Two kinds of hundred-watt-level random distributed feedback Raman fiber have been demonstrated. The optical efficiency can reach to as high as 84.8%. The reported power and efficiency of the random laser is the highest one as we know. We have also demonstrated that the developed random laser can be further used to pump a Ho-doped fiber laser for mid-infrared laser generation. Finally, 23 W 2050 nm laser is achieved. The presented laser can obtain high power output efficiently and conveniently and opens a new direction for high power laser sources at designed wavelength.
© 2015 Optical Society of America
Recently, random distributed feedback (DFB) Raman fiber lasers, which operate via distributed Rayleigh scattering and amplified through the Raman effect, have drawn more and more attentions in generation of random laser [1, 2 ]. It is a new way to generate random laser in fiber with a simple and efficient setup. Combining different kinds of system structures, tunable, multi-wavelength, narrow band, and cascaded random laser output can be realized [3–8 ]. The most obvious feature of the random DFB Raman fiber laser is the lack of cavity mirror, such as fiber Bragg grating (FBG), making the laser systems more simple and compact. Moreover, the gain of such laser comes from stimulated Raman scattering, which hold the properties of small quantum defect in 1μm range, low background spontaneous emission, and absence of photo-darkening effect that is a problem in rare-earth-doped active fibers . The optical efficiency of this random fiber laser (RFL) can reach the value of 70% or higher [10, 11 ]. Consequently, this kind of laser would be an alternative choice to obtain high brightness laser, opening a new direction for high power laser sources at designed wavelength.
For traditional random DFB Raman fiber laser, the passive fiber used to provide feedback and gain is usually several or several-ten kilometers, which has relatively low high-order Raman Stokes threshold and limits the power increasing. The most direct way to obtain high power output is to shorten the fiber length, which would obviously increase the lasing threshold but the output power of first-order Stokes wave can also be increased [10, 12, 13 ]. Another advantage of this kind of laser to achieve high power is that the maximal optical conversion efficiency would increase with the decrease of passive fiber length due to the reduction of the attenuation of pump and Stokes waves and the efficiency can even approach to the quantum limit when the fiber is short enough [11, 12 ].
In this paper, we will make a comprehensive study on the generation of high power random DFB Raman fiber laser with short fiber length. Two kinds of cavities (open and half cavity) will be investigated; the characters of the output power and spectra will be discussed in detail. More than 100 W random laser output at 1150 nm is obtained by both cavities and the optical efficiency can be higher than 84%. The temporal trace also confirm that there is no longitudinal mode beating signal corresponding the cavity length and the random laser can operate in the CW regime without pulsation. For further exploring the application of such high power random fiber laser, the random laser is used to core pump a Ho-doped fiber (HDF) for mid-infrared lasing. Evidences show that the RFL can stably work when pumping the HDF.
2. Experimental setup
The system setup of the high power random fiber laser is shown in Fig. 1 . The RFL is core pumped from one side by a 1090 nm Ytterbium-doped fiber laser (YDFL), which can export a maximum power of 160 W with a power fluctuation of less than 1%. The core diameter of the YDFL is 10 μm and the numerical aperture (NA) of the fiber is about 0.075. For the open cavity [the laser is constructed without cavity mirror, shown in Fig. 1(a)], the pump light is injected into one arm of a wavelength division multiplexer (WDM) and transmits along the followed 320-meter-long passive fiber with 10 μm core. The generated backward Stokes wave (relates to the pump direction) can be detected from another port of the WDM (backward direction). There is no point reflector in both direction of the passive fiber, which is checked by optical time-domain reflectometry, and the feedback is entirely provided by Rayleigh scattering. In the system all the fiber free ends are angle cleaved to avoid end reflection. To form the half cavity [Fig. 1(b)], we replace the WDM in the open cavity with a FBG centered at 1150 nm, whose reflectivity is 99% and 3 dB bandwidth is about 1.5 nm. By this configuration, the feedback is provided both by the point reflection and distributed Rayleigh scattering, which would reduce the threshold . Furthermore, in half cavity, the backward first-order Stokes wave would be totally reflected to the forward direction resulting in a high power output in only one direction.
3. Experimental results and discussion
3.1 Open cavity
Figure 2 shows the power evolution with the pump power in forward and backward directions. The laser generation threshold is as high as 92 W resulting from the short fiber length. Below the threshold the 1090 nm laser increases linearly and has a constant loss of about 6% when passes through the passive fiber. When the pump power is higher than the threshold, the power of Stokes wave grows rapidly in both directions and the slope efficiencies reach to 124% and 75% in forward and backward direction, respectively. Such high slope efficiency around threshold has also been reported in previous studies which is the result of nonlinear nature of the stimulated Raman scattering [10, 12 ]. When pump power is higher than the threshold, the power under threshold can also convent to signal. The power in forward direction is higher than that of the backward in this short-length RFL. The reason is that in our open cavity configuration, the WDM in the backward direction would introduce some reflection or wavelength related loss that is larger than the forward output end, which would result in more power coupling into the forward direction that is similar to the half-open cavity. Such difference is natural for asymmetry structure. However, the parasitic reflections are weak enough to induce oscillation, which can be confirmed by the frequency spectrum shown in Fig. 4. We can also find that the output power of the backward wave depends always linearly on pump power. It can be understood by that the backward wave is mainly amplified in the front part of the fiber (close to the pump source) for this one-arm pumped cavity, so it always exists enough pump energy for the backward transmitted wave for power amplification and the backward Stokes wave would not suffer any saturation before the onset of high-order Stokes wave . At maximal pump power of about 157 W, the total output of the RFL is 124 W with 75 W, 49 W in forward and backward direction, respectively. The optical efficiency of the cavity is close to 79% showing that such short-length cavity can obtain high power efficiently.
The output spectrum exhibits two-peak structure when the power is around the threshold (see Fig. 3 ). As power increasing, the left peak, locating at 1146 nm, dominates the output. The 3 dB bandwidth is about 3.5 nm at the maximal power. There is nearly no difference in the forward and backward spectra, which can be understood by that the feedback is provided by Rayleigh scattering and the cavity has no specific wavelength selector.
In RFL, the random laser arises from thermal noise and amplified by Raman gain in both directions, which is an incoherent laser source essentially. This laser can strongly suppress self-pulsation and is free from random optical spectrum switching compared with rare-earth-doped laser . Figure 4 is the temporal behavior and its corresponding frequency spectrum at the output power of 40 W in backward direction. It can be found that there is no resonant signal corresponding to the mode spacing, which is about 0.32 MHz to a cavity with length of 320 m. In the time domain, the output laser can also work in the continual wave mode and there is no obvious pulse as the power increasing, which is nearly the same as Fig. 4 shown.
3.2 Half cavity
Replacing the WDM in the open cavity with an 1150 nm high reflectivity FBG, the cavity can be switched to a half-open structure, which can also produce random laser and emit from only one side . The output Stokes wave power and residual 1090 nm pump power are recorded in Fig. 5 . The power is obtained by multiplying the total output power and the proportion of corresponding spectral range, which is achieved by integrating the spectrum. The spectra are measured by optical spectrum analyzer via detecting the scattering light from power meter in the output end.The threshold of this cavity is only about 32 W that is about one third of the open cavity. With more power injected, the Stokes wave generates and propagates to forward and backward directions. For the backward Stokes wave, the part of laser within the reflection spectrum of the FBG can be reflect with a relatively higher reflectivity compared with Rayleigh scattering, so the threshold would dramatically reduce. When pump power is higher than 80 W, most of the output is the Stokes wave that centers at 1150 nm and there is nearly no pump light left. Such absolute conversion would result in high system efficiency. At the maximal pump power of 132 W, we achieved random laser output of 112 W with 0.1 W 1090 nm laser remained. The optical efficiency reaches a highest reported value of 84.8% comparing with previous power of several-ten watts and optical efficiency of 70%-80% [11,12 ]. Taking account 9% system loss, which can be estimated from the total output power below the threshold, such efficiency is close to the quantum limit (94.8%) corresponding to the absorbed pump power. The high optical conversion efficiency as well as relatively long fiber length may evidently weaken the thermal burden of a high power fiber laser system, which is an advantage for achieving high power laser at designed wavelength.
We can also find the process of power conversion from the output spectra (Fig. 6(a) ). The peak value of the Stokes wave is 25 dB higher than that of the 1090 nm laser at the maximal power. However, the second-order Stokes wave (around 1214 nm) appears at this power level, which limits the power further increasing. But higher power can be achieved by using shorter length fiber or larger mode area fiber, despite the increase of threshold and the sensitivity to parasitic feedback. The 3 dB bandwidths at the threshold and maximal power are 0.8 nm and 2 nm, respectively. Obviously spectral broadening is observed as a result of the complicated interaction of nonlinear Kerr effect [2, 16,17 ]. At low power level, the bandwidth of random laser is narrow that is defined by the FBG; when the 1150 nm laser power is higher than 90 W, the bandwidth begins to be broader than that of the FBG, which is about 1.5 nm. Theoretically, narrow band random laser can be realized by using narrow bandwidth FBG or inserting special wavelength filter . The temporal behavior and its corresponding frequency spectrum of the half cavity at the full power are shown in Fig. 6(b). The output is so stable due to the free of longitudinal mode beating signal. Consequently, this laser source also can be the seed of a high power laser amplifier.
4. Random laser pumped Ho-doped fiber laser at 2050 nm
Mid-infrared fiber laser, which can be found various applications in medical care, nonlinear frequency conversion, spectroscopy, eye-safe LIDAR, and remote sensing, has been widely studied recently . The developed RFL also can be applied to generate mid-infrared fiber laser. Here we try to demonstrate one of the applications of pumping HDF to generate laser operated at mid-infrared band. HDF has an absorption band in 1150 nm range and experiments pumped by 1150 nm fiber lasers or LDs have been widely reported [18–22 ]. Nevertheless, the output power of Ho-doped fiber laser (HDFL) will be limited due to the low brightness of 1150 nm LD and the small net-gain of YDFL for 1150 nm wavelength. The high-power RFL may solve the problem mentioned above.
The experimental setup of the random laser pumped HDFL is schematically shown in Fig. 7 . The HDF has a core diameter of 25 μm and the central wavelength of the FBG is 2050 nm. The reflectivity of the high reflectivity and output FBG are 99% and 10%, respectively. The total nominal absorption of the HDF at 1150 nm is about 20 dB. The output laser of the random laser is core launched into the HDF via a mode field adapter (MFA). The output end of the HDFL is angle cleaved to avoid unwanted reflection. In the output, a dichroic mirror is used to separate the pump and 2 μm lasers.
Firstly, in order to analyze the influence of the HDFL on the random laser cavity, we pumped the HDFL with the forward output of the open-cavity RFL, and monitored the backward output simultaneously. The power and spectra are nearly the same as that case of the direct output of the open cavity. We also measured the temporal behavior and no difference was found. So, it is believed that the random laser can stably pump the HDFL and the fused points would have little impact on the random laser generation. To obtain higher power, we pumped the HDFL with the half-cavity RFL. We finally achieved about 23 W laser output at 2050 nm. There were still 20 W 1150 nm laser left due to the insufficient pump absorption. The slope efficiency is about 33%, which is comparable to previous demonstrations [18, 22 ]. Spectral broadening can be observed from Fig. 8(b) that has been reported in  and other kinds of fiber laser owing to nonlinear effects [23,24 ]. The HDFL can stably work under the pump of the RFL and can be further optimized to acquire higher power.
In this paper we studied in detail the short-cavity random fiber laser to obtain high power output. For the open cavity, we obtained 75 W and 49 W random laser output at forward and backward direction, respectively. The corresponding optical efficiency was 79%. For the half cavity, we obtained 112 W 1150 nm output from one side at the pump power of 132 W corresponding the optical efficiency of 84.8%. This efficiency was close to the quantum limit if the insert loss of the FBG and passive fiber was took into account. These power level and efficiency are both the highest report as we know. And higher power can be achieved by using shorted fiber or larger mode area fiber opening a new direction for high power laser sources. In addition, the RFLs were further applied to pump a HDF to obtain mid-infrared laser. We had demonstrated that such pump structure could be stable. Power of 23 W at 2050 nm was achieved with the pump of the half-cavity RFL and the slope efficiency is about 33%. It is the first report of random laser pumped HDFL to generate mid-infrared fiber laser. Our results show that such RFL can emit high power laser efficiently and can be further applied to many other cases.
The authors would like to acknowledge the support of the National Natural Science Foundation of China (grant NO. 61322505) and Graduate Student Innovation Foundation of the National University of Defense Technology (Grant No. B130702).
References and links
1. S. K. Turitsyn, S. A. Babin, A. E. El-Taher, P. Harper, D. Churkin, S. I. Kablukov, J. D. Ania-Castañón, V. Karalekas, and E. V. Podivilov, “Random distributed feedback fibre laser,” Nat. Photonics 4(4), 231–235 (2010). [CrossRef]
2. S. K. Turitsyna, S. A. Babinb, D. V. Churkin, I. D. Vatnikc, M. Nikulinc, and E. V. Podivilov, “Random distributed feedback fibre lasers,” Phys. Rep. 542(2), 133–193 (2014). [CrossRef]
3. S. A. Babin, A. E. El-Taher, P. Harper, E. V. Podivilov, and S. K. Turitsyn, “Tunable random fiber laser,” Phys. Rev. A 84(2), 021805 (2011). [CrossRef]
4. A. M. R. Pinto, M. Bravo, M. Fernandez-Vallejo, M. Lopez-Amo, J. Kobelke, and K. Schuster, “Suspended-core fiber Sagnac combined dual-random mirror Raman fiber laser,” Opt. Express 19(12), 11906–11915 (2011). [CrossRef] [PubMed]
6. W. L. Zhang, Y. J. Rao, J. M. Zhu, Z. X. Yang, Z. N. Wang, and X. H. Jia, “Low threshold 2nd-order random lasing of a fiber laser with a half-opened cavity,” Opt. Express 20(13), 14400–14405 (2012). [CrossRef] [PubMed]
7. S. Sugavanam, Z. Yan, V. Kamynin, A. S. Kurkov, L. Zhang, and D. V. Churkin, “Multiwavelength generation in a random distributed feedback fiber laser using an all fiber Lyot filter,” Opt. Express 22(3), 2839–2844 (2014). [CrossRef] [PubMed]
8. H. Tang, W. Zhang, Y. Rao, Y. Zhu, and Z. Wang, “Spectrum-adjustable random lasing in single-mode fiber controlled by a FBG,” Opt. Laser Technol. 57, 100–103 (2014). [CrossRef]
11. H. Zhang, P. Zhou, H. Xiao, and X. Xu, “Efficient Raman fiber laser based on random Rayleigh distributed feedback with record high power,” Laser Phys. Lett. 11(7), 075104 (2014). [CrossRef]
12. Z. Wang, H. Wu, M. Fan, L. Zhang, Y. Rao, W. Zhang, and X. Jia, “High Power Random Fiber Laser With Short Cavity Length: Theoretical and Experimental Investigations,” IEEE J. Sel. Top. Quantum Electron. 21(1), 900506 (2015).
13. I. D. Vatnik, D. V. Churkin, E. V. Podivilov, and S. A. Babin, “High-efficiency generation in a short random fiber laser,” Laser Phys. Lett. 11(7), 075101 (2014). [CrossRef]
14. D. V. Churkin, S. A. Babin, A. E. El-Taher, P. Harper, S. I. Kablukov, V. Karalekas, J. D. Ania-Castanon, E. V. Podivilov, and S. K. Turitsyn, “Raman fiber lasers with a random distributed feedback based on Rayleigh scattering,” Phys. Rev. A 82(3), 033828 (2010). [CrossRef]
15. A. A. Fotiadi, “Random lasers: An incoherent fibre laser,” Nat. Photonics 4(4), 204–205 (2010). [CrossRef]
17. D. V. Churkin, I. V. Kolokolov, E. V. Podivilov, I. D. Vatnik, M. A. Nikulin, S. S. Vergeles, I. S. Terekhov, V. V. Lebedev, G. Falkovich, S. A. Babin, and S. K. Turitsyn, “Wave kinetics of random fibre lasers,” Nat. Commun. 2, 6214 (2015). [CrossRef] [PubMed]
18. S. D. Jackson, F. Bugge, and G. Erbert, “High-power and highly efficient Tm3+-doped silica fiber lasers pumped with diode lasers operating at 1150 nm,” Opt. Lett. 32(19), 2873–2875 (2007). [CrossRef] [PubMed]
19. A. S. Kurkov, E. M. Sholokhov, V. B. Tsvetkov, A. V. Marakulin, L. A. Minashina, O. I. Medvedkov, and A. F. Kosolapov, “Holmium fibre laser with record quantum efficiency,” Quantum Electron. 41(6), 492–494 (2011). [CrossRef]
20. S. D. Jackson, “Towards high-power mid-infrared emission from a fibre laser,” Nat. Photonics 6(7), 423–431 (2012). [CrossRef]
21. A. Hemming, S. Bennetts, N. Simakov, A. Davidson, J. Haub, and A. Carter, “High power operation of cladding pumped holmium-doped silica fibre lasers,” Opt. Express 21(4), 4560–4566 (2013). [CrossRef] [PubMed]
22. X. Wang, P. Zhou, Y. Miao, H. Zhang, H. Xiao, X. Wang, and Z. Liu, “Raman fiber laser-pumped high-power, efficient Ho-doped fiber laser,” J. Opt. Soc. Am. B 31(10), 2476–2479 (2014). [CrossRef]
24. S. A. Babin, D. V. Churkin, A. E. Ismagulov, S. I. Kablukov, and E. V. Podivilov, “Turbulence-induced square-root broadening of the Raman fiber laser output spectrum,” Opt. Lett. 33(6), 633–635 (2008). [CrossRef] [PubMed]