Expand this Topic clickable element to expand a topic
Skip to content
Optica Publishing Group

Broadband flat-amplitude multiwavelength Brillouin-Raman fiber laser with spectral reshaping by Rayleigh scattering

Open Access Open Access

Abstract

In this letter, we propose a novel configuration for generating multiwavelength Brillouin-Raman fiber laser (MBRFL). The spectral reshaping effect introduced by Rayleigh scattering in a 50km single mode fiber unifies the generated Brillouin comb in terms of both power level and linewidth. As a consequence, we are able to obtain a 40nm flat-amplitude MBRFL with wide bandwidth from 1557nm to 1597nm covering >500 Stokes lines. This is, to the best of our knowledge, the widest flat-amplitude bandwidth of MBRFL with uniform Stokes combs using just a single Raman pump laser. The channel-spacing is 0.08nm and the measured OSNR is higher than 12.5dB. We also demonstrate that the output spectrum of the MBRFL is nearly unaffected over 14dB range of Brillouin pumping power.

© 2013 Optical Society of America

1. Introduction

Multiwavelength fiber laser has potentials for wide applications in optical communications, fiber-optic sensing, optical component characterization, etc. Among various ways to generate multiwavelength lasing, multiwavelength Brillouin-Raman fiber laser (MBRFL) which combines broadband Raman gain with narrow-band Brillouin linewidth has attracted a lot of attentions due to its advantages, such as a large number of channels, low-cost and simplicity. In the last few years, some attempts have been carried out to optimize the performance parameters of MBRFL, e.g., number of output Stokes lines [1,2], flat-amplitude bandwidth [3,4], optical signal-to-noise ratio (OSNR) [5], channel spacing [6,7] and stability [8].

One of the major challenges for MBRFL is to achieve uniform Stokes lines in a wide bandwidth. The MBRFLs in a linear cavity formed by reflector at each end of fiber [3] and half-open cavity [9] have been shown to generate uniform Stokes combs and the flat-amplitude bandwidth of each of these two configurations is about 17nm. A viable way to further widen the flat-amplitude bandwidth is utilizing the multiple Raman pumping scheme [1,3], because the Raman gain spectrum can be broadened and flattened by using several pumps with different wavelengths. However, the need of multiple high-power Raman pump units will increase the cost considerably and result in a more complex configuration.

Recently, R. Sonee Shargh et al. proposed a simple MBRFL with a single pump wavelength in a linear cavity without employing any feedback mirrors at the ends of the cavity [5], and later they demonstrated a MBRFL using double-pass structure [4]; the flat-amplitude bandwidth of generated Stokes combs are 37nm and 30nm, respectively. But note that the uniformity of the Brillouin combs is not very well: there are discrepancies in power levels (2.3dB) and linewidths between the odd and the even channels.

On the other hand, Rayleigh scattering (RS) can serve as virtual mirrors and provide distributed feedback for random lasing [1015]. As for the Brillouin fiber laser, Rayleigh distributed feedback has been proved to significantly narrow the Stokes lines [16], and Rayleigh-scattering-based narrow linewidth Brillouin lasing has been demonstrated recently [17,18]. Moreover, the self-gain narrow linewidth fiber laser can also be achieved based on stimulated RS [19]. RS also plays an important role in Brillouin-Raman fiber laser and has been found to assist the generation of Brillouin Stokes combs [20,21]. In this work, we propose a novel mirror-less cavity for MBRFL by combining the “active” dispersion compensating fiber (DCF) and the “passive” single mode fiber (SMF) to enhance the RS effect on the Stokes combs. It turns out RS in a 50km SMF can significantly improve the performance of MBRFL and over 500 Stokes lines across 40nm bandwidth are generated, without obvious discrepancies in power level and linewidth between neighboring channels that exist in the previous works with mirror-less open cavity [4,5]. This is also, to the best of our knowledge, the widest flat-amplitude bandwidth (3dB range) of MBRFL with uniform Stokes combs using just a single Raman pump laser.

2. Experimental setup

Figure 1 shows the experimental setup for the MBRFL. A wavelength-tunable laser with 150MHz linewidth and maximum 10dBm power is used as the Brillouin pump (BP). The power fluctuation of BP is less than 0.01dB over 1 hour. The BP is injected into DCF through an optical circulator (insertion loss: 0.3dB; isolation from port 2 to port 1: 29dB) and a 3dB optical coupler (OC) (excess loss: 0.6dB). The 1480nm Raman pump (RP) has 1nm spectral bandwidth at 1W output level, and the power fluctuation is less than 1% over 1 hour. The RP is launched into the fiber through the isolator (isolation: 60dB) and the 1455/1550nm WDM (insertion loss: 0.5dB), providing Raman gain for the BP. The 10km DCF (YOFC DCF G.652-C/250) with 9.7GHz Brillouin frequency shift acts as both Brillouin and Raman gain medium with random distributed Rayleigh feedback. The dispersion and dispersion slope of the DCF are −130ps/nm/km and −0.438ps/nm2/km at 1550nm, respectively, and the numerical aperture (NA) is 0.27. The DCF has 0.45dB/km loss at 1550nm, and the total insertion loss of the section of DCF after being spliced with SMF is about 7.2dB at 1550nm. Note that the parasitic feedback from the fiber connections is avoided by splicing or using angled connectors, while the splicing loss is controlled to less than 0.01dB and the connecting loss is below 0.1dB. The right tip of the DCF is attached with an isolator with 50dB isolation to avoid the Fresnel reflection on the fiber end. A 50km SMF (Corning SMF 28e + ) with 11.1GHz Brillouin frequency shift, 0.18dB/km loss at 1550nm and 0.14 NA is connected to the port 2 of the OC as a distributed mirror, in order to partially reflect the Stokes light-waves back into the DCF and reshape the spectrum at the same time. The port 3 of circulator is connected to the port 4 of the OC in order to utilize larger portion of the DCF output into the SMF. We have measured the output power of the OC’s port 2 when fixed input power is injected into the OC’s port 3, and found that the output power is 2.5dB higher in the case of connecting the circulator’s port 3 and the OC’s port 4 together, comparing with no connection between the two ports. The output spectrum is monitored at the end of the 50km SMF by an optical spectrum analyzer (OSA) with 0.01 nm resolution.

 figure: Fig. 1

Fig. 1 Experimental setup for MBRFL.

Download Full Size | PDF

3. Results and discussions

The spectrum evolution with different RP powers is firstly investigated when BP is set to be 1556.6nm with 0dBm output. Without the RP, the total BP power measured at the output is only −45dBm. When the RP is injected into the DCF, the BP power is boosted via Raman gain and the 1st Stokes line will appear when the RP power is above certain threshold. The 1st Stokes line will act as the new BP and generate the 2nd Stokes line as the RP grows. This cascaded process will continue and a wide Stokes combs appears when RP power is 585mW. As shown in Fig. 2, at first the even-order lines which are the Rayleigh components have lower power level than Brillouin components (odd-order lines). The Rayleigh components increase rapidly and reach nearly the same power level as the Brillouin components when the RP power is increased to 765mW. However, the linewidth of Brillouin components is still significantly wider than that of Rayleigh components, which were observed in previous works [5,20,21]. By further increasing the RP power to 1030mW, the power of Rayleigh components grow faster and they become higher than odd-order lines by ~4dB. Finally, the power of odd-order lines can also increase to the same level of even-order lines when the RP power is 1150mW; and the linewidth discrepancies among the neighboring channels are diminished. It should be noted that the maximum input power of OSA is 20dBm and the total output power of MBRFL is about 10dBm, so it is clear that the OSA is working in a non-saturated regime.

 figure: Fig. 2

Fig. 2 Spectrum evolution with different RP powers.

Download Full Size | PDF

The widest flat-amplitude bandwidth of the MBRFL is obtained when BP is set to be 1556.6nm with 0dBm output, while RP power is 1360mW, as shown in Fig. 3.More than 500 uniform Brillouin Stokes lines with 0.08nm spacing are obtained. The flat-amplitude bandwidth is from 1557 to 1597nm. Figures 4(a)4(c) represent magnified views of the Stokes lines at the left margin, the center and the right margin of whole 40nm spectrum, respectively.

 figure: Fig. 3

Fig. 3 Optimized flat-amplitude Stokes spectrum with 1360mW RP.

Download Full Size | PDF

 figure: Fig. 4

Fig. 4 Magnified view of the Stokes spectrum at 1360mW RP power: (a) near the left margin of the spectrum; (b) near the center of the spectrum; (c) near the right margin of the spectrum.

Download Full Size | PDF

It can be clearly seen that for all three cases, the peak power of Stokes lines is within the 3dB range and there is no obvious discrepancies in linewidth among neighboring channels. The measured OSNRs are 16.5dB, 12.5dB, 14.5dB for these regions, respectively. Beyond the flat amplitude bandwidth, significant power discrepancies exist among neighboring channels, which is owing to the relative low Raman gain near 1600nm for the 1480nm RP.

In order to give prominence to the role of the 50km SMF as the virtual mirror, we remove the 50km SMF in Fig. 1 and perform the experiment with the open cavity formed by the 10km DCF only. With the optimization, the generated Stokes spectrum is shown in Fig. 5(a). It is clearly seen there are significant power discrepancies between neighboring channels in the whole span. From the magnified view of the Stoke lines as depicted in Fig. 5(b), the power discrepancies of about 2.2dB between neighboring channels is observed, and the linewidth of Brillouin components is wider than that of Rayleigh components. The linewidth discrepancy is originated from the fact that the characteristic decay time (τR) giving rise to RS is longer than the phonon lifetime of SBS (τp), and the linewidths of Brillouin components and Rayleigh components are inversely proportional to τp and τR, respectively [22].

 figure: Fig. 5

Fig. 5 Optimized Stokes spectrum without spectral reshaping by 50km SMF: (a) the whole span; (b) the magnified view of the Stoke lines.

Download Full Size | PDF

It can be inferred that the result of linewidth-equalized multiwavelength output can be mainly attributed to RS effect inside the 50km SMF and 10km DCF. With relative high Raman pump power, the RS effect on both the even- and odd-order lines inside the SMF is significant, thus the Stokes combs will be reflected by RS. As discussed in the above paragraph, the reflected Brillouin components will be affected by the spectral narrowing effect of RS [16,19,23], thus the reflected Brillouin components will have similar bandwidth as the reflected Rayleigh components. The bandwidth equalized components are then fed back into the DCF, acting as the seed light-waves for the subsequent oscillation. After multiple reciprocating inside the cavity, each Stokes line will reach the same linewidth at last. It should also be noted that, there is no significant SBS inside the 50km SMF, because the Brillouin frequency shift of each fiber span is different (11.1GHz vs 9.7GHz). The channel spacing of the MBRFL is determined by the Brillouin frequency shift of the DCF; and each Stokes line is too weak to build up significant Brillioun scattering of its own inside the passive SMF.

The influence of BP power level on the comb generation at fixed BP wavelength and RP power is shown in Fig. 6. It is shown that for the BP power range from −4dBm to 10dBm, both the output power level and the flat-amplitude bandwidth are similar. Comparing with some other MBRFL configurations that are heavily dependent on the BP power level [3,9], our scheme has its advantage on the systematic robustness.

 figure: Fig. 6

Fig. 6 Output spectra with different BP power: (a) −4dBm; (b) 4dBm; (c) 10dBm.

Download Full Size | PDF

Figure 7(a) shows the repeated scanning spectra of the MBRFL in the span of 1570.1nm to 1570.4nm. There are no observable fluctuations in both the peak power and central wavelength over the time period. We also record the peak power variation of the fixed wavelength at 1570.196nm over 30mins with the time interval of 5mins and find the power fluctuation is less than 0.5dB (see Fig. 7(b)). The results indicate the output stabilities of the MBRFL.

 figure: Fig. 7

Fig. 7 Stability measurement of the MBRFL: (a) repeated scanning spectra with 5 mins interval; (b) power fluctuation of a fixed wavelength over 30 mins.

Download Full Size | PDF

4. Conclusions

In this paper, we experimentally demonstrate a novel multiwavelength Brillouin-Raman random fiber laser with 40nm flat amplitude bandwidth, with only one Raman pump laser. Thanks to the spectral reshaping effect in the 50km SMF, the discrepancies in power level and linewidth between neighboring channels are diminished, and over 500 Brillouin Stokes lines with at least 12.5 dB OSNR are obtained. The laser’s performance is also almost immune to the Brillouin pump power variation from −4dBm to 10dBm. We suppose our work can make contribution to the further understanding of Rayleigh-feedback-based lasing. Also, the proposed laser may have potential applications in optical communication and sensing, etc [24].

Acknowledgments

This work is supported by Natural Science Foundation of China (61205048, 61106045, 61290312), Research Fund for the Doctoral Program of Higher Education of China (20120185120003), Fundamental Research Funds for the Central Universities (ZYGX2012J002, ZYGX2011J001, ZYGX2011J002), and PCSIRT (IRT1218).

References and links

1. B. Min, P. Kim, and N. Park, “Flat amplitude equal spacing 798-channel Rayleigh-assisted Brillouin-Raman multi-wavelength comb generation in dispersion compensating fiber,” IEEE Photonics Technol. Lett. 13(12), 1352–1354 (2001). [CrossRef]  

2. A. K. Zamzuri, M. I. Md Ali, A. Ahmad, R. Mohamad, and M. A. Mahdi, “Brillouin-Raman comb fiber laser with cooperative Rayleigh scattering in a linear cavity,” Opt. Lett. 31(7), 918–920 (2006). [CrossRef]   [PubMed]  

3. A. K. Zamzuri, M. A. Mahdi, A. Ahmad, M. I. Md Ali, and M. H. Al-Mansoori, “Flat amplitude multiwavelength Brillouin-Raman comb fiber laser in Rayleigh-scattering-enhanced linear cavity,” Opt. Express 15(6), 3000–3005 (2007). [CrossRef]   [PubMed]  

4. R. Sonee Shargh, M. H. Al-Mansoori, S. B. A. Anas, R. K. Z. Sahbudin, and M. A. Mahdi, “OSNR enhancement utilizing large effective area fiber in a multiwavelength Brillouin–Raman fiber laser,” Laser Phys. Lett. 8(2), 139–143 (2011). [CrossRef]  

5. R. Sonee Shargh, M. H. Al-Mansoori, S. B. A. Anas, R. K. Z. Sahbudin, A. K. Zamzuri, and M. A. Mahdi, “Improvement of comb lines quality employing double-pass architecture in Brillouin-Raman laser,” Laser Phys. Lett. 8(11), 823–827 (2011). [CrossRef]  

6. H. Ahmad, M. Z. Zulkifli, N. A. Hassan, and S. W. Harun, “S-band multiwavelength ring Brillouin/Raman fiber laser with 20 GHz channel spacing,” Appl. Opt. 51(11), 1811–1815 (2012). [CrossRef]   [PubMed]  

7. G. Mamdoohi, A. R. Sarmani, A. F. Abas, M. H. Yaacob, M. Mokhtar, and M. A. Mahdi, “20 GHz spacing multi-wavelength generation of Brillouin-Raman fiber laser in a hybrid linear cavity,” Opt. Express 21(16), 18724–18732 (2013). [CrossRef]   [PubMed]  

8. Y. Liu, D. Wang, and X. Dong, “Stable room-temperature multi-wavelength lasing oscillations in a Brillouin-Raman fiber ring laser,” Opt. Commun. 281(21), 5400–5404 (2008). [CrossRef]  

9. H. Wu, Z. N. Wang, X. H. Jia, P. Y. Li, M. Q. Fan, Y. Li, and Y. Y. Zhu, “Flat amplitude multiwavelength Brillouin-Raman random fiber laser with a half-open cavity,” Appl. Phys. B 112(4), 467–471 (2013). [CrossRef]  

10. S. K. Turitsyn, S. A. Babin, A. E. El-Taher, P. Harper, D. V. Churkin, S. I. Kablukov, J. D. Ania-Castañón, V. Karalekas, and E. V. Podivilov, “Random distributed feedback fiber laser,” Nat. Photonics 4(4), 231–235 (2010). [CrossRef]  

11. 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]  

12. D. V. Churkin, S. A. Babin, A. E. El-Taher, P. Harper, S. I. Kablukov, V. Karalekas, J. D. Ania-Castañón, 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]  

13. Y. Y. Zhu, W. L. Zhang, and Y. Jiang, “Tunable multi-wavelength fiber laser based on random Rayleigh back-scattering,” IEEE Photonics Technol. Lett. 25(16), 1559–1561 (2013). [CrossRef]  

14. Z. N. Wang, Y. J. Rao, H. Wu, P. Y. Li, Y. Jiang, X. H. Jia, and W. L. Zhang, “Long-distance fiber-optic point-sensing systems based on random fiber lasers,” Opt. Express 20(16), 17695–17700 (2012). [CrossRef]   [PubMed]  

15. Z. N. Wang, H. Wu, M. Q. Fan, Y. J. Rao, X. H. Jia, and W. L. Zhang, “Third-order random lasing via Raman gain and Rayleigh feedback within a half-open cavity,” Opt. Express 21(17), 20090–20095 (2013). [CrossRef]   [PubMed]  

16. A. A. Fotiadi and R. V. Kiyan, “Cooperative stimulated Brillouin and Rayleigh backscattering process in optical fiber,” Opt. Lett. 23(23), 1805–1807 (1998). [CrossRef]   [PubMed]  

17. M. Pang, S. Xie, X. Bao, D. P. Zhou, Y. Lu, and L. Chen, “Rayleigh scattering-assisted narrow linewidth Brillouin lasing in cascaded fiber,” Opt. Lett. 37(15), 3129–3131 (2012). [CrossRef]   [PubMed]  

18. M. Pang, X. Bao, and L. Chen, “Observation of narrow linewidth spikes in the coherent Brillouin random fiber laser,” Opt. Lett. 38(11), 1866–1868 (2013). [CrossRef]   [PubMed]  

19. T. Zhu, X. Bao, and L. Chen, “A self-gain random distributed feedback fiber laser based on stimulated Rayleigh scattering,” Opt. Commun. 285(6), 1371–1374 (2012). [CrossRef]  

20. K. D. Park, B. Min, P. Kim, N. Park, J.-H. Lee, and J.-S. Chang, “Dynamics of cascaded Brillouin-Rayleigh scattering in a distributed fiber Raman amplifier,” Opt. Lett. 27(3), 155–157 (2002). [CrossRef]   [PubMed]  

21. A. K. Zamzuri, M. H. Al-Mansoori, N. M. Samsuri, and M. A. Mahdi, “Contribution of Rayleigh scattering on Brillouin comb line generation in Raman fiber laser,” Appl. Opt. 49(18), 3506–3510 (2010). [CrossRef]   [PubMed]  

22. R. W. Boyd, Nonlinear Optics, 3rd ed. (Academic, 2008).

23. T. Zhu, F. Y. Chen, S. H. Huang, and X. Y. Bao, “An ultra-narrow linewidth fiber laser based on Rayleigh backscattering in a tapered optical fiber,” Laser Phys. Lett. 10(5), 055110 (2013). [CrossRef]  

24. Y. Liu, W. Zhang, T. Xu, J. He, F. Zhang, and F. Li, “Fiber laser sensing system and its applications,” Photonic Sens. 1(1), 43–53 (2011). [CrossRef]  

Cited By

Optica participates in Crossref's Cited-By Linking service. Citing articles from Optica Publishing Group journals and other participating publishers are listed here.

Alert me when this article is cited.


Figures (7)

Fig. 1
Fig. 1 Experimental setup for MBRFL.
Fig. 2
Fig. 2 Spectrum evolution with different RP powers.
Fig. 3
Fig. 3 Optimized flat-amplitude Stokes spectrum with 1360mW RP.
Fig. 4
Fig. 4 Magnified view of the Stokes spectrum at 1360mW RP power: (a) near the left margin of the spectrum; (b) near the center of the spectrum; (c) near the right margin of the spectrum.
Fig. 5
Fig. 5 Optimized Stokes spectrum without spectral reshaping by 50km SMF: (a) the whole span; (b) the magnified view of the Stoke lines.
Fig. 6
Fig. 6 Output spectra with different BP power: (a) −4dBm; (b) 4dBm; (c) 10dBm.
Fig. 7
Fig. 7 Stability measurement of the MBRFL: (a) repeated scanning spectra with 5 mins interval; (b) power fluctuation of a fixed wavelength over 30 mins.
Select as filters


Select Topics Cancel
© Copyright 2024 | Optica Publishing Group. All rights reserved, including rights for text and data mining and training of artificial technologies or similar technologies.