DINIZ, J. F.B.; http://lattes.cnpq.br/6618902993266324; DINIZ, Jacqueline Félix de Brito.
Résumé:
In recent years, with the growing awareness of environmental preservation and pollution control, interest in the use of natural fibers in polymer composites has increased significantly. However, vegetable fibers are highly hygroscopic and their use, when wet, in the manufacture of composites, strongly affects the properties of these materials. Seeking improvement to the drying process of vegetable fibers, this work presents a numerical and experimental study of the drying process of fibrous bodies. A three-dimensional transient mathematical model was developed to predict the heat transfer and mass to simulate the distribution of moisture content, water vapor concentration and temperature inside the fibrous body in the form of a parallelepiped, considering the symmetry around the center of the solid, constant thermophysical properties and volume. The numerical solution of the diffusion equation was obtained by numerical method of the finite volumes, considering the condition of convective contour on the surface of the solid. For validation of the numerical methodology drying experiments were carried out with sisal fibers of the variety Agave sisalana with an average moisture content of 11,19% (d.b.). The fibers were subjected to oven drying with forced air circulation at 50, 60, 70, 80 e 90°C. Thin layer mathematical models (exponential and logarithmic) were proposed to represent experimental data on the drying of sisal fibers, and their adjustment was performed by non - linear regression analysis. Experimental results of drying and heating kinetics of the fibers were presented and analyzed. It was verified that the curves of moisture loss and temperature of the sisal fibers were influenced by the drying temperature, showing a gradual variation with the drying time, being more accentuated in the higher temperatures of the drying air. Theoretical results of the moisture content and temperature of the sisal fibers were compared to the experimental results, allowing the determination of the mass diffusion coefficient and the convective co-efficients of mass and heat. A good agreement was observed between the predicted and experimental results, characterized by correlation coefficients close to 1.0.
