Open Access
Add to BibSonomyAbstract
We report on a 25 W continuous wave narrow linewidth (< 2.3 MHz) 589 nm laser by efficient (> 95%) coherent beam combination of two narrow linewidth (< 1.5 MHz) Raman fiber amplifiers with a Mach-Zehnder interferometer scheme and frequency doubling in an external resonant cavity with an efficiency of 86%. The results demonstrate the narrow linewidth Raman fiber amplifier technology as a promising solution for developing laser for sodium laser guide star adaptive optics.
©2009 Optical Society of America
1. Introduction
Fiber lasers and amplifiers are attractive because they are generally robust and compact. For the same reason, fiber laser based sources for 589 nm sodium laser guide star to be used in astronomical adaptive optics systems have been actively researched in past years [1–10]. There is currently no fiber gain medium lasing directly at 589 nm, so all approaches use fiber laser sources in the near infrared and frequency double or sum to reach 589 nm.
Yb doped silica fiber can lase at 1178 nm, but the gain is very low, suffering from amplified spontaneous emission at shorter wavelength where the gain is much higher. Photonic bandgap fiber laser [8] or Yb fiber laser pumped Yb fiber laser [11] have been studied to overcome the problem, but have not yet demonstrated the necessary performance for high power narrow linewidth operation. Bi-doped fiber lasers at wavelength range of 1100 nm to 1300 nm have been demonstrated [9,12]. But the loss of Bi-doped fibers is still too high for efficient narrow-line amplifier operation. Moreover, the physical mechanism of emission is still not clear. 589 nm lasers can also be generated by sum-frequency of a 1583 nm Er-doped fiber laser and 938 nm Nd-doped fiber laser, but the quasi-three-level nature of 938 nm laser has limited so far its output power [12].
Raman fiber laser or amplifier is another approach that has attracted several groups [1–7]. Raman fiber lasers and amplifiers are usually not considered as ways for generating high power narrow linewidth lasers. Because typically hundreds of meters of fiber are necessary to provide enough gain for high efficient operation, stimulated Brillouin scattering (SBS) is likely to set in and limit the achievable power. However, we had demonstrated multiple watts of narrow linewidth lasers in a simple strongly-pumped non-saturated single mode Raman fiber amplifier [5]. Later, by applying some SBS suppression techniques, a 20.7 W narrow linewidth 1178 nm Raman fiber amplifier was demonstrated [6]. Based on that work, 14.5 W of 589 nm laser has been achieved by external cavity resonant frequency doubling [7]. Although there is still room for scaling up the output power from a single amplifier, coherent beam combination (CBC) has been pursued in our group for future guide star lasers.
In this paper, we report our recent progress on Raman fiber amplifier based guide star laser research. Efficient (> 95%) coherent beam combination of two narrow linewidth (< 1.5 MHz) Raman fiber amplifiers is demonstrated, followed by efficient (86%) frequency doubling in an external resonant cavity, which produces up to 25.4 W continuous wave narrow linewidth (< 2.3 MHz) 589 nm laser. The power level is more than that is required by the 4LGSF project currently carried on at European Southern Observatory [13]. The results demonstrate that the narrow linewidth Raman fiber amplifier technology is promising for developing laser for sodium laser guide star adaptive optics.
2. Raman fiber amplifiers
A schematic diagram of the experimental setup is shown in Fig. 1 . The seed laser is an external cavity diode laser from Toptica Photonics, which can generate up to 36 mW, ~100 KHz bandwidth, linearly polarized fiber coupled laser at 1178nm. It is split by a 50/50 fused fiber coupler, and then independently amplified in two Raman fiber amplifiers.
Fig. 1 A schematic diagram of the optical setup for coherent beam combination and external cavity resonant frequency doubling.
The pumps for the 1178 nm Raman fiber amplifiers are home-built 75 W 1120 nm Raman fiber lasers, which are pumped by Yb fiber lasers at 1070 nm. Two Raman fiber amplifiers can generate up to 17 W and 20 W, respectively. Both are limited by SBS onset. Figure 2 shows a curve of the 1178 nm power versus 1120 nm power for the 20 W amplifier. The amplifiers’ spectra are measured with a Fabry-Perot interferometer of 1 GHz free spectral range. It is found that the FWHM linewidth is less than 1.5 MHz, as shown in Fig. 3 . We cannot resolve the spectrum due to the device resolution limit. The linewidth is narrower than what was reported in our previous works [5–7], because a different and narrower diode seed laser is used (~100 KHz).
Fig. 2 1178 nm power as a function of 1120 nm pump power from a single narrow linewidth Raman fiber amplifier.
Fig. 3 Spectrum of the 1178 nm laser measured with a Fabry-Perot interferometer, which is resolution limited. The inset is a full scan over a free spectral range.
Although non-polarization-maintaining fibers are used in these amplifiers, the laser outputs can be highly linearly polarized with a polarization extinction ratio >25 dB by polarization control with a λ/4 and λ/2 waveplate pair after the amplifier output. This suggests that no significant depolarization effect takes place inside the fiber amplifier. The polarization state may rotate or become elliptical inside the amplifier, but the polarization purity is kept. An automatic polarization control loop is setup by monitoring the power at the other polarization port of the isolator and rotating the waveplates accordingly. The following LINOS Faraday isolators give > 85% transmission and ~30 dB isolation at 1178 nm.
3. Coherent beam combination
Coherent beam combination takes place at a nominal 50/50 splitting mirror. A Mach-Zehnder interferometer is formed by the 50/50 fused fiber coupler, two independent Raman fiber amplifiers, and the 50/50 splitting mirror. A pair of fiber stretchers are inserted between one of the amplifiers and the 50/50 fused fiber coupler. A short stretcher (with 12 m fibers) is used for compensating fast phase jitter; a longer one (60 m) is used for compensating slow relative thermal drift between two amplifiers. Both fiber stretchers are from OptoPhase. By monitoring optical power of either output ports from the splitting mirrors and accordingly phasing the amplifier with stretchers, with a closed loop at 60 KHz, we can force most of the light into one of the ports, if the two laser beams are spatially overlapped and timely coherent.
Spatial overlap of two beams is ensured by mode matching both beams to the resonant doubling cavity. Since mode matching is necessary for efficient resonant cavity frequency doubling anyway, the CBC doesn’t impose on us extra complexity in alignment. We build the two amplifiers with fibers of about the same length. Also the free space path lengths for two beams, from the fiber output to the combing mirror, are set to almost identical, so that we can use lenses of same focus length for collimating two beams, and let two beams share the same mode matching lens as well.
We have obtained maximum 29.5 W coherently combined beam with ~1 W at the dark port. The missing optical power is lost mostly at the two optical isolators. The combining efficiency, which is defined by power at bright port divided by sum of powers at both ports, varies slightly as a function of input powers, but always > 95%, if we raise the powers from both amplifiers evenly. The spectrum of the combined beam is also checked with the Fabry-Perot interferometer. Again it is resolution-limited and FWHM width is < 1.5 MHz.
During the warming-up process, the relative phase between the two amplifiers drifts. When the phase drift exceed the range of the piezo fiber stretcher, the servo control loop will snap back the piezo voltage automatically to the starting point. Figure 4 shows a typical phase drift starting two hours after turning on the laser. In this measurement, the slow fiber stretcher is set to snap back to middle point when approaching about 500 waves. The relative phase drift becomes slower gradually. So free time between two snap back events become longer. When the system is warmed up, we observe no snapback for tens of hours. We believe the warming-up time associated with this relative phase drift can be reduced significantly in a fully engineered laser system.
Fig. 4 Typical relative phase drift between two amplifiers, starting two hours after turning on the laser, calculated from the voltage applied on the slow stretcher.
4. Second harmonic generation
The combined beam is then coupled into an external doubling cavity with a pair of steering mirrors and a mode matching lens. As shown in Fig. 1, the doubling cavity has a bowtie configuration. A 3x3x30 mm3 non-critically phase matched LBO crystal is placed between two curved mirrors (radius 46 mm) for frequency doubling. The phase marching temperature is about 40 °C. The beam waist in the crystal is about 48 μm. The generated yellow beam is extracted through a dichroic mirror.
The cavity is locked using the well-established Pound-Drever-Hall method [14]: the cavity length is adjusted by acting on the piezo-mounted mirror, based on a signal measured using a fast photodiode, D2. The photodiode D1 serves to monitor the intra-cavity IR power. Both the original cavity and associated electronics were purchased as off-the-shelf products from Toptica Photonics AG. However, we have slightly adapted them for high power operation.
Figure 5 shows the generated 589 nm laser power and conversion efficiency as a function of input 1178 nm power, when an incoupling mirror of 90% reflectivity is used. We obtain up to 25.4 W 589 nm laser with a conversion efficiency of 86%. The efficiency is among the highest ever reported at such power level.
Fig. 5 Power and efficiency of frequency doubled 589 nm laser generation as a function of 1178 nm laser power.
We measure the spectrum of the visible output with a visible Fabry Perot interferometer which has less finesse at 589 nm than the infrared device at 1178 nm. Again we cannot resolve the spectrum. As shown in Fig. 6 , the 589 nm laser has a measured FWHM linewidth of less than 2.3 MHz.
Fig. 6 589 nm laser spectrum measured with a Fabry Perot interferometer. It is device resolution limited, which suggests FWHM width is < 2.3 MHz. The inset is a full scan over a free spectral range.
A preliminary longterm test shows the laser is already very stable in a half-engineered experimental setup. In a warmed-up condition, it runs continuously over 15 hours at specified 20W with a power fluctuation of only 170 mW standard deviation and without CBC snapback, as shown in Fig. 7 .
Fig. 7 Power and phase stability test during a 15 hour test in a warmed-up condition. Power fluctuation is only 170 mW standard deviation. Phase drift is within the stretcher range. No CBC snapback event happens.
5. Summary and perspective
In summary, we have coherently combined two narrow linewidth Raman fiber amplifiers with an efficiency > 95%, which have produced up to 29.5 W 1178 nm laser with a linewidth < 1.5 MHz. Up to 25.4 W, <2.3 MHz linewidth, 589 nm laser has been generated by frequency doubling in an external resonant cavity.
In the coming months we will work on power scaling of a single amplifier, and developing polarization maintaining narrow linewidth Raman fiber amplifiers which will allow us in-fiber coherent beam combination. The demonstrated highly efficient coherent beam combination and frequency doubling in external resonant cavity and stability show that the narrow linewidth Raman fiber amplifiers have excellent performance in terms of intensity and phase stability, and robustness. In view of the simplicity of the approach compared to other available technology, the Raman fiber amplifier based technology is proved to be promising for developing laser for sodium guide star adaptive optics. In addition we’d like to note that the technology can be used for producing lasers at other wavelengths that are not easily accessible, since it is based on Raman fiber amplifier.
References and links
1. Y. Feng, S. Huang, A. Shirakawa, and K. Ueda, “589nm light source based on Raman fiber laser,” Jpn. J. Appl. Phys. 43(No. 6A), L722–L724 (2004). [CrossRef]
2. D. Georgiev, V. P. Gapontsev, A. G. Dronov, M. Y. Vyatkin, A. B. Rulkov, S. V. Popov, and J. R. Taylor, “Watts-level frequency doubling of a narrow line linearly polarized Raman fiber laser to 589nm,” Opt. Express 13(18), 6772–6776 (2005). [CrossRef] [PubMed]
3. D. Bonaccini Calia, W. Hackenberg, S. Chernikov, Y. Feng, and L. Taylor, “AFIRE: fiber Raman laser for laser guide star adaptive optics,” Proc. SPIE 6272, 62721M (2006). [CrossRef]
4. P. Dupriez, C. Farrell, M. Ibsen, J. K. Sahu, J. Kim, C. Codemard, Y. Jeong, D. J. Richardson, and J. Nilsson, “1 W average power at 589 nm from a frequency doubled pulsed Raman fïber MOPA system,” Proc. SPIE 6102, 61021G (2006). [CrossRef]
5. Y. Feng, L. Taylor, and D. Bonaccini Calia, “Multiwatts narrow linewidth fiber Raman amplifiers,” Opt. Express 16(15), 10927–10932 (2008). [CrossRef] [PubMed]
6. Y. Feng, L. Taylor, and D. Bonaccini Calia, ““20W CW, 4MHz linewidth Raman fiber amplifier with SHG to 589nm,” Photonics West 2009 post-deadline paper,” Proc. SPIE 7195, xvii–xviii (2009).
7. L. Taylor, Y. Feng, and D. B. Calia, “High power narrowband 589 nm frequency doubled fibre laser source,” Opt. Express 17(17), 14687–14693 (2009). [CrossRef] [PubMed]
8. A. Shirakawa, H. Maruyama, K. Ueda, C. B. Olausson, J. K. Lyngsø, and J. Broeng, “High-power Yb-doped photonic bandgap fiber amplifier at 1150-1200 nm,” Opt. Express 17(2), 447–454 (2009). [CrossRef] [PubMed]
9. A. B. Rulkov, A. A. Ferin, S. V. Popov, J. R. Taylor, I. Razdobreev, L. Bigot, and G. Bouwmans, “Narrow-line, 1178nm CW bismuth-doped fiber laser with 6.4W output for direct frequency doubling,” Opt. Express 15(9), 5473–5476 (2007). [CrossRef] [PubMed]
10. J. Dawson, A. Drobshoff, R. Beach, M. Messerly, S. Payne, A. Brown, D. Pennington, D. Bamford, S. Sharpe, and D. Cook, “Multi-watt 589nm fiber laser source,” Proc. SPIE 6102, 61021F (2006). [CrossRef]
11. A. S. Kurkov, “Oscillation spectral range of Yb-doped fiber lasers,” Laser Phys. Lett. 4(2), 93–102 (2007). [CrossRef]
12. E. M. Dianov, A. V. Shubin, M. A. Melkumov, O. I. Medvedkov, and I. A. Bufetov, “High-power cw bismuth-fiber lasers,” J. Opt. Soc. Am. B 24(8), 1749–1755 (2007). [CrossRef]
13. European Southern Observatory, Four Laser Guide Star Facility (4LGSF), http://www.eso.org/projects/aot/DSM/4LGSF.html.
14. R. W. Drever, J. L. Hall, F. V. Kowalski, J. Hough, G. M. Ford, A. J. Munley, and H. Ward, “Laser phase and frequency stabilization using an optical resonator,” Appl. Phys. B 31(2), 97–105 (1983). [CrossRef]
