In this work, on-line calibration methods were applied for compensation for matrix effects in Inductively Coupled Plasma-Optical Emission Spectrometry (ICP-OES) using three novel multiple sample introduction systems based on Flow Blurring Ò technology. The methods were compared with conventional calibration methods, using a Conikal nebulizer and a cyclonic spray chamber (i.e., Standard Sample Introduction (SSI) system). Experiments were carried out with synthetic samples containing different matrices. The total liquid flow through the multinebulizers was 400 mL min À1 whereas in the SSI system it was 1000 mL min À1 . One type of calibration method tested was external calibration. By using this calibration method, the mean of absolute values corresponding to the relative error values of different multiple sample introduction systems and all the matrices was 14% and uncertainty was 0.6%. When on-line internal standard calibration was used, the mean relative error value dropped to 3% and uncertainty was 0.6%. With on-line standard addition calibration, relative error values went down to 2%. However, uncertainty values increased to 2% in all cases. With all the calibration methodologies, the accuracy and uncertainty of the obtained results were very similar for both standard and multiple sample introduction systems. The main difference was a significant reduction in resource consumption (i.e., samples, reagents and time) when multinebulization systems were used. Sensitivity, precision and limits of detection were evaluated for the different Flow Blurring Ò based systems and SSI system. For most of the emission lines evaluated, all the Flow Blurring Ò based systems gave higher precision values and lower limits of detection than SSI system. A certified reference material (Estuarine Water, LGC6016), without prior sample treatment (i.e., dilution), was analyzed using external calibration with the SSI system and on-line standard addition calibration with Flow Blurring Ò based systems. The certified reference material analysis gave relative error values ranging between +20% and À30% for the SSI system, and between +4% and À2% for Flow Blurring Ò based systems.

No abstract

No abstract

No abstract

No abstract