Fractionation, precision and accuracy in 14C and 13C measurements

Abstract

During the past ten years, AMS has become a powerful tool in radiocarbon dating. In some applications, an overall accuracy comparable to that of conventional high precision low level counting is required. To achieve this accuracy, fractionation during sample preparation and measurement has to be constant. Comparison of series of 13C 12C ratios measured with AMS and a conventional mass spectrometer indicates that systematic errors in the carbon isotope ratio measurements are in the order of 0.1 to 0.2%. With mg size samples prepared by catalytic reduction on iron, delivering beam currents of about 12 μA for at least 1 h, statistical uncertainties for 14C of 0.3% and a total error of 0.4 to 0.5% for the 14C 12C ratio (including the calibration error and uncertainty in the background subtraction) for modern carbon has been obtained.