Time has always had a special status in physics because of its fundamental role in specifying the regularities of nature and because of the extraordinary precision with which it can be measured. This precision enables tests of fundamental physics and cosmology, as well as practical applications such as satellite navigation. Recently, a regime of operation for atomic clocks based on optical transitions has become possible, promising even higher performance. We report the frequency ratio of two optical atomic clocks with a fractional uncertainty of 5.2 x 10(-17). The ratio of aluminum and mercury single-ion optical clock frequencies nuAl+/nuHg+ is 1.052871833148990438(55), where the uncertainty comprises a statistical measurement uncertainty of 4.3 x 10(-17), and systematic uncertainties of 1.9 x 10(-17) and 2.3 x 10(-17) in the mercury and aluminum frequency standards, respectively. Repeated measurements during the past year yield a preliminary constraint on the temporal variation of the fine-structure constant alpha of alpha/alpha = (-1.6+/-2.3) x 10(-17)/year.

The 1S0-3P0 clock transition frequency nuSr in neutral 87Sr has been measured relative to the Cs standard by three independent laboratories in Boulder, Paris, and Tokyo over the last three years. The agreement on the 1 x 10(-15) level makes nuSr the best agreed-upon optical atomic frequency. We combine periodic variations in the 87Sr clock frequency with 199Hg+ and H-maser data to test local position invariance by obtaining the strongest limits to date on gravitational-coupling coefficients for the fine-structure constant alpha, electron-proton mass ratio mu, and light quark mass. Furthermore, after 199Hg+, 171Yb+, and H, we add 87Sr as the fourth optical atomic clock species to enhance constraints on yearly drifts of alpha and mu.

We report a frequency measurement of the 1 S0 − 3 P0 transition of 87 Sr atoms in an optical lattice clock. The frequency is determined to be 429 228 004 229 879 (5) Hz with a fractional uncertainty that is comparable to state-of-the-art optical clocks with neutral atoms in free fall. Two previous measurements of this transition were found to disagree by about 2 × 10 −13 , i.e. almost four times the combined error bar, instilling doubt on the potential of optical lattice clocks to perform at a high accuracy level. In perfect agreement with one of these two values, our measurement essentially dissipates this doubt.PACS numbers: 06.30. Ft,42.50.Hz,42.62.Fi Recent advances in the field of optical frequency metrology make measurements with a fractional accuracy of 10 −17 or better a realistic short term goal [1]. Among other possible applications (e.g. a redefinition of the S.I. second, optical very long baseline interferometry in space, direct mapping of the earth gravitational using the Einstein effect,...), a very interesting prospect with measurements at that level is a reproducible test of Einstein Equivalence Principle by the repeated determination of the frequency ratio of different atomic and molecular transitions [2,3,4,5,6,7]. The topicality of such a test has recently been renewed by measurements at the cosmological scale which seem to indicate a slow variation of the electron to proton mass ratio [8]. The richness of the test directly depends on the performance of the clocks that are used but also on the variety of clock transitions and atomic species on which high accuracy frequency standards are based.In that context, optical lattice clocks are expected to play a central role in the future of this field. They use a large number of atoms confined in the Lamb-Dicke regime by an optical lattice in which the first order perturbation of the clock transition cancels [9]. Due to the lattice confinement motional effects, which set a severe limitation to standards with neutral atoms in free fall [10,11], essentially vanish [12]. This gives hope for a ultimate fractional accuracy better than 10 −17 . On the other hand, the large number of atoms in an optical lattice clock in principle opens the way to a short term fractional frequency stability significantly better than 10 −15 τ −1/2 with τ the averaging time in seconds. In this regime the coherence time of the laser frequency locked to the clock transition would be several seconds, possibly tens of seconds [25]. Such a long coherence time could for instance be used to reduce the width of the optical resonances in single ion clocks down to or below the 0.1 Hz range opening new prospects for these devices also. Finally, the optical lattice clock scheme is in principle applicable to a large number of atomic species (Sr, Yb, Hg, Ca, Mg,...) which is a key feature for the fundamental test discussed above. It is then particularly problematic that the frequency delivered by the first two evaluated optical lattice clocks, which both use 87 Sr, disagree by about 2 × ...

We report the observation of the higher order frequency shift due to the trapping field in a 87 Sr optical lattice clock. We show that at the magic wavelength of the lattice, where the first order term cancels, the higher order shift will not constitute a limitation to the fractional accuracy of the clock at a level of 10 −18 . This result is achieved by operating the clock at very high trapping intensity up to 400 kW/cm 2 and by a specific study of the effect of the two two-photon transitions near the magic wavelength.PACS numbers: 06.30. Ft,42.50.Hz,42.62.Fi The recent proposal and preliminary realizations of optical lattice clocks open a promising route towards frequency standards with a fractional accuracy better than 10 −17 [1,2,3,4]. A large number of atoms are confined in micro-traps formed by the interference pattern of laser beams which in principle allows both the high signal to noise ratio of optical clocks with neutral atoms [5] and the cancellation of motional effects of trapped ion devices [6, 7, 8, 9]. In contrast to the ion case an optical lattice clock requires high trapping fields. The evaluation of their effects on the clock transition is a major concern and is the subject of this letter. For a Sr optical lattice clock, the typical requirement in terms of trapping depth is about 10 E r with E r the recoil energy associated to the absorption of a lattice photon [10]. The corresponding frequency shift of both clock states amounts to 36 kHz at 800 nm, while a relative accuracy goal of 10 −18 implies a control of the differential shift at the 0.5 mHz level, or 10 −8 in fractional units.The frequency of the clock transition in a laser trap of depth U 0 is shifted with respect to the unperturbed frequency ν 0 according towith ν 1 and ν 2 proportional to the (dynamic) polarizability and hyperpolarizability difference between both states of the clock transition [1]. By principle of the optical lattice clock, ν 1 cancels when the laser which forms the lattice is tuned to the "magic wavelength" λ m . Although this remains to be demonstrated experimentally, a control of this first order term to better than 1 mHz seems achievable [1]. The higher order term is a priori more problematic with no expected cancellation. A theoretical calculation of the effect is reported in Ref.[1] predicting a frequency shift of −2 µHz/E 2 r for a linear polarization of the lattice. The calculation however was performed at the theoretical magic wavelength of 800 nm. The actual value [17], λ m = 813.428 (1) effect and impede the realization of an accurate clock. The first one couples 5s5p 3 P 0 to 5s7p 1 P 1 (Fig. 1) and is at a wavelength of 813.36 nm, or equivalently 30 GHz away from the magic wavelength. Although this J = 0 → J = 1 two-photon transition is forbidden to leading order for two photons of identical frequencies [11], it is so close to the magic wavelength that it has to be a priori considered. The second one resonantly couples 5s5p 3 P 0 to 5s4f 3 F 2 at 818.57 nm and is fully allowed. We report here an expe...

We report an uncertainty evaluation of an optical lattice clock based on the 1 S 0 ↔ 3 P 0 transition in the bosonic isotope 174 Yb by use of magnetically induced spectroscopy. The absolute frequency of the 1 S 0 ↔ 3 P 0 transition has been determined through comparisons with optical and microwave standards at NIST. The weighted mean of the evaluations is ͑ 174 Yb͒ = 518 294 025 309 217.8͑0.9͒ Hz. The uncertainty due to systematic effects has been reduced to less than 0.8 Hz, which represents 1.5ϫ 10 −15 in fractional frequency.

We report the first direct excitation of the strongly forbidden 5s 2 1 S0 −5s5p 3 P0 transition in 87 Sr. Its frequency is 429 228 004 235 (20) kHz. A resonant laser creates a small leak in a magneto-optical trap (MOT): atoms build up to the metastable 3 P0 state and escape the trapping process, leading to a detectable decrease in the MOT fluorescence. This line has a natural width of 10 −3 Hz and can be used for a new generation of optical frequency standards using atoms trapped in a light shift free dipole trap.

We report the frequency measurement with an accuracy in the 100 kHz range of several optical transitions of atomic Sr : 1 S0 − 3 P1 at 689 nm, 3 P1 − 3 S1 at 688 nm and 3 P0 − 3 S1 at 679 nm. Measurements are performed with a frequency chain based on a femtosecond laser referenced to primary frequency standards. They allowed the indirect determination with a 70 kHz uncertainty of the frequency of the doubly forbidden 5s 2 1 S0 − 5s5p 3 P0 transition of 87 Sr at 698 nm and in a second step its direct observation. Frequency measurements are performed for 88 Sr and 87 Sr, allowing the determination of 3 P0, 3 P1 and 3 S1 isotope shifts, as well as the 3 S1 hyperfine constants.

No abstract