Biotechnology Advances 30 (2012) 13641380

Contents lists available at SciVerse ScienceDirect

Biotechnology Advances
journal homepage: www.elsevier.com/locate/biotechadv

Research review paper

Membrane technology as a promising alternative in biodiesel production: A review

Siew Hoong Shuit, Yit Thai Ong, Keat Teong Lee, Bhatia Subhash, Soon Huat Tan
School of Chemical Engineering, Universiti Sains Malaysia, Engineering Campus, Seri Ampangan, 14300 Nibong Tebal, Pulau Pinang, Malaysia

a r t i c l e

i n f o

Available online 16 February 2012

Keywords:
Biodiesel
Membrane reactor
Catalytically inert membrane
Catalytically active membrane

a b s t r a c t
In recent years, environmental problems caused by the use of fossil fuels and the depletion of petroleum
reserves have driven the world to adopt biodiesel as an alternative energy source to replace conventional
petroleum-derived fuels because of biodiesel's clean and renewable nature. Biodiesel is conventionally
produced in homogeneous, heterogeneous, and enzymatic catalysed processes, as well as by supercritical
technology. All of these processes have their own limitations, such as wastewater generation and high energy
consumption. In this context, the membrane reactor appears to be the perfect candidate to produce biodiesel
because of its ability to overcome the limitations encountered by conventional production methods. Thus, the
aim of this paper is to review the production of biodiesel with a membrane reactor by examining the fundamental concepts of the membrane reactor, its operating principles and the combination of membrane and
catalyst in the catalytic membrane. In addition, the potential of functionalised carbon nanotubes to serve
as catalysts while being incorporated into the membrane for transesterication is discussed. Furthermore,
this paper will also discuss the effects of process parameters for transesterication in a membrane reactor
and the advantages offered by membrane reactors for biodiesel production. This discussion is followed by
some limitations faced in membrane technology. Nevertheless, based on the ndings presented in this review,
it is clear that the membrane reactor has the potential to be a breakthrough technology for the biodiesel
industry.
2012 Elsevier Inc. All rights reserved.

Contents
1.
2.
3.
4.
5.

6.

7.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Limitations in conventional biodiesel conversion technology . . . . . . . . .
Process intensication technologies in biodiesel production . . . . . . . . . .
Concept of membrane reactor . . . . . . . . . . . . . . . . . . . . . . . .
Membrane technology in biodiesel production . . . . . . . . . . . . . . . .
5.1.
Concepts and principles . . . . . . . . . . . . . . . . . . . . . . .
5.1.1.
Membrane separation based on oil droplet size . . . . . . . .
5.1.2.
Membrane separation based on catalytic membrane . . . . .
5.1.3.
Membrane separation based on pervaporation . . . . . . . .
5.2.
Possible combinations of membrane and catalyst in biodiesel production
5.2.1.
Membrane without incorporated catalyst . . . . . . . . . . .
5.2.2.
Membrane with incorporated catalyst . . . . . . . . . . . .
Effect of process parameters in biodiesel production by membrane reactor . . . .
6.1.
Effect of reaction temperature . . . . . . . . . . . . . . . . . . . .
6.2.
Effect of methanol to oil ratio . . . . . . . . . . . . . . . . . . . .
6.3.
Effect of catalyst concentration . . . . . . . . . . . . . . . . . . . .
6.4.
Effect of reactant ow rate . . . . . . . . . . . . . . . . . . . . . .
6.5.
Effect of trans-membrane pressure (TMP) . . . . . . . . . . . . . . .
6.6.
Effect of membrane pore size and thickness . . . . . . . . . . . . . .
Advantages of catalytic membrane reactor in biodiesel production . . . . . . . .
7.1.
Environmentally friendly process . . . . . . . . . . . . . . . . . . .
7.2.
Lower investment cost . . . . . . . . . . . . . . . . . . . . . . . .

Corresponding author. Tel.: + 60 4 5996475; fax: + 60 4 5941013.

E-mail address: chshtan@eng.usm.my (S.H. Tan).
0734-9750/$ see front matter 2012 Elsevier Inc. All rights reserved.
doi:10.1016/j.biotechadv.2012.02.009

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

1365
1366
1367
1367
1368
1368
1368
1368
1368
1369
1369
1370
1371
1371
1372
1372
1372
1373
1373
1373
1373
1375

S.H. Shuit et al. / Biotechnology Advances 30 (2012) 13641380

7.3.
Overcoming the limitation caused by chemical equilibrium
7.4.
High process exibility of feedstock conditions . . . . . .
7.5.
Complying with international standards . . . . . . . . .
8.
Membrane life-time and fouling in biodiesel production . . . . .
9.
Limitations in membrane technology for biodiesel production . .
10.
Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . .
Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

.
.
.
.
.
.
.
.

1. Introduction
Human civilisation has always relied on the utilisation of energy.
As illustrated in Fig. 1, the industrial sector, consisting of diverse
industrial groups that include manufacturing, agriculture, mining
and construction, accounted for 52% of global energy used in 2007;
the residential sector for household activities comprise 14% of the
total; the transportation sector, providing services, such as moving
people and goods by road, rail, air, water and pipeline, uses 27%;
and the commercial sector, which consists of businesses, institutions,
and organisations that provide services, comprises 7%. Although the
global economic recession drove a reduction of energy consumption
by 1.1% in 2009, the International Energy outlook 2010 (IEO, 2010)
projections still predicted an increase of global energy consumption
by 49%, or 1.4% every year until 2035 (EIA, 2010). This predicted
increase forecasts increasing demand of resources for energy production. According to the statistical review conducted by British Petroleum
(BP) (BP, 2009), global energy production depends heavily on oil (35%),
coal (29%) and natural gas (24%) to satisfy the global energy demand, as
shown in Fig. 2. Fossil fuels are the world's slowest-growing source of
energy, and their supplies are decreasing daily. The increasing demand
for energy production throughout the projection period will lead to an
increase in the price of these resources. In addition, the growing emission of carbon dioxide, sulphur dioxide, hydrocarbons and volatile
organic compounds (VOCs) from the combustion of fossil fuels could
result in air pollution, global warming and climate change. These negative
impacts on the environment are the target of current energy policies that
emphasise cleaner, more efcient and environmentally friendly technologies to increase the supply and usage of energy (Hammond et al., 2008;
Hoekman, 2008; Monni and Raes, 2008; Sawyer, 2009). Thus, developments in alternative renewable energy sources have become indispensable for sustainable environmental and economic growth. Among the
explored alternative energy sources, considerable attention has been
focused on biodiesel because it is widely available from inexhaustible
feedstocks that can effectively reduce its production cost.
Biodiesel, which is also known as fatty acid methyl ester (FAME),
is a mixture of monoalkyl esters of long-chain fatty acids derived
from renewable lipid feedstocks, such as vegetable oil and animal
fats. Because biodiesel has similar physical properties to diesel fuels,

Fig. 1. Global energy consumption in 2007 (EIA, 2010).

.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.

1365

.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.

1376
1376
1377
1377
1378
1378
1378
1378

it has established its commercial value in the automobile markets of

Europe, the US, Japan, Brazil and India (Janaun and Ellis, 2010).
Moreover, the implementation of the directive on the promotion of
the use of biofuels for transport in the EU (Directives 2003/30/EC)
mandated the increased use of biofuels to power transportation
from 2% to 5.75% between 2005 and 2010, triggering a huge demand
for biodiesel (Mabee, 2007). Unlike conventional diesel fuel, biodiesel
offers several advantages, including renewability, higher combustion
efciency (Fazal et al., 2011), cleaner emission (Janaun and Ellis,
2010), higher cetane number, higher ash point, better lubrication
(Lin et al., 2011) and biodegradability (Wardle, 2003). Depending
on the climate, local soil conditions and availability, various biolipids
have been used in different countries as feedstocks to produce biodiesel.
Biolipid feedstocks can be divided into four categories: virgin vegetable
oils, waste vegetable oils, animal fats and non-edible oils. Virgin vegetable oil feedstock refers to rapeseed, soybean, sunower and palm oil
(Demirbas, 2008), while waste vegetable oil refers to these oils that
have been used in cooking and are no longer suitable for human
consumption (Conservation ADoE, 2011; Lam et al., 2010). Animal fats
include tallow, lard and yellow grease (Atadashi et al., 2010) while the
non-edible oils include Jatropha (Shuit et al., 2010b; Yee et al., 2009),
neem oil, castor oil, tall oil (Demirbas, 2008) and microalgae (Ahmad
et al., 2011).
Several modication techniques, such as dilution, microemulsion,
pyrolysis and transesterication have been used to reduce the viscosity
of vegetable oil (Andrade et al., 2011). Of these processes, transesterication is the most widely used; this method involves the alcoholysis of
vegetable oil to produce alkyl ester. Generally, the mechanism consists
of three consecutive reversible reaction steps. The rst step involves the
conversion of triglycerides (TG) to diglycerides (DG) and later to monoglycerides (MG). Subsequently, the monoglycerides are converted to
glycerol. Each reaction step produces an alkyl ester. Thus, a total of
three alkyl esters are produced in the transesterication process
(Sharma and Singh, 2008). The overall reaction that occurs in transesterication is simplied in Fig. 3 (Lim and Teong, 2010).

Fig. 2. World energy production in 2009 (BP, 2009).

1366

S.H. Shuit et al. / Biotechnology Advances 30 (2012) 13641380

O
CH2

O
R1

O
CH

Triglyceride

R1

O
R2

+ 3 R OH

O
CH2

R2

CH2

OH

CH

OH

CH2

OH

O
R

R3
Alcohol

R3

Fatty Acid Alkyl Ester

Glycerol

Fig. 3. Transesterication of triglycerides with alcohol (Lim and Teong, 2010).

Common transesterication reactions that are used to produce

biodiesel include homogeneous catalysed transesterication, heterogeneous catalysed transesterication, enzymatic catalysed transesterication and supercritical technology (Marchetti et al., 2007). Each of
these methods has its own disadvantages that eventually limit the
economic feasibility and low environmental impact of the entire
biodiesel production process. The limitations of conventional biodiesel
production technology will be discussed in this paper. In addition,
certain process intensication technologies in biodiesel production will
also be discussed. Because of its ability to overcome those limitations,
membrane technology has the potential to be a promising alternative
for biodiesel production. Therefore, the aim of this paper is to present
the production of biodiesel via membrane reactor, which includes the
operation principles, possible combinations of membrane and catalyst
and the effect of the process parameters of the transesterication
reaction on the membrane reactor. Finally, the advantages offered by
membrane technology in biodiesel production will be discussed. The
discussion will also focus on the ability of the membrane reactor to
produce biodiesel in a more economical and environmentally friendly
manner.
2. Limitations in conventional biodiesel conversion technology
In the conversion of vegetable oil by the transesterication process,
the reversible reaction between the reactant and product indicates
that the formation of biodiesel is highly dependent on the proportion
of the reactant and the conditions of the transesterication process.
According to Le Chatelier's principle, large quantities of alcohol are
needed to shift the equilibrium of the reaction to the product side
and increase the yield of biodiesel (Othman et al., 2010). Unfortunately,
high consumption of alcohol is associated with higher production cost.
The consumption of alcohol could be reduced by using acid or
alkaline catalysts, which could improve the reaction rate and biodiesel
yield. However, homogeneous acid solutions that catalyse transesterication processes, such as sulphuric (Sahoo et al., 2007), hydrochloric
(Boucher et al., 2008), or sulphonic acids (Guerreiro et al., 2006) have
been largely ignored because they increase the time consumption of
the process, require a higher reaction temperature and are corrosive
by nature. Although the use of homogeneous alkaline catalysts, such
as sodium (Rashid et al., 2008) and potassium hydroxide (Rashid
and Anwar, 2008) could overcome these limitations, it has been
reported that the alkaline catalysed reaction is sensitive to the purity
of the reactant. The presence of water and free fatty acids in the raw
feedstock could induce a saponication process in which the free
fatty acid produced by the hydrolysis of triglycerides reacts with the
alkaline catalyst to form soap. The dissolved soap in the glycerol
phase would increase the solubility of methyl ester in the glycerol
and complicate the subsequent separation process (Vicente et al.,
2004). Also, the removal of either the homogeneous acidic or alkaline

catalyst using hot distilled water would eventually result in the need
to dispose of wastewater (Xie and Li, 2006).
Heterogeneous catalyst has been viewed as an alternative solution
to replace the homogeneous catalyst because it is non-corrosive and
environmentally benign. However, the heterogeneous catalytic reaction
usually faces a mass transfer resistance problem because the constitution of the three-phase system (triglycerides, alcohol and solid catalyst)
in the reaction mixture limits the pore diffusion process and reduces the
active site availability for the catalytic reaction, thereby decreasing the
reaction rate (Mbaraka and Shanks, 2006). Catalyst support can minimise the mass transfer limitation, but the active species in the supported
catalyst can easily be corroded by alcohol, shortening the catalyst lifecycle (Liu et al., 2008). The biodiesel obtained from the biocatalytic
transesterication process that uses an enzyme as a catalyst seems
attractive and encouraging for because the product is easily separable
without side reactions (Jegannathan et al., 2008), but biodiesel from
this process is not yet commercially viable because of the requirement
of longer reaction time and the unfavourable reaction yield in comparison to the alkaline catalyst. It has been reported that the enzymatic
transesterication process requires 24 h to achieve a biodiesel yield of
90% (Oda et al., 2005). Most importantly, the major obstacle to this
process is the high cost of the enzyme. The enzyme also requires very
specic reaction conditions because the denaturation of the enzyme
and its deactivation as a result of feed impurity could decrease its
efciency (Dizge et al., 2009).
Supercritical alcohol transesterication provides a new path for the
production of biodiesel without the aid of a catalyst. The supercritical
condition could overcome the mass transfer limitation by enabling the
mixture of triglyceride and alcohol to become a homogeneous phase
(Pinnarat and Savage, 2008). However, the major drawbacks of this
non-catalytic process are its large energy requirement and its infeasibility for large-scale industrial application because of the increased
production cost imposed by the high reaction temperature and pressure
(Yin et al., 2008). Moreover, the supercritical process is potentially
hazardous and requires attention to personal risk and safety.
Both the catalytic and non-catalytic transesterication downstream processes will receive a mixture of product, biodiesel and
glycerol, as well as unreacted reactant and catalyst. Ineffective biodiesel
separation and purication may cause severe diesel engine problems,
such as plugging of lters, coking on injectors, carbon deposits, excessive engine wear, oil ring sticking, engine knocking, and thickening
and gelling of lubricant oil (Demirbas, 2007). In order to obtain highpurity biodiesel, the downstream of the transesterication process
will undergo various complementary separation stages, such as glycerol
separation, catalyst neutralisation and biodiesel purication. The multiple downstream processes are time-consuming and require additional
cost. A recent report revealed that the current downstream processing
alone constituted over 60-80% of the total cost of a transesterication
process plant (Tai-Shung, 2007). In addition, the multiple separation

S.H. Shuit et al. / Biotechnology Advances 30 (2012) 13641380

and purication stages could cause loss of the biodiesel, resulting in a

decrease in the pure biodiesel yield.
3. Process intensication technologies in biodiesel production
The problems described above, including the wastewater issue
(Xie and Li, 2006), limited mass transfer (Mbaraka and Shanks,
2006), thermodynamic equilibrium (Cao et al., 2008b), high energy
requirement (Yin et al., 2008) and multiple downstream processing
steps (Tai-Shung, 2007) in biodiesel production can be improved by
process intensication technologies. These technologies involve the
use of novel reactors or coupled reaction/separation processes to
enhance the reaction rate and to reduce the residence time (Qiu et
al., 2010).
Recently, novel reactors, such as the micro-channel reactor, oscillatory
ow reactor, rotating/spinning tube reactor and cavitational reactor, have
been developed and applied to improve the mixing and mass/heat transfer between the oil and methanol in biodiesel production (Qiu et al.,
2010). The micro-channel reactor can achieve a rapid reaction rate
because it has a high volume/surface ratio, short diffusion distance and
fast and efcient heat dissipation and mass transfer (Qiu et al., 2010;
Wen et al., 2009). The micro-channel reactors used in biodiesel production include micro-channel reactors with T- or Y-ow structures, zigzag
micro-channel reactors (Wen et al., 2009) and slit channel reactor (Kalu
et al., 2011). Because of the high heat transfer rate, it has been reported
that the micro-channel reactor consumed less energy than the conventional stirrer reactor (Qiu et al., 2010). However, the micro-channel reactor suffers from the drawback of low production throughput, which is
attributed to the limitations of the micro-fabrication technology that is
used to produce the micro-channel. Furthermore, the high investment
cost of the micro-channel reactor prohibits the addition of more reactors
in parallel to amplify the production of biodiesel (Kalu et al., 2011).
The oscillatory ow reactor is a type of continuous plug ow reactor
(PFR) (Phan et al., 2011) in which the orice plate bafes are equally
spaced, and a piston is used to produce oscillatory ow (Qiu et al.,
2010). The combination of bafes and oscillatory motion intensies
the radial mixing by the formation of periodic vortices in the bulk
uid, causing an increase in mass and heat transfer while maintaining
plug ow (Phan et al., 2011; Qiu et al., 2010). In addition, the oscillatory
ow reactor can also improve the residence time distribution (RTD) and
multi-phase suspension (Zheng et al., 2007). Because the oscillatory
ow reactor can achieve long residence times, it can be designed with
a smaller length to diameter ratio, which eventually helps to improve
the economy of biodiesel production because of the smaller footprint, lower capital, reduced pumping cost and ease of control (Qiu et
al., 2010).
The rotating/spinning tube reactor is a shear reactor containing
two tubes. The inner tube rotates rapidly within the concentric
stationary outer tube. Both tubes are separated by a narrow annular
gap, which produces Couette ow when the reactants are introduced.
Because of the high shear rate, the reactants are mixed and move
through the gap as a coherent thin lm. This thin lm provides a
large interfacial contact area to enhance the reaction rate between
the oil and the methanol. As a result, less mixing power and reaction
time are required to produce biodiesel using a rotating/spinning reactor
compared to a conventional reactor. This type of reactor is suitable to
handle feedstocks with high FFA because the residence time is short
(Qiu et al., 2010).
The cavitational reactor is another type of novel reactor that has
been used successfully in biodiesel production (Gogate and Kabadi,
2009; Kelkar et al., 2008; Pal et al., 2010; Qiu et al., 2010). Cavitation
is dened as the generation of cavities followed by their growth and
violent collapse, causing high local energy densities, temperatures
and pressures (Gogate and Kabadi, 2009; Qiu et al., 2010). Cavitation
enhances the mass transfer rate of the reaction by creating conditions
of local intense turbulence and liquid micro-circulation currents in

1367

the reactor (Gogate and Kabadi, 2009; Kelkar et al., 2008; Qiu et al.,
2010). Cavitational reactors can be classied into two types: hydrodynamic cavitation and acoustic cavitation (Kelkar et al., 2008; Qiu et al.,
2010). Hydrodynamic cavitation can be generated by using a restriction
component, such as an orice plate, a throttling valve or a venture,
placed in a liquid ow (Gogate and Kabadi, 2009; Kelkar et al., 2008).
At the constriction area, the kinetic energy or velocity of the liquid
increases, but the local pressure decreases. (Gogate and Kabadi, 2009).
A hydrodynamic cavitation reactor is more effective for mixing of
immiscible liquids (Pal et al., 2010). The mixing efciency of a hydrodynamic cavitation reactor has been reported to be 160400 times higher
than that of the conventional mixing method (Qiu et al., 2010). Therefore, the hydrodynamic cavitation reactor consumes half of the energy
required by conventional mechanical stirring (Pal et al., 2010). A reactor
that generates cavitation by ultrasound is known as a sonochemical
reactor (Gogate and Kabadi, 2009) or an acoustic cavitation reactor
(Qiu et al., 2010; Wu et al., 2009). Ultrasound causes a series of
compression and rarefaction cycles by alternately compressing and
stretching the molecular spacing of the medium (Colucci et al., 2005).
Low-frequency ultrasound irradiation is useful for the emulsication
of immiscible liquids, such as methanol and oil. Emulsication is a result
of the induced collapse of cavitation bubbles that disrupt the phase
boundary of methanol and oil (Rokhina et al., 2009). Emulsions with
large interfacial areas provide more reaction sites for transesterication
and eventually increase the reaction rate (Chand et al., 2010). It has
been reported that the operating parameters, such as temperature,
pressure, reaction time and catalyst concentration, are signicantly
reduced in ultrasound-assisted transesterication (Deshmane et al.,
2008; Kalva et al., 2008). However, sonochemical reactors suffer from
erosion and particle shedding at the delivery tip surface because of
the high surface energy intensity (Gogate and Kabadi, 2009). Also, the
scale-up of a sonochemical reactor is relatively more difcult than it is
for a hydrodynamic cavitation reactor because the former relies on a
source of vibration (Qiu et al., 2010).
The microwave reactor is another intensication technology for
biodiesel production. The main function of a microwave reactor is
not to improve the mixing of oil and methanol but to use its irradiation
to transfer energy directly into the reactants and thus accelerate the
transesterication. Because both polar and ionic components are
available in the mixture of oil and methanol/alcohol, a microwave
reactor plays an important role in the more efcient heating of reactants
to the desired temperature because of the energy interactions at the
molecular level (Barnard et al., 2007). Compared to a conventional
thermal heating reactor, a microwave reactor is able to achieve similar
biodiesel conversion with a shorter reaction time and in a more
energy-efcient manner (Qiu et al., 2010).
All of the above mentioned novel reactors intensify the transesterication by either enhancing the mixing of oil and methanol or
improving the heat transfer between the two liquid phases. However,
none of these novel reactors, except the membrane reactor, is able to
overcome the limitation caused by chemical equilibrium in transesterication. Therefore, the membrane reactor offers another interesting
process intensication technology for biodiesel production that will be
discussed in detail in this paper.
4. Concept of membrane reactor
A membrane reactor is also known as a membrane-based reactive
separator (Sanchez Marcano and Tsotsis, 2002). According to IUPAC, a
membrane reactor is dened as a device that combines reaction and
separation in a single unit (Caro, 2008). Generally, the classication
of a membrane reactor is based on four concepts (Ertl et al., 2008):
the reactor design (extractor, distributor or contactor), the membrane
used in the reaction (organic, inorganic, porous or dense membrane),
whether it is an inert or catalytic membrane reactor and the reaction
that occurs in membrane reactor (such as dehydrogenation (Caro,

1368

S.H. Shuit et al. / Biotechnology Advances 30 (2012) 13641380

2008), esterication (Buonomenna et al., 2010; Caro, 2008), water

dissociation (Caro, 2008) or wastewater treatment (Drioli et al.,
2008)). In addition to providing the separation, a membrane reactor
also enhances the selectivity and yield of the reaction (Sanchez
Marcano and Tsotsis, 2002). As illustrated in Fig. 4, there are two
basic congurations of membrane reactor (Lipnizki et al., 1999a).
Fig. 4A shows the layout of a membrane reactor system in which the
membrane reactor appears as an external process unit. On the other
hand, the membrane reactor shown in Fig. 4B combines the reactor
and membrane separator into a single unit. In comparison to the
conventional biodiesel production process, the main advantage
offered by the membrane reactor, especially the integrated membrane
system, is the reduction of the capital and operating costs because of
the elimination of the intermediate processing steps (Sanchez
Marcano and Tsotsis, 2002). Recently, the membrane reactor has
been applied as a promising technology in biodiesel production
(Baroutian et al., 2010; Cao et al., 2008b; Dub et al., 2007).
5. Membrane technology in biodiesel production
5.1. Concepts and principles
Membrane separation involves the use of a selective barrier
(membrane) to regulate the transport of substances, such as gases,
vapours and liquids, at different mass transfer rates (Bddeker,
2008; Sirkar and Ho, 1992). The mass transfer rates of different
substances are controlled by the permeability of the barrier toward
the feed components (Bddeker, 2008). In the production of biodiesel,
the membrane plays an important role by removing glycerol from the
product (biodiesel) stream (Guerreiro et al., 2006; Saleh et al., 2010)
or retaining the un-reacted triglycerides within the membrane
(Baroutian et al., 2011; Cao et al., 2008b; Dub et al., 2007) as shown
in Figs. 5 and 6 respectively. There are two basic principles of operation
in biodiesel production via membrane technology; separation based on
oil droplet size (Cao et al., 2008a, 2008b) or catalytic membrane
(Guerreiro et al., 2006, 2010; Shao and Huang, 2007). Pervaporation
also seems applicable to biodiesel production.
5.1.1. Membrane separation based on oil droplet size
Membrane separation based on oil droplet size requires a microporous membrane, which is typically a ceramic membrane (Baroutian
et al., 2010, 2011; Cao et al., 2008a) or a carbon membrane (Dub et
al., 2007). The operation principle of the membrane used in a membrane
reactor for biodiesel production is illustrated in Fig. 7 (Dub et al., 2007).
Because of the difference in polarity, methanol is immiscible with oils
and lipids (Cao et al., 2008a; Shuit et al., 2010a). Therefore, a mixture
of methanol and lipid will exist in a two-phase system or as an emulsion
of lipid droplets suspended in a methanol rich phase (Cao et al., 2008a;
Dub et al., 2007). The immiscibility of the lipid and the methanol is the
main cause of the mass transfer limitation in the transesterication

Glycerol
Biodiesel
Methanol
Triglycerides

Fig. 5. Schematic diagram of membrane to remove glycerol from the product stream.

reaction, but this emulsied system is favoured in the operation of a

membrane reactor (Dub et al., 2007). In the emulsied system, transesterication is believed to occur at the interface between lipid droplets
and the continuous methanol phase in which they are dispersed
(Ataya et al., 2006). It has been reported that biodiesel and glycerol, as
well as the catalysts (both acid and alkaline catalysts), are soluble in
methanol (Cao et al., 2008a; Zhou et al., 2006). Thus, the unreacted
lipids will be suspended and dispersed in a mixture of methanol, biodiesel, glycerol and catalyst on the membrane retentate side (Cao et al.,
2008a). Because of its smaller molecular size, methanol and other
soluble components, such as biodiesel, glycerol and catalysts, are able
to pass through the microporous membrane into the permeate stream
when the transmembrane pressure (TMP) is increased (Baroutian et
al., 2010). Meanwhile, the emulsied lipid droplets with larger molecular
size are trapped within the membrane to be continuously converted into
biodiesel (Baroutian et al., 2010; Cao et al., 2008a; Dub et al., 2007).
5.1.2. Membrane separation based on catalytic membrane
Membrane separation based on the catalytic membrane involves a
non-porous dense polymer membrane, such as poly(vinyl alcohol)
(PVA) (Guerreiro et al., 2006, 2010; Shi et al., 2010). The operation of
this type of membrane is based on the interaction between the target
component and the polymer functional groups of the membrane
(Guerreiro et al., 2006). In biodiesel production via this type of catalytic
membrane, glycerol and methanol are able to form hydrogen bonds
with the OH groups in the polymer membrane (Guerreiro et al.,
2006). Therefore, the glycerol and methanol are continuously removed
from the mixture during the reaction (Guerreiro et al., 2006; Saleh et al.,
2010). Meanwhile, the unreacted lipids and the produced biodiesel are
retained within the membrane because of their difference in chemical
properties with the polymer group of the membrane. In this case, the
separation is carried out under atmospheric pressure (Guerreiro et al.,
2006).
5.1.3. Membrane separation based on pervaporation
Separation by pervaporation does not rely on the relative volatilities
of the components but on the relative rates of permeation through a
membrane. Pervaporation is also performed with a non-porous dense
membrane that is usually made from a polymer or zeolite (Shao and
Huang, 2007; Sharma et al., 2004). Therefore, pervaporation has always
been hailed as clean technology to replace conventional energyintensive separation processes, such as evaporation and distillation
(Sae-Khow and Mitra, 2010). Pervaporation is most often applied to

Glycerol
Biodiesel
Methanol
Triglycerides

Fig. 4. Basic layout of membrane reactor (A) a conventional membrane reactor system
(B) an integrated membrane reactor system (Lipnizki et al., 1999a).

Fig. 6. Schematic diagram of membrane to retain un-reacted triglycerides within the

membrane.

S.H. Shuit et al. / Biotechnology Advances 30 (2012) 13641380

Fig. 7. Separation of oil and FAME by micro-porous membrane (Dub et al., 2007).

the dehydration of organic solvents, the removal of organic compounds

from aqueous solutions and the separation of organicorganic mixtures
(Khayet et al., 2004). However, based on the concept of the catalytic
membrane, pervaporation is believed to be a possible operation principle
in biodiesel production. The concept and applications of pervaporation
have been reviewed in detail in several articles (Lipnizki et al., 1999a,
1999b; Pangarkar and Pal, 2008; Sae-Khow and Mitra, 2010; Shao and
Huang, 2007).
The pervaporation process is distinct from other membrane processes
because it combines permeation and evaporation in a single module.
A phase change occurs for molecules that permeate through the
membrane toward the downstream site (Pangarkar and Pal, 2008). By
applying a lower pressure at the permeate side of the membrane, a
driving force is created to remove target solutes from the solution
mixture (Sae-Khow and Mitra, 2010). Solution-diffusion is the wellrecognised mechanism to describe mass transport through nonporous membranes (Lipnizki et al., 1999b; Sae-Khow and Mitra, 2010;
Shao and Huang, 2007). The permeation of solute molecules through
the membrane occurs in ve main steps, which are shown in Fig. 8
(Sae-Khow and Mitra, 2010). First, the solutes in the reaction mixture
diffuse through the liquid boundary layer of the membrane feed (PL1
to PL2). At the membrane-liquid interface, specic solutes are selectively
partitioned into the membrane (PL2 to PM1). Under a pressure difference,
the solute molecules diffuse across the membrane (PM1 to PM2)
(Sae-Khow and Mitra, 2010). Next, the desorption of solute molecules
into the vapour phase occurs at the downstream surface of the lm
(PM2 to PV1) (Pangarkar and Pal, 2008; Sae-Khow and Mitra, 2010).
Lastly, the gas molecules of the solute diffuse away from the membrane
through the boundary layer on the permeate side (PV1 to PV2)
(Sae-Khow and Mitra, 2010). The sorption of solutes into the membrane
depends on the interaction between the solutes and the polymer groups

Fig. 8. Permeation of solute molecules through non-porous dense membrane (SaeKhow and Mitra, 2010).

1369

in the membrane (Pangarkar and Pal, 2008). Therefore, glycerol

molecules have a high probability of being selectively partitioned by
the membrane because hydrogen bonds are formed between the glycerol
molecules and the OH groups of the polymer membrane (Guerreiro et al.,
2006).
Flux and selectivity are the two most important parameters in
pervaporation. Flux is expressed in terms of the partial pressure
difference across the two sides of the membrane, and the concentration
gradient or vapour pressure difference is maintained either by keeping
a constant vacuum on the permeate side or by introducing a sweep gas
to depress the partial pressure (Sae-Khow and Mitra, 2010). The selectivity of solutes is governed by sorption and diffusion, depending on the
solute. Sorption depends on the solubility parameter of the solutes and
the membrane material. Apart from the physical properties of the
solutes, such as the size, shape and molecular weight, the availability
of inter/intramolecular free space in the polymer also affects the
diffusion coefcient (Pangarkar and Pal, 2008). The last two steps in
pervaporation, the desorption step (PM2 to PV1) and the diffusion of
gas phase from the membrane through the boundary layer (PV1 to
PV2), are rapid and nonselective, offering the least resistance in the
overall transport process (Pangarkar and Pal, 2008).
5.2. Possible combinations of membrane and catalyst in biodiesel
production
Catalytic membranes in biodiesel production can be classied into
two categories: membranes that do not incorporate catalyst and
membranes that do incorporate catalyst. In addition, the potential
application of mixed matrix membrane (MMM) with embedded
functionalised carbon nanotubes (CNTs) in biodiesel production will
be discussed. The role of the membrane in this particular conguration
is as a medium to provide intimate contact between the oil and the
alcohol (Buonomenna et al., 2010).
5.2.1. Membrane without incorporated catalyst
This type of noncontact conguration between the membrane and
the catalyst is also known as the catalytically inert membrane
(Buonomenna et al., 2010) in which the catalysts are added to the
reactants but not embedded inside the membrane. The most common
catalytically inert membranes in biodiesel production are the TiO2/
Al2O3 in ceramic membrane (Baroutian et al., 2010, 2011), ltanium
ceramic membrane (Cao et al., 2008a, 2008b) and carbon membrane
(Dub et al., 2007) with the separation concept based on oil droplet
sizes. The pore sizes of these membranes range from 0.02-0.05 m
(Baroutian et al., 2010; Cao et al., 2008b; Dub et al., 2007). The
catalysts used for catalytically inert membranes include sulphuric
acid (H2SO4) (Dub et al., 2007) or potassium hydroxide/sodium
hydroxide (KOH/NaOH) (Baroutian et al., 2010; Cao et al., 2008a,
2008b). The schematic diagram for the transesterication reaction
via catalytically inert membrane is shown in Fig. 9 (Baroutian et al.,
2010; Cao et al., 2008a, 2008b). Initially, a pre-determined amount
of oil and a homogeneous mixture of methanol/KOH are charged
into a mixing vessel for pre-mixing. Next, the reaction mixture is
heated to the desired reaction temperature before charging into the
membrane reactor. The permeate stream consists of FAME (biodiesel),
glycerol, methanol and catalysts (Baroutian et al., 2010; Cao et al.,
2008a; Dub et al., 2007). Oil droplets with molecular size of 12 m
(larger than the pore size of membrane) (Cao et al., 2008b;
DeRoussel et al., 2001) are trapped on the retentate side and recycled
back into the mixing vessel (Cao et al., 2008b). The backpressure valve
and cooler bring the permeate stream to atmospheric conditions (Cao
et al., 2008b). The permeate stream can subsequently be separated
into non-polar and polar phases (Cao et al., 2008a). The non-polar
phase (collectively known as the FAME-rich phase) consists of more
than 85% FAME, and the remainder consists of methanol, trace amount
of DG and catalysts (Cao et al., 2008b). In order to comply with the

1370

S.H. Shuit et al. / Biotechnology Advances 30 (2012) 13641380

Safety and relief valves

Heat exchanger
Purge and
quench tank

Coriolis meter

Retentate stream
Mixing vessel

Membrane module
Back pressure valve
Vacuum pump

Biodiesel

Feed pump
Permeate stream

Phase separation
Cooler

FAME phase
Polar phase

Oil feed tank

Methanol/catalyst
feed tank

Purge
Recycled back

Fig. 9. Schematic diagram of transesterication reaction via catalytically inert membrane (Baroutian et al., 2010; Cao et al., 2008a, 2008b).

American Society for Testing and Materials (ASTM) or European

Standards (EN) standards for biodiesel, further purication of the
FAME-rich phase is conducted to remove methanol, DG and catalysts
(Cao et al., 2008a). Meanwhile, the polar phase, which is also known
as the glycerine-rich phase, contains a mixture of glycerol, methanol,
catalysts and FAME (Cao et al., 2008b). Results have shown that this
catalytic membrane reactor was capable of achieving a high oil-toFAME conversion of more than 90% for both H2SO4 and KOH catalysts
(Dub et al., 2007).
Methanol that permeates through the membrane is recycled back
to the membrane reactor in order to reduce the overall methanol to
oil molar ratio (Cao et al., 2006). There are two ways to recycle
methanol back to the reactor: recycling of the methanol from the
distillation of the FAME-rich phase (Baroutian et al., 2010) and direct
recycling of the glycerine-rich phase (polar phase) (Cao et al., 2008b).
It has been reported that more FAME is distributed into the FAMErich phase by direct recycling of the glycerine-rich phase. Therefore,
this process can facilitate the production of higher purity FAME and
reduce the amount of water required for the FAME purication
process (Cao et al., 2008b).
A packed bed membrane reactor consisting of a catalyst supported
by activated carbon (Baroutian et al., 2011) was used to avoid the
permeation of catalysts through the membrane. The catalysts were
prepared by adding activated carbon ranging from 550810 m in
size into a potassium hydroxide solution. The mixture was subsequently agitated at a temperature of 25 C for 24 h. Next, the catalysts
were packed inside the tubular TiO2/Al2O3 ceramic membrane reactor
(Baroutian et al., 2011). The highest oil to FAME conversion for this
packed bed membrane reactor was 93.5% (Baroutian et al., 2011),
which was comparable to the conversion achieved by the membrane
reactor with the addition of H2SO4 or KOH catalysts. Moreover, it has
been reported that high-quality biodiesel was produced from such a
reactor without washing or purication steps (Baroutian et al.,
2011).
5.2.2. Membrane with incorporated catalyst
A membrane that incorporates catalyst in which the catalyst is
immobilised in the polymeric matrix is commonly known as a catalytically active membrane (Buonomenna et al., 2010). Polymeric
membranes (Guerreiro et al., 2006, 2010; Zhu et al., 2010) are usually
used as catalytically active membranes (Sarkar et al., 2010). A

membrane can be made catalytically active by heterogenisation of

homogeneous catalysts or incorporation of heterogeneous catalysts
inside the polymer matrix (Buonomenna et al., 2010). The catalytically
active membrane combines reaction and separation in a single step,
realising the concept of reactive separation (Buonomenna et al.,
2010); for this reason, the membrane is known as a separative reactor
(Stankiewicz, 2003). Presently, poly(vinyl alcohol) (PVA) membranes
are the only reported polymer membranes that have been tested in
biodiesel production (Guerreiro et al., 2006, 2010; Sarkar et al., 2010;
Zhu et al., 2010) because of their high hydrophilicity, good thermal
properties and good chemical resistance (Guan et al., 2006).
A PVA membrane must be modied before it can be transformed
into a catalytically active membrane. There are two important steps
in preparing a catalytic PVA membrane: crosslinking of PVA followed
by esterication of the free PVA-OH groups (Guerreiro et al., 2006,
2010). Sulphosuccinic acid (Guerreiro et al., 2006), succinic acid
(Castanheiro et al., 2006), fumaric acid (Guan et al., 2006), maleic
acid (Figueiredo et al., 2008) and glutaraldehyde (GA) (Wang and
Hsieh, 2010) can be used as the membrane crosslinking agents.
Higher degrees of crosslinking can enhance the thermal stability of the
membrane (Guan et al., 2006), but they can also cause the membrane
to be less hydrophilic and more brittle (Kim et al., 1994). In biodiesel
production, increased crosslinking can reduce the degree of membrane
swelling in oil and methanol, thereby reducing the biodiesel yield because oil and methanol are prohibited from diffusion into the membrane in the catalytic reaction (Guerreiro et al., 2006). Esterication of
the free PVA-OH groups by 5-sulphosalicylic acid (SA) (Castanheiro et
al., 2006; Guerreiro et al., 2006) is an important step in making the
membrane catalytically active because sulphonic (SO3H) functional
groups are introduced into the polymer matrix (Castanheiro et al.,
2006). The modied PVA membrane shows higher catalytic activity
with larger amounts of SA because of the increased amount of SO3H
groups in the polymer matrix (Castanheiro et al., 2006). In addition to
functionalisation with SA, PVA can be transformed into a catalytic
membrane by blending with poly(styrene sulphonic acid) (PSSA),
which contains strong acidic groups (Zhu et al., 2010). It has been
reported that 92% conversion of oil into FAME can be achieved by a
PSSA/PVA membrane in 8 h of reaction time. In addition to functionalisation with sulphonic groups, the annealing temperature is also of
critical importance during the synthesis of the membrane because it
controls the degree of crosslinking and the number of SO3H groups

S.H. Shuit et al. / Biotechnology Advances 30 (2012) 13641380

available in the membrane. Such a blended membrane also showed a

stable conversion of 80% after 5 repeated runs (Zhu et al., 2010).
Heterogeneous catalysts can also be embedded into the polymer
matrix in place of homogeneous catalysts. Hydrotalcite, Mg6Al2
(OH)16(CO32-).4H2O is a suitable solid base catalyst for biodiesel
production because of its large specic surface area (Bastiani et al.,
2004) and strong Lewis basicity (Roelofs et al., 2000). This catalytic
membrane is prepared by dispersing 1 g of hydrotalcite into a 10%
PVA solution. The membrane showed a promising yield of biodiesel
(more than 95%) (Guerreiro et al., 2010). Polyacrylonitrile (PAN) is
another potential polymer membrane material that can be applied
to the biodiesel process. However, the only reported use of a PAN
membrane has been the purication of the produced biodiesel by
removing free glycerol (Saleh et al., 2010). Because of the lack of
functionality in the PAN membrane, it must be modied either by
crosslinking with SA or blending with a polymer containing SO3H
groups before it can be transformed into a catalytic membrane.
Polymeric membranes suffer from a lack of chemical and thermal
stability (Ismail et al., 2009) and are easily broken (Guerreiro et al.,
2006). These deciencies and the high fabrication cost of inorganic
membranes (Ismail et al., 2009) have encouraged the development
of the more capable mixed matrix membrane (MMM). In comparison
to competing materials, MMM offers promising fabrication cost,
mechanical strength and chemical and thermal stability. MMM is a
heterogeneous membrane that incorporates an inorganic ller in a
polymer matrix. MMM combines the superior permeability and
selectivity of inorganic membranes with the economical processing
capabilities of polymeric membranes. The rigid, adsorptive and
porous inorganic phase in the MMM offers good separation properties,
and the presence of the exible polymer makes membrane forming
easier, solving the fragility problems encountered by inorganic
membranes (Ismail et al., 2009). The most common inorganic llers
for MMMs include zeolites (Mahajan et al., 1999), carbon molecular
sieve (CMSs) (Peng et al., 2006) and CNTs (Peng et al., 2007). The
unique properties of CNTs, such as their high aspect ratio and surface
area, simple functionalisation and dispersion in organic polymer,
enhance the mechanical strength of the MMM with minimal ller
content. The simplication of pore dimension control at the nanometre
scale (Ismail et al., 2009) has made CNTs a suitable inorganic ller
material in the polymer matrix. In addition to improving the physical
properties of the membrane, the functionalised CNTs can also act as a
catalyst for the transesterication reaction. Recent studies have shown
the capability of amino-functionalised CNTs to serve as a solid base
catalyst for transesterication reactions (Villa et al., 2009, 2010). Aminofunctionalised CNTs are simply known as nitrogen-functionalised CNTs.
Different amide groups, including triethylamine, ethylamine and pyrrolidine, can be readily grafted into CNTs (Villa et al., 2009). Triethylamine
has higher basicity than other amide groups, making it the most active
catalyst, which is able to achieve almost complete conversion under certain reaction conditions (Villa et al., 2009). These amino-functionalised
CNTs have shown extremely stable catalytic activity, which could obtain
greater than 90% conversion, even after 6 reaction cycles (Villa et al.,
2009). Therefore, mixed matrix polymer membranes with embedded
functionalised CNTs could represent a breakthrough for applications in
biodiesel production.
Such properties as the membrane thickness, swelling capability
and active site concentration play an important role in enhancing
the biodiesel yield. It has been observed that membranes that are
capable of swelling in oil give higher biodiesel yields (Guerreiro et
al., 2006, 2010). This improvement is caused by the increased oil
concentration in the membrane, which leads to higher catalytic activity
(Guerreiro et al., 2006). The concentration of active sites and the thickness of the catalytic polymer membrane were reported to fall within the
range of 1.263.80 mmol/g and 0.040.13 mm, respectively (Guerreiro
et al., 2006; Zhu et al., 2010). The reported basic site concentration of
the heterogeneous catalysts (CNTs and activated carbon) that have

1371

the potential to be incorporated into the membranes were found to be

between 1.00 (Villa et al., 2009) and 1.58 (Baroutian et al., 2011)
mmol/g.
A schematic diagram of the transesterication reaction via catalytically active membrane is shown in Fig. 10. A pre-determined amount
of oil and methanol were mixed, heated and pumped into a membrane
reactor. Glycerol was continuously removed from the reaction mixture
once it was produced. The permeate stream contained a binary mixture
of glycerol/methanol, which was recovered in cold trap immersed in
liquid nitrogen. Meanwhile, the retentate that contained unreacted oil
was returned to the mixing vessel to be circulated back into the
membrane reactor for further reaction (Figueiredo et al., 2008;
Guerreiro et al., 2006).
Glycerol and methanol are able to permeate through the PVA
membrane because of the hydrogen bonds formed between the glycerol
and methanol molecules and the OH groups in the polymer. It has been
reported that no oil or FAME were detected in the permeate stream,
indicating that product loss can be avoided (Guerreiro et al., 2006). As
compared to the catalytically inert membrane, the advantage of this
catalytically active membrane is the elimination of the purication
process for the post-reaction permeate stream.
6. Effect of process parameters in biodiesel production by membrane
reactor
In addition to the typical process parameters (reaction temperature, methanol to oil ratio and catalyst concentration), other process
variables, such as the reactant ow rate, trans-membrane pressure,
membrane thickness and pore size (for membrane separation based
on oil droplet selection), also have a great inuence on the biodiesel
yield and need to be taken into consideration when biodiesel is
produced by membrane technology. In order to produce biodiesel in
a more sustainable and cost effective manner, the important process
parameters that should be taken into consideration will be discussed
in the following section.
6.1. Effect of reaction temperature
In order to reduce the total reaction time, a higher reaction
temperature (without evaporation of methanol) is required for transesterication (Cheng et al., 2010). Therefore, for transesterication to
take place in a stirring batch reactor, the reaction temperatures for
homogeneous acid and base catalysed transesterication should be
80 C and 25120 C, respectively (Marchetti et al., 2007). However,
for a membrane reactor in which the separation is based on the oil
droplet size, a lower reaction temperature between 5070 C is used
to synthesise biodiesel (Baroutian et al., 2011; Cao et al., 2008a;
Dub et al., 2007). The reason that the reaction temperature is kept
as low as possible in the membrane reactor is to encourage the existence of a two-phase system between the methanol and the lipid. At elevated temperatures, the system tends to be homogeneous. As
mentioned earlier, transesterication is believed to occur at the surfaces
of the oil droplets that are suspended in the methanol stream; therefore,
the resulting heterogeneous phases are needed for the operation of the
membrane reactor. Moreover, the purity of FAME is reduced as the
reaction temperature is increased because the solubility of oil and
other intermediates in FAME increase with temperature (Cheng et al.,
2010).
Transesterication with a minimum temperature of 60 C has
been reported with the use of a dense membrane (Guerreiro et al.,
2006, 2010). A possible reason for this result is the thermal mobility
of the molecules inside the membrane, which increases at higher
reaction temperatures and thus generates extra free volume space
(Ong et al., 2011), enhancing the permeation of larger molecules
such as glycerol. Moreover, glycerol demonstrates a signicant
decrease in viscosity at higher reaction temperatures. If the viscosity

1372

S.H. Shuit et al. / Biotechnology Advances 30 (2012) 13641380

Heat exchanger

Mixing vessel

Retentate stream

Membrane module

Permeate stream
Three way valve
Feed pump
Cold trap immersed
in liquid nitrogen

Oil feed tank

Cold trap immersed

in liquid nitrogen

Methanol feed tank

Vacuum pump

Fig. 10. Schematic diagram of transesterication reaction via catalytically active membrane (Figueiredo et al., 2008; Guerreiro et al., 2006).

is reduced, the circulation of the mixture becomes easier (Gomes et

al., 2010).
Irrespective of the type of membrane used to synthesise biodiesel,
the conversion of oil to FAME was found to be positively affected by
increasing the reaction temperature (Dub et al., 2007). This increase
can be easily justied because transesterication is an endothermic process (Samart et al., 2009). Therefore, from Le Chatelier's principle, by
increasing the temperature, the equilibrium of the reaction is shifted
to the forward direction, which favours the conversion of oil into FAME.
6.2. Effect of methanol to oil ratio
It has been proven that at higher methanol to oil ratios, the time
required for the reaction mixture to achieve a homogeneous liquid
system increases. This relationship indicates that at higher methanol
to oil ratios, the two-phase system can be maintained so that the
oil-rich phase could not pass through the membrane (Cheng et al.,
2010). Unlike the reaction temperature, a higher methanol to oil
ratio favours the production of biodiesel in a membrane reactor. A
methanol to oil ratio of 24:1(in molar ratio) or 1:1 (in volume ratio)
is usually used for the production of biodiesel in a membrane reactor
in which the separation is based on oil droplet size (Baroutian et al.,
2011; Cao et al., 2008a; Cheng et al., 2010). The methanol to oil
ratio in a catalytic based membrane was reported to be 6:1 (Shi et
al., 2010), 26:1 (Zhu et al., 2010), 106:1 (Guerreiro et al., 2006) and
254:1 (Guerreiro et al., 2010). However, no clear explanation was
given of the need for such a large amount of methanol to produce
biodiesel in the former method. Regardless of the operational principle
of the membrane, the conversion increased in proportion to the ratio of
methanol to oil because the reaction was driven towards biodiesel
production (Shi et al., 2010).
6.3. Effect of catalyst concentration
The emulsion of oil molecules in the alcohol stream may cause a
mass-transfer problem in a membrane reactor, especially at a high
methanol to oil ratio. Therefore, a higher catalyst concentration is
required to achieve the complete conversion to oil. It has been
reported that for 20 min of reaction time, oil conversions of 61.1%
and 100% were achieved for catalyst concentrations of 0.05 and
0.5 wt.%, respectively. Despite the low catalyst concentration of 0.05%,

the reaction was still capable of achieving complete oil conversion but
required a longer residence time (1 h) (Cheng et al., 2010; Tremblay
et al., 2008). For the acid catalyst (H2SO4), oil conversion was signicantly increased when H2SO4 concentration was increased from 0.5 to
2%. However, oil conversion was not signicantly different (less than
10%) when H2SO4 concentration was further increased to 4 and
6 wt.%. This result implies that the high concentration of the acid catalyst was not necessary in the membrane reactor (Dub et al., 2007).
For similar concentrations and reaction times, base catalysts provided
much higher oil conversion than acid catalysts because of the faster reaction of the base catalyst (Dub et al., 2007). The catalytic membrane
also exhibited the same results; the membrane showed higher catalytic
activity with an increased concentration of SO3H groups embedded in
the polymer matrix (Castanheiro et al., 2006). The amount of SO3H
groups in the polymeric membrane depends on the degree of membrane crosslinking with succinic acid (Guerreiro et al., 2006) and the
percentage of free OH groups in the membrane to be esteried with
5-sulphosalicylic acid (Castanheiro et al., 2006).
6.4. Effect of reactant ow rate
There has been no detailed study on the effect of the reactant ow
rate on biodiesel production in a membrane reactor. However, a
signicant increase in the conversion of oil to FAME could be observed
as the ow rate of reactants increases (Baroutian et al., 2011; Dub et
al., 2007). This could be caused by an improvement of the mixing
intensity at higher ow rate or greater ow circulating velocity
(Baroutian et al., 2011) because the reactants (oil/methanol) and
catalyst will be owing in a turbulent ow regime (Vospernik et al.,
2004). Mixing is a crucial factor for the increase of the reaction rate in
transesterication because oil is immiscible with methanol. Without
mixing, the reaction only occurs at the interface between the layers of
methanol and oil (Kumar et al., 2010; Meher et al., 2006).
Concentration polarisation is a common problem for membrane
separation. Concentration polarisation is caused by the accumulation
of retained solute at the membrane interface, forming a secondary
layer that restricts the transport of the permeating species (Porter,
1972). Concentration polarisation reduces the permeation rate of
the more permeable component but favours the permeation of the
less permeable components (Bakhshi et al., 2006). This problem can
be easily solved by increasing the reactant ow rate. With the

S.H. Shuit et al. / Biotechnology Advances 30 (2012) 13641380

increase of this rate, the resulting turbulent ow will reduce the

thickness of the boundary layer caused by the retained solute. Therefore,
the mass transfer resistance at the boundary layer on the upstream of
the membrane can be reduced, leading to an increase in the total permeation ux (Peng et al., 2007).
6.5. Effect of trans-membrane pressure (TMP)
TMP is dened as the pressure difference between the feed side
and the permeate side of the membrane (Nishimoto et al., 2010).
TMP is the driving force for membrane separation (TakhtRavanchi
et al., 2008) and is normally kept constant throughout the process
(Nishimoto et al., 2010). The TMP used in the production of biodiesel
ranges between 37.9 and 173.4 kPa (Cao et al., 2008a, 2008b; Dub et
al., 2007). The permeate ux increases with higher TMP because a
greater driving force is applied for separation (Gomes et al., 2010).
In dense polymeric pervaporation membranes, TMP affects the
sorption and desorption rates across the membrane. At higher TMP,
the driving force drawing the molecules across the membrane and
sweeping them away from the permeate side is increased, resulting
in higher mass transfer rates (Ong et al., 2011).
In biodiesel production via membrane reactor, TMP is possitively
affected by the concentration of unreacted oil in the emulsion within
the reactor. If the residence time of the reaction is insufcient, the
concentration of the unreacted oil in the reactor will increase. Therefore,
it was hypothesized that TMP is dominated by the residence time of the
transesterication reaction. A residence time of 60 min for rened,
bleached and degummed (RBD) oil and 80 min for waste cooking oil is
required to produce a stable and constant TMP. Under these residence
times, the resulted TMP generated for RBD oil and waste cooking oil
was 80 and 90 kPa respectively (Falahati and Tremblay, 2012).
Besides, TMP is correlated with the viscosity of the reaction
mixture. This relation is clear when the glycerine-rich phase is
recycled back to the reactor for methanol recovery. In other words,
an increased content of glycerol in the reaction mixture would
require higher TMP. However, the increase of TMP from increased
glycerol content does not cause any negative effect in the membrane
reactor. Assuming complete oil conversion in the batch reaction, the
glycerol content would be 8.5% in the FAME/glycerol/methanol mixture,
and the corresponding TMP would only be 90 kPa. This TMP value is far
below the recommended membrane operating pressure of 1000 kPa
(Cao et al., 2008b).
The TMP proles in the membrane reactor can also act as an
indicator to check the progress of the transesterication reaction. A
sharp increase in the TMP indicates that transesterication is not
occurring, and the oil has become a continuous phase within the
membrane reactor. On the other hand, a constant and stable TMP
prole for all operating times reveals that a sufcient amount of oil
has been transesteried into FAME, allowing the continuous operation
of the membrane reactor (Tremblay et al., 2008).
6.6. Effect of membrane pore size and thickness
For a membrane reactor with separation based on the molecular
size of the components, the selection of a membrane with a suitable
pore size becomes extremely important in order to forbid oil droplets
from passing through the membrane. It has been proven that no oil is
found in the permeate stream when a membrane with a pore size
between 0.051.4 m is used to produce biodiesel. This separation
occurs because the reported average size of the oil droplets falls in
the range of 12400 m, which is much larger than the membrane
pore size (Cao et al., 2006).
The thickness of the membrane is an additional vital factor that
needs to be taken into consideration for the use of catalytic membranes
in biodiesel production. Zhu et al. (2010) reported that the membrane
thickness used in transesterication falls in the range of 0.04

1373

0.26 mm. They also reported that the transesterication rate increased
at the initial stage of the reaction, as the membrane thickness
decreased. However, the same oil conversion could be achieved by
membranes of any thickness by the end of the reaction because the
catalytic membrane had become swollen by reactants, and the active
sites of the catalyst contained by the membrane that could be exposed
to reactants for catalytic conversion (Zhu et al., 2010).
7. Advantages of catalytic membrane reactor in biodiesel production
The catalytic membrane reactor is a new technology for biodiesel
production. This technology can offer an alternative to overcome the
common limitations arising from conventional biodiesel production
processes. The advantages of the catalytic membrane reactor for
biodiesel production will be discussed in the following section.
7.1. Environmentally friendly process
The production of biodiesel via catalytic membrane reactor is
undeniably an environmentally friendly process because of its low
energy consumption. Transesterication in a catalytic membrane
reactor is carried out under mild operating conditions. The highest
reported reaction temperature in a membrane reactor was 70 C
(Dub et al., 2007), which is quite similar to the conventional homogeneous transesterication (65 C) (Berchmans and Hirata, 2008) but
much lower than either heterogeneous or supercritical transesterication. The reaction temperature for transesterication using a solid
basic catalyst, such as magnesium oxide (MgO), calcined hydrotalcite
(CHT), zinc oxide (ZnO), KNO3/KL zeolite and KNO3/ZrO2, falls in the
range of 180200 C (Di Serio et al., 2006; Jitputti et al., 2006). It has
been reported that at approximately, 100 C, and alkaline catalysts
exhibited very low catalytic activity that only produced a FAME
yield of 20% (Di Serio et al., 2006). Additionally, transesterication
by solid acid catalysts like tungstated zirconia (WO3/ZrO2), sulphated
tin oxide (SO4/SnO2), sulphated zirconiaalumina (SZA) and sulphated
zirconia (SO42 /ZrO2) were carried out in the range of 200300 C
(Chen et al., 2007; Furuta et al., 2004; Jitputti et al., 2006). In addition,
there is more hidden energy required in the synthesis of heterogeneous
catalysts because most heterogeneous catalysts must be calcined at
high temperatures, ranging from 200500 C (Albuquerque et al.,
2008; Lu et al., 2009). Unlike solid catalysts, catalytically active membranes (for example, functionalised PVA membranes with sulphonic
groups) are fabricated in a low temperature environment (Guerreiro
et al., 2006). Of all of the reported biodiesel production methods, supercritical transesterication requires the most extreme reaction temperature (240340 C) and reaction pressure (5.78.6 MPa) (Hawash et al.,
2009). As compared to biodiesel produced in a catalytic membrane
reactor (70 C and 173.4 kPa), the reaction temperature and pressure
required for the supercritical process are 5 and 50 times higher,
respectively.
From the perspective of chemical requirements, the catalytic
membrane reactor could reduce the usage of solvents and chemicals
that are harmful to the environment. For the conventional production
method, the reported concentration for the alkaline catalyst is in the
range of 0.51% (NaOH) (Marchetti et al., 2007). The concentration
of the acid catalyst varied from 14%, depending on the FFA content
in the oil (Narasimharao et al., 2007; Wang et al., 2006b). Compared
to the catalyst concentration in the conventional methods, the use
of catalysts in the catalytic membrane reactor is lower: 0.05% for the
basic catalyst (Tremblay et al., 2008) and 2% for the acid catalyst
(Dub et al., 2007). The catalytically inert membrane reactor and
some catalytically active membranes also consume much less methanol than supercritical technology in which the methanol to oil ratio is
normally higher than 40 (Barnwal and Sharma, 2005; Sharma and
Singh, 2009). Table 1 summarises the reaction conditions and performances for various types of transesterication processes.

1374

S.H. Shuit et al. / Biotechnology Advances 30 (2012) 13641380

Table 1
Comparison of reaction conditions and performances of various biodiesel production methods.
Transesterication Reaction conditions
processes
Catalyst used

Catalyst
Temperature, Methanol
concentration,
C
to oil
molar ratio wt.%

Purication
Free
of FAME
fatty acid
limitation, %

Yield, %

Reference

NaOH

60

6:1

b1

Repeated
washing
(Atadashi et al.,
2011)

97.1

(Rashid et al.,
2008)

Homogeneous acid
Waste cooking oil H2SO4

95

20:1

No
limitation

Conversion > 90%

(Wang et al.,
2006b)

Soybean oil

H2SO4

65

30:1

No
limitation

Conversion > 99%

(Narasimharao
et al., 2007)

Heterogeneous
base

MgO (III)

200

11:1

Unknown

>95

(Di Serio et al.,

2006)

CHT

200

11:1

Unknown

>95

(Di Serio et al.,

2006)

KNO3/KL zeolite

200

6:1

Unknown

77.2

(Jitputti et al.,
2006)

KNO3/ZrO2

200

6:1

Unknown

65.5

(Jitputti et al.,
2006)

ZnO

200

6:1

Unknown

Repeated
washing
(Atadashi et al.,
2011)
Repeated
washing
(Atadashi et al.,
2011)
Simple washing
(Atadashi et al.,
2011)
Simple washing
(Atadashi et al.,
2011)
Simple washing
(Atadashi et al.,
2011)
Simple washing
(Atadashi et al.,
2011)
Simple washing
(Atadashi et al.,
2011)

77.5

(Jitputti et al.,
2006)

WO3/ZrO2

250

40:1b

6.7b

No
limitation

Conversion > 90%

(Furuta et al.,
2004)

SO4/SnO2

200

6:1

No
limitation

80.6

(Jitputti et al.,
2006)

SZA

300

40:1b

6.7b

No
limitation

80

(Furuta et al.,
2004)

SO42 /ZrO2

230

12:1

No
limitation

>90

(Chen et al.,
2007)

320

43:1

No
limitation

100

(Hawash et al.,
2009)

65

24:1

0.5

b1

A packed bed membrane

reactor with activated
carbon supported catalyst
H2SO4

70

24:1

unknown

70

Flow rate:
6.1 ml/min

143.75 mg/cm3 (mass

of catalyst per unit
volume reactor)
2

No
limitation

Simple washing Conversion = 90%

(Saleh et al.,
2010)

(Dub et al.,
2007)

Zr(SO4)2

65

6:1

Conversion > 90%

64

26:1

Unknown

Conversion > 90%

Succinic acid as crosslinking

agent

60

106:1

No
limitation
No
limitation
No
limitation

Unknown

PSSA

Zr(SO4)2:SPVAa =
1:1(mass ratio)
PSSA:PVA = 1:2 (mass
ratio)
PVA membrane is 20%
crosslinked with
succinic acid

No washingc

94.3b

(Shi et al.,
2010)
(Zhu et al.,
2010)
(Guerreiro et
al., 2006)

Homogeneous
base

Heterogeneous
acid

Supercritical

methanol
Catalytically inert
membrane
Base
NaOH

Acid

Simple washing
(Atadashi et al.,
2011)
Simple washing
(Atadashi et al.,
2011)
Simple washing
(Atadashi et al.,
2011)
Simple washing
(Atadashi et al.,
2011)
No washing

(Cao et al.,
Simple washing 97.7
2008a)
(Saleh et al.,
2010)
No washing
Conversion = 93.5% (Baroutian et
al., 2011)

Catalytically
active
membrane

a
b
c

SPVA = sulphonated poly(vinyl alcohol).

Self-estimation.
No washing is performed because glycerol is removed during transesterication.

S.H. Shuit et al. / Biotechnology Advances 30 (2012) 13641380

Additionally, glycerol is separated from the membrane as it is formed,

eliminating the need for the washing step to remove free glycerol
content from the FAME phase.

Even though the operating parameters are similar to those of

homogeneous transesterication, the problem of wastewater generation can be greatly reduced if biodiesel is produced through membrane
technology. In homogeneous transesterication, the crude biodiesel
produced after separation from the glycerol phase still contain catalysts,
unreacted alcohol, soaps and free glycerol (Leung et al., 2010). The common approaches to the purication of the biodiesel include washing
with distilled water or ether or the use of a solid adsorbent. Adsorbent
such as Magnesol can selectively adsorb those hydrophilic materials
such as glycerol, MG and DG. Other solid absorbents like activated
clay, activated carbon and activated bre can also be used to purify
biodiesel (Atadashi et al., 2011). However, washing with hot distilled
or deionised water is the best way to purify biodiesel because both
glycerol and methanol are highly soluble in water (Karaosmano lu et
al., 1996; Leung et al., 2010). In both acid and base-catalysed transesterication, the washing process consists of two steps: neutralisation and
water washing. Hot distilled/deionised water at 6080 C showed
promising performance for washing the FAME phase because of the
higher diffusivity of glycerol from FAME to the water phase at higher
washing temperature (Atadashi et al., 2011). In order to achieve less
than 0.02% glycerol content in the biodiesel, seven consecutive washing
steps have been reported by Gomes et al., 2010. In the acid-catalysed
transesterication, H2SO4 was neutralised by CaO, followed by a gravity
separation to remove the produced CaSO4. CaO was used because its
price is lower than those of other alkali substances. After the removal
of the acid catalyst, the FAME phase undergoes the same washing
process as the material from the base-catalysed transesterication to
remove glycerol, methanol and other contaminants (Zhang et al.,
2003). In the production of biodiesel via catalytically inert membrane,
washing would still be needed to remove the catalyst in the permeate
stream, but fewer washing steps would be required because a lower
catalyst concentration (0.05%) could be used (Tremblay et al., 2008).
In the conventional separation method, 10 L of water are consumed to
wash 1 L of biodiesel. In contrast, only 0.002 L of water per litre
biodiesel would be needed to purify biodiesel produced via the
membrane method (Saleh et al., 2010). Assuming a biodiesel production of 20 million tonnes per year (Licht, 2007) and a biodiesel density
of 900 kg/m 3 (Knothe et al., 2005), approximately 59 billion gallons of
wastewater are produced by the conventional separation method, and
this amount of wastewater could be signicantly reduced to only 12
billion gallons by applying membrane separation to biodiesel production
and purication. The catalytically active membrane has the potential to
eliminate the wastewater problem because the washing step is not
required. As mentioned in Section 5.2.2 the catalyst is embedded in the
polymer matrix, thus, the neutralisation step is not required.

7.2. Lower investment cost

Typical process ow diagrams of biodiesel production in the
conventional process and in the catalytic membrane reactor are
shown in Figs. 11 and 12, respectively. In the catalytic membrane
reactor, both the separation and catalysis processes are combined in
single unit operation (Vankelecom, 2002). The integration of these
processes into a catalytic reactor is able to reduce the number of
operating units, as well as the number of processing steps, thereby
is the leading to a reduction in the size and complexity of the plant
and a consequent reduction of the investment cost (Dittmeyer et al.,
2004). One of the main factors contributing to the high production
cost of biodiesel is the need for downstream processes, which include
biodiesel separation and purication (Hasheminejad et al., 2011). The
catalytically inert membrane has the advantage of not requiring
decantation to separate the two phases obtained after transesterication (Gomes et al., 2011). The catalytically active membrane reactor
has the potential to simplify FAME and glycerol separation, catalyst
neutralisation.
Even though the phase separation between FAME and water can be
easily carried out, the equilibrium solubility of water in FAME after
washing is higher than the water content stated in the international
standard (Gomes et al., 2010) (500 ppm for both ASTM and EN
standard (Knothe et al., 2005). Therefore, vacuum drying is usually required to remove water from the FAME before storage (Gomes et al.,
2010). The neutralisation unit in a conventional production plant
(Sdrula, 2010) could also be eliminated because the catalyst is embedded inside the polymer matrix and would not mix with the reactant.
A combination of centrifugation and water washing is used to
enhance the separation of glycerol and impurities from the FAME
phase. For this method, sufcient residence time is required for the
less dense oil to oat to the surface of the water, thereby resulting
in the preferential separation of the heavy phase. Because glycerol is
fully miscible with water and insoluble in the FAME phase, almost
all of the glycerol is easily removed by this separation method. However, there are several disadvantages to this method, such as its high
initial investment cost, high power consumption and the requirement
for considerable maintenance (Saleh et al., 2010). This separation step
is unnecessary if biodiesel is produced via catalytically active
membrane because glycerol can be simultaneously removed from
Upper layer

Catalyst

Catalyst
mixing

Bottom layer

Neutralising
acid/alkaline

Vacuum
drying

FAME

Washing

Methanol

Oil sources

1375

Transesterification

Crude
biodiesel

Methanol
Distillation

Methanol
Distillation

Phase separation
Neutralisation

Pharmaceutical
glycerin

Gravity settling
Centrifuge

Glycerin
purification

Neutralisation

Crude
glycerin

Fig. 11. Process ow diagram of conventional homogeneous acid/alkaline-catalysed transesterication reaction (Saleh et al., 2010; Sdrula, 2010).

1376

S.H. Shuit et al. / Biotechnology Advances 30 (2012) 13641380

Crude
biodiesel

Methanol
Distillation

FAME

Transesterification in
catalytically active
membrane

Methanol

Crude
glycerin
Oil sources
Methanol
Distillation

Pharmaceutical
glycerin

Glycerin
purification

Fig. 12. Proposed process ow diagram of transesterication reaction in catalytically active membrane reactor.

the reaction mixture, and the need for phase separation between
FAME and glycerol is eliminated.
Furthermore, both the catalytically inert membrane and the catalytically active membrane eliminate the inter-stage temperature and
pressure changes (Dittmeyer et al., 2004) required in biodiesel process,
such as supercritical technology in which the mixture needs to be
cooled before separation of glycerol from the biodiesel can be carried
out (Demirbas, 2007). The elimination of this step indicates that both
the energy and the number of heat exchangers required for the process
could be reduced.

the reaction mixture, while for a dense polymeric pervaporation

membrane, glycerol and methanol are separated into the permeate
stream (Guerreiro et al., 2006). The catalytic membrane reactor can
enhance and increase the overall reaction rate when an enzyme (lipase)
is used as the catalyst. In the conventional lipase-catalysed transesterication, glycerol is easily adsorbed onto the surface of the lipase, reducing the activity and operational stability of the lipase (Su et al., 2007).
Therefore, the continuous removal of glycerol by membrane technology
can decrease the inhibition of lipase, thereby increasing the overall
reaction rate (Vankelecom, 2002).

7.3. Overcoming the limitation caused by chemical equilibrium

7.4. High process exibility of feedstock conditions

Another attractive benet offered by catalytic membrane reactors

for biodiesel production is the ability of the process to overcome the
limitation imposed by chemical equilibrium and achieve complete
conversion. As noted, the transesterication reaction is a reversible
reaction that can never reach 100% completion (Cao et al., 2008b).
The typical conversion for transesterication is 98% or lower
(Knothe et al., 2005). The existence of chemical equilibrium in transesterication is proven by monitoring the progress of a transesterication reaction by bre-optic near-IR spectroscopy with correlation
to 1H nuclear magnetic resonance (NMR). The experiment showed that
in the rst 30 min, approximately 85% of the reaction was completed,
while in the following 30 min, the reaction only proceeded by approximately 7%, indicating that the reaction began to approach chemical
equilibrium (De Boni and Lima da Silva, 2011).
According to Le Chatelier's principle, the equilibrium of the transesterication reaction can be shifted toward higher conversion by
having one reactant in excess or by selectively removing of one of
the products generated in the reaction (Castanheiro et al., 2006).
Therefore, a higher methanol to oil ratio is needed to increase oil
conversion (Shi et al., 2010). Unlike other conventional methods,
the catalytic membrane reactor is capable of driving the transesterication reaction further toward completion by simultaneously removing
the products from the reaction mixture. The separation depends on the
type of membrane used in the catalytic membrane reactor. With a
micro-porous membrane, FAME and glycerol (Baroutian et al., 2010,
2011; Cao et al., 2008a, 2008b; Dub et al., 2007) are separated from

Water and free fatty acid (FFA) found in oil sources can create
signicant problems in transesterication (Atadashi et al., 2011). As
shown in Fig. 13, the presence of water or moisture in the feedstock
can cause hydrolysis of the formed methyl esters back to FFA (Van
Gerpen and Knothe, 2005), resulting in reduced product. At the
same time, water will also hydrolyse triglyceride to diglyceride and
FFA, especially at higher temperatures (as shown in Fig. 14)
(Atadashi et al., 2011). It has been reported that 0.1% water in an oil
source is sufcient to reduce the conversion of oil to FAME during the
transesterication reaction (Demirbas, 2007). In short, the presence of
water will result in the production of more FFA and reduce the FAME
yield.
FFA in the reaction mixture will react with water and an alkaline
catalyst, such as KOH or NaOH, to form a saponied product (soap),
as shown in Fig. 15. The saponied product tends to be strengthened
at ambient temperatures, forming a gel-like mixture that is difcult to
recover. Excessive soap formation increases catalyst consumption,
reduce its effectiveness, and causes difculties in glycerol separation
and crude biodiesel purication (Atadashi et al., 2011). The recommended level of FFA in oil for homogeneous base-catalysed transesterication is reported to be less than 1% (Lam et al., 2010).
The homogeneous two-step acidbase catalysed transesterication reaction has been proposed as one of the biodiesel production
methods for oil with high FFA content. First, the high FFA oil is
subjected to acid esterication to remove the FFA from the oil. The
acid esterication is carried out at a temperature of 50 C for 1 h to

O
R

Methyl ester

CH3

H2 O

Water

FFA

OH

3HC

OH

Methanol

Fig. 13. Hydrolysis of methyl ester to form FFA (R = alkyl) (Van Gerpen and Knothe, 2005).

S.H. Shuit et al. / Biotechnology Advances 30 (2012) 13641380

1377

O
H2 C

R1

H 2C

OH

O
HC

O
R2

H2 O

HC

O
H2 C

O
R2

R1

OH

R3

Triglyceride

H 2C

Water

R3

Diglyceride

FFA

Fig. 14. Hydrolysis of triglyceride to form diglyceride and FFA (R = alkyl) (Atadashi et al., 2011).

convert FFA to esters using an acid catalyst (H2SO4, 1% w/w), thereby

reducing the FFA concentration to below 2%. The second step is alkaline based catalysed transesterication, which is carried out at 65 C
for 2 h using a NaOH catalyst. The reported FAME yield for this twostep transesterication is approximately 90%, which is much higher
than the one-step alkaline base transesterication (Berchmans and
Hirata, 2008). Although this process can achieve high FAME yield, it
involves a greater number of processing steps and reagents. After
acid esterication, the reaction mixture must be allowed to settle for
2 h; next, the methanolwater fraction in the top layer is removed before alkaline base transesterication. More NaOH is needed in twostep process because it not only serves as a catalyst but also neutralises
H2SO4 in the acid esterication. Although the heterogeneous acid catalyst and supercritical technology could also be used to produce biodiesel from oil with high FFA content, those processes are energy
intensive and would increase the operating cost of biodiesel production,
as mentioned in Section 7.1.
The catalytic membrane reactor, especially with a catalytically
active membrane, appears to be a suitable alternative to produce
biodiesel from oil with high FFA content because it can be easily
modied into an acidic membrane by introducing SO3H as a functional
acid group into the polymer matrix. Furthermore, the water or moisture
content found in oil sources can be separated by polymer membranes,
such as PVA and PAN, during the pervaporation process (Chapman et
al., 2008), thereby preventing the water from hydrolysing the produced
FAME back to FFA. Therefore, cheaper feedstocks such as non-edible
oils, waste cooking oils and even unrened crude oils with high FFA
content can be used in biodiesel production (Hasheminejad et al.,
2011).

7.5. Complying with international standards

Such impurities as glycerol, MG and DG are unfavourable for
engine performance and have negative effects on the environment.
High free glycerol content in biodiesel can cause gum formation
around injector tips and valve heads, causing problems in the fuel
system. In addition, the burning of glycerine produces the toxic compound acrolein (Hasheminejad et al., 2011). Therefore, the produced
biodiesel should be separated from these impurities. Of all the
reported biodiesel rening and purication methods, the membrane
separation technology has been proven to be a promising technology

that can produce and purify high-quality biodiesel that meets international standards (Sdrula, 2010).
Experimental results indicate that membrane separation technology
is able to reduce the free glycerol content in biodiesel to a level below
0.02 mass percent, which fulls the international standards (Gomes et
al., 2011; Saleh et al., 2010). In membrane separation, only 0.225% of
water by mass was added to FAME to improve separation efciency
(Saleh et al., 2010). Furthermore, compared to conventional biodiesel
purication methods, membrane separation can produce biodiesel
with higher purity and reduce the loss of ester during the rening
process. Moreover, the water content, density at 20 C and kinematic
viscosity of the biodiesel puried by the membrane technology were
also found to comply with the international standards (He et al., 2006).
8. Membrane life-time and fouling in biodiesel production
Because catalytically inert membranes come in contact with
strong acid or base catalysts during operation, it is vital to select a
membrane with high resistance to degradation and corrosion. Carbon
membranes are able to resist the harsh environment in the production
of biodiesel when H2SO4 or NaOH is used as a catalyst. It has been
reported that no tangible evidence of degradation of the carbon
membrane is observed, even after 10 months of operation and contact
with acid or base solution (Dub et al., 2007). The blended PSSA/PVA
(a kind of catalytically active membrane) showed a stable conversion
of 80% after 5 repeated transesterication runs with 8 h of reaction
time (Zhu et al., 2010). The polyethersulphone used in the biocatalytic
membrane microreactor also showed good stability with no decay of
its catalytic activity for at least 12 days of continuous operation
(Machsun et al., 2010).
Fouling is one of the major challenges in membrane processes.
Fouling of membranes is attributed to the accumulation and deposition
of solutes or particles in the feed onto the membrane surface and into
the membrane pores (Pagliero et al., 2007). In biodiesel production,
the agglomeration size of glycerol is inuenced by the alcohol concentration in the emulsion. Increased alcohol concentration favours the
formation of smaller glycerol agglomerates. Therefore, when a ceramic
membrane (catalytically inert membrane) is used to synthesise or
purify biodiesel, the presence of excess alcohol, soap and catalyst in
the reaction mixture favours membrane fouling and decreases the
permeate ux. This phenomenon probably occurs because the greater
amount of alcohol used in the reaction enables the glycerol and other

O
HO

O
(CH2)7CH

Oleic acid

CH(CH2)7CH3

KOH

K+ O

Potassium hydroxide

(CH2)7CH

CH(CH2)7CH

Soap

Fig. 15. Soap formation by using oleic acid as example (Atadashi et al., 2011).

H2O
Water

1378

S.H. Shuit et al. / Biotechnology Advances 30 (2012) 13641380

alcohol soluble substances to pass through the membrane pores easily,

causing pore plugging (Gomes et al., 2011). However, in the typical biodiesel production or rening process in which the alcohol concentration
in the reaction mixture is lower, the serious fouling of the membrane
that is caused by the accumulation of oil and other submicron particles
on the inverted membrane surface is not observed (Cheng et al., 2009;
Wang et al., 2009). Unfortunately, there has not been much study on
the fouling of catalytically active membranes, especially for the polymeric membranes used in biodiesel production. However, catalytically
active membranes show an entirely different behaviour from catalytically inert membranes: the accumulation of glycerol in the membrane
increases the catalytic activity of the membrane. The accumulation of
glycerol in the catalytically active membrane blocks the polymer OH
groups in the membrane, preventing interchain hydrogen bonding.
Eventually, the transport of oil or TG into the membrane is improved,
leading to a higher yield of biodiesel (Guerreiro et al., 2010). For
dense membranes that deal with an oily environment, an increase of
the membrane surface hydrophilicity prohibits the adsorption of oil
particles to the membrane surface and reduces membrane fouling
(Wang et al., 2006a). In addition, the fouling of the membrane can be
reduced by increasing the temperature, regardless of the type of
membrane used in biodiesel production. This phenomenon occurs
because of the reduction in solution viscosity that is observed when
the reaction temperature is increased (Pagliero et al., 2007).

9. Limitations in membrane technology for biodiesel production

Undeniably, membrane reactors can be considered to be an
emerging technology for biodiesel production. In order to successfully
develop and commercialise membrane reactors in the biodiesel
industry, knowledge is required in three major elds: catalysis,
membrane technology and reactor engineering. However, the desired
properties, such as the mechanical properties and surface morphology
of the membrane (especially in polymeric membranes), used in
biodiesel production have not been fully studied. Also, most of the
reported transesterication reactions via catalytic polymeric membranes have been performed in situations in which the membrane
was cut into small squares and loaded together with the reactants
(Guerreiro et al., 2010; Shi et al., 2010; Zhu et al., 2010). Therefore,
the ability of the synthesised polymeric membrane to separate glycerol from the product stream remains unstudied. Additionally, the engineering aspects of the membrane reactor have been minimally studied
because most publications have only offered proofs of concepts. Biodiesel production using membrane reactors is still running under
non-optimal conditions. Therefore, it is a challenge to choose the
best possible combination between catalyst and membrane. Optimisation studies and modelling will be needed to advance the membrane
reactor into commercial operation.
Although high biodiesel yield can be obtained via catalytically
inert membrane (mainly the micro-porous ceramic and carbon membranes), a water-washing step is still needed to purify the produced
biodiesel. The purication problem can be reduced by using catalytically active membranes (constructed from polymeric membranes)
in the ow conguration studied by Guerreiro et al., 2006, 2010;
Sarkar et al., 2010 and Zhu et al., 2010. However, the polymeric membranes face the problem of low mechanical strength. It has been
reported that the tested polymer membranes break before a high
conversion of biodiesel could be achieved (Guerreiro et al., 2006).
Therefore, more attention is needed to the selection of membranes
and operating conditions to avoid membrane failure.
The success of functionalised CNTs as a catalyst in biodiesel
production and their capability to enhance the mechanical strength
of membranes has made CNTs a suitable membrane ller. However,
no study had been conducted to investigate the potential of CNTMMM in biodiesel production. Furthermore, the compatibility of

functionalised CNTs with the membranes used in transesterication

reactions also remains uninvestigated.
10. Conclusion
The obstacles faced by conventional biodiesel production processes
have hindered the production of biodiesel in a way that is fully economically feasible and environmentally friendly. The emergence of
membrane technology offers a solution for future biodiesel production
that has the potential to be developed into large-scale commercial
processes. The main advantages of membrane technology concern the
production of biodiesel in a more environmentally friendly and costeffective manner. The two operation principles for membrane technology in biodiesel production are based on oil droplet size and the catalytic
membrane. The production of biodiesel via catalytically inert membrane
requires further purication because the permeate stream contains a
mixture of glycerol, methanol, catalysts and FAME. Therefore, the
catalytically active membrane seems to be a better option for biodiesel
production because less purication will be required. Published ndings
indicate that membrane reactors have the potential to be a breakthrough technology in the biodiesel production industry. However, the
application of membrane technology to the biodiesel industry can be
achieved only if the process parameters have been optimised. In
addition to the typical parameters, such as reaction temperature,
methanol-to-oil ratio and catalyst concentration, other process parameters, for example, the reactant ow rate, trans-membrane pressure,
membrane thickness and pore size, are reported to have great impacts
on the yield of biodiesel as well, and they cannot be ignored.
In addition, previous reports on the use of CNTs as a support for
catalysts in transesterication has prompted the idea of incorporating
the functionalised CNTs into the membrane, thereby fabricating a
CNT-MMM for biodiesel production. However, to the best of our
knowledge, there are no reports on the application of a CNT-MMM
to biodiesel production. Therefore, more experimental studies are
required to demonstrate the application of a CNT-MMM to the practical
biodiesel industry.
Acknowledgments
Shuit, S. H. and Ong, Y. T. acknowledge the MyPhD fellowship
support from the Ministry of Higher Education of Malaysia. This
research work is supported by USM Membrane Cluster Grant, the
Fundamental of Research Grant Scheme (FRGS), Universiti Sains
Malaysia Research University (RU) grant and the Postgraduate
Research Grant Scheme (PRGS).
References
Ahmad AL, Yasin NHM, Derek CJC, Lim JK. Microalgae as a sustainable energy source for
biodiesel production: a review. Renewable Sustainable Energy Rev 2011;15:58493.
Albuquerque MCG, Jimnez-Urbistondo I, Santamara-Gonzlez J, Mrida-Robles JM,
Moreno-Tost R, Rodrguez-Castelln E, et al. CaO supported on mesoporous silicas
as basic catalysts for transesterication reactions. Appl Catal A 2008;334:3543.
Andrade JE, Prez A, Sebastian PJ, Eapen D. A review of bio-diesel production processes.
Biomass Bioenergy 2011;35:100820.
Atadashi IM, Aroua MK, Aziz AA. High quality biodiesel and its diesel engine application: a review. Renewable Sustainable Energy Rev 2010;14:19992008.
Atadashi IM, Aroua MK, Aziz AA. Biodiesel separation and purication: a review. Renew
Energy 2011;36:43743.
Ataya F, Dub MA, Ternan M. Single-phase and two-phase base-catalyzed transesterication of canola oil to fatty acid methyl esters at ambient conditions. Ind Eng
Chem Res 2006;45:54117.
Bakhshi A, Mohammadi T, Nik OG, Aroujalian A. Effect of operating conditions on pervaporation of methanolwater mixtures: part 2. Membr Technol 2006;2006:7-11.
Barnard TM, Leadbeater NE, Boucher MB, Stencel LM, Wilhite BA. Continuous-ow
preparation of biodiesel using microwave heating. Energy Fuel 2007;21:177781.
Barnwal BK, Sharma MP. Prospects of biodiesel production from vegetable oils in India.
Renewable Sustainable Energy Rev 2005;9:36378.
Baroutian S, Aroua MK, Raman AAA, Sulaiman NMN. Methanol recovery during transesterication of palm oil in a TiO2/Al2O3 membrane reactor: experimental study
and neural network modeling. Sep Purif Technol 2010;76:5863.

S.H. Shuit et al. / Biotechnology Advances 30 (2012) 13641380

Baroutian S, Aroua MK, Raman AAA, Sulaiman NMN. A packed bed membrane reactor
for production of biodiesel using activated carbon supported catalyst. Bioresour
Technol 2011;102:1095102.
Bastiani R, Zonno IV, Santos IAV, Henriques CA, Monteiro JLF. Inuence of thermal
treatments on the basic and catalytic properties of Mg, Al-mixed oxides derived
from hydrotalcites. Braz J Chem Eng 2004;21:193202.
Berchmans HJ, Hirata S. Biodiesel production from crude Jatropha curcas L. seed oil with
a high content of free fatty acids. Bioresour Technol 2008;99:171621.
Bddeker KW. Liquid separations with membranes. Berlin: Springer-Verlag Berlin
Heidelberg; 2008. p. 34.
Boucher MB, Unker SA, Hawley KR, Wilhite BA, Stuart JD, Parnas RS. Variables affecting
homogeneous acid catalyst recoverability and reuse after esterication of concentrated omega-9 polyunsaturated fatty acids in vegetable oil triglycerides. Green
Chem 2008;10:13316.
BP. BP statistical review of world energy. Retrieved 10th February, 2011, from; 2009.
http://www.bp.com/liveassets/bp_internet/globalbp/globalbp_uk_english/reports_and_publications/statistical_energy_review_2008/STAGING/local_assets/
9_downloads/statistical_review_of_world_energy_full_report_9.pdf.
Buonomenna MG, Choi SH, Drioli E. Catalysis in polymeric membrane reactors: the
membrane role. Asia Pac J Chem Eng 2010;5:2634.
Cao P, Tremblay AY, Dub MA, Morse K. Effect of membrane pore size on the performance of a membrane reactor for biodiesel production. Ind Eng Chem Res
2006;46:528.
Cao P, Dub MA, Tremblay AY. High-purity fatty acid methyl ester production from canola, soybean, palm, and yellow grease lipids by means of a membrane reactor.
Biomass Bioenergy 2008a;32:102836.
Cao P, Dub MA, Tremblay AY. Methanol recycling in the production of biodiesel in a
membrane reactor. Fuel 2008b;87:82533.
Caro J. Catalysis in Micro-structured membrane reactors with nano-designed membranes. Chin J Catal 2008;29:116977.
Castanheiro JE, Ramos AM, Fonseca IM, Vital J. Esterication of acetic acid by isoamylic
alcohol over catalytic membranes of poly(vinyl alcohol) containing sulfonic acid
groups. Appl Catal A 2006;311:1723.
Chand P, Chintareddy VR, Verkade JG, Grewell D. Enhancing biodiesel production from
soybean oil using ultrasonics. Energy Fuel 2010;24:20105.
Chapman PD, Oliveira T, Livingston AG, Li K. Membranes for the dehydration of solvents by pervaporation. J Membr Sci 2008;318:5-37.
Chen H, Peng B, Wang D, Wang J. Biodiesel production by the transesterication of cottonseed oil by solid acid catalysts. Front Chem Eng Chin 2007;1:115.
Cheng L-H, Cheng Y-F, Yen S-Y, Chen J. Ultraltration of triglyceride from biodiesel
using the phase diagram of oilFAMEMeOH. J Membr Sci 2009;330:15665.
Cheng L-H, Yen S-Y, Su L-S, Chen J. Study on membrane reactors for biodiesel production by phase behaviors of canola oil methanolysis in batch reactors. Bioresour
Technol 2010;101:66638.
Colucci J, Borrero E, Alape F. Biodiesel from an alkaline transesterication reaction of
soybean oil using ultrasonic mixing. J Am Oil Chem Soc 2005;82:52530.
Conservation ADoE. How can I dispose of used cooking oils and fats? . Retrieved 13th
March, 2011, from. http://www.dec.state.ak.us/eh/docs/fss/cooking.pdf.
De Boni LAB, Lima da Silva IN. Monitoring the transesterication reaction with laser
spectroscopy. Fuel Process. Technol. 2011;92:10016.
Demirbas A. Progress and recent trends in biofuels. Prog Energy Combust Sci 2007;33:1-18.
Demirbas A. New liquid biofuels from vegetable oils via catalytic pyrolysis. Energy Educ
Sci Technol 2008;21:1-59.
DeRoussel P, Khakhar DV, Ottino JM. Mixing of viscous immiscible liquids. Part 2:
overemulsication-interpretation and use. Chem Eng Sci 2001;56:55317.
Deshmane VG, Gogate PR, Pandit AB. Ultrasound-assisted synthesis of biodiesel from
palm fatty acid distillate. Ind Eng Chem Res 2008;48:79237.
Di Serio M, Ledda M, Cozzolino M, Minutillo G, Tesser R, Santacesaria E. Transesterication of soybean oil to biodiesel by using heterogeneous basic catalysts. Ind Eng
Chem Res 2006;45:300914.
Dittmeyer R, Svajda K, Reif M. A review of catalytic membrane layers for gas/liquid reactions. Top Catal 2004;29:3-27.
Dizge N, Aydiner C, Imer DY, Bayramoglu M, Tanriseven A, Keskinler B. Biodiesel production from sunower, soybean, and waste cooking oils by transesterication
using lipase immobilized onto a novel microporous polymer. Bioresour Technol
2009;100:198391.
Drioli E, Fontananova E, Bonchio M, Carraro M, Gardan M, Scorrano G. Catalytic membranes and membrane reactors: an integrated approach to catalytic process with a
high efciency and a low environmental impact. Chin J Catal 2008;29:11528.
Dub MA, Tremblay AY, Liu J. Biodiesel production using a membrane reactor. Bioresour Technol 2007;98:63947.
EIA. International Energy Outlook. Chapter 1: world energy demand and economic outlook. Retrieved 25th January, 2011, from. http://www.eia.gov/oiaf/ieo/pdf/world.
pdf2010.
Ertl H, Knozinger H, Schuth F, Weitkamp J. Catalytic membrane reactors. Handbook of
heterogeneous catalysis. Weinheim: Wiley-VCH; 2008. p. 2198241.
Falahati H, Tremblay AY. The effect of ux and residence time in the production of biodiesel from various feedstocks using a membrane reactor. Fuel 2012;91:12633.
Fazal MA, Haseeb ASMA, Masjuki HH. Biodiesel feasibility study: an evaluation of material compatibility; performance; emission and engine durability. Renewable Sustainable Energy Rev 2011;15:131424.
Figueiredo KCdS, Salim VMM, Borges CP. Synthesis and characterization of a catalytic
membrane for pervaporation-assisted esterication reactors. Catal Today 2008:
1335. :80914.
Furuta S, Matsuhashi H, Arata K. Biodiesel fuel production with solid superacid catalysis in xed bed reactor under atmospheric pressure. Catal Commun 2004;5:7213.

1379

Gogate PR, Kabadi AM. A review of applications of cavitation in biochemical engineering/biotechnology. Biochem Eng J 2009;44:6072.
Gomes MCS, Pereira NC, Barros STDd. Separation of biodiesel and glycerol using ceramic membranes. J Membr Sci 2010;352:2716.
Gomes MCS, Arroyo PA, Pereira NC. Biodiesel production from degummed soybean oil
and glycerol removal using ceramic membrane. J Membr Sci 2011;378:45361.
Guan H-M, Chung T-S, Huang Z, Chng ML, Kulprathipanja S. Poly(vinyl alcohol) multilayer mixed matrix membranes for the dehydration of ethanol-water mixture. J
Membr Sci 2006;268:11322.
Guerreiro L, Castanheiro JE, Fonseca IM, Martin-Aranda RM, Ramos AM, Vital J. Transesterication of soybean oil over sulfonic acid functionalised polymeric membranes. Catal Today 2006;118:16671.
Guerreiro L, Pereira PM, Fonseca IM, Martin-Aranda RM, Ramos AM, Dias JML, et al. PVA
embedded hydrotalcite membranes as basic catalysts for biodiesel synthesis by
soybean oil methanolysis. Catal Today 2010;156:1917.
Hammond GP, Kallu S, McManus MC. Development of biofuels for the UK automotive
market. Appl Energy 2008;85:50615.
Hasheminejad M, Tabatabaei M, Mansourpanah Y, far MK, Javani A. Upstream and
downstream strategies to economize biodiesel production. Bioresour Technol
2011;102:4618.
Hawash S, Kamal N, Zaher F, Kenawi O, Diwani GE. Biodiesel fuel from Jatropha oil via
non-catalytic supercritical methanol transesterication. Fuel 2009;88:57982.
He H, Guo X, Zhu S. Comparison of membrane extraction with traditional extraction
methods for biodiesel production. J Am Oil Chem Soc 2006;83:45760.
Hoekman SK. Biofuels in the U.S. Challenges and opportunities. Renew Energy
2008;34:1422.
Ismail AF, Goh PS, Sanip SM, Aziz M. Transport and separation properties of carbon
nanotube-mixed matrix membrane. Sep Purif Technol 2009;70:1226.
Janaun J, Ellis N. Perspectives on biodiesel as a sustainable fuel. Renewable Sustainable
Energy Rev 2010;14:131220.
Jegannathan KR, Abang S, Poncelet D, Chan ES, Ravindra P. Production of biodiesel
using immobilized lipase a critical review. Crit Rev Biotechnol 2008;28:25364.
Jitputti J, Kitiyanan B, Rangsunvigit P, Bunyakiat K, Attanatho L, Jenvanitpanjakul P.
Transesterication of crude palm kernel oil and crude coconut oil by different
solid catalysts. Chem Eng J 2006;116:616.
Kalu EE, Chen KS, Gedris T. Continuous-ow biodiesel production using slit-channel reactors. Bioresour Technol 2011;102:445661.
Kalva A, Sivasankar T, Moholkar VS. Physical mechanism of ultrasound-assisted synthesis of biodiesel. Ind Eng Chem Res 2008;48:53444.
Karaosmano lu F, C zo lu KB, Tter M, Ertekin S. Investigation of the rening step of
biodiesel production. Energy Fuel 1996;10:8905.
Kelkar MA, Gogate PR, Pandit AB. Intensication of esterication of acids for synthesis
of biodiesel using acoustic and hydrodynamic cavitation. Ultrason Sonochem
2008;15:18894.
Khayet M, Villaluenga JPG, Godino MP, Mengual JI, Seoane B, Khulbe KC, et al. Preparation and application of dense poly(phenylene oxide) membranes in pervaporation.
J Colloid Interface Sci 2004;278:41022.
Kim K-J, Lee S-B, Han N-W. Kinetics of crosslinking reaction of PVA membrane with
glutaraldehyde. Korean J Chem Eng 1994;11:417.
Knothe G, Gerpen JV, Krahl J. The Biodiesel Handbook. Champaign: AOCS; 2005. p. 2703.
Kumar G, Kumar D, Singh S, Kothari S, Bhatt S, Singh C. Continuous low cost transesterication process for the production of coconut biodiesel. Energies 2010;3:4356.
Lam MK, Lee KT, Mohamed AR. Homogeneous, heterogeneous and enzymatic catalysis
for transesterication of high free fatty acid oil (waste cooking oil) to biodiesel: a
review. Biotechnol Adv 2010;28:50018.
Leung DYC, Wu X, Leung MKH. A review on biodiesel production using catalyzed transesterication. Appl Energy 2010;87:108395.
Licht FO. World Biodiesel Market: The Outlook to 2010. Tunbridge Wells: Agra
Informa; 2007.
Lim S, Teong LK. Recent trends, opportunities and challenges of biodiesel in Malaysia:
an overview. Renewable Sustainable Energy Rev 2010;14:93854.
Lin L, Cunshan Z, Vittayapadung S, Xiangqian S, Mingdong D. Opportunities and challenges for biodiesel fuel. Appl Energy 2011;88:102031.
Lipnizki F, Field RW, Ten P-K. Pervaporation-based hybrid process: a review of process
design, applications and economics. J Membr Sci 1999a;153:183210.
Lipnizki F, Hausmanns S, Ten P-K, Field RW, Laufenberg G. Organophilic pervaporation:
prospects and performance. Chem Eng J 1999b;73:11329.
Liu X, He H, Wang Y, Zhu S, Piao X. Transesterication of soybean oil to biodiesel using
CaO as a solid base catalyst. Fuel 2008;87:21621.
Lu H, Liu Y, Zhou H, Yang Y, Chen M, Liang B. Production of biodiesel from Jatropha curcas
L. oil. Comput Chem Eng 2009;33:10916.
Mabee WE. Policy options to support biofuel production. Adv Biochem Eng Biotechnol
2007;108:32957.
Machsun A, Gozan M, Nasikin M, Setyahadi S, Yoo Y. Membrane microreactor in biocatalytic transesterication of triolein for biodiesel production. Biotechnol Bioprocess
Eng 2010;15:9116.
Mahajan R, Koros WJ, Thundyil M. Mixed matrix membranes: important and challenging! Membr Technol 1999;1999:68.
Marchetti JM, Miguel VU, Errazu AF. Possible methods for biodiesel production. Renewable Sustainable Energy Rev 2007;11:130011.
Mbaraka IK, Shanks BH. Conversion of oils and fats using advanced mesoporous heterogeneous catalysts. J Am Oil Chem Soc 2006;83:7991.
Meher LC, Vidya Sagar D, Naik SN. Technical aspects of biodiesel production by transesterication a review. Renewable Sustainable Energy Rev 2006;10:24868.
Monni S, Raes F. Multilevel climate policy: the case of the European Union, Finland and
Helsinki. Environ Sci Policy 2008;11:74355.

1380

S.H. Shuit et al. / Biotechnology Advances 30 (2012) 13641380

Narasimharao K, Lee A, Wilson K. Catalysts in production of biodiesel: a review. J Biobased Mater Bioenergy 2007;1:1930.
Nishimoto A, Yoshikawa S, Ookawara S. A model for transport phenomena in a
cross-ow ultraltration module with microchannels. Membranes 2010;1:1324.
Oda M, Kaieda M, Hama S, Yamaji H, Kondo A, Izumoto E, et al. Facilitatory effect of
immobilized lipase-producing Rhizopus oryzae cells on acyl migration in
biodiesel-fuel production. Biochem Eng J 2005;23:4551.
Ong YT, Ahmad AL, Zein SHS, Sudesh K, Tan SH. Poly(3-hydroxybutyrate)-functionalised
multi-walled carbon nanotubes/chitosan green nanocomposite membranes and
their application in pervaporation. Sep Purif Technol 2011;76:41927.
Othman R, Mohammad AW, Ismail M, Salimon J. Application of polymeric solvent resistant nanoltration membranes for biodiesel production. J Membr Sci 2010;348:
28797.
Pagliero C, Mattea M, Ochoa N, Marchese J. Fouling of polymeric membranes during
degumming of crude sunower and soybean oil. J Food Eng 2007;78:1947.
Pal A, Verma A, Kachhwaha SS, Maji S. Biodiesel production through hydrodynamic
cavitation and performance testing. Renew Energy 2010;35:61924.
Pangarkar V, Pal S. Pervaporation. Handbook of Membrane Separations: Chemical,
Pharmaceutical, and Biotechnological Applications. Boca Raton: CRC Press; 2008.
p. 10738.
Peng F, Jiang Z, Hu C, Wang Y, Xu H, Liu J. Removing benzene from aqueous solution
using CMS-lled PDMS pervaporation membranes. Sep Purif Technol 2006;48:
22934.
Peng F, Hu C, Jiang Z. Novel ploy(vinyl alcohol)/carbon nanotube hybrid membranes
for pervaporation separation of benzene/cyclohexane mixtures. J Membr Sci
2007;297:23642.
Phan AN, Harvey AP, Rawcliffe M. Continuous screening of base-catalysed biodiesel
production using new designs of mesoscale oscillatory bafed reactors. Fuel Process Technol 2011;92:15607.
Pinnarat T, Savage PE. Assessment of noncatalytic biodiesel synthesis using supercritical reaction conditions. Ind Eng Chem Res 2008;47:68018.
Porter MC. Concentration polarization with membrane ultraltration. Prod Res Dev
1972;11:23448.
Qiu Z, Zhao L, Weatherley L. Process intensication technologies in continuous biodiesel production. Chem Eng Process 2010;49:32330.
Rashid U, Anwar F. Production of biodiesel through optimized alkaline-catalyzed transesterication of rapeseed oil. Fuel 2008;87:26573.
Rashid U, Anwar F, Moser BR, Ashraf S. Production of sunower oil methyl esters by optimized alkali-catalyzed methanolysis. Biomass Bioenergy 2008;32:12025.
Roelofs JCAA, van Dillen AJ, de Jong KP. Base-catalyzed condensation of citral and acetone at low temperature using modied hydrotalcite catalysts. Catal Today
2000;60:297303.
Rokhina EV, Lens P, Virkutyte J. Low-frequency ultrasound in biotechnology: state of
the art. Trends Biotechnol 2009;27:298306.
Sae-Khow O, Mitra S. Pervaporation in chemical analysis. J Chromatogr A 2010;1217:
273646.
Sahoo PK, Das LM, Babu MKG, Naik SN. Biodiesel development from high acid value
polanga seed oil and performance evaluation in a CI engine. Fuel 2007;86:44854.
Saleh J, Tremblay AY, Dub MA. Glycerol removal from biodiesel using membrane separation technology. Fuel 2010;89:22606.
Samart C, Sreetongkittikul P, Sookman C. Heterogeneous catalysis of transesterication
of soybean oil using KI/mesoporous silica. Fuel Process Technol 2009;90:9225.
Sanchez Marcano JG, Tsotsis TT. The coupling of the membrane separation process with
a catalytic reaction. Catalytic Membranes and Membrane Reactor. Weinheim:
Wiley-VCH; 2002. p. 5-14.
Sarkar B, Sridhar S, Saravanan K, Kale V. Preparation of fatty acid methyl ester through
temperature gradient driven pervaporation process. Chem Eng J 2010;162:60915.
Sawyer RF. Science based policy for addressing energy and environmental problems.
Proc Combust Inst 2009:4556.
Sdrula N. A study using classical or membrane separation in the biodiesel process. Desalination 2010;250:10702.
Shao P, Huang RYM. Polymeric membrane pervaporation. J Membr Sci 2007;287:
16279.
Sharma YC, Singh B. Development of biodiesel from karanja, a tree found in rural India.
Fuel 2008;67:17402.
Sharma YC, Singh B. Development of biodiesel: current scenario. Renewable Sustainable Energy Rev 2009;13:164651.

Sharma A, Thampi SP, Suggala SV, Bhattacharya PK. Pervaporation from a dense membrane: roles of permeantmembrane interactions, Kelvin effect, and membrane
swelling. Langmuir 2004;20:470814.
Shi W, He B, Ding J, Li J, Yan F, Liang X. Preparation and characterization of the organic
inorganic hybrid membrane for biodiesel production. Bioresour Technol 2010;101:
15015.
Shuit SH, Lee KT, Kamaruddin AH, Yusup S. Reactive extraction of Jatropha curcas L.
seed for production of biodiesel: process optimization study. Environ Sci Technol
2010a;44:43617.
Shuit SH, Lee KT, Kamaruddin AH, Yusup S. Reactive extraction and in situ esterication
of Jatropha curcas L. seeds for the production of biodiesel. Fuel 2010b;89:52730.
Sirkar KK, Ho WSW. Membrane Handbook. New York: Van Nostrand Reinhold; 1992.
p. 1-20.
Stankiewicz A. Reactive separations for process intensication: an industrial perspective. Chem Eng Process 2003;42:13744.
Su E-Z, Xu W-Q, Gao K-L, Zheng Y, Wei D-Z. Lipase-catalyzed in situ reactive extraction
of oilseeds with short-chained alkyl acetates for fatty acid esters production. J Mol
Catal B: Enzym 2007;48:2832.
Tai-Shung NC. Development and purication of biodiesel. Sep Purif Technol 2007;20:
37781.
TakhtRavanchi M, Kaghazchi T, Kargari A. Immobilized liquid membrane for
propylene-propane separation. World Acad Sci Eng Technol 2008:7981.
Tremblay AY, Cao P, Dub MA. Biodiesel production using ultralow catalyst concentrations. Energy Fuel 2008;22:274855.
Van Gerpen J, Knothe G. Biodiesel Production. In: Knothe G, Van Gerpen J, Krahl J, editors. The biodiesel handbook. Champaign: AOCS; 2005. p. 2656.
Vankelecom IFJ. Polymeric membranes in catalytic reactors. Chem Rev 2002;102:
3779810.
Vicente G, Martnez M, Aracil J. Integrated biodiesel production: a comparison of different homogeneous catalysts systems. Bioresour Technol 2004;92:297305.
Villa A, Tessonnier J-P, Majoulet O, Su DS, SchlAgl R. Amino-functionalized carbon
nanotubes as solid basic catalysts for the transesterication of triglycerides.
Chem Commun (Camb) 2009:44057.
Villa A, Tessonnier J-P, Majoulet O, Su DS, Schlgl R. Transesterication of triglycerides
using nitrogen-functionalized carbon nanotubes. ChemSusChem 2010;3:2415.
Vospernik M, Pintar A, Bercic G, Batista J, Levec J. Potentials of ceramic membranes as
catalytic three-phase reactors. Chem Eng Res Des 2004;82:65966.
Wang Y, Hsieh Y-L. Crosslinking of polyvinyl alcohol (PVA) brous membranes with
glutaraldehyde and PEG diacylchloride. J Appl Polym Sci 2010;116:324955.
Wang S, Chu L, Chen W. Fouling-resistant composite membranes for separation of
oil-in-water microemulsions. Chin J Chem Eng 2006a;14:3745.
Wang Y, Ou S, Liu P, Xue F, Tang S. Comparison of two different processes to synthesize
biodiesel by waste cooking oil. J Mol Catal A: Chem 2006b;252:10712.
Wang Y, Wang X, Liu Y, Ou S, Tan Y, Tang S. Rening of biodiesel by ceramic membrane
separation. Fuel Process Technol 2009;90:4227.
Wardle DA. Global sale of green air travel supported using biodiesel. Renewable Sustainable Energy Rev 2003;7:1-64.
Wen Z, Yu X, Tu S-T, Yan J, Dahlquist E. Intensication of biodiesel synthesis using zigzag micro-channel reactors. Bioresour Technol 2009;100:305460.
Wu Z, Ondruschka B, Zhang Y, Bremner DH, Shen H, Franke M. Chemistry driven by
suction. Green Chem 2009;11:102630.
Xie W, Li H. Alumina-supported potassium iodide as a heterogeneous catalyst for biodiesel production from soybean oil. J Mol Catal A: Chem 2006;255:19.
Yee KF, Tan KT, Abdullah AZ, Lee KT. Life cycle assessment of palm biodiesel: revealing
facts and benets for sustainability. Appl Energy 2009;86:S18996.
Yin JZ, Xiao M, Song JB. Biodiesel from soybean oil in supercritical methanol with
co-solvent. Energy Convers Manage 2008;49:90812.
Zhang Y, Dub MA, McLean DD, Kates M. Biodiesel production from waste cooking oil:
1. Process design and technological assessment. Bioresour Technol 2003;89:1-16.
Zheng M, Skelton RL, Mackley MR. Biodiesel reaction screening using oscillatory ow
meso reactors. Process Saf Environ Prot 2007;85:36571.
Zhou H, Lu H, Liang B. Solubility of multicomponent systems in the biodiesel production by transesterication of Jatropha curcas l. oil with methanol. J Chem Eng
Data 2006;51:11305.
Zhu M, He B, Shi W, Feng Y, Ding J, Li J, et al. Preparation and characterization of
PSSA/PVA catalytic membrane for biodiesel production. Fuel 2010;89:2299304.