Heat Transfer In Food Processing 6u6h30

Heat transfer in food processing: ensuring product quality and safety

Heat transfer to foods is commonplace but critical; heating develops flavour and texture and ensures product safety. The food industry must ensure that all parts of the product have all been processed sufficiently, without unacceptable loss of quality. Conventionally, food is significantly over-processed to ensure safety. This paper reviews some problems which heat-transfer engineers face in the food industry, and recent progress in understanding them. Examples are taken in (i) safety: the problems of predicting and validating food processes, using a combination of modelling, visualisation and enzyme temperature–time indicators which through a process and show how much cooking they have received, (ii) quality: the combination of reaction engineering and heat transfer in the generation of flavours and colours for ales, and (iii) process operation: how to clean food process plant.

1. Introduction 1.1. Types of thermal process in the food industry Thermal processing is ubiquitous in food processing. Many of the commonest food-processing operations, such as canning, baking and pasteurisation, rely on heating: For preservation, i.e. to kill bacteria and inactivate enzymes, such as the pasteurisation of milk and the sterilisation of canned food. Here the aim is to deliver the required microbial kill with as little damage as possible. • To develop taste and flavour, such as when cooking meats and vegetables, where in addition to sterilisation heat is required to carry out physical changes to the food. • To develop the structure of the material, such as in baking of bread or biscuits, where heating acts both to change the starch structure and function and also to develop the bubble structure within the material (for examples, see [11]).

There are also processes, such as drying and frying, where heat and mass transfer are coupled. Many processes were developed empirically, and only after many years have attempts been made to develop an engineering understanding. The consumer desires minimally processed products that retain nutrition and flavour, display an acceptable shelf life and are convenient to prepare. 1.2. Heat transfer and reactions in foods Foods are physically complex fluids and soft solids, with rheological and material properties that are time- and process-dependent and which have a natural variation. It is thus difficult to specify the physics in such a way as to allow accurate prediction of the process. Some safety margin is added to ensure that every part of the material is safe. Modelling methods have been developed to help in process design or as part of a control specification; purely empirical correlations or graphical solutions were followed by models based around very simple approximations, such as simple geometries (spheres or cylinders) and uniform physical properties. In food processing, the effect of thermal processing, where T is a function of time t, is commonly described through the integrated lethality calculated using ………………… where z is the increase in temperature that increases the rate by a factor of 10, and Tref is a reference temperature. F has units of time; it is the length of time that food has to be held at the reference temperature to obtain the same effect as the actual process. Estimates of F values were first proposed by Ball [5], where for canned low-acid foods (pH < 4.5), the proposed reference temperature (Tref) was 121.1 C (250 F) and z = 10 C was used, based on estimates of the death of Clostridium botulinum. A process was then considered safe if the slowest heating point of a can reached an F value of 3 min. Eq. (1) is an experimental fit, a local approximation to the Arrhenius expression: although only accurate over a narrow temperature range, it fits microbial behaviour in the narrow range of temperature appropriate to cooking. The equation assumes that microbial death, which is not strictly true. More accurate models for microbial growth and death are available and can be used in predicting the results of processes, for example Bellara et al.,

[6] modelled the growth of pathogenic bacterial species after thermal treatment within agar cylinders. Investigations modelled the inactivation of the bacteria and explored the notion that slow heating rates (62 C/min) may increase thermotolerance in potentially pathogenic bacteria [21]. Any type of model requires accurate data, often lacking in the food industry. This paper reviews some of the recent attempts that have been made to understand heat transfer and its consequences in food processing, using examples from recent work at Birmingham. 2. Prediction of food safety: models and validation 2.1. Models for food processes Before personal computers became common, calculating integrated lethality from Eq. (1) was complex and time consuming even for simple products. In place of repetitive manual calculations ‘‘simpler’’ methods were used, for example, F could be estimated by measuring temperature, calculating lethality and using graphical solutions (see [20]). Real products are rarely regular, have thermal properties which vary with temperature and have different heat resistances along the boundary. For example, in retorts, condensing steam is used and condensation will affect the uniformity of heat transfer to the product surface [37]. Complex thermal effects in conduction can result from the polymorphic nature of food components; for example Tewkesbury et al. [32] used a FE model to predict the cooling of chocolate (a complex mixture of triglycerides with six polymorphic forms) within a polycarbonate mould. Both conduction cooling of chocolate, and the change in the effective specific heat capacity of chocolate were modelled as a function of temperature and cooling rate. For most commercially processed products heating is rarely by pure conduction. In canning, for example, heat transfer occurs by a combination of conduction and natural convection within the can. CFD can solve for the flow field within the can. Kumar et al. [18] describe the heating of canned viscous liquids that have temperature-dependant viscosity. In their model an initial conductive heat-transfer phase reduces the viscosity of the liquid at the periphery of the can. This less viscous and more buoyant hot liquid rises and then re-circulates to the centre of

the can. Abdhul Ghani et al. [1,2] provided estimates of the axial velocity of the fluid in cans and modelled bacterial inactivation. In recent work, Cox et al. [15] and Bakalis et al. [4] used positron emission particle tracking (PEPT) to examine the three-dimensional flow of liquids within rotating cans (including cans under heat-transfer conditions). The method can measure flow velocities of tracers even through several centimetres of stainless steel. Fig. 1 shows the flow pattern inside a can with a 10% headspace (volume of air in the can). The flow pattern is three-dimensional, as fluid elements travel from the centre of the can to the walls and back. These flow patterns be followed by PEPT thanks to its temporal and spatial resolution through metal, and the isokinetic nature of the flow followers. PEPT thus allows the mixing patterns within the material to be quantified, leading to new ways of validating model predictions. The next stage is to model the system in which cooking takes place. Varga et al. [36] used FE methods to model horizontal cascading water retorts, and verified their model with industrial scale measurements. It was stated that a reliable mathematical solution required a model combined with an appropriate statistical method to estimate real world variability, as well as an understanding of the quantitative implications of the variance upon the system. Similarly, Verboven et al. [38] used CFD to examine temperature deviations within an industrial scale forced convection oven. The limitations of conduction heat transfer in processing large solids are obvious: by the time the centre is cooked the outside is overcooked. Alternative processes, such as microwave or electrical resistance (ohmic) cooking heat volumetrically. The problem is in ensuring uniform heating and in validation. Excellent discussions of the problems inherent in modelling microwave processes, and in validating them for commercial production, are given by Bows [7] and Bows et al. [8]. The interaction between the applied field and the food material is key. Multimode resonant applicators, used in all domestic ovens and almost all industrial applications produce one resultant heating pattern from the complex interaction of the field, material and applicator. Strategies to minimise temperature differences (rotating turntables, mode stirrers or moving the product) are not

always sufficient to overcome undesirable effects. For electrical heating, complex temperature distributions can be found, such as by Kemp et al. [17]. Local heating and cooling effects have been shown by thermocouple measurements and in elegant MRI experiments by Ruan et al. (1999) 2.2. Time–temperature indicators One technology that allows for the verification of processes is the use of time– temperature indicators (TTIs). Taoukis [31] reviewed the use of TTIs for product safety assurance and shelf life during transportation, storage and sale. The other aspect of TTIs use is as thermal process indicators to define the efficiency of pasteurisation and sterilisation processes. A successful system allows the thermal history of processed food to be evaluated without temperature loggers, whilst mimicking the thermal characteristics of any given target attribute (e.g. microbiological load or vitamin retention). Biochemical TTIs offer such a technology. The most familiar and widely used are encapsulated spores. Non-pathogenic analogues of important food contaminating organisms (e.g. Bacillus stearothermophilus in place of Clostridium botulinum) are encapsulated into alginate spheres, which are placed at the centre of the foodstuff, and the product processed. Comparison of pre- and postprocess numbers of spores allow determination of the overall kill for the process (as with Eq. (3)). The preparation and analysis is non-trivial [34]. More recently spores are being replaced with chemical or biochemical indicators, such as purified bacterial enzymes. TTI particles are placed in the food and after heating the proportion of denatured enzyme can be assayed and applied to Eq. (2), which is equivalent to Eq. (1). ……………………….. where F is the integrated lethality for the process (in minutes), DT is the decimal reduction time, the process time which reduces the value by a factor of 10, Afinal is the concentration of enzyme that has not been denatured and Ainitial is the activity of unprocessed enzyme.

Tucker [34] describes one such example (Bacillus amyloliquifaciens a-amylase) whose thermal characteristics (DT = 18.7 min and z = 9.7) closely match those of Clostridium botulinum. It is of great for examining pasteurisation treatment, as it ensures a sufficient cook to kill most aerobic pathogens (e.g. Listeria and Salmonella) as well as the psychrotrophic Clostridia. Maesmans et al. [22] and Loey et al. [35] describe other potential measurement systems. For conductioncooked products bacterial enzyme based TTIs have provided a rapid method of process validation [34].In continuous processing such an approach can be difficult, although considerable efforts have been made to determine temperature– time distributions in convective and flowing food systems. For TTIs to be truly useful in such processes the trajectories of the TTI particles through the equipment (in relation to the heating surfaces) needs to be known, perhaps by using methods such as PEPT. Combination of model and validation is needed to be able to demonstrate product safety to the food industry. 3. A DETERMINATION METHOD OF THE ENERGETIC BALANCE AND COEFICIENT OF THE HEAT TRANSFER In the drying process, the total invested energy is spent on: water evaporation, heating of drying material and heat losses. Energetic balances show appropriate relations between the total invested energy, utilized energy and heat losses during the drying process. The energetic balances can be useful when showing the dryer condition diagnosis. 5. Heat transfer in food processing This paper has concentrated on the heating of foods. The problems of predicting the rate of heating and the resulting changes in product quality and sterility are significant, in part because of the complexity of food materials and their imperfectly known physical properties. Other areas have not been described; for example, refrigeration and freezing are significant industrial processes which are of particular importance in the safe storage of foods, see, for example, [3]. The industry is a very conservative one. Developing new processes, or using the results of models to change processing schedules, is very difficult to do, and requires a lot of data for safety before any process can be approved. As the margins in the industry are very low, few food manufacturers can afford to do the sort of modelling that would be commonplace in, for example, the chemical industry. The problems of the food industry are similar to those of other industries that deal with complex fluids. The above does not claim to be comprehensive.

The interested reader should try some of the journals in the area, such as the Journal of Food Engineering, Innovative Food Science and Emerging Technologies, and Transactions Part C of the Institution of Chemical Engineers for papers which go into more detail. ////////////////////////////////////////////////////////////////////////////////////////////////// ////////////////

Convection Drying in the Food Industry Experimental and theoretic research was conducted and the results were implemented in a real industrial environment on a convection dryer with pneumatic transport of material. The numeric values are given for optimum parameters of drying, energetic characteristics and balances as well as the models of heat transfer. Accomplishment of the heat transfer in these systems is based on the principle of direct of dried material and warm air. Then, an intensive transfer of heat and mass is accomplished. This work presents the results of research which can be useful in deg and construction of such dryers in the food industry. It refers to the technical and engineering characteristics of the dryer, energetic balances and coefficients of heat transfer. 1. INTRODUCTION Application of the convection pneumatic dryers is represented especially in food industry in plants for industrial processing of grains (wet milling processing of wheat and corn). Generally, such dryers can be used for drying of meal-like and fine-kernel materials. Simple construction and a relatively low consumption of energy have enabled successful application of such dryers in the above stated industrial branches. The construction of the convection dryer enables simultaneous pneumatic transport of wet material and its drying. In these dryers, a continuous drying of loose materials is being made, the concentration being (0,052,00) kg of material per 1 kg of air. Average particle size of the drying material can be (0,05-2,00) mm. The circulation speed of the heated agent of drying (air or gas) in the dryer is (10-30) ms1. The initial moisture of the material dried can be w1= (35-40) %, and the remaining moisture after drying is usually w2 = (1015) %. The specific consumption of energy is usually (3900-5040) kJ kg -1, of evaporable water. Efficiency of such dryers is evaluated according to the thermal degree of utilization which is within the limits of (66-75) %, depending of the drying system (indirect or direct drying). The quantity of evaporated moisture in the dryer pneumatic pipe is approximately (300-350) kg m -3h-1. The drying time in these dryers is very short, only several

seconds, therefore they can be used for drying of the materials susceptible to high temperatures in a short drying period of time. 2. DESCRIPTION OF EXPERIMENTAL PLANT Experimental research is made in the convection pneumatic dryer, Figure 1. Drying agents are heated with the gas burner (1). Drying is performed in the direct of warm gases with the moist material. The principle of direct drying is represented here. The drying material is corn bran. Dosing of moist material to the dryer is performed through the rotation dozer (2) with the capacity of m 1 = 1640 kg h-1, through the auger conveying system, as given in the scheme of experimental equipment in Figure 1. Auger conveyers have a role of mixing the moist material with the dried material from the pipeline for recirculation (8). In such a way, a homogenous moist material is obtained at the dryer inlet. In the table 1, the characteristics of convection pneumatic dryer are given. Moist material is transported via hot air – the drying agent through the dryer pneumatic pipe (3), it es through the dryer head (4) and goes to the cyclone separators (5) for separation of dried material, and the hot gases are expelled by a ventilator (6), into the atmosphere. The material which is not dried completely falls into the conical bottom of the dryer head (4) due to the higher content of moisture, the gravitation takes it back to additional drying through the pipeline (8). The dried material is transported from the cyclone via auger conveyors and through a separate line of pneumatic transport up to the material warehouse department. During drying, the determined fuel - gas consumption is B = 293 m3h-1. Table 2 contains average values of the results of measuring the air temperature – the drying agent and moisture of dried material. Experimental measuring is being made in the approximate stationary conditions of the dryer operation. The stationary conditions mean the stationary conditions during a longer period of the dryer operation and greater number of measuring (where non-stationary conditions of the process are excluded during the realistic conditions of the dryer operation) 3. A DETERMINATION METHOD OF THE ENERGETIC BALANCE AND COEFICIENT OF THE HEAT TRANSFER In the drying process, the total invested energy is spent on: water evaporation, heating of drying material and heat losses. Energetic balances show appropriate relations between the total invested energy, utilized energy and heat losses during the drying process. The energetic balances can be useful when showing the dryer condition diagnosis. The difference in the air enthalpy at the inlet and at the outlet of the dryer:

……………. Quantity of evaporated water: Total heat quantity Quantity of drying air: Specific consumption of energy: Total heat power of drying: Total coefficient of the heat transfer Heat for drying, i.e. its convective part consists of the heat for water evaporation (Qw) and heat for heating of the drying material (Qs), meaning, without heat losses (Qg): During convective drying the following equation of the heat transfer is applied as well: Coefficient of the heat transfer through convection (hc) is a relevant parameter of Nusselt criteria of heat transfer: Based on the analysis of influential parameters of heat transfer, the following equation is being acquired of Nusselt type: Based on it, the research results can be shown with the help of correlation equation of Nusselt type: Reynolds number is being determined with the following equation: The constant (K) and the exponent (a) are being determined by the method of the least difference squares. 4. THE RESULTS OF EXPERIMENTAL RESEARCH OF ENERGETIC BALANCE HEAT TRANSFER COEFFICIENTS AND DISCUSSION Experimental research on the convection pneumatic dryer, Figure 1., was aimed at determining the energetic balance, specific consumption of energy, thermal degree of utilization and other relevant parameters of drying, according to the literature (Prvulovic, 2004). The results of the energetic balance are given in the Table 3. 5. CONCLUSION This work presents the experimental and theoretic research of relevant parameters of drying on the convection pneumatic dryer in the food industry. Based on the analysis of energetic balance, the heat force of drying has been determined Q u = 2210 kW, specific consumption of energy q =3918 kJ kg -1 of evaporable water, as well as the thermal degree of utilization ηT = 0,74. Energy balance of the dryer can serve in evaluation of power condition of the dryer as well as in reviewing of the possibility of rational consumption of energy. A significant share of the energy

during drying is forwarded to transfer of heat to the material, necessary for evaporation of moisture and heat for the breaking of connection forces of moisture with the basis of the material to be dried. Specific consumption of energy and quality of dried material are basic data which characterize the results of drying on the convection dryer. By following and control of these parameters in the drying process, the optimum consumption of energy is provided as well as the quality of dried material. On the basis of the results of research of energetic balance and the results of measuring the temperature of the drying agent, the total coefficient of the heat transfer is being determined in the convection dryer in the amount of h u = 342 Wm-2K-1, and the coefficient of the heat transfer by convection h c = 242 Wm-2K-1. The effects of the heat losses during drying are expressed through the separate value hu – hc = 100 Wm-2K-1, so called coefficient of the heat transfer for the heat losses together with the outlet air and the heat transfer by conduction and radiation through the dryer pipe. In such a way the effects of the heat transfer are determined as well as the basic parameters of the heat transfer. The acquired results of research are based on the experimental data from the industrial dryer. Based on that, the results of research have a value of use, i.e. they are useful to the designers, manufacturers and beneficiaries of these and similar drying systems as well as for the educational purposes. The results of research can also be used for: determination of dependence

Modelling of Heat and Mass Transfer in Food Products Abstract: The use of the finite-element method for understanding and analyzing the freezing and drying processes of food products is in focus in this paper. The objective of this study is to develop a model that can predict temperature distribution and weight loss of food products during the freezing and drying processes. The problem was solved by utilizing heat, mass transfer and moving mesh model. In order to predict the heat transfer, the thermo physical properties of food products are evaluated by a compositionbased prediction method where the thermal properties of a given food product are evaluated based on major components found in the food. The mass transfer problem in the unfrozen region is solved by the Fickian model and in the frozen region with the receding front model that assumes that the water movement becomes immobilized and that the moisture sublimates first from the surface then through the dried area in the food.

From the model the temperature, moisture and evaporation rates are obtained as a function of location and time.

1. Introduction Food drying and food freezing are the most common and widely used methods to preserve food for later use. Nevertheless, the processing of food is complex and still little understood. The processing and storage conditions of the food products are highly related to the product end quality and also to economic factors. The main objective of this study is to develop a general model that can for the food characteristics, the processing and storage conditions on the freezing and drying times of the food products. That is done by applying some of the methods suggested in the literature.

2. Methods 2.1 Thermal properties of food Knowledge of the thermal properties of food is required in order to evaluate the heat transfer in food products. Therefore, a composition-based prediction method was used to obtain the thermal properties of food. The thermal properties are described by major components found in food such as: proteins, fat, carbohydrate, fibers, ash and water/ice. The composition data is stated as mass fraction and given in [1] and [2] for different food products such as vegetables, fruits, meat, etc. The thermo physical properties of each component are subsequently described by tempe-rature dependent mathematical models in the temperature range of -40oC to 150oC. The properties were evaluated in the global expression window in Comsol. Knowing the composition as well as the thermal properties of the different components makes it possible to find the apparent properties of the food product. A different approach is used for the evaluation of the properties in the unfrozen and frozen region of the food. In order to evaluate which approach is to be used, determination of the initial freezing point is required. Pham Q.T. [3] has suggested the following equation for evaluation of the initial freezing point based on the composition data of the product: