Simplified method for microlitre deuterium measurements in water and urine by gas chromatography–high‐temperature conversion–isotope ratio mass spectrometry

• Published in: Rapid Communications in Mass Spectrometry
• Main Authors: Gucciardi, Antonina, Cogo, Paola E., Traldi, Umberto, Eaton, Simon, Darch, Tegan, Simonato, Manuela, Ori, Carlo, Carnielli, Virgilio P.
• Format: Article in Journal/Newspaper
• Language: English
• Published: Wiley 2008
• Subjects: Antarctic; Greenland; Antarc*; GISP; Ice Sheet
• Online Access: http://dx.doi.org/10.1002/rcm.3593; https://api.wiley.com/onlinelibrary/tdm/v1/articles/10.1002%2Frcm.3593; http://onlinelibrary.wiley.com/wol1/doi/10.1002/rcm.3593/fullpdf

Abstract Deuterium ( 2 H) in water and urine can be measured by off‐line and, more recently, on‐line techniques using isotope ratio mass spectrometry (IRMS). We describe a new simple on‐line pyrolysis method for the analysis of 2 H/ 1 H in water and urine samples by continuous flow IRMS, normally used for 2 H/ 1 H measurements in organic compounds. A deactivated column connected the split injector to a high‐temperature conversion reactor (TC HD), and 0.5 µL of sample was injected. Accuracy and precision were determined with Vienna Standard Mean Ocean Water (VSMOW), Standard Light Antarctic Precipitation (SLAP), and Greenland Ice Sheet Precipitation (GISP). The range of linearity was measured with a calibration curve of enriched water from 0 up to 0.1 atom percent excess (APE) (i.e. −72 up to 6323 delta per mil ( δ D ‰)) with a precision of <5‰ and accuracy ranging between 1 and 55‰. Blinded reanalysis of urine samples by an equilibration device (Gas Bench) and by a dedicated pyrolysis system (TC/EA) was performed and results compared by the Bland‐Altman test. Enrichments ranged between 600 and 2400‰ δD VSMOW with a precision of ±5‰. Urine enrichments described by our method were strongly correlated with values obtained by Gas Bench and TC/EA ( p < 0.0001). There was a significant memory effect that was reduced by injecting the sample 15 times and discarding the first 10 injections, together with accurate furnace conditioning and appropriate cleaning of the syringe. Data indicate that the method is accurate, and that it can be used for water and urine deuterium determination when a Gas Bench or TC/EA instrument is not available and the amount of sample is limited. Copyright © 2008 John Wiley & Sons, Ltd.