Silicon Based Photonic Components for High Performance Computing (HPC) Networks

High performance computing networks (HPC) require very high data transmission rates with large bandwidth. Fiber optic networks can meet the data transmission and bandwidth requirements. However, new component technology is needed to provide the interface to computer logic and memory systems, with are primarily silicon based electronics. Direct bandgap III-V semiconductor materials such as gallium arsenide (GaAs) and indium phosphide (InP) have excellent photonic properties, but are not easily integrated with silicon based microelectronics and fabrication techniques.This SBIR project will develop and demonstrate photonic components based on silicon- germanium-tin (SnGeSi). SnGeSi can be readily integrated with silicon microelectronics at the chip level, very similar to existing technology for SiGe. The addition of Sn induces a direct bandgap in this material, and enables high performance photonic devices based on SnGeSi. In Phase I, the SBIR/STTR team will demonstrate silicon compatible fabrication technology for SnGeSi based photonic devices. Phase I will demonstrate epitaxial thin film deposition of SnGeSi on silicon substrates by chemical vapor deposition (CVD). The Phase I project will also fabricate and demonstrate infrared emitter and detector devices, which are the fundamental building blocks for photonic components. Modeling and computational techniques for SnGeSi based devices will be established in Phase I, as well as the pathway forward to SnGeSi photonic component prototype demonstration in Phase II. The successful conclusion of this SBIR program will result in a new generation of silicon based photonic devices, which will enable computers to exchange data at high transmission rates over fiber optic networks. The resulting high performance computing networks will enable advanced computational work in commercial, scientific and military applications. Commercial Applications and Other Benefits: High performance computing networks are essential for processing, storing and analyzing vast amounts of data for commercial, scientific and military applications. High performance networks in scientific applications provide scientists remote access to instruments and facilities, while also allowing citizens access to the data and knowledge that has been produced. The full integration of silicon logic and memory devices with photonic networks will enable the long awaited dawn of the next generation in semiconductor electronics and photonics on one common platform; revolutionizing a multi-billion dollar industry.