Performance test, calibration and validation of a novel optical analyzer for continuous and high precision CO 2 isotope ratio measurements Performance test, calibration and validation of a novel optical analyzer for continuous and high precision CO 2 isotope ratio measurements B. Tuzson 1* , J. Mohn 1 , L. Emmenegger 1 , R.A. Werner 2 , M.J. Zeeman 2 1) Empa, Swiss Federal Laboratories for Materials Testing and Research, Dübendorf, Switzerland 2) ETH Zurich, Institute of Plant Sciences, Switzerland *bela.tuzson@empa.ch B. Tuzson 1* , J. Mohn 1 , L. Emmenegger 1 , R.A. Werner 2 , M.J. Zeeman 2 1) Empa, Swiss Federal Laboratories for Materials Testing and Research, Dübendorf, Switzerland 2) ETH Zurich, Institute of Plant Sciences, Switzerland *bela.tuzson@empa.ch The instrument was developed for continuous and high precision CO 2 isotope ratio (both 13 C/ 12 C and 18 O/ 16 O) measurements at ambient air concentration [1,2]. A single-mode, pulsed quantum cascade laser (QCL) operating near 4.3 μm at quasi-room temperature is employed as ligth source. The emitted IR-radiation, is after collimation, divided in two equal beams by a wedged ZnSe beamsplitter and then directed through a dual multi-pass cell assembly. After passing through the cells, the outcoming beams are detected by two TEC photodiodes. A removable Ge-etalon allows for accurate frequency calibration. Fig.1 Schematic diagram of the laser spectrometer optics. The dual- cell arrangement allows for simultaneous sample-reference gas determination which improves the instrument accuracy. Uncertainties in the spectral fitting procedure are considerably reduced by applying the spectral analysis to the ratio of the sample and reference spectra. Introduction Introduction Instrumental Setup Instrumental Setup Isotope ratios of carbon dioxide are highly valuable to investigate its sources, sinks and fate at local and global scales. However, such studies generally require extensive and long-term measurements under field conditions, which may not be feasible with standard isotope ratio mass spectrometers (IRMS). Here we present an alternative analytical tool based on direct absorption spectroscopy employing quantum cascade lasers (QCL). The instrument is compact, mobile and inherently has high temporal resolution. It is capable to continuously analyze air samples in situ, because it does not require any specific sample preparation. Acknowledgements: We gratefully acknowledge Alpes Lasers (CH) for providing us excellent QCLs at the desired wavelength. M. Zahniser, D. Nelson and B. McManus from Aerodyne Research Inc. (USA) are acknowledged for their continuous support.  Precision: The long-term stability as well as the short term precision of the instruments was evaluated using the Allan variance technique. This indicates an achievable precision in the CO 2 isotope ratios at ambient air concentrations of 0.03 ‰ and 0.05 ‰ for δ 13 C and δ 18 O, respectively [2].  Accuracy: The linearity and the accuracy of the δ-scale was determined employing several air tanks which have similar CO 2 mole fractions, but differ in their stable isotope ratio. These gases were previously analyzed by high precision IRMS [3] and then measured by QCLAS. Accuracy of <0.2‰ was achieved for δ 13 C, whereas the δ 18 O showed a factor two higher scatter in the correlation plot. Laboratory investigations Laboratory investigations This poster illustrates the development of a QCL based spectroscopic technique for the simultaneous and high precision measurement of 12 CO 2 , 13 CO 2 and 12 C 16 O 18 O in atmospheric carbon dioxide. Long-term and continuous measurements are made possible by employing cryogen-free components for both the laser and detector. Furthermore, an adequate gas handling and calibration unit has been developed, which assures a rigorous control of gas temperatures, pressures and flow rates. At the moment the instrument is used for a feasibility study on a high alpine site (Jungfraujoch, Switzerland, 3580 masl). Conclusion & Outlook Conclusion & Outlook Calibration & Sampling Calibration & Sampling [3] Werner, R.A, M. Rothe, W.A. Brand, Rapid Commun. Mass Spectrom., 15, 2152-2167, (2001) [4] Zeeman, M.J., B. Tuzson, L. Emmenegger, A. Knohl, N. Buchmann, W. Eugster, Biogeosciences Discuss., 6, 3481-3510 (2009). Fig.2 Picture of the complete instrument. The modular construction results in high mobility and compactness. The use of novel laser and detectors assures cryogen-free operation, thus facilitating long-term measurements. 10 -10 2 4 10 -9 2 4 10 -8 2 4 10 -7 Allan Variance (σ 2 ) 1 10 100 1000 Integration Time (s) 0.843 0.842 0.841 Isotopic Ratios 3500 3000 2500 2000 1500 1000 500 0 Collected Data Time (s) 0.930 0.929 0.928 [ 13 CO 2 ]/[ 12 CO 2 ] σ (1s) = 0.24 ‰ σ min = 0.03 ‰ [ 12 C 16 O 18 O]/[ 12 CO 2 ] σ (1s) = 0.29 ‰ σ min = 0.05 ‰ -40 -30 -20 -10 0 δ IRMS (‰) 0.840 0.830 0.820 [ 13 CO 2 ]/[ 12 CO 2 ] QCL -0.2 -0.1 0.0 0.1 0.2 Residual (‰) R 2 = 0.99996 Fig.3 Time series and associated Allan plots of the simultaneously retrieved spectroscopic ratio of 13 CO 2 / 12 CO 2 and 12 C 16 O 18 O/ 12 CO 2 , respectively. Fig.4 Comparison of carbon isotope ratios measured by high precision IRMS and QCLAS. Beside the instrumental development, the issue of air sampling and calibration has also been considered. The unit includes gas handling, drying, temperature stabilization and automatic calibration system. No further sample preparation is needed. Fig.5 Schematics and picture of the gas handling unit designed for the QCL spectrometer. The abbreviations are as follows: Vi – 3-way solenoid valve, MV – manual precision valve, MFC – mass flow controller, PC – pressure controller. Validation & Field Application Validation & Field Application The instrument was successfully operated in various field campaigns, including grass-land ecosystem – atmosphere exchange (gradient and eddy-flux method) and forest soil carbon dynamics studies, and delivered continuous mixing ratio data of the three main CO 2 isotopologues [2,4]. Here we show some relevant results of the field experiments. Fig.6 a) Time series of the CO 2 mixing ratio measured by the QCLAS (line) and with the standardized IRGA (dots). The corresponding δ 13 C and δ 18 C values measured by the laser spectrometer are shown in b) and c). δ -values measured by IRMS on collected flask samples are also given for comparison. Diurnal variations in the CO 2 concentration indicate ecosystem activity (photosynthesis and respiration). The closed symbols for the δ 18 O values indicate sampling issues with small volume metal flasks and possible isotope exchange effects between water and carbon dioxide. 2000 1600 1200 800 400 CO 2 (ppm) 17.08.2007 19.08.2007 21.08.2007 23.08.2007 Date & Time -25 -20 -15 -10 -5 δ 13 C (‰) -15 -10 -5 0 5 δ 18 O (‰) a) b) c) -25 -20 -15 -10 -5 δ 13 C (‰) 0.003 0.002 0.001 1/[CO 2 ] (ppm -1 ) y QCLAS = 7062(± 29)*x - 26.88(± 0.06) y QCLAS = 7043(± 67)*x - 26.87(± 0.13) y IRMS = 6955(±182)*x - 27.14(± 0.39) QCLAS ( 1s) QCLAS (55 s) IRMS -6 -4 -2 0 2 4 6 δ 18 O (‰) 0.003 0.002 0.001 1/[CO 2 ] (ppm -1 ) y QCLAS = 3470(± 32)*x - 6.56(± 0.06) y QCLAS = 3357(± 79)*x - 6.34(± 0.14) y IRMS = 2810(±346)*x - 7.70(± 0.73) QCLAS ( 1s) QCLAS (55s) IRMS Fig.7 Keeling plots for data collected during 24 hours at measurement height of 0.1 m. Isotope ratios are plotted against the inverse of CO 2 concentration. The intercept and the slope of the linear regression is given together with their standard deviation. Besides the QCLAS data points (circles), the field-collected flask samples are also plotted (rectangles). The open symbols represent the day-time, while the closed symbols the night-time measurements. The high temporal resolution (one second) QCLAS data are also presented in the background (dots). [1] Nelson D.D., J.B. McManus, S.C. Herndon, M.S. Zahniser, B. Tuzson, L. Emmenegger, Appl. Phys. B 90, 301–309, (2008). [2] Tuzson B., J. Mohn, M.J. Zeeman, R.A. Werner, W. Eugster, M.S. Zahniser, D.D. Nelson, J.B. McManus, L. Emmenegger, Appl. Phys. B 92, 451–458, (2008). 15 th WMO/IAEA Meeting of Experts on Carbon Dioxide, Other Greenhouse Gases, and Related Tracer Measurement Techniques, Jena (Germany), September 7-10, 2009