CO2 isotope analyses using large air samples collected on intercontinental flights by the CARIBIC Boeing 767.

Analytical details for 13C and 18O isotope analyses of atmospheric CO2 in large air samples are given. The large air samples of nominally 300 L were collected during the passenger aircraft-based atmospheric chemistry research project CARIBIC and analyzed for a large number of trace gases and isotopic composition. In the laboratory, an ultra-pure and high efficiency extraction system and high-quality isotope ratio mass spectrometry were used. Because direct comparison with other laboratories was practically impossible, the extraction and measurement procedures were tested in considerable detail. Extracted CO2 was measured twice vs. two different working reference CO2 gases of different isotopic composition. The two data sets agree well and their distributions can be used to evaluate analytical errors due to isotope measurement, ion corrections, internal calibration consistency, etc. The calibration itself is based on NBS-19 and also verified using isotope analyses on pure CO2 gases (NIST Reference Materials (RMs) and NARCIS CO2 gases). The major problem encountered could be attributed to CO2-water exchange in the air sampling cylinders. This exchange decreased over the years. To exclude artefacts due to such isotopic exchange, the data were filtered to reject negative delta18O(CO2) values. Examples of the results are given.

On the N2O correction used for mass spectrometric analysis of atmospheric CO2.

To obtain accurate values of delta(13)C(CO(2)) and delta(18)O(CO(2)) on environmental CO(2) by mass spectrometry, the raw isotope data must be corrected for the isobaric N(2)O contribution. This is one of the analytical problems limiting inter-laboratory delta(13)C(CO(2)) data consistency. The key parameter, the N(2)O relative ionisation efficiency (E(N2O)), cannot be determined with sufficient accuracy by direct measurements of pure N(2)O. The determination of (E(N2O)) by analyses on N(2)O--CO(2) mixtures of known isotope composition and mixing proportions has been recently suggested. In this work we propose a new method of N(2)O correction which uses the m/z 30 signal as a measure of the N(2)O/CO(2) ratio, so that determinations of (E(N2O)) and N(2)O content are not required. The method uses the fact that fragment-ion spectra of N(2)O and CO(2) are very specific. The formalism of the correction is considered. Various tests demonstrate that the new method is robust, stable and easy to implement in practice. The effective value (E(N2O)) (the key parameter for the new correction) has to be calibrated on known N(2)O--CO(2) mixtures by measuring (30)R signals only. The method accuracy we presently achieved is around 2.5% and any error which appears to come mostly from our N(2)O--CO(2) mixture preparation. Based on our tests and error considerations, the error of the proposed method that may be achieved is as low as +/-1.5% (relative to the correction magnitude). For tropospheric CO(2) this means +/-0.003 per thousand and +/-0.005 per thousand for delta(13)C(CO(2)) and delta(18)O(CO(2)), respectively. The proposed method may be valuable for small samples where no separate N(2)O determinations are available (e.g. ice core samples and CF-IRMS measurements) as well as for determination of (E(N2O)) and testing the 'traditional' N(2)O correction based on mass balance calculations.

Greenhouse gases: low methane leakage from gas pipelines.

Using natural gas for fuel releases less carbon dioxide per unit of energy produced than burning oil or coal, but its production and transport are accompanied by emissions of methane, which is a much more potent greenhouse gas than carbon dioxide in the short term. This calls into question whether climate forcing could be reduced by switching from coal and oil to natural gas. We have made measurements in Russia along the world's largest gas-transport system and find that methane leakage is in the region of 1.4%, which is considerably less than expected and comparable to that from systems in the United States. Our calculations indicate that using natural gas in preference to other fossil fuels could be useful in the short term for mitigating climate change.

Reporting small Delta 17O values: existing definitions and concepts.

The three-isotope tracer Delta(17)O is increasingly used in atmospheric chemistry and other research areas. Thanks to the development of isotope-ratio mass spectrometry (IRMS), delta(17)O and delta(18)O can be determined with a precision of a few 0.01 per thousand, and values for Delta(17)O may be calculated with similar precision. However, interpreting small and precisely determined Delta(17)O values as a deviation from an expected mass-dependent fractionation process is not straightforward. Several aspects are of high importance. In the present paper we review existing definitions, formulas and some other aspects of Delta(17)O reporting. One of the most confusing aspects is a variance of definitions and corresponding formulas. While Delta(17)O is traditionally defined to characterise a data point, i.e. Delta(17)O is considered as a deviation from an expected mass-fractionation line, the recently introduced definition (Miller MF. Geochim. Cosmochim. Acta 2002; 66: 188) characterises a fractionation line itself, in terms of its ordinate intercept. The formulas corresponding to this definition gives a characteristic for a specific process. When the 'traditionally defined' Delta(17)O is in use, an expected fractionation processes--the key point for Delta(17)O reporting--should be defined and parameterised with the same accuracy as intended for reporting Delta(17)O. When Delta(17)O is reported for a data point, not only a value for lambda but an ordinate intercept of a reference fractionation line should be given with high accuracy. We note that defining a single fractionation process is hardly possible for many natural compounds. For such compounds we propose to use a phenomenological reference line, namely an isotope composition range of natural sources. Next, aspects of Delta(17)O comparison and mass-balance calculations are considered. All the aspects considered for Delta(17)O may be relevant for others three-isotope tracers, e.g. Delta(33)S.

Isotope analysis of hydrocarbons: trapping, recovering and archiving hydrocarbons and halocarbons separated from ambient air.

It is argued that isotope analysis of atmospheric non-methane hydrocarbons (NMHCs) and, in particular, the analysis of the deuterium/hydrogen (D/H) ratio is valuable because the dominant self-cleansing property of the troposphere is based on the OH radical which removes, e.g., CH4 and other alkanes by H-atom abstraction, which induces large kinetic isotope effects. The major obstacle in applying D/H isotope analysis to atmospheric NMHCs is not only the low abundance of D itself but, in particular, the low concentrations of NMHCs in the parts per trillion range. We show how a selection of NMHCs can be quantitatively separated from 300 L air samples together with CO2 as carrier gas matrix, by using high efficiency cryogenic traps. After diluting the extracted NMHC mixtures with hydrocarbon free air, and determining the mixing ratios, good agreement with original whole air sample analysis exists for alkanes and several halocarbons. For unsaturated hydrocarbons and some other halocarbons the extraction and recovery yield under the given conditions fell considerably, as a function of boiling point. Furthermore, the mixture of NMHCs in the CO2 matrix is proven to remain unchanged over several years when conveniently stored in glass ampoules. The 'extracts' or 'concentrates' of condensables extracted from larger air samples will enable the D/H isotope analysis of ultra trace gases in the atmosphere.

A new method to determine the 17O isotopic abundance in CO2 using oxygen isotope exchange with a solid oxide.

This paper discusses a simple method to determine 17O isotope excess or deficiency ('mass-independent isotopic composition') in CO2 gas. When applying conventional mass spectrometry of CO2 (m/z 44, 45 and 46) to determine the 17O/16O ratio, the 13C/12C ratio has to be established separately. This can be achieved by analysing an aliquot of sample CO2 before and after subjecting it to oxygen isotope exchange with a pool of oxygen with 'normal' 17O/16O ratio, i.e. with Delta17O approximately equal to delta17O-0.516 x delta18O = 0. Cerium oxide has been shown to be practically well suited for the exchange of CO2 oxygen; the reagent is safe and does not produce any contamination. The CO2-CeO2 exchange reaction has 99.8 +/- 0.7% recovery yield. At 650 degrees C this reaction reaches equilibrium in 30 min and, as tested, results in complete oxygen replacement. Delta17O determinations depend on accuracy of CO2 delta measurements: the repeatability of +/-0.015 per thousand (1sigma) in delta(45)R and delta(46)R determination relative to the working reference results in an error of Delta17O as small as +/-0.33 per thousand. Such a precision is sufficient for Delta17O determination in stratospheric CO2. The calculated Delta17O value systematically depends on absolute 17R and 13R ratios in isotopic reference materials, which are presently not yet known with certainty (the 17R value is most important), and may be inadequate for 17O-correction with a = 0.516. Within the present uncertainty, Delta17O determined in 17O-enriched CO2 agrees with the value directly measured in the enriched O2 from which this CO2 was produced. Besides Delta17O determination, investigated CO2-CeO2 equilibration may have several other implications. Fast, complete isotopic exchange of CO2 by reaction with CeO2 may also be employed to get reproducible 17O-correction and, hence, to better monitor small delta13C shifts and to isotopically equilibrate mixtures of CO2 gases.