Freeze-Drying Process Optimization

Kisialiou Yury/Shutterstock.com

Freeze-drying is widely used in the pharmaceutical industry to stabilize drug products. During freeze-drying, solvent is removed from a frozen product by sublimation to produce a dry product that can be readily reconstituted to its original liquid formulation by the addition of solvent. The process prolongs shelf life of the product by inhibiting chemical and physical degradation pathways that occur in the presence of moisture. The process, however, is one of the most expensive unit operations in a complete manufacturing line because of the slow drying rate arising from inefficient heat transfer, and the high investment and operating costs (1, 2). Cycle optimization is, therefore, crucial, especially for products produced on a large scale, where the cost factor increases significantly with time spent in the freeze dryer (1). A fully optimized cycle uses only the energy and time required, resulting in shorter process time and higher product throughput (3). Most importantly, it ensures optimum product quality and consistency between batches. 

Freeze-drying involves three basic steps-freezing (solidification), primary drying (ice sublimation), and secondary drying (moisture desorption)-which can take several days to complete. A fundamental goal of freeze-drying is to produce a well-dried product with elegant cake appearance, short reconstitution time, long shelf life, and complete recovery of activity on rehydration (3, 4), among other requirements. Two major considerations for optimization of a freeze-drying process are the stability of the drug product, during the process itself and during the storage period after the cycle is terminated (5). Designing an optimized freeze-drying cycle requires identifying the critical properties of the formulation so that the processing conditions can be specifically tailored to the formulation’s freeze-drying characteristics, such as the collapse temperature (T_c), glass transition temperature (T_g), and eutectic temperature (T_eu) (3). As Arnab Ganguly, technology manager, IMA Life North America, underlines, thermal characterization of the product is essential to understand product limits.

The importance of proper freezing

The freezing step is of paramount importance in a freeze-drying cycle because it is a known equivalent dehydration step. Further, it determines the ice crystal morphology and pore size distribution of the product, which in turn affect downstream steps such as primary and secondary drying, observes Paritosh Pande, research scientist, IMA Life North America. Proper freezing creates the foundation for efficient and consistent freeze-drying cycles, according to T.N. Thompson, president of Millrock Technology (6).

“For freezing, we want to make sure that the product is thoroughly solidified,” says Steven Nail, principal scientist at Baxter Biopharma Solutions. He explains that freezing is often carried out at a shelf temperature of -40 °C to -60 °C for a period of time long enough to establish a steady state. 

Critical parameters in primary drying 

Once freezing is complete, primary drying begins. “The chamber pressure is reduced, and shelf temperature is raised to supply heat for the ice to sublime,” says Vaibhav Kshirsagar, associate scientist, IMA Life North America. “Normally, most of the water is removed at this stage.” 

Primary drying is usually the longest step, taking up the largest fraction of the freeze-drying cycle time. Optimizing this process step has been shown to have significant economic impact (5). The temperature and pressure during primary drying will determine the rate of solvent removal and product temperature, according to Ganguly. “At or close to the edge of failure, increasing the temperature and pressure will speed up the primary drying process, but it may adversely affect product quality,” he says. 

“Shelf temperature and chamber pressure are critical because of their influence on heat transfer, mass transfer, and ultimately product temperature-which is the most critical process variable, even though we can’t control it directly,” Nail explains. “Ideally, you want the product temperature to be a safe distance below the upper product temperature limit (usually the collapse temperature), without being so low that it unnecessarily slows down the process.” 

Pande cautions that drastic increase of the shelf temperature in favor of more rapid ice sublimation during primary drying can lead to the collapse of an amorphous product or result in a eutectic melt for a formulation containing a crystalline product. “Both the collapse and eutectic melt can have significant consequences on product stability as well as for product appearance,” he says. 

Kshirsagar further adds that heat transfer to the product is not only dependent upon temperature and pressure, but also on the product containments such as whether they are contained in vials or syringes. He notes that increasing the pressure in the chamber will generally increase the heat transfer to the product, thus, increasing the mass transfer to the condenser. 

Finding the optimum process conditions

The optimal temperature and pressure during primary drying are determined based on the product collapse temperature and the freeze dryer’s capabilities. “The optimum process conditions will prevent loss of pressure control in the chamber and maintain the product temperature below the collapse temperature,” says Kshirsagar. Pande highlights that there are also other variables that can influence the primary drying process such as fill volumes, vial size, freeze dryer geometry, nature of the solvent, and product concentration. 

“To ensure successful primary drying, the product, process, and equipment characteristics must be factored in when deciding the appropriate conditions,” Ganguly explains. “The process of identifying appropriate conditions for primary drying typically begins with thermal characterization of the product of interest. During thermal characterization, the critical collapse temperature of a product is identified by factoring in its glass transition temperature (T_g’), as determined by differential scanning calorimetry, and identifying its collapse temperature through freeze-drying microscopy. The critical collapse temperature determines the maximum temperature that a product can withstand during primary drying without undergoing melt or collapse.”

From an equipment standpoint, Kshirsagar notes that aggressive cycles may be possible for products with high collapse temperature. “However, such aggressive cycles may sometimes lead to condenser overload or choked flow in the freeze dryer, if not well characterized,” he says. “In such situations, the equipment capability will be a limiting factor, and the temperature and pressure would be chosen accordingly.” 

According to Pande, appropriate primary drying conditions can be determined with good accuracy using a combination of thermal characterization data and steady-state mathematical modeling performed using first principle methods. “This approach eliminates the need for trial and error,” he says. 

Baxter’s approach involves constructing a graphical design space to determine primary drying conditions. “For us, that is a plot with sublimation rate on the y-axis, chamber pressure on the x-axis, and two sets of isotherms-one set for shelf temperature and the other for product temperature,” Nail explains. “We do this by measuring the vial heat transfer coefficient as a function of pressure and the resistance of the dried product layer to flow of water vapor. One of the boundaries design space is the upper product temperature limit, and the other is the equipment capability curve. The highest sublimation rate within the resulting ‘space’ gives the optimum shelf temperature and chamber pressure.” 

The most widely used tools to monitor primary drying are pressure sensors, including the capacitance manometer and Pirani gauge. Temperature probes, which include thermocouples, resistance temperature detectors, or wireless sensors, are typically used during engineering runs, observes Ganguly. “Residual gas analysers and mass spectrometers can be used to determine the gas composition inside close environments and have proven to be extremely useful to monitor trace concentrations of contaminants such as silicone oils or even for process monitoring in primary and secondary drying (7). Tunable diode laser absorption spectroscopy is a well-established tool for monitoring the vapor flow rate between the chamber and condenser in real time. Heat flux sensors mounted on the freeze dryer shelves can be used to quantitatively obtain the vapor flow rate if prior heat transfer coefficient (Kv) measurements are performed,” he says. “These process-monitoring techniques gather enormous amount of data that can be used to determine the end point of a primary drying process.”

Shelf temperature during secondary drying 

The purpose of secondary drying is to remove remaining unfrozen water that is either loosely associated with the crystal surfaces or buried within the glassy phase (1). The relatively small amount of water is removed by desorption, “typically to a final moisture content of 1–3%,” says Pande. “As the moisture content approaches 1%, the choice and concentration of excipients play an important role in preventing damage to the activity of any biological product,” he continues. “Excipients such as trehalose, for example, can better offset the effects of dehydration for a freeze-dried protein like immunoglobulin G by offering better H-bonding and a phase-homogeneous matrix with more suppressed β-relaxation, compared to large-sugar excipients such as dextran or inulin.” 

Nail highlights that there is only one critical process variable during secondary drying, and that is the shelf temperature. “Both published literature and our own experience show that the chamber pressure doesn’t seem to have any effect on either the rate or the extent of secondary drying, probably because the rate-limiting step in secondary drying is diffusion of water through the glassy mixture of partly-dried solids,” he adds.

“We nearly always use a secondary drying temperature that is room temperature or above,” Nail says. “As a general rule, we like to keep the secondary drying shelf temperature as high as we can, because it results in a faster rate of secondary drying, and probably a greater extent of secondary drying.” 

In addition, Kshirsagar points out that the temperature of the product needs to be maintained well below the T_g during secondary drying. The rate of water desorption during secondary drying step is sensitive to the specific surface area of solid, he notes, which is impacted by the ice nucleation event occurring in freezing stage. 

Kshirsagar explains that the rate of water desorption during secondary drying can also be influenced by the primary drying step. “Aggressive primary drying above the glass transition temperature can generate macro- or micro-collapse in the product cake,” he says. “To avoid product collapse during secondary drying, it is imperative that maximum water sublimation is achieved during the preceding primary drying step.” He reiterates the importance of maintaining shelf temperature during secondary drying several degrees below the T_g of the product. “Because the primary driver of water removal during secondary drying is the shelf temperature, as one example, it can be optimized by choosing a shelf temperature close to the T_g while maintaining a safety buffer,” he says. “In addition, a slow shelf temperature ramp rate is preferred to avoid exceeding the T_g and leading to viscous flow in the product.”

Nail explains that the appropriate combinations of shelf temperature and drying time are determined by taking “thief” samples during secondary drying and measuring the residual moisture content relative to a target residual moisture limit.

According to Kshirsagar, the Pirani and capacitance manometer pressure sensors alongside thermocouples can be used in the early stage of secondary drying. Pande adds that the use of mass spectrometers can greatly improve the secondary drying process monitoring, by which the exact gas composition in the chamber can be known and potentially related back to the batch average residual moisture (7). 

Conclusion

Determining the right process conditions for a freeze-drying cycle requires an understanding of the effect of each step on the drug product. A confirmation run should be conducted at the end of cycle design to ensure that the process is robust and efficient and that it produces acceptable products.

References

1. J. Schwegman, Technical Note: Basic Cycle Development Techniques for Lyophilized Products (Nov. 11, 2009).
2. D. Fissore et al., Pharmaceutical Engineering 29 (5) 58–68 (2009).
3. K. Scoffin and L. Ciccolini, BioPharm International 26 (3) 46–51 (2013).
4. F.K. Bedu-Addo, Pharmaceutical Technology 27 (1) 10–16 (2004).
5. X. Tang and M. Pikal, Pharmaceutical Research 21 (2) 191–200 (2004).
6.A Siew, Pharmaceutical Technology 37 (5) 36–40 (2013).
7. A. Ganguly et al., European Journal of Pharmaceutics and Biopharmaceutics127, 298-308 (2018). 

Article Details

Pharmaceutical Technology
Vol. 42, No. 5
May 2018
Pages: 18–23

Citation

When referring to this article, please cite it as A. Siew, “Freeze-Drying Process Optimization,” Pharmaceutical Technology 42 (5) 2018.