The variation of a magnetic field in the vicinity of a zero-energy resonance allows highly vibrationally excited molecules (‘Feshbach molecules’) to be produced from an ultracold atomic gas. In this thesis, we study the dynamics of this process. We begin by studying the dissociation of Feshbach molecules, showing that in the limit of a sudden jump the shape of the spectrum of dissociated atoms can act as a probe of the zero-energy resonance. For some resonances, such jumps are within reach of current experiments. We also study the intermediate region between sudden jumps and asymptotically wide, linear ramps. It is shown from a precise derivation how the latter limit leads to a universal spectrum with a shape independent of the implementation of the two-body physics, provided that the near-resonant scattering properties are correctly modelled. We then turn to the dynamics of Feshbach molecule production from thermal and condensed gases. Our microscopic quantum dynamics approach includes the exact twobody evolution as an input to the many-body calculations. We show that in the long-time limit, and the Markov limit for the interactions, the non-Markovian Boltzmann equation (NMBE) we derive for the one-body density matrix reduces to the normal Boltzmann equation. In the limit of short times and small depletion of the atomic gas, the molecule production efficiency can be calculated by thermally averaging the two-body transition probability density. This thermal averaging technique is applied to studies of the formation of Feshbach molecules using a magnetic field modulation that is near-resonant with the molecular bound state energy. The continuum is shown to have a significant effect on both the dynamics and efficiency of this process. We examine the dependence of the molecule production efficiency on the duration, amplitude and frequency of the modulation, as well as the temperature and density of the gas. This method of producing molecules is effective for a wide range of bound state energies, but requires sufficient variation of the two-body energy levels with magnetic field. Lastly, we implement the NMBE for the case of a fast linear ramp across a Feshbach resonance. The solution of this equation is made feasible by including a large part of the required computation in the kernel, which is calculated in advance. The NMBE allows predictions of the molecule production efficiency which go beyond the thermal averaging technique by accounting for the depletion and rethermalisation of the continuum. In the limit of small depletions, the two approaches give the same results. As the depletion increases, the two approaches differ due to many-body effects limiting the maximum possible molecule production efficiency. We have observed this in our simulations by considering higher-density gases. We have therefore shown the suitability and practicability of this beyond mean-field approach for application to further problems in the production of Feshbach molecules from ultracold gases.

We obtain the characteristic relaxation rates and relaxation modes of a Bose gas in three regimes. The classical regime corresponds to a classical gas of hard spheres and the quantum regime corresponds to an interacting quantum Bose gas with no Bose-Einstein condensate present. In the condensed regime a Bose-Einstein condensate is present and modifies the behavior of the gas. In each regime there is a different kinetic equation that describes the evolution of the relevant distribution function. The classical kinetic equation is the Boltzmann equation and the quantum kinetic equation with no condensate present is the Uehling-Uhlenbeck equation. When a condensate is present, we derive a new kinetic equation that describes the evolution of the momentum distribution of Bogoliubov excitations or bogolons. For each of the three kinetic equations, we linearize the collision integral and use it to generate the elements of a collision matrix. The eigenvalues of this matrix give us the characteristic relaxation rates and the eigenvectors give us the relaxation modes. We report numerical results for the eigenvalues in each regime as the particle species, density and temperature of the gas are varied. / text

Bose-Einstein condensates (BECs), with their superfluid behavior, quantized vortices, and high-level of control over trap geometry and other system parameters provide a compelling environment for studies of quantum fluid dynamics. Recently there has been an influx of theoretical and numerical progress in understanding the superfluid dynamics associated with two-dimensional quantum turbulence, with expectations that complementary experiments will soon be realized. In this dissertation I present progress in the development of an experimental toolkit that will enable such experimental studies of two-dimensional quantum turbulence. My approach to developing this toolkit has been twofold: first, efforts aimed at the development of experimental techniques for generating large disordered vortex distributions within a BEC; and second, efforts directed towards the design, implementation, and characterization of a quantum vortex microscope. Quantum turbulence in a superfluid is generally regarded as a disordered tangle of quantized vortices in three dimensions, or a disordered planar distribution of quantized vortices in two dimensions. However, not all vortex distributions, even large disordered ones, are expected to exhibit robust signatures of quantum turbulence. Identification and development of techniques for controlled forcing or initialization of turbulent vortex distributions is now underway. In this dissertation, I will discuss experimental techniques that were examined during the course of my dissertation research, namely generation of large disordered distributions of vortices, and progress towards injecting clusters of vortices into a BEC. Complimentary to vortex generation is the need to image these vortex distributions. The nondeterministic nature of quantum turbulence and other far-from-equilibrium superfluid dynamics requires the development of new imaging techniques that allow one to obtain information about vortex dynamics from a single BEC. To this end, the first vortex microscope constructed as part of my dissertation research enabled the first in situ images of quantized vortices in a single-component BEC, obtained without prior expansion. I have further developed and characterized a second vortex microscope, which has enabled the acquisition of multiple in situ images of a lattice of vortex cores, as well as the acquisition of single in situ images of vortex cores in a BEC confined in a weak hybrid trap. In this dissertation, I will discuss the state-of-the-art of imaging vortices and other superfluid phenomena in the University of Arizona BEC lab, as indicated by the examined performance of the quantum vortex microscope.

Bose-Einstein condensates of dilute gases provide a rich and versatile platform to study both single-particle and many-body quantum phenomena. This thesis describes several experiments using a Bose-Einstein condensate of Rb-87 as a model system to study novel matter-wave effects that traditionally arise in vastly different systems, yet are difficult to access. We study the scattering of a particle from a repulsive potential barrier in the non-asymptotic regime, for which the collision dynamics are on-going. Using a Bose-Einstein condensate interacting with a sharp repulsive potential, two distinct transient scattering effects are observed: one due to the momentary deceleration of particles atop the barrier, and one due to the abrupt discontinuity in phase written on the wavepacket in position-space, akin to quantum reflection. Both effects lead to a redistribution of momenta, resulting in a rich interference pattern that may be used to reconstruct the single-particle wavefunction. In a second experiment, we study the response of a particle in a periodic potential to an applied force. By abruptly applying an external force to a Bose-Einstein condensate in a one-dimensional optical lattice, we show that the initial response of a particle in a periodic potential is in fact characterized by the bare mass, and only over timescales long compared to that of interband dynamics is the usual effective mass an appropriate description. This breakdown of the effective mass description on fast timescales is difficult to observe in traditional solid state systems due to their large bandgaps and fast timescale of interband dynamics. Both these experiments make use of the condensate's long coherence length, and the ability to shape and modulate the external potential on timescales fast compared to the particle dynamics, allowing for observation of novel matter-wave effects.

Bose-Einstein condensates of dilute gases provide a rich and versatile platform to study both single-particle and many-body quantum phenomena. This thesis describes several experiments using a Bose-Einstein condensate of Rb-87 as a model system to study novel matter-wave effects that traditionally arise in vastly different systems, yet are difficult to access. We study the scattering of a particle from a repulsive potential barrier in the non-asymptotic regime, for which the collision dynamics are on-going. Using a Bose-Einstein condensate interacting with a sharp repulsive potential, two distinct transient scattering effects are observed: one due to the momentary deceleration of particles atop the barrier, and one due to the abrupt discontinuity in phase written on the wavepacket in position-space, akin to quantum reflection. Both effects lead to a redistribution of momenta, resulting in a rich interference pattern that may be used to reconstruct the single-particle wavefunction. In a second experiment, we study the response of a particle in a periodic potential to an applied force. By abruptly applying an external force to a Bose-Einstein condensate in a one-dimensional optical lattice, we show that the initial response of a particle in a periodic potential is in fact characterized by the bare mass, and only over timescales long compared to that of interband dynamics is the usual effective mass an appropriate description. This breakdown of the effective mass description on fast timescales is difficult to observe in traditional solid state systems due to their large bandgaps and fast timescale of interband dynamics. Both these experiments make use of the condensate's long coherence length, and the ability to shape and modulate the external potential on timescales fast compared to the particle dynamics, allowing for observation of novel matter-wave effects.

In the present thesis, “Hydrodynamics of Binary Bose-Einstein Condensates and Hydro-elasticity of the Inner Crust of Neutron Stars”, the hydrodynamic effects, instabilities and superfluid turbulence in binary immiscible ultracold gases, and hydro-elastic macroscopic coupled modes and microscopic structure of the inner layers of the crust of neutron stars, are studied. The ultracold gas dynamics can be realized in the laboratory. The excitation modes of the inner crust determine a number of observable properties such as elasticity, thermal properties and mass transport properties. Here we focus on expanding the details, rather than repeating the results presented in the published articles. In the part of the thesis related to atomic ultracold gases, we utilize the physical parameters in the experimentally realizable parameter region. We numerically simulate the coupled non-linear Schrödinger equations, and calculate observable quantities, such as phase and modulus of the order parameter, conditions needed for observation of the Rayleigh-Taylor instability and for turbulence generation. The numerical calculations are accompanied by analytical description of the processes. The dispersion relation for capillary-gravitational waves at the interface between two ultracold gases, is derived straightforwardly from the superfluid Lagrangian. The equations of motion for centre-of-mass of the superfluids are derived, and then used in our model of the quantum swapping of immiscible superfluids pressed by a strong external force. By numerical simulation, we find that the Kelvin-Helmholtz instability which occurs at the non-linear stage of the Rayleigh-Taylor instability, can generate quantum turbulence with peculiar properties. We find that two-dimensional superfluid systems with weak inter-component repulsion are different from previously studied strongly repulsive binary superfluids, because the quantum Kelvin-Helmholtz instability in weakly repulsive superfluids rolls up the whole interface forming a vortex bundle, similarly to dynamics of the shear fluid layers in the classical hydrodynamics. Production of vortex bundles favours the Kolmogorov spectrum of turbulence, and we find that the Kolmogorov scaling indeed is present in a freely decaying turbulence. In the part of the thesis related to neutron stars, we study the inner crust of neutron stars, where the fully ionized atomic nuclei coexist with a superfluid of neutrons. The interaction between superfluid neutrons and the crystallized Coulomb plasma is due to the interaction between density perturbations (interaction of the scalar type), and between the current - the non-dissipative entrainment effect (interaction of the vector type). We calculate velocities of the collective modes of the crystal coupled to superfluid neutrons. As an input we use the results of microscopic nuclear calculations in the framework of the compressible liquid drop model (the Lattimer and Swesty equation of state), and more recent effective Thomas-Fermi calculations with shell corrections (N. Chamel, and the Brussels theoretical nuclear physics group). Knowledge of velocities as functions of the matter density in the inner crust is important for calculation of a number of dynamic and transport properties. The heat transport properties of the inner crust are directly observable in accreting binary systems (low-mass x-ray binaries). The mass transport properties of the inner crust are directly linked to the rotational evolution, being a key physical ingredient of the pulsar glitch phenomenon. The elastic properties are related to the vibrational modes of the star, and to the breaking stress of the crust. In the second part of our work on neutron stars we investigate the microscopic structure of the inner crust treating the structure as an anisotropic crystal coupled to s-wave superfluid neutron liquid. As we analyse dynamics of the elementary excitations at higher wavenumbers (smaller scales), we reach the edge of the first Brillouin zone. The Lattimer-Swesty data is applicable for wavenumbers much smaller than the edge of the first Brillouin zone. We extrapolate the data through the whole first Brillouin zone to calculate the fastest growth rate of the unstable modes. The crucial step is to calculate the mode velocities in anisotropic crystal incorporating both the induced neutron-proton interactions, and the electron screening properties. We find that the combined influence of these two effects leads to softening of the longitudinal phonon of the lattice above about the Thomas-Fermi screening wavenumber of the electrons. The critical wavenumber when the frequency becomes purely imaginary is about  1/5 - 2/3  of the reciprocal lattice vector, thus validating our assumption. The imaginary mode frequency implies instability at finite wavenumbers. Our calculations suggest that the mode at the first Brillouin zone edge is the most unstable, and thus the structure experiences a displacive phase transition when the central ion of a unit cell of the body-cubic-centred lattice, is displaced to the cube face. Thus, the electronic structure of matter at densities above the neutron drip [1], is richer than previously appreciated, and new microscopic calculations of nuclear structure are necessary which take into account the high-wavenumber physics. Such calculations will provide crucial input to models interpreting the quasi-periodic oscillations in Soft Gamma Repeaters as magnetar x-ray flares, and to the theory of glitches of neutron stars. [1] The neutron drip density is ~3×1011 g cm-3.

Targeted Energy Transfer is a resonance phenomenon in coupled anharmonic oscillators. In this thesis we investigate if the concept of Targeted Energy Transfer is applicable to Bose-Einsteain condensates in optical lattices. The model used to describe Bose-Einstein condensates in optical lattices is based on the Gross-Pitaevskii equation. Targeted Energy Transfer in these systems would correspond to energy being transferred from one lattice site to another. We also try to expand the concept of Targeted Energy Transfer to a system consisting of three sites, where one of the sites are considered a perturbation to the system. We have concluded that it is possible to achieve Targeted Energy Transfer in a three-site system. The set-up of the system will in some of the cases studied lead to interesting properties, such as more energy being transferred to the acceptor site than what was initially localized on the donor site.

Ce manuscrit présente mes travaux de recherche depuis la fin de ma thèse début 2005. J'essaye en particulier de replacer mes recherches dans leur contexte, et d'expliquer mes choix scientifiques de façon chronologique. Les principaux résultats ont donne lieu a des publications. Celles-ci ont été écrites avec soin et c'est pourquoi, j'ai choisi de ne décrire les résultats principaux que de fa con succincte. Pour avoir plus de détails et pour retrouver les gures expérimentales, mes articles sont reproduits a la n de chaque chapitre et cites à l'endroit adéquat. J'ajoute parfois des parties plus techniques qui ne sont pas explicitées dans les publications. Le Chapitre 2 est une courte introduction au domaine des atomes ultrafroids, s'adressant a un lecteur non spécialiste. Cela permet de présenter mon domaine de recherche, c'est a dire l'utilisation des gaz d'atomes ultra-froids en tant que systèmes modèles pour l'étude des propriétés quantiques des systèmes a N-corps. Le chapitre 3 est consacre a mes recherches durant mon séjour postdoctoral de deux ans dans le groupe de T. Esslinger a l'ETH Zurich. J'ai participé a une série d'expériences utilisant une cavité de haute nesse pour détecter individuellement les atomes issus d'un gaz ultra-froids. J'explique la théorie de ce système quantique ouvert et sa résolution numérique qui permet de comprendre quantitativement le processus de mesure de la présence d'un atome dans la cavit e. Experimentalement, nous avons mis en œuvre une technique consistant a extraire deux faisceaux d'atomes en deux points distincts du nuage d'atomes pour avoir accès a la fonction de corrélation g1 en perturbant peu le système. Nous avons ainsi étudie la dynamique de la transition de Bose-Einstein de fa con beaucoup plus ne que cela n'avait et e 15 fait auparavant. Nous avons pu comparer la dynamique de croissance de la densite a la dynamique d'apparition de la cohérence. Nous avons ensuite observe le comportement de la fonction g1 dans le régime critique tr es proche de la condensation et extrait une valeur expérimentale de l'exposant critique associe a la longueur de cohérence. En n, nous avons mis en place un ascenseur a atomes constitue de deux lasers contra-propageants contrôles en phase. Le condensat de Bose-Einstein a ainsi et e transport e vers la cavite de haute nesse pour créer un système quantique couple atome-rayonnement dans un régime de couplage extrême et nouveau. Le chapitre 4 est d edi e a mes trois premières années (2007-2009) au laboratoire Charles Fabry de l'Institut d'Optique dans le groupe d'optique atomique. J'explique d'abord le d em enagement du syst eme exp erimental a Palaiseau ainsi que les differentes ameliorations apportees au piege magn etooptique 2D. Nous avons realise un condensat de Bose-Einstein de rubidium par une methode entierement optique. Notre laser de piegeage est un laser a bre dop ee erbium de puissance a 1565 nm. Cette longueur d'onde n'avait jamais et e utilisee auparavant dans des experiences avec des atomes ultrafroids. Nous avons utilise les specifcites de ce laser, et notamment le fort decalage lumineux de la transition optique pour demontrer une technique de tomographie du champ lumineux vu par les atomes. Cette compr ehension des d ecalages lumineux nous a guide vers une melasse tres fortement decalee (environ 200 MHz) pour charger les atomes le plus e cacement possible dans le piege optique. Nous avons montre qu'une nouvelle geometrie permet de contrôler independamment la profondeur du piege et sa raideur, et ainsi d'optimiser l' evaporation dans le piege optique. En n, nous avons utilise notre dispositif pour une premiere application qui consiste a faire rebondir les atomes sur une onde stationnaire dans le but d'allonger le temps d'interrogation dans les interferometres o u les atomes sont en chute libre. Le trampoline a atomes peut fonctionner dans un r egime quantique o u les interferences entre chemins quantiques permettent d'allonger le temps de levitation. Le chapitre 5 est consacre a mes recherches sur les gaz 2D et desordonnes qui se poursuivent encore aujourd'hui. Lorsque les interactions entre atomes sont negligeables, la physique est a un corps. Le phenomene de diffusion a et e mis en evidence et caracteris e pour la premi ere fois avec des atomes ultra-froids dans un potentiel conservatif. En allant vers le regime quantique ou la longueur d'onde de DeBroglie des atomes devient de l'ordre de la taille caracteristique des grains de desordre, on s'attend alors a voir des effets de localisation d'Anderson, li es aux interferences entre ondes de mati ere. Les conditions necessaires pour observer la localisation d'Anderson dans un gaz en expansion sont detaillees. Pour des gaz pieges, les interactions jouent un r^ole predominant a 2D et la transition de Bose-Einstein est alors rem- 16 placee par une transition super uide de type Berezinskii-Kosterliz-Thouless. Exp erimentalement, nous avons etudie cette transition via la distribution en impulsion qui permet de caract eriser les propri et es de coherence du gaz. Ensuite, nous avons observe quantitativement l'in uence du d esordre sur la transition super uide. Nous formons alors un systeme quantique complexe pour lequel il n'y a pas de pr ediction th eorique pr ecise et dont la physique est liee a celle de certains materiaux de matiere condensee. J'insiste sur le rôle de la longueur de correlation du desordre en la comparant aux longueurs caracteristiques du gaz. Dans le regime d'un desordre correle a longue portee, l'approximation de densite locale est valable dans le desordre et alors des predictions quantitatives sur le diagramme de phase du systeme sont possibles. En n, je propose des directions pour nos recherches futures. Le refroidissement du potassium par un m ethode enti erement optique est un d e experimental mais permettra d'avoir acces a des resonances de Feshbach larges et ainsi de controler la force des interactions. De plus, un champ magnetique effectif donnera un parametre de controle suppl ementaire sur le syst eme. Ces deux outils seront utiles non seulement pour l' etude de la physique a un corps mais aussi pour une etude plus pr ecise du gaz de Bose 2D en pr esence de d esordre et d'interactions. On cherchera notamment a mettre en evidence le diagramme de phase et a observer une phase isolante exotique, le verre de Bose.

In this thesis we study the effects of the dipole-dipole interaction on the physics of ultracold quantum gases, both bosonic and fermionic, within the theoretical framework provided by the mean-field regime. This kind of interaction takes place in ultracold atomic gases (for instance 52Cr or 164Dy) due to their atomic magnetic dipole moment, and in ultracold molecular gases due to the magnetic or electric dipole moment. In the case of quantum gases of bosonic atoms, or Bose-Einstein condensates, the dipole-dipole interaction can be studied within mean-field approximation using the Gross-Pitaevskii equation, which now contains a new non-linear term due to the dipole-dipole interaction. We investigate, on the one hand dipolar condensates confined in harmonic traps, and on the other dipolar condensates confined in toroidal traps. In the harmonic geometry, our focus is on the study of the ground state and the quantized vortex state, where the density profile is characterized as well as some properties leading to the process of vortex formation, such as the critical frequency and the energy barrier that has to be overcome to bring the vortex from the surface to the centre of the gas. We finish the study of dipolar condensates in harmonic traps by dynamically simulating the precession frequency of an off-center vortex in a non-rotating condensate. In the toroidal geometry the dipolar effects are strongly magnified when the polarization axis of the dipoles is perpendicular to the trap symmetry axis. In this case, the anisotropic structure of the density can be understood as the response of the system to the double-well effective potential along the ring. We have studied the dynamics of this system when the initial number of atoms in the left and right wells is imbalanced, predicting Josephson and self-trapping oscillations depending on the initial condition. This has led us to name this new system as Self-induced Josephson Junction. We have studied in detail the self-trapping regime and we have seen that the particle flux inversion is closely related to the crossing of vortices across the Josephson junctions. This result opens the door to establishing a more direct connection between the phase-slip regime, widely addressed in superfluid helium, and the self-trapping regime of condensates. In the case of quantum gases of fermionic dipolar particles, we have studied how the radial quadrupole mode allows one to distinguish between hydrodynamic and collisionless regimes. We have analytically calculated the frequency of this mode in the mean-field approximation, generalizing the results from the Thomas-Fermi approximation for trapped ideal Fermi gases. On the one hand, we observe that the frequency in the hydrodynamic regime is smaller than in non-dipolar Fermi gases, while in the collisionless regime the frequency is larger or smaller than that corresponding to the non-interacting system depending on the geometry of the harmonic trap. On the other hand, we predict that reducing the trap deformation (aspect ratio) an observable jump in the frequency of the radial quadrupole mode would take place, which would correspond to the transition between the collisionless and hydrodynamic regimes, for instance when the gas undergoes the transition to the superfluid state. / En aquesta tesi s’estudien els efectes de la interacció dipol-dipol en la física dels gasos qu`antics ultrafreds, tant de caràcter bosònic com fermiònic, i dins del marc teòric del règim de camp mig. En el primer cas considerem condensats de Bose-Einstein dipolars confinats tant en trampes harmòniques com toroidals, descrivint-ne la geometria de l’estat fonamental i de l’estat de vòrtex quantitzat.En la geometria toroidal els efectes dipolars es veuen fortament magnificats quan l’eix de polarització dels dipols és perpendicular a l’eix de simetria de la trampa. Aquesta configuració ens permet introduïr el concepte de Junció de Josephson Autoinduïda (Self-induced Josephson Junction), en la qual hem predit oscil•lacions de Josephson i d’autoatrapament (self-trapping) depenent de la condició inicial. Estudiant en detall el règim d’autoatrapament hem vist que la inversió del flux de partícules està fortament lligada al creuament de vòrtexs quantitzats a travès de les unions de Josephson. Aquest resultat obre les portes a establir una relació més directa entre el règim dinàmic de salts de fase (phaseslips), àmpliament estudiat en heli superfluid, i el règim d’autoatrapament propi dels condensats. Finalment, en el cas de gasos quàntics de partícules dipolars fermiòniques, hem estudiat com les excitacions col•lectives, en concret el mode quadrupolar radial, permeten distingir entre els règims hidrodinàmic (que pot ser tant degut a la rapidesa de les interaccions com a la superfluidesa) i nocol•lisional (que té lloc quan les interaccions són a tan baixa freqüència que efectivament es poden negligir).

This thesis investigates thermalisation, correlations and entanglement in Bose-Einstein condensates. Bose-Einstein condensates are ultra-cold collections of identical bosonic atoms which accumulate in a single quantum state, forming a mesoscopic quantum object. They are clean and controllable quantum many-body systems that permit an unprecedented degree of experimental flexibility compared to other physical systems. Further, a tractable microscopic theory exists which allows a direct and powerful comparison between theory and experiment, propelling the field of quantum atom optics forward at an incredible pace. Here we explore some of the fundamental frontiers of the field, examining how non-classical correlations and entanglement can be created and measured, as well as how non-classical effects can lead to the rapid heating of atom clouds. We first investigate correlations between two weakly coupled condensates, a system analogous to a superconducting Josephson junction. The ground state of this system contains non-classical number correlation arising from the repulsion between the atoms. Such states are of interest because they may lead to more precise measurement devices such as atomic gyroscopes. Unfortunately thermal fluctuations can destroy these correlations, and great care is needed to experimentally observe non-classical effects. We show that adiabatic evolution can drive the isolated quantum system out of thermal equilibrium and decrease thermal noise, in agreement with a recent experiment [Esteve et al. Nature 455, 1216 (2008)]. This technique may be valuable for observing and using quantum correlated states in the future. Next, we analyse the rapid heating that occurs when a condensate is placed in a moving periodic potential. The dynamical instability responsible for the heating was the subject of much uncertainty, which we suggest was due to the inability of the mean-field approximation to account for important spontaneous scattering processes. We show that a model including non-classical spontaneous scattering can describe dynamical instabilities correctly in each of the regimes where they have been observed, and in particular we compare our simulations to an experiment performed at the University of Otago deep inside the spontaneous scattering regime. Finally, we proposed a method to create and detect entangled atomic wave-packets. Entangled atoms are interesting from a fundamental perspective, and may prove useful in future quantum information and precision measurement technologies. Entanglement is generated by interactions, such as atomic collisions in Bose-Einstein condensates. We analyse the type of entanglement generated via atomic collisions and introduce an abstract scheme for detecting entanglement and demonstrating the Einstein-Podolsky-Rosen paradox with ultra-cold atoms. We further this result by proposing an experiment where entangled wave-packets are created and detected. The entanglement is generated by the pairwise scattering that causes the instabilities in moving periodic potentials mentioned above. By careful arrangement, the instability process can be controlled to to produce two well-defined atomic wave-packets. The presence of entanglement can be proven by applying a series of laser pulses to interfere the wave-packets and then measuring the output populations. Realising this experiment is feasible with current technology.