B51A-0400
A new derivatization method for δ18O analysis of individual carbohydrates with GC-Pyrolysis-IRMS

Friday, 18 December 2015
Poster Hall (Moscone South)
Marco Marius Lehmann1, Rolf T Siegwolf1, Matthias Saurer1, Jan Blees1, Maria Fischer2 and Michael Zech3, (1)Paul Scherrer Institute, Laboratory of Atmospheric Chemistry, 5232 Villigen, Switzerland, (2)Empa, Laboratory of Nanoscale Materials Science, 8600 Dübendorf, Switzerland, (3)Martin-Luther-University Halle Wittenberg, 3Department of Soil Biogeochemistry, 06120 Halle, Germany
Abstract:
Compound specific isotope analysis (CSIA) with gas chromatography coupled to an isotope ratio mass spectrometer (GC-Pyr-IRMS) is nowadays a powerful tool that is widely used by a broad spectrum of research fields to investigate the isotopic signature of diverse metabolites. While many CSIA methods for carbon, hydrogen, and nitrogen isotopes are known, CSIA methods for the analysis of oxygen isotopes (δ18O) are still not widely established. Especially, reliable and precise methods for the δ18O analyses of individual carbohydrates are scarce, which is caused by the highly sensitive nature of the sugars. However, carbohydrates are important components of living organisms, source for many biochemical reactions, and can be found in all organisms, in soils, sediments, and in air. Thus, a method, allowing the investigation of the 18O/16O ratio in carbohydrates will enhance the scope of research using isotopes. We developed a new and easy to handle derivatization method to determine δ18O in carbohydrates with GC-Pyr-IRMS that consists of a catalyzed one-pot reaction in acetonitrile, resulting in complete methylation of all sugar hydroxyl groups within 24 hours, with silver oxide as the proton acceptor and methyl iodide as the methyl group carrier. Results derived from standard material show unrivalled δ18O precision ranging from about 0.2 to 1.1 ‰ for different individual carbohydrates of different classes and a generally very good accuracy, with a narrow range of 0.2 ‰ around the reference value, despite of high area variations. We applied this method on real samples, demonstrating that the method can commonly be used for analyzing honey samples, and for the analyses of more complex carbohydrate mixtures from plant leaves, including glucose, fructose, pinitol, and sucrose. Our new method may be used for food, beverage, and medical applications, as well as for biogeochemical and paleoclimatic sciences.