Journal of Membrane Science 352 (2010) 189196

Contents lists available at ScienceDirect

Journal of Membrane Science

journal homepage: www.elsevier.com/locate/memsci

Pilot-scale production of oxygen from air using

perovskite hollow bre membranes
Xiaoyao Tan a,b, , Zhigang Wang b , Bo Meng b , Xiuxia Meng b , K. Li c
a
b
c

School of Environmental and Chemical Engineering, Tianjin Polytechnic University, Tianjin 300160, China
School of Chemical Engineering, Shandong University of Technology, Zibo 255049, China
Department of Chemical Engineering and Technology, Imperial College London, South Kensington, London SW7 2AZ, UK

a r t i c l e

i n f o

Article history:
Received 19 October 2009
Received in revised form 1 January 2010
Accepted 4 February 2010
Available online 11 February 2010
Keywords:
Oxygen production
Hollow bre
Perovskite
Mixed conducting membrane

a b s t r a c t
La0.6 Sr0.4 Co0.2 Fe0.8 O3 (LSCF) hollow bre membranes prepared by a phase-inversion/sintering technique were assembled into a membrane system to produce oxygen of high purity (>99%) from air at
elevated temperatures. The separation performances, stability, scaling-up effect and the energy consumption of the membrane system were investigated both theoretically and experimentally. The membrane
system containing 889 hollow bres could yield maximum 3.1 L(STP) min1 oxygen with the purity of
99.9% at 1070 C and 98.5 kPa vacuum degree, but the temperature higher than 1070 C would lead to
the system failure. It showed the potential of much higher production rates only if the high-temperature
sealing problem could be solved. When operated at around 960 C, the system exhibited more than 1167 h
longevity with the oxygen production rate of 0.84 L(STP) min1 and oxygen purity of 99.4%. The energy
consumption of the system increased with operating temperature but the energy consumption per unit
oxygen product decreased with increasing the operating temperature and the effective membrane areas.
In order to reduce the oxygen cost to commercial level, heat exchangers have to be integrated in the
membrane system to recover the heat energy in both the exhaust gas and the oxygen product. The oxygen recovery should be limited within 2040% for the sake of both energy and membrane area savings.
2010 Elsevier B.V. All rights reserved.

1. Introduction
Oxygen production from air is of great importance in both
environmental and chemical industries. It is usually achieved
by cryogenic distillation, pressure swing adsorption (PSA) or by
polymeric membrane separation. These processes are either of
high energy consumption or unable to produce oxygen of high
purity. Alternately, dense mixed conducting membranes such as
La1x Srx Co1y Fey O3 (LSCF) perovskite which exhibit appreciable oxygen ionic and electronic conductivity have become of great
interest as a potentially economical, clean and efcient means of
high pure oxygen production [17]. When an oxygen partial pressure gradient is imposed across such dense membranes at a high
temperature (usually >700 C), the oxygen may be transferred from
the high partial pressure side to the low partial pressure side without the need of electrodes and external electrical loadings, making
the membrane system and operation much simplied.
In recent years, perovskite hollow bre membranes have
been successfully prepared by a phase-inversion/sintering process

Corresponding author. Tel.: +86 533 2786292; fax: +86 533 2786292.
E-mail address: cestanxy@yahoo.com.cn (X. Tan).
0376-7388/$ see front matter 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.memsci.2010.02.015

[812]. Such hollow bre membranes can provide much larger

areas per unit volume than other congurations such as at sheet
or tubular, thus it is possible to reduce the membrane system
size remarkably. Furthermore, the hollow bre conguration also
makes the high-temperature sealing less problematic in fabricating membrane modules [1315]. In this work, we developed
a perovskite hollow bre membrane system that could produce
over 3 L(STP) min1 oxygen gas with the purity of >99%. The separation performances, the stability, the scaling-up effect and the
energy consumption of the perovskite membrane systems have
been investigated both theoretically and experimentally.
2. Experimental
2.1. LSCF powder and hollow bre membranes
La0.6 Sr0.4 Co0.2 Fe0.8 O3 (LSCF) hollow bre membranes were
prepared by a phase-inversion/sintering technique using the powders synthesized through a solgel combustion process which were
described elsewhere [8,12]. The resultant hollow bre membranes
were usually ca. 2832 cm in length and around 0.12/0.18 cm in
i.d./o.d. One end of the hollow bres was closed with the same
LSCF material and the gas-tight property was achieved after sin-

190

X. Tan et al. / Journal of Membrane Science 352 (2010) 189196

Fig. 1. Photos of (a) the LSCF hollow bre membranes; (b) the hollow bre bundles; and (c) the hollow bre membrane module.

tering. Surface modication was conducted on the hollow bres by

coating catalysts on the outer surface so as to improve the permeation ux of the hollow bre membranes. The catalyst consisted of
a perovskite with high Co content and an Ag paste with the particle size <7 m and was mixed in weight ratio of 6:1. The detailed
coating procedures were described elsewhere [16].
2.2. Hollow bre membrane module and system
In practical applications, the hollow bres must be assembled
into a membrane module. For the sake of sealing and facile replacement of broken bres, we rst bundled 7 bres together in a
quartz tube (8 mm in diameter and 70 mm in length) with a hightemperature silicone sealant (1592, purchased from Tonsan New
Materials and Technol. Co., Beijing) that is able to withstand up to
350 C. The bre bundles as required were then placed in a stainless steel holder. Fig. 1 shows the LSCF hollow bre membranes,
the bre bundles and the membrane module. It is very important
to avoid any leakage in order to obtain highly pure oxygen product.
Therefore, each hollow bre was individually tested to be gas-tight
prior to bundling and every bundle was again tested to be free
of leakage before integrating them into a module. In this work,
three modules containing 1 bundle (7-bre module), 9 bundles
(63-bre module) and 129 bundles (889-bre module) respectively
were fabricated to study the scaling-up effect. The effective membrane areas by subtracting the sealing length from the whole bre
length for these modules were ca. 78 cm2 , 702 cm2 and 9914 cm2 ,
respectively.
The hollow bre membrane system for oxygen production is
schematically shown in Fig. 2. The hollow bre membrane module was placed under a vertically positioned tubular furnace. An
insulator was placed between the bottom inlet of furnace and the
stainless steel holder so that the temperature at the sealing points
was lower than that the sealant can withstand. A vacuum was
applied to the membrane module using a claw rotor oil-free vacuum pump (2ZBL, purchased from Beijing Huaxiajia Sci & Tech Inc.).
The operating vacuum degree was controlled using a frequency
modulator (Suny3200, from Sunye Electric Co.) tted to the vacuum
pump and was measured with a digital pressure sensor (PSA, from
Autonics Co.). The oxygen concentration and the ow rate of the

product were measured using an oxygen analyzer (TG-J310, from

Xian Taige analysis instrument Inc.) and a digital lm ow meter
(SF-1U/2U, purchased from Horiba Stec Co. Ltd.), respectively. The
operating temperature was controlled using a temperature controller with a K-type thermocouple positioned in the middle of the
furnace. Meanwhile, a temperature gauge tted with another Ktype thermocouple was used to monitor the temperature at the
sealing points. In the operation, the oxygen product ow rate and
the concentration of the oxygen product were recorded with time

Fig. 2. Hollow bre membrane system for oxygen production (A) ow chart; and
(B) photo of the setup.

X. Tan et al. / Journal of Membrane Science 352 (2010) 189196

191

under different operating conditions. All the values of volumetric

ow rates in this study were calibrated to the standard temperature
and pressure (STP).
3. Results and discussion
3.1. Production capacity of the membrane system
Production capacity of a membrane system is mostly determined by the membranes effective permeation areas as well as
the permeability of single hollow bre membrane. In this work, we
mainly used the membrane module containing 889 hollow bres
(called as 889-bre system) to study the separation performances
of the perovskite membrane systems because it can reect most of
the problems involved in practical applications. On the other hand,
since the silicone sealant can only withstand maximum 350 C, the
sealing points of the hollow bre module or the joints between the
hollow bres and the quartz tube have to be placed out of the furnace. As the hollow bre module is inserted deeper in the furnace
tube, the effective bre length for oxygen permeation would get
longer and the oxygen production rate could be increased due to
a larger permeation area provided. Meanwhile, the sealing points
would also be closer to the furnaces bottom inlet leading to higher
temperature and failure of sealing on these points. Fig. 3 shows the
oxygen production rate of the 889-bre system with different air
gaps, which is referred to the distance from the sealing points to
the bottom inlet of the furnace. As can be seen, the oxygen production rate was increased noticeably by decreasing the air gap.
At the air gap of 1.5 cm, the maximum oxygen production rate
at 1070 C reached to 3.10 L min1 with the oxygen concentration
of 99.88%. However, the temperature at the sealing points in this
case reached to as high as 235 C and the operation only lasted
for about 2 h. A fast decrease in oxygen purity was observed after
2 h operation. The post-mortem analysis indicated that the leakage
occurred at the sealing points of several bre bundles. In addition,
it was also found that there were several hollow bres collapsed
due to the sucking force of vacuum under such a high temperature.
After the air gap was increased to 3.0 cm, the highest temperature at the sealing points decreased to 173 C but the maximum
oxygen production rate at 1070 C also decreased to 2.36 L min1
with the oxygen concentration of 99.71%. Even now, the separation lasted only for about 20 h after which leakage at the sealing
points occurred again. In order to prevent the membrane system
from failure due the leakage, the air gap was set to 4.5 cm so that
the temperature at the sealing points would not be higher than
130 C. Under this air gap, the maximum oxygen production rate at
1070 C only reached 1.79 L min1 with the oxygen concentration

Fig. 3. Oxygen production rate of the 889-bre membrane system as a function of

temperature with different air gaps (vacuum degree = 98.9 kPa).

Fig. 4. Performance of the membrane system at different temperatures (889-bre

module, vacuum degree = 97.4 kPa) (,  for the rst cycle; ,  for the second cycle).

of 99.58%. Since there existed a noticeable temperature gradient

from the center to the bottom inlet of the furnace tube, the effective heating length of the bres for oxygen permeation could be
much shorter than the apparent length of the bres. As the air gap
was increased from 1.5 cm to 4.5 cm, the effective heating length
would also be shortened by 3 cm. Consequently, the actual permeation area was signicantly decreased leading to the noticeable
decrease in oxygen production rate although the air gap was small
compared to the total bre length. In the following experiments,
the operating temperature would not be higher than 1060 C so as
to avoid the membrane breakage due to over-sintering.
3.2. Effect of operation parameters
Performances of the 889-bre membrane system at different
temperatures are shown in Fig. 4 where the applied vacuum was
kept at 97.4 kPa. Two operating cycles, which is dened as a whole
operating process from heating up the membrane system from
room temperature, collecting experimental data at high temperatures to cooling the system back to ambient temperature, were
conducted under the same conditions. As can be seen, the two
cycles exhibited the same increasing trend with temperature for
both the production rate and the product purity, and the data for
the same temperatures had almost the same values. This indicated that the repeatability of the system operation was good.
As is expected, the oxygen production rate and the product concentration increased with increasing temperature since a higher
temperature facilitated both the surface exchange reactions and
the bulk diffusion. When the system was operated at 960 C, the
oxygen production rate and the oxygen purity reached 0.86 L min1
and 99%, respectively. As the temperature was increased to 1060 C
the oxygen production rate also increased to 1.66 L min1 with the
corresponding oxygen concentration of 99.67%. The oxygen concentration in product did not reach to the theoretical value (100%)
because of minor leakage of the membrane system which led to
the entry of some nitrogen into the product stream. At room temperature, oxygen permeation in the perovskite membrane would
not occur but there was still some gas could be collected from the
outlet of the membrane module under vacuum operation. This gas
was mostly the air entering into the product side due to the system
leakage. The ow rate of this air leaking was measured at room temperature using a gas bubble ow meter to be about 4.7 mL min1 .
But in operation, the amount of air leaking to the product stream
was calculated from the production rate and the product concentration as 615 mL min1 depending on the operating temperature.
The leakage may be produced either from the tiny defects presented on the hollow bre membranes or from the silicone sealing
joints. For the present membrane system the oxygen production
rate and the product concentration could reach above 99% and

192

X. Tan et al. / Journal of Membrane Science 352 (2010) 189196

above 1.1 L min1 , respectively, only if the operating temperature

was higher than 1050 C.
Fig. 5 shows the oxygen production rate and the product purity
as a function of vacuum degree under different temperatures. Both
the oxygen production rate and product purity at a given temperature increased with increasing the applied vacuum degree.
This result can be expected because the driving force for oxygen permeation was increased. At lower temperatures (<900 C),
the product concentration increased rapidly with increasing the
vacuum degree. But the promotion of product purity by increasing vacuum degree would be weakened as the temperature was
increased. On the other hand, at lower vacuum degrees, the oxygen
production rate can be improved noticeably with increasing vacuum degree. However, when the vacuum degree was higher than
99 kPa, the increase in oxygen production rate would be lessened
especially at higher temperatures. Therefore, it is not necessary
to apply the vacuum degree higher than 99 kPa because it contributes little to improving the oxygen production rate but leads
to appreciable increase in operation cost.
3.3. Long-term test
In order to measure the stability of the hollow bre membrane system, it was operated continuously for 1067 h at around
960 C and 9798 kPa applied vacuum degree with the recorded
data shown in Fig. 6. It should be mentioned that it had already
taken totally about 101 h with more than 20 cycles in 30 days for
the performance measurement prior to the longevity test. It can
be seen that the production rate was kept at around 0.84 L min1
and the oxygen concentration in product was around 99.4%. When
the operating temperature was a bit lowered, the oxygen product
purity and the production rate also decreased. But once the temperature was recovered to 960 C, both the oxygen product purity
and the production rate were also recovered to the original values. Therefore, the present LSCF hollow bre membrane system
possessed a good stability.
3.4. Scaling-up effect
In order to examine the scaling-up effect of the hollow bre
membrane system, three modules containing 7 bres (1 bre bun-

Fig. 5. Effect of vacuum degree on the performance of the hollow bre membrane
system (889-bre module).

Fig. 6. Performance of the 889-bre membrane system as a function of operation

time.

dle in 20 mm 280 mm furnace tube), 63 bres (9 bundles in

60 mm 350 mm furnace tube) and 889 bres (129 bundles in
180 mm 600 mm furnace tube) were fabricated respectively.
Fig. 7 shows the product concentration of the membrane systems at
different temperatures. It can be seen that the larger module could
yield purer oxygen product at the same temperature. For example,
the oxygen product concentration of the 7-bre membrane module only reached maximum 97.14% at 1060 C, but the maximum
product concentrations of the 63-bre and the 889-bre membrane system reached 98.9% and 99.71% at 1060 C, respectively.
This implies that the relative leakage to oxygen production rate in
the larger membrane system was lower than that in the smaller
one. Furthermore, for all the three modules the product concentration always increased with increasing temperature. Therefore, it is
possible to obtain highly pure oxygen products by increasing the
production rate of the perovskite membrane systems.
However, the efciency of the hollow bre membrane system
would be reduced remarkably due to the use of larger membrane
modules. That is to say, the scaling-up effect of the hollow bre
membrane system was signicant. As can be seen from Fig. 8, the
average oxygen production rate of a single hollow bre membrane
in the 889-bre system is much lower than that in the 7-bre
system especially at higher temperatures. For example, the average oxygen production rate per hollow bre in the 7-bre system
was 10.37 mL min1 at 1060 C but in the 889-bre system it was
only 1.86 mL min1 at the same temperature, which was only 18%
of the 7-bre modules average production rate. It is noted that
even in the 7-ber system the average production rate was still
much lower than that calculated by the single bre permeation test
(1.85 mL cm2 min1 at 1000 C [16]). Such signicant decrease in
the average oxygen production rate in a single bre mostly originated from the uneven temperature distribution in the furnace

Fig. 7. Oxygen product concentration as a function of temperature for different

hollow bre membrane systems.

X. Tan et al. / Journal of Membrane Science 352 (2010) 189196

Fig. 8. Average oxygen production rate of a single bre as a function of temperature

in different hollow bre membrane systems.

tube. Due to the unavoidable temperature gradient from the center

to the bottom inlet of the furnace tube, the length of the hollow
bres served for oxygen permeation is actually much shorter than
the total bre length. Such loss of the effective permeation length
due to the uneven distribution of temperatures would get more as
the furnace tube is enlarged. On the other hand, it is difcult to
keep all the hollow bre membranes in a large module containing hundreds of bres to have the same length, which can be seen
in Fig. 1b. The shorter bres actually contributed little to oxygen
permeation especially in the large furnace having a signicant temperature gradient. Consequently, the average oxygen permeation
rate over the whole bre length in the large hollow bre module
would be much lower than that in the smaller one. In addition,
although the furnace tube was increased with the membrane module size, the top outlet for exhaust gas was not enlarged as much
as the bottom inlet so as to achieve and keep the high temperature in the furnace tube for permeation. As the permeation rate
increases, the oxygen concentration in air stream in the large membrane system may be lower than that in the smaller one leading to
a reduced permeation ux. Theoretically, if the whole hollow bre
length can be utilized for oxygen permeation, the oxygen production rate of the 889-bre module can reach up to 22.9 L min1 (or
1.38 m3 h1 ) at 1000 C, which is 19.5 times the present production
rate, 1.16 L min1 . Therefore, in order to improve the production
capacity of the system, it is essential to eliminate or at least to lessen
the uneven distribution of temperatures for the membrane module. A feasible way to utilize the whole membrane area in a hollow
bre module for oxygen permeation is to preheat the air feed to the
permeation temperature before it is introduced into the system.
But the high-temperature sealing will become a great challenge to
assemble intact hollow bre membrane modules.

193

Fig. 9. Heating power of the furnace in the 889-bre membrane system at different
operating temperatures.

exhaust gas. Consequently, the volume of the exhaust gas as well as

the heat loss to produce a volume of oxygen product also decreased
noticeably. Therefore, the heating power per unit oxygen product
decreased with operating temperature as shown in Fig. 9. It suggests that increasing the operating temperature favors reducing
the cost of oxygen product. However, the energy consumption to
produce 1 Nm3 O2 for the present hollow bre membrane system
was still much higher (12 kWh even at 1050 C) than that by the
PSA process (0.7 kWh/Nm3 ). It contradicts to the general sense
that the ceramic membrane process can save a lot of oxygen production cost [18]. The primary reason for that is due to the failure
to recover the heat energy from exhaust gas which will be analyzed
in details as follows.
Fig. 10 shows the operation of the membrane system combined
with heat exchangers for heat recovery where the parameters for
energy calculation are also presented. The oxygen balance in the
membrane system is given by:
0.21FAir = FOx + (FAir FOx ) xe

(1)

3.5. Energy consumption

The energy consumption of the membrane system which determines the cost of oxygen product mainly includes the power for
the furnace to heat air feed to permeation temperature and the
power for the vacuum pump to yield oxygen partial pressure gradient. Generally, the power consumed by the vacuum pump is much
lower than that by the furnace. This was also found to be true in
the operation of the present hollow bre membrane system. Fig. 9
shows the heating power of the furnace as a function of operating temperature. As can be seen, the heating power of the furnace
increased with increasing operating temperature. This is because
the oxygen production rate increased with increasing temperature
and hence more energy was required to heat the air feed. Moreover, the heat loss also increased at a higher operating temperature.
On the other hand, the higher oxygen production rate at a higher
temperature would lead to decrease in oxygen concentration in the

Fig. 10. Diagram of the membrane system with heat exchange (A) between the air
feed and the oxygen stream; (B) between the air feed and the exhaust stream; and
(C) between air feed and both the exhaust and oxygen streams.

194

X. Tan et al. / Journal of Membrane Science 352 (2010) 189196

Table 1
Parameters for calculating the energy consumption of oxygen product.
Membrane area
Outer diameter of the bre membrane
Inner diameter of the bre membrane
Enhancement factor by catalytic modication
Operating vacuum degree
Air feed ow rate
Air feed temperature
Temperature of exhaust gas
Operating temperature
Efciency of vacuum pump

Am = 1 m2
0.18 cm
0.12 cm
= 2.0
98.3 kPa
0.2 Nm3 min1
25 C
50 C
8501050 C
0.8

Diffusion coefcient of oxygen vacancy [7]

DV = 1.58 102 exp 8852.5

T

Forward reaction rate constant of Eq. (3) [7]

kf

Reverse reaction rate constant of Eq. (3) [7]

kr = 2.07 10 exp

where FAir and FO2 are the air feed ow rate and the oxygen production rate, respectively; xe is the oxygen fraction in the furnace
tube and exhaust stream. For the catalytically modied hollow bre
membrane, the oxygen production rate may be given by [17]:
kr [(p1 xe )

FO2 = Am



(cm2 /s)


0.5
= 1.85 104 exp 27291
(cm/Pa
s)
T
 29023

4
2

[(Rm /Ro ) p0.5

+(Rm /Rin ) (p1 xe )
2

0.5

0.5

(p2 )

0.5

]+[((2kf (Ro Rin ))/DV ) (p1 xe p2 )

0.5

(mol/cm s)

Using the parameters in Table 1, the energy consumption of the

membrane system with different membrane areas to produce one
molar oxygen product was calculated and plotted against operating temperature in Fig. 11. For comparison, the oxygen cost by the
PSA process is also presented with a dash line in the gure. As can

(2)
where Am is the membrane area of the hollow bre module; the
permeation enhancement factor by catalytic modication; p1 and
p2 are the upstream and downstream pressures (p2 = pa operating
vacuum degree where pa is atmospheric pressure), respectively; Rm
the logarithmic radius, Rm = (Ro Rin )/ln(Ro /Rin ), in which Ro and Rin
are respectively the outer and the inner radius of the bre; DV is the
diffusion coefcient of oxygen vacancy; kf and kr are, respectively,
the forward and the reverse reaction rate constants for the surface
exchange reaction:
kf /kr

x + 2h
+ VO OO

1
O
2 2

(3)

With the assumption of 10% heat loss of the membrane system,

the heating power of the furnace may be given, respectively by,
(A): heat exchange between the air feed and the oxygen product
only
Q1 =

FO2 CpO2 (50 25) + [(0.21FAir FO2 )CpO2 + 0.79FAir CpN2 ] (T 25)
0.9

(4a)
(B): heat exchange between the air feed and the exhaust stream
only
Q2 =

FO2 CpO2 (T 25) + [(0.21FAir FO2 )CpO2 + 0.79FAir CpN2 ] (50 25)
0.9

(4b)
(C): heat exchange between the air feed and both the exhaust and
oxygen streams
Q3 =

FO2 CpO2 (T  25) + [(0.21FAir FO2 )CpO2 + 0.79FAir CpN2 ] (50 25)
0.9

(4c)
where Cpi is the specic heat capacity of gas species i; T is
the temperature of the air feed after heating by the exhaust
stream,
T =

(0.21FAir FO2 )CpO2 + 0.79FAir CpN2

(0.21CpO2 + 0.79CpN2 )FAir

(T 50) + 25

(5)

The power of the vacuum pump corresponding to the oxygen

product rate may be estimated by:
W=

pa VO2 ln(pa /p2 )



2.48 103 FO2



ln

p 
a

p2

where  is the efciency of the vacuum pump.

(6)

Fig. 11. Energy consumption of the oxygen product by the membrane system with
heat exchange (A) between the air feed and the oxygen stream; (B) between the air
feed and the exhaust stream; and (C) between air feed and both the exhaust and
oxygen streams.

X. Tan et al. / Journal of Membrane Science 352 (2010) 189196

195

tion cost of the perovskite membrane systems, the oxygen recovery

should be limited within 2040%.
Acknowledgements
The authors gratefully acknowledge the research funding provided by the National High Technology Research and Development
Program of China (No. 2006AA03Z464), the National Natural Science Foundation of China (No. 20676073) and EPSRC in the United
Kingdom (EP/E032079/1).

Nomenclature
Fig. 12. Plot of the energy consumption of oxygen product and membrane area
against oxygen recovery (air feed = 0.2 Nm3 min1 ).

be seen, the oxygen production cost of the membrane process cannot be reduced to the PSA cost level unless the heat in exhaust
gas is recovered by performing heat exchange between the air
feed and the exhaust stream. In addition, the energy consumption
per unit oxygen product by the membrane system can be reduced
by either increasing the operating temperature or increasing the
membrane areas. Since the energy consumption of the membrane
system is mainly the residual heat along with the exhaust and product streams, it can be reduced by decreasing the amount of exhaust
gas for a given air feed. In another word, it is necessary to increase
the oxygen recovery so as to reduce the oxygen cost. This can be
achieved by increasing the operating temperature and the membrane areas. Fig. 12 plots the energy consumption per unit oxygen
product and the required membrane area against oxygen recovery
which is dened as:
=

FOx
100%
0.21FAir

(7)

It can be seen that the energy consumption per unit oxygen

product decreases remarkably with increasing the oxygen recovery at lower recovery levels (<20%). However, after the oxygen
recovery is higher than 40%, further increasing the oxygen recovery only leads to a slight decrease in energy consumption, but the
membrane areas to achieve this recovery have to be increased signicantly especially for a lower operating temperature, resulting
in the remarkable increase of membrane cost. Therefore, the oxygen recovery should be limited within 2040% so as to reduce
the overall oxygen production cost of the perovskite membrane
systems.
4. Conclusions
A LSCF perovskite hollow bre membrane system has been
successfully scaled-up to produce large quantity of oxygen with
high purity. The maximum oxygen production rate reached to
3.1 L min1 with the corresponding oxygen concentration of above
99.8%. The system possessed a good stability and can be operated
for more than 1167 h at around 960 C to produce 99.4% oxygen
product in production rate of 0.84 L min1 . The operating temperature for the LSCF hollow bre membrane system cannot exceed
1070 C otherwise the system breakage would occur. Increasing
vacuum degree favors the increase in oxygen production rate, but
this effect is alleviated after the vacuum degree is higher than
99 kPa. For the commercialization of ceramic membrane technology, heat exchangers have to be integrated in the membrane system
to recover the heat energy in exhaust gas and oxygen products. The
operation cost of the perovskite membrane system can be reduced
by either increasing the operating temperature or increasing the
membrane areas. In order to reduce the overall oxygen produc-

Am

effective membrane area for oxygen permeation

(cm2 )
Cp
specic heat capacity of gas species (J/(mol K))
DV
effective diffusivity of oxygen vacancy (cm2 /s)
FAir
air feed ow rate (mol/s)
oxygen permeation rate (mol/s)
FO2
kr
reverse surface exchange reaction rate constant
(mol/(cm2 s))
kf
forward surface exchange reaction rate constant
(cm/(Pa0.5 s))
pa
atmospheric pressure (1.013 105 Pa)
upstream and downstream pressure (Pa)
p1 , p2
Q1 , Q2 , Q3 heating power of the furnace (W)
Rm
algorithmic
radius
of
bre,
Rm = (Ro Rin )/ln(Ro Rin )
inner and outer radius of hollow bre (cm)
Rin , Ro
T
operating temperature (K)
T
temperature of the air feed after heating by the
exhaust stream (K)
W
power of the vacuum pump (W)
xe
oxygen fraction in the furnace tube and exhaust
stream

permeation improvement factor due to the surface

modication

efciency of the vacuum pump

oxygen recovery (%)

References
[1] V.V. Kharton, A.A. Yaremchenko, A.V. Kovalevsky, A.P. Viskup, E.N. Naumovich,
P.F. Kerko, Perovskite-type oxides for high-temperature oxygen separation
membranes, J. Membr. Sci. 163 (1999) 307317.
[2] J. Sunarso, S. Baumann, J.M. Serra, W.A. Meulenberg, S. Liu, Y.S. Lin, J.C.D. da
Costa, Mixed ionic-electronic conducting (MIEC) ceramic-based membranes
for oxygen separation, J. Membr. Sci. 320 (2008) 1341.
[3] M. den Exter, J.F. Vente, D. Jansen, W.G. Haije, Viability of mixed conducting
membranes for oxygen production and oxyfuel processes in power production,
Energy Procedia 1 (2009) 455459.
[4] P.N. Dyer, R.E. Richards, S.L. Russek, D.M. Taylor, Ion transport membrane technology for oxygen separation and syngas production, Solid State Ionics 134
(2000) 2133.
[5] S. Li, W. Jin, N. Xu, J. Shi, Synthesis and oxygen permeation properties of
La0.2 Sr0.8 Co0.8 Fe0.2 O3 membranes, Solid State Ionics 124 (1999) 161170.
[6] D. Stefan, J.V. Herle, Oxygen transport through dense La0.6 Sr0.4 Fe0.8 Co0.2 O3
perovskite-type permeation membranes, J. Eur. Ceram. Soc. 24 (2004)
13191323.
[7] S.J. Xu, W.J. Thomson, Oxygen permeation rates through ion-conducting perovskite membranes, Chem. Eng. Sci. 54 (1999) 38393850.
[8] X. Tan, Y. Liu, K. Li, Preparation of La0.6 Sr0.4 Co0.2 Fe0.8 O3 hollow bre membranes for oxygen production by a phase-inversion/sintering technique, Ind.
Eng. Chem. Res. 44 (2005) 6166.
[9] K. Li, X. Tan, Y. Liu, Single-step fabrication of ceramic hollow bres for oxygen
permeation, J. Membr. Sci. 272 (2006) 15.
[10] S. Liu, G. Gavalas, Oxygen selective ceramic hollow bre membranes, J. Membr.
Sci. 246 (2005) 103108.

196

X. Tan et al. / Journal of Membrane Science 352 (2010) 189196

[11] T. Schiestel, M. Kilgus, S. Peter, K.J. Caspary, H. Wang, J. Caro, Hollow bre
perovskite membranes for oxygen separation, J. Membr. Sci. 258 (2005) 14.
[12] Z. Wang, N. Yang, B. Meng, X. Tan, Preparation and oxygen permeation properties of highly asymmetric La0.6 Sr0.4 Co0.2 Fe0.8 O3 (LSCF) perovskite hollow
bre membranes, Ind. Eng. Chem. Res. 48 (1) (2009) 510516.
[13] X. Tan, Y. Liu, K. Li, Mixed conducting ceramic hollow bre membranes for air
separation, AIChE J. 71 (2005) 19912000.
[14] X. Tan, K. Li, Oxygen production using dense ceramic hollow ber membrane
modules with different operating modes, AIChE J. 53 (2007) 838845.
[15] X. Tan, Z. Pang, K. Li, Oxygen production using La0.6 Sr0.4 Co0.2 Fe0.8 O3 (LSCF)
perovskite hollow bre membrane modules, J. Membr. Sci. 310 (2008) 550556.

[16] X. Tan, Z. Wang, H. Liu, S. Liu, Enhancement of oxygen permeation through

La0.6 Sr0.4 Co0.2 Fe0.8 O3 hollow bre membranes by surface modications, J.
Membr. Sci. 324 (1/2) (2008) 128135.
[17] Z. Wang, H. Liu, X. Tan, Y. Jin, S. Liu, Improvement of the permeation through
perovskite hollow bre membranes by surface acid-modication, J. Membr. Sci.
345 (1/2) (2009) 6573.
[18] P.A. Armstrong, E.P. Foster, D.A. Horazak, H.T. Morehead, V.E. Stein, Ceramic
and coal: ITM oxygen for IGCC, in: Proceedings of the 22nd International
Pittsburgh Coal Conference, Pittsburgh, Pennsylvania, 2005, September 1115,
2005.