Physical and Biological Efficiency Testing of ImpactAir® Microbiological Air Sampler Using Techniques Described in ISO14698-1

Page 2 of 28

The work presented herein was performed by the Biosafety Investigation Unit of Public Health England (PHE). This document has been prepared using the original report (Report No. 15/020) provided to Pinpoint Scientific and is not to be taken as an endorsement or recommendation by PHE.

Public Health England

Microbiology Services

Porton Down

Salisbury SP4 0JG

Tel: 01980 612392

http://www.gov.uk/phe

Page 3 of 28

Contents

Contents 3

About Public Health England Biosafety Investigation Unit 4

Executive summary 5

Introduction 6

Materials and Method 8

Samplers 8

Test Micro-organisms 12

Microbiological Assays 12

Test Environments 13

Test Procedures 15

Results 18

TABLE 1: The experimentally determined sizes of particles containing

Bacillus atrophaeus spores generated by the STAG from aqueous solutions of

KI in 80% ethyl alcohol and the associated physical efficiency of ImpactAir® for

each particle size 18

TABLE 2: Collection of airborne Bacillus atrophaeus spores in the PHE

Environmental Room 20

TABLE 2 (Continued) 21

TABLE 2 (Continued) 22

TABLE 2 (Continued) 23

TABLE 2 (Continued) 24

TABLE 3: Mean Mass Diameter determined by collection with the Cascade

impactor 25

TABLE 4: Biological efficiency of ImpactAir® compared with the Casella slit

sampler. 26

Discussion 27

References 28

Page 4 of 28

About Public Health England Biosafety

Investigation Unit

The Public Health England Biosafety Investigation Unit at Porton Down has been

carrying out independent evaluations of infection control interventions in laboratories,

health care, containment, workplace and domestic settings for over twenty years. Their

expertise is in air and water microbiology applied to nosocomial, pharmaceutical and

containment situations. They have developed and offer standard techniques for the

determination of the efficacy of filters and air disinfection units, the performance of

safety cabinets, sealed centrifuges, rotors and air samplers. They are also able to

assess liquid and gaseous disinfectants and the microbial air quality of healthcare

facilities, workplaces and other environments.

The Biosafety Investigation Unit provides specialist bespoke research, testing and

evaluation services for commercial customers that delivers independent analysis and

reports. However as a public sector body they are not able to endorse any particular

products or recommend them for use by the NHS or others.

Page 5 of 28

Executive summary

The physical efficiency of the ImpactAir® air sampler for collecting small bacteria-laden

particles of various sizes has been compared with membrane filter samplers using the

techniques described in ISO14698-1. The samplers were operated simultaneously in a

controlled room where they were challenged with airborne bacteria. Uniform sized

particles of different diameters containing bacterial spores were generated into the

room. The results showed that the ImpactAir® sampler was more effective than the

reference samplers for smaller particles up to 6.5 micron, and as effective as the

reference samplers for the largest particles of 10.5 micron. When compared with using

a paired t test, it was found that for the range of 0.8 to 2.8 micron the impact air was

significantly more efficient than the test sampler.

The biological efficiency of the ImpactAir® sampler was compared to that of the Casella

slit sampler, a commonly used reference sampler. The biological efficiency was

measured as the comparative efficiency of collection of Staph epidermidis, a common

human-associated clean room contaminant, and the extremely aerostable B.

atrophaeus spore. The biological efficiency measured was found to be 125% that of the

Casella sampler and when compared using a paired t test there was no significant

difference, indicating the sampler is effective at sampling bacteria-laden particles,

without undue loss of viability.

Page 6 of 28

Introduction

Determination of the microbiological quality of air is vital in a number of sites such as

areas where pharmaceuticals and medical devices are manufactured, operating

theatres and other critical areas in hospitals and food processing facilities. In many

cases only sparse concentrations of airborne micro-organisms are present in these

locations and this means that large volumes of air (1m3) have to be sampled to collect

sufficient numbers of micro-organisms for a proper quantitative assessment to be made.

Accurate measurement of microbial contamination of air is also dependent on obtaining

a representative sample from the air and limiting any losses that may occur between

the sampler and the assay system. Losses can occur, either due to a failure of the

sampler to capture particles containing micro-organisms (physical loss) or due to

inactivation of viable micro-organisms during collection so that formation of visible

colonies on agar surfaces will not occur (biological loss).

The ImpactAir® sampler is a slit impactor type of instrument based on the principle

described by Bourdillon(1) in which air is aspirated through a single slit above a rotating

plate. The resulting air stream containing microbial particles is directed onto the agar

surface of the plate, as it rotates. The system employs an integrated sensor to maintain

the critical slit to agar distance and the constant rotation of the plate ensures that once

particles have been collected, the surface is moved away from the desiccating effects

of the airstream, helping maintain viability. When the pre-set sampling cycle is

completed the plates are removed and incubated. Viable organisms which form visible

colonies are then counted.

Efficient removal of particles containing micro-organisms from the air and their collection

onto medium for identification often depends on the sizes of the particles. At present no

sampling system or device has been considered a suitable reference method to which

other samplers can be compared. ISO 14698-1(2) recommends using a membrane filter

sampler as the standard method. In filtration systems, accurately measured volumes of

air are drawn through filter material of low pore size so that all particles containing

micro-organisms are deposited by impaction and interception. Provided that the

micro-organisms are resistant to desiccation by drawing air through the filter material,

Page 7 of 28

this simple method can be used as a standard method by which the physical sampling

efficiencies of other samplers can be determined. Bacterial spores of Bacillus

atrophaeus NCTC 10073 are selected as the challenge micro-organisms because they

are known to remain viable in aerosols and are not easily inactivated by desiccation

during collection on membrane filters. They also form very characteristic orange

colonies which are easily identified after overnight incubation. The number of bacteria

collected on membrane filters can be counted by simply placing the filters on an agar

surface and incubating. The physical efficiency of the ImpactAir® sampler to collect

airborne particles of various sizes can therefore be determined by comparison with the

membrane filtration samplers operating side-by-side. This has been carried out in a

controlled environmental chamber by generating the bacterial spores in particles of

uniform size. The sizes of the particles containing the bacterial spores generated are

determined using a cascade impactor (3) which fractionates the particles of different

sizes during collection.

The biological efficiency of a sampler is a measure of how effectively it can collect micro-

organisms on an agar plate in such a way as the micro-organism will subsequently form

a colony. The biological efficiency depends on many factors, including the micro-

organism used, how it is grown and aerosolised, what it is aerosolised from and the

stresses endured during the sampling process. ISO 14698-1 suggests that the common

human derived environment contaminant Staphylococcus epidermidis is used as an

indicator of biological efficiency. In this study the biological efficiency of the ImpactAir®

sampler is compared to that of the reference Casella slit sampler. To measure the

biological efficiency, the ratio of recovery of Staph. epidermidis to the aerostable B.

atrophaeus was determined for the ImpactAir® and Casella sampler.

Page 8 of 28

Materials and Method

Samplers

ImpactAir®

ImpactAir® air samplers (figure 1) were provided by Pinpoint ScientificTM (Serial

Numbers Proto 1, Proto 2 and Proto 3). The samplers were operated as described in

the Manual provided, operating at 30 litres min-1. The flow rate was confirmed prior to

testing using calibrated TSI 4040 mass flow meter (Serial No. 40401104030, UR 36401,

Cal Date. 28-04-2015).

Sampler Serial No. Proto 1 was used for undertaking the physical efficiency tests and

samplers Serial No. Proto 2 and Proto 3 were used for the Biological efficacy tests.

Figure 1. ImpactAir® sampler

Page 9 of 28

For the physical tests, total counts were recorded after 24 hours incubation. However,

for the biological efficiency tests described below, a two stage incubation process was

used, with the plates being initially examined after a reduced incubation time, allowing

for the identification of the separate species of colonies, prior to any possible overgrowth

by the quicker growing B. atrophaeus. The plate were then re-incubated to allow any

further growth and re-counted after a day’s incubation.

Membrane Filter Samplers

These consist of aluminium membrane filter holders (figure 2) each incorporating an

80mm diameter, 3µm pore size Sartorius gelatine filter (Sartorius Stedim Biotech, part

no. 12602--80----ALK). The filters were mounted on a sterile back-up support and

connected to a vacuum pump to provide a measured flow rate of ca. 30 litres min-1. The

actual flow rates were determined and recorded before each set of tests using a

calibrated TSI 4040 mass flow meter (Serial No. 40401104030, UR 36401, Cal Date.

28-04-2015). After sampling, the membrane filters were carefully removed and placed

on an agar growth medium with the exposed side facing upwards and incubated at 37

(±2) °C for at least 18 hours.

Figure 2. Membrane filter sampler

Page 10 of 28

Cascade Sampler

The four-stage Cascade sampler(3) (figure 3) was used to determine the mass mean

diameters of the aerosols generated. The sampler was fitted with four microscope slides

which were ground down to a width of 25mm in order to fit the impactor. The slides were

decontaminated in a 10% sodium hypochlorite solution, washed with distilled water,

dried in an oven and then autoclaved at 131oC for 11 minutes. The slides were covered

with a gel made up of a mixture of 5g Bovine Skin Gelatin (Type B Sigma) and 10ml of

glycerol made up to 100ml with distilled water. This solution was heated to 120°C twice

in an oven before it was applied thinly to the microscope slides. After sampling, the gel

was dissolved in 5ml of warm sterile distilled water, diluted and spread on Tryptone

Soya Agar plates (Biomerieux). The sampling rate was controlled by placing a 17.5 litres

min-1 critical orifice between the sampler and the vacuum source, the flow being

confirmed using a TSI 4040 mass flow meter (Serial No. 40401104030, UR 36401, Cal

Date. 28-04-2015).

Figure 3. Cascade Sampler

The recoveries from each stage of the sampler were calculated using the assay

method above and from this the cumulative percentage loading on each stage was

calculated. This data was then plotted on a log-log scale against the calculated d-50

cut-offs for each stage (the particle size at which 50% of the particles are collected).

A linear regression was drawn to best fit the plot using SigmaPlot 12.0 and from this

Page 11 of 28

the Mass Mean Diameter (MMD) can then be read as the particle size corresponding

to the 50% point of loading.

Casella slit sampler

A calibrated low volume Casella slit sampler (Casella, London) was operated at 30 litres

min-1 for two minutes during the biological efficiency testing. TSA plates (Becton

Dickinson) were used in the sampler for all of these tests. The flow was calibrated using

a TSI 4040 mass flow meter (Serial No. 40401104030, UR 36401, Cal Date. 28-04-

2015).

Figure 4. Casella sampler

Page 12 of 28

Test Micro-organisms

Bacillus atrophaeus

Suspensions of washed B. atrophaeus spores (NCTC 10073) in distilled water were

prepared(4)

. This suspension was the source of the B. atrophaeus used in the physical

and biological efficiency testing. Suspensions of spores (ca. 1 x 105 colony forming units

(cfu) per ml) in 0%, 0.007%, 0.07%, 0.7% and 7% of potassium iodide (KI) in 80%

aqueous ethanol were prepared for use in the physical efficiency testing.

Staphylococcus epidermidis

Staph. epidermidis (NCTC 11047) was grown up in liquid TSB-F (Biomerieux) at 37oC

(±2) in a static incubator for 24 (± 1) hours. All assays with Staphylococcus epidermidis

were carried out on Tryptone Soya Agar plates.

Microbiological Assays

Media

Media Item Batch Number Manufacturer Expiry Date

TSA – 150mm for test samplers

5110311

5278225

5205054

Becton Dickinson

09-10-15

29-03-16

22-01-16

TSA – 90mm for reference samplers

5110362

5278222 Becton Dickinson

28-10-15

11-04-16

Gelatine Filter

0615 12602 150057

0915 12602 150109

0915 12602 150122

Sartorius

06-2020

09-2020

09-2020

Nutrient Broth 431025 Biomerieux 06-05-16

Sterile distilled water 3011793 Versol 08-2017

Microbiological assays

All plates used in the samplers were incubated at 37(±2)°C for the required length of

time before counting the colonies.

Page 13 of 28

Test Environments

Environmental room

The environmental room has a volume of 20m3 with an optional flow of a horizontal

clean air supplied through a bank of HEPA filters (figure 5). Microbial aerosols were

generated in still air in the chamber as described below. At the end of the sampling

period, the room was vented by supplying a horizontal flow of clean air for 10 minutes.

After re-setting of the samplers another microbial cloud containing bacterial spores was

generated and the experiment was repeated. The ability to flush the room with clean air

to remove airborne organisms allows a large number of bacterial aerosol challenges to

be carried out over relatively short periods.

Figure 5. Test chamber

A spinning top aerosol generator (STAG, figure 6) was used to produce an aerosol of

controlled particle size containing bacterial spores (dependent on rotational speed of

the spinning disc)(5). After formation, the particles reduced by evaporation to a size

related to the solid content (KI) of the suspension aerosolized. The STAG Mark 2 (Bristol

Industrial and Research Associates Ltd, Portishead, Bristol, UK) was used to generate

particles of different sizes containing viable B. atrophaeus spores. This was done by

preparing suspensions (ca. 1 x 105 cfu per ml) of the spores in 0 - 7% (w/v) solutions of

potassium iodide (KI) in 80% aqueous ethanol. The suspensions were injected into the

STAG, operating at ca. 48,000 rpm, using a peristaltic pump. The size of the particles

produced depended on the concentration of the KI in the suspension. The liquid

associated with the generated droplets evaporates completely in less than 1 second at

Page 14 of 28

the test relative humidity (ca. 40 - 60%). The concentration of spores in the suspension

was low enough to ensure that the vast majority of particles were mono-dispersed.

Figure 6. Spinning top aerosol generator (STAG)

Class III Cabinet

The biological efficiency testing carried out with Staph. epidermidis was undertaken in

a Class III microbiological safety cabinet (internal volume 0.865m3, figure 7) to allow for

greater control of the mixing and concentration of the test aerosol. Modified ports were

used to allow the supply of compressed air to the nebuliser. The cabinet generates 6 air

changes a minute when the fan unit is operated and is supplied with HEPA filtered air.

This allowed the aerosol generated to be rapidly removed after each test.

Figure 7. Class 3 safety cabinet

Page 15 of 28

A three jet Collison nebuliser(6) (figure 8) operating at a pressure of 26 pounds per

square inch (psi) was used to generate the mixed microbial aerosol for the biological

efficiency testing. The spray suspensions were made up using a stock 105 sterile

aqueous dilution of the B. atrophaeus spore suspension and suitably diluted recent

liquid culture of Staph epidermidis.

Figure 8. Three jet Collison nebuliser

Test Procedures

Physical Efficiency

The aerosols were generated from the various spore suspensions by the STAG Mark 2

as described above. The samplers were arranged in a semi-circle at a distance of 1

metre from the STAG in the environmental room. The STAG was 15 cm off the floor,

placed above a small fan unit. Above the STAG a larger secondary fan was also

operated to ensure effective distribution of the aerosol throughout the test chamber. The

heads of the samplers were 0.8 metres from the floor.

The nebulisation, sampling regimes and the fan speed were chosen after a number of

preliminary experiments were carried out. Conditions were chosen so that a reasonable

Page 16 of 28

number of colony forming units (40-200 per plate) would be produced following

incubation. The nebulisation was started remotely five seconds before the sampling

process was started. The nebulisation and sampling times were two minutes.

The plate counts obtained respectively with the tested sampler and membrane filter

samplers are standardized in colony forming units per cubic meter in order to calculate

the physical efficiency.

When the Cascade impactor was used, the flow rate was controlled at 17.5 litres min-1

by the insertion of a critical orifice in the line between the sampler and the vacuum

source.

Biological Efficiency

The ImpactAir® sampler, a Casella slit sampler and the Collison nebuliser were placed

in the cabinet. A mixed suspension of B. atrophaeus and Staphylococcus epidermidis

was made up immediately prior to the experiment and 30 ml of this suspension was

placed into the Collison nebuliser. The samplers were loaded with the agar plates and

the cabinet fan unit was switched off. A small fan was used to mix the air within the

cabinet while the cabinet ventilation system was switched off. The Collison nebuliser

was operated at 26 psi for two minutes, starting and finishing 20 seconds before and

after the samplers, respectively. The samplers were operated for a total of two minutes

after this the cabinet was vented.

During initial studies the plates were removed from the samplers and stored at 4(±2)°C

until the end of the working day when they were moved to a 37(±2)°C incubator and

incubated for no more than 17 hours before the orange colonies of B. atrophaeus (BA)

and white colonies of Staph. epidermidis (SE) were counted separately. These studies

showed the organisms to be slow growing. Therefore for all test runs the plates were

removed from the samplers and placed in a 37(±2)°C incubator and incubated

overnight before the orange colonies of B. atrophaeus (BA) and white colonies of

Staph. epidermidis (SE) were counted separately. The plates were further incubated

for between 5 and 6 hours to allow for any further colonies to develop.

Page 17 of 28

Due to the inherent variability encountered with the Staph. epidermidis, it is normal to

undertake runs over a number of days to ensure that effective counts can be achieved.

Each day’s runs are then combined to give the overall results.

The comparative biological efficiency of the ImpactAir® sampler for sampling Staph.

epidermidis (SE) was calculated as follows: -

𝐵𝑖𝑜𝑙𝑜𝑔𝑖𝑐𝑎𝑙 𝐸𝑓𝑓𝑖𝑐𝑖𝑒𝑛𝑐𝑦

= 𝑅𝑎𝑡𝑖𝑜 𝑜𝑓 𝑆𝐸 𝐵𝐴⁄ 𝑠𝑎𝑚𝑝𝑙𝑒𝑑 𝑏𝑦 𝑡ℎ𝑒 𝐼𝑚𝑝𝑎𝑐𝑡𝐴𝑖𝑟® 𝑆𝑎𝑚𝑝𝑙𝑒𝑟

𝑅𝑎𝑡𝑖𝑜 𝑜𝑓 𝑆𝐸 𝐵𝐴⁄ 𝑠𝑎𝑚𝑝𝑙𝑒𝑑 𝑏𝑦 𝑡ℎ𝑒 𝐶𝑎𝑠𝑒𝑙𝑙𝑎 𝑆𝑎𝑚𝑝𝑙𝑒𝑟 𝑥 100

Page 18 of 28

Results

The results of the physical tests are summarised in Table 1 and Figure 9 and shown in

more detail in Table 2a-2e.

Mean mass diameter determinations are shown in Table 3.

The biological results are shown in Table 4.

TABLE 1: The experimentally determined sizes of particles containing Bacillus

atrophaeus spores generated by the STAG from aqueous solutions of KI in 80% ethyl

alcohol and the associated physical efficiency of ImpactAir® for each particle size

Percentage KI in suspension

Mass mean diameter determined experimentally by Cascade impactor

(microns)

% Efficiency of ImpactAir®

0.0 <1 (theoretical 0.8 micron)* 130.7

0.007 1.4 123.7

0.07 2.8 120.8

0.7 6.5 112.2

7 10.7 99.8

*As the washed spores have been aerosolised from 80% alcohol w/c distilled water, aerodynamic particle size

can be assumed to be that of the naked spore - ca. 0.8 micron.

Page 19 of 28

Figure 9. Physical Efficiency of the ImpactAir® Sampler for a Range of Particle Sizes

Particle size (Micron)

0 2 4 6 8 10 12

% E

ffic

iency

0

20

40

60

80

100

120

140

160

Page 20 of 28

TABLE 2: Collection of airborne Bacillus atrophaeus spores in the PHE

Environmental Room

(a) Bacillus atrophaeus spores aerosolized from 80% aqueous ethanol

Test NO

cfu per m3 collected by % efficiency to filter average

ImpactAir®

ImpactAir® Filter 1 Filter 2

1 3532.42 2524.75 2838.50 131.73

2 3276.45 2409.24 2724.31 127.65

3 2986.35 2161.72 2088.09 140.54

4 3464.16 2541.25 2463.30 138.44

5 2986.35 3019.80 2707.99 104.28

6 3361.77 2359.74 2283.85 144.79

7 3430.03 2079.21 2104.40 163.97

8 2883.96 2854.79 2300.16 111.89

9 3242.32 2722.77 2626.43 121.23

10 3088.74 2574.26 2316.48 126.31

11 2918.09 2557.76 2381.73 118.15

12 2935.15 2161.72 2446.98 127.37

13 3412.97 2178.22 2610.11 142.55

Average 3193.75 2472.71 2453.26 130.69

Standard Deviation

235.41 286.44 236.73 15.66

Page 21 of 28

TABLE 2 (Continued)

(b) Bacillus atrophaeus spores aerosolized from 0.007% KI in 80% aqueous

ethanol

Test NO

cfu per m3 collected by % efficiency to filter average

ImpactAir® ImpactAir® Filter 1 Filter 2

1 2721.09 1960.46 1733.55 147.32

2 2687.07 2191.10 2487.96 114.86

3 2551.02 2273.48 1765.65 126.32

4 2295.92 1927.51 2263.24 109.57

5 2346.94 1630.97 1990.37 129.62

6 2465.99 1960.46 2134.83 120.43

7 2465.99 2059.31 2327.45 112.43

8 2346.94 2289.95 1829.86 113.93

9 2942.18 1812.19 2231.14 145.53

10 2329.93 2125.21 1845.91 117.34

11 2482.99 1960.46 2118.78 121.74

12 2448.98 2042.83 1861.96 125.43

Average 2507.09 2019.49 2049.22 123.71

Standard Deviation

191.65 189.08 246.62 12.19

Page 22 of 28

TABLE 2 (Continued)

(c) Bacillus atrophaeus spores aerosolized from 0.07% KI in 80% aqueous

ethanol

Test NO

cfu per m3 collected by % efficiency to filter average

ImpactAir® ImpactAir® Filter 1 Filter 2

1 3282.31 3139.53 2934.43 108.08

2 2840.14 2541.53 2639.34 109.64

3 1955.78 1960.13 2213.11 93.73

4 2585.03 2076.41 1918.03 129.43

5 1683.67 1644.52 1655.74 102.03

6 2074.83 1810.63 1901.64 111.78

7 2057.82 1893.69 2704.92 89.50

8 1921.77 1710.96 1540.98 118.19

9 1836.73 2076.41 1540.98 101.55

10 2074.83 1295.68 1819.67 133.20

11 2380.95 1810.63 1229.51 156.63

12 2465.99 1129.57 1311.48 202.04

13 2942.18 2657.81 2491.80 114.27

Average 2315.54 1980.58 1992.43 120.78

Standard Deviation

482.82 546.16 556.92 30.23

Page 23 of 28

TABLE 2 (Continued)

(d) Bacillus atrophaeus spores aerosolized from 0.7% KI in 80% aqueous

ethanol

Test NO

cfu per m3 collected by % efficiency to filter average

ImpactAir® ImpactAir® Filter 1 Filter 2

1 1791.81 2474.92 2524.75 71.68

2 1040.96 2391.30 2541.25 42.21

3 1313.99 2207.36 2491.75 55.93

4 1996.59 2274.25 2194.72 89.35

5 1672.35 1989.97 2244.22 78.99

6 3037.54 2140.47 1930.69 149.22

7 2406.14 2324.41 1815.18 116.25

8 2064.85 1672.24 1831.68 117.86

9 2491.47 1822.74 1831.68 136.35

10 2474.40 2525.08 1650.17 118.53

11 2679.18 1722.41 1996.70 144.08

12 2559.73 2157.19 1650.17 134.46

13 1518.77 1103.68 759.08 163.07

14 1058.02 936.45 957.10 111.75

15 1006.83 719.06 940.59 121.33

16 1075.09 668.90 874.59 139.31

17 904.44 702.34 726.07 126.64

18 955.63 903.01 808.58 111.67

19 1092.15 785.95 973.60 124.14

20 716.72 852.84 709.57 91.75

Average 1692.83 1618.73 1572.61 112.23

Standard Deviation

720.5 697.9 663.8 31.7

Page 24 of 28

TABLE 2 (Continued)

(e) Bacillus atrophaeus spores aerosolized from 7% KI in 80% aqueous

ethanol

Test NO

cfu per m3 collected by % efficiency to filter average

ImpactAir® ImpactAir® Filter 1 Filter 2

1 2013.65 2176.87 2399.33 88.01

2 1928.33 1496.60 1862.42 114.82

3 1160.41 1870.75 1543.62 67.97

4 1996.59 1751.70 1493.29 123.06

5 1877.13 1819.73 1694.63 106.83

6 1467.58 1870.75 1728.19 81.56

7 2423.21 1683.67 1610.74 147.11

8 1535.84 2142.86 1442.95 85.66

9 1689.42 1938.78 1728.19 92.14

10 1518.77 1564.63 1778.52 90.86

Average 1761.09 1831.63 1728.19 99.80

Standard Deviation

359.11 221.75 270.08 23.28

Page 25 of 28

TABLE 3: Mean Mass Diameter determined by collection with the Cascade

impactor

Percentage KI in suspension Stage Cumulative % of total

0.007

1 100

2 99.43

3 88.83

4 51.58

Mass mean diameter (MMD) = 1.4

0.07

1 100

2 99.03

3 84.14

4 34.79

Mass mean diameter (MMD) = 2.8

0.7

1 100

2 92.98

3 44.42

4 4.13

Mass mean diameter (MMD) = 6.5

7

1 100

2 61.36

3 9.10

4 0.91

Mass mean diameter (MMD) = 10.7

Page 26 of 28

TABLE 4: Biological efficiency of ImpactAir® compared with the Casella slit

sampler.

TEST

Casella ImpactAir® Total Count

% Efficiency cfu cfu ratio cfu cfu ratio

Ba Se Se/Ba Ba Se Se/Ba

1 73 30 0.41 160 75 0.47 114.06

2 50 45 0.90 96 117 1.22 135.42

3 36 39 1.08 110 99 0.90 83.08

4 29 36 1.24 46 100 2.17 175.12

5 14 33 2.36 45 102 2.27 96.16

6 124 65 0.52 354 147 0.42 79.22

7 98 48 0.49 278 124 0.45 91.07

8 74 45 0.61 232 128 0.55 90.73

9 70 55 0.79 223 139 0.62 79.33

10 125 47 0.38 293 103 0.35 93.49

11 93 51 0.55 188 123 0.65 119.31

12 88 38 0.43 195 81 0.42 96.19

13 67 31 0.46 117 96 0.82 177.34

14 76 34 0.45 156 115 0.74 164.78

15 71 38 0.54 137 83 0.61 113.20

16 66 39 0.59 82 111 1.35 229.08

17 80 34 0.43 120 73 0.61 143.14

18 56 31 0.55 84 86 1.02 184.95

19 43 37 0.86 71 72 1.01 117.85

Average 125.45

Standard Deviation 42.79

Page 27 of 28

Discussion

The results showed that the ImpactAir® sampler was more effective than the reference

samplers for smaller particles up to 6.5 micron, and as effective as the reference

samplers for the largest particles of 10.5 micron. When compared with using a paired t

test, it was found that for the range of 0.8 to 2.8 micron the ImpactAir® was significantly

more efficient than the reference samplers, and for the other particle sizes it was as

efficient as the reference samplers. This makes the sampler highly efficient across the

range of particle sizes tested.

The biological efficiency of the ImpactAir® sampler was compared to that of the Casella

slit sampler, a commonly used reference sampler. The biological efficiency was

measured as the comparative efficiency of collection of Staph epidermidis, a common

human-associated clean room contaminant, and the extremely aerostable B.

atrophaeus spore. The biological efficiency measured was found to be 122% that of the

Casella sampler and when compared using a paired t test there was no significant

difference, indicating the sampler is effective at sampling bacteria laden particles,

without undue loss of viability.

The combination of high physical efficiency and good biological recovery reflects the

advantages of rotating plate slit samplers compared to sieve or fixed slit sampler types.

The rotating plate allows for high impaction velocities, whilst minimising the time in which

the captured particles are exposed to the direct airflow. The reference filter samplers do

not use impaction to capture the test aerosol and hence the particle capture is inherently

less stressful. However, the filter membrane is not an ideal media to then maintain the

viability of the captured particles and the greater than 100% efficacy of the ImpactAir®

may reflect the limitations of the reference sampler system.

Page 28 of 28

References

1. BOURDILLON, R.B., LIDWELL, O.M. and THOMAS, J.C. (1941). A slit sampler

for collecting and counting airborne bacteria. Journal of Hygiene, Cambridge 41,

197-226.

2. ISO 14698-1, Cleanrooms and associated controlled environments--

Biocontamination control, Part 1: General principles and methods.

3. MAY, K.R. (1945). The cascade impactor: An instrument for sampling coarse

aerosols. Journal of Scientific Instruments, 22, 187-195.

4 SHARP, R.J., SCAWEN, M.D. and ATKINSON, A. (1989). Fermentation at

downstream processing of Bacillus. In "Bacillus" Edited by Colin R Harwood,

Plenum Publishing Corporation.

5. FOORD, N. and LIDWELL, O.M. (1975). Airborne infection in a fully

air-conditioned hospital. II. Transfer of particles between rooms resulting from the

movement of air from one room to another. Journal of Hygiene, 75, 31-44.

6. MAY, K R (1973). The Collision nebuliser: Description, performance and

application. Aerosol Science, 4, 235-243.