A fringe visibility enhanced fiber-optic Fabry-Perot interferometer based ultrasonic sensor is proposed and experimentally demonstrated for seismic physical model imaging. The sensor consists of a graded index multimode fiber collimator and a PTFE (polytetrafluoroethylene) diaphragm to form a Fabry-Perot interferometer. Owing to the increase of the sensor’s spectral sideband slope and the smaller Young’s modulus of the PTFE diaphragm, a high response to both continuous and pulsed ultrasound with a high SNR of 42.92 dB in 300 kHz is achieved when the spectral sideband filter technique is used to interrogate the sensor. The ultrasonic reconstructed images can clearly differentiate the shape of models with a high resolution.
© 2018 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
Ultrasonic seismic physical model imaging technology is a method to replace the seismic wave by ultrasound and simulate the real geological structures by constructing seismic physical models in the laboratory. By means of large data acquisition, establish different types of data models to provide information guidance for resource extraction. Compared to the field detection, the seismic physical model imaging is cost effective, controllable and efficient [1–3]. In the seismic physical model detection, piezoelectric transducer (PZT) is widely used as ultrasonic source and receiver [4,5]. However, PZT is very sensitive to ambient electromagnetic disturbances due to the electrical loading effects. And its sensitivity decreases with decreasing size and the corresponding electrical capacitance [6,7]. In addition, performance degradation in harsh environment also limits its filed application.
Because of the outstanding performances on these issues, fiber-optic sensors have been widely researched in ultrasonic detecting [8,9], including fiber gratings , fiber interferometers  and fiber lasers  based structures. The spectral sideband filter technique is commonly used to achieve signal demodulation for ultrasonic detecting [13,14]. The larger slope of the sensor’s spectrum, the higher sensitivity of the ultrasonic response. Therefore, a phase-shifted fiber grating with a narrower central wavelength region was used to replace a fiber Bragg grating to achieve a higher ultrasonic response [15,16]. For the fiber-optic Fabry-Perot (F-P) sensor, the larger cavity length results in a smaller free spectral range (FSR). In order to obtain a larger spectral slope, a smaller FSR and a greater interference fringes visibility (FV) is needed. However, the small FV induces large optical loss. To solve this problem, a short section of graded index fiber (GIF) is spliced to a single mode fiber (SMF) for beam collimation to reduce optical loss of the F-P cavity was proposed . This simple structure will transform the converging light into parallel light output with low loss and high stability.
In addition, a fiber-optic F-P ultrasonic sensor is generally composed of a diaphragm and a fiber-optic end-face as two reflectors for a higher acoustic response. The properties of the diaphragm material directly definite the performance of the sensor. Metal diaphragms , polymer membranes [19,20], photonic-crystal membrane  and multilayer graphene diaphragms  etc. have been proposed. However, the complicated preparation, poor chemical stability and heat resistance of the diaphragm materials which limit those sensor’s application. And the larger Young’s modulus causes the lower acoustic response.
In this paper, we propose and experimentally demonstrate a fringe visibility enhanced fiber-optic F-P functionalized diaphragm based ultrasonic sensor for seismic physical models imaging. The sensor consists of a fiber-optic collimator and a PTFE diaphragm to form an F-P interferometer. Spectral sideband filter technique is used to interrogate the sensor and a high Signal to noise ratio (SNR) is achieved, making it can clearly distinguish the interface information when the sensor is applied in seismic physical model scanning and imaging.
2. Sensor design and operation principle
2.1. Sensor design
As shown in Fig. 1(a), a short section of GIF is spliced to a SMF for beam collimation to reduce optical loss. The refractive index of GIF is gradually increased in the radial direction, and therefore the diverged output light can be collimated owing to the period focusing effect. In previous work , the smallest divergence angle is obtained when the length of GIF is set to be 260 μm. As shown in Fig. 1(b), a fiber cleaver (Fujikura CT-32) was installed on a basement with a micrometer to precisely control the length of GIF to form a collimator. First, the spliced fiber was fixed on the basement and mounted on the cleaver by fiber holder with the blade perpendicular to the fiber axis. Then adjusted the relative position between the blade and fiber by micrometer under the microscope. After aligning the SMF-GIF splicing point to be coincidence with the blade, pulled the fiber to 260 μm by adjusting the micrometer again and finally completed the precise cleaving by activating the blade.
Afterwards, the collimator and PTFE diaphragm (thickness of 30 μm) was combined to form a fiber-optic F-P interferometer. As shown in Fig. 2(a), the PTFE diaphragm was adhered to one end of the aluminum tube (inner diameter of 2 mm) using the T530 glue. Then the collimator was inserted into a plexiglass tube (inner and outer diameters are 0.4 mm and 1.6 mm, respectively) and then into an aluminum tube. Next, the collimator and aluminum tube were respectively fixed on the basement with a micrometer. To obtain a desired interference spectrum, the distance between collimator and PTFE diaphragm was adjusted by the micrometer and real-time observing the reflective spectrum on a fiber grating demodulator. AB epoxy adhesive was used to bond the moving parts and the picture of the packaged sensor is shown in Fig. 2(b).
The use of the cost-effective PTFE diaphragm extends the application range of the sensor. Moreover, the extremely high corrosion resistance, temperature resistance (−196∼260 °C) and anti-aging (has the best aging life in plastic) of PTFE endow the sensor’s ability of long-term use in harsh environment. In addition, the Young’s modulus of PTFE (1.14∼1.42 GPa) is much lower than other materials such as polymer (4∼5 GPa), fused-silica fiber (70 GPa) and metal (Tens to hundreds GPa), which also improves the acoustic response of the proposed sensor.
Figure 3(a) shows the interference spectrum of the proposed sensor (red line), with a FSR of 4 nm and a max extinction ratio (ER) of 14.96 dB, respectively. Compared with the sensor’s spectrum without collimator (blue line), with a FSR of 4.82 nm and a max ER of 9.30 dB, the FSR is almost same and the ER is significantly improved. And the FV is given byFig. 3(b), the linear slope of the sensor with collimator is 15.98 dB/nm, which is 2.56 times over the slope of 6.24 dB/nm of the one without collimator. It can be seen the spectral slope of the proposed sensor has been greatly enhanced. In order to analyze the characteristics of the interference pattern, the spectrum with collimator is transformed to the spatial frequency, which is shown in Fig. 3(c). The three peaks are caused by the reflection light from end face of GIF and two surfaces of PTFE diaphragm, which results in the three waves interference.
2.2. Operation principle
When ultrasound is applied to PTFE diaphragm, it will lead to the vibration of diaphragm and the change of diaphragm thickness h. Then, the change of F-P cavity length l results in a phase shift Φ and causes a shift of the interference pattern. For spectral sideband filter demodulation technique, the initial static operating point of sensor should be set at the quadrature point which gives a linear response of the reflected optical power P0 to ultrasonic pressure P. The variation in P0 can be express as20] 19]
3. Ultrasonic measurement and discussion
3.1. System setup
Figure 4 shows the schematic diagram of the seismic physical model ultrasonic imaging system. Plexiglas blocks are placed in a water tank as the physical model. Four copper cylinders (depth of 1.20 cm) are sandwiched between Plexiglas block and water tank bottom. A PZT is driven by a function generator to generate ultrasound. The PZT and fiber-optic sensor are fixed on a scanning platform (SMC100) which is located above the water surface and controlled by a personal computer (PC) via the RS232 interface. The fiber-optic sensor is illuminated by a tunable laser (Santec, TSL-710) with a line width of 100 kHz and power of 20 mW. The reflected light of the sensor reaches a balanced photodetector (PD: New focus, 2117-FC) with a peak responsivity of 1 A/W at 1550 nm for photoelectric conversion. To obtain the optimal ultrasonic response, the wavelength of laser is tuned to the quadrature point of one linear side of the interference spectrum. When ultrasound is loaded on the sensor, the cavity length of the F-P interferometer will be changed, resulting in the shift of the interference spectrum. Spectral sideband filtering method is employed to transfer the spectral shift to the change of optical power, then converted to voltage signal by PD and displayed on PC.
3.2. Characterization of ultrasonic sensor
A continuous sinusoidal wave of 300 kHz was input into the PZT to test the proposed sensor. Figure 5(a) shows the temporal time domain responses and can be seen that the sensor can clearly detect ultrasound with a peak-to-peak voltage of 3.4 V. Figure 5(b) shows the single frequency spectrum of the sensor, which is calculated by taking the Fourier transform of Fig. 5(a).
In order to observe the reflection surfaces of physical models, a pulse square wave of 300 kHz is used to drive the PZT. Figure 6(a) shows the ultrasonic signal reflected from a rectangular Plexiglas block with a thickness of 6 cm. The reflected wave from the upper and lower surface of rectangular Plexiglas block and water tank bottom can be clearly differentiated. The sensor with collimator acquired a signal peak-to-peak voltage of 1.4 V and noise is only 10 mV, resulting in a SNR of 42.92 dB. Compared with the sensor without collimator (peak-to-peak voltage is only 0.47 V, SNR is 33.44 dB), the ultrasonic response sensitivity has been enhanced. In our test, the ultrasonic velocities are 1473 m/s and 2692 m/s in water and Plexiglas block, respectively. The travel time of ultrasonic reflected waves can be calculated as 105.60 μs, 150.18 μs and 166.47 μs, respectively, which is in accordance with the experimental results shown in Fig. 6(a). To further analyze the temporal response, the time domain signal (red line) is transformed to frequency spectrum, as shown in Fig. 6(b). It can be seen that the main frequency response is 300 kHz and the proposed sensor has a broadband response.
3.3. Seismic physical model imaging
As shown in Fig. 7(a), a rectangular Plexiglas block (50 cm×50 cm×6 cm) was used as a physical model to test the sensor’s ability of ultrasound imaging. In experiment, the horizontal distance between the sensor and PZT were 3.5 cm and the vertical distance between the upper surface of Plexiglas block and water surface were 4.5 cm, respectively. The sensor and PZT were programmed to move 5 cm along the water surface with a step of 1 mm. Figure 7(b) shows the ultrasonic image reconstructed by time-of-flight approach using our imaging system after filtering processing and signal amplification. Each layer can be clearly observed. The surface of the water tank bottom with the thickness of 1 cm and ground can be clearly distinguished, which indicates that the sensor has a high resolution. The spatial resolution is mainly determined by the frequency of ultrasound. In addition, the spectral envelope of the ultrasonic reflection wave will also limit the spatial resolution of imaging. The duration of envelope is related to the relative position of the sensor and PZT in ultrasonic detection. Therefore, the spatial resolution can express as Fig. 6(a) (red line) and ultrasonic velocity, the spatial resolution can be inferred up to 3.7 mm.
Afterwards, as shown in Fig. 8(a), a sunken block with a center depression area (3 cm×3cm×5.2 cm) to simulate geological faults was scanned and imaged. The horizontal distance between the sensor and PZT were 4.5 cm and the vertical distance between the upper surface of sunken block and water surface were 3.2 cm. As the ultrasonic imaging result shown in Fig. 8(b), the fault and each layer can be clearly observed. As a result of the different ultrasonic propagation velocity in water and physical model, the image of water tank bottom also occurred fault. And due to the diffraction of ultrasound at the fault, the imaging of the water tank bottom is slightly deformed, which is consistent with the actual situation.
In addition, further test was also done by using a semi-cylindrical block (radius of 5 cm) as shown in Fig. 9(a) to simulate geological dome structure. In experiment, the horizontal distance between the sensor and PZT were 5 cm and the vertical distance between the highest point of semi-cylindrical block and water surface were 3.5 cm, respectively. The ultrasonic image reconstructed is shown in Fig. 9(b) after scanning 7 cm. With the same as before, due to the different ultrasonic velocity in water and model, the image of the model bottom and water tank bottom are also arched. This is in good consistent with shape of the model and the actual situation.
We proposed and experimentally demonstrated a sensitivity enhanced optical fiber ultrasonic sensor for seismic physical models imaging. The sensor consists of a PTFE diaphragm and a fiber-optic collimator to form an F-P interferometer. PTFE endow the sensor with the ability to long-term use in harsh environments and its lower Young’s modulus improves the ultrasonic response. Moreover, by using an optical fiber beam collimator, a high ultrasonic response was obtained. The results show that the proposed sensor exhibits a high response to both continuous and pulsed ultrasound with a high SNR of 42.92 dB in 300 kHz. By scanning seismic physical models, the reconstructed images by time-of-flight approach can clearly differentiate the shape of models with a high resolution. The proposed sensor is simple to fabricate, ultra-compact in size and cost effective, it can replace the use of PZT in harsh environment for long-term measurement.
National Natural Science Foundation of China (NSFC) (61505163, 61327012, 61735014); Natural Science Basic Research Plan in Shaanxi Province of China (No. 2017JM6076).
References and links
1. D. H. Sherlock, J. A. Mcdonald, and B. J. Evans, “Seismic imaging of sandbox models,” Freien Universität Berlin 11(1), 1371–1374 (1997).
2. M. S. Vassiliou, M. Abdelgawad, and B. R. Tittmann, “Ultrasonic physical modeling of seismic wave propagation from graben-like structures,” Final Report, 19 Feb. 1985–30 Jun. 1987 Rockwell International Corp. Thousand Oaks, CA.
3. A. Bakulin, V. Grechka, and I. Tsvankin, “Estimation of fracture parameters from reflection seismic data–Part I: HTI model due to a single 313 fracture set,” Geophysics 65(6), 1788–1802 (2000). [CrossRef]
5. K. Toda and A. Sawaguchi, “Ultrasonic imaging system using a leaky surface acoustic wave transducer composed of piezoelectric ceramic and fused quartz,” Journal of Applied Physics 69(1), 103–108 (1991). [CrossRef]
6. S. Lin and S. Wang, “Radially composite piezoelectric ceramic tubular transducer in radial vibration,” IEEE Trans. Ultrason. Ferroelectr. Freq. Control 2011 . 58, 2492–2498 (2011).
7. D. Kerseyetal, “Fiber grating sensors,” J.Lightw.Technol. 15(8), 1442–1460 (1997).
9. K. S. Kim, Y. Mizuno, and K. Nakamura, “Fiber-optic ultrasonic hydrophone using short Fabry-Perot cavity with multilayer reflectors deposited on small stub,” Ultrasonics 54(4), 1047–1051 (2014). [CrossRef] [PubMed]
10. Z. Yu, Q. Jiang, H. Zhang, and J. Wang, “Theoretical and experimental investigation of fiber Bragg gratings with different lengths for ultrasonic detection,” Photonic Sensors 6(2), 187–192 (2016). [CrossRef]
11. J. Ma, Y. Yu, and W. Jin, “Demodulation of diaphragm based acoustic sensor using Sagnac interferometer with stable phase bias,” Opt. Exp. 23(22), 29268–78 (2015). [CrossRef]
12. S. Foster and A. Tikhomirov, “A fiber laser hydrophone,” Proceedings of Optical Fiber Sensors 5855, 627–630 (2005).
13. J. R. Lee and H. Tsuda, “Fiber optic liquid leak detection technique with an ultrasonic actuator and a fiber Bragg grating,” Opt. Lett. 30(24), 3293–3295 (2005). [CrossRef]
14. C. Broadway, D. Gallego, G. Woyessa, A. Pospori, O. Bang, and H. Lamela, “Polymer optical fibre sensors for endoscopic optoacoustic imaging,” Opto-Acoustic Methods and Applications in Biophotonics II 9539, 953907 (2015). [CrossRef]
15. A. Rosenthal, D. Razansky, and V. Ntziachristos, “High-sensitivity compact ultrasonic detector based on a pi-phase-shifted fiber Bragg grating,” Opt. Lett. 36(10), 1833–1835 (2011). [CrossRef] [PubMed]
16. J. Guo, S. Xue, Q. Zhao, and C. Yang, “Ultrasonic imaging of seismic physical models using a phase-shifted fiber Bragg grating,” Opt. Exp. 22(16), 19573 (2014). [CrossRef]
17. R. Wang, Z. Liu, and X. Qiao, “Fringe visibility enhanced Fabry-Perot interferometer and its application as gas refractometer,” Sensors and Actuators, B: Chemical 234, 498–502 (2016). [CrossRef]
19. P. C. Beard and T. N. Mills, “Extrinsic optical-fiber ultrasound sensor using a thin polymer film as a low-finesse Fabry-Perot interferometer,” Applied Optics 35(4), 663–675 (1996). [CrossRef] [PubMed]
20. Q. Y. Wang and Q. X. Yu, “Polymer diaphragm based sensitive fiber optic Fabry-Perot acoustic sensor,” Chinese Optics Letters 8(3), 266–269 (2010). [CrossRef]
21. W. Jo, O. C. Akkaya, O. Solgaard, and M. J. F. Digonnet, “Miniature fiber acoustic sensors using a photonic-crystal membrane,” Optical Fiber Technology 19(6 PART B), 785–792 (2013). [CrossRef]
22. J. Ma, H. Xuan, H. L. Ho, and W. Jin, “Fiber-optic fabry-perot acoustic sensor with multilayer graphene diaphragm,” IEEE Photonics Technology Letters 25(10), 932–935 (2013). [CrossRef]
23. A. Bar-Zion, C. Tremblay-Darveau, O. Solomon, A. Dan, and Y. C. Eldar, “Fast vascular ultrasound imaging with enhanced spatial resolution and background rejection,” IEEE Transactions on Medical Imaging 36(1), 169–180 (2016). [CrossRef] [PubMed]