Freeze-Drying Process Design

22 Freeze-drying Process Design Cristina Ratti

Introduction Freeze-drying or lyophilization is often regarded as the best method of water removal to obtain final products of the highest quality. Because of the absence of liquid water and the low temperatures required for the process, most of the deterioration reactions and microbiological activities are prevented, which gives a final product of excellent quality. The solid state of water during freeze-drying protects the primary structure and the shape of the product, with minimal reduction in volume. Freezedried products have a long shelf-life without refrigeration, 2 years for a product with a 2% residual moisture content being usual (Williams-Gardner, 1971). This technique has been applied with success to diverse biological material, such as meat, coffee, juices, dairy products, cells and bacteria and is now standard practice in the production of penicillin, protein hydrolysates, hormones, blood plasma and vitamin preparations. The application of freeze-drying to food products has traditionally been confined to the production of heat- or oxygen-sensitive foodstuffs, or those foods having a special end-use such as space foods, military or extreme-sport foodstuffs and instant coffee (Ratti, 2001). Recently, however, the market for “natural” and “organic” products has been increasing strongly, along with consumer demand for foods with minimal processing and high quality but without the presence of preservatives. The market for higher

Handbook of Food Process Design, First Edition. Edited by Jasim Ahmed and Mohammad Shafiur Rahman. © 2012 Blackwell Publishing Ltd. Published 2012 by Blackwell Publishing Ltd.

622

Handbook of Food Process Design: Volume I

quality food powders or ingredients is not only increasing in volume but also diversifying (Brown, 1999). Despite many advantages, freeze-drying has always been recognized as the most expensive process for manufacturing a dehydrated product. The high operating and maintenance costs are the main problems with the process. The long drying times under continuous vacuum increases energy consumption enormously and makes this process considerably more expensive compared with drying at atmospheric pressure. These reasons have limited the wide application of freeze-drying to the food industry. In this chapter, the basis of freeze-drying will be analyzed, plus some hints about process design regarding operating parameters such as temperatures and pressure. It is important to note that this chapter is intended to help people working primarily in the food industry. Practical use of the glass transition temperature concept to interpret food quality, particularly in relation to the freeze-drying process, is also discussed. Finally, the latest innovations in freeze-drying and their application to food materials are analyzed in order to draw conclusions on the state-of-the-art and future of the process.

Underlying Principles of Freeze-drying Figure 22.1 presents the water phase diagram (pressure versus temperature), which indicates the conditions for the existence of the liquid, vapor and solid phases of water. Three important lines shown in this figure mark the passage from solid to vapor (sublimation), from liquid to vapor (evaporation) and from solid to liquid (fusion). Point T in Figure 22.1 represents the triple point of water (at 0.01 °C and

Pressure (kPa) C

Liquid Solid

A

101.330 EVAPORATION FUSION

Vapor

0 612 0.612 T SUBLIMATION

D 0.01

100 T Temperature (oC)

Figure 22.1

Phase diagram of water. T, triple point; C, critical point.

Freeze-drying Process Design

623

0.612 kPa) where the three phases coexist, while point C is the critical point of water (374 °C and 22 060 kPa). Freeze-drying mainly uses the sublimation phenomenon (at temperatures lower than 0.01 °C and vapor pressures below 0.612 kPa) to eliminate most of the water in a product. In Figure 22.1, if a product at the pressure and temperature corresponding to ambient conditions (point A) is to be freeze-dried, it will follow the path from point A to point D, i.e. the product should first be frozen by decreasing its temperature, then the water vapor pressure should be lowered below the pressure corresponding to the triple point and finally some heat should be supplied to help the ice to convert into vapor by sublimation. After all the ice has been sublimated (and during sublimation), desorption of nonfreezable water occurs. Therefore, we may say that three important steps characterize the freeze-drying process: freezing, sublimation (or primary drying stage) and desorption (or secondary drying stage). Figure 22.2 shows a schematic diagram of a food product during freeze-drying at different stages of the process. Although freeze-drying could take place at atmospheric pressure in an inert gas atmosphere, most freeze-drying operations are carried out under vacuum. In Figure 22.2, the frozen food is placed in a vacuum chamber on a shelf plate (or heating plate) which supplies the necessary energy for sublimation and desorption by conduction (qc). Also, the product can receive heat from the top shelf and the surroundings by radiation (qR). Convection is rare because very few fluid molecules are available under vacuum so the purely convective heat transfer coefficient should be negligible in high vacuum situations (i.e. freeze-drying). Thus, heat in the freeze-drying chamber is mainly transferred to the product by radiation and/or conduction from the shelf plates. However, it should be taken into account that conduction

VACUUM Heating plate

qR

Frozen F core

Frozen food

qC

t = t0

Dry layers

Heating plate

t = t1

t = t2

Figure 22.2 Schematic representation of a food product during freeze-drying at initial time (t = t0), during sublimation (t = t1), and when desorption takes place (t = t2). qc, conduction heating; qR, radiation.

624


heat transfer from the bottom shelf plate to the product can be reduced significantly if there is not good contact between the product and the heating plate, since conduction will only take place through contact points. On the other hand, heat transfer by conduction is predominant within the product. As the product receives heat, sublimation is initiated (t1). Drying is faster during primary drying due the availability of large amounts of unbound water in the frozen state. Ice sublimation leaves a porous dry layer that increases as drying proceeds (receding front). During sublimation, two distinct phases separated by a receding front are present in the product: the dry layers and a frozen core (Figure 22.2). During secondary drying (t2), bound water has to be lost. A major portion of the bound water is in the unfrozen sate and the drying rate is very slow (Mellor, 1978; Vega-Mercado et al., 2001). Figure 22.3 shows typical moisture loss (a) and product temperature (b) curves during vacuum freeze-drying (the example shown is apple juice, with and without pulp, freeze-dried at 50 °C heating plate temperature; data taken from Raharitsifa, 2003). The traditional exponential loss of moisture during freeze-drying can be observed in Figure 22.3(a), particularly after 200 min. Initially, unbound water can easily leave the matrix and thus the drying rate is almost constant; however, as the matrix dries out, the kinetics slow down, primarily because sublimated vapor passes through a dry layer with increasing thickness over time and, at the end of the process, because water is progressively more bound as drying proceeds. When only desorption of highly bound water takes place, freeze-drying kinetics are very slow. The evolution of product temperature during freeze-drying is well visualized in Figure 22.3(b). At the start of the process, the frozen product is placed inside the freezedryer and vacuum is applied. Although the heating plate temperature is high in this case (50 °C), product temperature remains low (approximately −30 °C) during primary drying due to the significant amount of heat expended for sublimation, which protects the sample from heating up. In Figure 22.3(b), the temperature plateau that extends for 460 min is the “sublimation” stage. When there is no ice remaining, the temperature of the product suddenly increases and secondary drying (desorption) starts. It is interesting to note in Figure 22.3 that although sublimation ends at approximately the same time for both types of juice samples (with or without pulp), juice with pulp showed a slower drying rate (higher water content marked with a dashed line and arrow at the end of sublimation), probably due to a higher amount of bound water present in juice with pulp. In the example shown in Figure 22.3, the samples could be considered dry after approximately 800 min when the relative moisture content is near zero and the product temperature (46 °C) is constant and close to the heating plate temperature (50 °C). In general at the end of freeze-drying, the product has a temperature 3–6 °C below the heating plate temperature, the difference depending on the geometry of the freeze-dryer and thermal properties of the foodstuff. Product temperature evolution during freeze-drying is an interesting parameter to follow closely, since it provides useful information for determining the end of sublimation, estimating the end point


625

(a) 1

relativ ve water con ntent, X/X o

0.9

Without p pulp p

0.8

With pulp

0.7 0.6 0.5 0.4 03 0.3 0.2 0.1 0 0

200

(b)

Tp

400

600

800

1000

800

1000

time (min) 50 40

o

temperature ( C C)

30 20 10 0 -10 -20 -30 -40

desorption

sublimation

-50 0

200

400

600

time (min)

Figure 22.3 Freeze-drying of apple juice (with and without pulp) at Tp = 50 °C (shelf temperature), 30 mTorr (total pressure), and product thickness 8 mm: (a) moisture loss and (b) product temperature at the center of the product. (Data from Raharitsifa, 2003.)

of the freeze-drying cycle, assessing potential quality-related problems and predicting which transfer, heat or mass, controls the process.

Process Design Two types of products are usually the focus of food freeze-drying applications: solid foodstuffs and homogeneous solutions such as juice and liquid coffee. Although small volume is common in the pharmaceutical and nutraceutical industry, processing these

626


types of products in small volumes (i.e. vials, ampoules) is not often seen in the food industry where the products are mainly processed in bulk on trays. Many parameters play an important role in obtaining a premium-quality freeze-dried product but in this chapter only some of them are explored, namely product pretreatments, sample thickness and process parameters. Chamber pressure and shelf temperature are the main operational parameters. Although in the pharmaceutical industry freeze-drying is run following complicated temperature and pressure cycles, an efficient freeze-drying operation for food applications can be achieved using a single-step cycle, where the shelf temperature is set for secondary drying and the product temperature for primary drying is controlled by adjusting chamber pressure (Chang and Fischer, 1995).

Product Pretreatments Pretreatment of a material prior to drying has long been used as a technique to accelerate drying rates as well as improve final product quality (Ratti, 2008a). The most common pretreatments in freeze-drying are concentration (to reduce the amount of water to sublimate) and grinding (to reduce the size of the particles). Readers are encouraged to acquire in-depth knowledge of the subject by researching the numerous articles in the literature. Hard-to-dry samples, such as oil or sugar-rich foodstuffs, represent technical problems for processing by freeze-drying. Foaming the sample prior to freeze-drying could be beneficial in these cases (Kudra and Ratti, 2006).

Thickness Thickness of the product to be freeze-dried is an important parameter to take into account before starting the process. Equation 22.1 is one of the most well-known representations of freeze-drying sublimation time as a function of different parameters (Karel, 1975) and was developed for strict application in freeze-drying cases where heat and mass are transferred through the dry layers, i.e. when radiation from both surfaces or radiation in one surface and “contact” convection in the other are the main heat transfer mechanisms (shown schematically in Figure 22.2b): t=

L2 ρ ( Xo − X f ) ΔHs 8kd (Ts − Tice )

(22.1)

where t is the freeze-drying time, L the thickness of the slab, ρ the bulk density of the solids, Xo and Xf the initial and final moisture contents, ΔHs the latent heat of sublimation, kd the thermal conductivity of the dry layer and Ts and Tice the maximum permissible surface temperature and ice temperature, respectively. Equation 22.1 was developed by considering slab geometry with negligible end effects, as well as the assumption that the maximum allowable surface temperature is reached instantaneously and that it remains constant during freeze-drying (Karel et al., 1975). As seen from Equation 22.1, the sublimation drying time should increase dramatically with


627

thickness and this is the reason why in most practical situations a maximal sample thickness of 1 cm is used. In order to use Equation 22.1, several parameters must be known, such as thermal conductivity of the dry layer, density of bulk solids, or ice sublimation heat. Thermal conductivity of the dry layer is a property not only dependent on total pressure, nature of the surrounding gas and temperature, but also on porosity and total solid concentration. Information on average thermal conductivity of freeze-dried foodstuffs (i.e. beef, whole milk, apple, peach, tomato juice, coffee, avocado) at the very low pressures used in freeze-drying can be found in Kessler (1975). However, more scarce information on experimental values for thermal conductivity of freeze-dried foods as a function of total pressure or other variables, as well as models to represent this property, can be found in the literature (Harper, 1962; Qashou et al., 1972; Fito et al., 1984; Sagara and Ichiba, 1994; Lombraña and Izkara, 1996). Heat of sublimation varies slightly with temperature and is in the order of 2839 kJ·kg−1 (Ratti, 2008a). Bulk solids density can be considered to be in the region of 300–330 kg·m−3 (Karel et al., 1975). Figure 22.4 shows freeze-drying kinetic curves (a) and product temperature (b) of apple juice at different thicknesses (Raharitsifa and Ratti, 2010). As can be seen, thickness has a significant impact (P < 0.01) on juice freeze-drying kinetics. For instance, the time for a sample of 4-cm thickness to reach X/Xo = 0.1 is four times higher than that for a sample of 1 cm. The shape of the temperature curves is similar to those obtained for most products during the freeze-drying process (Sagara and Ichiba, 1994). The initial temperature of both the heating plate and product after freezing was −40 °C. As the temperature of the heating plate increases to 20 °C, product temperature increases with a delay corresponding to the sublimation time. After all the ice sublimated, the temperature increased gradually to reach the heating plate temperature. A simplified energy balance for the frozen region of a product undergoing freezedrying can be estimated by assuming a lumped parameter approach due to the higher thermal conductivity of frozen materials: miCpi

∂T = Qinput − ΔHs N w ∂t

(22.2)

where T is product temperature, t is time, mi and Cpi the mass and specific heat of frozen material, respectively, Qinput the heat that gets into the frozen control volume by conduction and Nw the water flux due to sublimation. In Equation 22.2, if all the heat received in the control volume is used for sublimation, then ∂T/∂t = 0 and consequently T is constant. In this case, neither heat nor mass transfer can be considered separately as the controlling step. On the other hand, the process can be considered to be mass transfer limited when Nw is small and thus the derivative ∂T/∂t is positive. This is the case for juice samples of 4-cm and 6-cm thickness for which the temperature curves shown in Figure 22.4(b) are above −40 °C (initial temperature). Thus in most cases, juice freeze-drying seems to be mass-transfer controlled. In addition, from Figure 22.4(a,b) it can be observed that the time taken to achieve final product

628


(a) 1,0 6cm 4cm 1cm

0,8

X//Xo

0,6

0,4

0,2

0,0 0

10

20

30

40

50

Time (h) (b) 30 20

Temperrature (°C)

10 0 -10 -20 -30 6cm 4cm 1cm

-40 -50 0

5

10

15

20

25

30

35

40

45

50

55

60

65

70

Time (h)

Figure 22.4 Freeze-drying of apple juice at different sample thickness at Tp = 20 °C (shelf temperature) and 30 mTorr (total pressure).


629

temperature (close to shelf temperature) is shorter than the time necessary for a dry product, confirming that in this case freeze-drying is controlled by mass transfer. Finally, limited heat transfer occurs in the situation where the derivative in Equation 22.2 is negative. Thus, temperature profiles have proven helpful in understanding the type of controlling transfer (heat or mass) during freeze-drying. Performing simple calculations from data shown in Figure 22.4(a), it can be concluded that the relationship between drying time and thickness is linear for apple juice samples. Also and despite the relationship shown in Equation 22.1, Shishehgarha et al. (2002) showed that the freeze-drying time of strawberry slices correlated better with a linear function on thickness than with a quadratic one. In this study, experimental freeze-drying times for strawberries of 5 and 10 mm thickness were determined at several water content values. Prediction of freeze-drying time for 10-mm slices, calculated as if it varied proportionally to the smallest thickness of the product (in this case 5 mm), were also presented. It was observed that, for most water content values, predicted and actual drying times for the 10-mm thickness were very close, indicating that the freeze-drying time of strawberries can be considered proportional to thickness. Other authors have previously found a linear relationship between freeze-drying time and thickness, for example Sharma and Arora (1995) for yoghurt under different heating transfer modes and Saravacos (1967) for apple and potato. As shown in Equation 22.1, freeze-drying time is often presented in the literature as proportional to the square of the piece size, since the process is usually explained from diffusion theory. However, King (1968) explained the anomaly with respect to pure diffusion theory based on an externally controlled (boundary layer) freeze-drying process, which clearly describes the linear relationship between freeze-drying time and product thickness.

Freezing Rate, Process Temperatures, and Chamber Pressure Freezing rate controls the size of ice crystals and therefore the porosity of the dry layer, which could have an impact on drying time (Hammami and René, 1997). From porous media theory, vapor removal will be easier from a material having larger pore size. In this regard, a slow freezing rate would be preferable since it forms larger ice crystals. In addition, freezing rate has a marked impact on food quality, most of the published information indicating that preservation of quality in cellular food systems is only enhanced by rapid cooling (de Kock et al., 1995; Allan-Wojtas et al., 1999; Boonsumrej et al., 2007). The size and shape of ice crystals are critical for the final quality of frozen foodstuffs, the rate of heat removal being one of the main factors determining crystal growth rate (Fernandez et al., 2006). Slow freezing promotes the formation of large extracellular ice crystals that damage vegetable tissues, whereas rapid freezing promotes intensive nucleation and formation of small intracellular ice crystals (Fernandez et al., 2006). Thus, determination of the “optimal” freezing rate for a product that is to be freeze-dried poses an interesting compromise between acceleration of dehydration and enhancement of final quality. According to King et al.

630


(1968), the appearance of freeze-dried turkey meat seemed to depend on freezing conditions. Quick-frozen meat samples maintained a whiter color than those frozen slowly. Similar results were found by Karel et al. (1975) and Flink (1975) for freeze-dried coffee. In another study, however, Genin and René (1996) showed that freezing rate had no influence on the final quality of the freeze-dried product nor on the dehydration time, a finding supported by Hammami and René (1997) for strawberries. On the other hand, process temperature is the main parameter affecting the quality of freeze-dried products. Increasing the freezing or shelf temperature certainly reduces costs associated with energy consumption during the whole process, but it could lead in turn to product deterioration. Volume reduction during freeze-drying is minimal if operating pressures and temperatures are appropriate (Jankovié, 1993; Hammami and René, 1997; Krokida and Maroulis, 1997; Shishehgarha et al., 2002). However, collapse may occur causing the sealing of capillaries, which also leads to reduced dehydration and puffiness. Thus, in the case of the freeze-drying process, both freezing and drying temperatures have an impact on final product quality (Khalloufi and Ratti, 2003). Therefore, the control and optimization of operating parameters during product manipulation and processing could prove essential in achieving a viable and efficient operation and one might expect that the optimal operating conditions are influenced by the type of product being processed.

Collapse, “Scorch” and Glass Transition Temperatures In order to achieve an efficient freeze-drying operation, process parameters (chamber pressure, freezing and heating plate temperatures) should be carefully chosen. First, the “target” temperatures for the specific food product should be determined. Collapse temperature is the single most important parameter determining conditions during freezing and primary drying (Shalaev and Franks, 2002). Collapse results in loss of structure and porosity, significant decrease in water sublimation rate, increase in product density and residual water content, change in color and even loss of aroma and nutrients. To avoid collapse during freezing and primary drying, product temperature should be below its collapse temperature. During freeze-drying and for a specific food product for which moisture permeability and thermal conductivity of the dry layer are fixed, this can be achieved by adjusting the chamber pressure as explained later. On the other hand, “scorch” temperature is the maximum allowable temperature for the dry layer, the value of which is based on quality considerations (often browning) that mark the transition from an acceptable to an unacceptable product (Flink et al., 1974). To avoid scorching during secondary drying, product temperature should be below its “scorch” temperature. Table 22.1 shows some literature values for collapse and “scorch” temperatures of different types of foods. Collapse temperatures are related, among other product properties, to composition and structure of the food matrix. From Table 22.1, it is interesting to note that collapse temperatures are much lower for products having a “weak” structure, such as juices or tomato. Also, it should be pointed out that “scorch” tem-


631

Table 22.1 Collapse and “scorch” temperatures for selected foods. Product

Strawberry Potato Tomato Sweetcorn Beef, quick frozen Beef, slow frozen Chicken Salmon Cheddar cheese Apple juice (22%) Grape juice (16%) Orange juice Guava juice Coffee extract (25%)

Collapse temperature (°C)

“Scorch” temperature (°C)

Reference

−15 −12 −41 −8 to −15 −14 −17 −20 −29 −24 −41.5 −46 −43 −37 −20

70 — — — 60 60 60 80 — — — 49 43 —

Karel et al. (1975) Fellows (2002) Fellows (2002) Fellows (2002) Karel et al. (1975) Karel et al. (1975) Karel et al. (1975) Karel et al. (1975) Fellows (2002) Fellows (2002) Fellows (2002) Karel et al. (1975) Karel et al. (1975) Fellows (2002)

peratures for most foodstuffs are higher than 40 °C, as shown for some selected foods in Table 22.1. Most of the values shown in the table were probably determined by trial and error, following visual observations. Other methods, such as intruded porosity or specific volume, could also be used to determine collapse (Levi and Karel, 1995). However, it could be of practical use to find out if these “target” temperature values could actually be predicted. Collapse phenomenon is closely related to glass transition phenomenon. Glass transition temperature, Tg, is a product property linked to deterioration during thermal processing (Karel, 1993; Sapru and Labuza, 1993; Chuy and Labuza, 1994; Taoukis et al., 1997). It can be defined as the temperature at which an amorphous system changes from a glassy state to a rubbery state (Roos and Karel, 1991; Karmas et al., 1992), which is mainly a function of water content, molecular weight and nature of the dry matter compounds (e.g. sugars) in a given substance (Slade and Levine, 1991; Genin and René, 1995; Roos, 1995). The effect of moisture on the Tg of foods has been extensively reported in the literature (Roos, 1987; Pääkkönen and Roos, 1990; Khalloufi et al., 2000). The Gordon–Taylor equation (Gordon and Taylor, 1952) is commonly used to fit experimental data on Tg of food products as a function of water content and composition: Tg =

x1Tg1 + kx 2Tg 2 x1 + kx 2

(22.3)

where Tg and x are the glass transition temperature and mass fraction, k is a parameter determined from experimental data and subscripts 1 and 2 correspond to dry solids and water, respectively.

632


When cooling a solution, ice crystallizes out at temperatures below the freezing temperature, resulting in concentration of the remaining material. With further reductions in temperatures more ice crystallizes and the material becomes increasingly concentrated until it forms a glass at Tg′, the temperature of maximum freezeconcentration where viscosity is such that it is impossible to form more ice (Hatley and Franks, 1991). This temperature of maximum freeze-concentration is also known as the absolute glass transition temperature, which occurs at the moisture content of the freeze-concentrate (amount of unfrozen water at Tg′). Although sometimes much higher, collapse temperatures during freeze-drying are linked to Tg′. The reported value for potato collapse temperature (Table 22.1) is 33 °C higher than the Tg′ value for fresh potato given by Karathanos et al. (1996). For strawberry, Tg′ has been reported as −35 °C (Hammami and René, 1997), while collapse temperature is reported as −15 °C (Table 22.1). This is because collapse is a dynamic process not only dependent on the specific foodstuff but also on the difference between its proper temperature and Tg′ as well as on the time that the material is under this temperature difference condition. In a work on collapse of freeze-dried carbohydrates, Levi and Karel (1995) showed that volume reduction in freeze-dried sucrose/raffinose mixture (3 : 2) increases as (T − Tg) and time increases. Thus, Tg′ could be seen as a “theoretical” maximum temperature limit that should not be surpassed during primary drying, while collapse temperature is the limit from a practical standpoint. The differential scanning calorimeter and freeze-drying microscope are two specialized techniques for determining important product properties related to freeze-drying. Hatley and Franks (1991) indicated that Tg′, w ′g (unfrozen water at Tg′) and Tgs (glass transition of dry-cake) are properties that provide sufficient information to optimize a freeze-drying cycle and can be measured using the differential scanning calorimeter. To complete these measurements, a freeze-drying microscope provides real-time images of freezing, melting, crystallization, collapse and melt-back during freezing and freeze-drying processes (Wang, 2004). It is possible that if the vacuum level in the freeze-dryer is low enough, then this first thermal limit (product temperature lower than collapse temperature) is always achieved. The second thermal limit is accomplished when the final temperature of the product is lower than the maximum permissible surface temperature, i.e. “scorch” temperature. This latter limit could be assumed to be the glass transition temperature of dry solids (Khalloufi and Ratti, 2003). Shrinkage and Tg are interrelated in that significant changes in volume and collapse are noticed only if the temperature of the process is higher than the Tg of the material at that particular moisture content (Genin and René, 1995). Khalloufi and Ratti (2003) showed that equal freeze-drying conditions had different impacts on the quality attributes of freeze-dried strawberry, apple and pear. In this paper, shrinkage and quality changes during freeze-drying were related to glass transition temperature and microstructure of the samples. Knowledge of the microstructural arrangement of a heterogeneous food can also help in understanding quality deterioration and collapse during processing. Pore formation during freezedrying of apples and dates was studied at different shelf temperatures (Sablani and


633

Rahman, 2002), from which it was concluded that glass transition theory alone could not explain pore formation during freeze-drying. In a review article on porosity prediction during drying, Rahman (2001) described several physical mechanisms, in addition to glass transition changes, which may play an important role in the control of collapse during drying (i.e. pore pressure, moisture transfer regime, mechanical strength of the matrix, environmental pressure, etc.). Further research on the relationship between freeze-drying and collapse is required to fully predict the phenomenon from a practical standpoint.

Chamber Pressure Figure 22.5 shows pressure–temperature data for water from −70 to 0 °C. During primary drying, sublimation rate (g/h) can be represented by: N w = km ( pice − Pchamber )

(22.4)

where km is a coefficient quantifying the easiness of the dry layer for vapor transfer, pice is the pressure at the receding front, which should be at the target temperature to avoid collapse and Pchamber is the chamber pressure. Thus from Equation 22.4 it is obvious that the driving force during primary drying depends directly on the difference in pressure between the ice in the product and the chamber pressure. Historically, it was believed that lowering the vacuum in the chamber and the condenser temperature as much as possible would accelerate freeze-drying. However, it is now well understood that allowing the condenser temperature to rise and bleeding air or inert gas into the freeze-drying chamber can actually accelerate sublimation (Rowe, 1976). This

5000 4500 Pressure (mTorr)

4000 3500 3000 2500 2000 1500 1000 500 0 -60

-50

-40

-30

-20

-10

0

Temperature (ºC)

Figure 22.5

Pressure–temperature data for pure water at temperatures below 0 °C.

634


is because lowering the vacuum level too much can cause large heterogeneity in heat transfer (Tang and Pikal, 2004). Tang and Pikal (2004) proposed an equation for determining “optimal” chamber pressure (Pchamber, in Torr): Pchamber = 0.29 × 10( 0.019Tt )

(22.5)

where Tt is the target temperature for the product. As an example, strawberry has a collapse temperature of −15 °C (Table 22.1), so a margin of 5 °C below the collapse temperature is acceptable. Thus the target temperature (Tt) at which the product should be kept during primary drying is −20 °C. From Figure 22.5, we can determine the pice for the product at this temperature as 750 mTorr (100 Pa). And from Equation 22.5, the chamber pressure to achieve goal temperature during primary drying can be calculated as 120 mTorr (16 Pa). In trial-and-error freeze-drying/quality experiments on strawberries, Hammami and René (1997) determined an optimal chamber pressure of 200 mTorr (26.7 Pa). In the case of grape juice, for which collapse temperature is much lower (−46 °C from Table 22.1), a low chamber pressure of 30 mTorr (4 Pa) could be optimal. Condenser temperatures in commercial freeze-dryers range between −45 and −60 °C. Specialized freeze-drying equipment have condensers operating up to −95 °C.

Heating Plate Temperature Shelf temperature is an important freeze-drying parameter, especially at the end of drying when desorption of bound water has to be accomplished. Tang and Pikal (2004) indicated that one of the most time-consuming tasks in freeze-drying process design is the determination of shelf temperature. They provide some interesting equations and guidelines for determining the “optimal” shelf temperature as a function of primary drying conditions so as to maintain product temperature during sublimation always below its collapse temperature. Nevertheless, other authors have shown that although heating plate temperature has an effect on product temperature during sublimation, this effect is not as dramatic as that of chamber pressure. On the other hand, sublimation rate can be noticeably increased by rising shelf temperature during primary drying. In an article on the development of a single-step freeze-drying cycle for a recombinant human IL-1ra formulation, Chang and Fischer (1995) showed that a very efficient freeze-drying cycle can be obtained when shelf temperature is set as high as possible depending on the system and product stability and, at the same time, product temperature is maintained below its collapse temperature by reducing chamber pressure. The design of shelf temperature during a secondary drying step is based on the principle that product dry-cake temperature should be lower than the “scorch” temperature. Also, as Franks (1990) stated, freeze-drying should be controlled such that


635

the temperature of the product never exceeds the glass transition temperature at that particular moisture content. Thus, the maximal heating plate temperature should be chosen based on the glass transition temperature of the dry product. During freezedrying of cabbage, strawberry and pear, Giasson and Ratti (2000) showed that the amorphous dry portion of the solid was in contact with a shelf plate maintained at a high temperature for long periods. This dry matrix has low moisture content and a Tg corresponding to that of the dry cake. This suggests that the glass transition temperature of the dry layer (Tgs) could be an interesting optimization parameter for the freezedrying process. This parameter is also a useful tool for determining the maximum water content at the end of the process for stable storage of freeze-dried products. When freeze-drying apple, pear and strawberry, Khalloufi and Ratti (2003) also showed a relationship between quality loss and glass transition temperature of dry cake, although they indicated that glass transition theory alone cannot explain all the observed quality changes. In a detailed work on freeze-drying of pear and apple juices, Raharitsifa (2003) developed a shelf plate temperature selection method based on glass transition temperature of dry cake. The procedure is shown schematically in Figure 22.6, where both glass transition temperature of the dry cake and final product temperature (Tf) are plotted as a function of heating plate temperature. The cross-point between curves helps determine the maximum heating plate temperature for avoiding quality problems during secondary drying. Using this method, the optimal shelf temperature for juice freeze-drying was determined as 54.6 °C for apple juice and 45.9 °C for pear juice (whose dry cake has a lower glass transition temperature than apple juice) (Raharitsifa, 2003). In this work, the optimal shelf temperature parameters were in agreement with freeze-dried product quality determinations.

70

Tg, Tf (°C)

Tf Tg

60

50

40

30

TsMAX 20 20

30

40

50

60

Heating plate temperature (oC)

Figure 22.6

Selection of maximum shelf temperature, TsMAX.

70

636


Modeling the Process One of the earliest and simplest models considers a receding front inside the product during freeze-drying and the energy used solely for ice sublimation (Sandall et al., 1967; Karel, 1975). If heat and mass transfer takes place through the dry layer, then Equation 22.1 is applied. This model is usually applied in the case of heating by radiation to both faces of the solid undergoing freeze-drying. If conduction through the frozen layer is the prevailing heating mechanism, another equation is used (Karel, 1975) where the temperature and vapor pressure at the interface should be evaluated simultaneously by iteration in order to obtain the drying time. Ratti (2008b) compiled some of the commonly used simple equations representing heat and/or mass transfer during vacuum and freeze-drying. Although easier to use than complex mathematical models, these equations make several key assumptions not usually applicable: (i) the maximum allowable surface temperature, Ts, is reached instantaneously; (ii) the heat output of the external supply is adjusted to maintain Ts constant throughout the drying cycle; (iii) partial pressure in the drying chamber is constant; and (iv) all the heat is used for sublimation of water vapor (Karel, 1975; Khalloufi et al., 2005). Numerical models employing highly detailed freeze-drying equations have been developed (Liapis and Bruttini, 1995a; Lombraña and Izkara, 1996; Lombraña et al., 1997; Brülls and Rasmuson, 2002; George and Datta, 2002; Khalloufi et al., 2005; Nastaj and Ambrozek, 2005). However, in most cases adjustable parameters are needed to match the model predictions to experimental data (Liapis and Marchello, 1984; Millman et al., 1985; Sharma and Arora, 1995; Sadikoglu and Liapis, 1997; Sheehan and Liapis, 1998; George and Datta, 2002; Nastaj and Ambrozek, 2005). In other cases, no comparison with experimental data is presented (Liapis and Bruttini, 1995b; Nastaj, 1991). In addition, most of the models were developed for liquids and not for solid products such as foodstuffs (Sadikoglu and Liapis, 1997; Sheehan and Liapis, 1998; Brülls and Rasmuson, 2002). Khalloufi et al. (2005) developed a freeze-drying model for solid foods based on microscopic energy and mass balances in the dried and frozen regions of the product. All the parameters involved in the model (i.e. thermal conductivity, permeability, heat transfer coefficients, etc.) were obtained independently from actual experimental data. Simulation results agreed closely with apple and potato freeze-drying data (Khalloufi et al., 2005). The model presented by Nastaj and Ambrozek (2005) is interesting since it deals with multicomponent freeze-drying (simultaneous desorption of water and other organic compounds), which could be applied to the simulation of aroma retention during the process.

Industrial Freeze-drying Figure 22.7 shows a schematic diagram of a typical batch freeze-dryer, in which the three main design components are as follows:


637

INSULATED WALL

SHELVES

DOOR

CONDENSERS

VACUUM PUMP

REFRIGERATOR OIL TRAP MOTOR

Figure 22.7

COLD TRAP

Batch freeze-dryer showing main components.

1. A vacuum system for evacuating air from the apparatus before and during drying. Vacuum levels range from 30 to 200 mTorr (4–26.7 Pa). Most commercial freezedryers for food applications work at a pressure of 100 mTorr (13.3 Pa). 2. A heat transfer system, which allows cooling to −50 °C or heating up to 70 °C. 3. A condenser operating at −60 °C or lower. Heat transfer is usually performed through hollow and fluid-filled shelves, whose freezing or heating temperatures can be controlled. Condensers are needed due to the enormous quantity of vapor generated during primary drying that cannot be extracted solely by the vacuum system. Condensers are critical “pumps” maintaining the freezedrying conditions (Sutherland, 2000), while the vacuum pump just removes the noncondensable gases of the environment. Condensers can be located inside the drying chamber (less expensive option, although some dried products such as juices and high sugar content foods could reconstitute during secondary drying), or outside in the path prior to the vacuum pump. In order to work correctly, the ice formed in the condenser should have a maximum thickness of 1–1.5 cm. Some auxiliary components of industrial freeze-dryers include (i) a defrost system to rapidly melt the condensed ice once

638


the freeze-drying operation is finished; (ii) a sterilization system to kill contaminant microorganisms (in-place pressurized steam sterilization, at 121 °C or higher temperature, is presently the first choice for industrial applications); and (iii) a cleaning-inplace system with sterile water sprayed at high pressure from internal nozzles (Liapis and Bruttini, 1995a). A description of the different technical procedures for operation of cleaning-in-place/sterilization-in-place for freeze-drying equipment can be found in Beurel (2004). Because freeze-drying is performed under vacuum, processing of food is frequently undertaken in batch (Figure 22.7), which is a major drawback for industry. An industrial batch freeze-dryer can function with trays or multi-cabinets and can process up to 2000 kg where there is a tray surface of 150 m2 (Lombraña, 2009). Tunnel freezedryers utilize large vacuum cabinets where trolleys carrying the trays are loaded at intervals through a large vacuum lock located at the entrance to the freeze-dryer and discharged in a similar way at the exit (Liapis and Bruttini, 1995a). The food industry uses this type of freeze-dryer for processing cottage cheese and coffee. Recently, great interest has been shown in developing continuous freeze-dryers for handling a single product, which can be in trays if delicate or with agitation for bulk materials for improving heat transfer (Liapis and Bruttini, 1995a). The “dynamic” freeze-drying continuous method is used for fluid/free-flowing products in direct contact with the heating surface. Figure 22.8 shows such a system for fluid/free-flowing products

Frozen liquid

VACUUM CONDENSERS

Dried product

Figure 22.8 2010.)

Continuous “dynamic” freeze-dryer. (Adapted from ALD Vacuum Technologies GMBH,


639

(ALD Vacuum Technologies GMBH, 2010). The frozen and granulated product is brought into the sublimation/drying tunnel by a special lock device. Inside the drying tunnel, the product is uniformly dry in a short time. At the end of the belt transport system, the freeze-dried product passes an outlet vacuum lock and is filled into storage containers.

Costs Freeze-drying costs vary depending on the type of raw material, the product, the packaging, the capacity of the plant and duration of cycle (Lorentzen, 1979; Sunderland, 1982a). Compared with air drying, freeze-drying costs are four to eight times higher (Ratti, 2001). Table 22.2 shows the fixed and operating costs of freeze-drying compared with other drying methods when applied to lactic acid bacteria (data from Santivarangkna et al., 2007). The costs of freeze-drying (both fixed and operational) are double those for vacuum drying and are 75% higher than for other dehydration methods. Although the differences in costs between freeze-drying and other drying methods are considerable, it is important to include all energy use when evaluating or comparing different processes. For example, the costs of freeze-drying compared with other methods of food preservation (e.g. freezing) are quite advantageous if the energy of the home storage freezer is taken into account (calculations based on Flink, 1977 and Judge et al., 1981). Also, the energy expended in the freeze-drying process itself becomes insignificant when dealing with high-value raw materials. Freeze-drying should therefore not be regarded as a prohibitively expensive preservation process if it gives a reasonable added value to the product or if it maintains its high value compared with other preservation methods (Lorentzen, 1979). It is important that a freeze-drying investigation should aim to reduce operation times and consequently lower energy consumption, analyze ways of controlling heat intensity and vacuum pressure and investigate approaches for optimizing the freezedrying process. Several studies have been carried out in laboratory- and pilot-scale plants (Sagara and Ichiba, 1994; Kuu et al., 1995; Liapis et al., 1996). Simulation has also been used as a preliminary tool for evaluating the freeze-drying process. As Table 22.2 Comparison of fixed and operating costs of different dehydration methods for lactic acid bacteria. Drying process Freeze-drying Vacuum drying Spray-drying Rotating drum Fluidized bed Hot air

Fixed costs (%)

Operating costs (%)

100.0 52.2 12.0 9.3 8.8 5.3

100.0 51.6 20.0 24.1 17.9 17.9

Source: data from Santivarangkna et al. (2007).

640


mentioned previously, several theoretical models concerning heat and mass transfer phenomena during freeze-drying can be found in the literature (Karel, 1975; Mellor, 1978; Liapis and Bruttini, 1995a; Lombraña and Izkara, 1996; Lombraña et al., 1997; Khalloufi et al.. 2005). From an energy point of view, the freeze-drying process comprises four main operations: freezing, vacuum, sublimation and condensing. Each of these operations shares the total energy consumption but while sublimation consumes almost half of the total energy of the process, the freezing step is not highly energy consuming. The energy consumption of vacuum and condensation is practically the same (Ratti, 2001). Any technological improvement to classical vacuum freeze-drying in order to reduce energy costs should address the following goals: (i) improve heat transfer in order to help sublimation; (ii) cut drying times in order to reduce vacuum; and (iii) avoid using condensers.

Unconventional Freeze-drying Microwave Heating Microwave heating provides an energy input that is not only essentially unaffected by the dry layers of the material undergoing vacuum or freeze-drying, but is absorbed mainly in the humid region (Sunderland, 1982b). Since the humid region has high thermal conductivity, microwave energy aids sublimation so that freeze-drying times are decreased by up to 60–75% (Peltre et al., 1977; Rosenberg and Bögl, 1987). In addition, compared with conventional freeze-drying, microwave-assisted freeze-drying leads to products of similar or even higher quality (Rosenberg and Bögl, 1987; Barrett et al., 1997). Nevertheless, microwave freeze-drying is still not widely used in industry since many technical problems can be encountered, some related to the extreme low pressures used during freeze-drying (i.e. corona discharges, melting and overheating of the frozen kernel, nonuniform heating, etc,) so it remains of academic interest only.

Adsorption Freeze-drying Adsorption freeze-drying uses a desiccant (e.g. silica gel) to create a high vapor drive at low temperatures (Bell and Mellor, 1990a). The adsorbent replaces the condenser and leads to a reduction of 50% in total costs compared with traditional freeze-drying. Despite the many advantages over regular freeze-drying (Bell and Mellor, 1990b), the quality of adsorption freeze-dried foods is slightly reduced and sometimes poor as compared with that obtained by traditional freeze-drying.

Atmospheric Freeze-drying Another method that has been developed and which is becoming popular is the fluidized atmospheric freeze-dryer. This process can be summed up in three words: adsorp-


641

tion, fluidization and atmospheric pressure (Wolff and Gibert, 1987). It is a freeze-drying operation at atmospheric pressure that utilizes a fluidized bed of adsorbent particles (Di Matteo et al., 2003). The adsorbent particles should be compatible with the material to be freeze-dried since it could be difficult to separate the adsorbent from the freeze-dried product (Kudra and Mujumdar, 2009). Approximately 34% energy reduction can be obtained with the use of this method (Wolff and Gibert, 1990). However, drying times are increased by up to threefold since the use of atmospheric pressure changes the process from one involving heat transfer to one involving mass transfer, which renders the kinetics extremely slow. In addition, other studies have shown that the quality of products is inferior when atmospheric pressure is used instead of vacuum, since the risk of product collapse is increased (Lombraña and Villarán, 1996; 1997). After a study of heat and mass transfer during atmospheric freeze-drying in a fluidized bed, Di Matteo et al. (2003) concluded that choosing the proper set of variables (sample size, bed temperature and nature of the adsorbent are the main ones) is key to success in the application of this technique in the food industry. Recently, Mujumdar (personal communication, April 2007) has reported ongoing work on the potential for use of a vibrated bed atmospheric freeze-dryer for costcompetitive drying of heat-sensitive materials like fruits and vegetables. Using a vortex tube to provide the cooled air and combined conduction and radiation modes for supplying the heat of sublimation, their results on a laboratory-scale unit show that a vibrated bed dryer can operate successfully without using the large volumes required for fluidization in conventional manner. By ensuring that product temperature is always above the triple point (considering the freezing point depression caused by soluble sugars or salts), they were able to obtain dried product quality characteristics (e.g. color, porosity and rehydration) that closely matched those obtained in vacuum freeze-drying. By addition of suitable adsorbents to the bed of model materials they tested (carrot and potato cubes and slices), they showed that the drying time can also be reduced by up to 20%. This work may lead to cost-competitive atmospheric freeze-drying processes that can compete with vacuum drying in general, which tend to be generally expensive in capital and operating costs.

Conclusions Freeze-drying is an expensive process used to manufacture high-quality food products and powders. Because of the intricate relationship between process variables and product properties, the freeze-drying cycle is usually determined by trial and error. In this chapter, simple advice for designing this complicated process has been given in order to guide users on how to predict an efficient cycle in order to obtain maximum quality in freeze-dried foodstuffs in an optimal time. Analyzing world trends on foods and eating habits, some predictions can be made about the future of freeze-drying as a method of preserving foods. Recently, the market for “natural” and “organic” products has been increasing strongly, as has consumer demand for foods with minimal

642


processing and high quality. With this in mind, the demand for freeze-dried foodstuffs and ingredients will certainly increase in the future.

References ALD Vacuum Technologies GMBH (2010) Vacuum systems and technologies for vacuum freeze-drying. Promotional brochure, page 9. Available at http://web.ald-vt. de/cms/fileadmin/pdf/prospekte/freezedrying.pdf (accessed July 5, 2010). Allan-Wojtas, P., Goff, H.D., Stark, R. and Carbyn, S. (1999) The effect of freezing method and frozen storage conditions on the microstructure of wild blueberries as observed by cold-stage scanning electron microscopy (Cryo-SEM). Scanning 21: 334–347. Barrett, A.H., Cardello, A.V., Prakash, A., Mair, L., Taub, I.A. and Lesher, L.L. (1997) Optimization of dehydrated egg quality by microwave assisted freeze-drying and hydrocolloid incorporation. Journal of Food Processing and Preservation 21: 225–244. Bell, G.A. and Mellor, J.D. (1990a) Adsorption freeze-drying. Food Australia 42: 226–227. Bell, G.A. and Mellor, J.D. (1990b) Further developments in adsorption freeze-drying. Food Research Quarterly 50: 48–53. Beurel, G.A. (2004) Technical procedures for operation of cleaning-in-place/sterilizationin-place process for production freeze-drying equipment. In: Freeze-Drying/ Lyophilization of Pharmaceutical and Biological Products, 2nd edn (eds L. Rey and J.C. May). Marcel Dekker, New York. Boonsumrej, S., Chaiwanichsiri, S., Tantratian, S., Suzuki, T. and Takai, R. (2007) Effects of freezing and thawing on the quality changes of tiger shrimp (Penaeus monodon) frozen by air-blast and cryogenic freezing. Journal of Food Engineering 80: 292–299. Brown, M. (1999) Focusing on freeze-drying. Food Manufacture September, 34–36. Brülls, M. and Rasmuson, A. (2002) Heat transfer in vial lyophilization. International Journal of Pharmaceutics 246: 1–16. Chang, B.S. and Fischer, N.L. (1995) Development of an efficient single-step freezedrying cycle for protein formulations. Pharmaceutical Research 12: 831–837. Chuy, L.E. and Labuza, T.P. (1994) Caking and stickiness of dairy-based food powders as related to glass transition. Journal of Food Science 59: 43–46. de Kock S., Minnaar, A., Berry, D. and Taylor, R.N. (1995) The effect of freezing rate on the quality of cellular and non-cellular par-cooked starchy convenience foods. LWT Food Science and Technology 28: 87–95. Di Matteo, P., Donsì, G. and Ferrari, G. (2003) The role of heat and mass transfer phenomena in atmospheric freeze-drying of foods in a fluidised bed. Journal of Food Engineering 59: 267–275.


643

Fellows, P.J. (2002) Freeze drying and freeze concentration. In: Food Processing Technology. Principles and Practice. Woodhead Publishing, Cambridge, pp. 401–414. Fernandez, P.P., Otero, L., Guignon, B. and Sanz, P.D. (2006) High-pressure shift freezing versus high-pressure assisted freezing: effects on the microstructure of a food model. Food Hydrocolloids 20: 510–522. Fito, P.J., Piñaga, F. and Aranda, V. (1984) Thermal conductivity of porous bodies at low pressure: Part I. Journal of Food Engineering 3: 75–88. Flink, J.M. (1975) The influence of freezing conditions on the properties of freeze-dried coffee. In: Freeze-drying and Advanced Food Technology (eds S.A. Goldblith, L. Rey and W.W. Rothmayr). Academic Press, London, chapter 2. Flink, J.M. (1977) A simplified cost comparison of freeze-dried food with its canned and frozen counterparts. Food Technology 31: 50. Flink, J.M., Hawkes J., Chen H. and Wong, E. (1974) Properties of the freeze drying “scorch” temperature. Journal of Food Science 39: 1244–1246. Franks, F. (1990) Freeze-drying: from empiricism to predictability. Cryo-letters 11: 93–110. Genin, N. and René, F. (1995) Analyse du rôle de la transition vitreuse dans les procédés de conservation agroalimentaires. Journal of Food Engineering 26: 391–408. Genin, N. and René, F. (1996) Influence of freezing rate and the ripeness state of fresh courgette on the quality of freeze-dried products and freeze-drying time. Journal of Food Engineering 29: 201–209. George, J.P. and Datta, A.K. (2002) Development and validation of heat and mass transfer models for freeze-drying of vegetable slices. Journal of Food Engineering 52: 89–93. Giasson, J. and Ratti, C. (2000) Glass transition temperature of dry solids: a crucial optimization parameter for freeze-drying. In: Proceedings of the International Conference on Engineering and Food: ICEF 8, Vol. I (eds J. Welti-Chanes, G.V. Barbosa-Canovas and J.M. Aguilera). Technomic Publishing Co., Lancaster, PA, pp. 202–206. Gordon, M. and Taylor, J.S. (1952) Ideal co-polymers and the second order transitions of synthetic rubbers. Journal of Applied Chemistry 2: 493–500. Hammami, C. and René, F. (1997) Determination of freeze-drying process variables for strawberries. Journal of Food Engineering 32: 133–154. Harper, J.C. (1962) Transport properties of gases in porous media at reduced pressures with reference to freeze-drying. AIChE Journal 8: 298–302. Hatley, R.H.M. and Franks, F. (1991) Applications of DSC in the development of improved freeze-drying processes for labile biologicals. Journal of Thermal Analysis 37: 1905–1914. Jankovié, M. (1993) Physical properties of convectively dried and freeze-dried berrylike fruits. Faculty of Agriculture, Belgrade 38: 129–135.

644


Judge, M.D., Okos, M.R., Baker, T.G., Potthast, K. and Hamm, R. (1981) Energy requirements and processing costs for freeze-dehydration of prerigor meat. Food Technology 35: 61–62, 64–67. Karathanos, V.T., Anglea, S.A. and Karel, M. (1996) Structural collapse of plant materials during freeze-drying. Journal of Thermal Analysis 47: 1451–1461. Karel, M. (1975) Heat and mass transfer in freeze drying. In: Freeze Drying and Advanced Food Technology (eds S.A. Goldblith, I. Rey and W.W. Rothmayr). Academic Press, London. Karel, M. (1993) Temperature-dependence of food deterioration processes. Journal of Food Science 58(6): ii. Karel, M., Fennema, O.R. and Lund, D.B. (1975) Principles of Food Science. Part II: Physical Principles of Food Preservation. Marcel Dekker, New York. Karmas, R., Buera, M.P. and Karel, M. (1992) Effect of glass transition on rates of nonenzymatic browning in food systems. Journal of Agricultural and Food Chemistry 40: 873 –879. Kessler, H.G. (1975) Heat and mass transfer in freeze drying of mixed granular particles. In: Freeze Drying and Advanced Food Technology (eds S.A. Goldblith, I. Rey and W.W. Rothmayr). Academic Press, London. Khalloufi, S. and Ratti, C. (2003) Quality deterioration of freeze-dried foods as explained by their glass transition temperature and internal structure. Journal of Food Science 68: 892–903. Khalloufi, S., El Maslhui, Y. and Ratti, C. (2000) Mathematical model for prediction of glass transition temperature of fruit powders. Journal of Food Science 65: 842–848. Khalloufi, S., Robert, J.-L. and Ratti, C. (2005) A mathematical model for freeze-drying simulation of biological materials. Journal of Food Process Engineering 28: 107–132. King, C.J. (1968) Rates of moisture sorption and desorption in porous, dried foodstuffs. Food Technology 22: 165–171. King, C.J., Lam, W.K. and Sandal, O.C. (1968) Physical properties important for freezedrying poultry meat. Food Technology 22: 1302. Krokida, M.K. and Maroulis, Z.B. (1997) Effect of drying method on shrinkage and porosity. Drying Technology 15: 2441–2458. Kudra, T. and Mujumdar, A.S. (2009) Atmospheric freeze-drying. In: Advanced Drying Technologies. CRC Press, Boca Raton, FL, pp. 327–335. Kudra, T. and Ratti, C. (2006) Foam-mat drying: energy and cost analyses. Canadian Biosystems Engineering 48: 3.27–3.32. Kuu, W.Y., McShane, J. and Wong, J. (1995) Determination of mass transfer coefficient during freeze drying using modeling and parameter estimation techniques. International Journal of Pharmaceutics 124: 241–252. Levi, G. and Karel, M. (1995) Volumetric shrinkage (collapse) in freeze-dried carbohydrates above their glass transition temperature. Food Research International 28: 145–151.


645

Liapis, A.I. and Bruttini, R. (1995a) Freeze drying. In: Handbook of Industrial Drying, 2nd edn (ed. A.S. Mujumdar). Marcel Dekker, New York. Liapis, A.I. and Bruttini, R. (1995b) Freeze-drying of pharmaceutical crystalline and amarphous solutes in vials: dynamic multi-dimensional models of the primary and secondary drying stages and qualitative features of the moving interface. Drying Technology 13: 43–72. Liapis, A.I. and Marchello, J.M. (1984) Freeze drying a frozen liquid in a phial. Drying Technology 2: 203–217. Liapis, A.I., Pikal, M.J. and Bruttini, R. (1996) Research and development needs and opportunities in freeze-drying. Drying Technology 14: 1265–1300. Lombraña, J.I. (2009) Fundamentals and tendencies in freeze-drying of foods. In: Advances in Food Dehydration (ed. C. Ratti). CRC Press, Boca Raton, FL. Lombraña, J.I. and Izkara, J. (1996) Experimental estimation of effective transport coefficients in freeze drying for simulation and optimization purposes. Drying Technology 14: 743–763. Lombraña, J.I. and Villarán, M. (1996) Interaction of kinetic and quality aspects during freeze drying in an adsorbent medium. Industrial and Engineering Chemical Research 35: 1967–1975. Lombraña, J.I. and Villarán, M. (1997) The influence of pressure and temperature on freeze-drying in an adsorbent medium and establishment of drying strategies. Food Research International 30: 213–222. Lombraña, J.I., De Elvira, C. and Villarán, M. (1997) Analysis of operating strategies in the production special foods in vials by freeze drying. International Journal of Food Science and Technology 32: 107–115. Lorentzen, J. (1979) Freeze-drying of foodstuffs. Quality and economics in freezedrying. Chemistry and Industry 14: 465–468. Mellor, J.D. (1978) Fundamentals of Freeze Drying. Academic Press, London. Millman, M.J., Liapis, I.A. and Marchello, J.M. (1985) An analysis of lyophilization process using a sorption–sublimation model and various operational policies. AIChE Journal 31: 1594–1604. Nastaj, J. (1991) A mathematical modeling of heat transfer in freeze drying. In: Drying 91 (eds A.S. Mujumdar and I. Filkova). Elsevier, Amsterdam, pp. 405–413. Nastaj, J.F. and Ambrozek, B. (2005) Modeling of vacuum desorption in freeze-drying process. Drying Technology 23: 1693–1709. Pääkkönen, K. and Roos, Y.H. (1990) Effects of drying conditions on water sorption and phase transitions of freeze-dried horseradish roots. Journal of Food Science 55: 206–209. Peltre, R.P., Arsem, H.B. and Ma, Y.H. (1977) Applications of microwave heating to freeze-drying: perspective. AIChE Symposium Series 73(163): 131–133. Qashou, M.S., Vachon, R.I. and Touloukian, Y.S. (1972) Thermal conductivity of foods. ASHRAE Transactions 78: 165–183. Raharitsifa, N. (2003) Freeze-drying of fruit juice: an optimization problem. Masters thesis, Université Laval, Québec, Canada.

646


Raharitsifa, N. and Ratti, C. (2010) Foam-mat freeze-drying of apple juice. Part 1: experimental data and ANN simulations. Journal of Food Process Engineering 33 (Suppl. S1): 268–283. Rahman, M.S. (2001) Towards prediction of porosity in foods during drying: a brief review. Drying Technology 19: 3–15. Ratti, C. (2001) Hot air and freeze-drying of high-value foods: a review. Journal of Food Engineering 49: 311–319. Ratti, C. (2008a) Avances récents dans les prétraitements des aliments avant le séchage. Presented at Progrès Récents en Génie Alimentaire Symposium, 76th ACFAS Congress, May 8, 2008. Ratti, C. (2008b) Freeze and vacuum drying of foods. In: Drying Technologies for Food Processing (ed. A.S. Mujumdar). Blackwell Publishing Ltd., Oxford. Roos, Y.H. (1987) Effect of moisture on the thermal behavior of strawberries studied using differential scanning calorimetry. Journal of Food Science 52: 146–149. Roos, Y.H. (1995) Phase Transitions in Foods. Academic Press, London. Roos, Y.H. and Karel, M. (1991) Applying state diagrams to food processing and development. Food Technology 45: 66–70, 107. Rosenberg, U. and Bögl, W. (1987) Microwave thawing, drying and baking in the food industry. Food Technology 41: 85–91. Rowe, T.W.G. (1976) Optimization in freeze-drying. Developments in Biological Standards 36: 79–97. Sablani, S.S. and Rahman, M.S. (2002) Pore formation in selected foods as a function of shelf temperature during freeze-drying. Drying Technology 20: 1379–1391. Sadikoglu, H. and Liapis, A.I. (1997) Mathematical modeling of the primary and secondary drying stages of bulk-solution freeze-drying in trays: parameter estimation and model discrimination by comparison theoretical results with experimental data. Drying Technology 15: 791–810. Sagara, Y. and Ichiba, J.-I. (1994) Measurement of transport properties for the dried layer of coffee solution undergoing freeze drying. Drying Technology 12: 1081–1103. Sandall, O.C., King, J. and Wilke, C.R. (1967) The relationship between transport properties and rates of freeze-drying of poultry meat. AIChE Journal 13: 428–438. Santivarangkna, C., Kulozik, U. and Foerst, P. (2007) Alternative drying processes for the industrial preservation of lactic acid starter cultures. Biotechnology Progress 23: 302–315. Sapru, V. and Labuza, T.P. (1993) Glassy state in bacterial spores predicted by polymer glass-transition theory. Journal of Food Science 58: 445–448. Saravacos, G. (1967) Effect of the drying method on the water sorption of dehydrated apple and potato. Journal of Food Science 32: 81–84. Shalaev, E. and Franks, F. (2002) Solid–liquid state diagrams in pharmaceutical lyophilisation: crystallisation of solutes. In: Amorphous Food and Pharmaceutical Systems (ed. H. Levine). Royal Society of Chemistry, Athenaeum Press, London.


647

Sharma, N.K. and Arora, C.P. (1995) Influence of product thickness, chamber pressure and heating conditions on production rate of freeze-dried yoghurt. International Journal of Refrigeration 18: 297–307. Sheehan, P. and Liapis, A.I. (1998) Modeling of primary and secondary drying stages of freeze-drying of pharmaceutical products in vials: numerical results obtained from solution of dynamic and spatially multi-dimensional lyophilization model for diffrent operational policies. Biotechnology and Bioengineering 60: 712–728. Shishehgarha, F., Makhlouf, J. and Ratti, C. (2002) Freeze-drying characteristics of strawberries. Drying Technology 20: 131–145. Slade, L. and Levine, H. (1991) Beyond water activity: recent advances based on an alternative approach to the assessment of food quality and safety. Critical Reviews in Food Science and Nutrition 30: 115–360. Sunderland, J.E. (1982a) An economic study of microwave freeze-drying. Food Technology 36: 50–52, 54–56. Sunderland, J.E. (1982b) Microwave freeze-drying. Journal of Food Process Engineering 4: 195–212. Sutherland, D.T. (2000) Notes on lyophilization technology. The Theory and Practice of Freeze Drying course (Volume 2, section E). Center for Professional Development (CfPA). Tang, X.C. and Pikal, M.J. (2004) Design of freeze-drying processes for pharmaceuticals: practical advice. Pharmaceutical Research 21: 191–200. Taoukis P.S., Labuza T.P. and Saguy, I.S. (1997) Kinetics of food deterioration and shelf-life prediction. In: Handbook of Food Engineering Practice (eds K.J. Valentas, E. Rotstein and R.P. Singh). CRC Press, Boca Raton, FL, chapter 9. Vega-Mercado, H., Gongora-Nieto, M. and Barbosa-Canovas, G.V. (2001) Advances in dehydration of foods. Journal of Food Engineering 49: 271–289. Wang, D.Q. (2004) Formulation characterization. In: Freeze-drying/Lyophilization of Pharmaceutical and Biological Products, 2nd edn, revised and expanded (eds L. Rey and J. May). Marcel Dekker, New York. Williams-Gardner, A. (1971) Industrial Drying. Gulf Publishing Company, Houston, TX, pp. 275–281. Wolff, E. and Gibert, H. (1987) Lyophilisation sous pression atmosphérique. In: Collection Récents Progrès en Génie des Procédés (eds A. Storck and G. Grévillot). Lavoisier, Paris. Wolff, E. and Gibert, H. (1990) Atmospheric freeze-drying. Part 2: modelling drying kinetics using adsorption isotherms. Drying Technology 8: 405–428.