Quantifying and empirically correcting apparent gas matrix effects:Example measurements for two CRDS analyzers for CO2 and CH4 amount fractions and 13/12C isotope ratios

Accurate measurements of amount fractions and isotopic compositions of greenhouse gas such as carbon dioxide (CO₂) and methane (CH₄) provide valuable insights on their atmospheric composition and origin. Commonly used field deployable commercial laser spectrometers that measure amount fractions and isotopic ratios are often calibrated with reference gases with certified amount fractions and/or isotopic composition. Reference gases, also known as calibration reference materials (CRMs), can be for example synthetic mixtures of e.g. CO₂ in N₂, where the gas matrix N₂ does not match that of the sample (e.g. ambient air) to be measured. A mismatch in the composition of the gas matrix of a CRM and sample can lead to a considerable bias in the amount fraction or isotopic ratio results of the sample due to changes in the measured spectra which e.g. are not perfectly captured by the analysers’ fitting routine.

In this work, we demonstrate the quantification of matrix effects for two commercial CRDS analysers measuring CO₂ and CH₄ amount fractions and isotope ratios. In our experiments with synthetic air gas matrix where the O₂ concentration was varied, we measured (for a 1 % change in the O₂ concentration in the gas matrix) a relative change of 0.15 % for the amount fractions of two major CO₂ isotopologues and 0.07 % for the amount fractions of two major CH₄ isotopologues. Similarly, in terms of isotopic δ¹³C values, we found matrix effects larger than 0.2 for both CO₂ and CH₄. We present options for correcting the gas matrix effects and discuss the underlying assumptions made during the analysis. Amount fraction results for CO₂ and CH₄ are reported including δ¹³C isotope ratio results. Our work concludes that a matrix mismatch when using a commercial laser spectrometer can lead to considerable biases in amount fraction and isotope ratio results, and appropriate correction approaches have to be applied in order to achieve accurate and reliable results.

Acknowledgements: This work has received partial funding from the EMPIR programme (19ENV05 STELLAR project) co-financed by the Participating States and from the European Union's Horizon 2020 research and innovation programme. Part of this work has also received funding from the European Partnership on Metrology (21GRD04 isoMET project), co-financed from the European Union’s Horizon Europe Research and Innovation Programme and by the Participating States.