Open Access
Add to BibSonomy
Abstract
This paper presents and demonstrates a novel in-fiber integrated high sensitivity temperature sensor based on long Fabry-Perot (FP) resonator. In a quartz capillary, the FP resonator was composed of two single mode fibers (SMFs) whose end faces were coated by gold films. The temperature can be obtained by measuring the length variation of the FP cavity. A white light interference demodulation system was used to measure the length variation of the FP cavity. By the multiple reflections of light in the FP cavity, we achieved the sensitivity multiplication of the sensor. The proposed sensor measured the temperature up to 350°C for 2 hours, and the sensitivity of the sensor is six times that of the traditional interference temperature sensor. Due to the advantages of low cost, high sensitivity and simple fabrication, this temperature sensor can be widely used in high temperature applications.
© 2019 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
In recent years, the frequent occurrence of mountain fires has brought tremendous losses to human beings [1,2]. Therefore, the prevention and monitoring of mountain fires have been paid much attention. Excessive temperature is also the most important cause of equipment fire. For most of the equipment used in high temperature environment, temperature monitoring is extremely necessary [3–5]. Optical fiber sensors are widely studied in high temperature sensing due to its good high temperature characteristics, small size, and anti-electromagnetic interference [6,7]. Therefore, it has potential application prospects in high temperature environment, e.g., structural fire engineering [3], and engine health monitoring [4,5].
In recent years, optical fiber temperature sensors usually have the following forms: temperature sensor based on sapphire fiber [8,9], fiber Bragg gratings (FBGs) [8,10,11], and in-line interferometer [6,12–15]. The grating-based sensors usually have high resolution and wide dynamic measurement range [8,10,11], but the fabrication of FBGs usually requires complex fabrication process and extremely expensive equipment (including light source and phase mask, etc.). Sapphire fiber-optic sensors can measure extremely high temperatures (even higher than 1500°C), but their materials and fabrication cost are still relatively high. In-line interferometric sensors include several interferometer structures as follows: Michelson interferometer [16–19], Mach-Zehnder interferometer [15,20], and Fabry-Perot interferometer [12–14,21]. The in-line interferometric sensors are usually simple in structure, easy in fabrication and low in cost. However, low sensitivity is usually a fatal disadvantage for in-line interferometric sensors.
In this paper, we proposed and demonstrated a low cost and high sensitivity temperature sensor based on in-fiber long FP resonator. The sensitivity multiplication of the sensor is achieved by multiple reflections of light in the FP cavity which consists of two SMFs. The end face of the two SMFs are coated with gold film. The proposed sensor can measure temperature up to 350°C for 2 hours, and the sensitivity of the sensor is six times more than the conventional interferometric temperature sensor. Compared with other temperature sensors, the sensor proposed in this paper has the advantages of low cost, high sensitivity and simple preparation without any complex equipment.
2. Sensor structure and theoretical analysis
2.1 Sensor structure and fabrication
The structure of the long FP resonator based high temperature sensor is shown in Fig. 1(a). The sensor is fabricated as follows. Firstly, an ion sputtering instrument (JS-1600) is used for coating gold film on the end surface of SMF1 and SMF2. Secondly, SMF2 is put into a quartz capillary. The inner and outer diameters of the quartz capillary are about 127 μm and 1 mm, respectively. Then SMF1 is inserted from the left side of the capillary. And the opposite end of the capillary is inserted by SMF3 whose tip is cut into an oblique angle to reduce the tip reflection. SMF3 has two functions in the sensor. Firstly, SMF3 can further help the FP cavity to be immobilized in capillary. The second and most important point, SMF3 can effectively protect the gold film of SMF2 end face, thus ensuring the normal operation of the temperature sensor. Finally, the quartz capillary and the SMFs are fused together by a fiber fusion splicer (Fujikura FSM-100P + ). The main parameters of the fusion splicer are set as follows. The splice mode is the factory preset mode “SM 1000”. In this mode, we modified it slightly. Prefuse Power: 30.0 mA, Prefuse Time: 7000 ms, Main Arc Power: 15.0 mA, and Main Arc Time: 6000 ms. The main parameters of the sensor are as follow: the length of SMF2 is 16 mm, the reflectivity of the gold film on end face of SMF1 and SMF2 are 50% and 90%, respectively. The distance (d1) between the right end of SMF1 and the left end of SMF2 is about 10 μm. The total length of the sensor head is about 30 mm.
Fig. 1 Configuration and principle of high sensitivity temperature sensor based on long FP resonator. (a) Schematic diagram of sensor structure. (b) Operation principle diagram of the FP resonator.
2.2 Operation principle
The configuration and operation principle of the temperature sensor are shown in Fig. 1. The sensor consists of two gold-coated SMFs and another SMF with a beveled end face. The two transflective films (TFs) at the tips of SMFs form a FP resonator as shown in Fig. 1(b). When the light injects into the sensor, part of the light is reflected at TF1, and the remaining part passes through the TF1 and enters the SMF2. The vast majority of the light is reflected at the TF2, and only a small fraction of the light injects into SMF3. Similarly, the reflected light will be partial reflected again at TF1, too. Other light goes back into the SMF1. The light goes there and back like this in the FP cavity. The intensities of light propagated in the sensor are shown in Fig. 1(b), respectively. And they could be calculated as follows:
where I0 is the input light intensity, R1 and R2 are the reflectivity of the TF1 and TF2, respectively. β is the transmission loss when the light passes through the TF1. The intensity of light that goes back to the input terminal from the TF1 for the k times is Ik. That is,Supposing that k beams of light return to the input terminal. According to white light interference principle, the k beams have different optical paths (OPs). Therefore, every two beams of light can interfere with each other by adjusting a white light interference demodulation interferometer (WLIDI) when white light interference conditions are satisfied [6,17,21,22]. The complete equation of the white light interference is calculated by:
where φ1k is the phase difference between 1st beam and kth beam. The DC term of the interference result is neglected, and only its AC term is retained, which is shown as following.The OPs of the k beams of light change with the temperature [6,17,21,22]. In addition to the above, the light of Ik and Ij (1≤j<k) would be also interference, and they are named auxiliary interference peak [23]. These auxiliary peaks coincide with the interference peaks mentioned in Eq. (7).
The light beams reflected by two gold films of the sensor have a certain optical path difference (OPD). A variable OPD is obtained by scanning a mirror in the white light interferometer. When the variable OPD equals the OPD of the light returned by the sensor, the white light interference condition is satisfied. That is to say, the position of the scanning mirror was used to obtain the OP information of the sensor. Then the temperature information in the OP variation is demodulated [6,17,21,22]. The relationship between the displacement of the scanning mirror and the change of temperature is as follows,
where ΔX is OP variation tested by WLIDI, ST is temperature sensitivity of the sensor, T0 and T are initial temperature and real-time temperature, respectively.where αT is the thermal expansion coefficient (αT = 0.055 × 10−5 /°C), and CT is the thermo-optic coefficient (CT = 0.811 × 10−5 /°C) [6], n is the refractive index of the fiber core and L is the length of the sensor.2.3 Temperature sensitivity enhancement mechanism
According to Eq. (9), the temperature sensitivity ST is closely related to the length of the sensor. As shown in Fig. 1(b), with the multiple reflection of light wave in FP resonator, the equivalent length of the sensor has a several fold increase, too. For instance, the equivalent lengths of I2, I3 and I4 are 2L, 4L, and 6L, respectively. Correspondingly, according to Eq. (9), the temperature sensitivities for them are ST, 2ST, and 3ST, respectively, i.e., with the reflection of light wave k times in the FP resonator, the sensitivity of temperature sensing increases k times.
3. Temperature sensing experiments
The block diagram of the temperature sensing experiment of the high sensitivity temperature sensor is shown in Fig. 2(a). It consists of a WLIDI system [6], an optical fiber circulator, and a tubular furnace (Chengyi: CHY-1200), and the schematic diagram of the WLIDI system is shown in Fig. 2(b). The sensor was put into the tubular furnace and connected to WLIDI by a circulator as shown in Fig. 2(a). The temperature of furnace was set to increase from 22°C (room temperature) to 350°C, remain for 5 minutes and then decrease to room temperature.
Fig. 2 (a) The block diagram of the temperature sensing experiment. (b) The schematic diagram of the WLIDI system. The WLIDI system consists of the following components: a superluminescent diode (SLD), a 3-port optical fiber circulator, an optical fiber collimator whose end face is coated with transflective film, a photodiode (PD), a data acquisition (DAQ) card and a scanning mirror.
Figure 3(a) shows the WLIDI measurement results of the sensor at room temperature. The middle highest peak and two lowest peaks beside it are system inherent in the WLIDI, which have nothing to do with the signals of the temperature sensor. And there are six interference peaks inside the right dashed line frame which represent six temperature sensing signals. As mentioned above, these six interference peaks are caused by multiple reflections of light in the FP resonator. The positions of the interference peaks are related to the OPs of light passing through the FP resonator. And therefore, these peaks would shift with the varieties of OPs when the temperature changes. When the temperature rises, the peaks shift to the right (the OPs become longer) as shown in Fig. 3(b), and vice versa. In Fig. 3(b), in addition to the highest peak, there are some lower interference peaks. These peaks are the auxiliary interference peaks as mentioned in section 2.
Fig. 3 The temperature sensing results measured by WLIDI. (a) The measured results of the sensor at room temperature, (b) the OP responses of the sensor (interference peak 1) at different temperature.
In order to verify the sensitivity multiplication effect of the sensor, optical path variations of the six interference peaks at different temperatures are presented in Fig. 4. And Figs. 4(a) and (b) show the results of temperature rise and decrease, respectively. As mentioned before, the temperature sensitivity is related to the equivalent length of the sensor. Due to the multiple reflections of light in the FP resonator, the OP of the equivalent sensor is k times of the original OP, i.e., the sensitivity is k times of the original sensor. In Fig. 4, the slope of curve is the temperature sensitivity of temperature sensing signal. The temperature sensitivity ratio of sensors is 0.3014:0.6241:0.9293:1.248:1.562:1.861≈1:2:3:4:5:6 when the furnace temperature is in the rising stage. Similarly, the temperature sensitivity ratio at cooling is 0.3065:0.6339:0.9396:1.272:1.568:1.885≈1:2:3:4:5:6, too. The experimental results are in good agreement with those calculated in Eq. (10). The OP variations versus temperature show good linearity.
Fig. 4 Contrast curves of sensitivity of temperature sensor (6 peaks). (a) Temperature increases. (b) Temperature decreases.
With the increase of sensor sensitivity, there are some negative effects on noise transmission. The whole system consists of two parts, a WLIDI system and a sensor. Therefore, the measurement error Etotal can be divided into two parts, too. One is the error caused by the demodulation process of white light interferometer system (Es). The other is the equivalent length uncertainty caused by the thermal noise of the sensor (Et).
where Es depends on the characteristics of the WLIDI system, so there is no change in the process of sensor sensitivity multiplication. In comparison, Et does increase with the increase of sensitivity in this process since the equivalent length of the sensor becomes longer. With the sensitivity increases six times, the measurement error Etotal' isHowever, the sensor length uncertainty caused by thermal noise is small [24]. From Ref [24], the phase uncertainty could be calculated by
where L is the equivalent length of the sensor, r is the radius of the fiber core. V1 and V2 are convenient shorthand for the collections of thermodynamic quantities, and they depend only on the material properties of the fiber and not on the dimensions of the fiber. The phase uncertainty of the sensor can be estimated by Eq. (14), <φ2>ρ1/2 = 1.92 × 10−3 rad, <φ2>T1/2 = 1.18 × 10−4 rad. The equivalent length uncertainty of the sensor is about 0.4 nm. Although this part of the sensor noise increases several times with the increase of sensitivity, the uncertainty of equivalent length is still only a few nanometers, which has little effect on the whole system. Considering these two factors, the result is that the sensor resolution is a little worse than before, However, this degradation has little effect on the whole sensor system.The sensor was put into the tube furnace for 2 hours at 350°C to test its stability. The test results are shown in Fig. 5. The experimental results show that the six interference peaks generated by the proposed sensor can work normally in 120 minutes.
4. Discussion
Some factors may affect the sensor signals: the reflectivity of TFs (R1 and R2), the distance between the two SMFs (d1) and the ambient temperature.
4.1 Effect of TF reflectivity on interference intensity after multiple reflections
The reflectivity of TF has a great influence on the interference signals of sensor. In experiment, the reflectivity of the two TFs cannot be exactly the same, so the influence of different reflectivity of the TFs on the interference signal of the sensor needs to be discussed separately. The influence of the reflectivity of two TFs on the interference signal intensity is simulated, and the simulation model is shown as Eq. (7) in section 2.2. Moreover, in this simulation process, we only consider the influence of the reflectivity of two TFs on the interference signal, so we assume that the parameter d1 = 0, that is, the influence of the distance d1 on the interference signal is neglected. In Fig. 6(a), the higher the reflectivity of TF2 is, the greater the interference intensity is. The same phenomenon also occurs in other several interference peaks. Thus, in the experiment, the reflectivity of TF2 should be as high as possible. However, the effect of reflectivity of TF1 on interference intensity is somewhat special. When the reflectivity of TF1 is about 40%, the interference intensity of the signal is maximum for the interference peak 1. Similarly, when the reflectivity of TF2 is the highest (100%), the influence of the reflectivity of TF1 on the intensity of several other interference peaks is shown in Fig. 6(b).
Fig. 6 (a) The influence of reflectivity of two TFs on interference peak 1. The horizontal and vertical coordinate axes are the reflectivity of TF1 and TF2 (0~100%), respectively. The color bar represents interference intensity. (b). The relationship between the reflectivity of TF1 and the interference intensity of every peaks.
4.2 Effect of the distance (d1) on the interference intensity of each interference peak
The distance (d1) between TF1 and the left end of SMF2 affects the interference intensity a lot. As shown in Fig. 7, except for reference peak, the intensities of other interference peaks decrease with the increase of spacing. Moreover, as the number of reflections increase, the influence of distance (d1) increases, too. That is, the intensity of interference peak 6 decreases most rapidly with the increase of spacing.
4.3 Effect of temperature on gold film reflectivity
The reflectivity of gold film varies with temperature as shown in Fig. 8. An optical low-coherence domain reflectometry (OLCR) was used to test the reflectivity of gold film. The temperature of the tube furnace was set to rise from room temperature to 600°C. The reflectivity of the gold film decreases rapidly after 350°C. The reflectivity is only half of the maximum when temperature is 600°C. Moreover, the reduction of reflectivity is irreversible when the temperature is restored to room temperature.
There are also other factors that have some effects on the sensor. For instance, the length of the linear translation stage in the WLIDI limits the scanning range of the system, i.e., the length of the sensor and the sensitivity of the sensor are also greatly limited by it.
5. Conclusions
In summary, an in-fiber integrated high sensitivity temperature sensor based on FP interferometer was proposed and demonstrated. Sensitivity multiplication principle of the sensor has been theoretically deduced and verified by experiments. The influence of temperature on the reflectivity of gold film and the influence of the reflectivity of gold film on the signal of the sensor are discussed, respectively. The experimental results are in good agreement with the theoretical simulation. In addition, it is found that the reflectivity of gold film on end face of SMF is also related to its temperature characteristics. Consequently, further study on relevant properties of gold film coated on the end face of SMF will be summarized in our next study. The advantages of the proposed sensor are high sensitivity, compact structure and simple fabrication. Thus, the proposed high sensitivity temperature sensor can be very suitable for temperature sensing at high temperatures, e.g., structural fire engineering, and engine health monitoring.
Funding
National Nature Science Foundation of China (NSFC) (61535004, 61735009, 61827819); Guangxi project (AD17195074); National Defense Pre-Research Foundation of China (CN) (6140414030102).
Disclosures
The authors declare that there are no conflicts of interest related to this article.
References
1. M. Abdollahi, T. Islam, A. Gupta, and Q. Hassan, “An Advanced Forest Fire Danger Forecasting System: Integration of Remote Sensing and Historical Sources of Ignition Data,” Remote Sens. 10(6), 923 (2018). [CrossRef]
2. M. R. Ahmed, K. R. Rahaman, and Q. K. Hassan, “Remote Sensing of Wildland Fire-Induced Risk Assessment at the Community Level,” Sensors (Basel) 18(5), 1570 (2018). [CrossRef] [PubMed]
3. Y. Bao, Y. Huang, M. S. Hoehler, and G. Chen, “Review of Fiber Optic Sensors for Structural Fire Engineering,” Sensors (Basel) 19(4), 877 (2019). [CrossRef] [PubMed]
4. L. Polz, A. Zeisberger, H. Bartelt, and J. Roths, “Total Temperature Measurement of Fast Air Streams With Fiber-Optic Bragg Grating Sensors,” IEEE Sens. J. 16(17), 6596–6603 (2016). [CrossRef]
5. P. Jia, H. Liang, G. Fang, J. Qian, F. Feng, T. Liang, and J. Xiong, “Batch-producible MEMS fiber-optic Fabry-Perot pressure sensor for high-temperature application,” Appl. Opt. 57(23), 6687–6692 (2018). [CrossRef] [PubMed]
6. W. Li, Y. Yuan, J. Yang, and L. Yuan, “In-fiber integrated quasi-distributed high temperature sensor array,” Opt. Express 26(26), 34113–34121 (2018). [CrossRef] [PubMed]
7. W. Li, Y. Yuan, J. Yang, H. Deng, and L. Yuan, “In-Fiber Integrated Sensor Array With Embedded Weakly Reflective Joint Surface,” J. Lightwave Technol. 36(23), 5663–5668 (2018). [CrossRef]
8. S. Yang, D. Hu, and A. Wang, “Point-by-point fabrication and characterization of sapphire fiber Bragg gratings,” Opt. Lett. 42(20), 4219–4222 (2017). [CrossRef] [PubMed]
9. H. Chen, M. Buric, P. R. Ohodnicki, J. Nakano, B. Liu, and B. T. Chorpening, “Review and perspective: Sapphire optical fiber cladding development for harsh environment sensing,” Appl. Phys. Rev. 5(1), 011102 (2018). [CrossRef]
10. C. Wang, J. Zhang, C. Zhang, J. He, Y. Lin, W. Jin, C. Liao, Y. Wang, and Y. Wang, “Bragg Gratings in Suspended-Core Photonic Microcells for High-Temperature Applications,” J. Lightwave Technol. 36(14), 2920–2924 (2018). [CrossRef]
11. S. C. Warren-Smith, L. V. Nguyen, C. Lang, H. Ebendorff-Heidepriem, and T. M. Monro, “Temperature sensing up to 1300°C using suspended-core microstructured optical fibers,” Opt. Express 24(4), 3714–3719 (2016). [CrossRef] [PubMed]
12. Z. Chen, S. Xiong, S. Gao, H. Zhang, L. Wan, X. Huang, B. Huang, Y. Feng, W. Liu, and Z. Li, “High-Temperature Sensor Based on Fabry-Perot Interferometer in Microfiber Tip,” Sensors (Basel) 18(1), 202 (2018). [CrossRef] [PubMed]
13. J. Mathew, C. Hauser, P. Stoll, C. Kenel, D. Polyzos, D. Havermann, W. N. MacPherson, D. P. Hand, C. Leinenbach, A. Spierings, K. Koenig-Urban, and R. R. J. Maier, “Integrating Fiber Fabry-Perot Cavity Sensor Into 3-D Printed Metal Components for Extreme High-Temperature Monitoring Applications,” IEEE Sens. J. 17(13), 4107–4114 (2017). [CrossRef]
14. C. L. Lee, J. M. Hsu, J. S. Horng, W. Y. Sung, and C. M. Li, “Microcavity Fiber Fabry Prot Interferometer With an Embedded Golden Thin Film,” IEEE Photonic. Tech. L. 25(9), 833–836 (2013). [CrossRef]
15. H. Sun, H. Luo, X. Wu, L. Liang, Y. Wang, X. Ma, J. Zhang, M. Hu, and X. Qiao, “Spectrum ameliorative optical fiber temperature sensor based on hollow-core fiber and inner zinc oxide film,” Sens. Actuators B Chem. 245, 423–427 (2017). [CrossRef]
16. L. Yuan, “Optical path automatic compensation low-coherence interferometric fibre-optic temperature sensor,” Opt. Laser Technol. 30(1), 33–38 (1998). [CrossRef]
17. L. Yuan, L. Zhou, and J. Wu, “Fiber optic temperature sensor with duplex Michleson interferometric technique,” Sens. Actuators A Phys. 86(1), 2–7 (2000). [CrossRef]
18. Y. Zhao, A. Zhou, H. Guo, Z. Zheng, Y. Xu, C. Zhou, and L. Yuan, “An Integrated Fiber Michelson Interferometer Based on Twin-Core and Side-Hole Fibers for Multiparameter Sensing,” J. Lightwave Technol. 36(4), 993–997 (2018). [CrossRef]
19. Y. Zhang, Y. Zhang, Z. Wang, Z. Liu, Y. Wei, E. Zhao, X. Yang, J. Zhang, J. Yang, and L. Yuan, “A novel Michelson Fabry–Perot hybrid interference sensor based on the micro-structured fiber,” Opt. Commun. 374, 58–63 (2016). [CrossRef]
20. S. Yuan, Z. Tong, J. Zhao, W. Zhang, and Y. Cao, “High temperature fiber sensor based on spherical-shape structures with high sensitivity,” Opt. Commun. 332, 154–157 (2014). [CrossRef]
21. L. Yuan and J. Yang, “Tunable Fabry-Perot-resonator-based fiber-optic white-light interferometric sensor array,” Opt. Lett. 33(16), 1780–1782 (2008). [CrossRef] [PubMed]
22. Y. Yuan, B. Wu, J. Yang, and L. Yuan, “Tunable optical-path correlator for distributed strain or temperature-sensing application,” Opt. Lett. 35(20), 3357–3359 (2010). [CrossRef] [PubMed]
23. J. Yang, Y. Yuan, B. Wu, A. Zhou, and L. Yuan, “Higher-order interference of low-coherence optical fiber sensors,” Opt. Lett. 36(17), 3380–3382 (2011). [CrossRef] [PubMed]
24. W. H. Glenn, “Noise in interferometric optical systems: an optical Nyquist theorem,” IEEE J. Quantum Electron. 25(6), 1218–1224 (1989). [CrossRef]
