FOMs extraction for quantum-well lasers and low-noise amplifiers with a modified genetic algorithm technique

Abstract

Semiconductor market trends indicate the rapid development of new integrated systems containing both optoelectronic and wireless-communication functionalities [1]. For projection-display devices and high-density optical disks in new integrated systems, the green light can be generated from solid-state lasers, in combination with second-harmonic generation (SHG) crystal, owing to their distinct features such as low noise, high-frequency-modulation capability, compactness, and low-power consumption [2]. The 1060 nm InGaAs quantum-well (QW) laser diode is one of applicable active components for this application [3]. However, the internal loss (α j ), the internal quantum efficiency (η j , and the transparency current density (J th ), which are fig-ures-of-merit (FoMs) for the QW laser under high-power-laser operation, are needed to be further improved. In the other respect, the low-noise amplifier (LNA), for wireless-communication functionalities in new integrated systems, is exploited to overcome the noise produced in subsequent stages by amplifying incoming radio-frequency (RF) signals, while introducing minimum amount of noise [4]. Nowadays, the SiGe LNA employing HBTs has been a main contender among the Si-based IC industry due to their advantages of higher cutoff frequency, lower noise, and better power performance. In general, the performance of SiGe LNAs relies on the geometrical-scaling issue, e.g., the emitter-length scaling [5]. For the HBT-based LNA design, attention should be devoted to optimizing FoMs such as the minimum noise figure (NF mm ) and the associated available gain (G A, assoc ) [6].