Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
PROCESS AND APPARATUS FOR FREEZING IN PLATES TO OBTAIN ICE MATRIXES WITH A UNIFORM DISTRIBUTION OF SOLUTES
Document Type and Number:
WIPO Patent Application WO/2012/070962
Kind Code:
A2
Abstract:
This invention is a controlled freezing process and an apparatus that aims to attenuate the effect of cryoconcentration of solutes that occurs during the freezing of aqueous solutions, by means of a freezing method that gives the user an adequate control of the speed and of the direction of the interface between the solid and the liquid phases in order to incorporate the solutes uniformly in the ice matrix. This invention has applications in the pharmaceutical industry, the food industry and also in the industry for development and commercialization of cryogenic technology.

Inventors:
RODRIGUES MIGUEL ANGELO JOAQUIM (PT)
BALZAN GUSTAVO BRUZUAL (SE)
FERNANDEZ VITOR MANUEL GERALDES (PT)
MATOS HENRIQUE ANIBA SANTOS DE (PT)
AZEVEDO EDMUNDO JOSE SIMOES GOMES DE (PT)
Application Number:
PCT/PT2011/000040
Publication Date:
May 31, 2012
Filing Date:
November 25, 2011
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
INST SUPERIIOR TECNICO (PT)
UNIV DO ALGARVE (PT)
UNIV LISBOA (PT)
RODRIGUES MIGUEL ANGELO JOAQUIM (PT)
BALZAN GUSTAVO BRUZUAL (SE)
FERNANDEZ VITOR MANUEL GERALDES (PT)
MATOS HENRIQUE ANIBA SANTOS DE (PT)
AZEVEDO EDMUNDO JOSE SIMOES GOMES DE (PT)
International Classes:
A23B4/06
Domestic Patent References:
WO2003037083A12003-05-08
WO2000072902A12000-12-07
Foreign References:
US6855354B22005-02-15
JP3188903B22001-07-16
US4438634A1984-03-27
EP1407202B12010-01-27
Other References:
WANG W.; SINGH S.; ZENG D.; KING K.; NEMA S., J. PHARM. SCI., 2007, pages 96
BUTLER M.F., CRYSTAL GROWTH & DESIGN, vol. 2, no. 6, 2002, pages 541
MIYAWAKI O., FOOD SCI. TECHNOL. RES., vol. 7, no. 1, 2001, pages 1
GU. X.; SUZUKI T.; MIYAWAKI O., J. FOOD SCIENCE, vol. 70, no. 9, 2005, pages 546
BHATNAGAR B. S.; PIKAL M. J.; BOGNER R. H., J. PHARM. SCI., vol. 97, no. 2, 2008, pages 798
SINGH SK; KOLHE P; WANG W; NEMA S., BIOPROCESS INT., vol. 7, no. 10, 2009, pages 32 - 44
PARAG KOLHE; ELISABETH HOLDING; ALANTA LARY; STEVEN CHICO; SATISH K. SINGH, BIOPHARM INTERNATIONAL, vol. 23, no. 7, 2010
WILKINS J; SESIN D; WISNIEWSKI R, INNOVATIONS IN PHARMACEUTICAL TECHNOLOGY, vol. 1, no. 8, 2001, pages 174 - 180
WEBB SD; WEBB JN; HUGHES TG; SESIN DF; KINCAID AC, BIOPHARM, vol. 15, no. 5, 2002, pages 22 - 34
BUTLER MF, CRYSTAL GROWTH & DESIGN, vol. 1, no. 3, 2000, pages 213
WAKISAKA, M.; SHIRAI, Y.; SAKASHITA, S., CHEMICAL ENGINEERING AND PROCESSING, vol. 40, 2001, pages 201 - 208
SHAMLOU PA; BREEN LH; BELL WV; POLLO M; THOMAS BA, BIOTECHNOL APPL BIOCHEM, vol. 46, 2007, pages 13 - 26
Attorney, Agent or Firm:
SERRA, António Manuel da Cruz (Av. Rovisco Pais, 1049-001 Lisboa, PT)
Download PDF:
Claims:
CLAIMS

1. Apparatus for freezing in plates to obtain ice matrixes with uniform distribution of solutes characterized in that it comprises :

a) one or more plate cavities adapted to receive the solution to freeze (10), each of these cavities has an active heat transfer surface (2), which is the one in the bottom of the cavity, and it is perpendicular to the force of gravity, it being that in each plate more than 90% of the heat released during freezing happens through the aforementioned active heat transfer surface (2);

b) three inlets/outlets for input and output of fluid in that cavity adapted to receive the liquid (5), (7) and (8), cavities with baffles (3) for circulating a cryogenic fluid through the active heat transfer surfaces and their inlets for cryogenic fluid circulation (4);

c) thermal insulation material on the external surfaces of the vertical walls in contact with liquid (9) and on the surfaces that separate these plates when arranged vertically in series (6) ;

d) metal enclosure around the plates (14);

e) base with a leveling system that supports the metal enclosure, which contains the aforementioned plates (13) ;

f) mechanical oscillation system;

g) cryostat that controls the temperature of the cryogenic fluid.

2. Apparatus according to the previous claim characterized in that the cryostat (that controls the temperature of the cryogenic fluid) , also controls the rate of progression of the solid/liquid interface being above 8 mm per hour during the freezing of the solution.

-1-

3. Apparatus according to previous claims characterized in that the insulation surface (9) comprises more than one layer of thermal insulation materials with different thermal properties .

4. Apparatus according to previous claims characterized in that it comprises side walls of low thermal conductivity, with values lower than 0.61 W nf1 K"1 (1) in contact with the aqueous solution to freeze.

5. Apparatus according to previous claims characterized in that it comprises side walls with a reduced aspect ratio, typically from 0.01 m to 0.1 m.

6. Apparatus according to previous claims characterized in that it has surfaces in contact with the solution (1) and (2) covered with a thin layer of metal that is mildly oxidizable, such as chromium gold and platinum.

7. Apparatus according to previous claims characterized in that said plates (10) are placed in the vertical plane (11) or the horizontal plane.

8. Apparatus according to claim 7 characterized in that the containers have a common cavity for circulation of cryogenic fluid (1) .

9. Freezing process using the apparatus described in claims 1- 8, characterized in that it comprises the following steps:

a) loading an aqueous solution to freeze in plates with inlets adapted to receive the liquid;

b) freezing of this solution through heat transfer surfaces which are perpendicular to the force of gravity, where over

-2- 90% of the heat released during freezing is transferred through this active heat transfer surface (2) due to the circulation of a cryogenic fluid through the cavities adapted to the flow of cryogenic fluid;

c) formation of a solid-liquid interface which is perpendicular to the force of gravity;

d) unidirectional progression of solid-liquid interface in the opposite direction of gravity and at a linear velocity of progression than 8 mm per hour;

e) stopping the freezing process by unloading the unfrozen liquid fraction;

f) transportation and/or storage of the solution or frozen in these plates;

g) thawing of the solution through the circulation of a cryogenic fluid.

10. Process according to claim 9, characterized in that thawing comprises the mechanical oscillation of the apparatus.

11. Process according to claims 9 and 10, characterized in that the liquid fraction of the aqueous solution to freeze, is diluted during freezing by adding a less concentrated aqueous solution .

12. Process according to claims 9-11, characterized in that the control of heat transfer is made by varying the temperature and flow of a cryogenic fluid that circulates inside the base metal (1) .

13. Process according to claims 9-12, characterized in that it produces frozen solutions where the concentration of solutes, by a sampling of 0.5 cm3 from anywhere in the ice matrix,

-3- exhibits a typical variation of less than 20% of the concentration of the solution thawed.

14. Process according to claims 9-13, characterized in that it is composed of multiple plates arranged in a horizontal, vertical, or both.

15. Use of the process described in claims 9-14, characterized in that it applies to solutions containing peptides or proteins and other constituents such as salts, buffering agents, sugars, anti-foam agents, antioxidants, anti-microbial agents and other agents necessary to the stability of proteins and peptides in solution or frozen.

16. Use of the process according to claim 15, characterized in that it applies to freezing solutions containing peptides or proteins for pharmaceutical or food applications.

17. Use of the process according to claims 15 and 16, characterized in that it applies to freezing solutions containing peptides or proteins and excipients for decoupling of processes, for long-term storage (typically between 6 months to 5 years) or for lyophilization .

18. Use of the process according to claims 15 to 17, characterized in that it applies to freezing aqueous solutions contained in flexible and elastically deformable. packages.

Date: November, 25th, 2011

-4-

Description:
DESCRIPTION

PROCESS AND APPARATUS FOR FREEZING IN PLATES TO OBTAIN ICE MATRIXES WITH A UNIFORM DISTRIBUTION OF SOLUTES

Field of the invention

This invention relates to methods and apparatus with application in the cryogenic freezing of aqueous solutions with the aim of improving the stability of proteins during the period in which they remain in the frozen state. In particular, the invention relates to methods and apparatus for freezing of aqueous solutions containing physiologically active molecules for decoupling of processes, for long-term storage (typically between 6 months and 5 years) or freeze- drying) . The present invention has applications in the pharmaceutical industry, the food industry and also in the industry for development and commercialization of cryogenic technology . Description of the Related Art

The proteins and polypeptides of therapeutic interest are a significant portion of emerging drugs. Under this context, monoclonal antibodies (MAb) are an ever more representative example comprising over 1/5 of the drugs currently in clinical trials with numerous therapeutic applications including cancer, allergies and cardiovascular diseases [1] . As the industrial scale production develops, new needs arise, related to production, and also with the proper handling, transport and storage of these products. Freezing of solutions is a common and very convenient process for decoupling processes, transportation, storage or production of particles by lyophilization . However, phenomena of denaturation and aggregation of proteins may occur during freezing, thawing and during storage in the ice matrix. These phenomena create severe obstacles to production at industrial scale because they have a direct impact on . production costs and are limiting of freezing, particularly when the proteins are intended to injectable formulations.

Cryoconcentration is one of the phenomena that occurs during freezing of solutions that is likely to cause the aggregation and denaturation of proteins. This phenomenon is the heterogeneous distribution of solutes during freezing, wherein the fraction that freezes first ends up more diluted than the initial solution and the fraction that freezes last, ends up more concentrated than the initial solution. Cryoconcentration can be interpreted in a micro-scale (not dependent on the sample), i.e. a phenomenon governed solely by thermodynamics, or in a macro-scale, in which case it depends on the kinetics of the process and on the size of the sample size. Only macro- scale cryoconcentration will be discussed in this document. The crystallization of water is a difficult process to control, particularly in buffers with high concentration of proteins and excipients. Nucleation and ice crystal growth are complex phenomena involving heat · and mass transfer. These depend on both thermodynamic and kinetic variables, such as solute interactions with the ice crystals, molecular diffusion in the ice matrix, supercooling, etc [2].

Some processes have been developed to take advantage of the cryoconcentration phenomena, particularly in the food industry to produce food concentrates, such as the progressive freeze- concentration process [3-7]. Even though the freezing process is common in the food industry, restrictions imposed by pharmaceutical compliance, the complexity, the sensitivity and the high value of protein drugs all make it incompatible with conventional freezing technologies. It is known that slow freezing promotes cryoconcentrat ion of solutes [2, 8] . Cryoconcentration is the result of the growth of ice crystals that remove preferentially water molecules from the unfrozen solution leading to an increase of solute concentration and formation of concentration gradients throughout the unfrozen solution that displace solutes further by convection and diffusion [9, 10] . Regions with high protein concentration can lead to protein denaturation and to the formation of aggregates which, for example, are unacceptable for injectable drug formulations [11] . Cryoconcentration is therefore an undesirable phenomenon during the process of freezing therapeutic substances. However, there are no effective methods to successfully avoid cryoconcentration because it is a result of variables that are, traditionally, not controlled, such as speed and direction of the solid-liquid interface and natural convection of solutes. Various authors have reported the occurrence of cryoconcentration in freezing methods used in the pharmaceutical industry (see [9, 10, 12]), which demonstrates the urge to find technical solutions to overcome this problem.

The properties at the solid-liquid interface during freezing are essential for the solute capture process. Two types of structure have been addressed: dendritic interfaces (with the formation of dendrites) and planar interfaces. Dendrites are interface protuberances formed by growing ice crystals [13] .

Several authors (for example [11]) have linked dendrites with the breakup of the solute transport phenomena to the inside of the solution, i.e., to the capture of the solutes. Therefore, the freezing under dendritic regime (with dendrite formation) is crucial to reduce the criocentration . As showed by Butler

[2], the velocity of the solid-liquid interface is decisive for the formation of planar or dendritic interfaces. The same author described that the interface forms dendrites whenever the interface grows with a linear velocity (normal to the surface) above 2 micrometer per second. On the contrary, for velocities below this number, the interface remains planar. Actually, planar interfaces are described as relevant to the production of concentrates due to intense crioconcentration they promote. In the production of concentrates the formation of the planar surface is induced by the increasing of convection phenomena by the use of mechanic stirrer [14] . Convection is therefore one of the most influent variables for crioconcentration. Contrarily to concentrate production processes, there are no forced convection mechanisms in freezing solution processes. In the latter processes, natural convection is still determinant for the solute dispersal in the ice matrix. Using interferometry, Butler [2] showed that a very rich-solute limiting layer is formed near the freezing interface, because the solute diffusion is very small when compared to the velocity of the interface growth (Peclet number is around 10 3 -10 4 ) that typically it is in the range of 1(Γ 6 m/s to 10 ~4 m/s.

As measured by Butler [2], the concentration of several solutes in the liquid layer adjacent to the interface is 10 to 15 times larger than that in the bulk of the solution. These sharp concentration gradients originate meaningful density gradients and promote therefore the natural convection. In this invention it is purposed a means to control both the velocity and the direction of the interface towards the minimization of the solute partitioning of solutes between the liquid and the solid phases, suppressing therefore the natural convection phenomena. The magnitude of the temperature gradient is controlled in order to allow velocities of interface growth above 2xl0 -6 m/s. In this way, dendritic interfaces are promoted and the solute partition is reduced. The direction of the temperature gradient is the opposite of that of gravity in order to originate a unidirectional freezing against gravity so that the denser limiting layer remain stabilized and parallel to the interface, thus eliminating the transport of the solutes by natural convection. Other authors have already addressed the unidirectional freezing that has the advantage of simplifying the scale-up of the process [15]. Yet, in that case it addresses an horizontal freezing process, whereas the freezing direction is perpendicular to gravity and therefore does not attenuate or reduce the natural convection.

Summary of the Invention

This invention is a process of freezing and a device that aims to attenuate the effect of the solutes' cryoconcentration that occurs during freezing of aqueous solutions containing physiologically active molecules. This procedure is applicable in the decoupling of processes for long-term storage (typically between 6 months and 5 years) or lyophilization . The process uses a freezing method which allows the control of the speed and of the direction of the interface between solid and liquid phases to incorporate solutes uniformly in the ice matrix. The current invention has applications in the pharmaceutical and the food industries, and or to those that develop and commercialize cryogenic technology.

The present invention is characterized in that it performs the freezing of aqueous solutions using a container which is characterized in that it comprises a reduced height (in the order of 0.01 m to 0.1 m) consisting of a metal base which is the heat transfer surface active (for internal circulation of a cryogenic fluid) and isolated side walls, which enhance the formation of a unidirectional temperature gradient (more than 90% of the heat generated by freezing is transferred in the direction of gravity) and oriented from bottom to top, i.e. in the opposite direction of the force of gravity.

Detailed description, of the invention

This invention is a process of freezing and a device that aims to attenuate the effect of solutes' cryoconcentration that occurs during freezing of aqueous solutions containing physiologically active molecules. This procedure is applicable in the decoupling of processes for long-term storage (typically between 6 months and 5 years) or lyophilization . The process uses a freezing method which allows to control the speed and the direction of the interface between solid and liquid phases to incorporate solutes uniformly in the ice matrix .

This invention allows to obtain frozen . solutions with a uniform distribution of solutes, i.e. frozen solutions in which the concentration of solutes (C) , based on 0.5 cm 3 samples taken from anywhere in the ice matrix, exhibit a typical variation of less than 20% of the initial concentration (Co) before freezing.

In most industrial processes used for freezing, the solutions are frozen in cylindrical tanks by circulating a cryogenic fluid, through a jacket in the side walls of the cylinder - such as the equipment Cryo edge™ commercialized by Sartorius Stedim Biotech S.A.. This type of configuration has two cryoconcentration effect promoters: first, causes the formation of a vertical interface, because the active heat transfer surface, i.e. the surface of the container that is in contact with the cryogenic fluid and the solution, is vertical. Consequently, the ice surface progresses in the horizontal plane, causing a density gradient in the liquid phase also oriented in the horizontal plane due to the exclusion of solutes by the solid phase. This density gradient enhances the natural convection and the consequent transport of solutes away from the interface. The second effect is the radial and concentric geometry of the freezing direction (from the outside to the center of the cylinder) which causes a quadratic reduction in the volume of fluid as the interface progresses towards the center. Therefore, all solutes not embodied in the ice will converge radially to a single point - the center - which amplifies the effect of cryoconcentration . In the present invention, the unidirectional freezing takes place in the reverse direction of gravity. The condition for unidirectional freezing consists in that the temperature gradient of the solution inside the container is also unidirectional .

Towards this end, this invention provides that all heat transfer surfaces are perpendicular to gravity and that the side walls are designed and built so that more than 90% of the heat transfer (enthalpy of melting of the solid solution) occurs through surfaces that are perpendicular to gravity.

This feature, within the present invention, ensures that the ice-solution interface is normal to the gravity force and advances in the direction from the bottom to the top. The control of the progression of the interface in this direction serves to counteract the two disadvantages mentioned above. On the one hand, freezing in this direction (opposite to gravity) ensures the formation of a solute concentration gradient (and density) that has the direction of gravity, i.e. the solute is more concentrated in the deeper layer of liquid (adjacent to the ice surface) where the partition of solutes occurs, and diluting towards the top. This gradient minimizes the gravitational potential of the aqueous solution and hence attenuates the natural convection. On the other hand, as in this case the freezing takes place in the axial direction, it is linear the reduction of the volume of the liquid with the freezing rate, because the area of the interface is constant - even when the freezing unit is a vertical cylinder.

Other cryopreservation systems also have heat transfer surfaces at the bottom of containers - one example is the method "Cryopreservation method and system with controlled dendritic freezing front velocity" [16] . In this case, however, even if the liquid can freeze in the direction from bottom to top in some areas, freezing is not unidirectional in the opposite direction of gravity, because the active heat transfer surfaces are not exclusively perpendicular to gravity. Therefore, in that example it is not possible to obtain a unidirectional temperature gradient and oriented in the opposite direction of gravity, as it is in this invention. The freezing process and the apparatus described in this invention comprise the use of containers characterized by a low height (between 0.01 m to 0.1 m) . These containers are called herein "freezing plates." Figure 1 shows the schematic representation of a container or a freezing plate. Each freezing plate (10) is characterized by comprising a cavity adapted to receive the solution to freeze, with a base containing a metallic surface (2) with a cavity adapted to the internal circulation of a cryogenic fluid (1) and side walls (3) with thermal insulation (9) and/or with low thermal conductivity (less than the water thermal conductivity of. 0.61

W rrf 1 K "1 ) . The metallic base (2) allows the heat transfer between the solution and the cryogenic fluid circulating inside the base. The high-thermal resistivity of the side walls associated with a large active heat transfer surface area, induce the formation of a uniform temperature gradient oriented from bottom to top, i.e., in the opposite direction of the force of gravity. This occurs because the heat removed by the side walls, which are also cooled by being in contact with the ice, is irrelevant compared to the heat removed by the active heat transfer surface (normal to gravity) . The intensity of the temperature gradient is controlled to ensure that the rate of progression of the interface is greater than 2xl0 "6 m/s, in order to trigger the formation of micro dendritic ice-solution interfaces. At the same time, the direction of the temperature gradient is opposite to the gravity force direction, to ensure that the denser thermal boundary layer stabilizes parallel to the interface, thus avoiding the transport of solutes by natural convection.

To obtain the desired temperature gradient, the design of the freezing plates takes into account the temperature of the cryogenic fluid (which circulates in the pan for freezing) and also the insulation thermal resistance of the side walls. The temperature of the cryogenic fluid may be constant or time- dependent. If the cryogenic fluid temperature is constant, the freezing interface growth decelerates with time due to: 1) the increase of the thickness of the frozen solution; and 2) the increase of heat input through the lateral walls of the freezing plates as more heat may enter as more surface contacts the ice (which has a lower temperature than the liquid) . For this reason, the increase in height of the ice layer causes the progressive lowering of the efficiency of this phenomenon. This effect lowers the linear velocity of the ice front. However, in this invention the reduction of ice front velocity can be compensated by progressively lowering the temperature of the freezing base. . Nevertheless, this compensation is limited to operational values (for example in the range from -10 °C to -60 °C) . Therefore, the volume of freezable liquid in each plate by active freezing surface is limited, by the characteristics of this invention, to a small value (when compared to other freezing methods) . This is because, in this invention, freezing occurs unidirectionally, the characteristic ratio of volume per transfer area is equivalent to the liquid height in each plate.

Figure 2 shows the influence of the height of the liquid column in the cryoconcentration of solutes during the freezing of bovine hemoglobin (Hgb) solutions 1 mg/ml using a cryogenic fluid at -10 °C and also at -20 °C. In this case, when the liquid column reaches a critical height (z=0.017 m for -10 °C and z=0.051 m for -20°C) the crioconcentration factor, C/C 0 , ratio between the concentration obtained by ice sampling (C) and the initial concentration in the solution Co, decreases markedly. This means that there was a considerable loss of ability to include the solutes in the ice matrix. In this case, as the temperature lowers, higher is the critical height, i.e. the height at which a substantial drop in the concentration of solutes in the ice is observed.

Figure 3 shows that the concentration decay happens when the ice front reaches 2 micrometers per second, reason why this velocity is considered the minimum value.

Figure 4 shows an example of an ice matrix with uniform concentration of solutes (less than 20% variation compared to the initial concentration before freezing measured in ice samples of 0.5cm 3 taken from various regions of the ice matrix) , produced by freezing of an hemoglobin solution with the initial concentration of 1 mg/ml in PBS buffer (1/15 M pH 4). In this example, the height frozen was only 80% of the critical height, to make sure that the velocity of the progressing interface was always above 2xl0 ~6 m/s .

The mass balance implicit to Figure 4 implies that even though this invention attenuated considerably the effect of crioconcentration, part of the solutes were still be displaced towards the last fraction to freeze, the ice top. For this reason, in this invention the process is interrupted before the solution is completely frozen, by diluting and purging or recycling the remaining liquid fraction not frozen, using a solution with a lower solute concentration (such as water or a buffer) . Summarizing, this invention enables the freezing of solutions while .attenuating the influence- of natural convection and the crioconcentration factor caused by concentric freezing in cylindrical containers, which happens in most industrial freezing processes. This enables the production of frozen products with low crioconcentration. However, it has the disadvantage of requiring multiple containers (freezing plates) to obtain substantial capacity, due to the low height of liquid frozen in each container. Nevertheless, the freezing plates have a large heat transfer area and as a result the freezing or thawing process is faster and better controlled, what constitutes additional advantages. Another advantage is that the freezing is unidirectional and therefore the dimensioning of the freezing plates is not dependent of the heat transfer area, since the height of liquid is not dependent of the active heat transfer surface (area of the base) dimensioning and scale-up are simplified compared to other freezing processes.

Unidirectional freezing in plates, as in this invention, also has the advantage of having application to freezing of aqueous solutions contained inside flexible and elastically deformable packages. In this case, the packages should be placed inside cavities adapted with surfaces for heat transfer normal to the force of gravity.

Detailed description of the apparatus and process

The apparatus used in this invention is a modular installation of containers of freezing containers characterized by a reduced height of liquid (in the order of 0.01 m to 0.1 m) . Containers are hereby named freezing plates. Each freezing plate consists essentially of: a heat transfer surface in the active base (2), which is perpendicular to the force of gravity and includes a cavity (1) for internal circulation of a cryogenic fluid; side walls (3) with insulation (9), inlets for loading and circulating the aqueous solution (5, 7, 8) and inlets for cryogenic fluid circulation (4) .

The refrigerated metal base .(e.g. copper, aluminium or steel) is where a temperature-controlled cryogenic fluid flows

(typically between -10 °C and -60 °C) to cool the solution and freeze it. The base has internal baffles to facilitate heat exchange . The side walls (3) can be a non-metallic low thermal conductivity material (such as Teflon, acrylic, PVC and others) . Figure 5 a) shows a base plate with fittings for tubular side walls. In the case of base outlined in Figure 5, the tubular side wall mates with a seal that prevents the solution from leaking to the outside of the system.

The solutions to. be frozen are introduced into each plate and frozen by circulating a cryogenic fluid at a constant temperature (or variable temperatures, typically between -10 °C and -60 °C) .

The temperature of the cryogenic fluid is controlled by a cryostat . Freezing processes can be performed in multiple plates, by composing a modular assembly of several plates. Figure 5 illustrates a base with slots for installation of three plates. The number of plates can be increased by placing them horizontally as shown in Figure 6, or vertically as illustrated in Figure 7.

The system is leveled to ensure that freezing takes place perpendicular to gravity, which is a necessary condition for the formation of a unidirectional temperature gradient.

The freezing process involves the following steps: (a) prior cooling of the system and of the aqueous solution until the initial temperature is attained (typically between 1 °C to 10

°C for the initial temperature aqueous solution, and -10 °C to -60 °C to the initial temperature base) , (b) filling the plates with the aqueous solution to be frozen, (c) starting of the freezing process with constant temperature of the base (typically between -10 °C and -60 °C) or alternatively with a variable temperature of the base (typically in the range -10 °C to - 60 °C) , (d) interruption of the freezing process through a purging or recycling the liquid (typically the last 5% to 10% of the initial solution) through the circulation of. a less concentrated solution (e.g. a buffer solution with excipients) and/or movement of a compressed gas.

To facilitate the process of purging or recycling of the concentrated residue, the plates comprise inlets (5, 7 and 8) at different heights, as shown in Figure 1, to facilitate the loading of the aqueous solution and its unloading after it is thawed. The height of these inlets takes into account the volume expansion of the aqueous solution due to the freezing process .

The thawing of the solution can be facilitated by mechanical oscillation of the system. Description of figures

Figure 1 Schematic of a freezing plate (10), consisting of a base metal surface (2) with a cavity for circulation of a cryogenic fluid (1), inlets for cryogenic fluid circulation

(4), the side walls (3) with isolation (9), isolation of the base (6), and inlets / outlets for the introduction or removal of the aqueous solution (7, 8 and 5) . Figure 2 Variation of the cryoconcentration factor of hemoglobin (C/C 0 ) in a frozen solution (1 mg/ml in PBS buffer pH4 1/15 M) analyzed by absorbance at 400 nm by sampling discs (0.005 m) of the final frozen at several heights (z) . The open squares refer to a freezing temperature with constant basis at -10 °C and the solid squares refer to a freezing temperature of the base with a constant -20 °C.

Figure 3 Variation of the cryoconcentration factor (dashed) hemoglobin (C/C 0 ) in a frozen solution (1 mg/ml in PBS buffer 1/15M pH4) with constant temperature of the base at -20 °C.

The frozen was analyzed by absorbance at 400 nm by sampling discs (0.005 m) at various heights (z); and speed of the interface of ice (U) was calculated for various heights (squares) using a computational model described in Example 3. The open squares refer to a freezing temperature with constant basis at -10 °C and the solid squares refer to a freezing temperature of the base with a constant -20 °C.

Figure 4 Production of frozen protein with a uniform distribution of solutes using a uniform liquid height of 4 cm and a base temperature constant at -20 °C. Variation of the cryoconcentration factor of hemoglobin (C/Co) in frozen solution (1 mg/ml . in PBS buffer pH4 1/15 M) analyzed by absorbance at 400 nm by sampling discs (0.005 m) of the final frozen at several heights (z) .

Figure 5 a) 3D Schematic description of the metal base with slots for the side walls (15) used in freezing of protein solutions described in Examples 1 to 4.

Figure 5 b) Schematic description of the base metal used in the freezing of protein solutions described in Examples 1 to 4. The base has three active heat transfer surfaces (2) to assemble three plates, the cavity for circulation of the cryogenic fluid (1) is the same for all three heat transfer surfaces . Figure 6 Schematic of an apparatus with multiple freezing plates disposed horizontally.

Figure 7 Schematic of an apparatus with multiple freezing plates disposed in the vertical plane (11) comprising: a base support level (12) with rollers (13) and a protective outer casing ( 14 ) .

Examples :

Example 1

This example demonstrates the determination of the maximum freezable height (critical height) in order to obtain a frozen solution with uniform distribution of solutes, applying a constant temperature at the refrigerated base - 10 °C. The procedure consisted of freezing 60 ml of a bovine hemoglobin solution (Hgb) lg/dm 3 in a phosphate saline buffer (PBS) 1/15M pH =7.4. The freezing was carried in an apparatus with three freezing plates arranged horizontally. The metallic base (1) (made of tin) is described schematically in Figure 5. The 60 ml were distributed evenly by the three freezing plates. The side walls (3) of the freezing plates (with low thermal diffusivity) were made of acrylic and fitted the refrigerated base with a Viton™ o-ring sealant. A double insulation was assembled in the side walls (9), the first layer made of polyurethane (5 cm thickness) and the second layer made of expanded polypropylene (5 cm thickness) . The thermal insulation of the base (6) consisted in expanded polypropylene (5 cm thickness) . The 60 ml corresponded to a height of 0.06 m in each plate. The solution was frozen by circulating a cryogenic fluid (ethanol - in this example) at -10°C. After 20h of freezing the remaining liquid (approximately 0.5 ml) was purged and the freeze was removed from the apparatus and cut (radially) into slices with 0.5 cm thickness. The slices were thaw and the concentration was determined by spectrophotometry (absorbance at 400 nm) . Figure 2 Shows the crioconcentration profile i.e. the ration between the concentration of the sample and the initial concentration of the solution (C/C 0 ) (-10°C curve) . This example shows that crioconcentration accented beyond the height of 0.017 m, what demonstrates that this is the maximum height at -10°C in order to obtain an ice matrix with uniform distribution of solutes.

Example 2

This example demonstrates the determination of the maximum freezable height (critical height) in order to obtain a frozen solution with uniform distribution of solutes, applying a constant temperature at the refrigerated base - 20 °C. The procedure consisted of freezing 80 ml of a bovine hemoglobin solution (Hgb) lg/dm 3 in a phosphate saline buffer (PBS) 1/15M pH =7.4. The freezing was carried in an apparatus with three freezing plates arranged horizontally. The metallic base (1) (made of tin) is described schematically in Figure 5. The 80 ml were distributed evenly by the three freezing plates. The side walls (3) of the freezing plates (with low thermal diffusivity) were made of acrylic and fitted the refrigerated base with a Viton™ o-ring sealant. A double insulation was assembled in the side walls (9), the first layer made of polyurethane (5 cm thickness) and the second layer made of expanded polypropylene (5 cm thickness) . The thermal insulation of the base (6) consisted in expanded polypropylene

(5 cm thickness). The 60 ml corresponded to a height of 0.08 m in each plate. The solution was frozen by circulating a cryogenic fluid (ethanol - in this example) at -10°C. After 20h of freezing the remaining liquid (approximately 0.5 ml) was purged and the freeze was removed from the apparatus and cut (radially) into slices with 0.5 cm thickness. The slices were thaw and the concentration was determined by spectrophotometry (absorbance at 400 nm) . Figure 2 Shows the crioconcentration profile i.e. the ration between the concentration of the sample and the initial concentration of the solution (C/C 0 ) (-20°C curve). This example shows that crioconcentration accented beyond the height of 0.05 m, what demonstrates that this is the maximum height at -20°C in order to obtain an ice matrix with uniform distribution of solutes.

Example 3

This example demonstrates the importance of the linear velocity of interface growth in dimensioning the critical height. Three temperature probes (thermocouples type T) were inserted at various distances from the refrigerated base during the freezing of a bovine hemoglobin solution (Hgb) lg/dm3 in a phosphate saline buffer (PBS) 1/15M pH =7.4 using the same apparatus described in examples 1 and 2. The temperature profiles obtained were used to fit a computational fluid dynamics model that allows to calculate the linear velocities of interface growth for pure water, described in Figure 3. The model determines the time required for temperature to reach -1 °C (for which it is assumed the freezing occurred) depending of the height. The model solves the equation for heat transfer eq. (1) :

^- (p - Cp - T) - V 2 (k - T) - S Q = 0

dt eq. (1) where p is the density [kg/m J ] ; Cp is the thermal capacity [J/ (kg K) ] ; k is thermal conductivity [W/ (m K) ] and S Q is a heat source that corresponds to the latent heat release as a result of freezing. To model considers the following approximations:

1) Convection does not exist and therefore Navier-Stokes equations are not required;

2) Eq. 1 is solved only in vertical direction (unidirectional freezing) ;

3) Heat transfer occurs at the base (at the temperature of the cryogenic fluid) and at the insolated side walls at room temperature.

The mass transfer equation is solved for each time interval; in a freezing cycle. In this model the process of freezing implies to satisfy three conditions:

1) The temperature of the control volume is lowering;

2) The control volume it is not totally frozen i.e. the fraction of ice (ctl) is below 1;

3) The temperature of the control volume is lower than the equilibrium freezing temperature. When the conditions described above are satisfied, a heat source S Q is calculated by equation eq. (2) : where S Q is the heat generated per unit of volume [W/m 3 ] , p2 [kg/m 3 ] is the density of liquid water, Cpi [J/ (kg K) ] is the specific heat capacity of the control volume, Ti [K] is the temperature of the control volume T eq [K] is the freezing equilibrium temperature; At [s] is the time interval between the last iteration and the current iteration.

The mass of ice formed per unit of volume [kg/m 3 ] S M is calculated by S Q and the enthalpy of melting Ah = 330000 J/kg, according to equation eq. (3) :

eq. (3)

Finally, the equation VOF, eq. (4), is solved, where ccl is the ice domain, S M the mass of ice produced and p the density of the region containing both ice and liquid: eq. (4)

The time integration of eq. (4) is described in equation eq. (5) :

al New =al old +^-At

eq. (5) Example 4

This example demonstrates the production of frozen solutions with uniform distribution of solutes · by freezing a height of liquid of 4 cm, i.e. 89% lower than the critical height using the refrigerated base temperature constant at -20 °C. This height corresponded to a total volume of 42 ml of a bovine hemoglobin solution (Hgb) lg/dm 3 in a phosphate saline buffer (PBS) 1/15M pH =7.4 using the same apparatus described in examples 1 and 2. The volume was evenly distributed by the three plates. The concentration profile obtained in this example is shown in Figure 4.

References [1] - Wang W., Singh S., Zeng D., King K., Nema S., J. Pharm. Sci., 2007, 96.

[2] - Butler M.F., Crystal Growth & Design, 2002: 2 (6) : 541.

[3] - Miyawaki 0., Food Sci. Technol. Res., 2001, 7(1) :1.

[4] - Gu. X., Suzuki T., Miyawaki 0., J. Food Science, 2005, 70 (9) : 546) .

[5] - Patente US6855354.

[6] - Patente JP3188903.

[7] - Patente US4438634.

[8] - Bhatnagar B. S., Pikal M. J., Bogner R. H. , J. Pharm. Sci. , 2008, 97 (2) : 798.

[9] - Singh SK, Kolhe P, Wang W, Nema S. Bioprocess Int. 2009/7 (10) :32-44. [10] - Parag Kolhe, Elisabeth Holding, Alanta Lary, Steven Chico, Satish K. Singh BioPharm International 2010, 23(7).

[11] - Wilkins J, Sesin D, Wisniewski R 2001. Innovations in Pharmaceutical Technology .1 (8 ): 174-180. [12] - Webb SD, Webb JN, Hughes TG, Sesin DF, Kincaid AC 2002. BioPharm 15 (5) : 22-34.

[13] - Butler MF 2000. Crystal Growth & Design 1(3):213.

[14] - Wakisaka, M . , . Shirai, Y. and Sakashita, S. (2001) Chemical Engineering and Processing 40, 201-208). [15] - Shamlou PA, Breen LH, Bell WV, Polio M, Thomas BA 2007. Biotechnol Appl Biochem 46:13-26.

[16] - Patente EP1407202B1.

Date: November, 25 th , 2011