Sisyphus cooling in a continuously loaded trap

Our paper on Sisyphus cooling during the continuous loading of our hybrid magnetic/dipole trap has just appeared in New Journal of Physics (open access, check out -> here ).

Sketch of the continuous loading and simultaneous cooling mechanism.

Sketch of the continuous loading and simultaneous cooling mechanism.

In this paper we demonstrate continuous Sisyphus cooling combined with a continuous loading mechanism used to efficiently slow down and accumulate chromium atoms from a guided beam. While the loading itself is based on a single slowing step, applying a radio frequency field forces the atoms to repeat this step many times resulting in a so-called Sisyphus cooling.

This extension allows efficient loading and cooling of atoms from a wide range of initial beam conditions. We study the interplay of the continuous loading and simultaneous Sisyphus cooling in different density regimes. In the case of a low density flux we observe a differential gain in phase-space density of nine orders of magnitude. This makes the presented scheme an ideal tool for reaching collisional densities enabling evaporative cooling—in spite of unfavourable initial conditions.

V. V. Volchkov, J. Rührig, T. Pfau, A. Griesmaier: „Sisyphus cooling in a continuously loaded trap“, New J. Phys. 15, 093012 (2013)

Workshop on long range interactions in the ultracold

To be held in Stuttgart (Germany), September 3 – 5, 2013

Satellite workshop to the 2013 BEC meeting in Sant Feliu (Spain)

It is a tradition to expand the program of the BEC-Conference in Sant Feliu with satellite workshops on related topics. With this workshop we want to focus on the physics of long range interacting systems (Coulomb or dipole-dipole), strongly interacting systems (Rydberg atoms) and light mediated interactions (atoms in cavities). Under the term exotic interactions we gather all other types of unusual interaction mechanism (next neighbor coupling, 1/r4, 3-body, 4-body, …).

Invited speakers:

  • Antoine Browayes
  • Hans-Peter Büchler
  • Patrick Cheinet
  • Tilman Esslinger
  • Francesca Ferlaino
  • Johannes Hecker-Denschlag
  • Matt Hummon
  • Bruno Laburthe-Tolra
  • Hanns Christoph Nägerl
  • Silke Ospelkaus
  • Igor Ryabtsev
  • Kazik Rzazewski
  • Luis Santos
  • Ferdinand Schmidt-Kaler

Organizers: Robert Löw and Axel Griesmaier

More info and registration → here

Spectroscopy of a narrow-line optical pumping transition in dysprosium


Saturated absorption spectrum around 684nm of the most abundant isotopes of dysprosium: the normalized intensity is plotted against the frequency difference to the highest peak (164Dy). The spectrum is averaged over 100 single shots.

We have measured the hyperfine coefficients and isotope shifts of the Dy I 683.731 nm transition using saturated absorption spectroscopy on an atomic beam. Since this is a J →J’=J transition, it is in principle suited for optical pumping with circular light into a dark state in the extreme Zeeman states. This is an important tool eg. for purifying the polarization of a trapped ultra cold gas or demagnetization cooling. The other prerequisite is that the branching ratio between transitions back to the ground state and into other states is not too big such that several photons can be scattered in a quasi-closed manner. There was previously not much known about this transition. In particular about the linewidth, the branching and the hyperfine structure of the transition for the fermionic isotopes. We have now determined the specific mass shift 164−162, sms = −534 ± 17MHz, the excited state lifetime (1.68(5) μs corresponding to a linewidth of 95 ± 3 kHz) and an upper limit of the branching ratio between the two decay channels from the excited state showing that this transition is useable for optical pumping into a dark state and demagnetization cooling.

Read the full story here and the paper here

Workshop on Continuous Sources of Quantum Matter

Workshop from 11. – 13. March 2013, Freudenstadt (Black Forrest / Germany)

The recent achievements made by several groups that implement continuous and non-standard methods to slow down and cool atoms and molecules has triggered the idea of starting a dialog, especially focusing on these methods and their prospects. With this workshop we want to address groups working on all kinds of such (quasi-) continuous or in other respect novel approaches to the production of quantum degenerate and ultra-cold samples of atoms and molecules. Our goal is to bring together people from these different fields to get in touch, share their ideas, and find synergies and possibilities for collaborations. The general idea is that we have a number of longer introductory / tutorial talks from the PIs of the participating groups, some shorter contributions by students, a poster session and of course enough time for discussions. The topics of the workshop can cover everything that is interesting and related to the field, e.g.deceleration of molecules and atoms

  • ultracold guided beams
  • sisyphus cooling techniques
  • continuous trap loading mechanisms
  • all-optical approaches to BEC

Invited speakers:

  • Georg Raithel (University of Michigan)
  • Ed Narevicius (Weizmann Institute)
  • Thomas Vanderbruggen (ICFO / University of California)
  • David Guéry-Odelin (Université Paul Sabatier, Toulouse)
  • Sotir Chervenkov (MPQ Garching)
  • Mark Raizen (University of Texas)
  • Barbara Englert (MPQ Garching)
  • Börge Hemmerling (Havard University)
  • Simon Stellmer (IQOOI Innsbruck)
  • Nicolas Vanhaecke (Fritz Haber Institut, Berlin)

Organzers: Carsten Klempt and Axel Griesmaier

More info and registration →here

Evaporation-limited loading in our atom beam experiment

Schematic illustration of the setup underlying the continuous loading mechanism. Atoms are traveling to the right and are funneled into the radially confining ODT potential. Axially, the barrier coils create a state-dependent potential that is switched from a wall to a well by means of the optical pumping beam.

Recently, we have experimentally demonstrated a continuous loading mechanism for an optical dipole trap from a guided atomic beam . The observed evolution of the number of atoms and temperature in the trap are consequences of the unusual trap geometry. In our recent paper that we published this week in Physical Review A  present paper, we develop a model based on a set of rate equations to describe the loading dynamics of this mechanism. We consider the collision statistics in the nonuniform trap potential
that leads to two-dimensional evaporation. The comparison between the resulting computations and experimental data allows to identify the dominant loss process and suggests ways to enhance the achievable steady-state atom number. Concerning subsequent evaporative cooling, we find that the possibility of controlling axial and radial confinement independently allows faster evaporation ramps compared to single beam optical dipole traps.
The results of this paper are important because they sugest that our trap is not limited by inelastic processes like light assisted collisions or multiple scattering of the pump-light involved in the cw loading scheme but rather by the evaporation of atoms out of the trap volume. Therefore modifications of e.g. the properties of the guided atomic beam, the trap potential or e.g. the additional use of an rf field during the loading process could be used to further increase the steady state atom number in the trap and thus improve the starting conditions for our further preparation steps.

Read the full story here

Light Assisted Collisions and Superradiance in Bose-Einstein Condensates

Collective excitation through binary atom-atom collisions spoils matter wave coherence in superradiant Bose-Einstein condensates

SR Threshold behavior depending on detuning from atomic resonance.

SR Threshold behavior depending on detuning from atomic resonance. The (blue) detuning in GHz determines the threshold scattering rate required to initiate superradiance. This happens through excitations of atom pairs rather than single atoms at short distances where scattering of blue light is enhanced at the Condon point of the repulsive interaction potential of an excited-state/ground-state pair of atoms.

Rayleigh scattering of photons is the elastic process where light that is detuned from any resonance is scattered e.g. from atoms. I a Bose-Einstein condensate (BEC) this process can create long-lived modulations in the form of regular ripples in the density distribution of the atomic cloud. Bosonic stimulation then leads to a positive feedback mechanism which enhances the formation of a matter-wave grating that scatters photons coherently predominantly along the directions of high optical depth and in turn leads to even stronger density modulation in this direction and so on. This self enhancing directed Rayleigh scattering is well known as Rayleigh superradiance and has been studied in BECs already in the „early days“ of BECs about ten years ago [1]. The coupled dynamics of superradiant scattering and the simultaneous buildup of recoiling matter waves and light fields have at that time been successfully described using Maxwell-Schro¨dinger equations as well as rate equations derived from those.
Recently, however, those established descriptions of superradiant scattering have been challenged by new models that predicted a peculiar asymmetry in the dependence of the dynamics on detuning of the drive light [2] and also experimental evidence of such an asymmetry – pronounced at high atomic density and vanishing at low densities – was reported in [3].
This has sparked an ongoing controversial debate [8] about the validity of the presented theoretical expalanations and the „real“ physics behind those observations. It appears that the mechanism leading to the different dynamics for below resonance (red) and above resonance (blue) tuning of the incident light is not yet fully understood. To address this open question we have conducted a detailed experimental study of the threshold behavior of SR light scattering for a wide range of detunings up to 35 GHz. The observed detuning dependence rules out previous explanation attempts based on the action of dipole forces (suggested in [3] as the explanation for the observed behavior).This studies that we have conducted at the Niels Bohr Institute, Copenhagen have now been published in Physical Review Letters. We offer a physically motivated explanation for the asymmetry in the threshold behavior of SR scattering based on detuning dependent loss of matter-wave coherence. Resonant excitation of close pairs of atoms to excited state molecular potentials and subsequent spontaneous decay provides a source of frequency shifted photons, which for blue detuned drive light can be trapped inside the BEC for a long time.

Read the full story here.

[1] see e.g. S. Inouye, A. P. Chikkatur, D.M. Stamper-Kurn, J. Stenger, D. E. Pritchard, and W. Ketterle, Science 285, 571 (1999)
[2] L. Deng, M. G. Payne, and E.W. Hagley, Phys. Rev. Lett. 104, 050402 (2010)
[3] L. Deng, E.W. Hagley, Q. Cao, X. Wang, X. Luo, R. Wang, M. G. Payne, F. Yang, X. Zhou, X. Chen, and M. Zhan, Phys. Rev. Lett. 105, 220404 (2010)
[4] W. Ketterle, Phys. Rev. Lett. 106, 118901 (2011)
[5] L. Deng, M. G. Payne, and E.W. Hagley, ibid. 106, 118902 (2011)

BEC in our atom guide

Redemption for Sisyphus:
A Single Dissipative Step To Load Atoms Into A Conservative Trap

We have recently demonstrated the fast accumulation of <sup>52</sup>Cr atoms in a conservative potential from a guided atomic beam. Without laser cooling on a cycling transition, a single dissipative step involving optical pumping allows us to load atoms at a rate of 2×10<sup>7</sup> atoms/s in the trap. Within less than 100 ms we reach the collisionally dense regime, from which we produce a Bose-Einstein condensate with subsequent evaporative cooling. This constitutes a new approach to degeneracy where Bose-Einstein condensation can be reached without a closed cycling transition, provided that a slow beam of particles can be produced [1].

A diode for atoms

The underlying idea of our scheme is to use a continuous beam of guided ultra-cold atoms and accumulate them in a conservative trap after taking away the directed kinetic energy in one step. This is achieved by letting them climb up a potential barrier in the form of a local magnetic field, converting the „directed“ kinetic energy along the guide axis—given by the velocity of the center of mass motion of the beam— into potential energy. Subsequently, when the atoms have reached the maximum of the barrier, the internal state of the atoms is changed from the initial low-field-seeking state to a high-field-seeking state, trapping them at the bottom of a potential „valley“ and leaving them with virtually no directed kinetic energy. The remaining kinetic energy is much lower now and—to a great extend—determined by the initial temperature of the atoms in the co-moving frame of the guided beam.

Some people have suggested to call this process a „single Sisyphus cooling step“ because it is reminiscent of the process that has to be undergone by atoms in the so-called Sisyphus laser-cooling scheme over and over again. But in fact the process couldn‘t be less Sisyphus. Unlike this poor Figure of Greek mythology, the atoms are redeemed already after climbing up the hill once.

How it works

As the source of ultra-cold guided atoms, we use our continuous chromium source which we describe here and here. The Figure on the left shows an illustration of the dissipative loading scheme. A superposition of an optical dipole trap (ODT) and a magnetic field confines the atoms in radial direction(xy-plane). This is necessary because the magnetic field we superimpose to the field of the magnetic guide produces a saddle potential: Along the z-direction (symmetry axis of the guide), the hybrid potential can be either repulsive or attractive depending on the atom’s magnetic sublevel, and in radial direction the confining magnetic field of the guide is weakened. Arriving atoms are in the low-field seeking mJ = 3 state. The magnetic field thus acts as a barrier. At the position of the barrier’s maximum, the atoms are pumped to the high-field seeking absolute ground state mJ = -3 while the directed kinetic energy is dissipated

Once the atoms are pumped to the high-field-seeking state, they feel a potential valley in which they are trapped. The ODT plus magnetic barrier field provide a strong confinement for the high-field-seeking state in all directions. For the scheme to work, the spatial extension of the magnetic barrier has to be of the same size as the Rayleigh range of the ODT because the optical trap has to provide sufficient radial confinement in the regions where the barrier field has a significant strength. Thus we use tiny coils with a diameter of only 1mm to produce the magnetic field. The optical pumping scheme which is used is illustrated on the left. The atoms have to scatter on average 4.5 photons to be pumped to the mJ=-3 state. We observe the fast accumulation of atoms in the ODT. And already after 20ms of loading, the temperatures of the trapped atoms in radial and longitudinal direction start to equilibrate by elastic collisions.

A novel way to reach quantum degeneracy

On the left you can see fluorescence images of the atoms in the trap region at different stages of the experiment from top to bottom: I) the guided atom beam without the superimposed dipole trap and barrier II) the beam when we switch on the optical trapping potential III) with dipole trap and magnetic barrier IV) when we start to pump the atoms optically V) the fully loaded hybrid optical/magnetical trap.

Starting from this steady state situation of atom number and temperature in the trap which is reached after approximately 100ms with our current flux of atoms in the guide, we are able to produce chromium Bose-Einstein condensates by further evaporative cooling within less than five seconds. Not only is this a significant improvement of the standard production scheme of a chromium BEC [2] but also a new approach where laser cooling is not a prerequisite. We therefore think that the scheme in combination with other methods to produce ultra-cold guided beams could be used for the production of dense ultra-cold vapors with species that can not be cooled by laser-cooling.


[1] Read the full story in our recent PRL 106, 163002 (2011) or on arXiv

[2] J. Phys. B 40, R91–R134 (2007)

[3] Homepage of the project

Ein Gas aus winzigen Magneten

Neue Experimente mit stark dipolaren Quantengasen

Gegenstand unserer Forschungsarbeit ist die experimentelle Untersuchung der Eigenschaften eines Bose-Einstein-Kondensats (BEK) aus Chrom-Atomen. In unserem Labor an der Universität Stuttgart gelang uns im November 2004 weltweit erstmals die Erzeugung eines solches Kondensats mit Chrom. 1995 war dies den späteren Nobelpreisträgern des Jahres 2001 Wolfgang Ketterle (MIT), Carl Wieman und Eric Cornell (beide JILA)  mit Alkali-Atomen gelungen. In diesem besonderen Materiezustand nehmen viele tausend Atome denselben Quantenzustand ein, was gegenüber klassischen Gasen zu fundamental anderen Eigenschaften führt.

Abbildung 1 Übergang vom klassischen Gas zum BEK. Die Falschfarben bilder aus Fotografien, bei denen die Dichte des Gases durch die Höhe und die Farbe dargestellt sind, zeigen die Gaswolke um die kritische Temperatur von etwa einem halben millionstel Kelvin.

Der Phasenübergang in das Bose-Einstein-Kondensat geschieht dabei bei extrem niedrigen Temperaturen, weniger als ein millionstel Grad über dem absoluten Nullpunkt (0Kelvin entsprechend ca. -273,15°C). Die Realisierung solch tiefer Temperaturen in einem  atomaren Gases, das zunächst Temperaturen von weit über 1000°C besitzt, ist nur im Ultrahochvakuum durch den Einsatz verschiedener Kühlverfahren, unter anderem der Laserkühlung, möglich. Im Unterschied zu den davor erzeugten BEK, vorwiegend mit Alkaliatomen, zeichnet sich das Chrom Kondensat durch eine vielfach stärkere magnetische (Dipol-Dipol) Wechselwirkung zwischen den Atomen aus, die die experimentelle Untersuchung neuartiger Quanteneffekte ermöglicht. So gelang uns mit dem neuen BEK im  Februar 2005 die erste Beobachtung einer mechanischen Auswirkung der magnetischen Wechselwirkung zwischen Atomen in einem Gas. Die weitere Erforschung dieser Art der Wechselwirkung in Quantensystemen ist von großem Interesse, weil sie z.B. in Festkörpern beim Auftreten von Supraleitung, dem widerstandslosen, also verlustfreien Leiten von elektrischem Strom, eine wichtige Rolle spielen, und ein Gas mit derartigen Eigenschaften zum modellhaften Studium dieser Phänomene dienen könnte. Zum anderen basieren auch Vorschläge zur Realisierung von Quantencomputern auf der Verknüpfung mehrerer Quantenobjekte, wie z.B. einzelner Atome, durch lang-reichweitige Wechselwirkungen, wie sie durch die magnetische Dipol-Dipol Wechselwirkung repräsentiert werden.

Inzwischen konnten wir weitere Experimente durchführen, die die außergewöhnlichen Eigenschaften eines Bose-Einstein Kondensats mit Dipol-Dipol Wechselwirkungen zeigen. Durch externe Magnetfelder sind wir z.B. in der Lage, andere Wechselwirkungen völlig zu unterdrücken, und dadurch ein rein dipolares Quantengas zu erzeugen.

Abbildung 2 Falschfarbenbilder eines implodierenden Chrom BEK, aufgenommen zu verschiedenen Zeiten nach der Initiierung der Implosion. Die Struktur der Gaswolke kommt durch die magnetische Wechselwirkung zwischen den Chrom Atomen zustande.

Für ein solches Gas waren bereits zehn Jahre zuvor Stabilitäts-Kriterien in unterschiedlichen Fallengeometrien hergeleitet worden, die wir nun in sehr guter Übereinstimmung mit den theoretischen Berechnungen komplett vermessen konnten. Hierbei ist die Geometrie der Falle dafür verantwortlich, ob das Gas stabil ist oder aufgrund des anziehenden Teils der Dipol-Dipol Wechselwirkung implodiert. Dadurch lässt sich eine solche Implosion auch gezielt initiieren und wir konnten deren Dynamik genau studieren. Aufgrund ihrer Ähnlichkeiten mit einer Supernova werden solche Implosionen auch als „Bose Nova“ bezeichnet. Der anisotrope Charakter der magnetischen Wechselwirkung ist bei unseren Experimenten direkt an der räumlichen Struktur der Gaswolke zu erkennen.

Seit unseren ersten experimentellen Arbeiten steigt die Zahl theoretischer Veröffentlichungen, die sich mit dieser Art von Wechselwirkungen in Quantengasen beschäftigen, um etwa 50 pro Jahr an und der Katalog vorgeschlagener Experimente wächst ständig. Das zeigt, dass die Erforschung der dipolaren Gase erst am Anfang steht und das Gebiet auch in Zukunft weitere äußerst interessante Entdeckungen verspricht.

Als weitere Besonderheit besitzt Chrom im Gegensatz zu den meisten Bose-kondensierbaren Materialien darüber hinaus auch technische Relevanz. Da Chrom als Standardmaterial in lithographischen Strukturierungsverfahren eingesetzt wird, besteht hier weiteres Potenzial. In unserer aktuellen Arbeit beschäftigen wir uns daher auch mit der Realisierung eines so genannten Atomlasers auf Basis des Chromkondensats, um damit lithographisch geordnete Nanostrukturen auf einer Oberfläche zu erzeugen.

Pancake prevents collapse

Press Release: Physicists in Stuttgart measure the border between stable and unstable states of a quantum gas of magnets

Instabilities in clouds of attracting matter are a well known phenomenon. In astrophysics, they lead to spectacular effects such as supernovae. But also in a gas, when it is made up of tiny atomic magnets, the magnetic forces lead to instabilities. Such a gas cloud implodes due to the attractive interaction between the magnetic atoms. The research group of Professor Tilman Pfau of the 5th Institute of Physics at the University of Stuttgart have recently measured the complete stability diagram of such a gas of magnets which maps exactly the border between stable and unstable states of the gas. These results have now been published in the latest issue of Nature Physics*.

The finding that attracting matter is instable is known to everybody who has ever played with a bunch of magnets: they simply clump together. Even when they are aligned in parallel the same happens – unless the magnets are, at the same time, forced to stay in a plain. Such an arrangement in a pancake-like shape – in contrast to a spherical or cigar-like geometry – is stable. In the case of these spheres or cigars, only an additional repulsive interaction between the atoms that keeps them apart could prevent the whole system from collapsing (clumping together).

Tilman Pfau’s group are investigating the properties of so called “quantum gases” in the framework of their trans-regional collaborative research centre (SFB/TRR21 “Control of quantum correlations in tailored matter –Co.Co.Mat.” ). The gas in which the above experiments were performed consists of chromium atoms that are forced to undergo a phase transition to a so called Bose-Einstein condensate at extremely low temperatures. In this special state of quantum matter, the interactions as well as the shape of the trap that holds the atomic cloud and squeezes it into the desired shape can be controlled very precisely. Already ten years ago Tilman Pfau, together with a group of polish scientists, predicted theoretically the border between stable and unstable conditions. Only now could his group proof experimentally that the gas behaves as had been predicted and that it is indeed stabilized by forcing it into a plain pancake shape.

In their ongoing research project, the group are having a closer look at the dynamics of the collapse itself. Due to its similarity to a supernova, this collapse is sometimes called “Bose Nova”. During the collapse the researchers expect for certain parameters the occurrence of new states of quantum matter caused by the magnetic interaction. An exactly controlled and triggered collapse could – on the other hand – also be used to deposit precisely focussed chromium atoms on a surface.

*) Tobias Koch, Thierry Lahaye, Jonas Metz, Bernd Fröhlich, Axel Griesmaier, Tilman Pfau: „Stabilizing a purely dipolar quantum gas against collapse“, Nature Physics (2008), DOI number 10.1038/nphys887 [arXiv:cond-mat 0710.3643]