Theoretical and experimental analysis of the drying kinetics of bananas

Abstract

This work reports a theoretical and experimental study on the drying kinetics of bananas, under different air drying conditions. The experimental data of moisture content during drying were treated according to a diffusional model considering constant diffusion coefficient and including successively, through different models, the effects of shrinkage of the fruit and convective mass transfer on its surface. The diffusion and convective coefficients were determined by fitting the model with experimental data, minimising the sum of square residuals, in successive trials. The numerical results, relative to each one of the effects introduced, are compared with experimental data, giving interesting conclusions about the influence of each one of the phenomena considered. The best fitting was obtained when only the convective boundary condition was included in the diffusional model with a constant diffusion coefficient, although this model did not describe the physical phenomena well. Bananas shrink by about 43–47% their original diameter during drying. Therefore, the shrinkage assumption in the model provides greater reliability on the calculated diffusion and convective coefficients.

Introduction

The drying of agricultural products with high initial moisture content, such as fruits and vegetables, always produces a considerable shrinkage effect. This phenomenon must be included in the modelling in order to improve the physical representation of the process and to increase the confidence on the coefficients obtained, such as the diffusion coefficient.

Shrinkage has been treated theoretically in several ways in the literature. For some authors, shrinkage could be considered as directly related to the water volume removed during the process Aregba et al., 1990, Kechaou and Roques, 1989, Vagenas and Marinos-Kouris, 1991a, Balaban, 1989, Mulet et al., 1989. Other authors have proposed a further component to the shrinkage phenomenon during drying besides the volume reduction due to the loss of moisture: the mechanical forces Misra and Young, 1980, Ketelaars et al., 1992. However, the mechanical shrinkage could be neglected if the analysis is focused on drying kinetics (Ketelaars et al., 1992).

Another controversial point of diffusional drying models are the boundary conditions adopted on the material surface. The most common boundary condition used in agricultural products is the first-order condition: assuming that the material surface is in equilibrium with the air drying throughout the process. Probably, this assumption is made due to the difficulty of obtaining the mass transfer convective coefficient for biological products. However, some authors adopt a model that combines the liquid diffusion theory inside the solid, with a convective boundary condition at the material surface (Husain et al., 1973, Misra and Young, 1980, Lamberg, 1989, Kechaou and Roques, 1989, Yapar et al., 1990, Sereno and Medeiros, 1990, Parti and Dugmanics, 1990, Haghighi et al., 1990, Vagenas and Marinos-Kouris, 1991a, Vagenas and Marinos-Kouris, 1991b, Jayas et al., 1991).