The first experimental observation of a dynamic grating in polarization-maintaining erbium-doped fiber (PM-EDF) is reported and a novel fiber-optic strain- and temperature-sensing mechanism based on the dynamic grating in PM-EDF is demonstrated experimentally. The dynamic grating is written with light beams in one primary polarization axis of the PM-EDF, and read with a light beam in the other primary polarization axis. The readout Bragg reflection wavelength of the grating differs from the writing wavelength and the wavelength difference is proportional to the birefringence between the two polarization axes. Making use of the dependence of the birefringence on strain or temperature, strain- and temperature-sensing is realized by measuring the Bragg reflection wavelength (frequency) shift. In order to detect the weak reflection from the dynamic grating, a dual-stage synchronous detection scheme is adopted in the experiment. The results show a strain-sensitivity of 1.4 MHz/με and a temperature-sensitivity of 60 MHz/°C, respectively.
©2006 Optical Society of America
Distributed fiber-optic strain sensors and temperature sensors have become important devices for monitoring the condition of structures and structural materials with their capabilities of spatially continuous measurement. A typical technique for distributed fiber optic sensors is based on the stimulated Brillouin scattering (SBS) process, such as the Brillouin optical time-domain analysis (BOTDA)  and the Brillouin optical correlation-domain analysis (BOCDA) . Recently, we have proposed a novel scheme based on forming a dynamic grating in polarization-maintaining erbium-doped fiber (PM-EDF) . In this scheme, the dynamic grating is written with light beams in one primary polarization axis of the PM-EDF, and read with a light beam in the other primary polarization axis. The readout Bragg reflection wavelength of the grating differs from the writing wavelength and the wavelength difference is proportional to the birefringence of two polarization axes. Strain- and temperature- sensing can be realized by measuring the Bragg reflection wavelength (frequency) shift, since the birefringence is sensitive to the strain or temperature on the fiber. The dynamic grating is localized and swept along the fiber by using the technique of synthesis of optical coherence function (SOCF)  to realize fully distributed sensing. It has been predicted by simulation that the performance of strain-sensitivity and temperature-sensitivity comparable or even better than SBS-based technology is expected with the new scheme .
However, dynamic gratings reported up to date are all formed in single mode EDF [5–8]. These gratings’ Bragg reflection wavelength is just the same as the wavelength of the lightwaves that form the grating. In this letter, we report, for the first time to the best of our knowledge, a successful experimental observation of the dynamic grating in PM-EDF, and demonstrate a novel fiber-optic strain- and temperature-sensing mechanism based on the grating.
As shown schematically in Fig. 1, when two counter-propagating coherent light beams (referred to as writing beams hereafter) are launched into a pumped erbium-doped fiber (EDF), they interfere to each other and form stationary interference fringes in the fiber. The interference fringes create a periodical gain structure per the phenomenon of gain saturation and hence produce a dynamic grating in the EDF [5–8]. The period of the grating is the half of the writing beam’s wavelength in the fiber. When a third beam is launched into the fiber (referred to as reading beam hereafter), it is reflected by the dynamic grating when its optical frequency (wavelength) is the same as the writing beams’. In other words, the reading beam is reflected when it satisfies the Bragg condition of the dynamic grating.
Here, we form the dynamic grating in PM-EDF. As shown in Fig. 2, the polarization direction of the writing beams is in one primary polarization axis (x-axis) of the PM-EDF, and the reading beam’s in the other primary polarization axis (y-axis). Due to the birefringence between these two primary polarization axes, the Bragg reflection frequency is different from that of the writing beams. The Bragg frequency is given by
where fx denotes the writing beams’ frequency, and nx and ny the refractive indices of the two primary polarization axes, respectively.
When a strain or temperature change is applied to the fiber, the refractive indices change because of the photo-elastic effect, which results in a change in the birefringence. Consequently, the detected Bragg reflection frequency shifts as shown in Fig. 3. The shift of the Bragg frequency is in proportion to the birefringence change, which is proportional to the strain or temperature:
where ∆f and ∆B denote the shift of the Bragg frequency and the birefringence change caused by strain and temperature variation, respectively. Therefore, the magnitude of strain or temperature change can be calculated by measuring the ∆f:
where ε and ∆T denote the strain and temperature change, and α and β the strain coefficient and temperature coefficient of the birefringence, respectively. Our simulation has shown a strain sensitivity of 0.426 MHz/με under a conservative assumption in the parameters .
3. Experimental setup
The experimental setup is shown in Fig. 4. Two light beams from LD1, a distributed feedback laser diode (DFB-LD), are launched into the PM-EDF (50-cm Nufern PM-ESF-7/125 high-doped PM-EDF) to write the dynamic grating in polarization direction X. The PM-EDF is pumped with a 980-nm pump laser diode. A light beam from LD2, a wavelength-tunable DFB-LD, is used to read the grating in polarization direction Y. Three adjustable attenuators are used for obtaining optimum writing and reading intensities. The intensity of the writing beams is around 0 dBm, while the reading beam about -5 dBm, respectively. Three polarizers are used to enhance the polarization extinction ratio of the writing/reading beams, and a polarization beam splitter/combiner (PBS/PBC) is used to introduce the reading beam to the PM-EDF and to output the reflection from the dynamic grating in polarization direction Y.
Since the PM fiber and the polarization components have only a limited extinction ratio, the light beam in one primary polarization axis can couple to the other primary polarization axis. Therefore, when we measure the reflection from the dynamic grating in polarization direction Y, the detector receives not only the reflection of the reading beam but also a part of writing beam that coupled to polarization direction Y. In order to remove the leakage of the writing beam from the reflected reading beam, we chop the reading beam and use a lock-in amplifier LIA1 for synchronous detection. A LiNbO3 electro-optic intensity modulator (IM) is used as a chopper, where the input polarization is controlled by a polarization controller (PC). The chopping frequency is selected as high as 37 MHz to avoid the gain modulation effect in EDF .
Even after the synchronous detection by LIA1, there is still quite strong noise received that results in a lot of spurious peaks in the observed reflection spectrum of the dynamic grating. The noise is not rejected by LIA1 because it is modulated at the same frequency as the chopping frequency. This noise is directly related to the chopped reading beam as shown schematically in Fig. 5, including the amplified Rayleigh backscattering of the reading beam, and the amplified reflection of the reading beam at splicing points or connection points. There is another type of noise coming from the beating between the above reading-beam-related noise and the other light components received at the photo-detector. These components include the amplified spontaneous emission (ASE), which is relatively less significant, and the writing-beam-related light. Through combinations of reflection and polarization-mode-coupling, some part of the writing-beam-related light beams go through the PBS/PBC in Y polarization direction and reach the photo-detector. To make matters worse, these writing-beam-related leakage are amplified when they pass through the pumped EDF, some even in round-trip.
All above noises make it impossible to observe the reflection spectrum of the dynamic grating even after LIA1 becasuse the reflection from the dynamic grating is quite weak and buried under the noises. In order to get rid of these noises, we develop a dual-stage synchronous signal processing method. We design a scheme to switch the dynamic grating ON and OFF. In the ON-state, the dynamic grating is formed, and in the OFF-state is not. Then the difference between the two states is purely related to the reflection at the dynamic grating; all other reflections and beatings, which are common between the both states, are canceled out. The OFF-state is realized here by modulating the light frequency of LD1 in sinusoidal wave at 47 kHz to make the two writing beams not interfere inside the PM-EDF. The ON-OFF is switched at 57 Hz in the experiment, which is decided after considering the trade-off relationship of dynamic grating setting-up time and the environmental noise. In fact, all these modulation frequencies are selected to be prime numbers in order to avoid any correlation relationship. The differnece between ON-OFF states is obtained with LIA2. All the data are recorded by computer via GPIB.
4. Experimental results
With the dual-stage synchronous signal processing method, we observed the reflection from the dynamic grating in PM-EDF successfully. To measure the reflection spectrum of the dynamic grating, we tune the wavelength of the reading beam by changing the injection current to LD2 and keep its intensity constant with a saturated EDFA (not drawn in Fig. 4) after LD2. Figure 6 shows the reflection spectrum of the dynamic grating. The peak is 47-GHz deviated from the frequency of the writing beams, corresponding to a birefringence of 3.6×10-4. Gaussian fitting shows a bandwidth of 140 MHz. The bandwidth is in inverse proportion to the length of the grating , which is here the full length of the PM-EDF.
The experiment of strain-sensing is performed as we apply a strain to the whole PM-EDF. The PM-EDF is fixed on two translation stages with epoxy adhesive and stretched by adjusting one of the stages. On the other hand, the temperature sensing experiment is performed when we put the PM-EDF into a temperature-controllable water bath. Figures 7 and 8 show the strain dependence and temperature dependence of the Bragg frequency shift, respectively. These results exhibit good linearity and demonstrate a strain-sensitivity of 1.4 MHz/με and temperature-sensitivity of 60 MHz/°C, which are 28 and 60 times higher than those of SBS technique, respectively.
Another factor to compare the performance of the two sensing mechanisms is the strain-or temperature-resolution, which is basically defined as the spectral bandwidth divided by the sensitivity. SBS has a fixed 3-dB bandwidth of about 30 MHz. For the dynamic grating, there is a tradeoff relationship between the dynamic grating’s length and its bandwidth: the longer the dynamic grating is, the smaller its bandwidth . In the experiments described above, the length of the dynamic grating is the whole length of the PM-EDF. When the PM-EDF is longer than 8 cm for strain sensing, or longer than 4 cm for temperature sensing, respectively, the dynamic grating method has a better resolution than SBS-based method.
As described in Eq. (2), the Bragg frequency shift is due to the change in birefringence. Our measured strain- and temperature-sensitivity correspond to 1.06×10-8/με and 4.56×10-7/°C in birefringence change, respectively. Using a PM-EDF with higher birefringence change is favorable for better sensing performance.
In this paper, we report the first experimental observation of a dynamic grating in PM-EDF and demonstrate a novel fiber-optic strain- and temperature-sensing mechanism based on the grating. Our experimental results show a strain-sensitivity of 1.4 MHz/με and temperature-sensitivity of 60 MHz/°C, which are 28 and 60 times higher than those of SBS-based technique, respectively. By using the technique of SOCF  to localize and scan the position of the dynamic grating along the PM-EDF, we expect this scheme can be developed as a new technique for distributed strain and temperature sensors.
References and links
1. T. Horiguchi, K. Shimizu, T. Kurashima, M. Taneda, and Y. Koyamada, “Development of a distributed sensing technique using Brillouin scattering,” J. Lightwave Technol. 13, 1296–1302 (1995). [CrossRef]
2. K. Hotate and M. Tanaka, “Distributed fiber Brillouin strain sensing with 1-cm spatial resolution by correlation-based continuous-wave technique,” IEEE Photonics Technol. Lett. 14, 179–181 (2002). [CrossRef]
3. X. Fan, Z. He, and K. Hotate, “Novel distributed fiber-optic strain sensor by localizing dynamic grating in polarization-maintaining erbium-doped fiber: proposal and theoretical analysis,” Jpn. J. App. Phys. 44, 1101–1106 (2005). [CrossRef]
4. Z. He and K. Hotate, “Distributed fiber optic stress location measurement by arbitrary shaping of optical coherence function,” J. Lightwave Technol. 20, 1715–1723 (2002). [CrossRef]
6. B. Fischer, J. L. Zyskind, J. W. Sulhoff, and D. J. DiGiovanni, “Nonlinear four-wave mixing in erbium-doped fiber amplifiers,” Electron. Lett. 29, 1858–1859 (1993). [CrossRef]
8. X. Fan, Z. He, Y. Mizuno, and K. Hotate, “Bandwidth-adjustable dynamic grating in erbium-doped fiber by synthesis of optical coherence function,” Opt. Express 13, 5756–5761 (2005). http://www.opticsexpress.org/abstract.cfm?URI=OPEX-13-15-5756 [CrossRef] [PubMed]