Vertical-cavity semiconductor optical amplifiers (VCSOAs) are attractive as a low-cost alternative to existing amplifier technologies for use in fiber-optic communication systems such as metro and access networks. In contrast with in-plane SOAs, the surface-normal operation of vertical-cavity SOAs gives rise to a number of advantages including a high coupling efficiency to optical fiber, polarization insensitive gain, the potential to fabricate high fill-factor two-dimensional arrays, and the ability to test devices on wafer.; Due to their narrow gain bandwidth, VCSOAs function as amplifying filters. In these devices the inherent spectral filtering of the high-finesse Fabry-Perot cavity leads to the elimination of out-of-band noise and results in channel-selective amplification. For multi-wavelength communications systems, it is of great interest to develop widely tunable VCSOAs that can be dynamically adjusted to match the signal wavelength. A promising approach to achieve wide wavelength tuning in VCSOAs is micromechanical, or MEMS-based tuning. Here, mechanical alteration of the effective cavity length gives rise to tuning ranges greater than those that can be achieved via refractive index modulation.; This dissertation outlines the development of three generations of MEMS-tunable VCSOAs (MT-VCSOAs), with the initial generation of devices being noteworthy as the first demonstration of a micromechanically-tunable VCSOA. In contrast with temperature tuning, the AlGaAs-based electrostatic actuator used in these devices allows for rapid, low power, and wide wavelength tuning. In the final generation, the MT-VCSOA utilizes a bottom-emitting configuration in which the MEMS-tuning element serves as the high-reflectivity back mirror. By suppressing the variation in reflectance with tuning, this configuration exhibits a two-fold increase in the effective tuning range as compared with the initial generation of devices---with a minimum of 5 dB fiber-to-fiber gain (12 dB on-chip gain) over a wavelength span of 21 nm, from 1557.4 nm to 1536.4 nm, while requiring a maximum tuning bias of 10.5 V (a five-fold reduction when compared with the first generation of MT-VCSOAs). Furthermore, these devices exhibit properties comparable to state-of-the-art fixed-wavelength VCSOAs, with a maximum fiber-coupled saturation output power of -1.4 dBm and an average gain bandwidth and noise figure of 65.2 GHz and 7.5 dB respectively.
展开▼