Open Access
Add to BibSonomy
Abstract
A novel optical fiber liquid level sensor based on a hollow core Bragg fiber (HCBF) was proposed and demonstrated. The HCBF was first designed and successfully fabricated with periodic transmission band in the spectrum and a transmission loss of ~3.48 dB/cm. An inline optical fiber liquid-level sensor was fabricated by simply sandwiching a piece of HCBF between two single mode fibers. The sensing performance was experimentally tested. A linear liquid-level sensitivity of ~1.1 dB/mm, and fast response time less than 3s was obtained by the intensity demodulation measurement. The temperature and refractive index cross-sensitivities were also investigated. The experimental results indicate that our proposed structure has tiny temperature and RI dependence, which makes it a promising liquid level sensing platform for different liquids.
© 2018 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
Liquid level sensing is of great importance in various fields such as fuel or petroleum storage systems, bio-chemical progressing, pharmaceutical industries and environment monitoring. Although the traditional sensors based on mechanical and electrical techniques show excellent performance in normal liquid level measurement [1,2], the consumption of electricity limits their applications, especially for conductive, flammable and explosive environments. Optical fiber sensors with advantages of electromagnetic immunity, chemical-erosion resistance, high sensitivity and compact size, are suitable for sensing applications in harsh conditions. In recent years, various fiber-optic level sensing systems have been proposed based on diverse principles. Long period grating(LPG) as a common structure for external refractive index (RI) measurement have been demonstrated for liquid level sensing [3,4], whose transmission spectra are dependent on the fraction of LPG surrounded by the liquid. Such sensors show good liquid-level response, but always suffer from high temperature cross-sensitivity [3,4]. As a compact sensing structure, fiber Bragg grating (FBG) has also been proposed for liquid level sensing [5,6]. However, chemical etch process should be exploited to make the FBG sensitive to the external variance, resulting in the structural vulnerability. Recently, fiber modal interferometers (FMIs) have also been proposed as liquid level sensors [7–9]. The spectrum of the FMI is determined by the destructive/constructive phase condition of the input light, which is much dependent on the fusion splicing process and the length of the FMI. Thus, it is hardly to control the resonance wavelengths in each experiment and the repeatability of alternative samples is low [8].
Basically speaking, the liquid sensing principle of the above mentioned structures is based on the refractive index change of the ambient media surrounding the sensing interface. So these kinds of liquid level sensors also have RI cross-sensitivity, which means, the calibration should be taken first when the liquid under test changes. Hollow core waveguides or pipe waveguides, based on anti-resonant reflecting guidance (ARRG) [10–13] mechanism, have been demonstrated as an excellent platform for deposited monolayers [11], powders [12] or industrial oil [13] detection. The resonance condition of the waveguides will be affected by the surrounding environments, causing certain wavelength leaked out of the cladding. Recently, a liquid level sensor based on an alcohol-filling hollow core photonic crystal fiber (AL-HCPCF) is proposed and demonstrated [14]. The anti-resonant reflecting guidance mechanism was introduced into the HCPCF, by femtosecond laser assisted selectively filling technology, which increases the cost of the fabrication equipment. A much simple optical fiber liquid level sensor based on silica tube structure has been reported lately, which shows very low RI dependency [15]. However, the transmission spectrum of the sensor is a combination of anti-resonant reflecting and fiber modal interference, which leads to a somehow random spectral pattern. Also the loss of the silica tube is nearly 10 dB/cm, limiting the sensing length.
In this paper, we first design and successfully fabricate a weak-light-confined Bragg fiber, with a low transmission loss of 3.48 dB/cm. An inline optical fiber liquid-level sensor was proposed by simply sandwiching a piece of Bragg fiber between two single mode fibers. A series of sensing performance was experimentally tested, and the results indicate that our proposed structure has a linear liquid-level sensitivity of ~1.1 dB/mm, and fast response time less than 3s. The sensor also exhibits very low temperature and RI dependence based on the intensity modulation measurement, making it a promising platform for liquid level sensing applications.
2. Bragg fiber design, fabrication and characterization
2.1 Bragg fiber design
Hollow core Bragg fiber (HCBF) is a type of photonic crystal fiber, which confines light in the hollow core by the one-dimensional photonic bandgap of the multiple bilayer cladding. The light guided in air core can significantly reduce the nonlinear optical effects. Thus, most reporters focus on high power light delivery in the HCBF, including fiber design [14,15], modal characteristics [16,17], and most importantly the transmission loss [18]. Different from the application of high power delivery, for sensing applications, the guided light in the HCBF is expected to spread out of the cladding, so as to guarantee the sensitivity to the external change [19]. In this work, we designed a weak-light-confined HCBF, for liquid level sensing application. The physical structure of the fiber is depicted in Fig. 1(a), including an air core with radius of 16 μm (),four bilayers and an outer cladding with a radius of 62.5 μm. The refractive index profile of the bilayers is shown in Fig. 1(b). The low index layer and outer cladding layers are pure silica with RI of 1.444 () at the wavelength of 1550 nm. The high index layers are Ge-doped silica with RI of 1.454 () .The average thicknesses of the high and low index layers are 1.06 μm () and 3.07 μm (), respectively.
Fig. 1 (a) Schematic illustration of the physical structure and (b) Refractive index profile of the proposed HCBF.
Simulation was carried out, by using a full-vector finite element method (FEM) combined with perfectly matched layer (PML), to analyze the mode propagation property of the HCBF. By wavelength scanning, the transmission loss spectrum is calculated and given in Fig. 2(a). The four bilayers functioning as Bragg reflectors exhibits strong bandgap effect and there are periodic transmission band around 1550 nm, in which the light transmission loss is relatively low. The corresponding mode distributions at wavelength of 1510 nm and 1521 nm, are shown as inset in Fig. 2(a). It can be seen that the fundamental mode spreads into the cladding region at the wavelength of 1521 nm, corresponding to the peak in the transmission loss spectrum, with a confinement loss higher than 30.74 dB/cm, while the confinement loss at the wavelength of 1510 nm is lower than 0.29 dB/cm with a well confined fundamental mode.
Fig. 2 Simulated transmission loss spectrum (a) in air and (b) in water, respectively. The insert is the corresponding modal distribution of the light at wavelength of 1510 nm and 1521 nm, respectively.
When the outer media of the HCBF is set as water with RI ~1.33, the simulated transmission loss of the wavelengths in the transmission band increases significantly as shown in Fig. 2(b). Meanwhile, the bandgap effect of the bilayer weakens a lot, due to the boundary condition change for the Bragg reflector. Thus, it is possible to realize liquid level sensing by detecting the light transmission power at specific wavelength in the transmission band.
2.2 Bragg fiber fabrication
The multilayer structure of the Bragg fiber preform was deposited by using the modified chemical vapor deposition (MCVD) equipment. The high purity quartz tube (Hereaus F300) with inner/ outer diameter of 25/31mm was cleaned and dried; then placed to the CVD equipment. Firstly, deposed a portion of pure silica layer on the inner side; then deposed a layer of germanium doped glass with the thickness of 380 μm. Repeating the deposition after 4 cycles, a tube contains multiple bi-layers with higher/lower refractive indices was obtained. To ensure the Bragg reflection, the thickness of the higher and lower refractive index layers must be precisely controlled during the sample preparation. After the Rod-in-tube (RIT) process, the preform with outer diameter of 50 mm and inner diameter of 13mm was fabricated.
The preform was then drawn into optical fibers via modified fiber drawing tower. The top of the preform was sealed using in-house designed connection equipment, and negative pressure of the hollow core was carefully controlled under a precision of ± 0.1mbar, during fiber draw process. In order to minimize the temperature-induced diffusion, the drawing temperature was chosen to be 100°C lower than that of normal silica glass preform. Finally the HCBF obtained has an inner diameter 32 ± 0.3μm and 125 ± 0.5μm coating diameter.
2.3 Characterization
The dark field microscopy image of the fabricated HCBF was taken and presented as Fig. 3(b). Four bi-layers with higher/lower refractive indices were periodically surrounding the center air core. The cross section was basically the same as that we designed and the uniform structure confirms the well process control during the fabrication procedure. The transmission property of the fabricated HCBF was tested by using the cut-back method. A HCBF sample with length of 8.5 cm was cleaved and splicing between two single mode fibers (SMFs). The arc parameters of the fusion splicer were optimized to avoid air core collapsing in the HCBF. Figure 3(c) shows the microscopy image of the splicing joint. The interface between the HCBF and SMF was clearly existed with no structure distortion. The transmission spectrum was recorded and then the HCBF was cut 1 cm off to repeat the transmission measurement.
Fig. 3 (a) Schematic illustration of the proposed structure. (b) Microscopy image of the cross section view of the fabricated HCBF. (c) Microscopy image of the splicing joint between the HCBF(left) and the SMF(right).
A series of experiments was conducted with fiber lengths from 8.5 cm to 0.5 cm. The recorded transmission spectra were presented in Fig. 4(a), which exhibit strong periodicity with transmission band and stop band. Compared to Fig. 2(a), the center wavelengths of the transmission band of the fabricated Bragg fiber in Fig. 4(a) coincide well with the simulated ones. The bandwidth of the stopband somehow broadens in Fig. 4, due to the inhomogeneous thickness of the fabricated multilayer structure. The light at 1515.97 nm and 1538.57 nm were confined well and the transmission loss increased slowly as the length of the inserted HCBF increased from 0.5 cm to 8.5 cm. But in some wavelengths like 1525.00 nm and 1545.00 nm, the transmission loss of the light increased significantly when the HCBF length increased. Meanwhile, the center wavelengths of the transmission band and stop band keep also constant as the fiber length changed, which confirms that the stable output spectrum was caused by the bandgap structural properties of the HCBF but not the modal interference. The normalized transmission loss of the light at 1538.57 nm, was calculated by linearly fitting each data points in Fig. 4(b). The transmission loss was obtained as ~3.48 dB/cm, which was just one-third of the loss of the silica tube [19].
Fig. 4 (a) Transmission spectra of the HCBF with different lengths. (b) Normalized transmission loss as a function of HCBF lengths at wavelength of 1538.57 nm.
3. Liquid-level sensor
The liquid-level sensor is simply an inline structure with a segment of HCBF sandwished between two SMFs, as Fig. 3(a) shown. During the fabrication, a HCBF with length of 8 cm was cleaved and spliced with two SMFs. The investigation of the sensing performance of the fabricated sample was tested by using two sets of systems switched by the optical switches (see Fig. 5). For liquid-level response test, the incidence light from a broadband source (BBS) was launched into the leading-in SMF and the transmission spectrum was collected by the Optical Spectrum Analyzer (OSA). For dynamic response measurement, both optical switches were adjusted to channel 2. A tunable laser (TL) instead of BBS was launched into the sensing system as a probing signal at a selected wavelength in the transmission band. The transmission signal was then received by a photodiode (PD). A trigger signal was used to synchronous control the data acquisition and liquid-level changing. The fabricated sample was fixed onto a homemade fiber holder fabricated by 3D printing technology to keep the sensor straight during the test. The liquid-level was precisely controlled by an injection pump.
3.1 Liquid level sensing
The liquid level sensing was performed at the room temperature. Firstly, the deionized water with RI~1.33 was injected into the container, with a liquid level interval of 2 mm. At each liquid level, the transmission spectrum was recorded with a resolution of 0.02 nm. The response spectra were presented in Fig. 6(a). With the rise of the liquid level, more and more HCBF was immersed into the deionized water which changed the boundary condition of the bandgap of the HCBF, causing the increasing of the transmission loss. The experimental results indicate that the proposed structure has good response to liquid-level changing.
Fig. 6 (a) Spectral revolution of the sensor with the rise of the liquid level when deionized water was used. (b) Normalized transmission loss as a function of liquid-level in deionized water (black) and sucrose solution (red), respectively.
In order to test the RI dependency of the proposed liquid level sensor, the sucrose solution with RI~1.40 was chosen as another liquid to test the response of the same sensor. The liquid level was precisely controlled by the injection pump, rising from 0 to 18 mm, with the same interval of 2 mm. The normalized transmission loss at wavelength of 1538.57 nm as a function of liquid level for both in deionized water and sucrose solution was given in Fig. 6(b). One can see that the recorded data for each test is almost the same. By linear fitting process, the sensitivities for both in deionized water and sucrose solution are obtained as ~-1.1055 dB/mm and −1.1143 dB/mm, which have a tiny difference of less than 1% and are much better than that of silica tube as ~- 0.4 dB/mm [19]. Considering that the minimal power detection of the OSA is 0.01 dB, we can estimate that the minimal detectable change in the liquid level is ~0.01 mm. It is experimentally demonstrated that our proposed sensor is RI independent for liquid-level sensing application and it can be widely used for different liquid solutions, for example some industrial oils with high RI values [13].
3.2 Dynamic response
The dynamic response of the sensor was tested by using the setup of channel 2 in Fig. 5. A tunable laser with a selected wavelength at 1538.57 nm was launched into the sensing system as a probing signal. The transmission signal was then received by a photodiode and recorded via a data acquisition card using LabVIEW program with a sampling frequency of 1000 Hz. During the test, the liquid level rose by a step of 2 mm and the interval time for each adjustment was set as 60 s to test the stability of sensor. As shown in Fig. 7(a), the relative intensity collected by the PD has a stepped response, with sharp decent edges and stable plateaus. The insert of Fig. 7(a) is the enlarged part of the response curve between 322.3 s and 325.7s, which responds to the process of liquid level changing from 8 mm to 10 mm. The average response time in our experiment is calculated as ~3.4 s. It should be mentioned that, the relatively long response time mainly stems from the injection process when changing the liquid level which costs nearly 3 s. Thus, the real response time of the sensor should be less than 0.4 s. The relationship between the intensity variation and the lengths of HCBF merged in the liquid are also given in Fig. 7(b). The linear fitting was conducted and the sensitivity for liquid level sensing was calculated as −1.097 dB/mm, which coincides well with the previous test results in Section 3.1. Compared with other sensors based on resonance wavelength shifting which rely on broadband light source and expensive wavelength demodulation device like OSA, this scheme can realize the same performance with relatively low cost, thus is more suitable for practical use.
Fig. 7 (a) Dynamic response for different liquid levels. The inset is the enlarged part of the response curve as liquid level changed from 8 mm to 10 mm. (b) Normalized transmission loss as a function of the immersed sensor length in water.
3.3 Temperature response
The temperature response of the sensor was also investigated. The fabricated sample was heated from 30°C to 70°C with an increment interval of 10°C in a heating oven. The evolution of transmission spectrum is shown in Fig. 8(a), which exhibits a red shift with temperature increment. This is attributed to the thermo-expansion of the Bragg layers and the thermo-optic effect of the silica material. The center wavelength of the transmission band around 1539 nm (red line and squares) and the corresponding transmission loss (black line and squares) as a function of temperature are presented in Fig. 8(b). The sensor shows a linear spectral response with the variation of temperature, and the sensitivity is obtained as 0.044 nm/°C. However, the corresponding transmission loss keeps almost constant with maximum fluctuation of 0.31 dB in the temperature range of 40 °C, which is equivalent to a cross liquid level error of 0.07 mm/10 °C. Thus, by using the intensity demodulation method, the temperature effect on the proposed sensor for liquid level sensing can be ignored. In the practical applications, one can no longer characterize the liquid-level sensing at a fixed wavelength, due to the spectrum shift. However, it should be noted that the spectral shape of the transmission band keeps almost constant as temperature changes. Thus, by monitoring the average intensity at two wavelengths which are symmetrically set with center wavelength in transmission band, one can remove the influence of temperature change, during the liquid-level sensing process.
Fig. 8 (a) Spectral revolution of the fabricated sensor as the temperature rose from 30 °C to 70 °C. (b) Center wavelength of the transmission band (red line and squares) and the corresponding transmission loss (black line and squares) as a function of temperature.
4. Conclusion
In conclusion, we successfully designed and fabricated a weak-light-confined HCBF for sensing application. The fabricated HCBF was characterized with an excellent spectral property and transmission loss of 3.48dB/cm. An inline optical fiber liquid level sensor was proposed and formed by inserting a piece of HCBF between two SMFs. Sensing performance was tested and the experiment results show that, our proposed sensor has a liquid –level sensitivity of ~-1.1 dB/mm, with temperature and RI independence based on the intensity modulation measurement. Such good sensing ability makes our proposed sensor suitable for liquid level sensing in different liquid solutions.
Funding
National Natural Science Foundation of China Grant No. 61605170.
References and links
1. B. W. Northway, N. H. Hancock, and T. Tran-Cong, “Liquid level sensors using thin walled cylinders vibrating in circumferential modes,” Meas. Sci. Technol. 6(1), 85–93 (1995). [CrossRef]
2. F. N. Toch, G. C. M. Meijer, and M. Van-der-Lee, “A new capacitive precision liquid-level sensor,” Proceedings of the conference on Precision Electromagnetic Measurements, (Institute of Electrical and Electronics Engineers, Piscataway, N.J., 1996) pp.356–357.
3. S. Khaliq, S. W. James, and R. P. Tatam, “Fiber-optic liquid-level sensor using a long-period grating,” Opt. Lett. 26(16), 1224–1226 (2001). [CrossRef] [PubMed]
4. H. Y. Fu, X. W. Shu, A. P. Zhang, W. S. Liu, L. Zhang, S. L. He, and I. Bennion, “Implementation and characterization of liquid-level sensor based on a long-period fiber grating Mach-Zehnder interferometer,” IEEE Sens. J. 11(11), 2878–2882 (2011). [CrossRef]
5. A. F. Obaton, G. Laffont, C. Wang, A. Allard, and P. Ferdinand, “Tilted fibre Bragg gratings and phase sensitive-optical low coherence interferometry for refractometry and liquid level sensing,” Sensor Actuat. A 189, 451–458 (2013). [CrossRef]
6. B. Yun, N. Chen, and Y. Cui, “Highly sensitive liquid-level sensor based on etched fiber Bragg grating,” IEEE Photonics Technol. Lett. 19(21), 1747–1749 (2007). [CrossRef]
7. L. Li, L. Xia, Z. Xie, and D. Liu, “All-fiber Mach-Zehnder interferometers for sensing applications,” Opt. Express 20(10), 11109–11120 (2012). [CrossRef] [PubMed]
8. Y. Dong, S. Xiao, H. Xiao, H. Xiao, J. Liu, C. Sun, and S. Jian, “An Optical Liquid-Level Sensor Based on D-Shape Fiber Modal Interferometer,” IEEE Photonics Technol. Lett. 29(13), 1067–1070 (2017). [CrossRef]
9. J. E. Antonio-Lopez, J. J. Sanchez-Mondragon, P. LiKamWa, and D. A. May-Arrioja, “Fiber-optic sensor for liquid level measurement,” Opt. Lett. 36(17), 3425–3427 (2011). [CrossRef] [PubMed]
10. A. M. Zheltikov, “Ray-optic analysis of the (bio)sensing ability of ring-cladding hollow waveguides,” Appl. Opt. 47(3), 474–479 (2008). [CrossRef] [PubMed]
11. B. You, J. Y. Lu, J. H. Liou, C. P. Yu, H. Z. Chen, T. A. Liu, and J. L. Peng, “Subwavelength film sensing based on terahertz anti-resonant reflecting hollow waveguides,” Opt. Express 18(18), 19353–19360 (2010). [CrossRef] [PubMed]
12. B. You, J. Y. Lu, C. P. Yu, T. A. Liu, and J. L. Peng, “Terahertz refractive index sensors using dielectric pipe waveguides,” Opt. Express 20(6), 5858–5866 (2012). [CrossRef] [PubMed]
13. J. Li, H. Qu, and M. Skorobogatiy, “Simultaneous monitoring the real and imaginary parts of the analyte refractive index using liquid-core photonic bandgap Bragg fibers,” Opt. Express 23(18), 22963–22976 (2015). [CrossRef] [PubMed]
14. S. Février, R. Jamier, J.-M. Blondy, S. L. Semjonov, M. E. Likhachev, M. M. Bubnov, E. M. Dianov, V. F. Khopin, M. Y. Salganskii, and A. N. Guryanov, “Low-loss singlemode large mode area all-silica photonic bandgap fiber,” Opt. Express 14(2), 562–569 (2006). [CrossRef] [PubMed]
15. L. Yang, J. Li, Y. Wu, and C. Xiao, “Mode classification and loss mechanism in air-core Bragg fibers,” Opt. Commun. 285(13-14), 3066–3074 (2012). [CrossRef]
16. H. Y. Yao, J. Y. Jiang, Y. S. Cheng, Z. Y. Chen, T. H. Her, and T. H. Chang, “Modal analysis and efficient coupling of TE01 mode in small-core THz Bragg fibers,” Opt. Express 23(21), 27266–27281 (2015). [CrossRef] [PubMed]
17. K. J. Rowland, S. V. Afshar, and T. M. Monro, “Novel Low-Loss Bandgaps in All-Silica Bragg Fibers,” J. Lightwave Technol. 26(1), 43–51 (2008). [CrossRef]
18. S. Liu, Y. Wang, M. Hou, J. Guo, Z. Li, and P. Lu, “Anti-resonant reflecting guidance in alcohol-filled hollow core photonic crystal fiber for sensing applications,” Opt. Express 21(25), 31690–31697 (2013). [CrossRef] [PubMed]
19. S. Liu, J. Tian, N. Liu, J. Xia, and P. Lu, “Temperature Insensitive Liquid Level Sensor Based on Antiresonant Reflecting Guidance in Silica Tube,” J. Lightwave Technol. 34(22), 5239–5243 (2016). [CrossRef]
