A new palette for optical fibers

Picasso went blue for some time: mainly monochromatic hues with few bright and warm colors. Yet the so-called Blue Period lasted only a few years. Then pinks and reds and greens began to recur in his paintings, and without those colors—without this broad, festive palette—many of his masterpieces, including Le Rêve, the most expensive of his works auctioned to date, would lose their vitality. Great paintings were created in blue, but with the whole spectrum, even more. 

 

This is essentially what ECE researcher Peter D Dragic and his collaborator at Clemson University, John Ballato, have been discovering with optical fibers. While fiber structure has become increasingly complex in recent years, the materials used to construct them have been stuck in one section of the periodic table, within a few spaces of silicon (silica—SiO₂—being the predominant fiber component). The materials are monochromatic, in other words, like Picasso’s blues. 

 

“We’re saying, look at all this table,” Dragic said one day recently, indicating a periodic table pulled up on his computer screen. Each element and crystal combination has a unique suite of properties—different refractive indices, different thermal-optic effects—and by utilizing these materials in the right combination, specific optical properties can be tailored into the fibers with relatively simple fabrication techniques.  

 

Take, for example, Brillouin scattering, a phenomenon that arises in high-power applications where the optical waves transmitting information through a fiber are interrupted by acoustic waves that stem from pressure-induced changes in the fiber. The optical signal is scattered backwards and can, in turn, disrupt additional optical transmission. The information is distorted or even lost.    

 

To minimize Brillouin scattering, Dragic and Ballato knew that, relative to silica, they would have to design a fiber with a lower photoelastic constant—a ratio that indicates how much the fiber medium is influenced by the light. “The question though is, how does one start from a crystal, and from that understand what the glass is going to do,” Dragic said, referring to the heating process whereby crystalline solids melt and form glasses. “Glasses are little bit of an enigma because they’re disordered—they’re amorphous—and there are also different levels of order and disorder.”

 

The team spent some time establishing trends between crystals and glasses, and using that model, they can calculate certain properties that can be achieved in fibers, including, in the case of Brillouin scattering, an improved photoelastic constant. That model led initially from yttrium aluminum garnet (Y₃Al₅O₁₂, aka YAG) to sapphire (Al₂O₃) to spinel (MgAl₂O₄) to barium oxide (BaO). With the last three materials, the model suggests that a full removal of Brillouin scattering is possible. “We haven’t actually observed the zero,” Dragic said, “but we’ve got a factor of one hundred decrease in its strength, so that’s really outstanding and raised a lot of eyebrows.” In fact, their sapphire-derived fibers have set records for the lowest observed Brillouin scattering. 

 

Conventional fibers are limited in a large part by conventional fabrication methods. This process starts with a long silica tube, generally about four or five centimeters in diameter. The tube is clamped with both ends inside of a sealed vacuum and then heated. Any dopants—that is, any other compounds that are desired in the fiber, often a chloride form of phosphorous, boron, or germanium—are blown inside as vapor. The heat causes the vapor to react chemically, and a soot is deposited. The tube is then compressed into a solid glass rod, leaving a small core of the dopant in the center. This preform, as it is called, is taken to a draw tower where it is heated, and gravity pulls the glass downward, stretching it to the desired thickness (from centimeters to micrometers). 

 

Using this method, purity of the preform depends on large differences in vapor pressure between the dopant and any detrimental impurities. A halide solution (SiCl₄), for example, might have small amounts of iron chloride contamination (Fe₂Cl₆), but because their vapor pressures are significantly different, when the solution is heated, pure halide gas can be extracted and pumped into the silica tube. The fiber components must also form a stable glass at the high temperatures employed in the deposition process. Both of these factors constrain the combinations of materials that can be used in fiber. 

 

“There may also be a role played by the classic human condition whereby something done the same way for decades can paralyze creativity,” Dragic and Ballato write in a forthcoming article featured on the September cover of the Journal of the American Ceramic Society, admitting that this is a provocative assertion. “From a materials perspective, modern optical fibers are boring.” 

 

To expand the materials palette, Ballato’s fabrication team at Clemson has created fibers using a molten core approach. With this method, a thin rod of the desired core material is slid inside of the silica tube, instead of injecting gases. When this is heated in the draw tower, the core melts, dissolving and fusing with the inside of the silica tube. In the resulting fiber, a silica cladding surrounds the unique core material. By controlling this process, in terms of how much of the cladding is dissolved, the temperature, and the amount of time for which heat is applied, optical characteristics, previously unachievable, are possible. 

 

This molten core approach, in addition to producing exceptional fibers from a wider variety of materials, is a radically simple technique, especially when compared to recent research, which has focused on building structural wave-guides into the fiber: a honeycomb of holes, for example. These are known as photonic crystal fibers, and the optical improvements they offer are the result of changes to the micro-arrangement of conventional materials, rather than reconsidering the materials themselves. While that process is expensive, necessitating substantial infrastructure that is only practical for large-scale manufacturers, the molten core approach could be adopted by small companies. “It’s not something that requires a huge investment to implement commercially,” Dragic said.

 

Beyond the technical realm, Ballato pointed out the unique nature of the teamwork itself. “It is a model for inter-institutional and interdisciplinary collaborations,” he said. While his research is well established in material sciences—optical fibers in particular—Dragic’s background is in remote sensing lasers, a field which, though not normally associated with fiber fabrication, requires precise testing and modeling of optical materials. On this project, their expertise are perfect complements. “We iterate back and forth in a very open and productive way,” Ballato added. 

 

Based on their model for crystals and glasses, Dragic will suggest new materials to study. And after considering specific adjustments to optimize the molten core process (temperatures and dimensions among other things), Ballato will have a demonstration fiber boxed and ready to send to Illinois within a week or two.

 

Dragic then tests the fiber using optical breadboard apparatuses, custom-built at Illinois, checking for Brillouin scattering and other phenomena and properties, and upon processing the data with their model, he can suggest even better materials to try. “It’s all predictable and designable,” he said. “It’s not like we say, let’s just make something and see what happens.” That back-and-forth between modeling and fabrication is how sapphire led to spinel and so forth. And now, even with records set, their work continues. The palette is still being explored.