School of Engineering and Science - Jacobs University

cps, then 3000 cps must be subtracted from any signal intensity measured at mass 156
(e.g., 156 Gd or 156 Dy). For a detailed treatment of possible REE interfering species and
examples of calculations of the appropriate correction ratios the reader is referred to
Dulski (1994). Considering the number of isotopically different REE-oxyhydroxide
species that may be generated (Table 2), such corrections may seem unwieldy, but
determination of these correction ratios is not necessary with each individual ICPMS
analysis. The approach adopted here, and following the methods of Dulski (1994),
requires periodic determination of these correction ratios at very specific ICPMS
instrument settings which consistently produce similar REEO(H) + /REE + values, and
then ensuring that sample analyses are always performed at identical ICPMS
instrument settings. For the analyses performed within the JUB Geochemistry Lab
this entails tuning and optimization of the ICPMS so that CeO + /Ce + does not deviate
from 0.029 ±0.001.
4.2. External calibration
The determination of trace metal concentrations in sample solutions is
performed using laboratory prepared external calibration standards. This procedure
consists of using commercially available 1000 mg/l single element standards
(Inorganic Ventures, USA) that are combined to form 1 mg/kg (parts-per-million,
ppm) multi-element stock solutions. The 1 mg/kg stock solutions are prepared in acid
matrices that are similar to those present in the 1000 mg/l single element standards,
which are typically ~0.5 M HNO 3 (2.2% v/v). These multi-element stock solutions are
further diluted immediately prior to an individual ICPMS analysis to produce 10 and
20 μg/kg calibration standards, which are analyzed in conjunction with the sample
solutions. Preparation of these external calibration standards is presented
schematically in Figure 2.
Ideally, the determination of elemental concentrations in sample solutions
would proceed using the highly accurate method of standard addition, in which the
sample solutions are split into a minimum of three aliquots, two of which would be
spiked with the elements of interest at different concentrations. By comparing the
ICPMS response for these three solutions it is possible to calculate the concentrations
of the spiked elements in the non-spiked sample solution. The greatest advantage of
the standard addition method is that it minimizes effects caused by matrix differences
9