Open Access
Add to BibSonomyAbstract
1178 nm single-frequency amplification by Yb doped photonic bandgap fiber has been demonstrated. 24.6 W output power and 12 dB gain were obtained without parasitic lasing and also stimulated Brillouin scattering. 1.8 dB suppression of Brillouin gain by an acoustic antiguiding effect has been found in the Yb doped photonic bandgap fiber.
©2012 Optical Society of America
1. Introduction
Recently rare-earth doped solid-core photonic bandgap fibers (PBGFs) are attracting a great deal of attention. They are showing a promising solution for new wavelength sources which have never been possible by conventional step-index type active fibers [1–8]. In PBGFs, high-index inclusions arranged periodically in the cladding create photonic bandgap guidance in the low-index rare-earth doped signal core. Since the guidance is obtained only in the spectrally-limited photonic bandgaps, huge attenuation is featured in the outside regions of the transmission windows. Therefore undesired gain that creates strong amplified spontaneous emission (ASE) can be eliminated and the gain profile can be optimized for a specific wavelength operation like the long wavelength edge (1150-1200 nm) of the broad gain band of an Yb doped active fiber. We have been investigating the Yb doped PBGF (Yb-PBGF) sources operating at 1178 nm [4]. We reported a fiber amplifier with as high as 167 W output power [6,7], as well as a fiber oscillator with an output power of 53.6 W [8] at this wavelength.
Since the most efforts of these PBGF researches [1–8] have been made on the subject of power extraction, all of the PBGF lasers and amplifiers reported have been limited to broad linewidth operation in which stimulated Brillouin scattering (SBS) can be neglected, because its gain spectrum is much narrower than the laser linewidths. However, single-frequency or narrow linewidth sources are desired for various applications: coherent beam-combining, frequency conversion by enhancement cavity, high resolution spectroscopy, and so on. We are focusing on the 1178 nm Yb-PBGF sources for laser guide star adaptive optics, in which narrow linewidth and high power 589 nm light is required to effectively illuminate sodium. This narrow linewidth 1178 nm sources can be efficiently converted to 589 nm by frequency doubling technology [9]. However, as we mentioned, it has still not been clarified if PBGFs will work in narrow linewidth, single-frequency operation. Here we present a single-frequency Yb-PBGF source based on a master-oscillator power amplifier configuration and a characterization of the Yb-PBGF in the single-frequency operation by measuring the Brillouin gain spectrum.
2. Properties of the photonic bandgap fiber
The Yb-PBGF for single frequency amplifier is similar to those reported in [6,7]. Microscope images of the fiber cross section are shown in Fig. 1(a) and 1(b). The signal core consists of an ytterbium-doped rod and is index-matched to the silica background and surrounded by a PBG cladding structure composed of eight rings of high-index germanium-doped rods (Δnmax ~2.3%) with a pitch of 10.1 μm. In Fig. 1(a) the lighter regions are the germanium-doped rods and the two darker regions are boron-doped rods for inducing a birefringence on the order of 10−4 in the fiber. The low refractive index of the boron rods (Δn ~-0.4%) results in confinement by total internal reflection instead of PBG guiding in this direction. The mode field diameter derived from a near field image (Fig. 1(c)) is ~10.3 μm at 1178 nm. The fiber is a double clad fiber with an airclad and the cladding diameter is ~220 μm (Fig. 1(b)), the numerical aperture of the pump cladding is 0.6, the pump absorption is ~1.1 dB/m at 976 nm and the propagation loss of the core mode is ~0.03 dB/m at 1178 nm.
Fig. 1 Microscope images of (a) the photonic bandgap structure around the core and (b) the pump-cladding structure. (c) CCD image of the near field at 1178 nm.
3. High power 1178 nm single-frequency fiber amplifier
The configuration of the single-frequency Yb-PBGF amplifier is shown in Fig. 2 . The 1178 nm seed source was an external-cavity semiconductor laser diode (EC-LD). The output was incident into the fiber Raman amplifier (FRA) passing through a free space isolator and an anamorphic prism pair for beam shaping. The isolator also works as a monitor of the SBS power from the amplifier. The FRA was incorporated to increase the seed power above the saturation power of the Yb-PBGF derived from the relation Psat = (hνs /σesτ) · AMFA, where h is Plank’s constant, νs is the signal frequency, σes is the emission cross section at the signal wavelength λs, τ is lifetime, and AMFD is the mode field area at λs. By use of σes = 0.01 × 1020 cm−2 and τ = 0.84 ms [7] the saturation power is calculated to be ~1.7 W at 1178 nm. The power amplifier is based on the Yb-PBGF pumped by a 976 nm fiber coupled LD.
Fig. 2 Experimental setup of 1178 nm single-frequency amplification. ISO: isolator; YFL: Yb-doped fiber laser; BS: beam sampler.
3.1 1178 nm seed source of an external cavity laser diode
Figure 3 shows the configuration of our EC-LD. It was composed of a gain controlled quantum-dot InAs/GaAs chip (Innolume). The ASE spectra of this chip are shown in Fig. 4 and indicating its gain band. Both surfaces of the gain chip were AR coated. The reflectance of the laser output facet was suppressed to 3% and the other facet was further suppressed to 0.01% by the AR coating and angled output due to the curved waveguide. This low reflecting structure can prevent lasing in the cavity constructed by the LD surfaces and thereby realize an external cavity. A diffraction grating (1600 lines/mm) was used as the wavelength selecting mirror in a Littrow configuration. Since the diffraction efficiency is proportional to the groove number in the beam spot, we placed the grating with its grooves orthogonal to the fast axis of the gain chip (vertical to the paper). The λ/2 plate was used to match the polarization with the higher diffraction efficiency axis of the grating which is orthogonal to the grooves.
Fig. 3 External cavity laser diode composed of the quantum-dot based InAs/GaAs gain chip and the diffraction grating (1600 lines/mm).
Fig. 4 ASE spectra from the angled facet of the quantum-dot based InAs/GaAs gain chip at different LD currents.
The output property is shown in Fig. 5(a) . The maximum output power in single-frequency operation is about 150 mW (the LD current ~400 mA). However, above 300 mA mode-hopping occurs frequently when increasing the LD current and the output power is fluctuating. Therefore through the experiments the EC-LD was operated with 300 mA. The laser spectrum measured by an optical spectrum analyzer (OSA) is shown in Fig. 5(b). The linewidths of the spectra were limited by the resolution (0.02 nm). The exact linewidth was measured by a delayed self-heterodyne interferometer with a 4 km delay fiber. The measured FWHM was around 175 kHz.
Fig. 5 (a) EC-LD output power as a function of the LD current. (b) Laser spectra at different LD currents obtained by an OSA with a resolution of 0.02 nm.
3.2 Fiber Raman preamplifier
The FRA setup was shown in the left part of Fig. 2. The seed light was coupled into the single-mode (SM) Raman gain fiber passing through free space with the coupling efficiency of about 15%. The Raman gain spectra of the fibers used here are shown in Fig. 6(a) . The red dashed curve shows the spectrum of a phosphosilicate fiber (called as PDF) supplied from the Fiber Optics Research Center [10] and the black curve shows a well-known spectrum of a silica fiber [11]. Because of its broad gain spectrum, a FRA is able to amplify relatively arbitrary wavelengths and 1178 nm amplification was also reported [9,12,13]. However, in general FRAs the output power is seriously limited by the SBS threshold in case that the seed source is narrow linewidth or single frequency. Not only for the FRAs but also for optical communications it is a serious problem, and thus the SBS mitigation techniques have been investigated for a long time. There are several common and simple ways, for instance imposing strain distribution [14] or temperature distribution [15] to gradually change the Brillouin shift frequency along the fiber and thus reducing the peak of the total Brillouin gain of the fiber. But these methods still require complicated systems to impose such inhomogeneous distributions. Here we demonstrated a much easier way to suppress SBS that we call the hybrid Raman fiber configuration.
Fig. 6 (a) Normalized Raman gain spectra of silica fiber (black curve) and phosphosilicate fiber (red dashed curve). (b) Normalized Brillouin gain spectra of HI1060 (black curve) and PDF (red dashed curve). The pump wavelength is 1178 nm.
We simply spliced two different types of fiber as the Raman gain fiber, in which the former fiber was the PDF and the latter fiber was HI1060 (Corning). Each fiber has a different Brillouin shift frequency as shown in Fig. 6(b). Therefore we can obtain the similar effect when we impose the strain or temperature distribution in two steps. The pump source for this fiber Raman preamplifier was a home-built nonpolarized Yb-doped fiber laser (YFL) with 50 W of maximum output power at 1120 nm, and thus we can amplify the seed light by the Raman gain around 440 cm−1of both fibers. We optimized the length of each Raman fiber for the highest output power without SBS at the maximum pump power of the YFL by the following steps: Firstly, only the PDF fiber was used for the FRA and the optimum length of 130m was simply determined. Secondly the HI1060 fiber was spliced to the pump side of the PDF fiber. Since the signal power in this fiber was high, the optimum fiber length was so short and carefully determined to be 14m by cut-backing. Since there is no overlap between the Brillouin gain spectra, splicing the HI1060 fiber does not change the SBS threshold in the first PDF fiber.
The output power properties of the FRA are shown in Fig. 7 . With suppressing SBS by our hybrid Raman fiber configuration, we were able to amplify the signal power up to 4.4 W, 1.8 times higher than that from the FRA without suppressing SBS. This suppression effect is not so strong, but we consider this method will be much more efficient if you combine it with the conventional SBS mitigation methods. Because the amount of stress or heat you can impose on fibers is limited by the strength of the fiber or its coating, it is practically difficult to widely change the Brillouin shift frequency by GHz. The linewidth measured by the delayed self-heterodyne interferometer was 320 kHz. The beat spectra of the EC-LD and the FRA output are shown in Fig. 7(b). The linewidth was twice as broad as that of the EC-LD output. Since the linewidth of the seed light monitored at WDM-1 also similarly broadened with increase of backward power by SBS, we conclude recoupling of the backward light causes linewidth broadening of the seed source and thereby of the overall output.
Fig. 7 (a) Output power properties of the FRA with hybrid Raman fiber configuration. The forward (red filled squares) and backward (blue filled circles) output powers are shown. (b) Beat spectra of the seed (black curve) and the amplifier (red curve) measured by delayed self-heterodyne detection.
3.3 Yb-PBGF single-frequency amplification
The setup was already shown in Fig. 2. The configuration of the Yb-PBGF amplifier was a simple counter pumping setup. The gain fiber was a 20 m Yb-PBGF. Since the cutoff wavelength of the transparent spectral band can to be tuned by changing the coiling diameter of the Yb-PBGF, the fiber was coiled in the optimum diameter of 26 cm for 1178 nm operation [6,7]. Both fiber ends were sealed and angle-polished by ~10 deg to inhibit the occurrence of parasitic lasing at very low threshold. The output from the FRA passed through an inline isolator (Shinkosha Co. Ltd) and was injected to the Yb-PBGF via free space. The isolator loss, Fresnel-reflection loss, and low coupling efficiency to the Yb-PBGF (~50%) limited the incident seed power to about 1.6 W. The amplifier output was taken by a dichroic mirror, and the core mode was selected by an iris pinhole. The residual pump light through the Yb-PBGF was removed from the beam line by two dichroic mirrors. After the dichroic mirrors a little part of the backward output power was picked by a beam sampler. The actual backward power was obtained by calibration.
The amplifier output power is shown in Fig. 8(a) , and the output spectrum measured by the OSA is shown in Fig. 8(b). The single-frequency output of 24.6 W was obtained at a pump power of 190 W. Since the incident seed power was not high enough, the output power was saturated. The gain was 12 dB. Since there were no exponential growth in the backward output and also no sign of the SBS in the backward spectra checked throughout the experiments, we conclude SBS did not occur. The signal intensity is 45 dB higher than ASE (Fig. 8(b)), indicating the ASE was efficiently suppressed by the spectral filtering effect. Figure 9 shows the beat spectra of the seed and the amplified seed signal, which were obtained by the delayed self-heterodyne interferometer. The signal linewidth after pumping was determined to be 320 kHz, and thus there was no spectral broadening by amplification.
Fig. 8 (a) Signal output power and the backward output power from the Yb-PBGF amplifier. (b) Output spectra of the seed (black curve) and the amplifier (blue curve).
Fig. 9 Measured beat spectra of the seed (black curve) and the amplified output (red curve) by the self-delayed heterodyne detection.
Let us compare the Yb-PBGF amplifier with a FRA. If we consider a Raman fiber with the same MFD (10.3 μm) and the same length (20 m), the gain of the FRA is estimated to be only 8.5 dB, even with core pumping at the same power of 190 W. Here we use the Raman gain coefficient of 0.43 × 10−13 m/W, evaluated from the experiments of the fiber Raman preamplifier with only a standard Ge-doped fiber (Nufern 1060XP). The higher gain of the Yb-PBGF amplifier (12 dB) is an advantage. Another advantage to be noted is that Yb-ions can store energy and thereby enable pulse operation.
4. Measurement of Brillouin gain spectrum of Yb-PBGF
We measured the Brillouin gain spectrum of the Yb-PBGF by a pump-probe method (Fig. 10 ). Two 1178 nm EC-LDs were used in the measurement. The wavelength of the probe EC-LD was tuned by changing its grating angle. The pump EC-LD was operated at a fixed wavelength and amplified by the aforementioned FRA. They were both incident into the test fiber in counter pumping configuration to induce the Brillouin gain and amplify the probe light. The probe EC-LD was perfectly protected from the pump light by two isolators. Using the λ/4 plate the probe light was incident into the Yb-PBGF in circular polarization, and its polarization state was strongly scrambled in the polarization maintaining Yb-PBGF. In this case, the Brillouin gain of the fiber is independent of the polarization state of the pump light and reduced by a factor (called as polarization factor) of 1/2 [16]. We derived the Brillouin gain by measuring the amplified power of the probe light from Coupler-3. The measured Brillouin gain coefficient was on the form fP · CB(νΒ). Here CB(νΒ) is the Brillouin gain coefficient of the test fiber [17,18], fP is the aforementioned polarization factor determined from the polarization states of the pump and the probe [16,19]. The Brillouin shift frequency νΒ was determined by the beat frequency between the probe and the pump which were combined by Coupler-1 and Coupler-2. Coupler-4 was used to attenuate the pump light incident to the photo detector.
Fig. 10 Setup of the Brillouin gain spectrum measurement by a pump-probe method. Coupler-1 to −3 were 3 dB couplers and Coupler-4 was a 30 dB tap coupler for pump power attenuation. FR: Faraday rotator, PD: photo detector (cutoff frequency 16 GHz), RF-SA: radio frequency spectrum analyzer.
The measured Brillouin gain spectra of the Yb-PBGF and a conventional SM fiber of germanosilicate fiber (Nufern, 1060XP) are shown in Fig. 11 . The spectrum of the SM fiber was measured for comparing the acoustic properties with the Yb-PBGF. The spectrum of 1060XP fiber was obtained by a similar pump-probe setup. The polarization state of the incident pump light was adjusted by a polarization controller to be matched with that of the probe light at the position of ISO-2. The polarization factor was 2/3 because there is a weak birefringence in the fiber [19]. The corresponding Brillouin gain coefficient CB(νΒ), Brillouin shift frequency νΒ and linewidth (FWHM) of Yb-PBGF and 1060XP are 0.12 m−1W−1 and 0.41 m−1W−1, 14.38 GHz and 14.23 GHz, 56 MHz and 42 MHz, respectively. To discuss the inherent Brillouin gain (in the form of gB), we multiply CB by the mode field area (MFA). The effective gain gB,eff is 1.0 × 10−11 m/W for Yb-PBGF and 1.5 × 10−11 m/W for 1060XP. Hence there is 1.8 dB suppression in the Brillouin gain of the Yb-PBGF compared to that of the 1060XP fiber. The measured acoustic properties of both fibers are summarized in Table 1 .
Table 1. Acoustic Properties of Yb-PBGF and 1060XP
It is to be noted that the spectrum of the Yb-PBGF shows a small peak in lower frequency. This small peak was originated from cladding acoustic modes, because the Yb-PBGF has Ge-doped rods with the slower sound velocity [20,21] in the clad. Hence some acoustic antiguiding effect [17,20] exists and causes the 1.8 dB suppression in the Brillouin gain. Optimizing the design of the PBG structures will enhance the acoustic antiguiding and reduce the Brillouin gain more efficiently.
5. Conclusion
The single-frequency amplification by an Yb-PBGF was demonstrated at the wavelength of 1178 nm. The seed source was an EC-LD composed of a quantum-dot based semiconductor gain chip, for which the maximum output power was 150 mW and the linewidth was 175 kHz in the single transverse mode. For efficient amplification in the Yb-PBGF, the seed power was increased by a fiber Raman preamplifier in which SBS was suppressed by the hybrid Raman fiber method. This realized 1.8 times higher output power from the preamplifier. Finally, ASE suppressed amplification to the output power of 24.6 W with a 12 dB gain was successfully obtained by the Yb-PBGF amplifier without SBS. The SNR of the output was more than 45 dB and the linewidth was 320 kHz.
To characterize the acoustic properties we measured the Brillouin gain spectrum by a pump-probe set up. From the obtained spectrum and by comparing it with a conventional SM fiber, it was found that there is 1.8 dB Brillouin gain suppression by the acoustic antiguiding effect of the PBG cladding.
We believe that the Yb-PBGF amplifier can be the most productive source for high power single-frequency generation at this wavelength region, because it has high gain and potential for pulse operation. Enhancement of the SBS suppression effect is the next-stage investigation.
Acknowledgments
This research was partly supported by Grant-in-Aid Scientific Research and the Photon Frontier Network Program of Ministry of Education, Culture, Sports, Science and Technology of Japan.
References and links
1. A. Wang, A. K. George, and J. C. Knight, “Three-level neodymium fiber laser incorporating photonic bandgap fiber,” Opt. Lett. 31(10), 1388–1390 (2006). [CrossRef] [PubMed]
2. V. Pureur, L. Bigot, G. Bouwmans, Y. Quiquempois, M. Douay, and Y. Jaouen, “Ytterbium-doped solid core photonic bandgap fiber for laser operation around 980 nm,” Appl. Phys. Lett. 92(6), 061113 (2008). [CrossRef]
3. C. B. Olausson, C. I. Falk, J. K. Lyngsø, B. B. Jensen, K. T. Therkildsen, J. W. Thomsen, K. P. Hansen, A. Bjarklev, and J. Broeng, “Amplification and ASE suppression in a polarization-maintaining ytterbium-doped all-solid photonic bandgap fibre,” Opt. Express 16(18), 13657–13662 (2008). [CrossRef] [PubMed]
4. 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]
5. R. Goto, E. C. Mägi, and S. D. Jackson, “Narrow-linewidth, Yb3+-doped, hybrid microstructured fibre laser operating at 1178 nm,” Electron. Lett. 45(17), 877–878 (2009). [CrossRef]
6. C. B. Olausson, A. Shirakawa, M. Chen, J. K. Lyngsø, J. Broeng, K. P. Hansen, A. Bjarklev, and K. Ueda, “167 W, power scalable ytterbium-doped photonic bandgap fiber amplifier at 1178 nm,” Opt. Express 18(16), 16345–16352 (2010). [CrossRef] [PubMed]
7. A. Shirakawa, C. B. Olausson, H. Maruyama, K. Ueda, J. K. Lyngsø, and J. Broeng, “High power ytterbium fiber lasers at extremely long wavelengths by photonic bandgap fiber technology,” Opt. Fiber Technol. 16(6), 449–457 (2010). [CrossRef]
8. X. Fan, M. Chen, A. Shirakawa, K. Ueda, C. B. Olausson, J. K. Lyngsø, and J. Broeng, “High power Yb-doped photonic bandgap fiber oscillator at 1178 nm,” Opt. Express 20(13), 14471–14476 (2012). [CrossRef] [PubMed]
9. Y. Feng, L. R. Taylor, and D. B. Calia, “25 W Raman-fiber-amplifier-based 589 nm laser for laser guide star,” Opt. Express 17(21), 19021–19026 (2009). [CrossRef] [PubMed]
10. I. A. Bufetov, M. M. Bubnov, Y. V. Larionov, O. I. Medvedkov, S. A. Vasiliev, M. A. Melkoumov, A. A. Rybaltovsky, S. L. Semjonov, E. M. Dianov, A. N. Gur’yanov, V. F. Khopin, F. Durr, H. G. Limberger, R.-P. Salathe, and M. Zeller, “Highly Efficient One- and Two- Cascade Raman Laser Based on Phosphosilicate fibers,” Laser Phys. 13, 234–239 (2003).
11. R. H. Stolen, C. Lee, and R. K. Jain, “Development of the stimulated Raman spectrum in single-mode silica fibers,” J. Opt. Soc. Am. B 1(4), 652–657 (1984). [CrossRef]
12. Y. Feng, L. R. Taylor, and D. Bonaccini Calia, “Multiwatts narrow linewidth fiber Raman amplifiers,” Opt. Express 16(15), 10927–10932 (2008). [CrossRef] [PubMed]
13. C. Vergien, I. Dajani, and C. Robin, “18 W single-stage single-frequency acoustically tailored Raman fiber amplifier,” Opt. Lett. 37(10), 1766–1768 (2012). [CrossRef] [PubMed]
14. T. Horiguchi, T. Kurashima, and M. Tateda, “Tensile strain dependence of Brillouin frequency shift in silica optical fibers,” IEEE Photon. Technol. Lett. 1(5), 107–108 (1989). [CrossRef]
15. J. Hansryd, F. Dross, M. Westlund, P. A. Andrekson, and S. N. Knudsen, “Increase of the SBS threshold in a short highly nonlinear fiber by applying a temperature distribution,” J. Lightwave Technol. 19(11), 1691–1697 (2001). [CrossRef]
16. T. Horiguchi, M. Tateda, N. Shibata, and Y. Azuma, “Brillouin gain variation due to a polarization-state change of the pump or Stokes fields in standard single-mode fibers,” Opt. Lett. 14(6), 329–331 (1989). [CrossRef] [PubMed]
17. A. Kobyakov, S. Kumar, D. Q. Chowdhury, A. B. Ruffin, M. Sauer, S. R. Bickham, and R. Mishra, “Design concept for optical fibers with enhanced SBS threshold,” Opt. Express 13(14), 5338–5346 (2005). [CrossRef] [PubMed]
18. V. Lanticq, S. Jiang, R. Gabet, Y. Jaouën, F. Taillade, G. Moreau, and G. P. Agrawal, “Self-referenced and single-ended method to measure Brillouin gain in monomode optical fibers,” Opt. Lett. 34(7), 1018–1020 (2009). [CrossRef] [PubMed]
19. M. Oskar van Deventer and A. J. Boot, “Polarization Properties of Stimulated Brillouin Scattering in Single-Mode Fibers,” J. Lightwave Technol. 12(4), 585–590 (1994). [CrossRef]
20. M.-J. Li, X. Chen, J. Wang, S. Gray, A. Liu, J. A. Demeritt, A. B. Ruffin, A. M. Crowley, D. T. Walton, and L. A. Zenteno, “Al/Ge co-doped large mode area fiber with high SBS threshold,” Opt. Express 15(13), 8290–8299 (2007). [CrossRef] [PubMed]
21. P. D. Dragic, “Brillouin spectroscopy of Nd–Ge co-doped silica fibers,” J. Non-Cryst. Solids 355(7), 403–413 (2009). [CrossRef]
