flight region. The 80 K atom interrogation environment reduces the uncertainty due to the blackbody radiation shift by more than a factor of 50. Also, the Ramsey microwave cavity exhibits a high quality factor (>50 000) at this low temperature, resulting in a reduced distributed cavity phase shift. NIST-F2 has undergone many tests and improvements since we first began operation in 2008. In the last few years NIST-F2 has been compared against a NIST maser time scale and NIST-F1 (the US primary frequency standard) as part of in-house accuracy evaluations. We report the results of nine in-house comparisons since 2010 with a focus on the most recent accuracy evaluation. This paper discusses the design of the physics package, the laser and optics systems and the accuracy evaluation methods. The type B fractional uncertainty of NIST-F2 is shown to be 0.11 × 10 −15 and is dominated by microwave amplitude dependent effects. The most recent evaluation (August 2013) had a statistical (type A) fractional uncertainty of 0.44 × 10 −15 .

The evaluation procedure of a new laser-cooled caesium fountain primary frequency standard developed at the National Institute of Standards and Technology (NIST) is described. The new standard, NIST-F1, is described in some detail and typical operational parameters are discussed. Systematic frequency biases for which corrections are made -second-order Zeeman shift, black-body radiation shift, gravitational red shift and spin-exchange shiftare discussed in detail. Numerous other frequency shifts are evaluated, but are so small in this type of standard that corrections are not made for their effects. We also discuss comparisons of this standard both with local frequency standards and with standards at other national laboratories.

In the last several years we have made many improvements to NIST-F1 (a laser-cooled Cs fountain primary frequency standard at the National Institute of Standards and Technology (NIST) in Boulder, Colorado) resulting in a reduction by over a factor of 2 in the uncertainty of the realization of the SI second. The two most recent accuracy evaluations of NIST-F1 had combined standard fractional uncertainties of 0.61 × 10 −15 (June 2004) and 0.53 × 10 −15 (January 2005), which were submitted to the Bureau International des Poids et Mesures with total fractional uncertainties (including time-transfer contributions) of, respectively, 0.88 × 10 −15 and 0.97 × 10 −15. Here we discuss the improvements and evaluation methods and present an updated uncertainty budget.