Abstract

Two methods of thallium quantification by inductively coupled plasma mass spectrometry were validated: isotope dilution (ID) after spiking by ²⁰³Tl and external calibration (EC) where the sum of ²⁰⁵Tl + ²⁰³Tl was measured. Both methods are almost specific. The results are significantly influenced only by 1000-fold excess of tungsten due to the interference of ¹⁸⁶W¹⁷O. The limits of detection of both techniques are 0.15 ng g⁻¹ and 0.20 ng g⁻¹ for ID and EC, respectively. The ID technique is robust with respect to instrument parameters settings (rf power, nebulizer Ar flow and lens voltage). The results of EC analysis can be slightly affected by alteration of rf power. For the verification of accuracy of results, three approaches were applied. (1) The analysis of samples prepared from standard thallium solution gave results in agreement with theoretical values. (2) The analysis of spiked plant samples with a very low (wheat flour) and high (defatted rape-seed meal) thallium content proved the accuracy of the ID method (recovery values of 99.5 and 99.8%, respectively do not differ significantly from 100%). On the other hand, EC analysis gave recovery values of 99.2 and 101.6%, respectively, which are more than satisfactory for routine trace analysis, but owing to high precision of measurement these values are statistically different from 100%. (3) Analysis of several CRMs by ID gave results in agreement with certified values except for SRM NIST 2709 San Joaquin Soil. Results of EC analysis can be affected by the choice of internal standard. The usage of rhodium was the best choice. The repeatabilities of both analytical procedures are comparable; relative standard deviations ranged from 0.4 to 1.0%.