Abstract

Compound-specific oxygen stable isotope analysis has been commercially available for about 20 years but the technique has been infrequently applied and practical operating knowledge is limited. Existing reactor designs and chemistry fail to fully exclude contaminating oxygen, are chromatographically flawed, and are generally fabricated from expensive platinum tubing. As a new interface between the gas chromatograph and isotope ratio mass spectrometer, we used a nickel tube reactor (containing no platinum wires) in a ceramic sheath with countercurrent helium flushing, operated at 1250°C. We analyzed suites of four long-chain alcohols (LCAs), defining the least and most 18 O-enriched LCAs as internal standards. The other two LCAs, co-injected in varying amounts, were treated as unknowns and amplitude-corrected during data reduction. With this design, signals from non-oxygenated compounds are minimized with a hydrogen trickle, while excessive graphitization is counterproductive. A balance of hydrogen and graphite is required to maintain a proper, stable redox state in the reactor. With low extraneous oxygen, the exact hydrogen flow has little effect on final δ18 O data quality. Overall average δ18 O values precisions and errors for the unknowns were better than 0.4‰ for injections between 2 and 7nmol per component. Our actively flushed sheath reduces oxygen infiltration to negligible levels and results in a chromatographic flow devoid of dead or poorly swept volumes. The improved design allows measurement of LCAs with accuracy and precision sufficient to enable investigating real processes, and may also have wider applications to other isotopes.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call