Industrial Health 2009, 47, 436–442 Field Report

Application of Traditional Cyclone with Spray

Scrubber to Remove Airborne Silica Particles
Emitted from Stone-crushing Factories
Abdulrahman BAHRAMI1*, Farshid GHORBANI1, Hossien MAHJUB2,
Farideh GOLBABEI3 and Mohsan ALIABADI1

1Department of Occupational Health, Faculty of Health and Center for Health Research, Hamadan University
of Medical Science, P.O.Box 4171-65175, Hamadan, Iran
2Department of Biostatistics & Epidemiology, Faculty of Health and Center for Health Research, Hamadan

University of Medical Science, Hamadan, Iran

3Department of Occupational Health, Faculty of Health, Tehran University of Medical Science, Tehran, Iran

Received May 19, 2007 and accepted March 6, 2009

Abstract: The traditional cyclone with spray scrubber was developed for the removal of air-
borne silica particles from local exhaust ventilation (LEV). The objective of this research is to
evaluate the efficiency of this process for removing silica particles in LEV. After designing and
installing a traditional cyclone and spray scrubber, air samples were obtained at the inlet and
outlet of the apparatus. The mass of each collected sample was determined gravimetrically using
EPA method. The efficiency of the cyclone with spray scrubber for the removal of dust parti-
cles from the LEV system was determined to be in the range of 92–99%. There was a high cor-
relation between the inlet concentration of dust particles and the efficiency of the apparatus. The
total pressure across the system was 772.17–1120.90 Pa. It was concluded that a traditional
cyclone with a spray scrubber can effectively remove a very high percentage of the incoming sil-
ica particles from an LEV. The total pressure drop across the current process is less than the
pressure drop across other treatment equipment, which means that our process can effectively
remove silica particles while using less electricity than other processes.

Key words: Cyclone, Spray scrubber, Silica, Stone crushing, Local exhaust ventilation

Introduction factories produce powder that contains 80% silica. Stone

Azandarian, an area in the Hamadan Province in the
west of Islamic Republic of Iran, is a center for produc-
ing silica powder for glass factories, sand blasting, fur-
nace linings, ceramics, refractories, and porcelain articles
throughout the country. There are a total of 29 factories
that crush quartz stones, and the process type used
depends on the desired purity of the product. One process
crushes the quartz and uses screening techniques to
acquire a product that is 98% silica. The other process
crushes the quartz and uses screening techniques to
acquire a product that is 80% silica. Ten of the factories
produce a product that contains 98% silica, and 19 of the
*To whom correspondence should be addressed.
cutting is done in the mining area, and the stones are taken
to the factory by trucks. The main processes used in each
factory are stone crushing, initial storage, grinding,
screening I, and screening II. In the first stage, which is
stone crushing, stone blocks are fed manually to the hop-
per from which they are dumped into a jaw crusher. The
crushed stones are carried by a conveyor belt to the rotary
grinder for grinding, and, finally, the ground stone is
transferred to screening I and screening II for sieving,
according to the size of quartz required. The particles
that are emitted from each of these processes contain a
very high amount of free silica. The workers are exposed
to the silica dust, and high rates of pneumoconiosis have
been reported in the Azandarian Area1, 2). The major
adverse effects of exposure to crystalline silica include
CONTROL OF AIRBORNE SILICA PARTICLES 437

silicosis, chronic bronchitis, certain connective tissue dis-

orders, and lung cancer3, 4). Also, dust generated at the
work place gets dispersed in the vicinity of the factory
and causes health risks to the public and degradation of
the environment.
Cyclones are among the oldest types of industrial par-
ticulate control equipment and air sampling devices.
Their simple design, low maintenance costs, and adapt-
ability to a wide range of particles sizes and flow rates
make cyclones one of the most widely used particle
removal devices5).
Spray scrubbers consist of an empty cylindrical or rec-
tangular chamber in which the gas stream is contacted
with liquid droplets generated by spray nozzles. A com-
mon design has the particle-laden gas flow upward Fig. 1. The picture of two traditional cyclone and one spray
scrubber in first and second branch at one of stone crush-
through a tank or successive tanks that have spray noz-
ing unit.
zles. Such scrubbers generally consume relatively small
A) Traditional cyclone in first branch, B) Spray scrubber in first
amounts of energy, because the pressure drops the scrub- branch, C) Traditional cyclone in second branch, D) Spray
bers are generally less than 2.5 cm (1 in) of water5, 6). scrubber in second branch.
Until now, many experimental investigations have been
performed either on industrial cyclone separators or on
scrubbers7–14). However, no information is available on
the application of the cyclone with a spray scrubber for
the removal of silica compounds from air.
According to Islamic Republic of Iran regulations, the where:
emissions of free silica particles in workplaces and to the Q = Flow rate (m3/h)
environment should not exceed the permissible limits. Dc = Diameter of cyclone (m)
We installed LEV devices equipped with conventional The diameters of the cyclones in the first and second
cyclones and spray scrubbers at the 29 factories between branches were determined to be 1.2 and 1.4 m, respec-
March 2004 and July 2006. The objective of this research tively. Figure 2 shows the cyclones designed for
is to evaluate the performance of a traditional cyclone branch 1. The main dimensions of the device, which are
with a spray scrubber for the removal of airborne silica illustrated in Fig. 2 and in part A of Fig. 3, will be referred
particles in LEV devices designed for stone crushers. to throughout the remainder of this text as follows: the
cylindrical body has a diameter Dc and a height h; the
Subjects and Methods conical section has a height Z and ends at the solids out-
let that has a diameter B; the gas entry port is rectangu-
Two LEV systems were conceived, designed, fabricat- lar, and its dimensions are height a and width b; the exit
ed, and installed for each industrial unit according to the duct, also called the vortex finder, has a diameter De and
recommendation given by America Conference starts at a distance S from the top of the cylindrical body;
Governmental Industrial Hygiene (ACGIH)15). The first the cyclone has a total height H.
branch of the LEV system was designed for a rotary The dimensions of the cyclones in the first and second
grinder and a hopper. The second branch was designed branches, based on the Lapple model, are shown in
for screenings I and II. Two traditional cyclones and one Table 1.
spray scrubber were designed for the removal of silica The spray scrubber was designed to allow a gas veloc-
particles in each branch. Figure 1 shows a photograph ity of 1.01–1.52 ms–1 inside the scrubber. The diameter
of the two systems in a stone crushing unit in the of the cyclone was calculated from the following equa-
Azanderian area. The flow rates in first and second tion:
branch were 19,800 and 27,000 m3/h, respectively. The
traditional cyclones were designed according to the
Lapple model16), and two were designed for each branch.
The diameter of each cyclone was calculated as shown in where:
the following equation, which is the basis of the Lapple Q = Flow rate (m3s–1)
model: d = Diameter of scrubber (m)
438
Fig. 2. The main dimensions of the traditional cyclone in
first branch.
Dc; Body Diameter, h; Body height, Z; Length of cone, B;
Diameter of Dust Outlet, De; Diameter of Gas Exit, S; Length
of Vortex Finder.
V = velocity inside scrubber (ms–1).
The diameters of the scrubbers in the first and second
branches were 2.1 and 2.7 m, respectively, as shown in
Fig. 3 part B for the first branch. The water consump-
tion in each scrubber was 37.85 liters per 0.472 m3s–1 gas
at STP.
Two type I stone crushers and two type II stone crush-
ers were selected to obtain isokinetic samples for the eval-
uation of the system. In both LEV branches, samples
were obtained before the cyclone, after the cyclone, and
in the stack. A total of 96 isokinetic samples was obtained
in one month. Four replications were conducted for each
sample, to obtain an average emission concentration. The
points and site position of the samples in each branch
were selected based on EPA recommendations17, 18).
Isokinetic sampling probes were used to measure the
representative particle concentrations upstream and down-
stream from the cyclones, and the particle concentrations
downstream from the scrubber. The inlet and outlet con-
centrations of dust particles were measured by using a
small pump to aspirate the sample into a probe that was
connected to the fiberglass filter holder. For each test,
the sampling time was two minutes. The mass of all sam-
ples collected was determined gravimetrically using EPA
method17). Filters were equilibrated in desiccators for a
minimum of 24 h prior to obtaining the final tare weight.
A BAHRAMI et al.
A pitot tube with a manometer was used to monitor sta-
tic pressure and velocity pressure at the sampling site.
The total collection efficiency for the cyclone and
scrubber was calculated on the basis of total mass col-
lected as shown in the following equation
where
EF = overall collection efficiency
W1 = total inlet loading (mg)
W2 = total emissions (mg)
The data were analyzed using the statistical package
SPSS Version 13.0 for Windows.
To evaluate the normality distribution of variables, a
one-sample Kolmogorov-Smirnov test was performed. To
determine the effect of pre-cyclone concentration on
cyclone efficiency, a quadratic regression model was
applied.
Results
The results of the analysis of isokinetic samples in both
LEV systems in industrial units are shown in Table 2.
The efficiency of the cyclone in removing dust particles
from the LEV system was 81–97 percent, and the effi-
ciency increased to 92–99 percent when the spray scrub-
ber was used (Table 2). There was a significant differ-
ence (p<0.05) between the concentration of particles at
the outlets of the cyclones in LEV1 and LEV2 the type
I stone crusher and the type II stone crusher. However,
no significant difference was observed between the con-
centration of particles at the inlet of the cyclones in the
two systems. The efficiency of the traditional cyclone
used in LEV2 was greater than the traditional cyclone
used in LEV1. The total pressure drop for the tradition-
al cyclone and spray scrubber combination was in the
range of 772.17–1,120.90 Pa.
Figure 4 shows the scatter diagrams between inlet con-
centration to the cyclone and the efficiency of the cyclone.
There was a significant correlation between the concen-
tration in the inlet to the cyclone and the efficiency of the
cyclone. As inlet concentration increased, the removal
efficiency also increased. The coefficient of the fitted
regression model between inlet concentration to the
cyclone and the efficiency of the cyclone was r2 = 0.85.
The equation for the fitted model is: Cyclone efficiency
= 75.08 + 6 × 10–3 (inlet concentration to cyclone)
–4 × 10–7 (inlet concentration to cyclone)2.
Discussion
This study was undertaken to evaluate the performance
Industrial Health 2009, 47, 436–442
CONTROL OF AIRBORNE SILICA PARTICLES 439

Fig. 3. Two traditional cyclone and spray scrubber (Elevation map) in first branch.
A; Cyclone (h; Body height, Z; Length of cone, a; Height of Inlet, b; Width of Inlet), B; Scrubber (d; Diameter of scrubber).

of a cyclone with a spray scrubber for the capture and
removal of silica particles in from the various processes
in the stone crushing industry. The results of this study
showed that using a traditional cyclone with a scrubber
has significant effect on the removal of particles from the
LEV system. The mean value of outlet concentration
from the stack for the type I stone crusher was approxi-
mately 44–130 mg m–3, and, for the type II stone crush-
er, this value was approximately 64–130 mg m–3 during
8–10 working hours, which is less than the one-year stan-
dard permissible limit of 120 mg m–3 over 24 h recom-
mended by the Iranian Environmental Protection
Agency19). The mean outlet concentration of the cyclone
in most stone crushing plants is more than 250 mgm–3,
but adding a spray scrubber reduces the concentration to
a level that is below the permissible limit.
In addition to the system’s high particulate removal
efficiency, its low pressure drop minimizes the system’s
use of electricity during the removal of the particulate
matter, which is of prime importance.
The total pressure drop for a traditional cyclone and
spray scrubber was 772.17–1,120.90 Pa for control of par-
ticles in stone crushing, which is less than pressure drop
than other equipment used to remove particles from air
streams. Other important equipment used to control dust
particles in the air are high efficiency cyclones, venturi
wet scrubbers, fabric filters, and electrostatic precipita-
tors5). If any of these techniques were designed for the
stone crushing industry, the pressure drop would be sig-
nificantly greater and the electricity cost would be high-
440 A BAHRAMI et al.

Table 1. The diamentions of cyclones in first and second branches

Dimension in first Dimension in second

Dimension based on Lapple model
branch (Dc=1.2) branch (Dc=1.4)

Body Diameter ( ) 1.2 1.4

Body height ( ) 2.4 2.8

Length of cone ( ) 2.4 2.8

Diameter of Dust Outlet ( ) 0.30 0.35

Height of Inlet ( ) 0.6 0.7

Width of Inlet ( ) 0.3 0.35

Diameter of Gas Exit ( ) 0.6 0.7

Length of Vortex Finder ( ) 0.6 0.7

Table 2. The mean of total concentration dust in different site of duct at system

Stone crushers Type I Stone crushers Type II

LEV1 LEV2 LEV1 LEV2

Pre to cyclone (mgm–3) 5,393 ± 986.2 5,210.75 ± 468.8 1,587.5 ± 239.3 1,299.0 ± 146.97
After cyclone (mgm–3) 389.5 ± 27.1 135.0 ± 14.7 301.5 ± 48.3 213.4 ± 30.00
φ1 (%) 92.65 97.37 80.7 83.6
Stack (After Scrubber) (mgm–3) 131.0 ± 30.5 43.7 ± 17.5 127.8 ± 51.3 64.3 ± 28.2
φ2(%) 66.4 67.9 58.5 70.0
φ3(%) 97.5 99.1 92.1 95.0

φ1 Efficiency of Cyclone, φ2 Efficiency of Scrubber, φ3 Efficiency of Cyclone and Scrubber.

The efficiency of the traditional cyclone used in the

Fig. 4. Correlation between pre-cyclone concentration of parti-
cles and efficiency of cyclone.
er than those of the traditional cyclone with spray scrub-
ber.
LEV2 system was greater than the efficiency of the tra-
ditional cyclone used in the LEV1 system, which may be
related to the dimensions of the cyclone. The ratio of the
height of the cylinder to the height of the cone for LEV2
and LEV1 was 1.17. The cyclone collection efficiency
was directly affected by the height of the cylinder and the
height of the cone of the cyclone9, 12). According to the
Lapple model, an increase in the height of the cyclone
causes an increase in the number of effective turns, which
creates more space for the particles to be collected in the
cyclone16).
In recent research20), we used X-ray diffraction (XRD)
to evaluate the respirable dust and quartz in stone crush-
ing factories after the installation of the LEV system. The
efficiency of the LEV system in controlling particles
inside industrial units was greater than 99%. The results
of personal sampling showed that the concentrations of
respirable dust that workers were exposed to in the
process, hopper, and driver areas were 1.90, 2.22, and

Industrial Health 2009, 47, 436–442

CONTROL OF AIRBORNE SILICA PARTICLES
1.41 times greater than the permissible exposure limit
(PEL) established by the Occupational Safety and Health
Administration (OSHA). The amount of free silica in
stone quartz contributes 85 to 97 percent of the particu-
late emissions that affect the health of workers.
The inlet and outlet concentrations and the sizes of the
particles in the cyclone were determined in isokinetic
sampling probes using a cascade impactor with 21-mm
fiberglass filters connected to a sampling pump21). The
results of measurement of particle size in the cyclone
showed that, for diameters of <1, 1–5, and 5–10 µm, the
cyclone’s collection efficiencies were 60, 88, and 90,
respectively.
There has been some research on the efficiency of tra-
ditional cyclone, but we could not find any research con-
cerned with the traditional cyclone combined with a spray
scrubber. According to the research conducted by Texas
A&M University (TAMU), the 2D2D cyclone (tradition-
al cyclone) is the most efficient collector of fine dust. In
another research project, Wang, et al. reported that the
2D2D cyclone had a high efficiency for both fine dust
and large trash10).
There was a significant correlation between the inlet
concentration of particles entering the cyclone and the col-
lection efficiency of the cyclone. As the inlet concentra-
tion increased, the collection efficiency also increased.
This can be related to the facts that more particles col-
lide with the walls of the cyclone22) and more particles
collide with each other inside the cyclone. When the par-
ticles collide with each other, their velocities are
decreased, and, therefore, they are more likely to settle in
the collection bin of the ncyclone.
In the present study, we were limited in our ability to
analyze particle sizes at the inlet and outlet of the cyclones
and at the outlet of the scrubbers because the particle size
counter was not available. This will be dealt with in later
research21).
Conclusions
It is concluded from this research that applying a tra-
ditional cyclone with a spray scrubber has a significant
effect on the removal of silica particles from LEV sys-
tems. Total pressure drop of the current process was less
than the pressure drops for other control equipment, which
is significant because this reduces the consumption of
electricity and reduces the associated operational costs.
Acknowledgements
The authors thank the Union of Industrial Units in the
Azandarian area for supporting this research.
441
References
1) Bahrami AR, Mahjub H (2003) Comparative study of
lung function in Iranian factory workers exposed to sil-
ica dust. East Mediterr Health J 3, 390–8.
2) Mohebbi I, AbdiRad I (2007) Secondary spontaneous
pneumothorax in rapidly progressive forms of silicosis;
characterization of pulmonary function measurements
and clinical patterns. Toxicol Ind Health 23, 125–32.
3) International Agency for Research on Cancer (1997)
IARC Monographs on the Evaluation of the
Carcinogenic Risk of Chemicals to Humans: silica and
some silicates. International Agency for Research on
Cancer, Lyon.
4) Parks CG, Conrad K, Cooper GS (1999) Occupational
exposure to crystalline silica and autoimmune disease.
Environ Health Perspect 107, 793–802.
5) Wang LK, Pereira NC, Hung YT (2004) Air Pollution
Control System, Human Press Inc, New Jersey.
6) Environmental Protection Agency U.S. (2002) Daniel
Mussatti, Paula Hemmer, Section 6 Particulate Matter
Controls, Chapter 2, Wet Scrubbers for Particulate
Matter, EPA/452/B-02-001.
7) Hesketh HE (1995) Predict particle collection in spray
towers. Chem Eng Prog 91, 98–100.
8) Vaaraslahti K, Laitinen A, Keskinen J (2002) Spray
charging of droplets in a wet scrubber. Air Waste Manag
Assoc 2, 175–80.
9) Rongbiaho X, Park SH, Lee KW (2001) Effect of cone
dimension on cyclone performance. J Aerosol Sci 32,
549–61.
10) Wang L, Parnell Jr CB, Shaw BW (2002) A study of
the cyclone fractional efficiency curves. Agricultural
Engineering International: the CIGR J Scientific
Research and Development. IV, 1–7.
11) Kim HT, Lee KW, Kuhman MR (2001) Exploratory
design modification for enhancing cyclone performance.
J Aerosol Sci 32, 1135–6.
12) Lozia Dl, Leith D (1989) Effect of cyclone dimension
on gas flow pattern and collection efficiency. Aerosol
Sci and Technol 10, 490–5.
13) Kim HT, Jung CH, Oh SN, Lee KW (2001) Particle
removal efficiency of gravitational wet scrubber con-
sidering diffusion, interception, and impaction. Environ
Eng Sci 18, 125–36.
14) Funk PA, Hughs SE, Holt GA (2000) Entrance veloci-
ty optimization for modified dust cyclones. J Cotton Sci
4, 178–82.
15) ACGIH (2001) Industrial ventilation, A manual of rec-
ommended practice. 24th Ed., American Conference of
Governmental Industrial Hygienists, Cincinnati.
16) Lapple C (1951) Processes use many collection types.
Chem Eng 58 144–51.
17) Clarke AG (1998) Industrial air pollution monitoring,
Chapman & Hall, London.
18) California Environmental Protection Agency (1997) Air
Resources board, Determination of particulate matter
442
from stationary sources, Method 5.
19) Iran Environmental Protection Agency, Regulation of
Protection Agency (1995) No 2435, Environmental
Protection Agency publication, Tehran.
20) Bahrami AR, Golbabai F, Mahjub H, Qorbani F,
Aliabadi M (2008) Determination of exposure to res-
pirable quartz in the stone crushing units at Azendarian-
West of Iran. Ind Health 46, 404–8.
A BAHRAMI et al.
21) Bahrami AR, Mahjub H, Qorbani F, Aliabadi M (2008),
Effects of velocity and particles load on efficiency of
cyclone in the stone crushing units at Azendarian Area
J Res Health Sci 8, 12–7.
22) Parnell CB (1996) Cyclone design for air pollution
abatement associated with agricultural operations.
Proceeding of the Beltwide Cotton production
Conferences. National Cotton Council, Nashville.
Industrial Health 2009, 47, 436–442