As laser designers seek ever-higher powers from single-mode fiber lasers, they encounter a fundamental roadblock. A small core in a fiber laser forces single-mode oscillation, but the spatial intensity in a small-core, high-power laser is great enough to damage the fiber. A larger core alleviates the intensity problem, but it allows higher-order modes to oscillate.A solution to this quandary has been to bend the fiber because high-order modes experience greater bending loss than the fundamental mode. A large-core fiber coiled in a circle produces high power in a single transverse mode.But this solution is good only up to a point. As the core gets larger, the entire fiber gets bigger, and the bigger the fiber is, the harder it is to bend. Eventually, fibers get to the point where they cannot be bent sharply enough to curtail high-order modes.Recently, scientists at the University of Southampton in the UK finessed this problem by fabricating a fiber with a helical core (Figure 1). The key is that the total internal reflection physics is the same whether an entire fiber is bent or just the core is bent inside a straight fiber. A much sharper bend is possible, however, when only the core is bent — in this case, in a helix.Figure 1. The core of the fiber laser was bent into a helix (spiral), causing greater bending loss to high-order modes than to the fundamental mode. ro: offset from center, P: pitch of helix. Images ©OSA.The researchers fabricated the double-clad, helical-core fiber by constructing a preform whose core was offset from center and then by spinning the preform around its axis as the fiber was drawn. Because the drawing speed was constant, the pitch of the helix was determined by the perform’s spinning rate. The experimental fibers had an offset of 100 μm, a core diameter of 30 μm and an inner-cladding diameter of 275 μm. The pitch of various fibers ranged from 8.3 to 9.6 mm.The scientists pumped the 2.8-m-long, Yb-doped, helical-core fiber with a diode-laser stack at 976 nm (Figure 2). The resonator comprised a high-reflecting mirror on one end and the cleaved face of the fiber at the output end.Figure 2. The investigators coupled up to 95 W of pump power into the inner cladding of the experimental fiber laser. Mechanical alignment issues precluded actively cooling the ends of the fiber, but they believe that much higher pump powers and correspondingly higher outputs would be possible with active cooling. Because the core was not orthogonal to the perpendicularly cleaved face of the fiber, the mirror had to be offset, as indicated in the figure. Likewise, the cleaved end of the fiber that served as the output coupler was somewhat misaligned, but it provided sufficient feedback to support lasing.By using fibers with different helical pitches in the laser, the investigators observed that a relatively long pitch was sufficient to extinguish high-order modes at low pump powers. At higher powers, however, the gain for all modes increased, and only fibers with a short pitch lased in a single mode. They obtained the best results with an 8.5-mm-pitch fiber: 60.4 W at 1043 nm from 92.6 W of launched pump power. They measured the M2 beam quality parameter to be 1.30 and 1.37 in the two transverse planes.For comparison, the scientists also fabricated a straight-core fiber and evaluated it in the same experimental configuration. This laser had a lower threshold than the helical-core lasers, but the maximum output — 63 W from 92.6 W of launched pump power — was only slightly greater. The straight-core fiber was multimode, with an M2 of 3.6 at all power levels.The researchers believe that the higher threshold of the helical-core laser is due to the low feedback efficiency of the output end of the fiber and that it could be remedied by cleaving the fiber at the appropriate angle so that the face would be orthogonal to the core.