Abstract :
[en] Apple is the second most popular fruit in the market worldwide, and its production in China represents 46.85% of the total world production. Aroma is a critical quality attribute consisting of not only one specific component but a complex mixture of volatiles compounds. Different varieties or cultivars of apples present great differences in aroma profile, and their composition and content can objectively reflect their flavor characteristics. Although three-hundreds and more apple volatile compounds have been identified in the literature, the characteristic compounds, the discrimination of apple geographical region and botanical origins, and the extraction methods of apple volatile compounds have not been yet studied in detail. The aim of this work was to study apple volatiles through apple discrimination and extraction methods. Therefore, the following parts had been investigated in the present thesis.
First of all, thirty-five apple varieties from the same germplasm in Liaoning China were selected to investigate characteristic volatiles over two consecutive harvest years by HS-SPME-GC-MS. Taking odor-active value into account, ethyl 2-methylbutyrate, hexanal, 1-hexanol, E-2-nonenal and linalool were the critical characterized odor-active compounds over two harvest years. Partial least squares discriminant analysis (PLS-DA) revealed a good apple varieties classification. According to VIP values, ethyl 2-methylbutyrate, 2-methyl-1-butanol, Z-3-hexenyl acetate, E-2-hexen-1-ol, linalool and dodecanol were the most important variables to discriminate apple cultivars. Moreover, through the aroma data from two consecutive years, it was found that the aroma is greatly affected by seasonal factors.
Secondly, in order to take into account the geographical region of apples and find a more rapid detection method, a flash GC E-nose was carried out to assess forty-one apple samples from seven geographical regions (Shandong, Shanxi, Sinkiang, Hebei, Gansu, Liaoning and Shaanxi in China) and three botanical origins (cv. Golden Delicious, cv. Fuji and cv. Ralls). A total of twenty-nine volatile compounds have been detected, which were seventeen esters, five alcohols, three aldehydes, one ketone and three others. However, only seven compounds in this paper were consistent with the previous chapter, which were ethyl butyrate, butyl acetate, ethyl 2-methylbutyrate, pentyl acetate, butyl butanoate, ethyl octanoate, and hexyl 2-butenoate. From the point of view of compound identification, flash GC E-nose may not be satisfactory. Compared to classification methods of PLS-DA, stepwise linear discriminant analysis (SLDA) and decision tree, SLDA indicated the best discrimination results both in geographical regions and botanical origins, which could reach 88.2% and 88.9%, respectively.
Finally, to explore the effect of extraction methods on volatile compounds, four extraction methods, involving solid-phase micro-extraction (SPME), stir bar sorptive extraction (SBSE), dynamic headspace sampling (DHS) and solvent-assisted flavor evaporation (SAFE, with three conditions) were employed to extract apple flavor compounds. A total of one-hundred and twenty volatiles were extracted with forty-six esters, twenty-six alcohols, fourteen acids, ten aldehydes, four ketones, four ethers, three phenols, four olefins and nine others. SBSE allowed the identification of the greater number of compounds (fifty-two), followed by DHS (forty-six), SPME (thirty-three) and SAFE (thirteen, twenty-five and twenty-seven). SBSE and DHS both revealed a high extraction capability for low-volatile and alcohol compounds, respectively. The combined use of the DHS and SBSE methods was more comprehensive, which had eighty-two compounds in total and extracted 68.33% of the total compounds. Moreover, these two methods could identify compounds that had not been previously determined in apples or other products, and be used as a complement to SPME results. While current SAFE extraction conditions were not effective for apple flavor analysis in the present study.
