Monica M. Arienzo

Measurement of δ18O and δ2H values of fluid inclusion water in speleothems using cavity ring-down spectroscopy compared with isotope ratio mass spectrometry

RATIONALE The hydrogen and oxygen isotopic analyses (δ(2)H and δ(18)O values) of water trapped within speleothem carbonate (fluid inclusions) have traditionally been conducted utilizing dual-inlet isotope ratio mass spectrometry (IRMS) or continuous-flow (CF)-IRMS methods. The application of cavity ring-down spectroscopy (CRDS) to the δ(2)H and δ(18)O analysis of water in fluid inclusions has been investigated at the University of Miami as an alternative method to CF-IRMS. METHODS An extraction line was developed to recover water from the fluid inclusions consisting of a crusher, sample injection port and an expansion volume (either 100 or 50 cm(3)) directly connected to the CRDS instrument. Tests were conducted to determine the reproducibility of standard water injections and crushes. In order to compare results with conventional analytical methods, samples were analyzed both at the University of Miami (CRDS method) and at the Vrije Universiteit Amsterdam (CF-IRMS method). RESULTS The analytical reproducibility of speleothem samples crushed on the Miami Device demonstrates an average external standard deviation of 0.5 and 2.0 ‰ for δ(18)O and δ(2)H values, respectively. Sample data are shown to fall near the global meteoric water line, supporting the validity of the method. Three different samples were analyzed at Vrije Universiteit Amsterdam and the University of Miami in order to compare the performance of each laboratory. The average offset between the two laboratories is 0.7 ‰ for δ(18)O and 2.5 ‰ for δ(2)H. CONCLUSIONS The advantage of CRDS is that the system is a low-cost alternative to CF-IRMS for fluid inclusion isotope analysis. The CRDS method demonstrates acceptable precision and good agreement with results from the CF-IRMS method. These are promising results for the future application of CRDS to fluid inclusion isotope analysis.