Researchers from Intel and the University of California, Santa Barbara, on Monday said they have built the world’s first electrically powered hybrid silicon laser using standard silicon manufacturing processes.
This development addresses what Intel sees as one of the last major barriers to producing low-cost, high-bandwidth silicon photonics devices for use inside and around future computers and data centers. Photonics is the technology of transmission, control, storage and manipulation of information using light.
“We believe this technology is volume-scalable and will enable us to produce hundreds of lasers on a single bonding step with no need for alignment,” said Mario Paniccia, director of Intel’s photonics technology lab. Intel is trying to “siliconize” photonics to enable a technology that will allow the company to drive optical communications to the mass market, he added.
The Photonics Challenge
While widely used to mass produce affordable digital electronics today, silicon can also be used to route, detect, modulate and even amplify light, but not to effectively generate light. In contrast, indium phosphide-based lasers are commonly used today in telecommunications equipment. The need to individually assemble and align them has made them too expensive to build in the high volumes and at the low costs needed by the PC industry, according to Intel.
The researchers were able to combine the light-emitting properties of indium phosphide with the light-routing capabilities of silicon into a single hybrid chip. When voltage is applied, light generated in the indium phosphide enters the silicon waveguide to create a continuous laser beam that can be used to drive other silicon photonic devices. Waveguides are structures that guide electromagnetic waves along their lengths. They are used for transferring power and communication signals.
A laser based on silicon could drive wider use of photonics in computers because the cost can be greatly reduced by using high-volume silicon manufacturing techniques.
“We are able now to start integrating photonic devices together to actually build form factors at sizes and costs that fundamentally weren’t available previously,” Paniccia explained. The long-term vision, he added, is to integrate the electronics to converge communications and computing all on one platform.
The Public-Private Solution
The hybrid silicon laser involves a design employing indium phosphide-based material for light generation and amplification while using the silicon waveguide to contain and control the laser. The key to manufacturing the device is the use of a low-temperature, oxygen plasma — an electrically charged oxygen gas — to create a thin oxide layer — roughly 25 atoms thick — on the surfaces of both materials.
When heated and pressed together, the oxide layer functions as a “glass-glue” fusing the two materials into a single chip. When voltage is applied, light generated in the indium phosphide-based material passes through the oxide “glass-glue” layer and into the silicon chip’s waveguide, where it is contained and controlled, creating a hybrid silicon laser. The design of the waveguide is critical to determining the performance and specific wavelength of the hybrid silicon laser.
“This is not what most people are doing, which is to bond lasers on top of silicon, and you couple it into silicon,” John Bowers, a professor of electrical and computer engineering at UC Santa Barbara, told TechNewsWorld. “The light is propagating and guided and controlled by the silicon, and we amplify the tail of it to extend into the indium phosphide. That’s what makes for very efficient devices.”