This paper proposed a nonlinear chemical fingerprint method for simultaneous determination of urea and melamine in milk powder using "H++Ce4++BrO3-+malonic acid" as reaction system. A multiple linear relationship was obtained between the adulterants content in milk powder and inductive time of corresponding mixed milk powder. System analysis model established with classical least squares (CLS) method was then used to calculate the content of urea and melamine in milk powder. The method was successfully applied to milk powder samples and had good recoveries in the range of 99.17-100.25%, with the relative standard deviation (RSD) in the range of 0.60-4.12%. The limits of detection for urea and melamine were 0.33μg·g-1 and 0.05μg·g-1, respectively. The limits of quantification were 1.11μg·g-1 and 0.18μg·g-1, respectively. The results indicated that the new method was feasible and had the advantages of low cost, simple operation and without pretreatment of samples.
Keywords: Classical least squares method; Melamine; Multiple linear regression model; Nonlinear chemical fingerprint; Urea.
Copyright © 2016 Elsevier Ltd. All rights reserved.