Deep frying

Deep frying is one of the most used methods in the food processing industry. Though practically any food can be fried, French fries are probably the most well-known deep fried products. The popularity of French fries stems from their unique taste and texture, a crispy outside with a mealy soft interior, but also because of the ease and speed of preparation. However, despite being a practical and easy method, the fundamental phenomena that occur during frying are very complex. This thesis aimed at gaining a deeper understanding of the frying of French fries. This was done at the product level, with regards to heat transfer, moisture loss, oil uptake and crust formation, and at the process level, which encompasses the oil movement in a frying unit and the consequent oil-fry interactions. Firstly a numerical model was developed to describe the water evaporation during frying (<strong>Chapter 2</strong>). Though various models exist for describing moisture loss, they all use constant values for the heat transfer coefficient. However, the heat transfer coefficient actually varies greatly due to the varying degrees of turbulence, induced by the vapour bubbles escaping from the fry surface. Therefore, the model in this thesis incorporated an evaporation rate dependent heat transfer coefficient. Other than the varying heat transfer coefficient, the model was heat transfer dependent, with a sharp moving evaporation boundary and Darcy flow describing the flow of water vapour through the crust. The model was successfully validated against experimental results for moisture loss and temperature profiles in the fry. For oil uptake during frying, a pore inactivation model from membrane technology was adopted (<strong>Chapter 3</strong>). In membranes, pores will inactivate when the transmembrane pressure becomes too low. In fries, this can be translated as pores in the crust inactivating when the evaporation rate becomes too low. As pores stop expelling water vapour, oil can migrate into the fry. The model also took into account the lengthening of the pores with increasing crust thickness, allowing for more oil uptake in inactivated pores. The model fitted well with experimental data for oil uptake during frying. Also, the pore inactivation model better described oil uptake during the initial stages of frying, where the evaporation rate is still relatively high, compared to the linear relation between oil uptake and moisture content, which is usually assumed in literature. Both the influences of frying temperature and moisture content on crust structure and consequent textural properties were studied (<strong>Chapter 4</strong>). The crust structure was visualized and quantified using X-ray tomography (XRT), which uses multiple 2D X-ray pictures of a rotated sample to reconstruct a 3D density map. Textural properties, like hardness and crispness, were quantified using force deformation curves from a texture analyser. Moisture loss was shown to greatly increase porosity and pore size in fries. More crispy behaviour was also shown for higher moisture loss, though not significantly at moisture contents close to the initial moisture content. Though increased frying temperatures also showed an increased porosity and pore size, there was no significantly observed increase in crispness. This is most likely because the texture analysis was not sensitive enough to discern any increased crispness for porosities below a certain degree. Strikingly, for frying temperature around 195 &deg;C, a decrease in crispness was observed. These samples visually also showed more plastic behavior. The most likely cause for this is degradation of sucrose, which happens around 186 &deg;C, and consequent caramelization of glucose, thus increasing the glass transition temperature. At the process level, oil flow and fry quality distribution were investigated using a pilot scale cross-flow fryer (<strong>Chapter 5</strong>). Oil circulation velocities were varied to observe the initial fluidization behavior of the fry bed through an observation window. This fluidization behavior was well described by the Ergun equation, modified for non-spherical particles. The distribution in moisture content of the fries was used as an indicator for quality distribution. Though increased oil circulation initially increased the homogeneity of the moisture content, upon fluidization the homogeneity actually decreased. Image analysis of fries before and after frying showed local packing of fries around their fluidization point. This was due to the non-spherical shape of the fries, making them more sensitive to channelling. The results obtained in this thesis were finally discussed, together with the possibility to also model the process scale of the frying process (<strong>Chapter 6</strong>). The possibility of modelling the oil flow through a packed bed of fries, and the free-convective heat transfer during frying, using a CFD software package (STARCCM<strong>+</strong>) was shown. Additionally, the possibility of linking oil flows computed using CFD to the general models developed in this thesis was discussed. Modelling the momentum transfer of the expelled vapour bubbles to the oil, but also the movement of the fries themselves is still a faraway goal. However, a multiphase model that can describe both the entire frying setup as the consequent individual fry parameters would be invaluable. &nbsp