10877

IDENTIFICATION AND QUANTIFICATION OF SERUM AND STARCH IN MILK POWDER FROM THE ALLIANCE OF X-RAY FLUORESCENCE WITH CHEMOMETRICS

Favoritar este trabalho

Powdered milk is a food that has an important place in the popularion's diet, also presenting significant economic importance in society. As a consequence there is a high quality demand which requires fast analysis results even in ports and airports.
The quality of powdered milk must obey strict rules of identity and status; frequent frauds are related to the addition of serum, starch, maltose, sucrose, and caustic soda. The most common range of ingredient addition level is 20-25%, as at this level the consumer does not identify the difference in flavor.
In this context, this work proposes a methodology to detect and quantify serum and starch like adulterants in powdered milk with chemometric treatment of data from X-ray fluorescence (XRF). To obtain the spectra cells XRF were completely filled with powdered milk to simulate adulteration with starch and serum in the range from 0 to 30% (w/w) for model construction and analysis of multivariate calibration by the method of regression partial least squares (PLS). The PLS models were constructed with 3 latent variables, the values of rcal and rval obtained were 0.99 and 0.97 respectively and the capacity of prevision to external samples with errors less than 10%. From the results obtained, we propose the application of the PLS models developed to the identification and quantification of serum and starch in powdered milk.