Professional Documents
Culture Documents
pubs.acs.org/ac
■ EXPERIMENTAL SETUP
The aim of the presented work was the optimization of the
built identically to the first PA cell and refilled every six months.
The PA signal of this cell is demodulated at 3f, resulting in a
zero PA signal at the absorption maximum, which again allows
analytical performance of a PA system with a clear focus on the for a precise identification of the maximum. This information is
routine application in the rough environment of WWTPs with used to control the QCL wavelength via the QCL temperature
respect to sensitivity and selectivity as well as robustness and and the in-house made control software written in LabView
costs. The complete analytical system for the quantification of 2013 (National Instruments).
N2O emissions from WWTPs comprises a gas sampling hood, a The second laser system consists of a 2.9 μm DFB laser
PA system, and a pump providing a continuous gas flow from diode (Nanoplus, Germany) with an optical power of 5.5 mW.
the hood to the detector. We compared two different laser An integrated thermoelectric cooling device and a negative-
systems, operating at two different emission wavelengths, temperature-coefficient thermistor was attached to the laser
different PA cell designs, and different signal detection and diode (LD). Temperature and current were controlled by a
demodulation systems. LDC1000 (Profile, U.S.A.), leading to a variable emission from
Laser Systems. The PA setup basically comprises laser 2893 to 2900 nm. The laser is modulated with a simple
source including the corresponding driver and thermostatic transistor-to-transistor logic signal at 4 kHz. The laser beam is
control and a PA cell with an acoustic resonator (see Figure 1). focused into the PA cell described above by an aspherical lens
One laser source is a QCL (AdTechoptics, U.S.A.) purchased ( f = 25.4 mm). For this setup, a second PA cell was not
from EQ Photonics (Eching, Germany) with an emission band necessary. Temperature regulation was precise enough to
ranging from 4.53 to 4.54 μm and a maximum output power of ensure a high stability and therefore no wavelengths drift.
B DOI: 10.1021/acs.analchem.7b00491
Anal. Chem. XXXX, XXX, XXX−XXX
Analytical Chemistry Article
PA Cells. The PA cells are conventional PA cells featuring valve is connected to the top of the floating hood to prevent air
longitudinal acoustic resonances in a central optical resonator from entering.
formed by a cylindrical glass tube (length 40 mm, inner To prevent foam, which can cover the reactor’s liquid surface
diameter 4 mm) equipped in its middle with a condenser as high as 20 cm, from entering the sampling line, a sieve cloth
microphone (EK-23029, Knowles, U.S.A.) for the PA signal (60 μm mesh) is placed between the gas-outlet and the tubing.
detection and a suitable preamplifier, as described elsewhere.24 A pump (VDE0530, KNF Neuberger GmbH, Germany)
In earlier experiments, we successfully used a Helmholtz type provides a continuous gas-flow of 700 mL/min through the
resonator as the PA detection cell in combination with a whole system of the sampling hood and PA analyzer. During
QCL.23 This cell type is particularly suitable for laser sources the aerated phases of the process, excess air can escape from the
with limited beam quality. The system essentially consists of ball check valve, which is connected to a mass flow meter (red-
two coaxial cylinders (diameter 22 mm, length 2 mm), y, Vögtlin Instruments AG, Switzerland). The concentration
combined by a connecting duct (diameter 4 mm, length 2 measured by our PA device, in combination with the total
mm). The laser source is positioned directly at the center of emitted volume, which is the sum of the flow through our
one of the cell inlets without any focusing element. Aluminum instrument and the volume emitted through the ball check
was chosen as the cell material due to its high reflectivity for the valve, allows for a precise calculation of the total N2O
laser light; additionally, the inner surface of the cell was emissions. During unaerated phases, ambient air is aspirated
polished. A gas inlet is positioned laterally into the same cell, through the gas-inlet for keeping the system at ambient
while a microphone sits in the other cell right opposite to the pressure. The low flow rate in the sampling hood allows for
connecting tube. Finite element simulation (Comsol Multi- equilibration between the liquid-phase and the gas phase N2O
physics, Germany) was performed to estimate the eigenfre- concentration. The sampling hood is held in position by an
quency of the cell (see Figure S2, SI). aluminum bar of 4 m length, connected to the hood by a hinge
As an alternative to the microphone described above, which joint enabling us to follow the water surface even during
was used in longitudinal PA cells, a microelectro-mechanical turbulent aeration phases. Simultaneous determination of the
dissolved N2O concentration is performed using an electro-
system (MEMS) microphone from Knowles (SPM0408LE5H-
chemical N2O sensor with a detection limit of 0.5 μM, a
TB, Knowles, U.S.A.) was tested. The advantage of this
response time below 45 s and a working range up to 300 μM
microphone is its high sensitivity of 126 mV/Pa and an output
(Unisense Environment A/S, Denmark), fixed in the center
impedance of 200 Ω, which makes higher amplification
part of the hood approximately 20 cm below the liquid surface.
possible. As an all solid-state device, it is particularly robust, As an alternative to the measurement using the sampling hood,
chemically as well as mechanically. the total N2O emission of the reactor could be sampled via the
Signal Recording and Demodulation. Three different headspace, as this reactor is covered by a nearly gastight hatch.
approaches for signal recording and demodulation were An exhaust tube on its top allows for a time-resolved, spatially
compared. The most conventional system uses a commercial integrating analysis of the gas emission.
lock-in amplifier (SR 830, Stanford Research, U.S.A.) to detect Calibration Gases. For the calibration of the PA-systems,
the preamplified PA signal. In the second approach, the signals commercial test gases of different concentrations were used:
are recorded using an analogue to digital converter (ADC) (NI- 98%, 1%, 200 ppmv and 20 ppmv N2O in nitrogen (Westfalen
PCI 4474 dynamic signal acquisition card, National Instru- AG, Germany). The test gases were diluted with pure nitrogen
ments, U.S.A.) recording four channels at 102.4 kS/S and a 24- (99.99% N2) using rotameters. To assess the influence of
bit resolution. The signal lock-in is performed by an in-house humidity on the PA signal, water-saturated test gases were
software, also programmed in LabView 2013. The third achieved by bubbling the gases through a washing bottle and
approach employs a simple PC soundcard (24-bit resolution, then diluted to the required humidity.
■
Xonar DG, ASUS, Taiwan) instead of the ADC for the signal
recording, but the same software-based demodulation. Temper- RESULTS AND DISCUSSION
ature and humidity of the gas are measured by analogue sensors Comparison of the Recording and Demodulation
(HIH4010−004, Honeywell, U.S.A. and TMP36GZ, Analog Systems. A comparison of the three data recording systems
Devices, Inc. U.S.A.) connected to the measurement computer was performed using the 2.9 μm DFB laser source. It was found
via the Arduino. The whole setup is integrated in a mobile that the Stanford Research SR830 lock-in amplifier yields an
housing, allowing field-measurements. approximately 10% lower standard deviation, resulting in a
Gas Sampling System for WWTPs. For the collection of slightly lower LOD than the combination of the ADC with the
the emitted off-gas at a specific spot in a full-scale wastewater software demodulator (see Figure S4, SI).
treatment plant (more specifically in the nitritation step of a The comparison of the ADC with the soundcard yielded the
two-stage deammonification process being intermittently surprising result that there was no significant difference
aerated by aeration membranes installed at the bottom of the between the two with respect to the LOD and the relative
reactor with an aeration duration of 45 min followed by an standard deviation of the signals. A major limitation of the
unaerated phase of 15 min), a floating sampling hood was soundcard approach is that it has only two channels, one for the
placed in the reactor (see Figure S3, SI). The square-shaped PA signal and one for the modulation reference required for the
hood is made up of stainless steel (V2A) with floating bodies lock-in, meaning that the detection of two independent PA
on each side and a cross-sectional surface of 0.5 m2. A gas-inlet channels is not possible (i.e., the signal from two PA cells
and a gas-outlet are located on two opposite corners of the cannot be recorded simultaneously). The SR830 has two input
hood. The gas-outlet is connected via gastight tubing to the PA- channels plus a separate channel for the modulation reference,
detection system, which actively sucks in the sample air. Short but these two channels can only be used in a differential way,
distances between the hood and the PAS system were not for independent measurements. Only the ADC has four
emphasized to enable a quick response time. A ball check independent channels, of which one is required for the
C DOI: 10.1021/acs.analchem.7b00491
Anal. Chem. XXXX, XXX, XXX−XXX
Analytical Chemistry Article
■ AUTHOR INFORMATION
Corresponding Author
*E-mail: Haisch@tum.de. Tel. +49 89 2180-78242. Fax -99
78242.
Author Contributions
The manuscript was written through contributions of all
authors. All authors have given approval to the final version of
the manuscript.
Notes
The authors declare no competing financial interest.
■
concentration (e.g., at 12.10 h) is due to local fluctuations.
The bioreactor content is not homogeneous and has strong REFERENCES
spatial variations. A mixing device installed at medium height of (1) Myhre, G., Shindell, D.; Bréon, F.-M., Collins, W.; Fuglestvedt, J.;
the reactor leads to local fluctuations. The total emission Huang, J.; Koch, D.; Lamarque, J.-F.; Lee, D.; Mendoza, B.; Nakajima,
sampling at the top of the reactor cover avoids this problem. An T.; Stephens, G.; Takemura, T.; Zhang, H.. In Climate Change 2013:
overnight analysis is plotted in Figure S8 (SI). Qualitatively, the The Physical Science Basis. Contribution of Working Group I to the Fifth
behaviors of both, gas-phase and liquid-phase N2O concen- Assessment Report of the Intergovernmental Panel on Climate Change;
trations are identical. The variations over the whole analysis Stocker, T. F., Qin, D., Plattner, G.-K., Tignor, M., Allen, S. K.,
time can be attributed to spatially variable conditions in the Boschung, J., Nauels, A., Xia, Y., Bex, V., Midgley, P. M., Eds.;
reactor, such as different aeration intensities causing not only Cambridge University Press: Cambridge, U.K., 2013.
differential stripping effects, but also varying dissolved oxygen (2) Ravishankara, A. R.; Daniel, J. S.; Portmann, R. W. Science 2009,
326, 123−125.
concentration affecting the biological N2O production.
■
(3) Wuebbles, D. J. Science 2009, 326, 56−57.
(4) Machida, T.; Nakazawa, T.; Fujii, Y.; Aoki, S.; Watanabe, O.
CONCLUSIONS Geophys. Res. Lett. 1995, 22, 2921−2924.
For the detection and quantification of gaseous N2O (5) World Meteorological Organization (WMO). GAW Report No.
concentration, different PA setups were tested and compared, 185. Guidelines for the Measurement of Methane and Nitrous Oxide and
namely, a QCL system at 4.5 μm and a DFB system at 2.9 μm. their Quality Assurance; WMO: Geneva, Switzerland, 2009.
The LOD was improved by the use of 2f modulation. The costs (6) Hamilton, J. F.; Lewis, A. C. In Analytical Techniques for
of the setups were reduced using a PC soundcard as an analog- Atmospheric Measurement; Dwayne, H., Ed.; Blackwell Publishing:
to-digital converter and a software lock-in amplifier instead of a Oxford, 2006.
(7) Hammer, S.; Griffith, D. W. T.; Konrad, G.; Vardag, S.; Caldow,
much larger instrumental lock-in amplifier. This also allows for C.; Levin, I. Atmos. Meas. Tech. 2013, 6, 1153−1170.
the development of a robust and highly mobile setup, which (8) Griffith, D. W. T.; Deutscher, N. M.; Caldow, C.; Kettlewell, G.;
was tested at a WWTP. The results regarding the comparison Riggenbach, M.; Hammer, S. Atmos. Meas. Tech. 2012, 5, 2481−2498.
of different PA setups is valid for other analyst accessible in the (9) Rapson, T. D.; Dacres, H. TrAC, Trends Anal. Chem. 2014, 54,
IR range and can thus be transferred to other applications, such 65−74.
as automotive exhaust gas monitoring or industrial process (10) Yu, Y.; Sanchez, N. P.; Griffin, R. J.; Tittel, F. K. Opt. Express
monitoring. The specific optimum performance can be selected 2016, 24, 10391−10401.
according to the specific demands. (11) Dong, L.; Tittel, F. K.; Li, C. G.; Sanchez, N. P.; Wu, H. P.;
■
Zheng, C. T.; Yu, Y. J.; Sampaolo, A.; Griffin, R. J. Opt. Express 2016,
ASSOCIATED CONTENT 24, A528−A535.
(12) Haisch, C. Meas. Sci. Technol. 2012, 23, Article No. 012001.
*
S Supporting Information
(13) Yang, X. K.; Petrillo, C. J.; Noda, C. Chem. Phys. Lett. 1993, 214,
The Supporting Information is available free of charge on the 536−540.
ACS Publications website at DOI: 10.1021/acs.anal- (14) Schramm, D. U.; Sthel, M. S.; da Silva, M. G.; Carneiro, L. O.;
chem.7b00491. Souza, A. P.; Vargas, H. Rev. Sci. Instrum. 2003, 74, 513−515.
Additional data including absorption coefficients of N2O, (15) Grossel, A.; Zeninari, V.; Parvitte, B.; Joly, L.; Courtois, D. Appl.
finite element simulation of acoustic pressure and sound Phys. B: Lasers Opt. 2007, 88, 483−492.
(16) Rocha, A. M.; Sthel, M. S.; de Castro, M. P. P.; Mothe, G. A.;
pressure, image of gas sampling hood, relative standard Silva, W. C.; Perez, V. H.; da Silva, M. G.; Miklos, A.; Vargas, H. Energy
deviation of a commercial lock-in amplifier compared to Fuels 2014, 28, 4028−4032.
a software lock-in amplifier, simulated absorption spectra (17) Lima, J. P.; Vargas, H.; Miklos, A.; Angelmahr, M.; Hess, P.
of the relevant biogas components, calibration of the Appl. Phys. B: Lasers Opt. 2006, 85, 279−284.
QCL-PA-system, image of PA prototype instrument, and (18) Castleden, S. L.; Kirkbright, G. F.; Spillane, D. E. M. Anal. Chem.
N2O concentration in WWTP (PDF) 1981, 53, 2228−2231.
F DOI: 10.1021/acs.analchem.7b00491
Anal. Chem. XXXX, XXX, XXX−XXX
Analytical Chemistry Article
(19) Peng, Y.; Zhang, W.; Li, L.; Yu, Q. X. Spectrochim. Acta, Part A
2009, 74, 924−927.
(20) Schilt, S.; Kosterev, A. A.; Tittel, F. K. Appl. Phys. B: Lasers Opt.
2009, 95, 813−824.
(21) Santoni, G. W.; Daube, B. C.; Kort, E. A.; Jiménez, R.; Park, S.;
Pittman, J. V.; Gottlieb, E.; Xiang, B.; Zahniser, M. S.; Nelson, D. D.;
McManus, J. B.; Peischl, J.; Ryerson, T. B.; Holloway, J. S.; Andrews,
A. E.; Sweeney, C.; Hall, B.; Hintsa, E. J.; Moore, F. L.; Elkins, J. W.;
Hurst, D. F.; Stephens, B. B.; Bent, J.; Wofsy, S. C. Atmos. Meas. Tech.
2014, 7, 1509−1526.
(22) Couto, F. M.; Sthel, M. S.; Castro, M. P. P.; da Silva, M. G.;
Rocha, M. V.; Tavares, J. R.; Veiga, C. F. M.; Vargas, H. Appl. Phys. B:
Lasers Opt. 2014, 117, 897−903.
(23) Lendl, B.; Ritter, W.; Harsek, M.; Niessner, R.; Haisch, C.
Proceedings of IEEE Sensors 2006, EXCO; Daegu, Korea, October 22−
25, 2006; pp 338−341.
(24) Beck, H. A.; Niessner, R.; Haisch, C. Anal. Bioanal. Chem. 2003,
375, 1136−1143.
(25) The HITRAN Database. http://www.cfa.harvard.edu/
HITRAN/.
(26) Frison, N.; Chiumenti, A.; Katsou, E.; Malamis, S.; Bolzonella,
D.; Fatone, F. J. Cleaner Prod. 2015, 93, 126−133.
(27) Lackey, J. C.; Peppley, B.; Champagne, P.; Maier, A. Waste
Manage. Res. 2015, 33, 767−771.
(28) Du, Z.; Zhen, W.; Zhang, Z.; Li, J.; Gao, N. Appl. Phys. B: Lasers
Opt. 2016, 122, 100.
(29) Rooth, R. A.; Verhage, A. J. L.; Wouters, L. W. Appl. Opt. 1990,
29, 3643−3653.
(30) Kosterev, A. A.; Mosely, T. S.; Tittel, F. K. Appl. Phys. B: Lasers
Opt. 2006, 85, 295−300.
(31) Avramides, E.; Hunter, T. F. J. Chem. Soc., Faraday Trans. 2
1979, 75, 515−527.
(32) Avramides, E.; Hunter, T. F. Chem. Phys. 1981, 57, 441−451.
G DOI: 10.1021/acs.analchem.7b00491
Anal. Chem. XXXX, XXX, XXX−XXX