Multi-resonator photoacoustic spectroscopy

Multi-resonator photoacoustic spectroscopy

Sensors and Actuators B 251 (2017) 632–636 Contents lists available at ScienceDirect Sensors and Actuators B: Chemical journal homepage: www.elsevie...

1MB Sizes 0 Downloads 41 Views

Sensors and Actuators B 251 (2017) 632–636

Contents lists available at ScienceDirect

Sensors and Actuators B: Chemical journal homepage: www.elsevier.com/locate/snb

Multi-resonator photoacoustic spectroscopy Kun Liu a,∗ , Jiaoxu Mei a , Weijun Zhang a , Weidong Chen b , Xiaoming Gao a,∗ a b

Anhui Institute of Optics and Fine Mechanics, Chinese Academy of Sciences, Hefei 230031, China Laboratoire de Physicochimie de l’Atmosphère, Université du Littoral Côte d’Opale, 189A, Av. Maurice Schumann, 59140 Dunkerque, France

a r t i c l e

i n f o

Article history: Received 5 January 2017 Received in revised form 3 May 2017 Accepted 19 May 2017 Available online 22 May 2017 Keywords: Photoacoustic spectroscopy Sensing Multi-component Multi-laser

a b s t r a c t In this letter, we report on the development of an innovative multi-gas sensor based on multi-resonator photoacoustic spectroscopy (MR-PAS). This novel technique offers multi-laser operation to simultaneously monitor multiple pollutant species using a single photoacoustic spectrophone. A photoacoustic cell including three acoustic resonators operating at different resonant modes was designed, a single microphone was used to listen the photoacoustic signal in each resonator simultaneously. Feasibility and performance of the innovated MR-PAS sensor was demonstrated by simultaneous trace gas detection of H2 O vapor, CH4 and CO2 using three near infrared distributed feedback diode lasers. 1␴ normalized noise equivalent absorption coefficients (NNEA) of 2.1 × 10−9 cm−1 W/Hz1/2 , 2.9 × 10−9 cm−1 W/Hz1/2 and 6.5 × 10−9 cm−1 W/Hz1/2 were respectively achieved for H2 O, CH4 and CO2 detection at normal atmospheric pressure. © 2017 Elsevier B.V. All rights reserved.

1. Introduction Photoacoustic spectroscopy (PAS) is a sensitive, selective and well established method for trace gas detection which has been successfully employed in numerous applications. Amid spectroscopy-based optical sensors, PAS based on photo-acoustic convention effect offers several intrinsic attractive features including ultra-compact size, free from cross-response of light scattering, zero-background and broad band wavelength-independent acoustic signal measurements (free of wavelength-dependent photo-detector). Up to date, various concepts have been explored with respect to PAS-based sensing platform, such as resonant PAS, cantilever enhanced PAS and quartz enhanced PAS [1–7]. There is strong need for the development of PAS sensor capable of multi-wavelength and multi-component detection. One such application, the original motivation for the present work, is the simultaneous measurement of several green house gases and aerosol optical absorption. However, most of the reported PAS sensors are not able to offer multi-component detection capability, except for using widely tunable laser sources. In fact, narrow tuning range of most commercially available lasers make multicomponent detection needing multi-laser sources. For example, recently, H. Wu et al. reported a dual-gas QEPAS sensor by exciting two different quartz tuning fork resonance modes with two lasers

∗ Corresponding authors. E-mail addresses: [email protected] (K. Liu), [email protected] (X. Gao). http://dx.doi.org/10.1016/j.snb.2017.05.114 0925-4005/© 2017 Elsevier B.V. All rights reserved.

[8]. In conventional PAS configuration, a single acoustic resonator operates at its own resonant frequency f0 and it is difficult to simultaneously retrieve the PAS signals resulting from different lasers for multi-component analysis (in this case, monitoring of multigas should be made by consequently switching corresponding laser source). 2. Experimental As an alternative method, we developed an innovative multi-resonator phtoacoustic spectroscopy (MR-PAS) scheme for simultaneous monitoring of multi-component with multi-laser coupled to multi-resonator in a single phtoacoustic spectrophone. Geometric configuration of the MR-PAS cell is shown in Fig. 1. The MR-PAS cell includes three cylindrical acoustic resonators with their own specific lengths offering different acoustic resonant frequencies. These acoustic resonators were set in a cylinder with a diameter of 50 mm and distributed in a circle of 25 mm diameter, as shown in Fig. 1. The inner diameter of each resonator was 10 mm. The length of acoustic resonator 1 (indicated as AR1), acoustic resonator 2 (AR2) and acoustic resonator 3 (AR3) was 120 mm, 110 mm and 100 mm, respectively, such that the resonant frequency of each resonator was separated by a gap of ∼100 Hz that was larger than a FWHM (full width at half maximum) of the resonance profile of a typical cylindrical resonator with a length of 100 mm, which ensures no signal cross-talking during phasesensitive signal demodulation for each resonator. A buffer volume was set at both sides of the cylinder, and the total length of the PAS

K. Liu et al. / Sensors and Actuators B 251 (2017) 632–636

633

Fig. 1. Sectional view of the developed MR-PAS sensor: geometric configuration of multi-resonators PAS cell (left) and geometric distribution of three multi-resonators integrated in a cylinder (right).

Table 1 Summarized parameters of the developed MR-PAS sensor. Parameters

Length, mm Resonant frequency, Hz Molecule Target line, cm−1 Laser wavelength, ␮m Laser power, mW NNEA, cm−1 W/Hz1/2

Acoustic resonator AR1

AR2

AR3

120 1400 CO2 4989.97 2.004 1.5 6.5 × 10−9

110 1510 CH4 6046.95 1.653 8.6 2.9 × 10−9

100 1610 H2 O 7161.41 1.396 7.8 2.1 × 10−9

controller. Acoustic signals from each resonator were detected with the same microphone (BSWA, MP201) and demodulated by using three lock-in amplifiers separately. A unique time constant of the lock-in amplifier of 10 ms in combination with an 18 dB/octave slope (leading to a detection bandwidth of 9.375 Hz) was used. The demodulated signals were subsequently digitalized with a DAQ card (NI-USB-6212) and displayed on a laptop via a LabVIEW interface. The used acoustic resonators and their corresponding specific geometric parameters are given in Table 1. 3. Results and discussion

cell was 220 mm. A hole with an inner diameter of 3 mm was set in the middle of each resonator and used as a sound wave guide tube between acoustic resonator and microphone detector. Such a design allows for the detection of PAS signal in each resonator using a single microphone, as illustrated in Fig. 1. The signal from each resonator was demodulated at its own resonant frequency using a lock-in amplifier. Validation of the developed MR-PAS sensor has been carried out by the measurements of three important green house gases: H2 O vapor, CH4 and CO2 . Selection of the absorption lines for each species was made based on the HITRAN04 database [9]. The selected interference-free lines are given in Table 1. The line for CH4 detection is a R3 triplet consisting of 3 lines: F1 , F2 and A2 , overlapped and unresolvable even at low pressure under Doppler broadening conditions. A schematic of the MR-PAS experimental setup is shown in Fig. 2. Three fiber-coupled distributed feedback (DFB) diode lasers operating at 1.396 ␮m (NEL), 1.653 ␮m (NEL) and 2.004 ␮m (eblana photonics) have been used for detection of H2 O vapor, CH4 and CO2 , respectively. Three commercial diode laser controllers (ILX Lightwave LDC-3724) were used for laser current and temperature controls. Laser wavelength scans were realized by feeding an external voltage ramp from a function generator (SPF05, NANJING SAMPLE INSTRUMENT) to the laser diode current which swept the laser wavelengths back and forth across the absorption feature of each target molecule at a rate of 1 Hz, simultaneously. Wavelength modulation and second harmonic detection was employed in the present work for sensitive trace gas detection. The wavelength modulation was achieved by adding a sine wave to the DFB laser diode current. The sine form wave was supplied by the sinusoidal signal output of a lock-in amplifier (Stanford Research Systems, Model SR 830 DSP). The voltage ramp and the sine wave were combined with a home-made adder and then fed to the laser

Characteristics of each acoustic resonator were experimentally investigated to determine the specific resonant frequency for each resonator at atmospheric environment. Measurements of H2 O vapor absorption signal using a 1.396 ␮m laser was performed for this study. Fig. 3 shows the square of the signal amplitude as a function of frequency. The data were fitted to a Lorentz contour, which describes power in a classical driven oscillator as a function of frequency. From the results shown in Fig. 3, optimum resonant frequencies f10 = 1400 Hz, f20 = 1510 Hz and f30 = 1610 Hz were found for resonators AR1, AR2 and AR3, respectively. These values correspond to the first longitudinal resonant mode of each resonator [1]. To evaluate the capacity of multi-gas detection of the developed MR-PAS proof-of-concept, a mixture of 610 ppm CH4 , 1430 ppm H2 O vapor and 7042 ppm CO2 in N2 and air was filled in the MRPAS sensor at atmospheric pressure. The concentration of each molecule in the mixture was determined with direct absorption spectroscopy. The lasers emitting at 2.004 ␮m (CO2 ), 1.653 ␮m (CH4 ) and 1.396 ␮m (H2 O) were modulated at 700 Hz (half of the optimum frequency, f10 /2), 755 Hz (f20 /2) and 805 Hz (f30 /2), respectively. Modulation amplitude applied to each laser was optimized by investigating detected signal amplitudes at different modulation amplitudes. Fig. 4(a) depicts the 2f-signals of H2 O, CH4 and CO2 obtained simultaneously using optimum modulation amplitudes (850 mV, 600 mV and 800 mV for H2 O, CH4 and CO2, respectively, The 2f-signal of H2 O was divided by 10). The noise level, defined as standard deviation in the baseline fluctuation of a non-absorption portion in the 2f-signal spectrum, was found to be ∼20 ␮V for all resonators. The minimum detection limit were then estimated to be 1.3 ppm for H2 O, 4.4 ppm for CH4 and 140 ppm for CO2 , respectively. Taking into account the used lasers power and the detection bandwidth of 9.37 Hz (10 ms time constant combined with an 18 dB/octave slope filter), 1␴ normalized noise equiva-

634

K. Liu et al. / Sensors and Actuators B 251 (2017) 632–636

Fig. 2. Schematic diagram of the MR-PAS experimental setup.

Fig. 3. Square of the MR-PAS signals amplitude as a function of frequency. The data are fitted to a Lorentz profile.

Fig. 4. (a) 2f photoacoustic signals of H2 O, CH4 and CO2 absorptions. The signal of H2 O absorption was divided by 10. (b) One hour time series measurements of H2 O, CH4 and CO2 absorption levels.

lent absorption coefficients (NNEA) of 2.1 × 10−9 cm−1 W/Hz1/2 , 2.9 × 10−9 cm−1 W/Hz1/2 and 6.5 × 10−9 cm−1 W/Hz1/2 were found for H2 O, CH4 and CO2 detection at normal atmospheric pressure, respectively. Long-term stability and achievable minimum detection limit of the developed MR-PAS system were investigated via one hour continuous measurement of H2 O, CH4 and CO2 . The measured results are shown in Fig. 4(b), in which each point was obtained in 1 s. The best achievable sensitivity and long-term stability of the system were evaluated based on an Allan-Werle deviation analysis as shown in Fig. 5. AR1 (for CO2 detection) and AR3 (for H2 O) had an optimum integration time of about 100 s, while AR2 (for CH4 ) exhibited an longer integration time of up to 400 s due to its optimum buffer length (equal to half of the resonator length) for noise elimination [1], which allowed for more effective fil-

Fig. 5. Allan deviation plots of the developed MR-PAS sensing platform.

K. Liu et al. / Sensors and Actuators B 251 (2017) 632–636

635

Fig. 6. Top signals obtained from resonator AR1, AR2 and AR3, respectively; Bottom: signals from resonator AR1, AR2 and AR3 when light injection into AR2 was blocked.

tering of background fluctuation noise than AR1 and AR3 due to geometry limitation of the MR-PAS spectrophone. According to the results of Allan-Werle deviation analysis showed in Fig. 5, minimum detectable concentrations of 12 ppm and 0.1 ppm could be obtained for detection of CO2 and H2 O in 100 s integration time, respectively, while a minimum detectable concentration of 0.2 ppm could be obtained for CH4 in 400 s. Finally, investigation of cross-talk effects among resonators was experimentally carried out. When blocking the light injected into one of the resonators (for example AR1) while other two resonators keeping in operation, there is no signal detected in the resonator without light injection. This demonstrated that there is no signal cross talking among the resonators. As an example, Fig. 6 shows a result obtained with no laser injection into the AR2, which clearly shows no cross-talk coming from AR1 or AR3 to AR2. 4. Conclusion In conclusion, we have demonstrated a proof-of-concept of an innovative multi-gas sensor based on multi-resonator photoacoustic spectroscopy (MR-PAS) coupled to multi-laser excitations using a single microphone. The MR-PAS sensing capacity was validated using simultaneous measurements of CH4 , CO2 and H2 O with minimum detection limits of 0.2 ppm, 12 ppm and 0.1 ppm, respectively. The MR-PAS performance could be further improved by optimization of its configuration, the parameters of the multiresonators and higher laser powers. The reported development would significantly reduce the size of multi-laser based PAS sensor for multi-pollutant analyses, which is, in particular, very attractive for application in multi-wavelength measurements of wavelengthdependent aerosol optical absorption coefficients. Acknowledgements This work was financially supported by the National Science Foundation of China (No. 41475023, No. 41575030, No. 21307136, NO. 41330424 and No. 41205120) and partially by the French Agence Nationale de la Recherche (ANR) under the LABEX-CaPPA (ANR-10-LABX-005) contracts.

References [1] A. A.Miklos, P. Hess, Z. Bozoki, Application of acoustic resonators in photoacoustic trace gas analysis and metrology, Rev. Sci. Instrum. 72 (2001) 1937–1955. [2] H. Moser, B. Lendl, Cantilever-enhanced photoacoustic detection of hydrogen sulfide (H2S) using NIR telecom laser sources near 1.6 ␮m, Appl. Phys. B 122 (2016) 1–11. [3] A. Kosterev, Y.A. Bakhirkin, R.F. Curl, F.K. Tittel, Quartz-enhanced photoacoustic spectroscopy, Opt. Lett. 27 (2002) 1902–1904. [4] K. Liu, X. Guo, H. Yi, W. Chen, W. Zhang, X. Gao, Off-beam quartz-enhanced photoacoustic spectroscopy, Opt. Lett. 34 (2009) 1594–1596. [5] S. Borri, P. Patimisco, I. Galli, D. Mazzotti, G. Giusfredi, N. Akikusa, M. Yamanishi, G. Scamarcio, P. De Natale, V. Spagnolo, Intracavity quartz-enhanced photoacoustic sensor, Appl. Phys. Lett. 104 (2014) 091114. [6] H. Wu, L. Dong, H. Zheng, X. Liu, X. Yin, W. Ma, L. Zhang, W. Yin, Su. Jia, F.K. Tittel, Enhanced near-infrared QEPAS sensor for sub-ppm level H2S detection by means of a fiberamplified 1582 nm DFB laser, Sens. Actuators B: Chem. 221 (2015) 666–672. [7] Z. Li, C. Shi, W. Ren, Mid-infrared multimode fiber-coupled quantum cascade laser for off-beam quartz-enhanced photoacoustic detection, Opt. Lett. 41 (2016) 4095–4098. [8] H. Wu, X. Yin, L. Dong, K. Pei, A. Sampaolo, P. Patimisco, H. Zheng, W. Ma, L. Zhang, W. Yin, L. Xiao, V. Spagnolo, S. Jia, F.K. Tittel, Simultaneous dual-gas QEPAS detection based on a fundamental and overtone combined vibration of quartz tuning fork, Appl. Phys. Lett. 110 (2017) 121104. [9] L.S. Rothman, D. Jacquemart, A. Barbe, D. Chris Benner, M. Birk, L.R. Brown, M.R. Carleer, C. Chackerian, K. Chance, L.H. Coudert, V. Dana, V.M. Devi, J.-M. Flaud, R.R. Gamache, A. Goldman, J.-M. Hartmann, K.W. Jucks, A.G. Maki, J.-Y. Mandin, S.T. Massie, J. Orphal, A. Perrin, C.P. Rinsland, M.A.H. Smith, J. Tennyson, R.N. Tolchenov, R.A. Toth, J. Vande Auwera, P. Varanasi, G. Wagner, The HITRAN 2004 molecular spectroscopic database, J. Quantum Spectrosc. Radiat. Tadiat. Transfer 96 (2005) 139–204.

Biographies Kun Liu received his Ph.D. degree in optics from Anhui Institute of Optics & Fine Mechanics, Chinese Academy Sciences in 2010. Now he is an associate professor of Anhui Institute of Optics & Fine Mechanics, Chinese Academy Sciences. His research interests include photoacoustic spectroscopy and laser spectroscopy for application in atmospheric photochemistry and environmental science. Jiaoxu Mei received his Ph.D. degree in optics from Anhui Institute of Optics & Fine Mechanics, Chinese Academy Sciences in 2015. Now he is an assistant research fellow of Anhui Institute of Optics & Fine Mechanics, Chinese Academy Sciences. His research interests include optic-electric sensor technology, laser spectroscopy and application. Weijun Zhang is Professor at Anhui Institute of Optics & Fine Mechanics, Chinese Academy of Sciences. He received his Ph.D degree from Anhui Institute of Optics &

636

K. Liu et al. / Sensors and Actuators B 251 (2017) 632–636

Fine Mechanics, Chinese Academy Sciences. His research interests include mechanism of second pollution and aerosol formation. Weidong Chen received his PhD from the University of Sciences & Technologies of Lille (France) in 1991. He obtained his State PhD (HDR) in 2001 and became full Professor of University of Sciences & Technologies of Lille in 2003. His research interests include developments of photonic instrumentation for applied spectroscopy, Optical metrology of trace gases for applications in atmospheric photochemistry and environmental science.

Xiaoming Gao is Professor at Anhui Institute of Optics & Fine Mechanics, Chinese Academy of Sciences. He received his Ph.D degree from Anhui Institute of Optics & Fine Mechanics, Chinese Academy Sciences in 1998. He studied and worked in MaxPlanck- Institute for Quantumoptics, German as postdoctoral from 1998 to 2000. His research interests include atmospheric molecules absorption spectroscopy, high sensitive diode laser spectroscopy and application.