OR WAIT null SECS
Mattias Haag is Manager Mechanical Engineering at Ramboll, Stockholm, Sweden Tel. +46 (0)10 615 62 49 Mattias.Haag@ramboll.se
The author presents a method to calculate the relationship between supply air volume flow and airborne particle concentrations.
Particulate contaminants are a very important factor in the pharma industry, and understanding particulate contamination sources and their behaviour is critical to controlling their spread within production areas.
Ambient particles, potentially carrying microbial contaminants (viable particles), and nonviable particles can spread from surfaces to the surrounding air volume of a proposed clean area through airborne contamination and/or via persons, contaminating both products and the manufacturing area. The main source of viable particles to the surroundings is the operator. Other potential sources of microorganisms are pointed out in guidelines from regulatory agencies.1,2
One way of reducing contamination risk through airborne particles is to reduce the particle source strength. This can be achieved by capturing contaminants at the source; however, it is difficult to capture all contaminants and different types of protection are often required. In critical areas, for example, both barrier technology and controlled airflow directions are usually employed.
High Efficiency Particulate Air (HEPA) filters are used to remove particles from incoming air and displace those that may be spread into the controlled area. Large amounts of HEPAfiltered air are often required, but this solution is not always reliable. The actual contamination risk can be quantified by better understanding the potential for contamination, as well as how effective HEPA filtration is in mitigating contamination risks.
The model described in this article enables the contamination risk to be calculated.
Large amounts of HEPAfiltered air can produce vortices because of the high energy levels created by the air movements. A vortex can exacerbate a contamination source so it is important to reduce the occurrence of vortices to maintain a cleanroom or clean zone.
A vortex can be described as a rotational motion of a fluid that is maintained by an energy source; for example, a cleanroom's air supply. Investigation by Ljungqvist in 1979 showed that in ventilated rooms, vortex regions move as a rigid body or as a free vortex.4 Ljungqvist and Reinmüller describe a vortex as having streamlines that are closed within a region.3 Vortices can accumulate high concentrations of particles that are released when the vortex is disturbed. Disturbances can occur when people cross the vortex or when there is a major change in pressure or airflow routes, such as when a door is opened. This is significant with both free and rigid vortex structures.
Figure 1: Convective deposition of particles.
Another type of air disturbance phenomenon is turbulence. Turbulence can be described as a fluctuating, unsteady flow of fluid with large velocity differences perpendicular to the net flow direction. The science regarding turbulence is complex and not fully understood, but it can create difficulties regarding particle dispersions.
Particles spread within a volume because they are affected by an external force, such as air volume mass flow (diffusion and or convective spreading, i.e., through vortices or turbulence), gravitational deposition, electrostatic forces or thermal differences.
Gravity creates a downward motion on particles based on mass, and there is a ratio between the mass weight of the particle and the settling time. Free-floating particles within an air volume will also have a gravity effect against each other; however this force is very small and, in a pharmaceutical environment, it can be neglected.
Particles and surfaces can be positively or negatively charged via static electricity. Particles with a positive charge will be drawn to a particle or surface with a negative charge. The potential for static charge also depends on the material; for example, glass has fewer tendencies to be charged than plastic materials. A material's tendency to develop static electrical charge is an important issue to consider when it comes to using them in a cleanroom. A surface with a positive charged material can attract negatively charged particles, which will accumulate on the surface. If a surface gets grounded, particles may also come loose and spread to the surroundings.
The energy in airflow influences particles to move parallel with the airflow vector, i.e., particles are spread in the same direction as the air movement (Figure 1). This knowledge is commonly used in the pharmaceutical industry to create clean areas and air barriers. Unidirectional airflow (UDF) units have a parallel airflow such that airborne particles are transported away in the flow vector direction. The flow in a UDF unit is either laminar or turbulent. Ljungqvist and Reinmüller tell us that "it is assumed in a parallel flow field that, next to surfaces along the main flow direction, there is a thin sub layer (boundary layer) in which the transfer of momentum is dominated by viscous forces, and the effect of weak turbulent fluctuations can be neglected. The situation is quite different for particle diffusion. In this case, even weak fluctuations in the viscous sub layer contribute significantly to transport".3
Figure 1: Convective deposition of particles.
Gas molecules colliding with airborne particles give rise to diffusive spreading (Figure 2). The effect of molecular diffusion may be considered when airborne particles are small (microscopic), which may occur in cleanrooms and clean areas.
Figure 2: Diffusion of particles.
Acknowledging the fact that cleanroom operators are often the main source of microorganisms, it can be useful to study the probability that microorganisms will contaminate certain areas associated with operators. To calculate the gravitation deposition of a concentration of airborne viable particles, Whyte describes an equation (Equation 1).5
The size of viable airborne particles from humans in a cleanroom environment is in the range of 5–20 µm. An assumption can be made that the average viable particle size is 12 µm. Using Stokes Law, Whyte shows that the settling velocity for a 12 µm particle in air is 0.462 cm/s (Equation 2).
The knowledge of settling velocity is useful in volumes where the particles are only affected by gravity and not from major airflow routes such as in cleanrooms. In a pharmaceutical environment, however, air is used to create barriers with air velocities much higher than 0.462 cm/s. In critical areas using UDF units, common air velocities range between 0.3–0.5 m/s or 30–50 cm/s. Compared with the settling time of a 12 µm particle (0.462 cm/s) affected only by gravity, the convective effect on the particle in a critical area will be dominating and the effect of gravity can often be neglected.3
If only the convective effect is considered, the equation can be used to calculate the number of deposited particles in UDF, which enables the contamination risks for different types of processes to be calculated. Whyte gives examples of how to calculate contamination rates for different container sizes and exposure times, regarding gravity settling time (Table 1). In most sterile processes, 1.350 bacteria/m3 is considered to be a high value. For example, a Grade D cleanroom environment according to EU GMP is defined as having a limit less than 200 CFU (viable particles)/m3 . Whyte reports an average value of 0.2 bacteria/m3 in a wellworking UDF unit (corresponding to Grade A clean room designation per EU GMP). This demonstrates a significant change in increased product safety in the calculation.
Table 1: Calculated contamination rate in containers - settling velocity from gravity.
A comparison of contamination rates can be conducted if the settling velocity from gravity in the calculation is replaced by convective settling velocity. Results in Table 2 present calculations for the same containers as Table 1, except for their placement under a UDF unit (Grade A), with a UDF velocity of 40 cm/s. Calculations are based on the assumption that gravity force is negligible. Grade A, according to the EU GMP recommended limit of less than 1 CFU/m3 is used in the calculation, but so too is an average value of 0.2 CFU/m3 because this more accurately represents a true UDF environment.
Table 2: Calculated Contamination Rate in Containers - settling velocity from UDF.
In Table 2, where the calculations are adapted to UDF, there is a higher potential risk of contamination of the containers. With a wellworking UDF unit and operation routines; however, the amount of airborne viable particles (CFU/m3 ) is most likely lower than 0.2 CFU/m3 . When applying Equation 1 in a volume with completely mixing air or with no air movements, the Stokes law (Equation 2) must be used to calculate settling velocity for particles. In a UDF, the settling velocity is equal to the air velocity.
As demonstrated by the calculations, contamination rates are dependent on different factors. It is not only the amount of airborne viable particles that must be considered; so too must the exposure time and the exposed area to those viable particles. Sundström, Ljungqvist and Reinmüller have used this model to calculate particle concentrations in small-volume parenterals produced by blow-fill-seal technology.6 They have also performed experimental studies with analyses of the amount of particles that disperse into the ampoule from the surroundings as a comparison to the theoretical method.
This knowledge can be used for various purposes. One example is during investigations of deviations in the amount of airborne viable particles (CFU/m3 ) within a cleanroom. The model can be used to calculate the risk of contaminating a container or surface, although an investigation as described has to be more complex, and other practical and experimental methods will need to be considered.
As a practical example, the model can be used to calculate the contamination rate if the exposure time for ampoules in a blowfillseal filling machine is increased. Many types of blowfillseal filling machines generate a high amount of particles during plastic extrusion. Often, these particles are removed through local exhaust devices inside the machine region. To improve the local exhaust devices and thereby remove more airborne particles, the cycle time was increased, allowing the mould to stay at the extrusion position for a longer time. In this case, the concentration of airborne particles in the filling machine was reduced. However, the increased cycle time and exposure time must also be taken into account when evaluating the overall risk situation. The calculated contamination rate was increased 20% due to increased exposure time (20% longer). The concentration of airborne particles was decreased by 25%. In the end, considering both the increased exposure time and the reduced concentration of airborne particles, the contamination risk decreased by 5%.
As another example, comparisons can be made between different filling techniques. If vials in filling process A have the same exposed area, airborne concentration and settling velocity as ampoules in process B, but twice the exposure time, the calculated contamination rate is doubled for process A. The guidelines1,2 do not make any exceptions between these two techniques, but in theory, the airborne concentration at process B could be twice as high as for process A and still have equal contamination risk.
The calculation models presented in this article can be helpful in calculating risk factors and increasing the understanding of contamination caused by airborne particles. The method described shows that it is possible to calculate the risk factor or the rate of contaminations to a certain area. The theoretical risk to contaminate, for example, a container or ampoule, depends on the exposed neck area, airborne particle concentration and exposure time.
Mattias Haag is Validation Manager at Energo, Stockholm (Sweden).
Tel. +46 10 470 60 00
1. EU GMP European Commission, The Rules Governing Medicinal Products in the European Union, Vol. 4, EU Guidelines to Good Manufacturing Practice, Annex 1, Manufacture of Sterile Medicinal Products, 1997 (revised 2008).
2. Food and Drug Administration. Guidance for Industry: Sterile Drug Products Produced by Aseptic Processing – Current Good Manufacturing Practice. Rockville, MD, 2004.
3. B. Ljungqvist and B. Reinmüller, Clean Room Design: Minimizing Contamination Through Proper Design (CRC Press LLC, Boca Raton, Florida, USA, 1997).
4. B. Ljungqvist, Some Observations on the Interaction Between Air Movements and the Dispersion of Pollution: Document D8:1979, Swedish Council for Building Research, Stockholm, Sweden (1979).
5. W. Whyte, Journal of Parenteral Science & Technology, 40(5) 188-197 (1986).
6. S. Sundström, B. Ljungqvist and B. Reinmüller, European Journal of parenteral & Pharmaceutical Science, 15(3) 87–92 (2010).