Freeze Drying Principles

Lyophilization is the sublimation of ice from a frozen material to leave behind a stable dessicated product. For practical purposes most of the lyophilization process is performed at sub-ambient temperatures under vacuum, to ensure that the material remains completely frozen and to enhance the sublimation rate while protecting the biological material from thermal damage. The lyophilization process can be seen as comprised of three basic steps (Figure 20.1):

• freezing, under conditions that will result in the conversion of the maximum amount of the water present in the sample to ice crystals. The biological material together with any excipients and some water is freeze concentrated to form an extremely viscous liquor, a glass, containing an amount of non-frozen (amorphous) water. This is discussed further below.

• primary drying under vacuum, at a temperature below the collapse or glass transition temperature (see Section 20.4 below for explanation). The frozen component of the water will be removed by sublimation during this stage. The amount of water remaining in the amorphous glass will vary from one formulation to another and may be relatively high.

• secondary drying, at an elevated temperature, still under vacuum, often a deeper vacuum than for the primary drying. This step removes much of the remaining amorphous water from the glass.

Freezing 1ry drying 2ry drying

Freezing 1ry drying 2ry drying

Time (hours)

Figure 20.1 Schematic of a freeze drying profile.

Time (hours)

Figure 20.1 Schematic of a freeze drying profile. Freezing

The successful freezing of biological materials prior to commencing sublimation is of crucial importance in obtaining a stable and visually acceptable freeze-dried product. Since reviews of this topic are available (Patapoff & Overcashier 2002) this stage will be mentioned only briefly here. As a mixture of biological macromolecules, inorganic salts and water are reduced in temperature below the freezing point (or more accurately, the melting point as the freezing point may be depressed by supercooling) of the mixture, water will be removed from the liquid phase as it crystallizes to ice. The temperature at which ice nucleation occurs is usually below the freezing point, the amount below being termed the degree of supercooling, and will depend upon a number of factors including the presence of nucleating agents, the composition of the solute material and influence of particulate matter. The degree of supercooling has been shown to be important in influencing the length of primary drying that is required (Searles et al. 2001a). This is thought to be due to the size and structure of the ice crystals formed. A small degree of supercooling results in slow crystallization with large crystals forming through crystal growth. Larger crystals allow large pores to remain as ice sublimes, and so there is less resistance to further sublimation from within the product cake and drying proceeds more quickly. Conversely, small pores may restrict the lyophilization rate and can result in poor product appearance. However, some materials may be insufficiently robust to withstand long freezing steps and so may require snap freezing.

Most proteins and carbohydrates do not crystallize on cooling. They are increasingly concentrated by the removal of water as ice. Some materials will crystallize when maximally concentrated and the temperature at which crystallization is maximized is termed a eutectic point. An isotonic saline solution will reach an effective concentration of 3M NaCl when frozen to its eutectic point and similar concentration of other components will also occur (Franks 1990). Other materials, including most biological materials, will continue to become concentrated as the temperature falls until they reach a maximally concentrated state and form a glass, stable at that temperature. The temperature at which this glass formation occurs is termed the glass transition temperature (7g). This glass will still contain some water, which remains uncrystallized and is amorphous, being a constituent of the glass (see Figure 20.2)

Primary freeze drying has to be undertaken at a product temperature (or rather the freeze drying interface temperature within the product) below 7g, otherwise the glass will be unstable and release fluid water that may be sufficient in quantity to 'collapse' the freeze-drying product. It is important to note that there may well be heterogeneity across the shelves and between shelves of the freeze drier in terms of temperature during freezing and sublimation. The shelf temperature in itself is not the critical factor; it is the temperature within the frozen product, which will be affected by the efficiency of heat transfer between the shelf and the product (i.e. the quality of the vial finish). Temperature mapping within the freeze dryer and examination of the distribution of complete freezing should be part of the process validation. However, temperature probes in containers cause the product in that container to behave in an atypical manner as the probe can be a route for additional heat and can induce nucleation. Typically, the duration of the freezing step should be continued for some time after freezing occurs in the containers with temperature probes.

Since crystallization occurs from the bottom of a container, slow freezing also gives rise to the exclusion of non-freezing material to the top of the liquid. In the extreme case, a meniscus skin can form a barrier to sublimation which compromises freeze-drying. Annealing

During the freezing stage the product temperature can be raised, held for a period, and subsequently lowered. This procedure is known as annealing. Some authors have suggested the use of 'annealing' stages to ensure greater uniformity in achieving a fully frozen state across the o




S /




i i i

Figure 20.2 Idealized phase diagram for a hypothetical small carbohydrate to show the relationship of phase to temperature and concentration. If a temperature between Tf1 and Tf2 is applied, the product can, above the Tg, recrystallize, start melting, or remain in the amorphous phase, depending upon the concentration of the dissolved substance. Below Tg, or at concentrations smaller than Cg, crystallization is possible. Key: A: amorphous solid, E: ice, S solution/liquid. Reproduced from Oetjen (1999) with kind permission from Wiley VCH

batch (Searles et al. 2001b). Annealing has been recommended by several practitioners and has the advantage that it can induce larger ice crystal formation through crystal growth - Ostwald ripening (Searles 2004) - which can speed the rate of subsequent sublimation. It can allow cycles to be shortened as the better pore structure allows faster primary drying and so this stage can be completed more quickly. By crystallizing out troublesome formulants the overall product Tg can be raised to make lyophilization a practical possibility. However, if these same formulants also provide lyoprotection to the biological (see Section, then the risk of increased lyophiliza-tion-induced damage must be weighed against these benefits. Primary drying

Once the sample to be freeze dried is fully frozen, a vacuum may be applied and the primary sublimation begun. As the sublimation rate is related to the vapour pressure of water above the product compared with that above the condenser where the ice re-forms, it is essential to have cooled the condenser initially to a temperature below that of the product, typically a temperature of —70 °C. In practice, most production-scale freeze-driers cannot attain a shelf temperature below —55 °C. This limits the freeze drying to formulations with a Tg greater than about —50 °C, although the use of liquid nitrogen to cool condensers may give some additional flexibility.

The choice of vacuum conditions is a compromise, as a deep vacuum will induce fast sublimation rates but this will be tempered eventually by the impact of a high vacuum in inhibiting heat transfer within the chamber and so reducing the heating capability. Indeed, it is preferable to have a weaker vacuum but to maintain efficient heat transfer within the chamber. This can be achieved by means of a vacuum bleed needle valve which allows air (or nitrogen if preferred) into the chamber to maintain the internal pressure at the programmed value (Rowe 2005).

For a given container, the heat transfer into the product depends on the shelf temperature and chamber vacuum. At 'low' pressures, the principal heat transfer is by conductance through the container's points of contact with the shelf, together with radiation from the metalwork within the chamber. At 'higher' pressures, heat transfer by convection of the gaseous molecules within the chamber is significant. The materials slowest to freeze dry are typically those in containers located in the middle of the shelf; a door with a large viewing port can also induce heterogeneity in speed of drying. Modification in the heat flow to the product can be affected more quickly by chamber pressure changes than by shelf temperature changes.

The temperature required to maintain the product in an immobile state (frozen glass) is determined by the glass transition point of the formulation, as explained in the section above. Product temperature should be maintained below the glass transition temperature, but for economic reasons should not be very much below this temperature, as the cost of maintaining production-scale freeze dryers at these temperatures is high.

The heat required for sublimation is taken from the frozen product. This is replaced by heat transfer from the shelves and shelf fluid. If the shelf temperature is maintained at the product freezing temperature, the product temperature will typically be less than that of the shelf due to the loss of heat of sublimation. To compensate, more heat can be put into the product by raising the temperature of the shelves. Heat will be transferred to the product by conduction from the shelves, by radiant heat from shelves and sides of the drier, and carried both to and from the product by convection via what atmosphere (of water vapour and gases) remains within the chamber.

As the sublimation proceeds so more heat can be put into the product, and for a larger bulk of freeze drying material the temperature of the shelf can be maintained above the glass transition as a gradient of heat will occur in the product cake itself. The rate at which the shelf temperature is raised (referred to as ramping), which can be used safely without causing collapse of the product cake, will need to be determined empirically for each depth of fill and container type. It may well vary as the size of the batch and possibly the dimensions of the freeze drier chamber are changed, and also the capacity and load on the condenser.

Once the crystalline water ice has sublimed, the product temperature may be allowed to rise above the original glass transition temperature. The end of primary drying can be determined in a number of ways. In the pressure rise (or barometric) test, the isolator valve between chamber and condenser is closed for a short period of time during primary drying, and the rise in chamber pressure measured as an indication of the rate of sublimation of water vapour. As primary drying reaches its end, the rate and amount of sublimation occurring will fall and this will be indicated by a negligible change in the chamber barometric pressure on closing the valve. This test is non-invasive but it does perturb the freeze drying process if primary drying is not yet complete, and as such is seen as suitable for validation tests but may be less appropriate in routine manufacture.

Another test for the completion of primary drying is to follow the temperature profile of containers fitted with thermocouples or resistance thermometers. As stated above, data from these containers must be treated with caution; they are atypical, freeze drying proceeding more quickly than in a container with no probe in a similar position. The temperature of the product is initially below that of the shelf due to cooling from the sublimation. However, as the primary drying nears completion the temperature rises to the shelf temperature or even exceeds it. An additional safety period should be implemented to allow the slower-drying containers of the batch to catch up.

Other more involved monitoring methodologies have been suggested including the monitoring of the vapour phase (via the vacuum pump exhaust) by mass spectrometry (Connelly & Welch 1993) and the comparison of pressure measurements as indicated by different barometric monitoring devices (Milton et al. 1997). Secondary drying

A product at the end of primary drying may still contain as much as 10 % residual moisture (Franks 1990) as amorphous water. If brought directly to room temperature and pressure this product would be unsuitable for long term storage. Collapse would be likely or hydrolytic degradation might proceed at an unacceptable rate. Such collapse may not be seen immediately as the release of water from the glass, although thermodynamically favoured, is kinetically slow. For this reason moisture content is further reduced by the use of secondary drying under vacuum at near ambient or even elevated temperatures, and hard vacuum used to remove the remaining water down to very low residual levels. Residual moisture in the lyophilized product can be a source of product instability, and dependent upon the material in question, and the intended shelf life and storage conditions, a maximum residual moisture level should be set and the secondary drying should be optimized to achieve this level or lower. For therapeutic products the acceptable levels of moisture may be 1-3 % by weight. For some applications even lower levels may be required, as for instance with biological standards, which need a long shelf life, where a level 1 % or lower is preferred.

As a guideline it has been suggested that a secondary drying period of one-third the length of primary drying be used (Murgatroyd 2001) with a deeper vacuum than is used in the primary drying, typically 1-30 |ibar. Some have claimed that pressure rise tests can be used as indicators of the end of secondary drying as in primary drying, although the pressure difference will be far smaller.

Was this article helpful?

0 0
Healthy Chemistry For Optimal Health

Healthy Chemistry For Optimal Health

Thousands Have Used Chemicals To Improve Their Medical Condition. This Book Is one Of The Most Valuable Resources In The World When It Comes To Chemicals. Not All Chemicals Are Harmful For Your Body – Find Out Those That Helps To Maintain Your Health.

Get My Free Ebook

Post a comment