Article 3.pdf

Vol.:(0123456789)
1 3
Waste and Biomass Valorization
https://doi.org/10.1007/s12649-018-0226-9
ORIGINAL PAPER
Effect of Operating Conditions on Separation of H2S from Biogas Using
a Chemical Assisted PDMS Membrane Process
Ebrahim Tilahun1
· Erkan Sahinkaya2
· Bariş Çalli1
Received: 28 October 2017 / Accepted: 5 February 2018
© Springer Science+Business Media B.V., part of Springer Nature 2018
Abstract
Hydrogen sulfide
(H2S) is an undesirable impurity that has to be removed from biogas to avoid the corrosion of co-generation
units. In the present study, we evaluated the potential of a gas–liquid membrane contactor process for selective removal of
H2S from biogas. The effects of biogas retention time (GRT), membrane thickness and liquid absorbent pH were investigated.
A dilute sodium hydroxide solution was used as absorbent. The results revealed that H2S removal efficiency (RE) improved
with increasing GRT and absorbent pH, and decreased with increasing membrane thickness. When GRT reduced from 19
to 3.4 min, the RE of
H2S and
CO2 decreased by over 2.5 and 5.2 times, respectively. In contrast, a higher desulfurization
selectivity was observed with lower GRT and thicker membranes. The
CH4 content of the treated biogas increased along
with increasing GRT and was enriched from 60% to a maximum of 87% with only 4.68% loss. The SEM–EDS analysis con-
firmed the deposition of inorganics such as Ca, Mg, S and Si on the membrane surface. However, any membrane clogging
and fouling problem was not observed. In summary, the novel gas–liquid polydimethylsiloxane membrane contactor tested
in this study has performed well in selective removal of
H2S from biogas and is expected to be a promising alternative to
conventional desulfurization processes.
Keywords Biogas · Desulfurization · Inorganics deposition · CH4 content · Membrane separation
Introduction
Biogas produced from anaerobic digestion of organic
wastes is a favorable energy source that plays a vital role
in the emerging renewable energy market, because of its
substrate flexibility and energy yield [1]. Biogas composi-
tion mainly depends on the raw materials used and process
conditions. Biogas consists of mainly methane
(CH4) and
carbon dioxide
(CO2) together with minor impurities such
as hydrogen sulfide
(H2S), siloxanes and aromatic, halo-
genated and other volatile organic compounds (VOCs) and
ammonia [2]. Biogas can be used directly to generate power
and heat; however, the high
CO2 content reduces its heat-
ing value and increases the compression and transportation
costs restricting its use at the point of production [3]. The
H2S is known to cause critical problems such as; toxicity,
corrosion in pipes, turbines or other units, as well as adverse
environmental pollution (sulfur dioxide emissions) [4, 5].
In order to obtain energy from biogas in a more productive
and cost-efficient way, the biogas has to be enriched and the
impurities should be removed.
There is an increasing demand for technologies capable
of separating
H2S and
CO2 from the biogas and enable the
delivery of a purified gas product suitable for use in differ-
ent applications [6, 7]. Conventional gas–liquid absorption
processes such as bubble column, packed and spray towers
are dependent on direct contact between the gas and liquid
phases, and cheap process when water is available. How-
ever, these processes require large space, high pressure and
energy consumption, low temperature, high investment cost
[8, 9]. The membrane gas–liquid contacting process, which
integrates the advantage of conventional absorption devises,
can potentially overcome those operational limitations and
* Ebrahim Tilahun
ebroti@gmail.com
Erkan Sahinkaya
erkansahinkaya@gmail.com
Bariş Çalli
baris.calli@marmara.edu.tr
1
Department of Environmental Engineering, Marmara
University, 34722 Istanbul, Turkey
2
Department of Bioengineering, Istanbul Medeniyet
University, 34700 Istanbul, Turkey

1 3
is considered as a better alternative for the treatment of both
H2S and
CO2 [10]. Membranes used in a gas–liquid contact-
ing process act as barrier between the gas and the liquid
phases without dispersing one phase into another. In the
membrane contactor, gas diffuses from the gas side across
the membrane and reaches into the gas–liquid interface,
where gas is absorbed in the liquid. Compared with conven-
tional gas absorption devices, the use of membrane contactor
offers a number of advantages such as high surface area per
unit contact volume, ease of adjusting the flowrate of liquid
and gas phases independently, high compactness and easy
to scale up or down [10–12]. Because of the advantages
offered by membrane contactors, numerous studies have
been carried out. Qi and Cussler [13, 14] first used micropo-
rous hollow fiber membranes for gas absorption. Follow-
ing their innovative findings, many others have focused on
gas–liquid membrane contactors for removal of
H2S, CO2
and other gases from gas streams [15, 16]. The prolonged
contact of microporous membrane with the liquid phase can
generate detrimental effects on the membrane material due
to swelling or gradual wetting of the hydrophobic membrane
pores, which then may decrease the absorption flux signifi-
cantly [17–19]. Similarly, other researchers also attributed
the performance deterioration of the membrane contactors
due to wetting problem. Wetting significantly affects the
mass transfer coefficients of the membrane module, thus
membrane resistance increases sharply and operation perfor-
mance declines soon [20, 21]. The possibility to ensure non-
wetting conditions for long operational period is of drastic
importance. Hence, to overcome the wetting problem and
avoid any penetration of the liquid into the membrane, a
material which represents a barrier for penetration of liquid
into the membrane is suggested. One possible solution is
to use a nonporous or dense polymeric membrane, which
is highly permeable to
H2S and absolutely impermeable to
the liquid permeation. Besides, the nonporous membrane
can enhance selectivity for
H2S over
CO2 and
CH4 since
separation is mediated by diffusion through the polymeric
membrane material [22]. Nevertheless, it was concluded in
a limited number of studies that the nonporous membrane
layer could not provide sufficient mass transfer compared to
microporous membranes [23, 24]. Therefore, a high mem-
brane permeability is necessary to increase the overall mass
transfer capacity of the nonporous membrane contactors. As
a result, reactive liquid absorbent such as quicklime, slaked
lime, sodium hydroxide, or amine solutions are preferably
used to increase the driving force for better absorption rate
and capacity. The high concentrations of
CO2 in the biogas
make the
H2S removal difficult because the
CO2 consumes
the alkalinity in the liquid phase. Due to its low cost and
rapid reaction rate with
H2S compared to other gases, a
sodium hydroxide solution was ideally used as an absorbent
liquid. In addition, hydrophobic nonporous membranes like
polydimethylsiloxane (PDMS) is suitable for use in mem-
brane contactors due to its low cost and availability. So
far, many studies investigated the effects of different liquid
absorbents and membrane materials on the performance of
H2S and
CO2 absorption. However, studies evaluating the
effects of different operating parameters on the selective
H2S
removal performance of membrane contactor was very lim-
ited. In this study, the effects of gas retention time (GRT),
membrane thickness and pH of the liquid absorbent on the
membrane contactor based biogas desulfurization process
were investigated experimentally.
Experimental
Materials
A gas cylinder containing
H2S (1%) and
CO2 (39%) in
balance of
CH4 (60%) was purchased from Hat Industrial
Gases PLC, Kocaeli, Turkey. Sodium hydroxide was pur-
chased from a local supplier (TEKKİM Kimya Sanayi,
Bursa, Turkey) and diluted as required with distilled water
to prepare the stock alkaline solution. All chemicals were
of the highest available purities and used without further
purification. A commercial tubular PDMS (polydimethylsi-
loxane) membrane having 1 mm thickness and 7 mm inter-
nal diameter was purchased from EUROFLEX (Germany),
while the PDMS having thickness of 1.5 and 2 mm with
internal diameter of 7 mm were purchased from DEUTSCH
& NEUMANN (Germany).
Membrane Contactor Setup
A laboratory scale gas–liquid membrane contactor was
designed and manufactured to perform biogas desulfurization
experiments under different operational conditions. The sche-
matic diagram and the details of the desulfurization setup was
shown in our previous paper [25]. The internal diameter and
height of the glass reactor were 120 and 200 mm, respectively.
The reactor active working volume and the tubular membrane
length immersed in the liquid were 1.5 L and 3 m, respectively.
The reactor was totally filled with tap water to minimize the
volatilization of the sulfur compounds. To reduce the oxy-
gen concentration inside the reactor the liquid absorbent was
flushed with nitrogen gas. Due to the high toxicity of
H2S, all
experiments were performed in a fume hood to confine any
accidental leakage of
H2S gas. The gas flow rate was adjusted
by a stainless steel mass flow controller on the line between
the gas cylinder and the set-up. Digital gas counter was used
to measure the inlet and outlet biogas volumetric flow rate
(MGC, Ritter, Germany). A pH range of 7–10 was maintained
by automated addition of 1N NaOH into the liquid absorbent
as needed. Almost all of
H2S and some of
CO2 were removed

1 3
from the simulated gas mixture by diffusing through the mem-
brane and then absorbing/reacting with the mildly alkaline
solution. For uniform distribution of gas and alkaline solu-
tion in the liquid phase magnetic stirrer was used throughout
the experiment. The experiments were carried out at ambient
pressure (approximately 1 atm), and the liquid temperature
was kept at 20 ± 1 °C using a heating blanket equipped with
electric coils wrapped on the wall of the reactor. The con-
ductivity, oxidation reduction potential (ORP) and dissolved
oxygen (DO) were monitored with electrodes on-line using a
digital multimeter (Multi 9430, WTW, Germany).
Techniques and Test Methods
Scanning Electron Microscopy with Energy Dispersive X-ray
Spectroscopy (SEM-EDS) analyses were carried out at Yildiz
Technical University Science and Technology Application and
Research Center (Istanbul, Turkey). The changing morpholo-
gies of the membrane were directly observed using a SEM
after Au–Pd coating. The semi quantitative elemental analyses
of the inorganics deposited on the external surface of the mem-
brane were performed using an EDS coupled with SEM at the
end of the experiment. The concentration of sulfide in the liq-
uid phase was determined spectrometrically (WTW PhotoLab
6100VIS) following the method described by Cord-Ruwisch
[26]. Sulfate (4500-SO4
2−
E) concentration was measured
according to standard methods [27]. The contents of
H2S,
CO2 and
CH4 gas in the entrance and exit of the membrane
contactor were analyzed by gas chromatography (GC-2014;
Shimadzu, Tokyo, Japan) equipped with a thermal conductiv-
ity detector and a stainless steel column. Argon gas was used
as the carrier gas. The temperatures of injection port, column
and detector were 150, 40 and 150 °C, respectively. When
the H2S concentration of the biogas in the exit was below the
detection limit of the GC-TCD, a second scrubbing liquid was
used at the end of the gas line to absorb all traces of
H2S before
venting, the detail of the measurement on the second scrubber
was described in a former study [28]. All samples were col-
lected and analyzed in triplicate following each change in the
process operation conditions and the averages were reported.
Calculations
In this study,
H2S, CO2 or
CH4 removal efficiencies (RE) from
the gas phase were calculated according to Eq. 1.
where Qg
in
and
Qg
out
are inlet and outlet biogas flowrates
(m3
/days), respectively.
Cg
in
and
Cg
out
are the inlet and
outlet concentrations in gas phase (mg/l). In the case of
(1)
RE (%) =
(
Qg
in
× Cg
in
)
−
(
Qg
out
× Cg
out
)
(
Qg
in
× Cg
in
) × 100
simultaneous removal of
H2S and
CO2, selectivity of the
desulfurization could be represented by selectivity factor
or separation factor (S), which could be expressed by Eq. 2.
where Xi,j are the mole fraction of components i and j in the
liquid phase and
Yi,j are the mole fraction of components i
and j in the feed gas. Surface removal rate or flux (J) of the
membrane contactor is another performance indicator, which
can be estimated as Eq. 3.
where J is the flux of the gas components (g/m2
/days) and A
is the membrane surface area
(m2
).
Results and Discussions
Impact of GRT and Membrane Thickness
on Desulfurization Performance
The gas retention time (GRT) is one of the crucial param-
eters determining the biogas desulfurization efficiency. The
influence of GRT on the removal efficiency (RE) of each
gas component is presented in Table 1. The pH of the liq-
uid absorbent was constant (pH 10) in the experiments. The
result shows that, the decrease in GRT reduces the mem-
brane surface contact per unit volume of gas, which in turn
reduces RE. The
H2S RE of the process was superior, espe-
cially when the GRT laid in the range of 19–10.4 min. The
H2S was completely removed at that GRT ranges. Moreover,
at GRT of over 10 min, an effluent
H2S concentration below
300 ppmv was achieved, which is safe to use in cogenera-
tion units [29]. However, lower GRT resulted in higher
effluent H2S concentrations which needs further treatment
before using the gas for various applications. In the same
way, CO2 removal declined significantly with the decreased
GRT. At all GRT tested, the RE of
H2S was much higher
than that of
CO2. Because
H2S has relatively high critical
temperature or gas condensability and expected to permeate
faster through the dense membrane than
CO2. Baker [30]
stated that the permeation of gas components through rub-
bery polymers depends mainly on the gas condensability.
Over all, the RE of
H2S and
CO2 improved by more than
2.5 and 5.2 times, when the GRT was raised from 3.4 to
19 min, respectively (Table 1). This finding is in agreement
with the results presented by Wang et al. [21]. They reported
that gas removal capacity decreased as the GRT declined
because of the shorter contact between the gas and liquid
(2)
S =
(
Xi∕Xj
)
(
Yi∕Yj
)
(3)
J =
(
Qg
in
× Cg
in
)
−
(
Qg
out
× Cg
out
)
A

1 3
phases. Nevertheless,
CH4, having a low diffusivity through
the PDMS membrane and low solubility in the liquid absor-
bent, could not permeate across the membrane. Figure 1 also
illustrated that,
CH4 content in the effluent stream increased
along with the GRT and enriched from 60% to a maximum
of 87% with only 4.68% loss. Heile et al. [31] performed a
similar study for upgrading of a biogas containing 20% of
CO2 and 80% of
CH4 using a PDMS membrane and they
improved the
CH4 content up to 88% in the outlet.
In addition, experiments with varying membrane thick-
nesses of 1, 1.5 and 2 mm were also performed. According
to the results shown in Table 1, thicker membrane reduces
the H2S and
CO2 transfer rates across the membrane due to
longer diffusion time. The negative impact of higher mem-
brane thickness in overall mass transfer of other permeates in
silicone membranes also reported by Raghunath and Hwang
[32]. They found significant mass transfer resistance when
the thickness of a silicone membrane was above 1.16 mm.
In a similar manner, Brookes and Livingston [33] used sili-
cone based membranes and they reported a reduction by
a factor of 1.5 in the overall mass transfer coefficient for
phenol when the membrane thickness was increased from
0.3 to 0.5 mm. Therefore, with thinner membranes, higher
loading rates could be attained. As a general trend, when the
membrane thickness increases, the
CO2 removal efficiency
decreases more significantly than that of
H2S even at high
GRT. The reduction was much more pronounced for lower
GRT. Nii and Takeuchi [34] also used PDMS hollow fıber
modules for the removal of
CO2 and achieved a higher per-
formance with thinner membranes. It can be concluded that
thicker membrane had introduced a considerable resistance
to the gas diffusion, which decreased the absorption flux
inevitably.
In the previous studies, separation factor has been exten-
sively used to indicate gas separation efficiency of mem-
brane based processes [35, 36]. In the gas–liquid membrane
contacting process used here
H2S in the feed gas diffused
through the membrane and was absorbed in the mildly alka-
line absorbent. For that reason, the permeate selectivity was
used to describe the process efficiency. As shown in Eq. 2,
the permeate selectivity depends on the difference between
the inlet and outlet gas concentrations which are influenced
by the membrane thickness, GRT, and pH of the absorbent
used. In order to examine the effect of different membrane
thicknesses and GRT on the selectivity of the membrane
contactor, the overall mass transfer ratio of
H2S to
CO2 and
CH4 were calculated. During each experimental tests, pH
of the liquid was adjusted to 10. As presented in Table 1,
a higher selectivity for the separation of
H2S from the gas
mixture was observed at lower GRT. A change in membrane
thickness from 1 to 2 mm had a positive influence on the
selectivity due to the higher permeability of
H2S in thicker
membrane compared to those of other gases. However, in
that case a lower RE was observed as a result of additional
Table 1 The membrane
contactor performance at
different GRT and membrane
thicknesses at liquid pH of 10
Thickness
(mm)
Gas retention time
(GRT) (min)
Removal efficiency (RE) (%) Selectivity (S)
H2S CO2 CH4 H2S/CH4 H2S/CO2
1 19 100 79.3 4.68 21.4 1.26
10.4 98.3 56.4 3.44 28.6 1.74
5.6 63.6 34.9 1.84 34.6 1.82
3.4 40.0 15.3 0.69 59 2.6
1.5 19 99.9 68.7 4.12 24.2 1.45
10.4 97.8 43.0 2.86 34.2 2.27
5.6 60.6 22.8 1.56 38.8 2.66
3.4 38.2 12.1 0.63 60.6 3.06
2 19 99.7 60.5 3.58 27.8 1.65
10.4 97.0 35.8 2.4 40.2 2.72
5.6 56.9 18.0 1.23 46.3 3.16
3.4 36.2 8.92 0.58 62.4 4.06
GRT (min)
3 6 9 12 15 18 21
Gas
effluent
(%)
0
20
40
60
80
100
CH4
CO2
H2S
Fig. 1 Impacts of GRT on the effluent gas contents at pH 10. (Color
figure online)

1 3
resistance of the membrane. Consequently, for a gas liquid
membrane contactor process used here, these two parameters
can considerably influence the
H2S selectivity and RE. Spe-
cific to the experimental results of this work, the maximum
separation factor for
H2S/CO2 and
H2S/CH4 were reached up
to 4 and 62, respectively, in the case of using thicker mem-
brane (2 mm). Stern and Bhide [37] pointed out that
H2S is
more permeable than
CO2 through PDMS by approximately
a factor of 1.8. Chatterjee et al. [38] also used cellulose ace-
tate membrane to clean the biogas having 6% of
H2S, 29%
of CO2 and 65% of
CH4 at 10 bar, and they reported the
H2S/CH4 separation factor as only 19. In another work on
separation of gases using Poly (ether urethane) membrane,
separation factor of
H2S/CO2 and
H2S/CH4 were reported as
3 and 21, respectively [38].
Impact of Liquid Absorbent pH on Desulfurization
Performance
The gas-liquid-membrane contactor was also tested with
liquid absorbent having different pH values for the removal
of H2S from the biogas stream. As displayed in Table 2,
the reduction of GRT negatively affected the RE of the gas
components. However, when the pH of the liquid absorbent
increased, the permeation of
H2S across the membrane
improved due to the increased concentration differences
between the two sides of the membrane, i.e. the driving
force. According to the report of earlier studies charged
ions existing in the liquid absorbent are unable to permeate
through the silicone rubber membrane [22, 39, 40]. Indeed,
when the pH decreased, the amount of undissociated
H2S
increased, resulting in an increase of the
H2S back diffusion
[25]. Hence, the pH is a key parameter in controlling
H2S
mass transfer through the membrane. The mildly alkaline
solutions give the greater fractional removal of
H2S com-
pared to
CO2 and
CH4. This is owing to the instantaneous
reaction rates
OH−
with
H2S rather than the other gases [13,
14]. At the highest GRT (19 min) changing of pH from 7
to 10 were not significantly affect
H2S RE (Table 2). To
establish the limiting process conditions, the transition in
GRT from 19 to 3.4 min was examined, and upon decreasing
of the GRT,
H2S RE dropped sharply. In general, at higher
liquid pH (> 8.5) a better RE was achieved. Because at high
pH values almost all of the dissolved
H2S was dissociated
to hydrosulfide ion, which increased the RE by continually
re-establishing the higher
H2S concentration gradient. Smet
et al. [41] also observed a sharp increase in hydrosulfide ion
concentration when the pH was higher than 7.04. Similarly,
González-Sánchez et al. [42] reported that a slightly alkaline
pH could improve the
H2S mass transfer from the gas phase
to the liquid phase.
According to the result shown in Table 2, the absorption
of CO2 into the liquid absorbent was considerably higher
than that of
CH4, as it chemically reacted with
OH−
. The
absorption of both
H2S and
CO2 in alkaline liquid was
assisted by agitation. The turbulence in the liquid increases
the diffusion of the molecules into the liquid phase due to
reduction of boundary layer resistance between membrane
and liquid interface. Under the tested conditions, the highest
RE of
CO2 was achieved at GRT of 19 min as 69.9, 71.1 and
79.3% at liquid pH of 7, 8.5 and 10, respectively. On the con-
trary, at the lowest GRT (3.4 min), the RE of
CO2 at liquid
pH of 7 and 8.5 were 12 and 14.8%, respectively, while as
the liquid pH raised to 10 its RE increased to 15.31%, which
is due to the high concentration of
OH−
ion at higher liquid
pH values. Likewise, the positive effect of absorbent con-
centration on the mass transfer rate were formerly reported
[43, 44]. In general, the pH of the liquid in this study was
not increased above 10, to avoid much carbonate formation
(CO2 absorption) and
CH4 loss.
The conductivity of a solution depends on the concen-
tration of all the ions present, the greater their concen-
trations, the higher the conductivity. During operation
of the system used here, the major cations are,
H+
, Na+
,
Ca2+
, and
Mg2+
. The major anions are,
OH−
, HS−
, S2−
,
HCO3
−
and
CO3
2−
. Hence, acidic or basic solution resulted
with high conductivity. Moreover, the conductivity is the
sum of the contribution of all ions present in the solu-
tion. For soft water samples, a pH of 7 will have the least
conductivity. On other hand, samples with a pH above 7
are likely to be higher conductivity. Figure 2 also con-
firmed that when the pH of the liquid absorbent increased
from 7 to 10, proportionally the conductivity raised up.
This happened due to higher dissociation of
H2S and
CO2
into sulfide and carbonate. Moreover, to kept constant pH
sodium hydroxide has been supplied to the liquid phase.
Table 2 Impact of different liquid pH on gas RE at 1 mm membrane
thickness
pH Gas retention time
GRT (min)
Removal efficiency (RE) (%)
H2S CO2 CH4
10 19.0 100 79.3 4.68
10.4 98.3 56.4 3.44
5.6 63.6 32.9 1.84
3.4 40.0 15.3 0.69
8.5 19.0 99.9 71.1 4.44
10.4 91.5 51.6 3.25
5.6 52.8 28.9 1.55
3.4 32.0 14.8 0.59
7.0 19.0 97.9 69.9 3.96
10.4 82.0 45.8 3.01
5.6 41.8 27.1 1.15
3.4 20.0 12.0 0.50

1 3
Our results were consistent with the finding of leveling
[45]. Figure 2 demonstrated that ORP decreased as the
pH of the liquid absorbent increased, which was due to
higher accumulation of sulfide. ORP values was around
− 200 mV at low sulfide accumulation, whereas at higher
loadings significant sulfide accumulation caused ORP to
drop below − 380 mV. Besides, at pH 10, in particular with
a GRT of 3.4 min, the ORP value decreased sharply as an
indication of sulfide accumulation. This behavior proves a
correlation between the ORP and
H2S absorption capacity
of the slightly alkaline liquid absorbent.
Membrane Morphology and Inorganics Deposition
The surface morphology of PDMS membrane was examined
using a scanning electron microscope (SEM). SEM images
of virgin membrane and dirty membrane are presented in
Fig. 3a and b, respectively. The SEM images confirms there
is no visible variation between the surface morphologies of
the two membranes samples. However, when the surface of
the membrane was magnified, the layer of deposits on the
membrane surface could be clearly realized. As illustrated
in Fig. 3b, it is observed that after each experimental work
a white crystal substances appeared on the surface of used
Fig. 2 ORP and conductivity
observation at different pH and
GRT. (Color figure online)
ORP
(mV)
-400
-300
-200
-100
Biogas retention time (min)
5 10 15 20
Conductivity
(µS/cm)
0
500
1000
1500
2000
pH 7 pH 8.5 pH 10
pH 7 pH 8.5 pH 10

1 3
membrane, which should be inorganic matters as confirmed
by EDS analysis, whereas the virgin membrane surface
viewed clean and smooth (Fig. 3a). The SEM images also
attributed to the fact that the pattern of inorganic matters is
an unevenly distributed over the membrane surface. Despite
minor deposition of inorganics on the membrane surface,
the structure was not suffered after being exposed to sodium
hydroxide aqueous solutions. In addition, the membrane
used here was less sensitive to wetting since there is no pores
that supports the liquid to penetrate through the membrane.
Thus, membrane wetting was not observed in our PDMS
membrane. However, other researchers attributed perfor-
mance deterioration of the microporous membrane contac-
tors due to wetting and blockage of the membrane pores.
Fig. 3 SEM image of the mem-
brane surface a before (virgin)
and b after the experiments
(used)

1 3
They reported significant drop in mass transfer capacity of
the contactors due to the developed membrane resistance
[20, 46]. Keshavarz et al. [19] used a microporous hollow
fiber membrane contactor to investigate the simultaneous
absorption of
CO2 and
H2S into the aqueous solution of
diethanolamine and they found that the RE of both gases
significantly decreased due to membrane wetting compared
with the non-wetted mode. Wang et al. [21] also studied
on CO2 absorption using a diethanolamine solution as an
absorbent in a polypropylene microporous membrane con-
tactor. They reported that with only 5% membrane pores
wetting the overall mass transfer coefficient of the contactor
may reduce by 20%.
With the use of scanning electron microscopy (SEM)
associated with an energy dispersive X-ray spectrometer
(EDS) often focus on the top surface deposits as indi-
cated in Fig. 4a and b, and it is possible to detect the exist-
ence of metal sulfide and carbonate salts deposition. The
Fig. 4 The semi quantitative elemental analyses of the inorganics deposited on the membrane surface a SEM image b EDS analysis after the
experiment

1 3
semi-quantitative EDS composition analysis indicated cer-
tain amounts of inorganic elements were accumulated on the
membrane surface. The presence of multivalent metal ions
in the tap water, which used in the system as an absorbent
liquid, was the origin of inorganics deposition. Thus, in the
presence of metal ions such as
Ca2+
and
Mg2+
that act as
cationic effects, there would be a strong interaction with
anionic sulfide, and carbonate, which were generated dur-
ing the dissociation of
H2S and
CO2, to form precipitates.
Furthermore Fig. 2 also approves that, when the pH of the
liquid phase increased the conductivity of the liquid absor-
bent linearly increased proportionally. The reason for this
observation was the greater accumulation of sulfide and car-
bonate in the reactor as the pH of the liquid raised up (when
the pH increased to 10). It can be observed that carbon (C)
and oxygen (O) were the major elements present in the spot
samples. The C and O peaks were likely due to the chemi-
cal structure of the membrane and entrapment of inorganic
matters. As discussed above, the multivalent ions have been
shown to form precipitates with sulfide and carbonates could
contribute to the calcium (Ca), magnesium (Mg) peaks and
their weight percent were about 36.1 and 0.5%, respectively.
The sulfur (S) peaks possibly indicating that S containing
compounds were able to penetrate through the membrane
with some retained on the membrane surface. Silicon (Si)
detected on the used membrane surface was mainly due to its
presence in the polydimethylsiloxane membrane layer. The
EDS spectra of the virgin membrane has also strong Si and
O peaks which originated from the membrane itself (data
not shown). Gold (Au) and palladium (Pd) were used dur-
ing coating procedures, due to its irrelevant to the elemental
analysis of the membrane foulants their peaks were removed.
In the present study regardless of inorganics deposition
discussed above, the abiotic experiments show selective
removal H2S than
CO2 and
CH4. Besides, in this gas–liquid
membrane contactor applications, membrane fouling and
clogging were not observed, which resulted in almost stable
flux during the operation of the membrane. Previous studies
also demonstrated nonporous membrane was more resistant
to fouling [47]. However, in the long run operation, an excess
deposition may have reduced the mass transfer efficiency of
the PDMS membrane. It was assumed that the accumulation
of inorganics on the membrane surface may decreased the
cross sectional area, subsequently reduced the gas retention
time and increased the pressure drop of the gas permea-
tion. Moreover, it may create an additional film resistance
for the gas to reach the liquid phase, which likely resulted
in limitation of
H2S transfer. Chuichulcherm et al. [40] used
silicone membrane combined with biological sulfide pro-
duction for the treatment of metal-containing wastewater.
They reported that the chemical reaction between sulfide
and multivalent ions in the wastewater enhanced the sulfide
mass transfer. However, accumulation of metal precipitates
on the membrane surface limited sulfide transfer and the
resistance due to the metal sulfide precipitates even exceeded
the membrane resistance. Other studies also demonstrated
the mass transfer reduction of membrane based reactors due
to the blockage of the membrane pores by the organic and
inorganics accumulation [46, 48, 49].
Conclusions
Results of the study performed to evaluate the effects of
operating parameters on
H2S removal performance of a
gas–liquid PDMS membrane contactor showed that the
H2S removal declined severely, when GRT decreased below
10 min, due to mass transfer limitation. In addition, increas-
ing membrane thickness reduced the rate of
H2S diffusion
through the membrane. By using a slightly alkaline absor-
bent, more
H2S was removed with respect to
CO2 and
CH4.
On other hand, lower GRT and higher membrane thickness
significantly increased the mass transfer resistance against
CO2, but showed a marginal influence on
H2S removal;
hence it favors a higher selectivity for
H2S. The maximum
selectivity for
H2S/CO2 and
H2S/CH4 were 4 and 62, respec-
tively. Despite some loss, the
CH4 content of the biogas
enriched up to 87%. Results of SEM-EDS analysis showed
that at higher pH values, sulfide and carbonate salts of Ca,
Mg, and Si deposit were observed on the membrane surface.
In spite of inorganics deposition, no membrane clogging and
fouling problems were observed. However, it is supposed
that in long run, the chemical precipitates on the outer sur-
face of membrane may behave as a secondary barrier and
reduce the diffusion of gasses. Finally, it is concluded that
the gas–liquid PDMS membrane contactor process tested in
this study is a promising alternative to conventional biogas
desulfurization processes.
Acknowledgements This study was financially supported by YTB
(Presidency for Turks Abroad and Related Communities) and Mar-
mara University Scientific Research Committee BAPKO (Project No.
FEN-C-DRP-070317-0109).
References
1. Holm-Nielsen, J.B., Al Seadi, T., Oleskowicz-Popiel, P.: The
future of anaerobic digestion and biogas utilization. Bioresour.
Technol. 100, 5478–5484 (2009). https://doi.org/10.1016/j.biort
ech.2008.12.046
2. Zhang, R., Brown, R.C., Suby, A., Cummer, K.: Catalytic destruc-
tion of tar in biomass derived producer gas. Energy Convers.
Manag. 45, 995–1014 (2004). https://doi.org/10.1016/j.encon
man.2003.08.016
3. Marzouk, S.A.M., Al-Marzouqi, M.H., Teramoto, M., Abdul-
latif, N., Ismail, Z.M.: Simultaneous removal of
CO2 and H2S
from pressurized
CO2–H2S–CH4 gas mixture using hollow fiber

1 3
membrane contactors. Sep. Purif. Technol. 86, 88–97 (2012).
https://doi.org/10.1016/j.seppur.2011.10.024
4. Poloncarzova, M., Vejrazka, J., Vesely, V., Izak, P.: Effec-
tive purification of biogas by a condensing-liquid membrane.
Angew. Chem. Int. Ed. 50, 669–671 (2011). https://doi.
org/10.1002/anie.201004821
5. Panza, D., Belgiorno, V.: Hydrogen sulphide removal from land-
fill gas. Process Saf. Environ. Prot. 88, 420–424 (2010). https://
doi.org/10.1016/j.psep.2010.07.003
6. Asri, O., Hafidi, I.E., Afilal, M.E.: Comparison of biogas purifi-
cation by different substrates and construction of a biogas purifi-
cation system. Waste Biomass Valoriz. 6, 459–464 (2015). https
://doi.org/10.1007/s12649-015-9378-z
7. Kurchania, A.K., Panwar, N.L., Pagar Savita, D.: Improved
biogas stove with scrubbing unit for household use. Waste Bio-
mass Valoriz. 2, 397–402 (2011). https://doi.org/10.1007/s1264
9-011-9080-8
8. Awe, O.W., Zhao, Y., Nzihou, A., Minh, D.P., Lyczko, N.: A
review of biogas utilisation, purification and upgrading tech-
nologies. Waste Biomass Valoriz. 8, 267 (2017)
9. Ryckebosch, E., Drouillon, M., Vervaeren, H.: Techniques
for transformation of biogas to biomethane. Biomass Bioen-
erg. 35, 1633–1645 (2011). https://doi.org/10.1016/j.biomb
ioe.2011.02.033

10. Gabelman, A., Hwang, S.-T.: Hollow fiber membrane contac-
tors. J. Memb. Sci. 159, 61–106 (1999). https://doi.org/10.1016/
S0376-7388(99)00040-X

11. Klaassen, R., Feron, P.H.M., Jansen, A.E.: Membrane contactors
in industrial applications. Chem. Eng. Res. Des. 83, 234–246
(2005). https://doi.org/10.1205/cherd.04196

12. Belaissaoui, B., Claveria-Baro, J., Lorenzo-Hernando, A.,
Albarracin Zaidiza, D., Chabanon, E., Castel, C., Rode, S.,
Roizard, D., Favre, E.: Potentialities of a dense skin hollow
fiber membrane contactor for biogas purification by pressurized
water absorption. J. Memb. Sci. 513, 236–249 (2016). https://
doi.org/10.1016/j.memsci.2016.04.037

13. Qi, Z., Cussler, E.L.: Microporous hollow fibers for gas absorp-
tion: I—mass transfer in the liquid. J. Memb. Sci. 23, 321–332
(1985). https://doi.org/10.1016/S0376-7388(00)83149-X

14. Qi, Z., Cussler, E.L.: Microporous hollow fibers for gas absorp-
tion: II—mass transfer across the membrane. J. Memb. Sci. 23,
333–345 (1985). https://doi.org/10.1016/S0376-7388(00)83150
-6

15. Karoor, S., Sirkar, K.K.: Gas absorption studies in microporous
hollow fiber membrane modules. Ind. Eng. Chem. Res. 32, 674–
684 (1993). https://doi.org/10.1021/ie00016a014

16. Poddar, T.K., Majumdar, S., Sirkar, K.K.: Membrane-based
absorption of VOCs from a gas stream. AIChE J. 42, 3267–3282
(1996)

17. Mavroudi, M., Kaldis, S.P., Sakellaropoulos, G.P.: A study of mass
transfer resistance in membrane gas-liquid contacting processes. J.
Memb. Sci. 272, 103–115 (2006). https://doi.org/10.1016/j.memsc
i.2005.07.025

18. Keshavarz, P., Fathikalajahi, J., Ayatollahi, S.: Analysis of
CO2
separation and simulation of a partially wetted hollow fiber mem-
brane contactor. J. Hazard. Mater. 152, 1237–1247 (2008). https
://doi.org/10.1016/j.jhazmat.2007.07.115

19. Keshavarz, P., Fathikalajahi, J., Ayatollahi, S.: Mathematical
modeling of the simultaneous absorption of carbon dioxide and
hydrogen sulfide in a hollow fiber membrane contactor. Sep. Purif.
Technol. 63, 145–155 (2008). https://doi.org/10.1016/j.seppu
r.2008.04.008

20. Atchariyawut, S., Jiraratananon, R., Wang, R.: Separation of
CO2
from CH4 by using gas-liquid membrane contacting process. J.
Memb. Sci. 304, 163–172 (2007). https://doi.org/10.1016/j.memsc
i.2007.07.030

21. Wang, R., Zhang, H.Y., Feron, P.H.M., Liang, D.T.: Influence of
membrane wetting on
CO2 capture in microporous hollow fiber
membrane contactors. Sep. Purif. Technol. 46, 33–40 (2005).
https://doi.org/10.1016/j.seppur.2005.04.007

22. Tilahun, E., Sahinkaya, E., Çalli, B.: A hybrid membrane gas
absorption and bio-oxidation process for the removal of hydro-
gen sulfide from biogas. Int. Biodeterior. Biodegrad. 127, 69–76
(2018). https://doi.org/10.1016/j.ibiod.2017.11.015

23. Al-Marzouqi, M.H., Marzouk, S.A.M., El-Naas, M.H., Abdul-
latif, N.:
CO2 removal from
CO2–CH4 gas mixture using differ-
ent solvents and hollow fiber membranes. Ind. Eng. Chem. Res.
48, 3600–3605 (2009). https://doi.org/10.1021/ie800977z

24. Al-saffar, H.B., Ozturk, B., Hughes, R.: A comparison of porous
and non-porous gas-liquid membrane contactors for gas sepa-
ration. Chem. Eng. Res. Des. 75, 685–692 (1997). https://doi.
org/10.1205/026387697524182

25. Tilahun, E., Bayrakdar, A., Sahinkaya, E., Çalli, B.: Perfor-
mance of polydimethylsiloxane membrane contactor process for
selective hydrogen sulfide removal from biogas. Waste Manag.
1–8 (2017). https://doi.org/10.1016/j.wasman.2017.01.011

26. Cord-Ruwisch, R.: A quick method for the determination of dis-
solved and precipitated sulfides in cultures of sulfate-reducing
bacteria. J. Microbiol. Methods 4, 33–36 (1985). https://doi.
org/10.1016/0167-7012(85)90005-3

27. AWWA A.: WEF, standard methods for the examination of
water and wastewater. (2005)

28. Bayrakdar, A., Tilahun, E., Calli, B.: Biogas desulfurization
using autotrophic denitrification process. Appl. Microbiol. Bio-
technol. 100, 939–948 (2016). https://doi.org/10.1007/s0025
3-015-7017-z

29. Ramos, I., Fdz-Polanco, M.: Microaerobic control of biogas sul-
phide content during sewage sludge digestion by using biogas
production and hydrogen sulphide concentration. Chem. Eng. J.
250, 303–311 (2014). https://doi.org/10.1016/j.cej.2014.04.027

30. Baker, R.W.: Membrane Technology and Application. Wiley, New
York (2004)

31. Heile, S., Rosenberger, S., Parker, A., Jefferson, B., McAdam,
E.J.: Establishing the suitability of symmetric ultrathin wall poly-
dimethylsiloxane hollow-fibre membrane contactors for enhanced
CO2 separation during biogas upgrading. J. Memb. Sci. 452,
37–45 (2014). https://doi.org/10.1016/j.memsci.2013.10.007

32. Raghunath, B., Hwang, S.T.: General treatment of liquid-phase
boundary layer resistance in the pervaporation of dilute aqueous
organics through tubular membranes. J. Memb. Sci. 75, 29–46
(1992). https://doi.org/10.1016/0376-7388(92)80004-4

33. Brookes, P.R., Livingston, A.G.: Aqueous-aqueous extraction of
organic pollutants through tubular silicone rubber membranes. J.
Memb. Sci. 104, 119–137 (1995). https://doi.org/10.1016/0376-
7388(95)00020-D

34. Nii, S., Takeuchi, H., Takahashi, K.: Removal of
CO2 by gas
absorption across a polymeric membrane. J. Chem. Eng. Jpn. 25,
67–72 (1992)

35. Stern, S.A.: Polymers for gas separations: the next decade. J.
Memb. Sci. 94, 1–65 (1994)

36. Mudler, M.: Basic Principles of Membrane Technology. Springer,
New York (1996)

37. Stern, A., Bhide, B.D.: Permeability of silicone polymers to
ammonia and hydrogen sulfide. J. Appl. Polym. Sci. 38, 2131–
2147 (1989)

38. Chatterjee, G., Houde, A.A., Stern, S.A.: Poly(ether urethane) and
poly(ether urethane urea) membranes with high
H2S/CH4 selectiv-
ity. J. Memb. Sci. 135, 99–106 (1997)

39. Livingston, A.G.: Extractive membrane bioreactors: a new pro-
cess technology for detoxifying chemical industry wastewaters.
J. Chem. Technol. Biotechnol. 60, 117–124 (1994). https://doi.
org/10.1002/jctb.280600202

1 3

40. Chuichulcherm, S., Nagpal, S., Peeva, L., Livingston, A.: Treat-
ment of metal-containing wastewaters with a novel extractive
membrane reactor using sulfate-reducing bacteria. J. Chem. Tech-
nol. Biotechnol. 76, 61–68 (2001)

41. Smet, E., Lens, P., Van Langenhove, H.: Treatment of waste gases
contaminated with odorous sulfur compounds. Crit. Rev. Environ.
Sci. Technol. 28, 89–117 (1998)

42. González-Sánchez A., Revah S., Deshusses M.A.: Alkaline bio-
filtration of
H2S odors. Environ. Sci. Technol. 42, (2008)

43. Wang, R., Li, D.F., Liang, D.T.: Modeling of
CO2 capture by
three typical amine solutions in hollow fiber membrane contac-
tors. Chem. Eng. Process. Process Intensif. 43, 849–856 (2004).
https://doi.org/10.1016/S0255-2701(03)00105-3

44. Yan, S.P., Fang, M.X., Zhang, W.F., Wang, S.Y., Xu, Z.K., Luo,
Z.Y., Cen, K.F.: Experimental study on the separation of
CO2
from flue gas using hollow fiber membrane contactors without
wetting. Fuel Process. Technol. 88, 501–511 (2007). https://doi.
org/10.1016/j.fuproc.2006.12.007

45. Leveling, T.: The relationship between pH and conductivity in a
lithium contaminated, de-ionized water system. fermilab-pbar-
note-674, 1–11 (2002)

46. Attaway, H., Gooding, C.H., Schmidt, M.G.: Comparison of
microporous and nonporous membrane bioreactor systems for the
treatment of BTEX in vapor streams. J. Ind. Microbiol. Biotech-
nol. 28, 245–251 (2002). https://doi.org/10.1038/sj/jim/7000235

47. Côté, P., Bersillon, J.L., Huyard, A.: Bubble-free aeration using
membranes: mass transfer analysis. J. Memb. Sci. 47, 91–106
(1989). https://doi.org/10.1016/S0376-7388(00)80862-5

48. Yurtsever, A., Çinar, Ö, Sahinkaya, E.: Treatment of textile waste-
water using sequential sulfate-reducing anaerobic and sulfide-oxi-
dizing aerobic membrane bioreactors. J. Memb. Sci. 511, 228–237
(2016). https://doi.org/10.1016/j.memsci.2016.03.044

49. Alvarez-Hornos, F.J., Volckaert, D., Heynderickx, P.M., Van
Langenhove, H.: Performance of a composite membrane biore-
actor for the removal of ethyl acetate from waste air. Bioresour.
Technol. 102, 8893–8898 (2011). https://doi.org/10.1016/j.biort
ech.2011.06.064

Article 3.pdf

Recommended

Recommended

More Related Content

What's hot

What's hot (20)

Similar to Article 3.pdf

Similar to Article 3.pdf (20)

Recently uploaded

Recently uploaded (20)

Article 3.pdf