Application of Electronic Nose in Tandem with Chemometric Analysis for Detection of Alicyclobacillus acidoterrestris-Spawned Spoilage in Apple Juice Beverage

Apple juice beverage spoilage spawned by Alicyclobacillus acidoterrestris can cause sensory defect by production of medicinal off-flavor, which could hardly be noticed in the early stage and would lead to the consumer complaints and, consequently, the economic and image loss of the manufacturers. Thus, it is necessary to detect the contamination at early stage in case of serious converse effect. The aim of this study was to evaluate the performance of an electronic nose (e-nose) in tandem with chemometric analysis as online, rapid screening technique for detection of the spoilage in apple juice beverage, using test panel evaluation as reference, as well as to predict the spoilage status of apple juice beverage based on the response to the specific volatiles in it. Coupled to linear discriminant analysis (LDA), e-nose could discern the contaminated beverage after 4 h, corresponding to the bacterial number of 200 cfu/mL, the level at which the test panel could not yet diagnose the spoilage, indicating that the signals of e-nose could be applied as early indicators for the onset of spoilage. Sensors 2, 6, 7, and 8 were determined to be quite sensitive to the alteration of flavor in apple juice beverage and had played the most important part in discrimination. Furthermore, based on the response of e-nose, A. acidoterrestris counts in unknown samples could be well predicted by established partial least squares (PLS) regression model with a high relation of 0.95. Based on these results, e-nose can be employed as a powerful tool for quick and reliable analysis of the flavor in apple juice beverages during processing or on the shelf to implement rapid routine inspection as well as realize the early detection of potential spoilage caused by A. acidoterrestris.