English abstract
Applications of advance technology in food science is important to minimize losses
due to spoilage and it could help consumers to determine shelf life. In this study,
pasteurized milk spoilage during storage was monitored using an electronic nose
(Cryanose 320, Sensigent, USA) with an array of 32 sensors. Milk samples were stored
for 2.3 days (56 hours) in an oven (at 25 ºC) and for 15 days in a refrigerator (at 4 ºC).
Samples were analyzed using e-nose and conventional tests used to determine quality of
milk (i.e. total bacterial count "TBC", titratable acidity "TA" and pH). The stored
samples at 25 oC were analyzed at 0, 8, 24, 32, 48 and 56 hours while samples stored at 4
ºC where analyzed every three days. Equilibrated volatile organic compounds (VOCs) in
the head space of e-nose vials (15 ml) were optimized using different extraction times (5,
15 and 30 minutes) and temperatures (30, 35 and 40 ºC). The best extraction of VOCs
corresponded the highest sensitivities of e-nose sensors, and it was achieved at 40 ºC for
30 minutes. Principal Component Analysis (PCA) was a satisfactory method for
analyzing e-nose data of milk stored at 25 ºC, while Linear Discriminant Analysis
(LDA) was more effective compared to PCA in discriminating milk samples stored at 4
ºC. A clear shift was identified in both smell prints as identified by e-nose, and total
bacterial count after 24 hours and 9 days for storage at 25 and 4 ºC, respectively. On the
other hand, TA exceeded the normal limits of 0.14-0.21 % after 24 hours for storage at
25 ºC (0.247 ± 0.006%) and after 15 days for storage at 4 ºC (0.25 ± 0.01%). The pH
was a good quality indicator just for storage at 25 ºC, while it had no clear trend for
storage at 4 ºC. It reached to pH value of 6.43, which was below the normal limits of
6.6-6.8 after 24 hours of storage at 25 ºC. Based on the microbial count data and e-nose
output, tested milk had a shelf life of 0.3 day (8 h) when milk stored at 25 °C, while it
was 9 days when milk was stored at 4°C. Finally, the quality of tested milk stored at 4 ºC
was classified based on microbial data and the three clear clusters of LDA plot into three
classes: Excellent (TBC ≤ 2.48 Log10 cfu ml-1
), good (TBC = 4.89 ± 0.21 Log10 cfu ml1
) and poor (TBC ≥ 5.00 Log10 cfu ml-1
).
