Since the early 1970s, the demand for higher communication and information traffic has caused the optical fiber network to burgeon. In fact, the modern fiber optic network comprises a large part of the Internet backbone. This includes long-distance communication cables containing optical fibers that are routed under the sea and through underground conduits – optical fibers that interconnect data centers around the world and bring fiber to the office (FTTO) and fiber to the home (FTTH).
In recent years, the growth of the fiber optic industry has been even more explosive. Fiber optic technology has expanded into short-distance connections between devices such as computer networking, high-definition televisions, and motherboards and devices within computers. Optical glass fiber, which is highly engineered in structure for multimode and singlemode, has evolved – as have most plastic optical fibers used for very-short-distance communications. Many companies have contributed to the proliferation of fiber’s uses and forms including (to name a few) Alcatel, AT&T, Ciena, Cisco, Corning, Finisar, JDS Uniphase, Lucent Technologies, and Tyco.
Making fiber optic cables for the huge diversity of the fiber network is a multibillion-dollar industry. Fiber optic cables – both outdoor and indoor patch cords and interlinks – need to assume a wide variety of custom form factors to accomplish all the different kinds of interconnection that is needed. A typical fiber optic patch cord is a glass optical fiber that is terminated at either end with connectors (such as SC, LC, and MTs), which allow the patch cord to be rapidly and reliably connected to other functional devices such as optical switches, optical couplers, amplifiers, and WDMs. The technology of fiber optic termination involves specialized epoxies, mirror polishing, inspection, and testing for continuity and performance.
The development of polymer waveguides as an alternate solution
In the early 1980s, engineers at DuPont foresaw the need for a more easily produced, customized fiber optic assembly. In 1985, they demonstrated a waveguide that was photo-imaged in a polymer film. Between 1985 and 1998, DuPont developed a technology – and later spun out the company Optical InterLinks (OIL) – to make high-performance, low-cost, manufacturable optical data devices.
The technology uses a polymer waveguide: a flexible, self-supporting, polymer film developed using a photo-image process, which allows rapid duplication. Producible features include splitters/combiners, graded or step index profiles, and optical shuffles. Easily incorporated are 90-degree interconnections with I/O mirrors, mirror/surface reflectors, and other optical surfaces. This technology enables highly compact customized optical probes and sensors. By using matrix processes, optical sensors can also be produced using this type of process where two intersecting fibers transfer optical power as pressure distorts their waveguide paths, thus creating a “cross-talk” function. Low-cost sensors in matrix format could be mass produced that easily measure pressure points, analogous to how a hand holds a glass with palm and finger pads.
This automated process could enable high-volume applications for short-distance communications such as the automotive data bus. As cars become more and more complex, the use of highly reproducible plastic optical fiber bus systems that are inexpensive to reproduce in quantities of thousands (or hundreds of thousands) are critical. In researching this topic, part of my process included contacting Wayne Kachmar through his company, Technical Horsepower Consulting, for consultation on the Optical InterLinks (OIL) polymer waveguides technology. Mr. Kachmar’s experience – over 38 years of optical fiber cable design, testing, specification creation and installation – gave me additional insight into overarching industry trends and specific opportunities regarding this technology. According to optical cable expert Wayne Kachmar, polymer waveguides could be a competing technology in the automotive industry for the reasons mentioned above.
Keep in mind that polymer waveguides are not a new technology. Between 1993 and 1997, the Defense Advanced Research Projects Agency (DARPA) supported a project in which Optical InterLinks provided the flexible polymer waveguide array between PD/VCSELs and its Parallel Optical Link Organization. Another project followed with DARPA, which is a US Department of Defense agency. Numerous publications, delivered prototypes, and publicity from DARPA projects generated market awareness and interest in polymer waveguides. In 30 years of working with this technology, Optical InterLinks has come to understand many of the subtleties of the photo-chemistry of this system to further refine the performance of the optical devices created.
Why hasn’t this technology caught on in the fiber optic industry?
Polymer waveguides have succeeded in generating interest, and some companies have watched this developing technology for quite some time. Every technology introduced into a market must find a niche and solve a problem. In which areas could this technology solve problems? Wayne Kachmar’s consultation with me included some of the most viable applications and opportunities, listed here:
- Computer motherboard applications for high-speed computing
- The ability to provide optical input to flexible OLED (organic LED) devices – In “wearable signage,” a screen on your shirt would receive and transmit data
- Automotive networking systems – The average communications load in automobiles is closely following Moore’s Law (the amount of computing power doubles every 2 years)
What is preventing a major adoption of polymer waveguides in the fiber optic industry?
In my consultations with Optical InterLinks’ engineers and industry experts such as Wayne Kachmar, I believe there are 4 key reasons why the polymer waveguides technology has not been fully integrated into our industry:
- Despite the fact that Optical InterLinks’ polymer waveguides technology is 30 years old, it is still very unique. In some ways, this technology is a solution that is ahead of its time. Although sensor technology, automotive data buses, and optical backplanes for computers are rapidly coming to fruition, they do not yet require the speed and simplicity that this technology represents. (Metallic conductive systems such as copper bus systems have significantly surpassed experts’ expectations, although we may be approaching the limits of copper.) Creating a photo-imaging template requires a library of photonic elements and rules for spacing, minimum radii of curvature, modal filling, and so forth. Optical InterLinks has developed many of these components for its polymer waveguides technology, but they have not yet found a solid foothold in the fiber optic industry.
- Alternate technologies using Si and Si-based materials seek to make waveguide structures that are smaller, denser, and cheaper. Methods to integrate light sources and amplification to silicon-based solutions are being addressed by a mix of companies including system houses and (photonic) integrated circuit manufacturers, with quite a variety of competing solutions. Companies involved in the push to make light penetrate further into integrated circuits prefer to develop and maintain innovative technologies themselves. The tendency is to keep this proprietary information in-house rather than collaborate with an outside partner. Due to the lack of intellectual property connected to the polymer waveguides technology, it’s effectively an open-source technology. This means, in all likelihood, no one organization would invest a large amount of money to further develop it. On the other hand, this means the door is open for anyone to embrace the technology and integrate it into their application.
- The advent of bend-insensitive fibers has severely limited the interest in Optical InterLinks’ polymer waveguides technology for any high-bandwidth applications. Also, the difficulty of connectorization of conventional fibers may become obsolete with the rapid advancement and acceptance of additive technology (aka 3D printing) to create (print) fiber optic connectors. These new, competing technologies – bend-insensitive fiber and additive technology – are likely to be more popular choices for applications such as optical backplanes for computers. These technologies are highly patentable, and there is potential to mass produce products using these technologies. In my discussion with Wayne Kachmar, he put forth an interesting idea: “In the automotive world, two technologies working together – additive technology and OIL’s polymer waveguides technology – could potentially be a killer solution for automotive data backbones. This niche potentially offers the volume, demand, and requirements to move polymer waveguides forward. This application could be a great match for the polymer waveguides technology.”
- Termination of polymer waveguides is still a manual process, and needs to become more automated. However, as waveguides penetrate further into electronic chips, standard connector termination with SC, LC, and MTs will continue outside the chip almost as a pigtail. But in attaching the pigtail to the chip emitters or detectors, other methods of coupling light such as additive technology may become more common on chip. This may allow for other methods including polymer waveguide devices to enter into the inter-chip area. However, it is unlikely to change the chip-to-glass fiber interface side of the connection due to many technical details such as index of refraction mismatch, flat-to-round waveguide shapes as well as a number of other physical challenges, beginning with the difference in thermal shrinkage of polymer and glass (and therefore the silicon substrate material).
Conclusion
Optical InterLinks’ polymer waveguides technology has not yet found a niche in the fiber optic industry. While many innovations throughout history have multiple advocates and follow roughly parallel paths – for example, the “electrical” feud of Nikola Tesla and Thomas Edison or the “format war” of Betamax and VHS tape – usually one emerges as the preferred choice and brings standardization to the industry. For some technologies and industries where computing power is just starting to exert its influence, such as sensors and automotive systems, polymer waveguides may gain a foothold. Wayne Kachmar summed it up nicely in our discussion, “There are opportunities for this particular technology. Polymer waveguides does have promise for the fiber optic industry.”
