19 March 2020
First bufferless 1.5μm III-V lasers grown directly on silicon wafers in Si-photonics
Hong Kong University of Science and Technology (HKUST) has reported what it reckons is the first 1.5μm III-V lasers grown directly, without a buffer layer, on industry-standard 220nm SOI (silicon-on-insulator) wafers using metal-organic chemical vapor deposition (MOCVD), potentially paving the way to interfacing with Si-based photonic devices and the subsequent realization of fully integrated silicon photonic circuits (Yu Han et al, ‘Bufferless 1.5µm III-V lasers grown on silicon photonics 220nm silicon-on-insulator platforms’, Optica, vol7, issue2, p148). Previous demonstrations required non-industry-standard bulk silicon or thick SOI wafers.
Bridging the active III-V light sources with the passive Si-based photonic devices, the development could be deployed as light sources in integrated circuits to greatly improve circuit speed, power efficiency and cost-effectiveness.
In other conventional approaches of integrating III-V lasers on silicon in the literature, III-V buffers up to a few microns thick are used to reduce the defect densities, which poses huge challenges for efficient light interfacing between the epitaxial III-V lasers and the Si-based waveguides.
Now, a team led by professor Lau Kei-May of HKUST’s Department of Electronic and Computer Engineering and post-doctoral fellow Dr Han Yu have devised (for the first time, it is reckoned) a novel growth scheme to eliminate the requirement of thick III-V buffers, promoting the efficient coupling of light into silicon waveguides. The bufferless feature points to fully integrated Si-based photonic integrated circuits.
Improvements in the efficiency of conventional electronic data systems cannot catch up with the soaring data traffic, which calls for the integration of photonic functionalities onto conventional Si-based electronic platforms. Integration could produce optoelectronic integrated circuits with unparalleled speed and functionalities, and enable new applications. Yet fundamental differences between silicon and III-V materials means it is extremely challenging to directly grow III-V functionalities on the silicon.
Lau’s group at HKUST’s Phonics Technology Center has endeavored to integrate III-V materials and functionalities on mainstream silicon wafers for over a decade, innovating and optimizing various approaches to improve the performance of III-V lasers grown on silicon, with the goal of progressively approaching the requirements of the industry. This work is part of their project on monolithic integration of III-V lasers on silicon.
Taking advantage of the constituent diffusivity at elevated growth temperatures, the reseachers first devised a unique MOCVD growth scheme for the direct hetero-epitaxy of high-quality III-V alloys on the 220nm SOI wafers through synergizing the conventional aspect ratio trapping (ART) and the lateral ART methods. In contrast to prevalent epitaxy inside V-grooved pockets, the method features epitaxy inside trapezoidal troughs, enabling the flexible integration of different III-V compounds on SOIs with different silicon device layer thicknesses.
Using indium phosphide (InP) as an example, the reseachers detailed the growth process and then characterized and evidenced the crystalline quality of the epitaxial III-V materials through extensive transmission electron microscopy and photoluminescence measurements. The team designed and fabricated designed and fabricated both pure InP and InP/InGaAs lasers with air-clad cavities based on numerical simulations. Testing the devices showed that the lasers could sustain room-temperature and low-threshold lasing in both the 900nm band and the technologically important 1.5μm band under pulsed optical excitation. The demonstration leads to the potential to monolithically integrate III-V lasers on the industry-standard 220nm SOI wafers in an economical, compact, and scalable way.
“If practically applied, our technology could enable a significant improvement of the speed, power consumption, cost-effectiveness and functionality of current Si-based integrated circuits,” says Lau. “Our daily electronic devices, such as smartphones, laptops and TVs - basically everything connected to the Internet - will be much faster, cheaper, using much less power and multi-functional,” she adds.
“The next step of our research will be to design and demonstrate the first electrically driven 1.5μm III-V lasers directly grown on the 220nm SOI platforms, and to devise a scheme to efficiently couple light from the III-V lasers into silicon waveguides and thereby conceptually demonstrate fully integrated silicon photonics circuits,” Han says.