Testing a New Model for Tiny Fiber Cores

Patricia Daukantas

OPEN ACCESSicon

 

Scatterings imageShahraam Afshar of the Institute for Photonics and Advance Sensing at the University of Adelaide.

An Australian team researching so-called “holey” optical fibers—with a narrow core surrounded by a ring of hollow capillaries—has discovered that light occupies much less space in the core than previously thought.

The experiments confirm one of the predictions of the group’s theory for signal propagation in subwavelength-sized waveguides, said Shahraam Afshar, a research fellow at the new Institute
for Photonics and Advanced Sensing, part of the University of Adelaide (Opt. Lett. 34, 3577).

According to the standard model for optical waveguides, which uses scalar mathematics, the contrast between the core and cladding should be small, and the core should be much larger than the operating wavelength of 1,550 nm. This model is found to work well for conventional fibers.

In holey fibers, however, the core diameter is much smaller than the signal wavelength (about one-third of the wavelength) and the core/cladding index contrast is much higher, so these fibers require a more exacting mathematical treatment of nonlinear pulse propagation.

In two papers published last year (Opt. Express 17, 2298; Opt. Express 17, 11565), Afshar and colleagues developed a general vector-based model for pulse propagation in tiny waveguides. One thing they noticed: At large core diameters, both new and standard models predicted similar values of the effective nonlinear coefficient of the waveguide, called γ, but as the diameter shrank below the operating wavelength, the models diverged by a factor of two.

Previous experiments could not distinguish between the two models, so the Adelaide team fabricated a specialized bismuth borosilicate fiber. Although its outer core’s diameter is still about 150 µm, its inner core is only 450 nm wide, surrounded by three fragile nanostructures. Testing the attenuation in the fiber demonstrated that its γ is indeed twice as large as the standard model predicted.

The divergence between models comes mainly from the smaller effective area of the mode in the core waveguide, Afshar said. The nonlinear coefficient is inversely proportional to the mode’s effective area. “We have found that the area that light occupies is smaller than what we used to think.”

Next, the Adelaide collaborators—institute director Tanya Monro, senior research fellow Heike Ebendorff-Heidepriem and student Wen Qi Zhang—will experimentally test another prediction of the vector-based waveguide model: that Raman gain is 2.5 times higher in subwavelength-sized fiber cores than previously expected.

According to Afshar, these small-core fibers could be useful in all-optical networking and signal processing devices. Holey fibers could also play a role in sensor technology.

[1 of 3]  next>>
View/add comments

No Comments Available