This thesis deals with elastic and inelastic collisions among ultra cold rubidium atoms. In particular, the impact of increased interaction on the stability of Bose-Einstein condensates is investigated both, experimentally and theoretically. A new decay mechanism is identified that significantly reduces the life time of condensates with respect to the expectation based on the known inelastic collision rates. Along with the interaction also the column density of the condensate is increased so that it eventually becomes impenetrable to products of exoergic collisions. These products then trigger avalanches of elastic collisions in which a significant part of the energy released in the initiatory exoergic collision is distributed among condensate atoms. Collisional avalanches are favoured by the isotropy of the differential scattering cross section in the low energy (s-wave) limit and occur as soon as the elastic collision rate exceeds a critical value. Since avalanches not only reduce the life time of the condensate but also disturb its thermodynamic equilibrium, their occurrence restricts the possibilities to explore the strongly interacting regime. A simple model is introduced to quantify the impact of avalanches on the decay and agrees well with experimental data. The thesis further contains the description and the characterisation of a new experimental apparatus which allows to create large atomic condensates with excellent reproducibility and high repetition rate complying with the requirements for the investigation of field-induced elastic scattering resonances.
This thesis deals with elastic and inelastic collisions among ultra cold rubidium atoms. In particular, the impact of increased interaction on the stability of Bose-Einstein condensates is investigated both, experimentally and theoretically. A new decay mechanism is identified that significantly reduces the life time of condensates with respect to the expectation based on the known inelastic collision rates. Along with the interaction also the column density of the condensate is increased so that it eventually becomes impenetrable to products of exoergic collisions. These products then trigger avalanches of elastic collisions in which a significant part of the energy released in the initiatory exoergic collision is distributed among condensate atoms. Collisional avalanches are favoured by the isotropy of the differential scattering cross section in the low energy (s-wave) limit and occur as soon as the elastic collision rate exceeds a critical value. Since avalanches not only reduce the life time of the condensate but also disturb its thermodynamic equilibrium, their occurrence restricts the possibilities to explore the strongly interacting regime. A simple model is introduced to quantify the impact of avalanches on the decay and agrees well with experimental data. The thesis further contains the description and the characterisation of a new experimental apparatus which allows to create large atomic condensates with excellent reproducibility and high repetition rate complying with the requirements for the investigation of field-induced elastic scattering resonances.
In this thesis we numerically compute the scattering lengths and bound states for sodium-rubidium collisions at low energy. This work was motivated by experiments which aim to produce Bose-Einstein condensates (BEC) mixtures of sodium-rubidium. Elastic collision properties are important for the rethermalization of the atoms during the evaporative cooling process. Inelastic processes, which we also discuss to some extent, cause trap losses in those experiments. In order to reach the required temperature and density the elastic collision rates should be sufficiently large compared to the inelastic rates. The scattering lengths, which completely specify the elastic collision parameters at low energy, determine the miscibility and phase diagram of the sodium-rubidium condensate mixture. We calculate the scattering lengths approximately and find agreement with previous calculations indicating that miscible phases of sodium and rubidium condensates do not appear to be feasible in the absence of external fields.
Summary form only given. In a Bose-Einstein condensate, due to the very low temperature, the s-wave scattering length can be used as a measure for the strength of the atom-atom interaction. Under typical experimental conditions, this interaction is weak and hence can be treated in terms of a mean field. However, when scattering length is large or the density is high, the mean field approximation breaks down. In this collisional (hydrodynamic) regime, effects of the interactions such as quantum depletion or shifts in the frequencies of the elementary excitations become large. It is therefore of great interest to study condensates close to or in the collisional regime. It has been demonstrated in recent experiments that the scattering length and thus the interactions among the atoms can be tuned by means of a Feshbach resonance (Inouye et al, 1998; Courteille et al., 1998; Vuleti et al., 1999; Cornish et al., 2000). In the vicinity of Feshbach resonances, however, the increase of the cross-section for elastic collisions is accompanied by a dramatic increase of particle losses. In this paper we report on the observation of anomalous losses from a /sup 87/Rb condensate with a high column density in the absence of an inelastic scattering resonance.
This thesis presents the development of an experimental apparatus to produce Bose-Einstein condensates (BECs) with tunable interparticle interactions. The ability to precisely control the strength of these interactions, and even to switch them from repulsive to attractive, allows one to probe novel regimes of condensate physics, from the collapse of attractively interacting BECs and the formation of solitary matter-waves to the observation of beyond mean-field effects in strongly repulsive condensates. The construction and characterisation of both a single and crossed beam optical dipole trap is presented. In the single beam case we develop a technique allowing the guided transport of atoms along the beam and up to a room-temperature surface; a technique which can be used to evaporatively cool the trapped atomic cloud. We produce Bose-Einstein condensates of 87Rb in the F=1, mF=-1 state in this trap, comparing the effect of beam waist on the evaporation trajectory. In the crossed beam trap Bose-Einstein condensation of 87Rb is realised in three distinct trapping configurations, along with a 1D optical lattice formed by changing the polarisation of the beams. A method of direct cooling of 85Rb atoms in the crossed trap is developed using a magnetic Feshbach resonance to precisely tune both the elastic and inelastic scattering properties of the atoms. The resonance used for this work occurs at 155G in collisions between atoms in the F=2, mF=-2 state of 85Rb. Bose-Einstein condensates of up to 40,000 85Rb atoms are formed in this trap and we demonstrate the presence of tunable interatomic interactions, exploring the collapse phenomenon associated with attractive condensates. By loading the 85Rb condensate into a quasi-1D waveguide we show that stable attractive condensates can be created, taking the form of bright solitary matter-waves. We observe a solitary wave of ~2,000 atoms which propagates, without dispersion, along the waveguide over a distance of ~1.1mm. The particle-like nature of the solitary wave is demonstrated by classical reflection of the wavepacket from a repulsive Gaussian barrier.
In this thesis we investigate theoretically the cold collision properties of metastable neon (Ne*). The collisions of metastable rare gases (Rg*) exhibit a plethora of interesting phenomena: Due to the high internal excitation energies of the metastable atoms ionization processes are very likely. The produced ions can be detected with single-ion efficiency in the experiment. In experiments with atomic Rg* gases at cold temperatures T~1 mK, this allows for in-situ monitoring of the real-time dynamics of the atomic gas and for example for the study of two-particle correlations. Furthermore, in cold Rg* gases one can examine the prospects of creating a Bose-Einstein condensate or a degenerate Fermi gas of Rg*. Ne* is a promising candidate for Bose-Einstein condensation and in the group of G. Birkl the properties of cold gases of Ne* are investigated experimentally. Two-body loss rate coefficients due to the ionization processes and elastic cross sections have been measured. It is useful to complement the experimental measurements with a theoretical study in order to obtain a better understanding of the cold Ne* gases and the collision physics of Ne*. At the low temperatures of the atomic gases, these gases are dilute with particle number densities of n~10^9 cm^{-3}. For these low densities, the behavior of the gas is determined mainly by two-body collision physics which is solved by quantum scattering theory in terms of the S matrix. The collision physics is given by short-range interactions and long-range interactions of the atoms. The short-range and long-range interaction potentials of Ne* have been calculated in the molecular basis of Ne2. In this thesis we first calculate short-range and long-range interactions of atoms and molecular basis states of diatomic molecules with a simpler electronic structure than Ne* and Ne2 in order to introduce the notation and to discuss the characteristics of the calculated molecular interaction potentials and in order to employ them in the scattering calculations for Ne*. Instead of solving the full scattering equations by taking into account all the molecular interaction potentials, we demonstrate that the collision physics of Ne* can be described in terms of a coupled two-channel model with a single interaction potential only which describes the elastic scattering of Ne*. In this model, ionization in Ne* collisions is described by the transition of the upper elastic interaction channel to the lower channel, representing the loss or ionization channel. We introduce two versions of the two-channel model in this work. In the first version, the two channels are given by square-well potentials. We can solve the scattering equations analytically and for complex wave numbers k and study the solutions in the complex k plane. With an expansion of the S matrix in terms of its poles in the complex k plane we find a parametrization of the two-body loss rate coefficients. We show that these coefficients describe ionizing collisions of Ne* by comparing them to the experimental measurements and to the numerical results obtained by the two-channel model in the second version. In this second version of the two-channel model, the elastic scattering channel is given by a realistic interaction potential of Ne*, consisting of a calculated short-range and long-range molecular potential, and the loss channel by a model ionization potential. For this model we calculate the two-body loss rate coefficients and the elastic cross sections for the isotope mixtures of Ne* which have been measured experimentally and demonstrate that the free potential parameters can be optimized to the experimental data to achieve very good agreement of the numerical results with the experimental measurements. We discuss the validity of the two-channel model by comparing the results to existing models for cold Rg* collisions and find that the two-channel model of this work is a useful extension to these models.
Collision physics plays a key role in understanding the properties of ultracold gases. First, collisions ensure thermalization in trapped atomic samples, which is essential for the success of evaporative cooling. Second, interactions between atoms determine the equilibrium shape and the dynamics of the condensate once it is formed. Finally collisions limit the achievable spatial densities via inelastic processes, which ultimately determine whether a given atomic species may lead to Bose-Einstein condensation. The purpose of the following lectures is to give a basic introduction to collision physics with a special emphasis on the low temperature aspect of these interactions. We will consider here only binary interactions between atoms since one of the remarkable prop-erties of the physics of ultra-cold dilute gases is that all the macroscopic properties of the system can be directly deduced from the two-body interaction potential. In particular the essential concept of scattering length will be introduced and developped for some simple potentials, using as practical examples experimental results obtained on atomic cesium and rubidium. The present lectures are complementary to the ones given by D. Heinzen where a detailed comparison between the collisional properties of the various alkali atoms will be found, together with explanations of the spectroscopic approaches used in the precise determination of interatomic potentials. 1. – Scattering theory: a brief reminder 1.1. Collision between two particles. – We consider here an elementary collision process between two particles 1 and 2 with the same mass M, interacting through the potential