Minimization of acid effects at low consumption rates in an axially viewed inductively coupled plasma atomic emission spectrometer by using micronebulizer-based sample introduction systems

Abstract

Two methods were applied in order to minimize acid effects at very low liquid uptake rates in an axially viewed ICP-AES system: aerosol solvent elimination and direct injection in the plasma base. The latter choice eliminated the use of a spray chamber. Four different sample introduction systems were evaluated based on the use of micronebulizers. The first system made use of a so-called microconcentric nebulizer (MCN) associated with a double-pass Scott spray chamber. Two other systems were based on the use of a desolvation unit that consisted of a first and a second stage in which the solvent was partially removed. The solvent elimination system was either two Liebig condensers connected in series or a porous PTFE membrane. In these cases two different MCNs were coupled. The last sample introduction system employed was a direct injection nebulizer (DIN) that eliminated the contribution of the spray chamber to the acid interference. The liquid flow rates ranged from 5 to 120 µl min^–1, depending on the sample introduction system employed. Nitric, hydrochloric and sulfuric acid of 0.9 mol l^–1 were tested and 3.6 mol l^–1 nitric acid solution was also studied in order to evaluate the acid concentration effect. The Mg II to Mg I line intensity ratio was used to evaluate any possible changes in the plasma conditions. The results showed that, for the MCN coupled to the spray chamber, the lower the liquid flow rate, the greater was the acid interference. The extension of this effect was dependent on the MCN configuration. For the desolvation systems and DIN the behavior was the opposite, the acid interference being more important as the liquid flow rate increased. For the four acids employed, and at very low liquid flow rates (below 30 µl min^–1), the acid effect was eliminated by using a conventional desolvation system with a heating temperature of 160 °C, the condensation temperatures for the two condensers being 10 and 0 °C, respectively. The use of a membrane was found to be advisable because, under some conditions, acids and water gave the same analytical signal. As regards the DIN, it was able to eliminate the interference for 0.9 mol l^–1 nitric and hydrochloric acid. Nevertheless, the two remaining solutions, i.e., 0.9 mol l^–1 sulfuric and 3.6 mol l^–1 nitric acid, gave rise to a decrease in the signal of around 20%.