International Journal of

Environmental Research
and Public Health

Article
Adsorption Performance of Activated-Carbon-Loaded
Nonwoven Filters Used in Filtering
Facepiece Respirators
Małgorzata Okrasa 1, * , Jörn Hitz 2 , Aleksandra Nowak 1 , Agnieszka Brochocka 1 ,
Christoph Thelen 2 and Zbigniew Walczak 3
1 Department of Personal Protective Equipment, Central Institute for Labour Protection, National Research
Institute, Wierzbowa 48, 90-133 Łódź, Poland; alnow@ciop.lodz.pl (A.N.); agbro@ciop.lodz.pl (A.B.)
2 Sub-Division 3.3 “Personal protective equipment against Chemical and Biological Substances”, Institute for
Occupational Safety and Health of German Social Accident Insurance (IFA), Alte Heerstraße 111,
53757 Sankt Augustin, Germany; joern.hitz@dguv.de (J.H.); christoph.thelen@dguv.de (C.T.)
3 Department of Theoretical Physics, University of Lodz, Pomorska 149/153, 90-236 Łódź, Poland;
z.walczak@merlin.phys.uni.lodz.pl
* Correspondence: maokr@ciop.lodz.pl; Tel.: +48-426-480-223




Received: 20 March 2019; Accepted: 31 May 2019; Published: 4 June 2019


Abstract: Filtering nonwovens loaded with activated carbon are among the most popular materials
used in the construction of filtering facepiece respirators (FFRs) with anti-odour properties that
can be used for respiratory protection at workplaces where the occupational exposure limits of
harmful substances are not exceeded. Such FFRs, in addition to a polymer filter material of varying
effectiveness, also contain a layer of activated-carbon-loaded nonwoven filter, which limits the
quantity of chemical compounds entering the breathing zone. The aim of this work was to analyse the
influence of challenge concentration (20–120 ppm), relative humidity (2–70%), flow rate (20–55 L/min),
and flow pattern (steady-state and pulsating) on the breakthrough of polymer/carbon nonwovens.
A commercial activated-carbon-loaded nonwoven filter was used in this study. Its morphology
and textural parameters were determined using optical microscopy, image processing, and nitrogen
adsorption/desorption measurements at 77 K. Breakthrough experiments were carried out using
cyclohexane vapours to assess adsorption characteristics of polymer/carbon media. The results
showed that the breakthrough times decreased with increasing challenge concentration (up to 30%),
relative humidity (up to 73%), and flow rate (up to 72%). The pulsating flow pattern was found to be
more favourable in terms of odour reduction efficiency (up to 30%). The results indicate that all of
these factors should be considered during selection and performance assessment of respirators used
for odour relief.

Keywords: activated-carbon-loaded nonwovens; respiratory protection; filtering facepiece respirators;

anti-odour properties; breakthrough time

1. Introduction
According to the generally accepted definition, odorous compounds are substances that at a
sufficient concentration may elicit a response in the olfactory and trigeminal system in the nose [1].
Regardless of its origin (natural or man-made), odour pollution has been associated with adverse
health effects in humans (mainly self-reported), including irritation of eyes and upper respiratory
tract, headaches, nausea, drowsiness, and sore throat [2–4]. Prolonged odour exposure has also been
connected with behavioural symptoms like stress, sleep disturbance, mood swings, and depression [4,5].
Due to the growing awareness of the general public concerning the influence of air quality on health and

Int. J. Environ. Res. Public Health 2019, 16, 1973; doi:10.3390/ijerph16111973 www.mdpi.com/journal/ijerph
Int. J. Environ. Res. Public Health 2019, 16, 1973 2 of 16

wellbeing, these issues have become the subject of regulatory interest world-wide [6]. Simultaneously,
extensive research on deodorization techniques for industrial and agricultural facilities [7–9] and
methods for determining the effectiveness of such techniques have been carried out [10–12].
Little attention was paid to the matter of exposure to odorous compounds in a work environment
in the past. In situations where the occupational exposure limits (OEL) of harmful substances are
exceeded at a workplace and there is no possibility of preventing occupational risk by eliminating it at
source or minimizing it by organizational solutions or technical protection measures, it is necessary to
use appropriate personal protective equipment [13]. In particular, protection against harmful vapours
and gases can be achieved by using appropriate respiratory protective devices, such as a facepiece
(halfmask/full-face mask) with gas filters or combined filters containing adsorbent materials with a
high surface area such as activated carbon [14]. The testing methods for such devices and the influence
of different operational conditions influencing their performance are widely discussed in the literature
and well established in international standards [12,14–16].
However, when the OEL values are not exceeded, disposable filtering facepiece respirators (FFRs)
with anti-odour properties are often in use. In addition to polymer filtering materials of varying
effectiveness, they also contain a layer (or several layers) of activated-carbon-loaded nonwoven, which
adsorbs some of the volatile chemical compounds that otherwise would enter the breathing zone.
Contrary to typical gas filters or combined filters, the performance assessment methods of anti-odour
FFRs have not been clearly established in the literature. Only limited research is available on the
adsorption characteristics or capabilities of such FFRs [17]. Because the basic principle of operation of
anti-odour FFRs is similar to gas filters (adsorption on activated carbon), it is quite straightforward to
use a similar performance assessment method. However, the nature of the factors determining the
adsorption kinetics on thin carbon-loaded nonwovens (usually thinner than 1 cm) may be different than
for densely-packed carbon beds of higher thickness (up to 10 cm) that are found in gas filters. This is
mainly due to the specific phenomena that occur during the two-way flow of the above-mentioned
substances through a flat and loosely-packed bed that is a mixture of polymer fibres and grains or
granules of activated carbon adsorbent.
Water vapour is another factor that profoundly influences the performance of activated carbon
beds found in gas filters [18–21], which may be of significance when considering breakthrough times of
polymer/carbon beds used in FFRs for odour relief. Studies using standard FFRs showed that relative
humidity under the facepiece can reach over 90% within a few minutes of its use [22]. Warm and humid
air exhaled through the thin polymer/carbon bed can easily interact with activated carbon, leading
to very poor adsorption conditions (high humidity and heating of the carbon). Those conditions
persist in the material over the whole breathing cycle, affecting the adsorption also during inhalation.
Moreover, the exhaled air can cause the migration of moisture into the facepiece structure and its
accumulation between the hydrophobic fibres of the nonwoven material [23]. This effect is negligible
in the case of typical gas filters used in combination with half-masks or full-face masks because they
are usually equipped with exhalation valves that prevent the humid exhaled air from entering the gas
filter. However, it may be significant in the case of anti-odour FFRs in which the flow is bidirectional,
even if an exhalation valve is installed.
Another of the factors that should be considered when analysing the process of uptake of volatile
compounds by these types of polymer/carbon beds is the composition of the absorbed mixture.
In general, carbon adsorbents have a greater affinity for less volatile substances, but there is no
unambiguous relationship between the single characteristics of an individual chemical compound and
the capacity of the sorption bed [24]. In the case of individual chemical compounds, the breakthrough
time of the activated carbon bed, under given ambient conditions, is usually described using the original
or modified Wheeler-Jonas equation [25–28]. But when there is a mixture of various compounds in
the air, as in the case of odour exposure, the problem of determination of breakthrough time becomes
much more complex [29,30].
Int. J. Environ. Res. Public Health 2019, 16, 1973 3 of 16

For densely-packed carbon beds, the breakthrough time was also found to be inversely proportional
to the volumetric flow rate of air/adsorbate mixture for a given organic vapour, concentration, and
relative humidity [31]. In this case the flow pattern (steady-state or pulsating) had no significant
influence on the breakthrough time even for high work rates and humidity conditions. However,
for carbon beds of low thickness (1.0–3.5 cm) the dependence of the breakthrough time on the flow
pattern has been confirmed and attributed to the influence of flow rate on the mass transfer from the
bulk gas phase to the surface of the adsorbent particles [32]. It can therefore be assumed that the flow
pattern will also have considerable influence on the breakthrough time of nonwoven filters loaded
with activated carbon. However, hitherto, no studies on the issue have been conducted.
The present study addresses some of the aspects connected with the adsorption characteristics
of FFRs with anti-odour properties. Namely, the influence of material morphology, challenge
concentration, relative humidity, flow rate, and flow pattern on the breakthrough of polymer/carbon
media is considered.

2. Materials and Methods

2.1. Activated-Carbon-Loaded Nonwoven Filters

A commercially available activated-carbon-loaded nonwoven filter (AC-NW composite, type
RK-270/C, Remark-Kayser, Batorowo, Poland) typically used in the construction of FFRs with anti-odour
properties was used in this study. It was manufactured from poly-(ethylene terephthalate) (PET) by
needle-punching (160 g/m2 ) and impregnated with 110 g/m2 of powdered activated carbon.

2.2. Morphology of Activated-Carbon-Loaded Nonwoven Filters

The basis weight of AC-NW media was determined according to methodology described in EN
29073-1:1992 standard [33] using R160P laboratory balance (Sartorius GMBH, Göttingen, Germany).
Their thickness was established according to ISO 9073-2:1995 standard [34] using the J-40-T digital
material thickness gauge (CheckLine, Bad Bentheim, Germany). The pressure drop was measured
according to EN 13274-3:2001 standard [35] using breathing resistance measuring apparatus comprising
a blower/sucker system connected via flowmeter to the filter holder and a manometer connected
upstream of the sample measuring the pressure drop relative to atmosphere at 30 and 95 L/min air
flow rates. Samples (N = 21 ÷ 76) for all of the measurements were cut out randomly from different
parts of AC-NW media.
Additionally, an Axiotech 100 HD optical microscope (Carl Zeiss CMP GmbH, Göttingen, Germany)
equipped with a Nikon DS-Fi1 digital camera and Nikon DS-U2 controller (Nikon Corporation, Tokyo,
Japan) was used to evaluate fibre diameter distribution and optical density of AC-NW media.
Micrographs taken at magnification of 250x were generated using NIS-Elements BR™ software
(Laboratory Imaging Ltd., Prague, Czech Republic) and then processed with the software package
Mathematica 9 (Wolfram Research, Inc., Champaign, IL, USA). In particular, the area of the regions
which had a darker shade (areas covered by the AC-NW media) compared to the rest of the image
was calculated numerically in the following way. First, we used the routine MorphologicalBinarize
that converts multichannel and colour images into grayscale images and then produces images in
which every pixel has value 0 (white pixel) or 1 (black pixel) (Figure 1). Next, we applied the routine
DominantColors that, for a given grayscale image, determines the fraction of the whole grayscale image
covered by clusters of black pixels. This method allowed us to roughly assess the uniformity of AC
agglomeration in AC-NW composite without considering the heterogeneity of the base NW nonwoven
because transparent PET fibres were omitted by the MorphologicalBinarize routine.
Int. J. Environ. Res. Public Health 2019, 16, 1973 4 of 16
Int. J. Environ. Res. Public Health 2019, 16, x 4 of 16

(a) (b)
Figure
Figure 1. Raw
1. Raw optical
optical microscope
microscope image
image (a) and
(a) and grayscale
grayscale representation
representation (b)activated-carbon-loaded
(b) of of activated-carbon-
loaded nonwoven
nonwoven (AC-NW)(AC-NW)
sample. sample.

TheThe uniformityofofthe
uniformity theparameters
parameters of
of interest
interest was
was assessed
assessedbased
basedon onthethecoefficient
coefficientof of
variance (%)(%)
variance
and
and index
index ofof dispersionI Idefined
dispersion definedas
asthe
theratio
ratio between
between the observed
observedvariance
varianceand andthe
themean
mean value. I <I1< 1
value.
indicated
indicated spatiallyuniform
spatially distribution, II==11random
uniformdistribution, randomdistribution,
distribution, I > 1I >
andand clumped
1 clumpeddistribution [36]. [36].
distribution

2.3.2.3. Textural
Textural PropertiesofofActivated-Carbon-Loaded
Properties Activated-Carbon-Loaded Nonwoven
Nonwoven Filters
Filters
The
The adsorption/desorption
adsorption/desorption isotherms
isotherms of2 N
of N at K77were
at2 77 K were measured
measured usingusing an Autosorb
an Autosorb iQ
iQ analyser
analyser (Quantachrome Instruments, Boynton Beach, FL, USA) to determine
(Quantachrome Instruments, Boynton Beach, FL, USA) to determine the textural properties of AC-NWthe textural properties
of AC-NWMoreover,
composite. composite.activated
Moreover,carbon
activated
(AC)carbon (AC) extracted
extracted from the from
AC-NW the AC-NW
compositecomposite
was testedwas to
tested to examine the influence of incorporation on those properties.
examine the influence of incorporation on those properties. Prior to the adsorption/desorption Prior to the
adsorption/desorption
measurements, each AC-NWmeasurements,
sample was each AC-NW sample
fragmented was fragmented
and homogenised. Next,and homogenised.
samples weighting
Next, samples weighting 1 g were prepared, placed in bulb cells, and subjected to a two-stage
1 g were prepared, placed in bulb cells, and subjected to a two-stage outgassing process in conditions
outgassing process in conditions compliant with the recommendations of the National Institute of
compliant with the recommendations of the National Institute of Standards and Technology (1st stage:
Standards and Technology (1st stage: 50 °C, 90 min; 2nd stage 120 °C, 240 min) [37] to remove any
50 ◦ C, 90 min; 2nd stage 120 ◦ C, 240 min) [37] to remove any contaminants already adsorbed on the
contaminants already adsorbed on the samples. Adsorption/desorption measurements were carried
samples. Adsorption/desorption measurements were carried out at 64 measuring points (P/P0 from
out at 64 measuring points (P/P0 from 10−7 to 0.995) and desorption measurements at 40 measuring
−7
10 points
to 0.995)
(P/P0 and
fromdesorption measurements at 40 measuring points (P/P0 from 0.995 to 0.025).
0.995 to 0.025).
The
The test results were analysed using
test results were analysed using ASiQwin
ASiQwin V. V. 3.01
3.01 software
software(Quantachrome
(Quantachrome Instruments,
Instruments,
Boynton Beach, FL, USA). The specific surface area was calculated using the
Boynton Beach, FL, USA). The specific surface area was calculated using the multipoint BET multipoint BET method
method
forfor
6 points
6 points atat
P/P 0 ranging
P/P 0 rangingfrom
from0.010.01to
to 0.15.
0.15. The volumeand
The volume andsize
sizedistribution
distributionofof micropores
micropores were
were
determined using the QSDFT (quenched solid density functional theory) model
determined using the QSDFT (quenched solid density functional theory) model based on a slit-pore based on a slit-pore
model
model(pore
(pore diameter<<22nm)
diameter nm)andandaa cylindrical pore model
cylindrical pore model(pore diameter> >
(porediameter 2 nm).
2 nm).

2.4.2.4.
Adsorption
AdsorptionCharacteristics
CharacteristicsofofActivated-Carbon-Loaded NonwovenFilters
Activated-Carbon-Loaded Nonwoven Filters
Breakthrough
Breakthrough experiments
experimentswerewerecarried
carried out
out using cyclohexane
cyclohexanevapours
vapoursasasa achallenge
challenge substance
substance
at aattest
a teststand
standshown
shownininFigure
Figure2.2.Cyclohexane
Cyclohexane waswas chosen
chosento toperform
performthe theexperiments
experiments due
due to to
thethe
factfact
that it is
that it the representative
is the representativeof of
thethe
protection
protection group
group AA (organic
(organicgases)
gases)named
namedininthe thetest
teststandard
standardEN
EN 14387:2004+A1:2008
14387:2004+A1:2008 [16]. [16]. Furthermore,
Furthermore, the test
the test setup
setup presented
presented in Figure
in Figure 2 was
2 was optimized
optimized forthe
for theuse
use of cyclohexane as a test substance due to that circumstance. Cyclohexane is
of cyclohexane as a test substance due to that circumstance. Cyclohexane is a good representative a good representative
of of
thisthis group
group ofofsubstances
substancesbecause
becauseititis
is non-polar
non-polar andand unreactive
unreactiveand andtherefore
thereforeshould
should bebe bound
bound to to
the carbon surface mainly by physical adsorption. Moreover, it is one of compounds
the carbon surface mainly by physical adsorption. Moreover, it is one of compounds that is used that is used in in
substantial amounts in industrial processes, thus it is often present in the work
substantial amounts in industrial processes, thus it is often present in the work environment. environment.
Int. J. Environ. Res. Public Health 2019, 16, 1973 5 of 16
Int. J. Environ. Res. Public Health 2019, 16, x 5 of 16

Figure 2. Diagram of breakthrough testing stand: 1—compressed air; 2—dust filter; 3—pressure
Figure 2. Diagram of breakthrough testing stand: 1—compressed air; 2—dust filter; 3—pressure
regulator; 4, 7—mass flow controllers; 5—humidifier; 6—heating/cooling system; 8—evaporator of the
regulator; 4, 7—mass flow controllers; 5—humidifier; 6—heating/cooling system; 8—evaporator of
test substance; 9—mixing chamber; 10—test chamber; 11—test specimen; 12—integrated temperature
the test substance; 9—mixing chamber; 10—test chamber; 11—test specimen; 12—integrated
and humidity sensor; 13, 14—gas detectors; 15—pressure sensor; 16—exhaust; 17, 18—control valves;
temperature and humidity sensor; 13, 14—gas detectors; 15—pressure sensor; 16—exhaust; 17, 18—
19—breathing machine. Elements 15–19 were used only during pulsating flow tests.
control valves; 19—breathing machine. Elements 15–19 were used only during pulsating flow tests.

During the test, dry compressed air was delivered to the system through a particulate filter
During the test, dry compressed air was delivered to the system through a particulate filter
equipped with a pressure reducer allowing an adjustment in the range of 0–8 bar. It was then divided
equipped with a pressure reducer allowing an adjustment in the range of 0–8 bar. It was then divided
into two lines (Figure 2): line A that was used for temperature and relative humidity regulation and
into two lines (Figure 2): line A that was used for temperature and relative humidity regulation and
line
lineB Bthat
thatwas
wasusedusedtotoadjust
adjust the
the concentration
concentration of of the
the test
testsubstance
substanceininthe thetest
test chamber.
chamber. AirAir
at aat a
given
given temperature and relative humidity from line A was mixed with challenge vapours from line Bline
temperature and relative humidity from line A was mixed with challenge vapours from
B in
in aa mixing
mixingchamber
chamberand and then
then directed
directed to the
to the testtest chamber
chamber withwith the sample
the sample (13 × 13(13cm)× 13 cm) locked
locked in a
inpneumatic
a pneumatic sample holder with a diameter of 11.3 cm (surface area of 100 cm 2 ). Additionally, a
sample holder with a diameter of 11.3 cm (surface area of 100 cm ). Additionally, a 2

breathing
breathingmachine
machineconnected
connected to to the
the outlet of the
outlet of thetest
testchamber
chamberwas wasused
used during
during pulsating
pulsating flow
flow tests.
tests.
The
The breathing
breathingmachine
machinewaswassetsetat
at 25
25 cycles/min
cycles/min and and22L/stroke,
L/stroke,which
whichcorresponded
corresponded to to25 25 inhalations
inhalations
and
and 2525exhalations
exhalationsperperminute
minute with
with a respiratory volumeofof22L.L.The
respiratory volume Theairflow
airflow pattern
pattern of of
thethe breathing
breathing
machine
machinewas wassinusoidal.
sinusoidal.TheThe challenge concentrationduring
challenge concentration duringpulsating
pulsatingflow flow tests
tests waswas compensated
compensated
toto ensurethat
ensure thatthe
theoverall
overallload
loadwas
was standardized
standardized to to the
the steady-state
steady-stateflow flowconditions.
conditions. TheThe challenge
challenge
vapour
vapour concentrationupstream
concentration upstream(C (Cinin) and downstream
downstream(C (Cout)
out of
) ofthe
thesample
sample was
wasrecorded
recorded using X-am
using X-am
7000 multi-gas meters (Dräger, Lübeck, Germany) and NGA 2000
7000 multi-gas meters (Dräger, Lübeck, Germany) and NGA 2000 MLT 3.1 (Fisher-Rosemount, Wesling,MLT 3.1 (Fisher-Rosemount,
Wesling, Germany).
Germany). The temperature
The temperature and relative andhumidity
relative humidity
in the test in chamber
the test chamber were monitored
were monitored using the
using the VelociCalc® Multi-Function Meter 9545 (TSI, Shoreview, MN,
VelociCalc®Multi-Function Meter 9545 (TSI, Shoreview, MN, USA). The experiments were performed USA). The experiments were
inperformed
four variants,in each
four assessing
variants, the eachinfluence
assessingof athe influence
different factorof (i.e.,
a different
challenge factor (i.e., challenge
concentration, relative
concentration, relative humidity, challenge flow rate, challenge flow pattern) on the adsorption
humidity, challenge flow rate, challenge flow pattern) on the adsorption characteristics of AC-NW
characteristics of AC-NW composite. Specific conditions at which the experiments were performed
composite. Specific conditions at which the experiments were performed are shown in Table 1.
are shown in Table 1.
Int. J. Environ. Res. Public Health 2019, 16, 1973 6 of 16

Table 1. Testing conditions.

Variable Range of the Variable Test Conditions

Challenge flow rate (steady-state): V = 30 L/min
Challenge concentration
{20, 40, 80, 100, 120} Temperature: T = 22 ± 3 ◦ C
Cin [ppm]
Relative humidity: RH = 70 ± 3%
Challenge concentration: Cin = 81 ± 3 ppm
Relative humidity
{2, 10, 20, 35, 55, 70} Challenge flow rate (steady-state): V = 30 L/min
RH [%]
Temperature: T = 22 ± 3 ◦ C
Challenge concentration: Cin = 81 ± 3 ppm
Challenge flow rate (steady-state)
{20, 25, 30, 35, 40, 45, 50, 55} Temperature: T = 22 ± 3 ◦ C
V [L/min]
Relative humidity: RH = 70 ± 3%
Challenge concentration: Cin = 81 ± 3 ppm
Challenge flow pattern
{steady-state, pulsating} Temperature: T = 22 ± 3 ◦ C
F
Relative humidity: RH = 70 ± 3%

The measurements were carried out until the concentration of the test substance behind the sample
reached the challenge concentration (i.e., the concentration upstream of the sample). Each experiment
was carried out in n = 5÷8 independent repetitions. Breakthrough times of 50-percent and 100-percent
were determined from the breakthrough curves and corrected using the following equation:

tbi ·Cin(measured)
tbi(corr) = , (1)
Cin

where: tbi(corr) —corrected breakthrough time [min], tbi —measured breakthrough time [min],
i—breakthrough percent, Cin (measured) —measured concentration [ppm], Cin —nominal challenge
concentration [ppm].
For breakthrough concentration equal to test concentration, the sorption capacity C [g] for studied
nonwovens was calculated according to the following equation:

C = V·Cin ·tb100(corr) , (2)

where: V—challenge flow rate [L/min], Cin —challenge concentration [mg/m3 ], tb100 (corr) —corrected
breakthrough time [min].

2.5. Statistical Analysis

Statistical analyses were performed using STATISTICA 13.1 software (Statsoft, Tulsa, OK, USA).
Descriptive statistics for all variables of interest were calculated. One-way analysis of variance
(ANOVA) at the significance level 0.05 was performed to identify statistical differences between
breakthrough times determined for each of the test conditions. When statistical differences were
detected (p < 0.05), mean values were compared using Tukey’s post hoc procedure at the significance
level 0.05.

3. Results and Discussion

3.1. Morphology of Activated-Carbon-Loaded Nonwoven Filters

The morphology of the nonwoven is shown in optical microscope images (Figure 3).
Int. J. Environ. Res. Public Health 2019, 16, x 7 of 16

3. Results and Discussion

Int. J.Morphology
3.1. Environ. Res. of
Public Health 2019, 16, 1973
Activated-Carbon-Loaded Nonwoven Filters 7 of 16

The morphology of the nonwoven is shown in optical microscope images (Figure 3).

(a) (b)
Figure 3.
Figure Optical microscope
3. Optical microscope images
images of
of AC-NW
AC-NW composite
composite at
at (a)
(a) 250x
250x and
and (b)
(b) 500x magnification.

Considerable accumulation
Considerable accumulation of of agglomerated
agglomerated powdered
powdered carbon carbon material
material in in the
the spaces
spaces between
between
the tangled
the tangled polymer
polymer fibres
fibres was
was observed. Agglomerates of
observed. Agglomerates of sizes
sizes not
not exceeding
exceeding the the fibre
fibre diameter
diameter
were present in smaller quantities. They were attached to elementary
were present in smaller quantities. They were attached to elementary fibres but not embedded fibres but not embedded in their
in
structure, which is important because the deposition of carbon adsorbent
their structure, which is important because the deposition of carbon adsorbent in the nonwoven in the nonwoven structure
can significantly
structure impact theimpact
can significantly sorption theproperties
sorption of the material.
properties of theThis is due to
material. Thistheisfact
duethat
to the
the presence
fact that
of binding substances or embedding carbon granules in polymer material
the presence of binding substances or embedding carbon granules in polymer material may lead to may lead to micropore
blocking, limiting
micropore blocking, thelimiting
resulting thecapacity
resulting of capacity
carbon material
of carbon [38].
material [38].
The basis
basis weight
weight of of AC-NW
AC-NW media media ranged 2 The index of dispersion
The ranged between
between 250 250 and
and 321321 g/m
g/m2.. The index of dispersion
was higher than 1 (Table 2), which indicated that the spatial
was higher than 1 (Table 2), which indicated that the spatial distribution of mass was distribution of mass was aggregated.
aggregated.
With such
With such non-uniformity,
non-uniformity, the the adsorption properties between
adsorption properties between different
different areas
areas of of AC-NW
AC-NW media media can
can
differ significantly due to the differences in available adsorption sites. At the
differ significantly due to the differences in available adsorption sites. At the same time, the thickness same time, the thickness
of the
of the nonwoven
nonwoven was was quite
quite stable,
stable, ranging
ranging between
between 2.1 2.1 and
and 2.72.7 mm,
mm, with
with low low aa dispersion
dispersion index,
index,
indicating uniformity.
indicating uniformity.
The AC-NW
The AC-NW media media had had very
very low
low pressure
pressure drops
drops (Table
(Table 2),2), which
which cancan be be associated
associated withwith low
low
breathing resistance when used in the construction of FFRs. According
breathing resistance when used in the construction of FFRs. According to the EN 149:2001+A1:2009 to the EN 149:2001+A1:2009
standard, the
standard, the breathing
breathing resistance
resistance of the FFR
of the FFR should
should notnot exceed
exceed 60–100
60–100 Pa Pa at
at the
the flow
flow rate
rate of
of 30
30 L/min
L/min
and 210–300 Pa at the flow rate of 95 L/min depending on the protection class
and 210–300 Pa at the flow rate of 95 L/min depending on the protection class [39]. The pressure drop [39]. The pressure drop
of AC-NW
of AC-NW composite
composite constituted
constituted less
less than
than 1%1% of
of the
the threshold
threshold values
values forfor both
both airair flow
flow rates,
rates, thus
thus the
the
increase of breathing resistance posed by this nonwoven in FFRs would
increase of breathing resistance posed by this nonwoven in FFRs would be negligible. Low values of be negligible. Low values
of pressure
pressure drop drop along
along with with
the the coefficients
coefficients of variance
of variance at level
at the the level
of 6% ofand
6% 9% andindicate
9% indicate thatnon-
that the the
non-uniformities
uniformities in terms
in terms of this
of this parameter
parameter should
should notnot affect
affect thethe sorptioncharacteristics
sorption characteristicsofofthe theAC-NW
AC-NW
samples cut from various parts of
samples cut from various parts of the material.the material.
Fibre diameters
Fibre diameters of of the
the AC-NW
AC-NW media media were
were typical
typical of of nonwovens
nonwovens obtainedobtained via via needle-punching
needle-punching
ranging from
ranging from 13 13to
to3030µm.µm.ThisThisparameter
parameteraffects
affects mainly
mainly thethe filtration
filtration of of particles
particles andand should
should notnot
be
be important in the case of sorption properties. However, the impregnation
important in the case of sorption properties. However, the impregnation with activated carbon could, with activated carbon
could,
to some to extent,
some extent,
depend depend
on the onfibre
the fibre diameter
diameter as more
as more carbon
carbon particles
particles wouldbebetrapped
would trappedin in the
the
structure of the nonwoven in the areas where where the
the thinner
thinner fibre
fibre agglomeration
agglomeration was was greater.
greater.
Optical densities
Optical densities of of the nonwoven
nonwoven varied between between 56% and 98%, which additionally confirms
carbon material
that the carbon material was was non-uniformly
non-uniformly distributed in the the polymer
polymer structure,
structure, i.e., areas with high high
and low
low agglomeration
agglomeration of activated
activated carbon were clearly visible. All All irregularities
irregularities in in distribution
distribution could
could
influence the uptake of volatile chemical compounds from the stream of flowing flowing air. air.
Int. J. Environ. Res. Public Health 2019, 16, 1973 8 of 16
Int. J. Environ. Res. Public Health 2019, 16, x 8 of 16

Table2.2.Morphological
Table Morphologicalparameters
parametersofofAC-NW
AC-NWcomposite.
composite.

Basis
BasisWeight,
Weight, Thickness,
Thickness, Pressure Drop,
Pressure Drop, Pa Pa Fibre
Fibre Optical
Optical
Parameter
Parameter g/m22
g/m mm
mm atat30
30 L/min
L/min atat95
95 L/min
L/min Diameter, µm
Diameter, µm Density,
Density,% %
NN 76
76 61
61 2121 21
21 9696 66 66
Mean
Meanvalue
value 275.64
275.64 2.35
2.35 1.43
1.43 5.48
5.48 22.75
22.75 81.82
81.82
Standard
Standard 17.55 0.11 0.13 0.32 4.63 9.09
deviation 17.55 0.11 0.13 0.32 4.63 9.09
deviation
Coefficient
Coefficient 6%
6% 4%
4% 9%
9% 6%
6% 20%
20% 11%
11%
ofof
variance
variance
Index
Indexofof 1.12
1.12 0.00
0.00 0.01
0.01 0.02
0.02 0.94
0.94 1.01
1.01
dispersion
dispersion

3.2.Textural
3.2. TexturalProperties
PropertiesofofActivated-Carbon-Loaded
Activated-Carbon-LoadedNonwoven
NonwovenFilters
Filters
Thenitrogen
The nitrogenadsorption
adsorption (lower
(lower branch)/desorption
branch)/desorption (upper
(upper branch)
branch) isotherms
isotherms of activated
of activated carbon
carbon and activated-carbon-loaded nonwoven filter are shown
and activated-carbon-loaded nonwoven filter are shown in Figure 4. in Figure 4.

Figure4.4.Nitrogen
Figure Nitrogenadsorption-desorption
adsorption-desorptionisotherms
isotherms(77
(77K)
K)of
ofAC
ACand
andAC-NW
AC-NWcomposite.
composite.

Isotherms
Isotherms of of activated
activated carbon
carbon and
and activated-carbon-loaded
activated-carbon-loaded nonwoven
nonwoven were were similar
similar to to the
the
theoretical
theoreticaltype
typeI(b) curve
I(b) curvecharacteristic for microporous
characteristic for microporousmaterials having
materials pore size
having distributions
pore over a
size distributions
broad
over arange
broadincluding wide micropores
range including (0.7–2.0 nm)
wide micropores and narrow
(0.7–2.0 nm) and mesopores (>2.5 nm) [40].
narrow mesopores (> 2.5Itsnm)
typical
[40].
feature is the long plateau extending over a very wide range of relative pressures
Its typical feature is the long plateau extending over a very wide range of relative pressures (P/P 0 ). In this
(P/Pcase,
0). In
instead of instead
this case, a plateau, of aa slow increase
plateau, a slowofincrease
the adsorbed
of thevolume
adsorbed(asvolume
for type(as
II isotherms) was observed,
for type II isotherms) was
which can be due to nitrogen physiosorption on nonporous polymer fibres or the
observed, which can be due to nitrogen physiosorption on nonporous polymer fibres or the adhesive adhesive used to
incorporate the carbon
used to incorporate theincarbon
the nonwoven structure structure
in the nonwoven [41]. A rapid
[41].increase
A rapidinincrease
the adsorbed
in the nitrogen
adsorbed
volume in the low P/P range indicates the presence of highly developed porous
nitrogen volume in the low P/P0 range indicates the presence of highly developed porous structures
0 structures in the
tested samples.
in the tested samples.
Textural
Texturalproperties
propertiesofofAC ACandandAC-NW
AC-NWcomposite
compositecalculated
calculatedusing
usingBET
BETand andQSDFT
QSDFTmodels
modelsare are
shown in Table
shown in Table 3. 3.
 Int. J. Environ. Res. Public Health 2019, 16, 1973 9 of 16

Int. J. Environ. Res. Public HealthTable 16,Textural

2019, 3. x properties of AC and AC-NW composite. 9 of 16

Calculation Method Parameter AC AC-NW

Table 3. Textural properties of AC and AC-NW composite.
BET Surface area, m /g2 419.4 153.2
Calculation Method Parameter AC AC-NW
BET Fitting area,
Surface error,m%2/g 0.15 153.2 0.16
419.4
Pore width,
Fitting error,nm
% 0.85 0.16 0.85
0.15
QSDFT
Pore width, nm
Average pore volume, cm /g 3 0.85
0.22 0.85 0.09
QSDFT
Average pore volume, cm /g 0.22
3 0.09
Surface area, m22/g 418.9 144.4
Surface area, m /g 418.9 144.4
BET—Brunauer–Emmett–Teller; QSDFT—Quenched Solid Density Functional Theory
BET - Brunauer–Emmett–Teller; QSDFT - Quenched Solid Density Functional Theory

Good fit
Good fit of experimental
experimentaldata datatotothe
thetheoretical
theoretical models
modelswaswasobserved
observed (correlation coefficient
(correlation close
coefficient
to 1 for
close to BET
1 forandBETlow fitting
and low error
fittingvalue
errorforvalue
QSDFT) (Table 3).(Table
for QSDFT) At the 3).
sameAt time the specific
the same time thesurface area
specific
of AC-NW
surface areacomposite
of AC-NWwas approx. was
composite 35% approx.
of the specific
35% ofsurface area ofsurface
the specific the extracted
area ofAC,
the which
extractedis due
AC,to
the high
which content
is due of non-porous
to the high content polymer fibres in polymer
of non-porous the testedfibres
samples [14].tested
in the Lower textural[14].
samples parameters
Lower
obtained
textural for AC-NW
parameters media could
obtained have also
for AC-NW resulted
media could from blocking
have of adsorbent
also resulted pores by of
from blocking theadsorbent
adhesives
used by
pores in the
the nonwoven
adhesives usedimpregnation process [38].
in the nonwoven Pore size distribution
impregnation process [38].ofPore AC size
and distribution
AC-NW compositeof AC
(Figure
and AC-NW 5) indicated
compositethat(Figure
the size5)ofindicated
most of the
thatpores ranged
the size fromof0.5
of most to pores
the 2.0 nmranged
with two
from peaks
0.5 toat
approx. 0.6 and 1.0 nm. Mesopores with dimensions ranging from 2.0 to
2.0 nm with two peaks at approx. 0.6 and 1.0 nm. Mesopores with dimensions ranging from 2.0 to50.0 nm were also observed,
which
50.0 nmmight
were have contributed
also observed, to themight
which presence
haveofcontributed
a small hysteresis loop (Figure
to the presence of a 4) generally
small associated
hysteresis loop
with metastability
(Figure 4) generallyof the adsorbed
associated withmultilayer,
metastabilitynetwork
of the effects,
adsorbedor pore blocking
multilayer, [37]. Macropores
network were
effects, or pore
not observed
blocking in either of the
[37]. Macropores samples.
were not observed in either of the samples.

Figure 5. Pore size distribution of AC and AC-NW composite based on QSDFT model.
Figure 5. Pore size distribution of AC and AC-NW composite based on QSDFT model.
3.3. Adsorption Characteristics of Activated-Carbon-Loaded Nonwoven Filters
3.3. Adsorption Characteristics of Activated-Carbon-Loaded Nonwoven Filters
3.3.1. Influence of Challenge Concentration
3.3.1. Influence
In Figure of
6aChallenge Concentration
the averaged concentrations of cyclohexane downstream of the AC-NW sample
obtained for upstream concentration in the range
In Figure 6a the averaged concentrations ofofcyclohexane
20-120 ppm downstream
at a steady-state of V = 30
flowAC-NW
of the L/min,
sample
T = 22 ± for
obtained
◦
3 C upstream = 70 ± 3% (seeinTable
and RH concentration 1) were
the range plottedppm
of 20-120 versus
at atime. Corresponding
steady-state breakthrough
flow of V = 30 L/min,
T = 22 ± 3 °C and RH = 70 ± 3% (see Table 1) were plotted versus time. Corresponding breakthrough
percentages of cyclohexane vapours through AC-NW media calculated as a ratio of downstream
concentration Cout to upstream concentration Cin (%) are shown in Figure 6b.
Int. J. Environ. Res. Public Health 2019, 16, 1973 10 of 16

percentages of Public
Int. J. Environ. Res. cyclohexane vapours
Health 2019, 16, x through AC-NW media calculated as a ratio of downstream
10 of 16
concentration
Int. J. Environ. Res.CPublic
out to upstream
Health 2019, concentration
16, x Cin (%) are shown in Figure 6b. 10 of 16

(a) (b)
(a) (b)
Figure 6. Downstream concentration (a) and breakthrough percentages (b) for AC-NW media in
Figure 6. Downstream concentration
time. concentration (a)
(a) and
and breakthrough
breakthroughpercentages
percentages(b)
(b)for
forAC-NW
AC-NWmedia
mediainintime.
time.
In contrast to
In contrast densely packed
to densely packed carbon
carbon beds
beds found
found in
in gas filters (depth
gas filters (depth greater
greater than
than 22 cm), an initial
cm), an initial
breakthrough (from
In contrast(from
breakthrough 0%
0% to
to densely 20%)
20%) occurred
to packed at
at the
carbon beds
occurred the very
found beginning
very in gas filtersof
beginning each
each experiment
of(depth greater than(within
experiment the
2 cm), an
(within first
theinitial
first
30 s), which
breakthrough resulted
(from from
0% tothe flow
20%) of the
occurred challenge
at the mixture
very through
beginning of the
each areas of
experiment
30 s), which resulted from the flow of the challenge mixture through the areas of the nonwoven withthe nonwoven
(within the with
first
lower
lower aggregation
30 s), which resulted
aggregation of AC
AC confirmed
offrom the flow ofby
confirmed thebasis
by basis weight
challenge
weight uniformity
mixture analysis.
through
uniformity After
the areas
analysis. that,nonwoven
of the
After that, an increase
an increasewithin
in
breakthrough
lower percentage
aggregation of AC with the
confirmed increasing
by basis upstream
weight concentration
uniformity of
analysis.the adsorbate
After
breakthrough percentage with the increasing upstream concentration of the adsorbate was observed, that, was
an observed,
increase in
which
which is consistent
breakthrough with
percentage
is consistent the
with
with the results obtained
the increasing
results for
obtained for densely
upstream packed carbon
denselyconcentration
packed carbon beds
of beds [42].
the adsorbate
[42]. was observed,
whichAdsorption
Adsorption characteristics
is consistent
characteristics of
of AC-NW
with the results obtained
AC-NW composite determined
for densely
composite based beds
packed carbon
determined based on the
on the breakthrough curves
[42].
breakthrough curves
are shown in
AdsorptionFigure 7 along with
characteristics ofthe results
AC-NW of the
compositestatistical analysis.
determined
are shown in Figure 7 along with the results of the statistical analysis. based on the breakthrough curves
are shown in Figure 7 along with the results of the statistical analysis.

(a) (b)
(a) (b)
Figure 7. Influence of challenge concentration
concentration on (a) corrected breakthrough time and (b) sorption
Figure 7.A–C,
capacity;
capacity; Influence of challengesignificant
A–C,a,a,b—statistically
b—statisticallyconcentration
significant on (a) between
corrected
differences
differences breakthrough
between
mean mean are time
values values andmarked
are
marked (b) different
with sorption
with
capacity;
different
letters; A–C, a,
letters;
ANOVA, p <b—statistically
ANOVA, p < 0.05;test,
0.05; Tukey’s p < 0.05.
significant
Tukey’s test,differences
p < 0.05. between mean values are marked with
different letters; ANOVA, p < 0.05; Tukey’s test, p < 0.05.
The 100-percent breakthrough times for AC-NW composite decreased with increasing
The 100-percent
concentrations (tb100(corr) breakthrough
ranged from 8.5 times
± 1.3 for
min AC-NW composite
to 6.0 ± 1.2 decreased
min). Statistical withindicates
analysis increasing
the
concentrations
existence (tb100(corr)
of some ranged significant
statistically from 8.5 ± 1.3 min to 6.0
differences ± 1.2 min).
between Statistical
averages analysis
obtained for indicates the
various test
existence of some statistically significant differences between averages obtained for various test
Int. J. Environ. Res. Public Health 2019, 16, 1973 11 of 16

The 100-percent
Int. J. Environ. breakthrough
Res. Public Health 2019, 16, x times for AC-NW composite decreased with increasing 11 of 16
concentrations (tb100(corr) ranged from 8.5 ± 1.3 min to 6.0 ± 1.2 min). Statistical analysis indicates
concentrations
the existence of(p < 0.05,
some Figure 7).
statistically In the case
significant of 50-percent
differences between breakthrough
averages obtained times,for a more
variousstable
test
diminishing
concentrations (p <
trend was observed
0.05, Figure with
7). In increasing
the case concentration
of 50-percent (tb 50(corr) ranged times,
breakthrough from 1.9 ± 0.3 min
a more to
stable
1.1 ± 0.2 min).
diminishing Thewas
trend sorption capacity
observed with increased, with rising challenge
increasing concentration (tb50(corr)concentrations,
ranged from 1.9for the min
± 0.3 entire
to
range
1.1 of concentrations
± 0.2 min). The sorption except for a minute
capacity increased,fall atwith
100 ppm.
rising challenge concentrations, for the entire
rangeThe results of the analyses
of concentrations except forarealargely
minuteinfall agreement
at 100 ppm. with previously published data. Nelson and
Harder postulated that the breakthrough time of
The results of the analyses are largely in agreement with previouslyan activated carbon published
bed, under given
data. ambient
Nelson and
conditions, is proportional
Harder postulated that the to the appropriate
breakthrough time of power of absorbed
an activated carbon substance concentration
bed, under given ambient[43].
Van Osdell is
conditions, et proportional
al. confirmedtothat thethis relationship
appropriate powercan be used to predict
of absorbed substancebreakthrough
concentration times[43].
of
densely
Van Osdellpacked carbon beds
et al. confirmed (depth
that this 2.54 cm)
relationship canfor
be usedlow tochallenge concentrations
predict breakthrough timesperforming
of densely
measurements
packed carbon at high
beds concentrations
(depth 2.54 cm) for [43].
lowThe power relationship
challenge concentrations between breakthrough
performing time and
measurements at
concentration does not
high concentrations [43].constitute
The power a good fit for tbbetween
relationship 100(corr) time (R2 = 0.61). time
breakthrough A better fit of the results
and concentration to
does
the
not theoretical
constitute acurve
good was found
fit for for tbtime
tb100(corr) 50(corr) 2 = 0.61).
(Rtime (R2 = 0.84).
A betterThese
fit ofdifferences
the resultsmay to thebetheoretical
due to thecurve
non-
uniformity
was of AC
found for concentration
tb50(corr) 2
time (R in = AC-NW
0.84). These nonwoven
differencescompared
may be todue
densely packed
to the carbon beds
non-uniformity of and
AC
the very low bed thickness. In the worst case scenario, the difference
concentration in AC-NW nonwoven compared to densely packed carbon beds and the very low bed in carbon content between two
100 cm2 samples 2/g in2available
thickness. In the could
worst be as scenario,
case high as 0.28 theg,difference
which would correspond
in carbon contenttobetween
approx.two115 m 100 cm samples
adsorption high as 0.28 g, which would correspond to approx. 115 m /g in available adsorption sites.
could be as sites. 2

3.3.2. Influence
Influence of Humidity

Breakthrough times and sorption capacities in relation to the relative humidity of the challenge
mixture are shown in Figure 8.

(a) (b)
Figure 8. Influence
Influence of
of relative
relative humidity
humidity on on (a)
(a) corrected
corrected breakthrough
breakthrough time
time and
and (b)
(b) sorption
sorption capacity;
capacity;
A-D, a, b—statistically significant differences
differences between mean
mean values
values are
are marked
marked with
with different
different letters;
letters;
ANOVA, pp <
ANOVA, test, p <
< 0.05; Tukey’s test, < 0.05.

The results
The results indicate
indicate aa significant
significant decrease
decrease of
of 100-percent
100-percent breakthrough
breakthrough timetime with
with an
an increase
increase in
in
the relative
the relative humidity
humidity of of the
the challenge
challenge mixture
mixture (from
(from approx.
approx. 3030 min
min for
for 2%
2% and
and 10%
10% to 8.2 ±
to 8.2 1.0 min
± 1.0 min
for 70%). In the case of 50-percent breakthrough, no statistically significant differences between
for 70%). In the case of 50-percent breakthrough, no statistically significant differences between times times
obtained at
obtained at different
different relative
relative humidity
humidity conditions
conditions were
were found
found (values
(values ranged
ranged from 3.3 ±
from 3.3 1.9 min
± 1.9 min for
for
10%
10% to 1.3 ±
to 1.3 0.4 min
± 0.4 min for
for 70%).
70%). AtAtthe
thesame
same time,
time, aa significant
significant reduction
reduction inin the
the sorption
sorption capacity
capacity of
of the
the
studied beds was observed (from 0.27 ± 0.06 g for 2% and 10% to 0.07 ± 0.01
studied beds was observed (from 0.27 ± 0.06 g for 2% and 10% to 0.07 ± 0.01 g for 70%). g for 70%).
The literature describes a number of models characterising the water adsorption process on the
surface of carbon materials [44–47]. In most cases, activated carbon, upon contact, adsorbs water
vapour from the challenge stream, thus reducing its capacity to adsorb non-polar organic vapours.
The research on respiratory protective equipment indicates that the reduction of sorption capacity of
Int. J. Environ. Res. Public Health 2019, 16, 1973 12 of 16

The literature describes a number of models characterising the water adsorption process on the
surface of carbon materials [44–47]. In most cases, activated carbon, upon contact, adsorbs water vapour
from the challenge
Int. J. Environ. stream,
Res. Public thus16,
Health 2019, reducing
x its capacity to adsorb non-polar organic vapours. The research
12 of 16
on respiratory protective equipment indicates that the reduction of sorption capacity of organic vapours
organic
with vapourshumidity
increasing with increasing
might behumidity might be
much stronger at much strongerconcentrations
low challenge at low challenge concentrations
[18,19]. This may
have significant implications in the case of carbon-loaded nonwoven filters found in FFRsfound
[18,19]. This may have significant implications in the case of carbon-loaded nonwoven filters with
in FFRs with
anti-odour anti-odour
properties, properties,
which can be which
used upcan
tobe
OEL used up to OEL concentrations.
concentrations.

3.3.3. Influence of Flow Rate

Breakthrough times and sorption capacities as a function of challenge flow rate are shown in
Figure 9.

(a) (b)
Figure 9.9. Influence
Influenceofof challenge
challenge flowflow
raterate oncorrected
on (a) (a) corrected breakthrough
breakthrough time
time and (b) and (b) capacity;
sorption sorption
capacity;
A, A, B, a-c—statistically
B, a-c—statistically significant between
significant differences differences
mean between mean
values are values
marked aredifferent
with markedletters;
with
ANOVA, p < 0.05;
different letters; ANOVA,
Tukey’s ptest, p < Tukey’s
< 0.05; 0.05. test, p < 0.05.

For
For tb
tb100(corr) no statistically significant decrease was observed with increasing flow rate. On the
100(corr) no statistically significant decrease was observed with increasing flow rate. On the
other
other hand,
hand, for for tbtb50(corr) such a relationship was statistically significant only for a flow rate range of
50(corr) such a relationship was statistically significant only for a flow rate range of
20–35
20–35 L/min. Subsequent semi-constant
L/min. Subsequent semi-constant valuesvalues ofof tb
tb50(corr) for the challenge flow rates ranging from
50(corr) for the challenge flow rates ranging from
35-55
35-55 L/min
L/min might
might have
have resulted
resulted from
from negligible
negligible differences
differences in in the
the adsorption
adsorption rates
rates inin the initial phase
the initial phase
of each experiment. For low flow rates, the increase in the test substance
of each experiment. For low flow rates, the increase in the test substance concentration downstream concentration downstream
of
of the
the sample
sample was was less
less rapid
rapid than
than atat higher flow rates,
higher flow which was
rates, which was due
due to
to the
the fact
fact that
that atat low
low flow
flow
rates, the time
rates, the timeof ofinteraction
interactionofofa agivengiven volume
volume of of
thethe challenge
challenge mixture
mixture with with the polymer/carbon
the polymer/carbon bed
bed was relatively long. This allowed more of the test substance
was relatively long. This allowed more of the test substance to be absorbed from to be absorbed from thethe challenge
challenge
mixture
mixture resulting
resultingininaagradualgradualincrease
increase ofofdownstream
downstream concentration.
concentration. With the the
With increase of the
increase offlow rate,
the flow
the
rate,time
the of interaction
time of the same
of interaction of thevolume of the challenge
same volume mixture with
of the challenge the bed
mixture withshortened causing a
the bed shortened
more rapid increase of downstream concentration.
causing a more rapid increase of downstream concentration.
Published
Published research
researchon onthe
thedependence
dependence of the breakthrough
of the breakthrough time time
on theonchallenge flow rate
the challenge mainly
flow rate
concerns densely-packed carbon beds found in gas filters. In their
mainly concerns densely-packed carbon beds found in gas filters. In their case, an inversely case, an inversely proportional
relation
proportionalbetween the breakthrough
relation time and thetime
between the breakthrough flowand ratetheis fulfilled
flow rateinisthe conditions
fulfilled in theof constant
conditions
concentration and humidityand
of constant concentration [31].humidity
In the case of carbon-loaded
[31]. In the case of nonwoven filtersnonwoven
carbon-loaded the relationfilters
between
the
those two variables might be even stronger (Figure 9). This means
relation between those two variables might be even stronger (Figure 9). This means that the that the differences in breakthrough
times for different
differences work intensities
in breakthrough times for willdifferent
be moreworkclearly visible and
intensities willshould
be more be clearly
considered during
visible and
assessment of the effectiveness of FFRs with anti-odour properties.
should be considered during assessment of the effectiveness of FFRs with anti-odour properties.
The sorption capacity increased with growing challenge flow rate over the entire flow range.
This result is consistent with the theoretical predictions, according to which, for a constant
concentration of the test substance, the capacity should increase proportionally to the product of the
flow rate and breakthrough time.
Int. J. Environ. Res. Public Health 2019, 16, 1973 13 of 16

The sorption capacity increased with growing challenge flow rate over the entire flow range.
This result is consistent with the theoretical predictions, according to which, for a constant concentration
of the test substance, the capacity should increase proportionally to the product of the flow rate and
breakthrough
Int. J. Environ. Res.time.
Public Health 2019, 16, x 13 of 16

3.3.4.
3.3.4. Influence
Influence of
of Flow
Flow Pattern
Pattern
In
In addition
addition to
to the
the challenge
challenge flow,
flow,the
the absorption
absorption process
process was
was also
also affected
affected by
by the
the challenge
challenge flow
flow
pattern.
pattern. Figure
Figure 10
10 presents
presents the
the results
results of
ofbreakthrough
breakthroughexperiments
experiments performed
performed for
for steady-state
steady-state and
and
pulsating
pulsatingflow.
flow.

(a) (b)
Figure 10.
Figure 10. Influence
Influence ofof challenge
challenge flow
flow pattern
pattern onon (a)
(a) corrected
corrected breakthrough
breakthrough time
time and
and (b)
(b) sorption
sorption
capacity;A,A,B,B,
capacity; a, b—statistically
a, b—statistically significant
significant differences
differences between
between mean are
mean values values are with
marked marked with
different
letters; t-test,
different p < t-test,
letters; 0.05. p < 0.05.

The
The breakthrough
breakthrough timestimes obtained
obtained forfor steady-state
steady-state flow
flow werewere nearly
nearly three
three times
times shorter
shorter than
than for
for
pulsating
pulsatingflow (p<< 0.05),
flow(p 0.05), which
which resulted
resulted from
from aa shorter
shorter interaction
interaction time time between
between the the challenge
challenge stream
stream
and
andactivated-carbon-loaded
activated-carbon-loadednonwovennonwovenfilter (only
filter (onlyin in
inhalation
inhalation phase).
phase).It cannot
It cannotbe ruled
be ruledout out
that that
the
observed phenomena were also influenced by the sinusoidal changes in
the observed phenomena were also influenced by the sinusoidal changes in the flow rate of the the flow rate of the challenge
mixture
challenge through
mixturethethrough
sample andthe the removal
sample and of theremoval
the adsorbate of from the polymer/carbon
the adsorbate material by the
from the polymer/carbon
exhaled air. This possibility was previously indicated in the literature in the context
material by the exhaled air. This possibility was previously indicated in the literature in the context of densely-packed
activated carbon beds
of densely-packed [32,48], however,
activated carbon beds to fully confirm
[32,48], this effect
however, for carbon-loaded
to fully confirm this effectnonwoven filters,
for carbon-
desorption
loaded nonwovenexperiments
filters,should
desorptionbe conducted
experiments at various
should be testing conditions.
conducted at variousThetesting
adverse effect of
conditions.
humidity, originating from the humidifying system located in the exhalation
The adverse effect of humidity, originating from the humidifying system located in the exhalation line of breathing machine,
on
linethe breakthrough
of breathing time on
machine, andthesorption capacity
breakthrough timewas
and smaller
sorption than the effect
capacity was of the pulsating
smaller flow.
than the effect
Hence, the reduction
of the pulsating flow.of Hence,
the breakthrough
the reductiontime ofdue to breakthrough
the the accumulation timeofduemoisture
to theinaccumulation
the structure of of
AC-NW composite was not observed. This result is interesting and requires
moisture in the structure of AC-NW composite was not observed. This result is interesting and further investigation, with
arequires
broader further
range ofinvestigation,
testing conditions,with as
a itbroader
might haverange vast
of implications in termsas
testing conditions, of it
themight
performance
have vast of
FFRs with anti-odour properties.
implications in terms of the performance of FFRs with anti-odour properties.

4.
4. Conclusions
Conclusions
Microscopic
Microscopicanalysis
analysishas
hasallowed
allowedthe
the identification
identificationof
of morphological
morphologicalfeatures
featuresthat
that may
may influence
influence
the adsorption characteristics of carbon-loaded nonwoven filters. It was found that the homogeneity
the adsorption characteristics of carbon-loaded nonwoven filters. It was found that the homogeneity of
carbon in the
of carbon polymer
in the fibre
polymer structure
fibre may
structure may have
havesignificantly
significantlyaffected
affectedthose
thoseproperties.
properties. Nitrogen
Nitrogen
adsorption/desorption measurements allowed the determination of textural properties of the
polymer/carbon media. The obtained results indicated the presence of a highly developed porous
structure with a minor contribution of mesopores. The breakthrough times decreased with increasing
concentration. With the increase in the relative humidity of the test mixture, a significant shortening
(even by 30% compared to the experiments performed in dry conditions) of the breakthrough times
Int. J. Environ. Res. Public Health 2019, 16, 1973 14 of 16

adsorption/desorption measurements allowed the determination of textural properties of the

polymer/carbon media. The obtained results indicated the presence of a highly developed porous
structure with a minor contribution of mesopores. The breakthrough times decreased with increasing
concentration. With the increase in the relative humidity of the test mixture, a significant shortening
(even by 30% compared to the experiments performed in dry conditions) of the breakthrough times
for the tested beds was observed. The relationship between the breakthrough time and the challenge
flow rate fitted a power trend line and was stronger than similar relations found in the literature
for densely-packed carbon beds. This may have a direct impact on estimating protection time for
respiratory protective equipment containing such materials during work. The test mixture flow mode
also had a significant influence on adsorption characteristics. The shortening of the interaction time of
the challenge mixture with the polymer/carbon bed was a dominant factor responsible for extending the
breakthrough time. However, in this case, the influence of increased humidity of exhaled air, sinusoidal
changes in flow, and the phenomena of test substance removal through the exhaled air on the sorption
process cannot be ruled out. Although, in general, it is difficult to establish the relationship between
odour and concentration of a substance due to interpersonal differences regarding odour thresholds,
the proposed approach extends the existing knowledge in the area of performance assessment of FFRs
containing activated-carbon-loaded nonwoven filters. Regardless of the limitations, the study results
show clearly which of the external factors need to be considered while assessing the efficiency of
such respiratory protective devices, which is valuable for health and safety professionals in terms
of the evaluation and selection of respirators, as well as establishing times that they can be safely
used by workers. Future research in this field should be aimed at developing objective methods for
the performance assessment of FFRs with anti-odour properties that would consider inter-individual
differences between individuals regarding odour thresholds.

Author Contributions: Conceptualization: M.O. and A.B.; data curation: M.O. and Z.W.; investigation: M.O.,
J.H., and A.N.; methodology: M.O and J.H.; supervision: A.B. and C.T.; visualization: M.O., A.N., and Z.W.;
writing—original draft: M.O.; writing—review & editing: M.O., A.N., A.B., J.H., C.T., and Z.W.
Funding: The publication is based on the results of CIOP-PIB statutory project funded in the years 2017-2019 by
the Polish Ministry of Science and Higher Education.
Acknowledgments: M.O. would like to express her gratitude to PEROSH for its support within the PEROSH
Research Exchange Fellowship Programme. This work was also partially supported by the University of Lodz.
Conflicts of Interest: The authors declare no conflict of interest. The funders had no role in the design of the
study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, and in the decision to
publish the results.

References
1. Doty, R.; Cometto-Muñiz, J.E.; Dalton, P. Assessment of upper respiratory tract and ocular irritative effects of
volatile chemicals in humans. Crit. Rev. Toxicol. 2004, 34, 85–142. [CrossRef] [PubMed]
2. Schiffman, S.S.; Williams, C.M. Science of odor as a potential health issue. J. Environ. Qual. 2005, 34, 129–138.
[PubMed]
3. Jaén, C.; Dalton, P. Asthma and odors: The role of risk perception in asthma exacerbation. J. Psychosom. Res.
2014, 77, 302–308. [CrossRef]
4. Blanes-Vidal, V.; Bælum, J.; Schwartz, J.; Løfstrøm, P.; Christensen, L.P. Respiratory and sensory irritation
symptoms among residents exposed to low-to-moderate air pollution from biodegradable wastes. J. Expo.
Sci. Environ. Epidemiol. 2014, 24, 388–397. [CrossRef] [PubMed]
5. Blanes-Vidal, V.; Bælum, J.; Nadimi, E.S.; Løfstrøm, P.; Christensen, L.P. Chronic exposure to odorous
chemicals in residential areas and effects on human psychosocial health: Dose-response relationships. Sci.
Total Environ. 2014, 490, 545–554. [CrossRef] [PubMed]
6. Nicell, J.A. Assessment and regulation of odour impacts. Atmos Environ. 2009, 43, 196–206. [CrossRef]
7. Burgess, J.E.; Parsons, S.A.; Stuetz, R.M. Developments in odour control and waste gas treatment
biotechnology: A review. Biotechnol. Adv. 2001, 19, 35–63. [CrossRef]
Int. J. Environ. Res. Public Health 2019, 16, 1973 15 of 16

8. Apgar, D.; Witherspoon, J.; Easter, C. Gas phase odor treatment. In Minimization of Odors and Corrosion in
Collection Systems; The Water Environment Research Foundation: Alexandria, VA, USA, 2007.
9. Bandosz, T.J. Activated Carbon Surfaces in Environmental Remediation; Academic Press: New York, NY,
USA, 2006.
10. Frechen, F. State of the art of odour measurement. Int Symp Odor Meas 2003, 3, 149–155.
11. Capelli, L.; Sironi, S.; Del Rosso, R. Odor sampling: Techniques and strategies for the estimation of odor
emission rates from different source types. Sensors 2013, 13, 938–955. [CrossRef]
12. Amid, H.; Maze, B.; Flickinger, M.; Pourdeyhimi, B. Dynamic adsorption of ammonia: Apparatus, testing
conditions, and adsorption capacities. Meas. Sci. Technol. 2017, 28, 1–7. [CrossRef]
13. Directive 89/656/EEC of 30 November 1989 on the Minimum Health and Safety Requirements for the Use by
Workers of Personal Protective Equipment at the Workplace. Available online: https://eur-lex.europa.eu/
legal-content/EN/TXT/PDF/?uri=CELEX:31989L0656&from=en (accessed on 17 March 2019).
14. Bansal, R.C.; Goyal, M. Activated Carbon Adsorption; CRC Press: Boca Raton, FL, USA.
15. Ruthven, D.M.M. Principles of Adsorption and Adsorption Processes; Wiley: New York, NY, USA.
16. European Standard EN 14387:2004+A1:2008. Respiratory Protective Devices–Gas Filters and Combined
Filters–Requirements, Testing, Marking; European Committee for Standardization: Bruxelles, Belgium, 2008.
17. Rozzi, T.; Snyder, J.; Novak, D. Pilot study of aromatic hydrocarbon adsorption characteristics of disposable
filtering facepiece respirators that contain activated carbon. J. Occup. Environ. Hyg. 2012, 9, 624–629.
[CrossRef] [PubMed]
18. Nelson, G.O.; Correia, A.N.; Harder, C.A. Respirator cartridge efficiency studies: VII. Effect of relative
humidity and temperature. Am. Ind. Hyg. Assoc. J. 1976, 37, 280–288. [CrossRef] [PubMed]
19. Nelson, G.O.; Correia, A.N. Respiratory cartridge efficiency studies: VIII. Summary and conclusions.
Am. Ind. Hyg. Assoc. J. 1976, 37, 514–525. [CrossRef] [PubMed]
20. Lodewyckx, P.; Vansant, E.F. Influence of humidity on adsorption capacity from the Wheeler-Jonas model for
prediction of breakthrough times of water immiscible organic vapors on activated carbon beds. Am. Ind.
Hyg. Assoc. J. 1999, 60, 612–617. [CrossRef]
21. Werner, M.D. The effects of relative humidity on the vapor phase adsorption of trichloroethylene by activated
carbon. Am. Ind. Hyg. Assoc. J. 1985, 46(10), 585–590. [CrossRef]
22. Majchrzycka, K.; Okrasa, M.; Skóra, J.; Gutarowska, B. Evaluation of the survivability of microorganisms
deposited on filtering respiratory protective devices under varying conditions of humidity. Int. J. Environ.
Res. Public Health 2016, 13, 98. [CrossRef]
23. Raynor, P.C.; Leith, D. The influence of accumulated liquid on fibrous filter performance. J. Aerosol. Sci. 2000,
31, 19–34. [CrossRef]
24. Nelson, G.O.; Harder, C.A. Respirator cartridge efficiency studies: V. Effect of solvent vapor. Am. Ind. Hyg.
Assoc. J. 1974, 35, 391–410. [CrossRef]
25. Yoon, Y.H.; Nelson, J.H. Application of gas adsorption kinetics — II. A theoretical model for respirator
cartridge service life and its practical applications. Am. Ind. Hyg. Assoc. J. 1984, 45, 517–524. [CrossRef]
26. Yoon, Y.H.; Nelson, J.H. Application of gas adsorption kinetics I. A theoretical model for respirator cartridge
service life. Am. Ind. Hyg. Assoc. J. 1984, 45, 509–516. [CrossRef]
27. Lodewyckx, P.; Wood, G.O.; Ryu, S.K. The Wheeler-Jonas equation: A versatile tool for the prediction of
carbon bed breakthrough times. Carbon 2004, 42, 1351–1355. [CrossRef]
28. Lodewyckx, P.; Verhoeven, L. Using the modified Wheeler-Jonas equation to describe the adsorption of
inorganic molecules: Chlorine. Carbon 2003, 41, 1215–1219. [CrossRef]
29. Wood, G.O. Reviews of Models for Adsorption of Single Vapors, Mixtures of Vapors, and Vapors at
High Humidities on Activated Carbon for Applications Including Predicting Service Life of Organic
Vapor Respirator Cartridges. 2000. Available online: http://gerryowood.com/uploads/3/4/3/1/3431344/
modelallreviews.pdf (accessed on 17 March 2019).
30. Yoon, Y.H.; Nelson, J.H.; Lara, J. Respirator cartridge service-life: Exposure to mixtures. Am. Ind. Hyg.
Assoc. J. 1996, 57, 809–819. [CrossRef] [PubMed]
31. Nelson, G.O.; Harder, C.A. Respirator cartridge efficiency studies. IV. Effects of steady-state and pulsating
flow. Am. Ind. Hyg. Assoc. J. 1972, 33, 797–805. [CrossRef]
Int. J. Environ. Res. Public Health 2019, 16, 1973 16 of 16

32. Linders, M.J.G.; Mallens, E.P.J.; Van Bokhoven, J.J.G.M.; Kapteijn, F.; Moulijn, J.A. Breakthrough of shallow
activated carbon beds under constant and pulsating flow. Am. Ind. Hyg. Assoc. J. 2003, 64, 173–180.
[CrossRef]
33. European Standard EN 29073-1:1992. Methods of Test for Nonwovens. Methods of Test for Nonwovens.
Determination of Mass per Unit Area; European Committee for Standardization: Bruxelles, Belgium, 1992.
34. International Standard ISO 9073-2:1995. Textiles. Test Methods for Nonwovens. Part 2: Determination of Thickness;
International Organization for Standardization: Geneva, Switzerland, 1995.
35. European Standard EN 13274-3:2001. Respiratory Protective Devices. Methods of Test. Part 3: Determination of
Breathing Resistance; European Committee for Standardization: Bruxelles, Belgium, 2001.
36. Amirnasr, E.; Shim, E.; Yeom, B.-Y.; Pourdeyhimi, B. Basis weight uniformity analysis in nonwovens. J. Text.
Inst. 2014, 105, 444–453. [CrossRef]
37. Klobes, P.; Meyer, K.; Munro, R.G. Porosity and Specific Surface Area Measurements for Solid Materials.
NIST Recommended Practice Guide. Available online: https://ws680.nist.gov/publication/get_pdf.cfm?pub_
id=854263 (accessed on 17 March 2019).
38. Amid, H.; Mazé, B.; Flickinger, M.C.; Pourdeyhimi, B. Hybrid adsorbent nonwoven structures: A review of
current technologies. J. Mater. Sci. 2016, 51, 4173–4200. [CrossRef]
39. European Standard EN 149:2001+A1:2009. Respiratory Protective Devices. Filtering Half Masks to Protect against
Particles. Requirements, Testing, Marking; European Committee for Standardization: Bruxelles, Belgium, 2009.
40. Thommes, M.; Kaneko, K.; Neimark, A.V.; Olivier, J.P.; Rodriguez-Reinoso, F.; Rouquerol, J.; Sing, K.S.W.
Physisorption of gases, with special reference to the evaluation of surface area and pore size distribution
(IUPAC Technical Report). Pure Appl. Chem. 2015, 87, 1051–1069. [CrossRef]
41. Sotomayor, F.J.; Cychosz, K.A.; Thommes, M. Characterization of micro/mesoporous materials by
physisorption: Concepts and case studies. Acc. Mater. Surf. Res. 2018, 3, 34–50.
42. Nelson, G.O.; Harder, C.A. Respirator cartridge efficiency studies: VI. Effect of concentration. Am. Ind. Hyg.
Assoc. J. 1976, 37, 205–216. [CrossRef] [PubMed]
43. Van Osdell, D.W.; Owen, M.K.; Jaffe, L.B.; Sparks, L.E. VOC Removal at low contaminant concentrations
using granular activated carbon. J. Air Waste Manag. Assoc. 1996, 46, 883–890. [CrossRef]
44. Wood, G.O. Estimating service lives of organic vapor cartridges. Am. Ind. Hyg. Assoc. J. 1994, 55, 11–15.
[CrossRef]
45. Švábová, M.; Weishauptová, Z.; Přibyl, O. Water vapour adsorption on coal. Fuel 2011, 90, 1892–1899.
[CrossRef]
46. Sullivan, P.D.; Stone, B.R.; Hashisho, Z.; Rood, M.J. Water adsorption with hysteresis effect onto microporous
activated carbon fabrics. Adsorption 2007, 13, 173–189. [CrossRef]
47. Gauden, P.A.; Terzyk, A.P.; Rychlicki, G.; Furmaniak, S.; Wesołowski, R.P.; Wiśniewski, M.; Kowalczyk, P.
Mechanisms of water adsorption on carbons. In Water Encyclopedia; Wiley: Hoboken, NJ, USA, 2005.
48. Tanaka, S.; Tsuda, Y.; Kitamura, S.; Shimada, M.; Arito, H.; Seki, Y. A simple method for detecting
breakthroughs in used chemical cartridges. Am. Ind. Hyg. Assoc. J. 2010, 62, 168–171. [CrossRef]

© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).