Reduction of Airborne Microbial Count in Sieve Impactor Samplers During Sampling

5 hours ago

19 min read

Peer Review Article | Open Access | Published 29th September 2026 | Submitted 1st June 2026

Reduction of Airborne Microbial Count in Sieve Impactor Samplers During Sampling

Authors: Joao Mario Negrao Duarte Santos, StechComply, Portugal, W Whyte, University of Glasgow, UK | EJPPS | 313, (2026) | Cite this article

Abstract

An investigation was carried out into a possible reduction of the concentration of airborne microbe-carrying particles (MCPs) measured by sieve samplers when there was an increase in the air volume sampled. Experiments were carried out in a cleanroom where a reasonably constant reduction of microbial growth, which averaged 26%, was found across a wide range of air sampling volumes. In experiments where Staphylococcus aureus bacteria were inoculated onto the nutrient agar surface of a Petri dish, and subjected to sampling across a range of air sampling volumes, there was also a reasonably constant reduction that averaged 35%. Reasons for the constant reduction in microbial growth are discussed.

Introduction

Air sampling is used to ascertain the concentration of airborne MCPs in a wide variety of locations, such as pharmaceutical and medical device cleanrooms, and hospitals, to ensure they do not exceed acceptable levels. The concentration of airborne MCPs is measured by volumetric air samplers that determine the number of colony forming units (CFU) per m³. This type of sampling is also known as ‘active’ sampling. In this paper we use the term ‘microbe carrying particles (MCPs)’ when discussing particles moving about in the air stream, but ‘colony forming units (CFUs)’ when discussing the count of microbial colonies obtained after incubation.

Several types of volumetric air samplers are available; the most popular type used in industrial cleanrooms and hospitals is the impactor type. This type of sampler will typically sample between 30L/min and 200L/min of air, which is drawn into the sampler through single or multiple inlet orifices, and accelerated to a velocity that is usually greater than 10m/s ¹. The air jet from the orifices flows towards the nutrient agar surface of a Petri dish, which varies in size, but can be 55mm, 90mm or 140mm diameter. The air turns as it approaches the nutrient agar, and MCPs are thrown onto the surface. The Petri dish is then incubated and, after incubation, the number of microbial colonies counted. Knowing the air sampling rate, the concentration of airborne MCPs is ascertained as CFUs/m³. This type of sampler is popular, as only incubation of the Petri dishes is required to obtain the concentration of MCPs, and no further microbiological processing is required, as occurs with other types of samplers.

Different types of impaction samplers are available for sampling airborne MCPs. One type, known as a slit sampler, draws air through a slit, and the air flows towards a revolving nutrient agar plate, where the MCPs are impacted on the surface. Another type is known as a sieve sampler, which has 200 to 400 small circular air intakes. The air from these orifices flows towards a stationary Petri dish containing nutrient agar, where the MCPs are impacted. Also available is a type that uses a revolving fan blade to draw air into the sampler and impart a centrifugal force to airborne MCPs that are impacted onto a nutrient agar surface.

Ideally, the airborne concentration of MCPs measured by an air sampler should be the same as the actual number of MCPs in the air, but volumetric air samplers underestimate the concentration of airborne MCPs ². A major loss occurs if the air velocity that impacts airborne MCPs onto the agar surface is too low, and losses can also occur if the distance of the nozzle to nutrient agar surface is excessive and MCPs fail to deposit ¹. Other deposition losses can occur at the orifice intake, and on internal surfaces within a sampler.

Microbes can be damaged during air sampling when they are impacted onto the nutrient agar surface, and may not grow into colonies ³. It has also been reported that as the volume of the air sample increases, a greater proportion of MCPs will fail to grow ⁴. This is considered to be caused by dehydration of the deposited microbes, as well as dehydration of the nutrient agar that causes it to be less supportive of microbial growth. Field studies have been carried out with an Andersen sampler, which is a sieve sampler that samples 1 cubic foot (28.3L) of air per minute, and it was reported that sampling up to 10 min gave a small reduction in the airborne concentration, but this was statistically insignificant ⁵,⁶. However, field tests have to contend with a variation in the concentration of airborne microbes at the sampling location, and this variation makes it difficult to distinguish between the losses caused by different air sampling volumes and those caused by a natural variation in airborne concentrations. It has been suggested that because of the dehydration problem, air sampling should be restricted to as short a time period as practical ⁴ but other research has shown that the volume of air sampled can be quite large before the concentration is significantly reduced ⁷,⁸.

The effect of the length of sampling time on growth in a 30L/min slit sampler has been investigated ⁹. This was carried out by using a wide variety of microbes that were either inoculated onto the surface of nutrient agar plates, or nebulised and deposited from air. When the sampling time was 75 minutes, and the reduction in water in the nutrient agar was 14%, the reduction in bacterial counts was about 8%, which was considered acceptable. These experiments were carried with a slit sampler, where the nutrient agar plate revolved during sampling. However, in sieve samplers, where the nutrient agar plate is stationary, and air continually impacts onto the nutrient agar directly below the intake orifices, there may be additional dehydration losses, and this possibility was investigated.

It should be noted that it is common to find in research papers that time is considered to be the variable that influences the loss of microbes during volumetric air sampling. We consider that air sampling volume and the velocity of impaction onto the nutrient agar are the relevant variables, and these are considered in this article.

Volumetric air samplers used in the experiments and their calibration

The experiments described in this article were carried out by use of four sieve samplers. These samplers sampled 100L/min of air through 400 intake holes of 0.7mm diameter. The air passed through these orifices at a velocity of 11m/s, and airborne MCPs were deposited by impaction onto the nutrient agar surface of 9cm diameter Petri dishes.

The collection efficiency of volumetric air samplers of airborne MCPs can be obtained from its d50% value, which is the size where 50% of MCPs are impacted onto the nutrient agar surface, and 50% pass through the sampler. Using the calculation method described in EN 17141 ¹⁰, and from the information given in the manufacturer’s manual, the d50% was 1.6µm. Experimental results have reported the d50% in this sampler to be 1.7µm ¹¹.

The Petri dishes used in these samplers were 9cm diameter plates and contained Tryptone Soya Agar (Oxoid Type No. 105458) prepared by adding 40g of medium to 1L of water. The Petri dishes were prepared no longer than 7 days before the experiments, and several plates were incubated to check their sterility. After each experiment, the nutrient agar plates were incubated at 30°C +/- 1°C for 7 days, and the microbial colonies counted. When counts exceeded 20, Feller’s coincidence correction factor ¹² was applied, using the manufacturer instructions for a sampling head that had 400 intake holes of 0.7mm diameter.

To ensure maximum counting accuracy, the air sampling rate of the four sieve samplers was calibrated by an independent test house. There was a small difference between the results obtained and the nominal air sampling rate of 100L/min. To compensate for this variation, the sampling volume was adjusted at each test, so the correct volume was sampled.

Although the batteries in the air samplers were recharged before every experiment, some sampling volumes could be unusually long, and a decrease in battery power might occur. This would be relevant to experiments where the microbial growth from a large sampling volume was compared to the sum of the growth from three sequential samples of one third of the air volume. Using a calibrated vane anemometer type TSI VelociCalc 5725, the maximum air sampling volume was measured, and in all four samplers there was no loss of the air sampling rate at the end of sampling.

Investigation of air sampling volumes on concentrations of airborne MCPs in a cleanroom

Russell et al ¹³ carried out experiments to investigate the effect of long sampling times on the microbial count obtained from settle plates. A single settle plate was used to sample air for 8 hours, and 16 plates were used in a sequential manner to sample air for 30 minutes each. The sum of all of the half hour sequential counts was found to be 78% of the count from the one 8-hour plate; this was considered acceptable. This approach was used in the following experiments to investigate the effect of the air sampling volume on the microbial concentration obtained by a sieve sampler in a cleanroom, and the experiments were designed to minimise the problem of the natural variation of the airborne MCP concentration affecting the result.

Description of cleanroom

Experiments were carried out in a cleanroom with a floor area of 3.8m by 4.0m, and a ceiling height of 2.70m, and therefore a room volume of 41.04m³. The cleanroom was a non-UDAF type with filtered air volume supply of 641m³/h (0.178m³/s) that was supplied through a terminal HEPA filter in the centre of the ceiling. Therefore, the air change rate per hour was 15.6. A single air extract was positioned at low level on a wall, and the pressure differential between the cleanroom and outside the room was 10Pa. This ventilation controlled the room temperature to between 19 ºC and 26ºC, and the relative humidity to between 40% and 70%

Air sampling experiments in the cleanroom

Experiments were carried out by sampling MCPs dispersed by personnel who continually walked about the cleanroom to a specified pattern. Four sieve samplers of the type previously described were used in the experiments. One sampler sampled a total air volume, which could be 0.1, 0.5, 1, 1.5, 3 or 6m³, and the other three samplers sequentially sampling one third of the total air volume. The sum of the counts from the three shorter sequential air samples was then compared to the count from the total air volume.

The four samplers were grouped together, with their distance apart being about 20cm. Air sampling was carried out at three defined locations, which were 1m above the floor, and after each test the samplers were moved to a different location, so as to minimise any influence of the location on the counts. Three tests and three locations were employed during each experiment, when the total air volume was either 0.1, 0.5, 1, or 1.5m³. However, because of the long walking time, the number of tests and locations were reduced to two for the 3m³ total sample size, and to one for the 6m³ total sample size. The temperature and relative humidity conditions during each of the experiments were measured.

The count of MCPs obtained from the samplers after incubation was regulated to between 40 and 133 microbial colonies per plate, irrespective of the total air volumes measured. This was obtained by varying the number of people walking in the cleanroom to between one and three, and having the air supply either on or off.

Calculation of time when air sampling should start

The airborne concentration in a cleanroom will be close to zero when empty but, when personnel enter the room, the concentration will rise to a steady-state condition, where airborne MCPs dispersed by personnel are balanced by their removal through the air exhaust. The air sampling experiments should be carried out when the airborne concentration of MCPs had reached a steady-state condition, and not when they were rising, and a time delay was required to allow the increase in airborne concentration of MCPs, which is exponential, to build up from zero to the steady state condition. This time delay was calculated by use of the following equation ¹⁴.

Where,

C = concentration of contaminants /m³;

D = assumed dispersion rate of airborne MCPs from people (no/s)

Q = air volume supply rate to cleanroom - m³/s

V_D = average deposition velocity of MCPs through air and onto surfaces (m/s) = 4.6 x 10⁻³

A = horizontal surface deposition area of cleanroom, which is assumed to be the floor area (m²);

N_D = equivalent air change rate/s owing to surface deposition (no/s);

N_V = air change rate of the cleanroom (no/s);

V = room volume (m³);

t = elapsed time (s).

Assuming the average size of MCPs is 12µm ¹⁵, Stokes Law can be used calculate the deposition velocity 16, which is 0.0046m/s ¹⁴. The equivalent air change rate caused by surface deposition has been shown to be 6 changes per hour (0.001704/s) ¹⁴. The rate of build-up of airborne MCPs is independent of the dispersion rate from people in the cleanroom, but a nominal dispersion rate of MCPs of 14/s was assumed.

By use of Equation 1, the increase in airborne MCP was determined and shown in Figure 1. From these results, it was ascertained that when the air supply was functioning, it would take about 8 minutes for the microbial concentration to build up from zero to 95% of the steady state concentration. Therefore, during the experiments, the apparatus was set up, walking started, and after a total of 8 minutes had elapsed, air sampling commenced. Some experiments were carried out with the air supply to the cleanroom switched off, and application of Equation 1 showed that the elapsed time to bring the microbial concentration up to the steady state condition was too long to be practical. Therefore, an elapsed time of at least 10 min was used.

Figure 1: Build-up of airborne MCPs in the cleanroom

Results of air sampling experiments in a cleanroom

Experiments were carried out in the cleanroom to compare counts obtained from sampling a total air volume of either 0.1, 0.5, 1, 3 or 6m³, to the sum of the counts from three sequential samples of one third of the air volume sampled over the same time. Shown in Table 1 are the temperature and relative humidity during the experiments, as well as the microbial colonies on the nutrient agar plate that obtained the total air sample count and, from that count, the concentration of airborne MCPs/m³ in the cleanroom.

Table 1: Conditions during the cleanroom experiments

Total air sample (m³)	Persons present (P) -and ventilation	Temperature(ºC)	Relative humidity (%)	Count on total sample agar plate	Airborne concentration/m³
0.1	3P - Off	20.8	55.7	116	1160
0.1	3P - Off	26.3	47.4	76	760
0.1	3P - Off	23.8	46.1	62	620
0.5	2P - Off	19.8	59.2	117	234
0.5	2P - Off	25.0	70.0	133	266
1	1P - On	20.0	57.5	40	40
1	1P - On	25.3	52.0	64	64
1.5	1P – On	19.0	61.7	44	29
3	1P – On	25.2	53.1	61	20
6	1P – On	24.9	43.9	71	12
6	1P – On	24.8	64.5	89	15

Shown in Figure 2 are the reductions in microbial growth obtained from the range of air sampling volumes. It can be seen that there was large variation in the test results. Inspection of these results by the eye suggests that there is no change in the reduction of the microbial count as the air sampling volume increased, but a fairly constant loss of about 26%. This was confirmed by a statistical analysis, which was carried out to obtain a trend line and regression coefficient by the method available in Excel. Shown on Figure 2 is the trend line and its equation, which shows that it had practically no gradient and, therefore, no correlation between the two variables. Also given is the regression coefficient (R2) of 8E-06, which showed there was no relation between the two variables.

Figure 2: Reduction in microbial count in relation to air sampling volume

Effect of air sampling volume on test microbes inoculated onto nutrient agar surfaces

The second set of experiments studied the effect of different air sampling volumes on test microbes inoculated onto the surface of nutrient agar plates.

A growth promotion test is described in both the European and USA pharmacopoeias to show that microbial media will support the growth of microbes. A known number of microbes is added to the surface of nutrient agar plates used for air sampling, and the plates incubated to show if the microbes can grow, or not. If between 50% and 200% of the test microbes grow, then the growth medium is considered to be satisfactory. The growth promotion test is also used to determine whether dehydration of the nutrient agar medium significantly reduces the count of airborne MCPs. However, to avoid some rehydration of the growth medium when the test microbes are added in an aqueous suspension, EN 17141 suggests that the test microbes should be added to the surface before air sampling, and not after; this approach was used in these experiments.

The microbes suggested for the growth promotion test are Staphylococcus aureus, Escherichia coli, Pseudomonas aeruginosa, Bacillus subtilis, and Candida albicans. However, in occupied rooms, such as pharmaceutical cleanrooms and hospital rooms, most of the airborne microbes are derived from the skin of personnel in the room, and the most common types are Gram-positive bacteria from the micrococcaceae family, with the staphylococcus and micrococcus genera making up more than 50% of those present in occupied rooms ¹⁷. In addition, experiments carried out with twelve types of micro-organisms that included Gram-positive and Gram-negative bacteria, demonstrated that Staphylococcus aureus gave typical growth reductions after being subjected to a long time period of sampling sterile air9. Therefore, from the types of microbes suggested for the growth promotion test, the bacterial species selected for these experiments was Staphylococcus aureus.

Effect of air sampling volume

A culture of Staphylococcus aureus (ATCC 6538) was prepared from a −80°C glycerol stock by performing three successive passages of a single colony on tryptic soy agar (TSA) to ensure purity. A single colony from the resulting pure culture was suspended in 5mL of sterile physiological saline (0.9% NaCl) and thoroughly vortexed to obtain a homogeneous bacterial suspension. The concentration of the suspension, which was between 1x10⁷ and 1x 10⁸ CFU/mL, was counted using a Neubauer counting chamber. Serial ten-fold dilutions were then carried out to obtain a working suspension that lay between 1x10³ and 1 x10⁴ CFU/mL. 10µL of this dilution was inoculated onto 9cm TSA plates to achieve a surface concentration of between 10 and 100 CFU per plate, and the inoculum was evenly spread across the nutrient agar surface of the Petri dishes, using a sterile spreader, and allowed to dry. It was observed that after incubation of the Petri dishes the colonies of the Staphylococcus aureus were evenly spread over the surface.

Two sieve samplers were loaded with nutrient agar plates inoculated with Staphylococcus aureus and placed inside an operating microbiological safety cabinet to obtain aseptic conditions. Three different air sample volumes, namely, 0.1, 1 and 3m³ of sterile air were collected each day for five separate days. After sampling, all plates, including inoculated control plates that had been retained to establish the pre-sampling surface count on the nutrient agar plates, were incubated under the previously defined conditions, and the colonies counted. The colony counts were not corrected by Feller's formula, as the microbes had been evenly spread over the nutrient agar.

Water loss from nutrient agar in relation to air sampling volume

During the tests that investigated the change in the surface count of Staphylococcus aureus, the water loss for each test was measured. Nutrient agar plates inoculated with Staphylococcus aureus were weighed before and after the tests to obtain the weight loss of water during sampling. The weight loss was obtained as an average of fifteen tests carried out for each air sampling rate of 0.1m³, 1m³ and 3m³. The average reductions are given in Table 2.

Table 2: Water loss caused by different air sampling volumes

Air volume sampled (m^3)	Water loss (%)
0.1	1.3
1	7.3
3	21.4

Reduction in microbial count on surface of nutrient agar

The reduction of microbial count on the surface of nutrient agar obtained from the three air sampling volumes is shown in Figure 3. Figure 3 shows a considerable variation in the reductions of counts, but there was no indication of a greater reduction in growth of the inoculated bacteria as the air sample volume and water loss increased. The average loss of MCPs for the 0.1, 1, and 3m³ air samples was 36.3, 30.8, 35.9%, respectively, and the overall average for all 45 tests was 35%. The results were subjected to a statistical analysis by a one-way ANOVA test, which showed no significant difference in the loss in bacterial growth between the three air sampling volumes (F (2, 44) = 0.23, p-value = 0.80).

Figure 3: Reduction in growth of microbial surface counts with respect to air sampling volume

Effect of sampling volume on the nutrient agar

Long sampling volumes can cause problems with shrinkage and cracking of the nutrient agar. For example, when Petri dishes with shrunken agar are inverted in the incubator, the agar can detach itself and the microbe growing surface fall onto the lower surfaces of the Petri dishes, and make it difficult to obtain a count of microbial colonies. Shown in Figure 4 is the effect of a 6m³ air sample on the nutrient agar. The airflow was imprinted on the agar surface, and the agar had shrunk well away from the sides of the Petri dish. No potential problems with shrinkage or cracking occurred when the air sampling volume was 3m³ and less.

Figure 4: Effect of 6m³ air sampling volume on the nutrient agar

Discussion and Conclusions

The airborne concentration of MCPs is measured in a wide variety of environments, such as industrial cleanrooms and hospitals, where an impaction type of volumetric microbial sampler is often used. The collection efficiency of these samplers is dependent on ensuring MCPs are efficiently collected on the nutrient agar surface, and losses through deposition on the intakes and other surfaces of the sampler is low 1. However, air samplers can have reduced counts caused by (a) failure of MCPs to grow when damaged by impaction on nutrient agar surfaces, (b) dehydration of MCPs lying on top off, or partially embedded, in nutrient agar, and (c) reduction of growth during incubation on dehydrated nutrient agar. The possibility and magnitude of these three losses was investigated in a sieve sampler as the air sample volume increased, and reported in this paper.

The effect of the air sample volume on the concentration of MCPs obtained by impaction air samplers has been studied, although information is conflicting. Mainelis and Tabayoyong ⁴ have reported a reduction in microbial growth and airborne concentration as air sampling progressed from 2 minutes to 30 minutes. However, Li and Lin ⁷ reported that air sampling could continue for 40 minutes without a significant drop in the microbial count. Kuehne and Decker ⁸ have also suggested that long sampling times can be used in impaction samplers, and reported that the loss in growth occurred when the MCPs were first deposited onto the nutrient agar, and not during the remainder of the sampling. However, Mainelis and Tabayoyong ⁴ reported that ‘both the desiccation of the collection media and the desiccation of already collected organisms play a role in a decreased microorganism recovery with increased sampling time’.

Two sets of experiments were carried out in this study. The first set was carried out in a cleanroom with sieve samplers, where one sampler measured the microbial count obtained from a total sample volume of either 0.1, 0.5, 1, 1.5, 3 or 6m³, and that count was compared to the sum of the counts from three other samplers that sequentially sampled one third of the total air volume during the same time period. It had been expected that these experiments would show a reduction of microbial growth as the difference increased between the air sampling volumes from total and sequential samples. However, as shown by the results given in Figure 2, this was not the case, and there was a fairly constant reduction across the total air volumes sampled of about 26%.

A second set of experiments was carried out by inoculating Staphylococcus aureus onto the surface of nutrient agar in Petri dishes, and these dishes were used to determine the reduction of growth when the sieve sampler sampled air volumes of 0.1, 1m and 3m³ of sterile air. The water loss from the nutrient agar when the air sampling volume was 0.1 m³, 1m³ and 3m³, was measured, and found to be 1.3%, 7.3% and 21.4%, respectively. It was again found that the reduction in growth over the different sampling volumes was fairly constant, at about 35%.

It had been anticipated that greater air sampling volumes would give greater losses in growth of the MCP and reduction in the concentration of airborne MCPs. However, this did not occur, and the loss of microbial growth was constant over different air sampling volumes. This constant loss could be explained if the loss occurred when MCPs are first deposited onto the nutrient agar, and not during the remainder of the sampling.

Investigations have been carried out into the effect of sampling time on microbes inoculated onto the nutrient agar surface of settle plates and exposed to airflow in a UDAF cabinet for 6.5 hours⁹ and 4 hours ¹⁸,¹⁹. It was reported in these experiments that the loss in water from the settle plates ranged from 11 to 16%, and the loss in microbial growth ranged from 8 to 12%. This water loss was similar to the loss in the volumetric sampler experiments report in this paper, but the loss of microbial growth in the volumetric samplers was much greater at 25% and 35%. This observation suggests that although dehydration could still cause losses during volumetric sampling, there is likely to be additional loss mechanisms.

Impaction of MCPs onto or into the nutrient agar has been reported to cause losses during volumetric sampling ³. As this will occur when the MCPs are first sampled, and there is likely to be protection against dehydration if the MCPs are submerged or partly embedded in the nutrient agar by the air jet, this could explain the reason for the constant loss in microbial growth in volumetric sieve samplers, irrespective of the air volume sampled.

The reasons for the results reported in this paper are discussed in the previous two paragraphs, which suggest that the constant loss of microbial growth in a sieve sampler could be largely caused by impaction damage when initially sampled, but the surrounding nutrient agar could protect them from further damage by dehydration. However, this possibility appears to be contrary to published research discussed in this article, and requires further research to obtain a clearer understanding as to why there is a constant loss of microbial growth during different sampling volumes in sieve air samplers.

It would appear from the results gathered in the experiments that there is no significant change in the airborne microbial concentration as the air volume sampled increases to 6m³, which suggests that large volumes of air could be sampled by the sieve sampler investigated. However, a shrinkage of the nutrient agar was observed at high air sampling volumes, which could lead to difficulties in obtaining a microbial count after incubation. Considering the values obtained in our experiment, 3m³ appears to be a safe maximum sampling volume for a sieve sampler of the type investigated, although under different environmental conditions, particularly at high temperatures and low relative humidities, the effect of water loss and shrinkage of nutrient agar could be more pronounced, and a lower air sampling volume selected.

References

1. Whyte W, Green G, Albisu A. Collection efficiency and design of microbial air samplers. Journal of Aerosol Science 2007;38(1):101-114.

2. Ljungqvist B, Reinmuller B. Active sampling of airborne viable particles in controlled environments: a comparative study of common instruments. European Journal of Parenteral and Pharmaceutical Science 1998;3(3):59-62.

3. Stewart SL, Grinshpun SA, Willeke K, Terzieva S, Ulevicius V, Donnelly J. Effect of impact stress on microbial recovery on an agar surface. Applied and Environmental Microbiology 1995;61(4):1232-9.

4. Mainelis G Tabayoyong M. The effect of sampling time on the overall performance of portable microbial impactors. Aerosol Science and Technology 2010; 44(1):75-82.

5. Folmsbee M, Strevett K, Stafford K, Evenson C. The effect of sampling time on the total efficiency of the Andersen microbial sampler: a field study. Journal of Aerosol Science;31(2):263-271.

6. Godish DR, Godish TJ. Relationship between sampling duration and concentration of culturable airborne mould and bacteria on selected media. Journal of Applied Microbiology 2007. 102(6):1479-1484.

7. Li CS, Lin YC. Sampling performance of impactors for bacterial bioaerosols. Aerosol Science and Technology 1999;30(3):280–287.

8. Kuehe RW Decker HM. Studies on continuous sampling of Serratia marcescens using a slit sampler. Applied Microbiology 1957; 5: 321-3.

9. Whyte W, Niven L. Airborne bacteria sampling: the effect of dehydration and sampling time. Journal Parenteral Science Technology 1986;40(5):182-8.

10. EN 17141. Cleanrooms and associated controlled environments - Biocontamination control. European Committee for Standardization (CEN), Brussels, Belgium.

11. Yao M, Mainelis G. Investigation of cut-off sizes and collection efficiencies of portable microbial samplers. Aerosol Science and Technology 2006;40(8):595–606.

12. Feller W. An introduction to the probability theory and its application. Volume 1 Third edition 1991, John Wiley and Sons, New York.

13. Russell MP, Goldsmith JA, Phillips I. Some factors affecting the efficiency of settle plates. Journal Hospital Infection 1984; 5(2): 189-99.

14. Whyte W, Whyte WM, Eaton T. The application of the ventilation equations to cleanrooms, Part 1: the equations. Clean Air and Containment Review 2012;12:4-8. Available at: https://eprints.gla.ac.uk/84186/1/84186.pdf

15. Whyte W, Hejab M. Particle and microbial airborne dispersion from people. European Journal of Parenteral Science and Technology 2007;12(2):39-46. Available at https://eprints.gla.ac.uk/84357/1/84357.pdf

16. Hinds WC. Aerosol technology; properties, behaviour, and measurement of airborne particles -Second edition 1998. Wiley-Interface, ISBN 0-471-19410-7.

17. Sandle TA. Review of cleanroom microflora: types, trends and patterns. PDA Journal Pharmaceutical Science Technology 2011;65(4):392-303.

18. Sandle TA. Settle plate exposure under unidirectional airflow and the effect of weight loss upon microbial growth. European Journal of Parenteral and Pharmaceutical Science 2015;20(2):45-50.

T Eaton, K Capper, A Jones, C Barnett, J Bright. Recovery of naturally occurring human borne microbial contamination from settle plates exposed in a unidirectional airflow workstation for 4 hours. European Journal of Parenteral and Pharmaceutical Science 2023.