Open Access
Add to BibSonomy
Abstract
An all-sapphire extrinsic Fabry-Perot interferometer (EFPI) optical fiber pressure sensor with ultra-wide pressure range and high temperature resistance is proposed and experimentally demonstrated. The sensor is fabricated by direct bonding three sapphire wafers, including the sapphire substrate, the sapphire wafer with a through hole, and the sapphire pressure-sensitive diaphragm. A femtosecond (fs) laser is used to inscribe a through hole in the center of the sapphire wafer and roughen the outer surface of the sapphire pressure-sensitive diaphragm. By using original polished surfaces of sapphire wafers with low surface roughness as reflective surfaces of the Fabry-Perot (FP) cavity, the high-quality interference signal can be obtained, thereby improving the measurement accuracy of the sensor. The optical cavity length (OCL) of the proposed sensor changes linearly with the applied pressure in the wide range of 0 - 50 MPa at room temperature, and the pressure sensitivity is 0.0921 µm/MPa. The pressure measurement accuracy reaches 0.31%FS (full scale). High temperature experiments show that the sensor can work stably at 1000 ℃.
© 2023 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
In-situ high pressure monitoring in high temperature environments plays an indispensable role in a variety of industrial applications, ranging from aerospace, chemical engineering, and petroleum power. Electronical sensors have been used to measure high pressure [1]. However, due to the creep deformation of the silicon and the increased leakage current at high temperature, the operating temperature of electronical sensors is lower than 600 °C [2].
Optical fiber sensors have attracted much interest for physical quantities monitoring in harsh environments due to their advantages of small size, light weight, high temperature resistance, intrinsic safety and immunity to electromagnetic interference [3–5]. Technical schemes for pressure monitoring are mainly based on Michelson interferometer [6], Mach-Zehnder interferometer [7], and Fabry-Perot interferometer [8]. Among them, optical fiber EFPI pressure sensors have been proven to be the optimal scheme because of the reflective structure, fast response, high reliability and ease of fabrication. Optical fiber EFPI pressure sensors are mainly divided into two types: diaphragm-based type and diaphragm-free type. According to the principle that the refractive index of the gas in the open cavity varies linearly with the pressure, diaphragm-free optical fiber EFPI pressure sensors have been developed to measure gas pressure [9,10]. However, common problems with the diaphragm-free type are that the pressure sensitivity decreases significantly as the temperature increases and can only measure gas pressure. Diaphragm-based optical fiber EFPI pressure sensors effectively solve above problems [11]. The operating temperature of the diaphragm-based type is mainly determined by the material of the pressure-sensitive diaphragm. For example, due to the softening of the silica diaphragm at high temperature, the maximum operating temperature of optical fiber pressure sensors based on silica is lower than 800 °C [12,13]. For pressure monitoring in ultra-high temperature environments above 1000 °C, a higher temperature-resistant material is needed to fabricate the diaphragm.
Sapphire (α-Al2O3) has attracted much attention as an ideal material for the construction of sensors used in ultra-high temperature environments because of its promising characteristics, such as high melting point (up to 2040 °C), wide transmission spectral range, corrosion resistance, thermal stability. Currently, sealed sapphire FP cavities have been developed for pressure monitoring based on etching and bonding processes. Mills et al. [14] proposed a sapphire pressure sensor, which used platinum as an intermediate layer to bond sapphire. However, due to the intermediate layer between the sapphire substrate and the sapphire diaphragm, the sensor has problems in sealing at high temperature, limiting the operating temperature to 900 °C. Yi et al. [15] proposed an all-sapphire FP cavity for pressure sensing constructed by combining reactive ion etching with direct wafer bonding. The sensor displayed a linear response with a pressure range of 0.04 - 1.38 MPa at room temperature. Then, to increase the operating temperature of the sensor, the interference signal of the FP cavity was picked up by the sapphire fiber [16]. The stability tests verified that the FP cavity is fully sealed at 1000 °C. Shao et al. [17] proposed an all-sapphire-based FP pressure sensor based on wet etching. The pressure-sensitive diaphragm was fabricated by wet etching solutions with different volume ratios of H2SO4 and H3PO4 at 280 °C. When the roughness of the etched surface was 0.39 nm, the cavity length fluctuation was decreased to ± 5 nm. However, compared to the original polished surface of the sapphire wafer, the surface roughness after reactive ion etching or wet etching will increase, resulting in lower demodulation accuracy. In addition, limited by the structure and the bonding strength, the pressure range of sapphire FP pressure sensors is limited to 0 - 5 MPa [16,18].
In this paper, we demonstrate a sapphire FP pressure sensor with an ultra-wide pressure range based on fs laser micromachining and direct bonding. The sensor head is composed of three directly bonded sapphire wafers. The FP cavity, formed by two polished surfaces of the first sapphire wafer, is used for the temperature measurement. To form a sealed FP air cavity for the pressure measurement, the second sapphire wafer is inscribed with a through hole by a fs laser. The third sapphire wafer is the pressure-sensitive diaphragm. Experimental results show that the proposed sensor has no leakage at an ultra-high pressure of 50 MPa and can simultaneously measure pressure and temperature within the temperature range of 25 - 1000 °C.
2. Design and principle
The schematic diagram of the sapphire pressure sensor is shown in Fig. 1. The sensor head adopts a three-layer structure, including a sapphire substrate, a sapphire wafer with a through hole and a sapphire pressure-sensitive diaphragm. To ensure that the sapphire substrate hardly deforms in the ultra-high pressure environment and the sapphire pressure-sensitive diaphragm deforms linearly with the applied pressure, the thicknesses of the substrate and the diaphragm are 400 µm and 100 µm, respectively. To reduce the signal loss, the thickness of the sapphire wafer with a through hole is 175 µm. The FP pressure sensing cavity is fabricated by directly bonding three sapphire wafers. The outer surface of the sapphire pressure-sensitive diaphragm is roughened by a fs laser. As shown in Fig. 1, the FP cavity composed of R1 and R2 is used for measuring temperature, while the FP cavity composed of R2 and R3 is used for measuring pressure.
The interference spectrum formed by three reflected beams can be expressed as:
When the applied pressure changes, the OCL of the air cavity is determined by the center deflection of the sapphire pressure-sensitive diaphragm, which can be expressed as:
where P is the pressure applied on the diaphragm, E is the Young’s Modulus, µ is the Poisson’s ratio, r and h are the radius and thickness of the diaphragm, respectively. The Young's modulus of the sapphire is 380 GPa, and the Poisson's ratio of the sapphire is 0.27. From Eq. (2), the deflection of the diaphragm is proportional to the change of the pressure. Hence, the pressure can be measured by monitoring the OCL of the air cavity. According to Eq. (2), the pressure sensitivity is deduced to be:When the temperature changes, the OCL of the sapphire substrate and the air cavity will change due to the thermal-optical and thermal-expansion effect. The ambient temperature can be obtained by monitoring the OCL of the sapphire substrate. By the measured temperature and the temperature response of the air cavity, the change of the OCL of the air cavity caused by the temperature (ΔOCLa(T)) can be calculated. Then, the pressure after temperature compensation can be expressed as:
Equation (3) illustrates that increasing the aperture of the through hole and reducing the thickness of the pressure-sensitive diaphragm can increase the pressure sensitivity, but it will reduce the range of the pressure measurement. To ensure the linear deformation of the sapphire pressure-sensitive diaphragm within the pressure range of 50 MPa, the radius of the through hole and the thickness of the diaphragm are designed to be 0.65 mm and 100 µm, respectively. From Eq. (3), the theoretical pressure sensitivity of the proposed sensor is 0.082 µm/MPa.
3. Sensor fabrication
The fabrication process of the proposed sensor is shown in Fig. 2, which is mainly divided into three steps: 1. Fs laser micromachining sapphire wafers; 2. Sapphire direct bonding; 3. Sensor packaging.
Fig. 2. Fabrication process of the sensor. (a) Fs laser micromachining; (b) Sapphire wafer cleaning; (c) Direct bonding; (d) Sensor packaging; (e) Photos of the sensor.
3.1 Fs laser micromachining sapphire wafers
Fs laser has characteristics of high machining accuracy and small heat affected area. Using fs laser to inscribe sapphire wafers can ensure the smoothness of cutting edges and effectively improve the success rate of direct bonding. The parameters of the fs laser we used are 800 nm central wavelength, 35 fs pulse width, and 1 kHz repetition rate. Firstly, the sapphire wafer is inscribed into square pieces with a side length of 5 mm by the fs laser, as shown in Fig . 2(a). The sapphire wafer is fixed on a six-dimensional micro motion platform (M-840, Physik Instruments). The laser pulse power is attenuated to 20 mW through a neutral density filter. Then, the laser beam is focused perpendicularly on the surface of the sapphire wafer by a plano-convex lens with a focal length of 100 mm. By controlling the micro motion platform, the fs laser inscribes straight lines at an interval of 5 mm on the surface of the sapphire wafer. Secondly, a circular through-hole with a radius of 0.65 mm is inscribed in the center of the square sapphire wafer with a thickness of 175 µm. To prevent the wafer fragmentation caused by excessive laser energy, the laser power is adjusted to 5 mW. The laser beam is focused by an objective lens (MPlan FL N, Olympus) with an amplification factor of 20X and a numerical aperture of 0.45. The laser scans repeatedly along a circle with a radius of 0.65 mm at the center of the sapphire wafer until the internal wafer automatically falls off. The micrograph of the edge of the through hole is shown in Fig . 2(a). It can be seen that the edge is flat, and the wafer can be used for direct bonding. Finally, the outer surface of the sapphire pressure-sensitive diaphragm is roughened by fs laser. In order to not change the thickness of the diaphragm, the laser power is adjusted to 1 mW. The laser scans the diaphragm line by line with an interval of 50 µm.
3.2 Sapphire direct bonding
Before direct bonding, hydrophilic layers are deposited on surfaces of sapphire wafers, as shown in Fig. 2(b). Organics and particles on surfaces are eliminated through RCA cleaning (1. Piranha solution (98% H2SO4 : 30% H2O2 = 4 : 1), 130 °C, 15 min. 2. RCA-1 solution (30% NH4OH : 30% H2O2 : H2O = 1 : 2 : 5), 80 °C, 30 min). Then wafers are immersed in 85% H3PO4 and heated to 150 °C for 30 min to remove residual oxides. Cleaned wafers are immersed in diluted H2SO4 for 30 min to deposit hydrophilic OH— layers. Finally, sapphire wafers direct bonding is performed, as shown in Fig. 2(c). Three sapphire wafers are aligned and placed between two ceramic plates. Sapphire wafer pairs are pressurized by the weight of 2 kg. The entire mold is heated to 1300 °C for 20 h.
3.3 Sensor packaging
The single mode fiber (SMF) is used to pick up the interference signal of the sapphire EFPI pressure sensor. The coating layer of the SMF in the high temperature region is peeled off. Then, the SMF is inserted into a ceramic ferrule and fixed by the high temperature resistance adhesive. To remove the reflection of the end face of the SMF, the ceramic ferrule is fixed on the grinder to polish the end face to 8° angle. Finally, the ceramic ferrule is inserted into a ceramic tube and the ceramic tube is fixed in the center of the sapphire wafer by high temperature resistance adhesive. The schematic and photos of the packaged sensor are shown in Fig . 2(d) and Fig. 2(e), respectively.
4. Experiment and discussion
4.1 Signal demodulation
The EFPI interference signal collected by the homemade white-light interference demodulator with a center wavelength of 1550 nm [19] is shown in Fig. 3(a). It can be seen that the reflection spectrum is formed by three-beam interference. Figure 3(b) shows the Fourier transform frequency spectrum of the interference signal. Combined with Fig. 1, three peaks correspond to frequencies of interference signals formed by R2 and R3 (cavity 1), R1 and R2 (cavity 2), R1 and R3 (cavity 3), respectively. Cavity 1 is used to measure pressure, and cavity 2 is used to measure temperature. Although cavity 3 can be used to measure pressure, the OCL of cavity 3 is greatly affected by temperature because cavity 3 contains the sapphire substrate, which will increase the measurement error. A bandpass filtering process is performed on the frequency spectrum. The main frequency (black line) is extracted by the bandpass filter (red line) of the center frequency at the main frequency position. Interference spectrums of two FP cavities are obtained by the inverse Fourier transform, as shown in Fig. 3(c) and Fig. 3(d). Then, corresponding OCLs of two FP cavities are calculated by demodulating the phase information [19].
Fig. 3. Signal demodulation. (a) Reflection spectrum of the sensor; (b) Frequency spectrum (black line) and bandpass filter (red line); (c) Interference signal of cavity 1 after inverse Fourier transform; (d) Interference signal of cavity 2 after inverse Fourier transform.
4.2 Ultra-high pressure test
A hydraulic source is used to generate ultra-high pressure, as shown in Fig. 4. The sensor is fixed in the stainless steel package and connected to the pressure outlet of the hydraulic source. The sensor is connected to the demodulator through the SMF. The water is injected into the hydraulic source from the water inlet. The hydraulic pressure is controlled to increase from 0 to 50 MPa with a step of 1 MPa and the OCL of cavity 1 (OCL1) is recorded real time by the demodulator. The pressure response of the sensor is shown in Fig. 5. The OCL1 decreases linearly in the pressure range of 0 - 50 MPa, and the sensitivity is 0.0921 µm/MPa, which is close to the theoretical value of 0.082 µm/MPa.
To verify performance indicators of the sensor, three pressure cycle tests are carried out. Six pressure values are selected as calibration points. Corresponding OCLs are shown in Table 1. Relevant calibration data are shown in Table 2, where $\overline {{Y_\textrm{i}}} $ is the average OCL at the calibration point during the pressure increase process, $\overline {{Y_\textrm{d}}} $ is the average OCL at the calibration point during the pressure decrease process, $\overline {{Y_\textrm{a}}} $ is the total average OCL. According to the total average value $\overline {{Y_\textrm{a}}} $ at each calibration point, the fitted operating straight line of the sensor is:
where X is the applied pressure, Y is the OCL. The value of Y at the calibration point is recorded in Table 2.
Table 1. The OCLs corresponding to three pressure cycle
Table 2. Calibration data
The non-linearity is the inconsistency between the calibration curve and the operating straight line of the sensor, which is expressed as:
The hysteresis is the maximum deviation between the positive and negative stroke calibration curves of the sensor, which is expressed as:
The repeatability is the inconsistency of calibration curves during pressure increase or pressure decrease, which is expressed as:
The uncertainty is the limit range under the specified confidence of the deviation between output values at calibration points and the operating straight line, which is expressed as:
4.3 Pressure test at high temperature
To investigate the pressure response of the proposed sensor at high temperature, a pressure test system is set up, as shown in Fig. 6. The sensor is inserted into a ceramic tube and placed in the constant temperature zone of the muffle furnace. The ceramic tube is connected to the gas chamber through a stainless steel tube. The SMF is led out from the gas chamber and connected to the demodulator. The pressure response within the temperature range of 25 - 1000 °C is shown in Fig. 7. At each temperature point, the OCL1 decreases linearly with the increase of the applied pressure, which proves that the sensor is well sealed at high temperatures. Limited by the experimental equipment, the gas pressure is limited to 5 MPa.
Fig. 7. Pressure response at high temperature. (a) Pressure test within the range of 25 - 1000 °C; (b) Pressure sensitivity changes with temperature.
As shown in Fig. 7, the OCL1 increases with the ambient temperature. In addition, due to the decrease of the Young’s Modulus with the increase of the temperature, the pressure sensitivity of the sensor increases at high temperatures. Therefore, it is necessary to measure the ambient temperature and remove the influence of the temperature on the pressure measurement. The temperature responses of cavity 1 and cavity 2 at 0 MPa are shown in Fig. 8. As mentioned above, the OCL of cavity 2 (OCL2) is only related to the temperature. The measured temperature can be expressed as:
Then, the measured pressure after temperature compensation can be calculated by using Eq. (4).
4.4 Sensor test
The comprehensive test of the sensor is shown in Fig. 9. Figure 9(a) shows the zero drift, which is less than 0.05%FS in the whole temperature range. Figure 9(b) shows the temperature test of the sensor. Figure 9(c) and Fig. 9(d) show pressure tests at room temperature and high temperature, respectively. It can be seen that the proposed pressure sensor can accurately measure pressure at high temperature without additional temperature sensors.
Fig. 9. Sensor test. (a) Zero drift; (b) Temperature test; (c) Pressure test at room temperature; (d) Pressure test at high temperature.
Finally, the OCL1 is continuously recorded for 300 s at room temperature and atmospheric pressure. As shown in Fig. 10, the measurement resolution is about 2 nm. According to the pressure sensitivity, the pressure resolution is about 22 kPa (0.044%FS).
5. Conclusion
In conclusion, an all-sapphire-based optical fiber pressure sensor with an ultra-wide pressure range is demonstrated. The fs laser is used to inscribe and roughen sapphire wafers. A sealed FP cavity based on sapphire is fabricated by direct bonding, which has no leakage at the ultra-high pressure of 50 MPa. The pressure and temperature can be measured simultaneously by monitoring OCLs of two FP cavities. Experimental results show that the OCL of the pressure cavity varies linearly in the ultra-wide range of 0 - 50 MPa with an accuracy of 0.31% FS and the maximum operating temperature of the sensor reaches 1000 °C. The proposed sensor has characteristics of ultra-high pressure resistance, high temperature resistance, and intrinsic safety, and can provide technical support for in-situ pressure monitoring in harsh environments of high temperature and high pressure.
Funding
National Natural Science Foundation of China (U20B2057).
Disclosures
The authors declare no conflicts of interest.
Data availability
Data underlying the results presented in this paper are not publicly available at this time but maybe obtained from the authors upon reasonable request.
References
1. Z. Zhu, J. Wang, C. Wu, et al., “A Wide Range and High Repeatability MEMS Pressure Sensor Based on Graphene,” IEEE Sensors J. 22(18), 17737–17745 (2022). [CrossRef]
2. S. Guo, H. Eriksen, K. Childress, et al., “High temperature smart-cut SOI pressure sensor,” Sensors Actuat A-Phys. 154(2), 255–260 (2009). [CrossRef]
3. C. Liao, S. Liu, L. Xu, et al., “Sub-micron silica diaphragm-based fiber-tip Fabry-Perot interferometer for pressure measurement,” Opt. Lett. 39(10), 2827–2830 (2014). [CrossRef]
4. Y. Yu, W. Zhou, J. Ma, et al., “High-Temperature Sensor Based on 45° Tilted Fiber End Fabricated by Femtosecond Laser,” IEEE Photon. Technol. Lett. 28(6), 653–656 (2016). [CrossRef]
5. L. Zhang, Y. Jiang, J. Jia, et al., “Micro all-glass fiber-optic accelerometers,” Opt. Eng. 57(08), 1 (2018). [CrossRef]
6. S. Yuan, F. Ansari, X. Liu, et al., “Optic fiber-based dynamic pressure sensor for WIM system,” Sensors and Actuators A: Physical 120(1), 53–58 (2005). [CrossRef]
7. Y. Liu, H. Lin, Y. Dai, et al., “Fiber In-Line Mach–Zehnder Interferometer for Gas Pressure Sensing,” IEEE Sensor J. 18(19), 8012–8016 (2018). [CrossRef]
8. G. Liu and M. Han, “Fiber-optic gas pressure sensing with a laser-heated silicon-based Fabry-Perot interferometer,” Opt. Lett. 40(11), 2461–2464 (2015). [CrossRef]
9. H. Liang, P. Jia, J. Liu, et al., “Diaphragm-Free Fiber-Optic Fabry-Perot Interferometric Gas Pressure Sensor for High Temperature Application,” Sensors 18(4), 1011 (2018). [CrossRef]
10. H. Gao, Y. Jiang, L. Zhang, et al., “Antiresonant mechanism based self-temperature-calibrated fiber optic Fabry–Perot gas pressure sensors,” Opt. Express 27(16), 22181 (2019). [CrossRef]
11. Y. Jiang, J. Li, Z. Zhou, et al., “Fabrication of All-SiC Fiber-Optic Pressure Sensors for High-Temperature Applications,” Sensors 16(10), 1660 (2016). [CrossRef]
12. Y. Zhang, L. Yuan, X. Lan, et al., “High-temperature fiber-optic Fabry – Perot interferometric pressure,” Opt. Lett. 38(22), 4609–4612 (2013). [CrossRef]
13. W. Ma, Y. Jiang, and H. Gao, “Miniature all-fiber extrinsic Fabry-Pérot interferometric sensor for high-pressure sensing under high-temperature conditions,” Meas. Sci. Technol. 30(2), 025104 (2019). [CrossRef]
14. D. A. Mills, D. Alexander, G. Subhash, et al., “Development of a sapphire optical pressure sensor for high-temperature applications,” in Sensors for Extreme Harsh Environments (2014).
15. J. Yi, E. Lally, A. Wang, et al., “Demonstration of an All-Sapphire Fabry–Pérot Cavity for Pressure Sensing,” IEEE Photon. Technol. Lett. 23(1), 9–11 (2010).
16. J. Yi, “Sapphire Fabry–Perot Pressure Sensor at High Temperature,” IEEE Sensors J. 21(2), 1596–1602 (2021). [CrossRef]
17. Z. Shao, Y. Wu, S. Wang, et al., “All-sapphire-based fiber-optic pressure sensor for high-temperature applications based on wet etching,” Opt. Express 29(3), 4139–4146 (2021). [CrossRef]
18. Z. Shao, Y. Wu, Z. Sun, et al., “Excellent repeatability, all-sapphire Fabry Perot optical Pressure sensor based on wet etching and direct bonding for Harsh Environment Applications,” Opt. Express 29(13), 19831–19838 (2021). [CrossRef]
19. Y. Jiang, “Fourier Transform White-Light Interferometry for the Measurement of Fiber-Optic Extrinsic Fabry–Pérot Interferometric Sensors,” IEEE Photon. Technol. Lett. 20(2), 75–77 (2008). [CrossRef]
