Advances in Chemical Engi neering and Science , 2011, 1, 72-76
doi:10.4236/aces.2011.12012 Published Online April 2011 (http://www.scirp.org/journal/aces)
Kinetics and Modeling of H2S Removal in a Novel Biofilter
Zarook M. Shareefdeen*, Wasim Ahmed, Ahmed Aidan
Department of Chemical Engineering, American University of Sharjah, Sharjah, UAE
Received March 13, 2011; revised March 24, 2011; accepted April 2, 2011
Abstract
Biofiltration has become a widely accepted technology for the removal of hydrogen sulfide (H2S) which is
one of the major odor causing gases present in the air streams of municipal wastewater treatment facilities. In
addition to odorous nature, H2S is toxic and corrosive. In this study, a biofilter which uses a novel media was
employed in a pumping station which is closely located at the University City, Sharjah, UAE. The H2S re-
moval performance data were collected and subsequently used in the determination of kinetics and modeling
of H2S. The data were best represented by a first order biofilter model. Based on the first order kinetic con-
stant, a correlation is developed to predict concentrations at the biofilter outlet. Based on the predicted outlet
concentrations and dispersion (gaussian and US-EPA AERMOD) models, a study on H2S dispersion is con-
ducted. The dispersion study confirmed a biofilter installation at the pumping station site would significantly
reduce H2S levels in the University community and would provide cleaner air.
Keywords: Biofilter Media, Hydrogen Sulphide, Kinetics, Modeling
1. Introduction
Hydrogen sulfide (H2S) is one of the major compounds
that are emitted from municipal industries [1]. Hydro-
gen sulfide is odorous and highly toxic. It is heavier
than air so it tends to accumulate in poorly ventilated
spaces. Exposure to lower level concentrations of this
gas can result in eye irritation, sore throat and cough,
shortness of breath, and fluid in the lungs. Long-term
low-level exposure may result in fatigue, loss of appe-
tite, headaches, irritability, poor memory, and dizziness.
Between 150 - 250 ppm levels, the olfactory nerve is
paralyzed after a few inhalations, and the sense of smell
disappears. Concentrations over 1000 ppm cause im-
mediate collapse with loss of breathing.
Biofiltration has become a widely accepted technolo-
gy for treating air streams containing odorous com-
pounds. There are three main variations of this technol-
ogy: biofilters, bio-scrubbers, and biotrickling filters [2].
Of these technologies, biofilter technology is the most
popular and widely used at municipal industries. In the
case of biofilters and bio-trickling filters, microorgan-
isms are immobilized on support materials or media.
Thus, biofilter packing media plays a major role for
many reasons such as: providing a higher surface area
for biofilm growth, low pressure drop, long term physi-
cal stability, and good moisture retention.
Most biofiltration processes are aerobic, employing
heterotrophic microbes which are effective in removing
H2S for a wide range of potential applications with re-
spect to pH 4.0 - 8.0 [3]. The pH in a biofilter may also
change during operation. In some cases, the pH has to be
regulated using buffer substances mixed with the pack-
ing media or using alkaline or acid solutions. Tempera-
ture control is also very essential. The temperature range
for biofilter performance is about 15℃ - 40℃. Recent
literature [4-6] on biofiltration of H2S emphasizes that
there is a growing need for development of this technol-
ogy.
This work briefly reviews methods of a novel bio-
filter media preparation, pilot biofilter column set-up
and H2S removal performance data collection at a
pumping station located in the University City, Sharjah,
UAE. The main objectives of this work are to deter-
mine the kinetics of H2S removal using the perfor-
mance data [7], modeling of a biofilter which is packed
with the novel media and to study the dispersion ef-
fects of H2S in the vicinity of the University commu-
nity.
2. Experiments: Media Preparation, Pilot
Biofilter and Performance Data
In our recent work [7], we described in details the experi-

Z. M. SHAREEFDEEN ET AL. 73
mental procedure used in the development of a novel bio-
filter media, the pilot biofilter unit set-up and H2S perfor-
mance data collection. Since the experimental data are
used in the determination of kinetics and modeling, a brief
description on the novel media preparation and experi-
mental set-up is given below.
Using a specific material that is used in building con-
struction, hollow cylindrical particles were made as me-
dia base materials and subsequently coated with nutrients
and microbes. Several sets of media samples with dif-
ferent compositions were prepared and analyzed. The
mixing ratios of the nutrients were adjusted accordingly
so that the desired amount of nutrients was coated. A
mobile pilot biofilter unit (as shown in Figure 1) was
constructed, packed with the novel media and installed at
a pumping station for field data collection. The unit con-
sisted of the followings: (a) biofilter column, (b) humi-
difier, (c) diaphragm pump, (d) gas compressor, (e) ro-
tameter, (f) OdaLog (App-Tek International Pty Ltd,
Australia), OdaLog instrument is specifically designed
for the wastewater industry to measure H2S from pump-
ing stations and other operations (g) trolley and (h) ma-
nometers.
Hydrogen sulphide (H2S) performance data were
collected at different empty bed residence time (EBRT),
which is defined as the ratio of the volume of media to
the volumetric air flow rate. A detailed H2S perfor-
mance data for all EBRTs tested were presented in our
previous work [7]. Figure 2 shows a sample data of
H2S performance data at 30 second EBRT.
3. Theory: Biofilter Models
In a biofilter, the pollutant in the gas phase is first trans-
ferred to the biofilm by diffusion and then biodegraded
along the depth of the biofilm which is formed on the
media particles. Hence, two processes affect the removal
of pollutants: diffusion and reaction. For a zero order
reaction assumption, one of these processes limits the
overall removal. If the rate of diffusion is slower than the
rate of reaction, the removal process will be limited by
diffusion. Similarly, if the rate of reaction is slower than
the rate of diffusion, the process would be reaction li-
mited. The three biofilter models based on these concepts
are known as “Ottengraf and van den Oever” models
which are widely used by researchers, engineers and en-
vironmental professionals in describing performance data
and designing biofilter systems [8].
Case 1. Zero Order: Reaction Limited

outin0 EBRTCCk (1)
Where, Cin: inlet concentration of pollutant in air
3
(kg m)

, Cout: outlet concentration of pollutant in air
Figure 1. Mobile pilot-biofilter unit.
Figure 2. H2S data at 30 second-EBRT.
3
(kg m)

, k0 = zero order reaction rate constant
(31 kg ms



) defined as *
VS
X
AY

 ; μ*:specific growth
rate (s–1), Xv: biofilm density (3 kg m ), AS: biofilm sur-
face area per unit volume of biofilter (m–1), δ: biofilm
depth (m), Y: yield coefficient which is equal to the
amount of biomass produced/substrate consumed,
EBRT
g
H
u

; H: height of biofilter column (m) and ug:
velocity of air (ms–1).
Case 2. Zero order: Diffusion limited
In this case, it is assumed that the diffusion limits the
overall removal in the biofilm. The diffusion limited
model is based on the idea that the pollutant reaches its
maximum biodegradation in the biofilm at a depth that is
less than the actual biofilm thickness, implying that the
biofilm is not fully active [8]. The equation for the gas
phase concentration for this case is given by:
2
out in1
in
EBRT
1CC C






(2)

Z. M. SHAREEFDEEN ET AL. 74
Where,

2
0,
12
VHSW
S
kf XD
Am


;

V
f
X: ratio of
diffusivity of a compound in the biofilm to that in water,
2,
H
SW
D: diffusivity of H2S in water (21 ms

), and m:
dimensionless Henry’s constant of the pollutant.
Case 3. First order Reaction
For a steady state operation and first order reaction
assumption, the model equation is given as follows;

out in1
exp EBRTCC k (3)
Where, k1= first order reaction rate constant (s-1) defined
as

2
*
,VVHSW
SXfXD
A
mKY
  tanh

2

,

*
2
2
,
;
V
VHSW
X
KYf XD

  and K: Monod kinetics con-
stant 3
(kg m)

 [2].
4. Results and Discussion
To determine the kinetics and the best theoretical model
that fits the experimental data, concentrations were av-
eraged at each of the four EBRTs tested (Table 1). The
average concentrations (Cin and Cout) and EBRT data
were used to find the kinetics of H2S removal in the bio-
filter.
Equations (1), (2) and (3) were first re-arranged to ob-
tain linear plots. A best fit line was drawn through the
points for each case. Figures 3(a)-3(c) show the plots for
all the three cases. From the correlation coefficients and
Figure 3(c), it is clear that the data of novel biofilter
media follows a first order kinetics. Thus, Equation (3)
based on the fitted parameter now can be written as:

outin exp0.055*EBRTCC (4)
The removal efficiencies that were determined from the
model Equation (4) were compared with the removal
efficiencies calculated from the experimental data, as
shown in Figure 4. It can be seen that the values of re-
moval efficiency predicted by the model are in good
agreement with the experimental data.
In most industrial countries, air pollution control laws
require the prediction of pollutant dispersion in ambient
Table 1. Average concentrations at each EBRT.
EBRT (s) Cin (ppm) Cout (ppm)
20 3.33 1.00
30 9.84 1.06
45 9.17 0.981
60 13.7 0.597
(a)
(b)
(c)
Figure 3. (a) zero-order reaction limited (b) zero-order dif-
fusion limited and (c) first order kinetics.
air resulting from industrial emissions [9]. There are
several atmospheric dispersion models (i.e. AERMOD,
ISC-PRIME, ISCST3) that are used in the prediction of

Z. M. SHAREEFDEEN ET AL. 75
Figure 4. Comparison between the model and the experi-
mental data.
pollutants in the ambient atmosphere [10]. In this work,
downwind concentration calculations were made to ob-
serve the effect of H2S dispersion from the pumping sta-
tion located near the University City. First, a simple
Gaussian model [9] was used for two conditions: (a) the
dispersion of H2S from the stack without the use of a
biofilter and (b) the dispersion of H2S with the use of a
biofilter. For both cases, it was assumed that the average
wind speed of 3 m/s and the stability class of moderate
atmosphere. For case A, it was assumed that the stack
gas (without biofilter) emits H2S concentration at 100
ppm. In case B, it was assumed that the biofilter outlet
emits H2S concentration at 1.5 ppm (with 98.5% removal
efficiency). Based on the experimental data and the cor-
relation (4), it is clear that a biofilter packed with the
novel media can remove 98.5% H2S. Table 2 shows the
measured and assumed parameters for the stack. The
coordinates (x, y, z) were obtained by selecting seven
specific locations around the university and finding their
approximate distances from the stack using a GPS. The
point source (stack) was used as the reference point. Ta-
ble 3 shows the coordinates for each location and the
corresponding dispersion coefficients.
Downwind concentrations for each case were calcu-
lated from the Gaussian plume equation, using the values
presented in Tables 2 and 3. Figure 5 shows the com-
parison of downwind concentrations at each location
between case A and case B. Comparing the results, it can
be seen that the predicted downwind H2S concentrations
are substantially lower in all 7 locations. The highest
possible downwind concentration would be 0.0023 ppb
with the use of a biofilter. Using the AERMOD model
and with the available meteorological data base, the cal-
culations were repeated for the same point source and the
results are presented in the Figures 6(a)-6(b).
Table 2. Stack specifications.
Diameter 0.13 m
Height 3.24 m
Average plume velocity 3.56 ms–1
Wind speed 3 ms–1
H2S concentration (case A) 100 ppm
H2S concentration (case B) 1.5 ppm
Table 3. Coordinates and dispersion coefficients (σy and σz )
for each location.
#1 #2 #3 #4 #5 #6 #7
x(m) 462 467 395 397 245 614950
y(m) 140 23 32 34 247 51942
z(m) 3 3 10 10 0 0 0
σy (m) 78 79 67 68 44 101149
σz (m) 47 48 40 40 24 64 104
5. Conclusion
In conclusion, a kinetic study of H2S removal for a
novel biofilter media was performed. The data were
best represented by a first order model. The kinetic
constant was then used in developing a correlation


outin exp0.055* EBRT.CC This correlation can be
used to predict outlet concentration and sizing a biofilter
which is packed with the novel media. When compared
with the predicted concentration using the correlation,
experimental data were in good agreement with the
model. A basic gaussian dispersion and US-EPA
AER-MOD models were subsequently used “with” and
“without biofilters” to determine the effect of dispersion
in the University City. The H2S dispersion data shows a
biofilter installation at the pumping station would signif-
icantly reduce H2S levels and would give cleaner air
Figure 5. Comparison of downwind concentrations.

Z. M. SHAREEFDEEN ET AL. 76
(a)
(b)
Figure 6. AERMOD model predicted H2S concentration
profiles in (μg/m3) (a) without a biofilter (top) and (b) with
a biofilter (bottom).
to the University community. Based on this study, a local
environmental company in UAE has funded a research
project for an industrial application. The results of such
study will be our future contribution.
6. Acknowledgment
The authors acknowledge the support of American Uni-
versity of Sharjah, United Arab Emirates (UAE).
7. References
[1] Z. Shareefdeen, B. Herner, D. Webb and S. Wilson, “Hy-
drogen Sulphide (H2S) Removal in Synthetic Media Bio-
filters,” Environmental Progress, Vol. 22, No. 3, 2003,
pp. 207-213.
[2] Z. Shareefdeen and A. Singh, “Biotechnology for Odor
and Air Pollution Control,” Springer, Berlin, 2005.
doi:10.1007/b138434
[3] Y. Cohen, “Biofiltration—The Treatment of Fluids by
Microorganisms Immobilized into the Filter Bedding
Material: A Review,” Bioresource Technology Review,
Vol. 77, 2001, pp. 257-254.
doi:10.1016/S0960-8524(00)00074-2
[4] A. Martinez, et al., “Biofiltration of Wastewater Lift Sta-
tion Emissions: Evaluation,” Clean Technologies and
Environmental Policy, Vol. 10, No. 1, 2008, pp. 81-87.
doi:10.1007/s10098-007-0110-y
[5] D. Ramírez-Sáenz et al., “H2S and Volatile Fatty Acids
Elimination by Biofiltration: Clean-Up Process for Bio-
gas Potential Use,” Journal of Hazardous Materials, Vol.
163, No. 2-3, 2009, pp. 1272-1281.
10.1016/j.jhazmat.2008.07.129
[6] W. Lia, et al.. “Sulfide Removal by Simultaneous Auto-
trophic and Heterotrophic Desulfurization-Denitrification
Process,” Journal of Hazardous Materials, Vol. 162,
2009, pp. 848-853.
doi:10.1016/j.jhazmat.2008.05.108
[7] Z. Shareefdeen, A. Aidan, W. Ahmed, M. B. Khatri, M.
Islam, R. Lecheheb and F. Shams, “Hydrogen Sulphide
Removal Using a Novel Biofilter Media,” World Acade-
my of Science, Engineering and Technology, Vol. 62,
2010, pp. 13-16.
[8] M. A. Deshusses and Z. Shareefdeen, “Modeling of Bio-
filters for Odor and VOC Control Applications,” In: Z.
Shareefdeen and A. Singh, Ed., Biotechnology for Odor
and Air Pollution Control, Springer-Verlag, Berlin, 2005,
pp. 213-229.
doi:10.1007/3-540-27007-8_10
[9] N. De Nevers, “Air Pollution Control Engineering,”
McGraw-Hill, Boston, 2000.
[10] L. T. Jesse, L. T. Cristioane and A. J. Michel, “AER-
MOD View: Interface for the US EPA AERMOD Mod-
el,” 2009.