Since 1967, the Second has been defined by the CGPM as 9,192,631,770 periods of oscillation corresponding to a transition between two hyperfine levels in a Cesium atom with zero external fields. Key to this definition was the fact that as experimentalists found new and more precise ways to measure this frequency, the definition of the second would become more and more accurate with time. However, in the last 30 years, new technologies based on tunable lasers addressing optical transitions in atoms, ions, and molecules have offered an entirely new approach to defining the Second with significantly higher precision and accuracy. Here, I will show that by trapping thousands of atoms inside a specially engineered optical lattice one can create an extremely accurate frequency standard with 2 orders of magnitude improvement over current Cs standards. Furthermore, I will explain that standards based on this technology are fundamentally more stable than the primary standard by 3 orders of magnitude. Leapfrogging the currently held accuracy records of ion clocks, this work documents the first optical lattice clock to best all other atomic clock implementations, a mere 8 years after the first proof of principle experiments. In this thesis I will describe how we have overcome a number of important systematic uncertainties to achieve these results. This revolution in accuracy and precision opens the door for new experiments utilizing the clock as a probe of quantum many-body physics. To this end, a new apparatus has been designed that combines the unprecedented precision of clock experiments with the amazing progress attained in quantum gas experiments.
Since 1967, the Second has been defined by the CGPM as 9,192,631,770 periods of oscillation corresponding to a transition between two hyperfine levels in a Cesium atom with zero external fields. Key to this definition was the fact that as experimentalists found new and more precise ways to measure this frequency, the definition of the second would become more and more accurate with time. However, in the last 30 years, new technologies based on tunable lasers addressing optical transitions in atoms, ions, and molecules have offered an entirely new approach to defining the Second with significantly higher precision and accuracy. Here, I will show that by trapping thousands of atoms inside a specially engineered optical lattice one can create an extremely accurate frequency standard with 2 orders of magnitude improvement over current Cs standards. Furthermore, I will explain that standards based on this technology are fundamentally more stable than the primary standard by 3 orders of magnitude. Leapfrogging the currently held accuracy records of ion clocks, this work documents the first optical lattice clock to best all other atomic clock implementations, a mere 8 years after the first proof of principle experiments. In this thesis I will describe how we have overcome a number of important systematic uncertainties to achieve these results. This revolution in accuracy and precision opens the door for new experiments utilizing the clock as a probe of quantum many-body physics. To this end, a new apparatus has been designed that combines the unprecedented precision of clock experiments with the amazing progress attained in quantum gas experiments.
The exquisite control exhibited over quantum states of individual particles has revolutionized the field of precision measurement, as exemplified by the most accurate atomic clock realized in single trapped ions. Whereas many-atom lattice clocks have shown advantages in measurement precision over trapped-ion clocks, their accuracy has remained 20 times worse. Here we demonstrate, for the first time, that a many-atom system achieves accuracy (6x10^{-18}) better than a single ion-based clock, with vastly reduced averaging times (3000 s). This is the first time a single clock has achieved the best performance in all three key ingredients necessary for consideration as a primary standard - stability, reproducibility, and accuracy. This work paves the way for future experiments to integrate many-body quantum state engineering into the frontiers of quantum metrology, creating exciting opportunities to advance precision beyond the standard quantum limit. Improved frequency standards will have impact to a wide range of fields from the realization of the SI units, the development of quantum sensors, to precision tests of the fundamental laws of nature.
Atomic clocks constitute a fundamental tool for time and frequency metrology and their application is widespread in many technological fields. In particular, the International System of units (SI) defines the second on a microwave transition of Caesium atoms. The realization is made by clocks reaching uncertainties of few parts in 10^−16, making the second to be the quantity realized with the smallest uncertainty in the SI. However, a new generation of atomic clocks, called optical clocks, have already demonstrated to surpass Caesium standards both in accuracy and stability. The research performed during my PhD activity has been focused on the development and characterization of Ytterbium (Yb) optical lattice clocks. These systems operate with a large number of ultra-cold neutral atoms having a clock transition in the visible region of the electromagnetic spectrum. The atomic sample is trapped in an periodical optical potential called optical lattice that gives the advantage to interrogate many quantum absorbers for an extended time, with small perturbations, allowing to achieve an unprecedented stability and accuracy. The main experimental work has been carried out in the laboratories of the Istituto Nazionale di Ricerca Metrologica (INRIM) in Turin, where several atomic clocks are present including the Italian primary frequency standard, the Caesium fountain ITCsF2, and where an Yb optical lattice clock is currently being developed. This thesis describes the functioning of the system along with the full characterization of systematic effects, the complete uncertainty budget and its first absolute frequency measurement against the primary frequency standard. The clock exhibited an accuracy of 1.6 × 10^−16 and the comparison with the Cs fountain resulted in a frequency of f171Yb = 518 295 836 590 863.59(31) Hz, limited by the fountain uncertainty. This measurement is in agreement with the ytterbium frequency recommended as a secondary representation of the second in the SI and constitutes the first measurement of a Yb clock in Europe and the second one in the world against a primary frequency standard. Several upgrades have been applied after the absolute measurement. In particular, the design and realization a system capable to frequency stabilize several lasers on a single optical cavity is illustrated. This cavity has been implemented to lock the lasers used to cool and trap the atomic sample at 399, 556 and 759nm using the offset sideband locking technique, a modified version of the Pound–Drever–Hall method that gives an extended frequency tunability. The system proved to be an easy-to-use and reliable tool for the experimental activity showing a linewidth below 300 Hz at 556nm, which is the wavelength that requires the most stringent performance, and a long term drift below 20 kHz per day at 759nm. That is suitable for operating the lattice laser with a light shift uncertainty below 1 × 10−18. During my PhD I have been guest researcher at the National Institute of Standards and Technology (NIST) of Boulder, Colorado, for nine months in 2016. In these laboratories two Yb optical lattice clocks are operative. I worked on the instability measurement of a composite system exploiting the two clocks to suppress the Dick effect, called zero-dead-time (ZDT) clock, which demonstrated a fractional instability of 3 × 10−17 at 1s. The two clocks can also be operated to extend the interrogation time obtaining a spectroscopic feature after 4 s of 120(20)mHz corresponding to a quality factor Q > 4 × 10^15. I also worked on the characterization of several systematic shifts that allowed to complete the uncertainty budget of the clocks at 1.6 × 10^−18. In particular, I contributed to the characterization of lattice light shifts considering the effect of atomic sample temperature and the identification of a metrological regime called operational magic frequency where frequency shifts are insensitive to changes in trap depth.
In 2013, for the first time ever, a many-atom optical lattice clock surpassed single ion clocks as the most stable, and accurate frequency standard in existence. Since this first result, continued improvements to the clock's accuracy and stability have been made. This work will focus on two of these major improvements. Firstly the frequency uncertainty was brought down to 2.5*10^-18, thanks in part to an accurate determination of the black body radiation shift, based on better characterization of the level structure of 87Sr. Secondly, a control modification will be demonstrated which provides 20% reduction in the standard deviation for the clock's frequency under differential measurement schemes. This represents a new record for clock accuracy, pushing performance towards the 10^-19 level.
Accurate time keeping is critical to the function and advancement of many human endeavours. Atomic clocks based on a microwave transition in atomic Cs have served as primary frequency standards to realise the SI second since their formal adoption in 1967. State-of-the-art, laser cooled, Cs fountains now routinely operate with fractional frequency uncertainties below the 10⁻¹⁵ level in laboratories throughout the world. Over the last decade, efforts to build frequency standards based on optical transitions have come to fruition and potential inaccuracies below the 10⁻¹⁷ level have been demonstrated. Such standards point to the future of time keeping and much development is being pursued to realise the potential of these 'next generation' optical clocks. One such approach, on which this work focusses, relies on a 'magic' wavelength optical lattice trap to provide (almost) Doppler- and recoil-free confinement of atoms during clock interrogation. At this 'magic' wavelength, light induced perturbation of the clock states is equal and so cancels precisely. This thesis describes the design and implementation of an optical lattice frequency standard apparatus based on 88Sr held in a 1D optical lattice. A review of the development of atomic frequency metrology is given from a historical viewpoint and the technology and limitations of current methods discussed. To tackle the blackbody radiation shift, which currently presents the greatest limitation to the Sr lattice clock, imposing at the 10⁻¹⁶ level, a scheme to measure the blackbody radiation shift coefficient at the 10⁻¹⁷ level is developed. Techniques to laser cool and trap atomic strontium are described and the implementation detailed. All laser systems required for clock operation are described including the design and implementation of extended-cavity diode lasers for precision spectroscopy. Details relevant to the eventual operation of the apparatus are discussed.
Time for a change Since the first practical caesium atomic clock was built in 1955 (by Louis Essen and John V. L. Parry, described in Nature 176; 280–285), work has gone on to improve the accuracy and stability of such devices. Trapped ions offer accuracy advantages, neutral atoms aid stability. A new device, the optical lattice clock, combines the best of both approaches and could point towards the next generation of atomic clocks. Using atoms trapped in an optical lattice as quantum references, this system has the potential to achieve a stability several orders of magnitude better than the caesium clock currently used to define the second as an SI unit.
8half a century ago and have provided the definition of the second since 1967. The international atomic time (TAI) is currently kept by an ensemble of Cs clocks with a fractional uncertainty of Δν/ν0 ≈ 10 –15 , and some Cs fountain clocks 9,10 operate at even smaller uncertainties. Although Cs atomic clocks operate at the microwave transition of ν0 ≈ 9.2 GHz, using optical transition frequencies of several hundred terahertz should improve the fractional uncertainty Δν/ν0 by 4–5 orders of magnitude, as most measurement uncertainties Δν are independent of ν0 (except the Doppler shift Δ νD = (v/c)ν0, where v is the atomic velocity and c is the speed of light). The invention of the laser and the subsequent improvement of laser frequency stabilities 11 has allowed the realization of optical clocks to become a reality 2,3,12