Research

Dept. of Physics & Astronomy
Physics S-114
Stony Brook University
Stony Brook, NY 11794-3800
Tel. 631 632-4497 and -8043
Fax 631 632-8176

Research

Introduction

Macroscopic objects tend to behave classically, but their behavior near zero temperature can be fundamentally different. When a trapped atomic gas is cooled down and the atoms' thermal velocities become small, the uncertainty principle dictates that their position becomes undefined -- the atoms smear out over the entire gas volume and, if bosonic, occupy a single quantum state. The resulting dilute-gas Bose-Einstein condensate, or BEC, forms a giant coherent matter wave with superfluid properties.

In our laboratory, we load ultracold atoms from BECs into optical lattices, i.e. potentials realized by standing waves of laser light, in order to create and explore ultracold quantum systems. With the help of such optical lattices, it thus becomes possible to address questions in various areas of physics, ranging from condensed-matter physics to quantum optics, through direct quantum simulation. For example, the behavior of atoms in an optical lattice closely mimics that of electrons in a solid, but at a length scale that is three orders of magnitude larger and with exquisite control over all relevant parameters in a naturally defect-free system. Moreover, atoms in optical lattices can act as localized spins (qubits) with controllable interactions, making lattice-based atomic quantum systems a versatile platform for studying the fundamental science driving the development of modern quantum technologies.

Over the past decade, we have pioneered the manipulation of atomic mixtures in state-dependent optical lattices to create ultracold quantum systems that combine particle-like/strongly-correlated and wave-like/superfluid behavior. Our research addresses topics in condensed-matter physics and, dissipative/open quantum systems. Recently we have made novel connections to quantum optics, in particular waveguide quantum electrodynamics, with atomic matter waves.

Research highlights

Formation of matter-wave polaritons. We realized interacting polaritonic quasiparticles in a lattice

and investigated their behavior across the superfluid-to-Mott transition. The observed behavior is consistent with a combination of the Bose-Hubbard and Weisskopf-Wigner models. A matter-wave polariton can be visualized as a hopping atom surrounded by a cloud of matter-wave excitations. Analogous features are known from the formation of exciton polaritons, where excitons are dyamically bound to guided photons. The coupling hybridizes the dispersion relation of the heavier constituent with that of the lighter one, giving rise to renormalized transport and interaction properties. In our system, we prepared the lower polariton branch dominated by the heavier constituent (the atom hopping in the lattice). We directly detected the presence of polaritons by probing their excitation spectrum and found very good agreement of the inferred hopping with our theoretical expectations. In contrast to their photonic counterparts, matter-wave polaritons are dissipation-free and tunable and thus can serve as an ideal testbed for polariton physics. [article][SharedIt] Press:[SBUNews][PHYS.ORG(feature)][N+1|ru (transl)]

In preceding theoretical work, we calculated the details of the polaritonic band

structure based on a ormalism that captures generalized situations in which both the radiation and the matter-wave excitations in the lattice are subject to lattice transport. The resulting band structure of the matter-wave polaritons is exotic and cannot be obtained by from a simple periodic potential. With the flexibility to tune the parameters, it becomes possible to realize a wider platform for studying low-dimensional frustrated systems.[article]

Quantum emitters in a tunable band structure. We studied the dynamics of matter-wave quantum emitters coupled t

o the band structure of an optical lattice. Contrary to free-space decay (which provides an analog to optical decay near a single band edge), the dynamics in a band, which is bounded by two edges, can change from exponential decay to fully coherent oscillations. We investigated the interplay between the emitters' excitation energy, the vacuum coupling, and the bandwidth of the lattice, and characterized the two contributing bound states of the matter-wave radiation field. The understanding how quantum emitters interact with the modified vacuum is at the heart of waveguide QED, one of the frontiers in quantum optics, and this concerns in particular the emergence and properties of atom-photon bound states in photonic crystals. Our experiment sheds light on the link between coherent emitter dynamics, fractional decay, and the ability of such bound states to interfere. [article]

In follow-up theoretical work, we generalized our earlier single-emitter model to emitter

arrays with a single excitation and radiative coupling to a full band structure. Our analytic model successfully reproduces the observed momentum distributions for emission into free space well as into the bands of an attractive and repulsive lattice potential, corresponding to co-located and shifted wells. Moreover, the model fully captures the observed decay in the crossover regime, reproducing propagation effects and reabsorption by neighboring emitters that lead to pronounced oscillations at long times [article]

Spontaneous emission of matter waves. We demonstrated the spontaneous emission of matter waves from a tunable open quantum system. The quantitative understanding of spontaneous emission harks back to the early days of quantum electrodynamivcs (QED), when in 1930 Weisskopf and Wigner, using Dirac's radiation theory, calculated the transition rate of an excited atom undergoing radiative decay. Their model, which describes the emission of a photon through coherent coupling of the atom s dipole moment to the continuum of vacuum modes, reflects the view that spontaneous emission into free space, driven by vacuum fluctuations, is inherently irreversible. We realized the Weisskopf-Wigner model in a novel context that allowed us to go beyond the model's usual assumptions. For this purpose, we created an array of microscopic atom traps in an optical lattice that, driven by tunable microwave radiation, individually emit single atoms, rather than single photons, into the surrounding vacuum. Our ultracold system, which provides a matter-wave analog of photon emission in a photonic-bandgap material, revealed behavior beyond standard exponential decay with its associated Lamb shift. It includes partial backflow of radiation into the emitter, and the formation of a long-predicted, exotic bound state in which the emitted particle hovers around the emitter in an evanscent wave. - Our system provides a flexible experimental platform for simulating open-system quantum electrodynamics and for studying dissipative many-body physics with ultracold atoms. [further details] * [article][SharedIt] * Press:[NatureN&V][pro-physik.de][SBUNews][Gizmodo][Investigacion y Sciencia|es]

In preceding theoretical work, we used the Weisskopf-Wigner formalism to revisit a proposal by the group of J.I.Cirac at MPQ and determine the details of the (non-)Markovian decay and emission dynamics considering an isolated matter-wave emitter in our waveguide geometry. In our work, we also calculated the spatial structure of the bound state for negative excitation energy, which is in excellent agreement with our measurements. [further details] * [article]

Diffracting matter waves without a potential. We explored a novel regime of diffraction arising in the dynamical

response of an atomic Bose-Einstein condensate coupled to a optical standing wave. Throughout the long history of diffraction, it has generally been taken for granted that the diffracting object can be described through a (position-dependent) potential with absorptive or dispersive character. This includes diffraction from material structures and gratings, ponderomotive potentials, as well as Kapitza-Dirac diffraction of atoms from a standing wave of light. The latter can be described via an optical potential, and is one of the standard tools in atom optics and metrology. We demonstrated that Kapitza-Dirac diffraction (i.e. the diffraction of matter from light) can in fact occur without a potential being present. In our experiments, we studied the dynamics of a condensate after switching on a resonant driving to the orbitals of a deep, state-selective optical lattice. We observed diffractive dynamics that remains coupled to coherent oscillations of the internal degree of freedom, violating the usual adiabatic (Born-Oppenheimer) approximation needed for the introduction of an optical potential. This represents a hitherto unexplored regime in which the dressed states undergoing diffraction are strongly mixed by zero-point motion. Our study sheds new light on the role of potentials in diffraction, and on the nature of dressed states arising in Rabi-type dynamics. We explored its utility for atom interferometry and future realization of dissipative quantum systems [further details]. * [article]

Abating the effects of disorder with weak interactions. In studies of the dynamics of a Bose-Einstein co

ndensate in a tilted incommensurate optical lattice, we demonstrated a nontrivial interplay between interactions and disorder. When a weak constant force is applied to atoms moving in a lattice potential, they undergo periodic motion known as Bloch oscillations (rather than accelerating as free particles). It is well established that for condensates these oscillations damp out due to either the presence of disorder or collisional interactions. However, we observed that when interactions and disorder are present simultaneously, they can compete with each other such that more interactions actually mean less, rather than more, damping. This screening effect vanishes once the interactions become stronger than the disorder (strong interactions cooperate with disorder, cf. eg. our earlier work). Our results are of relevance for the description of systems ranging from superfluid helium in porous materials to superconductivity in granular or disordered materials. [further details] * [article]

Crossing the quantum-to-classical boundary with a coupled-rotor system The understanding of how classical dynamics can emerge in closed quantum systems is a problem of fundamental importance. Remarkably, while classical behavior usually arises from coupling to thermal fluctuations or random spectral noise, it may also be an innate property of certain isolated, periodically driven quantum systems. By exposing a Bose-Einstein condensate to a periodically-flashed, two-color standing wave of light, we have for the first time realized a system of two coupled, kicked quantum rotors, and have discovered a quantum-to-classical transition. The simple quantum kicked rotor, which one obtains when using a single-color standing wave only, is a well-studied paradigm system for quantum chaos. Here, unless very specific conditions are met, the atomic velocities remain frozen in time due to quantum interference, despite continuous periodic kicking. However, already when just two such off-resonantly kicked rotors are coupled, new physics can appear. In our work, we found that the ability of each rotor to freeze the atoms' velocities can be destroyed, and that the coupling between the two rotors instead produces classical momentum diffusion. Our work sheds light on the fundamental question how the quantum and classical worlds are connected. [further details] * [article] * Press:[PHYS.ORG(feature)]

Probing an ultracold-atom crystal using atomic matter waves. We explored the scattering of atomic matter waves from ultracold atoms confined in an optical lattice. By shining a one-dimensional Bose gas (probe) onto a Mott insulator (target), we observed Bragg diffraction peaks that reveal the spatial ordering and localization of atoms on individual lattice sites. Our in-situ interrogation technique is non-destructive, and it precludes limitations on spatial resolution since the de Boglie wavelength of the incident atoms can be tuned freely. For weak confinement, we observed inelastic excitations of atoms in the target, which connect to a quantum "Newton's cradle" in the free-atom limit. Furthermore, we used atomic de Broglie waves to detect forced antiferromagnetic ordering in an atomic spin mixture, demonstrating the suitability of our method for the non-destructive detection of spin-ordered phases in strongly correlated atomic gases. Our work extends matter-wave diffraction, a technique that has already proven useful in various scientific disciplines, to the realm of ultracold quantum matter. Our method is similar to the diffraction of neutrons for the characterization of solid-state systems, but at energies that are a billion times lower. It also provides a nice example of wave-particle duality, where ultracold atoms serve both as localized particles and as coherent waves diffracting from them. [further details] * [article] * Press:[PHYS.ORG]

Impurities make an ultracold-atom crystal behave like an amorphous glass. We studied one-dimensional gases of bosonic atoms subject to localized atomic impurities and observed behavior consistent with an exotic quantum phase called a Bose glass. Two-species mixtures of quantum gases can form so-called "quantum emulsions" (see next paragraph). When one of the two species is pinned in place, its frozen emulsion pattern can effectively act as a static disorder potential for the more mobile species. It is predicted that such atomic impurities can cause a disorder-driven quantum phase transition from a perfectly-ordered Mott insulator crystal to an amorphous Bose glass. In our experiment, we created "frozen" atomic impurities by loading the atoms of one species onto the sites of a deep, species-specific optical lattice. We found that the other, more mobile species is driven to a localized, insulating state with a flat excitation profile, consistent with the expectation for a Bose glass. One key motivation for the use of atomic impurity disorder, as opposed to the standard optical disorder potentials, is the difference in spatial correlation properties. Indeed, we found that using uncorrelated atomic disorder much more easily drives the system into the amorphous phase. [further details] * [article]

Dressings, dips, and emulsions in ultracold mixtures. We reported novel many-body effects for a two-component mixture of ultracold bosonic atoms in an optical lattice. In our setup, one component feels a much deeper lattice than the other, which confines these atoms to the lattice sites, while the other component forms a frictionless superfluid background. An atom on a given site locally displaces the superfluid around it, thereby dressing itself with a density dip. Such a dressed atom ("polaron") is predicted to be less mobile than a bare atom, since it has to carry its dip along as it hops from site to site. Also, when two or more dressed atoms get close, the dips may cause mutual attraction and clustering, similar to what would happen to bowling balls placed on a mattress. On the other hand, the density of the superfluid surrounding the atoms can be made to dip so strongly that the superfluid effectively breaks up into many tiny droplets, leading to a "quantum emulsion" akin to a shaken vinaigrette. Consistent with these two scenarios, we observed that the apparent superfluid coherence of a given component near the superfluid-to-Mott transition is reduced by a second component, most significantly when the latter either experiences a strongly localizing lattice potential or none at all. [further details] * [article]

Adding a spin to nonlinear atom optics. We have observed collinear four-wave mixing of two-component matter-waves. Even though the first atomic four-wave mixing (FWM) experiment was performed a decade ago, no experiment to date has exploited the possibilities arising from the internal degrees of freedom. We report the first experimental demonstration of atomic FWM with two internal states and distinct, macroscopically occupied momentum modes. Our system is promising for the creation and observation of counter-propagating atomic wave packets with macroscopic spin entanglement and/or spin squeezing. The collinearity should be especially useful for possible applications in sub-shot-noise atom interferometry. Moreover, we find that FWM can also occur after the release of a homo-nuclear bosonic mixture trapped in a state-dependent optical lattice. We show that FWM can mimic in-situ interaction effects, and thus needs to be taken into account in such studies. [further details] * [article]

Poster gallery

Our laboratory

Experimental gallery click to animate