PRECISION AND REPRODUCIBILITY OF IN-SITU OXYGEN-ISOTOPE RATIO MEASUREMENTS ON QUARTZ OBTAINED USING AN ISOLAB-54 ION MICROPROBE

Ion probe techniques for measuring in situ O-18/O-16 ratios from insul ating materials (particularly quartz) have been developed using an Iso lab 54 ion microprobe. O-18/O-16 ratio measurements were obtained from a range of quartz standards for which O-18/O-16(SMOW) ratios had been determined by conventional fluorination techniques. The correlation b etween the ratios obtained by ion probe and by conventional fluorinati on methods is high. Accurate absolute O-18/O-16(SMOW) ratios can be ob tained by normalisation to a well-characterised standard and the Is de viation between the ion probe ratio and the conventionally determined ratio is 1.5 parts per thousand for a single measurement on the standa rd and a single measurement on the unknown sample. This measured scatt er in normalised ratios is consistent with the typical measured uncert ainty on a spot measurement being of the order of 1 parts per thousand and the uncertainty in the absolute ratio being derived from uncertai nties added in quadrature. The data show no large systematic errors, d emonstrating that the trueness of the mean of a number of ion probe de termined ratios may be improved by repeated measurements. Detector sta bility and instrumental fractionation yield measured ratios (absolute in the sense that they are not normalised from day to day) which show a scatter of between 1 and 2 parts per thousand over periods of many w eeks, an instrument stability which is an order of magnitude better th an data published elsewhere. This level of stability opens the possibi lity of ''standardless'' ion probe oxygen isotope ratio measurements, certainly to an accuracy of 2 parts per thousand or better.