Final Report Summary - ULPPIC (Ultralow power photonic integrated circuits for short range interconnect networks)
The aim of the ULPPIC project is to develop novel optical devices such as lasers and switches, with very low power consumption. The primary application being targeted is very short range optical interconnections, e.g. between different electronic chips, but also other applications such as low power optical sensors and quantum optics are being considered. The main approach followed in the project is to embed novel optical active materials within a high quality factor optical cavity and integrated these with a high performance optical backbone network based on silicon or silicon nitride. Two types of optical active materials are being considered: III-V semiconductor materials, which are epitaxially grown on silicon and colloidal quantum dots, which are formed first in solution using chemical methods.
The difficulty in epitaxially growing optically active materials, such as III-V semiconductors, on silicon lies in the large lattice mismatch between these materials and the silicon material. Therefore we rely on a technique recently developed by the electronics industry for realizing the next generation high speed transistors and which allows us to define small islands of high quality III-V semiconductor on silicon. The challenge now resides in designing a high performance optical cavity around these islands. We managed to do exactly this and demonstrated lasing, first from InP-nanowires, then from more practical InP-DFB like devices grown in trenches on a silicon substrate. Finally we also showed laser operation from InGaAs-on-InP-on-Si structures, allowing to shift the laser wavelength to 1300nm. This wavelength is used in optical telecom and interconnect applications and compatible with classical silicon photonics devices.
Colloidal quantum dots are formed in solution using chemical methods and through their composition and size the emission wavelength can be tuned. They can easily be deposited on top of a substrate and there is no issue of lattice mismatch as with the epitaxially grown materials but it is not straightforward to embed them directly in a high index contrast cavity. Within ULPPIC we showed that we can embed them in Silicon Nitride cavities and demonstrated disk based devices with embedded quantum dots. In the final stage we could even show lasing from such devices, demonstrating this technology platform is ready for practical applications.
The difficulty in epitaxially growing optically active materials, such as III-V semiconductors, on silicon lies in the large lattice mismatch between these materials and the silicon material. Therefore we rely on a technique recently developed by the electronics industry for realizing the next generation high speed transistors and which allows us to define small islands of high quality III-V semiconductor on silicon. The challenge now resides in designing a high performance optical cavity around these islands. We managed to do exactly this and demonstrated lasing, first from InP-nanowires, then from more practical InP-DFB like devices grown in trenches on a silicon substrate. Finally we also showed laser operation from InGaAs-on-InP-on-Si structures, allowing to shift the laser wavelength to 1300nm. This wavelength is used in optical telecom and interconnect applications and compatible with classical silicon photonics devices.
Colloidal quantum dots are formed in solution using chemical methods and through their composition and size the emission wavelength can be tuned. They can easily be deposited on top of a substrate and there is no issue of lattice mismatch as with the epitaxially grown materials but it is not straightforward to embed them directly in a high index contrast cavity. Within ULPPIC we showed that we can embed them in Silicon Nitride cavities and demonstrated disk based devices with embedded quantum dots. In the final stage we could even show lasing from such devices, demonstrating this technology platform is ready for practical applications.