Open Access
Add to BibSonomy
Abstract
We experimentally demonstrate a femtosecond laser enabled selective micro-holes drilling technique on the multicore-fiber facet. The precise position of individual cores at the seven-core fiber facet is initially locked by the image processing algorithm, and then six micro-holes are successfully fabricated after the pulse energy of femtosecond laser is optimized. Meanwhile, the use of fabricated seven-core fiber for the application of reflective intensity-modulated fiber optics displacement sensor (RIM-FODS) is comprehensively investigated. By using the beam propagation method (BPM), we theoretically investigate the effect of micro-hole depth on the RIM-FODS performance, in terms of both dead zone and measurement range. We identify that, with the increase of micro-hole depth, the dead zone range can be substantially reduced at the expense of measurement range reduction. However, multiple micro-holes with a successive depth difference can overcome such problem. When the micro-holes with depths of 5, 10, 15, 20, 25, 30 μm are fabricated on the seven-core fiber facet, and the dead zone range can be substantially reduced from 150 μm to 20 μm, together with an extension of measurement range from 250 μm to 400 μm.
© 2019 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
Nowadays, due to the capability to introduce the localized material physical change like permanent refractive index change [1], femtosecond laser micromachining is widely used to fabricate novel fiber optical devices. According to the micromachining position, those devices can be roughly divided into two categories, fiber-lateral fabrication and fiber-facet fabrication. The material physical change arising in the fiber-lateral fabrication is enabled by a focused femtosecond laser to inscribe fiber grating [2–4], Fabry-Perot interferometer [5], Mach-Zehnder interferometer [6], and Michelson interferometer [7] along the longitudinal direction. As for the fiber-facet fabrication, the material physical change occurs at the cleaved fiber facet, leading to new opportunities for novel photonic devices including the micro Fresnel plate [8,9], Michelson interferometer [10], Fabry-Perot interferometer [11], and surface-enhanced Raman scattering fiber probe [12]. Generally, by adjusting the transverse position between optical fiber and the focus of femtosecond laser, the fiber-lateral fabrication is realized. However, the cylindrical shape of optical fiber may bring an aberration, leading to a complicated adjustment to realize the specific fabrication at the desired position. Moreover, the use of bridge fiber for the interferometric device, such as Fabry-Perot or Mach-Zehnder interferometer, from the fiber-lateral fabrication may bring the performance penalty under the environmental perturbation. Alternatively, femtosecond laser can be directly focused onto the facet plane, making the collimation of laser focus convenient. Since the corresponding material physical change arising in the fiber-facet fabrication has a straight-forward impact on the guided light within the optical fiber, its compact footprint is suitable in the harsh environment. However, to recognize the precise location of fiber core is still challenging. When the fiber facet is fixed at the stage, the femtosecond laser fabrication system needs to identify the position of fiber facet and adjust the laser focus accordingly. In particular, the collaborative ability of femtosecond laser fabrication system is ideally desired. When the fiber facet structure needs to be maintained together with micromachining at the specific area, the motion of fiber sample and the interaction time with the femtosecond laser need to be coordinated.
Fiber optics displacement sensor (FODS) plays a critical role in the development of physical parameters measurement [13], not only for the displacement or vibration sensing but also as a secondary transducer for parameters like temperature, bending, pressure and speed. Generally, there exist two kinds of FODSs, phase modulated one and reflective intensity modulated one [14]. For the phase modulated type, a lot of configurations have been reported by using the microfiber coupler [15], Fabry-Perot interferometer [16], multimode interference [17], and whispering gallery mode [18]. The operation principle is based on the fringe counting method, with obvious advantages of high sensitivity, but suffers from high cost and slow response. Alternatively, the reflective intensity modulated type needs a much simpler implementation to achieve high sensitivity. The commonly used configuration of reflective intensity modulated fiber optics displacement sensor (RIM-FODS) is based on the fiber bundle structure. In order to achieve small dead zone range, high sensitivity, and large measurement range simultaneously, fiber with the large numerical aperture (NA) and complicated fiber arrangement are compulsory, from the initial bifurcated fiber to nowadays fiber matrix arrangement [19–21]. Although the optimization of fiber arrangement with larger NA can definitively promote the RIM-FODS performance, it is bulky with a complex signal processing procedure. Furthermore, theoretical investigation reveals that, under a fixed arrangement of RIM-FODS sensor head, there occurs a trade-off among the dead zone range, system sensitivity and measurement range [22].
Recently, we have experimentally demonstrated a multicore-fiber based RIM-FODS [23]. Although the use of multicore-fiber with larger core radius and small core spacing is helpful to improve the system sensitivity and reduce the dead zone range, it brings obvious challenging to the corresponding multicore-fiber fan-in/fan-out device fabrication. In this submission, we introduce a lateral offset between the transmitted and the received channels by the femtosecond laser enabled micro-holes drilling on the seven-core fiber facet. We theoretically identify that the multiple micro-holes with a successive depth difference on the received cores cannot only reduce the dead zone range but also extend the measurement range, in comparison with the traditional seven-core fiber based RIM-FODS. Finally, we experimentally verify that the dead zone range is reduced from 150 μm to 20 μm, and the displacement measurement range is extended from 250 μm to 400 μm.
2. Operation principle
Figure 1(a) presents the optical microscopic image of conventional seven-core single mode fiber facet. The fiber cladding is 150 μm, and the core spacing between two adjacent cores is 42 μm. Each core with a diameter of 9 μm is surrounded by a low refractive index trench. The trench configuration is helpful to reduce the crosstalk among adjacent fiber cores and has almost no effect on the performance of the RIM-FODS. Figure 1(b) shows the schematic sensor head of seven-core fiber based RIM-FODS, where S is the spacing between the transmitted and received cores, Rt and Rr are the core radii of the transmitted and received cores, respectively, d is the measured displacement between the fiber facet and the target surface, and h is the introduced lateral offset between the transmitted and received channels. In our previous work [23], we have theoretically derived the normalized power transfer function M of multicore fiber based RIM-FODS as follows:
where ϭ is the reflection coefficient of the target surface, w(d') is the waist radius at the received fiber plane, as shown in Eq. (2)where NA is the numerical aperture of the transmitted fiber core. When the waist radius w(d') of reflected light at the received core d' = 2d + h is smaller than S-Rr, no light can be collected by the received fiber cores, resulting in the occurrence of dead zone. As the displacement d keep increasing, w(d') becomes larger accordingly. Especially when w(d') is equal to S-Rr, the dead zone vanishes and light can be coupled into the received cores. With the increment of displacement d, the received light power can reach a peak in case w(d') is equal to S + Rr. The value d difference at the maximum and the minimum power is defined as the displacement measurement range. With the further increment of displacement d, the received power slowly reduces, leading to a rear slope of power transfer function, which is not usually considered for the practical application. Although both large fiber core radius and small core spacing are helpful to reduce the range of dead zone [23], the fabrication of strongly coupled multicore fiber and its corresponding fan-in/fan-out devices are challenging. Therefore, we propose to utilize femtosecond laser enabled micro-hole drilling to introduce an offset between the transmitted fiber core and received fiber core, for the purpose of reducing the dead zone range.
Fig. 1 (a) Optical microscopic image of seven-core fiber, (b) schematic design of the RIM-FODS sensor head.
With the help of beam propagation method (BPM), we numerically investigate the micro-hole depth at the transmitted core with respect to the normalized power transfer function. In order to simplify the simulation, the reflective coefficient of target surface ϭ is set 1, so that we can take the mirror image of the received fibers to evaluate the light power under the condition of forward transmission, as shown in Fig. 2(a). Figure 2(a) shows the numerical model of micro-hole on the transmitted fiber. The micro-hole structure is a cylinder type with a fixed diameter equals to single core diameter, and the depth is variable during the simulation. A light with the normalized power from the transmitted core goes through the micro-hole at the fiber facet and captured by the mirror image of the received cores. Then, the light power at the received cores is recorded and processed to obtain the received power with respect to the displacement d. Figure 2(b) shows the received power with respect to the displacement under conditions of various micro-hole depth of the transmitted core. As shown in Fig. 2(b), the whole power transfer function shifts towards the small displacement area with the growing micro-hole depth, resulting in a reduction of both dead zone range and measurement range. Before the simulation, an experiment of RIM-FODS without the micro-holes is conducted, and a dead zone of 150 μm is observed. During the simulation, we firstly calculate the multicore fiber without the micro-holes, and record the optical power when the displacement value is 150 μm. Next, when the received power is higher than the recorded power, we take the corresponding value as the initial point of measurement range. The simulation results reveal that the use of micro-hole is helpful to reduce the dead zone range. However, as for the measurement range, the penalty occurs with the growing micro-hole depth. Thus, there exists a trade-off between the dead zone range and measurement range. Moreover, the use of micro-hole may bring a damage to the waveguide of transmitted core, leading to a sharp reduction of the received power.
Fig. 2 (a) Model of seven-core fiber based RIM-FODS with one micro-hole at the transmitted core. (b) Received power with respect to the micro-hole depth of the transmitted core.
In order to enhance the received power, we carry out another numerical simulation of six micro-holes at the received cores, as shown in Fig. 3(a). Figure 3(b) shows the corresponding power transfer function with respect to the micro-holes depth of the received fiber cores. Unfortunately, the effect of micro-holes depth on the received cores is almost the same as the circumstance of the transmitted core. The reduction of dead zone range is obtained at the expense of measurement range reduction. Next, in order to solve the trade-off between dead zone range and measurement range, we propose to fabricate six micro-holes with a successive depth difference of 5 μm, so that we can take the advantage of various micro-hole depth response for the performance enhancement of both dead zone range and measurement range. By introducing more micro-holes, we can obtain a wide range of depth under the condition of fixed depth difference. Therefore, a transfer function with a relative wide range of displacement response can be expected. Figure 4(a) shows the theoretical model of six micro-holes with a successive depth difference at the received cores.
Fig. 3 (a) Model of seven-core fiber based RIM-FODS with six micro-holes at the received cores. (b) Received power with respect to the micro-holes depth of the received fiber cores.
Fig. 4 (a) Model of seven-core fiber based RIM-FODS with a successive depth difference of micro-holes at the received cores. (b) Received power comparison between seven-core fiber and seven-core fiber with a successive depth of 5 μm based RIM-FODS.
Figure 4(b) shows the received power comparison between traditional seven-core fiber and seven-core fiber with a successive depth difference of 5 μm based RIM-FODS. The RIM-FODS with the use of micro-holes having a successive depth difference is equipped with a good response at the small displacement range, at the expense of small received power. Furthermore, in comparison with the results in Fig. 3(b), micro-holes with a successive depth difference have a smoother power transfer function over the whole measurement range, due to the contribution of various micro-hole depths. The use of micro-holes with a successive depth difference prevents the oscillation arising in the power transfer function, which is observed for six micro-holes with the same depth of 20 μm in Fig. 3(b). However, compared with micro-holes with fixed depth of 20 μm, multicore fiber with a successive depth difference provides a smaller dead zone range, because it can utilize the response of different micro-hole depths. As a result, the rising slope of power transfer function keeps a monotonous increasing, which is extremely important for the practical sensing application.
3. Experimental setup and results
Experimental setup of femtosecond laser enabled the micro-holes drilling on the seven-core fiber facet is schematically shown in Fig. 5. A femtosecond laser (Satsuma, Amplitude System) with operation wavelength of 1030 nm, pulse-width of 270 fs, and repetition rate of 1 kHz, is used as the fabrication source. The energy of laser pulse can be attenuated by a combination of half wave plate and Glan prism, and then focused by a 20X objective (NA = 0.4) onto the seven-core fiber facet placed on the three-dimensional (3D) motion stage. Two CCDs are used to monitor the fabrication process and provide a reference to the image processing algorithm. The flowchart of Fig. 5 shows the image processing algorithm and Fig. 6 shows the details of calibration points and seven-core position identification process. Firstly, two calibration points are arbitrarily drilled on the fiber facet away from each core, and the coordinate of calibration points on the 3D motion stage is recorded. Then, the fiber facet image is captured by CCD and undergoes a series of image processing, including image dilation and erosion, to identify individual fiber cores and calibration points. Finally, the relative distance between seven cores and two calibration points is calculated, and the coordinate of seven cores on the 3D motion stage can be precisely locked. The process of all core position recognition takes less than 1min.
Fig. 6 (a) Image of seven-core fiber facet with two calibration points captured by CCD. Calculated absolute position of (b) individual cores, and (c) calibration points after the image processing.
Figure 7(a) presents the microscope of seven-core fiber with the micro-holes drilling at the position of surrounding cores. Two holes within the yellow circle are used to identify the location of the received cores within the 3D motion stage. During the micro-holes drilling, the movement of 3D motion stage is programmed in order to automatically drill the micro-holes with a constant speed. As shown in Fig. 7(b), the side view of micro-holes presents a shape like a conic, due to the laser focus having narrow and long ellipsoid type along beam axis [24].
Figure 8 presents the relationship between the micro-holes with a successive depth difference of almost 5 μm and corresponding pulse energy of femtosecond laser, under the condition of 1μm/s drilling speed. The depths of these micro-holes are 5.36 μm, 10.43 μm, 15.05 μm, 20.25 μm, 24.42 μm and 30.53 μm located on the six surrounding cores of seven-core fiber, when the corresponding pulse energy are 220 nJ, 410 nJ, 590 nJ, 760nJ, 1.03 μJ, and 1.19 μJ, respectively. Compared with normal multicore fiber based RIM-FODS, the existence of six micro-holes in RIM-FODS with a successive depth difference may bring 3 dB insertion loss under the same input optical power.
Figure 9 illustrates the characterization setup for the seven-core fiber based RIMOFDS. A light from an amplified spontaneous emission (ASE) light source (AlS-CL-20-B-FA, Amonics) with a wavelength range from 1500 nm to 1600 nm is coupled into the central core of the seven-core fiber through a self-fabricated fan-in/fan-out device, and reflected by a mirror (BB1-E04, Thorlabs) mounted on a computer-controlled motion stage with a movement precision of 1μm. When the motion stage functions, the reflected light is collected by the other six cores of seven-core fiber and coupled into the standard single mode fibers (SSMFs) through the same fan-in/fan-out device. Finally, the individual photonic-to-electronic conversion is realized by a power meter (PMSII-B, AcceLink), in order to finally obtain the power transfer function of seven-core fiber based RIMOFDS. The whole measurement is based on the direct-current (DC) power monitoring, together with stable light source. Then, six power values corresponding to the single specific displacement are collected, in order to generate an accurate look-up table. Finally, multiple experiments of displacement sensing are verified with a good repeatability.
For the ease of performance comparison, we carry out a displacement sensing test based on both traditional seven-core fiber and seven-core fiber with the multiple micro-holes drilling RIM-FODS, when the output power of ASE is fixed as 8.93 mW. Firstly, after the system calibration, two power transfer functions of RIM-FODS are obtained and stored with individual look-up tables (LUTs). Next, we are able to obtain the measured displacement value. As shown in Fig. 10, the measured displacement values are compared with the computer-controlled motion stage values with a step of 10 μm, under the use of two different sensor heads of RIM-FODS. Basically, both RIM-FODS configurations present a good linear relationship between the measured displacement and the pre-set displacement. Figure 10 intends to illustrate the repeatable operation of the proposed RIM-FODS, and present the performance promotion of proposed RIM-FODS. We observe that, both displacement sensors possess good linear response, and both measurements have an almost the same linear slope of 1. The deviation from the computer-controlled motion stage value is small, and all measurement errors are within ± 0.5 μm. We infer that the measurement error comes from the calibration error of power transfer function. As shown in Fig. 10(a), owing to the small core radius and large core spacing, RIM-FODS with traditional seven-core fiber possesses a relatively large dead zone range of 150 μm. However, when the seven-core fiber with six micro-holes having a successive depth difference are drilled by the femtosecond laser, the received cores laterally move backward in comparison with the transmitted core, leading to a sharp reduction of dead zone range to 20 μm. Meanwhile, since the multiple micro-holes have an equal-difference depth of 5 μm, the corresponding RIM-FODS can obtain an extension of measurement range from 250 μm to 400 μm.
4. Conclusion
We have experimentally demonstrated a selective micro-hole drilling technique by femtosecond laser on the seven-core fiber facet. With the help of an image processing algorithm, precise location identification of individual cores can be secured. By optimizing the laser pulse energy, six micro-holes with a successive depth difference from 5 μm to 30 μm are successfully fabricated. Experimental results show that, compared with traditional seven-core fiber based RIM-FODS, seven-core fiber with a successive depth difference of almost 5 μm based RIM-FODS can bring a substantial reduction of dead zone range from 150 μm to 20 μm, together with an extension of measurement range from 250 μm to 400 μm.
Funding
National Natural Science Foundation of China (61711530043, 61575071), Key project of R&D Program of Hubei Province (2018AAA041).
References
1. R. R. Gattass and E. Mazur, “Femtosecond laser micromachining in transparent materials,” Nat. Photonics 2(4), 219–225 (2008). [CrossRef]
2. E. Ertorer, M. Haque, J. Li, and P. R. Herman, “Femtosecond laser filaments for rapid and flexible writing of fiber Bragg grating,” Opt. Express 26(7), 9323–9331 (2018). [CrossRef] [PubMed]
3. C. Zhang, L. Cheng, S. Fu, M. Tang, and D. Liu, “Long period fiber grating fabrication by two-step infrared femtosecond laser exposure,” IEEE Photonics J. 9(4), 1–7 (2017). [CrossRef]
4. S. Antipov, M. Ams, R. J. Williams, E. Magi, M. J. Withford, and A. Fuerbach, “Direct infrared femtosecond laser inscription of chirped fiber Bragg gratings,” Opt. Express 24(1), 30–40 (2016). [CrossRef] [PubMed]
5. P. Chen and X. Shu, “Refractive-index-modified-dot Fabry-Perot fiber probe fabricated by femtosecond laser for high-temperature sensing,” Opt. Express 26(5), 5292–5299 (2018). [CrossRef] [PubMed]
6. P. Chen, X. Shu, and K. Sugden, “Ultra-compact all-in-fiber-core Mach-Zehnder interferometer,” Opt. Lett. 42(20), 4059–4062 (2017). [CrossRef] [PubMed]
7. J. Liu and D. N. Wang, “In-fiber beam splitters for construction of in-line Michelson interferometers,” Opt. Lett. 43(17), 4304–4307 (2018). [CrossRef] [PubMed]
8. J. K. Kim, J. Kim, K. Oh, I. Sohn, W. Shin, H. Y. Choi, and B. Lee, “W, Shin, H. Choi, and B. Lee, “Fabrication of micro Fresnel zone plate lens on a mode-expanded hybrid optical fiber using a femtosecond laser ablation system,” IEEE Photonics Technol. Lett. 21(1), 21–23 (2009). [CrossRef]
9. T. Gissibl, M. Schmid, and H. Giessen, “Spatial beam intensity shaping using phase masks on single-mode optical fibers fabricated by femtosecond direct laser writing,” Optica 3(4), 448–451 (2016). [CrossRef]
10. L. Yuan, T. Wei, Q. Han, H. Wang, J. Huang, L. Jiang, and H. Xiao, “Fiber inline Michelson interferometer fabricated by a femtosecond laser,” Opt. Lett. 37(21), 4489–4491 (2012). [CrossRef] [PubMed]
11. Z. Ran, Y. Rao, J. Zhang, Z. Liu, and B. Xu, “A miniature fiber-optic refractice-index sensor based on laser machined Fabry-Perot interferometer Tip,” J. Lightwave Technol. 27(23), 5426–5429 (2009). [CrossRef]
12. X. Lan, Y. Han, T. Wei, Y. Zhang, L. Jiang, H. L. Tsai, and H. Xiao, “Surface-enhanced Raman-scattering fiber probe fabricated by femtosecond laser,” Opt. Lett. 34(15), 2285–2287 (2009). [CrossRef] [PubMed]
13. B. Lee, “Review of the present status of optical fiber sensors,” Opt. Fiber Technol. 9(2), 57–79 (2003). [CrossRef]
14. H. Huang and U. Tata, “Simulation, implementation, and analysis of an optical fiber bundle distance sensor with single mode illumination,” Appl. Opt. 47(9), 1302–1309 (2008). [CrossRef] [PubMed]
15. Y. Chen, S. C. Yan, X. Zheng, F. Xu, and Y. Q. Lu, “A miniature reflective micro-force sensor based on a microfiber coupler,” Opt. Express 22(3), 2443–2450 (2014). [CrossRef] [PubMed]
16. Q. Sun, N. Chen, Y. Ding, Z. Chen, and T. Wang, “Distance detection with optical fiber extrinsic Fabry-Perot interference ultrasonic sensor,” IET International Communication Conference on Wireless Mobile and Computing (IET), pp. 441–443, (2009).
17. A. Mehta, W. Mohammed, and E. G. Johnson, “Multimode Interference-Based Fiber-Optic Displacement sensor,” IEEE Photonics Technol. Lett. 15(8), 1129–1131 (2003). [CrossRef]
18. N. Liu, L. Shi, S. Zhu, X. Xu, S. Yuan, and X. Zhang, “Whispering gallery modes in a single silica microparticle attached to an optical microfiber and their application for highly sensitive displacement sensing,” Opt. Express 26(1), 195–203 (2018). [CrossRef] [PubMed]
19. R. O. Cook and C. W. Hamm, “Fiber optic lever displacement transducer,” Appl. Opt. 18(19), 3230–3241 (1979). [CrossRef] [PubMed]
20. G. He and F. W. Cuomo, “Displacement Response, Detection limit, and Dynamic Range of Fiber-optic lever Sensors,” J. Lightwave Technol. 9(11), 1618–1625 (1991). [CrossRef]
21. B. Guzowski and M. Lakomski, “Realization of fiber optic displacement sensors,” Opt. Fiber Technol. 41, 34–39 (2018). [CrossRef]
22. A. Shimamoto and K. Tanaka, “Geometrical analysis of an optical fiber bundle displacement sensor,” Appl. Opt. 35(34), 6767–6774 (1996). [CrossRef] [PubMed]
23. T. Lan, C. Zhang, S. Fu, B. Zhu, M. Tang, and W. Tong, “Spatial division multiplexing-based reflective intensity-modulated fiber optics displacement sensor,” IEEE Photonics J. 10(4), 1–7 (2018). [CrossRef]
24. Y. Wang, D. N. Wang, M. Yang, and C. R. Liao, “Asymmetric micro-hole structured long period fiber gratings,” Sens. Actuators B Chem. 160(1), 822–825 (2011). [CrossRef]
