Offline oxygen isotope analysis of organic compounds with high N:O

Although the advantages of online δ¹⁸O analysis of organic compounds make its broad application desirable, researchers have encountered NO⁺ isobaric interference with CO⁺ at m/z 30 (e.g. ¹⁴N¹⁶O⁺, ¹²C¹⁸O⁺) when analyzing nitrogenous substrates. If the δ¹⁸O value of inter‐laboratory standards for substrates with high N:O value could be confirmed offline, these materials could be analyzed periodically and used to evaluate δ¹⁸O data produced online for nitrogenous unknowns. To this end, we present an offline method based on modifications of the methods of Schimmelmann and Deniro (Anal. Chem. 1985; 57: 2644) and Sauer and Sternberg (Anal. Chem. 1994; 66: 2409), whereby all the N₂ from the gas products of a chlorinated pyrolysis was eliminated, resulting in purified CO₂ for analysis via a dual‐inlet isotope ratio mass spectrometry system. We evaluated our method by comparing observed δ¹⁸O values with previously published or inter‐laboratory calibrated δ¹⁸O values for five nitrogen‐free working reference materials; finding isotopic agreement to within ±0.2‰ for SIGMA® cellulose, IAEA‐CH3 cellulose (C₆H₁₀O₅) and IAEA‐CH6 sucrose (C₁₂H₂₂O₁₁), and within ±1.8‰ for IAEA‐601 and IAEA‐602 benzoic acids (C₇H₆O₂). We also compared the δ¹⁸O values of IAEA‐CH3 cellulose and IAEA‐CH6 sucrose that was nitrogen‐'doped' with adenine (C₅H₅N₅), imidazole (C₃H₄N₂) and 2‐aminopyrimidine (C₄H₅N₃) with the undoped δ¹⁸O values for the same substrates; yielding isotopic agreement to within ±0.7‰. Finally, we provide an independent analysis of the δ¹⁸O value of IAEA‐600 caffeine (C₈H₁₀N₄O₂), previously characterized using online systems exclusively, and discuss the reasons for an average 1.4‰ enrichment in δ¹⁸O observed offline relative to the consensus online δ¹⁸O value. Copyright © 2010 John Wiley & Sons, Ltd.