Abstract

Although the first arrays employing anti-guiding were reported almost 10 years ago, renewed interest in leaky-mode arrays commenced only two years ago with the fabrication of arrays with considerably improved performance.¹ These devices lased in the fundamental mode with a diffraction-limited on-axis far-field peak at currents as much as 2.5 times threshold. Since that time, further improvements have been made, both in our understanding of these devices and in our understanding of their emission characteristics. Theoretically, work began with the use of the effective-index method, which has provided important insights into the nature of leaky array modes.² Newer work has employed both analytic and numerical methods that include the full two-dimensional cross-sectional structure of these devices.^3,4 The two most significant achievements of this body of theoretical work have been (1) the illumination of the discrimination mechanisms present^2-4 and (2) the discovery of resonance effects in device design.² Recent experimental achievements include pulsed operation to 10 times threshold, cw operation of 250 mW per facet, and the design and fabrication of a new generation of devices. These new devices have lower internal losses than older devices and should eventually demonstrate improved efficiencies and emission quality.

© 1990 Optical Society of America