A kind of magnetic field sensor composed of magnetic fluid surrounding a segment of singlemode fiber is proposed. The taper-like and lateral-offset fusion splicing techniques are employed. The sensing principle is based on cladding mode interference. The interference valley wavelength or transmission loss of the sensing structure is sensitive to the external magnetic field, which is utilized for magnetic field sensing. The linear response regions are obtained in the range of 38-225 Oe and 250-475 Oe. For the valley-wavelength-shift-type sensing, the sensitivities are 14.1 pm/Oe and 26 pm/Oe at low and high field ranges, respectively. For the transmission-loss-variation-type sensing, the sensitivity of −0.024 dB/Oe is achieved for the magnetic field strength ranging from 250 to 475 Oe.
© 2014 Optical Society of America
Magnetic fluid (MF) is a kind of stable colloidal system consisting of surfactant-coated magnetic nanoparticles dispersed in a suitable liquid carrier. It possesses both the features of magnetic property of solid magnetic materials and fluidity of liquids, which presents versatile magneto-optical properties. The magneto-optical properties of MFs have been utilized to design many unique optical devices, such as optical switches , optical gratings [2, 3], tunable optical capacitors , modulators [5, 6], tunable slow light , and magnetic field sensors [8–14]. Recently, the magnetic traction  and magneto-volume variation  of MFs are also exploited for magnetic field sensing. Fiber-optic magnetic field sensors based on MFs have the advantages of fiber-compatibility, compactness and high sensitivity. Besides, due to the fluidity of MFs, they are easy to be integrated with fibers. So, MF-based magnetic field sensors are attracting consistent research interests. The schemes of this kind of sensors include immersing microfiber knot inside the MFs , injecting MFs into the air holes of the photonic crystal fiber (PCF) [18, 19] or polarization-maintaining PCF , and using MFs as the claddings of the special fiber or fiber structures (e.g. S-tapered fiber , singlemode fiber (SMF) with up-tapered joints , singlemode-multimode-singlemode (SMS) fiber structures [10–12], PCF  and SMF Michelson interferometer ). Though many promising results have been achieved, fabricating these fiber-optic magnetic sensors is considerably complicated, which involves tapering or microfabrication techniques. Moreover, the tapered structures are easily broken in practical applications. Simultaneously, many of these structures need special optical fibers and then the cost is relatively high, especially for the PCF-based magnetic field sensors.
In this work, a novel sensing structure based on SMFs was fabricated by the taper-like and lateral-offset fusion splicing techniques. The fundamental sensing principle of our proposed structure is that the high-order modes will be excited in the SMF cladding and then interfere with each other. The effective refractive indices (RIs) of these interference modes could be influenced by the external environment/stimuli as well.
2. Fabrication and operating principle
The proposed structure consists of three pieces of SMF connected by two special joints in series, which is formed under a series of simple operations with a commercial electric-arc fusion splicer (AV4671) [see Fig. 1]. The main operations are taper-like [see Figs. 1(a)-1(d)] and lateral-offset [see Figs. 1(e)-1(f)] fusion splicing. For the taper-like fusion splicing, the initial distance between two cleaved flat fiber end faces is about 135.4 μm [see Fig. 1(a)]. Then, arc discharging is applied. The arc discharging parameters are set at arc duration of 15, arc current of 30 and Z push distance of 0 (all of them are relative values defined by the AV4671 fusion splicer). After arc discharging, the two cleaved flat fiber end faces are all melt into spherical shape and the distance between the two fiber tips is shortened to 129.6 μm [see Fig. 1(b)]. The two spherical fiber end faces are moved inward to physically contact [see Fig. 1(c)] and then manually fusion spliced together by setting arc duration at 15, arc current at 30 and Z push distance at 7. The waist diameter of the splicing joint is 84.3 μm [see Fig. 1(d)]. Now, the taper-like joint is formed. For the lateral-offset fusion splicing [see Fig. 1(e) and 1(f)], the offset distance is 13.9 μm and the fusion splicing parameters are set at arc duration of 10, arc current of 15 and Z push distance of 5. The relatively small splicing parameters are used to avoid damaging the internal fiber structure during the lateral-offset fusion splicing. The length between the taper-like and lateral-offset joints is 19 mm, which acts as the sensing arm.
Figure 2 shows the sensing structure schematically. The taper-like joint structure is cylindrically symmetric, so the interference pattern is independent of the lateral-offset orientation. This avoids the demand of strictly keeping the malposition direction between two joints consistently to obtain high-quality interference spectrum . When the light from SMF1 approaches the taper-like joint, the fundamental core mode LP01 will spread out widely and then the high-order modes LP0j will be excited within the cladding of SMF2. These high-order cladding modes will interfere with each other, which will be coupled into the core of SMF3. Because the lateral-offset distance (13.9 μm) is larger than the core diameter (9 μm) of SMF, SMF3 only receives the interference between the cladding modes rather than the interference between the core mode and cladding modes. The transmission spectrum is monitored with the optical spectrum analyzer (OSA) and can be expressed as [13, 14, 22]Eq. (1), the wavelength corresponding to the interference valley can be expressed by
When the environmental RI (ERI) changes, the influence of effective RIs of high-order modes will be much greater than those of the low-order modes. This is assigned to the larger leakage field of high-order mode outside the fiber comparing with that of the low-order mode. So, different modes will experience different effective RI change. Moreover, the ERI variation can affect the transmission loss of the proposed structure according to Eq. (1). Considering the magnetically tunable RI of MF, the valley wavelength and transmission loss may be sensitive to the external magnetic field when using MF as the cladding of the structure. This can be employed for magnetic field sensing.
3. Experimental details
The experimental setup for investigating the magnetic field sensing properties is shown in Fig. 3. Light from the broadband amplified spontaneous emission source (ASE, wavelength ranging from 1525 to 1610 nm) is coupled into the lead-in SMF. The transmission spectrum is measured and analyzed by the OSA (AQ6370C). The resolution of the OSA is set at 0.02 nm for all experiments. The sensing structure is placed between two poles of an electromagnet, which generates a uniform magnetic field with nonuniformity of less than 0.1% within the sample region. The strength of the magnetic field is adjusted by tuning the magnitude of the supply current. The magnetic field direction is perpendicular to the optical fiber axis. In our experiments, the oil-based Fe3O4 MF (EXP.08105, 1.87% in volume fraction, saturation magnetization of 10mT) provided by Ferrotec Corporation is employed. The diameter of the magnetic nanoparticles is around 10 nm. The MF is diluted with n-dodecane (C12H26). The volume ratio of EXP.08105 to n-dodecane is 1:3. During our experiments, the ambient temperature is kept at 19 °C. The saturation magnetization of the diluted MF is measured to be about 2.18 mT with the vibrating specimen magnetometer (EV9, MicroSense).
4. Results and discussion
Before proceeding with investigating the magnetic field sensing properties, the response and applicability of our sensing structures in a wide range of RI variation will be characterized firstly. The glycerol-water solutions with different mass fractions are prepared, which are listed in Table1.
The typical transmission spectra of the proposed sensing structure surrounded with glycerol-water solutions of different mass fractions are depicted in Fig. 4. The corresponding valley wavelength as a function of ERI is shown in the inset of Fig. 4. Figure 4 indicates that the valley wavelength shifts to long wavelength side with the ERI increase. The mode effective RI is closely related with the mode energy distribution. The value of mode effective RI is closer to that of the mode field concentrated region. The SMF RI is constant and larger than ERI. Most of the mode field is confined inside the SMF in our cases. If the order of stimulated cladding mode is relatively high, the ERI variation can exert a considerable influence on the mode field distribution. In this case, the variation trend of effective RI is mainly related with the variation trend of mode field fraction inside the fiber. The radius of mode field will increase when the ERI becomes larger. Then the fraction of mode field inside the fiber will decrease. This will result in the decrease of effective RI with the ERI. If the order of stimulated cladding mode is relatively low, the mode field distribution is hardly affected by the variation of ERI. The variation trend of effective RI is mainly related with the variation trend of ERI. This will result in the increase of effective RI with the ERI. The red-shift of wavelength valley in Fig. 4 indicates that the Δnij increases with the ERI. This may be assigned to the interference between low-order and high-order modes (or high-order and high-order modes). In , Yang et al. also made a similar discussion concisely.
In addition, it is clear from Fig. 4 that the sensing structure has a very high RI sensing sensitivity when the ERI is approaching the SMF RI. In our experiments, the MF RI is around 1.435 which is very close to the fiber RI, so our magnetic field sensing structure may have a high sensitivity when MF was used as the cladding of the sensing structure. The transmission spectra of our proposed sensing structure at magnetic field strength ranging from 38 to 250 Oe are shown in Fig. 5. The lowest applied magnetic field strength equals the remanence of the electromagnet, which is 38 Oe in our experiments. Our experimental observations indicate that interference valley around 1585 nm (referred as Valley A) disappears gradually when the magnetic field ranging from 225 to 250 Oe. Meanwhile, another interference valley around 1557 nm (referred as Valley B) becomes obvious gradually, which will be utilized to monitor the magnetic field change at high field (H≥250 Oe). From Fig. 4 and Table 1, it is easily deduced that the RI of MF should be between 1.4353 and 1.4429, which is lower than that of SMF. When the RI of MF is lower than the effective RIs of cladding modes, the cladding modes are effectively confined, which will be easily manifested in transmission measurements. So the interference valley will be obvious. However, when the RI of MF is larger than the effective RIs of cladding modes, the cladding modes will be leaked out in the MF, which will exhibit low spectral signature due to leakage and absorption. Then, the interference valley will be unobvious, or even disappear. Therefore, Valley A may represent the coupling to cladding modes with effective RIs close and below to the RI of MF. When the magnetic field is applied, the RI of MF will increase and may surpass the effective RIs of the specific cladding modes, which will result in the leakage of the involved cladding modes. As a result, Valley A will be almost invisible in transmission spectra at high field as shown in Fig. 5. In addition, the RI change of MF with magnetic field will also affect the beat-length between the cladding modes and the scattering losses of the MF also can be modified. The phenomena of Valley B can be discussed similarly. The wavelength corresponding to Valley A as a function of magnetic field strength is plotted in the upper left section of Fig. 6. Figures 5 and 6 display that the wavelength corresponding to Valley A increases linearly with the magnetic field strength, which is assigned to the RI increase of MF with external magnetic field [24, 25]. The sensitivity is 14.1 pm/Oe, which is ~6 times higher than that using MF as the cladding of PCF (~2.367 pm/Oe) , and more than twice larger than that using MF as the cladding of SMF-based Michelson interferometer (~6.49 pm/Oe) .
Figure 7 shows the transmission spectra of the proposed sensing structure under magnetic field strength larger than 225 Oe (225-700 Oe). An interference valley (Valley B) around 1557 nm became obviously on the OSA when the external magnetic field strength increases to 250 Oe from 225 Oe. The interference Valley B also has a red shift with the magnetic field strength and tends to be stable gradually at relatively high field (>475 Oe). The wavelength corresponding to Valley B as a function of magnetic field strength is plotted in the lower right section of Fig. 6. The linear response region appears in the range of 250-475 Oe. The sensitivity of the wavelength shift is around 26 pm/Oe which is higher than Valley A. Besides the wavelength shifts of interference valley with magnetic field, the intensity at any specific wavelength also changes with magnetic field, which can be employed as an additional or complementary sensing means. As an example, the intensity at 1566.36 nm (referred as Site C) is selected for investigating. Similarly, the intensity at any other wavelength can be monitored for sensing purpose. Moreover, sensing based on intensity variation needs a relatively simple interrogation system. Figure 5 shows that the intensity at 1566.36 (Site C) changes slightly with the increment of magnetic field at low field (H≤225 Oe). While Fig. 7 indicates that the intensity of Site C varies remarkably with the magnetic field monotonously at high field (H≥225 Oe), which implies a good response to magnetic field. Figure 8 shows the intensity of Site C has a good linear relationship with the magnetic field strength in the range of 250-475 Oe. The sensitivity is about −0.024 dB/Oe. Recently, the compact magnetic field sensors based on SMS structures attract a lot of interests. In contrast, our sensor has a higher sensitivity [10, 12] or wider linear response range [9–12]. Moreover, Miao et al. presented an S-tapered fiber structure based on the conventional SMF . Their sensor has higher sensitivities of 56 pm/Oe and 0.13056 dB/Oe but the linear response range is relatively narrow (25-200 Oe).
For practical applications, the reproducibility and stability of the sensing performance are critical. Due to the intrinsic nature of MF, fast consecutively change of magnetic field may result in the hysteresis of the optical response of MF. In addition, due to the relatively large thermo-optical coefficient of MF, the sensing structure may be temperature-sensitive. Therefore, the applicability of the proposed sensing structure is limited for fast dynamic sensing or temperature variation case. Under these circumstances, the compensation or calibration techniques are necessary. Besides, it may be appropriate for investigating the polarization effect . In our experiments, the unpolarized light source is utilized and the polarization issues are ignored for simplification.
In summary, a kind of magnetic field sensor based on MF-clad SMF with taper-like and lateral-offset fusion splicing is proposed. It was found that both of the interference valley wavelength and transmission loss are highly sensitive to the external magnetic field. The shift of the valley wavelength has a good linear relationship with magnetic field strength in the wide ranges (38-225 Oe and 250-475 Oe). The corresponding sensitivities at low and high field ranges are 14.1 pm/Oe and 26 pm/Oe, respectively. The intensity at wavelength of 1566.36 nm decreases linearly with magnetic field strength in the range of 250-475 Oe. The sensitivity is obtained to be −0.024 dB/Oe. Considering the simplicity and low cost of the proposed structure, it is a good candidate for implementing magnetic field sensing.
This research was supported by the Shanghai Natural Science Fund (No. 13ZR1427400), the Innovation Fund Project for Graduate Student of Shanghai (No. JWCXSL1302) and the Hujiang Foundation of China (B14004).
References and links
1. S. Xia, J. Wang, Z. Lu, and F. Zhang, “Birefringence and magneto-optical properties in oleic acid coated Fe3O4 nanoparticles: application for optical switch,” Int. J. Nanosci. 10(03), 515–520 (2011). [CrossRef]
2. A. Candiani, W. Margulis, C. Sterner, M. Konstantaki, and S. Pissadakis, “Phase-shifted Bragg microstructured optical fiber gratings utilizing infiltrated ferrofluids,” Opt. Lett. 36(13), 2548–2550 (2011). [CrossRef] [PubMed]
3. S. Pu, X. Chen, L. Chen, W. Liao, Y. Chen, and Y. Xia, “Tunable magnetic fluid grating by applying a magnetic field,” Appl. Phys. Lett. 87(2), 021901 (2005). [CrossRef]
6. P. Zu, C.-C. Chan, L. W. Siang, Y. Jin, Y. Zhang, L. H. Fen, L. Chen, and X. Dong, “Magneto-optic fiber Sagnac modulator based on magnetic fluids,” Opt. Lett. 36(8), 1425–1427 (2011). [CrossRef] [PubMed]
7. S. Pu, S. Dong, and J. Huang, “Tunable slow light based on magnetic-fluid-infiltrated photonic crystal waveguides,” J. Opt. 16(4), 045102 (2014). [CrossRef]
8. Y. Miao, J. Wu, W. Lin, K. Zhang, Y. Yuan, B. Song, H. Zhang, B. Liu, and J. Yao, “Magnetic field tunability of optical microfiber taper integrated with ferrofluid,” Opt. Express 21(24), 29914–29920 (2013). [CrossRef] [PubMed]
9. S. Pu and S. Dong, “Magnetic field sensing based on magnetic-fluid-clad fiber-optic structure with up-tapered joints,” IEEE Photonics J. 6, 5300206 (2014).
10. H. Wang, S. Pu, N. Wang, S. Dong, and J. Huang, “Magnetic field sensing based on singlemode-multimode-singlemode fiber structures using magnetic fluids as cladding,” Opt. Lett. 38(19), 3765–3768 (2013). [CrossRef] [PubMed]
11. Y. Chen, Q. Han, T. Liu, X. Lan, and H. Xiao, “Optical fiber magnetic field sensor based on single-mode-multimode-single-mode structure and magnetic fluid,” Opt. Lett. 38(20), 3999–4001 (2013). [CrossRef] [PubMed]
12. W. Lin, Y. Miao, H. Zhang, B. Liu, Y. Liu, and B. Song, “Fiber-optic in-line magnetic field sensor based on the magnetic fluid and multimode interference effects,” Appl. Phys. Lett. 103(15), 151101 (2013). [CrossRef]
13. P. Zu, C.-C. Chan, W. S. Lew, L. Hu, Y. Jin, H. F. Liew, L. H. Chen, W. C. Wong, and X. Dong, “Temperature-insensitive magnetic field sensor based on nanoparticle magnetic fluid and photonic crystal fiber,” IEEE Photonics J. 4(2), 491–498 (2012). [CrossRef]
16. S. Dong, S. Pu, and J. Huang, “Magnetic field sensing based on magneto-volume variation of magnetic fluids investigated by air-gap Fabry-Pérot fiber interferometers,” Appl. Phys. Lett. 103(11), 111907 (2013). [CrossRef]
18. P. Zu, C.-C. Chan, T. Gong, Y. Jin, W. C. Wong, and X. Dong, “Magneto-optical fiber sensor based on bandgap effect of photonic crystal fiber infiltrated with magnetic fluid,” Appl. Phys. Lett. 101(24), 241118 (2012). [CrossRef]
20. H. V. Thakur, S. M. Nalawade, S. Gupta, R. Kitture, and S. N. Kale, “Photonic crystal fiber injected with Fe3O4 nanofluid for magnetic field detection,” Appl. Phys. Lett. 99(16), 161101 (2011). [CrossRef]
21. Z. Tian, S. S.-H. Yam, and H.-P. Look, “Single-mode fiber refractive index sensor based on core-offset attenuators,” IEEE Photon. Technol. Lett. 20(16), 1387–1389 (2008). [CrossRef]
22. J. Zhang and S. Peng, “A compact SMS refractometer based on HF corrosion scheme,” in Proceedings of IEEE Conference on Photonics and Optoelectronics, China, June 19–21, 2010.
24. C.-Y. Hong, H. E. Horng, and S. Y. Yang, “Tunable refractive index of magnetic fluids and its applications,” Phys. Status Solidi 1(7c), 1604–1609 (2004). [CrossRef]
25. C.-Y. Hong, S. Y. Yang, H. E. Horng, and H. C. Yang, “Control parameters for the tunable refractive index of magnetic fluid films,” J. Appl. Phys. 94(6), 3849–3852 (2003). [CrossRef]
26. P. Childs, A. Candiani, and S. Pissadakis, “Optical fiber cladding ring magnetic field sensor,” IEEE Photon. Technol. Lett. 23(13), 929–931 (2011). [CrossRef]