The master equation for optical bistability in the bad cavity limit is solved directly when the number of atoms is small. A necessary condition for bistability is given. In contrast to the semi-classical analysis, there is no sense in which there is bistability for 4 atoms; rather the order of 12 atoms is needed.
In a theoretical discussion of optical bistability, it is often necessary to resort to some approximation scheme to discuss time-dependent properties. The standard method consists in assuming a very good cavity (κ«γ⊥, γ11) Or a very bad cavity (γ11 <<κ1, γ⊥ ) and to perform an adiabatic elimihation of the fast varying variables. In the good cavity limit /1/ the atomic variables are eliminated and one is left with an equation for the field. In the bad cavity limit /2/ the field and the atomic polarization are eliminated and one is left with an eouation for the atomic inversion.
In the general framework of the system size expansion of Van Kampen and Kubo, we consider the Fokker-Planck equation for a model of absorptive bistability in the bad-cavity limit. The physical system is described by the reduced atomic density operators after adiabatic elimination of the cavity field variables. Mapping of the master equation into c-number form according to the normal-ordering mapping scheme yields known results for the atomic fluctuations and correlation functions; however, it also leads to a Fokker-Planck equation with a non-positive-definite diffusion matrix. The symmetrical-order-mapping scheme eliminates this difficulty. The leading contribution to the system size expansion yields a Fokker-Planck equation for the symmetrical-ordered density function having a positive-definite diffusion matrix. The atomic expectation values and fluctuations previously derived from the quantum Langevin equations emerge naturally from this Fokker-Planck equation.
We study the spectrum of squeezing in the light transmitted by a system of two-level atoms in an optical cavity driven by a coherent field, as recently observed by Kimble and collaborators. Our analysis does not introduce any restriction on the detuning parameters or on the relative order of magnitude of the atomic and cavity relaxation rates. Using the dressed-mode formalism, we derive explicit analytic expressions for the spectrum of squeezing and for the second-order intensity correlation function. We describe how the spectrum depends on the various parameters of the system; this study provides general indications on how to optimize the conditions to observe squeezing in optical bistability. A special attention is devoted to the changes of the spectrum under variation of the detuning parameters or on the ratio of cavity to atomic damping rate. In all cases, the changes are continuous and smooth, with quantitative variations in the level of squeezing. This indicates the existence of a single avenue to squeezing in this system. The bad-cavity direction is identified as especially convenient to obtain a significant amount of squeezing.