Systematic Error in Isotopic Analysis of Noble Gas Mixtures

摘要

Analyses of known standard mixtures of helium-neon, helium-argon, and helium-xenon, show significant biases in measured ratio which change with sample (inlet) pressure. Bias magnitude appears dependent on component partial pressure and not total sample pressure. Quantitative gas analysis assumes a linear relationship between measured ion current, I/sub i/, and partial pressure, PP/sub i/, of a corresponding component in the inlet system: PP/sub i/ = I/sub i//S/sub i/ where S/sub i/ is the sensitivity parameter (assumed constant). For this mass spectrometer, sensitivity is not constant but varies with pressure and is different for each gas. Ion current from a mole fraction, X/sub i/, is related to source pressure, P/sub S/, and ion production factor, K/sub i/, by I/sub i/ = K/sub i/X/sub i/P/sub i/ = K/sub i/(F/sub I//F/sub S/)(l + (F/sub S//Fsub H/))X/sub i/ P/sub I/. Identifying X/sub i/P/sub I/ as PP/sub i/, sensitivity is given by S/sub i/ = K/sub i/(F/sub I//F/sub s/)(l + (F/sub S//F/sub H/)). Nonlinear operation can arise from three factors determining sensitivity. Factor K/sub i/, if it varies at all, should depend on total pressure, not partial pressure. Conductance effects of the molecular leak (F/sub I/) and source exit ports (F/sub S/) should have positive slopes for all gases if deviations from molecular flow were significant. Changes in diffusion pumping speed (F/sub H/) by varying the power produce sensitivity changes. Reciprocal of sensitivity changes reflect change in F/sub H/. The variation shown is attributed to changes in F/sub H/ from a cooling effect on pumping fluid with increased gas load. Independent flow of gas components implies that the use of a sensitivity, S/sub i/, appropriate for the partial pressure of the component should minimize systematic errors in ratios: S/sub i/ = S/sub o//sub i/ (l + S/sub i/ (X/sub i/ P/sub I/)). Applying this model to the data reduced the systematic error. (ERA citation 06:023965)