EXPERIMENTAL DETAILSOur experimental setup is depicted by the simple drawing in Fig. 1A of the manuscript, with many of the technical aspects described in more detail in Refs.[S1, S2]. After releasing a cloud of atoms from a magnetooptical trap (MOT) above the cavity, transverse cooling beams illuminate the cavity region, at which point an atom can be loaded into the intracavity far-off resonance trap (FORT), which is matched to a standingwave, TEM 00 mode along the cavity axis. The trap depth is U 0 /k B = 2.3 mK (47 MHz), and because its wavelength is λ F = 935.6 nm, the potential for the atomic center-of-mass motion is only weakly dependent on the atom's internal state [S2]. The cavity length is actively stabilized with an auxiliary laser at wavelength λ C = 835.8 nm that does not interfere with the trapping or the cQED interactions. Relevant cavity parameters are length l 0 = 42.2 µm, waist w 0 = 23.6 µm, and finesse F = 4.2 × 10 5 at 852 nm. For our system, the Rabi frequency 2g 0 for a single quantum of excitation is given by g 0 /2π = 16 MHz, where g 0 is based upon the reduced dipole moment for the 6S 1/2 , F = 4 ↔ 6P 3/2 , F = 3 transition in atomic Cs (Fig 1B). The amplitude decay rates (κ, γ) due to cavity losses and atomic spontaneous emission are κ/2π = 4.2 MHz, and γ/2π = 2.6 MHz. Since g 0 (κ, γ), strong coupling is achieved, resulting in critical photon and atom numbers n 0 ≡ γ 2 /(2gWith an atom loaded into the intracavity FORT, our protocol for the generation of single-photon pulses consists in illuminating the atom with a sequence of laser pulses according to the timing diagram shown in Fig. 1(c) of the manuscript. Within each trial, the first pulse Ω 3 (t) contains light tuned 10 MHz blue of F = 3 → F = 3 , which initiates the adiabatic transfer F = 3 → 4 between the ground hyperfine levels, with the emission of a photon into the cavity mode. This transformation is principally accomplished via "dark" eigenstates of the atom-cavity system, with no contribution from the excited level F = 3 , and hence with a concomitant reduction of fluorescent loss [S3, S4, S5]. The second pulse Ω 4 (t) is tuned 17 MHz blue of F = 4 → F = 4 and recycles the atom back to the F = 3 ground state through spontaneous decay F = 4 → F = 3. Each Ω 3,4 field consists of two orthogonal pairs of counter-propagating beams in a σ + −σ − configuration. The detuning between the 3 → 4 transition at ω 43 and the cavity resonance ω C is ∆ CA ≡ ω C − ω 43 = 2π × 9 MHz [S6].We now provide some additional details on the optical path from the cavity to the detectors. After emerging from the vacuum chamber window, the path includes a polarizing beam splitter (PBS), several dichroic mirrors and two interference filters. The light is next coupled into a single-mode fiber, and then split using a 50/50 fiber coupler. The two output fibers of the coupler are connected to fiber-coupled avalanche photodiodes (APD), labelled D A and D B . LOSSES AND EFFICIENCIESPhotons generated in the cavity are subject to various types of loss along thei...

No abstract

Single cesium atoms are cooled and trapped inside a small optical cavity by way of a novel far-off-resonance dipole-force trap, with observed lifetimes of 2-3 s. Trapped atoms are observed continuously via transmission of a strongly coupled probe beam, with individual events lasting approximately 1 s. The loss of successive atoms from the trap N>/=3-->2-->1-->0 is thereby monitored in real time. Trapping, cooling, and interactions with strong coupling are enabled by the trap potential, for which the center-of-mass motion is only weakly dependent on the atom's internal state.

The whispering gallery modes ͑WGMs͒ of quartz microspheres are investigated for the purpose of strong coupling between single photons and atoms in cavity quantum electrodynamics ͑cavity QED͒. Within our current understanding of the loss mechanisms of the WGMs, the saturation photon number n 0 and critical atom number N 0 cannot be minimized simultaneously, so that an ''optimal'' sphere size is taken to be the radius for which the geometric mean ͱn 0 N 0 , is minimized. While a general treatment is given for the dimensionless parameters used to characterize the atom-cavity system, detailed consideration is given to the D 2 transition in atomic cesium at 0 ϭ852 nm using fused-silica microspheres, for which the maximum coupling coefficient g a /(2)Ϸ750 MHz occurs for a sphere radius aϭ3.63 m corresponding to the minimum for n 0 Ϸ6.06 ϫ10 Ϫ6 . By contrast, the minimum for N 0 Ϸ9.00ϫ10 Ϫ6 occurs for a sphere radius of aϭ8.12 m, while the optimal sphere size for which ͱn 0 N 0 is minimized occurs at aϭ7.83 m. On an experimental front, we have fabricated fused-silica microspheres with radii aϳ10 m and consistently observed quality factors Qу0.8 ϫ10 7 . These results for the WGMs are compared with corresponding parameters achieved in Fabry-Perot cavities to demonstrate the significant potential of microspheres as a tool for cavity QED with strong coupling.

The spatial resolution of a conventional imaging laser radar system is constrained by the diffraction limit of the telescope's aperture. We investigate a technique known as synthetic-aperture imaging laser radar (SAIL), which employs aperture synthesis with coherent laser radar to overcome the diffraction limit and achieve fine-resolution, long-range, two-dimensional imaging with modest aperture diameters. We detail our laboratory-scale SAIL testbed, digital signal-processing techniques, and image results. In particular, we report what we believe to be the first optical synthetic-aperture image of a fixed, diffusely scattering target with a moving aperture. A number of fine-resolution, well-focused SAIL images are shown, including both retroreflecting and diffuse scattering targets, with a comparison of resolution between real-aperture imaging and synthetic-aperture imaging. A general digital signal-processing solution to the laser waveform instability problem is described and demonstrated, involving both new algorithms and hardware elements. These algorithms are primarily data driven, without a priori knowledge of waveform and sensor position, representing a crucial step in developing a robust imaging system.