Keywords

1 Introduction

Food is a multicomponent system that mainly comprises protein, carbohydrate, fat, and water. During food processing and preservation, various physical changes (e.g., melting, crystallization, glass transition) occur in food products, affecting their quality. This chapter specifically examines the effects of physical changes on the quality of dry and frozen food products. In the first part, effect of water content on the glass transition temperature (T g) of dry food products (dry fruits and cookie) and trial to predicting T g is presented for understanding physical changes of dry food products induced by water sorption. In the second part, physical changes of frozen food products during cooling and frozen storage, such as ice crystallization, freeze concentration , eutectic separation , glass transition, recrystallization of ice crystal , and sublimation of ice , and their effects on quality of frozen food are reviewed.

2 Glass Transition of Dry Food Products

2.1 Role of Glass Transition in Dry Food Products

Dry food products are commonly in an amorphous state. Therefore, they undergo a glass to rubber transition (glass transition) during dehydration–rehydration processing. This transition involves a physical change between a solid-like (glassy) state and a liquid-like (rubbery) state and causes a drastic change in rheological properties. The temperature at which glass transition occurs is known as the glass transition temperature (T g). Glass transition can also occur with no change in temperature if the water content changes because T g of hydrophilic amorphous solids decreases with increased water contents. The water content at which T g is 25 °C is described as the critical water content (W c) . The glass transition behavior can be described as a T g curve (effect of water content on T g), as presented in Fig. 21.1. In the area under the T g curve, the food product is in a glassy state. Glassy food products have a hard or brittle texture because of their high elasticity (Payne and Labuza 2005a, b). In addition, glassy food products are expected to have greater physical stability than rubber ones because of their low molecular mobility. In the area above the T g curve, the food products are in a rubber state. Rubbery food products, which have a soft or ductile texture, undergo various physical changes (e.g., stickiness, caking, collapse, sugar crystallization) because of their high molecular mobility (Roos 1995; Jaya and Das 2004; Cano-Chauca et al. 2005; Palzer 2005; Harnkarnsujarit and Charoenrein 2011a, b; Zou et al. 2013). To control their texture and physical stability, it is important to ascertain the T g curve of dry food products.

Fig. 21.1
figure 1

Effect of water content on the T g (T g curve) of dry food products. The water content at which T g is 25 °C is described as the critical water content (W c)

Full size image

2.2 Glass Transition of Dry Fruits

Dry fruits include large amounts of low-molecular-weight carbohydrates (e.g., sucrose, fructose, glucose). Because these carbohydrates have a low T g , dry fruits readily undergo a glass transition because of water sorption.

The T g of dry fruits is commonly evaluated using differential scanning calorimetry (DSC) . Glass transition can be detected as an endothermic shift in a DSC thermogram. However, glass transition observed in the first scanning shows an endothermic peak because of the enthalpy relaxation effect depending on the thermal history of glassy samples (Kawai et al. 2005). Enthalpy relaxation is a process wherein the excess enthalpy of glass in a non-equilibrium thermodynamic state decreases spontaneously toward its equilibrium value. The T g value is known to be affected to various degrees by the enthalpy relaxation effect (Haque et al. 2006). To cancel the thermal history of glassy samples on T g, DSC measurement is taken again after the first scanning. Therefore, T g can be determined from the onset point of the shift in the second scanning.

Effect of water content on the T g of dry mango pulp is shown in Fig. 21.2. The T g decreased with increased water content because of water plasticizing effect. This behavior can be summarized as a T g curve, as depicted in Fig. 21.2. The solid line represents fitting of the Gordon–Taylor equation (Eq. 21.1) to the T g data.

$$ {T}_{\mathrm{g}}=\frac{W_{\mathrm{s}}{T}_{\mathrm{g}\mathrm{s}}+k{W}_{\mathrm{w}}{T}_{\mathrm{g}\mathrm{w}}}{W_{\mathrm{s}}+k{W}_{\mathrm{w}}} $$
(21.1)

Therein, W s and W w, respectively, represent the weight fractions of mango pulp and water. T gs and T gw are T g for anhydrous mango pulp and water, respectively, whereas k is a constant. T gs was evaluated experimentally. T g2 was set to 136 K, as reported in the literature (Johari et al. 1987; Sastry 1999). When k was determined as a fitting parameter, the T g curve presented in Fig. 21.2 was obtained. The k represents the sensitivity to water plasticizing: higher k is associated with greater T g depression caused by water sorption. From the T g curve, W c can be evaluated.

Fig. 21.2
figure 2

Effect of water content on the T g of dry mango pulp. The T g was evaluated using DSC . The solid line was given by the Gordon–Taylor equation with two parameters (anhydrous T g and k). The anhydrous T g was evaluated experimentally and k was determined as a fitting parameter

Full size image

For low-molecular-weight carbohydrates including polyol, a linear relation is known to exist between anhydrous T g (°C) and k (Roos 1995) as

$$ k=0.0293\times \mathrm{anhydrous}{T}_{\mathrm{g}}+3.61 $$
(21.2)

The major constituent of dry fruits is low-molecular-weight carbohydrates. Therefore, similar behavior would be expected in dry fruits. The relation between anhydrous T g and k for dry fruits of various types (Bai et al. 2001; Fabra et al. 2011; Kasapis et al. 2000; Khalloufi et al. 2000; Kurozawa et al. 2012; Moraga et al. 2004, 2006, 2011; Mosquera et al. 2010, 2012; Oikonomopoulou & Krokida 2012; Silva et al. 2006; Sobral et al. 2001; Sonthipermpoon et al. 2006; Syamaladevi et al. 2009; Telis 2006; Telis & Martínez-Navarrete 2009; Vásquez et al. 2013; Wang et al. 2008; Zotarelli et al. 2017) is presented in Fig. 21.3. Although some outliers (e.g., camu camu, persimmon, plum) were observed, there was a roughly linear relation.

$$ k=0.0825\times \mathrm{anhydrous}\ {T}_{\mathrm{g}}+0.005\pm 1.5749 $$
(21.3)

As described above, a T g curve can be described by the Gordon–Taylor equation with two parameters (anhydrous T g and k). W c can be evaluated from the T g curve. Consequently, Eq. 21.3 is expected to be useful for the prediction of T g depression and physical changes of dry fruits induced by water sorption.

Fig. 21.3
figure 3

Relation between anhydrous T g and k for various types of dry fruits. The solid line is a mean linear relationship. The upper and lower dotted lines are the variation of the mean linear relationship. Most of the data exist in the gray zone. The applied ranges of anhydrous T g and k were 21–75 °C and 2.0–8.0, respectively

Full size image

2.3 Physical Modification of Dry Fruits Based on T g

As described above, dry fruits undergo physical deterioration of various types at temperatures higher than T g . To improve their physical stability, it is important to elevate the T g of dry fruits by additives with high-T g materials. Maltodextrin (MD) , which has a much higher T g (100–243 °C) than low-molecular-weight carbohydrates (Goula and Adamopoulos 2008), has been used as the physical modifier of dry fruits. For example, some related reports on it are being used in camu camu (Silva et al. 2006), tomato pulp (Goula and Adamopoulos 2008), grapefruit juice powder (Telis and Martínez-Navarrete 2009), borojó powder (Mosquera et al. 2010), orange juice powder (Goula and Adamopoulos 2010), strawberry (Mosquera et al. 2012), and mango powder (Zotarelli et al. 2017). In an earlier study conducted by the authors (Fongin et al. 2017), the effect of MD addition on the glass transition properties of dry mango pulp was investigated. It was demonstrated that the anhydrous T g increased with an increase in MD content (Fig. 21.4). A systematic study showed that an abrupt anhydrous T g change occurred between 60% and 70% MD. Similar observations were also obtained for the T g change for glucose–MD and maltose–MD mixtures (Kawai and Hagura 2012). These results suggest that amorphous mixtures have heterogeneous molecular dynamics. MD is plasticized by the mango pulp above 70% MD content. Consequently, the anhydrous T g decreases continually with increasing mango pulp content. In the region, glass transition of the mango pulp–MD mixture will occur cooperatively. Part of the mango pulp is excluded from the mango pulp–MD domain when the MD content decreases to less than 60%. Then, not only a mango pulp–MD domain but also a mango pulp-rich domain (lower anhydrous T g ) is formed in the amorphous system. Consequently, an abrupt anhydrous T g change can be observed at MD contents of 60–70%.

Fig. 21.4
figure 4

Effect of maltodextrin (MD) addition on the anhydrous T g of mango pulp. An abrupt anhydrous T g change occurred between 60% and 70% MD. The suggested turning point is indicated by dotted line. Inserted drawing is a model of heterogeneous molecular dynamics in the amorphous system

Full size image

The k value of dry mango pulp also increased with an increase in MD content. The relation between anhydrous T g and k obeys Eq. 21.3 below MD content of 60%. By the further addition of MD , the relation between anhydrous T g and k deviates from Eq. 21.3 (for dry fruits) and approaches Eq. 21.2 (for low-molecular-weight carbohydrates). This can be related to the suggestion described above: the mango pulp-rich domain is lost from the amorphous mixture system at MD contents between 60% and 70%.

2.4 Glass Transition Properties of Dry Bakery Products

The T g of amorphous materials has commonly been evaluated using DSC . However, when a more complex multicomponent system (e.g., a cookie) is subjected to DSC measurement, a continuous thermal response is observed. Also, the endothermic shift associated with the glass transition is overlapped. In addition, because starch-based food intrinsically shows a small and broad endothermic shift, it is difficult to evaluate the T g from the DSC thermogram. In such cases, thermomechanical approaches such as thermomechanical compression test (TMCT), thermal mechanical analysis (TMA), and dynamic thermal mechanical analysis (DMTA) are useful. For example, T g values of barley (van Donkelaar et al. 2015), rice (Thuc et al. 2010), dairy powder (Hogan et al. 2010), peas (Pelgrom et al. 2013), chocolate wafers (Payne and Labuza 2005b), and abalone (Sablania et al. 2004) have been found using thermomechanical approaches. An earlier study conducted by the authors (Kawai et al. 2014) established thermal rheological analysis (TRA) , which is almost equivalent to TMCT and TMA. The TRA curve shows a clear force drop associated with the glass transition: mechanical T g can be determined from the onset point. It is confirmed that TRA is an effective tool to evaluate T g of bakery products including cookies. It is noteworthy that T g determined by thermomechanical approaches does not always agree with that determined by DSC (Sandoval et al. 2009). Thermal T g observed by DSC has a clear physical meaning, representing the temperature at which the viscosity is approximately 1012 Pa∙s (Angell et al. 1994). Mechanical T g is more sensitive to experimental conditions such as mechanical force, heating rate, and sample quantity (Lacík et al. 2000; Ross et al. 2002; Boonyai et al. 2007).

Cookie dough mainly comprises wheat flour, sugar, butter, and egg. Three types of sugar composition (sucrose alone, sucrose containing 40% trehalose, and sucrose containing 40% sorbitol on a dry weight basis) were used and baked for cookie production. Effect of water content on the mechanical T g of the handmade cookie samples is presented in Fig. 21.5. The mechanical T g of the cookie samples decreased with an increase in water content because of water plasticizing effect, similar to the case for dry fruits. The onset point would be independent of water content if the force drop observed in the TRA curve resulting from the melting of fat. The fact that the force-drop point decreased with an increase in water content reveals that the rheological response resulted from glass transition of the cookie samples and not simply by the melting of fat. The water content of cookies is approximately 3–4% under the initial conditions. Consequently, it was noted that normal cookies (sucrose alone) were in a glassy state at 25 °C, which explains why they had a brittle texture. When the water content of normal cookies became higher than W c (5.04 g-H2O/100 g-DM), they transitioned into a rubber state. Therefore, the brittle texture changed to a ductile one. Trehalose-added cookies had higher mechanical T g and W c than normal cookies because trehalose has higher anhydrous T g (114 °C) than sucrose (62 °C). Consequently, it was concluded that trehalose can improve the physical stability of cookies; trehalose led to the brittle–ductile texture change occurring at higher water content than in normal cookies. By contrast, sorbitol-added cookies had lower mechanical T g and W c than normal cookies had because sorbitol has lower T g (−9 °C) than sucrose. Considering these findings, the sorbitol-added cookies showed ductile texture, even at low water content. This texture is also important to control the cookie texture.

Fig. 21.5
figure 5

Effect of water content on the mechanical T g of cookie samples with varying sugar compositions. The mechanical T g was evaluated using TRA

Full size image

It is noteworthy that anhydrous mechanical T g for cookies increased with an increase in the T g for the sugar composition used for cookie preparation. When the anhydrous mechanical T g for cookies (extrapolated value) was shown against anhydrous T g for sugar composition (experimental value obtained by DSC ), a linear relation (anhydrous mechanical T g for cookie = 0.3245 × anhydrous T g for sugar composition +71.62) was identified, which indicates that the mechanical T g of cookies can be characterized by the T g of sugar composition. Cookies comprise a continuous glassy sugar or toffee-like matrix containing embedded starch granules, an undeveloped gluten network, and fat (Slade and Levine 1994; Chevallier et al. 2002). Consequently, the physical properties of cookies depend strongly on the sugar composition. Numerous publications describe the T g of carbohydrate materials and their mixtures. They are expected to be useful for predicting the T g of cookies.

3 Physical Changes of Frozen Food Products During Cooling and Frozen Storage

3.1 Ice Crystallization in Frozen Food Products During Cooling

During cooling of food, crystallization of water component occurs after supercooling and nucleation. The size of ice crystal in frozen food strongly affects its quality. Rapid freezing gives smaller ice crystals and even distribution of ice crystals in food . The rate of ice nuclei formation is larger at lower temperatures. Ice crystal growth is most accelerated around the equilibrium ice melting temperature, a zone of maximum ice crystal formation (0 to −5 °C). Therefore more nuclei occur, and ice crystal growth is suppressed when a food is passed quickly through a maximum ice crystal production zone by rapid cooling. Rapid freezing is desirable for most foods. Slow freezing produces larger ice crystals, inhomogeneous locations of ice crystals, shrunken appearance of the microstructure, and usually lower quality than rapid freezing. For cryopreservation of living cells and microorganism, slow freezing might sometimes be preferred (Franks 1985). Rapid freezing usually causes intracellular ice crystal formation, which is lethal for cells. Slow freezing initiates the formation of nuclei outside cell and extracellular ice crystallization occurs. Furthermore, cells are dehydrated during the process of extracellular ice crystallization. Extracellular crystallization and dehydration of cells suppress intracellular crystallization. Practically speaking, an aid of cryoprotectant such as dimethyl sulfoxide (DMSO), glycerol, and trehalose is needed for successful cryopreservation of living cells and microorganisms. Cryoprotectants permeate into a cell or dehydrate a cell by osmotic pressure action. Consequently, water contents in the cell are reduced, which suppresses intracellular crystallization (Franks 1985).

3.2 Freeze Concentration

Upon cooling food products below 0 °C, ice formation and separation of water from food solids occur. Consequently, solute concentration of unfrozen part increases as ice contents increase with declining temperatures. This phenomenon is called “freeze concentration .” Generally, a rate of chemical reaction decreases at lower temperatures. However, the chemical or biochemical reaction rate increases sometimes because of lowered temperatures when the effect of increase of reactant by freeze concentration overcomes that from lowered temperatures (Fennema et al. 1973; Franks 1985). The freeze concentration engenders a change of pH or electrolyte concentration in food, which can cause protein denaturation (Fennema et al. 1973; Franks 1985).

Freeze concentration has been used for the concentration of liquid foods such as fruit juices (Deshpande et al. 1982; Bayindirli et al. 1993; Miyawaki et al. 2016), vegetable juices (Miyawaki et al. 2005), and dairy products (Hartel and Espinel 1993). Among the methods of concentrating liquid food, freeze concentration presents several benefits: low energy requirements, low process temperature preventing undesirable chemical and biochemical changes, and minimal loss of flavors and aromas (Ramteke et al. 1993; Liu et al. 1997).

3.3 Eutectic Separation

Lowering the temperature continuously, the concentration of freeze-concentrated phase reaches its solubility at last. At this point, the concentration of solute cannot become any higher without increasing the temperature (Franks 1985). When more ice is formed, the amount of dissolved solute is too high and solute crystals start to form simultaneously. This phenomenon is called “eutectic separation .” The temperature at which eutectic separation occurs is known as the eutectic point. The eutectic point of sodium chloride is −21.1 °C. Practical eutectic separation probably occurs only rarely in most frozen foods because of the complex nature of food materials (Roos 1995). However, it is well known that lactose in ice cream is often crystallized during frozen storage by supersaturation. It does not crystallize immediately after freezing because of kinetic constraints, but it crystallizes after a certain period of storage. Subsequently, the crystals grow (Roos 1995). When lactose crystals grow so large that they can be detected in the mouth, the smooth texture of ice cream is transformed into a sandy texture: an unacceptable product (Marshall and Arbuckle 2000).

3.4 Glass Transition by Freeze Concentration

It is quite often true that the solute does not crystallize at the eutectic point from aqueous solution system within a practically realistic time scale. Consequently, freeze concentration proceeds continuously by lowering the temperature. Finally, the freeze-concentrated phase turns into a glassy state at a specific temperature (T g′). Ice crystallization ceases at T g′ because the high viscosity of the freeze-concentrated matrix suppresses diffusion of water molecules to the surface of existing ice crystal (Franks 1985; Roos 1995). Frozen food below T g′ is a mixture of the glassy amorphous substance and ice crystals. It is generally acknowledged that the deterioration of frozen food during storage is minimized by maintaining it below T g′ (Franks 1985; Agustini et al. 2001). It is noteworthy that exposure of a freeze-drying material to temperatures higher than T g′ causes melting of ice crystals and softening of the freeze-concentrated matrix (Roos 1995). Because the freeze-concentrated matrix cannot support its own weight, collapse, reduced water removal rate, and inferior product quality occur (Roos 1995).

3.5 Ice Recrystallization During Frozen Storage

Ice crystal size and shape are unstable during frozen storage even if the temperature is constant. They change continuously through the process of “recrystallization,” which includes any change in number, size, shape, orientation, or perfection of crystals following the completion of initial solidification (Fennema et al. 1973). Recrystallization proceeds as a result of minimization of surface free energy of the entire crystal phase (Fennema et al. 1973; Hartel 1998). Recrystallization of ice crystals during storage and distribution is a major cause of deterioration in frozen foods, especially frozen desserts. Generally, the recrystallization of ice crystals in frozen foods is characterized as an increase in the mean size of ice crystals (Fennema et al. 1973; Hartel 2001) (Fig. 21.6).

Fig. 21.6
figure 6

Example of ice recrystallization during storage. Sample, 25% glucose; storage temperature, −10 °C; storage time (a) 0 min, (b), 54 min, (c) 109 min, (d) 252 min; observation procedure, 2 μl sample solution enclosed between two coverslips (diameter, 12 mm) was placed on the sample stage to be frozen at −22 °C. After 10 min, the temperature was elevated to the observation temperature (−10 °C) at a rate of 1.5 °C/min. Finally, after the observation temperature was reached, ice crystal images were periodically photographed (Hagiwara et al. 2006)

Full size image

Three recrystallization mechanisms are likely to occur during the conventional storage of frozen food: migratory, isomass, and accretion (Fennema et al. 1973; Hartel 2001). Migratory recrystallization refers to the trend for larger crystals in a polycrystalline system to grow in size at the expense of smaller crystals. Smaller crystals cannot bind their surface water molecules as tightly as larger ones because of the higher curvature and higher surface free energy. Therefore, the water molecules on the surface of smaller crystals tend to transfer to the surface of larger ones through the freeze-concentrated matrix , engendering the growth of larger crystals and disappearance of smaller ones. A similar process called isomass recrystallization can occur in a single crystal. Consider a single separated crystal with a rough surface. The part of the surface with higher curvature cannot bind surface water molecules as tightly as a smoother surface. As a result, the rougher surface becomes smoother. Accretion recrystallization is the process by which two crystals that are in mutual contact grow together into one larger crystal. Because the contact point has apparently high curvature and because it is not as stable as the rest, neck formation occurs by transportation of water molecules to this region, which engenders growth into one crystal.

When temperature fluctuations occur during storage, the recrystallization process is enhanced by melt–refreeze recrystallization, which is more important for ice cream texture/shelf life in practical situations than during isothermal processes (Hartel 2001).

To describe a change of ice crystal size by isothermal recrystallization, the following two equations have been used.

$$ {R}^3={R_0}^3+ kt $$
(21.4)
$$ {R}^2={R_0}^2+ kt $$
(21.5)

Therein, R stands for the averaged size of ice crystals, R 0 signifies the averaged size of ice crystals at time t = 0, and k denotes the isothermal recrystallization rate constant. The recrystallization process in ice cream and its model solutions can be described well by Eq. 21.4 (Sutton et al. 1996; Hagiwara et al. 2006; Klinmalai et al. 2017). Equation 21.5 was applied for ice recrystallization in frozen beef (Bevilacqua and Zaritzky 1982).

Recrystallization of ice crystals in frozen foods has been studied extensively. The group of Zaritzky et al. studied the recrystallization of frozen beef and its model system (Bevilacqua and Zaritzky 1982; Martino and Zaritzky 1988, 1989). The effects of storage temperature, temperature fluctuation, sweeteners, and stabilizers on the recrystallization rate of ice crystals in ice cream have also been reported (Donhowe and Hartel 1996a, b; Hagiwara and Hartel 1996; Miller-Livney and Hartel 1997).

To elucidate recrystallization phenomena, it is necessary to comprehend not only the correlation between production conditions and recrystallization rate constant but also the molecular-level mechanisms accounting for different recrystallization rates. A better understanding of the molecular mechanisms of recrystallization is expected to facilitate the systematic prediction and control of recrystallization behavior for various frozen foods. Recently it has been recognized that the concept of water mobility is useful for predicting and controlling the recrystallization rate in frozen foods (Ablett et al. 2002; Hagiwara et al. 2006, 2009; Klinmalai et al. 2017). Reportedly, recrystallization rates in a series of frozen saccharide solutions (Table 21.1) increased concomitantly with increasing the diffusion coefficients of water component in the freeze-concentrated matrix. A direct relation was found between the recrystallization rate and the diffusion coefficient (Hagiwara et al. 2006, 2009) (Fig. 21.7). This relation indicates that the self-diffusion coefficient of water component in freeze-concentrated matrix is a useful parameter for predicting and controlling recrystallization. The 1H spin–spin relaxation time T2 of water components in freeze-concentrated matrix also shows good correlation with the recrystallization rate constant (Ablett et al. 2002; Klinmalai et al. 2017).

Table 21.1 Sample sugar solutions. Values of ice content and concentration of freeze-concentrated matrix were calculated by using literature values of freezing point depression versus solute concentration (Young and Jones 1949; Young et al. 1952; Young 1957; Lide 2003)
Full size table
Fig. 21.7
figure 7

Plots of recrystallization rate constant K of ice crystals in various frozen saccharide solutions as a function of (a) temperature and (b) diffusion coefficient of water component in a freeze-concentrated matrix . Sample saccharide solutions are listed in Table 21.1. The recrystallization rate constant K of ice crystals was obtained by using Eq. 21.4 (Hagiwara et al. 2006). The diffusion coefficient of water component was measured by pulse field gradient stimulated gradient echo proton NMR method (Hagiwara et al. 2006)

Full size image

3.6 Effects of Glass Transition on Recrystallization of Ice Crystals

Based on the concept of water mobility in freeze-concentrated matrix , the recrystallization of ice crystals is expected to be strongly suppressed below T g′ because the molecular mobility is strongly restricted. Hagiwara et al. (2005) investigated the recrystallization of ice crystals in a 30% sucrose solution at temperatures of −21 to −50 °C, including temperatures around T g′ (−32 °C). As the storage temperature decreased, a rapid decline in the recrystallization rate was observed between −29 and −35 °C (Fig. 21.8). This result is consistent with the concept of the glass transition of the freeze-concentrated matrix . It is noteworthy that even at −50 °C, at which the freeze-concentrated matrix is regarded as being in glassy state, an increase in the mean crystal size was observed within 20 h storage. This observation suggests that, over a realistic storage period, deterioration by recrystallization might occur, even in the glassy state. In general, it is believed that food in a glassy state is very stable because its molecular motion is restricted severely. However, in the field of polymer science, it is well-recognized that molecular movement leading to macroscopic structural relaxation over a practical period remains below the glass transition temperature because of the non-equilibrium nature of glassy substances (Matsuoka 1992; Yoshida 1995; Tiemblo et al. 2002). Molecular mobility in glassy food and food component carbohydrates with low moisture content has also been investigated recently (Hancock et al. 1995; Urbani et al. 1997; Kim et al. 2003; Hashimoto et al. 2004; Kawai et al. 2005). Regarding frozen food systems, molecular mobility in the freeze-concentrated phase of trehalose (Pyne et al. 2003) and sucrose solutions (Inoue and Suzuki 2005) below T g′ was investigated on the concept of enthalpy relaxation. Molecular motion in a freeze-concentrated solute matrix might be sufficient to cause ice recrystallization over a realistic storage period, even in a glassy state.

Fig. 21.8
figure 8

Arrhenius plot of recrystallization rate of ice crystals in frozen 30% sucrose solutions

Full size image

3.7 Recrystallization of Ice Crystals in the Presence of Antifreeze Protein (AFP)

Recently, many antifreeze proteins (AFPs) have been extracted from a variety of sources (Davies and Hew 1990; Graether et al. 2000; Griffith and Yaish 2004; Regand and Goff 2006; Kawahara et al. 2009). They are now anticipated as additives for suppressing ice recrystallization process. AFPs suppress ice crystal growth by an adsorption–inhibition mechanism (Raymond and DeVries 1977). They inhibit thermodynamically favored ice crystal growth by their adsorption to specific planes of ice crystals (Raymond et al. 1989).

To put AFP into practical use as a recrystallization suppressor, evaluation of their suppression ability is fundamentally important. A typical AFP extracted from polar fish, AFP type I, actually suppresses the recrystallization of ice crystals efficiently, with quite a low concentration (Fig. 21.9) (Hagiwara et al. 2011). The recrystallization rate constant of 33 wt% sucrose solution containing 1 μg/ml AFP type I was about 13% of that of sample without AFP (Table 21.2). No marked reduction of recrystallization rate was observed for samples containing 0.1 μg/ml or 0.01 μg/ml AFP type I.

Fig. 21.9
figure 9

Effects of AFP type I addition on recrystallization of ice crystals in 33% sucrose solution at −10 °C. The concentrations of AFP type I; (a) 0 μg/ml, (b) 0.01 μg/ml, (c) 0.1 μg/ml, (d) 1 μg/ml

Full size image
Table 21.2 Summary of recrystallization rate constants for the samples with different AFP type I concentrations
Full size table

3.8 Sublimation of Ice in Frozen Food Products

When frozen foods are stored without appropriate moisture-proof package, the food surface is dehydrated. Consequently the surface has an opaque appearance called “freezer burn” (Fennema et al. 1973). Freezer burn results from the sublimation of ice on the surface of frozen foods when the water vapor pressure of the ice is higher than that in the environmental air (Fennema et al. 1973). The temperatures of specific parts of the freezer shelf, such as the heat exchanger, are lowest: lower than the stored frozen foods. Therefore, the situation described above normally occurs in most cases. Vapor from the frozen food surface adheres to the surface of the lower temperature part in the shelf of freezer, which is familiar to us as “frost.” The sublimation of ice enhances many deteriorative processes such as lipid oxidation, color change, protein denaturation, odorous emissions, and the decline of nutritional value. Therefore, suppression of ice sublimation is necessary for the long-term preservation of frozen foods.

4 Conclusion

As reviewed above, melting, crystallization, and glass transition occur in the food products during food processing and preservation and affect their quality. This chapter will help to control and predict the quality of dry and frozen food products. In addition, it is important to understand the adaptation mechanisms of organism in extreme cold and desiccation as seen in the other chapters.