852-nm triggered single-photon source based on trapping and manipulation of a single cesium atom confined in a microscopic optical dipole trap

Abstract

Single-atom-based single-photon source has several advantages, such as narrow bandwidth, wavelength matching with the absorption line of the same atomic ensemble, and insensitivity to the environment disturbing, and it is very important not only for basic researches in quantum optic field but also for applications in quantum information processing. In this paper, we report the generation of a 10-MHz-repetition-rate triggered single-photon source at 852 nm based on a trapped single cesium atom in a far-off-resonance microscopic optical dipole trap (FORT). To generate an optical dipole trap, a far-red-detuned 1064 nm laser beam is tightly focused by using a high numerical aperture lens, a typical trap depth is 2 mK and trap waist is 2.3 m. To obtain a maximum probability of pulsed excitation, the frequency of the pulsed laser should be resonant with the atomic energy levels and the trapped single atom must be excited with a -pulse. However, the interaction between the FORT laser and the atoms causes AC Stark shifts of the atomic energy levels. Thus, in order to demonstrate the resonant pulsed excitation, it is important to calculate and measure the shift of 6S1/2|Fg=4,mF=+4-6P3/2|Fe=5,mF=+5 cyclical transition in the FORT. For a two-level system, the probability of pulsed excitation can be described by Rabi oscillations with a characteristic Rabi frequency . With an optimized time sequence, we experimentally demonstrate the Rabi oscillation between the ground state and the excited state, and the peak power of -pulse laser is about 1.25 mW. We also measure the temporal envelope of single photons after a -pulse excitation. A gated pulsed excitation and cooling technique are used to reduce the possibility that atoms are heated by -pulse laser. The typical trapping lifetime of single cesium atom is extended from~108 ups to~2536 ms. The corresponding number of excitations is improved from 108 to 360000. The second-order intensity correlations of the emitted single-photon are characterized by implementing Hanbury Brown-Twiss setup. The statistics shows a strong anti-bunching with a value of 0.09 for the second-order correlation at zero delay. In the future, we will perform a Hong-Ou-Mandel two-photon interference experiment to analyze the indistinguishability of the single photons. We will also trap single atoms in a magic-wavelength optical dipole trap where the ground and the excited states have the same shift.