Ann. Occup. Hyg., Vol. 57, No. 6, pp.

728739, 2013
The Author 2013. Published by Oxford University Press
on behalf of the British Occupational Hygiene Society
doi:10.1093/annhyg/mes102

Ultrafine Particle Characteristics in a Rubber

Manufacturing Factory
BOOWOOKKIM1, JONG SEONGLEE1, BYUNG-SOONCHOI1,
SO-YOUNGPARK1, JIN-HAYOON1 and HYUNWOOKKIM2*
1
Occupational Lung Diseases Institute, Korea Workers Compensation and Welfare Service, Ansan,
South Korea; 2Department of Preventive Medicine, Catholic University of Korea, Seoul,
South Korea

Received 23 October 2011; in final form 6 December 2012; Advance Access publication 9 January 2013

Background:According to epidemiological research, exposure to rubber fumes can cause

various types of cancer and can lead to an increase in death rate because of cardiovascular
diseases.
Objectives: In this study, we have assessed the characteristics of ultrafine particles emitted
into the air during the manufacturing of rubber products using waste tires.
Methods: To assess the aerosol distribution of rubber fumes in the workplace from a prod-
uct during curing, we have performed particle number concentration mapping using a hand-
held condensation particle counter.
The particle number concentration of each process, count median diameter (CMD), and
nanoparticle ratio (<100nm) were determined using an electrical low-pressure impactor
(ELPI), and the surface area concentration was determined using a surface area monitor. The
shape and composition of the sampled rubber fumes were analyzed using an ELPItrans-
mission electron microscopy grid method. Further, the rubber fume mass concentration was
determined according to the Methods for the Determination of Hazardous Substances 47/2.
Results:The results of particle mapping show that the rubber fumes were distributed
throughout the air of the workplace. The concentration was the highest during the final pro-
cess of the work. The particle number concentration and the surface area concentration were
545 000cm3 and 640m2 cm3, respectively, approximately 10- and 4-fold higher than those in
the outdoor background.
During the final process, the CMD and the nanoparticle ratio were 26nm and 94%, respec-
tively. Most of the rubber fume particles had a compact shape because of the coagulation
between particles. The main components of these fumes were silicon and sulfur, and heavy met-
als such as zinc were detected in certain particles. The filter concentration of the rubber fumes
was 0.22mg m3, lower than the UK workplace exposure limit of 0.6mg m3.
Conclusions: Therefore, the rubber manufacturing process is a potentially dangerous pro-
cess that produces a high concentration of specific nanoparticles.

Keywords: cardiovascular diseases; nanoparticle; rubber fume; tire particle; ultrafine particle

Introduction Agency for Research on Cancer (IARC), with

The rubber industry is classified as an industry
with carcinogenic potential by the International
*Author to whom correspondence should be addressed.
Tel:+82-2-2258-7363; fax: +82-2-590-3820;
e-mail: hwkim@catholic.ac.kr
high risks of bladder cancer, lung cancer, leuke-
mia, and laryngeal cancer (IARC, 1982, 1987;
Kogevinas et al., 1998). Further, according to
some research, the death rate due to cardiovas-
cular diseases is also increased by certain activi-
ties of this industry (Holmberg et al., 1983;
Gustavsson etal., 1986).

728
 Rubber factory ultrafine particle characteristics 729
An epidemiological survey has been conducted
of the numerous recent deaths caused by ischemic
heart diseases (IHDs) among the workers of the
tire manufacturing industry in Korea. The workers
had a lower death rate from overall diseases com-
pared with the normal control [standardized mor-
tality ratio (SMR) = 84] but had a higher death
rate (SMR=141) from IHDs, although this was
not statistically significant. However, substances
known to be related to cardiovascular diseases,
such as carbon monoxide, carbon disulfide, sty-
rene, and butadiene, were not detected or detected
at a very low concentration; however, during cer-
tain curing processes, the concentration of rubber
fumes was higher than the UK workplace exposure
limit of 0.6mg m3 (Korea Occupational Safety
and Health Agency, KOSHA, unpublished data).
A rubber-producing plant is environmentally
characterized by the presence of rubber dust
and fumes generated by the production process
(Kromhout etal., 1994; Dost etal., 2000).
Rubber fumes are formed during the refining
and curing processes (HSE, 2005), but a chemi-
cal composition analysis is difficult, and there are
insufficient data on their adverse health effects.
For these reasons, rubber fumes are not classified
as a chemical ingredient, but the concentration
of rubber fumes itself is managed (Chaier, 2001).
Fumes from hot processes comprise particles hav-
ing a size in the range of ~1m to a minimum of
a few nanometers (Vincent etal., 2000).
Nanoparticles are very small and, once depos-
ited, can easily translocate to other tissues. It
is known that the smaller the particle size, the
easier is the translocation (Kreyling et al., 2002;
Oberdrster et al., 2002). Because smaller nan-
oparticles have higher toxicity per unit mass
(Donaldson et al., 2000), even if the component
is identical, nanoparticles are known to be inti-
mately associated with lung and cardiovascular
diseases (von Klot et al., 2002; Pekkanen et al.,
2002). Therefore, unlike the analysis of micropar-
ticles based on mass and chemical components,
nanoparticles are assessed in terms of various fac-
tors such as particle size, shape, coagulation sta-
tus, and surface area concentration (Maynard and
Kuempel, 2005).
Recently, concerns about the health effects of
manufactured and incidental nanoparticles pro-
duced in the traditional workplace have increased
(Wake et al., 2002; Elihn et al., 2008; Elihn and
Berg 2009; Heitbrink et al., 2009), but there has
been scant research on rubber fumes produced
during a rubber product manufacturing process.
Therefore, this study has been conducted to
assess the particle number concentration, sur-
face area concentration, particle size, shape, and
coagulation state of rubber fumes, a suspected
risk factor for cardiovascular diseases in the rub-
ber industry, produced during the rubber product
manufacturing process.
FACILITY DESCRIPTION
The workplace in which this study was conducted
manufactures rubber speed bumps and parking
stops. One cycle of the curing process requires
914kg of waste tire shreds to be fed into the
mold of the hot press and heating it at 150C for
1015min.
After curing, the product is moved to the work-
bench placed 1.5 m in front of the hot press, and
the final process of drilling holes with an electric
drill and cutting the ends of the product with scis-
sors is performed for 10min. After this, the product
is collected on the workplace floor, and after cool-
ing, the product is wrapped in plastic and shipped.
The product is very hot after curing, and a
considerable amount of rubber fumes is produced
from the product surface. Workers are exposed
to a high concentration of rubber fumes starting
from the time that the product is taken from the
mold to the end of the last task. While the product
is cured in the hot press, workers are involved in
other types of work such as cleaning the work-
place floor and cutting the used tires into pieces.
The dimensions of the workplace were 25 m
(width) 15 m (length) 15 m (height). Although
there were seven hot presses totally, only four were
in operation during this study. A local exhaust
was present on the upper part of the hot press,
and rubber fumes produced during the process
were appropriately collected and vented outside.
However, the rubber fumes produced during the
final stages on the workbench were dispersed into
the entire workplace. This workplace operated in
three shifts, including a night shift, but only two
hot presses were in operation during the night
shift, and output during the nightshift was half
as much as that during the day. There were four
workers in total, and none wore respirators.
METHODS
Particle number concentration mapping
To determine the aerosol distribution of rub-
ber fumes in the workplace, we conducted a
730 B. Kim etal.
mapping analysis based on the particle num-
ber concentration, using methods similar to
that used by Peters etal. (2006). The difference
between our method and that used by Peters
etal. is that they used both a condensation par-
ticle counter (CPC, model 3007, 101000nm,
upper limit 100 000cm3) and an optical particle
counter to assess the particle number concen-
tration of particles <300nm and the respir-
able mass concentration, whereas we assessed
the particle number concentration using only
CPC (model 8525, 201000nm, upper limit 500
000cm3). Particle number concentration was
measured at 24 points, at 1min per point, with a
CPC (model 8525, TSI Inc., MN, USA; Fig.1),
which counted particles with a diameter rang-
ing from 0.02 to 1.0m. Amap was created by
the filled contour function of SigmaPlot version
10.0 (SSI Inc., CA, USA) on the basis of the
means of the measured values after such meas-
urements were performed five times. Five sepa-
rate measurements were conducted at intervals
of 30min in the morning; each measurement
took ~40min. Furthermore, as outlined by Park
etal. (2010), the coefficient of variability (CV)
of five repetitions of the mapping was calcu-
lated from the geometric mean and the geomet-
ric standard deviation (GSD) of each mapping
measurement.
Particle sizing, number concentration, and
surface area concentration
We used the electrical low-pressure impactor
(ELPI, Dekati Ltd, Finland) apparatus, which
determines the particle number concentration of
particles between 7nm and 10m present in the
air, divided into 12 stages. The data were saved
every second. To minimize the particle-bouncing
phenomenon, greased aluminum filters were used
as the impaction plate of each stage. The results
from the ELPI measurements were transferred
to a spreadsheet, and the count median diameter
(CMD) and the GSD were calculated (Hinds,
1999). We drew the cumulative distribution curve
according to the particle size and calculated the
ratio of particles <100nm. Using a nanoparticle
surface area monitor (NSAM; model Aerotrak
9000, TSI Inc., USA), the surface area concen-
trations of the alveolar-deposit particles between
10 and 1000nm were calculated every second.
The two apparatuses were placed on mobile
carts, and the sampling inlet was placed 1.5 m
from the ground.
ELPI with transmission electron microscopy
grid sampling
To analyze the samples by transmission electron
microscopy (TEM), along with real-time ELPI
measurements, a TEM grid was placed on each of
the stages of ELPI between Stage 1 and Stage 5
(Fig. 2). The TEM grid used was a lacey 400-
mesh copper grid with only an ultrathin carbon
film. The TEM grid was not preprocessed, and
the TEM analysis was conducted immediately.
Overlapping particles from oversampling were
excluded from the analysis. To determine the valid-
ity of this study, a pilot study of NaCl nanopar-
ticles produced by the atomization technique was
conducted in the laboratory. Brich et al. (2011)
used a method similar to ours, where TEM grids
placed on a Sioutas cascade impactor were used
for analyzing the shape of aerosol carbon fibers.
Particle shape and chemical composition
analysis
We selected a total of 193 individual particles,
evenly distributed from each of the stages of
ELPI, and analyzed them by using TEM. The
shape of the particles was classified as compact
or fractal. The composition of the particles was
analyzed using energy-dispersive X-ray analy-
sis (EDX). The electron microscope used in this
study was Hitachi H-7650, using a voltage of
100kV, with an EDX spot of 60100nm, and the
data acquisition was carried out for 100 s.
Rubber fume filter sampling
The rubber fume filter concentration was measured
for approximately 6h, according to the Health and
Safety Executive methods for the determination
of hazardous substances (MDHS) 47/2 method
(HSE, 1999). The personal samples were collected
in the breathing zone using SKC XR5000 pumps
at a flow rate of 2.0 l min-1 for the four workers.
The area samples were taken from a mobile cart
placed 1.5 m in front of the workbench.
Sampling day and sampling location
The concentrations were measured for two days.
On the first day, the mapping analysis, the rub-
ber fume filter analysis, and the ELPI and NSAM
analyses in front of the workbench were carried
out (Fig.1). On the second day, the sampling loca-
tion was changed to spots inside the workplace,
air outside the workplace, and air of the resi-
dential area approximately 15 km away from the
workplace (Fig.1).
 Rubber factory ultrafine particle characteristics 731

Fig.1. Facility layout of the rubber curing department. The width of the facility is 25 m, and the depth is 15 m.
indicates electrical low pressure impactor (ELPI) and nanoparticle surface area monitor (NSAM) measurement points:
(a) in front of the worktable, (b) at the entrance door of the workplace, and (c) the outdoor background sampling point.
The + symbol indicates mapping sampling point.

Fig.2. (a) Carbon-coated transmission electron microscope (TEM) grid; (b) TEM grid on the aluminum filter assembly
of the electrical low pressure impactor (ELPI) stage.

The residential area far from the workplace workplace was within an industrial complex with
where the study was conducted may not be an a high concentration of air pollutants, we included
appropriate place to measure the background the residential area to compare the air of the resi-
concentration of nanoparticles. However, as the dential area with that of the industrial area.
732 B. Kim etal.
Fig.3. Map of mean particle number concentrations, measured by a condensation particle counter (CPC) (201000nm).
60475 particles cm3 were measured in the outdoor background. The + symbol indicates mapping point.
Data analysis
A distribution of the ELPI and the NSAM meas-
urement results on a log probability plot resulted
in linearity. Thus, the geometric mean concentra-
tion and the GSD were calculated after the loga-
rithmic conversion. To compare differences in
concentration depending on measurement points,
an analysis of variance and Scheffes multiple
comparisons were performed. An analysis of the
correlation between the particle number concen-
tration (ELPI) and the surface area concentration
(NSAM) was performed.
RESULTS
Particle number concentration mapping
Figure 3 shows the number concentration map
of the rubber manufacturing plant measured
using the CPC. The concentration was the
highest at a distance of 300 000cm3 near the
workbench. However, the particle number con-
centration of parts of the workplace was not
very different (240 000300 000cm3), except
near the entrance where the outside air entered.
This demonstrated the easily dispersible char-
acteristic of rubber fumes. The particle number
concentration outside the workplace was 60
475cm3, which is approximately a quarter of
the concentration inside the workplace. The CV
from the five repetitions of the mapping analysis
was28%.
Aerosol characterization bytask
The particle number concentration increased at
the final process of the work and decreased during
the waiting period (during curing process; Fig.4).
As time passed, the background concentration
increased because the rubber fumes that were
continuously produced from the rubber product
accumulated within the workplace and were not
appropriately vented outside.
The average particle number concentration
during the measurement was 184 000 cm3
(surface area concentration = 271 m2 cm3),
~1.7-fold higher than the average concentra-
tion during the waiting period (127 000cm3).
The 15-min peak concentration during the final
process was 545 000cm3 (surface area concen-
tration=640m2 cm3); the CMD was 26nm;
and the nanoparticle ratio was 95% (Table 1).
The average particle number concentration
during the lunch hour (when work was stopped)
was similar to the average concentration dur-
ing the waiting period because of the accumu-
lated rubber fumes, but it slowly decreased as
time passed to a level similar to the outdoor
concentration.
 Rubber factory ultrafine particle characteristics 733

Fig.4. Variation in particle number concentration levels over the sampling period, indicating that workers at the finishing
stages are exposed to high concentrations of rubber fumes. indicates peak concentration at the finishing stages of the
work; represents waiting period.

Table1. Particle number concentration, CMD, nanoparticle ratio, and surface area concentration by tasks.
Task Step ELPI SA
NC (GSD) CMD (GSD) Nanoparticle ratio
Final process A 215103 (1.3) 32 (2.0) 90 315
B 545103 (1.3) 26 (1.9) 95 640
Waiting period 131103 (1.8) 33 (2.1) 90 180
Lunch period 127103 (1.8) 31 (2.2) 92 173
Mean 184103 271
A, total final process; B, peak concentration for 15min, acquired from total final process; Lunch period, 12:3013:30h;
NC, geometric mean of number concentration, particles per cubic centimeter; GSD, geometric standard deviation;
CMD, count median diameter, nm; nanoparticle ratio, <100 nm, % SA, surface area concentration, micrometer squared
per cubic centimeter.

Aerosol characterization by sampling location
The particle number concentration of the air
was statistically significantly different depend-
ing on the sampling location. The concentration
was the highest inside the workplace, followed by
those at the workplace entrance, in the air out-
side the workplace, and in the residential area.
The particles inside the workplace were smaller
(CMD = 31 nm), and the nanoparticle ratio
(<100nm) was higher, at 90% (Table2).
The particle number concentration of the residen-
tial area was approximately half of the concentra-
tion in the air outside the workplace. This signified
that the air of the industrial complex contained a
considerable amount of nanoparticle pollutants.
Figure5 shows the particle number concentra-
tion measured by ELPI and the particle size dis-
tribution at the workplace entrance, showing an
intermittent detection of a high concentration of
particles suspected to be rubber fumes (there were
no other sources of particles). The particle size of
these peaks was mainly <30nm, and the particle
number concentration was very high. This result
indicated that the rubber fumes produced in the
rubber plant could be expelled outside, retaining
the nanoparticlesize.
734 B. Kim etal.

Table2. Particle number concentration, CMD, nanoparticle ratio, and surface area concentration by sampling places.
Place ELPI NSAM
NC (GSD) CMD (GSD) nanoparticle m2 cm3 G
3 ratio
# cm G
BG1 53103 (1.3) 3 51 (2.7) 72 146 (1.2) 3
At entrance 59103 (1.4) 2 48 (2.7) 72 174 (1.3) 2
At plant 264103 (1.9) 1 31 (2.1) 90 366 (1.5) 1
BG2 23103 (1.1) 4 35 (2.4) 84 39 (1.2) 4
<0.001 <0.001
ELPI, electrical low-pressure impactor; NSAM, nanoparticle surface area monitor; BG1, factory outside background;
BG2, residential area background; NC, geometric mean of particle number concentration; GSD, geometric standard
deviation; CMD, count median diameter; nanoparticle ratio, <100 nm, %; SA, surface area concentration; , comparison
of particle number concentration and surface area concentration by places, using a general linear model; G, grouping
codes from Scheffes multiple range tests for multiple comparisons.

Fig.5. Characteristics of rubber fumes in the factory entrance according to time passage: (a) particle number
concentration, (b) particle size distribution with particle number concentration; dn/dlogDp, normalized concentration.

Particle number concentration and surface area
concentration
The correlation coefficient of the particle num-
ber concentration and the surface area concen-
tration, which were derived from the ELPI and
NSAM measurements in the workplace, and that
of the outside atmosphere were 0.82 and 0.58,
respectively, which had statistical significance
(P < 0.001; Table 3). Therefore, the correlation
between the results obtained with the two instru-
ments was higher when they were used to meas-
ure rubber fume particles generated during the
working process than when they were used for
measurement in the outside atmosphere where
 Rubber factory ultrafine particle characteristics 735

Fig.6. Particle shape of rubber fumes collected by a TEM grid on the electrical low-pressure impactor (ELPI) stage: (a)
compact, (b) fractal-like.

Table3. Correlation coefficient between particle the remaining 8% being of fractal shape (Fig.6;
number concentration by particle sizes and surface area Table4). Larger particle sizes related to a slightly
concentration at the workplace and outdoors.
higher proportion of fractal-shaped particles
ELPI channel (nm) Workplace Outdoors (Table4). Analysis of their components with EDX
21 0.80 0.55 revealed that silicone and sulfur were characteris-
40 0.83 0.58 tically present (Fig.7a). In addition, some rubber
73 0.84 0.60 fumes contained heavy metal particles, such as
123 0.75 0.56 iron, manganese, and zinc (Fig.7b). Carbon was
203 0.48 0.50 present in both the rubber particles and the car-
317 0.20 0.43 bon film of thegrids.
485 0.14 0.47
764 0.11 0.10 Rubber fume filter concentration
1233 0.06 N/A The average concentration from the personal
1956 0.06 N/A samples of the workers were 0.22mg m3 (range :
3092 0.06 N/A 0.160.25mg m3), approximately one-third
6302 0.05 0.05 of the UK workplace exposure limit. The area
Total 0.82 0.58 sample concentration was 0.1mg m3, approxi-
ELPI channel, geometric mean of particle size of each mately half of the concentration in the personal
channel; Workplace, sampling for only 90min in front of samples.
workbench (10:0011:30h).

relatively larger (CMD = 51nm, <100nm, 72%)
particles existed. In summary, a stronger correla-
tion between the particle number concentration
and the surface area concentration is expected
in the presence of ultrafine particles at high con-
centrations. This result is consistent with that
reported by Heitbrink etal. (2009).
Analysis of particle shape and chemical
compositions
The morphological analysis of 193 particles
showed that 92% were of compact shape, with
Discussion
In this study, we performed particle mapping
and determined particle number concentration,
particle surface area concentration, CMD, and
nanoparticle ratio using real-time instruments
at a plant manufacturing rubber products from
used tires. Further, by using the ELPITEM grid
method that we developed, we assessed the par-
ticle number concentration and the shape of the
particles simultaneously. Furthermore, we deter-
mined the rubber fume filter concentration by
using the MDHS 47/2 method.
736 B. Kim etal.

Fig.7. Microanalysis of individual rubber fumes. (a) The co-presence of Si and S serves as a significant marker for
identifying rubber fumes. (b) Some rubber fumes contained heavy metal particles such as Zn and Fe. Cu is derived from
the Cu-400 mesh grid used to support the presence of particles, and Mo is derived from the TEM column.

Table4. Estimated volumetric percentage of particles CPC with a low upper limit was used or when
from the shape type of each ELPI stage. sampling for a short time in a location where
ELPI Particle shape, % nanoparticle concentration was very high. In this
Stage GM (nm) Compact Fractal like study, the upper limit of the CPC was 500 000cm
3
1 40 93 7 ; we did not measure the concentrations directly
2 73 96 4 next to the workbench, and the 1-min average was
3 123 94 6 determined; the measurements did not exceed the
4 203 93 7 upper limit.
5 317 83 17
ELPITEM grid sampling
Mean 92 8
In the pilot study, minimal change was observed in
Totally, 193 particles were analyzed; GM, geometric mean
size of each stage. the NaCl concentration measured before and after
attachment of the TEM grid onto the greased alu-
minum filter (overall variation: +1%). Therefore,
Particle mapping the ELPI with TEM grid sampling method could
A study by Park et al. (2010) reported that the be also used to collect ultrafine particles for meas-
CV of repeated measurements of mapping from uring the particle number concentration and con-
a restaurant was 7080% and that from a die ducting TEM analysis. The NaCl concentration
casting facility was ~20%. The CV of repeated changed little even if the grid was attached prob-
measurements in this study was low, at 28%. This ably because the TEM grid was composed of Cu
was attributed to the fact that the rubber fumes and was coated with a carbon film. The high con-
remained inside the workplace. This finding is an ductivity of the two materials allows the electric
advantage for mapping and is consistent with the charge of the particles collected through the grid
results of the studies conducted by Peters et al. to flow into the ammeter of the ELPI. Moreover,
(2006), Heitbrink etal. (2007), Evans etal. (2008), neither overload nor buildup of NaCl particles
Liu and Hammond (2010), and Park etal. (2010). and rubber fume particles, which could cause a
On the other hand, the use of a diluter was sug- change in ELPI particle diameter distribution,
gested, as in the study by Peters etal. (2006), when was observed on the grid surface. However, as
 Rubber factory ultrafine particle characteristics 737
reported by Gulijk etal. (2003), if fractal-shaped
particles, such as those of diesel engine fumes, are
collected for a prolonged time, some overload and
buildup takes place in the filter; therefore, it can
become difficult to perform morphological analy-
sis using TEM. Thus, it is necessary to perform
further on-site feasibility studies on the ELPI with
the TEM grid sampling method in various work
environments.
Limitation of area sampling method in
nanoparticle assessment
Nanoparticle assessment instruments appropriate
for personal sampling are still rare or nonexistent.
Therefore, real-time instruments for nanoparticle
assessment were used for the area sampling. This
was a limitation with respect to the measurement
of the personal exposure concentration of the
nanoparticles.
In this study, the highest concentration of nan-
oparticles that the workers were exposed to was
observed when the product was taken from the
hot press and during the final process. However, as
the mobile cart with the real-time instruments was
bulky, we could carry out the measurements only
at distances of 3 and 1.5 m from the hot press and
the workbench where the final process was carried
out, respectively, indicating limited proximity. As
demonstrated by the traditional mass concentra-
tion analysis of a filter in this study, the personal
sampling and the area sampling concentrations
could have up to 2-fold differences.
Carbon black and silica nanoparticles as fillers
During the tire manufacturing process, various
other materials (including fillers, vulcanization
accelerators, and oil) are added to the raw material,
namely rubber. Nanoparticles such as carbon black
and silica are used as fillers (ICBA 2005; IARC
2010). During processes such as refining and cur-
ing, these additives are chemically coupled to the
rubber, forming a stable bound rubber, and carbon
black and silica no longer exist as pure nanoparti-
cles (Mark and Erman, 2005). Similarly, the rubber
fumes produced by the product after the curing pro-
cess are believed to be physically and chemically dif-
ferent from the pure nanoparticles of carbon black
or silica. The TEM/EDX analysis of the rubber
fumes in this study also supported our hypothesis,
showing that C, Si, and S were present together.
Rubber fumes are believed to be bound rub-
ber, similar to rubber tires, but we were unable to
determine this conclusively.
Background measurement
In many guidelines for assessing the workplace
nanoparticle concentration, methods such as
comparison of the nanoparticle concentrations
before and after the process or a comparison of
the concentrations in the workplace with those in
the air outside the workplace are recommended
for measuring the background concentration
(Kuhlbusch et al., 2011; Ramachandran et al.,
2011; Brouwer et al., 2012). In plants that are
operated round-the-clock, such as the plant where
this study was conducted, it is difficult to meas-
ure the background concentration of nanoparti-
cles before the process begins. Further, in plants
within an industrial complex, the outside air con-
tains a high concentration of nanoparticle pollut-
ants, making the background concentration high.
Therefore, unless the process produces a high con-
centration of nanoparticles, such as in this study,
it may be difficult to distinguish the background
concentration from the concentration attributed
to the manufacturing process. Therefore, it is
important to analyze the images of the particles
and their composition in parallel.
Particle shape and chemical composition
In the current study, most of the particles of rub-
ber fumes are present in compact shape. Because
the rubber fumes are generated at high tempera-
tures, the interparticle coagulation is completed
as sintering coagulation to produce a compact
shape. This result could be supported by a simula-
tor study of the ultrafine particle emissions from
pavementtire interface (Dahl etal., 2006).
In the EDX analysis, Si and S were character-
istically detected, and some rubber fume particles
contained heavy metals such as Zn, Fe, and Mn. In
this study, Si, S, and Zn resulted from the additive
components used in the tire manufacturing process,
which is a typical EDX pattern of dust arising from
tire abrasion (Gualtieri etal., 2005). Therefore, Si,
S, and Zn can be used as markers for tire particles
in the TEM analysis. The prevalence of heavy metal
particles in tires was also evident in a study in which
two types of tire dust particles, both of which could
be inhaled, were instilled into the lungs of rats. The
concentrated exposure induced cardiac oxidative
stress (Gottipolu etal., 2008).
Occupational exposure to particulates and
cardiovascular disease
This studys findings on occupational exposure
to ultrafine particles are consistent with those of
738 B. Kim etal.
other studies on the relativity between the expo-
sure levels to background particles and ultrafine
particle at the workplace ( Elihn etal., 2008; Evans
etal., 2008; Baxter etal., 2010; Fang etal., 2010).
Arecent mortality study examining the effects of
occupational exposure to particulates showed an
increased rate of IHD (Toren et al., 2007; Fang
etal., 2010). Therefore, results of previous stud-
ies and this study support the hypothesis that
exposure to ultrafine particles at high levels may
represent a potential additional risk for IHD in
employees of the rubber manufacturing industry.
Our findings demonstrate that ultrafine particles
are generated in high concentrations, increasing
relative to background values, during the finish-
ing stages after the rubber curing process. These
results further emphasize the necessity for sub-
sequent studies to examine whether the exposure
to ultrafine particles is a risk factor for IHD in
employees of the rubber manufacturing industry.
Limitation
As this study was conducted at one workplace for
two days, it is difficult to generalize the result of
this study to all rubber manufacturing processes.
Moreover, as the measurements were taken in
autumn (October 2010) all of the windows of
the workplace were open. However, in the win-
ter, when all windows are closed, the rubber fume
concentration would be even higher. Factors that
affect the rubber fume concentration of the work-
place include ventilation conditions, air exchange
rate, and emission factors such as mold pressure
and temperature of the hot press (Kromhout
et al., 1994), the type of rubber product, curing
frequency, and the duration of the final process
(Dost etal., 2000).
Conclusions
In this study of the workplace rubber fume con-
centration of a rubber product manufacturing
plant, the filter concentration measured by using
the MDHS method was lower than UK work-
place exposure limit, but the nanoparticle con-
centration measured by real-time instruments was
high. From the mapping analysis, we confirmed
that rubber fumes were dispersed throughout
the workplace. During the final process, workers
were exposed to the highest concentration of rub-
ber fumes. The peak 15-min concentration dur-
ing the final process was 545 000cm3; surface
area concentration was 640 m2 cm3; the CMD
was 26nm; and the nanoparticle ratio was 94%.
Most of the rubber fumes were compact in shape.
The composition analysis mainly detected Si and
S; heavy metals such as Zn, Fe, and Mn were
detected in some particles. Therefore, the rubber
manufacturing process is a potentially dangerous
process that produces a high concentration of
specipic nanoparticles.
References
Baxter CS, Ross CS, Fabian T etal. (2010) Ultrafine par-
ticle exposure during fire suppressionis it an important
contributory factor for coronary heart disease in fire-
fighters? J Occup Environ Med; 52: 7916.
Birch ME, Ku BK, Evans DE etal. (2011) Exposure and
emissions monitoring during carbon nanofiber produc-
tionPart I: elemental carbon and iron-soot aerosols.
Ann Occup Hyg; 55: 101626.
Brouwer D, Berges M, Virji MA etal. (2012) Harmonization
of measurement strategies for exposure to manufactured
nano-objects; report of a workshop. Ann Occup Hyg; 56:
19.
Chaier N. (2001) Health and safety in the rubber industry.
Rapra review reports: Expert overviews covering the sci-
ence and technology of rubber and plastics. Vol. 12 (No.
6). Shawbury, UK: Rapra Technology Ltd.
Dahl A, Gharibi A, Swietlicki E. et al. (2006) Traffic-
generated emissions of ultrafine particles from pave-
ment-tire interface. Atmos Environ; 40: 131423.
Donaldson K, Stone V, Gilmour PS. etal. (2000) Ultrafine
particles: mechanisms of lung injury. Phil Trans R Soc
Lond A; 358: 27419.
Dost AA, Redman D, Cox G. (2000) Exposure to rub-
ber fume and rubber process dust in the general rubber
goods, tyre manufacturing and retread industries. Ann
Occup Hyg; 44: 32942.
Elihn K, Berg P. (2009) Ultrafine particle characteristics
in the seven industrial plants. Ann Occup Hyg; 53:
47584.
Elihn K, Ulvestad B, Hetland S etal. (2008) Exposure to
ultrafine particles in asphalt work. J Occup Environ Hyg;
5: 7719.
Evans DE, Heitbrink WA, Slavin TJ etal. (2008) Ultrafine
and respirable particles in an automotive grey iron
foundry. Ann Occup Hyg; 52: 921.
Evans DE, Ku BK, Birch ME etal. (2010) Aerosol moni-
toring during carbon nanofiber production: mobile
direct-reading sampling. Ann Occup Hyg; 54: 51431.
Fang SC, Cassidy A, Christiani DC. (2010) A systematic
review of occupational exposure to particulate matter
and cardiovascular disease. Int J Environ Res Public
Health; 7: 1773806.
Gottipolu RR, Landa ER, Schladweiler MC et al. (2008)
Cardiopulmonary responses of intratracheally instilled
tire particles and constituent metal components. Inhal
Toxicol; 20: 47384.
Gualtieri M, Andrioletti M, Mantecca P et al. (2005)
Impact of tire debris on in vitro and in vivo systems. Part
Fibre Toxicol; 2: 1.
Gulijk CV, Marijnissen JCM, Makkee M. etal. (2003) Oil-
soaked sintered impactors for the ELPI in diesel particu-
late measurements. J Aerosol Sci; 34: 63540.
 Rubber factory ultrafine particle characteristics 739
Gustavsson P, Hogstedt C, Holmberg B. (1986) Mortality
and incidence of cancer among Swedish rubber workers,
19521981. Scand J Work Environ Health; 12: 53844.
Heitbrink WA, Evans DE, Ku BK etal. (2009) Relationships
among particle number, surface area, and respirable mass
concentrations in automotive engine manufacturing. J
Occup Environ Hyg; 6: 1931.
Heitbrink WA, Evans DE, Peters TM et al. (2007)
Characterization and mapping of very fine particles in an
engine machining and assembly facility. J Occup Environ
Hyg; 4: 34151.
Hinds W. (1999) Aerosol technology: properties, behavior,
and measurement of airborne particles. 2nd edn. New
York, John Wiley and Sons, Inc. ISBN 0471194107.
Holmberg B, Westerholm P, Maasing R et al. (1983)
Retrospective cohort study of two plants in the Swedish
rubber industry. Scand J Work Environ Health; 9
(Suppl.2): 5968.
HSE. (1999) Determination of hazardous substances.
Health and Safety Laboratory. MDHS 47/2. London,
UK: Health and Safety Executive Publication. ISBN
0717624196.
HSE. (2005) EH40: workplace exposure limits. HSE pub-
lication EH40/2005. London, UK: HSE Books. ISBN
0717629775.
IARC. (1982) Monographs on the evaluation of carcino-
genic risk of chemicals to humans. Vol. 28. The rub-
ber industry. Lyon, France: International Agency for
Research on Cancer.
IARC. (1987) Monographs on the evaluation of carcino-
genic risk of chemicals to humans. Overall evaluation of
carcinogenicity: an updating of IARC monographs. Vol.
142 (Suppl.7). Lyon, France: International Agency for
Research on Cancer.
IARC. (2010) Monographs on the evaluation of carcino-
genic risk of chemicals to humans. Vol. 93. Carbon black,
titanium dioxide, and talc. Lyon, France: International
Agency for Research on Cancer.
ICBA. (2005) International carbon black association.
Available at www.carbon-black.org
Kogevinas M, Sala M, Boffetta P etal. (1998) Cancer risk
in the rubber industry: a review of the recent epidemio-
logical evidence. Occup Environ Med; 55: 112.
KOSHA. (2008) Report on the epidemiologic investiga-
tion in the Hankook Tire Company. Incheon, Korea:
Occupational Safety and Health Research Institute,
Korea Occupational Safety and Health Agency. In
Korean. (unpublished).
Kreyling WG, Semmler M, Erbe F etal. (2002) Translocation
of ultrafine insoluble iridium particles from lung
epithelium to extrapulmonary organs is size dependent but
very low. J Toxicol Environ Health Part A; 65: 151330.
Kromhout H, Swuste P, Boleij JS. (1994) Empirical model-
ling of chemical exposure in the rubber-manufacturing
industry. Ann Occup Hyg; 38: 322.
Kuhlbusch TA, Asbach C, Fissan H et al. (2011)
Nanoparticle exposure at nanotechnology workplaces: a
review. Part Fibre Toxicol; 8: 22.
Liu S, Hammond SK. (2010) Mapping particulate matter
at the body weld department in an automobile assembly
plant. J Occup Environ Hyg; 7: 593604.
Mark JE, Erman B. (2005) Science and technology of
rubber. 3rd edn. Academic Press, California. ISBN
0124647863.
Maynard AD, Kuempel ED. (2005) Airborne nanostruc-
tured particles and occupational health. J Nanoparticle
Res; 7: 587614.
Oberdrster G, Sharp Z, Atudorei V et al. (2002)
Extrapulmonary translocation of ultrafine carbon parti-
cles following whole-body inhalation exposure of rats. J
Toxicol Environ Health Part A; 65: 153143.
Park JY, Ramachandran G, Raynor PC et al. (2010)
Determination of particle concentration rankings by
spatial mapping of particle surface area, number, and
mass concentrations in a restaurant and a die casting
plant. J Occup Environ Hyg; 7: 46676.
Pekkanen J, Peters A, Hoek G et al. (2002) Particulate
air pollution and risk of ST-segment depression dur-
ing repeated submaximal exercise tests among subjects
with coronary heart disease: the Exposure and Risk
Assessment for Fine and Ultrafine Particles in Ambient
Air (ULTRA) study. Circulation; 106: 9338.
Peters TM, Heitbrink WA, Evans DE et al. (2006) The
mapping of fine and ultrafine particle concentrations in
an engine machining and assembly facility. Ann Occup
Hyg; 50: 24957.
Ramachandran G, Ostraat M, Evans DE et al. (2011) A
strategy for assessing workplace exposures to nanomate-
rials. J Occup Environ Hyg; 8: 67385.
Toren K, Bergdahl IA, Nilsson T etal. (2007) Occupational
exposure to particulate air pollution and mortality due
to ischaemic heart disease and cerebrovascular disease.
Occup Environ Med; 64: 5159.
Vincent JH, Clement CF. (2000) Ultrafine particles in work-
place atmospheres. Phil Trans R Soc Lond A; 358: 267382.
von Klot S, Wlke G, Tuch T etal. (2002) Increased asthma
medication use in association with ambient fine and
ultrafine particles. Eur Respir J; 20: 691702.
Wake D, Mark D, Northage C. (2002) Ultrafine aerosol in
the workplace. Ann Occup Hyg; 46 (Suppl.): 2358.