Research Article Simultaneous Aerodynamic and Thermal Analysis during Cooling of Stacked Spheres inside Ventilated Packages Aerodynamic and thermal analysis during forced convection cooling of produce was conducted by modeling coupled airflow and heat transfer. Air velocities and heterogeneity indexes were predicted for different configurations of package openings at an airflow rate of 0.022 m 3 /s. Predicted temperature profiles were compared with experimental data for model validation. Good agreement between model prediction and measured data was obtained. The results showed that air- flow distribution during the process was not homogeneous. More uniform air- flow distribution was obtained by increasing the vent area from 2.4 to 12.1 %. The highest cooling heterogeneity index (108 %) was recorded at 2.4 % vent area whereas the lowest heterogeneity index (0 %) was detected in a package with 12.1 % vent area. Proper package vent design is necessary to provide more uni- form cooling operation during the forced convection cooling process. Keywords: Airflow, Convection, Heat transfer, Modeling, Package design Received: June 09, 2008; accepted: August 12, 2008 DOI: 10.1002/ceat.200800290 1 Introduction Fruits and vegetables are cooled from ambient temperature to an optimal temperature before storage or shipment in order to minimize their deterioration after harvest. The forced convec- tion cooling process is the most common technology used for this purpose [1, 2]. However, it induces strong heterogeneities of the thermal treatment due to heterogeneous airflow distri- bution at different locations in ventilated packages [3–5]. Commodities located behind blind walls may not be suffi- ciently cooled while others exposed to higher velocities are over-cooled. Therefore, package vent design is a very critical factor influencing airflow and heat transfer patterns during the forced convection cooling process and, therefore, affecting the efficiency of the system [4, 6, 7]. Package vents must be suffi- ciently large and properly distributed on package walls to pro- vide uniform cooling operation, ensuring adequate mechanical resistance [7–10]. The difficulty in measuring air velocity within ventilated packages has caused limited progress in improving the design and efficiency of the forced convection cooling process. Some of the previously attempted strategies can not accurately mea- sure or predict air velocity inside ventilated packages during the process [3, 7, 10]. However, recent advances in computa- tional resources have provided powerful tools to obtain de- tailed aerodynamic and thermal analysis by modeling coupled airflow and heat transfer during the process. There are two typical methods for modeling forced convec- tion cooling of produce, namely the porous medium approach (single-phase or two-phase) and direct numerical simulation (DNS). Severe simplifications of the porous medium ap- proach, such as local thermal equilibrium in single-phase models and continued lumping of transport processes in the two-phase models, obscure the physical basis of the models leading to considerable errors [11–13]. Additionally, the two- phase porous medium approach neglects the internal produce gradient. This is questionable when the difference between center and surface temperature of produce are sufficiently large; as the case in the transient forced-air precooling process [13]. The porous medium assumption is also questionable when the package-to-product diameter ratio is below 10 [13, 14], which is a common case for fruits and vegetable packages [15]. Direct numerical simulation performs a complete time-de- pendent solution of the Navier-Stokes and continuity equa- tions. The value of such simulations is obvious. They can be viewed as an additional source of experimental data that are taken with inconspicuous measuring techniques. This is espe- cially desirable in obtaining information about essentially im- measurable properties such as pressure and velocity fluctua- © 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim http://www.cet-journal.com Jalal Dehghannya 1 Michael Ngadi 1 Clement Vigneault 1,2 1 Department of Bioresource Engineering, McGill University, Montreal, Canada. 2 Agriculture and Agri-Food Canada, Saint-Jean-sur- Richelieu, Canada. – Correspondence: Dr. M. Ngadi (michael.ngadi@mcgill.ca), Department of Bioresource Engineering, McGill University, Ste-Anne de Bellevue, Montreal, QC, Canada H9X 3V9. Chem. Eng. Technol. 2008, 31, No. 11, 1651–1659 1651