Open Access
Add to BibSonomyAbstract
A refractive index (RI) insensitive temperature sensor based on specialty triple-clad fiber (STCF) is proposed. Based on coupling mode theory, the STCF can be equivalent to a rod waveguide and two tube waveguides. Then the cladding mode resonance characteristic of STCF is analyzed by calculating different mode dispersion curves, which indicates that it works only on the mode resonance from core to the fluorine-doped silica cladding, and finally a resonance wavelength can be obtained. Two straightforward experiments are performed to prove its sensing properties. Experimental results show that it has sensitivities of 72.17 pm/°C at temperature range from 35°C~95°C with characteristics of insensitive to external RI in the range from 1.3450 to 1.4607. Thus, this proposed sensor can be used for solution temperature monitoring in real time.
© 2015 Optical Society of America
1. Introduction
Compared with the other techniques based on the mechanical and electrical methods, optical fiber sensor has attracted great attentions for its well-known advantages, such as small size, fast response, high sensitivity, anti-EM interference and corrosion resistance etc. A majority of fiber sensors for temperature measurement by using specialty optical fiber or different kinds of micro-structure fiber such as photonic crystal fibers [1], multicore fiber [2], double cladding fiber [3], fiber gratings [4–6], tapered structure [7], core-offset structure [8] and so on. Each of them has advantages and disadvantages. Fiber Bragg gratings (FBGs) are particularly attractive for external RI immunity, but its temperature sensitivity is low, only 10 pm/°C [9]. Long period fiber gratings (LPFGs) has high temperature sensitivity, almost 100pm/°C [10], but it is easy effected by bending and external RI, also it can be erased in certain circumstances. Other of these sensors which are mentioned above are mostly sensitive to external RI and temperature, however, the RI of solution always changes with its temperature, so that the cross sensitivity of RI and temperature is a critical disadvantage when these sensors are used for solution temperature monitoring. In a word, the application of all these sensors is limited more or less by its fragility, complicated fabricating process, expensive cost and cross sensitivity.
In this paper, a novel RI insensitivity temperature fiber sensor by using a section of specialty triple-clad fiber (STCF) has been proposed and investigated. The STCF is spliced between two sections of single mode fiber (SMF), and then a SMF-STCF-SMF (SSS) fiber structure is constructed. The corresponding mode properties of STCF are investigated theoretically based on the coupled mode theory, and the RI and temperature sensing characteristic of this SSS-based sensor have been studied experimentally. RI insensitivity and high temperature sensing sensitivity of 72.17 pm/°C have been achieved in the range of 35 °C~95 °C.
2. Theory of the STCF
The STCF that we used in our experiments is fabricated by Yangtze Optical Fiber and Cable Joint Stock Limited Company, and the RI distribution of it is shown in Fig. 1.
In Fig. 1, we can see that it consists of four sections, including a pure silica core, a fluorine-doped silica cladding, a single annular air-hole and a pure silica outer cladding. Equaling to the RI of the outer cladding n4, the RI of the core is higher than that of the two different kinds of inner claddings; the RI of the air-hole cladding n3 is the lowest; n5 is the RI of external environment. a, b, c and d are the radius of core, the fluorine-doped silica inner cladding, the annular air-hole cladding and the pure silica outer cladding, respectively.
As a multi-clad fiber (MCF), the analysis method of the STCF is more complicated than two-layer uniform circular waveguide, such as conventional SMF, multi-mode fiber and so on. So far, normal mode theory [11] and coupled mode theory [12] are two main methods that have been proposed to analyze the MCF. Normal mode theory can be considered as the extension of two-layer uniform circular waveguide, and it treats the MCF as a multi-layer structure during the whole analysis. However, the MCF can be considered as the combination of rod waveguide and tube waveguide by coupled mode theory. Both theories have their advantages and characteristics. The normal mode theory can strictly analyze the mode propagation characteristics of the STCF specifically, while the coupled mode theory is more likely to get the resonant wavelength and more intuitive to explain the cladding mode resonance phenomenon of the STCF. Therefore, in order to better illustrate the resonance phenomenon of the sensor and resonant wavelength, the coupled mode theory is introduced to analyze the STCF in this paper.
According to the coupled mode theory, the STCF can be equivalent to the sum of a rod waveguide and two tube waveguides (I and II) as shown in Fig. 2.
In Fig. 2, similar with the SMF, the core size of the STCF is small, only 8.54 µm, and its normalized frequency is lower than 2.405, so the rod waveguide can be regarded as a SMF which can just propagates fundamental core mode, while there are a variety of cladding modes in two tube waveguides. Based on the coupled mode theory, the core mode in rod waveguide and cladding mode in tube waveguide I will interact by evanescent wave coupling due to such a small spacing. However, this phenomenon will not happen between the rod waveguide and the tube waveguide II because of so long spacing between them, although it also has the ability to support guiding mode. In summary, the optical power from an ASE optical source near a certain wavelength which propagates in the core and the fluorine-doped silica inner cladding will couples each other. For a fabricated STCF, its structure is fixed, so the coupling coefficient between different modes has a defined value. To a large extent, the transmission energy in core is affected by the length of sensor. And only when the length reaches an appropriate value, the transmission energy in core will completely couple to the fluorine-doped silica inner cladding. This phenomenon is the so called cladding mode resonance. And at this time this wavelength and the length of the STCF is defined as the resonance wavelength and beat length, respectively.
Based on the coupled mode theory, the mode field distribution of the three waveguides and their characteristic equation [13] can be easily obtained. Due to the similar structure, the mode field distribution and characteristic equation of two tube waveguides is almost the same despite the different parameters. The dispersion curves of each mode in the three waveguides can be calculated by solving their characteristic equation with n1 = n4 = 1.458, n2 = 1.454, n3 = 1.442, 2a = 8.54μm, 2b = 26.4μm, 2c = 45.6μm, 2d = 125μm, then the resonance wavelength can be easily obtained. The dispersion curves of different modes are shown in Fig. 3.
Fig. 3 (a) Dispersion curves of rod waveguide and tube waveguide I; (b) dispersion curves of tube waveguide II.
As shown in Fig. 3(a), we can see that dispersion curves of fundamental core mode HE11(rod) in rod waveguide and cladding mode HE11(tube) in tube waveguide I cross over at point A, where the two modes have the same propagation constant and meet the phase-matching condition. That is to say, λ = 1.5637μm is the resonant wavelength, and the light can completely couples from the core to the fluorine-doped silica cladding after transmitted through the STCF with its length is equal to the beatlength. Thus, the cladding mode resonance phenomenon occurs and a band-stop filter spectrum will be obtained. Obviously, there is no intersection among dispersive curves in tube waveguide II and rod waveguide, because of the big difference of their RI index which can be observed in Fig. 3(a) and Fig. 3(b), especially in the effective wavelength range of 1.520~1.610μm that we actually used in experiments. Therefore, there is no cladding mode resonance phenomenon between rod wavelength and tube waveguide II.
Based on the analysis that mentioned above, the resonant spectrum of the sensor also can be simulating calculated. PCore, PTube and PSTCF are the light energy among the rod waveguide, tube waveguide I and the STCF, respectively. Since the mode coupling occurs between the rod waveguide and tube waveguide I and energy that excited in tube waveguide II is so little as to be almost neglected, energy in tube wavelength, the energy of the sensor satisfies the equation of PSTCF = PCore + PTube, according to the conservation of energy of non-destructive optical waveguide. Therefore, based on coupled mode theory, the expression of PCore can be represented as follows [14]:
where is the difference in propagation constants between the core mode and the cladding mode; κ is the coupling coefficient which is calculated from the fields overlap integral of the core mode and the cladding mode; z is the length of the STCF. After a brief approximate calculation, the resonant spectrum with different lengths of the STCF can be obtained, and it is depicted in Fig. 4.As shown in Fig. 4, there’s a great relationship between the resonance spectrum and the length z of the sensor. If z is equal to the beat length, the light will completely couples from the core to the cladding at the resonance wavelength as noted before; but if it is longer than the beatlength, the energy exchange between the core mode and cladding mode will occurred more than once, then multiple loss peaks of the resonance spectrum will be observed. A resonance spectrum with a litter loss peak will be obtained on the contrary if z is shorter than the beatlength. Considering the fringe visibility and demodulate feasibility of the sensor, the length of the sensor in the experiments is selected as a beat length.
3. Experiments and discussion
The schematic diagram of the SSS-based sensor is shown in Fig. 5(a). Similar with the sandwich structure proposed in Ref [15], the sensor is also easy to be fabricated. It is fabricated by just splicing a short, few-centimeter-long segment of STCF between two standard SMFs (Corning, SMF-28e) with a fusion splicer (FITEL S178). In order to obtain a resonance spectrum with low insertion loss and high fringe visibility, the discharge current and discharge time of the fusion splicer are set to 84mA and 500ms, respectively. Finally, the SSS-based sensor is fabricated successfully. The length of the STCF is 10 mm and it is really closed to the beatlength. The corresponding resonance spectrum is depicted in Fig. 5(b). The resonant wavelength of the SSS fiber structure is 1563.7 nm and without the transmission loss of nearly 5dB caused by the two splicing points, it shows great similarity with the simulated resonance spectrum in Fig. 4 within the margin of error.
Fig. 5 (a) The schematic diagram of SSS fiber structure (b) Resonance spectrum of the sensor in air at room temperature.
3.1 RI sensing properties
The experimental setup for measurement of external RI sensitivity of the SSS-based-sensor that we fabricate is shown in Fig. 6.
As shown in Fig. 6, the SSS-based-temperature-sensor is connected with an ASE (Amplified Spontaneous Emission, ASE) optical source (Shanghai Fsphotonics Technology Limited, ASE3700) and an optical spectrum analyzer (OSA) (YOKOGAWA, AQ6317C). To evaluate the capability of RI measurement, the SSS-based-sensor is tested by using solutions of glycerol and water mixture as samples. Nine different mass concentrations of glycerol in water are prepared, and its RI ranges from 1.3450 to 1.4607. The RI of the samples is all measured by using an Abbe refractometer whose accuracy is 0.001. During the whole experiment, the sensor head is mounted on a sample pool with glue to keep it straight, and it is immersed into the mixed solutions, the resonance spectrum of the sensor will not be recorded until it is stabilized by observing from the OSA. The entire measurement is carried out under a constant temperature to avoid the impact of the temperature, and it is carefully cleaned by alcohol and water for more than once to ensure the accuracy of the experiment after each RI measurement. The RI sensing experimental result is shown in Fig. 7.
Fig. 7 (a)The resonance spectrum corresponding to the different refractive index; (b) the fitting relationship between refractive index and wavelength.
The resonance spectrum corresponding to the different external RI is monitored by OSA. As shown in Fig. 7, we can find that the resonance spectrum has small irregular changes with the RI of the glycerol-water solution increasing. During the RI measurement experiment, the mixed solutions with different RI are added by a dropper, so the liquid surface tension of them is different which finally caused these small changes. However, these changes are so small that the SSS-based-sensor can be just considered insensitive to the external RI.
Combined the experimental results with the simulation data, it is not hard to see that the RI insensitive characteristics of this sensor is inevitable. As mentioned above, due to the isolation effect of the outer cladding, the changes of external RI will not affect the effective RI of modes in rod wavelength and tube waveguide I except tube waveguide II. However, the changes of modes in tube waveguide II is too small, and the dispersion curve of it still has no intersection with core mode by calculating. Thus, we can observe that the transmission spectrum of the sensor has no changes with the change of external RI in the experiment.
3.2 Temperature sensing properties
Figure 8 shows the schematic diagram of the temperature experimental setup. The SSS-based-sensor is put into a temperature-controlled cabinet for temperature measurements. To eliminate the bending cross sensitivity, two ends of the interferometer are fixed to keep it up-straight. The temperature is increased with an interval of 10 °C from 35 °C to 95 °C, and maintained for more than 10 minutes each step to ensure to stabilize before the resonance spectrum is recorded. The change of the resonance spectrum with different temperatures is depicted in Fig. 9(a).
Fig. 9 (a)The resonance spectrum corresponding to the different temperatures (b) Temperature response characteristic of the resonance spectrum.
As shown in Fig. 9, with the increasing of the temperature, the resonance spectrum has obvious drift towards long wavelength. It shifts about almost 4.3nm within the whole temperature arising process, and it exhibits high sensitivity to temperature, almost 72.17 pm/°C.
In contrast with RI sensing experiment, the SSS-based-sensor shows good response to temperature increasing, for temperature variations make big difference to all of the three waveguides due to the thermal expansion effect and thermo-optic effect of the fiber. That is to say, temperature increasing will increase the effective RI of the modes in rod waveguide and tube waveguide I. Because of the fluorine-doped in STCF, the thermal expansion coefficient and thermal optical coefficient of tube waveguide I are lower than that of rod waveguide [16]. With temperature increasing, the effective RI variation of HE11(rod) mode is greater than that of HE11(tube) mode. Thus it is not difficult to find that the resonance point A shown in Fig. 3 shifts toward longer wavelength. By measuring the shift of resonance point A, the variation of the external temperature will be determined.
As we know, the polarization sensitivity of device is related to the polarization state of the fiber [17]. The fluctuation in the input polarization to an interferometry optical fiber sensor can result in the variation of the fringe visibility. Generally, with the fringe visibility increasing, the polarization sensitivity decreases. In fiber-optic sensor based on STCF, the fringe visibility can reach 21 dB and maintain stability. Thus, it has the advantage of low polarization sensitivity and can ensure the accuracy of experimental results.
In summary, this kind of temperature sensor exhibits good temperature sensitivity, almost seven times higher than that of FBGs, and has no cross sensitive of the external RI and won’t be easily erased as LPFGs. Compare to others based on special fiber which also have the characteristics of temperature sensitive and external RI insensitive [18], it shows better linearity and lower external RI response characteristics, so higher-accuracy measurement for solutions’ temperature can be realized by utilizing it. Moreover it possesses more less transmission dips which is more convenient to demodulate. Therefore, with all these good advantages, we initially determined that the SSS-based-sensor can be better applied to detect variation of temperature within different solutions, also can replaces the FBGs or LPFGs to achieve optical switch function in some aspects.
4. Conclusion
A new fiber-optic sensor based on STCF is proposed. The sensing properties are investigated theoretically based on the coupled mode theory. Experimental results indicate that the RI insensitive sensor with an only 10 mm long STCF can be operated with high temperature sensing sensitivity of 72.17 pm/°C within the range of 35 °C~95 °C. This type of sensor has the features of small size, high sensitivity, high stability, simple structure, and low cost. It is expected to be applied in temperature sensing in different solution.
Acknowledgments
This work was supported by the National Natural Science Foundation of China (Nos. 61205068 and 61475133), College Youth Talent Project of Hebei Province (No. BJ2014057), Hebei Provincial Science and Technology Program (Nos.13273305 and 12963550D), “XinRuiGongCheng” Talent Project and Excellent Youth Funds for School of Information Science and Engineering, Yanshan University(No. 2014201). Zeng acknowledges the Program for Eastern Scholar at Shanghai Institutions of Higher Learning.
References and links
1. Y. Lu, M. T. Wang, C. J. Hao, Z. Q. Zhao, and J. Q. Yao, “Temperature sensing using photonic crystal fiber filled with silver nanowires and liquid,” IEEE Photon. J. 6(3), 6801307 (2014). [CrossRef]
2. A. Van Newkirk, E. Antonio-Lopez, G. Salceda-Delgado, R. Amezcua-Correa, and A. Schülzgen, “Optimization of multicore fiber for high-temperature sensing,” Opt. Lett. 39(16), 4812–4815 (2014). [CrossRef] [PubMed]
3. F. F. Pang, W. C. Xiang, H. R. Guo, N. Chen, X. L. Zeng, Z. Y. Chen, and T. Y. Wang, “Special optical fiber for temperature sensing based on cladding-mode resonance,” Opt. Express 16(17), 12967–12972 (2008). [CrossRef] [PubMed]
4. L. L. Xian, P. Wang, and H. P. Li, “Power-interrogated and simultaneous measurement of temperature and torsion using paired helical long-period fiber gratings with opposite helicities,” Opt. Express 22(17), 20260–20267 (2014). [PubMed]
5. W. Huang, Y. G. Liu, Z. Wang, B. Liu, J. Wang, M. M. Luo, J. Q. Guo, and L. Lin, “Multi-component-intermodal-interference mechanism and characteristics of a long period grating assistant fluid-filled photonic crystal fiber interferometer,” Opt. Express 22(5), 5883–5894 (2014). [CrossRef] [PubMed]
6. L. Chen, W. G. Zhang, Y. J. Liu, L. Wang, J. Sieg, B. Wang, Q. Zhou, L. Y. Zhang, and T. Y. Yan, “Real time and simultaneous measurement of displacement and temperature using fiber loop with polymer coating and fiber Bragg grating,” Rev. Sci. Instrum. 85(7), 075002 (2014). [CrossRef] [PubMed]
7. Z. B. Tian and S. S.-H. Yam, “In-line abrupt taper optical fiber Mach–Zehnder interferometric strain sensor,” IEEE Photon. Technol. Lett. 21(3), 161–163 (2009). [CrossRef]
8. B. Dong and E. J. Z. Hao, “Core-offset hollow core photonic bandgap fiber-based intermodal interferometer for strain and temperature measurements,” Appl. Opt. 50(18), 3011–3014 (2011). [CrossRef] [PubMed]
9. O. Frazao and J. L. Santos, “Simultaneous measurement of strain and temperature using a Bragg grating structure written in germanosilicate fibres,” J. Opt. A, Pure Appl. Opt. 6(6), 553–556 (2004). [CrossRef]
10. A. Singh, D. Engles, A. Sharma, and M. Singh, “Temperature sensitivity of long period fiber grating in SMF-28 fiber,” Optik (Stuttg.) 125(1), 457–460 (2014). [CrossRef]
11. J. W. Attrigde, J. R. Cozens, and K. D. Leaver, “Coxial fiber sensors,” J. Lightwave Technol. 3(5), 1084–1091 (1985). [CrossRef]
12. J. W. Fleming, “Dispersion in step-index silicone-clad fibers,” Appl. Opt. 18(23), 4000–4002 (1979). [CrossRef] [PubMed]
13. Z. N. Xu and Z. J. Liu, “Light propagation in a circular double clad fiber using coupled mode method,” Acta Photonica Sinica 39(10), 1857–1860 (2010). [CrossRef]
14. R. V. Schmidt and R. C. Alferness, “Directional coupler switches, modulators and filter using alternating Δβ techniques,” IEEE Trans. Circ. Syst. CAS-26(12), 1099–1108 (1979). [CrossRef]
15. J. E. Antonio-Lopez, Z. S. Eznaveh, P. LiKamWa, A. Schülzgen, and R. Amezcua-Correa, “Multicore fiber sensor for high-temperature applications up to 1000°C,” Opt. Lett. 39(15), 4309–4312 (2014). [CrossRef] [PubMed]
16. A. Koike and N. Sugimoto, “Temperatrue dependences of optical path length in fluorine-doped silica glass and bismuthate glass,” Proc. SPIE 6616, 61160Y (2006). [CrossRef]
17. H. Z. Lin, L. N. Ma, Y. M. Hu, and W. Wang, “Elimination of polarization-induced signal fading and reduction of phase noise in interferometric optical fiber sensor using polarization diversity receivers,” Optik (Stuttg.) 124(21), 4976–4979 (2013). [CrossRef]
18. A. Zhou, Q. Xu, T. Zheng, J. Yang, Y. Y. Yang, and L. B. Yuan, “In-fiber modal interferometer based on coaxial dual-waveguide fiber for temperature sensing,” IEEE Photon. Technol. Lett. 26(3), 264–266 (2014). [CrossRef]
