Selecting a freeze dryer

The most important consideration when choosing a freeze dryer is to ensure the system is fit for both today's applications and future needs. For the sake of this discussion, we will focus on freeze dryers with fluidfilled shelves, which does not include lowend manifold or heatonly shelf freeze dryers.

An understanding of the available features of a freeze dryer can facilitate the choice. The following are some considerations:

• shelf size (m² )

• shelf style (bulk or stoppering)

• condensing rate (L/h)

• condensing capacity (L)

• condenser location (internal versus external)

• material (304 versus 316SS)

• 21 CFR Part 11 compliance.

The freeze dryer manufacturer will also want to know what space is available for the freeze dryer and what utilities, such as electrical, air, chilled water and air conditioning, are available.

The following are a few examples of some of the various options:

• Cylindrical or rectangular product chambers — a cylindrical product chamber is less expensive than a rectangular chamber; however, it may occupy more floor space depending on the configuration of the shelf assembly.

• Internal or external condenser — an internal condenser is cheaper and provides unrestricted vapour flow. An external condenser is supplied with an isolation valve to separate the product from the condenser, which protects the product from reconstitution during power loss, and keeps the condensate out of the clean room environment.

• Pirani or capacitance manometer — piranis, the least expensive vacuum measurement device, read the relative vacuum inside a freeze dryer because they are affected by vapour. The more vapour present, the higher the pressure reading. A capacitance manometer reads the absolute vacuum level and the reading is unaffected by vapour pressure. Most production systems use a capacitance manometer for measurement and control of vacuum level. The best method for determining the 'end of primary drying' is to compare a pirani reading to a capacitance manometer reading. When they read the same, there is no vapour present and the product is dry. A quick test can be conducted by lowering the vacuum level to see if the pirani reading tracks the capacitance manometer. If water is present, the capacitance manometer will drop faster. If no water is present, they will drop at the same rate.

• Proportional vacuum control — the least expensive vacuum control system bleeds gas into the chamber using a solenoid valve, providing ±10 mT stability at 100 mT. For better stability, a proportional vacuum controller can be used that regulates the gas bleed through a proportional valve. The result is ±0.5 mT or better control.

Main freeze drying categories

Freeze dryer selection falls into two main categories: laboratory versus production, and non-sterile versus sterile.

Laboratory freeze dryers are used for a large variety of applications, including removal of solvent from a material, Phase I clinical trials and protocol development for scaleup production. A typical laboratory system will have a shelf area of 0.1–1 m² and a condensing capacity of up to 30 L.

Laboratorystyle systems can be simple freeze dryers with only standard features, such as a pirani gauge for vacuum level measurement and thermocouples for temperature monitoring, or they can incorporate more advanced instrumentation:

• capacitance manometer for vacuum measurement

• proportional vacuum control for fine vacuum control

• isolation valve between the product chamber and condenser for pressure rise testing

• liquid nitrogen traps for organic solvent trapping

• additional product thermocouples for monitoring product temperature.

Pilot and production systems offer shelf areas from 1 m² up to more than 40 m² . Production systems are used for Phases II and III clinical trials, and tend to be used for the same or a limited number of products in high-volume production. Recently, there has been a shift from using 10–50-mL vials, to 2mL and 5mL vials for smaller volume, highpotency biotech and proteinrelated products. The result is smaller freeze dryers with expensive payloads.

The type of processing will determine whether stoppering is required. Bulk applications can have fixedinplace shelves, but vial applications require stoppering where the shelves move and are squeezed together to press the partially inserted stoppers into the vial.

Pharmaceutical and other applications may also need to be sterilised between cycles, which can add significant complications and costs to a freeze dryer. A freeze dryer is normally rated for vacuum and the most common method of sterilisation is pressurized steam, which requires the freezedryer chambers to be certified pressure vessels rated to 2 atm at 131 °C.

An alternative sterilisation technique, which is growing in popularity for laboratory and small production systems, uses hydrogen peroxide (H₂0₂). H₂0₂ does not require a pressurerated vessel, which helps to minimise costs.

Validation and compliance

The most current demands on freeze dryers are validation and the ability to be 21 CFR Part 11compliant. These factors are a significant part of the cost of pharmaceutical processing freeze dryers.

For validation, a full component catalogue must be supplied. An installation qualification, operational qualification (IQOQ) document is generated that outlines the proper validation process, and a factory acceptance test (FAT) and site acceptance test (SAT) are implemented to verify that the system is supplied as ordered and performs within the required specifications.

To be 21 CFR Part 11compliant, the freezedryer software must encrypt all data to prevent tampering, and must log every change and entry on the computer control system using user log ins and password protection.

Misconceptions

The most common oversight is the concept that "all freeze dryers are the same". The choice of components, materials, construction and instrumentation create a wide variety in cost versus performance.

Older systems tend to have undersized compressors/condensers, as well as restrictions between the product chamber and the condenser, which limits the rate of freeze drying and often causes the freezedrying process to be extended. Today, freeze dryers are much better designed to accommodate the maximum load that may be placed in the system and the freeze dryer is not the limit to the process. For example, compressor reliability has significantly improved during the last 15 years. In small freeze dryers, the use of scroll compressors has virtually eliminated the failures common with reciprocating compressors.

As there are so many possible variations in size and features, advanced freeze dryers are 'built to order' where the end user works with the manufacturer to obtain a system suitable for their application requirements.

Innovations in freeze dryer technology

Thermal analysis and freeze drying microscopes have helped improve the understanding of the critical temperature — the temperature at which the product may collapse or melt-back — of the product being freeze dried. This knowledge provides the information required to produce a robust and efficient freeze drying cycle.

Classic freeze drying control is open loop, where the shelf temperature and chamber pressure are controlled based on a predetermined profile. It is assumed that the temperature of the product stays below its critical temperature, and the result is a reproducible — but very conservative and long — freezedrying cycle.

Closed loop control of the shelf temperature is required to both prevent collapse and minimise the length of the freezedrying cycle. The latest control systems use critical temperature information to dynamically control shelf temperature, which both protects against collapse and meltback whilst optimising the freezedrying cycle.

Methods that use an average measurement of the product in the chamber, such as calculated via pressure rise testing, adjust the temperature of the shelves a few times throughout the first half of the primary drying process. This process is limited to the first half of primary drying and only provides a conservative protocol. However, it is not optimised and does not take into account variations inside the chamber.

The latest control systems take into account both average and specific measurements to ensure there is no melt back and constantly control the shelf temperature throughout the entire cycle to produce a userselectable conservative or aggressive protocol.

In the future, more advanced closed loop control systems will be available that offer improved process control. Today, most applicable measuring instruments, such as tunable diode laser absorption spectroscopy, near infrared and mass spectrometry, are expensive and provide only marginal process improvement, which means they are not economically feasible for process control. As instrumentation and techniques advance, they will be incorporated into real-time process control systems.