Open Access
Add to BibSonomyAbstract
We demonstrated a method to trim the beat frequency of dual-polarization fiber grating lasers by side-exposing the laser cavity to UV beam. The UV-side-illumination induces an additional birefringence of the cavity fiber and therefore permanently changes the beat frequency of the laser. The beat frequency can be trimmed up or trimmed down, depending on the UV incident direction relative to the intrinsic polarization axis of the active fiber. A trimming range as much as ~700MHz has been achieved. This method allows us to actively control the beat frequency of dual-polarization fiber grating lasers. A 6-channel RF-frequency division multiplexed polarimetric fiber grating laser array has been demonstrated.
©2010 Optical Society of America
1. Introduction
Fiber grating lasers have been attracting a great deal of interest for sensing applications because of their advantages of compact size, narrow bandwidth, high signal-to-noise ratio, and multiplexing capability. According to the working principle, fiber grating laser sensors can be classified into two types: wavelength encoding sensor and polarimetric sensor. The former converts measurands into shift in the operation wavelength of the laser [1–3], which is similar to that of fiber grating sensor. The latter converts measurands into change in the beat frequency between the two orthogonal polarization modes from the laser [4–6]. Because the beat frequency is in radio frequency (RF) range, polarimetric sensor has distinctive advantages of ease of interrogation and avoidance of expensive wavelength measurement that is required for wavelength encoding sensor. During the last decade, polarimetric fiber grating laser sensors have been demonstrated to measure a number of parameters, such as hydrostatic pressure [5], lateral force [6], temperature and strain [7], displacement [8], acoustic wave [9], and electric current [10].
Multiplexing capability is an attractive advantage of fiber grating laser sensors. So far, 16-channel wavelength encoding fiber grating laser sensor array has been demonstrated for detection of underwater acoustic wave by using wavelength-division-multiplexing technique [11]. However, little work has been reported on the multiplexing of polarimetric fiber grating laser sensors. To realize the multiplexing of polarimetric fiber grating laser sensors, each laser in the array is required to has not only a given wavelength but also a given beat frequency. One can easily write fiber grating lasers with different wavelength by using phase masks with different period. The difficulty lies in how to fabricate dual-polarization fiber grating lasers with different beat frequency. Although using different active fibers to fabricate dual-polarization fiber grating lasers can obtain different beat frequency due to the difference in intrinsic birefringence of the active fibers, the type of active fibers suitable for fabricating short fiber grating lasers for sensor applications is quite limited. Furthermore, the fiber dependent beat frequency is entirely uncontrollable. It would be highly desirable if we can actively control the beat frequency of the laser.
In this paper, we demonstrate a method to trim the beat frequency of dual-polarization fiber grating lasers by side-exposing the laser cavity to a UV laser beam. The UV-side-illumination induces an additional birefringence of the cavity fiber and therefore permanently changes the beat frequency of the fiber laser. This allows us to trim up or trim down the laser beat frequency according to the requirements. Based on this method, we fabricated six dual-polarization fiber grating lasers with different wavelength and different beat frequency, and multiplexed them on a single fiber.
2. Principle
Polarimetric fiber grating laser sensor is based on the beating of two orthogonal polarization modes from the laser. The two orthogonal polarization modes, denoted by x- and y-polarization, generate a beat signal in RF domain with frequency given by [7]:
where λ 0 is Bragg wavelength of the fiber grating, n 0 is average index of the fiber, B = ny − nx (ny >nx) is the fiber birefringence, and nx, y are the effective refractive index of the two polarization modes. The beat frequency is highly dependent on the birefringence of the cavity fiber. Any perturbations that modulate the fiber birefringence can be measured by detecting the beat frequency. This is the principle of the polarimetric fiber grating laser sensors. Because the beat frequency is in RF range, the sensor can be interrogated with mature and cheap techniques that are commonly used in the RF test and measurement. The advantages of polarimetric fiber grating laser sensor include ease of interrogation, absolute frequency encoding, and capability to multiplex a number of sensors on a single fiber by use of frequency division multiplexing technique.To realize the multiplexing of polarimetric fiber grating laser sensors, the key is to fabricate fiber grating laser with target beat frequency. Because the beat frequency depends on the fiber birefringence, if we can permanently change the fiber birefringence, the beat frequency can be controlled. It has been demonstrated that [12], [13], UV-side-illumination results in axially unsymmetrical index change of the optical fiber, and the light polarized along the UV incident direction sees a larger index change than that polarized perpendicular to the UV incident direction. The UV induced birefringence has been used to control the polarization properties of distributed feedback (DFB) fiber lasers [14], [15]. This also provides a way to tune the beat frequency of the dual-polarization fiber grating lasers by side-exposing the laser cavity to a UV beam. The final beat frequency is determined by the superposition of the intrinsic birefringence of the cavity fiber and the UV-induced birefringence. The UV-induced birefringence can either enhance or counteract the intrinsic birefringence, depending on the UV incident direction relative to the intrinsic polarization axis of the fiber. When the UV beam is incident along the slow axis (y- axis), as shown in Fig. 1(a) , the UV-induced birefringence will strengthen the intrinsic birefringence, resulting in the beat note shift to higher frequency. Contrarily, when the UV beam is incident along the fast axis (x- axis), as shown in Fig. 1(b), the UV-induced birefringence will counteract the intrinsic birefringence, resulting in the beat note shift to lower frequency. Therefore, we can trim up or trim down the laser beat frequency according to the requirements.
3. Experimental results and discussions
The experiment setup is showed in Fig. 2 . The same setup was used for writing dual-polarization fiber grating lasers and trimming the laser beat frequency. We wrote distributed Bragg reflector (DBR) fiber lasers by inscribing two wavelength-matched Bragg gratings with appropriate separation in a short piece of erbium (Er)-doped fiber by using 193 nm excimer laser and phase mask method. The Er-doped fiber used in the experiments was manufactured by Fibercore Ltd (DF1500L). The fiber has typical absorption of ~15 dB/m at 980 nm and ~26 dB/m at 1530 nm. All DBR fiber lasers are formed with 7-mm-long high reflectivity grating, 6-mm-long low reflectivity grating, and 10-mm grating separation. We then remove the phase mask and expose the middle of the laser cavity to 3-mm-width UV beam from the same 193 nm laser to trim the beat frequency. The dimensions of the UV beam from the laser are 3mm × 6mm (V × H). A slit is used to confine the width of the UV beam on the fiber to 3 mm. A cylindrical lens is used to vertically focus the UV beam on the fiber to 0.3 mm. The UV spot size on the fiber is 0.2mm × 3mm.
Before the UV exposing process, the most important thing is to find out the polarization axes of the cavity fiber. The fast and slow axis was identified by this way [16]: we axially rotate the fiber laser step by step by using two rotators that hold both sides of the DBR laser as shown in Fig. 2, each time after rotating a small angle, we load a constant lateral force to the laser cavity and monitor the beat frequency shift. The direction where the maximum shift toward higher frequency occurs corresponds to the fast axis; whereas the direction where the maximum shift toward lower frequency occurs corresponds to the slow axis.
In the first experiment, we fixed the pulse energy of the 193nm laser to 2 mJ and the repeat frequency to 10 Hz. Three DBR fiber lasers were tested with three different exposing directions: exposing along slow axis, exposing along fast axis, and exposing at 45°. We recorded the beat frequency after each 50 pulses. Figure 3 shows the beat frequency as a function the number of UV pulse in these three situations.
We tested the first DBR fiber laser by exposing to UV beam along the slow axis. As expected, the beat note shifted to higher frequency and a trimming rang as much as 700 MHz was achieved after 1400 pulses. As observed from Fig. 3, the shift of the beat frequency is much faster at the beginning especially in the first 50 pulses which denotes a highly effective trimming range.
Then we tested the second DBR laser with the same experiment parameters, but we aligned the fast axis of the fiber to the UV incident direction. As expected, the beat note shifted to lower frequency, as shown in Fig. 3. The third DBR laser was tested by exposing to UV beam in direction near 45° between the fast and slow axis. In theory, the beat frequency should do not change with exposure. However, a small trimming range of ~150MHz was obtained in the experiment. The most possible reason is that the exposing direction was not exactly set to 45°.
Figure 4 shows the beat note spectra for the second DBR laser before and after 1400 UV pulses. A decrease of ~3dB in the strength of the beat note is observed. The signal-to-noise ratio (SNR) of the beat note after 1400 UV pulses is still better than 40 dB.
In the above experiments, the shift of the beat frequency is very fast at the beginning of the exposing process. As shown in Fig. 3, the beat frequency shifts more than 200MHz after first 50 pulses when the UV beam is incident along the fast or slow axis. We can realize refined trimming be setting the UV laser to lower energy density and lower repeat frequency, and setting the exposing direction near 45°. Figure 5 shows the refined trimming of the laser beat frequency. We set the 193nm excimer laser’s repeat frequency to 5 Hz and increase the energy step by step from 0.1 mJ. This allows us to finely control of the beat frequency in a small range from 478MHz to 617MHz.
During the side illumination, we monitored the operation wavelength of the laser with an optical spectrum analyzer (OSA). No obvious wavelength shift was observed. This indicates that UV-side-illumination has no effect on the laser wavelength.
We measured the transmission spectra of the three laser cavities. The laser cavity supports 3 or 4 longitude modes, but these modes see different grating reflectivity and therefore have different cavity loss. The mode next to the grating reflection peak has much lower cavity loss than other modes and has absolute advantage in oscillation. Therefore, although the laser cavity supports 3 or 4 modes, the laser can stably operate in single longitude mode. In other experiments, we measured the long-term stability of the DBR fiber lasers with parameters similar to that used in the paper at 300°C for two weeks. No mode hopping was observed. The beat note showed high stability and the frequency fluctuation was less than 0.3 MHz. However, when the fiber lasers are subjected to strongly uneven perturbations such as localized heating or stretching on the subsection of the grating, mode hopping may occur due to the grating spectrum distortion.
Based on the above method, we fabricated six dual-polarization DBR fiber lasers with different wavelength and different beat frequency, and multiplexed them on a single fiber. The laser array was pumped by a 980 nm laser diode with output of ~200 mW. Figure 6(a) show the optical spectra of the 6-channel multiplexed dual-polarization fiber laser array measured with OSA. The wavelength of the six lasers ranged from 1523 to 1546nm. Figure 6(b) shows beat note spectra of 6-channel laser array in RF domain measured with RF spectrum analyzer. These six DBR fiber lasers were written by using three different period phase masks. By stretching the fiber, each phase mask was used to write two DBR lasers. Limited by the number of the phases mask in the experiments, we realized six wavelength different DBR fiber lasers.
Fig. 6 (a) Optical spectrum of 6-channel multiplexed DBR laser array measured with optical spectrum analyzer; (b) Beat note spectrum of 6-channel DBR laser array measured with RF spectrum analyzer.
4. Conclusion
In this paper, we have demonstrated a method to trim the beat frequency of the dual-polarization DBR fiber lasers by side-exposing the laser cavity to UV light. The beat frequency trimmed up when the UV beam is incident upon the laser cavity in direction along the slow axis of the fiber, whereas the beat frequency trimmed down when the UV beam is incident upon the laser cavity in direction along the fast axis of the fiber. Therefore, the beat frequency can be actively controlled in a large frequency range. Based on this method, we have fabricated six dual-polarization DBR fiber lasers with different wavelength and different beat frequency and demonstrated 6-channel RF-frequency division multiplexing of polarimetric fiber grating laser sensors.
Acknowledgment
This work was supported by the Key Project of National Natural Science Foundation of China (60736039), the Research Fund for the Doctoral Program of Higher Education (20070141041), and the Fundamental Research Funds for the Central Universities (21609102).
References and links
1. D. J. Hill, B. Hodder, J. D. Freitas, S. D. Thomas, and L. Hickey, “DFB fibre-laser sensor developments,” in Proc. 17th Int. Conf. Optical Fiber Sensors, Bruges, Belgium, 904–907 (2005).
2. G. H. Ames and J. M. Maguire, “Erbium fiber laser accelerometer,” IEEE Sens. J. 7(4), 557–561 (2007). [CrossRef]
3. G. A. Cranch, G. M. H. Flockhart, and C. K. Kirkendall, “Distributed feedback fiber laser strain sensors,” IEEE Sens. J. 8(7), 1161–1172 (2008). [CrossRef]
4. G. A. Ball, G. Meltz, and W. W. Morey, “Polarimetric heterodyning Bragg-grating fiber-laser sensor,” Opt. Lett. 18(22), 1976–1978 (1993). [CrossRef] [PubMed]
5. K. Bohnert, A. Frank, E. Rochat, K. Haroud, and H. Brändle, “Polarimetric fiber laser sensor for hydrostatic pressure,” Appl. Opt. 43(1), 41–48 (2004). [CrossRef] [PubMed]
6. J. T. Kringlebotn, W. H. Loh, and R. I. Laming, “Polarimetric Er(3+)-doped fiber distributed-feedback laser sensor for differential pressure and force measurements,” Opt. Lett. 21(22), 1869–1871 (1996). [CrossRef] [PubMed]
7. O. Hadeler, E. Rønnekleiv, M. Ibsen, and R. I. Laming, “Polarimetric distributed feedback fiber laser sensor for simultaneous strain and temperature measurements,” Appl. Opt. 38(10), 1953–1958 (1999). [CrossRef]
8. Y. Zhang and B. O. Guan, “High sensitivity distributed Bragg reflector fiber laser displacement sensor,” IEEE Photon. Technol. Lett. 21(5), 280–282 (2009). [CrossRef]
9. B. O. Guan, Y. N. Tan, and H. Y. Tam, “Dual polarization fiber grating laser hydrophone,” Opt. Express 17(22), 19544–19550 (2009). [CrossRef] [PubMed]
10. B. O. Guan and S. N. Wang, “Fiber grating laser current sensor based on magnetic force,” IEEE Photon. Technol. Lett. 22(4), 230–232 (2010). [CrossRef]
11. S. B. Foster, A. Tikhomirov, M. Englund, H. Inglis, G. Edvell, and M. Milnes, “A 16 Channel Fibre Laser Sensor Array,” 18th Int. Conf Optical Fiber Sensors, Cancún, Mexico, FA4 (2006)
12. A. M. Vengsarkar, Q. Zhong, D. Inniss, W. A. Reed, P. J. Lemaire, and S. G. Kosinski, “Birefringence reduction in side-written photoinduced fiber devices by a dual-exposure method,” Opt. Lett. 19(16), 1260–1262 (1994). [CrossRef] [PubMed]
13. H. Renner, “Effective-index increase, form birefringence and transition losses in UV-side-illuminated photosensitive fibers,” Opt. Express 9(11), 546–560 (2001). [CrossRef] [PubMed]
14. H. Storøy, B. Sahlgren, and R. Stubbe, “Single polarization fibre DFB laser,” Electron. Lett. 33(1), 56–58 (1997). [CrossRef]
15. J. L. Philipsen, M. O. Berendt, P. Varming, V. C. Lauridsen, J. H. Povlsen, J. Hubner, M. Kristensen, and B. Palsdottir, “Polarisation control of DFB fibre laser using UV-induced birefringent phase-shift,” Electron. Lett. 34(7), 678–679 (1998). [CrossRef]
16. Y. Zhang, B. O. Guan, and H. Y. Tam, “Characteristics of the distributed Bragg reflector fiber laser sensor for lateral force measurement,” Opt. Commun. 281(18), 4619–4622 (2008). [CrossRef]
