Open Access
Add to BibSonomy
Abstract
An innovative trace gas-sensing technique utilizing a single quartz crystal tuning fork (QCTF) based on a photoelectric detector and dual-frequency modulation technique was demonstrated for the first time for simultaneous multi-species detection. Instead of traditional semiconductor detectors and lock-in amplifier, we utilized the piezoelectric effect and resonant effect of the QCTF to measure the light intensity. A fast signal analysis method based on fast Fourier transform (FFT) algorithm is proposed for overlapping signal extraction. To explore the capabilities of this technique, a gas-sensing system based on two lasers having center emission wavelength of 1.653 µm (a DFB laser diode in the near-IR) and 7.66 µm (an EC QCL in the mid-IR) is successfully demonstrated for simultaneous CH4 spectroscopy measurements. The results indicate a normalized noise equivalent absorption (NNEA) coefficients of 1.33×10−9 cm−1W·Hz−1/2 at 1.653 µm and 2.20×10−10 cm−1W·Hz−1/2 at 7.66 µm, were achieved. This proposed sensor architecture has the advantages of easier optical alignment, lower cost, and a compactness compared to the design of a conventional TDLAS sensor using multiple semiconductor detectors for laser signal collection. The proposed technique can also be expanded to common QEPAS technique with multi-frequency modulation for multiple species detection simultaneously.
© 2020 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
The quartz crystal tuning fork (QCTF) [1] has been a versatile component for developing various transducers in many fields, such as mass sensors for biochemical and microbiological applications [2,3], stress sensors for measuring liquid density [4] and high-resolution atomic force microscopy [5], as well as pressure and temperature sensors in low-temperature physics [6–8]. In 2002, a standard 32-kHz tuning fork (QCTF) was first demonstrated as an acoustic wave transducer by Kosterev et al. [9], instead of a microphone used in traditional photoacoustic spectroscopy, i.e. quartz enhanced photoacoustic spectroscopy (QEPAS). Since then, many novel detection strategies have been successively proposed, such as an off-beam spectrophone with a convenient beam arrangement, a micro-resonator configuration and a multi-QCTF for signal enhancement, as well as a custom QCTF with larger prongs for diverging light sources (like LED and terahertz (THz)) [10–12]. On the other hand, QCTF was utilized as an optical detector for standoff laser spectroscopy in 2008 [13]. Unlike the standard QEPAS technique, this technique exploits QCTFs as photoelectric detectors to convert an optical signal into an electrical signal via quartz piezoelectric properties, and has been successfully demonstrated for direct absorption spectroscopy [14]. Moreover, the QCTF has been used as a point sensor in the open environment for detection of trace amounts of chemical explosive [15]. Most recently, the QCTF based detection technique was demonstrated with a superior sensing capability compared with a traditional TDLAS sensor and a conventional QEPAS sensor [16]. To date, QCTF based gas sensors have been successfully demonstrated for trace detection of many atmospheric species, such as NH3, NO, CO2, N2O, CO, H2S, CH2O, and chemical volatile organic compounds (VOCs) [11–18]. Multi-gas analysis plays an important role in many fields, such as including atmospheric environment monitoring, industrial process control, combustion diagnosis, clinical breath analysis [19–21]. In many practical applications, it is desirable to detect multiple gas species using a single compact system [22,23].
Mid-infrared (MIR) laser spectroscopy is a powerful tool for qualitative and quantitative gas sensing with high selectivity and sensitivity, due to the fundamental ro-vibrational absorption bands in this spectral region, which are significantly more intense than overtone transitions in the near infrared. With the rapid development of laser technology, quantum cascade laser (QCL) and interband cascade laser (ICL) have been proved to be versatile MIR laser sources for spectroscopic applications, and now are commercially widely available. However, they are hardly suitable for ordinary user due to the rather high costs, whereas the near-infared distributed feedback diode laser is significantly more cost-efficient. For MIR spectroscopy, an infrared mercury cadmium telluride (MCT) detector is commonly used for optical signal detection (typical VIGO Systems, Poland). However, the detectivity of this kind of semiconductor photodetector shows a significant inverse proportion to the wavelength response bandwidth. In case of NIR spectroscopy, an InGaAs photodetector (typical New Focus, USA) is usually used for signal detection. That means multiple different type of photodetectors have to use when laser sources with broad electromagnetic radiation span, which will make gas sensor system more cost and complex.
In this paper, an ultra compact sensor architecture for multi-gas species detection scheme is reported. A single QCTF with a high Q-factor based on a photoelectric detector and dual-frequency modulation technique is employed to collect laser spectra signals with spectral coverage from NIR to MIR. An adaptive signal processing algorithm based on a Fast Fourier transform (FFT) was developed for signal separation. Methane (CH4) was selected as the target species to demonstrate the reported sensor performance due to its important applications for atmospheric chemistry, natural gas leakage, combustion analysis and medical diagnosis [24,25].
2. Experimental setup
The experimental configuration of the multi-species sensor system is shown in Fig. 1. A near-infrared (NIR) continuous-wave (CW) distributed feedback (DFB) diode laser near 1.653 µm and a mid-infrared (MIR) external cavity quantum cascade laser (ECQCL) with emission wavelength between 6.96 and 8.85 µm (Block Engineering, USA) were employed to excite the rotational-vibrational and fundamental absorption of CH4 molecule, respectively. The CW-DFB diode laser yielded a spectral tuning range from 6045 cm−1 to 6058 cm−1, through which the R3 triplet of the 2v3 band of CH4 near 6046.95 cm−1 can be targeted. A custom flat mirror with a high antireflection film for MIR and a high high-reflecting film for NIR was specially used to couple both the NIR and MIR laser beams into a gas cell (with two CaF2 windows and a path length of 25 cm) for absorption spectral measurements. After exiting the gas sample cell, the laser beams were focused on the QCTF detector by a CaF2 lens. The optimal focusing point was experimentally determined near the base of the QCTF prongs, which is significantly dependent on the QCTF internal structure. The optical power of NIR and MIR lasers focused on the QCTF were about 1.8 mW and 1.5 mW, respectively. For a more detailed description of the QCTF based photoelectric detector and laser sources we refer the reader to [26]. The piezoelectric currents of QCTF detector are converted into a voltage, amplified by a low-noise preamplifier (SR560, Stanford Research Systems), and digitized by a data acquisition card (NI USB-6212) with 16-bits of resolution and finally transferred to a computer and a custom-written LabView-based program for real-time signal processing. Gas samples with different concentrations are prepared with a gas handling system, which mainly comprises some needle valves and three-way connectors, two pressure controllers (PC3-Series ALICAT) and an air pump. The gas pressure is monitored by using a digital gauge (Testo 552, Germany).
3. Results and discussion
In view of weak absorption features at 1.653 µm, wavelength modulation spectroscopy (WMS) with 2nd harmonic detection was employed for sensitive improvement of NIR CH4 detection, which was realized by adding a modulated sine wave to the DFB laser ramp with the modulation frequency of f/2, where f is the QCTF resonance frequency. The ECQCL is operated in a pulsed-mode and room temperature, calibration-free direct absorption spectroscopy (DAS) was selected for MIR CH4 detection, which is realized by setting the laser pulse repetition rate to the QCTF resonance frequency f. In theory, the piezoelectric current of the QCTF detector is well described by a sine function:
a multi-signal simultaneously excited by multi-laser at different wavelengths and different modulation frequencies, which propagate independently and the total signal can be expressed as: where ${f_i} \in \Delta f$ are all the noise frequencies falling into the acceptance band of the QCTF. Fast Fourier transform (FFT) is a powerful tool for frequency analysis. By using FFT algorithm, each frequency component can be easily determined. Generally, the modulation frequency should be selected as close as the central resonance frequency of the QCTF in order to achieve the maximal signal enhancement factor. From Eq. (2), it predicts that there is no cross-talk between each vibration mode of the QCTF theoretically, even multiple vibration modes are excited simultaneously.To achieve this detection scheme, the resonance characteristics of a QCTF was first investigated. A near-infrared diode laser model with fixed emitting wavelength at 1550 nm was used as the excitation light source for testing its resonance property. We performed this experiment by applying a sinusoidal modulation to the laser current with a constant modulation amplitude of 1 V, while scanning the modulation frequency with 0.5 Hz interval around the QCTF resonance frequency to observe its resonant profile. Generally, the resonant profile of the QCTF can be described by a Lorentzian function:
The Q value is defined as the ratio of the central resonance frequency ${f_0}$ to the square root of one half of the frequency bandwidth (full width at half maximum) $\Delta f$, which represents the amount of signal enhancement: The experimental data and theoretical fitting were shown in Fig. 2. Finally, a central resonance frequency of f0=32.765 kHz at ambient pressure and a quality factor Q of 9069 were obtained. In this work, dual-frequency modulation scheme was demonstrated by selecting two frequencies with a 1 Hz interval near the center frequency, mainly due to the resolution limit of the pulse repetition rate of the ECQCL. To illustrate the dual-frequency detection scheme, the pulse repetition rate of the ECQCL was set at f1=32.764 kHz and the DFB diode laser was modulated at f2/2 (f2=32.766 kHz for the second harmonic). In this case, the frequency shift effect on the loss of sensitivity is only 1.2 for f1 and 1.1 times for f2, respectively, which is acceptable for us. For example, Fig. 3(a) is the experimentally measured QCTF signal in the time domain solely excited by the MIR ECQCL laser (wavelength fixed at 1160 cm−1) at f1=32.764 kHz with the DFB diode laser beam blocked and the result of theoretical fitting. Similarly, Fig. 3(b) is the experimentally measured QCTF signal in time domain solely excited by the NIR DFB diode laser (wavelength fixed at 6046.95 cm−1) at f2=32.766 kHz with the MIR ECQCL laser beam blocked and the result of theoretical fitting. Figure 3(c) is the result simultaneously excited by both lasers, and the corresponding frequency spectrum is shown in Fig. 3(d). For clear comparison, all the fitted results were also presented. Obviously, the independent component cannot be extracted from the total signal in the time domain, but they can be easily distinguished in the frequency domain. Thus, a spectral analysis model based on a FFT algorithm was developed to extract the CH4 absorption spectra at different wavelengths.
Fig. 2. Experimentally measured resonant profile of the QCTF at ambient air and the fitted curve with a Lorentzian distribution, the inset shows the dual-frequency modulation scheme.
Fig. 3. The QCTF signals recorded in different conditions and the results of theoretical fitting. (a) and (b) are experimentally measured signal in time domain solely excited by the MIR ECQCL laser f1=32.764 kHz and the DFB diode laser at f2=32.766 kHz, respectively, (c) and (d) are the results simultaneously excited by both lasers, and the corresponding frequency spectrum, respectively. (Details see text)
The WMS-2f signal and the pulsed ECQCL output power show a strong dependence on the modulation amplitude and the pulse width, respectively. Therefore, the wavelength modulation depth of the NIR diode laser and the pulse width of the MIR ECQCL were optimized to further improve the WMS-2f signal amplitude and the DAS signal amplitude, respectively. Figure 4(a) shows the WMS-2f signal amplitudes as a function of laser modulation amplitude (in voltage) at 1 atm pressure for 1.08% CH4. Figure 4(b) presents the relationship between the ECQCL output power at 1305 cm−1 and laser pulse duration, and depicts a good linear dependence of the laser output power on the laser pulse duration in the range from 50 to 450 ns. Finally, a sinusoidal modulation amplitude of 40 mV for the NIR diode laser and a pulse width of 400 ns for the MIR ECQCL were used in this study.
Fig. 4. (a) The WMS-2f signal amplitudes as a function of the DFB diode laser modulation amplitude; (b)The relationship between the ECQCL laser power and its pulse duration.
To further verify the sensor system performance, the NIR and MIR CH4 absorption spectra were simultaneously recorded for different CH4:Air gas mixtures. In this experiment, we restricted both lasers at the approximate same wavelength scanning time interval, and a data sampling rate of 1 Hz was used without any signal averaging. In order to acquire the NIR CH4 absorption spectra, a dc voltage (between 200 and 575 mV) superimposed with a sine modulation signal (40 mV) within the laser threshold voltage range is injected into the DFB diode laser driver step by step. The ECQCL is a Littrow configuration with a grating as the tuning element. The angle of the grating selects the wavelength of the diffracted light, which couples back into the QCL chip and creates a laser at a single wavelength. The QCL chip is driven with pulsed current to realize wavelength tuning.
For example, the demodulated WMS-2f spectra from the NIR DFB diode laser at 1.653 µm and the corresponding 2f signal amplitude as a function of CH4 gas concentration are depicted in Figs. 5(a) and 5(b), respectively. The calculated linear regression coefficient R2 value was equal to 0.999, exhibiting an excellent linearity response to CH4 concentrations. The raw DAS spectra of CH4 and background signal extracted from the MIR ECQCL laser are shown in Fig. 6(a) (upper panel). According to Lambert-Beer’s Law, the normalized absorption coefficient is presented in Fig. 6(a) (middle panel). For comparison, the simulated spectrum between 1150 cm−1 and 1425 cm−1 based on the HITRAN database [27] is also presented in Fig. 6(a) (lower panel). As can be seen, a good agreement between experimental data and theoretical simulation was obtained, although the NIR CH4 spectrum was recorded simultaneously. To verify the linear concentration response, the absorption coefficients at the strongest absorption line near 1305 cm−1 are plotted as a function of CH4 concentrations in Fig. 6(b). Linear regression yields to a regression coefficient R2 of 0.994. It is worth noting that the results have been compared with the CH4 spectra recorded only using individual laser sources, and no any obvious cross interference was found, except comparison to the theoretical simulation.
Fig. 5. Results extracted from the NIR diode laser. (a) WMS-2f absorption spectra under different CH4 mixing ratio, (b) WMS-2f signal amplitude at as a function of CH4 gas concentration.
Fig. 6. Results extracted from the MIR ECQCL laser. (a) DAS spectra under different CH4 mixing ratio before and after normalizing process (upper panel and middle panel), and HITRAN simulation (lower panel), (b) Absorption coefficient (at 1305 cm−1) as a function of CH4 gas concentration.
Finally, the sensor stability and precision were evaluated by continuous measurement with a constant gas sample. In this experiment, both lasers were operated with fixed wavelength mode, the NIR diode laser at 6046.95 cm−1 and the MIR ECQCL laser at 1305 cm−1, respectively, and the gas sample cell was operated in a static mode with a constant CH4 concentration. The measured time series data in units of % are shown in Fig. 7 (upper panel), which are calculated from the raw data in units of V according to the calibration curve. The calculated standard deviation (sd) for the NIR and MIR time series data are 0.005% (1σ) and 0.003% (1σ), respectively. For further analysis, the Allan-Werle variance analysis method was applied [28], as shown in Fig. 7 (bottom panel). The Allan deviation plot indicates that a measurement precision of 38 ppm for the NIR diode laser and 25 ppm for the MIR ECQCL laser were achieved with a 1-s integration time, respectively, which can be further improved to 5 ppm and 3.5 ppm by increasing the averaging time up to 1000 s and 400 s, respectively. Such a long optimal integration time indicates that the sensor system is mainly affected by the white noise. Increase of the averaging time decreased the Allan variance as predicted, following the theoretical expectation. After the optimal integration time, other factors influence the system noise. Moreover, this analysis shows that, to minimize measurement error in practical applications, calibrations should be performed within the optimal integration time to realize high-precision measurement. As the piezoelectric current of the QCTF generated by light radiation excitation is very weak, therefore, a low noise, high precision preamplifier used for the piezoelectric current amplification and conversion is very crucial for obtaining of high SNR and high sensitivity. Currently, the QCTF detector is mainly affected by the home-made preamplifier. We would like to improve the preamplifier used for current amplification and voltage conversion in our next work. Then the noise level will be reduced, which in turn decreases the integration time.
Fig. 7. Allan-Werle plot analysis for WMS-2f detection with the NIR diode laser and DAS detection with the MIR ECQCL laser.
Although a comparison of detection limits with other sensor systems developed for CH4 or other trace gases detection is not straightforward, in view of rather different experimental configurations and definitions of sensitivity and sometimes lack of specific information. However, a summary is made here with some recently reported systems employing QCTF as a detector. The calculated normalized noise equivalent absorption (NNEA) coefficient were 1.33×10−9 cm−1W·Hz−1/2 and 2.20×10−10 cm−1W·Hz−1/2, respectively, which indicates that the proposed sensing technique is superior to the common multiple QCTFs based QEPAS technique [29,30] and the recently reported result based on lock-in a amplifier technique [31], as listed in Table 1. Compared to Ref. [16], the degraded performance of our sensor system is mainly due to the ultra-high Q-factor of 50177 achieved in their work, which was realized by sealing the quartz tuning fork in a low pressure chamber. As demonstrated in our previous work [26], the QCTF detector shows a good linear response on the incident laser power. Thus, the detection sensitivity of the sensor system can be further improved by using laser sources with a higher emitting power, e.g. by combining with an erbium-doped fiber amplifier [32,33]. Improvements in sensitivity and detectivity can also be obtained by fabricating the QCTF detector in a vacuum cavity to enhance the Q factor.
Table 1. Comparison of sensitivity achieved with previous publications.
4. Conclusion
In conclusion, we reported a compact sensor architecture for multi-gas species sensing technique based on a single QCTF and dual-frequency modulation scheme, which is successfully demonstrated as a photoelectric detector for simultaneously measuring CH4 spectroscopy in the NIR and MIR regions. Instead of a traditional lock-in amplifier technique, a fast signal analysis method based on a FFT algorithm is proposed for overlapping signal extraction. The measured results indicate a normalized noise equivalent absorption (NNEA) coefficients of 1.33×10−9 cm−1W·Hz−1/2 at 1.653 µm and 2.20×10−10 cm−1W·Hz−1/2 at 7.66 µm, respectively, which indicates that the proposed sensing technique is superior to the common multiple QCTFs based QEPAS technique. Most important of all, the sensor architecture has the advantages of easier optical alignment, lower cost, and more compactness compared with a conventional TDLAS sensor using multiple detectors [34] and multi-quartz-enhanced photoacoustic spectroscopy. Note that this technique can also be extended to other wavelength ranges, and other spectroscopic techniques, such as the QEPAS, with a QCTF as an acoustic transducer, with multi-frequency modulation for multiple species detection simultaneously.
Funding
The National Program on Key Research and Development Project of China (2016YFC0302202); National Natural Science Foundation of China (41875158, 61675005, 61705002).
Disclosures
The authors declare no conflicts of interest.
References
1. J. M. Friedt and E. Carry, “Introduction to the quartz tuning fork,” Am. J. Phys. 75(5), 415–422 (2007). [CrossRef]
2. K. Waszczuk, G. Gula, M. Swiatkowski, J. Olszewski, W. Herwich, Z. Drulis-Kawa, J. Gutowicz, and T. GotszalkaL, “Evaluation of Pseudomonas aeruginosa biofilm formation using piezoelectric tuning fork mass sensors,” Sens. Actuators, B 170, 7–12 (2012). [CrossRef]
3. X. Su, C. Dai, J. Zhang, and S. J. O’ Shea, “Quartz tuning fork biosensor,” Biosens. Bioelectron. 17(1-2), 111–117 (2002). [CrossRef]
4. J. Zhang, C. Dai, X. Su, and S. J. O’ Shea, “Determination of liquid density with a low frequency mechanical sensor based on quartz tuning fork,” Sens. Actuators, B 84(2-3), 123–128 (2002). [CrossRef]
5. D. Hussain, J. Song, H. Zhang, X. Meng, Y. Wen, and H. Xie, “Optimizing the Quality Factor of Quartz Tuning Fork Force Sensor for Atomic Force Microscopy: Impact of Additional Mass and Mass Rebalance,” IEEE Sens. J. 17(9), 2797–2806 (2017). [CrossRef]
6. M. Človečko, M. Grajcar, M. Kupka, P. Neilinger, M. Rehák, P. Skyba, and F. Vavrek, “High Q value Quartz Tuning Fork in Vacuum as a Potential Thermometer in Millikelvin Temperature Range,” J. Low Temp. Phys. 187(5-6), 573–579 (2017). [CrossRef]
7. J. Söderkvist, “Micromachined gyroscopes,” Sens. Actuators, A 43(1-3), 65–71 (1994). [CrossRef]
8. J. A. Hedberg, A. Lal, Y. Miyahara, P. Grütter, G. Gervais, M. Hilke, L. Pfeiffer, and K. W. West, “Low temperature electrostatic force microscopy of a deep two-dimensional electron gas using a quartz tuning fork,” Appl. Phys. Lett. 97(14), 143107 (2010). [CrossRef]
9. A. A. Kosterev, Y. A. Bakhirkin, R. F. Curl, and F. K. Tittel, “Quartz-enhanced photoacoustic spectroscopy,” Opt. Lett. 27(21), 1902 (2002). [CrossRef]
10. K. Liu, X. Guo, H. Yi, W. Chen, W. Zhang, and X. Gao, “Off-beam quartz-enhanced photoacoustic spectroscopy,” Opt. Lett. 34(10), 1594–1596 (2009). [CrossRef]
11. H. Zheng, L. Dong, H. Wu, X. Yin, L. Xiao, S. Jia, R. F. Curl, and F. K. Tittel, “Application of acoustic micro-resonators in quartz-enhanced photoacoustic spectroscopy for trace gas analysis,” Chem. Phys. Lett. 691, 462–472 (2018). [CrossRef]
12. P. Patimisco, A. Sampaolo, L. Dong, F. K. Tittel, and V. Spagnolo, “Recent advances in quartz enhanced photoacoustic sensing,” Appl. Phys. Rev. 5(1), 011106 (2018). [CrossRef]
13. C. W. Van Neste, L. R. Senesac, and T. Thundat, “Standoff photoacoustic spectroscopy,” Appl. Phys. Lett. 92(23), 234102 (2008). [CrossRef]
14. J. Sun, H. Deng, N. Liu, H. Wang, B. Yu, and J. S. Li, “Mid-infrared gas absorption sensor based on a broadband external cavity quantum cascade laser,” Rev. Sci. Instrum. 87(12), 123101 (2016). [CrossRef]
15. C. W. Van Nestea, M. E. Morales-Rodríguez, L. R. Senesac, S. M. Mahajan, and T. Thundat, “Quartz crystal tuning fork photoacoustic point sensing,” Sens. Actuators, B 150(1), 402–405 (2010). [CrossRef]
16. Y. He, Y. Ma, Y. Tong, X. Yu, and F. K. Tittel, “Ultra-high sensitive light-induced thermoelastic spectroscopy sensor with a high Q-factor quartz tuning fork and a multipass cell,” Opt. Lett. 44(8), 1904–1907 (2019). [CrossRef]
17. S. Borri, P. Patimisco, I. Galli, D. Mazzotti, G. Giusfredi, N. Akikusa, M. Yamanishi, G. Scamarcio, P. De Natale, and V. Spagnolo, “Intracavity quartz-enhanced photoacoustic sensor,” Appl. Phys. Lett. 104(9), 091114 (2014). [CrossRef]
18. M. Giglio, P. Patimisco, A. Sampaolo, A. Zifarelli, R. Blanchard, C. Pfluegl, M. F. Witinski, D. Vakhshoori, F. K. Tittel, and V. Spagnolo, “Nitrous oxide quartz-enhanced photoacoustic detection employing a broadband distributed-feedback quantum cascade laser array,” Appl. Phys. Lett. 113(17), 171101 (2018). [CrossRef]
19. R. F. Curl, F. Capasso, C. Gmachl, A. A. Kosterev, B. McManus, R. Lewicki, M. Pusharsky, G. Wysocki, and F. K. Tittel, “Quantum cascade lasers in chemical physics,” Chem. Phys. Lett. 487(1-3), 1–18 (2010). [CrossRef]
20. J. S. Li, W. Chen, and H. Fischer, “Quantum cascade laser spectrometry techniques: a new trend in atmospheric chemistry,” Appl. Spectrosc. Rev. 48(7), 523–559 (2013). [CrossRef]
21. A. Schwaighofer, M. Brandstetter, and B. Lendl, “Quantum cascade lasers (QCLs) inbiomedical spectroscopy,” Chem. Soc. Rev. 46(19), 5903–5924 (2017). [CrossRef]
22. W. Ren, W. Jiang, and F. K. Tittel, “Single-QCL-based absorption sensor for simultaneous trace-gas detection of CH4 and N2O,” Appl. Phys. B 117(1), 245–251 (2014). [CrossRef]
23. J. S. Li, H. Deng, J. Sun, B. Yu, and H. Fischer, “Simultaneous atmospheric CO, N2O and H2O detection using a single quantum cascade laser sensor based on dual-spectroscopy techniques,” Sens. Actuators, B 231, 723–732 (2016). [CrossRef]
24. C. Liu and L. Xu, “Laser absorption spectroscopy for combustiondiagnosis in reactive flows: A review,” Appl. Spectrosc. Rev. 54(1), 1–44 (2019). [CrossRef]
25. K. Liu, L. Wang, T. Tan, G. Wang, W. Zhang, W. Chen, and X. Gao, “Highly sensitive detection of methane by near-infrared laser absorption spectroscopy using a compact dense-pattern multipass cell,” Sens. Actuators, B 220, 1000–1005 (2015). [CrossRef]
26. J. Li, N. Liu, J. Ding, S. Zhou, T. He, and L. Zhang, “Piezoelectric effect-based detector for spectroscopic application,” Opt. Laser. Eng. 115, 141–148 (2019). [CrossRef]
27. I. E. Gordon, L. S. Rothman, C. Hill, R. V. Kochanov, Y. Tan, P. F. Bernath, M. Birk, V. Boudon, A. Campargue, K. V. Chance, B. J. Drouin, J. M. Flaud, R. R. Gamache, J. T. Hodges, D. Jacquemart, V. I. Perevalov, A. Perrin, K. P. Shine, M. A. H. Smith, J. Tennyson, G. C. Toon, H. Tran, V. G. Tyuterev, A. Barbe, A. G. Csaszar, V. M. Devi, T. Furtenbacher, J. J. Harrison, J. M. Hartmann, A. Jolly, T. J. Johnson, T. Karman, I. Kleiner, A. A. Kyuberis, J. Loos, O. M. Lyulin, S. T. Massie, S. N. Mikhailenko, N. Moazzen-Ahmadi, H. S. P. Mueller, O. V. Naumenko, A. V. Nikitin, O. L. Polyansky, M. Rey, M. Rotger, S. W. Sharpe, K. Sung, E. Starikova, S. A. Tashkun, J. Vander Auwera, G. Wagner, J. Wilzewski, P. Wcislo, S. Yu, and E. J. Zak, “The HITRAN2016 molecular spectroscopic database,” J. Quant. Spectrosc. Radiat. Transfer 203, 3–69 (2017). [CrossRef]
28. P. Werle, R. Miicke, and F. Slemr, “The limits of signal averaging in atmospheric trace-gas monitoring by tunable diode-laser absorption spectroscopy (TDLAS),” Appl. Phys. B 57(2), 131–139 (1993). [CrossRef]
29. Q. Zhang, J. Chang, Z. Cong, J. Sun, and Z. Wang, “QEPAS Sensor for Simultaneous Measurements of H2O, CH4, and C2H2 Using Different QTFs,” IEEE Photonics J. 10(6), 1–8 (2018). [CrossRef]
30. Y. Ma, X. Yu, G. Yu, X. Li, J. Zhang, D. Chen, R. Sun, and F. K. Tittel, “Multi-quartz-enhanced photoacoustic spectroscopy,” Appl. Phys. Lett. 107(2), 021106 (2015). [CrossRef]
31. Y. Ma, Y. He, Y. Tong, X. Yu, and F. K. Tittel, “Quartz-tuning-fork enhanced photothermal spectroscopy for ultra-high sensitive trace gas detection,” Opt. Express 26(24), 32103–32110 (2018). [CrossRef]
32. Y. Ma, Y. He, L. Zhang, X. Yu, J. Zhang, R. Sun, and F. K. Tittel, “Ultra-high sensitive acetylene detection using quartz-enhanced photoacoustic spectroscopy with a fiber amplified diode laser and a 30.72 kHz quartz tuning fork,” Appl. Phys. Lett. 110(3), 031107 (2017). [CrossRef]
33. H. Wu, L. Dong, H. Zheng, X. Liu, X. Yin, W. Ma, L. Zhang, W. Yin, S. Jia, and F. K. Tittel, “Enhanced near-infrared QEPAS sensor for sub-ppm level H2S detection by means of a filev amplified 1582 nm DFB laser,” Sens. Actuators, B 221, 666–672 (2015). [CrossRef]
34. Y. Yu, N. P. Sanchez, F. Yi, C. Zheng, W. Ye, H. Wu, R. J. Griffn, and F. K. Tittel, “Dual quantum cascade laser based sensor for simultaneous NO and NO2 detection using a wavelength modulation division multiplexing technique,” Appl. Phys. B 123(5), 164 (2017). [CrossRef]
