Final Report Summary - CLOCKLIGHT (Light for Clocks)
Many definitions of the International System (SI) of Units have been changed since the first definitions were made. These redefinitions are essentially reflections on improvements in the accuracy and reproducibility with which the definitions could be realised. At present, the definition of the second is evolving, and the future definition will be based on an optical frequency. The primary motivation for research on optical frequency standards is the unique combination of laser cooling and tight confinement during the interrogation cycle that provides a spectroscopic system permitting observation of nearly unperturbed and very narrow atomic transitions at high clock frequencies.
As these clocks have an unprecedented level of accuracy, the lowest reported fractional frequency uncertainties are 3.3E-18 for an 171Yb+ octupole-transition ion clock and 6.4E-18 for an 87Sr lattice clock. However, reliable measurements validating these claims can only be made by directly comparing optical frequencies between local optical clocks. The primary reason behind this is the uncertainty in the shape of the geoid that impacts all clocks through the gravitational red shift. A height difference of one meter causes a fractional frequency shift of ~1E-16 in any clock, and the uncertainty of the geoid shape is typically 0.3 meters and can be much more in mountainous regions. Thus, only transportable optical clocks can fulfil the accuracy requirements of validation, necessary prerequisite for the future redefinition of the second. This motivates the objective to study novel, transportable light sources for a 88Sr+ single-ion optical clock and to contribute, thereby, to the future redefinition of the second through establishment of a method for a low uncertainty frequency comparison between institutes operating optical clocks.
Successful operation of any optical clock necessarily requires a relatively complicated laser ensemble, while long integration times are necessary for frequency comparisons at relevant uncertainty. Thus, the requirements for operational reliability of the light sources are most stringent, while spectroscopic properties of the lasers need to meet the purpose at hand. Ultimately, the requirement of completely autonomous operation necessitates advanced re-locking schemes for the lasers and possibly an automatic re-loading of the ion trap, should the ion be lost e.g. due to collisions with background gas molecules. Although this ultimate goal is beyond the scope of the project, significant progress towards improved light sources for the 88Sr+ single-ion optical clock was achieved.
Of particular concern in all ion clocks is the above mentioned trap loading process. Traditional methods based on e.g. electron bombardment of neutral strontium atoms usually result in contamination of the trap surfaces due to the necessarily high atom flux. In this project the photoionization loading method was studied and improved upon. For this method it is characteristic that the lasers are not operated continuously, but need to be readily available, should the ion be lost. Photoionization is a two-step process in which the strontium atoms are first excited to higher electronic state with a narrow-band laser and then to an auto-ionizing state using a broadband light source. The necessary requirements for the broadband light source can be relatively easily met with multi-mode Fabry-Pérot –type blue laser diodes, while excitation to the electronic state requires a carefully controlled single-mode light source, whose frequency needs to be compensated for Doppler-shift that depends on the geometrical arrangement of ionization and atom beams. Several different frequency stabilization schemes for the narrowband excitation were studied in this project, including but not limited to, measurement of ionization current, detection of fluorescence of atoms in atom beam, absorption spectroscopy, and referencing to a frequency comb generator. A combination of the two latter techniques was found to yield a working solution.
Once an ion is successfully trapped, the ion needs to be cooled to mK-level temperature to reach the operational regime of a single-ion optical clock. In 88Sr+ ion clocks standard Doppler-cooling is used, where a laser red-detuned from a strong cyclic transition provides cooling while a re-pumper prevents decay into a low-lying metastable state, out of the cooling cycle. Efficient Doppler cooling requires the cooling laser to have a linewidth and frequency drifts much smaller than the natural linewidth of the cooling transition. The re-pumping transition forms dark states for any fixed re-pumper polarization, which inhibits the cooling and fluorescence detection of the ion. Conventionally, these dark states have been avoided by some form of polarization randomization technique. In this work, a novel approach to avoid dark states was studied and a novel polarization-incoherent re-pumper laser was developed. Further, similar to the photoionization, diverse frequency stabilization approaches for the cooling laser were studied. The final working combination was also found to be similar to that of photoionization, but instead of strontium beam reference, a rubidium gas-cell absorber is used. Additionally, locking to a frequency comb generator was also studied for linewidth narrowing of the laser.
Much effort was put onto testing and verification of the novel polarization-incoherent re-pumper. To this end the laser was built onto a robust package that allowed shipment of the laser source to a collaborating partner, the National Research Council (NRC) of Canada, using a standard courier service. The light source was tested at, and in collaboration with, NRC using their operational Sr+ single-ion clock, thus allowing direct comparison with previous generation techniques. This endeavour was highly successful and demonstrated both the viability of the technique and robustness of the developed light source.
Last but not least, an operational clock needs an interrogation laser for probing the clock transition. As the clock transition is very narrow, complex multi-stage laser stabilization techniques of laser diodes are required to reach sufficient spectral purity. The final stabilization stage is conventionally a high-Q mechanically stable optical resonator, whose properties are transferred onto the properties of the clock laser. To this end, schemes for pre-stabilization and final stabilization to the optical resonator were studied and successful locking to the cavity was achieved. However, final evaluation of the performance could not be performed, as the single-ion clock at MIKES did not reach operational status during the lifetime of the project.
As these clocks have an unprecedented level of accuracy, the lowest reported fractional frequency uncertainties are 3.3E-18 for an 171Yb+ octupole-transition ion clock and 6.4E-18 for an 87Sr lattice clock. However, reliable measurements validating these claims can only be made by directly comparing optical frequencies between local optical clocks. The primary reason behind this is the uncertainty in the shape of the geoid that impacts all clocks through the gravitational red shift. A height difference of one meter causes a fractional frequency shift of ~1E-16 in any clock, and the uncertainty of the geoid shape is typically 0.3 meters and can be much more in mountainous regions. Thus, only transportable optical clocks can fulfil the accuracy requirements of validation, necessary prerequisite for the future redefinition of the second. This motivates the objective to study novel, transportable light sources for a 88Sr+ single-ion optical clock and to contribute, thereby, to the future redefinition of the second through establishment of a method for a low uncertainty frequency comparison between institutes operating optical clocks.
Successful operation of any optical clock necessarily requires a relatively complicated laser ensemble, while long integration times are necessary for frequency comparisons at relevant uncertainty. Thus, the requirements for operational reliability of the light sources are most stringent, while spectroscopic properties of the lasers need to meet the purpose at hand. Ultimately, the requirement of completely autonomous operation necessitates advanced re-locking schemes for the lasers and possibly an automatic re-loading of the ion trap, should the ion be lost e.g. due to collisions with background gas molecules. Although this ultimate goal is beyond the scope of the project, significant progress towards improved light sources for the 88Sr+ single-ion optical clock was achieved.
Of particular concern in all ion clocks is the above mentioned trap loading process. Traditional methods based on e.g. electron bombardment of neutral strontium atoms usually result in contamination of the trap surfaces due to the necessarily high atom flux. In this project the photoionization loading method was studied and improved upon. For this method it is characteristic that the lasers are not operated continuously, but need to be readily available, should the ion be lost. Photoionization is a two-step process in which the strontium atoms are first excited to higher electronic state with a narrow-band laser and then to an auto-ionizing state using a broadband light source. The necessary requirements for the broadband light source can be relatively easily met with multi-mode Fabry-Pérot –type blue laser diodes, while excitation to the electronic state requires a carefully controlled single-mode light source, whose frequency needs to be compensated for Doppler-shift that depends on the geometrical arrangement of ionization and atom beams. Several different frequency stabilization schemes for the narrowband excitation were studied in this project, including but not limited to, measurement of ionization current, detection of fluorescence of atoms in atom beam, absorption spectroscopy, and referencing to a frequency comb generator. A combination of the two latter techniques was found to yield a working solution.
Once an ion is successfully trapped, the ion needs to be cooled to mK-level temperature to reach the operational regime of a single-ion optical clock. In 88Sr+ ion clocks standard Doppler-cooling is used, where a laser red-detuned from a strong cyclic transition provides cooling while a re-pumper prevents decay into a low-lying metastable state, out of the cooling cycle. Efficient Doppler cooling requires the cooling laser to have a linewidth and frequency drifts much smaller than the natural linewidth of the cooling transition. The re-pumping transition forms dark states for any fixed re-pumper polarization, which inhibits the cooling and fluorescence detection of the ion. Conventionally, these dark states have been avoided by some form of polarization randomization technique. In this work, a novel approach to avoid dark states was studied and a novel polarization-incoherent re-pumper laser was developed. Further, similar to the photoionization, diverse frequency stabilization approaches for the cooling laser were studied. The final working combination was also found to be similar to that of photoionization, but instead of strontium beam reference, a rubidium gas-cell absorber is used. Additionally, locking to a frequency comb generator was also studied for linewidth narrowing of the laser.
Much effort was put onto testing and verification of the novel polarization-incoherent re-pumper. To this end the laser was built onto a robust package that allowed shipment of the laser source to a collaborating partner, the National Research Council (NRC) of Canada, using a standard courier service. The light source was tested at, and in collaboration with, NRC using their operational Sr+ single-ion clock, thus allowing direct comparison with previous generation techniques. This endeavour was highly successful and demonstrated both the viability of the technique and robustness of the developed light source.
Last but not least, an operational clock needs an interrogation laser for probing the clock transition. As the clock transition is very narrow, complex multi-stage laser stabilization techniques of laser diodes are required to reach sufficient spectral purity. The final stabilization stage is conventionally a high-Q mechanically stable optical resonator, whose properties are transferred onto the properties of the clock laser. To this end, schemes for pre-stabilization and final stabilization to the optical resonator were studied and successful locking to the cavity was achieved. However, final evaluation of the performance could not be performed, as the single-ion clock at MIKES did not reach operational status during the lifetime of the project.