(611a) Evaluation of Ultrasound By Direct Contact on Convective Drying of Banana (Musa paradisiaca) Slices Using Principal Component Analysis and Response Surface Methodology

Conventional convective drying is one of the latest techniques used to reduce the amount of water in foods, however, due to the long times of exposure to high temperatures, these techniques can reduce the sensory quality, in addition to generating high costs associated with consumption of energy. An alternative to improve efficiency in these processes is to use innovative tools such ultrasound (US) application by using direct contact during drying to improve quality and keep nutritional content of the product with reduction in time, cost and temperatures of operation.

In this study, a Box-Behnken experimental design was used to study the influence of three independent variables: US time-frequency (t), US power level (Pot), and air temperature (T) on the drying process time, apparent density, total polyphenol content and color difference as responses of the banana slices. Seventeen experiments were performed which involved 12 three-level experimental points and 5 repetitions in the central point. Fisher's statistical testing was performed for the analysis of variance (ANOVA) for quadratic regression equations. Results of each response variable were analyzed according to the Response Surface Methodology, analyzing tendencies and regions of operation with contour graphs. In addition, models were proposed for each response variable, which included the input factors of the experimental design proposed. These models were reduced in order to eliminate non-significant terms and were carried out validation of the statistical models proposed. After performing the reduction of non-significant terms, the ANOVA analysis showed a good fit for most of the response variables. PCA was performed to evaluate effect of ultrasound on quality characteristic of banana-dried slices. The PCA analysis showed that the drying time and the apparent density of the samples was strongly related to the temperature of the process. Moreover, power level of the ultrasound influenced the color variation of the dry sample respecting fresh fruit. Finally, US time-frequency affected total polyphenol content of dried samples. Analysis of both tools like PCA and RSM allowed knowing the effect of the independent variables on the drying operation and quality parameters on banana dried slices.