containing bacteria;suchparticles havea diameterof 1m or more.Tests have
beencarriedout in whichsurfaces weresmeared witha solution ofBacillus
subtilisvar.globigiiand then cleanedand the airborne dispersalof the organisms
around a roomwas detectedusing settleplates(Burfootand Brown, 2004b).
Thesetests showed that contaminationis easilyspreadby hosing(Fig. 5.3a), with
contaminationtravelling many metres. In this example, platesdirectlyin front of
the hose becameso wetthat they could not be used to assess microbial
contamination.It is expected that the counts on suchplateswouldhavebeenvery
high.Muchof the contaminationfroma bootscrubber(Fig.5.3b)settledwithin
2 m and fromhand-washing(Fig.5.3c)within1 m of the sink.Contamination
froma floor scrubber witha vacuumwas verylow and detected onlynext to the
scrubber(Fig. 5.3d).In all cases,much of the contaminationfell closeto the
cleaning operation but therewas alwayssomecontamination spreadthroughout
the room, as evidenced by the low countsawayfromthe mainsourcesin Fig. 5.3.
Thisspread results fromthe dispersalof smallparticles.
Measurementsin factories and controlled environment roomsand the use of
computer models haveled to someimportantconclusions:
∑ The risk of product contaminationis greatest whenthe direction of the air
flowis froma source of contamination towards the food.
∑ The smallerthe particlethe greaterthe flighttimeand the distance it may
travel.
∑ Generally, in high-care production areas, less than 1% of the particles
generated will settle.Mostwill be removedby the filtration system or escape
throughdoorwaysand hatches.
∑ The temporal changein concentration of verysmallparticles,around 1m
diameter, is affectedby the air exchange rate,the efficiency of the filtration,
the rate of generationof particles and the leakageof the room.
Fig. 5.2 Particlesizesproducedby a low-pressure(100 psi/0.69MPa)trigger-typehose
(takenfromBurfootet al., 2003b).
Aerosolsas a contamination risk 97