By Mark Andrews
Mastery in advanced technology is established at its core through materials science and materials processing. In optical communications, some extraordinary achievements in the purity and dispersion characteristics of glass optical fiber have been recorded over the last 40 years. These advances are attributable to innovations in glass science and processing.
Not surprisingly, glass fiber has fostered its own cult of transparency, and ironically, an almost blind allegiance to silica. Photonics companies that concentrate only on silica do so at the risk of hitting the proverbial glass ceiling, while those that have control over a broad spectrum of materials and processes will enjoy significant advantage in future photonics markets.
At this stage of the photonics game, I believe that it is a mistake for any company to back complacency by adopting a single material or process. One just needs to look at the variety of solutions being touted at the recent Optical Fiber Conference exhibition at Anaheim, CA, to see a small hint of the diversity of material possibilities. OFC attracted an abundance of photonics companies offering a striking range of materials and/or processes to make similar products.
Take the lowly waveguide, for example. We saw devices fabricated from plastics, organic liquid crystals, silica, silicon, semiconductors, hybrid glasses, ceramics, and even metals. In some instances, the same material (silica) was used to make the same device, but by a different process. Nature, even more than OFC, instructs us that with the proper selection of materials and processing, one can conduct a photon in almost anything. While this fact translates into broadened opportunities, it does not mean that all solutions are viable.
It is unlikely that a "silicon revolution" will take place in photonics, like it did in microelectronics. That there will be dielectric and semiconductor hybrid and monolithic integration is mundane and only part of the photonics story. Probably the worst thing that could happen to this nascent industry is for one material or concept to dominate and obscure our optical thinking. Consider, for example, the potential of nanotechnology and its promise of the optical analog of the semiconductor in nanostructured photonic bandgap solids and other exotica.
Materials Exploration
We need to achieve integration, but doing so does not mean that photonics can be miniaturized or rationalized within a generic material platform. The printed circuit board did not vanish with the onset of miniaturization. Expect in the future to see recyclable plastic photonics and large integrated optical backplanes and motherboards made of composite waveguides, patterned by molding, light, extrusion, and ablation.
Companies need to explore a range of materials for competitive and technical advantage, niche success, ease of manufacturing, industry acceptance, potential for integration, cost of fabrication, option for design wins, and so on. Among the inorganic materials for passive integrated optics, silica may offer the lowest loss, for example, and may enjoy "name recognition," but crystalline media like the Bookham Technology (Abingdon, England) ASOC silicon bench offer unrivaled atom-by-atom precision in defining waveguides and other optical elements required for future integration.
Silica and semiconductor processes for planar lightwave circuits (PLC), however, come at the cost of complex and expensive equipment, significant expenditure in specialized personnel and facilities, and questionable integration yields. Despite claims to the contrary, few PLC products today are truly developed with manufacturing yields in sight, and even fewer can be touted as manufacturing successes alongside the daily output of millions of packaged lasers that enjoy the true benefits of mature semiconductor processing.
But progress is taking place. For instance, Lumenon Innovative Lightwave Technology (Montreal, Canada), with its core competencies in materials science and processing, has developed hybrid optical material platforms of glass and polymers specifically designed to overcome these manufacturing and high-yield obstacles to deliver products across both active and passive PLCs.
Optical switching
I mentioned at the outset how different perspectives on materials might lead to competitive products in similar application domains. Optical switching is among the new technological waves to hit the industry, and several candidate switch fabrics have rushed into the gap. These are primitive solutions, however, that for the most part cannot escape the millisecond time domain. Silica and polymers have approached optical switching through the thermo-optic effect. But electroholography offered by Trellis Photonics (Columbia, MD) is a bold assertion of ferroelectrics in switching that offers the potential of speed, density, throughput and multiplexing in a single fabric.
Control of multiphoton events for true "light-by-light" all-optical switching in integrated optics continues to elude us. The pursuit does remind us of the huge potential of harnessing active optical functions that are difficult or impossible to achieve in conventional silica for meaningful product development. Functions like parametric oscillation, harmonic generation, ultrafast modulation, and controlled wave-mixing can be obtained in semiconductor and disordered nonlinear optical composite media like polymers and hybrid polymer-silica glasses. It is among these elite materials that I believe true all-optical subpicosecond switching can be achieved.
In summary, the message for photonics is simple: to ride the optical wave all the way to the marketplace, companies must master the materials and processes that make light work.
Mark Andrews is cofounder, vice president, and chief technology officer at Lumenon Innovative Lightwave Technology Inc., 8851 Trans-Canada Highway, St-Laurent (Québec), H4S 1Z6 CANADA; e-mail: [email protected].