Climate reconstructions using stable isotopes from tree-rings are steadily increasing. The investigations concentrate mostly on cellulose due to its high stability. In recent years the available amount of cellulose has steadily decreased, mainly because micro-structures of plant material have had to be analyzed. Today, the amounts of cellulose being studied are frequently in the milligram and often in the microgram range. Consequently, homogeneity problems with regard to the stable isotopes of carbon and oxygen from cellulose have occurred and these have called for new methods in the preparation of cellulose for reliable isotope analyses. Three different methods were tested for preparing isotopically homogenous cellulose, namely mechanical grinding, freezing by liquid nitrogen with subsequent milling and ultrasonic breaking of cellulose fibres. The best precision of isotope data was achieved by freeze-milling and ultrasonic breaking. However, equipment for freeze-milling is expensive and the procedure is labour-intensive. Mechanical grinding resulted in a rather high loss of material and it is also labour-intensive. The use of ultrasound for breaking cellulose fibres proved to be the best method in terms of rapidity of sample throughput, avoidance of sample loss, precision of isotope results, ease of handling, and cost.

A new method for stable carbon isotope ratio analysis of anhydrosugars from biomass burning aerosol particle source filter samples was developed by employing Thermal Desorption--2 Dimensional Gas Chromatography--Isotope Ratio Mass Spectrometry (TD-2DGC-IRMS). Compound specific isotopic measurements of levoglucosan, mannosan, and galactosan performed by TD-2DGC-IRMS in a standard mixture show good agreement with isotopic measurements of the bulk anhydrosugars, carried out by Elemental Analyzer--Isotope Ratio Mass Spectrometry (EA-IRMS). The established method was applied to determine the isotope ratios of levoglucosan, mannosan, and galactosan from source samples collected during combustion of hard wood, softwood, and crop residues. δ(13)C values of levoglucosan were found to vary between -25.6 and -22.2‰, being higher in the case of softwood. Mannosan and galactosan were detected only in the softwood samples showing isotope ratios of -23.5‰ (mannosan) and -25.7‰ (galactosan). The isotopic composition of holocellulose in the plant material used for combustion experiments was determined with δ(13)C values between -28.5 and -23.7‰. The difference in δ(13)C of levoglucosan in biomass burning aerosol particles compared to the parent fuel holocellulose was found to be -1.89 (±0.37)‰ for the investigated biomass fuels. Compound specific δ(13)C measurements of anhydrosugars should contribute to an improved source apportionment.

The chemical stability of levoglucosan was studied by exploring its isotopic fractionation during the oxidation by hydroxyl radicals. Aqueous solutions as well as mixed (NH4)2SO4‐levoglucosan particles were exposed to OH. In both cases, samples experiencing different extents of processing were isotopically analyzed by Thermal Desorption‐Gas Chromatography‐Isotope Ratio Mass Spectrometry (TD‐GC‐IRMS). From the dependence of levoglucosan δ13C and concentration on the reaction extent, the kinetic isotope effect (KIE) of the OH oxidation reactions was determined to be 1.00187±0.00027 and 1.00229±0.00018, respectively. Both show good agreement within the uncertainty range. For the heterogeneous oxidation of particulate levoglucosan by gas‐phase OH, a reaction rate constant of (2.67±0.03)·10−12 cm3 molecule−1S−1 was derived. The laboratory kinetic data, together with isotopic source and ambient observations, give information on the extent of aerosol chemical processing in the atmosphere.

Abstract. In many environments organic matter significantly contributes to the composition of atmospheric aerosol particles influencing its properties. Detailed chemical characterization of ambient aerosols is critical in order to understand the formation process, composition, and properties of aerosols and facilitates source identification and relative contributions from different types of sources to ambient aerosols in the atmosphere. However, current analytical methods are far from full speciation of organic aerosols and often require sampling times of up to one week. Offline methods are also subjected to artifacts during aerosol collection and storage.In the present work a new technique for quasi on-line compound specific measurements of organic aerosol particles was developed. The Aerosol Collection Module (ACM) focuses particles into a beam which is directed to a cooled sampling surface. The sampling takes place in a high vacuum environment where the gas phase from the sample volume is removed. After collection is completed volatile and semi-volatile compounds are evaporated from the collection surface through heating and transferred to a detector.For laboratory characterization the ACM was interfaced with a Gas Chromatograph Mass Spectrometer, Flame Ionization Detector system (GC/MS-FID), abbreviated as ACM GC-MS. The particle collection efficiency, gas phase transfer efficiency, and linearity of the ACM GC-MS were determined using laboratory generated octadecane aerosols. The ACM GC-MS is linear over the investigated mass range of Correspondence to: A. Kiendler-Scharr (a.kiendler-scharr@fz-juelich.de) 10 to 100 ng and a recovery rate of 100% was found for octadecane particles.The ACM GC-MS was applied to investigate secondary organic aerosol (SOA) formed from β-pinene oxidation.Nopinone, myrtanal, myrtenol, 1-hydroxynopinone, 3-oxonopinone, 3,7-dihydroxynopinone, and bicyclo[3,1,1]hept-3-ene-2-one were found as products in the SOA. The ACM GC-MS results are compared to quartz filter samples taken in parallel to the ACM GC-MS measurements. First measurements of ambient atmospheric aerosols are presented.

Compound specific carbon isotopic measurements (δC) of levoglucosan were carried out for ambient aerosol sampled during an intensive biomass burning period at different sites in Guangdong province, China. The δC of ambient levoglucosan was found to be noticeably heavier than the average δC of levoglucosan found in source C3-plant-combustion samples. To estimate the photochemical age of sampled ambient levoglucosan, back trajectory analyses were done. The origin and pathways of the probed air masses were determined, using the Lagrangian-particle-dispersion-model FLEXPART and ECMWF meteorological data. On the other hand, the isotopic hydrocarbon clock concept was applied to relate the changes in the field-measured stable carbon isotopic composition to the extent of chemical processing during transport. Comparison of the photochemical age derived using these two independent approaches shows on average good agreement, despite a substantial scatter of the individual data pairs. These analyses demonstrate that the degree of oxidative aging of particulate levoglucosan can be quantified by combining laboratory KIE studies, observed δC at the source and in the field, as well as back trajectory analyses. In this study, the chemical loss of levoglucosan was found to exceed 50% in one-fifth of the analyzed samples. Consequently, the use of levoglucosan as a stable molecular tracer may underestimate the contribution of biomass burning to air pollution.