Abstract
High-resolution continuum source graphite furnace atomic absorption spectrometry (HR-CS-GFAAS) methods with solid and solution sampling were developed for the determination of the Bi dopant in high-purity lithium niobate (LiNbO(3)) crystals. Samples were cut from the cylindrical crystal bulks, cleaned, and pulverized. A HF-HNO(3) mixture was applied for microwave digestion of LiNbO(3) (≈0.07 g per sample). Atomization transients, pyrolysis, and atomization curves were studied with various media and chemical modifiers, e.g., triammonium citrate (TAC), Pd-Mg-(NO(3))(2). For solid samples, the optimal pyrolysis and atomization temperatures of Bi were found at 1000 and 1800 °C, respectively, whereas, for solution samples, much lower values of 500 and 1300 °C were obtained. TAC slightly, but Pd-Mg considerably increased the optimal pyrolysis and atomization temperatures, i.e., up to 1300 and 2100 °C, respectively. The dissolved LiNbO(3) matrix acted as an internal modifier, exhibiting optimal pyrolysis and atomization temperatures for Bi. For solid sample analysis, 0.05-0.4 mg (average: 0.1 mg) of LiNbO(3) powder was dosed into graphite boats, while conventional standard addition and three-point estimation were used for calibration. The limit of detection (LOD) was 0.4 and 0.3 μg/g for solid and solution analysis, respectively, using Bi I 227.6580 nm and Bi I 223.0608 nm lines, respectively. Utilizing the latter for solid sampling, an LOD of 0.03 μg/g can be attained. The analytical results for all methods were in good agreement (mean bias: <12%). The precision of the solid and solution sample methods was at 6-16% (average: 12%) and 1-13% (average: 4.4%), respectively. The Bi content of the crystals ranged from 56 to 311 μg/g. The characteristic mass for solid and solution sampling was 220 and 17 pg, respectively. The accuracy of the method was checked against the GBW07407 (laterite soil) certified reference material.