Estimation of the quantification uncertainty from flow injection and liquid chromatography transient signals in inductively coupled plasma mass spectrometry

Abstract

The quality of the quantitative results obtained from transient signals in high-performance liquid chromatography–inductively coupled plasma mass spectrometry (HPLC–ICPMS) and flow injection–inductively coupled plasma mass spectrometry (FI–ICPMS) was investigated under multielement conditions. Quantification methods were based on multiple-point calibration by simple and weighted linear regression, and double-point calibration (measurement of the baseline and one standard). An uncertainty model, which includes the main sources of uncertainty from FI–ICPMS and HPLC–ICPMS (signal measurement, sample flow rate and injection volume), was developed to estimate peak area uncertainties and statistical weights used in weighted linear regression. The behaviour of the ICPMS instrument was characterized in order to be considered in the model, concluding that the instrument works as a concentration detector when it is used to monitorize transient signals from flow injection or chromatographic separations. Proper quantification by the three calibration methods was achieved when compared to reference materials, although the double-point calibration allowed to obtain results of the same quality as the multiple-point calibration, shortening the calibration time. Relative expanded uncertainties ranged from 10–20% for concentrations around the LOQ to 5% for concentrations higher than 100 times the LOQ.