The calibration of the intramolecular nitrogen isotope distribution in nitrous oxide measured by isotope ratio mass spectrometry

Two alternative approaches for the calibration of the intramolecular nitrogen isotope distribution in nitrous oxide using isotope ratio mass spectrometry have yielded a difference in the 15N site preference (defined as the difference between the delta15N of the central and end position nitrogen in NNO) of tropospheric N2O of almost 30 per thousand. One approach is based on adding small amounts of labeled 15N2O to the N2O reference gas and tracking the subsequent changes in m/z 30, 31, 44, 45 and 46, and this yields a 15N site preference of 46.3 +/- 1.4 per thousand for tropospheric N2O. The other involves the synthesis of N2O by thermal decomposition of isotopically characterized ammonium nitrate and yields a 15N site preference of 18.7 +/- 2.2 per thousand for tropospheric N2O. Both approaches neglect to fully account for isotope effects associated with the formation of NO+ fragment ions from the different isotopic species of N2O in the ion source of a mass spectrometer. These effects vary with conditions in the ion source and make it impossible to reproduce a calibration based on the addition of isotopically enriched N2O on mass spectrometers with different ion source configurations. These effects have a much smaller impact on the comparison of a laboratory reference gas with N2O synthesized from isotopically characterized ammonium nitrate. This second approach was successfully replicated and leads us to advocate the acceptance of the site preference value 18.7 +/- 2.2 per thousand for tropospheric N2O as the provisional community standard until further independent calibrations are developed and validated. We present a technique for evaluating the isotope effects associated with fragment ion formation and revised equations for converting ion signal ratios into isotopomer ratios.