English Abstract The aim of this project was to investigate whether
ICP-MS is a suitable method for determining aluminium at low concentration
levels in ground and drinking water. The determination was performed at
m/z=27 with indium as an internal standard. The calibration curve was
linear up to at least 100 mug/l A1 ; the detection limit (3s; n=25) was
0.4 mug/l. This detection limit could only be achieved if labware was
thoroughly cleaned with 1% HNO3, if solutions were kept in polyethene
bottles and if the metal parts of the autosampler were cleaned or replaced
by synthetic materials. The short term stability (RSD) was 1.8% and the
long term stability (RSD) was 3.0%. Spectral interferences and
matrixeffects due to the most abudant ions in ground and drinking water and
interferences like 11B16O+, 1H12C14N+ and 26Mg1H+ could be neglected. Only
the addition of 400 mg/l sulphate resulted in an 10% positive shift in
sensitivity. ICP-MS results compared well with ICP-AES results for
concentrations above the detection limit of ICP-AES (20 mug/l). 70% of the
109 analyzed ground and drinking water samples (from the Netherlands) had
concentrations within the linear working range (0.8-200
mug/l).