Partenaires

Menu

• Présentation
• Atomes, cavités et photons
  • Principes de l’electrodynamique quantique en cavité
  • Techniques expérimentales
    • Atomes
    • Cavité
    • Dispositif expérimental.
    • Deux opérations fondamentales
  • Information quantique avec des atomes et des cavités.
  • Mesure de l’état quantique du champ
    • Mesure QND du nombre de photons et détermination de l’état du champ
    • Distributions de probabilité
  • Explorations de la limite classique/quantique.
• Puces à atome cryogéniques
  • Piège électrodynamique pour état de Rydberg
    • Piégeage des états de Rydberg
    • Comment charger un et un seul atome
    • Comment garder de long temps de cohérence durant le piégeage
  • Puce à atomes supraconductrice
    • Dispositif expérimental
    • Piégeage magnétique d’atomes ultrafroids
• Microsphères et microtores de Silice
  • Structure des modes de galerie
  • Motivation générale
• Chaire de Physique quantique, Collège de France
• À la une
• Publications
• Téléchargements
• Recrutements
• LKB

• Annuaire

Rechercher

We measure directly the lifetime of a classical coherent field stored in the cavity. We use atomic probes to detect the field amplitude decay [1].

In order to minimize surface defects, the superconducting mirrors do not have coupling irises at their apex. It is still possible to couple microwave into the cavity mode trough the residual diffraction loss channels. A fraction of the stored photons finally escape through the gap between the mirrors, diffused by surface defects. In the same way, a small fraction of the photons from an intense coherent source sent though this gap are coupled inside the mode. We can in this way feed a rather large coherent state into the mode.

The field later diffused out of the cavity is however to weak to be directly detected. We use instead an atomic probe to monitor the field decay in the cavity. We first inject a large field in the cavity. We then send after a delay a sample of atoms prepared in the level. These atoms absorb the cavity resonant microwave. A large d.c. electric field applied across the cavity mirrors makes the absoption broadband via the Stark effect. The atomic absorption is thus efficient whatever the cavity frequency. The transfer rate from the initial level to a nearby Stark multiplicity is thus a good measure of the intra-cavity field intensity.

We first map the frequency position of the cavity modes by monitoring the intra-cavity power while tuning the injection source. We find as expected two modes with a 1.2 MHz splitting, corresponding to the two orthogonal linear polarizations. We also check the spatial structure of the modes. We send velocity-selected atomic samples, whose position at any time is well-known and we tune them in resonance with the cavity field by the Stark effect at a precise position. We obtain a gaussian shape with the expected 6 mm waist.

We then determine the cavity field lifetime (see APL 90, 164101). For a given injected power, , we measure the atomic transfer rate as a function of the time delay (see figure above, red dots). This transfer rate is a monotonic function of the field power. It thus decreases slowly with increasing . The solid line in the figure is a fit on a simple model including saturation of the transition.

We thus resume the experiment with an injected power and ( is the base of natural logarithms). The corresponding transfer decay curves (green and black dots respectively) are simply translated by and from the first ones, since the transfer depends only upon the field power. The distance between these curves thus provides a direct determination of the cavity damping time.

We have measured the damping time for a given mode as a function of temperature (see figure above - note that the vertical scale is logarithmic and the horizontal one reciprocal). Above 1.5 K, decreases exponentially with temperature. This corresponds to the BCS limit for a superconductor. The superconducting gap inferred from this decay (20 K) is in good qualitative agreement with other measurements.

Below 1 K, the cavity lifetime reaches a saturation at an extremely high value (110 ms for this mode). We have tested two mirror sets and four modes. All damping times are above 70 ms. The longest one (corresponding to the mode used in the recent experiments) is 0.13 s.

This damping time corresponds to a factor of and to a finesse of . This is the highest finesse reported so far for a Fabry Perot cavity, in any frequency range. During its lifetime, a photon travels 40000 km between the cavity mirrors, bouncing more than a billion times on them. It is quite remarkable that the damping time is comparable to that of the best closed cavities in the same frequency range.