With the booming interest in and use of high-power fiber lasers, the drive to further boost their power has spurred much research into the development of large-mode-area (LMA) optical fibers as gain fibers. In a LMA fiber, the core is physically large enough to support more than one mode, but the physical characteristics of the fiber are designed to suppress higher-order modes. As a result, the desired transverse fundamental lasing mode can be spread across a large cross-sectional area, lowering the potential of light-related damage to the fiber (as well as optical nonlinearities).
Efforts to create practical LMA fibers have typically centered on the use of step-index or microstructured designs. One example of successful single-mode operation via mode-selective attenuation in a step-index design occurs in a gain-guided and index-antiguided (GG + IAG) fiber. However, this fiber has an inherent leakage loss, reducing lasing efficiency.
Now, two researchers, Hyun Su Kim from the Department of Photonic Engineering at Chosun University (Gwangju, South Korea) and Seongwoo Yoo from the Photonics Institute, School of Electrical and Electronic Engineering, Nanyang Technological University (Singapore), have taken a different approach by changing the GG + IAG fiber design, which is conventionally step-index, so that it has a graded-index core; the change reduces leakage loss while maintaining the higher-order-mode suppression of the GG + IAG design (because of its raised-index cladding). A graded-index core fiber by itself, without the GG + IAG configuration, cannot maintain single-mode output under amplification (see figure).
Depressed graded-index core
"We investigated an alternative LMA design that can inherently suppress higher-order modes (HOMs), thus pushing up the threshold of the transverse mode instability," Yoo explains. "An antiwaveguide structure with a gain-guiding mechanism was proposed as a new route to selectively upscale the fundamental mode power. However, it was recognized that optical gain in this type of structure required achieving not only amplification but also overcoming the antiguide loss, which was difficult to do."
This fundamental limitation was overcome in the new theoretical design, in which the depressed graded-index greatly reduces the antiguide loss of the fundamental mode while the raised cladding index concurrently provides the mechanism to suppress HOMs. "The design offers power scaling of a Gaussian beam output via the inherent HOM suppression," Yoo adds.
The two researchers modeled the leakage losses of a step-index GG + IAG fiber and the new fiber design (with no gain) for core diameters of 50 and 100 μm, finding that the new design with the graded-index core reduced the leakage loss by more than a factor of 1000 as compared to the step-index core. They also found that, while the leakage loss of a step-index GG + IAG does not depend on the beam-radius size, for the new design, there is a dependency. So, they determined, for example, that for the optimum beam size for a 25 μm core radius, the leakage loss was <0.003/cm, which is again three orders of magnitude less than for the GG + IAG fiber.
Simulations with a pure graded-index fiber without the inverse index showed that, with amplification, the beam could not keep a Gaussian profile and, indeed, could damage the fiber core via beam distortion and self-focusing.
"Noting that the simulation suggests a practical range of operation, our next goal is to demonstrate the concept in the lab," Yoo notes. "This will include fabrication of a depressed graded-index ytterbium (Yb)-doped core, verification of Gaussian-mode amplification, and performance under fiber bending. Although the bending distortions, such as squeezing mode area and shift of the mode profile, seem not significant in a graded-index profile (also supported by our simulation), experimental demonstration will be more convincing in confirming robustness and adaptability to various amplifier configurations."
1. H. S. Kim and S. Yoo, Opt. Express (2017); https://doi.org/10.1364/oe.25.021935.