Open Access
Add to BibSonomy
Abstract
We report an ytterbium-doped 20/400 µm double cladding fiber with an internally modified circular inner cladding. Through four fluorine-doped (F-doped) low refractive index quartz rods embedded into the inner cladding, the cladding pump absorption is considerably increased by 1.5 times. It is found that stimulated Raman scattering (SRS) threshold improves after using shorter fiber lengths. More than 2.1 kW laser output with good beam quality (M2 = 1.38) has been obtained with a suitable pump power injected, and the slope efficiency of the all-fiber laser oscillator is about 75.4%, when pumped at 915 nm. The present results suggest that low refractive index quartz rods embedded into the inner cladding proves to be an effective way for enhanced cladding pump absorption even with a circular inner cladding, which facilitates fiber splicing and provides a novel and robust fiber design of industry grade ytterbium-doped double cladding fibers for high power fiber laser applications.
© 2019 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
In recent years, fiber lasers based on large mode area double cladding fibers (DCF) have been widely used in industrial processing, military, medical care, scientific research and other fields [1–5] due to their improved output power, excellent beam quality, high efficiency, good heat dissipation, small size, and portability [1,2,5–7]. With the development of high-power laser diode sources, double cladding fibers, pump coupling techniques and so on [1,3,6,8–11], power scaling of fiber lasers has shown a remarkable increase, reaching ∼20 kW [12] laser output with single mode operation and 100 kW [13] laser output under multimode operation. Fiber lasers have become the leader of industrial lasers [8]. However, with the ever increasing demands for high output power and high beam quality, the range of limitations for the development of industrial high power fiber laser also extends. More specifically, long-term stability with photodarkening effects [14,15], short-term stability with transverse modal instabilities (TMI) [14,16] and nonlinear effects such as stimulated Raman scattering (SRS) and stimulated Brillouin scattering (SBS) [12,15,16] are among the main limiting factors that need to be carefully dealt with.
Increasing the cladding pump absorption hence shortening the length of DCF is considered to be an effective way to increase the nonlinear threshold of fiber lasers. For the same material compositions, a series of schemes with broken circular symmetry of fiber cross sections have been proposed to enhance the cladding pump absorption efficiency of DCF, including hexagonal [10], octagonal [17], decagon [2], circular offset [11], rectangular [11,18], stadium [4], D-shaped [19] and spiral cladding [20]. These fiber designs improve the cladding pump absorption by eliminating and/or increasing the intersection of cladding light and doped core. DCF fiber with an octagonal inner cladding is currently the dominant fiber design for commercially available active fibers. However, optical fibers with non-circular shaped cladding require additional complicated manufacturing processes and grinding lead to additional core/cladding offset. Moreover, splicing between shaped active fibers and circular passive fibers is less straightforward and may result in added splicing-losses [5], because the splicing quality relies on a more difficult core instead of cladding alignment that affected by resolution and edge angle. It is important to note that any excessive core/cladding offset and the splicing-loss increase will reduce the high power output stability and output beam quality. Incorporating boron-doped stress elements into the inner cladding is considered to be an effective method to improve pump absorption of polarization maintaining DCF [15], nevertheless, the preparing process of boron rod is complex, quantity limited and position required for preform inner cladding which may hinder its further development.
In this paper, for the purpose of reducing the splicing loss between conventional non-circular active fibers and circular passive fibers, a circular active fiber design is proposed. In order to overcome the disadvantage of helical light that does not intersect the doped core in a circular fiber, the inner cladding is internally modified by inserting mode scrambling elements. Particularly in this study, four low refractive index rods are embedded in the inner cladding. This method can reduce the effective reflection area, increase the reflection frequency and shorten the reflection path of pump light in the inner cladding, and then improve the cladding pump absorption of the designed optical fibers. This design avoids the concentricity deviation caused by conventional polishing method, reduces the fusion loss between conventional special shaped fibers and passive fibers, and improves the cladding pump absorption of double cladding fibers. In this study, ytterbium-doped double cladding fiber is prepared by Modified Chemical Vapor Deposition (MCVD) technology in combination with a low refractive index quartz rods embedded into the inner cladding process. As comparison, the octagonal 20/400 µm fiber is also prepared by a mechanical grinding method. The cladding pump absorption of the F-doped low refractive index quartz rods embedded (LCA) 20/400 µm fiber and octagonal 20/400 µm fiber are measured respectively. In addition, the SRS and power stability are tested by a 2.1 kW all-fiber oscillator. This work may provide new insights into the smart design and preparation of new high-cladding pump absorption ytterbium-doped double cladding fiber.
2. Experimental details
2.1 Preform and fiber fabrication
The Yb/Al co-doped phosphosilicate core preform is prepared by MCVD and conventional solution doping process. Firstly, the inner wall of a deposition tube is deposited with a highly silica porous soot layer with small quantities of other dopants. Secondly, the soot layer is immersed in a Yb-containing alcoholic solution, followed by drying and sintering steps. Finally, the deposited tube is collapsed into a solid rod at higher temperatures.
Circular 20/400 µm LCA DCF: After diameter and refractive index testing (the numerical aperture of the preform is ∼0.063), the core preform prepared in the previous step is jacketed with a suitable silica tube to achieve the design core/cladding ratio (1/20). Four holes are drilled symmetrically in the cladding area of the preform, and F-doped low refractive index quartz rods are inserted in the holes separately. Then the preform assembly is drawn into a 20/400 µm fiber with a circular cladding at temperatures of 2050-2250°C and coated with a low index polymer which provides a pump NA of 0.46.
Octagonal 20/400 µm DCF: As comparison, the preform of an octagonal 20/400 µm DCF is also prepared under the identical MCVD deposit and solution doping conditions, achieving the same rare earth concentration and numerical aperture in the core. After being jacketed with a suitable silica tube and grinded into an octagonal shape with a 1/20 core/cladding ratio, the preform is drawn into a 20/400 µm diameter fibers with octagonal cladding and coated with a low index polymer which provides a pump NA of 0.46.
2.2 Laser performance testing
Laser performances of the Circular 20/400 µm LCA DCF and Octagonal 20/400 µm DCF are tested by an all-fiber oscillator configuration, which is reported in our previous work [1] and has a small difference, as showed in Fig. 1. In brief, the laser cavity consists of the fiber Bragg gratings (FBGs) and large-mode-area (LMA) gain fiber, and is pumped by ten pump lasers (output power of a single pump laser is ∼280 W) with central wavelength of 915 nm. The pump delivery fibers of semiconductor lasers are 200/220 µm with NA 0.22. They are fused with the signal fiber of a (6 + 1) x1 fiber coupler directly. The fibers of the (6 + 1) x1 coupler output, high-reflection (HR) FBG and low-reflection (LR) FBG are made of 20/400 µm passive fibers. The reflectivity of HR FBG is about 99.5% and about 10% for the LR FBG is with the central wavelength of 1080 nm. The output fibers of the fiber couplers are fused with the tail fiber of FBGs. Rare earth doped fibers are fused between HR and LR FBGs, the gain fiber length is determined by the 18 dB total cladding pumping absorption at 915 nm. Gain fiber length of Circular 20/400 µm LCA DCF and Octagonal 20/400 µm DCF are 30 m and 45 m, respectively. Leaked signal light and unabsorbed pump light is removed by a cladding mode stripper (CMS). The fiber laser is output by a quartz block holder (QBH) and detected by a power meter at the end. The splice spots of the fiber components take advantage of low fuse-loss as 0.15 dB and are coated with a low index polymer which provides a pump NA of 0.46 in the all-fiber oscillator system. The laser output efficiency, signal beam quality and output power stability are also recorded for further discussion.
3. Experimental results and discussion
3.1 Fiber characteristics
Figure 2(a) shows the cross-section of a Circular 20/400 µm LCA DCF with a core of 20.1 µm and a cladding of 401.2 µm in diameters. Unlike conventional commercial Octagonal 20/400 µm fiber, the cladding of fabricated Circular 20/400 µm LCA DCF has a circular structure with four embedded regions. These embedded regions are situated symmetrically around the doped core, with a diameter of 118.9 µm. The core/cladding offset of Circular 20/400 µm LCA DCF is 0.7 µm. Different from the conventional grinding resulting in the concentricity variation of optical fibers, it shows that there is no deterioration of concentricity with drilling and inserting rods in cladding. Circular structure of optical fiber cladding and better core/cladding offset control of Circular 20/400 µm LCA DCF can help to reduce the splicing loss and improve the output stability of high power fiber laser.
For further study and analysis, the measured 2D refractive index profile of the Circular 20/400 µm LCA DCF is given in Fig. 2(b). It is clear found that four rods are filled in inner cladding with four directions and this result is consistent with the cross-section test results. The refractive index of the four F-doped quartz rods embedded in the drill hole is obviously lower than that of the silica cladding, it can reduce the light entering the hole better. The refractive index of the interface between the F-doped quartz rod and the cladding is lower than that of the middle of the F-doped quartz rod, which is due to the evaporation of fluorine at the interface during drawing process of the combined preform in high temperature. Fiber core is not found in the refractive index profile because it is difficult to detect the light passing through the four embedded F-doped low refractive index quartz rods under side injection.
The cladding pump absorption is measured by an optical spectrum analyzer (OSA) with a white light source. A length of test fiber is wound onto a mandrel with a fixed coiling diameter. The pump absorption coefficient is determined by a standard cutback method. The measured pump absorption spectra of (a) Octagonal 20/400 µm DCF and (b) Circular 20/400 µm LCA DCF are shown in Fig. 3. Pump absorption of Octagonal 20/400 µm fiber at 915 nm and 976 nm are 0.40 dB/m and 1.36 dB/m respectively, similar to commercially available 20/400 um active fibers. Meanwhile, the pump absorption of Circular 20/400 µm LCA DCF at 915 nm and 976 nm are 0.60 dB/m and 2.27 dB/m respectively. Characteristic absorption peaks at 915 nm and 976 nm indicate successful incorporation of Yb ions in the fiber core [21,22].
Circular 20/400 µm LCA DCF shows a significant increase in pump absorption which is larger than that of homemade Octagonal 20/400 µm fiber by a factor of 1.50 (915 nm) and 1.67 (976 nm). For conventional fiber, the fraction of rays that encounter the core gives an estimation of pump absorption efficiency that roughly depends on the ratio of ${A_{core}}/{A_{clad}}$ [4,7,9] with the assumption that the ray evolution is a random process. The ${A_{core}}$ and ${A_{clad}}$ are the areas of the core and inner cladding, respectively. The embedded low index regions in the inner cladding essentially reduce the pump light guiding area relative to the core, although the refractive index is not low enough to significantly impact the pump NA relative to the outer cladding. Therefore, we propose a modified model to assess the cladding pump absorption of Circular 20/400 µm LCA DCF, which roughly depends on the ratio of ${A_{core}}/({A_{clad}} - N \cdot PI \cdot {(D/2)^2})$, where N and D are the number and the diameter of the embedded quartz rods, respectively. ${A_{core}}$ is almost constant in the 20/400 µm fibers, ${A_{clad}}$ of the typical 401.2-diameter circular optical fiber is 0.126 mm2, ${A_{clad}} - N \cdot PI \cdot {(D/2)^2}$ of the 20/401.2 µm fiber by F-doped low refractive index quartz rods embedded into the inner cladding calculated is 0.082 mm2, the effective pump area of Circular 20/400 µm LCA DCF is reduced by a factor of 1.54 (comparing with 401.2-diameter circular optical fiber) which is the same as enhancement coefficient of cladding pump absorption (comparing with Octagonal 20/400 µm DCF) basically. Obviously, octagonal cladding can help to avoid the helix pump light and therefore increase pump absorption and slope efficiency for laser output compared with circular optical fiber. Therefore, compared with circular cladding fibers, the pump absorption enhancement factor of Circular 20/400 µm LCA DCF is obviously greater than 1.5 times.
The core background attenuation of the Circular 20/400 µm LCA DCF is evaluated by a Photo Kinetics Optical Fiber Analysis System PK2500 from 1100 nm to 1600 nm. This measurement is done with a standard cutback method with a free-space launched light source, as shown in Fig. 4. The values of the core background attenuation at 1200 and 1300 nm are 6.5 and 4.3 dB/km, respectively. Low background loss of rare earth doped optical fibers is beneficial to the use in high power fiber lasers. As can be seen from the Fig. 4, bending loss occurs slightly when the wavelength is higher than 1400 nm, which is due to the low core numerical aperture of the optical fiber. For comparison, the homemade Octagonal 20/400 µm DCF shows similar core loss results under the above core attenuation measure conditions.
3.2 Laser performance and power stability
The all-fiber laser oscillator system and the pump lasers are cooled by a water chiller during the experiments and long-term working. Octagonal 20/400 µm DCF fabricated by ourselves has been tested by the laser output spectrum is centered at 1080 nm. Nonlinearity such as SRS begins to appear at 1.6 kW, and 1.7 kW which becomes more serious. The signal light of the fiber oscillator is analyzed by an optical spectrum analyzer (OSA) at the same time as shown in Fig. 5, spectrum of the SRS light observed from the output beam of the all-fiber laser oscillator. The laser output spectrum is centered at 1080 nm and the peak is centered at 1133.5 nm which is generated by SRS, it shows strong SRS effect. The Circular 20/400 µm LCA DCF is also tested by the all-fiber laser oscillator system, output power of the all-fiber laser oscillator with a suitable 915 nm pump power injected as shown in Fig. 6, the maximum laser power is 2.1 kW at 1080 nm and linearly-fitted slope efficiency is 75.4% (Fig. 7), and there is no SRS effect as shown in Fig. 6. Its description of the Circular 20/400 µm LCA DCF has a higher SRS threshold.
This Raman threshold is usually carried out using the conventional formula published by R.G. Smith in 1972: Pth = 16Aeff / gR Leff [3,23]. The SRS threshold is mainly dependent on the fiber characteristics such as fiber mode-field-area Aeff and the fiber effective length Leff, and gR is the peak Raman gain coefficient (gR = 1 × 10*13 m/W in fused silica at a pump wavelength of 1 µm). In this paper, in order to ensure that the pump light from pump lasers is adequately absorbed by the fiber core, the total cladding pump absorption is calculated to be 18 dB. The length of Octagonal 20/400 µm DCF is calculated to be 45 meters (As shown in Fig. 3, cladding pump absorption at 915 nm is 0.4 dB/m), and the Circular 20/400 µm LCA DCF is calculated to be 30 meters (As shown in Fig. 3, cladding pump absorption at 915 nm is 0.6 dB/m) with 33% shorter than before. Octagonal 20/400 µm DCF begins to show SRS in 1.6 kW, by classical Raman threshold formula calculation, the SRS threshold of Circular 20/400 µm LCA DCF can be increased to 2.4 kW by shortening the length of the gain fiber. We have achieved stable single mode 2.1 kW power output without non-linear Raman effect, and SRS effect appears in the all-fiber laser oscillator gradually when the power exceeds 2.1 kW, there may be other reasons.
The beam quality of signal light does not decrease significantly at high power level. Signal beam quality (M2) of the fiber oscillator at 2.1 kW is analyzed by an Ophir’s BeamSquared system at the same time as shown in Fig. 8, and giving M2 = 1.36 and 1.38 for X and Y directions respectively, it shows excellent beam quality of the signal light in high power level. The excellent beam quality signal light in the 2.1 kW of the all-fiber laser oscillator system is attributed to the high SRS threshold, low splicing-loss of fibers and elimination of higher-order modes by appropriate bending of optical fibers (bending diameter is ∼130 mm) of the oscillator [24]. The laser stability of Circular 20/400 µm LCA DCF using all-fiber laser oscillator at 2.1 kW as shown in Fig. 9. After 60 hours of full power working, power stability is maintained at 2.1 kW, no significant signal power decrease is observed (Power reduction < 1.0%). During the experiment, there is no mode instability phenomenon of the signal light which shows long term stability. The beam quality of laser from Octagonal 20/400 µm DCF is also analyzed by an Ophir’s BeamSquared system, which is 1.40 under the situation that the output power is 1.6 kW.
3.3 Comparison with PLMA fiber
It is worthy to note that in polarization-maintaining large mode area (PLMA) fibers, the stress applying members are typically boron-doped and have a refractive index lower than silica glass. Therefore, the cladding pump absorption of PLMA fiber is naturally higher than the corresponding non-PM version for the similar reasons. However, in our study the emphasis is placed on taking the full advantage of the internally modified inner cladding to enhance the cladding pump absorption without significantly influence the polarization states of the signal light. We believe that inserting the low refractive index regions in four directions around the core strikes a delicate balance between minimizing the impact on the polarization state and the manufacturing costs. Using fluorine doped silica rods further reduces that impact, because only small amount of fluorine doping is needed to achieve low refractive index and fluorine doped silica rods are readily commercially available.
4. Conclusion
In conclusion, reducing the effective area of the circular inner cladding is demonstrated to be a new and efficient method for the enhanced cladding pump absorption of DCF. Through low refractive index quartz rods embedded into the inner cladding, the result of the experiments shows that the cladding pump absorption of the fabricated fibers is improved by 1.5 times compared with conventional Octagonal optical fibers. The SRS threshold can be improved by using this type of optical fiber, more than 2.1 kW signal power with good beam quality (M2 = 1.38) of the all-fiber laser oscillator is obtained by using a suitable pump power injected. This work may provide a new strategy for the design and development of high-cladding pump absorption DCF. Influence of the embedded rods composition, numerical aperture, size, number and position of the double cladding fiber will be studied in future works.
Funding
National Key Research and Development Program of China (2016YFB0402200, 2017YFB1104400).
Disclosures
The authors declare that there are no conflicts of interest related to this article.
References
1. J. Wang, D. Yan, S. Xiong, B. Huang, and C. Li, “Influence of the fiber Bragg gratings with different reflective bandwidths in high power all-fiber laser oscillator,” Opt. Commun. 383, 355–358 (2017). [CrossRef]
2. D. Young and C. Roychoudhuri, “Results and comparison of a cladding pumped fiber simulation using a decagon-shaped fiber,” Opt. Express 11(7), 830–837 (2003). [CrossRef]
3. J. Wang, D. Yan, S. Xiong, B. Huang, and C. Li, “High power all-fiber amplifier with different seed power injection,” Opt. Express 24(13), 14463–14469 (2016). [CrossRef]
4. P. Koska, P. Peterka, and V. Doya, “Numerical Modeling of Pump Absorption in Coiled and Twisted Double-Clad Fibers,” IEEE J. Sel. Top. Quantum Electron. 22(2), 55–62 (2016). [CrossRef]
5. S. Jeong, S. Ju, and W. T. Han, “Effect of bending on emission characteristics of large core Yb/Al doped optical fiber with depressed cladding structure,” Laser Phys. 29(2), 025102 (2019). [CrossRef]
6. N. A. Mortensen, “Air-clad fibers: pump absorption assisted by chaotic wave dynamics?” Opt. Express 15(14), 8988–8996 (2007). [CrossRef]
7. V. Doya, O. Legrand, and F. Mortessagne, “Optimized absorption in a chaotic double-clad fiber amplifier,” Opt. Lett. 26(12), 872–874 (2001). [CrossRef]
8. J. Nilsson and D. N. Payne, “High-Power Fiber Lasers,” Science 332(6032), 921–922 (2011). [CrossRef]
9. V. Filippov, Y. Chamorovskii, J. Kerttula, K. Golant, M. Pessa, and O. G. Okhotnikov, “Double clad tapered fiber for high power applications,” Opt. Express 16(3), 1929–1944 (2008). [CrossRef]
10. L. F. Wang, D. B. He, L. L. Hu, and D. P. Chen, “Nd3+-doped soft glass double-clad fibers with a hexagonal inner cladding,” Laser Phys. 25(4), 045108 (2015). [CrossRef]
11. A. P. Liu and K. Ueda, “The absorption characteristics of circular, offset, and rectangular double-clad fibers,” Opt. Commun. 132(5-6), 511–518 (1996). [CrossRef]
12. B. Shiner, “The Impact of Fiber Laser Technology on the World Wide Material Processing Market,” CLEO: Applications and Technology, AF2J.1 (2013).
13. E. A. Shcherbakov, V. V. Fomin, A. A. Abramov, A. A. Ferin, D. V. Mochalov, V. P. E. D. H. G. Gapontsev, and P. Moulton, “Industrial grade 100 kW power CW fiber laser,” Advanced Solid-State Lasers Congress, ATh4A.2 (2013).
14. C. G. Ye, L. Petit, J. J. Koponen, I. N. Hu, and A. Galvanauskas, “Short-Term and Long-Term Stability in Ytterbium-Doped High-Power Fiber Lasers and Amplifiers,” IEEE J. Sel. Top. Quantum Electron. 20(5), 188–199 (2014). [CrossRef]
15. M. N. Zervas and C. A. Codemard, “High Power Fiber Lasers: A Review,” IEEE J. Sel. Top. Quantum Electron. 20(5), 219–241 (2014). [CrossRef]
16. C. Jauregui, J. Limpert, and A. Tunnermann, “High-power fibre lasers,” Nat. Photonics 7(11), 861–867 (2013). [CrossRef]
17. L. Liao, Y. B. Wang, Y. B. Xing, H. Q. Li, J. G. Peng, N. L. Dai, and J. Y. Li, “Fabrication, measurement, and application of 20/400 Yb-doped fiber,” Appl. Opt. 54(21), 6516–6520 (2015). [CrossRef]
18. A. Liu, J. Song, K. Kamatani, and K. Ueda, “Rectangular double-clad fibre laser with two-end-bundled pump,” Electron. Lett. 32(18), 1673–1674 (1996). [CrossRef]
19. R. Narro-Garcia, M. Arronte, J. Guerrero-Contreras, and E. Rodriguez, “Study of the pump absorption efficiency in D-shaped double clad optical fiber,” Opt. Appl. 42(3), 587–596 (2012). [CrossRef]
20. D. Kouznetsov and J. V. Moloney, “Efficiency of pump absorption in double-clad fiber amplifiers. II. Broken circular symmetry,” J. Opt. Soc. Am. B 19(6), 1259–1263 (2002). [CrossRef]
21. D. J. Richardson, J. Nilsson, and W. A. Clarkson, “High power fiber lasers: current status and future perspectives,” J. Opt. Soc. Am. B 27(11), B63–B92 (2010). [CrossRef]
22. J. Kirchhof, S. Unger, A. Schwuchow, S. Grimm, and V. Reichel, “Materials for high-power fiber lasers,” J. Non-Cryst. Solids 352(23-25), 2399–2403 (2006). [CrossRef]
23. R. G. Smith, “Optical power handling capacity of low loss optical fibers as determined by stimulated Raman and brillouin scattering,” Appl. Opt. 11(11), 2489–2494 (1972). [CrossRef]
24. J. M. Wang, C. Li, and D. P. Yan, “High power composite cavity fiber laser oscillator at 1120 nm,” Opt. Commun. 405, 318–322 (2017). [CrossRef]
