Air-segmented, 5-mL flow injection associated with a 200 degrees C heated chamber to minimize plasma loading limitations and difference of behaviour between alkanes, aromatic compounds and petroleum products in inductively coupled plasma atomic emission spectrometry

A method based on the use of a high temperature single pass spray chamber and the injection of a sample plug into an air carrier gas stream was developed to mitigate non spectral interferences caused by organic samples and petroleum products and to reduce plasma loading in Inductively Coupled Plasma Atomic Emission Spectrometry (ICP-AES). The studied solvents were eleven alkanes (hexane, heptane, octane, nonane, decane, undecane, dodecane, tridecane, tetradecane, pentadecane, hexadecane), xylene, kerosene and tetralin. As regards to the real samples two gasolines, a kerosene and a diesel sample were taken. The evaluated sample introduction systems were a 12 cm(3) inner volume single pass spray chamber (also called Torch Integrated Sample Introduction System, TISIS) with and without heating and a 40 cm(3) inner volume cyclonic spray chamber that was taken as a reference device. A characterization of the matrix effect in continuous aspiration mode at a 30 mu l min(-1) flow rate was initially performed. Drop size distributions were measured for the aerosols generated by the nebulizer (primary aerosols) and those leaving the spray chamber (tertiary aerosols). The results proved that the median of the aerosol volume drop size distribution (D-50) for primary aerosols took values from 13.2 to 15.3 mu m. Meanwhile, tertiary ones changed more significantly as a function of both the solvent nature and the chamber temperature. They went from 2 to 4 mu m for the TISIS at room temperature, whereas at 100 degrees C D-50 was included within the 0.7 to 3.1 mu m range. The analyte mass transported towards the plasma was also measured and it was confirmed that this parameter was directly related to the solvent volatility. Thus, at room temperature, efficiencies went from 20 to 60% for hexadecane and octane, respectively. ICP-AES sensitivities changed significantly as a function of the solvent. For real samples, heating of the chamber walls mitigated the interferences, thus, while at room temperature, gasoline samples provided more than one order of magnitude higher signals than diesel samples, at 100 degrees C this signal improvement factor was only of five. All these problems were mostly overcome when the segmented injection of a 5 mu l sample plug was performed. It was concluded that, for all the solutions at 200 degrees C heating temperature the injected sample volume (c. a., 5 mu l) evaporated completely before its further introduction into the plasma. Therefore, differences in analyte mass transported as a function of the solution matrix were mitigated.