Integrated photonics advancing on multiple fronts
Article By : Bill Schweber
Researchers are making progress toward reducing the size and number of discrete elements by integrating electro-optical elements onto a common substrate.
In the last decade, there have been major university and corporate research efforts and tangible advances in integrated photonics. The objective has been to develop building blocks that will transform systems and components from “electronics plus optical” to a more seamless merging of the two technologies, which have a lot in common and exhibit major differences, as defined by the laws of physics.
A few examples show the range of progress. In one case study, researchers at the prestigious Swiss-based École Polytechnique Fédérale de Lausanne (EPFL) have built a compact waveguide amplifier by incorporating rare-earth ions into integrated photonic circuits.
Erbium-doped fiber amplifiers (EDFAs) have been used since the 1980s to provide power gain for the photons in optical fibers (Figure 1). They are critical in long-distance communication cables and fiber-based lasers to boost the optical-signal power; remember that optical power is boosted by increasing the number of photons at a given wavelength, not the “amplitude” of the photons themselves, which is a fixed quantity and solely a function of wavelength. Erbium ions are used because they can amplify light in the 1.55 mm wavelength region, which is where silica-based optical fibers have the lowest transmission loss.
Figure 1 In a schematic setup of a simple erbium-doped fiber amplifier, two laser diodes (LDs) provide the pump power for the erbium-doped fiber. The pump light is injected via dichroic fiber couplers, while optical isolators reduce the sensitivity of the device to back-reflections. Source: RP Photonics
There have been attempts to use an erbium-doped optical waveguide instead of distinct fiber, but power output was too low and production issues were difficult to overcome. Now, the EPFL team has built and tested an integrated circuit–based erbium amplifier which provides 145 milliwatts of output power and more than 30 decibels of small-signal gain, comparable to commercial fiber amplifiers (A photonic integrated circuit–based erbium-doped amplifier), as shown in Figure 2.
Figure 2 Researchers at École Polytechnique Fédérale de Lausanne in Switzerland have developed a compact waveguide amplifier by successfully incorporating rare-earth ions into integrated photonic circuits. Source: EPFL
The device is based on ion implantation with an ultra-low-loss silicon nitride (Si3N4) photonic integrated circuit and boasts a waveguide structure about a half meter long on a millimeter-scale footprint. This project is another step toward reducing the size and number of discrete elements by integrating electro-optical elements onto a common substrate—somewhat analogous to how discrete transistors and passive components were subsumed by ICs.
Intel’s laser array
There’s also progress in multiwavelength integrated optics. Intel Labs has demonstrated an eight-wavelength distributed feedback (DFB) laser array that is fully integrated on a wafer, fabricated using the company’s 300-millimeter silicon-photonics manufacturing process. It delivers robust output power uniformity of ±0.25 dB and wavelength spacing uniformity of ±6.5% that exceed industry specifications.
Such co-packaged optics using dense wavelength division multiplexing (DWDM) technology offer the potential of greatly increased bandwidth while significantly reducing the physical size of photonic chips. However, it has been very difficult to produce DWDM light sources with uniform wavelength spacing and power until now (Figure 3).
Figure 3 The illustration shows eight micro-ring modulators and an optical waveguide, with each micro-ring tuned to a different wavelength of light with even spacing and each micro-ring can be individually modulated. Source: Intel
The construction of the Intel device ensures consistent wavelength separation of light sources while maintaining uniform output power, resulting in meeting one of the requirements for optical compute interconnect and DWDM communication. Intel uses advanced lithography to define the waveguide gratings in silicon prior to the III-V wafer-bonding process, yielding better wavelength uniformity when compared to conventional semiconductor lasers manufactured in 3-inch or 4-inch III-V wafer fabs.
In addition, due to the tight integration of the lasers, the array maintains its channel spacing when the ambient temperature shifts—always a major consideration in optical devices, as temperature-induced drifts play havoc with the basic consistent performance.
Optical chiplets
While these are lab developments and potentially important steps toward commercial products, some integrated photonic advances are on the market. Ayar Labs, which has Intel among its backers, is delivering monolithic in-package optical I/O (OIO) chiplets. These integrated silicon-photonic devices are built with CMOS processes inside a multi-chip package (MCP). They eliminate electrical I/O bottlenecks, providing power efficiency, low latency, and high bandwidth density in the never-ending quest for more and faster performance, as shown in Figure 4.
Figure 4 The hierarchy of approaches to electrical optical integration shows the ascending range of options, with fully integrated photonics at the top. Source: Ayar Labs
The above solution combines TeraPHY (an in-package optical I/O chiplet) with SuperNova (a multi-wavelength optical source) merging silicon photonics with standard CMOS manufacturing processes to deliver up to 1000x bandwidth density improvement at 1/10th the power compared to electrical I/O (Figure 5).
Figure 5 In this advanced assembly, TeraPHY OIO chiplets, each containing up to eight 256-Gbps optical ports, are flip-chip attached to simplify in-package integration of many optical ports and automate assembly. Source: Ayar Labs
I wonder: Are we seeing enough advancement across multiple facets of integrated photonics, so we are reaching an “inflection point” where the technology will break open in the next few years, and we will transition rapidly to widespread design-in of these devices? Or will it be a case of slow, steady, incremental progress and correspondingly paced adoption of such devices and the architectures they enable? Or will challenging obstacles and issues related to mass production hold things back?
Let’s look back in five to 10 years and see what has happened. It will be interesting to compare today’s predictions, extrapolations, and expectations with that reality.
This article was originally published on Planet Analog.
Bill Schweber is an electronics engineer who has written three textbooks on electronic communications systems, as well as hundreds of technical articles, opinion columns, and product features. In past roles, he worked as a technical website manager for multiple EE Times sites and as both Executive Editor and Analog Editor at EDN. At Analog Devices, he was in marketing communications; as a result, he has been on both sides of the technical PR function, presenting company products, stories, and messages to the media and also as the recipient of these. Prior to the marcom role at Analog, Bill was Associate Editor of its respected technical journal, and also worked in its product marketing and applications engineering groups. Before those roles, he was at Instron Corp., doing hands-on analog- and power-circuit design and systems integration for materials-testing machine controls. He has a BSEE from Columbia University and an MSEE from the University of Massachusetts, is a Registered Professional Engineer, and holds an Advanced Class amateur radio license. He has also planned, written, and presented online courses on a variety of engineering topics, including MOSFET basics, ADC selection, and driving LEDs.