This thesis investigates dipolar gases of dysprosium (Dy) atoms at cold and ultracold temperatures. Dy is an open-shell lanthanide atom and one of the most magnetic elements in the periodic table. In the context of ultracold science, Dy has attracted a lot of interest because its large magnetic dipole moment allows us to investigate quantum gases with strong dipole–dipole interactions and new phases of matter, such as dipolar supersolids. In the first part of this thesis, I will present the design and development of a new apparatus for the study of ultracold dipolar quantum gases of Dy. The project entailed the construction of an experimental apparatus that includes ultra-high vacuum components, laser systems that allow first to laser cool and trap Dy in a magneto-optical trap and then to transfer laser-cooled Dy atoms into a conservative optical potential for further evaporative cooling. Thanks to this, we are able to observe the onset of Bose–Einstein condensation in our gas of Dy atoms. In the second part of this thesis, I will present a spectroscopic investigation of electronically excited states in a beam of Dy atoms. Differently from other atomic species, the electronic spectrum of Dy is rather dense, with many levels being closely spaced. Interestingly, it has been pointed out by Ref. [1] that Dy can acquire a large electric dipole moment, in addition to the magnetic one, when prepared in a doublet of long-lived opposite parity states that lies around 17513 cm−1 above the ground state. This opportunity could allow us to realize a unique quantum gas with two competing dipolar interactions, one magnetic and one electric, that can host new complex many-body phases. Using the supersonic beam machine, I performed a detailed spectroscopic characterization of the excited doublet, which allowed us to accurately measure the doublet spacing and to find an optical scheme for efficiently populating it in the ultracold experiment. Finally, using electric fields up to 150 kV/cm, I was able to experimentally confirm an induced electric dipole moment larger than 1 Debye, comparable to those found in polar molecule experiments but without dealing with the complex laser cooling methods to bring molecules in the ultracold regime.
This thesis investigates dipolar gases of dysprosium (Dy) atoms at cold and ultracold temperatures. Dy is an open-shell lanthanide atom and one of the most magnetic elements in the periodic table. In the context of ultracold science, Dy has attracted a lot of interest because its large magnetic dipole moment allows us to investigate quantum gases with strong dipole–dipole interactions and new phases of matter, such as dipolar supersolids. In the first part of this thesis, I will present the design and development of a new apparatus for the study of ultracold dipolar quantum gases of Dy. The project entailed the construction of an experimental apparatus that includes ultra-high vacuum components, laser systems that allow first to laser cool and trap Dy in a magneto-optical trap and then to transfer laser-cooled Dy atoms into a conservative optical potential for further evaporative cooling. Thanks to this, we are able to observe the onset of Bose–Einstein condensation in our gas of Dy atoms. In the second part of this thesis, I will present a spectroscopic investigation of electronically excited states in a beam of Dy atoms. Differently from other atomic species, the electronic spectrum of Dy is rather dense, with many levels being closely spaced. Interestingly, it has been pointed out by Ref. [1] that Dy can acquire a large electric dipole moment, in addition to the magnetic one, when prepared in a doublet of long-lived opposite parity states that lies around 17513 cm−1 above the ground state. This opportunity could allow us to realize a unique quantum gas with two competing dipolar interactions, one magnetic and one electric, that can host new complex many-body phases. Using the supersonic beam machine, I performed a detailed spectroscopic characterization of the excited doublet, which allowed us to accurately measure the doublet spacing and to find an optical scheme for efficiently populating it in the ultracold experiment. Finally, using electric fields up to 150 kV/cm, I was able to experimentally confirm an induced electric dipole moment larger than 1 Debye, comparable to those found in polar molecule experiments but without dealing with the complex laser cooling methods to bring molecules in the ultracold regime.
Laser cooling and trapping of the most magnetic fermionic atom, dysprosium (Dy), may provide a framework to explore quantum liquid crystal (QLC) theory (Chapter 1). This thesis presents details of the Dy laser cooling and trapping apparatus including the laser systems at 421, 741, and 1064 nm, the ultra-high vacuum (UHV) chamber, and the computer control that has produced a magneto-optically (MOT) and magneto-statically (MT) trapped Dy gas (Chapters 3, 4, 5). Despite the fact that Dy has a complex energy level structure with nearly 140 metastable states (Chapter 2), Dy MOT at 421-nm transition with 32-MHz linewidth was realized without any rempumper, exploiting its large magnetic moment, which brought a strong magnetic confinement of metastable states of Dy. This unique MOT/MT dynamics is discussed and its quantitative measurements are shown in Chapter 6. When the Dy atoms dropped from the MOT were adsorptively imaged, it was observed that Dy MOT had a bimodal temperature distribution in contrast to the usual MOT described by a single temperature (Chapter 7). Such novel anisotropic sub-Doppler laser cooling of Dy, which breaks the symmetry in cooling, is due to Dy's large magnetic spin aligned along a strong axis of the quadrupole field of the MOT, and we further support this plausible conjecture with the velocity selective resonance (VSR) theory. The MOT at ~ 1 mK was cooled to ~ 10 uK by narrow-line cooling at 741 nm with a linewidth of 2 kHz, and we were able to load the optical dipole trap (ODT) at 1064 nm. By loading two isotopes of 164Dy and 163Dy in sequence to the MOT and narrow-line cooling them simultaneously, ultracold Bose-Fermi mixtures of Dy in the ODT were realized (Chapter 8). This thesis is concluded with a discussion of prospect on the Bose-Fermi mixtures of Dy.
This thesis discusses the experiments on 2D dipolar Bose gas of ultracold erbium atoms. Long-range and anisotropic dipole-dipole interaction induces complex orders in quantum many-body systems. Highly magnetic lanthanide atoms such as dysprosium and erbium have the largest magnetic moment in the periodic table, which has attracted significant attention in the past decades. Starting from quantum degenerate dysprosium and erbium atoms, exotic phases like quantum droplets, supersolid, dipolar quantum solid in the Hubbard regime have been studied. However, the 2D dipolar superfluid still remains understudied, despite being theoretically discussed for over two decades. Although tremendous theoretical proposals predict different phenomena in 2D dipolar superfluids, our first interest is to know how the superfluidity emerges in the 2D dipolar gases at finite temperatures. Our observations suggest that the superfluid transition can be well described by the Berezinskii-Kosterlitz-Thouless transition with an effective contact interaction as a function of dipolar orientation. With the 2D dipolar superfluid with strong DDI, we characterized the elementary excitation of the matter: the collective breathing mode. Our measurements establish how the competition between local and nonlocal interaction in strongly dipolar 2D superfluid retains quasi-scale invariance or strongly breaks scale invariance. By further searching the interaction parameters, we created stripes with global coherence, which are the first supersolid stripes in 2D. Experiments including measuring density fluctuation, coherence, and collective mode have been performed to understand the 2D supersolid. At last, we also observe unconventional light scattering from round-disk dipolar BECs, which provides new possibilities for atom optics.
We experimentally show that an electric dipole moment of more than 1 Debye can be induced in the dysprosium (Dy) atom, in a long-lived state that is about 17 513 cm − 1 above the ground state. This metastable state is part of a strongly coupled opposite-parity doublet. Using optically detected microwave spectroscopy in an atomic beam, we determine the approximately 1.12 cm − 1 doublet spacing for the five stable bosonic isotopes of Dy with kilohertz-level accuracy. From the shift of the microwave transition frequency in low electric fields (below 150 V/cm) and from optical spectra in high electric fields (up to 150 kV/cm), a reduced transition dipole moment of 7.65 ± 0.05 Debye between the doublet states is extracted. In high electric fields the doublet interacts with a third state at 17 727 cm − 1 that connects to the ground state via an electric-dipole transition. The three-state Stark interaction enables preparation of Dy atoms in the metastable state via single-photon excitation from the ground state. Our work experimentally validates predicted characteristics of the opposite-parity doublet and provides an effective route to the first doubly polar quantum gas of atoms.
Ultracold dysprosium gases, with a magnetic moment 10 times that of alkali atoms and equal only to terbium as the most magnetic atom, are expected to exhibit a multitude of fascinating collisional dynamics and quantum dipolar phases, including quantum liquid crystal physics. We report the first laser cooling and trapping of half a billion Dy atoms using a repumper-free magneto-optical trap (MOT) and continuously loaded magnetic confinement, and we characterize the trap recycling dynamics for bosonic and fermionic isotopes. The first inelastic collision measurements in the few partial wave, 100 microK-1 mK, regime are made in a system possessing a submerged open electronic f shell. In addition, we observe unusual stripes of intra-MOT <10 microK sub-Doppler cooled atoms.
This thesis addresses several experimental studies carried out on ultracold Bose gases of dysprosium. The complex electronic structure of this species gives rise to a rich optical spectrum, remarkable magnetic properties, and complex underlying collision physics. This manuscript is divided into three parts. The first part is devoted to the experimental system used to produce ultracold and degenerate gases of 162Dy. We present the main steps leading from laser cooling to Bose–Einstein condensation, as well as several optical-control tools developed around a J′ = J − 1 transition. This transition supports the existence of dark states and, together with the large tensor polarizability of dysprosium, makes it possible to generate substantial spin-dependent light shifts of the energy levels with limited heating. The second part focuses on the interactions that coexist in this system, namely contact interactions and magnetic dipole-dipole interactions. We introduce the elements of scattering theory relevant to ultracold gases, followed by Fano-Feshbach resonances, which allow one to control two-body interactions through the scattering length. Since these resonances are accompanied by atomic losses, we first study their macroscopic modeling at low magnetic field, in the case of narrow resonances of 162Dy. We then address another aspect of these scattering resonances: their connection to Tan’s contact. Introducing this quantity makes it possible to describe the behavior of the Bose gas in the strongly interacting regime. In particular, the use of light shifts allowed us to rapidly modify the dimer energy associated with a relatively narrow resonance, in order to microscopically study the dynamics of Tan’s contact through losses induced by molecular photodissociation. The final part is devoted to the production and stabilization of spin mixtures in a dipolar Bose gas. We show how spin-dependent light shifts, combined with a two-photon Raman transfer, make it possible to prepare degenerate gases in different Zeeman sublevels and controlled spin mixtures. We then characterize the scattering lengths associated with several spin-polarized states using the expansion of dipolar condensates. Finally, we study dipolar relaxation, which generally limits the lifetime of spin mixtures, and show that a destructive-interference process, related to contact interactions, makes it possible to reduce this relaxation.
This thesis presents the development of an experimental apparatus designed to investigate the ultracold collisional properties for mixtures of Cs and Yb, with a long-term view to the creation of ultracold CsYb molecules via indirect cooling methods. The unpaired electron spin that is inherent to molecules of this form gives rise to a magnetic dipole moment in addition to a ground state electric dipole moment. This enables extra control over molecular interactions and should enable the experimental simulation of spin lattice models. We focus on the implementation of a system designed to controllably laser cool and optically trap Yb. The first step in this system is the production of a magneto-optical trap (MOT) on the triplet 1S0 to 3P1 transition of Yb. With careful control over the cooling beam detunings and power, gravitational-assisted Doppler cooling allows samples of Yb to be prepared at 22 uK. This regime of enhanced Doppler cooling is investigated and proves to be a crucial step to ensuring good transfer of cold Yb to optical traps. The construction and characterisation of single and crossed beam optical dipole traps for Yb are discussed. The single beam optical trap has been used to verify a model for the optical trapping of Yb in its ground state. This trap has also been utilised as a tool for the measurement of the light shift on the 1S0 to 3P1 transition at a wavelength of 1070~nm. In the main experimental sequence, Yb atoms are loaded from the magneto-optical trap into the crossed optical dipole trap, allowing evaporative cooling ramps to quantum degeneracy to be performed. This highly-reproducible system typically forms Bose-Einstein condensates with 2 x 10^5 174Yb atoms. This thesis additionally reports on the progress made towards measurements of the interspecies scattering length for 133Cs and Yb isotopes. We present two approaches that are being developed in tandem: rethermalisation in a conservative trap, and two-photon photoassociation. Progress towards rethermalisation measurements has focussed on developing systems for the efficient transfer of Cs to an optical trap. For photoassociative measurements, a laser system has been developed and tested by producing one-photon photoassociation spectra of Cs2.