Project Topics
Get Complete Project Material File(s) Now! »
Metrology for the in-situ quantitative measurement of CO2 and CH4 concentrations in soil and subsoil.
Il est largement admis qu’une mesure précise et continue du CO2 du sol et du sous-sol est très importante pour mieux comprendre le processus de transfert de gaz entre la géosphère, la biosphère et l’atmosphère. La mesure des ces concentrations repose avant tout sur une calibration fine en laboratoire, des capteurs utilisés. Pour cela, ce travail décrit une technique d’étalonnage spécifique appliquée en laboratoire pour la mesure quantitative des gaz du sous-sol à l’aide d’un spectromètre infrarouge portable connecté à une cellule à gaz à faible trajet optique (l = 5 cm) et à résolution modérée (1 cm-1). Les concentrations de gaz ont été validées de manière croisée en combinant un spectromètre infrarouge portable, une spectroscopie à anneau de cavité, une chromatographie en phase gazeuse et des capteurs NDIR de faible puissance. Les courbes d’étalonnage par spectroscopie infrarouge à transformée de Fourier (FTIRS) du CO2 ont été établies dans une plage de concentration allant 100 à 60 000 ppm. De la même manière, les courbes d’étalonnage par FTIRS du CH4 ont été établies dans la plage de concentration allant de 50 à 5000 ppm. Dans ces conditions de mesures en faible réolution (Res =1cm-1) les lois de proportionalités entre l’aire d’une bande (A) et sa concentration (CCO2, CCH4) sont du type polynomiales à savoir : CCO2=−0.0286 3 + 4.67 2 + 450 − 2.17( 2 = 1) et CCH4 = −1.06 3 + 39.4 2 + 1130 − 1.13 ( 2 = 0.9999). Même si pendant les mesures de terrain, la concentration de CH4 a toujours été inférieure à la limite de sensibilité de notre capteur IR (<5 ppm), elle était toujours détectable avec une analyse ex-situ de micro-GC. Le bon accord entre les différentes valeurs de concentration données par les différents capteurs utilisés (FTIRS, Non Dispersive infared Sensor – NDIR, Gas Chromatography – GC et Cavity Ring Down Spectroscopy – CRDS) valide l’utilisation d’un seul capteur IR sur le terrain, limitant ainsi les consommations d’énergie tout en simplifiant le dispositif expérimental. Une telle stratégie de surveillance pour la mesure des gaz offre de grandes possibilités de surveillance à long terme pour dériver la distribution verticale du CO2 et du CH4 ainsi que pour aider à créer des données empiriques sur les flux de gaz à partir du sous-sol sur plusieurs cycles saisonniers.
Abstract
It is a widely held view that accurate and continuous measurement of the subsoil CO2 is fundamental to understand better the gas transfer process between the geosphere, biosphere, and atmosphere. This work describes a specific calibration technique applied in the laboratory for quantitative subsoil gas measurement using a portable infrared spectrometer connected to a gas cell with low path optical (l = 5 cm) and moderate resolution (1 cm-1). Gas concentrations were cross-validated by combining portable infrared spectrometer, cavity ring-down spectroscopy, gas chromatography, and low-power NDIR sensors. The FTIR calibration curves of CO2 have been established in the range of 100 to 60,000 ppm. Similarly, the FTIR calibration curves of CH4 were created in the range of 50 to 5000 ppm fit a simple polynomial fitting on CCO2= −0.0286 3 + 4.67 2 + 450 − 2.17( 2 = 1) and CCH4 = −1.06 3 + 39.4 2 + 1130 − 1.13 ( 2 = 0.9999), where CCO2 and CCH4 are the CO2 and CH4 concentration and A is a unit area of absorbance band consecutively. Even though CH4 concentration under 5 ppm was always below the sensibility limit of our IR sensor during the field test, it was still detectable with a micro-GC ex-situ analysis. The excellent agreement of CO2 concentration derived from FTIR, NDIR, GC, and CRDS justifies using a single IR sensor in the field, resulting in low energy consumption and less complexity. Such a monitoring strategy for gas measurement offers great long-term monitoring possibilities to derive CO2 and CH4’s vertical distribution and help create empirical gas fluxes data from the subsoil.
Fundamental and Instrumental Generalities: Application to Gas Studies
The accurate and continuous measurement of the gas exchange between the geosphere, biosphere, and atmosphere is essential to detect any slight increase of atmospheric CO2 and CH4 (Frey et al., 2015). Several factors are known to influence the precision and accuracy of greenhouse gas measurement, including the analyzer’s quality, the calibration strategy, and the experimental protocol.
FTIR Spectroscopy (FTIRS)
The term Infrared (IR) refers to electromagnetic radiation that falls in the region spanning from 0.7 µm to 1000 µm. The so-called mid-IR that falls in the region between 2.5 µm to 25 µm (4000 to 400 cm-1) is considered the most helpful region for analyzing and identifying chemical functional groups (Larkin, 2011). spectroscopy is widely viewed as very reliable in molecular studies for analyzing quantitatively and qualitatively. It can be used for understanding the interaction of IR radiation and samples that can be either solid, liquid, or gas by measuring the frequency and intensity of IR absorption (Maria, 2012; Mitsuo, 2014).
The discovery of interferometers by Albert Michelson and the Fourier mathematical equation by Lord Rayleigh leads to a massive breakthrough in IR spectroscopy. Known as FTIR (Fourier Transform Infrared) spectrophotometer, it allows the measurement and the calculation of the IR spectrum (Naseska, 2016). It is based on the idea of the radiation interferences between two beams to generate interferograms, a function of the path line change of the two-beam signal derived using the mathematical method of Fourier-transformation (Stuart, 2004). Some significant advantages of using FTIR are (i) a small number of the optical element; (ii) Excellent sensitivity and stability, (iii) fast data collection and near real-time operation, (iv) low maintenance (Doyle, 1992; Naseska, 2016). However, in terms of its operating condition, FTIR spectroscopy has some limitations, including (i) only recognizing heteronuclear gases, (ii) low resolution (1 cm-1) that requires specific laboratory calibration for every single gas, (iii) it is superimposed by the presence of impurities (Smith, 2011; Taquet et al., 2013)
There are generally some significant components of the FTIR that includes; (i) an Infrared beam source, (ii) a Michelson Interferometer consisting of a beam splitter and a compensating plate, and (iii) optical elements. The difference in path length between the two arms is intentionally varied mechanically. It produces the interferogram, which is transmitted optical intensity that rises to a time-dependent. Principally, the spectral point’s power can be identified by bypassing the electrical signal obtained by the infrared radiation detector through the electronic filter. At the same time, the complete spectrum can be determined by applying various filter frequencies. This processing time, nowadays, is shortened by performing a Fourier transformation using a digital computer. Hence, the composite spectrum can be yielded directly from the instrument, the source, and the sample along the optical path (Doyle, 1992). The schematic diagram of the FTIR is provided in Fig 4.
FTIR’s approach relies on the fact that most of the molecules absorb infrared radiation on which the concentration of the sample is proportional to the absorption intensity. Therefore, the sample’s concentration can be determined using pre-constructed calibration curves (Maria, 2012; Frey et al., 2015). The FTIR calibration for gas analysis, mainly CO2 and CH4, has been carried out by (CAILTEAU, 2008; TAQUET, 2012; Frey et al., 2015; Hase et al.
The portable FTIR spectrometers (version alpha) manufactured by Bruker were used in this research project. It would be calibrated to measure the precise and continuous observation of the CO2 and CH4. Besides its portable size, another outstanding factor distinguishes the Alpha Bruker FTIR Spectrometers from other FTIR devices on the market. Instead of using the classical Michelson type of interferometers, the Alpha Bruker FTIR utilizes ROCK SOLID interferometers. The use of a simple Michelson interferometer that depends on the flat mirrors is practically prone to mirror tilt resulting in a decrease in the stability, resolution, and spectral quality. This problem occurred since the light returning to the beam splitter must be accurately recombined. This problem, however, can be resolved by using an approach called dynamic alignment. The ROCK SOLID Interferometer integrates a dual pendulum arrangement in which each consists of retroreflective cube corner mirrors. The mechanism of a wear-free pivot and the comparison between those techniques that are illustrated in Fig 5 prevents mirror tilt and avoid mirror shear.
Fig 5. The ROCKSOLID Interferometer (1. Beam splitter, 2. Cube Corner reflectometer, 3. Wear-free pivot mechanism) [a] The Optical Comparison between the Flat mirror and Retroreflecting Cube Corners. Adapted from Bruker Optik GmbH (no date, p.1).
The FTIR Bruker alpha utilizes a permanently aligned ROCKSOLID interferometer that enables low polarization effects and relatively large energy throughput. Moreover, the spectrometer’s resolution defines the degree of the ability of an instrument to distinguish peaks that are close together. If the peak changes between two spectra are 32 cm-1, the minimum resolution needed is 32 cm-1, which is considered a low solution. The influence of FTIR resolution on spectral is provided in Fig 6.
Fig 6. The comparison between the spectrum of water vapor measured at a resolution of 0.5 cm-1 (top) and the spectrum of water vapor was measured at a resolution of 32 cm-1 (bottom). Adapted from Smith (2011, p.36).
Nonetheless, there is a significant trade-off between resolution and noise level where the high resolution of measurement produces a high level of noise. Smith (2011) has recommended 8 or 4 cm-1 for solid and liquid states and 2 cm-1 or higher for samples in a gas state. Besides, the higher accuracy of the FTIR is needed to show the absorption spectrum because of the rotational vibration, Although FTIR is widely used to analyze organic chemistry qualitatively, the band intensities and path length can also measure the quantity since they are highly correlated to the sample’s concentration through the BEER-Lambert Law be seen in the equation below.
A is the absorbance, ε is the molar absorptivity in L/(mol cm), l is the length of the path (cm), and C is the concentration of the sample moles/L. It is expected that the function will be in linear dependence format, = × + . According to the equation, the number and range of the IR (Infrared) spectrum are highly dependent on the range of the standard gas concentration. If the absorbance of the known standard gas series is measured, a plot of the function of the absorbance against the concentration can be generated using least-square analysis.
Cavity Ring-Down Spectroscopy (CRDS)
Compared with conventional spectroscopy, Cavity Ring-Down Spectroscopy (CRDS) is considered to have higher sensitivity. The measurement is based on the absorption rate instead of the magnitude of absorption on which the light circulates an optical cavity with a high Q factor (Berden, Peeters, and Meijer, 2000; Lehmann, Berden, and Engeln, 2009). It detects even very weak transitions and trace species (Jong, 2012). The other advantages of using CRDS are the very long effective path length that can be done in stable optical cavities. Secondly, the fluctuation of light source intensity has little effect on intrinsic sensitivity. (Berden, Peeters and Meijer, 2000). In the last decade, there is some evidence to suggest that the CRDS approach is potent in gas-phase spectroscopy both for the measurement of trace amount containing strong absorption species and a large concentration of weak absorption species (Crosson, 2008). The Schematic diagram of the CRDS is provided in Fig 7.
The principal mechanism of CRDS starts with a small (35 cc) optical resonator cavity containing the gas sample that receives the light from a tunable semiconductor diode laser. After the laser completes the build-up, the light will circulate in the cavity traveling more than 20 km. There are three mirrors inside the cavity on which the photodetector located behind each mirror identifies the light intensity. A laser pulse is intentionally trapped in an optical cavity wherein the presence of scattering and absorbing compound the intensity of the trapped laser pulse decays exponentially (Jong, 2012). The decay time, also known as cavity ring-down time, is inversely proportional to the losses inside the cavity. The rate of absorption is determined by measuring this cavity ring-down time, resulting in an absolute scale of the losses in which introducing the sample causes a shorter ring-down time due to more extensive cavity loss (Lehmann, Berden, and Engeln, 2009). It is a widely held view that the CRDS can provide high absorption intensity with noise equivalent absorption limits of ~10−9 – 10−11 cm−1 Hz-1, which further allows detection of the gas concentration as low as parts-per-trillion depending on the details of the optical scheme and target species (McHale, Hecobian, and Yalin, 2016).
Gas Chromatography (GC)
Gas Chromatography (GC) is one of the most widely-used analytical methods for identifying and quantity measurement of the organic substances in much-advanced research worldwide (Sparkman, Penton and Kitson, 2011; Mohd, 2012). Chromatography is a separation technique parting the samples’ components into two samples, including a stationary bed with a relatively large surface area and a gas spreading gradually through the stationary bed (McNair and Miller, 2009). Thus, this technique allows the mass spectra of individual compounds obtained for qualitative purposes and quantitative information in the same compound (Sparkman, Penton and Kitson, 2011). There are some majors advantages to using GC analysis on the complex mixture, which includes: (1) excellent resolution shown by symmetric and sharp peaks, (2) high reproducibility and repeatability, (3) High accuracy and high precision, and (4) minimum catalytic and thermal decomposition especially for sensitive sample component (Al-Bukhaiti et al., 2017). The analyzed mixture may be either a solid, a liquid, or a gas in which one of the requirements is that the sample components be stable, have a vapor pressure around 0.1 Torr at operating condition, and interact with column material and the mobile phase (Grob and Barry, 2004). The critical parts of a GC can be seen in Fig 8, which includes a gas source as the mobile phase, an inlet for delivering the sample to the column. The column where separations were taking place; an oven is acting as a thermostat for the column; a detector for registering the presence of chemical data in the column effluent; and a data system to display and proceed the chromatogram (Eiceman, 2006).
Fig 8. The Simplified Block Diagram of Gas Chromatograph. Adapted from Eicemen (2006, p.1)
The time taken from injection to emergence, also known as the Retention Time (Rt), varies between substances under given conditions. It also depends on the volatility of the substance and the column’s temperature, length, and diameter. Some methods for detecting and measuring the individual component have been developed. The most common type of detector is thermal conductivity (TCD). It measures changes in heat conduction and thermal ionization (FID), which measures pyrolysis changes. TCD and FID have approximate limit detection of 10-5 to 10-6 gs-1 and 10-12 gs-1. As for data processing, most modern chromatographic instruments are equipped with integrated hardware and software to estimate peak areas of analyte substances (Al-Bukhaiti et al., 2017).
Laboratory Gases Analysis
The detailed description of gas analyzers, gas standard preparation, and calibration technique is extensively described in this section. The works include leakage tests, cross-validation, and reproducibility test. These metrology protocols were done to allow a monitoring campaign in Montiers-sur-Saulx.
Gas Analyzers and Gas standard Preparation
Commercially available FTIR alpha produced by Bruker optic alpha was first calibrated in the laboratory of Georesources at the University of Lorraine and used throughout the monitoring campaign in Montiers-sur-Saulx. This IR spectrometer is equipped with RockSolidtm interferometer, which is currently considered very stable and permanently aligned. Equipped with a standard KBr beam splitter makes it possible to analyze a sample in the region between 375 to 7500 cm-1. It has a slightly better spectral resolution compared with standard FTIR. Moreover, it has wavenumber accuracy and precision <0.05 cm-1 and <0.0005 cm-1 respectively at 2000 cm-1. The default company setting was used throughout this metrological experiment. The FTIR alpha Bruker’s data output is the IR absorption spectrum transformed into an interferogram spectrum integrating a range of bands specifying the identified gas in a unit area. Furthermore, the illustration of the main metrological experimental apparatus is provided in Fig 9.
Fig 9. Real view and schematic diagram of experimental devices used for CO2 gas calibration: A) schematic diagram of gas circulation, B) realistic view, and C) gas standard mixing diagram.
Some gas analyzers were also employed to cross-validate standard gas and collected gases sample. Therefore, the reproducibility of the gases measurement results can be maintained across the devices, which include a) GC-TCD, b) a low-power NDIR sensor engineered by Solexperts allowing the continuous monitoring of CO2; c) CP-4900 Micro-Gc (Gas Chromatography) manufactured by Varian BV where a six levels calibration was carried out for each gas of interest including Ar, O2, N2, CH4, and CO2 at INRAE laboratory, and d) G2201-i Greenhouse Gas Analyzer (Picarro Inc., Santa Clara, CA, USA).
As shown in Fig 10, GC-TCD and CRDS gas analyzers were pre-calibrated and utilized to cross-validate the gas standard. Single reference gas was supplied by Air Liquide S.A. and certified with a relative uncertainty of 2%. The various mixture of binary gas of CO2, N2, Ar, and CH4 was produced using an on-site customized gas standard preparation system of AlyTech GasMix™by (mixing, diluting, and injecting single references gas standard). It has a dimension of 42×22×46 cm (L×W×H) connected through Swagelok 1/8” with the error of less than 2% of the flow over the scale. The gas standard was controlled in the provided software interface allowing the user to define the gas concentration and flow rate.
The sub-soil gas measurement depends heavily on NDIR and FTIR due to its simplicity and low power consumption. This measurement system is supported by ex-situ gas laboratory analysis using a CRDS analyzer and CP-4900 Micro-GC. CRDS G2201-i measures H2O, CO2, and CH4 concentrations and isotopic ratios of those gases. The default setting of the CRDS G2201-I was used throughout the experiment. It can also measure CO2 and its isotopic composition up to 4000 ppm with measurement intervals of 3 to 5 seconds. To obtain considerable accuracy, the sample pressure that was introduced should be no more than 1.315 atm. Moreover, a more detailed explanation of FTIR and CP-4900 Micro-G’s calibration technique is given in the following subchapter.
Table of contents :
Chapter 1: General Introduction
1.1. Global Carbon cycle
1.2. Scientific Context of the Thesis: DEEPSURF Project
1.3. Aims of the thesis
1.4. Chapters Structure
1.5. CO2 and CH4 Dynamic in Critical Zone
1.5.1. Humidity
1.5.2. Temperature
1.5.3. Soil pH Values
1.5.4. Nutrients
1.5.5. Exposure and Air Pressure
1.5.6. Land Use Changes
1.5.7. Stand and Plant types
1.5.8. Hydrobiogeochemical Processes
1.5.9. Measurement methods for gas exchange
Chapter 2: Metrology
2.1. Fundamental and Instrumental Generalities: Application to Gas Studies
2.1.1. FTIR Spectroscopy (FTIRS)
2.1.2. Cavity Ring-Down Spectroscopy (CRDS)
2.1.3. Gas Chromatography (GC)
2.2. Laboratory Gases Analysis
2.2.1. Gas Analyzers and Gas standard Preparation
2.2.2. Calibration Strategy
2.3. Results
2.3.1. Calibration results of CO2 and CH4
2.3.2. Cross-Validation
2.4. Intermediate Conclusion
Chapter 3: Subsoil CO2 and Micrometeorological Monitoring
3.1. Introduction
3.2. Materials and Methods
3.2.1. Study site description
3.2.2. In-situ Subsoil Gases and Auxiliaries Data Measurements
3.2.3. Micrometeorological Measurement
3.2.4. Principal Component and Wavelet Analysis
3.2.5. In-situ Subsoil Gases Measurement
3.2.6. Seasonal, Daily, and Hourly Variations
3.3. Discussion
3.3.1. CO2 molar fraction dynamics
3.3.2. Analysis of Wind Turbulence and Soil moisture Content
3.3.3. Analysis of Pressure pumping Effect due to Wind Turbulence
3.4. Conclusion
Chapter 4: Long-Term Eddy Covariance Analysis
4.1. Introduction
4.2. Materials and Methods
4.2.1. The Fundamental of Eddy Covariance Technique
4.2.2. Instrumentation and Measurements
4.2.3. Data Processing
4.2.4. The flux footprint
4.2.5. Drought Index and Vegetation State
4.3. Results
4.3.1. Seasonal variations in environmental variables
4.3.2. Variations of Carbon Exchange
4.3.3. Vegetation State
4.3.4. Thinning and Fluxes
4.4. Discussion
4.4.1. Soil and atmospheric dryness across the years
4.4.2. Ecosystem Respiration Responses
4.4.3. Gross Primary Productivity Response
4.5. Intermediate Conclusion
Chapter 5: Subsoil Gases and Eddy Covariance Analysis
5.1. Introduction
5.2. Materials and Methods
5.2.1. CO2 storage calculation
5.2.2. Ecosystem Respiration and Flux Footprint
5.2.3. Soil Gases Sampling and Analysis
5.3. Results
5.3.1. Ecosystem Respiration, CO2 Storage, and Environmental Driver Dynamics
5.3.2. Isotopic Composition
5.3.3. Relationship between soil CO2 and other gas compositions
5.4. Discussion
5.4.1. Change in Soil CO2 storage Due to Subsoil Ventilation
5.4.2. Controlling Processes on Subsoil CO2 in Temperate Forest
5.5. Conclusion
Chapter 6: General Discussion and Conclusion
