We propose and demonstrate a thinned fiber based Mach-Zehnder interferometer for multi-purpose sensing applications. The sensor head is formed by all-fiber in-line singlemode-multimode-thinned-singlemode (SMTS) fiber structure, only using the splicing method. The principle of operation relies on the effect that the thinned fiber cladding modes interference with the core mode by employing a multimode fiber as a mode coupler. Experimental results showed that the liquid refractive index information can be simultaneously provided from measuring the sensitivity of the liquid level. A 9.00 mm long thinned fiber sensor at a wavelength of 1538.7228 nm exhibits a water level sensitivity of −175.8 pm/mm, and refractive index sensitivity as high as −1868.42 (pm/mm)/RIU, respectively. The measuring method is novel, for the first time to our knowledge. In addition, it also demonstrates that by monitoring the wavelength shift, the sensor at a wavelength of 1566.4785 nm exhibits a refractive index sensitivity of −25.2935 nm/RIU, temperature sensitivity of 0.0615 nm/°C, and axial strain sensitivity of −2.99 pm/με, respectively. Moreover, the sensor fabrication process is very simple and cost effective.
© 2012 OSA
Optical fiber sensors have attracted great attentions in the applications of biological, chemical and environmental industries, including the measurements of the liquid level, refractive index (RI), temperature and strain. Compared to the other techniques that based on the mechanical and electrical methods, the optical fiber sensors have many advantages, such as electromagnetic immunity, resistance to erosion, high sensitivity and capability of remote sensing. In the past years, several types of fiber optical sensors have been developed. For example, the sensors based on the long period fiber gratings [1–3] and fiber Bragg gratings [4,5] have been widely reported. These sensors possess advantages such as absolute response parameter, large dynamic range and high sensitivities. However, they have large cross sensitivities and the fabrication requires the expensive ultraviolet light laser, phase masks, etc. Sensors based on all fiber interferometer have also been typically used [6–8], owing to high sensitivity, simple fabrication process and un-limited measuring of wavelength range. Such as the in-line Fabry-Perot interferometer sensor based on the endlessly photonic crystal fiber , but it still requires expensive photonic crystal fiber. Furthermore, the sensor head is very fragile.
Recently, sensors based on all-fiber Mach-Zehnder interferometer (MZI) have received considerable attention [10–13], they are compact and robust. Linh et al. presented a multimode-singlemode-multimode (MSM) fiber structure based MZI sensor for high temperature measurement , but the lengths of the MMFs should be perfectly controlled. Wu Q, et al. reported a high sensitivity singlemode-multimode-singlemode (SMS) fiber structure based refract-meter . Nevertheless, the sensors based on the etching process have to use poison chemicals in the fabrication process and the diameter of the exposed core is hard to be controlled.
In this letter, we propose a novel approach, employing a small core and cladding diameters thinned fiber, with singlemode-multimode-thinned-singlemode (SMTS) fiber structure based Mach-Zehnder interferometer for the sensor application of liquid level, refractive index, temperature and axial strain. Since the mode field diameter of the multimode fiber is much larger than that of the thinned fiber, the core mode and cladding modes of the thinned fiber could be excited at the same time. The principle of our structure is different from the modal interferometers [16–21] in which just the guided modes in core were excited and the beating between the first two fiber modes to exhibit an oscillatory pattern [14,19]. Moreover, compared with the modal interferometers in temperature sensitivity (less than 0.01nm/°C) [20,21], the sensitivity of our structure (as high as 0.0615nm/°C) is higher. It is due to the fact that the guided modes of the modal interferometers propagate primarily in the same core, thus they have similar thermal coefficients . Theoretically, the interference between the eigenmodes of the multimode fiber is also included in the interference spectrum. However, by taking the fast Fourier transformation (FFT), we find that only the low-order cladding mode of the thinned fiber is the dominant and mainly interfered with the core mode. It may be due to the reason that the interference between the multimode fiber eigenmodes has very large free spectral range (FSR), and it does not fall into the measured transmission spectra range. In this paper, a new method based on liquid level sensitivity monitoring for the measurement of refractive index is also proposed. Experimental results indicate that the sensitivity of the sensor can reach up to −1868.42 (pm / mm) / RIU. It is also notable that the fabrication of the sensor is simple and inexpensive, including only the fusion splicing.
2. Schematic diagram and properties of the sensor
The schematic of the sensor is set up in Fig. 1 . A section of a thinned fiber (TF) is spliced to a multi-mode fiber (MMF), these sections are, in their turn, spliced between two single-mode fibers (SMF). The TF has a core refractive index of 1.4735 with high Ge-doped and mode field diameter about 4.5μm at 1550nm, and it also has a small cladding diameter of 80μm. The core / cladding diameters of the standard SMF, step index MMF are 9.2/125μm and 50/125μm, respectively. These used fibers are all produced by the Yangtze Optical Fiber and Cable Company Ltd. When the light is launched into the MMF through the lead-in SMF, at the MMF-TF spliced point, a part of the power is coupled to the cladding modes of the TF due to the mode field mismatch. It will excite multiple cladding modes propagating within the cladding of the TF. Similarly, at the TF-SMF splice point, part of the TF cladding modes that coupled back into the core of the SMF interfere with the TF core mode. Consequently, they are coupled into the fundamental mode of the lead-out fiber. The fibers are all fusion spliced with the AUTO MODE in the splicer menu (FSM 60S). It should be pointed out that although the splicing method is so easy, it still needs carefully cleaving and fusion splicing procedures. The splicing loss could reduce the coupling ratio, which influences the fringe visibility and transmission loss of the interference spectrum. It is well known that when the cladding diameter is reduced, a thinned fiber can enhance the fraction of power in the evanescent wave (in the cladding), thus the light propagating from the cladding of the TF will interfere with the core mode more effectively. Furthermore, the effective refractive index of the cladding will show a change depending on the surrounding environment. There will be an interaction between the evanescent wave of the fundamental guided mode and the surrounding environment, leading to a shift in the transmission spectrum .
In order to improve the effect of the MMF in the structure, experimental results based on the singlemode-thinned-singlemode (STS) fiber structure and the SMTS fiber structure are shown in Figs. 2(a) and 2(b), respectively. The length of the TF in both cases is kept at 48.38mm. Compared to the STS, the interference fringe visibility of the SMTS is improved by several times. It is accredited to the mode field diameter of the MMF being much larger than SMF, thus, the power of the light injected to the cladding mode of the TF is dramatically enhanced. Moreover, the cladding modes of the TF could be excited due to the using of MMF, which is different from the STS structure in  that the core modes LP01 and LP11 were excited through the mode field mismatch between two SMFs and the thinned fiber.
Additionally, we also have researched the influence of the MMF length to the sensor performance during our experiment. Theoretically, the length of the MMF determines the transverse field distribution at the interface MMF-TF and influences the coupling strengths of MMF core modes to the TF cladding / core modes . However, when this length is long enough, such as it is above several centimeters, we find that this influence on the interference fringe is very slight through our experiments. It can be seen from Fig. 3 with a sensor TF length of 18.02mm in the air, the length of the MMF is changed as 22cm, 32cm and 40cm, respectively. It is due to the fact that the interference spectrum is mainly formed by the TF cladding modes interfering with the TF core mode. The interference between the eigenmodes in the MMF core has very large FSR , and it may not fall into the measured transmission spectra range. Thus, it only modifies the envelope of the interference when the length of the MMF increases. From Refs. [15,23], we also can obtain that the length of the MMF has no significant influence on the sensor sensitivity
Sensors with TF length of L = 4.62mm, 9.00mm, 27.42mm and 54.54mm SMTS fiber structure are fabricated, and the transmission spectra are shown in Fig. 4 . It can be seen that the FSR will decrease as the interferometer length L increases, and the interference fringe of the sensor is not uniform but looks like having several frequency components in the interference fringe periods.
In order to analyze the number and power distribution of the interference modes, the wavelength spectra in Fig. 4 are Fourier transformed  to get the spatial frequency spectra in Fig. 5 . The dominant intensity peak at zero relates to the core modes. The powers are primarily distributed in the core mode and the low-order cladding mode as the length varies. It means that the mode coupling of different length mainly occurs between the core mode and the low-order cladding mode. The multiple minor intensity peaks in the inset of Fig. 5 correspond to the high-order cladding modes. Those interferences between the core mode and the high-order cladding modes also modify the envelope of the interference in Fig. 4. It is also observed that when the sensor is put in the index matching oil to remove the cladding modes, the interference phenomenon almost disappears, which can be seen from Fig. 3 with the MMF length of 22cm. It is again believed that the cladding modes are excited different from the core mode LP11 in Ref. .
Since the interference is mainly formed by two modes (the dominant low-order cladding mode and the core mode). The structure still can be considered as a Mach-Zehnder interferometer . When the sensor is in the air, the transmission spectrum of MZI can be expressed asEq. (1), when the interference signal reaches its minimum, the wavelength of the mth order attenuation peak can be written as:Eq. (2), the wavelength spacing between two interference minima (FSR) can be approximated as:Fig. 3 are consistent with the theory.
3. Sensing applications and discussion
In the Section 2, the design and principle of the SMTS fiber structure sensor has been presented. In the following part, we will describe and discuss the applications of this sensor in the measurement of liquid level, refractive index, temperature and axial strain with the TF length L of 9.00mm. The transmission spectrum of the sensor in the air is shown in Fig. 4(b). In these experiments, a broadband optical source (C and L band) is used to inject light into the structure, and an optical spectrum analyzer (YOKOGAWA AQ6370C) is used to measure the transmission spectral response of the sensor.
3.1 Liquid level sensor
When a section of the TF is surrounded by a liquid, the effective refractive index of the cladding mode increases and that the core mode hardly disturbs, therefore the difference of the effective refractive indices between the core and the cladding modes will decrease . From Eq. (2), the attenuation peaks wavelength will shift to a short wavelength. Nevertheless, the sensor is not entirely covered with a liquid. The phase difference between the core and the cladding modes is determined by the contribution of the two sections. The wavelength of the mth order interference valley can be written as:Eq. (4), it is known that when the sensor is placed in the vertical direction, the change in the liquid level can be easily detected by monitoring the wavelength shifts. The attenuation peak wavelengths will shift to shorter wavelengths as the liquid level is increased. Because the refractive indices of the liquid is larger than that of air, and the difference of the effective refractive indices between the core and cladding modes becomes smaller with increasing liquid level. According to Eq. (4), the contribution of the second part will increase, leading to the peak shifts to the shorter wavelength. Additionally, the detecting liquid level sensitivity of the sensor is related to the refractive index of the certain liquid. Therefore, the RI information of the liquid also can be provided from measuring the level sensitivity of the certain liquid.
We experimentally examine the liquid level and refractive index with the length L of 9.00mm. Figure 6(a) shows the schematic diagram of the experimental system. The sensor is set on a fixing skeleton, which is placed vertically inside the beaker. The lead-in and lead-out fibers are connected with a broad band optical source and an optical spectrum analyzer, respectively. The liquid level is increased by using a buret step by step, which is metered by the vernier caliper. The transmission spectrum is measured as the liquid level rises slowly. When the spectrum starts to shift, it is chosen as the initial state, and the level is marked as the reference liquid level.
The response of the sensor to the liquid level is firstly investigated in the water with refractive index of n = 1.3345, while the temperature is kept at the room temperature (20°C). Figure 6(b) shows the transmission spectra measured with liquid level in a range of 0.00mm to 9.00mm. As expected, the attenuation peak wavelengths shift to shorter wavelengths as the water level is increased. The peak at a wavelength of 1538.7228nm (order 116) is chosen to record. The result is shown in the inset of Fig. 6(b). A wavelength blueshift about 1.5761nm occurs with the water level changing from 0.00mm to 9.00mm. The sensitivity of this peak to the water level is −175.8 pm / mm using the linear regression fits. The coefficients of the attenuation peak to the different refractive indexes n = 1.3440 and n = 1.3775 are also analyzed by using linear regression fits in Fig. 6(c). It is noted that they exhibit a similar linear response, and the sensitivity of the sensoris related to the refractive index of the liquid.
The relationship between the sensitivityand the refractive indexis also investigated in detail by putting the sensor into different liquids, at range from 1.3345 to 1.3775. The temperature is still kept at 20 °C. Figure 7 shows the linear fitting lines of the level sensitivity with respect to the refractive index, and the regression equation can be expressed as26]. An average sensitivity of −1868.4233 (pm / mm) / RIU is obtained in our experiment with the sensing range from 1.3345 to 1.3775. The linearity of the -curve is about 0.99278, indicating a high linearity of the SMTS sensor. This provides another advantage to our sensor, because we can measure the surrounding liquid refractive index by monitoring the sensitivity of the sensor. This method is quite different from the traditional way to measure the RI based on monitoring the wavelength shift  or the changing of the fringe visibility , for the first time to our knowledge.
Therefore, the procedure of measurements of liquid level and refractive index can be the following: Firstly, the RI information can be obtained by dipping the sensor entirely into the liquid for the measurement of the liquid level sensitivity, since the cavity length L of MZI and the wavelength shifts of both in the air and in the liquid are all known. And then, the sensor is placed vertically and fixed for the monitoring the changes of liquid level. The largest sensing range is limited by the cavity length L. We believe that this range can be as large as up to tens of centimeters, for we observe that the interference fringes still exist there. When the outside refractive index approaches to that of the thinned fiber cladding (around 1.45), the interference fringes will disappear, thus it limits the maximum sensing range.
3.2 Refractive index sensor
From the Section 3 part 1, it is known that the attenuation peak wavelength will shift to a short wavelength when the refractive index surrounding the thinned fiber rises. The response of the sensor to the external refractive index is investigated by putting the sensor entirely into the NaCl solution, while the temperature is kept at 20 °C. The schematic diagram of the experimental setup for the refractive index is shown in Fig. 6(a) with TF length L of 9.00mm. The RI of the solutions is calibrated by an Abbe refractometer with the resolution of 0.0001.
Figure 8(a) shows the wavelength shift of transmission spectra as the external refractive index increases from 1.3345 to 1.3775. As shown in the inset, the selected order (114) experiences 1.0739 nm blueshift as the surrounding refractive index increases. It is due to the fact that the accumulated difference of the effective refractive indices between the core and the cladding modes will decrease more for a higher outside refractive index. Therefore, the attenuation peak will shift to a shorter wavelength. The measured RI responses to different orders of attenuation peaks are analyzed in Fig. 8(b). The sensitivities of the selected order attenuation peaks are analyzed by using linear regression fits, and they exhibit good linear responses. For the wavelengths selected at 1538.7228nm (order 116), 1551.5819nm (order 115) and 1566.4785nm (order 114), the average corresponding sensitivities are −17.5224nm/RIU, −22.5952nm/RIU and −25.2935nm/RIU, respectively. The experimental results show that the lower order attenuation peak exhibits a larger wavelength shift than the higher order peak.
3.3 Temperature sensor
When the external temperature around the TF rises, both the effective refractive index of the core and cladding modes increase. While that of the core mode changes more since the thermo-optic coefficient of the Ge-doped silica core is higher than that of the cladding consisting of fused silica. As result, the effective refractive index indices between the core and cladding modes increase , and from Eq. (2), it is known that the attenuation peak wavelength will change to a longer wavelength.
For the temperature measurement, Fig. 9(a) illustrates the schematic diagram of the experimental setup with TF sensor length of 9.00mm. The sensor is put into the water bath with temperature increasing from 20°C to 80°C. The attenuation peak at a wavelength of 1566.4785nm (order 114) is chosen to record, and the peak experiences 3.8923nm redshift as the temperature increases which is shown in the Fig. 9(b). The Fig. 9(c) shows the measured temperature response of the selected order peak. The coefficient of the peak is analyzed using the linear regression fits, as 0.0667nm/°C.
It is known that the refractive index (RI) of the water decreases when the temperature increases . So the coefficient of the temperature is the sum of the direct contribution of the temperature change and the indirect contribution of the induced change of the RI of the water.
Assuming the temperature sensitivity of the sensor can be linearly approximated as :29] when the temperature increases from 20°C to 80°C, thus the pure sensitivity of the temperature can be calculated from Eq. (6) as 0.0615nm/°C.
3.4 Axial strain sensor
Besides the above sensing applications, the sensor also can be used in the axial strain measurement. The schematic diagram of the experimental setup is the same as Fig. 9(a), while the container is filled with air, and the experiment is conducted at the room temperature (20°C). When the axial strain applied on the SMTS fiber structure sensor is increased from 0 με to 1000με, the selected attenuation peak shifts to a shorter wavelength. It is known that when the axial strain increases, the length of sensor will increase. From Eq. (2), the attenuation peak wavelength will shift to a longer wavelength. At the same time, the decreases in the ratio of the fiber core and cladding result in the difference in the effective refractive indices between the core and the cladding modes reduce. The influence of increase in the sensor length has a weaker impact than the decrease in the, so the peak wavelength will have a blueshift with axial strain increasing. Figure 10 shows the axial strain response of the sensor using the linear regression fits. It is noted that the average axial strain sensitivity of the selected peak is about −2.99pm / με,.
Finally, it is worth noting that the specific peak order for the measurement can be achieved from accurately controlling the length of the TF from Eq. (2). The resolution of the liquid level, refractive index, temperature and axial strain response also depends on the interference order and the length of the MZI. The low order and long length of the TF can increase the sensitivity, which can be obtained from Eq. (2). However, it should be noted that longer TF sections increase the interferometer sensitivity, but the sensing range decreases accordingly because the fringes become much more closely spaced in wavelength. From the Ref  and our experimental process, it is also known that the layout of MMF does not influence the measurement results in practical application. If the length of MMF is long, after the part of MMF (~8cm) is placed straightly, the extra MMF still can be coiled. In fact, all the sensors [14,16,25] that exploited by the same effect from the difference in effective index of Eq. (2) will also exhibit the similar sensitivity. It is simply based on the fact the sensitivity of the interferometer is a function of the group index difference induced by the interaction of the propagated mode(s) with the environment. The sensitivity depends therefore trivially on the length of the interferometer arm and on the fraction of the mode volume that propagates in the evanescent wave. Long interferometers propagating high-index cladding modes (or core modes through a nanowire) are therefore more sensitive.
In conclusion, we have proposed and demonstrated a novel and simple in-line all-fiber sensor for multi-purpose sensing applications. The sensor is based on the thinned fiber Mach-Zehnder interferometers by using singlemode-multimode-thinned-singlemode (SMTS) fiber structure. Due to the MMF in the structure, the cladding modes of the TF could be excited, and the fringe visibility of the spectrum could be improved by several times. The principle of the operation relies on the interference of the core mode and cladding modes of the thinned fiber. Experiments show that by measuring the shifts of the arbitrarily selected interference order with the changing of the external environment, the sensitivities of the liquid level, refractive index, temperature and axial strain can be experimentally measured. It should be noted that the fiber sensor can also detect the refractive index by monitoring the sensitivity of the sensor to the liquid level, for the first time to our knowledge. Moreover, the proposed sensor also can operate in the reflection mode by coating the end of the TF or lead-out SMF with a mirror. Finally, since the fabrication is so easy, safe and cost effective, includes only the fusion splicing that it makes the device properly attractive for physical, biological and chemical sensing in practical applications.
This work is supported by the National Natural Science Foundation of China (60937002), International Cooperation Projects between China and Singapore (2009DFA12640) and Professional Talents Fund (0124182015, Huazhong University of Science and Technology).
References and links
1. A. P. Zhang, L. Y. Shao, J. F. Ding, and S. He, “Sandwiched long-period gratings for simultaneous measurement of refractive index and temperature,” IEEE Photon. Technol. Lett. 17(11), 2397–2399 (2005). [CrossRef]
2. H. J. Patrick, A. D. Kersey, F. Bucholtz, K. J. Ewing, J. B. Judkins, and A. M. Vengsarkar, ““Chemical sensor based on long-period fiber grating response to index of refraction,” Proc. Lasers and Electro-Optics. 11, 420–421 (1997).
3. P. L. Swart, “Long-period grating Michelson refractometric sensor,” Meas. Sci. Technol. 15(8), 1576–1580 (2004). [CrossRef]
4. A. Iadicicco, S. Campopiano, A. Cutolo, M. Giordano, and A. Cusano, “Nonuniform thinned fiber bragg gratings for simultaneous refractive index and temperature measurements,” IEEE Photon. Technol. Lett. 17(7), 1495–1497 (2005). [CrossRef]
5. P. Lu and Q. Chen, “Fiber Bragg grating sensor for simultaneous measurement of flow rate and direction,” Meas. Sci. Technol. 19(12), 125302–125309 (2008). [CrossRef]
6. Y. J. Rao, “Recent progress in fiber optic extrinsic Fabry-Perot interferometric sensors,” Opt. Fiber Technol. 12(3), 227–237 (2006). [CrossRef]
7. Z. L. Ran, Y. J. Rao, W. J. Liu, X. Liao, and K. S. Chiang, “Laser-micromachined Fabry-Perot optical fiber tip sensor for high-resolution temperature-independent measurement of refractive index,” Opt. Express 16(3), 2252–2263 (2008). [CrossRef] [PubMed]
8. Y. J. Rao, M. Deng, T. Zhu, and H. Li, “In-line Fabry-Perot Etalons based on hollow-core photonic bandgap fibers for high temperature applications,” J. Lightwave Technol. 27(19), 4360–4365 (2009). [CrossRef]
11. J. H. Lim, H. S. Jang, K. S. Lee, J. C. Kim, and B. H. Lee, “Mach-Zehnder interferometer formed in a photonic crystal fiber based on a pair of long-period fiber gratings,” Opt. Lett. 29(4), 346–348 (2004). [CrossRef] [PubMed]
12. L. Jiang, J. Yang, S. Wang, B. Li, and M. Wang, “Fiber Mach-Zehnder interferometer based on microcavities for high-temperature sensing with high sensitivity,” Opt. Lett. 36(19), 3753–3755 (2011). [CrossRef] [PubMed]
13. Z. Tian, S. S. H. Yam, J. Barnes, W. Bock, P. Greig, J. M. Fraser, H. P. Loock, and R. D. Oleschuk, “Refractive index sensing with Mach-Zehnder interferometer based on concatenating two single-mode fiber tapers,” IEEE Photon. Technol. Lett. 20(8), 626–628 (2008). [CrossRef]
14. 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), 11369–11375 (2008). [CrossRef] [PubMed]
17. P. R. Horche, M. Lopez-Amo, M. A. Muriel, and J. A. Martin-Pereda, “Spectral behavior of a low-cost all-fiber component based on untapered multifiber unions,” IEEE Photon. Technol. Lett. 1(7), 184–187 (1989). [CrossRef]
18. X. Daxhelet, J. Bures, and R. Maciejko, “Temperature-independent all-fiber modal interferometer,” Opt. Fiber Technol. 1(4), 373–376 (1995). [CrossRef]
19. J. Villatoro, V. P. Minkovich, and D. Monzon-Hernandez, “Compact modal interferometer built with tapered microstructured optical fiber,” IEEE Photon. Technol. Lett. 18(11), 1258–1260 (2006). [CrossRef]
20. J. Villatoro, V. Finazzi, V. P. Minkovich, V. Pruneri, and G. Badenes, “Temperature-insensitive photonic crystal fiber interferometer for absolute strain sensing,” Appl. Phys. Lett. 91(9), 091109 (2007). [CrossRef]
22. C. Zhou, L. Ding, D. Wang, Y. Kuang, and D. Jiang, “Thinned fiber Bragg grating magnetic field sensor with magnetic fluid,” Proc. SPIE 8034, 803409, 803409-6 (2011). [CrossRef]
23. S. M. Nalawade and H. V. Thakur, “Photonic crystal fiber strain-independent temperature sensing based on modal interferometer,” IEEE Photon. Technol. Lett. 23(21), 1600–1602 (2011). [CrossRef]
24. H. Y. Choi, G. Mudhana, K. S. Park, U. C. Paek, and B. H. Lee, “Cross-talk free and ultra-compact fiber optic sensor for simultaneous measurement of temperature and refractive index,” Opt. Express 18(1), 141–149 (2010). [CrossRef] [PubMed]
25. P. Lu, L. Men, K. Sooley, and Q. Chen, “Tapered fiber Mach-Zehnder interferometer for simultaneous measurement of refractive index and temperature,” Appl. Phys. Lett. 94(13), 131110 (2009). [CrossRef]
28. Q. Jiang, D. Hu, and M. Yang, “Simultaneous measurement of liquid level and surrounding refractive index using tilted fiber Bragg grating,” Sens. Actuators A Phys. 170(1-2), 62–65 (2011). [CrossRef]
29. D. R. Lide, Handbook of Chemistry and Physics, 70th ed. (CRC Press, 2004), Chap. 6.
30. J. Yan, A. P. Zhang, L. Y. Shao, J. F. Ding, and S. He, “Simultaneous measurement of refractive index and temperature by using dual long-period gratings with an etching process,” IEEE Sens. J. 7(9), 1360–1361 (2007). [CrossRef]