Measuring both major and trace element concentrations using LA-ICP-MS

    All mass spectrometry can only measure isotopic ratios. To obtain element abundances from isotopic ratios, we need two kinds of standards–external standards and internal standards. External standards, with known composition, provide the coefficients that convert signal strength (counts per second) to chemical concentration. These coefficients are then applied to samples to determine the concentrations of unknowns. However, these coefficients may vary from standards to samples. For LA-ICP-MS, such variation is caused by the inconstant laser ablation rate on different matrix and instrumental drift. This is why we need internal standards, which correct the variation (Fig. 1). An internal standard is usually an element with known concentration in both standards and samples. To learn the concentration of an internal standard, EMPA is applied prior to LA-ICP-MS analysis.

Fig. 1 A schematic flow of LA-ICP-MS analysis.

    Following Liu et al. (2008), I am now developing a new LA-ICP-MS analytical method that replaces the conventional internal standard, which is one element, with the sum of major element oxides. This enables us to measure both major and trace elements by LA-ICP-MS alone in that the sum of major element oxides is approximately 100% (assuming H₂O, CO₂, halogens, etc. contribute little). Since the major difference lies in the internal standard, my task is to examine whether this new internal standard can effectively correct the variation in signal-strength/concentration coefficients, and which external standard would be the best with regard to both accuracy and precision.

Fig. 2 Plot of deviation using BHVO as external standard for major and rare earth elements. The pink shaded area denotes 10% deviation area from referred values.

    Five reference glass materials, including BHVO, BCR-2G, BIR-1G, KL-2G and ML-3B, were analyzed in our work. In each session, one of the glasses was assigned as the external standard while the rest as samples. As for precision, 1 σ uncertainty is better than 4% for most elements. Accuracy, however, is not as good. In Fig. 2, deviation relative to referred values frequently falls outside the ballpark (±10%), especially for rare earth elements. In order to monitor potential isobaric interferences, many elements were measured by more than one isotope. Isotopic ratios were plotted in Fig. 3, with true values plotted at the end of each series. Obviously, isotopic ratios obtained in my experiments agree well with the true values, which indicates that isobaric interference may not be a significant issue in my experiments. So far, it remains unknown what caused the accuracy problem in this work.

Fig. 3 Plot of isotopic ratios as monitors of isobaric interference.

Reference

Liu, Y-S., Hu, Z-C., Gao, S., Gϋnther, D., Xu, J., Gao, C-G., Chen, H-H., (2008) In situ analysis of major and trace elements of anhydrous minerals by LA-ICP-MS without applying an internal standard. Chem. Geol. 269, 237-251.