226

OPTICS LETTERS / Vol. 31, No. 2 / January 15, 2006

Efcient single-mode operation of a claddingpumped ytterbium-doped helical-core ber laser

P. Wang, L. J. Cooper, J. K. Sahu, and W. A. Clarkson
Optoelectronics Research Center, University of Southampton, Higheld, Southampton SO17 1BJ, UK Received September 14, 2005; accepted September 15, 2005 A novel approach to achieving robust single-spatial-mode operation of cladding-pumped ber lasers with multimode cores is reported. The approach is based on the use of a ber geometry in which the core has a helical trajectory within the inner cladding to suppress laser oscillation on higher-order modes. In a preliminary proof-of-principle study, efcient single-mode operation of a cladding-pumped ytterbium-doped helical-core ber laser with a 30 m diameter core and a numerical aperture of 0.087 has been demonstrated. The laser yielded 60.4 W of output at 1043 nm in a beam with M2 1.4 for 92.6 W launched pump power from a diode stack at 976 nm. The slope efciency at pump powers well above threshold was 84%, which compares favorably with the slope efciencies achievable with conventional straight-core Yb-doped double-clad ber lasers. 2006 Optical Society of America OCIS codes: 140.3510, 140.3570, 140.3480.

Over the past few years there has been rapid progress in scaling output power from ber lasers.1,2 One of the key issues for further power scaling of single-mode operation will be to increase the core area to prevent the need for excessively long bers for efcient pump absorption and to raise the threshold for unwanted nonlinear processes and optical damage while maintaining robust single-mode operation. This is generally achieved by employing a multimode large-mode-area core design with a low numerical aperture (NA) in combination with bend-loss suppression of higher-order modes2,3 to promote single-mode operation. However, scaling to higher power levels will require an even larger core diameter and a larger inner-cladding diameter to accommodate more pump power, making it increasingly difcult to coil the ber with a small enough diameter to use bend-induced loss as an effective means for suppressing higher-order modes. An alternative approach to achieving single-mode operation in a largecore ber laser is to employ a ber with a helical core trajectory within the inner cladding. Helical-core bers have a number of potentially attractive features for power scaling: The offset core helps to promote efcient pump absorption in a cladding-pumped ber conguration, and, more importantly, the helical loss that is due to the core trajectory is greater for higher-order modes and hence can be used as an effective means for suppressing high-order modes in a large-core ber device without resort to bending. Recently we demonstrated singlemode operation of a helical-core ber with a multimode ytterbium-doped core of 26 m diameter and a NA of 0.14 in a simple core-pumped laser conguration.4 The output power 350 mW was limited by the power available from our single-mode pump source. In a preliminary demonstration of a cladding-pumped Yb-doped helical-core ber laser5 with a nonoptimum core design, we obtained 24 W of output power in a multimode beam with M2 3 with a slope efciency of 50%. In this Letter we report on a double-clad Yb-doped helical-core ber laser in a cladding pumped conguration. The ber laser
0146-9592/06/020226-3/$15.00

yielded 60 W of single-mode output M2 1.4 at 1043 nm for 92.6 W of launched pump power from a diode stack at 976 nm and had a slope efciency at pump powers well above threshold of 84%. We manufactured the helical-core bers used in our experiments in-house by rst fabricating a preform with an offset Yb-doped aluminosilicate core and a pure silica inner cladding and then spinning the preform during the ber drawing process to obtain the helical-core trajectory within the inner cladding. For this preliminary proof-of-principle study, the core diameter, the core NA, the core offset ro, and the inner-cladding diameter were 30 m, 0.087, 100 m, and 275 m, respectively. The ber was coated with a low-refractive-index polymer outer cladding to provide a high NA of 0.49 (calculated) for the inner-cladding pump guide. The pitch, P, of the helical path (dened in Fig. 1) is determined by the preform spinning speed (in radians per second) and the ber pulling speed v, from the expression P 2v / . In practice, to maintain a constant ber diameter the pulling speed must remain xed, so one can adjust the pitch by varying the spinning speed. Fibers with a range of pitches (8.3, 8.5, 8.8, 9.2, and 9.6 mm) were fabricated by careful adjustment of the preform spinning speed. These pitches were chosen to yield propagation losses for the LP01 (fundamental) and LP11 modes (calculated by the analysis described in Ref. 6) in the regime of interest for efcient operation in the LP01 mode and suppression of higher-order modes (see Fig. 2). The results of this analysis suggested that helical-core bers with pitches from 8.5 to 8.8 mm would perform best, as they have a relatively low propagation loss 0.6 dB/ m for the LP01 mode and a relatively high

Fig. 1. Schematic diagram of a helical-core ber. 2006 Optical Society of America

January 15, 2006 / Vol. 31, No. 2 / OPTICS LETTERS

227

Fig. 2. Calculated propagation loss for the LP01 and LP11 modes versus pitch.

Fig. 3. Cladding pumped Yb-doped helical-core ber laser arrangement: T, transmission of the dichroic mirror.

tion such that light emerged from the core in the plane of the optical table, no attempt was made to actively cool the ber end sections by use of watercooled heat sinks. As a consequence, it was necessary to restrict the pump power to 95 W to prevent damage to the ber coating close to the ber end facet. Further modications to the ber holding arrangement to permit active cooling of the ber end sections should remedy this problem and allow much higher pump powers to be used without risk of damage. The pump absorption efciency was determined to be 5.8 dB/ m. This allowed a relatively short length of ber 2.8 m to be used. The ber laser output was extracted with the aid of a dichroic mirror (as shown in Fig. 3). We initially tested the bers at low pump powers (corresponding to a laser output power of 10 W) to determine the optimum value of pitch for efcient single-mode operation. In each case the beam propagation factor M2 of the output beam was measured. The results (shown in Fig. 4) indicate that the M2 parameter decreases from 1.96 to 1.2 as the pitch is reduced from 9.6 to 8.3 mm. This is roughly in agreement with our expectations and serves to illustrate the mode-ltering action of helical-core bers. The 8.3 mm pitch ber produced a beam with the lowest M2 parameter of 1.2 but had a higher threshold and a

propagation loss 9 dB/ m for the LP11 mode. In addition, a straight-core ber with the same core and cladding dimensions was pulled from the preform to allow a direct comparison between the performance of helical-core and that of straight-core ber lasers. The Yb-doped helical-core and straight-core bers were tested in the laser conguration shown in Fig. 3. A simple resonator design was employed that consisted of a length of the ber under investigation with feedback for lasing provided by a cleaved ber end facet (also serving as the output coupler) at the pump incoupling end of the ber and, at the opposite end of the ber, by an external cavity comprising an 8 mm focal-length aspheric collimating lens, a 100 mm focal-length focusing lens, and a plane mirror with high reectivity (HR; 99%) at the wavelength range 0.9 1.1 m. It should be noted that light emerges from a perpendicularly cleaved end facet on a helical-core ber at an angle arcn sini to the ber axis, which, in the case of the 8.5 mm pitch ber, is 6. Hence, careful alignment of the feedback cavity lenses and mirror at the appropriate angle to the ber axis (not shown in Fig. 3) is required. There is also a reduction in the feedback efciency at the pump incoupling end of the ber that is due to the angle of the core relative to the cleaved facet, but the feedback efciency is still sufcient for efcient laser oscillation with the high pump power available in our experiment. The pump light was provided by a diode-stack source at 976 nm and was coupled into the ber with the aid of two 30 mm focal-length aspheric lenses. With this arrangement a maximum of 120 W of pump could be launched into the ber. However, for the experiments described here, to facilitate adjustment of the ber end orienta-

Fig. 4. (Color online) Beam propagation factor M2 versus pitch at 10 W output power. a, Beam prole for an 8.5 mm pitch helical-core ber laser; b, beam prole for a straight-core ber laser.

Fig. 5. Output power versus launched pump power for the 8.5 mm pitch helical-core ber.

228

OPTICS LETTERS / Vol. 31, No. 2 / January 15, 2006

lower slope efciency than the other bers, indicating that the loss for the LP01 mode is too high for efcient laser operation. The calculated values for propagation loss versus pitch (shown in Fig. 2) suggest that an 8.8 mm pitch ber should have provided the best performance with a relatively low loss 0.13 dB/ m for the LP01 mode and a high loss 9.4 dB/ m for the LP11 mode. This ber, when it was tested in the laser conguration shown in Fig. 2, did in fact produce a nearly single-mode output at low powers 10 W with M2 = 1.3, but the beam quality degraded to M2 = 1.7 at an output power of 40 W, indicating that there was insufcient loss for discrimination against higher-order modes. The best overall performance was obtained with the 8.5 mm pitch ber. In this case, the M2 parameter was determined to be 1.26 at an output power level of 10 W. The output beam became slightly elliptical at high power levels, with the M2 parameters perpendicular to and parallel to the plane of incidence at the ber output facet, increasing to 1.30 and 1.37, respectively. The increase in the latter M2 value is thought to be due to thermally induced distortion of the ber end face, which is at an angle of 84 to the exiting laser beam. The beam prole (Fig. 4, inset a) conrms the single-mode nature of the output beam. In contrast, the output beam from the Yb-doped straight-core ber laser (the same resonator conguration was employed) was multimode, with M2 = 3.6 at all power levels (Fig. 4, inset b). The 8.5 mm pitch helical-core ber laser reached threshold at a launched pump power of 10 W and yielded a maximum output power of 60.4 W for 92.6 W of launched pump power (see Fig. 5). At pump powers well above threshold, the slope efciency increased to 84%, which compares favorably with the best slope efciencies routinely achievable with more-conventional straight-core Yb-doped doubleclad ber lasers. The relatively high threshold pump power and hence lower slope efciency at low pump powers can be attributed to the low feedback efciency for the outcoupling end of the ber. We could remedy this problem by polishing the output facet of the ber at the appropriate angle, such that the core axis is perpendicular to the facet or by using a second external-feedback cavity. By comparison, the straight-core ber laser had a lower threshold pump power of 4 W (launched) but produced only a slightly higher output power, 63 W for 92.6 W of launched pump power. Thus, with an optimized design, a cladding-pumped helical-core ber laser should yield a performance in terms of output power and efciency similar to that of a more-conventional straight-core ber laser. The center lasing wavelength and emission linewidth for the helical-core ber laser were 1043 and 9 nm, respectively. The former was a little shorter than is typical for cladding-pumped Yb-doped ber lasers owing to a combination of reduced helical loss at shorter wavelengths and the relatively short ber

length (i.e., with lower reabsorption loss). In contrast, the straight-core ber laser operated at a somewhat longer wavelength, with emission lines of 1050 and 1080 nm. The shorter lasing wavelength achievable with helical-core bers may well prove to be an important feature of this ber design, because efcient operation at shorter wavelengths will lead to reduced quantum-defect heating and hence will facilitate further power scaling. Note that the 8.5 mm pitch ber laser had a higher efciency than was expected based on the calculated value for the loss for the LP01 mode of 0.6 dB/ m (Fig. 2). These results indicate that the model6 slightly overestimates the propagation losses but is nevertheless a useful design aid for helical-core ber lasers and ampliers. It is also worth mentioning that the origin of the helical loss in the regime in which we are operating is the same as for bend-induced loss in a coiled straight-core ber. Thus to achieve the same level of discrimination against higher-order modes in the straight-core ber as in the 8.5 mm pitch helicalcore ber would require coiling the straight-core ber with a bend radius of 1.8 cm. Clearly, this would be extremely difcult to do with a ber with a large inner-cladding size 500 m selected for efcient incoupling of higher pump power. In summary, we have demonstrated a novel approach to achieving robust and efcient single-mode operation in a cladding-pumped multimode-core ber laser based on the use of a ber with helical-core geometry. In preliminary experiments we obtained 60.4 W of single-mode output from an Yb-doped helical-core ber laser at 1043 nm for 92.6 W of diode pump power, and, at pump powers well above threshold, the slope efciency 84% was comparable with slope efciencies achieved in conventional Yb-doped straight-core ber lasers. This approach should be applicable to much larger diameter bers and hence should facilitate further scaling of single-mode powers from ber lasers and ampliers. This research was funded by the European Ofce of Aerospace Research and Development under contract F61775-01-C0011 and by the Engineering and Physical Sciences Research Council (UK). P. Wangs e-mail address is puw@orc.soton.ac.uk. References
1. Y. Jeong, J. K. Sahu, D. N. Payne, and J. Nilsson, Electron. Lett. 40, 470 (2004). 2. J. Limpert, A. Liem, H. Zellmer, and A. Tunnermann, Electron. Lett. 39, 645 (2003). 3. J. P. Koplow, D. A. V. Kliner, and L. Goldberg, Opt. Lett. 25, 442 (2000). 4. P. Wang, L. J. Cooper, W. A. Clarkson, J. Nilsson, R. B. Williams, J. Sahu, and A. K. Vogel, Electron. Lett. 40, 1325 (2004). 5. W. A. Clarkson, L. J. Cooper, P. Wang, R. B. Williams, and J. K. Sahu, in Advanced Solid-State Photonics, J. Zayhowski, ed., Vol. 83 of OSA Trends in Optics and Photonics Series (Optical Society of America, Washington, D.C., 2003), pp. 261267. 6. D. Marcuse, J. Opt. Soc. Am. 66, 1025 (1976).