DETERMINATION OF THALLIUM IN CEMENT DUST AND SEDIMENT SAMPLES BY DIFFERENTIAL-PULSE ANODIC-STRIPPING VOLTAMMETRY - A CHEMOMETRIC APPROACH TO LINEAR CALIBRATION

Differential-pulse anodic stripping voltammetry (DPASV) was used to determine Tl in cement kiln dust and sediment samples. The separation of Tl from the sample matrix was achieved by the extraction of Tl3+ into diethyl ether from an HBr-Br2 medium. Initially, a mixture of HNO3 and HF was used to digest the cement dust samples to cause volatilization of silica as SiF4. The Tl3+ was then reduced to Tl+ using hydrazine sulfate prior to determination by DPASV. A calibration graph based on a deterministic relationship between the measured response y and the analyte concentration x (so that y = beta-0 + beta-x + r) is also discussed using classical (or reverse) regression. The precision of parameter estimates of the model (bo and b : y = b0 + bx) can be achieved by display of the diagonal elements of the hat matrix. Methods for computing the standard uncertainty in the predicted response s(yn)2 and in the concentration s(x)2 are also provided. Thallium concentrations of 700 +/- 3.07, 395 +/- 2.91 and 850 +/- 2.59 ng g-1 were found in the cement dust and sediment samples, respectively.