Open Access
Add to BibSonomy
Abstract
An ultra-short high-temperature fiber-optic sensor based on a silicon-microcap created by a single-mode fiber (SMF) and simple fusion splicing technology is proposed and experimentally demonstrated. A section of the SMF with a silicon-microcap at one end is connected to the “peanut” structure to build the microcap-based optical fiber improved Michelson interferometer (MI). The optimal discharge parameters of microcap and length of SMF has been investigated to achieve the best extinction ratio of 6.61 dB. The size of this microcap-based improved MI sensor is 560 µm and about 18 times shorter compared to the current fiber tip interferometers (about 10 mm). Meanwhile, it showed good robustness during the two heating-cooling cycles and the duration period stability test at 900 °C. This microcap-based improved MI sensor with the smaller size, simple fabrication, low cost, high reliability, and good linearity within a large dynamic range is beneficial to practical temperature measurement and massive production.
© 2021 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
In recent years, fiber-optic interferometers have been used in sensing applications of various physical [1–5], chemical [6–8], and biological [9–11] measurement due to their unique advantages, such as high-sensitivity, fast response, small size, and strong anti-electromagnetic interference. The measure or monitoring of high-temperature plays an essential role in modern industry, including metallurgy, oilfield exploitation, chemical manufactures, and monitoring space engine. Reflective optical fiber interferences, such as Fabry-Perot interferometers (FPI) sensors and MI sensors, are considered a practical solution in narrow space and remote distance.
Different photonic crystal fibers (PCF) were usually used as the sensing head of FPIs [12,13] to achieve a significant refractive index difference between the interference beams. Parallel structured optical fiber in-line FPIs for high-temperature sensing created by SMF with large lasers were also reported [14]. These high-temperature FPI sensors usually have abridged sensing lengths of tens of microns; consequently, they have strong robustness and anti-interference at high temperatures. However, these PCFs and fabrication technologies are expensive and complex to implement. Optical fiber MI high-temperature sensors utilize the refractive index (RI) difference between cladding mode and core mode to achieve beam interference, which was usually fabricated by simple fusion splicing technology and SMF. Unfortunately, the RI difference (≈0.0052) [15] between the core mode and cladding mode is tiny in SMF; hence the length of the interference material must be guaranteed, and the centimeter-level is almost the limit size of this type of sensors [16–22]. Some parameters of them are list in Table 3. FSR could become too large to be observed if we reduce these sensors’ length on this basis. As a result, it is hard to achieve millimeter-level or even micron-level MI high-temperature sensors in theory. However, repeatability and stability are two key targets of a temperature sensor for a wide measurement range. The size can seriously affect the stableness of optical fiber sensors [23].
In this article, we ingeniously designed a simple structure by fabricating a silicon-microcap located at the end face of SMF via electrical discharge. To the best of our knowledge, for the first time, a micron-level fiber-optic MI high-temperature sensor, which is 18 times shorter than this type of MI sensor (10 mm) [18] adopting this structure was proposed. The length of the SMF and discharge parameters at the end face with the best extinction ratio were obtained by the contrast experiment. In the two heating-cooling cycles and a long-period stability test at 900 °C, it showed good stability. Moreover, it also has good economic effects and is very easy to fabricate.
2. Device fabrication and principle
Peanut shape was chosen as the excitation and coupling unit (ECU) of our sensor because of its advantages of better robustness and simplicity than taper or other complex structures. As illustrated in Fig. 1, the structure’s fabrication process is as follows: Firstly, a section of SMF was placed in one of the fiber holders of a commercial fusion splicer (Fujikura 80s). Secondly, to ensure the fiber end melted to form the expected smooth microsphere under the action of electrode discharge, higher arc power of 80 bit and longer arc duration time of 2000ms were applied. Thirdly, such two sections of SMFs with a silicon-microsphere were placed in the two fiber holders of the fusion splicer, respectively. Before two SMFs were connected, the lower arc power of −20 bit and shorter arc duration time of 800 ms were applied to ensure the integrity of two microspheres during this process. Then the left and right fiber ends were moved toward each other until an overlap of 2r was achieved at the touching region of the two fiber ends via the fusion splicer’s left and right motors. The diameters of two microspheres are 207.2 µm, 202.5 µm, respectively, the overlap of two microspheres is 84.7 µm, the length of the SMF is L, and the microscope pictures are the structure with L = 560 µm, as shown in Fig. 1(c). Due to the existence of the microcap structure, it is difficult to find the interface between the microcap structure and the SMF. Therefore, the SMF without discharge at the end is used for measurement in this article, and the shortening of the SMF caused by the discharge is ignored.
Fig. 1. (a). The manufacture of ECU, (b). The manufacture of OEU, (c). The optical diagram and microscope picture.
The next part is the optical enhancement unit (OEU), which plays a crucial role in this microcap-based MI. A comparison experiment of arc times at the fiber end was established to obtain the perfect arc parameters. Before the experiment, lower arc power of −20 bit and a shorter arc duration time of 800 ms were applied to achieve a smaller shape change. As shown in Fig. 1(b), the structure was placed in one of the fiber holders of a fusion splicer, and the fiber end was flush with the electrodes by adjusting the motor of the fusion splicer. We established four consecutive discharge experiments on one structure with L = 560 µm, and after each discharge, the reflection spectrum and the microscope images were recorded, as shown in Fig. 2.
Fig. 2. The relationship between the different arc times at end face of fiber with Free Spectral Range (FSR) and Extinction Ratio (ER) (L = 560µm)
This is a typical MI fiber-optic sensor, using peanut shape as ECU. The intensity of the interference fringe can be written as:
The FSR of this microcap-based MI sensor can be obtained:
As shown in Fig. 3, we designed simulation experiments on the microcap-based MI sensors with different lengths of SMF to get the deflection angle of cladding mode. In the simulation experiment, we set the fiber's diameter as 125 µm, the diameter of the fiber core as 8.2 µm, the relative refractive index of the cladding as 1.445, and the relative refractive index of the fiber core as 1.4502. The light field's diameter is expanded at the peanut shape structure, and part of the light is coupled into the cladding fiber. As shown in the Fig. 1(c), the deflection angle between the cladding mode and the core mode is φ, which is measured as 3.4 in the simulation experiment. Because of the existence of φ, there is no interference fringe in Fig. 2(f). When the value of θ is also relatively small, formula (2) can be simplified as: $n_{eff}^{co}$− $n_{eff}^{cl}$, the effective refractive index difference between core and cladding of fiber, is approximately equal to 0.0052, which is quite smaller compare with $n_{eff}^{cl}$ ≈ 1.445 [15]. Though the L is about 10 times longer than d, the δ depends a lot on the value of 2(d − e)$n_{eff}^{cl}$ rather than the value of (L − D)($n_{eff}^{co}$− $n_{eff}^{cl}$). However, when the reflection angle of θ increases so that it cannot be ignored, although the increase of d leads to the increase of core mode optical path during the continuous discharge process, the increase of e also leads to increased cladding mode optical path. More importantly, the increase of θ will cause the interference to occur early, which will lead to a sharp drop in the optical path of the core mode, which leads to the increase of FSR in Fig. 2. As shown in Figs. 2(b)–2(e), the ER changes with the different curvature of microcap, and according to formula (1), it can be described as: where the Ico + Icl represent the total power of reflected light, which is a constant value. With the curvature of the microcap increasing, the cladding mode light decreases as the probability of light reflection entering the cladding becomes smaller. The core mode light becomes higher because part of the cladding mode light reflects the core mode. We can achieve the largest ER when the value of Ico and Icl are close to each other. In our comparison experiments, the microcap in Fig. 2(c) can achieve the best light intensity ratio, representing that twice discharge is the best discharge parameter. Based on formulas (2) and (5), we have verified the theory with data, as shown in Table 1. It can be seen from this that our theory can explain the phenomenon that appears in the experiment.Fig. 3. Light field simulation diagrams of microcap-based MI sensors with different lengths of SMF (Abscissa unit: (µm), ordinate unit: (µm); (a): 120 µm, (b): 310 µm, (c): 560 µm, (d): 680 µm, (e): 920 µm)
Table 1. Simulation data of different discharge parameters
After obtaining the best arc times, another interesting parameter is the L, representing the length of SMF remaining in the structure. As shown in Figs. 4(a)–4(e), we also established the comparison experiment of different L of 120 µm, 310 µm, 560 µm, 680 µm, and 920 µm with two arc times. In Fig. 4(a), we can hardly find obvious interference fringe, and as for Figs. 4(b) and 4(d), the ERs are not high. In Fig. 4(e), there is a huge total power loss accrued. It can be seen that FSR decreases with the increase of length, except for Fig. 4(e), and the reason can be found in Fig. 5.
Fig. 5. Different reflection of cladding modes corresponding to microcap-based MI sensors with different SMF lengths.
Figure 5 shows the respective reflections of the cladding modes in 5 sensors with different SMF lengths. Because these sensors have the same ECO, they have the same cladding mode deflection angle of φ. As the SMF length L increases, the reflection point of the cladding mode on the microcap structure is getting closer to the edge of the microcap, resulting in a decrease in e, which leads to the decrease of FSR in Fig. 4.
In the simulation experiment with the SMF length of 920 µm, due to the existence of θ, the cladding mode propagates to the boundary between the cladding and the air. Part of the light transmission results in the loss of the intensity of the reflected spectrum, as shown in Fig. 5. Meanwhile, the secondary reflection also results in the reduction of the optical path of the cladding mode. We have listed the simulation data based on formulas (2) and (5) in Table 2, from which it can be seen that the theory can be in good agreement with the experimental results. As a result, a structure with a length of 560 µm was selected for the next experiment. The sizes of some common MI sensors are listed in Table 3, in which we can find that this microcap-based MI sensor is much smaller than ordinary MI sensors.
The OEU allows this microcap-based MI sensor to achieve temperature sensing because of the thermo-optic effect and thermal expansion effect of the SMF. A dip like λdip in formula (4) is tranced to research its changes with the temperature changes:
Table 2. Simulation data of different SMF lengths
Table 3. The size of the current MI sensor
3. Experiment and results
We explored the response of the microcap-based improved MI sensor to applied a high-temperature test. Firstly, one end of the microcap-based MI sensor with an SMF length of 560 µm, as illustrated in Fig. 6, was fixed in a muffle furnace with a built-in thermocouple, and another end was attached with a 3 dB coupler. A broadband light source (BBS) and an optical spectrum analyzer (OSA, YOKOGAWA, AQ6370D) with a resolution of 0.01 nm were utilized as the incident light and monitored the reflection spectrum, respectively. Then, two-cycle tests including heating up and cooling down were established. As shown in Fig. 7, the shift of wavelength of the interference fringe around 1545 nm was monitored while the testing structure was heating up from 100 to 900 °C in the steps of 50 °C and cooling down back to 100 °C in the same steps. The stay time at each temperature step was about 30 minutes, which was to equilibrate spatial temperature distribution and avoid the noises’ effect. Besides, the sensor was maintained at 900 °C for 100 minutes to verify the robustness of this microcap-based MI sensor.
Figure 7 shows the reflection spectrum response of the microcap-based MI sensor. As the temperature increased from 100 °C to 900 °C, the wavelength of interference fringe around 1545 nm showed a redshift. The wavelength dip shift from 1544.302 nm to 1553.322 nm when reaching 900 °C. To ensure the rigor of this sensor's temperature test, the heating up and cooling down cycle has been repeated 2 times. The wavelength of the same dip was tranced, and the result obtained is illustrated in Fig. 8. The response of dip wavelength of the interference fringe shows a nonlinear characterization during the heating-cooling cycle process. As shown in formulas (2) and (4), the microcap-based MI sensor's response with temperature increasing depends on the RI and length of the SMF; moreover, both the two values change with temperature changes. Hence, the nonlinear phenomenon appeared in the temperature test.
The temperature range was divided into two parts with good linear fitting to research the temperature sensitivity of this microcap-based MI sensor, the first-temperature region of 100-500 °C, and the second-temperature region of 500-900 °C, respectively, as shown in Fig. 9. The reflectance spectrum of the microcap-based MI sensor had a redshift with the temperature sensitivity of 10.7 pm/°C, 10.79 pm/°C, 10.87 pm/°C, and 10.55 pm/°C with the linearity of 0.98139, 0.98167, 0.98426, and 0.98243 in 100-500 °C. As for the second-temperature range, the temperature sensitivities were 11.08 pm/°C, 11.19 pm/°C, 10.95 pm/°C, and 11.36 pm/°C with the linearity of 0.99735, 0.99437, 0.99681, and 0.99695 respectively, which were slightly higher than that of the first-temperature region. The maximum sensitivity repeatability errors of the two-cycle are respectively 0.32 pm/°C and 0.11 pm/°C for the first-temperature region and the second-temperature region, which indicates the good temperature repeatability in a large dynamic range of this microscope-based improved MI sensor.
Fig. 9. (a) Temperature response of microcap-based MI within 100-500 °C. (B) Temperature response of microcap-based MI within 500-900 °C.
In order to test the reliability of this microcap-based MI sensor in extreme high-temperature, a stability test within 100 minutes was performed at 900 °C. The same wavelength dip was tranced and recorded every 20 minutes, as illustrated in Fig. 10. The traced wavelength errors of the six experiments data are all within ${\pm} $0.2 nm, which revealed the microcap-based MI sensor's high robustness. The disordered shift of the tranced wavelength was observed, and it is proved that the cause of the wavelength shift is not the change of material properties but the result of the muffle furnace's cyclic heating process.
4. Conclusion
In summary, we proposed an ultra-compact microcap-based improved MI sensor for high-temperature measurement, which much shorter than the length of current MI has been created using a microcap structure. In contrast to other optical fiber sensors fabricated by expensive photonic crystal fiber or via femtosecond laser, our sensor possesses the advantages of low cost and ease of production. The ultra-short MI sensor's response with a length of 560 µm underwent two-cycle tests, and stably endured 900 °C high-temperature was obtained. The highest temperature sensitivity changed from 10.87 pm/°C in 100-500 °C to 11.36 pm/°C in 500-900 °C. The results show the sensor has good repeatability, reversibility, and reliability and possesses a potential application prospect for high-temperature measurement.
Funding
National Key Research and Development Program of China (2018YFB1801004); National Natural Science Foundation of China (61975084, 62005125, 61835005, 61822507, 61775098, 61727817, 61720106015, 61875248, 61935011, 61935005); Open Fund of IPOC (BUPT); Jiangsu talent of innovation and entrepreneurship; Jiangsu team of innovation and entrepreneurship; Postgraduate Research & Practice Innovation Program of Jiangsu Province (SJCX20_0296).
Disclosures
The authors declare no conflicts of interest.
References
1. X. Tan, Y. Geng, X. Li, R. Gao, and Z. Yin, “High temperature microstructured fiber sensor based on a partial-reflection-enabled intrinsic Fabry–Perot interferometer,” Appl. Opt. 52(34), 8195 (2013). [CrossRef]
2. S. Liu, Y. Wang, C. Liao, G. Wang, Z. Li, Q. Wang, J. Zhou, K. Yang, X. Zhong, J. Zhao, and J. Tang, “High-sensitivity strain sensor based on in-fiber improved Fabry–Perot interferometer,” Opt. Lett. 39(7), 2121 (2014). [CrossRef]
3. L. V. Nguyen, D. Hwang, S. Moon, D. S. Moon, and Y. Chung, “High temperature fiber sensor with high sensitivity based on core diameter mismatch,” Opt. Express 16(15), 11369 (2008). [CrossRef]
4. S. C. Warren-Smith, E. P. Schartner, L. V. Nguyen, D. E. Otten, Z. Yu, D. G. Lancaster, and H. Ebendorff-Heidepriem, “Stability of Grating-Based Optical Fiber Sensors at High Temperature,” IEEE Sensors J. 19(8), 2978–2983 (2019). [CrossRef]
5. J. Deng, D. N. Wang, and H. Zhang, “Femtosecond Laser Inscribed Multiple In-Fiber Reflection Mirrors for High-Temperature Sensing,” J. Lightwave Technol. 37(21), 5537–5541 (2019). [CrossRef]
6. S. Islam, H. Bakhtiar, Z. Haider, A. Alshoaibi, S. Riaz, and S. Naseem, “Silica-titania nanocomposite based fiber optic sensor for aromatic hydrocarbons detection,” Opt. Commun. 471, 125825 (2020). [CrossRef]
7. N. K. Radhika, B. S. Kavitha, S. Asokan, and S. S. Gorthi, “Detection of copper nanoparticles templated by DNA using etched Fibre Bragg Grating sensor,” IEEE Sensors J. 20(16), 9179–9186 (2020). [CrossRef]
8. J. Hribar and D. Donlagic, “Fiber-Optic Boiling Point Sensor for Characterization of Liquids,” IEEE Sensors J. 20(14), 7731–7739 (2020). [CrossRef]
9. X. Guo, J. Zhou, C. Du, and X. Wang, “Optical Fiber Ultrasound Probe for Radiofrequency Ablation Temperature Monitoring: In-Vitro Results,” IEEE Photonics Technol. Lett. 32(12), 689–692 (2020). [CrossRef]
10. S. Chen, Y. Liu, Q. Yu, and W. Peng, “Microcapillary-Based Integrated LSPR Device for Refractive Index Detection and Biosensing,” J. Lightwave Technol. 38(8), 2485–2492 (2020). [CrossRef]
11. C. Yue, Y. Lang, X. Zhou, and Q. Liu, “Sensitivity enhancement of an SPR biosensor with a graphene and blue phosphorene/transition metal dichalcogenides hybrid nanostructure,” Appl. Opt. 58(34), 9411 (2019). [CrossRef]
12. G. Zhang, X. Wu, S. Li, W. Liu, S. Fang, C. Zuo, W. Zhang, and B. Yu, “Miniaturized Fabry–Perot probe utilizing PMPCF for high temperature measurement,” Appl. Opt. 59(3), 873 (2020). [CrossRef]
13. H. Su, Y. Zhang, K. Ma, Y.-P. Zhao, and C. Yu, “Tip Packaged High-Temperature Miniature Sensor Based on Suspended Core Optical Fiber,” J. Lightwave Technol. 38(15), 4160–4165 (2020). [CrossRef]
14. X. L. Cui, H. Zhang, and D. N. Wang, “Parallel structured optical fiber in-line Fabry–Perot interferometers for high temperature sensing,” Opt. Lett. 45(3), 726 (2020). [CrossRef]
15. E. Li, “Design and test of multimode interference based optical fiber temperature sensors,” Proc. SPIE 7157, 71570F(2008). [CrossRef]
16. L. Xu, L. Jiang, S. Wang, B. Li, and Y. Lu, “High-temperature sensor based on an abrupt-taper Michelson interferometer in single-mode fiber,” Appl. Opt. 52(10), 2038 (2013). [CrossRef]
17. D. Wu, T. Zhu, and M. Liu, “A high temperature sensor based on a peanut-shape structure Michelson interferometer,” Opt. Commun. 285(24), 5085–5088 (2012). [CrossRef]
18. L. Yuan, J. Yang, Z. Liu, and J. Sun, “In-fiber integrated Michelson interferometer,” Opt. Lett. 31(18), 2692 (2006). [CrossRef]
19. Z. Tian, S. S.-H. Yam, and H.-P. Loock, “Refractive index sensor based on an abrupt taper Michelson interferometer in a single-mode fiber,” Opt. Lett. 33(10), 1105 (2008). [CrossRef]
20. F. Peng, J. Yang, X. Li, Y. Yuan, B. Wu, A. Zhou, and L. Yuan, “In-fiber integrated accelerometer,” Opt. Lett. 36(11), 2056 (2011). [CrossRef]
21. P. Fan, W. Yan, P. Lu, W. Zhang, W. Zhang, X. Fu, and J. Zhang, “High sensitivity fiber-optic Michelson interferometric low-frequency acoustic sensor based on a gold diaphragm,” Opt. Express 28(17), 25238 (2020). [CrossRef]
22. Z. Li, Y.-X. Zhang, W.-G. Zhang, L.-X. Kong, Y. Yue, and T.-Y. Yan, “Parallelized fiber Michelson interferometers with advanced curvature sensitivity plus abated temperature crosstalk,” Opt. Lett. 45(18), 4996 (2020). [CrossRef]
23. T. Zhu, K. S. Chiang, Y. J. Rao, C. H. Shi, Y. Song, and M. Liu, “Characterization of Long-Period Fiber Gratings Written by CO2 Laser in Twisted Single-Mode Fibers,” J. Lightwave Technol. 27(21), 4863–4869 (2009). [CrossRef]
