Multivariate optimization of a procedure employing microwave-assisted digestion for the determination of nickel and vanadium in crude oil by ICP OES

This work presents the optimization of a sample preparation procedure using microwave-assisted digestion for the determination of nickel and vanadium in crude oil employing inductively coupled plasma optical emission spectrometry (ICP OES). The optimization step was performed utilizing a two-level full factorial design involving the following factors: concentrated nitric acid and hydrogen peroxide volumes, and microwave-assisted digestion temperature. Nickel and vanadium concentrations were used as responses. Additionally, a multiple response based on the normalization of the concentrations by the highest values was built to establish a compromise condition between the two analytes. A Doehlert matrix optimized the instrumental conditions of the ICP OE spectrometer. In this design, the plasma robustness was used as chemometric response. The experiments were performed using a digested oil sample solution doped with magnesium(II) ions, as well as a standard magnesium solution. The optimized method allows for the determination of nickel and vanadium with quantification limits of 0.79 and 0.20 mu g g(-1), respectively, for a digested sample mass of 0.1 g. The precision (expressed as relative standard deviations) was determined using five replicates of two oil samples and the results obtained were 1.63% and 3.67% for nickel and 0.42% and 4.64% for vanadium. Bismuth and yttrium were also tested as internal standards, and the results demonstrate that yttrium allows for a better precision for the method. The accuracy was confirmed by the analysis of the certified reference material trace element in fuel oil (CRM NIST 1634c). The proposed method was applied for the determination of nickel and vanadium in five crude oil samples from Brazilian Basins. The metal concentrations found varied from 7.30 to 33.21 mu g g(-1) for nickel and from 0.63 to 19.42 mu g g(-1) for vanadium.