Real-time analysis of d13C- and dD-CH4 in ambient air with laser spectroscopy: Method development, validation and applications

Eyer, Simon (2016). Real-time analysis of d13C- and dD-CH4 in ambient air with laser spectroscopy: Method development, validation and applications (Unpublished). (Dissertation, Empa; Universität Bern, Philosophisch–naturwissenschaftliche Fakultät, Labor für Umwelttechnik und Luftfremdstoffe)

Text eyer16phd.pdf - Other Restricted to registered users only Available under License BORIS Standard License. Download (7MB)

About 20 % of the increase in radiative forcing due to human emitted greenhouse gases is attributed to CH4, and thus, reducing CH4 emissions offers a great short-term target in mitigating global warming. However, understanding what controls the global CH4 cycle remains a challenge, as many CH4 sources vary strongly on spatial and temporal scales. Therefore, measuring the isotopic composition of CH4 can provide valuable information to reduce uncertainties in the CH4 emission budget. Up till now, however, high precision measurements of CH4 isotopologues are scarcely available as they are limited to isotope ratio mass spectrometry (IRMS), which is generally a laboratory based technique.
This PhD thesis is about the development, validation and application of a field-deployable, fully automated, liquid nitrogen free measurement platform, called TREX-QCLAS, capable of performing real-time, high precision analysis of ambient air 12CH4, 13CH4 and 12CH3D mole fractions and their relative isotope ratios δ 13C- and δ D-CH4.
Chapter 1 provides an introduction to the topic, presenting the relevance of CH4 as a greenhouse gas, the value of isotopic information to pinpoint CH4 emission sources and the basic principles of mid infrared laser spectroscopy for high precision atmospheric trace gas and isotope measurements.
Chapter 2 is an in-depth study comparing six different adsorbent materials with respect to their effectiveness for CH4 preconcentration from ambient air. After an investigation of adsorption/desorption isotherms, HayeSep D was found to exhibit an outstanding selectivity for CH4 over N2.
Chapter 3 describes the TREX-QCLAS technique, which consists of two components: a fully automated preconcentration unit, called TRace gas EXtractor (TREX) and a high-precision dual-quantum cascade laser absorption spectrometer (QCLAS). The TREX system enhances the mole fraction of the target gas CH4 by a factor of up to 500 while simultaneously removing other trace gases such as carbon dioxide, nitrous oxide and water, which would otherwise interfere during spectral analysis. The QCLAS is capable of analyzing δ 13C- and δ D-CH4 values with a precision of 0.1 ‰ and 0.5 ‰, respectively, at 10 min averaging time. Chapter 3 also presents a two-week measurement campaign at the Empa Campus in Dübendorf, Switzerland, during which the newly developed TREX-QCLAS system was validated against flask samples analyzed by IRMS.
Chapter 4 presents results of a 5-month measurement campaign on CH4 isotopologues at the CESAR-tower in Cabauw, the Netherlands, where the TREX-QCLAS system was operated along with a dual-IRMS system of the University of Utrecht. Based on replicate measurements of compressed air, the repeatability of the TREX-QCLAS was determined to be 0.17 ‰ and 0.85 ‰ for δ 13C- and δ D-CH4, respectively, making it competitive with IRMS technique. Despite the dominant contribution of isotopically depleted ruminant emissions, events with elevated fossil or landfill contributions were distinguishable, demonstrating the value of in situ and high temporal resolution isotope ratio measurements. Comparison with atmospheric transport model simulations confirmed the temporal trends of different source contributions, but also indicated a possible overestimation of fossil fuel emissions by the EDGAR inventory. In summary, the measurement campaign demonstrates the feasibility of high precision long-term measurements of CH4 isotopologues and their capability in constraining emission inventories as well as verifying transport model predictions.

Interest & Impact

Downloads

1 since deposited on 22 Feb 2024
1 in the past 12 months

Citations

5 Citations in Web of Science ®
6 Citations in Scopus

Search

in Google Scholar™

Services

Actions (login required)

Edit item

Actions (login required)

Edit item