E-NOSE ANALYSIS OF MILK TO DETECT THE INCLUSION OF HYDROPONIC BARLEY FORAGE IN THE BUFFALO DIET

This study explores the use of an electronic nose (E-nose), equipped with ten metal oxide semiconductor thin-film sensors, to detect raw milk samples from buffaloes fed hydroponic barley forage as a substitute for maize silage. 108 samples were collected on three different days, from three groups of twelve Italian Mediterranean buffaloes each. The control group (C) was fed a diet based on maize silage, which in the other groups was replaced by hydroponic forage at 50 % (LH group) and 100 % (HH group). The data pattern from the E-nose sensor responses data was analysed using linear discriminant analysis (LDA) and principal component analysis (PCA) as pattern recognition methods, to investigate the discrimination capability of the sensor array. Confusion matrix, obtained from LDA, correctly classified C, LH and HH milk at 90 %. Solid-phase microextraction with gas chromatography– mass spectrometry revealed that C samples had the highest total level of volatile organic compounds, particularly ketones, volatile fatty acids and esters, while HH had the lowest quantity of volatile fatty acids and higher amounts of p-cresol, 1-octen-3-ol and dimethyl sulfone. This has resulted in a greater response intensity of most E-nose sensors for C, such as W1S, W2S, W3S and W6S, which react to high concentrations of broad-range and aliphatic compounds. E-nose proves effective in rapidly identifying hydroponic forage in buffalo diets.