Parkinson, I. J.; Bonnand, P.; Pearce, C. R. and Fehr, M.
Application of the generalised power law to double-spike measurements by MC-ICP-MS.
In: American Geophysical Union, Fall Meeting 2010., 13-17 December 2010., San Francisco.
(Click here to request a copy from the OU Author.
It is now well established that instrumental mass fractionation on MC-ICP-MS instruments does not always adhere to the exponential mass fractionation law and can be better described by the generalised power law (GPL). However, existing double-spike equations assume exponential mass fractionation behaviour. Therefore, deviations from the exponential law become important because of the large amount of fractionation in the MC-ICP-MS. We have collected Cr and Mo double-spike standard data over a number of years. These data plot with small but measurable mass dependent differences from zero per mil, which vary between analytical sessions. We interpret that these deviations are due to small variations in the mass fractionation behaviour of the MC-ICP-MS and are related to differences in the type of nebuliser and cone geometry. The GPL relies on deriving a value for the exponent term, n, and its value can be derived from the slope of the natural logarithms of the measured isotope ratios. However, the instrumental mass fractionation on modern MC-ICP-MS is very stable within an analytical session, so the spread in the isotope ratios of a standard is not very large, making it difficult to derive any statistically valid information on the value of the slope. For Cr, which only has four isotopes, we adopt a numerical technique for deriving the value of n. Offsets in the δ53Cr values can be reproduced by varying n between 0.058 and 0.090. Molybdenum has seven isotopes and is therefore possible to incorporate the GPL into the double-spike equations and solve directly for n, using four isotope ratios. The advantage of this technique is that allows the correct mass fractionation behavior to be applied to standards and samples within and between analytical sessions. This provides a means to generating high-precision data without recourse to ad hoc daily corrections.
Actions (login may be required)