Quasi-2D degenerate Fermi gases provide a rich physical environment to explore since many effects that are present do not exist in 3D. This thesis describes the first experiments carried out in a quasi-2D Fermi gas across the broad 6Li Feshbach resonance at 834 G. A description of the apparatus used to form molecular Bose-Einstein condensates and degenerate Fermi gases in the two lowest hyperfine states of 6Li, |F = 1/2,mf = 1/2i and |F = 1/2,mf = −1/2i, is presented. An anisotropic 2D optical potential is formed by an elliptical Gaussian beam produced by tightly focusing a circular Gaussian beam in one direction with a cylindrical lens. The trapping frequencies are ∼3 kHz in the axial direction and ∼50 Hz in the radial direction giving an aspect ratio of ∼60. For a non-interacting Fermi gas this allows ∼1700 atoms to populate the lowest transverse harmonic oscillator state and to be considered as quasi-2D. We present the transition from a three dimensional gas to a two dimensional gas at ∼2000 atoms agreeing within experimental uncertainty with the non-interacting atom number. The observations are made through measuring the radial and axial widths of the gas at three different magnetic fields, 810 G, 834 G, and 991 G. The widths σ are then plotted against the atom number N. The curve, log σ(N) = n logN+a, is then fit to the data in the 3D regime and in the 2D regime. The expansion of a non-interacting Fermi gas in the 2D regime changes from a scaling of N1/4 to N1/6 in the 3D regime and for a Bose gas it changes to N1/5. A value for n is extracted in each of these regimes. For each magnetic field we then obtain in the 2D regime: n = 0.27(0.04) at 810 G, n = 0.23(0.02) at 834 G and n = 0.26(0.01) at 991 G. In the 3D regime we obtain: n = 0.195(0.01) at 810 G, n = 0.154(0.01) at 834 G and n = 0.175(0.01 at 991 G. We also observe a change in the aspect ratio as we approach the 2D regime which is not present in our 3D single beam optical dipole trap and is absent in any 3D system. These first results indicate that we have a quasi-2D Fermi gas. The main experimental results are the first indications of a confinement induced resonance in a 6Li Fermi gas through the measurement of the cloud width in the tightly confined direction after a fixed expansion time. We observe a significant deviation in the transverse width of the 2D interacting Fermi gas from the 3D case due to the increased tight confinement. The confinement induced resonance should appear when the 3D scattering length approaches the transverse harmonic oscillator length, az = p~/mωz, where m is the mass and ωz is the tight transverse trapping frequency. For our experimental parameters this leads to a 3D scattering length of 1.03 μm, where the observation is at 819(1) G. In the weakly confined direction there is no such increase in the width due to the free motion in the xy plane. Instead the cloud width across the Feshbach resonance increases monotonically, similar to that of a 3D gas and as described theoretically for a quasi-2D Fermi gas by Lin et al. [1].
Quasi-2D degenerate Fermi gases provide a rich physical environment to explore since many effects that are present do not exist in 3D. This thesis describes the first experiments carried out in a quasi-2D Fermi gas across the broad 6Li Feshbach resonance at 834 G. A description of the apparatus used to form molecular Bose-Einstein condensates and degenerate Fermi gases in the two lowest hyperfine states of 6Li, |F = 1/2,mf = 1/2i and |F = 1/2,mf = −1/2i, is presented. An anisotropic 2D optical potential is formed by an elliptical Gaussian beam produced by tightly focusing a circular Gaussian beam in one direction with a cylindrical lens. The trapping frequencies are ∼3 kHz in the axial direction and ∼50 Hz in the radial direction giving an aspect ratio of ∼60. For a non-interacting Fermi gas this allows ∼1700 atoms to populate the lowest transverse harmonic oscillator state and to be considered as quasi-2D. We present the transition from a three dimensional gas to a two dimensional gas at ∼2000 atoms agreeing within experimental uncertainty with the non-interacting atom number. The observations are made through measuring the radial and axial widths of the gas at three different magnetic fields, 810 G, 834 G, and 991 G. The widths σ are then plotted against the atom number N. The curve, log σ(N) = n logN+a, is then fit to the data in the 3D regime and in the 2D regime. The expansion of a non-interacting Fermi gas in the 2D regime changes from a scaling of N1/4 to N1/6 in the 3D regime and for a Bose gas it changes to N1/5. A value for n is extracted in each of these regimes. For each magnetic field we then obtain in the 2D regime: n = 0.27(0.04) at 810 G, n = 0.23(0.02) at 834 G and n = 0.26(0.01) at 991 G. In the 3D regime we obtain: n = 0.195(0.01) at 810 G, n = 0.154(0.01) at 834 G and n = 0.175(0.01 at 991 G. We also observe a change in the aspect ratio as we approach the 2D regime which is not present in our 3D single beam optical dipole trap and is absent in any 3D system. These first results indicate that we have a quasi-2D Fermi gas. The main experimental results are the first indications of a confinement induced resonance in a 6Li Fermi gas through the measurement of the cloud width in the tightly confined direction after a fixed expansion time. We observe a significant deviation in the transverse width of the 2D interacting Fermi gas from the 3D case due to the increased tight confinement. The confinement induced resonance should appear when the 3D scattering length approaches the transverse harmonic oscillator length, az = p~/mωz, where m is the mass and ωz is the tight transverse trapping frequency. For our experimental parameters this leads to a 3D scattering length of 1.03 μm, where the observation is at 819(1) G. In the weakly confined direction there is no such increase in the width due to the free motion in the xy plane. Instead the cloud width across the Feshbach resonance increases monotonically, similar to that of a 3D gas and as described theoretically for a quasi-2D Fermi gas by Lin et al. [1].
We have studied the transition from two to three dimensions in a low temperature weakly interacting 6Li Fermi gas. Below a critical atom number N(2D) only the lowest transverse vibrational state of a highly anisotropic oblate trapping potential is occupied and the gas is two dimensional. Above N(2D) the Fermi gas enters the quasi-2D regime where shell structure associated with the filling of individual transverse oscillator states is apparent. This dimensional crossover is demonstrated through measurements of the cloud size and aspect ratio versus atom number.
The work described in this thesis can be divided into two parts. The first part is concerned with the production of degenerate Fermi gases and Bose-Fermi mixtures. The second part of the thesis describes studies of strongly interacting spin-mixtures of ultracold fermions. We have cooled a magnetically trapped spin-polarized 6Li Fermi gas into quantum degeneracy by sympathetic cooling with bosonic 23Na. In order to do this, we have converted our previously developed apparatus for Bose-Einstein condensation of 23Na into a two-species cooling facility. We have produced 6Li samples with up to 7 × 107 atoms at temperatures below 0.5TF , and up to 3× 107 atoms at temperatures as low as 0.05TF . Further, we have also produced stable quantum degenerate mixtures of 23Na and 6Li, with several million atoms of each species. In a second set of experiments, we have studied strongly interacting spin-mixtures of 6Li. The strength of interactions was resonantly controlled by applying a magnetic field and tuning to “Feshbach resonances”, which occur when the total energy of the colliding particles coincides with the energy of a bound molecular state. These studies are relevant for the continuing quest for a superfluid transition in a Fermi gas. By studying the inelastic decay of the cloud, we have experimentally identified two Feshbach resonances for the inter-state interactions of the two lowest ground states of 6Li. Most importantly, we found that the lifetime of the strongly interacting system is long enough to observe a superfluid transition in the future. The effects of the coherent collisions in 6Li spin-mixtures were studied using radiofrequency (RF) spectroscopy. We demonstrated the absence of mean field “clock shifts” in a two-state mixture of fermions, and have introduced a three-level method to observe the interaction energy shifts. In the strongly interacting regime, near a Feshbach resonance, we have observed universal behavior of the interaction energy, independent of the value of the scattering length. Finally, we have theoretically studied the effects of Pauli blocking during a free expansion of a degenerate Fermi gas. We have concluded that a strongly interacting normal gas will undergo a hydrodynamic expansion even at zero temperature, due to inefficiency of the Pauli suppression of collisions. Therefore, the already observed hydrodynamic expansion cannot be conclusively interpreted as a signature of superfluidity. Thesis Supervisor: Wolfgang Ketterle Title: John D. MacArthur Professor of Physics
This thesis describes the production of molecular Bose-Einstein condensates (BEC) of 6Li2 dimers and binding energy measurements of p-wave Feshbach molecules. A σ− Zeeman slower is used to produce a continuous beam of isotopically enriched 6Li atoms at speeds low enough to load a magneto-optical trap (MOT). We currently have a flux of slowed atoms of ~ 5 · 106 atoms/s loading the MOT with more than 108 atoms. We then transfer up to 106 atoms in an almost equal spin mixture of the two lowest hyperfine states into an optical dipole trap. We achieved condensates in three different crossed optical dipole trap geometries. The initial low power optical dipole trap was formed using light from a 25WVersaDisk Yb:YAG laser at 1030 nm. It consisted of a 15 W beam crossed with a 13 W beam at about 80 degrees with a waist of approximately 30 μm in each beam. By translating the focus of the second beam we could change the trap geometry from near symmetric to elongated. Nowadays, we produce condensates in a crossed dipole trap formed by a 100 W fibre laser. Both arms are focussed to a waist of 40 μm, cross each other at 14 degrees and have laser powers of ~80 W and ~70 W, respectively. Evaporative cooling is achieved by reducing the laser power near the broad s-wave Feshbach resonance at 834 G. By tuning to the low magnetic field side (770 G) of the Feshbach resonance molecules are formed through three-body recombination at sufficiently low temperatures. Further evaporation leads to the creation of a molecular BEC. After reducing the laser power by a factor of about 1000 in approximately 3 s we have observed more than 30 000 condensed molecules. During the evaporation the temperature decreases from about 100 μK to below 100 nK. We present measurements of the binding energies of 6Li p-wave Feshbach molecules formed in combinations of the |F = 1/2,mF = +1/2i (|1i) and |F = 1/2,mF = −1/2i (|2i) states. The binding energies scale linearly with magnetic field detuning for all three resonances. The relative molecular magnetic moments are found to be 113 ± 7 μK/G, 111 ± 6 μK/G and 118 ± 8 μK/G for the |1i − |1i, |1i − |2i and |2i − |2i resonances, respectively, in good agreement with theoretical predictions.
This thesis presents studies of quantum degenerate atomic gases of fermionic 6Li and bosonic 7Li. Degeneracy is reached by evaporative cooling of 7Li in a strongly confining magnetic trap. Since at low temperatures direct evaporative cooling is not possible for a polarized fermionic gas, 6Li is sympathetically cooled by thermal contact with 7Li. In a first series of experiments both isotopes are trapped in their low-field seeking higher hyperfine states. A Fermi degeneracy of is achieved for 105 fermions. For more than ~300 atoms, the 7Li condensate collapses, due to the attractive interatomic interaction in this state. This limits the degeneracy reached for both species. To overcome this limit, in a second series of experiments 7Li and 6Li atoms are transferred to their low field seeking lower hyperfine states, where the boson-boson interaction is repulsive but weak. The inter-isotope collisions are used to thermalize the mixture. A 7Li Bose-Einstein condensate (BEC) of 104 atoms immersed in a Fermi sea is produced. The BEC is quasi-one-dimensional and the thermal fraction can be negligible. The measured degeneracies are . The temperature is measured using the bosonic thermal fraction, which vanishes at the lowest temperatures, limiting our measurement sensitivity. In a third series of experiments, the bosons are transferred into an optical trap and their internal state is changed to , the lowest energy state. A Feshbach resonance is detected and used to produce a BEC with tunable atomic interactions. When the effective interaction between atoms is tuned to be small and attractive, we observe the formation of a matter-wave bright soliton. Propagation of the soliton without spreading over a macroscopic distance of 1.1~mm is observed.
This thesis reports on a study of the molecular Bose-Einstein condensation (mBEC) of the 6Li. More specifically, using model the Hartree-Fock (HF) model to fit the spatial density of the cloud, we demonstrate that the thermal cloud deplete the condensate, therefore a lower temperature is needed to condense the atoms in the ground state. Moreover the inter-atomic interactions can be controlled via Feshbach resonance (see section 2.2), allowing us to study the effects of the interactions on the thermodynamics properties of a mBEC, especially the critical temperature of the BEC that decreases when we increase the inter-atomic interactions. The description starts from the experimental procedure to produce a mBEC to its physical analysis. We draw a list describing the different chapters presented in the thesis: Chapter 1 Introduction: We present the fundamental difference between fermions and bosons, and we give some example of realisations of Bose-Einstein condensation and degenerate Fermi gas. Chapter 2 Interactions in an ultracold gas: We discuss the role of interactions depending on the type of particles, either fermions or bosons, and the phenomena of Feshbach resonance which is a powerful tool to change the interaction strength between particles. Chapter 3 Degenerate quantum gas: A brief historical section gives the fundamental ideas that initiate the long story of the Bose-Einstein condensation phenomena. We then discuss the mathematical description of a BEC without interactions, which is the starting point of the theoretical analysis used in the chapter 3 where we study the influence of the interactions. Then, we describe the statistics of Fermi gas and the different states that can be reached using Feshbach resonances, in particular the creation of the Feshbach molecules. Chapter 4 Basics of Cold atoms physics: A toolbox of the necessary knowledge to understand the underlying phenomena and concepts used in a cold atoms experiment. We describe the kind of atoms used, how these atoms interact with a magnetic and an electromagnetic field. Chapter 5 Guideline to making a 6Li BEC: This chapter describes the experimental methods used in our laboratory to cool down atoms to the regime of ultra-cold temperature T < 1µK. This section gives details about atomic beams production, the Zeemann slower (radiation force cooling), the magnetooptical trap (MOT) (cooling b molasses), optical dipole trap (evaporative cooling). Also we present the methods used to probe atomic clouds, such as in situ imaging methods. My contribution to this section has been to improve the stability of the evaporative cooling sequence (see section 5.4.4) and to optimize the shape of the the evaporative cooling ramp (see section 5.4.1) in order to maximise the phase space density (PSD), to reach Bose-Einstein condensation. Then, to make precise simulations and measurements to extract values of the trapping frequencies (see section 5.4.5) and scattering length (see section 5.5.4), the latter gives us the interaction strength of the system. These parameter are then used in the next chapter to extract thermodynamics properties of the system. Moreover, I developed the sequence for in situ absorption imaging (see section 5.5.3). Chapter 6 Study of an interacting 6Li mBEC: This chapter discusses the theory to understand the thermodynamics properties of an interacting Bose6 Einstein condensate. It is accomplished by fitting the spatial atomic density profiles with different models, the ideal gas (IG) model, the semi-Ideal (SI) gas model and the Hartree-Fock (HF) model. These three models vary in complexity from the most simple one the IG model to most elaborated one the HF model. The HF model requires a lot more of computational power than the other methods, therefore we present a new numerical method requiring less computational power proposed by Nathan Welch, and demonstrate experimentally through a statistical analysis that the HF model used to determine the thermodynamics properties of a BEC is more accurate to describe utracold atomic gas near the transition. Then, we present the results of the influence of the scattering length on the transition temperature and the chemical potential. My contribution to this section has been to develop tools to analyse the in situ atomic density profiles and improve the fitting program of Nathan (see section 6.3). Also, after taking the measurement of the in situ atomic density profiles, I developed and used the tools for the analysis of the thermodynamics properties of the mBEC (see section 6.4). I also adapted the models described in [1] to our experiment to extract the value of the radius of the condensate (see section 6.5). Finally, with N.Welch we present method based on energy conservation to verify the validity of the different model, my principal contribution to this was to modify the experiment to use this method (see section 6.7). Chapter 7 Initial work: Double well with 6Li molecules: This chapter presents the initial work to produce a double well potential to trap cold atoms and create a Josephson Junction. We present the state of the art of the different effects that have been observed, and discuss future measurements that would be achievable in our laboratory, in particular we would like to use a second specie 133Cs as impurities and observe the influence on Josephson oscillations of 6Li molecules. An experimental description of the double well trapping potential has been implemented and is also described. My contribution to this section was to design and implement the double well trapping potential.
Two sets of studies are described in this thesis. In the first set, an atom interferometry technique was developed for the measurement of the fine structure constant using a BoseEinstein Condensate. In the second set, degenerate Fermi gases were prepared and their properties explored in a regime of strong interactions. We have developed an atom interferometer which is capable of measuring with high precision the “photon recoil frequency” (ωrec). ωrec corresponds to the kinetic energy of an atom recoiling due to absorption of a photon. ωrec can be used to determine the quantity h/matom and the fine structure constant, α. A preliminary measurement using a 23Na BoseEinstein Condensate yielded ωrec with a precision of 7 × 10−6 which deviated by 2 × 10−4 from the currently accepted value. Plausible upgrades to the apparatus should produce a precision of 10−9 which would bring within reach a measurement of h/matom and α in the 10−9 range. Such accuracy would be of considerable scientific and metrological import. A quantum degenerate gas of 6Li fermions was produced by sympathetic cooling with 23Na bosons in a two-species atom trapping apparatus. The cooling strategy was optimized to enable production of fermions with atom numbers up to 7 × 107 at half the Fermi temperature (TF ), or temperatures down to 0.05TF ∼ 100 nK with 3× 107 atoms. We can also produce degenerate Bose-Fermi mixtures with several million atoms in each species. We studied the behavior of mixtures of fermi gases in regimes of strong interactions near “Feshbach” resonances. A study of system stability enabled the experimental observation of two such Feshbach resonances. We carried out a theoretical study to interpret the observation (by many experimental groups) of hydrodynamic behavior of fermi gases during expansion out of an atom trap in a strongly-interacting regime. The study concluded that this behavior was not a qualitative signature of fermionic superfluidity and could arise from classical collisions. Finally, radio-frequency (RF) spectroscopy was used to probe interaction in 6Li. We demonstrated the absence of mean field “clock” shifts of RF transitions in a two-(spin)state fermion system. Using a three-state system, we measured the interaction strength between different spin states. The measurements near Feshbach resonances indicate a saturation of the interaction strength at a large negative value. This result is of relevance in the continuing quest for fermionic superfluidity in atomic gases. Thesis Supervisor: David E. Pritchard Title: Cecil and Ida Green Professor of Physics Thesis Supervisor: Wolfgang Ketterle Title: John D. MacArthur Professor of Physics