Abstract

More than fifteen years since the first demonstration of a monolithic long-wavelength photoreceiver by Leheny et al.,¹ optoelectronic integrated circuit (OEIC) photoreceivers have come a long way and are today the most widely researched subject worldwide. The multiplicity of devices (both photodetectors and transistors) available in the InP material system, combined with the flexibility of epitaxial crystal growth has resulted in several excellent implementations of high-performance OEIC photoreceivers. Photodetectors such as p-i-n, M-S-M, and phototransistor have been integrated with several types of transistors (such as JFETs, HEMTs, MESFETs, HBTs) to realize monolithic photoreceivers operating at data rates ranging from 100 Mbit/s to 20 Gbit/s.^2–4 The overall performance of OEIC photoreceivers today is comparable to the best hybrid receivers. In fact, it has been demonstrated³ that at very high speeds, the sensitivity of an OEIC receiver is better than the best hybrid receiver as a result of reduced parasitics due to integration. The field has matured sufficiently so that issues such as yield, reliability, cost, and manufacturability are now being investigated actively in several laboratories. With the growth of the fiber-to-the-home market, there is expected to be a large demand for photonic components, and the monolithic photoreceiver has the performance and the capability to satisfy the requirements for this application.

© 1996 Optical Society of America