POTENTIAL LIMITS ON THE ACCURACY OF PB ISOTOPIC MEASUREMENTS BY LA-ICPMS DUE TO MATRIX EFFECTS

We present the development of an analytical method for the in situ measurement of Pb isotope ratios in relatively low-concentration samples (~1-40 ppm Pb) using a laser ablation system (193-nm excimer laser from Photon Machines^TM) with a double-focusing single-collector (SC) ICPMS (Nu AttoM ES^TM). Our method builds upon a published technique [1,2] that used different LA-SC-ICPMS instrumentation to demonstrate the benefits of fast-scanning ion-counting measurements combined with flat-top peaks. We have paid special attention to the characterization of instrumental artifacts using solutions of the NIST SRM981 Pb isotopic standard in “wet plasma” mode. This includes validating the correction for the interference of ²⁰⁴Hg on ²⁰⁴Pb, characterizing the effects of tails from thallium (with masses ²⁰³Tl and ²⁰⁵Tl) on all of the Pb peaks (for Tl/Pb ratios from ~0 to 10 at a constant Pb signal intensity), evaluating the stability of the instrumental mass bias, and maintaining linearity of the detector response over the full dynamic range. Using a suite of international glass standards and newly developed in-house mineral and glass reference materials, we have identified two non-spectral, mass-dependent matrix effects during laser ablation that may limit the accuracy of the measured Pb isotope ratios: (1) one is related to differences in laser spot size and (2) the other is related to differences in the bulk chemistry of the ablated materials (e.g., NIST610 vs. basaltic glasses). The first matrix effect, notable only at the smallest spot sizes (≤20 μm), can presumably be avoided by maintaining a constant spot size for standards and samples. The second matrix effect can potentially be minimized if a standard that closely matches the bulk chemistry of the sample is analyzed. These matrix effects, and the final precision and accuracy of the method, will be discussed in detail at the meeting. [1] Jochum et al. (2005) IJMS 242, 281-289. [2] Jochum et al. (2006) JAAS 21, 666-675.