Filling of Asthma Inhalers with Liquid Nitrogen: Cryogenic Product Cooling of Aerosol Propellants

October 1, 2003
Stefan Terkatz, Thomas Kraft, Thomas Kutz
Pharmaceutical Technology Europe

Volume 15, Issue 10

This case study describes how a major pharmaceutical manufacturer was equipped with four filling lines, for metered dose inhalers, supplied with a nitrogen cooling system to prevent spontaneous vaporization of the propellant gas. By doing so, a cost-effective and environmentally friendly solution was provided to a hazardous situation, which also complied with regulatory directives.

Aerosols inhaled from a metered dose inhaler (commonly referred to as an asthma pump) can help to reduce the discomfort experienced by asthmatics. Such aerosols are suspensions comprising a propellant gas with a mixture of active substances. These substances are not dissolved, but remain as solid particles suspended in the propellant gas. To fill metered dose inhalers (MDIs), the mixture is first homogenized in a mixing vessel and then held in suspension by a stirrer. During this process the pressure of the mixing vessels adjusts itself to the vapour pressure of the propellant gas at ambient temperature. This is a problem because the suspension must be filled at atmospheric pressure. To avoid spontaneous vaporization of the propellant gas during the filling process, the suspension must be cooled between the mixing vessel and the filling equipment.

In the past, this was achieved using conventional refrigerants. But, as a result of new directives concerning the use of ozone-depleting substances in the European Union (EU) that prohibit the use of halons, these units had to be shut down. In this article, the authors describe how Boehringer Ingelheim Pharma equipped four filling lines with a nitrogen cooling system, implementing a cost-effective and environmentally friendly solution to the problem.

Respiratory diseases

The respiratory disease of asthma is a chronic allergy that can be life threatening. In the developed world, 5–10% of the population suffers from asthma - and the proportion is rising. Boehringer Ingelheim Pharma is one of the three pharmaceutical companies developing therapies for respiratory diseases. Their research and development departments place particular emphasis on exploring the two topics of asthma and chronically obstructive pulmonary diseases (COPD). In today's treatment of asthma, both palliative and remedial medications are employed. Inhalable aerosols fall into the first category, relieving the symptoms. For this purpose, the suspension with the active agent is filled into small canisters (Figure 1). The patient uses the inhaler by activating the valve, which generates a puff of spray that is breathed into the lungs.

Figure 1: The pharmaceutical form of an aerosol canister with its inhaler (Boehringer Ingelheim Pharma).

Metered dose inhalers

Various gas mixtures can be used as aerosol propellants. Until recently, these frequently included chlorofluorocarbons (CFCs) such as freon, which now fall foul of the new directives limiting the use of ozone-depleting substances. Currently, appropriate substitutes such as tetrafluoroethane (R134a) are employed instead.

The aerosol propellant is mixed with the powdered active ingredients in a mixing vessel and the mixture is kept in suspension by a stirring system. The active agents are not dissolved in the liquid aerosol propellant. In the uninsulated mixing vessel, the pressure self-adjusts to the propellants' vapour pressure. At an ambient temperature of approximately 21 °C the vapour pressure of typical aerosol propellants is between 3 and 5 bar(g).

For some products, filling must occur at atmospheric pressure. Were the suspension to be filled at the vapour pressure present in the mixing vessel, the result would be the spontaneous vaporization of the propellant upon expansion to atmospheric pressure. In practical terms, this would mean that little or no aerosol propellant would reach the product canisters. The correct operation of the inhaler could not then be guaranteed. Moreover, the concentration of active agents in the suspension would be subject to unpredictable changes. Therefore, the mixture is first cooled to below its boiling point at atmospheric pressure prior to filling at ambient conditions.

Initial situation

In this study, the filling of the aerosol canisters occurs in four filling lines. Each line has its own mixing vessel in which the product is prepared. From here, it flows through the product cooler (a conventional cooler using trichlorofluoromethane [R11] as the refrigerant). The central cooling system is connected to the four product coolers by a closed circuit heat transfer system. The refrigerant for this circuit is also R11 and it is used to keep the temperature of the product coolant circuit constant at 248 °C. The product cooler is constructed as follows:

  • The room temperature suspension first flows through a cooling coil in which it is cooled down to a target temperature of 243 °C.

  • The cooling coil is mounted inside a vessel through which the cold refrigerant R11 flows.

  • Prior to filling, the suspension enters a small mixing vessel that is also actively cooled by the same refrigerant. This mixing vessel serves as a buffer for a small quantity of precooled suspension while ensuring a homogeneous temperature and a consistent composition of the active substances. Additionally, the suspension is kept thoroughly mixed so that no deposition of active ingredients can occur.

  • The product then flows through a cooled pipe to the filling head of the filling line.

New cooling system requirements

The new cooling system must operate with an environmentally friendly refrigerant and eliminate the disadvantages associated with the original system.

  • Recently developed MDIs with new propellant gas mixtures (R134a and ethanol, for example) require filling temperatures as low as 255 low as 255 °C. The new cooling system must also be able to operate at this temperature. The temperature control system has to maintain the product temperature within a relatively narrow band.

  • In future, the four filling lines will be required to simultaneously fill different products at different temperatures so the new system must be equipped with a decentralized control system. Thereby, a failure in one cooling system will only result in the shutdown of a single filling line. With the old centralized system, a breakdown in the cooling system resulted in a shutdown of all four lines.

  • To clean the product cooler after a filling campaign, the equipment first had to be warmed up to room temperature. The existing system is, however, not fitted with any heating facility. Because of the insulation on the system, the warm up through natural heat transfer takes up to 2 days, during which time, the system is out of service. The new cooling system must have a heating facility installed so that the heat up phase can be notably accelerated, thereby reducing the shutdown time.

Cooling with liquid nitrogen

Liquid nitrogen is the ideal refrigerant for low temperature applications. The following reasons are decisive for this argument:

  • The boiling point at atmospheric pressure is 2196 °C. Therefore, it provides an almost unrestricted potential to reach low temperatures.

  • As the primary component in air, it is very environmentally friendly.

  • Nitrogen is practically inert.

The new application

The suspension flows from the mixing vessel to the product cooler. The operating principle of the new liquid nitrogen cooled system is the same as that of the old cooler (Figures 2 and 3). The suspension is first cooled to the target temperature in a cooling coil. From here, it flows to the depressurized buffer vessel (within the product cooler) and the vapour pressure is flashed off. A stirrer fitted in this small buffer vessel ensures that the suspension is further homogenized with respect to composition and temperature. Then, the product flows through the outlet piping, which is also cooled with liquid nitrogen, to the filling step. The metering of the dose is achieved by the opening and shutting frequency of a solenoid valve fitted in the filling head (Figure 4). Different quantities can be filled by varying the opening and shutting times. Immediately after filling with the cold suspension, the canister is sealed with a valve cap. As it warms to ambient conditions, the pressure in the canister reaches the equilibrium vapour pressure of the propellant gas. The canister then passes through a number of quality control steps prior to packaging. In principle, there is a risk that during the heat exchange process between the suspension and the liquid nitrogen the liquid part of the suspension freezes because the melting point of the propellant gas is much higher than the boiling point of liquid nitrogen. For this reason, the heat exchanger has been constructed so that the piping filled with product does not come into direct contact with the cryogenic nitrogen. Most of the liquid nitrogen vaporizes away from the cooling coil. The cooling of the cooling coil takes place exclusively with cold nitrogen gas. In this manner, it can be guaranteed that even during stoppages the product is not undercooled, which would result in it freezing. The new liquid nitrogen cooling system now offers the user almost unlimited possibilities to vary the temperature of all four filling lines independently of one another. Now, as opposed to the old cooler, there is practically no lower temperature limit. The system is thus ready for future development and will be able to reliably cool suspensions and solutions that have not even been developed yet.

Figure 2: Process flow of the nitrogen-cooled aerosol filling system (H+P 2002-227).

Double use of nitrogen

The liquid nitrogen that is used for cooling is stored in a vacuum-insulated storage tank. From here it flows through a vacuum-insulated pipe to the product cooler where the nitrogen transfers its cold energy to the product indirectly and vaporizes without being contaminated. The nitrogen gas is then injected into the existing nitrogen inerting network. The application is particularly economical because of this double use of the nitrogen.

Figure 3: The new product coolers are compact and fit into the existing process perfectly.

Control system

Three control circuits are required to regulate the process.

  • The product level in the cooled buffer vessel controls the flow of suspension from the mixing vessel to the product cooler. When the level is too low, a solenoid valve in the feed line is opened and the product flows to the cooler.

  • The opening and closing cycles of the solenoid valve in the filling head are set according to the weight of suspension in each canister.

  • The regulating variable for the filling temperature control is the temperature of the suspension in the 'stirred' buffer. This temperature indirectly controls the refrigerant flow. A direct regulation of the nitrogen control valve would lead to large temperature swings in the suspension. To prevent this, the regulation of the nitrogen control valve is actuated by a cascade regulation, which uses an algorithm for an intermediate temperature. This temperature control functions stably and even maintains the temperature within the desired parameters during shutdowns.

Warm up facility

At the end of a filling campaign for a particular product, the entire system, including the product cooler, must be cleaned according to the relevant regulations. To do this, the low temperature product cooler has to be emptied and warmed up to ambient temperature. Previously, this process could take up to 2 days because the old cooler had no heating facility. With the new product cooler it is very easy to swap from the cryogenic liquid nitrogen supply to the warm nitrogen gas of the plant's network. Only a few valves in the nitrogen supply and vent lines need to be activated. Using the warm nitrogen (approximately 21 °C) from the plant network the heat up period has been reduced from 2 days to approximately 2 h. By so doing, the capacity of the filling system is better utilized.

Figure 4: Filling with the cooled suspension.

Application advantages

The following advantages swayed Boehringer's decision to purchase the system:

  • an environmentally friendly refrigerant

  • a broad temperature control range

  • a decentralized cooling system

  • better utilization of system capacity because of the heating facility

  • high operational reliability (no moving parts such as refrigerator compressors)

  • low investment costs

  • low operating costs.


This process demonstrates that the potential applications for liquid nitrogen as a cryogenic refrigerant are far from exhausted. The attainable temperatures and environmental protection being the main reasons for its use. In the chemical and pharmaceutical industries, nitrogen is already used for inerting purposes - so the cold energy content of the liquid nitrogen can be effectively used without causing high operating costs.

Project details

Messer developed the new product cooler in conjunction with Boehringer Ingelheim Pharma. A patent has been applied for. During the development process, Messer contributed its experience in cryogenic processes while Boehringer Ingelheim Pharma provided its extensive production experience concerning the filling of MDIs. A prototype was built during the development phase and was tested in Messer's pilot plant test facility in Krefeld (Germany), as well as under production conditions at the customer's site. After these test runs Messer managed the project planning as well as the construction of the four filling lines. This included the product coolers, the measuring and control system, the electrical cabinets and the entire nitrogen infrastructure. Since then, the units have been in continuous operation on a two or three shift basis. Occasional interruptions in the production process result in shutdowns that are kept under control by the fully automated temperature control system. The project team worked to tight deadlines on account of production requirements. Only 7 months passed from the first contact to the commissioning of all four filling lines, which included the development of the application. This was only made possible because of the intensive collaboration: systematic project management was essential on both sides as well as precise agreement concerning the numerous interfaces between the new system and the existing infrastructure.