Content uploaded by Richard Kerekes
Author content
All content in this area was uploaded by Richard Kerekes on Oct 02, 2017
Content may be subject to copyright.
420 Nordic Pulp and Paper Research Journal Vol 23 no. 4/2008
KEYWORDS: Wood pulp, Fibrillated fibres, Air filters, Nano-
scale particles, Aerosol particles, Freeze-drying, Filtration effi-
ciency
SUMMARY: This paper describes a method for producing air
filters from wood pulp to capture sub-micrometer aerosol parti-
cles. The filters were produced from softwood and hardwood
kraft pulp by wet beating to attain high surface fibrillation and
subsequent partial freeze drying to retain surface fibrillation
into the dry state. Filters produced by this method were evalua-
ted for efficiency as a function of particle size, as well as pres-
sure drop, at the upper and middle range of face velocity for
respirator filters (12 and 6 cm/s). It was found that pressure los-
ses and capture efficiencies of nebulized NaCl particles were
close to those typical of N95 commercial respirator filters. In
addition to these promising filtration characteristics, the wood-
pulp filters offer advantages in being produced from a renewa-
ble resource and in disposability by incineration as a carbon-
neutral fuel.
ADDRESSES OF THE AUTHORS: Jingliang Mao
(jmao@chml.ubc.ca) and Richard J. Kerekes
(kerekes@chml.ubc.ca): University of British Columbia,
Pulp and Paper Centre, Vancouver BC V6T 1Z4, Canada.
John F. Kadla (john.kadla@ubc.ca): University of British
Columbia, Forest Science Center, Vancouver BC V6T 1Z4,
Canada. Biljana Grgic (b.grgic@live.com) and Warren H.
Finlay (warren.finlay@ualberta.ca): University of Alberta,
Department of Mechanical Engineering, Aerosol Research
Laboratory, Edmonton AB, T6G 2G8, Canada.
Corresponding author: RichardJKerekes
1. Introduction
Concern over air quality, in particular airborne bacteria
and viruses, has increased the potential importance of air
filtration for health protection. Filters for this purpose
may take the form of face masks or cabin, household, and
building-air filters. These may have a short life and
therefore should be safely disposable in addition to
possessing high filtering capability.
Current air filters are typically made of fibers
produced from synthetic polymers such as polypropylene
or regenerated cellulose (Balazy et al. 2006a) Paper-
based filters made from wood pulp have been used in
some cases for face masks (Seto et al. 2003), but appea-
red to be ineffective. Nevertheless, wood-pulp based
filters may be attractive as inexpensive filters that are
also easily disposed of by incineration as a carbon-neu-
tral fuel. In the present study we examine the ability of
wood-fibre filters to removeaerosol particles in the most
penetrating size range when these filters are produced
using a production method aimed at yielding much hig-
her filtration quality than previous wood pulp filters.
2. Background
Filtration of respired air is a complex topic that has been
the subject of numerous studies. A recent review has
highlighted some of the key scientific aspects of this
field (Thiessen 2006). The major issue in filtering
pathogens, i.e., bacteria and viruses, is their small size
which ranges from a few nanometers to a few microns.
This size range determines the critical range of sizes in
aerosol filtration. Over this range, filtration efficiency
decreases with decreasing particle size down to a few
hundred nanometers and then increases as sizes become
smaller. This behavior is caused by differences in particle
capturing mechanisms for the different particles sizes,
with inertial impaction and interception operating for
large particles, while Brownian diffusion operates for
small particle sizes. Accordingly, the most penetrating
particle size (MPPS) (Thiessen 2006; Martin, Moyer,
2000) is in the range of 50-500 nm (Hinds 1999) and is
the critical size for gas filtering of airborne pathogens by
filters.
Various tests and standards have been established to
quantify filter effectiveness of health-related filters. A
common one is the rating system of the National Institute
of Occupational Safety and Health (NIOSH) based on the
removal of particles in the 300 nm size range. Another
standard, recommended for respirators by the US Center
for Disease Control and Prevention, is an N95 filter which
requires removal of 95% or more of particles for air flo-
wing at 85 liters per minute (NIOSH- 42 CFR Part 84).
Several methodologies exist to measure efficiency of
filters to meet the above as well as other standards
(Balazy et al. 2006a, 2006b; Qian et al. 1997a, 1997b,
1998; Huang et al. 1998). Some are based on nebulizing
suspended bacteria/viruses using average droplet size
much larger than the actual bacteria/virus sizes, and
measuring the captured mass or culturing and counting
captured material. However, this form of measurement
gives unduly large efficiencies because the large droplets
used in such studies exert a disproportionate influence on
the mass captured,and hence the efficiency (Thiessen
2006; Nicholson 2004; Wilkes 2004). Instead, a more
rigorous way of measuring filtration efficiency is by
measuring the amount and size of particles entering and
leaving the filter. Size distributions and shapes
representative of most viruses can be produced by
nebulizing and drying NaCl solutions into an aerosol.
This procedure is standardized by NIOSH for N95 filters
as NIOSH- 42 CFR Part 84.
Along with good filtration efficiency,asecond
requirement of filters is a small pressure drop. Large
pressure drop makes breathing difficult in respirator
masks, and in building filters, consumes large amounts of
Wood pulp based filters for removal of sub-micrometer
aerosol particles
Jingliang Mao,University of British Columbia, Canada, Biljana Grgic and Warren H. Finlay,University of Alberta, Canada, John F. Kadla and
Richard J. Kerekes, University of British Columbia, Canada
3708 08-11-26 16.20 Sida 420
Nordic Pulp and Paper Research Journal Vol 23 no. 4/2008 421
energy. Generally, for mechanical f iltration, as pressure
drop increases so does filtration efficiency, and therefore
there is a trade-off between filtration efficiency and pres-
sure drop (Hinds 1999).
Asstated earlier, paper filters have been used as face
masks to filter airborne pathogens, specifically to reduce
risk from viral infection (Seto et al. 2003). The paper
masks were considered ineffective because of wetting
issues, though no detail was given. On the other hand,
wood pulp fibers offer some potential advantages in fil-
tration: they are inexpensive, easily disposed of, and pro-
duced from renewable material in their manufacture.
Wood pulp fibers are bio-composites made up of spi-
rally-wound cellulosic fibrils which range in size from a
few microns to tens of nanometers (Donaldson 2007). It
is possible to expose the fibrils by mechanical beating,
that is, to “fibrillate” the f ibers. In doing so, a large sur-
face area is created both within the fibre cell wall and on
its surface. For example, upon mechanical beating, a wet
pulp fibre having a typical specific surface of about 1
m2/g can be fibrillated to up to 250 m2/g (Garner, Kerekes
1978). A good part of this new surface is external
fibrillation which should assist in particle capture.
However, upon drying, the fibrillated structure collapses
and re-bonds (Fernandes Diniz et al. 2004) leaving only
about 0.5 m2/g specific surface (Garner, Kerekes 1978).
Thus, it is apparent that if the wet fibrillated surface can
be retained into the drystate, a verylarge surface would
be available for capturing small particulate matter.
Moreover, the fibrils would be anchored to fibres which
mayact as a backbone to provide structural integrity for
the filter. This would prevent loose fibrils from accumu-
lating in the interstices of the fibrous network and there-
by increasing the pressure drop.
The major challenge in exploiting the above potential
lies in the fact that filters are sold and used in the dry
state, but fibrillation must be produced in the wet state
and is lost upon drying. Therefore, a way must be found
to retain wet-state fibrillation into the dry state if deve-
lopment of a high efficiency, low pressure drop wood
pulp filter is to be achieved. This is the major objective
of the present work.
3. Production of high surface pulp fibers
3.1 Wet beating
We produced high surface pulps by wet beating Northern
bleached softwood (NBSK) and hardwood (NBHK) kraft
pulps. This was accomplished in a Valley Beater (VB) as
well as an Esher-Wyss (EW) laboratory conical refiner.
The most extensive beating was achieved in the Valley
Beater with 480 minutes of operation time. Based on the
nameplate power of the beater, this time corresponds to a
specific energy of approximately20,000 kWh/t. The
specific beating energy for the NBSK and NBHK pulp in
the Esher-Wyss laboratoryconical ref iner was approxi-
mately250, 350, or 500 kWh/t, respectively.
3.2 Pulp drying
Freeze drying was employed to retain wet fibrillation into
the dry state. In this type of drying, water is sublimated
from ice into vapor, thereby avoiding the strong capillary
forces created by liquid menisci which cause horni-
fication (f ibril re-bonding). Freeze drying has been used
for many years in pulp fibre research as a laboratory tool
to preserve the structure of wet fibers into the dry state.
However, the only proposed commercial application for
freeze-dried wood pulp known to the authors is to
produce an absorbent (Chatterjee, Makoui 1984).
In our work, freeze-drying was carried out in a
Labconco FreeZone 4.5 Liter Benchtop Freeze Dry
System. A vacuum of 0.025 to 0.050 mbar was employed
in a drying chamber heated to temperatures of approxi-
mately 30ºC. Drying times ranged from 18 to 48 hours,
depending on the total freeze-drying sample charge, con-
sistency and thickness of the samples. For example, it
takes approximately 24 hours to freeze dry three pieces
of 120 g and 8 mm thick mats with initial consistency of
around 1.5% to above 95% consistency (mass
fibre/(mass f ibre+water)).
The aim of our research was retention of fibrillation on
the surface of fibres rather than within their cell wall. To
reduce drying time and cost, we explored the concept of
“partial freeze drying”, that is, freeze drying onlyin the
initial stage of drying where mostly surface water is
removed. The remaining water was evaporated byair
drying.
3.3 Forming filter pads
We made filter pads from pulp by two methods: dry for-
ming and wet forming. Dry-formed pads were produced
in a laboratory sheet former at the Weyerhaeuser
Technology Centre in Tacoma, Washington. After wet
beating, the pulp was freeze-dried. This lightly bonded
dry pulp was mechanically processed to break fibre
bundles into individual fibers in a Black & Decker
SmartGrind™ Coffee Grinder with a 30s treatment. The
fibers were then formed into pads in a laboratory sheet
former. The resulting pads were pressed for 5s at 2 bar,
with the exception of pads IV, V, and VI (see Table 1),
which were pressed at 2.5 bar.
Wet-formed pads were produced by placing a suspen-
sion of 1.3-2.4% pulp in water into a 150 mm Petri dish
to a depth of 8 mm, except filters B, G2, and G3 (see
Table 1), which had depths of 4, 14 and 16 mm,
respectively. The samples were then frozen at -70ºC and
freeze-dried. No pressing was imposed on the formed
pads. The dry-formed samples used in this study had
basis weights in the range 100-400 g/m2,and the wet-for-
med samples in the range 95-350 g/m2.The basis weight
and production condition for each filter is shown in
Tab l e 1 .They are all single-layer filters, with the excep-
tion of II, IV, and VI, which are three-layer filters with
beaten pulp as the middle layer and commercial fluff
pulp as the twoouter layers.
3708 08-11-26 16.20 Sida 421
422 Nordic Pulp and Paper Research Journal Vol 23 no. 4/2008
3.4 Testing of initial filter efficiency
We measured the initial filtration efficiency of the pulp
pads and their pressure drop in the Aerosol Research
Laboratory of the University of Alberta. The methodolo-
gy involves using solid, neutral sodium chloride particles
having a diameter in the most penetrating particle size
range for the fibrous f ilter media, which is 50-500 nm
(Hinds 1999). Fig 1 is a schematic diagram of the experi-
mental setup. The challenge aerosol was generated from
4% NaCl solution using a 6-jet Collison nebulizer (BGI
Inc., Waltham, MA), which was supplied by a compres-
sed air system. Before entering the nebulizer, the air was
purified by passing through a high-efficiency particulate
air (HEPA) quality filter and drying desiccant. Aerosol
output was passed through an 85Kr source charge neutra-
lizer (model 3054, TSI INC., Minneapolis, MN, USA) to
neutralize particles to Boltzman charge equilibrium.
Generated aerosol was mixed with the clean air to dry the
aerosol, with the clean air derived from the compressed
air system to allow 30 l/min of total flow rate. Pressure
drop was measured simultaneously using a low pressure
manometer (HHP-3, Omega Engineering Inc., CT, USA).
An electrical low pressure impactor (ELPI, DEKATI
Ltd., Finland) was used to determine the concentration
and aerodynamic particle size distribution in real time.
This instrument utilizes cascade impaction combined
with electrical current measurement to yield real time
aerosol sizing. The aerosol particles are charged by a
corona charger, and particles are detected by electrome-
ters inside the cascade impactor. The data were recorded
in 12 ELPI channels from 0.018 to 5.44 µm. We used the
first nine channels for our readings, while the last three
channels (>1.3 µm) were discarded due to low concentra-
tion readings. The ELPI requires 30 l/min of air flowrate,
so that the sizing of filter holder and filter samples was
done to accommodate this inlet flowcondition. The flow
was driven by a vacuum pump (Sogevac SV25, Leybold
S.A., France).
Our experimental protocol consisted of generating
aerosols and allowing the measured output concentration
to become steady, then measuring aerosol upstream of the
filter followed immediately by measurement of aerosol
downstream of the filter, for 60s each. The measured
aerosol concentration upstream (Cin) and downstream
(Cout) of the filter were incorporated into the equation
for the filter efficiency, E:
E= (1- Cin
Cout)×100 [1]
The experimental method was validated by testing
commercially available NIOSH-certified N95 filters
from a major manufacturer. Our efficiency tests agreed
well with published work (Janssen et al. 2003). Validation
was repeated throughout the duration of our subsequent
measurements for paper filters, with measured efficienci-
es of the most penetrating particle sizes always in the
range of 97-99% for NIOSH-certified N95 filters.
Since both filtration efficiency E and pressure drop
across a filter are important, a useful parameter for com-
paring different types of filters is the filter quality factor
qF,defined as:
qF =(-ln P)/ p [2]
Here Pis the fraction of aerosol penetrating a filter,P
=(1-E) and p is the pressure drop in Pa. Alarger qF
denotes better filter quality,that is, a greater capture effi-
Fig 1. Schematic diagram of the experimental setup for the filter testing.
Dry Formed Filter* Basis Weight (g/m2)Beating Energy and Pulp Species
I 200 VB 20, 000 kWh/t, NBSK
II 400 * VB 20, 000 kWh/t, NBSK (100)
0kWh/t, Fluff Pulp, NBSK (300)
III 200 VB 15, 000kWh/t, NBSK
IV 400* VB 15, 000kWh/t, NBSK (100)
0kWh/t, Fluff Pulp, NBSK (300)
V 200 EW 507 kWh/t, NBSK
VI 400* EW 507 kWh/t, NBSK (100)
0kWh/t, Fluff Pulp, NBSK (300)
VII 300 EW 352 kWh/t, NBSK
VIII 300 EW 250 kWh/t, NBSK
IX 100 EW 259 kWh/t, NBHK
X 200 EW 259 kWh/t, NBHK
XI 300 EW 259 kW/t, NBHK
XII 100 EW 501 kWh/t, NBHK
XIII 200 EW 501 kWh/t, NBHK
XIV 300 EW 501 kWh/t, NBHK
Wet Formed Filter* Basis Weight (g/m2)Beating Energy and Pulp Species
B 95 VB 15, 000kWh/t, NBSK
C 165 EW 507 kWh/t, NBSK
F 145 EW 259 kWh/t, NBHK
G165 EW 501 kWh/t, NBHK
G1 175 EW 501 kWh/t, NBHK
G2 295 EW 501 kWh/t, NBHK
G3 350 EW 501 kWh/t, NBHK
*3-layer filters: middle layer beaten pulp, two outer layers commercial fluff pulp; layer basis
weights shown in brackets; all other filters are single-layer
Table 1. Major production variables of the dry formed filters and wet formed filters.
3708 08-11-26 16.20 Sida 422
Nordic Pulp and Paper Research Journal Vol 23 no. 4/2008 423
ciency for smaller pressure drop. For a single
brand of synthetic commercially available
NIOSH certified N95 filters, we measured a
value of qF =0.037 for 12 cm/s face velocity.
Similar qF values have been found by others
for this filter, e.g. Janssen et al.(2003) measu-
red qF =0.045 to 0.055 for a 42.5 l/min flow
rate with an estimated face velocity smaller
than used in our work. This explains the some-
what different quality value measured by us.
The face velocity of the air flowing through
the filter was determined by dividing flow rate
by the exposed filter area. We used face veloci-
ties of 6 and 12 cm/s, which correspond to
NIOSH regulations for certification test of a
respirator and falls at the middle and upper end
of typical filtration face velocities of 4-12 cm/s
(Huang 1998).
Another consideration for filter samples tar-
geting respirator usage is the effect of humidity
on filter efficiency. For this reason, a separate
set of efficiency measurements was done using
preconditioned filter samples that had been
stored at 85% relative humidity and 38ºC for 25
hours according to NIOSH respirator approval
protocols, using a laboratoryhumidity chamber
(LH-1.5, Associated Environmental Systems,
MA).
4. Results and Discussion
4.1 Surface fibrillation
The specific beating energies used in this study for fiber
surface fibrillation by wet beating are large relative to
normal pulp beating. However, they are in the range of
specialty pulps. For example, to produce microfibrils,
Chakraborty et al. (2005) beat pulp in a PFI mill for
125,000 revolutions, which corresponds about 24,000
kWh/t based on the energy values reported by Welch and
Kerekes (1994). Turbak et al. (1983) homogenized spruce
pulp to produce microfibrillated cellulose using a speci-
fic energy consumption of approximately 27,000 kWh/t.
Fig 2 shows SEM images of the external surfaces of
air-dried (left) and freeze-dried (right) pulps. Compared
to air-drying, it is apparent that freeze-drying retains a
much greater amount of surface fibrillation.
In partial freeze drying, we found that it was necessary
to freeze-dry pulp up to a solids content of 85%
consistency(fibre/water mass ration = 5.67) or more in
order to retain surface fibrillation. This decreased the
time for freeze-drying by about 15%.
Although we subjected both hardwood and softwood
chemical pulps to beating, we found that only softwoods
developed extensive surface fibrillation. The softwood
surface is shown in Fig 3.Wepostulate that the softwood
gave better fibrillation because of the longer length of the
fibres, which permits more abrasion during the beating
process.
We found that significantly more surface fibrillation was
retained by wet forming followed by freeze drying as
opposed to dry forming previously freeze-dried pulp.
This is illustrated in Fig 4.We postulate that the necessa-
ry mechanical processing preceding dry forming, and
perhaps the dry forming process itself, sheared off some
of the surface fibrils.
4.2 Filtration results
The measured filtration efficiencies and pressure drops
for dry-formed filters (Sample I-XIV) in basis weight of
100-400 g/m2are shown in Fig 5,and for wet formed fil-
ters (Sample B-G) in basis weight 95-165 g/m2are shown
in Fig 6.Each experimental point in the following Figs 5
and 6represents a replicate of twotests. For equal basis
weight, Figs 5 and 6show that greater beating gives
higher capture efficiency and larger pressure drop, likely
due to the greater surface area and smaller diameter fibrils
produced by greater fibrillation. For pulps beaten at the
same energy, a larger basis weight also gives both increa-
sed capture efficiency and pressure drop. This can also be
attributed to increased surface area of filters, in this case
due to the larger mass (or thicker) sample mat.
From Figs 5 and 6it is not obvious which samples and
methodology give the best efficiency at lowest pressure
drop. This is instead more readilydetermined from the
“filter quality”, qF def ined earlier byEq (2).Values of
Fig 2. The surfaces of air dried (left) and freeze dried (right) beaten pulp fibres.
Fig 3. Surface fibrillation of softwood freeze-dried pulp at two SEM magnifications (X400 left and X3500 right).
Fig 4. Dry-formed (left) and wet-formed (right) pads of pulp ( X300).
3708 08-11-26 16.20 Sida 423
424 Nordic Pulp and Paper Research Journal Vol 23 no. 4/2008
qF for the dryformed and wet formed filters shown in
Table 1 are shown in Fig 7.The measured values of qF
for the wet formed filters B-G fall in the range 0.01<
qF<0.014, and are all larger than those of dry formed
samples I-XIV. This indicates that wet formed samples
have better retention ability than dry formed samples due
to higher fibrillation level on wet formed samples, as
discussed in a previous section.
Filtering efficiency is known to increase with increa-
sing filter thickness. For this reason, the process used to
produce the G samples, which is the highest quality filter
in Fig 7,was used to produce several mats of varying,
larger thickness. Increased efficiency with these thicker
filters is indeed seen in Fig 8.
It should be noted that all of the above tests were carri-
ed out using a face velocity of 12 cm/s, which is on the
upper range of face velocities expected in practice e.g. a
level commensurate with respiration occurring during
heavy work load. We also performed a small set of filtra-
tion measurements at a lower face velocity of 6 cm/s with
the highest quality factor samples (i.e. G samples). This
lower face velocity corresponds to light and moderate
work loads of the wearer, and, as expected (Lee, Mukund
2001), filtering efficiency is higher and pressure drops
are lower, as is seen in Fig 8.
As the efficiency of the wet formed 16 mm thick G3
sample in Fig 8 is high enough to potentially be used as
respirator, weperformed efficiency tests with these
samples after theyhad been exposed to humidity and
temperature preconditioning (85% RH, 38ºC for 25
hours). The efficiencyof these samples, for a 12 cm/s
face velocity, changed insignificantly with this pre-
conditioning, being 2-3% (absolute) higher than the non-
conditioned dry samples. This indicates that the wet for-
med samples may be suitable for respiratory protection
since their performance does not deteriorate in humid
conditions in our tests.
5. Summary and Conclusions
We have demonstrated that a process of wet beating, wet
forming, and freeze drying in situ can produce filters
from wood pulp capable of removing particles from air at
an efficiency and pressure drop comparable to that seen
with commercial synthetic N95 filters. The use of wood
pulp respirator filters for high efficiency obtained by
mechanical filtration is novel as it utilizes a renewable
resource and permits environmentally friendly disposal.
However, the economics of this process remain to be
determined. Freeze drying is a commercially available
process, but its economic feasibility for this application
must be assessed. Our preliminary estimates suggest that
it is feasible, in particular for wet-formed molded filters.
Acknowledgements
We gratefully acknowledge the financial support from the NSERC SENTINEL
Bioactive Paper Network and in-kind contributions of the Weyerhaeuser
TechnologyCentre for assistance in making the air-laid mats and FPInnovations,
Paprican Division for preparing the refined pulp.
Fig 7. Filter quality factor: dry formed (dark) and wet formed (gray) filter mats.
Fig 8. Dependence of filtration efficiency of thicker wet formed G3 samples at 12
cm/s (solid symbols) and 6 cm/s (open symbol) face velocity.
Fig 6. Filtration efficiency and pressure drop of wet-formed filters.
Fig 5. Filtration efficiency and pressure drops (p) of the dry formed filters.
3708 08-11-26 16.20 Sida 424
Nordic Pulp and Paper Research Journal Vol 23 no. 4/2008 425
Literature
Balazy,A., Toivola, M., Reponen, T., Podgorski, A., Zimmer, A. and Grinshpun,
S.A. (2006a): Manikin-based Performance Evaluation of N95 Filtering-Facepiece
Respirators Challenged with Nanoparticles, Ann Occup Hyg, 3, 259.
Balazy, A., Toivola, M., Adhikari, A., Sivasubramani, S.K., Reponen, T. and
Grinshpun, S.A. (2006b): Do N95 Respirators Provide 95% Protection Level
against Airborne Viruses, and How Adequate Are Surgical masks? Am J Infect
Control, 34 (2), 51.
Chakraborty, A., Sain, M. and Kortschot, M. (2005): Cellulose microfibrils: A
Novel Method of Preparation Using High Shear Refining and Cryocrushing,
Holzforschung, 59, 102.
Chatterjee, P.K. and Makoui, K.B. (1984): Freeze Dried Microfibrilar Cellulose,
United States Patent 4,474,949.
Donaldson L. (2007): Cellulose Microfibril Aggregates and their Size Variation
with Cell Wall Type, Wood Sci.Tech. 41(5), 443.
Fernandes Diniz, J.M.B., Gill, M.H. and Castro J.A.A.M. (2004): Hornification –
its Origin and Interpretation in Wood Pulps, Wood Sci. Tech. 37(6),489.
Garner, R.G. and Kerekes, R.J. (1978): Trans,Tech Section, CPPA, 4(3), TR82.
Hinds, C.W. (1999): Aerosol Technology: Properties, Behavior, and Measurement
of Airborne Particles, John Wiley&Sons, NY.
Huang, C., Willeke, K., Qian, Y., Grinshpun, S. A. and Ulevicius, V. (1998):
Method for Measuring the Spatial Variability of Aerosol Penetration through
Respirator Filters, Am Ind Hyg Assoc J. 59, 461.
Janssen, L.L., Bidwell, J.O., Mullins, H.E. and Nelson, T.J. (2003): Efficiency of
degraded electret filters: part 1 – laboratory testing against NaCl and DOP before
and after exposure to workplace aerosols, J. Int. Soc. Resp. Protec. 20, 71.
Lee, K.W. and Mukund, R. (2001): Filter Collection, In: Willeke, K. and Baron,
P.A., eds, Aerosol Measurement, Principles, Techniques and Applications, John
Wiley&Sons, NY.
Martin, S.B. Jr and Moyer, E.S. (2000): Electrostatic Respirator Filter Media:
Filter Efficiency and Most Penetrating Particle Size Effects, App Occup Environ
Hyg. 15, 609.
National Institute for Occupational Safety and Health (NIOSH) (1997): 42
CFR, Part 84, Respiratory Protective Devices: Final Rules and Notice, Federal
Register 60:110, US Centers for Disease Control and Prevention, National
Institute for Occupational Safety and Health.
Nicholson, M. (2004): Future Trends in Air Filtration for Medical Devices,
Business Briefing: Medical Device Manufacturing& Technology, 1.
Qian, Y., Willeke, K., Grinshpun, S.A. and Donnelly, J. (1997a): Performance of
N95 Respirators: Reaerosolization of Bacteria and Solid Particles, Am Ind Hyg
Assoc J. 58, 876.
Qian, Y., Willeke, K., Ulevicius, V. and Grinshpun, S.A. (1997b): Particle
Reentrainment from Fibrous Filters, Aerosol Sci Tech. 27,394.
Qian, Y., Willeke, K., Grinshpun, S.A., Donnelly, J. and Coffey, C.C. (1998):
Performance of N95 Respirators: Filtration Efficiency for Airborne Microbial and
Inert Particles, Am Ind Hyg Assoc. J. 59,128.
Seto, W.H., Tsang, D., Yung, R.W.H., Ching, T.Y., Ng, T.K., Ho, M., Ho, L.M. and
Peiris, J.S.M. (2003): Effectiveness of Precautions against Droplets and Contact
in Prevention of Nosocomial Transmission of Severe Acute Respiratory Syndrome
(SARS), Lancet, 361, 1519.
Thiessen, R.J. (2006): Filtration of Respired Gases: Theoretical Aspects, Respir
Care Clin, 12,183.
Turbak A.F., Snyder, F.W. and Sandberg, K.R. (1983): Microfibrillated Cellulose,
aNew Cellulose Product: Properties, Uses and Commercial Potential, J Appl
Polym Sci., Appl Polym Symp. 37, 815.
Welch, L.V. and Kerekes, R.J. (1994): Characterization of the PFI Mill by the C-
Factor,Appita, 47(5), 387.
Wilkes, T. (2004): An Assessment of the Performance of Breathing System
Filters, Business Briefing: Medical Device Manufacturing & Technology, 70.
Manuscript received June 19, 2008
Accepted September 18, 2008
3708 08-11-26 16.20 Sida 425
