Open Access
Add to BibSonomy
Abstract
We comprehensively study the effects of temporal and spectral optimization on single-mode Raman fiber amplifiers. Amplified spontaneous emission sources and ytterbium-doped fiber lasers are employed as seed or pump lasers for comparison, and passive fibers are utilized as gain media. The influences of various parameters of the laser on 2nd order Raman threshold and maximum output power are investigated experimentally, including bandwidth, seed power, wavelength separation between pump and seed laser, and temporal stability. With the 190 m passive fiber, the output power increases from 99.5 W to 142.4 W, corresponding to 43.1% improvement through the optimization of seed laser power, pump wavelength and temporal performance of pump source in this amplifier, which has guidance on the establishment of high-power single-mode Raman fiber amplifiers.
© 2020 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
High-power fiber lasers (HPFLs) have been worldwide studied over the past few decades which are broadly applied in the essential fields of industry and civil departments [1–3]. The performances of fiber lasers and amplifiers advance remarkably and the up-to-date power record has reached 20-kW level for single-mode fiber laser [4]. For continuous-wave (CW) HPFLs, the occurrence of various nonlinear effects are the limitations of power scaling, among which Stimulated Raman scattering (SRS) is one of the main obstacles [5–7]. When more pump laser is injected, there is red shift of the signal laser and the power boosting suspends [8]. From another point of view, SRS can be a direct and effective way to attain the required wavelength by adjusting the span and numbers of Stokes shift in Raman fiber lasers and amplifiers [9,10]. High power lasers have been reported based on gain scheme of SRS, however, the further power scaling is restrained by the threshold of higher order Raman light [11–13]. Consequently, considerable researches have been carried out to raise the threshold of higher order Raman light and boost the output power of signal light [14–16]. As for the Raman fiber lasers (RFLs) based on traditional passive fibers, discrete researches have been conducted so as to mitigate the nonlinear interplay, including optimizing parameters of the gain fibers such as the effective area and the fiber length, the bandwidth of fiber Bragg gratings (FBGs) and the temporal performance of seed laser [17–21]. On the other side, the pump light is the main energy source of the output laser, and its spectral properties will transfer to the output laser to some extent [22]. Previous studies show that the temporal performances of the pump laser have influence on the output characteristics of RFLs, and special pump sources are applied to improve the performance of RFLs, including the amplified spontaneous emission (ASE) which has relatively stable temporal performance [23,24]. However, the specific difference in power conversion and improvement of Raman threshold between traditional Re-doped fiber lasers and the ASE source are less discussed. Based on all these variable parameters, comprehensive and complete research is still in demand for better comparing and understanding the influence of input temporal and spectral characteristics on performance of RFLs.
In this paper, the influence of temporal and spectral performance of pump and seed laser on the output characteristics of high-power Raman amplifier is experimentally discussed. Raman fiber amplifier (RFA) with over 100 W single-mode output power is established as reference object, and both broadband ASE source and ytterbium-doped fiber lasers (YDFLs) are employed as pump and seed sources in the RFA. Various parameters are chosen to alter including the bandwidth, wavelength separation between pump and seed laser, temporal stability and the seed laser power. The influence of different parameters on the maximum output power and the threshold of 2nd order Stokes light in the RFA are discussed in detail.
2. Experimental setup
In order to study the influencing factors on the power scalability of RFA adequately, the temporal and spectral properties of laser sources are considered separately in the experiment, including bandwidth, wavelength separation between pump and seed laser, seed laser power and temporal stability. The configuration of the high-power RFA is illustrated in Fig. 1. The seed and pump sources are combined through a wavelength division multiplexing (WDM), and launched into a piece of passive fiber to achieve the Raman gain and amplification. To study the influences of the input temporal and spectral characteristics on output power and 2nd order Raman suppression in RFL, an ASE source with tunable central wavelength and bandwidth has been employed as the input seed or pump source respectively [25]. The operating central wavelength of ASE source can be tuned over 20 nm (from 1055 nm to 1075 nm), while the full width at half maximum linewidth (FWHM) ranges from 0.6 nm to 20 nm. Besides, the YDFLs at 1070 nm oscillating with and without temporal stability have been adopted as the pump and seed sources as well, comparing together with the non-oscillating ASE source. Accordingly, the YDFLs at 1018 nm or 1120 nm, and 314 m or 190 m GDF, are applied regarding to the 1070 nm-seed or 1070 nm-pump cases separately. All the fibers in the core-pumped RFA have the same core/cladding diameters of 10/125 µm. The output end of the GDF is angle cleaved to prevent the Fresnel reflection, and the amplifier system is water cooled on metal panels for thermal elimination.
Fig. 1. Schematic of the RFA, of which the pump and seed lasers are ASE source or YDFLs at various wavelengths.
Before the experiment, the output temporal performances of the ASE source and the YDFLs at 1070 nm are measured and characterized. The temporal intensity of the output laser is measured in the output end by a high-speed photo-detector with an oscilloscope, whose bandwidths are 5 GHz and 2 GHz, respectively. The sampling time interval is 0.2 ns, and the measurement time window is 40 µs. Two statistical parameters are defined to characterize the temporal stability of lasers including normalized range (Nr) and normalized standard deviation (Nstd), and the formulas are listed as follows:
Among which Imax and Imin are the maximum and minimum value of the temporal intensity respectively. Imean is the average value of the intensity and Std is the standard deviation. In the formula, lower Nr and Nstd indicate smaller fluctuation and better stability of the temporal intensity. Nr and Nstd at maximum output power for each light source are listed in Table 1, and Nr and Nstd of the non-resonant ASE source and resonant YDFL centered at 1070 nm is calculated and illustrated in Fig. 2. In the YDFL with low power generation, the intensity is more stable, while at high power (> 100 W) there is fluctuation in the temporal properties, which may result from the irregular self-pulsing features [26]. Then the temporal performance of the resonant YDFL is optimized by employing an additional weakly reflective FBG outside the cavity [27]. In this way, the regular round trip of the self-mode-locking (SML) pulses in the fiber laser cannot be built up and amplified due to the external feedback FBG, thus the temporal performance improves. Accordingly, the Nr and Nstd apparently decrease after optimization, which indicates the better temporal stability of the YDFL. The Nr and Nstd of the YDFLs lasing at 1018 nm and 1120 nm have similar varying tendency and magnitude with the non-stabilized 1070 nm YDFL. At the same central wavelength, the temporal intensity of ASE source with various FWHM is measured. The Nstd and Nr with different bandwidths all decrease with the increment of output power, and meanwhile remain lower than those of YDFLs. This comparison indicates the generally better temporal stability of ASE source than the YDFL, especially at high power levels. Furthermore, the ASE source with higher bandwidth owns lower Nr and Nstd, which means the reduction of self-pulsing and more stable performance in time domain.
Fig. 2. The statistical parameters of Nr and Nstd of the ASE source with different FWHM bandwidths from 0.6 nm to 20 nm and the conventional YDFLs with and without temporal optimization.
Table 1. Nr and Nstd of various optical sources
The output spectrum of the broadband ASE source and YDFL centered at 1070 nm after the output fiber of WDM are depicted in Figs. 3(a) and (b), respectively, and the output power remains ∼100 W. The FWHM of the ASE source ranges from 0.6 nm to 20 nm, while that of YDFL is about 0.16 nm. As the ASE spectrum broadening, the power density decreases gradually, and the peak intensity with 20 nm bandwidth is about 20 dB lower than that with 0.6 nm bandwidth.
Fig. 3. The output spectrum with the 100 W output power of (a) the broadband ASE source and (b) the YDFL at 1070 nm.
3. Experimental results and discussions
The output power and spectrum of the amplifier are monitored by power meter and optical spectrum analyzer. Typically, the power scaling of the RFA is limited by the 2nd order Raman generation. Thus, in this experiment, the threshold of the 2nd order Raman is regarded as the pump power where the relative intensity of 2nd order Stokes is 20 dB lower than that of the signal laser. Correspondingly, the power scaling is terminated until the 2nd order Raman threshold is reached.
Firstly, the output power of RFA pumped by 0.16-nm-YDFL and ASE source at 1070 nm with various bandwidths has been measured, as shown in Fig. 4(a). The seed power is 20 W at 1120 nm and the length of passive fiber is 314 m. When the pump source changes from the YDFL to the ASE source with 0.6 nm bandwidth, the output power rises from 85 W to 102 W, referring to 20% improvement. By broadening the bandwidth of the ASE further to 20 nm, the output power of the amplifier increases slightly to 115.6 W. In addition, the slope efficiencies of the amplifier by ASE pumping with different bandwidths are similar, while that with YDFL is lower by 20%. Figure 4(b) presents the dependence of 2nd order Raman threshold on pump bandwidth, and the 2nd order Raman threshold improves by 16% regarding to the YDFL and ASE pumping with bandwidths of 0.16 nm and 20 nm respectively. However, the threshold varies only by 3% when the ASE pumping bandwidth increases from 2 nm to 20 nm. Figure 4(c) shows the output spectra with various pump bandwidths, and with wider spectrum of pump light the Stokes conversion is more adequate, meanwhile the influence on the broadening of signal laser is insignificant. Combined with the statistical parameters in Fig. 2, one could find the output power scaling of RFA is mainly related to the pump intensity stability, rather than the pump bandwidth.
Fig. 4. The power of (a) output laser of the amplifier, (b) the corresponding 2nd order Raman threshold, and (c) the output spectrum at maximum power with various pumping bandwidths, including the YDFL (0.16 nm bandwidth) and ASE source (varied bandwidths from 0.6 nm to 20 nm).
Secondly, the temporal stability of the pump laser is improved for comparison, and the Nr and Nstd with and without optimization are also demonstrated in Fig. 2. After majorization of the pump laser, the temporal fluctuation of the laser reduces and the Nr and Nstd decrease from 1.143 to 0.4446 and from 0.1139 to 0.0476, respectively. In Fig. 5(a), the maximum output power of the amplifier before and after optimization of pumping temporal stability are 85.6 W and 98.9 W, respectively, and the slope efficiency of the signal laser has a notable improvement. Under this circumstance, the maximum output power increases by 15.5%. The 2nd order Raman threshold with these pump lasers are presented in Fig. 5(b), and there is improvement of threshold together with the increasing temporal steadiness of pump source.
Fig. 5. The power of (a) scaling characteristics of the amplifier and (b) the 2nd order Raman threshold of amplifier pumped by YDFLs with and without temporal optimization.
The strong intensity fluctuation is easier to trigger the 2nd Raman effect and foster the power amplification process, which limits the power scaling of signal laser. Comparing the different pump sources, it can be concluded that the intensity fluctuation of pump source affects the 2nd Raman threshold and thus the power scaling of the amplifier to some extent.
Except of the pump properties, the influence of the seed laser on output characteristics is also discussed. In this case, the wavelengths of pump and seed laser are altered to 1018 nm and 1070 nm, respectively. With the same fiber length of 314 m the output power are similar to each other, hence the fiber length is shortened to 190 m in order to scale up the overall threshold and better tell apart the amplification process. The output power characteristics and the spectrum at maximum output power of the amplifier are illustrated in Figs. 6(a) and (c) respectively, and the seed laser power remains 10 W centered at 1070 nm. When the bandwidth of the seed laser increases from 0.16 nm (YDFL) to 0.6 nm (ASE source), the maximum output power rises from 109 W to 113.8 W, corresponding to 4.4% improvement. Then with the gradual increment of the bandwidth of ASE source from 0.6 nm to 10 nm, the output power of the amplifier slightly improves to 115.2 W. Furthermore, with broader seed bandwidth, the amplified signal laser owns higher FWHM as well. In Fig. 6(b), the values of 2nd Raman threshold improves by 5.9% from seed of YDFL to ASE with 0.6 nm bandwidth, while the threshold with various bandwidths of ASE seed are all close to each other. The improvement of higher order Stokes suppression is more evident with higher difference of bandwidths, especially higher difference of temporal performance of the seed laser, which is in accordance with Ref. [28], in which the temporal performance of seed is optimized by increasing the bandwidth of the narrow band superfluorescent source seed laser, and higher order Raman threshold rises up. It can be inferred that the temporal performance of seed laser has more influence on the 2nd order Raman suppression of the RFA than the bandwidth of seed, though weaker than the influence of temporal characteristics of pump lasers. The slope efficiencies of the amplifier with different bandwidths of seed laser are similar, the reason is that the passive fiber in the amplifier is fixed, which means the same gain and loss are provided in the amplifier. The amplification results show that the temporal stability and bandwidth of pump laser has more influence than that of seed laser evidently, even under the circumstance that the fiber length of the second passive fiber is optimized.
Fig. 6. The power of (a) output laser of the amplifier, (b) the corresponding 2nd order Raman threshold, and (c) the output spectrum at maximum power with various bandwidths of seed, including the YDFL (0.16 nm bandwidth) and ASE source (varied bandwidths from 0.6 nm to 10 nm).
The next optimized parameter is the separation between pump and seed wavelength, and the wavelength-tunable ASE source with 2-nm bandwidth is applied as pump. The fiber length is fixed to 190 m and the seed power is 20 W. According to the seed wavelength at 1120 nm and the corresponding Stokes shifting in passive fibers, the most matched pump wavelength with peak Raman gain (∼13.2 THz) is ∼ 1067 nm. In the experiment, several wavelengths near 1067 nm are chosen to pump the amplifier, including 1055 nm (16.5 THz), 1057.5 nm (16 THz), 1060 nm (15.2 THz), 1065 nm (13.9 THz), 1070 nm (12.5 THz) and 1075 nm (11.2 THz). The amplification performance of the amplifier and the corresponding higher order Raman threshold with different pump wavelengths are illustrated in Figs. 7(a) and (b). The wavelengths of 1065 nm and 1070 nm are closer to the one with peak Raman gain, and the maximum output power are 110.7 W and 112.1 W, respectively. When the difference between pump wavelength and the one with peak Raman gain is higher, the amplification is boosted and the maximum output power reaches 142.4 W with pump laser at 1055 nm. By contrasting the output power with different pump wavelengths, the power of this amplifier increases by 28.6% at best. In previous studies, the extreme frequency shift between pump and seed wavelengths is discussed, and the wavelength-tunable YDFL is employed as pump source [29]. The conversion efficiencies of different pairs of wavelengths are close and the output signal power is limited by available pump power, however the maximum output power limited by the higher order Raman light is beyond consideration. By contrast, the 2nd order Raman threshold is reached in this system, which is convenient for comparison and discussion. Meanwhile, the highest frequency shift is 16.5 THz (1055 nm), which is out of the range of 10.6 THz ∼15.2 THz in Ref. [29] and realizes the peak output power in our system. The available pump wavelength is limited by the passing waveband of the WDM, and it can be inferred that the amplifier can reach higher output with further wavelength optimization.
Fig. 7. The power of (a) the amplification performance of the amplifier and (b) the corresponding higher order Raman threshold with the same seed wavelength at 1120 nm and varied pump wavelengths near 1067 nm.
After that, the power of seed laser is adjusted to various values for amplification. The pump wavelength is 1070 nm, and the fiber length is 190 m. The central wavelength of the seed laser remains at 1120 nm, and the power are set 10 W, 15 W, and 20 W to test the amplifying effects. The output power scaling properties and the higher order stokes threshold are shown in Figs. 8(a) and (b), respectively. With the increment of seed laser power, the maximum output power presents a descending trend, which is in accordance with Ref. [30]. When the seed laser power is 10 W, 15 W, and 20 W, the output power presents 115.6 W, 104.8 W and 99.5 W, respectively, and the higher order stokes threshold decreases gradually. By altering the seed laser power, the maximal output power of the amplifier increases by 16.2%, which still shows potential in power scaling and improvement of 2nd order Raman threshold with further adjustment of seed power. The decreasing of seed power can enhance the output power of RFAs, meanwhile it should be noticed that to certain laser system there is minimum limit of seed power to ensure the effective guidance of forward stokes light.
Fig. 8. With different seed laser power, (a) the output power scaling properties of the amplifier and (b) the higher order stokes threshold.
From the above experimental results, it can be concluded that aiming at certain laser system considering Raman suppression, the injected seed and pump laser can be optimized to improve the maximum output power and higher order Raman threshold to some extent, containing the better temporal stability of pump and seed laser, the relatively lower seed laser power, higher difference between pump wavelength and corresponding one with peak Raman gain, and broader spectrum of pump and seed laser. In our amplifier system, by contrast the influence of various parameters on the amplification effect with same length of passive fiber, the most effective individual parameter that improves the peak output power is the temporal performance of the pump laser, which differs from the rare-earth-doped fiber amplifier in Ref. [19]. The other parameters all have influence on the output characteristics of the laser system to varying degrees, though the specific experimental conditions and devices restrict the capacity of power scaling of various parameters, such as the passing bandwidth and wavelength of the WDM in this amplifier. Meanwhile, the power scaling of the fiber laser is limited by the available power of ASE source at this time. In general, the output power has 43.1% increment at best, corresponding to pump laser of YDFL at 1070 nm with 0.16-nm bandwidth (99.5 W output power), and pump laser of ASE source at 1055 nm with 2-nm bandwidth (142.4 W output power). In future high-power experimental test, the multiple factors of the single-mode high-power amplifier can be considered comprehensively, including the bandwidth, the central wavelength, temporal stability of pump and seed laser and the seed power to realize higher laser power and better Raman suppression.
4. Conclusions
In this paper, the influence of temporal and spectral performance of pump and seed laser on the output characteristics of Raman amplifier is comprehensively discussed. An all-fiber Raman amplifier is established as reference object, and both broadband ASE source and YDFLs are employed as pump or seed source for contrast. Various parameters are chosen for optimization, and the maximum output power as well as higher order Raman threshold of the RFA is compared and discussed. The experimental results show that aiming at certain single-mode laser system considering higher order Raman suppression, the injected seed and pump laser can be optimized to improve the maximum output power, containing the better temporal stability of pump and seed laser, the relatively lower seed laser power, higher difference between pump wavelength and corresponding one with peak Raman gain, and broader spectrum of pump and seed laser. In this amplifier, the output power has 43.1% increment at best, corresponding to pump laser of YDFL at 1070 nm with 0.16-nm bandwidth (99.5 W output power), and pump of ASE source at 1055 nm with 2-nm bandwidth (142.4 W output power). All these parameters can be considered altogether to enhance the output power and 2nd order Raman threshold, which has guidance on the establishment of high-power fiber lasers.
Funding
National Natural Science Foundation of China (11704409, 61605246, 61911530134); Hunan Provincial Innovation Foundation for Postgraduate (2019RS3017); Huo Yingdong Education Foundation (151062); State Key Laboratory of Pulsed Power Laser Technology (SKL2019ZR01).
Disclosures
The authors declare no conflicts of interest.
References
1. A. Tünnermann, T. Schreiber, and J. Limpert, “Fiber lasers and amplifiers: an ultrafast performance evolution,” Appl. Opt. 49(25), F71–F78 (2010). [CrossRef]
2. D. J. Richardson, J. Nilsson, and W. A. Clarkson, “High power fiber lasers: current status and future perspectives [Invited],” J. Opt. Soc. Am. B 27(11), B63–B92 (2010). [CrossRef]
3. J. Nilsson and D. Payne, “High-Power Fiber Lasers,” Science 332(6032), 921–922 (2011). [CrossRef]
4. B. Shiner, “The Impact of Fiber Laser Technology on the World Wide Material Processing Market,” CLEO paper AF2J.1 (2013).
5. J. W. Dawson, M. J. Messerly, R. J. Beach, M. Y. Shverdin, E. A. Stappaerts, A. K. Sridharan, P. H. Pax, J. E. Heebner, C. W. Siders, and C. Barty, “Analysis of the scalability of diffraction-limited fiber lasers and amplifiers to high average power,” Opt. Express 16(17), 13240–13266 (2008). [CrossRef]
6. J. Zhu, P. Zhou, Y. Ma, X. Xu, and Z. Liu, “Power scaling analysis of tandem-pumped Yb-doped fiber lasers and amplifiers,” Opt. Express 19(19), 18645–18654 (2011). [CrossRef]
7. M. N. Zervas, “Power scaling limits in high power fiber amplifiers due to transverse mode instability, thermal lensing, and fiber mechanical reliability,” Proc. SPIE 10512, 1051205 (2018). [CrossRef]
8. G. P. Agrawal, Nonlinear Fiber Optics (Academic, 2006).
9. T. Yao and J. Nilsson, “835 nm fiber Raman laser pulse pumped by a multimode laser diode at 806 nm,” J. Opt. Soc. Am. B 31(4), 882–888 (2014). [CrossRef]
10. L. Zhang, J. Dong, and Y. Feng, “High-Power and High-Order Random Raman Fiber Lasers,” IEEE J. Sel. Top. Quantum Electron. 24(3), 1–6 (2017). [CrossRef]
11. E. A. Evmenova, S. I. Kablukov, I. N. Nemov, A. A. Wolf, A. V. Dostovalov, V. A. Tyrtyshnyy, D. V. Myasnikov, and S. A. Babin, “High-efficiency LD-pumped all-fiber Raman laser based on a 100 µm core graded-index fiber,” Laser Phys. Lett. 15(9), 95101 (2018). [CrossRef]
12. Y. Glick, Y. Shamir, M. Aviel, Y. Sintov, S. Goldring, N. Shafir, and S. Pearl, “1.2 kW clad pumped Raman all-passive-fiber laser with brightness enhancement,” Opt. Lett. 43(19), 4755–4758 (2018). [CrossRef]
13. Y. Chen, T. Yao, Y. An, J. Song, J. Xu, H. Xiao, J. Leng, and P. Zhou, “2 kW High-efficiency Raman Fiber Amplifier Based on Graded-index Passive Fiber,” ASSL paperATu1A.3 (2019).
14. J. Ji, C. A. Codemard, M. Ibsen, J. K. Sahu, and J. Nilsson, “Analysis of the Conversion to the First Stokes in Cladding-Pumped Fiber Raman Amplifiers,” IEEE J. Sel. Top. Quantum Electron. 15(1), 129–139 (2009). [CrossRef]
15. V. R. Supradeepa and J. W. Nicholson, “Power scaling of high-efficiency 1.5 µm cascaded Raman fiber lasers,” Opt. Lett. 38(14), 2538–2541 (2013). [CrossRef]
16. V. R. Supradeepa, Y. Feng, and J. Nicholson, “Raman fiber lasers,” J. Opt. 19(2), 23001 (2017). [CrossRef]
17. C. Jauregui, J. Limpert, and A. Tünnermann, “On the Raman Threshold of Passive Large Mode Area Fibers,” Proc. SPIE 7914(1), 301–305 (2011). [CrossRef]
18. C. Jauregui, J. Limpert, and A. Tünnermann, “Derivation of Raman treshold formulas for CW double-clad fiber amplifiers,” Opt. Express 17(10), 8476–8490 (2009). [CrossRef]
19. W. Liu, P. Ma, H. Lv, J. Xu, P. Zhou, and Z. Jiang, “General analysis of SRS-limited high-power fiber lasers and design strategy,” Opt. Express 24(23), 26715 (2016). [CrossRef]
20. Y. Miao, P. Ma, W. Liu, H. Zhang, and Z. Pu, “Comprehensive Investigation on the Role of Temporal Property of Pump Laser in a Single-Frequency Raman Fiber Amplifier,” IEEE Photonics J. 10(5), 1–9 (2018). [CrossRef]
21. T. Li, W. Ke, Y. Ma, Y. Sun, and Q. Gao, “Suppression of stimulated Raman scattering in a high-power fiber amplifier by inserting long transmission fibers in a seed laser,” J. Opt. Soc. Am. B 36(6), 1457–1465 (2019). [CrossRef]
22. W. Liu, P. Ma, Y. Miao, H. Wu, P. Zhou, and Z. Jiang, “Intrinsic Mechanism for Spectral Evolution in Single-Frequency Raman Fiber Amplifier,” IEEE J. Sel. Top. Quantum Electron. 24(5), 1–8 (2018). [CrossRef]
23. J. Xu, P. Zhou, J. Leng, J. Wu, and H. Zhang, “Powerful linearly-polarized high-order random fiber laser pumped by broadband amplified spontaneous emission source,” Sci. Rep. 6(1), 35213 (2016). [CrossRef]
24. W. Pan, J. Zhou, X. Yang, and Y. Feng, “Raman dissipative soliton fiber laser pumped by an ASE source,” Opt. Lett. 42(24), 5162–5165 (2017). [CrossRef]
25. J. Ye, J. Xu, Y. Zhang, J. Song, J. Leng, and P. Zhou, “Spectrum-Manipulable Hundred-Watt-Level High-Power Superfluorescent Fiber Source,” J. Lightwave Technol. 37(13), 3113–3118 (2019). [CrossRef]
26. B. Upadhyaya, U. Chakravarty, A. Kuruvilla, S. M. Oak, M. Shenoy, and K. Thyagarajan, “Self-pulsing characteristics of a high-power single transverse mode Yb-doped CW fiber laser,” Opt. Commun. 283(10), 2206–2213 (2010). [CrossRef]
27. H. Xu, M. Jiang, P. Zhou, G. Zhao, and X. Gu, “Elimination of self-mode-locking pulses in high-power continuous-wave Yb-doped fiber lasers with external feedback,” Appl. Opt. 56(32), 9079 (2017). [CrossRef]
28. W. Liu, P. Ma, H. Lv, J. Xu, P. Zhou, and Z. Jiang, “Investigation of stimulated Raman scattering effect in high-power fiber amplifiers seeded by narrow-band filtered superfluorescent source,” Opt. Express 24(8), 8708 (2016). [CrossRef]
29. J. Song, H. Wu, J. Ye, H. Zhang, J. Xu, P. Zhou, and Z. Liu, “Investigation on extreme frequency shift in silica fiber-based high-power Raman fiber laser,” High Power Laser Sci. Eng. 6(2), 21–28 (2018). [CrossRef]
30. 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 (2016). [CrossRef]
