Comparison of drying process of plantain bananas under natural sunny and confined oven environments: experiments and assessment of mathematical models

ABSTRACT

In this paper, the drying process of plantain bananas from Democratic Republic of Congo (DRC) is investigated under two climatic operating conditions, the sunny and the confined oven environments. First, the effect of these environments at 40°C on the process evolution in terms of moisture content, particle size distribution of dried bananas and phytochemical properties (color, composition) is pointed out. Obtained results essentially show that natural sun drying is more favorable for all banana slice thicknesses used. The moisture content of plantain is reduced by 57.2% for natural drying in 28 h, while it is lowered by 52.76% for oven drying in 29 h. Phytochemical composition not detected for fresh bananas features an important quantity of sugar (433,4 ± 4,9 mg/g glucose) for sunny dried bananas and (60,8 ± 0,7 mg/g glucose) for oven drying. In contrast, the total polyphenol contents obtained using natural drying is less (1,78 ± 0,05 mg/g eq GA) and (3,05 ± 1,0 mg/g eq GA) for oven drying, even though the total lipids are mostly similar in both configurations (0,12 g/100 g) for natural drying and (0,13 g/100 g) for oven drying. Considering the impact of banana slice thickness, the use of small thicknesses is more affordable. Finally, the gathered data are used to fit in selected mathematical models for thin layer drying. It turns out that the models by Page (MR = exp(−0,1838t^0,9795) and MR = exp(−0,5611t^0,0025)) are the best fit to explain the moisture transfer in the DRC banana in the confined oven and in the natural sun environments, respectively. These results are novel as they determine for the first time the drying characteristics of the DRC plantain bananas (abb DRC) while providing their drying kinetics in the two environments under consideration.

KEYWORDS:

• Drying process
• banana slices
• natural sunny environment
• confined oven configuration
• mathematical model assessment

Acknowledgements

The authors would like to thank Prof. Dr.-Eng. Chrigui Mouldi, Director of the Mechanical Modeling, Energy and Materials (M²EM) Research Unit for his advice and support during the research that produced the results of this paper. The authors also thank Dr Eng. Hab. Ridha ENNETTA, Coordinator of the Research Master of Mechanical Engineering at the Higher Institute of Industrial Systems of Gabès (ISSIG) and member of the Mechanical Modeling, Energy and Materials (M²EM) Research Unit for his special efforts, advice, and relations, which have enabled us to access in different Laboratories in order to carry out the research which produced the results presented above.

Disclosure statement

No potential conflict of interest was reported by the authors.