Quantitative analysis of the adulteration of orange juice with sucrose using pyrolysis mass spectrometry and chemometrics

A IWO (w/v) beet sucrose solution was used to adulterate freshly squcexed orange juice over the range O-20% (or O-20 g I ’ of added sucrvsc). Samples were analyzed by the rapid automated screening technique of Curic-point pyrolysis mass spectrometry (PyMS). To deconvotve these spectra neural cognition-based methods of multilayer pcraptrons (MU’s) and radial basis functions (RBFr) and the line= regression technique of partial feast squat= (PLS) were studied. It was found that each of the methods could bc used to provide calibration models which gave exceknt predictions for the level of sucrose adulteration at levels below I% for samples, with an accuracy of f 1.3%, on whkb they had not been trained. The best results were obtained using PLS when 8 latent variabks were employed for predictions. Furthermore, the inputs to MLR could be reduad t&g principal comfmnents analysis (PC4) from 150 ~~~SSCS to 8 PC scores without any deterioration of the predictive ability of the modei, highlighting that PCA is an excelfont pep-g step which has the potential to speed up neural network tearuing as there arc fewer weights to update. Since any foodstuff can he pyroiysed in this way, the combination of PyMS with chtmome@ics constitutes a rapid, powerful and novel approach to the quantitative asXssmcnt of food adulteration generally.