Evaluation of collision-reaction interface ICP-MS for carbon effect correction on chromium determination

Complete text of publication follows. The use of mathematical correction equations, cool plasma, and dynamic reaction cells is frequent for correcting polyatomic interferences in ICP-MS. For carbon-rich matrixes, the 40Ar12C+ polyatomic ion affects the accuracy of Cr determination based on its main isotope. The aim of this work was to evaluate the employment of an ICP-MS equipped with a collision-reaction interface (CRI, Varian 820-MS) for carbon interference correction on Cr determination at m/z 52. The CRI allowed gas introduction directly into the plasma expansion zone, through modified sampler and skimmer cones, and H2 was tested as collisional and reactional gas. The gas flow-rate, point of introduction and plasma operational settings were systematically evaluated to achieve the lowest detection limits and robust method conditions. Based on typical residual carbon content in acid digestates, solutions containing from 0.1 to 2.0 % m v-1 carbon were prepared to evaluate carbon effect on Cr determination. Oxalic acid was used as carbon source. Preliminary results showed that H2 introduction through the skimmer, kept at 80 mL min-1 flow rate, presented better efficiency than through the sampler because lower BEC values (background equivalent concentration) were obtained. At flow rates lower than 60 mL min-1, Cr recovery value was around 170 % (2.0 % w v-1 carbon) due to the positive interference caused by 40Ar12C+. Using a H2 flow rate of 80 mL min-1, accurate recoveries for 5.0 μg L-1 Cr were achieved. The BEC obtained when using the CRI was ten fold lower than that obtained without using it. The effect of different RF powers (0.60-1.50 kW) and nebulizer flow-rates (0.90-1.10 L min-1) on sensitivity was also studied. The BEC and SBR values were similar for all tested RF powers. The analytical signal was improved using higher flow-rate. However, in order to obtain a low CeO+/Ce+ ratio for plasma (< 3%), plasma was operated at 1.50 kW and 0.90 L min-1. The CRI was fundamental to eliminate carbon effects, and its performance was improved by using robust operation conditions. The authors kindly acknowledge the financial support from the Fundacao de Amparo a Pesquisa do Estado de Sao Paulo (FAPESP, Grant 2006/59083-9), CNPq and CAPES.