Open Access
Add to BibSonomy
Abstract
We report a mid-infrared cavity-enhanced absorption sensor for atmospheric nitrous oxide (N2O) detection using a continuous-wave distributed-feedback quantum cascade laser (DFB-QCL) at 4.5 µm. The QCL beam is coupled to a short (78 mm) Fabry-Pérot (F-P) optical cavity, which consists of two plano-concave dielectric mirrors with a reflectivity of 99.84%. The Pound-Drever-Hall technique is used to lock the QCL to the optical cavity by directly modulating the injection current of the QCL at 4 MHz. Our mid-infrared gas sensor achieves a minimum detection limit of 0.32 ppb at 50 s integration time. We demonstrate a proof-of-concept absorption spectral measurement of ambient air. Our study provides a promising way of developing compact and sensitive gas sensors for environmental monitoring.
© 2021 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
Nitrous oxide (N2O), also known as laughing gas, is one of the top three greenhouse gases (GHGs) that accounts for both climate change and ozone depletion [1–3]. The 100-year global warming potential (GWP100) of N2O is over 200 times higher than that of carbon dioxide [4]. N2O is currently the biggest culprit responsible for the destruction of the stratospheric ozone layer [3]. Anthropogenic N2O, including that from agricultural fertilizer use, combustion exhaust and industrial processes, accounts for nearly 6% of the total anthropogenic GHG emissions. In general, the globally averaged N2O concentration in the atmosphere has kept increasing to the unprecedented level (332 ppb) than ever before [5], and it is unlikely to decrease due to the stability and long lifetime of N2O. The spatial distribution of N2O depends largely on the local climate and human activities [6,7]. Therefore, it is essential to monitor the concentration of N2O in ambient air so that we can identify N2O sources and sinks and better understand its impact on climate change [8].
In the last decade, laser-based absorption spectroscopic techniques with quantum cascade lasers (QCLs) have been extensively explored for sensitive detection of N2O [9–16]. By exploiting the strong fundamental absorption bands near 4.5 µm and 7.7 µm, sensitive N2O measurements at ppb-level were reported using mid-infrared (MIR) absorption spectroscopy combined with a multipass gas cell (MGC). For instance, a QCL absorption sensor was developed for atmospheric N2O and methane (CH4) detection using a MGC with a path length of 56 m and a volume of 0.5 L [9]. The sensor utilized two pulsed QCLs to achieve a measurement precision of 0.3 ppb for N2O at 4.5 µm and 4 ppb for CH4 at 7.9 µm at a gas pressure of 60 torr. Tao et al. [10] developed a field-deployable QCL-based N2O sensor at 4.5 µm using a MGC with an optical path length of 16 m. Multi-harmonic wavelength modulation spectroscopy (WMS) was used to accomplish a precision of 0.15 ppb N2O with 10 Hz sample rate. Simultaneous detection of N2O and CH4 was presented using a single QCL near 7.8 µm [11]. A compact MGC with a sampling volume of 225 mL was employed to realize an optical path length of 57.6 m. A minimal detection limit (MDL) of 2.6 ppb N2O was reported at 1 s averaging time.
The use of an optical cavity provides an alternative method for high-precision gas detection. The equivalent absorption path length of an optical cavity can reach thousands of times of the physical length, which mainly depends on the reflectivity of the cavity mirrors. The QCL-based cavity ring-down spectroscopy (CRDS) has been explored for ppb-level atmospheric N2O detection [12–14]. For instance, an external-cavity QCL (EC-QCL)-based CRDS sensor at 5.2 µm was reported to measure ambient N2O [12]. By adopting a cavity with a reflectivity of > 99.98% and length of 50 cm, the sensor achieved a MDL of 4.5 ppb at the atmospheric pressure. A similar work was reported by Tang et al. [13] to develop a MIR CRDS sensor with a MDL of 11 ppt N2O at 10.2 s averaging time. Later, Long et al. [14] proposed a CRDS spectrometer operating near 4.5 µm and achieved the quantum-noise-limited short-time performance. The length of the optical cavity was 1.5 m and the ring-down mirrors were designed with a power reflectivity of 99.99%. It should be noted that CRDS requires the use of high-speed electronics to measure the extremely short ring-down time. In addition to CRDS, Wojtas et al. [15] reported the development of off-axis cavity-enhanced absorption spectroscopy (OA-CEAS) for N2O detection. The pulsed-QCL-based N2O sensor implemented a 60 cm long cavity (reflectivity 99.98% at 4.53 µm) and achieved ppb-level sensitivity. Nadeem et al. [16] reported a MIR N2O sensor using off-axis integrated cavity output spectroscopy (OA-ICOS) at 7.7 µm, and achieved a MDL of 70 ppb in less than 10 s averaging time. Similar to MGCs, the MDL of these off-axis configurations may be limited by the partial overlap of the beam spots on the cavity mirrors [16]. A detailed comparison of the detection sensitivity for these absorption-based N2O sensors is summarized in Table 1 of Section 3.2, along with our results achieved in this work.
The use of cavity locking techniques in CEAS makes it promising for high-precision spectroscopic measurements. More importantly, the implementation of Pound-Drever-Hall (PDH) frequency locking strategy in cavity-enhanced experiments has been reported in previous studies [17–19]. The phase/frequency modulation is required in the PDH technique for optical heterodyne detection to generate the error signal [20,21]. Most of the previous studies used the near-infrared (NIR) lasers for demonstrations where phase modulators such as electro-optic modulators (EOMs) are commercially available. Dong et al. [17] developed a cavity-enhanced amplitude-modulated laser absorption spectrometer for the real-time detection of carbon dioxide (CO2) at 1.57 µm. The laser was PDH-locked to the scanning cavity to map out the CO2 absorption profile. He et al. [18] reported a repetitively mode-locked CEAS technique for NIR gas detection, where the laser was repetitively locked to the cavity during wavelength scanning. However, the application of the PDH-locked CEAS method in the MIR region is quite challenging due to the lack of appropriate phase modulators.
In this work, we report a MIR CEAS-based gas sensor for sensitive N2O detection. A continuous-wave (cw) distributed-feedback QCL (DFB-QCL) near 4.5 µm is used as the light source to target the N2O absorption line at 2190.35 cm−1. An optical cavity of 78 mm is employed to improve the detection sensitivity by achieving an effective absorption path length of 96 m. Rather than using a MIR phase modulator, we successfully lock the QCL to the optical cavity by directly modulating the laser injection current. By providing technical details of the sensor design, calibration and evaluation, this work provides a promising method of developing MIR optical gas sensors for environmental monitoring.
2. Sensor configuration
2.1 Laser characterization and line selection
The strongest fundamental vibrational band of N2O exists at 4.5 µm, which can be accessed by commercial QCLs. The cw DFB-QCL (Hamamatsu Photonics) used in this work can be tuned from 2187.76 to 2192.64 cm−1 by adjusting its operating temperature and injection current, as presented in Fig. 1(a). Based on the HITRAN database [22], Fig. 1(b) shows the absorption coefficient of the target N2O absorption line at 2190.35 cm−1 (4.565 µm) at 7 kPa and 296 K. The absorption profile of N2O can be well covered by a single current scan of the QCL. Besides N2O (300 ppb), some representative gas species (200 ppb CO, 400 ppm CO2 and 2.5% H2O) in ambient air are also added in the spectral plot shown in Fig. 1(b). A lower pressure of 7 kPa is chosen to isolate N2O absorption from absorption interference caused by ambient CO and H2O.
Fig. 1. (a) Wavelength tuning characteristics of the QCL provided by Hamamatsu Photonics. (b) Absorption spectra of 300 ppb N2O at 4.565 µm, together with 200 ppb CO, 400 ppm CO2 and 2.5% H2O at 296 K and 7 kPa.
2.2 Experimental setup
Figure 2 depicts the schematic of the MIR-CEAS sensor. The operating temperature and injection current of the QCL were precisely controlled by a low-noise laser driver (QubeCL system, ppqSense). The Faraday isolator (Electro-Optics Technology) used in the optical setup provides an isolation of around 30 dB for the reflected light. A polarizing beam splitter (PBS) combined with a quarter-wave plate (QWP) was used to collect the reflected light from the optical cavity by a high-bandwidth photodetector (PVI-4TE-6, VIGO Systems), which was used for wavelength locking.
Fig. 2. Schematic of the MIR cavity-enhanced absorption sensor for N2O detection. QCL, quantum cascade laser; ISO, isolator; PBS, polarizing beam splitter; QWP, quarter-wave plate; L1, L2, mode-matching lenses; M1, M2, high-reflectivity cavity mirrors; PZT, piezoelectric transducer; PD1, PD2, photodetectors; LO, local oscillator; DAQ, data acquisition card.
A compact Fabry-Pérot (F-P) optical cavity was used as the gas cell. The F-P cavity consists of two identical plano-concave dielectric mirrors separated by a distance of 78 mm. Such a short cavity enables the compact sensor configuration and reduces the gas sample volume. Each mirror has a radius of curvature of 125 mm and a reflectivity of 99.84%, corresponding to a finesse of 1931 and an effective absorption path length of 96 m. A piezo-electric transducer (PZT) was attached to the rear cavity mirror so that the cavity length can be adjusted. A pair of plano-convex lenses (focal length f1 = 50 mm, f2 = 75 mm) were placed in front of the F-P cavity for mode matching purposes. The transmission beam from the cavity was then focused by a concave mirror onto a 1-MHz bandwidth photodetector (PVI-2TE-6, VIGO Systems) for absorption measurements.
The PDH locking method was used in this study for MIR-CEAS. The schematic is also illustrated in Fig. 2. In conventional PDH techniques, an EOM is normally employed to generate heterodyne beat notes between the carrier and the sidebands for correcting the frequency shift [20,21]. As the MIR EOM is not always commercially available, here we propose a new method of generating optical sidebands for PDH locking by directly modulating the injection current of the QCL. The QCL current was modulated by a 4-MHz sinusoidal waveform from a function generator (Keysight Technologies). The reflected light from the F-P cavity was collected by the photodetector PD1 and then separated by a bias-tee circuit to obtain the direct current (DC) and radio-frequency (RF) components, respectively. The DC component was used to monitor the coupling efficiency of the QCL to the cavity. The RF component was mixed with the 4-MHz local oscillator from the function generator to generate the PDH error signal. The error signal was sent to the QCL driver via a servo (LB1005 Servo, New Focus). When the QCL is locked to the cavity mode, the laser frequency can be tuned by simply adjusting the cavity length via the PZT attached to the cavity mirror shown in Fig. 2.
Gas mixtures with different concentrations were prepared by diluting the certified 1 ppm (2% uncertainty) N2O/N2 mixture using two mass flow controllers. The mass flow controller has an accuracy of ±1.0% of the set point when the flow rate is equal or greater than 35% of the full scale, and ±0.35% of the full scale when the flow rate is less than 35% of the full scale. The two gas flows with designed flow rates were fed into a mixing tank and kept for about half an hour for thorough mixing. Then the gas mixture was introduced to the gas cell for absorption measurements.
3. Results and discussion
3.1 Test of locking performance
We first evaluated the locking performance between the injection-current-modulated QCL and the F-P cavity. Although a higher modulation frequency is preferred, here we adopted a modulation frequency of 4 MHz which was mainly limited by the bandwidth of the laser driver. Such a MHz-rate current modulation with a peak-to-peak amplitude of less than 0.9 mA was used to generate the first-order sidebands. Besides the 4-MHz sinusoidal waveform, the QCL injection current was scanned by a slowly varying triangle waveform (20 Hz) as well. Figure 3 depicts the representative PDH error signal, the reflected light detected by PD1, and the cavity transmission detected by PD2. The transmitted signal shown in Fig. 3(a) does not have a Gaussian profile, which is possibly due to the slight optical feedback existing in the system. At the moment of cavity resonance, the transmission signal appears with a peak voltage of 7.01 V, corresponding to the reduction of the reflected signal from 3.16 V to 0.66 V. Hence, we estimated a coupling efficiency of 79% when the QCL is in resonance with the cavity mode. Such a high coupling efficiency also indicates a relatively narrow laser linewidth with the PDH technique. The generated error signal is shown in Fig. 3(c), which is used as a feedback to lock the laser frequency to the cavity mode with a proportional-integral-derivative (PID) control servo loop.
Fig. 3. (a) Cavity transmission, (b) reflection and (c) PDH error signals measured during the laser current scan.
Figure 4 shows the corresponding cavity transmission, reflected light and PDH error signals after the PDH locking. The standard deviation (1σ) of the transmission signal was 0.48% of peak transmission for a measurement time of 200 ms. Such a small noise could come from the residual amplitude modulation (RAM) effect in PDH locking [23], which can be mitigated by reducing the mechanical vibration and the environmental temperature variation. Note that the F-P cavity was placed in the gas cell without using any temperature controller and vibration absorber.
3.2 Sensor performance
The performance of the MIR-CEAS gas sensor was then evaluated by measuring N2O/N2 mixtures at different concentrations. Gas samples were prepared by diluting the certified 1 ppm N2O/N2 mixture with pure N2. The transmission signal measured at each optical frequency was recorded for 1 s and averaged for spectral analysis. Figure 5 illustrates the representative absorption spectra of N2O measured at different N2O concentrations from 0.09 ppm to 1.0 ppm. The spectral fitting with Lorentzian profile is also plotted in Fig. 5, where the baseline was extracted from the non-absorption far wings of the transmission signal. Note that all the measurements were conducted at room temperature and pressure of 7 kPa.
Fig. 5. Representative cavity-enhanced absorption spectra of N2O measured at different concentrations. The measurement is fitted by the Lorentzian profile.
Figure 6 plots the measured peak absorbance of the MIR-CEAS signal varied with the N2O concentration. Over the tested concentration range, a linear fit (slope 0.269 ppm−1) is obtained with an R-square value of 0.999, indicating a good linear response of the current sensor. An Allan-Werle deviation analysis was conducted to evaluate the detection sensitivity and long-time stability, as shown in Fig. 7. By continuously monitoring pure N2 at the line-center with a sampling rate of 10 Hz, a MDL of 0.32 ppb was achieved at 50 s integration time, corresponding to a minimum detectable absorption coefficient (MDA) of 8.8×10−9 cm−1. As indicated in Table 1, our sensor shows a comparative performance among the state-of-the-art absorption-based gas sensing techniques. We believe that the MDL can be further improved by implementing an additional feedback control of the cavity length to compensate any slow drift with time [19].
Table 1. Comparison of typical N2O absorption sensors with quantum cascade lasers.a
Fig. 6. Variation of peak absorbance with N2O concentration. The uncertainties of concentrations are mainly contributed by the accuracy of mass flow controllers and the certified N2O/N2 mixture in the gas cylinder.
Fig. 7. (a) Continuous measurement of pure N2 at the absorption line-center. (b) Allan-Werle deviation analysis (in ppb) of the MIR-CEAS N2O sensor.
Finally, we demonstrated the atmospheric N2O measurement using the developed MIR-CEAS N2O sensor. The air sample was collected at the street using an aluminum foil gas collecting bag. Figure 8 shows the measured absorption spectrum of an air sample, showing a good agreement with the calculated absorption spectrum of 300 ppb N2O based on the HITRAN database [22]. The gas concentration can also be derived from the peak absorbance of the fitting profile and the calibration curve shown in Fig. 6. According to the fitting result, the standard deviation of the peak absorbance is 0.0018, leading to a measurement uncertainty of 2.2%. Note that the measured ambient N2O concentration level was slightly lower than the globally averaged atmospheric N2O concentration (332 ppb). One possible reason is that the sampling place is located in the downtown of Shenzhen City, which is far away from the main N2O emission sources such as the farmland and industrial district. Similar results were also reported in the previous studies [13,24].
Fig. 8. Measured absorbance of ambient N2O on May 9, 2021 in Shenzhen, a southern city in China. The absorption spectrum corresponds to a N2O concentration of 300 ppb.
4. Conclusion
We have demonstrated a MIR-CEAS gas sensor for the sensitive detection of atmospheric N2O using a DFB-QCL near 4.5 µm and an optical cavity of only 78 mm. The QCL was locked to the compact F-P cavity to reach an effective absorption path length of 96 m. When performing the PDH locking, the injection current of the QCL was modulated at 4 MHz to generate the sidebands. Our technique eliminates the use of an EOM, which is quite expensive or even commercially unavailable in the MIR region. We have achieved a MDL of 0.32 ppb N2O at 50 s integration time in this study. The sensor performance can be further improved by using a longer optical cavity or a higher finesse, as well as the active feedback control of the cavity length. Future work involves the field deployment of such a compact N2O sensor for environmental monitoring in the Pearl River Delta region of China.
Funding
Research Grants Council, University Grants Committee (14209220); Guangdong Science and Technology Department (2020A0505090010); Natural Science Foundation of Shenzhen City (JCYJ20200109143008165).
Disclosures
The authors declare no conflicts of interest.
Data availability
Data underlying the results presented in this paper are publicly available.
References
1. U. N. F. C. on C. C. (Organization), “Kyoto protocol to the United Nations framework convention on climate change,” Rev. Eur. Comp. Int. Environ. Law 7(2), 214–217 (2010).
2. D. J. Wuebbles, “Nitrous oxide: No laughing matter,” Science 326(5949), 56–57 (2009). [CrossRef]
3. A. R. Ravishankara, J. S. Daniel, and R. W. Portmann, “Nitrous oxide (N2O): The dominant ozone-depleting substance emitted in the 21st century,” Science 326(5949), 123–125 (2009). [CrossRef]
4. R. K. Pachauri and L. Meyer, “Climate Change 2014. Synthesis Report,” Environ. Policy Collect. 27(2), 408 (2014).
5. WMO, “ WMO Greenhouse Gas Bulletin No. 16: The State of Greenhouse Gases in the Atmosphere Based on Global Observations through 2019,” (2020).
6. B. Cai, C. Cui, D. Zhang, L. Cao, P. Wu, L. Pang, J. Zhang, and C. Dai, “China city-level greenhouse gas emissions inventory in 2015 and uncertainty analysis,” Appl. Energy 253, 113579 (2019). [CrossRef]
7. H. Sun, J. Jiang, M. Mohsin, J. Zhang, and Y. A. Solangi, “Forecasting Nitrous Oxide emissions based on grey system models,” Environ. Geochem. Health 42(3), 915–931 (2020). [CrossRef]
8. T. D. Rapson and H. Dacres, “Analytical techniques for measuring nitrous oxide,” Trends Anal. Chem. 54, 65–74 (2014). [CrossRef]
9. D. D. Nelson, B. McManus, S. Urbanski, S. Herndon, and M. S. Zahniser, “High precision measurements of atmospheric nitrous oxide and methane using thermoelectrically cooled mid-infrared quantum cascade lasers and detectors,” Spectrochim. Acta - Part A Mol. Biomol. Spectrosc. 60(14), 3325–3335 (2004). [CrossRef]
10. L. Tao, K. Sun, M. A. Khan, D. J. Miller, and M. A. Zondlo, “Compact and portable open-path sensor for simultaneous measurements of atmospheric N2O and CO using a quantum cascade laser,” Opt. Express 20(27), 28106–28118 (2012). [CrossRef]
11. 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]
12. G. D. Banik, S. Som, A. Maity, M. Pal, S. Maithani, S. Mandal, and M. Pradhan, “An EC-QCL based N2O sensor at 5.2 µm using cavity ring-down spectroscopy for environmental applications,” Anal. Methods 9(15), 2315–2320 (2017). [CrossRef]
13. J. Tang, B. Li, and J. Wang, “High-precision measurements of nitrous oxide and methane in air with cavity ring-down spectroscopy at 7.6µm,” Atmos. Meas. Tech. 12(5), 2851–2861 (2019). [CrossRef]
14. D. A. Long, A. J. Fleisher, Q. Liu, and J. T. Hodges, “Ultra-sensitive cavity ring-down spectroscopy in the mid-infrared spectral region,” Opt. Lett. 41(7), 1612–1615 (2016). [CrossRef]
15. J. Wojtas, Z. Bielecki, T. Stacewicz, J. Mikolajczyk, R. Medrzycki, and B. Rutecka, “Application of quantum cascade lasers in nitric oxide and nitrous oxide detection,” Acta Phys. Pol. A 120(4), 794–797 (2011). [CrossRef]
16. F. Nadeem, J. Mandon, S. M. Cristescu, A. Khodabakhsh, and F. J. M. Harren, “Experimental-based comparison between off-axis integrated cavity output spectroscopy and multipass-assisted wavelength modulation spectroscopy at 7.7 µm,” OSA Contin. 2(9), 2667–2682 (2019). [CrossRef]
17. J. Dong, T. T. Y. Lam, M. B. Gray, R. B. Warrington, E. H. Roberts, D. A. Shaddock, D. E. McClelland, and J. H. Chow, “Optical cavity enhanced real-time absorption spectroscopy of CO2 using laser amplitude modulation,” Appl. Phys. Lett. 105(5), 053505 (2014). [CrossRef]
18. Q. He, M. Lou, C. Zheng, W. Ye, Y. Wang, and F. K. Tittel, “Repetitively mode-locked cavity-enhanced absorption spectroscopy (RML-CEAS) for near-infrared gas sensing,” Sensors 17(12), 2792 (2017). [CrossRef]
19. Q. He, C. Zheng, M. Lou, W. Ye, Y. Wang, and F. K. Tittel, “Dual-feedback mid-infrared cavity-enhanced absorption spectroscopy for H2CO detection using a radio-frequency electrically-modulated interband cascade laser,” Opt. Express 26(12), 15436 (2018). [CrossRef]
20. R. W. P. Drever, J. L. Hall, F. V. Kowalski, J. Hough, G. M. Ford, A. J. Munley, and H. Ward, “Laser phase and frequency stabilization using an optical resonator,” Appl. Phys. B 31(2), 97–105 (1983). [CrossRef]
21. E. D. Black, “An introduction to Pound–Drever–Hall laser frequency stabilization,” Am. J. Phys. 69(1), 79–87 (2001). [CrossRef]
22. 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. Császár, 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. Müller, 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. Wcisło, S. Yu, and E. J. Zak, “The HITRAN2016 molecular spectroscopic database,” J. Quant. Spectrosc. Radiat. Transf. 203, 3–69 (2017). [CrossRef]
23. X. Shi, J. Zhang, X. Zeng, X. Lü, K. Liu, J. Xi, Y. Ye, and Z. Lu, “Suppression of residual amplitude modulation effects in the Pound–Drever–Hall locking,” Appl. Phys. B 124(8), 153 (2018). [CrossRef]
24. Q. Wei, B. Li, J. Wang, B. Zhao, and P. Yang, “Impact of residual water vapor on the simultaneous measurements of trace CH4 and N2O in air with cavity ring-down spectroscopy,” Atmosphere 12(2), 221 (2021). [CrossRef]
