Near infrared spectroscopy as an alternative quick method for simultaneous detection of multiple adulterants in whey protein-based sports supplement

Abstract

Results of the Dumas method and near infrared (NIR) spectroscopy were compared to detect and quantify adulterants (urea, L-taurine, L-histidine) and to evaluate protein content in whey protein powder samples. Three different NIR regions were compared using principal component analysis and partial least squares regression models determining the quantity of each component. Calibration models were cross-validated with leave-one-sample-out method and prediction models were tested using data of independent days. Protein values by the Dumas method showed significant differences to the actual protein content in most cases. Predictive models had a minimum R2 value of 0.97 in all cases and the lowest average prediction error was 0.16%, 0.41% and 0.23% with the respective limit of detection minimum values of 0.09%, 0.22% and 0.15% for urea, taurine and histidine. Results show, that NIR spectroscopy coupled with chemometrics can be a useful tool to detect multiple nitrogen-based adulteration simultaneously in whey protein powders.