Oxygen isotope analysis of levoglucosan, a tracer of wood burning, in experimental and ambient aerosol samples. (6th November 2017)

• Record Type: Journal Article
• Authors: Blees, Jan; Saurer, Matthias; Siegwolf, Rolf T.W.; Ulevicius, Vidmantas; Prevôt, André S.H.; Dommen, Josef; Lehmann, Marco M.
• Is Part Of: Rapid communications in mass spectrometry. Volume 31:Number 24(2017)
• Page Start: 2101
• Page End: 2108
• Publication Date: 2017-11-06
• Subjects: Mass spectrometry -- Periodicals; 543.65
• DOI: 10.1002/rcm.8005
• Languages: English
• ISSNs: 0951-4198

Abstract

Abstract : Rationale: Levoglucosan is formed from cellulose during biomass burning. It is therefore often used as a specific tracer to quantify the contribution of wood burning to the aerosol loading. The stable oxygen isotope composition (δ 18 O value) of biomass is determined by the water cycle and varies regionally, and hence the δ 18 O value of levoglucosan could help to identify source regions of organic aerosols. Methods: After solvent extraction of the organic fraction and concentration steps, a recently developed methylation derivatisation technique was applied on experimental (i.e. controlled wood‐burning experiments) and on ambient aerosol samples from Switzerland and Lithuania. The method achieves sufficient compound separation for isotope analysis in atmospheric particulate matter, enabling δ 18 O analysis of levoglucosan by gas chromatography/pyrolysis–isotope ratio mass spectrometry (GC/Pyr‐IRMS), with a precision better than 1.0 ‰ and an accuracy of 0.3 ‰. Results: The δ 18 O value of the levoglucosan released during controlled wood‐burning experiments was not significantly different from the cellulose δ 18 O values, which implies very little or no isotope fractionation during wood burning under the given conditions. While the δ 18 O values of levoglucosan in Swiss samples were as expected for the source region, those in Lithuania were 1–4 ‰ lower than expected. This may be due to differences in vegetation (grass vs wood) or burning conditions (high vs low temperatures). Conclusions: Low oxygen isotope fractionation between cellulose and levoglucosan and clear differences in levoglucosan δ 18 O values between the Swiss and Lithuanian ambient samples demonstrate that our new method is useful for source appointment studies on wood‐burning‐derived aerosols.