We report on the characteristics of an active fiber with core material made by sintering of Yb-doped silica powders as an alternative to a conventional modified chemical vapor deposition (MCVD) technique. This material provides the possibility to design very large and homogenously rare-earth doped active fiber cores. We have determined a fiber background attenuation of and measured a slope efficiency of 80%. These values are comparable to established fibers made by MCVD technology.
© 2011 Optical Society of America
Within the last ten years, fiber lasers [1, 2] have been established for many high power laser applications such as cutting and welding. The reason for this development is that the Ytterbium (Yb)-doped fiber lasers recently made tremendous progresses and the continuous wave (CW) power of these lasers has been scaled up to the multikilowatt output power level [3, 4]. The main advantages of the fiber laser in comparison to competing laser systems are the relatively low maintenance, the compactness of fiber lasers, the good thermal management of the active fibers, the high optical efficiency , and the excellent beam quality in combination with power scalability [5, 6]. To achieve high powers with high beam quality complex, concepts are used for commercial fiber lasers, e.g., the master oscillator power amplifier setup as a combination of a fiber seed laser and amplifier stages as well as tandem cladding pumping with multiple single mode (SM) fiber lasers [3, 7]. Single fiber concepts for achieving high powers in the multikilowatt range are possible for SM operation of large mode area (LMA) fibers, but also for the multimode operation with very large cores () in order to reduce nonlinear effects.
For the implementation of very large core fibers, no rare-earth doped fused bulk silica material of the required quantity and refractive index homogeneity has been available so far (to our knowledge). Modified chemical vapor deposition (MCVD) in combination with solution doping , the conventional state-of-the-art technique, has its well-known geometrical and homogeneity limitations . The sintering of Yb-doped fused silica powders is a new technique developed by the Institute of Photonic Technology and the Heraeus Quarzglas Company to produce very homogeneous rare-earth doped bulk silica rods, which are used as core material. The producible quantity by far overcomes the geometrical limitations of the MCVD. The production technique was first reported in . Another alternative non-CVD core material was demonstrated in , but with a low Yb concentration and a high background loss of . A slope efficiency of 74% has been reported, which is a surprisingly high value for the given fiber pa rameters regarding results of fiber laser modeling.
In this Letter, we will report on a highly improved core material with very low background loss that facilitates high slope efficiency for laser operation.
Core rod production. The Al/Yb-doped core material was produced by a powder sinter technology. An aqueous suspension of very pure particles is doped in a similar way to the above mentioned solution tech nique of MCVD layers . The dopants are added as a mixed solution of suitable compounds of Al and Yb. The main difference to MCVD is that the particles are homogenously dispersed in a liquid suspension and will not be directly deposited within a tube. This procedure is favorable to get enough material for green body forming with typical weights of 10 to .
After some additional processing and purification steps, the processed doped granulates are sintered into homogeneous and bubble-free Yb-doped bulk silica rods with diameters up to , without higher radial and axial doping level variations or fluctuations of the refractive index level (Figs. 1, 2).
The particle size of the used silica plays an important role in attaining these properties. It is necessary to distinguish between a suitable primary particle size in the range of 10 to and a typical agglomerate size of these particles in a range of 10 to containing open porosities.
Furthermore, the doping levels can be set very precisely, and a high batch-to-batch reproducibility of the desired doping levels and the refractive index is achieved. The refractive index profile of the core rod was measured by a York P104 preform analyzer and is shown in Fig. 1. The standard deviation of the refractive index fluctuations of the profile is . The fluctuations seem to increase in the center of the profile, but this is most likely a numerical artifact caused by the Abel transformation, which has been used for the calculation. In addition, the rounding of the profile edges is also a well-known numerical artifact caused by the Abel transformation. The real refractive index profile of the core is smooth without roundings.
In Fig. 2, the distribution of Yb and Al in the preform of the reported fiber detected by electron microprobe analysis is shown. The doping levels of the core material are per molecular and 15,000 ppm per molecular , respectively. The peaks between pure silica and the doped core are not real but caused by the examination route.
Preform production. The preform was fabricated by a combination of overcladding and the plasma outside deposition (POD) technique. The core rod was stretched and overcladded by a F300 fused silica tube to form the pump cladding with a cladding to the core ratio of 20. This overcladded preform was shaped by grinding in a so-called four-dimensional shape (see inset of Fig. 4), which enhances the pump absorption of the fiber by breaking the circular geometry of the pump cladding in order to avoid helical modes that do not pass through the core. After that, the pump cladding was surrounded by a second highly fluorine-doped synthetic fused silica cladding with a depressed refractive index. This second cladding is produced by the POD process  at Heraeus and features a depressed refractive index, typically in the range of with respect to the undoped silica, which results in a pump NA of 0.27. It is essential to guide the pump light inside the pump cladding by total internal reflection. For high power laser fibers, the F-doped silica cladding is preferred in comparison to an only plastic coating with a depressed refractive index because of its much higher power stability.
Fiber production. For our experiments, we used a fiber with a cladding and a core diameter.
The fiber was drawn in a typical fiber draw tower by heating the preform in a furnace to a temperature of about . For better mechanical stability, the fiber was coated with a commercial acrylate fiber coating of thickness. In Fig. 3, the refractive index profile of the fiber, which was measured using the refracted near-field method with a York S14 refractive index profiler, shows a very smooth step-index core profile and high depression of the F-doped silica cladding.
In Fig. 4, the attenuation spectrum of the fiber core is shown. It was measured by the cutback method using core-matched launching fibers to reduce coupling of the measuring signal only to the core. The background attenuation value at is . This is a very low value, which is comparable to typical attenuation values of Yb-doped MCVD fibers. The absorption band from 800 to is formed by the absorption, whereas the OH absorption is prominent at 1248 and . The inset of Fig. 4 shows the fiber cross section.
Laser experiments. To characterize the laser performance of the fiber, we used a Fabry–Perot setup (Fig. 5). The resonator was formed by a dichroitic mirror at the fiber input, reflecting nearly 100% at the lasing wavelength of the fiber and a perpendicular cleave at the end of the fiber with a reflectivity of about 4%. The fiber used was long and pumped by a pigtailed multimode laser diode at from LASERLINE GmbH. The core diameter of the pigtail is with an aperture of 0.22. Behind the fiber, the residual pump power and the laser power can be separated by a filter.
In Fig. 6, the laser characteristics are shown in a range of up to a output, which was limited by the available pump power. The slope efficiency depending on the absorbed pump power was determined to be 80%. The absorbed pump power was determined by measuring the pump power with cutback method from the whole fiber length back to a length of and calculating the launched pump power by means of the measured pump absorption value. As expected for a free-running laser, we have measured a broad emission spectrum from 1070 to depending on the laser power, which is shown in the inset of Fig. 6.
In conclusion, we were able to show that the sintering of Yb-doped silica powders can be used to fabricate laser fibers providing similar quality compared to conventional MCVD fibers. We have presented a very low attenuation fiber laser with excellent slope efficiency and a broad emission spectrum. Using this technology, an upscaled LMA fiber with similar four-dimensional design and a core was realized in  delivering CW power at emission wavelength. This progress gives rise to the possibility to design active fibers with very high core cladding ratio for high power multimode and LMA fibers. The possibility of additional codoping of the core material with elements like P and Ce opens new perspectives to design very low and homogenous index profiles for an SM operation and to minimize photodark ening simultaneously.
The authors acknowledge financial support from the German Federal Ministry of Education and Research within the research initiative Brilliant Diode Lasers and the Thuringian Ministry of Education, Science and Culture. We also acknowledge financial support from the European Regional Development Fund.
1. H. M. Pask, R. J. Carman, D. C. Hanna, A. C. Tropper, C. J. MacKechnie, P. R. Barber, and J. M. Dawes, IEEE J. Sel. Top. Quantum Electron. 1, 2 (1995). [CrossRef]
2. H. Zellmer, A. Tünnermann, H. Welling, and V. Reichel, in Optical Amplifiers and Their Applications, Vol. 16 of OSA Trends in Optics and Photonics Series (Optical Society of America, 1997), pp. 137–140.
3. D. J. Richardson, J. Nilsson, and W. A. Clarkson, J. Opt. Soc. Am. B 27, B63 (2010). [CrossRef]
4. Y. Jeong, A. J. Boyland, J. K. Sahu, S. Chung, J. Nilsson, and D. N. Payne, J. Opt. Soc. Korea 13, 416 (2009). [CrossRef]
6. J. Limpert, A. Liem, H. Zellmer, and A. Tünnermann, Electron. Lett. 39, 645 (2003). [CrossRef]
7. C. A. Codemard, J. K. Sahu, and J. Nilsson, IEEE J. Quantum Electron. 46, 1860 (2010). [CrossRef]
8. J. E. Townsend, S. B. Poole, and D. N. Payne, Electron. Lett. 23, 329 (1987). [CrossRef]
10. A. Langner, G. Schötz, M. Such, T. Kayser, V. Reichel, S. Grimm, J. Kirchhof, V. Krause, and G. Rehmann, Proc. SPIE 6873, 687311 (2008). [CrossRef]
12. A. Langner, M. Such, G. Schötz, S. Grimm, F. Just, M. Leich, C. Mühlig, J. Kobelke, A. Schwuchow, O. Mehl, O. Strauch, R. Niedrig, B. Wedel, G. Rehmann, and V. Krause, Proc. SPIE 7914, 79141U (2011). [CrossRef]