Calcite-CO2 mixed intoCO2-free air:: a new CO2-in-air stable isotope reference material for the VPDB scale

In order to generate a reliable and long-lasting stable isotope ratio standard for CO2 in samples of clean air, CO2 is liberated from well-characterized carbonate material and mixed with CO2-free air. For this purpose a dedicated acid reaction and air mixing system (ARAMIS) was designed. In the system, CO2 is generated by a conventional acid digestion of powdered carbonate. Evolved CO2 gas is mixed and equilibrated with a prefabricated gas comprised of N-2, O-2, Ar, and N2O at close to ambient air concentrations. Distribution into glass flasks is made stepwise in a highly controlled fashion. The isotopic composition, established on automated extraction/measurement systems, varied within very small margins of error appropriate for high-precision air-CO2 work (about +/- 0.015 parts per thousand for delta(13) C and +/- 0.025 parts per thousand for delta O-18). To establish a valid delta(18) O relation to the VPDB scale, the temperature dependence of the reaction between 25 and 47 degrees C has been determined with a high level of precision. Using identical procedures, CO2-in-air mixtures were generated from a selection of reference materials; (1) the material defining the VPDB isotope scale (NBS 19, delta C-13 = +1.95 parts per thousand and delta O-18=-2.2 parts per thousand exactly); (2) a local calcite similar in isotopic composition to NBS 19 ('MAR-J1', delta C-13 - +1.97 parts per thousand and delta O-18 = -2.02 parts per thousand). and (3) a natural calcite with isotopic compositions closer to atmospheric values ('OMC-J1', delta C-13 = -4.24 parts per thousand and delta O-18 = -8.71 parts per thousand). To quantitatively control the extent of isotope-scale contraction in the system during mass spectrometric measurement other available international and local carbonate reference materials (L-SVEC, IAEA-CO-1, IAEA-CO-8, CAL-1 and CAL-2) were also processed. As a further control pure CO2 reference gases (Narcis I and II, NIST-RM 8563, GS19 and GS20) were mixed with CO2-free synthetic air. Independently, the pure CO2 gases were measured on the dual inlet systems of the same mass spectrometers. The isotopic record of a large number of independent batches prepared over the course of several months is presented. In addition, the relationship with other implementations of the VPDB-scale for CO2-in-air (e.g. CG-99, based on calibration of pure CO2 gas) has been carefully established. The systematic high-precision comparison of secondary carbonate and CO2 reference materials covering a wide range in isotopic composition revealed that assigned delta-values may be (slightly) in error. Measurements in this work deviate systematically from assigned values, roughly scaling with isotopic distance from NBS 19. This finding indicates that a scale contraction effect could have biased the consensus results. The observation also underlines the importance of cross-contamination errors for high-precision isotope ratio measurements. As a result of the experiments, a new standard reference material (SRM), which consists of two 5-L, glass flasks containing air at 1.6 bar and the CO2 evolved from two different carbonate materials, is available for distribution. These 'J-RAS' SRM flasks ('Jena-Reference Air Set') are designed to serve as a high-precision link to VPDB for improving inter-laboratory comparability. (a) Copyrght (c) 2005 John Wiley & Sons, Ltd.