Research Article
Vol. 55, No. 4 / February 1 2016 / Applied Optics
673
Magnetic field sensor using the fiber loop ring-down technique and an etched fiber coated with magnetic fluid TAO SHEN,1,2,* YUE FENG,2 BINCHAO SUN,2
AND
XINLAO WEI1
1
Key Laboratory of Engineering Dielectrics and Its Application, Ministry of Education, Harbin University of Science and Technology, Harbin 150080, China 2 College of Applied Sciences, Harbin University of Science and Technology, Harbin 150080, China *Corresponding author: [email protected] Received 28 September 2015; revised 8 November 2015; accepted 5 December 2015; posted 15 December 2015 (Doc. ID 250664); published 21 January 2016
The fiber loop ring-down spectroscopy technique is introduced into the evanescent-field-based sensing scheme in order to create a new type of fiber-based magnetic field sensor. As a consequence, the sensitivity and stability of the magnetic field sensing system are significantly enhanced. The sensor head is constructed using a section of a single-mode fiber with its cladding partially etched. The process of fiber etching is described in detail, and the relationship between the diameter of the etched fiber and the etching time is experimentally investigated. After adopting the appropriate size of the etched fiber, the final experimental results show that the magnetic field strength has a well-defined linear relationship with the inverse of the ring-down time τ over a range of 30 mT with a sensitivity of 95.5 ns/mT. © 2016 Optical Society of America OCIS codes: (230.3810) Magneto-optic systems; (060.4005) Microstructured fibers; (060.2370) Fiber optics sensors. http://dx.doi.org/10.1364/AO.55.000673
1. INTRODUCTION Magnetic field sensing plays an important role in various fields such as electric power transmission [1], undersea measurements [2], and the military sector [3]. Sensors based on optical fibers employ a number of different sensor elements, including fiber grating sensors [4–6], fiber-optic Fabry–Perot sensors [7–9], multimode interference fiber sensors [10], fiber Michelson/ Mach–Zehnder interferometric sensors [11,12], and birefringence polarimetric fiber-optic sensors [13,14]. However, these sensors share a common drawback in that their fabrication is costly. On the other hand, evanescent-field- (EF) based fiber magnetic field sensors that use a section of an etched single-mode fiber (SMF) are less costly and are also simpler. Nevertheless, the usual response of the EF to the external magnetic field is relatively weak and is mainly limited by the intensity-based sensing scheme whose detection limit is constrained by light source fluctuations. Accordingly, in this paper, we introduce the fiber loop ring-down spectroscopy (FLRDS) technique into the EF-based fiber sensor in order to improve the detection sensitivity and system stability. The FLRDS technique, which is an absorption spectroscopic and detection technique, has become popular in recent years [15–19] because it can be directly adopted by many sensing schemes. It makes use of an optical cavity, which not only 1559-128X/16/040673-06$15/0$15.00 © 2016 Optical Society of America
realizes a long effective path length through a sample but also eliminates the effect of fluctuations in the light intensity. Through the FLRDS technique, the detection sensitivity of EF-based sensors has been improved significantly. To the best of our knowledge, no other EF-FLRDS-based magnetic field sensor has been reported. Moreover, in this study, a magnetic fluid (MF) is placed around the etched SMF to act as a medium for the magnetic field and evanescent field. MF is a new kind of colloidal liquid that consists of magnetic nanoparticles and a liquid carrier (usually water or an ester) [20]. As a kind of functional medium, MF has various magneto-optical properties; more specifically, MF exhibits a tunable refractive index [21], facilitates the Faraday effect [22], exhibits birefringence [23], and so on. When a varying magnetic field is introduced around the MF, the EF will vary in response to the transmissivity and refractive index of the MF. In this study, an EF-FLRDS-based magnetic field sensor is achieved with a considerable detection sensitivity of 95.5 ns/mT in the linear response range, which is a great improvement over the previously reported ferrofluid-based magnetic field sensors [24]; its advantages include easy fabrication, low cost, and an impressive stability owing to the characteristic that it is insensitive to the intensity fluctuations of the laser source of the FLRDS technique [25].
674
Research Article
Vol. 55, No. 4 / February 1 2016 / Applied Optics
2. PRINCIPLES The EF is generated in the medium beyond the reflecting interface when total internal reflection occurs at the interface between the two media with different refractive indices. The electric field decays exponentially with the distance from the interface. The depth of penetration, d p , is defined as follows: λ ; (1) d p pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 2 2π nc sin2 θ − n2M where θ is the incident angle of the laser, λ is the wavelength of the propagating light, and nc and nM are the refractive indices of the MF and the etched fiber, respectively. From the theory of EF absorption, γ denotes the bulk absorption coefficient of the MF. Therefore, the transmitted power of the sensor, P o , can be calculated according to the Lambert–Beer law: P o P i exp−gl ;
(2)
where P i is the launch power for the sensor head, l is the effective length of the sensor head, and γ corresponds to various properties of the MF, which are determined by the external magnetic field. When the FLRDS technique is introduced, the output signal becomes a periodic train of decaying pulses. Figure 6(a) shows the decaying laser pulses as detected without an external magnetic field in the experiment. As a development of the conventional cavity ring-down spectroscopy, the FLRDS technique replaces the two highreflectivity mirrors by a fiber loop consisting of two couplers and an SMF that is several tens of meters long [18]. The fiber loop represents the concept of the ring-down cavity, in which both 2 × 1 optical couplers have a split ratio of 0.5∶99.5. When pulsed light is injected into the fiber loop via the 0.5% leg of coupler 1, it then passes through the fiber loop for many round trips. Every round trip of the pulse sustains a loss caused by the fiber and the sensing part. Once one round trip is finished, a small fraction of the light beam is then decoupled from the fiber loop through the 0.5% leg of coupler 2 so that it can be detected by a photodetector (PD); the rest continues traveling in the fiber loop and experiences various sources of attenuation caused by the sensing device under the varying magnetic field and the transmission fiber. At this stage, a periodic train of decaying pulses is detected. The detected signal follows an exponential decay, which can be expressed as follows [26,27]: dI I Ac − ; (3) dt nL where I represents the light intensity at time t, c represents the speed of light in a vacuum, L represents the total length of the fiber loop, n represents the averaged refractive index of the fiber loop, and A represents the total fiber transmission loss of light during each round trip. The total loss contains the insertion losses of the fiber couplers, the fiber absorption loss, and the fiber scattering loss. From the solution of Eq. (3), we can obtain the relation between the light intensity at time t and the initial light intensity: c I I 0 exp − At : (4) nL
Equation (4) explains why the FLRDS technique is insensitive to the light source fluctuations of I 0 . In other words, this technique measures the light intensity decay rate rather than the change in absolute intensity. The ring-down time is defined as τ0 , which is the time at which I decreases to 1∕e of the value I 0 , and is given by nL (5) τ0 : cA When the sensing system is given, the loss induced by the sensor head should be considered. In this case, the ring-down time should be written as nL ; (6) τ cA V where V is the loss caused by the sensor. From Eqs. (5) and (6), we have nL 1 1 : (7) − V c τ τ0 Define V p as the loss of light caused by the sensor per round trip in the fiber loop, and we can get t p Δτ 1 1 1 Δτ − ; (8) V p tp τ0 τ τ τ0 m τ Δτ τ0 − τ;
(9)
where m is the number of round trips of the laser pulse within the ring-down time τ, and t p is the transmission time per round trip of light. Then the minimum detectable loss V min is defined as the 1σ detection limit can be described by 1 στ ; (10) V min mτ where τ is the averaged ring-down time, and σ τ is referred to as the 1σ standard deviation of the ring-down time. Thus, when a conventional single-pass, intensity-based EF optical fiber sensor has a detectable optical loss of V , a FLRD-based sensor can achieve a detection limit V min , improving the sensitivity of magnetic field detection by the factor of m. Therefore, a significant improvement in sensitivity can be achieved due to the FLRDS technique.
3. EXPERIMENTS AND RESULTS A. Fabrication of the Sensor Head
The optical microfiber has recently gained significant research interest in many fields on account of its low profile, high sensitivity, and robustness; for example, as sensors for chemical detection [28], temperature sensing [29], and magnetic field measurement [30]. In this paper, the sensor device was achieved by partially removing the cladding layer with a 44% hydrofluoric (HF) acid solution. When the concentration of the solution, the temperature, and the experimental parameter parallels were consistently maintained, the cladding diameter was experimentally found to gradually reduce. Figure 1 shows the linear fitting of the data measured in the etching process which can help us estimate the diameter of the etched fiber according to
Research Article
Vol. 55, No. 4 / February 1 2016 / Applied Optics
675
Fig. 4. Schematic diagram of the experimental setup. Fig. 1. Fiber diameter as a function of corrosion time.
Fig. 2. SEM picture of the etched fiber.
the etching time. In addition, pictures of the etched fiber under scanning electron microscopy (SEM) are provided in Fig. 2. Accordingly, the diameter of the fiber was controlled by adjusting the etching time. The length was controlled by a container made of nylon 66, which was inserted into the HF. The MF was then injected into the container (after the HF was cleaned out) in order to realize magnetic sensing functionality. Figure 3 shows a diagram of the sensor head. B. Experiments and Discussion
The experimental setup is shown in Fig. 4. The pulses were generated by a system that consisted of an amplified
Fig. 3. Schematic diagram of the sensor head.
spontaneous emission (ASE) device with a wavelength ranging from 1525 to 1565 nm (CONQUER KG-ASE), an intensity modulator (CONQUER KG-AM-15), and a function generator (TEKTRONIX AFG-3252). The ring-down fiber loop contained two SMF couplers with a splitting ratio of 0.5%:99.5%, an erbium-doped fiber amplifier (EDFA) produced by Conquer (CONQUER KG-EDFA-B), and a sensing device. According to the principle of FLRDS, the length of the fiber loop should be larger than the width of a pulse propagation one circle in the fiber loop, and the delay line in the setup is used to adjust the transmission time of light pulses in the fiber loop. The practical application length can vary from a few meters to several kilometers based upon different needs of practice [31]. In the practical application, the delay line can be chosen depending on the actual requirement. In our experiment, the pulse width is 1 μs, and then we can calculate that the minimum length of the fiber loop that we need is about 200 m. Considering the laboratory conditions, we choose a delay fiber with a length of 2 km. Also, such a length of delay fiber in the designed structure possesses a much greater potential for sensing remotely magnetic fields in the future depending on its effective path length up to thousands of meters and high accuracy. Furthermore, EDFA with automatic gain control (AGC) is employed to compensate the power loss of the pulse in the ring-down cavity, which will increase the fiber loop round trips and improve the precision of magnetic field detection. The sensor head had the following dimensions: 20 mm in length and 35 m in diameter. This allowed the sensor head to achieve an impressive level of sensitivity under laboratory conditions. The light pulse train (pulse width of 1 μs and a cycle of 100 μs) was
Fig. 5. System for the generation of the magnetic field.
676
Research Article
Vol. 55, No. 4 / February 1 2016 / Applied Optics
coupled into the fiber loop through the 0.5% leg of coupler 1, and then propagated around the fiber loop while its intensity decayed on account of losses in the fiber loop. The output periodic train of decayed pulses was detected by a PD and sampled by a digital oscilloscope (TEKTRONIX MSO70804C) and finally sent to a computer for data processing. The experiments for magnetic field sensing were accomplished by gradually changing the distance between the permanent iron magnets; a schematic diagram is shown in Fig. 5. A magnetic field intensity ranging from 0 to 254 mT can be successfully obtained by using this method. A Gaussmeter (WEITE WT10A) was employed for the intensity measurement. Figures 6(a) and 6(b) show the detected ring-down signal when the magnetic field intensity H was 0 mT and 254 mT, respectively. By comparing the two ring-down curves, we can see that a magnetic field with higher intensity results in fewer peaks, which indicates a shorter ring-down time. The ring-down time is obtained by exponentially fitting the peaks of the ring-down curves. In order to display the variation visually, the fitted curves under various magnetic field intensities from 0 to 254 mT are collected in Fig. 7. By calculating the time at which 1∕e of the initial value in Fig. 7 and by finding this point on the exponential curves, we obtained the relationship between the magnetic field strength and the ring-down time, as shown in Table 1. Accordingly, a scatter diagram was produced, as shown in Fig. 8. The detected signal varied at a very slow speed when the magnetic field intensity was smaller than 10 mT due to the magnetic field response time. After that, the ring-down time showed a good linear response [32] to the magnetic field intensity in the range of 10–40 mT with an average sensitivity of −95.5 ns∕mT. The linear fit curve is provided in Fig. 8. When the magnetic field reached ∼40 mT, the transmission loss tended to gradually saturate due to the magnetic saturation effect of the MF. The loss changes of the sensor this time are mainly induced by the extensive scattering effects related to both the alignment of the magnetic nanoparticles with the lines of the magnetic field and the refractive index changes still occurring within the agglomerated lines [33,34]. Owing to the magnetic field response time and saturation effect, the nonlinear result of the magnetic field sensor based on the MF is unavoidable. Generally, the data can be well fitted by the modified Langevin function which is also shown in Fig. 8 [21]. The magnetic saturation depends on parameters such as the temperature, the geometry size of the container, and the type of the MF. The MF employed in this experiment was P-230 (produced by Kegao Electronics, Ltd.), which has a saturation magnetization of 59 mT and a maximum refractive index of 1.446. The geometry of the container was as follows: 60 mm in length, 2 mm in height, and 2 mm in width. Under these circumstances, a measurement range of about 30 mT was possible at room temperature.
Fig. 6. (a) Ring-down signal at 0 mT. (b) Ring-down signal at 254 mT.
Fig. 7. Fitted curves under various magnetic field intensities from 0 to 254 mT.
Table 1. H (mT) τ (μs) H (mT) τ (μs)
Relationship between the Magnetic Field Strength and the Ring-Down Time 8 14.24 35 11.56
11 13.63 41 11.13
16 13.41 49 10.83
18 13.18 59 10.45
21 12.83 65 10.45
24 12.50 71 10.29
28 12.12 79 10.18
32 11.89 87 10.050
Research Article
Fig. 8. Relationship between the magnetic field strength and the ring-down time.
Furthermore, it has already been investigated that the loss effects in the light transmission through magnetic fluids is polarization dependent [21]. Accordingly, different sensitivities can be obtained by using a polarized light with respect to the application axis of the magnetic field. Consequently, the magnetic field sensor can be further enhanced through this method [33]. 4. CONCLUSIONS An EF-based magnetic field sensor utilizing a section of uncladded SMF coated with MF has been proposed and experimentally demonstrated. Furthermore, the sensitivity and stability of the system are increased by introducing FLRDS technology into the sensor scheme. The ring-down signal significantly changed with the magnetic field strength owing to the variation of the distribution of the magnetic nanoparticles in the MF. In this case, we can vary the ring-down time. The principles of the FLRDS technology and EF theory are discussed in detail, and the fabrication process of the sensor head is also provided. Ultimately, a novel magnetic field sensor with a sensitivity of 95.5 ns/mT is achieved. The effective measurement range of this sensor is about 30 mT due to the phenomenon of magnetic saturation. In addition, different sensitivities of the sensor can be obtained by using a polarized light with respect to the application axis of the magnetic field. Funding. National Natural Science Foundation of China (NSFC) (51307036); Natural Science Foundation of Heilongjiang Province (E201303). REFERENCES 1. K. Bohnert, H. Brandle, M. G. Brunzel, P. Gabus, and P. Guggenbach, “Highly accurate fiber-optic DC current sensor for the electrowinning industry,” IEEE Ind. Appl. 43, 180–187 (2007). 2. F. Bucholtz, C. A. Villarruel, A. R. Davis, C. K. Kirkendall, D. M. Dagenais, J. A. McVicker, S. S. Patrick, K. P. Koo, W. Gunnar, H. Valo, T. Lund, A. G. Andersen, R. Gjessing, E. J. Eidem, and T. Knudsen, “Multichannel fiber-optic magnetometer system for undersea measurements,” J. Lightwave. Technol. 13, 1385–1395 (1995). 3. A. Dandridge and G. B. Cogdell, “Fiber optic sensors for navy applications,” IEEE LCS 2, 81–89 (1991). 4. J. Mora, A. Diez, J. L. Cruz, and M. V. Andres, “A magnetostrictive sensor interrogated by fiber gratings for DC-current and temperature discrimination,” IEEE Photon. Technol. Lett. 12, 1680–1682 (2000).
Vol. 55, No. 4 / February 1 2016 / Applied Optics
677
5. A. Candiani, A. Bertucci, S. Giannetti, M. Konstantaki, A. Manicardi, S. Pissadakis, A. Cucinotta, R. Corradini, and S. Selleri, “Label-free DNA bio sensor based on apeptide nucleic acid-functionalized microstructured optical fiber-Bragg grating,” J. Biomed. Opt. 18, 057004 (2013). 6. Y. K. Dong, L. Chen, and X. Y. Bao, “Truly distributed birefringence measurement of polarization-maintaining fibers based on transient Brillouin grating,” Opt. Lett. 35, 193–195 (2010). 7. T. Yoshino, K. Kurosawa, K. Itoh, and T. Ose, “Fiber-optic Fabry– Perot interferometer and its sensor applications,” IEEE Trans. Microw. Theory 18, 1624–1633 (1982). 8. Y. J. Rao, “Recent progress in fiber-optic extrinsic Fabry–Perot interferometric sensors,” Technology 12, 227–237 (2006). 9. S. Liu, K. M. Yang, Y. P. Wang, J. Qu, C. R. Liao, J. He, Z. Y. Li, G. L. Yin, B. Sun, J. T. Zhou, G. J. Wang, J. Tang, and J. Zhao, “Highsensitivity strain sensor based on in-fiber rectangular air bubble,” Sci. Rep. 5, 7624 (2015). 10. Y. Y. Luo, L. Xia, C. Yu, W. Li, Q. Z. Sun, Y. W. Wang, and D. M. Liu, “Multi-parameter optical fiber sensor based on enhanced multimode interference,” Opt. Commun. 344, 120–124 (2015). 11. F. Bucholtz, C. A. Villarruel, A. R. Davis, C. K. Kirkendall, D. M. Dagenais, J. A. McVicker, S. S. Patrick, K. P. Koo, W. Gunnar, H. Valo, T. Lund, A. G. Andersen, R. Gjessing, E. J. Eidem, and T. Knudsen, “Multichannel fiber optic magnetometer system for undersea measurements,” J. Lightwave Technol. 13, 1385–1395 (1995). 12. M. Sedlar, I. PaulickaI, and M. Sayer, “Optical fiber magnetic field sensors with ceramic magnetostrictive jackets,” Appl. Opt. 35, 5340–5344 (1996). 13. L. Sun, S. Jiang, and J. R. Marciante, “All-fiber optical magnetic-field sensor based on Faraday rotation in highly terbium-doped fiber,” Opt. Express 18, 5407–5412 (2010). 14. S. Wang, P. Lu, and L. L. Mao, “Temperature sensing using the spectral interference of polarization modes in a highly birefringent fiber,” Opt. Laser Eng. 70, 51–56 (2015). 15. R. S. Brown, I. Kozin, Z. Tong, R. D. Oleschuk, and H. P. Loock, “Fiber-loop ring-down spectroscopy,” J. Chem. Phys. 117, 10444– 10447 (2002). 16. H. Waechter, J. Litman, A. H. Cheung, J. A. Barnes, and H. P. Loock, “Chemical sensing using fiber cavity ring-down spectroscopy,” Sensors 10, 1716–1742 (2010). 17. H. Tian, C. M. Zhou, D. Fan, Y. W. Ou, T. Tian, W. L. Liang, and M. M. Li, “Continuous-wave frequency-shifted interferometry cavity ringdown gas sensing with differential optical absorption,” IEEE Photon. J. 7, 1–10 (2015). 18. Y. Zhao, L. Bai, and Q. Wang, “Gas concentration sensor based on fiber loop ring-down spectroscopy,” Opt. Commun. 309, 328–332 (2013). 19. H. Berberoglu and H. Altan, “A simple single-mode fiber loss measurement scheme in the C-band based on fiber loop-cavity ring down spectroscopy,” Opt. Commun. 317, 29–33 (2014). 20. Y. Zhao, D. Wu, and R. Q. Lv, “Magnetic field sensor based on photonic crystal fiber taper coated with ferrofluid,” IEEE Photon. Technol. Lett. 27, 26–29 (2015). 21. P. Zu, C. C. Chan, W. S. Lew, Y. Jin, Y. Zhang, H. F. Liew, L. H. Chen, W. C. Wong, and X. Y. Dong, “Magneto-optical fiber sensor based on magnetic fluid,” Opt. Lett. 37, 398–400 (2012). 22. L. Martinez, F. Cecelja, and R. Rakowski, “A novel magneto-optic ferrofluid material for sensor applications,” Sens. Actuators A 123–124, 438–443 (2005). 23. M. Xu and P. J. Ridler, “Linear dichroism and birefringence effects in magnetic fluids,” Appl. Phys. 82, 326–332 (1997). 24. Y. G. Zheng, X. Y. Dong, C. C. Chan, P. P. Shum, and H. B. Su, “Optical fiber magnetic field sensor based on magnetic fluid and microfiber mode interferometer,” Opt. Commun. 336, 5–8 (2015). 25. J. Barnes, M. Dreher, K. Plett, R. S. Brown, C. M. Cruddena, and H.-P. Loock, “Chemical sensor based on a long-period fibre grating modified by a functionalized polydimethylsiloxane coating,” Analyst 133, 1541–1549 (2008). 26. C. Wang and S. T. Scherrer, “Fiber ring down pressure sensors,” Opt. Lett. 29, 352–354 (2004).
678
Vol. 55, No. 4 / February 1 2016 / Applied Optics
27. C. Wang and S. T. Scherrer, “Fiber loop ring down for physical sensor development: pressure sensor,” Appl. Opt. 43, 6458–6464 (2004). 28. C. L. Linslal, P. M. Syam Mohan, A. Halder, and T. K. Gangopadhyay, “Analysis and modeling of an optical fiber loop resonator and an evanescent field absorption sensor for the application for chemical detection,” Sens. Actuators A 194, 160–168 (2013). 29. X. H. Fu, H. Y. Xie, X. L. Zeng, G. W. Fu, and W. H. Bi, “Refractive index insensitive temperature sensor based on specialty triple-clad fiber,” Opt. Express 23, 2320–2327 (2015). 30. L. F. Luo, S. L. Pu, J. L. Tang, X. L. Zeng, and M. Lahoubi, “Highly sensitive magnetic field sensor based on microfiber coupler with magnetic fluid,” Appl. Phys. Lett. 106, 193507 (2015).
Research Article 31. D. Q. Tang, D. Yang, Y. Jiang, J. L. Zhao, H. Y. Wang, and S. Q. Jiang, “Fiber loop ring-down optical fiber grating gas pressure sensor,” Opt. Laser Eng. 48, 1262–1265 (2010). 32. L. Li, Q. Han, Y. F. Chen, T. G. Liu, and R. X. Zhang, “An all-fiber optic current sensor based on ferrofluids and multimode interference,” IEEE Sens. J. 14, 1749–1753 (2014). 33. A. Candiani, A. Argyros, S. G. Leon-Saval, R. Lwin, S. Selleri, and S. Pissadakis, “A loss-based, magnetic field sensor implemented in a ferrofluid infiltrated microstructured polymer optical fiber,” Appl. Phys. Lett. 104, 111106 (2014). 34. P. Childs, A. Candiani, and S. Pissadakis, “Optical fiber cladding ring magnetic field sensor,” IEEE Photon. Technol. Lett. 23, 929–931 (2011).
Vol. 55, No. 4 / February 1 2016 / Applied Optics
673
Magnetic field sensor using the fiber loop ring-down technique and an etched fiber coated with magnetic fluid TAO SHEN,1,2,* YUE FENG,2 BINCHAO SUN,2
AND
XINLAO WEI1
1
Key Laboratory of Engineering Dielectrics and Its Application, Ministry of Education, Harbin University of Science and Technology, Harbin 150080, China 2 College of Applied Sciences, Harbin University of Science and Technology, Harbin 150080, China *Corresponding author: [email protected] Received 28 September 2015; revised 8 November 2015; accepted 5 December 2015; posted 15 December 2015 (Doc. ID 250664); published 21 January 2016
The fiber loop ring-down spectroscopy technique is introduced into the evanescent-field-based sensing scheme in order to create a new type of fiber-based magnetic field sensor. As a consequence, the sensitivity and stability of the magnetic field sensing system are significantly enhanced. The sensor head is constructed using a section of a single-mode fiber with its cladding partially etched. The process of fiber etching is described in detail, and the relationship between the diameter of the etched fiber and the etching time is experimentally investigated. After adopting the appropriate size of the etched fiber, the final experimental results show that the magnetic field strength has a well-defined linear relationship with the inverse of the ring-down time τ over a range of 30 mT with a sensitivity of 95.5 ns/mT. © 2016 Optical Society of America OCIS codes: (230.3810) Magneto-optic systems; (060.4005) Microstructured fibers; (060.2370) Fiber optics sensors. http://dx.doi.org/10.1364/AO.55.000673
1. INTRODUCTION Magnetic field sensing plays an important role in various fields such as electric power transmission [1], undersea measurements [2], and the military sector [3]. Sensors based on optical fibers employ a number of different sensor elements, including fiber grating sensors [4–6], fiber-optic Fabry–Perot sensors [7–9], multimode interference fiber sensors [10], fiber Michelson/ Mach–Zehnder interferometric sensors [11,12], and birefringence polarimetric fiber-optic sensors [13,14]. However, these sensors share a common drawback in that their fabrication is costly. On the other hand, evanescent-field- (EF) based fiber magnetic field sensors that use a section of an etched single-mode fiber (SMF) are less costly and are also simpler. Nevertheless, the usual response of the EF to the external magnetic field is relatively weak and is mainly limited by the intensity-based sensing scheme whose detection limit is constrained by light source fluctuations. Accordingly, in this paper, we introduce the fiber loop ring-down spectroscopy (FLRDS) technique into the EF-based fiber sensor in order to improve the detection sensitivity and system stability. The FLRDS technique, which is an absorption spectroscopic and detection technique, has become popular in recent years [15–19] because it can be directly adopted by many sensing schemes. It makes use of an optical cavity, which not only 1559-128X/16/040673-06$15/0$15.00 © 2016 Optical Society of America
realizes a long effective path length through a sample but also eliminates the effect of fluctuations in the light intensity. Through the FLRDS technique, the detection sensitivity of EF-based sensors has been improved significantly. To the best of our knowledge, no other EF-FLRDS-based magnetic field sensor has been reported. Moreover, in this study, a magnetic fluid (MF) is placed around the etched SMF to act as a medium for the magnetic field and evanescent field. MF is a new kind of colloidal liquid that consists of magnetic nanoparticles and a liquid carrier (usually water or an ester) [20]. As a kind of functional medium, MF has various magneto-optical properties; more specifically, MF exhibits a tunable refractive index [21], facilitates the Faraday effect [22], exhibits birefringence [23], and so on. When a varying magnetic field is introduced around the MF, the EF will vary in response to the transmissivity and refractive index of the MF. In this study, an EF-FLRDS-based magnetic field sensor is achieved with a considerable detection sensitivity of 95.5 ns/mT in the linear response range, which is a great improvement over the previously reported ferrofluid-based magnetic field sensors [24]; its advantages include easy fabrication, low cost, and an impressive stability owing to the characteristic that it is insensitive to the intensity fluctuations of the laser source of the FLRDS technique [25].
674
Research Article
Vol. 55, No. 4 / February 1 2016 / Applied Optics
2. PRINCIPLES The EF is generated in the medium beyond the reflecting interface when total internal reflection occurs at the interface between the two media with different refractive indices. The electric field decays exponentially with the distance from the interface. The depth of penetration, d p , is defined as follows: λ ; (1) d p pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 2 2π nc sin2 θ − n2M where θ is the incident angle of the laser, λ is the wavelength of the propagating light, and nc and nM are the refractive indices of the MF and the etched fiber, respectively. From the theory of EF absorption, γ denotes the bulk absorption coefficient of the MF. Therefore, the transmitted power of the sensor, P o , can be calculated according to the Lambert–Beer law: P o P i exp−gl ;
(2)
where P i is the launch power for the sensor head, l is the effective length of the sensor head, and γ corresponds to various properties of the MF, which are determined by the external magnetic field. When the FLRDS technique is introduced, the output signal becomes a periodic train of decaying pulses. Figure 6(a) shows the decaying laser pulses as detected without an external magnetic field in the experiment. As a development of the conventional cavity ring-down spectroscopy, the FLRDS technique replaces the two highreflectivity mirrors by a fiber loop consisting of two couplers and an SMF that is several tens of meters long [18]. The fiber loop represents the concept of the ring-down cavity, in which both 2 × 1 optical couplers have a split ratio of 0.5∶99.5. When pulsed light is injected into the fiber loop via the 0.5% leg of coupler 1, it then passes through the fiber loop for many round trips. Every round trip of the pulse sustains a loss caused by the fiber and the sensing part. Once one round trip is finished, a small fraction of the light beam is then decoupled from the fiber loop through the 0.5% leg of coupler 2 so that it can be detected by a photodetector (PD); the rest continues traveling in the fiber loop and experiences various sources of attenuation caused by the sensing device under the varying magnetic field and the transmission fiber. At this stage, a periodic train of decaying pulses is detected. The detected signal follows an exponential decay, which can be expressed as follows [26,27]: dI I Ac − ; (3) dt nL where I represents the light intensity at time t, c represents the speed of light in a vacuum, L represents the total length of the fiber loop, n represents the averaged refractive index of the fiber loop, and A represents the total fiber transmission loss of light during each round trip. The total loss contains the insertion losses of the fiber couplers, the fiber absorption loss, and the fiber scattering loss. From the solution of Eq. (3), we can obtain the relation between the light intensity at time t and the initial light intensity: c I I 0 exp − At : (4) nL
Equation (4) explains why the FLRDS technique is insensitive to the light source fluctuations of I 0 . In other words, this technique measures the light intensity decay rate rather than the change in absolute intensity. The ring-down time is defined as τ0 , which is the time at which I decreases to 1∕e of the value I 0 , and is given by nL (5) τ0 : cA When the sensing system is given, the loss induced by the sensor head should be considered. In this case, the ring-down time should be written as nL ; (6) τ cA V where V is the loss caused by the sensor. From Eqs. (5) and (6), we have nL 1 1 : (7) − V c τ τ0 Define V p as the loss of light caused by the sensor per round trip in the fiber loop, and we can get t p Δτ 1 1 1 Δτ − ; (8) V p tp τ0 τ τ τ0 m τ Δτ τ0 − τ;
(9)
where m is the number of round trips of the laser pulse within the ring-down time τ, and t p is the transmission time per round trip of light. Then the minimum detectable loss V min is defined as the 1σ detection limit can be described by 1 στ ; (10) V min mτ where τ is the averaged ring-down time, and σ τ is referred to as the 1σ standard deviation of the ring-down time. Thus, when a conventional single-pass, intensity-based EF optical fiber sensor has a detectable optical loss of V , a FLRD-based sensor can achieve a detection limit V min , improving the sensitivity of magnetic field detection by the factor of m. Therefore, a significant improvement in sensitivity can be achieved due to the FLRDS technique.
3. EXPERIMENTS AND RESULTS A. Fabrication of the Sensor Head
The optical microfiber has recently gained significant research interest in many fields on account of its low profile, high sensitivity, and robustness; for example, as sensors for chemical detection [28], temperature sensing [29], and magnetic field measurement [30]. In this paper, the sensor device was achieved by partially removing the cladding layer with a 44% hydrofluoric (HF) acid solution. When the concentration of the solution, the temperature, and the experimental parameter parallels were consistently maintained, the cladding diameter was experimentally found to gradually reduce. Figure 1 shows the linear fitting of the data measured in the etching process which can help us estimate the diameter of the etched fiber according to
Research Article
Vol. 55, No. 4 / February 1 2016 / Applied Optics
675
Fig. 4. Schematic diagram of the experimental setup. Fig. 1. Fiber diameter as a function of corrosion time.
Fig. 2. SEM picture of the etched fiber.
the etching time. In addition, pictures of the etched fiber under scanning electron microscopy (SEM) are provided in Fig. 2. Accordingly, the diameter of the fiber was controlled by adjusting the etching time. The length was controlled by a container made of nylon 66, which was inserted into the HF. The MF was then injected into the container (after the HF was cleaned out) in order to realize magnetic sensing functionality. Figure 3 shows a diagram of the sensor head. B. Experiments and Discussion
The experimental setup is shown in Fig. 4. The pulses were generated by a system that consisted of an amplified
Fig. 3. Schematic diagram of the sensor head.
spontaneous emission (ASE) device with a wavelength ranging from 1525 to 1565 nm (CONQUER KG-ASE), an intensity modulator (CONQUER KG-AM-15), and a function generator (TEKTRONIX AFG-3252). The ring-down fiber loop contained two SMF couplers with a splitting ratio of 0.5%:99.5%, an erbium-doped fiber amplifier (EDFA) produced by Conquer (CONQUER KG-EDFA-B), and a sensing device. According to the principle of FLRDS, the length of the fiber loop should be larger than the width of a pulse propagation one circle in the fiber loop, and the delay line in the setup is used to adjust the transmission time of light pulses in the fiber loop. The practical application length can vary from a few meters to several kilometers based upon different needs of practice [31]. In the practical application, the delay line can be chosen depending on the actual requirement. In our experiment, the pulse width is 1 μs, and then we can calculate that the minimum length of the fiber loop that we need is about 200 m. Considering the laboratory conditions, we choose a delay fiber with a length of 2 km. Also, such a length of delay fiber in the designed structure possesses a much greater potential for sensing remotely magnetic fields in the future depending on its effective path length up to thousands of meters and high accuracy. Furthermore, EDFA with automatic gain control (AGC) is employed to compensate the power loss of the pulse in the ring-down cavity, which will increase the fiber loop round trips and improve the precision of magnetic field detection. The sensor head had the following dimensions: 20 mm in length and 35 m in diameter. This allowed the sensor head to achieve an impressive level of sensitivity under laboratory conditions. The light pulse train (pulse width of 1 μs and a cycle of 100 μs) was
Fig. 5. System for the generation of the magnetic field.
676
Research Article
Vol. 55, No. 4 / February 1 2016 / Applied Optics
coupled into the fiber loop through the 0.5% leg of coupler 1, and then propagated around the fiber loop while its intensity decayed on account of losses in the fiber loop. The output periodic train of decayed pulses was detected by a PD and sampled by a digital oscilloscope (TEKTRONIX MSO70804C) and finally sent to a computer for data processing. The experiments for magnetic field sensing were accomplished by gradually changing the distance between the permanent iron magnets; a schematic diagram is shown in Fig. 5. A magnetic field intensity ranging from 0 to 254 mT can be successfully obtained by using this method. A Gaussmeter (WEITE WT10A) was employed for the intensity measurement. Figures 6(a) and 6(b) show the detected ring-down signal when the magnetic field intensity H was 0 mT and 254 mT, respectively. By comparing the two ring-down curves, we can see that a magnetic field with higher intensity results in fewer peaks, which indicates a shorter ring-down time. The ring-down time is obtained by exponentially fitting the peaks of the ring-down curves. In order to display the variation visually, the fitted curves under various magnetic field intensities from 0 to 254 mT are collected in Fig. 7. By calculating the time at which 1∕e of the initial value in Fig. 7 and by finding this point on the exponential curves, we obtained the relationship between the magnetic field strength and the ring-down time, as shown in Table 1. Accordingly, a scatter diagram was produced, as shown in Fig. 8. The detected signal varied at a very slow speed when the magnetic field intensity was smaller than 10 mT due to the magnetic field response time. After that, the ring-down time showed a good linear response [32] to the magnetic field intensity in the range of 10–40 mT with an average sensitivity of −95.5 ns∕mT. The linear fit curve is provided in Fig. 8. When the magnetic field reached ∼40 mT, the transmission loss tended to gradually saturate due to the magnetic saturation effect of the MF. The loss changes of the sensor this time are mainly induced by the extensive scattering effects related to both the alignment of the magnetic nanoparticles with the lines of the magnetic field and the refractive index changes still occurring within the agglomerated lines [33,34]. Owing to the magnetic field response time and saturation effect, the nonlinear result of the magnetic field sensor based on the MF is unavoidable. Generally, the data can be well fitted by the modified Langevin function which is also shown in Fig. 8 [21]. The magnetic saturation depends on parameters such as the temperature, the geometry size of the container, and the type of the MF. The MF employed in this experiment was P-230 (produced by Kegao Electronics, Ltd.), which has a saturation magnetization of 59 mT and a maximum refractive index of 1.446. The geometry of the container was as follows: 60 mm in length, 2 mm in height, and 2 mm in width. Under these circumstances, a measurement range of about 30 mT was possible at room temperature.
Fig. 6. (a) Ring-down signal at 0 mT. (b) Ring-down signal at 254 mT.
Fig. 7. Fitted curves under various magnetic field intensities from 0 to 254 mT.
Table 1. H (mT) τ (μs) H (mT) τ (μs)
Relationship between the Magnetic Field Strength and the Ring-Down Time 8 14.24 35 11.56
11 13.63 41 11.13
16 13.41 49 10.83
18 13.18 59 10.45
21 12.83 65 10.45
24 12.50 71 10.29
28 12.12 79 10.18
32 11.89 87 10.050
Research Article
Fig. 8. Relationship between the magnetic field strength and the ring-down time.
Furthermore, it has already been investigated that the loss effects in the light transmission through magnetic fluids is polarization dependent [21]. Accordingly, different sensitivities can be obtained by using a polarized light with respect to the application axis of the magnetic field. Consequently, the magnetic field sensor can be further enhanced through this method [33]. 4. CONCLUSIONS An EF-based magnetic field sensor utilizing a section of uncladded SMF coated with MF has been proposed and experimentally demonstrated. Furthermore, the sensitivity and stability of the system are increased by introducing FLRDS technology into the sensor scheme. The ring-down signal significantly changed with the magnetic field strength owing to the variation of the distribution of the magnetic nanoparticles in the MF. In this case, we can vary the ring-down time. The principles of the FLRDS technology and EF theory are discussed in detail, and the fabrication process of the sensor head is also provided. Ultimately, a novel magnetic field sensor with a sensitivity of 95.5 ns/mT is achieved. The effective measurement range of this sensor is about 30 mT due to the phenomenon of magnetic saturation. In addition, different sensitivities of the sensor can be obtained by using a polarized light with respect to the application axis of the magnetic field. Funding. National Natural Science Foundation of China (NSFC) (51307036); Natural Science Foundation of Heilongjiang Province (E201303). REFERENCES 1. K. Bohnert, H. Brandle, M. G. Brunzel, P. Gabus, and P. Guggenbach, “Highly accurate fiber-optic DC current sensor for the electrowinning industry,” IEEE Ind. Appl. 43, 180–187 (2007). 2. F. Bucholtz, C. A. Villarruel, A. R. Davis, C. K. Kirkendall, D. M. Dagenais, J. A. McVicker, S. S. Patrick, K. P. Koo, W. Gunnar, H. Valo, T. Lund, A. G. Andersen, R. Gjessing, E. J. Eidem, and T. Knudsen, “Multichannel fiber-optic magnetometer system for undersea measurements,” J. Lightwave. Technol. 13, 1385–1395 (1995). 3. A. Dandridge and G. B. Cogdell, “Fiber optic sensors for navy applications,” IEEE LCS 2, 81–89 (1991). 4. J. Mora, A. Diez, J. L. Cruz, and M. V. Andres, “A magnetostrictive sensor interrogated by fiber gratings for DC-current and temperature discrimination,” IEEE Photon. Technol. Lett. 12, 1680–1682 (2000).
Vol. 55, No. 4 / February 1 2016 / Applied Optics
677
5. A. Candiani, A. Bertucci, S. Giannetti, M. Konstantaki, A. Manicardi, S. Pissadakis, A. Cucinotta, R. Corradini, and S. Selleri, “Label-free DNA bio sensor based on apeptide nucleic acid-functionalized microstructured optical fiber-Bragg grating,” J. Biomed. Opt. 18, 057004 (2013). 6. Y. K. Dong, L. Chen, and X. Y. Bao, “Truly distributed birefringence measurement of polarization-maintaining fibers based on transient Brillouin grating,” Opt. Lett. 35, 193–195 (2010). 7. T. Yoshino, K. Kurosawa, K. Itoh, and T. Ose, “Fiber-optic Fabry– Perot interferometer and its sensor applications,” IEEE Trans. Microw. Theory 18, 1624–1633 (1982). 8. Y. J. Rao, “Recent progress in fiber-optic extrinsic Fabry–Perot interferometric sensors,” Technology 12, 227–237 (2006). 9. S. Liu, K. M. Yang, Y. P. Wang, J. Qu, C. R. Liao, J. He, Z. Y. Li, G. L. Yin, B. Sun, J. T. Zhou, G. J. Wang, J. Tang, and J. Zhao, “Highsensitivity strain sensor based on in-fiber rectangular air bubble,” Sci. Rep. 5, 7624 (2015). 10. Y. Y. Luo, L. Xia, C. Yu, W. Li, Q. Z. Sun, Y. W. Wang, and D. M. Liu, “Multi-parameter optical fiber sensor based on enhanced multimode interference,” Opt. Commun. 344, 120–124 (2015). 11. F. Bucholtz, C. A. Villarruel, A. R. Davis, C. K. Kirkendall, D. M. Dagenais, J. A. McVicker, S. S. Patrick, K. P. Koo, W. Gunnar, H. Valo, T. Lund, A. G. Andersen, R. Gjessing, E. J. Eidem, and T. Knudsen, “Multichannel fiber optic magnetometer system for undersea measurements,” J. Lightwave Technol. 13, 1385–1395 (1995). 12. M. Sedlar, I. PaulickaI, and M. Sayer, “Optical fiber magnetic field sensors with ceramic magnetostrictive jackets,” Appl. Opt. 35, 5340–5344 (1996). 13. L. Sun, S. Jiang, and J. R. Marciante, “All-fiber optical magnetic-field sensor based on Faraday rotation in highly terbium-doped fiber,” Opt. Express 18, 5407–5412 (2010). 14. S. Wang, P. Lu, and L. L. Mao, “Temperature sensing using the spectral interference of polarization modes in a highly birefringent fiber,” Opt. Laser Eng. 70, 51–56 (2015). 15. R. S. Brown, I. Kozin, Z. Tong, R. D. Oleschuk, and H. P. Loock, “Fiber-loop ring-down spectroscopy,” J. Chem. Phys. 117, 10444– 10447 (2002). 16. H. Waechter, J. Litman, A. H. Cheung, J. A. Barnes, and H. P. Loock, “Chemical sensing using fiber cavity ring-down spectroscopy,” Sensors 10, 1716–1742 (2010). 17. H. Tian, C. M. Zhou, D. Fan, Y. W. Ou, T. Tian, W. L. Liang, and M. M. Li, “Continuous-wave frequency-shifted interferometry cavity ringdown gas sensing with differential optical absorption,” IEEE Photon. J. 7, 1–10 (2015). 18. Y. Zhao, L. Bai, and Q. Wang, “Gas concentration sensor based on fiber loop ring-down spectroscopy,” Opt. Commun. 309, 328–332 (2013). 19. H. Berberoglu and H. Altan, “A simple single-mode fiber loss measurement scheme in the C-band based on fiber loop-cavity ring down spectroscopy,” Opt. Commun. 317, 29–33 (2014). 20. Y. Zhao, D. Wu, and R. Q. Lv, “Magnetic field sensor based on photonic crystal fiber taper coated with ferrofluid,” IEEE Photon. Technol. Lett. 27, 26–29 (2015). 21. P. Zu, C. C. Chan, W. S. Lew, Y. Jin, Y. Zhang, H. F. Liew, L. H. Chen, W. C. Wong, and X. Y. Dong, “Magneto-optical fiber sensor based on magnetic fluid,” Opt. Lett. 37, 398–400 (2012). 22. L. Martinez, F. Cecelja, and R. Rakowski, “A novel magneto-optic ferrofluid material for sensor applications,” Sens. Actuators A 123–124, 438–443 (2005). 23. M. Xu and P. J. Ridler, “Linear dichroism and birefringence effects in magnetic fluids,” Appl. Phys. 82, 326–332 (1997). 24. Y. G. Zheng, X. Y. Dong, C. C. Chan, P. P. Shum, and H. B. Su, “Optical fiber magnetic field sensor based on magnetic fluid and microfiber mode interferometer,” Opt. Commun. 336, 5–8 (2015). 25. J. Barnes, M. Dreher, K. Plett, R. S. Brown, C. M. Cruddena, and H.-P. Loock, “Chemical sensor based on a long-period fibre grating modified by a functionalized polydimethylsiloxane coating,” Analyst 133, 1541–1549 (2008). 26. C. Wang and S. T. Scherrer, “Fiber ring down pressure sensors,” Opt. Lett. 29, 352–354 (2004).
678
Vol. 55, No. 4 / February 1 2016 / Applied Optics
27. C. Wang and S. T. Scherrer, “Fiber loop ring down for physical sensor development: pressure sensor,” Appl. Opt. 43, 6458–6464 (2004). 28. C. L. Linslal, P. M. Syam Mohan, A. Halder, and T. K. Gangopadhyay, “Analysis and modeling of an optical fiber loop resonator and an evanescent field absorption sensor for the application for chemical detection,” Sens. Actuators A 194, 160–168 (2013). 29. X. H. Fu, H. Y. Xie, X. L. Zeng, G. W. Fu, and W. H. Bi, “Refractive index insensitive temperature sensor based on specialty triple-clad fiber,” Opt. Express 23, 2320–2327 (2015). 30. L. F. Luo, S. L. Pu, J. L. Tang, X. L. Zeng, and M. Lahoubi, “Highly sensitive magnetic field sensor based on microfiber coupler with magnetic fluid,” Appl. Phys. Lett. 106, 193507 (2015).
Research Article 31. D. Q. Tang, D. Yang, Y. Jiang, J. L. Zhao, H. Y. Wang, and S. Q. Jiang, “Fiber loop ring-down optical fiber grating gas pressure sensor,” Opt. Laser Eng. 48, 1262–1265 (2010). 32. L. Li, Q. Han, Y. F. Chen, T. G. Liu, and R. X. Zhang, “An all-fiber optic current sensor based on ferrofluids and multimode interference,” IEEE Sens. J. 14, 1749–1753 (2014). 33. A. Candiani, A. Argyros, S. G. Leon-Saval, R. Lwin, S. Selleri, and S. Pissadakis, “A loss-based, magnetic field sensor implemented in a ferrofluid infiltrated microstructured polymer optical fiber,” Appl. Phys. Lett. 104, 111106 (2014). 34. P. Childs, A. Candiani, and S. Pissadakis, “Optical fiber cladding ring magnetic field sensor,” IEEE Photon. Technol. Lett. 23, 929–931 (2011).
