Nonlinear Semiconductor Microcavities

Abstract

A lossless Fabry-Pérot interferometer in its simplest form consists of two parallel mirrors, and it converts changes in the roundtrip phase into a modulation of the transmitted intensity. The general formula for a Fabry-Pérot depends also on the intensity reflectivity coefficient of each of the mirrors as well as the absorption within the spacer (Sect. 3.1.6 of [1]). If one knows the dependencies of the refractive index n and absorption coefficient α upon the intracavity intensity (output intensity over output mirror transmission), one can solve for the input intensity versus the output intensity. When that function is camelbacked, then the output is multi-valued over a certain range of inputs. If the upper and lower branches are stable the system is said to exhibit optical bistability (Chap. 23 of [1]). The first optical bistability was seen in lasers, where a hysteresis in lasing occurred as a function of cavity frequency (see Sect. 1.3 in [2]). If the spacer medium is not inverted then the self-action effect is referred to as passive optical bistability, the first demonstration in sodium vapor being motivated largely by Sam McCall’s quest for an optical transistor [2, 3]. Perhaps the first nonlinear semiconductor etalons were ordinary narrowband interference filters [4]; even commercial ones often exhibit low-power optical bistability based on the thermal increase of both n and the physical length L of the spacer. Electronic nonlinearities such as the carrier density dependent changes in band edge susceptibility of GaAs or InSb are much faster (≈ 1 ns), but the holding power is much higher. During this same period a systematic investigation of the nonlinear optical mechanisms in II–VI compounds, especially CdS, was also conducted (Chap. 20 of [1]).