Observations of a doubly driven V system probed to a fourth level in an N configuration are reported. A dressed-state analysis is also presented. The expected three-peak spectrum is explored in a cold rubidium sample in a magneto-optic trap. Good agreement is found between the dressed-state theory and the experimental spectra once light shifts and uncoupled absorptions in the rubidium system are taken into account.

Observation of transient effects in electromagnetically induced transparency ͑EIT͒ is reported in Rb 85 cooled in a magneto-optical trap. The transmission of a weak probe beam in resonance with the 5S 1/2 (Fϭ3) to 5P 3/2 (Fϭ3) hyperfine transition increased transiently when a relatively strong coupling field in resonance with the 5S 1/2 (Fϭ2) to 5 P 3/2 (Fϭ3) hyperfine transition was switched on rapidly using a Pockels cell. The probe transient showed an initial Rabi half-cycle overshoot before settling down to steady-state EIT. The results agreed well with computations using a three-state model of the ⌳ system. The computations also suggest that transient gain should be observed with coupling field power only four times larger than that presently available to us. ͓S1050-2947͑98͒03308-3͔PACS number͑s͒: 42.50. Gy, 32.80.Pj, 42.50.Md The steady-state absorption of a weak resonant probe beam can be reduced or even eliminated by the presence of a strong coherent coupling beam exciting a linked transition. This effect, known as electromagnetically induced transparency ͑EIT͒, was first observed in 1991 ͓1,2͔. Related work on coherent population trapping in three-level systems dates from the mid-1970s and has been recently reviewed ͓3͔. In the last few years steady-state EIT has been widely studied in vapor cells and more recently in laser-cooled samples, see reviews ͓4,5͔. Transient EIT effects, however, have received relatively little explicit attention. In the first EIT experiments ͓1,2͔ the coupling field was provided by a high power pulsed laser with pulse rise and fall times that were slow compared with the Rabi period of the field. The transient approach to transparency in this adiabatic regime has been analyzed by Harris and Luo ͓6͔ from the point of view of energy requirements for the establishment of the transparency. We are interested in the nonadiabatic regime where the coupling field is switched on in a time that is short compared with Rabi periods and relaxation times. The only reported experimental study in this regime is that of Fry et al. ͓7͔, who observed transient absorption of a probe pulse in a ⌳ system immediately after the coupling beam was rapidly switched off. This is the converse of our experiment where the coupling field is rapidly switched on. Three-state models of transient effects in probe transmission when the coupling field is switched on nonadiabatically have been reported ͓8,9͔. In these models the transmission of a weak resonant probe approaches the steady-state EIT via damped oscillations characterized by the Rabi frequency of the relatively strong coupling field ⍀ C and the excited state decay rate ⌫, and for a sufficiently strong coupling beam there are intervals of probe gain without inversion, even without incoherent pumping. Thus there are possible applications to fast optical switching and lasing without inversion.We report an observation of transient EIT effects in the nonadiabatic regime, using a 85 Rb sample cooled in a magneto-optical trap ͑MOT͒. A three-level ⌳ system wa...

This paper presents a wide-ranging theoretical and experimental study of nonadiabatic transient phenomena in a ⌳ electromagnetically induced transparency system when a strong coupling field is rapidly switched on or off. The theoretical treatment uses a Laplace transform approach to solve the time-dependent density matrix equation. The experiments are carried out in a 87 Rb magneto-optical trap. The results show transient probe gain in parameter regions not previously studied, and provide insight into the transition dynamics between bare and dressed states.

We show that for appropriate choices of parameters it is possible to achieve photon blockade in idealised one, two and three atom systems. We also include realistic parameter ranges for rubidium as the atomic species. Our results circumvent the doubts cast by recent discussion in the literature (Grangier et al Phys. Rev Lett. 81, 2833(1998), Imamoglu et al Phys. Rev. Lett. 81, 2836(1998) on the possibility of photon blockade in multi-atom systems.

No abstract