Food Engineering Chapter 1 PDF

Food Engineering Chapter 1 Physical Properties of Food Materials Dr. Asma Eskhan 1  1.1 Introduction The physical properties of foods can be defined as those properties that lend themselves to description and quantific...

Food Engineering Chapter 1 Physical Properties of Food Materials Dr. Asma Eskhan 1  1.1 Introduction The physical properties of foods can be defined as those properties that lend themselves to description and quantification by physical rather than chemical means. This seemingly obvious distinction between physical and chemical properties reveals an interesting historical fact. Indeed, until the 1960s, the chemistry and biochemistry of foods were by far the most active areas of food research. The systematic study of the physical properties of foods (often considered a distinct scientific discipline called ‘food physics’ or ‘physical chemistry of foods’ ) is of relatively recent origin. The physical properties of foods are of utmost interest to the food engineer, mainly for two reasons: 1. Many of the characteristics that define the quality (e.g. texture, structure, appearance) and stability (e.g. water activity) of a food product are linked to its physical properties. 2 2. Quantitative knowledge of many of the physical properties, such as thermal conductivity, density, viscosity, specific heat, enthalpy and many others, is essential for the rational design and operation of food processes and for the prediction of the response of foods to processing, distribution and storage conditions. These are sometimes referred to as ‘ engineering properties ’ , although most physical properties are significant both from the quality and engineering points of view. 3  1.2 Mechanical Properties 1.2.1 Definitions By mechanical properties, we mean those properties that determine the behavior of food materials when subjected to external forces. As such, mechanical properties are relevant both to processing (e.g. conveying, size reduction) and to consumption (texture, mouth feel). The forces acting on the material are usually expressed as stress, i.e. intensity of the force per unit area (N.m −2 or Pa). The dimensions and units of stress are like those of pressure. Very often, but not always, the response of materials to stress is deformation, expressed as strain. Strain is usually expressed as a dimensionless ratio, such as the elongation as a percentage of the original length. The relationship between stress and strain is the subject matter of the science known as rheology. 4 We define three ideal types of deformation: Elastic deformation : deformation appears instantly with the application of stress and disappears instantly with the removal of stress. For many materials, the strain is proportional to the stress, at least for moderate values of the deformation. The condition of linearity, called Hooke’s law is formulated in Eq. (1.1): where E = Young’s modulus, Pa F = force applied, N A0 = original cross-sectional area ΔL = elongation, m L0 = original length. 5  Plastic deformation: deformation does not occur as long as the stress is below a limit value known as yield stress. Deformation is permanent, i.e. the body does not return to its original size and shape when the stress is removed. Viscous deformation : deformation (flow) occurs instantly with the application of stress, and it is permanent. The rate of strain is proportional to the stress (Chapter 2). The types of stress are classified according to the direction of the force in relation to the material. Normal stresses are those that act in a direction perpendicular to the material’s surface. Normal stresses are compressive if they act towards the material and tensile if they act away from it. Shear stresses act in a direction parallel (tangential) to the material’s surface ( Figure 1.1 ). The increase in the deformation of a body under constant stress is called creep. Figure 1.1 Types of stress The decay of stress with time, under constant strain, is called relaxation. 6 7  1.2.2 Rheological models The stress–strain relationship in food materials is usually complex. It is therefore useful to describe the real rheological behavior of foods with the help of simplified, approximate models. Those models are constructed by connecting ideal elements (elastic, viscous, friction, rupture etc.) in series, in parallel or in combinations of both. Some of these models are shown in Figure 1.2. The physical models are useful in the development of mathematical models (equations) for the description and prediction of the complex rheological behavior of foods. The rheological characteristics of fluids are discussed in some detail in a subsequent section (Chapter 2, Section 2.3). Figure 1.2 Three rheological models 8 1.3 Thermal Properties Almost every process in the food industry involves thermal effects such as heating, cooling or phase transition. The thermal properties of foods are therefore of considerable relevance in food process engineering. The following properties are of particular importance: thermal conductivity, thermal diffusivity, specific heat, latent heat of phase transition and emissivity. In addition, theoretical or empirical methods have been developed for the prediction of these properties in the light of the chemical composition and physical structure of food materials. Specific heat cp (kJ.kg−1.K−1 ) is among the most fundamentals of thermal properties. It is defined as the quantity of heat (kJ) needed to increase the temperature of one unit mass (kg) of the material by one degree (°K) at constant pressure. The specification of ‘at constant pressure’ is relevant to gases where the heat input needed to cause a given increase in temperature depends on the process. It is practically irrelevant in the case of liquids and solids. 9 The definition of specific heat can be formulated as follows: The specific heat of a material can be determined experimentally by static (adiabatic) calorimetry or differential scanning calorimetry or calculated from measurements involving other thermal properties. It can be also predicted quite accurately with the help of a number of empirical equations. The simplest model for solutions and liquid mixtures assumes that the specific heat of the mixture is equal to the sum of the pondered contribution of each component. The components are grouped in classes: water, salts, carbohydrates, proteins, lipids. The specific heat, relative to water, is taken as: salts 0.2; carbohydrate 0.34; proteins 0.37; lipids 0.4; water 1. The specific heat of water is 4.18 kJ.kg−1.K−1. 10 The specific heat of a solution or liquid mixture is therefore: where X represents the mass fraction of each of the component groups (Rahman, 1995 ). For mixtures that approximate solutions of sugar in water (e.g. fruit juices), Eq. (1.3) becomes: Another frequently used model assigns to the total dry matter of the mixture a single relative specific value of 0.837. The resulting approximate empirical expressions for temperatures above and below freezing are given in Eq. (1.5): 11 1.4 Electrical Properties The electrical properties of foods are particularly relevant to microwave and ohmic heating of foods and to the effect of electrostatic forces on the behavior of powders. The most important properties are electrical conductivity and the dielectric properties. These are discussed in Chapter 3, in relation with ohmic heating and microwave heating. 1.5 Structure Very few foods are truly homogeneous systems. Most foods consist of mixtures of distinct physical phases, in close contact with each other. The heterogeneous nature of foods may be visible to the naked eye or perceived only when examined under a microscope or electron microscope. In foods, the different phases are seldom in complete equilibrium with each other and many of the desirable properties of ‘ fresh ’ foods are due to the lack of equilibrium between the phases. 12 Following are some of the different structural elements in foods: 1. Cellular structures: vegetables, fruits and muscle foods consist in large part of cellular tissue. The characteristics of the cells and, more particularly, of the cell walls determine the rheological and transport properties of cellular foods. One of the characteristics particular to cellular foods is turgidity or turgor pressure. Turgor is the intracellular pressure resulting from osmotic differences between the cell content and the extracellular fluid. This is the factor responsible for the crisp texture of fruits and vegetables and for the ‘fleshy’ appearance of fresh meat and fish. Cellular food structures may also be created artificially. Wheat bread consists of gas-filled cells with distinct cell walls. The numerous puffed snacks and breakfast cereals produced by extrusion owe their particular crispiness to their cellular structure with brittle cell walls. 13 2. Fibrous structures: in this context we refer to physical fibers, i.e. to solid structural elements with one dimension much larger than the other two and not to ‘dietary fiber’. The most obvious of the fibrous foods is meat. Indeed, protein fibers are responsible for the chewiness of muscle foods. The creation of a man- made fibrous structure is the main challenge of the meat analog developer. 3. Gels: gels are macroscopically homogeneous colloidal systems, where dispersed particles (generally polymeric constituents such as polysaccharides or proteins) have combined with the solvent (generally water) to create a semi-rigid solid structure. Gels are usually produced by first dissolving the polymer in the solvent, then changing the conditions (cooling, concentration, cross-linking) so that the solubility is decreased. Gelation is particularly important in the production of set yogurt, dairy desserts, custard, tofu, jams and confectionery. The structural stability of food gels subjected to shear and certain kinds of processing (e.g. freezing– thawing) is an important consideration in product formulation and process design. The gelling agent in the production of jam is pectin (a complex starch). 14 4. Emulsions: emulsions are intimate mixtures of two mutually immiscible liquids, where one of the liquids is dispersed as fine globules in the other ( Figure 1.3 ). In the case of foods the two liquid media are, in most cases, fats and water. Two possibilities exist for emulsions consisting of oil and water: a. The dispersed phase is oil (oil-in-water, o/w emulsions). This is the case of milk, cream, sauces and salad dressings. Figure 1.3 Schematic structure of oil- b. The dispersed phase is water (water-in-oil, w/o in-water and water-in-oil emulsions emulsions). Butter and margarine are w/o emulsions. Emulsions are not thermodynamically stable systems. They do not form spontaneously. Emulsification requires energy Depending on the agent used, a particular input (mixing, homogenization) in order to shear one of the emulsifier may be more soluble in one phase phases into small globules and disperse them in the continuous or another. If an emulsifier is more soluble phase. Emulsions tend to break apart as the result of in water, it is more likely to facilitate the creation of oil-in-water emulsions. coalescence (fusion of the disperse droplets into larger ones) Conversely, oil-soluble emulsifiers tend to and creaming (separation of the original emulsion into a more encourage the formation of water-in-oil emulsions. concentrated emulsion or cream, and some free continuous phase). Emulsions are stabilized with the help of surface-active agents known as emulsifiers. 15 4. Foams: foams are cellular structures consisting of gas (air) filled cells and liquid cell walls. Due to surface forces, foams behave like solids. Ice cream is essentially frozen foam, since almost half of its volume is air. Foams with specific characteristics (bubble size distribution, density, stiffness, stability) are important in milk-containing beverages and beer. On the other hand, the spontaneous excessive foaming of some liquid products (e.g. skim milk) during transportation and processing may create serious engineering problems. Undesired foaming is controlled by proper design of the equipment, mechanical foam breakers or through the use of food grade chemical antifoaming (prevention) and defoaming (foam abatement) agents such as oils and certain silicone based compounds. 5. Powders: solid particles, 10 to 1000 micrometers in size, are defined as powders. Smaller particles are conventionally called ‘ dust ’and larger particles are ‘ granules ’. Some food products and many of the raw materials of the food industry are powders. Powders are produced by size reduction, precipitation, crystallization or spray drying. One of the main issues related to powders in food engineering is the flow and transportation of particulate materials. 16  Food Engineering Chapter 1 Physical Properties of Food Materials Dr. Asma Eskhan 1 1.6 Water Activity 1.6.1 The importance of water in foods Water is the most abundant constituent in most foods. Indicative values of water content in a number of food products are shown in Table 1.1. Classification of foods into three groups according to their water content (high, intermediate and low moisture foods) has been suggested ( Franks, 1991 ). Fruits, vegetables, juices, raw meat, fish and milk belong to the high moisture category. Bread, hard cheeses and sausages are examples of intermediate moisture foods, while the low moisture group includes dehydrated vegetables, grains, milk powder and dry soup mixtures. The functional importance of water in foods goes far beyond its mere quantitative presence in their composition. On one hand, water is essential for the good texture and appearance of fruits and vegetables. In such products, loss of water usually results in lower quality. On the other hand, water, being an essential requirement for the occurrence and support of chemical reactions and microbial growth, is often responsible for the microbial, enzymatic and chemical deterioration of food. 2 3  It is now well established that the effect of water on the stability of foods cannot be related solely to the quantitative water content. As an example, honey containing 23% water is perfectly shelf stable while dehydrated potato would undergo rapid spoilage at a moisture content half as high. To explain the influence of water, a parameter that reflects both the quantity and the ‘effectiveness’ of water is needed. This parameter is water activity. 1.6.2 Water activity, definition and determination Water activity, aw , is defined as the ratio of the water vapor pressure of the food to the vapor pressure of pure water at the same temperature. where: p = partial pressure of water vapor of the food at temperature T p0= equilibrium vapor pressure of pure water at temperature T 4 The same type of ratio also defines the relative humidity of air, RH (usually expressed as a percentage): where: p′ = partial pressure of water vapor in air. If the food is in equilibrium with air, then p = p′. It follows that the water activity of the food is equal to the relative humidity of the atmosphere in equilibrium with the food. For this reason, water activity is sometimes expressed as the equilibrium relative humidity, ERH. Many of the methods and instruments for the determination of water activity are based on Eq. (1.8). A sample of the food is equilibrated with a small head-space of air in a close chamber and then the relative humidity of the headspace is measured by an appropriate hygrometric method such as the ‘chilled mirror’ technique ( Figure 1.4 ).5 Typical water activity values of some food products are given in Table 1.2. Figure 1.4 Measurement of water activity 1.6.3 Water activity: prediction The principal mechanisms responsible for the depression of vapor pressure of water in foods are solvent–solute interaction, binding of water molecules to the polar sites of polymer constituents (e.g. polysaccharides and proteins), adsorption of water on the surface of the solid matrix and capillary forces ( Le Maguer, 1987 ). In high moisture foods, such as fruit juices, the depression may be attributed entirely to water– solute interaction. If such foods are assumed to behave as ‘ideal solutions’, then their water vapor pressure obeys Raoult’s law (see Section 13.2), as in Eq. (1.9): where xw is the water content (in molar fraction) of the food. It follows that the water activity of an ideal aqueous solution is equal to the molar concentration of water xw. The water activity of high moisture foods (with aw of 0.9 or higher) can be calculated quite accurately by this method. 7 EXAMPLE 1.1 Estimate the water activity of honey. Consider honey as an 80% w/w aqueous solution of sugars (90% hexoses, 10% disaccharides). Solution: The composition of 100 g of honey is: Assuming Raoult’s law, the water activity of honey is equal to the molar fraction of water: 8  As the water content is reduced, water binding by the solid matrix and capillary forces become increasingly significant factors and overshadow water–solute interaction. Furthermore, the assumption of ideal solution behavior can no longer be applied because of the reduced concentration of the liquid phase. The relationship between water content and water activity, aw = f(X) becomes more complex. This is discussed in the next section. Water activity is temperature dependent. Considering the definition of water activity, as given in Eq. (1.6), one would be tempted to conclude the opposite. Temperature affects both p and p0 in the same manner (the Law of Clausius-Clapeyron), therefore their ratio should not be affected by the temperature. This is true for the liquid phase and, indeed, the water activity of high moisture foods is affected by temperature very slightly, if at all. The situation is different at lower levels of water content. Temperature affects not only the water molecules but also the solid matrix interacting with water. Therefore, temperature affects water activity at low moisture levels where adsorption and capillary effects are strong. The direction and intensity of temperature effects are not predictable. 9 1.6.4 Water vapor sorption isotherms The function representing the relationship between water content (e.g. as grams of water per gram of dry matter) and water activity at constant temperature is called the ‘water vapor sorption isotherm’ or a ‘moisture sorption isotherm’ of a food. The general form of a hypothetical sorption isotherm is shown in Figure 1.5. Sorption isotherms of a large number of foods have been compiled by Iglesias and Chirife (1982). Sorption isotherms are determined experimentally. Basically, samples of the food are equilibrated at constant temperature with atmospheres at different known relative humidities. After equilibration, the samples are analyzed for water (moisture) content. Each pair of ERH–moisture content data give one point on the isotherm. The experimental methods for the determination of sorption isotherms fall into two groups, namely, static and dynamic procedures. In static methods, weighed samples of food are placed in jars, over saturated aqueous solutions of different salts and left to equilibrate at constant temperature. At constant temperature, the concentration of saturated solutions is constant and so is their water vapor pressure. 10 Figure 1.5 General form of a sorption isotherm 11 12  The relative humidity of the atmosphere in equilibrium with saturated solutions of some salts is given in Table 1.3. In dynamic methods, the sample is equilibrated with a gas stream, the relative humidity of which is continuously changed. The quantity of moisture adsorbed or desorbed is determined by recording the change in the weight of the sample. The two curves shown in Figure 1.5 indicate the phenomenon of ‘hysteresis’ , often encountered. One of the curves consists of experimental data points where the food sample came to equilibrium by losing moisture (desorption). The other curve represents points obtained by the opposite path, i.e. by gain of moisture (adsorption). The physical explanation of the sorption hysteresis has been the subject of many studies. Generally, hysteresis is attributed to the condensation of some of the water in the capillaries ( Labuza, 1968 ; Kapsalis, 1987 ; deMann 1990 ). The observation that, depending on the path of sorption, food can have two different values of water activity at the same moisture content casts doubt on the thermodynamic validity of the concept of sorption equilibrium ( Franks, 1991 ). 13  Numerous attempts have been made to develop mathematical models for the prediction of sorption isotherms ( Chirife and Iglesias, 1978 ). Some of the models developed are based on physical theories of adsorption (see Chapter 12). Others are semi-empirical expressions developed by curve fitting techniques. One of the best-known models is the Brunauer-Emmett-Teller (BET) equation. The basic assumptions on which the BET model is based are discussed in Section 12.2. Applied to water vapor sorption, the BET equation is written as follows: where: X = water content, grams water per gram of dry matter Xm = a parameter of the equation, interpreted as the value of X for the saturation of one monomolecular layer of water on the adsorbing surface (the BET monolayer) C = a constant, related to the heat of adsorption. 14 To find Xm and C from experimental sorption data, the BET equation is written as follows: If the group Φ is plotted against aw, a straight line is obtained ( Figure 1.6 ). Xm and C are calculated from the intercept and the slope. The BET model has been found to fit well sorption isotherms, up to water activity values of about 0.45. 15 EXAMPLE 1.2 Following are 3 points from the sorption isotherm of potato at 20°C: Estimate the monolayer value of potato, based on the data. Solution: We calculate the group Φ in Eq. (1.11), as a function of aw. We find: 16 The linear plot of Φ versus a w is found to be: Φ = 16.8 aw + 0.57 Solving for C and Xm we find C = 30.47 and Xm 0.058 Figure 1.6 Linearization of the BET equation with 3 experimental points 17 Another equation which is often used to predict sorption isotherms is the Guggenheim- Anderson-de Boer (GAB) model shown below: where C and K are constants, both related to the temperature and heat of adsorption. The range of applicability of the GAB equation is wider than that of the BET model. 18 1.6.5 Water activity: effect on food quality and stability Bacterial growth does not occur at water activity levels below 0.9. With the exception of osmophilic species, the water activity limit for the growth of molds and yeasts is between 0.8 and 0.9. Most enzymatic reactions require water activity levels of 0.85 or higher. The relationship between water activity and chemical reactions (Maillard browning, lipid oxidation) exhibit more complex behavior with maxima and minima (Figure 1.7). Figure 1.7 Relative rate of deterioration mechanisms as affected by water activity. A: Lipid oxidation; B: Maillard browning; C: Enzymatic activity; D: Mold growth; E: Bacteria growth. The Maillard reaction is an organic chemical reaction in which reducing sugars react with amino acids to form a complex mixture of compounds. This reaction is responsible for the characteristic flavor and aroma of browned food. 19 Lipids (also known as fats) are components of plant (e.g., vegetable oils) and animal tissues (e.g., meat, eggs, milk). 20  Food Engineering Chapter 1 Physical Properties of Food Materials Dr. Asma Eskhan 1 1.7 Phase Transition Phenomena in Foods 1.7.1 The glassy state in foods With few exceptions, foods should be regarded as metastable systems capable of undergoing change. Stability is a consequence of the rate of change. In turn, the rate of change depends on molecular mobility. In recent years, molecular mobility has become a subject of strong interest among food scientists. The subject is particularly important in solid and semi-solid foods with low to intermediate water content. In the majority of foods belonging to this category, the interaction between polymeric constituents, water and solutes is the key issue in connection with molecular mobility, diffusion and reaction rates. Accordingly, concepts and principles developed by polymer scientists are now being applied to foods ( Slade and Levine, 1991, 1995 ). 2  Consider a liquid food product, such as honey, consisting of a concentrated aqueous solution of sugars. The physical properties and stability of such a solution depend on two variables: concentration and temperature. If the concentration is increased by slowly removing some of the water and the temperature is lowered gradually, solid crystals of sugar will be formed. If the process of concentration and cooling is carried out under different conditions, crystallization will not take place, but the viscosity of the solution will increase until a rigid, transparent, glass-like material will be obtained. The familiar transparent hard candy is an example of glassy (vitreous) food. The glassy state is not limited to sugar–water systems. Intermediate and low moisture foods often contain glassy regions consisting of polymer materials (e.g. gelatinized starch) and water. The phenomenon of passage from the highly viscous, rubbery semi-liquid to the rigid glass is called ‘glass transition’ and the temperature at which that occurs is the glass transition temperature, Tg. 3  Physically, a glass is an amorphous solid. It is also sometimes described as a supercooled liquid of extremely high viscosity. Conventionally, the viscosity assigned to a glass is in the order of 1011 to 1013 Pa.s., although it is practically impossible to verify this convention experimentally. The molecules of a glass do not have an orderly arrangement as in a solid crystal, but they are sufficiently close and sufficiently immobile to possess the characteristic rigidity of solids. Because of the negligible molecular mobility, the rate of chemical and biological reactions in glassy material is extremely low. The rigidity of the glassy regions affects the texture of the food. Staling of bread is due to the transition of the starch–water system from rubbery to glassy state. The crunchiness of many snack products is due to their glassy structure. Staling is a chemical and physical process in bread and similar foods that reduces their palatability. Staling is not simply a drying-out process due to evaporation. One important mechanism is the migration of moisture from the starch granules into the interstitial spaces, degelatinizing the starch; stale bread's leathery, hard texture results from the starch amylose and amylopectin molecules realigning themselves causing 4 recrystallisation. 1.7.2 Glass transition temperature The different physical states of an aqueous solvent–solute system capable of forming an amorphous solid and the processes of passage from one state to another have been described by Roos and Karel (1991a) in a frequently cited diagram ( Figure 1.8 ). Figure 1.8 State diagram of a carbohydrate solution. (Adapted from Roos and Karel, 1991a.) 5  Boiling of a liquid or melting of a crystal are ‘thermodynamic’ phase transitions, also known as ‘first order transitions’. They occur at a fixed, definite set of conditions (temperature, pressure), independent of rate. The phases in transition are mutually in equilibrium. In contrast, glass transition is of kinetic nature. It does not involve large step changes in properties and does not require a considerable latent heat of transition. The glass transition temperature of a given rubbery material is not a fixed point. It varies somewhat with the rate and direction of the change (e.g. rate of heating or cooling), therefore the procedure for its determination has to be specified exactly. Glass transition temperatures of pure dry sugars are given in Table 1.4. Glass transition temperature is strongly dependent on concentration. Dilute solutions have lower Tg. This led Roos and Karel (1991b) to conclude that water acts on the amorphous food as a plasticizer in a polymer system. The effect of concentration on Tg is shown in Figure 1.9 for a solution of sucrose. 6 7 Figure 1.9 Effect of temperature on T g of sucrose solutions 8  It has been suggested that the glass transition temperature of a blend can be predicted, using the Gordon–Taylor equation, borrowed from polymer science and shown in Eq. (1.13): where: Tg = glass transition temperature of the mixture Tg1, Tg2 = absolute glass transition temperatures (°K) of component 1 and 2, respectively. The subscript 2 refers to the component with the higher Tg. w1 , w2 = weight fractions of component 1 and 2, respectively k = a constant. A simpler approximate expression is the Fox equation ( Schneider, 1997 ): 9  According to Johari et al. (1987) , the Tg of water is 138°K or −135°C. Glass transition temperatures of some carbohydrates are shown in Table 1.4. The viscosity of solutions increases sharply as Tg is approached. Near Tg, the effect of temperature on viscosity does not comply with the Arrhenius law but follows the Williams–Landel–Ferry (WLF) relationship ( Roos and Karel, 1991c), shown in Eq. (1.15): where μ and μg stand for the viscosity at the temperatures T and Tg respectively. Just as water activity, glass transition temperature has become a key concept in food technology, with applications in quality assessment and product development. Since the glassy state is considered as a state where molecular mobility is at a minimum, it has become a custom to study food properties and stability, not as a function of the temperature T, but as a function of the difference T − Tg ( Simatos et al., 1995 ). 10  Several methods exist for the determination of Tg. The results may vary somewhat, depending on the method used. The most commonly applied method is differential scanning calorimetry (DSC). DSC measures and records the heat capacity (i.e. the amount of heat necessary to increase the temperature by 1 degree Celsius) of a sample and of a reference as a function of temperature. A sharp increase or decrease in heat capacity indicates an endothermic or exothermic phase transition at the temperature where this occurs. In the case of first order transitions, such as melting, the amplitude of the change is considerable. In contrast, glass transition is detected as a subtle inflexion in the heat capacity curve ( Figure 1.10 ). Since the change occurs over a temperature range and not sharply at one temperature, the decision where to read Tg is subject to interpretation. The two most common conventions are the mid-point of the step and the point representing the onset of the transition ( Simatos et al. 1995 ) 11 Figure 1.10 Glass transition temperature from DSC plot 12