ABSTRACT
The global data traffic is going up with an unprecedented growth rate over the years. While the continuous scaling of silicon (Si) electronic devices pushes ever higher computational speed, high dissipation electrical interconnects quickly become the bottleneck. The use of photonic integrated circuits (PICs) in communication systems is under extensive investigation, boosting the development of silicon photonics technology. In order to leverage the massive manufacturing capability and cost reduction provided by silicon foundries, it is of significant interest to realize the full suite of optical components with group IV semiconductors. However, despite the great electronic and waveguide properties, silicon cannot emit light efficiently due to its indirect bandgap properties. The lack of high-performance, long-lifetime and electrically injected silicon-based lasers continues to pose a major roadblock. In contrast to group-IV semiconductors (such as Si or Ge), most of the III-V compound semiconductors possess a direct bandgap with excellent photon absorption and emission efficiencies. Integrating a reliable electrically pumped III–V laser on Si would be ideal to overcome silicon’s deficiency. Yet the challenges are also enormous, primarily arising from differences in the lattice, thermal mismatch and crystal polarities, which result in severe non-radiative recombination and device early failures in lasers grown on silicon. Recent advances in advanced heteroepitaxy technologies have opened new opportunities to tackle these challenges associated with material incompatibility. In this chapter, we review progress in III–V quantum dot (QD) lasers and nano-lasers monolithically grown on silicon, together with a discussion of state-of-the-art device implementations towards PICs.
