Method optimization for heavy metal determination in milk powder: application to milk samples from Greece

Abstract

The scope of this study was the development, optimization and validation of an analytical method for the determination of selected heavy metals and trace elements (As, Hg, Se, Cd, Cu, Pb, Mn, Fe, Ni, Zn, Cr) in milk powder, using microwave-assisted digestion. A statistical experimental design approach using central composite design (CCD) was carried out, to investigate the effects of three independent pretreatment variables (final digestion temperature (°C), HNO₃ concentration (in % w/v), microwave hold time) on the heavy metal recovery of spiked undigested milk powder sample and to calculate the variable factor values which produce the optimum recovery. CCD results revealed that the optimum digestion conditions, with respect to maximum recovery were as follows: temperature 190 °C, HNO₃ 56.8% w/v, and digestion time of 8.47 min. The method was fully validated. Recoveries for all metals ranged between 92 and 108% while intra-day repeatability was below 6.59% (rsd). A certified reference material (ERM BD 150) that included 8 out of the total 11 heavy metals of the present study (Hg, Se, Cd, Cu, Pb, Mn, Ni, and Zn) was used to test the accuracy of the method where acceptable recovery values ranging between 96 and 107% were obtained. High heavy metal recoveries, short digestion time, and low acid consumption were the advantages of the pretreatment method. The analytical process was successfully applied for the determination of heavy metals in different milk samples from the Greek market. Heavy metal concentrations for Ni, Cr, Pb, Cd, Se, Mn, and Cu measured in this study reached 307, 102, 8.01, 5.96, 60.2, 519, and 438 μg/kg wet weight (ww), respectively. Zn and Fe were found at concentrations ranging 3.21–8.39 and 0.170–10.1 mg/kg ww, respectively. Risk assessment based on the WHO tolerable daily intake levels and the calculated target hazard quotients revealed that the consumption of the selected milk samples is considered safe.

http://ift.tt/2eFyVUr