Nv. Mccullough et al., COLLECTION OF 3 BACTERIAL AEROSOLS BY RESPIRATOR AND SURGICAL MASK FILTERS UNDER VARYING CONDITIONS OF FLOW AND RELATIVE-HUMIDITY, The Annals of occupational hygiene, 41(6), 1997, pp. 677-690
A variety of respirator filters and surgical masks were challenged wit
h three aerosolized bacteria: Mycobacterium abscessus (M.a.) (a rod),
Staphylococcus epidermidis (S.e.) (a sphere), and Bacillus subtilis (B
.s.) (a rod). Tests were conducted at two how rates (45 and 85 l./min)
and two humidity levels (30 and 70%). Aerosols were measured with a t
otal-particle, direct-reading, spectrometer and a viable particle casc
ade impactor. Measurements up-and downstream of the filter or mask wer
e used in determining aerosol penetration; respirator or surgical mask
fit was not evaluated. Bioaerosol penetration measured with two aeros
ol sampling instruments was found to correlate. Additionally, bioaeros
ol test parameters were evaluated with respect to their effect on pene
tration. Increasing flow resulted in increased penetration of all orga
nisms while an increase in relative humidity did not exert a consisten
t effect on all organisms. Of the respirators approved by the National
Institute for Occupational Safety and Health (NIOSH), filter efficien
cy was as expected with dust/mist respirators having the lowest and HE
PA filters the highest efficiency. Surgical masks were the least effic
ient of all filters tested; these are not certified by NIOSH. Bioaeros
ol penetration was compared to that of a polystyrene latex sphere (PSL
) aerosol. Penetration of the test aerosols was predicted on the basis
of particle aerodynamic diameter and was expected to be in this order
: PSL>M.a.>S.e.=B.s. The PSL aerosol was the most penetrating, as pred
icted. However, results showed that B.s. was more penetrating than S.e
. The aerodynamic diameter may not be the best parameter for predictin
g aerosol penetration of nonspherical particles in these filters. (C)
1997 British Occupational Hygiene Society. Published by Elsevier Scien
ce Ltd.