Open Access
Add to BibSonomy
Abstract
A multiple reflections-enhanced fiber-optic photoacoustic (PA) gas sensor for gas micro-leakage is introduced. Multiple reflections of the excitation laser occur on the inner surface of a reflective ring to enhance the PA signal. The PA signal is obtained by measuring the deflection of the gold-coated poly (phenylene sulfide) (PPS) diaphragm with a Fabry-Perot interferometer (FPI). The second harmonic wavelength modulation spectrum (2f-WMS) technology can essentially eliminate the fundamental frequency noise generated by the wavelength-independent absorption of the reflective ring. Experimental results show that the PA signal can be effectively enhanced 11.7 times by the multiple reflections optical path compare with the double-pass optical path. The minimum detection limit of the system is achieved to be 23.6 ppb. The designed PA gas sensor is suited for remote detection of gas micro-leakage.
© 2021 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
Spectroscopic technology is a fast response, minimal drift, and can perform in real-time monitoring with no sample consumption technology [1], divided into several categories according to the different techniques [2]. Photoacoustic spectroscopy (PAS) is one of the commonly used spectroscopic technology, which has high detection sensitivity because the intensity of the PAS signal is directly proportional to the optical power of the excitation source and the zero-background property [3,4]. However, the traditional PAS is usually using a capacitive microphone to detect the PA signal, which limits their applications in the circumstances with explosive environments [5,6], as well as electromagnetic interference (EMI) [7,8]. Different from the traditional PAS, the all-optical PAS uses optical acoustic sensing technologies to detect the PA signal. Due to the uncharged and high-sensitivity characteristics, all-optical PAS is an ideal substitute in the circumstances above [9].
Multiple optical acoustic sensing technologies have been used in all-optical PAS, such as intensity-modulated technology [10,11], fiber Bragg grating (FBG)-based technology [12], and interferometer-based technology [13–18]. Because of the compact structure and high sensitivity advantages, FPI has been widely used in all-optical PAS [16–18]. Benefit from the characteristics of small size and high sensitivity in PAS, the FPI-based fiber-optic PA sensor adapts well to remote detection [19–22], in-situ detection [23–26], and real-time monitoring [27].
Cao, et al. [19] first developed a fiber-optic PA gas sensor that is equipped with an FPI-based polymer diaphragm microphone, reaching a noise equivalent concentration (NEC) of 4.3 ppm for C2H2. However, using the same fiber to deliver excitation light and detection light will increase the complexity of the system structure. Gruca, et al. [20] separated the excitation light and detection light using two single-mode fibers. A microcantilever is used instead of the diaphragm, enhancing the NEC up to ∼1.5 ppm for C2H2. To enhance the sensitivity, Zhou, et al. [24] used a capillary tube as a PA cell, reaches an NEC of 24 ppb for C2H2, and shows a brilliant real-time in-situ monitor characteristic for headspace monitoring of CO2 production during yeast fermentation. Chen, et al. [21] reported a fiber-optic PA sensor for remote monitoring of gas micro-leakage, which shows a real-time response of 11.2 s and an NEC of 20 ppb for C2H2. This sensor has been tested for remote detection of C2H2 by using a 1 km conductive fiber. In Dissolved gas analysis (DGA), the fiber-optic PA gas sensor has an excellent application prospect because of the uncharged and high-sensitivity characteristics. Zhou, et al. [25] first developed an immersion photoacoustic spectroscopy (iPAS) for DGA in power transformers. The fiber-optic sensor head is mounted inside a small permeable chamber made of silicone-coated fiberglass sleeving, where dissolved gas diffuses while the power transformer oil is kept out, reaches an NEC of 47 ppb for C2H2. Optimizing the parameters can further miniaturize the sensor volume and enhance the PA signal [26].
Increasing the excitation light power can effectively improve the PA signal, directly proportional to the optical power. However, the excitation light is directly irradiated to the sound perceiving diaphragm in the fiber-optic sensor so far [19,27], which may heat the diaphragm and change its linear interval or even burning through the diaphragm. Another method to enhance the excitation light power is by using multiple reflections [28–34]. Chen, et al. [30] reported a PA sensor based on multiple reflections in a cylindrical PA cell, which is 6.4 times that of a single-pass cell. Tuzson, et al. [35] reported a compact multipass cell (MPC) design for laser absorption spectroscopy. Their previous work showed that this type of multiple reflections structure might have a good application prospect in PAS [34]. The annular reflector cavity has its unique advantages compared to retro-reflection cavity [28] and Herriott cell [32] in PAS, the top surface of the annular reflector cavity is a standard circle, which means that a big circular diaphragm can be used even with the same radius as the PA cell.
In this work, we present a multiple reflections-enhanced fiber-optic PA gas sensor for remote monitoring of gas micro-leakage. The collimated excitation light is incident from the sidewall of the reflective ring. Multiple reflections occur on the inner surface of the reflective ring to enhance the excitation light power. We used an F-P interferometer and a gold-coated PPS diaphragm to detect the PA signal owing to its excellent heat resistance, fatigue resistance, dimensional stability, thermal stability, and processability [36]. The PPS diaphragm radius is the same as the PA cell to maximize the diaphragm's sensitivity. The PA signal is measured by 2f-WMS technology to eliminate the fundamental frequency noise generated by the wavelength-independent absorption of the inner surface. The designed PA gas sensor has been tested for trace C2H2 gas detection with double-pass and multiple excitation light reflections. It is shown that the multiple reflection structure can significantly improve the sensitivity of the PA gas sensor.
2. Experimental details
2.1 Fiber-optic PA sensor head design
The construction details of the PA gas sensor is shown in Fig. 1(a). A membrane with a membrane ring, a reflective ring, and a stainless steel gasket make up the PA cell. [36]A gold-coated by magnetron sputtering PPS diaphragm is used as the sensing diaphragm, and a metal ring is used to support the sensing diaphragm. The thickness and effective radius of the PPS diaphragm are 5 µm and 5.25 mm, respectively. The gold-coated PPS diaphragm's reflectance is ∼80% over the wavelength range from 1.5 to 1.6 μm. Hence the excitation laser will reflect once at the diaphragm as a double-pass optical path. An aluminum ring with a notch to put the collimator 1 is used as a reflective ring. The radius of the collimator lens is ∼0.5 mm. The inner arc length of the V-shaped notch ln is ∼1 mm. The inner surface radius r0 of the reflective ring is 5.25 mm, polished by metallographic sandpaper with the grit size from 100# to 10000#. A stainless steel gasket with a thickness of 0.3 mm is fabricated by a laser marking machine. Two small holes with a diameter of ∼0.5 mm are opened on the stainless steel gasket for gas exchanging.
The above components together form a non-resonant PA cell. In a non-resonant PA cell, the PA signal is inversely proportional to the PA cell volume [37]. Besides, the diaphragm sensitivity is proportional to the fourth power of the diaphragm radius [38]. Therefore, increasing the diaphragm radius and reducing the PA cell volume can effectively enhance the PA signal in the non-resonant PA cell. The diaphragm radius is designed the same as the radius of the PA cell to make full use of the advantages of the reflective ring structure. With a short length design to reduce the PA cell volume, the PA signal can be effectively enhanced.
The PA cell is glued on a polypropylene holder with five vent holes. One of the vent holes is used to insert the collimator 2. A sponge filter is placed between the stainless steel gasket and polypropylene holder to prevent particulates from entering the PA cell and reduce the escape of PA signals from the small holes. The excitation laser is emitted into the PA cell through collimator 1 and collimator 2 separately by multiple passes and double-pass. The lens edges of collimator 1 and collimator 2 are tangent to the reflective ring and stainless steel gasket edges, respectively. The ceramic ferrule is used to transfer the detection light of the Fabry-Perot (F-P) acoustic sensor. The outer radius of the ceramic ferrule r1 is 0.625 mm. The air cavity between the end face of the ceramic ferrule and the PPS membrane constitutes an extrinsic F-P cavity, which is ∼100 µm. The length of the PA cell d is 2 mm, therefore the corresponding PA cell volume is only 171 µL. The overall structure of the PA gas sensor is shown in Fig. 1(b). Figure 1(c) is the ray-tracing simulation of the toroidal cell. Considering the small length of the light path and in order to simplify the calculation process, the influence of the divergence angle is ignored. Meanwhile, it is assumed that the excitation laser propagates in the meridian plane. The excitation laser that enters the PA cell at an angle θ with respect to the radial direction will be reflected back and forth with stepwise advances along the circumference of the ring surface until it reaches the notch again [35]. Therefore, the number of passes is determined by the angle of incidence, which can easily be calculated from the regular star polygon geometrical rule
where n is the number of vertices in the regular star polygon, q is the density of the p-sided regular star polygon [39], n and q are relatively prime. The path reflection times N is equal to n − 1. The larger the q, the longer the optical path of each reflection. Thus, to ensure the maximum excitation laser power, q should take the maximum value when n is constant.A simple rotation of the collimator 1 around the R-axis at the tangent of the reflective ring can induce a wide range of recirculation patterns. The radius of the laser spot r2 is ∼0.2 mm. The perimeter of the reflective ring is 33 mm. The path reflections ranging from 2 times to 81 times can be obtained without overlapping the spots on the reflective surface. However, the notch of the reflective ring can not reflect the excitation light. To prevent the laser from escaping from the notch, the spacing between each reflection point should not be less than ln / 2 + r2. The maximum times of reflection are 46 times without overlapping the spots on the reflective surface. Besides, the ceramic ferrule blocked part of the light path. The excitation laser should avoid contacting the ceramic ferrule. Hence, the smallest incident angle θs can be calculated by
In this system, the smallest θs is calculated as 9°. With the maximum times of reflection and the smallest θs, the incident angle θ is 9.6° when q equals 21.
The average optical power of multiple reflections can be expressed as:
where P is the power of the laser, R is the reflectivity of the inner surface of the reflective ring. Compared with the double-pass optical path, the absorption length in the multiple reflections will be significantly increased. The PA signal amplification factor of the multiple reflections relative to the double-pass optical path can be expressed as:In this system, R ∼87% [40], r0 = 5.25 mm, n = 47, d = 2 mm. With the incident angle of 9.6°, the PA signal amplification factor is calculated as ∼22.1. Different from tunable diode laser absorption spectroscopy (TDLAS), the light does not have to be focused onto the optical detector in PAS, which means the alignment of the optical path can allow a tiny offset. The reflected light can spread in the PA cell until the light power is attenuated to zero. Besides, the interference effect, which is often limiting the detection capability in absorption spectroscopy, is of little importance in PAS, allowing for longer optical path lengths compared to the absorption spectroscopy [34]. The light path is visualized by using a visible (650 nm) trace laser. Because the regular star polygon light path is symmetrically distributed, the incident angle can be effectively adjusted through 2-3 reflections by a low power fiber-optic visual fault locator (FO-VSL). Using the R-axis stage to adjust the collimator 1 to the specified angle and seal the notch with glue. Figure 1(d) is a photo of the PA gas sensor.
The multiple reflections-enhanced structures can increase the laser power and avoid the laser direct irradiate to the diaphragm [19,27], which may heat the diaphragm and change its linear interval or even burning through the diaphragm. Using a low-power laser can avoid this problem, but sacrificing laser power will also sacrifice the sensitivity, directly proportional to the optical power. The PPS maintains its thermal stability up to 450 °C and is rapidly pyrolyzed in the temperature range of 500–650 °C [36]. To prevent the laser from collimator 2 burning the diaphragm, using a different power laser to illuminate the diaphragm to verify the maximum laser power that the diaphragm can withstand. Clamping a piece of the ring with the same diaphragm as the PA gas sensor and adjusting the lens edges of a collimator to a distance of 2 mm from the diaphragm with a 3-axis stage. The collimator emitted different power laser with an erbium-doped fiber amplifier (EDFA) from 50 mW to 175 mW to the diaphragm separately. The laser of each power is irradiated on the diaphragm for 1 s and 60 s, respectively, as shown in Fig. 2. It can be seen from Fig. 2 that the laser burned through the diaphragm in a short time (< 1 s), and as the laser power increases, the part of the light spot that exceeds the heat-resistant temperature of the diaphragm increases. To protect the diaphragm from being burnt, the excitation laser from collimator 2 is limited to lower than 75 mW. It should be noted that although the laser below 75 mW will not burn through the diaphragm in a short time, it will still heat the diaphragm to change its linear interval and accelerate the aging rate in a long time.
Fig. 2. A photo of the diaphragm was irradiated by a collimated laser of different power from 50 mW to 175 mW with 1 s and 60 s.
2.2 Experimental setup
Figure 3 shows the experimental setup used to evaluate the performance of the PA gas sensor. A distributed feedback (DFB) excitation laser is modulated by a composite signal to produce 2f-WMS signal, which is summed by a sawtooth wave and a sine wave. The sawtooth wave and the sine wave are generated by a signal generator (Tektronix, AFG3022C) and a lock-in amplifier (SRS, SR830). The central wavelength of the DFB laser is chosen to be 1532.831 nm to avoid interference from water vapor and CO2 [41]. An EDFA (AEDFA-27-B-FA, Amonies) is used to amplify the power of the laser light. A 1×2 optical switch is used to compare the amplitude of the PA signal produced by multiple reflections and the double-pass of the optical path in the PA gas sensor. The PA gas sensor is placed in a gas chamber of ∼300 mL volume. C2H2/N2 gas mixtures of different concentrations are flowed into the gas chamber continuously, which were obtained by mixing a 500 ppm C2H2/N2 gas mixture with pure N2 at different flow rates controlled by two mass flow controllers (MFCs) (D07-19, SevenStar Electronics). The vibration of the membrane induced by the PA signal was detected by the demodulation module, which is consisted of a probe light (APEX, AP3350A), a photodetector (Newport, 2053-FC-M), and a circulator, coupled through the ceramic ferrule aligned with the membrane. The reflected probe light is received by the photodetector and demodulated by the lock-in amplifier at its second harmonic, and then transmitted to the computer.
3. Results and discussion
3.1. Comparison of multiple reflections and double-pass of the optical path
In order to evaluate the PA signal enhancement effect of the multiple reflections optical path, the collimated excitation laser was separately launched into the PA cell from the reflective ring and stainless steel gasket. A 500 ppm C2H2/N2 gas mixture was used as a sample to demonstrate the magnified multiple. The central wavelength of the DFB laser was fixed at 1532.831 nm, which is one of the central wavelengths of the Acetylene absorption line without interference from water vapor and CO2. The output laser power was set to 18 mW to ensure that the collimated laser would not burn the diaphragm. The sine wave modulation frequency was set to 300 Hz with the 2f-WMS signal of 600 Hz. The integration time of the lock-in amplifier was set to be 0.1 s. The gain factor of the photodetector was set to be 100. The frequency response of the photodetector was set from 100 Hz to 1000 Hz. Figure 4 shows the measured 2f-WMS signals with multiple reflections and a double-pass of the optical path. All the experiments were carried out at atmospheric pressure and room temperature. The peak values of the measured spectrums are 21.1 mV for multiple reflections optical path and 1.8 mV for a double-pass optical path, respectively. Therefore, the amplitude of the PA pressure signal using the multiple reflections optical path is 11.7 times that of the double-pass optical path, which is lower than the analysis result of ∼ 22.1. One reason is that the scattered light of the double-pass optical path might reflects more than once between the gold-coated diaphragm and the stainless steel gasket. Another reason may be that the reflectivity of the inner surface of the reflective ring is lower than 87%. As a result, the length of the actual light path decreases.
Fig. 4. 2f-WMS signals with multiple reflections and a double-pass optical path when the gas chamber was filled with 500 ppm C2H2/N2 gas mixture. Inset: a noise with multiple reflections and the double-pass optical path when the gas chamber was filled with pure N2.
The noise was measured with the gas chamber filled with pure N2 gas to compare the signal-to-noise ratio (SNR) of the multiple reflections and the double-pass optical path. In the whole measurement time of 200 s, the DFB laser was turned on and modulated with the central wavelength fixed to 1532.831 nm. The recorded values of the multiple reflections and the double-pass optical path are shown in the inset of Fig. 4. The deviations are calculated to be 5.19 μV (1σ) and 5.01 μV (1σ), respectively, which means that the interference from fundamental frequency signal generated by the wavelength-independent absorption of the reflective ring can be effectively eliminated.
3.2. Monitoring of gas micro-leakage
To verify the detection ability of the PA gas sensor with multiple reflections optical path, the responses of the PA gas sensor to C2H2 gas with different concentrations are presented. The gas mixture concentrations were adjusted using the mass flow controllers to pure N2, 50 ppm, 100 ppm, 200 ppm, 300 ppm, and 500 ppm, and the gas mixture flowed into the gas chamber successively. The central wavelength of the DFB laser was fixed at 1532.831 nm, and the output laser power was set to 18 mW. The modulation frequency of the sine wave was set to 300 Hz. The integration time of the lock-in amplifier was set to be 0.1 s. The peak value of the recorded spectrum as a function of concentration is plotted in Fig. 5. The recorded values of the 2f-WMS PA signal with different gas concentrations are shown in the inset of Fig. 5. The calculated R-square value, which represents the approximate degree of the regression line to the actual data point, is equal to 0.9996, which shows the linear responses of the PA gas sensor to C2H2 gas of less than 500 ppm.
Fig. 5. The peak value of the 2f-WMS spectrum as a function of concentration. Insets: 2f-WMS spectrum with different concentrations of C2H2/N2 gas mixture.
To verify the ability of continuous measurement of gas flow. Different concentrations of C2H2/N2 gas mixture were sequentially injected into the gas chamber. Figure 6 shows the peak value of the 2f-WMS spectrum collected every 1 s for 300 s with a flow rate of 50 sccm. The signal fluctuations in Fig. 6 are primarily due to air vibrations caused by airflow near the membrane and ambient sound noise coupled into the gas chamber, which can be weakened by adding a shell outside of the sensor head. Besides, the response time is measured to be about 30 s, which is longer than the result of the same size holes without a sponge filter [27]. This is because the gas takes time to through the sponge filter.
Fig. 6. The peak value of the 2f-WMS spectrum of continuous measurement results with changing concentrations of the C2H2/N2 gas mixture.
The PA sensor with multiple reflections optical path can benefit from high power excitation laser and without risk of burning through the diaphragm. The output laser power of the EDFA was set to 100 mW to verify the PA sensor detection sensitivity with high power excitation laser. The gas chamber was filled with a 500 ppm C2H2/N2 gas mixture. The integration time of the lock-in amplifier was set to be 0.1 s. The peak lock-in signal observed during the wavelength scan is 117.7 mV, as shown in Fig. 7. The noise with multiple reflections laser was measured with the gas chamber filled with pure N2 gas, as shown in the inset of Fig. 7. The responsivity is estimated to be 235.4 μV /ppm. The deviations are calculated as 5.56 μV (1σ), which can calculate that the PA sensor reaches an NEC of 23.6 ppb with a 100 mW excitation laser for C2H2 gas.
Fig. 7. 2f-WMS signals with multiple reflections when the excitation laser was set to 100 mW and the gas chamber was filled with 500 ppm C2H2/N2 gas mixture. Inset: a noise with multiple reflections when the gas chamber was filled with pure N2 gas.
4. Conclusion
In this paper, we report a multiple reflections-enhanced fiber-optic PA gas sensor for remote monitoring of gas micro-leakage. The volume of the PA cell is only 171 μL. A collimated laser with a central wavelength of 1532.831 nm is used as the excitation laser, which is an incident from the sidewall of the reflective ring. Multiple reflections occur on the inner surface of the reflective ring to enhance the excitation light power. The incident angle θ with respect to the radial direction is calculated as 9.6°. Besides, the excitation laser incidents from the sidewall can avoid direct irradiation on the diaphragm, which may burn through the diaphragm with high laser power. The PA signal is obtained by measuring the deflection of the membrane with an FPI. Experimental results show that the PA signal can be effectively enhanced 11.7 times by the multiple reflections optical path compare with the double-pass optical path. The 2f-WMS technology can essentially eliminate the fundamental frequency noise generated by the wavelength-independent absorption of the reflective ring. The designed PA gas sensor has an excellent linearity response to C2H2 concentrations of less than 500 ppm. The responsivity is estimated to be 235.4 μV /ppm, and the response time is detected to be ∼30 s. In addition, the minimum detection limit of the system is achieved to be 23.6 ppb by measuring the SNR of the system. The designed PA gas sensor has high sensitivity, small size, and intrinsic safety, which is suitable for remote detection and gas micro-leakage. The sensitivity of the PA gas sensor could be further enhanced by optimizing the PA cell dimensions and applying gold coating on the reflective ring inner wall.
Funding
National Natural Science Foundation of China (U1766217).
Disclosures
The authors declare that there are no conflicts of interest.
References
1. M. Lackner, “Tunable diode laser absorption spectroscopy (TDLAS) in the process industries–a review,” Rev. Chem. Eng. 23(2), 65–147 (2007). [CrossRef]
2. J. Hodgkinson and R. P. Tatam, “Optical gas sensing: a review,” Meas. Sci. Technol. 24(1), 012004 (2013). [CrossRef]
3. K. Liu, J. X. Mei, W. J. Zhang, W. D. Chen, and X. M. Gao, “Multi-resonator photoacoustic spectroscopy,” Sens. Actuators, B 251, 632–636 (2017). [CrossRef]
4. K. Chen, Z. F. Gong, and Q. X. Yu, “Fiber-amplifier-enhanced resonant photoacoustic sensor for sub-ppb level acetylene detection,” Sens. Actuators, A 274, 184–188 (2018). [CrossRef]
5. A. Hakonen, P. O. Andersson, M. S. Schmidt, T. Rindzevicius, and M. Käll, “Explosive and chemical threat detection by surface-enhanced Raman scattering: A review,” Anal. Chim. Acta 893, 1–13 (2015). [CrossRef]
6. J. Yinon, “Field detection and monitoring of explosives,” TrAC, Trends Anal. Chem. 21(4), 292–301 (2002). [CrossRef]
7. H. de Faria Jr, J. G. S. Costa, and J. L. M. Olivas, “A review of monitoring methods for predictive maintenance of electric power transformers based on dissolved gas analysis,” Renewable Sustainable Energy Rev. 46, 201–209 (2015). [CrossRef]
8. D.-E. A. Mansour, “Development of a new graphical technique for dissolved gas analysis in power transformers based on the five combustible gases,” IEEE Trans. Dielectr. Electr. Insul. 22(5), 2507–2512 (2015). [CrossRef]
9. T. H. Yang, W. G. Chen, and P. Y. Wang, “A review of all-optical photoacoustic spectroscopy as a gas sensing method,” Appl. Spectrosc. Rev. 20201–28 (2020). [CrossRef]
10. P. G. Westergaard and M. Lassen, “All-optical detection of acoustic pressure waves with applications in photoacoustic spectroscopy,” Appl. Opt. 55(29), 8266–8270 (2016). [CrossRef]
11. J. Suchanek, P. Janda, M. Dostal, A. Knizek, P. Kubat, P. Roupcova, P. Bitala, V. Nevrly, and Z. Zelinger, “Photoacoustic spectroscopy with mica and graphene micro-mechanical levers for multicomponent analysis of acetic acid, acetone and methanol mixture,” Microchem. J. 144, 203–208 (2019). [CrossRef]
12. S. Zhou, M. Slaman, G. Gruca, and D. Iannuzzi, “PAS-WRAP: a new approach to photoacoustic sensing, a new opportunity for the optical fiber sensor community,” in Seventh European Workshop on Optical Fibre Sensors (International Society for Optics and Photonics2019), p. 111992S.
13. J. Karhu, T. Tomberg, F. S. Vieira, G. Genoud, V. Hanninen, M. Vainio, M. Metsala, T. Hieta, S. Bell, and L. Halonen, “Broadband photoacoustic spectroscopy of (CH4)-C-14 with a high-power mid-infrared optical frequency comb,” Opt. Lett. 44(5), 1142–1145 (2019). [CrossRef]
14. T. Mikkonen, C. Amiot, A. Aalto, K. Patokoski, G. Genty, and J. Toivonen, “Broadband cantilever-enhanced photoacoustic spectroscopy in the mid-IR using a supercontinuum,” Opt. Lett. 43(20), 5094–5097 (2018). [CrossRef]
15. M. Köhring, A. Pohlkötter, U. Willer, M. Angelmahr, and W. Schade, “Tuning fork enhanced interferometric photoacoustic spectroscopy: A new method for trace gas analysis,” Appl. Phys. B: Lasers Opt. 102(1), 133–139 (2011). [CrossRef]
16. K. Chen, B. Zhang, M. Guo, Y. W. Chen, H. Deng, B. L. Yang, S. Liu, F. X. Ma, F. Zhu, Z. F. Gong, and Q. X. Yu, “Photoacoustic trace gas detection of ethylene in high-concentration methane background based on dual light sources and fiber-optic microphone,” Sens. Actuators, B 310, 127825 (2020). [CrossRef]
17. K. Chen, Z. H. Yu, Z. F. Gong, and Q. X. Yu, “Lock-in white-light-interferometry-based all-optical photoacoustic spectrometer,” Opt. Lett. 43(20), 5038–5041 (2018). [CrossRef]
18. Z. F. Gong, K. Chen, Y. W. Chen, L. Mei, and Q. X. Yu, “Integration of T-type half-open photoacoustic cell and fiber-optic acoustic sensor for trace gas detection,” Opt. Express 27(13), 18222–18231 (2019). [CrossRef]
19. Y. C. Cao, W. Jin, H. L. Ho, and J. Ma, “Miniature fiber-tip photoacoustic spectrometer for trace gas detection,” Opt. Lett. 38(4), 434–436 (2013). [CrossRef]
20. G. Gruca, K. Heeck, J. Rector, and D. Iannuzzi, “Demonstration of a miniature all-optical photoacoustic spectrometer based on ferrule-top technology,” Opt. Lett. 38(10), 1672–1674 (2013). [CrossRef]
21. K. Chen, M. Guo, S. Liu, B. Zhang, H. Deng, Y. H. Zheng, Y. W. Chen, C. Luo, L. Tao, M. Q. Lou, and Q. X. Yu, “Fiber-optic photoacoustic sensor for remote monitoring of gas micro-leakage,” Opt. Express 27(4), 4648–4659 (2019). [CrossRef]
22. T. Lauwers, A. Glière, and S. Basrour, “An all-optical photoacoustic sensor for the detection of trace gas,” Sensors 20(14), 3967 (2020). [CrossRef]
23. S. Zhou, M. Slaman, and D. Iannuzzi, “Demonstration of a highly sensitive photoacoustic spectrometer based on a miniaturized all-optical detecting sensor,” Opt. Express 25(15), 17541–17548 (2017). [CrossRef]
24. S. Zhou and D. Iannuzzi, “A fiber-tip photoacoustic sensor for in situ trace gas detection,” Rev. Sci. Instrum. 90(2), 023102 (2019). [CrossRef]
25. S. Zhou and D. Iannuzzi, “Immersion photoacoustic spectrometer (iPAS) for arcing fault detection in power transformers,” Opt. Lett. 44(15), 3741–3744 (2019). [CrossRef]
26. S. Zhou, “Fully coupled model for frequency response simulation of miniaturized cantilever-based photoacoustic gas sensors,” Sensors 19(21), 4772 (2019). [CrossRef]
27. Z. F. Gong, Y. W. Chen, T. L. Gao, K. Chen, Y. Jiao, M. Guo, B. Zhang, S. Liu, L. Mei, W. Peng, and Q. Yu, “Parylene-C diaphragm-based low-frequency photoacoustic sensor for space-limited trace gas detection,” Opt. Lasers Eng. 134, 106288 (2020). [CrossRef]
28. Y. F. Ma, S. D. Qiao, Y. He, Y. Li, Z. H. Zhang, X. Yu, and F. K. Tittel, “Highly sensitive acetylene detection based on multi-pass retro-reflection-cavity-enhanced photoacoustic spectroscopy and a fiber amplified diode laser,” Opt. Express 27(10), 14163–14172 (2019). [CrossRef]
29. S. D. Qiao, Y. F. Ma, Y. He, X. Yu, Z. H. Zhang, and F. K. Tittel, “A sensitive carbon monoxide sensor based on photoacoustic spectroscopy with a 2.3 µm mid-infrared high-power laser and enhanced gas absorption,” Sensors 19(14), 3202 (2019). [CrossRef]
30. K. Chen, B. Zhang, S. Liu, F. Jin, M. Guo, Y. W. Chen, and Q. X. Yu, “Highly sensitive photoacoustic gas sensor based on multiple reflections on the cell wall,” Sens. Actuators, A 290, 119–124 (2019). [CrossRef]
31. Y. F. Ma and H. T. Dang, “Corner cube prism-enhanced photoacoustic spectroscopy based gas sensing,” Infrared Phys. Technol. 109, 103386 (2020). [CrossRef]
32. B. Zhang, K. Chen, Y. W. Chen, B. L. Yang, M. Guo, H. Deng, F. X. Ma, F. Zhu, Z. F. Gong, W. Peng, and Q. X. Yu, “High-sensitivity photoacoustic gas detector by employing multi-pass cell and fiber-optic microphone,” Opt. Express 28(5), 6618–6630 (2020). [CrossRef]
33. M. Zhang, B. Zhang, K. Chen, M. Guo, S. Liu, Y. W. Chen, Z. F. Gong, Q. X. Yu, Z. Q. Chen, and M. F. Liao, “Miniaturized multi-pass cell based photoacoustic gas sensor for parts-per-billion level acetylene detection,” Sens. Actuators, A 308, 112013 (2020). [CrossRef]
34. A. Manninen, B. Tuzson, H. Looser, Y. Bonetti, and L. Emmenegger, “Versatile multipass cell for laser spectroscopic trace gas analysis,” Appl. Phys. B: Lasers Opt. 109(3), 461–466 (2012). [CrossRef]
35. B. Tuzson, M. Mangold, H. Looser, A. Manninen, and L. Emmenegger, “Compact multipass optical cell for laser spectroscopy,” Opt. Lett. 38(3), 257–259 (2013). [CrossRef]
36. T. J. Yoo, E.-B. Hwang, and Y. G. Jeong, “Thermal and electrical properties of poly (phenylene sulfide)/carbon nanotube nanocomposite films with a segregated structure,” Composites, Part A 91, 77–84 (2016). [CrossRef]
37. T. Kuusela and J. Kauppinen, “Photoacoustic gas analysis using interferometric cantilever microphone,” Appl. Spectrosc. Rev. 42(5), 443–474 (2007). [CrossRef]
38. D. Giovanni, Flat and corrugated diaphragm design handbook (Routledge, 2017).
39. H. S. M. Coxeter, Introduction to geometry (John Wiley & Sons, 1961).
40. K. Shanks, S. Senthilarasu, T. K. J. R. Mallick, and S. E. Reviews, “Optics for concentrating photovoltaics: Trends, limits and opportunities for materials and design,” Renewable Sustainable Energy Rev. 60, 394–407 (2016). [CrossRef]
41. K. Chen, S. Liu, B. Zhang, Z. F. Gong, Y. W. Chen, M. Zhang, H. Deng, M. Guo, F. X. Ma, F. Zhu, and Q. X. Yu, “Highly sensitive photoacoustic multi-gas analyzer combined with mid-infrared broadband source and near-infrared laser,” Opt. Lasers Eng. 124, 105844 (2020). [CrossRef]
