Internal standardization and use of microprobe data for internal standardization

LA-ICP-MS measurements require some other form of standardized analysis (either EPMA or XRF) in order to convert raw counts into compositions. For mineral and glass compositions, this is commonly achieved by first analyzing a region of interest by microprobe, and then using an element (typically Al or Ca, or any available element that is present at appreciable concentrations, ionizes well and is relatively unaffected by instrumental drift) to anchor counts per second to known concentrations.

For the purposes of producing high-precision, high-quality data, it is necessary to take all necessary steps to ensure that each LA-ICP-MS analysis is anchored to a quality EPMA measurement of a consistent element. For the purposes of reconnaissance, (and indeed for all intents and purposes of volcanic correlation) it is sufficient to make some first-order approximations:

  1. While Al and Ca are preferred elements for internal standardization, Si provides another alternative. In cases where datasets from different projects are being collated, the same internal standard is not always used. The effects of choosing a different internal standard are minimal compared to the underlying differences in trace elements in most cases (Fig. 1). The researcher analyzing results must consider whether these changes in internal standard choice are likely to affect their interpretation of volcanic correlation.

Fig. 1: Differences in trace element concentration as a function of internal standard chosen and concentration of internal standard.

  1. Intra-unit variation of a given major element can be assessed with a few microprobe analyses, in order to ensure that there is no major zonation in the internal standard element. In the case of Al and Si, the concentrations are relatively constant to ~1%. In such cases, it is possible to use a single, average value for internal standardization of all biotites in a single unit. Trace elements change linearly and proportionally to the concentration of the major element used as an internal standard, so a 0.5% increase in Al2O3 (13.5% instead of 13%) would impart a 4% increase in trace element concentrations. Meanwhile, the differences between units may approach an order of magnitude difference in concentration.
  2. Likewise, it is possible to conduct reconnaissance of a new unit by LA-ICP-MS even before undertaking any EPMA analyses by simply assuming an Al2O3 concentration of 13%. This value may be off by as much as 2%, imparting 15% relative difference in trace element concentrations, but this effect may anyways be swamped by the large differences in trace elements between units.