On-Line Standard Addition Technique for the Determination of Elements in the Samples with High Dissolved Solids by ICP-OES: Application to Determination of Rare Earth Elements in Geological Samples

Employing a switching valve to automate external standard calibration in inductively coupled plasma optical emission spectrometry

The traditional external standard calibration method (EC) is automated and simplified using a four-port switching valve (SV4) and a multi-signal approach that enables the generation of several calibration points from a single calibration solution. The SV4-EC method is applied to inductively coupled plasma optical emission spectrometry (ICP-OES) and is based on gradient dilution taking place within the instrument's sample introduction tubing. Both the calibration solution and the samples are diluted by a blank solution containing an internal standard species. Forty-five dilution points are collected over time while the solutions are mixed. Instrument responses from the calibration solution are then plotted against those from the samples, and the slope of the calibration curve is used to determine the unknown analyte concentrations in the samples. The method is used to determine Ba, Co, Cr, Cu, Fe, Mn, Ni, V and Zn in coconut water, creek water, green tea, mouthwash, soft drink, vinegar, and vodka. Limits of detection are in the 0.0002-0.009 mg L-1 (n = 10) range, with precision on the order of 0.4 %-3 % RSD. Analyte percent recoveries from a 0.5 mg L-1 spike are in the ranges of 88.4 %-111 %, 88.9 %-111 %, and 88.0 %-111 % for EC, SV4-EC, and the internal-standard-corrected method (SV4-EC/Sc), respectively. No statistically significant difference is observed between EC and SV4-EC recoveries for any of the sample matrices evaluated. Comparable results between EC and SV4-EC were also found for the analysis of two certified reference materials, Bovine Liver and Oyster Tissue. Based on a single calibration solution, the SV4-EC method requires caution when preparing the calibration standard to minimize measurement bias.