Fuel 235 (2019) 886–907

Contents lists available at ScienceDirect

Fuel
journal homepage: www.elsevier.com/locate/fuel

Review article

State of the art review on development of ultrasound-assisted catalytic T

transesterification process for biodiesel production
⁎ ⁎
Shiou Xuan Tana, Steven Limb, , Hwai Chyuan Onga, , Yean Ling Pangb
a
Department of Mechanical Engineering, Faculty of Engineering, University of Malaya, 50603 Kuala Lumpur, Malaysia
b
Department of Chemical Engineering, Lee Kong Chian Faculty of Engineering and Science, Universiti Tunku Abdul Rahman, 43000 Selangor, Malaysia

A R T I C LE I N FO A B S T R A C T

Keywords: Excessive utilization of petroleum diesel has led to severe environmental pollution. Biodiesel, which is greener
Ultrasonic and renewable, can be a potential alternative fuel. Biodiesel is produced through transesterification reaction
Reactive extraction between vegetable oil, animal fat or even waste cooking oil (WCO) and alcohol in the presence of catalyst. Under
Transesterification process intensification, ultrasonic irradiation is employed in the transesterification reaction to enhance the
Non-edible
agitation between immiscible reactants. Besides providing intensive mixing, it also offers uniform heating due to
Mechanism
the localized temperature increase and formation of micro jets from the transient collapse of cavitation bubbles,
thus reducing the energy consumption. The focus of this paper is to review the recent research progress on the
ultrasound-assisted catalytic transesterification of non-edible vegetable oils using homogeneous and hetero-
geneous catalysts. The primary factors that affect the operation and efficiency of ultrasound-assisted transes-
terification such as alcohol to oil molar ratio, catalyst loading, reaction time, reaction temperature, energy
consumption, phase separation time, ultrasonic pulse mode and biodiesel conversion or yield have been re-
viewed. The highlights of this review paper are the provisions on the mechanism of ultrasonic reactive extraction
(RE) in the biodiesel production, kinetic study and the existing pilot reactors on the ultrasound-assisted trans-
esterification which are still rarely reviewed in the current literature. Lastly, the challenges and feasibility for
future development in the process intensification of biodiesel production are also addressed.

1. Introduction
Biodiesel is a liquid fuel which comprises of mono alkyl ester of
long-chain fatty acid. In principle, it is a sustainable source of liquid
transportation fuels as it is synthesized from various types of renewable
lipids such as virgin vegetable oils, non-edible vegetable oils, waste
vegetable oils (WVOs) and animal fats. Biodiesel can be produced from
a variety of sources which include edible oils (e.g. canola, soybean,
sunflower and palm oils), non-edible oils (e.g. Jatropha curcas,
Daturametel Linn, Hevea brasiliensis and Calophyllum inophyllum oils),
waste oils as well as animal fats (e.g. chicken fat, beef tallow and
poultry fat) [1]. The increasing price volatility of fossil fuel has made
biodiesel appears to have significant economic potential as a renewable
fuel by lowering the dependency on crude oil foreign imports. Cheaper
and non-edible feedstocks have been employed to produce biodiesel in
order to alleviate the food crisis arising from the utilization of edible
feedstocks [2,3].
Biodiesel has negligible sulphur and aromatics content with about
11% built-in oxygen content, which helps it to undergo complete
combustion [4]. Therefore, it tends to produce less smoke and particles
during combustion, resulting in lower emissions of harmful pollutants
such as particulates, carbon monoxide (CO), unburned hydrocarbons
(HCs) and sulphur oxides (SOx) [5,6]. By employing biodiesel compared
to diesel, the release of particulates, CO, unburned HCs and SOx would
be lowered by 40, 44, 68 and 100%, respectively [7]. In contrast, diesel
fuel does not contain any oxygen element [8]. The sulphur content in
the diesel fuel contributes to the formation of SOx and sulphuric acid,
which lead to acid rain. The aromatic compounds in the diesel fuel also
increase particulate emissions and are considered carcinogens. Ben-
zene, toluene, ethylbenzene, and o-, m-, p-xylene, popularly known as
BTEX compounds, are well documented carcinogenic compounds
emitted from the exhaust of a diesel-powered compression ignition (CI)
engine [9]. Furthermore, biodiesel does not require any modifications
when it is utilized in diesel engine directly due to its similar properties
to diesel fuel, thus render it a suitable alternative to diesel fuel [10].
Transesterification is the most common process to synthesize bio-
diesel in the presence of alkali, acid or enzyme as catalyst. However,
there are two major challenges encountered in the transesterification

⁎
Corresponding authors.
E-mail addresses: stevenlim@utar.edu.my, stevenlim_my2000@hotmail.com (S. Lim), onghc@um.edu.my (H.C. Ong).

https://doi.org/10.1016/j.fuel.2018.08.021
Received 28 March 2018; Received in revised form 12 July 2018; Accepted 4 August 2018
Available online 28 August 2018
0016-2361/ © 2018 Elsevier Ltd. All rights reserved.
S.X. Tan et al.
Nomenclature
R1–COOH free fatty acid
NaOH sodium hydroxide
R1COONa soap
H2O water
triglyceride
reaction. Firstly, heterogeneous nature of the reaction system poses
challenges in mass transfer limitations. Secondly, higher alcohol to oil
molar ratio is required because of the reversible nature of transester-
ification reaction. These result in higher operating cost and energy
consumption which consequently lead to lower biodiesel production
efficiency. There are several other transesterification techniques to
produce biodiesel such as non-catalytic supercritical method, biox
process, microwave, membrane technology, reactive distillation and
ultrasonication. Non-catalytic supercritical method does not require
catalyst, thus it does not face the disadvantages encountered in the
catalytic routes such as the necessity of treating the free fatty acids
(FFAs) and triglycerides (TGs) in different reaction stages, the in-
hibitory effects of water molecules present in the mixture, catalyst de-
activation, separation of the catalyst from the product mixture, low
glycerol purity, and generation of waste water. However, it is an energy
intensive and economic infeasible process because it requires high op-
erating temperature (553–673 K) and pressure (10–30 MPa) as well as
high methanol-to-triglycerides molar ratio up to 42:1 [11]. For biox
process, it employs co-solvents such as tetrahydrofuran (THF) to solu-
bilize methanol for faster reaction. THF is commonly used due to its
boiling point (339 K) is similar to methanol (337.7 K) which allows the
removal of excess solvents in a single step once the reaction has com-
pleted [12,13]. This process is capable to convert oil feedstock con-
taining high FFA content (more than 10%) into biodiesel and can be
conducted under ambient temperature and pressure. However, eco-
nomical barrier arises due to their close proximity of boiling point be-
tween methanol and co-solvent which makes the product separation to
be very challenging [14]. In addition, complete removal of co-solvent
from both the glycerol and biodiesel is compulsory due to its hazardous
and toxicity natures [15,16]. Comparing to conventional transester-
ification, microwave method requires about 23 times lower energy
consumption. This is because microwave energy is transmitted directly
to the reactant, thus eliminating the preheating step. Furthermore,
microwave process has more effective heat transfer system than con-
ventional method. Conventional method transfers heat to the reaction
via thermal heat reflux whereas microwave transfers energy in the form
of electromagnetic wave [7]. Nonetheless, low penetration depth of
microwave radiation into the absorbing materials limits the scaled-up
to industrial scale [17]. On the other hand, membrane technology in-
tegrates reaction and separation into single step, thus minimizing se-
paration costs and recycling requirements. However, membrane tech-
nology still requires further downstream purification since the fatty
acid methyl ester (FAME)-rich phase will still contain methanol, gly-
cerol and water [11,18]. For reactive distillation, it is advantageous for
esterification process especially when dealing with high FFA feedstocks.
Compared to the conventional method, it does not require excess al-
cohol due to the continuous removal of by-product (water) and is able
to reduce the alcohol usage by 66% [11,19]. Even though reactive
distillation process has less number of connections between instruments
(due to smaller amount of equipment) which reduces the safety issues
[20], the requirement for reboiler and condenser will increase the
887
Fuel 235 (2019) 886–907
diglyceride
CH3OH methanol
R1COOCH3 fatty acid methyl ester
k reaction rate constant
X conversion of triglyceride at time,t
t time
CTG0 molar concentration of triglyceride at time zero in mol/L
capital investment and operating costs tremendously [21]. Among the
emerging transesterification techniques, ultrasonication appears to be a
promising approach as it can enhance mixing, heat and mass transfer
between different phases in alcoholysis process [22]. Consequently, it
can reduce the operating cost and energy consumption as it has lower
requirement on alcohol to oil molar ratio, catalyst amount and reaction
time as well as no external heating requirement. In addition, phase
separation of biodiesel from glycerol is simpler and shorter phase se-
paration time is needed. Its superiority in transesterification reaction
possesses a huge possibility to be scaled up from laboratory to industrial
scale. However, it is noteworthy to consider also the problems asso-
ciated with the utilization of biodiesel in engine. Biodiesel generally has
14–15% higher fuel consumption than diesel due to its lower calorific
value (37.9 MJ/kg) than diesel (42.7 MJ/kg) while NO emissions are
always 15–20% higher than diesel [23]. An overview of the advantages,
disadvantages and recommendations of various emerging biodiesel
production approaches is tabulated in Table 1.
This paper reviews in depth the ultrasound-assisted catalytic
transesterification for biodiesel synthesis, effect of different parameters
towards biodiesel conversion or yield in the ultrasound-assisted
homogeneous and heterogeneous base-catalyzed transesterification as
well as comparison between conventional stirring and ultrasonic cavi-
tation in biodiesel production. In addition, this paper also provides
fundamental understanding of ultrasonic mechanism and their effects
on transesterification reaction. Most importantly, kinetic study on the
ultrasound-assisted transesterification, latest pilot reactors for ultra-
sound-assisted biodiesel production and mechanism of ultrasonic re-
active extraction on solid seed for biodiesel production are also re-
viewed in details.
2. Introduction of ultrasound
Ultrasound refers to sound wave above human audibility limit
which is usually above 20 kHz [29]. It is categorized based on its fre-
quency into high-frequency (2–10 MHz) and low-frequency
(20–100 kHz) ultrasound [30]. Application of high-frequency operation
is more effective for chemical synthesis and wastewater treatment as
chemical effects induced are more intensive. Low-frequency operation
is useful for enhancing the mass transfer across the immiscible reactants
by increasing the interfacial surface area between them. Lower fre-
quency is suggested to be employed for biodiesel synthesis since
dominant physical effects are required for intensification of transes-
terification reaction [22]. High frequency will not benefit biodiesel
synthesis because of the cavitation bubble collapses are weaker than
impingement of the reactants [14]. Diverse forms of energy can be
generated through ultrasonic applicator, which is shown in Fig. 1. An
ultrasonic probe transforms electrical energy input into heat energy and
vibrational energy. The vibrational energy is then transformed into
cavitation energy while some of them is lost through sound reflection
[13]. The cavitation energy is further converted into chemical, physical,
and biological effects depending on the application and reaction
 Table 1
Comparison between different biodiesel production approaches.
Methods Description
Non-catalytic Biodiesel production in which transesterification
supercritical occurs by an alcohol at supercritical conditions
method (high temperature and pressure) without catalyst.
Biox process Non-catalytic biodiesel production with the use of
co-solvent (usually THF) that generates one phase
oil-rich system.
Microwave Continuous oscillation of polar ends of molecules
or ions by microwave irradiation causes the
collision and friction between the moving
molecules and generates localized superheating.
888
Reactive distillation Simultaneous chemical reaction and distillation
separation in a single unit.
Membrane Combination of reaction and membrane-based
technology separation by taking advantage of the
immiscibility of methanol and oil to create a better
purification process.
Ultrasonication Cavitation bubbles are produced by sound waves
in a liquid due to pressure variations.
Advantages
• Requires no catalyst. • Requires high operating temperature
• Achieves near complete conversion in a
relatively short time (2–4 min).
• Short reaction time.
• Able to handle feeds with high FFA.
• Recovery of both the excess alcohol and
the THF co-solvent in a single step.
• Absence of catalyst residues in either the
biodiesel phase or the glycerol phase.
• Short reaction time (10 times shorter
than conventional methods).
• Low oil to methanol ratio.
• Ease of operation.
• Higher quality and yield product.
• Drastic reduction in quantity of by-
products.
• Simultaneous separation of reactants and • Thermal degradation of the products
products boosts the conversion and
improves the selectivity by breaking the
reaction equilibrium restrictions.
• Eliminates the need for reheating
because the heat of vaporization
provides the heat of reaction in
exothermic reactions.
• Controlled contact of incompatible
reactants.
• Produces FAME-rich products.
• Eliminates undesired side reactions.
• Simple reaction setup.
• Shorter reaction time.
• Less energy consumption (only one-third
to half of the energy that is consumed by
mechanical agitation).
• Enhanced conversion.
• Higher oil extraction yield (30–40%
more when compared to the
conventional stirring reactor system).
• Higher quality of glycerol production.
Disadvantages
(553–673 K) and pressure (10–30 MPa)
resulting in much higher heating and
cooling costs.
• Requires high methanol-to-triglycerides
molar ratio up to 42:1.
• Completely removal of co-solvent is
compulsory due to its hazardous and
toxicity natures.
• Residual solvent in the biofuel product
can affect the compliance with the
international standards.
• Not suitable to be used in industrial
scale as high microwave output (power)
may cause damage to organic molecules
(TGs) and due to safety aspects.
occurs due to a higher temperature
profile in the column.
• High fabrication cost.
• Fouling can occur which reduces the
effectiveness over time.
• Necessity of downstream processing
which increases the purification cost.
• Reaction temperature is slightly higher
for long reactions.
• Ultrasonic power must be under control
due to the soap formation in fast
reaction.
• FAME yield can be reduced by higher
frequencies (40 kHz).
S.X. Tan et al.
Recommendations Ref.
Addition of co-solvents such as carbon dioxide [2,20,24,25]
(CO2), hexane, propane, calcium oxide (CaO) and
subcritical alcohol to reduce the reaction
temperature, pressure and the amount of
alcohol.*The supercritical condition of CO2 is
304 K and 7.38 MPa which it is lower than the
supercritical methanol conditions (512 K and
8.09 MPa).
− [14]
Combination effect of microwave (removal of heat [7,17,26]
transfer barrier and ultrasound (removal of mass
transfer barrier) can lead to synergistic effects
significantly which overcomes the limitation of
low penetration depth. However, this technique
leads to high energy consumption for maintaining
higher temperature condition which is not
applicable for industrial scale.
Reactive absorption could replace reactive [19,21,27]
distillation as it reduces capital investment and
operating costs due to the absence of the reboiler
and condenser, higher conversion and selectivity
as no products are recycled in the form of reflux or
boil-up vapours, as well as no occurrence of
thermal degradation of the products due to a
lower temperature profile in the column.
− [6,7,11,19,28]
− [7,17,22,26]
Fuel 235 (2019) 886–907
S.X. Tan et al.
Electrical
Heat Vibrational
Physical Chemical
Fig. 1. Energy forms, losses and effects in ultrasonic application.
environment [31]. Generally, the beneficial effects of ultrasound on
chemical reactions come from the intense energy produced from the
ultrasonic cavitation. Ultrasonic cavitation is the formation, growth and
implosive collapse of bubbles (cavities) in a liquid irradiated with ul-
trasound. Expansion and compression waves are present during the
introduction of ultrasound into the liquid, which are resulted from the
negative and positive pressures, respectively. These fluctuations will
form bubbles which are filled with solvent, solute vapour and dissolved
gases. These bubbles will then undergo further growth and re-
compression [30]. The bubbles have very short lifetimes which break
down after an approximate period of 400 µs [14]. When the cavitation
bubble collapses, it will produce extremely high local heating and
elevated pressures as well as enhancing the mass and heat transfer
within the reaction medium owing to the formation of small eddies
[30].
In heterogeneous liquid reaction systems, ultrasound can render
physical and chemical effects simultaneously through cavitation bub-
bles. The former emulsifies immiscible reactants by the micro-turbu-
lence generated due to the radial motion of cavitation bubbles, in-
creasing the interfacial area between them and subsequently the
reaction rate. The latter enhances the chemical reaction due to gen-
eration of free radicals during the sudden collapse of cavitation bubbles
[30]. The beneficial effects of ultrasound on transesterification are ex-
plained in more details in Section 3.
3. Ultrasound-assisted biodiesel productions
The physical effect of ultrasound on transesterification reaction had
been discussed by quite a number of researchers. In transesterification
reaction, ultrasonic irradiation causes bubble cavitations near the phase
boundary between the alcohol and oil phases, thus producing large
amount of micro bubbles. Some bubbles are stable for the next cycle
while the rest will be subjected to violent collapse upon expanding to
certain critical size. The asymmetric collapse of the cavitation bubbles
generates micro-turbulence and causes disruption at the phase
boundary. Micro jets generated from the liquid impingement can reach
to a speed up to 200 m/s, creating intimate mixing of the immiscible
reactants near the phase boundary and thus induce emulsification. With
the formation of such emulsion, the interfacial region and mass transfer
between the alcohol and oil phases increases, thus the reaction kinetics
will become faster [14,29,30,32,33]. According to He and Van Gerpen
[34], the formation and bursting of micro bubbles caused by ultrasonic
cavitation intensified the local energy transfer and energized the re-
actant molecules, thus enhanced the overall reaction rate. The ultra-
sound not only provides the mechanical energy for mixing but also the
activation energy (Ea) required for transesterification reaction [35].
Several researchers believed that high local temperature (≥5000 K)
and pressure (≥1000 atm) [36] would be generated from continuous
compression and rarefaction waves of ultrasonication which then in-
duced intense mixing between the immiscible reactants. This phe-
nomena automatically increased the temperature of the reaction
medium with disruption of the microbubbles which thus increased the
889
Fuel 235 (2019) 886–907
Cavitational
Biological Sound
heat and mass transfer in the medium and promoted the desired che-
mical reactions [37–39]. Therefore, employment of ultrasound in the
production of biodiesel eliminates the requirement of external agitation
and heating because of the formation of micro jets and increment of
localized temperature. For the chemical effect of ultrasound, highly
reactive species such as OH%, HO2% and H% radicals are produced from
the dissociation of solvent vapour trapped in the bubble during a
transient implosive collapse of bubbles, which can help to accelerate
the chemical reaction [33,36].
3.1. Ultrasound-assisted homogeneous alkaline-catalyzed
transesterification
Alkaline-catalyzed transesterification is commonly applied in the
production of biodiesel since alkaline catalysts can catalyze transes-
terification process faster than acid catalysts [40]. Alkaline metal alk-
oxides and hydroxides are most frequently used alkaline catalysts for
ultrasound-assisted transesterification. At low catalyst concentration
(0.5 mol%), the former was capable of achieving significantly high
yields (> 98%) in short reaction times (30 min). Therefore, alkaline
metal alkoxides were appraised as the most active catalysts [41]. Po-
tassium hydroxide (KOH) and sodium hydroxide (NaOH) are the most
commonly used alkaline metal hydroxides. They are normally in-
expensive and able to catalyze at low reaction temperatures (from room
temperature to around 333 K) as well as complete reaction in short
reaction time (30 min) [42,43]. On top of that, high conversion rate is
achieved without intermediate steps [44]. By employing ultrasound in
alkaline-catalyzed transesterification, cavitation events help to reduce
mass transfer resistance between immiscible reactants which then ac-
celerate the chemical reaction. Ea is also reduced which enables the
reaction to be conducted at lower temperature [45]. In addition,
homogeneous alkaline-catalyzed transesterification assisted by ultra-
sound requires lower alcohol to oil molar ratio, catalyst amount, reac-
tion temperature and energy consumption as well as shorter reaction
time and phase separation time while attaining higher biodiesel yield as
compared to those without ultrasound. The detailed beneficial effects of
ultrasound could be found in Section 4.
Nonetheless, alkaline-catalyzed transesterification can only be em-
ployed if the feedstock has FFA level less than 2 wt%, which corre-
sponds to acid value of 4 mg KOH/g [46]. Oil containing high FFA
content cannot be transesterified directly using alkaline catalyst since
the FFA reacts quickly with the alkaline catalyst through saponification
side reaction which is presented in Eq. (1) to produce soap and water.
Excessive soap formation tends to gel-up the reaction mixture as soli-
dification of soap tends to occur at ambient temperature [47]. The soap
formed binds with the catalyst and partially consumes it and hence a
higher cost will be incurred [44]. Consequently, viscosity of the reac-
tion mixture increases which then interferes with the separation of ester
from glycerol. This will reduce the biodiesel conversion and yield as
well as increases the product separation cost [10,48].
R1−COOH + NaOH → R1COONa + H2 O (1)
Furthermore, high percentage of water content in feedstock oil is
S.X. Tan et al.
also undesirable. The alkaline homogeneous catalysts must be handled
properly as they are highly hygroscopic and tend to absorb water from
air during storage. They can also form water when dissolved in the
alcohol reactant [44]. For instance, water molecules can be produced
from the reaction between the hydroxide and the alcohol even if water-
free alcohol/oil mixture is used [41]. Water can hydrolyze TGs to di-
glycerides (DGs) and forms more FFA, with subsequent soap formation
which is detrimental to the biodiesel yield. The typical hydrolysis re-
action is shown in Eq. (2) [44].
(2)
3.2. Ultrasound-assisted homogeneous acid-catalyzed transesterification
For oil or fat containing high FFA content (> 2 wt%), it was re-
ported that acid-catalyzed transesterification performed better than
base-catalyzed transesterification as acid catalysts could catalyze both
esterification and transesterification concurrently. This was because
acid could act as esterification reagent and solvent at the same time.
Unlike base-catalyzed transesterification, there is no soap formation in
the presence of acid catalyst [7,30]. Acid-catalyzed transesterification
has higher tolerance towards moisture content of feedstock, which
could otherwise trigger the hydrolysis of esters resulting in regeneration
of fatty acids [44]. For oil or fat with FFA less than 2 wt%, the kinetic
rate of homogeneous acid-catalyzed transesterification was about 4000
times lower than homogeneous base-catalyzed reaction which rendered
it unfavourable in commercial applications [49]. Production of FAME
through the reaction between FFA and methanol is presented in Eq. (3).
R−COOH + CH3 OH ↔ R1COOCH3 + H2 O (3)
Sulphuric acid (H2SO4) is generally used as a catalyst in the
homogeneous acid-catalyzed transesterification because of its acid
strength which is responsible for releasing more H+ species to proto-
nate the carboxylic moiety of the fatty acid [50]. It was found out that
mechanical stirring is not a feasible approach for homogeneous acid-
catalyzed transesterification as it requires high reaction temperature
(343–473 K), high pressure (6–10 bar) and long reaction time (> 48 h).
Biphasic nature of the reaction system is the main cause for slow ki-
netics of homogeneous acid-catalyzed transesterification which affects
the catalyst accessibility to the triglyceride molecules. Although reac-
tion kinetics could be improved by increasing alcohol to oil molar ratio
(20:1–45:1) and acid concentration (≈5 wt%), it is not practical for
large scale process as higher operating cost incurred from the excessive
usage of alcohol and reactor corrosion due to higher acid concentration
[51]. Therefore, ultrasound is a practical approach to increase the
miscibility of the biphasic nature of the reaction system through the
cavitation events.
Currently, researches on the homogeneous acid-catalyzed transes-
terification of non-edible oils under ultrasonic irradiation condition
were still limited. So far, only Choudhury et al. had conducted trans-
esterification of Jatropha oil using H2SO4 [52] and chlorosulfonic acid
[33] with the aid of ultrasonication. In their study, biodiesel yield of
80.12% was successfully obtained at optimum conditions of methanol
to oil molar ratio of 7, catalyst concentration of 6 wt%, temperature of
343 K and reaction time of 2 h. Simultaneous esterification and trans-
esterification with sonication did not succeed due to formation of water
in the esterification step, and sonication alone could not overcome this
obstacle [52]. In order to overcome the inability of H2SO4 to carry out
esterification and transesterification simultaneously, Choudhury et al.
[33] had utilized chlorosulfonic acid as catalyst instead. Chlorosulfonic
acid possessed superior ability as it was capable to catalyze both es-
terification and transesterification reactions. Furthermore,
890
Fuel 235 (2019) 886–907
chlorosulfonic acid also counteracted inhibition caused by water
formed during esterification, which was the primary cause for very slow
kinetics of acid catalyzed transesterification. With the usage of chlor-
osulfonic acid, 93% of Jatropha biodiesel yield was achieved at op-
timum conditions of 8.5 wt% catalyst, 20:1 methanol to oil molar ratio,
temperature of 331 K and reaction time of 4 h. The Ea for the process
(57 kJ/mol) was at least 3 times lower than H2SO4 catalyzed transes-
terification.
Apart from chlorosulfonic acid, sulfuryl chloride (SO2Cl2) is also
able to catalyze both esterification and transesterification reactions
concurrently. It had been employed in the preparation of biodiesel from
Daturametel Linn oil seeds containing high FFA content using ultrasonic
cleaning bath (40 kHz, 130 W) [53]. Strong acids such as hydrochloric
acid (HCl) and H2SO4 were formed through the vigorous reaction be-
tween water molecules in Daturametel Linn oil and SO2Cl2 under intense
mixing. The maximum biodiesel yield of 95.5% was attained at 8.0 wt%
SO2Cl2, 7:1 methanol to oil molar ratio and temperature of 318 K with
reaction time of 2 h.
3.3. Ultrasound-assisted heterogeneous base-catalyzed transesterification
In heterogeneous catalyzed transesterification, the catalyst will be
in solid phase in contrary to methanol and oil which are in immiscible
liquid phase. Hence, the overall system is regarded as a triphasic system
(liquid/liquid/solid) [54]. By applying ultrasound in the reaction
mixture, catalyst size can be reduced to smaller particles, creating new
active sites for subsequent reactions. Therefore, the solid catalyst is
expected to last longer in the ultrasound-assisted process [55]. The
effectiveness of heterogeneous catalytic transesterification depends on
the activity of the solid catalyst employed [56]. There are several
concerns for utilizing heterogeneous catalyst in ultrasound-assisted
biodiesel production such as durability, mechanical stability and de-
activation of catalyst. These concerns are crucial because the stability of
applying the catalysts might be affected in the long term operation
under vigorous mixing condition [57]. Catalyst reusability is often used
as the parameter to indicate the catalyst stability. In addition, it is one
of the critical criteria in determining the economic feasibility of bio-
diesel production using heterogeneous catalyst [58].
Various solid base catalysts had been reported for biodiesel synth-
esis and several selected studies reporting the maximum biodiesel yield
from non-edible feedstocks at optimum conditions have been tabulated
in Table 2. It can be observed from Table 2 that maximum biodiesel
yields were in the range between 90.0% and 98.5%. In the transester-
ification of WCO using calcium diglyceroxide (CaDG), 93.5% yield was
obtained in the first cycle. Nonetheless, significant decrement in the
yield was noticed in the second and third cycles in which the yields
dropped to 59% and 50.5%, respectively. This finding might be at-
tributed to better solubility of CaDG in glycerol–methanol mixture [59].
Basically, there are three primary reasons for basic catalyst deactivation
[60]. Firstly, absorption of products and by-products on the catalyst
surface reduces the contact between basic sites and reactants. This
claim can be verified through the study done by Deng et al. [60]. The
biodiesel yield decreased from 95.2% to 80.7% after 1st and 3rd cycle,
respectively. Absorption of viscous liquid (glycerol) on the catalyst
surface was found to be the primary suspect of catalyst deactivation.
Therefore, the catalyst was washed with ethanol for removal of viscous
liquid before subjected to the next cycle. At the 8th cycle reused, bio-
diesel yield of 89.1% was recorded. Similarly, the presence of calcium
soap and glycerol phase was suspected to cause a decrement in catalytic
activity of the Karabi biodiesel production by Yadav et al. [61]. The
catalyst maintained its catalytic activity without any considerable
change up to four reaction cycles (94.1% yield). In the 5th cycle, the
biodiesel yield decreased to 65.5%. In the transesterification of Jatropha
oil catalyzed by tripotassium phosphate (K3PO4), its service life only
could maintain up to three cycles due to the methanol deposition on the
catalyst [62]. Secondly, leaching of the active sites into solutions
S.X. Tan et al. Fuel 235 (2019) 886–907

contributes to the catalyst deactivation. The biodiesel yields in the

[37]

[60]

[63]

[58]
[59]
[61]
[62]

[64]
Ref.
transesterification of Silybum marianum oil for 1st and 5th cycle were
90.1% and 80.0%, respectively [63]. Leaching of potassium compounds
might be the reason for the drop in the yield. The last reason of catalyst
Catalyst reusability

deactivation can be attributed to the disappearance of active sites due

to the collapse of catalyst structure. This claim was verified through the
X-ray powder diffraction (XRD) characterization conducted by Deng
(times)

et al. [60]. On the other hand, reduction in the biodiesel yield by using
K3PO4 was explained by two possible causes: 1) agglomeration of the

−
3

4
1
4
3
catalyst in the subsequent cycle of transesterification reaction caused a
Yield (%)

decrement in the number of active sites; 2) Side reaction of K3PO4 with

98.53

95.2

90.1

92.0
93.5
94.1
98.0

95.4
methanol resulted in the formation of dipotassium phosphate (K2HPO4)
and monopotassium phosphate (KH2PO4) during the transesterification
reaction [58].
Reaction time

3.4. Ultrasound-assisted heterogeneous acid-catalyzed transesterification

(min)

120

Ideal solid acid catalyst for transesterification reaction should ex-

15

90

30

90
30

45

30

hibit characteristics such as an interconnected system of large pores, a

Alcohol type; alcohol to oil molar

hydrophobic surface and high concentration of strong acid sites.

However, research on direct usage of solid acid catalyst for transes-
terification reaction has not been widely explored because of its lim-
itations with slow reaction rate and undesirable side reactions.
Moreover, there is still a knowledge gap on the fundamental studies
Methanol; 16:1

12:1
12:1

Methanol; 10:1

dealing with reaction pathways of TGs on solid acid catalysts [56].

Methanol; 9:1

Methanol; 4:1

6:1
9:1

Reports on the role of heterogeneous acid catalyst in ultrasound-as-

Methanol;
Methanol;
Methanol;
Methanol;

sisted process are still relatively rare [65]. The biodiesel yields at op-
ratio

timum conditions by using different solid acid catalysts are tabulated in

Table 3. The highest biodiesel yield (98%) was achieved by using
Temperature (K)

composite of phosphotungstic acid encapsulated in a Fe(III)-based

metal organic framework (PTA-MOF) while the lowest yield (84%) was
achieved by using tungstophosphoric acid/gamma alumina (TPA/Al). It
318

333

323
333
333
323

338

is also noticeable from Table 3 that slightly higher biodiesel yield was
−

attained by using n-butanol (98%) than ethanol (96%) in the ester-

Catalyst type; loading (wt

ification of oleic acid catalyzed by PTA-MOF. This observation could be

due to higher solubility of n-butanol than ethanol in oleic acid.
KHC4H4O6/TiO2; 5.0

In order to find out the influence of ultrasound on the catalyst

stability (in terms of catalyst leaching), three studies were conducted by
SrO-CaO; 6.0
Na/SiO2; 3.0

Mg/Al; 1.0

K3PO4; 3.0

K3PO4; 1.0

Badday et al. on tungstophosphoric acid/activated carbon (TPA/AC),

CaDG; 1.0
CaO; 5.0

TPA/Al and tungstophosphoric acid/cesium (TPA/Cs) [65–67]. The

results revealed that only 11%, 9.6% and 9.2% of the FAME yield were
%)

achieved under the optimum conditions due to the minimal leaching of

3−
anions, for TPA/AC, TPA/Al and TPA/Cs, respectively. Thus,
Biodiesel yield at optimum conditions by using different solid base catalysts.

PW12 O40
Silybum marianum

all three catalysts had demonstrated good catalyst stability with low
Jatropha oil

Jatropha oil

Jatropha oil

Jatropha oil

leaching behaviour. For sulfonated solid carbon derived from cyclo-

Feedstock

Karabi oil

dextrin (SO3H-CD) without regeneration, the FAME yield decreased in

WCO
WCO

each cycle, from 90.8% (fresh) to approximately 81% (4th cycle). This
oil

observation could be attributed to the deactivation of the active sites

Power (W)

due to their poisoning by other impurities present in the reaction

mixture such as ion-exchange between alkali cations with protons and
200

210

250

375
120

600

210
50

natural adsorption of reactants and products in the system [68]. After

the regeneration process of SO3H-CD with repeating sulfonation, de-
Frequency (kHz)

activation problem was resolved and the catalyst could be reused for 4
cycles without any serious reduction in the FAME yield. For PTA-MOF,
it could be reused for 5 cycles and approximately 5 wt% of PTA was
20–30

leached during the recovery process in both esterification processes

24

40

22
22

20
−

using ethanol and n-butanol. Electrostatic interaction between PTA

Direct sonication (horn)
Direct sonication (horn)
Direct sonication (horn)
Ultrasonic reactor type

anions and Fe(III) network that was rich in amino groups played a
Indirect sonication

Indirect sonication

Indirect sonication

Indirect sonication

Indirect sonication

crucial part in minimizing leaching from the composite [69].

3.5. Two-step homogeneous and heterogeneous ultrasound-assisted

transesterification
(bath)

(bath)

(bath)

(bath)

(bath)

High initial acid content in non-edible oil will hamper transester-

Table 2

2010

2011

2014

2014
2015
2016
2016

2017
Year

ification approach using alkaline catalysts due to the soap formation.

On the other hand, acid catalyst can catalyze non-edible feedstock

891
 S.X. Tan et al.

Table 3
Biodiesel yields at optimum conditions by using different solid acid catalysts.
Year Ultrasonic reactor type Frequency (kHz) Power (W) Amplitude (%) Feedstock Catalyst type; loading (wt%) Temperature (K) Alcohol type; alcohol to oil molar ratio Reaction time (min) Yield (%) Ref.

2013 Direct sonication (probe) 20 400 60 Jatropha oil TPA/AC; 4.23 338 Methanol; 25:1 40 91 [66]
2013 Direct sonication (probe) 20 400 60 Jatropha oil TPA/Al; 4.44 338 Methanol; 19:1 50 84 [65]
2013 Direct sonication (probe) 20 400 60 Jatropha oil TPA/Cs; 2.9 338 Methanol; 25:1 34 90.5 [67]
2015 Direct sonication (probe) 25 − − WCO SO3H-CD; 11.5 390 Methanol; 20:1 8.8 90.8 [68]
2017 Indirect sonication (bath) 37 100 − Oleic acid PTA-MOF; 30.0 Ambient Ethanol; 16:1 15 96 [69]
2017 Indirect sonication (bath) 37 100 − Oleic acid PTA-MOF; 30.0 Ambient n-butanol; 16:1 15 98 [69]

892
Table 4
Optimum conditions for two-step homogeneous and heterogeneous ultrasound-assisted transesterification.
Year Ultrasonic reactor type Frequency (kHz); Power (W) Feedstock Reaction conditions Biodiesel conversion (%) Yield (%) Ref.

2014 Indirect sonication (bath) 35; 35 Jatropha oil 1st step: Methanol to oil (15:1), 5 wt% H2SO4, 338 ± 2 K, 60 min (mechanical stirring) − > 95 [54]
2nd step: Methanol to oil (11:1), 5.5 wt% CaO, 337 K, 120 min (ultrasonication)
2014 Direct sonication (probe) 30; 36 Karanja oil 1st step: Methanol to oil (9:1), 0.5 wt% H2SO4; 303 K, 90 min 84 − [73]
2nd step: Methanol to oil (9:1), 5 wt% Ba(OH)2·8H2O, 303 K, 60 min
2015* Direct sonication (horn) 20; 200 Neem oil 1st step: methanol to oil (0.45 vol%), 0.75 vol% H2SO4, 333 K, 10 min − 98.01 [74]
2nd step: methanol to oil (0.2 vol%), 1.2% KOH, 333 K, 10 min
2015* Direct sonication (horn) 20; 400 Neem oil 1st step: methanol to oil (6:1), 3 wt% H2SO4, 303 K, 30 min 92.72 − [75]
2nd step: methanol to oil (7:1), 1.2% KOH, 323 K, 30 min
*
2015 Direct sonication (horn) 20–30; 1000 Mahua oil 1st step: methanol to oil (0.35 mol%), 1 vol% H2SO4, 323 K, 5 min − 97.4 [71]
2nd step: methanol to oil (1.5 mol%), 0.75% KOH, 323 K, 5 min
2016* Direct sonication (horn) 28; 50 Nerium oleander oil 1st step: methanol to oil (0.4 vol%), 1 vol% H2SO4, 328 K, 10–15 min − 97 [76]
2nd step: methanol to oil (0.2 vol%), 1 v/w% KOH, 328 K, 10–15 min
2016 Direct sonication (probe) 20; 250 Schleichera triguga oil 1st step: Methanol to oil (4:1), 1 vol% H2SO4, 313 K, 20 min − 96.8 [77]
2nd step: Methanol to oil (9:1), 3 wt% Ba(OH)2, 323 K, 80 min
2017 Direct sonication (probe) 30; − Rubber seed oil 1st step: Methanol to oil (15:1), 1.5 wt% H2SO4; 303 K, 120 min − 97 [78]
2nd step: Methanol to oil (8.09:1), 5.38 wt% Ba(OH)2·8H2O, 303 K, 13.2 min

*
Note: denotes for two-step homogeneous ultrasound-assisted transesterification.
Fuel 235 (2019) 886–907
S.X. Tan et al.
containing high acid value but with lower reaction rate. Therefore, a
two-step process (esterification followed by transesterification) is an
efficient approach to address the undesirable saponification phenom-
enon and slow reaction rate while improving the biodiesel conversion
or yield. The feedstock was usually first treated with acid catalysts to
reduce FFA content to below 2 wt%, followed by transesterification
with alkaline catalysts. With the assistance of ultrasound in biodiesel
production, the processing cost can be greatly reduced due to lower
amount of methanol and operating temperature needed, compared to
the conventional two-stage approach [70–72]. Biodiesel conversions or
yields from different feedstocks at optimum conditions by employing
two-step homogeneous ultrasound-assisted transesterification are
summarized in Table 4. High biodiesel conversion or yield
(92.72%−98.01%) can be achieved as observed from Table 4.
Fig. 2. Interesterification reaction (Initial 3 equations show the initial steps whereas final equation gives the overall reaction). Adapted with permission from
(Maddikeri GL, Pandit AB, Gogate PR) [25]. Ultrasound assisted interesterification of waste cooking oil and methyl acetate for biodiesel and triacetin production.
Fuel Process Technol 2013;116:241–9. Copyright (2013) Elsevier Ltd.
893
Fuel 235 (2019) 886–907
3.6. Ultrasound-assisted interesterification
Interesterification is different from transesterification in term of
acryl acceptor used and the by-product produced. In biodiesel synthesis
via interesterification process catalyzed by base catalyst, alkyl acetate
will replace alcohol as acryl acceptor which then produces a short chain
triglyceride instead of glycerol. For instance, triacetin is yielded as by-
product instead of glycerol when methyl acetate is used during inter-
esterification [79]. This complex reaction is composed of three con-
secutive reversible reactions, which are shown in Fig. 2 [25]. In inter-
esterification reaction catalyzed by acid catalyst, FFA in vegetable oil
reacts with methyl acetate to form biodiesel and acetic acid [80].
Generally, there are several drawbacks associated with the trans-
esterification process. For enzyme catalyst, excess utilization of me-
thanol will lead to deactivation of enzyme and the by-product (glycerol)
will have inhibitory effects on the enzymatic activity. These
S.X. Tan et al.
unfavourable issues will restrict the application of enzymatic route at
industrial scale [81]. With the co-production of glycerol, neutralization
is required to prevent biodiesel contamination. Separation and pur-
ification steps are needed for recovery of glycerol. All these steps are
energy and resource intensive, decreasing the production efficiency and
increasing the production costs. In addition, glycerol is undervalued in
the world market due to its excessive supply derived from biodiesel
production [80,82]. With the employment of interesterification instead
of transesterification, enzyme deactivation could be avoided as triacetin
has no adverse effect on the activity of lipase enzyme [81]. Further-
more, production of triacetin is more preferable as it has higher com-
mercial value than glycerol. Triacetin can be used as plasticizer or ge-
latinizing agent in polymers and explosives and as an additive in
tobacco, pharmaceutical industries, and cosmetics. In addition, tria-
cetin is soluble in biodiesel and could be added into biodiesel (up to
10% by weight) as an additive which could help to improve oxidative
stability and cold flow properties of biodiesel [25,79].
By focusing on the non-edible feedstock, there are only several
studies conducted on the ultrasound-assisted interesterification. The
biodiesel yields obtained at optimum reaction conditions using different
catalysts are shown in Table 5. The superiority of ultrasound as com-
pared to conventional approach in interesterification had been proven
by researchers. For example, Maddikeri et al. [25] observed that higher
biodiesel yield was achieved by ultrasound (90%) than conventional
stirring (70%) under the same optimum reaction conditions. In the
study conducted by Subhedar and Gogate [81], they discovered that
ultrasound-assisted interesterification required lower methyl acetate to
oil ratio (9:1 versus 12:1), lower enzyme loading (3% versus 6%),
shorter reaction time (3 h versus 24 h) and achieved higher biodiesel
yield (96.1% versus 90.1%) as compared to conventional stirring. Based
on the several existing literatures, it is suggested that more studies
should be conducted on the ultrasound-assisted interesterification with
the utilization of different catalyst types to produce the optimum bio-
diesel yield (chemical and enzyme catalysts).
4. Primary factors for ultrasound-assisted transesterification
There are several factors that will influence the biodiesel conversion
or yield in the ultrasound-assisted transesterification such as alcohol to
oil molar ratio, catalyst loading, reaction time, reaction temperature,
energy consumption and ultrasonic pulse mode. Although mechanical
or magnetic stirring are commonly employed in the production of
biodiesel, ultrasonic cavitation appears to be more resource-effective
than conventional stirring approach in terms of alcohol to oil ratio,
catalyst loading, reaction time, reaction temperature, energy con-
sumption, phase separation time and biodiesel conversion or yield.
4.1. Alcohol to oil molar ratio
A lower alcohol to oil molar ratio will definitely minimize the en-
ergy consumption because alcohol separation using distillation is an
energy-intensive operation, which controls the overall economics of
biodiesel synthesis process [70]. By incorporating ultrasound in the
biodiesel synthesis, lower molar ratio will be required as compared to
conventional stirring. This is due to the intense micro-turbulence and
microstreaming generated during the cavitation which contributes to
Table 5
Optimum reaction conditions for ultrasound-assisted interesterification of non-edible oil.
Feedstock Catalyst Temperature (K) Methyl acetate to oil
molar ratio
WCO Potassium methoxide 313 12:1
WCO Thermomyces lanuginosus 323 9:1
(Lipozyme TLIM)
Crambe oil Novozym®435 333 12:1
894
Fuel 235 (2019) 886–907
the intensification of the transesterification or esterification processes
[84]. This claim had been verified by several researchers through their
studies such as Kumar et al. (6:1 versus 8:1) [46], Yin et al. (5:1 versus
6:1) [49], Mohod et al. (5:1 versus 6:1) [85] and Mohan et al. (4.5:1
versus 6:1) [84] in which the values in bracket represent the alcohol to
oil molar ratios required by ultrasonication and mechanical stirring
approaches, respectively. Based on their findings, it was revealed that
ultrasonication could achieve 16.67–25% reduction in alcohol to oil
molar ratio as compared to mechanical stirring.
Methanol is the most preferable alcohol in transesterification due to
its lower cost, industrial availability and superior reactivity as com-
pared to other alcohols. Its short molecular chain and polar properties
serve as its physical and chemical advantages. It can also react with TGs
efficiently and dissolve with the basic catalyst easily [2]. Molar ratio of
methanol to oil is one of the essential factors affecting the rate of for-
mation and yield of FAME. Theoretically, molar ratio of methanol to oil
for the transesterification is 3:1. However, large excess amount of me-
thanol is required to drive the reversible transesterification reaction
towards FAME [5]. Excess alcohol in transesterification can enhance
alcoholysis and ensure complete conversion of oils or fats into esters.
Furthermore, better ester conversion can be achieved in a shorter time
with a higher alcohol to triglyceride ratio [44]. For heterogeneous
catalyst, excess alcohol can promote transesterification rate in addition
to remove product molecules from the catalyst surface and regenerate
the active sites [86]. The effects of molar ratio of methanol to oil on
ultrasound-assisted alkaline transesterification using homogeneous and
heterogeneous catalysts are tabulated in Table 6. It is noticed from
Table 6 that a molar ratio of more than 6:1 was mostly required to
achieve optimum biodiesel yield.
Further increasing the molar ratio beyond the optimum will de-
crease the yield which might be caused by over dilution of biodiesel and
glycerol [38,62,87]. This is because higher dissolution of glycerol and
alcohol in biodiesel will affect the purity of biodiesel if the amount of
methanol increases in the mixture [87]. Moreover, the equilibrium re-
action might shift towards the reactant side, which reduces the effective
yield [58]. In addition, formation of undesirable suspension resulting
from the unreacted methanol and oil mixture has increased the diffi-
culty of separation and longer duration is required as compared to
lower methanol ratio [38].
The optimum methanol to oil molar ratio is also affected by the
phase of catalyst. Generally, heterogeneous catalysts require slightly
higher molar ratio than homogeneous catalysts due to slower rate
which stemmed from mass transfer limitation [88]. By comparing the
optimum molar ratio required by both homogeneous and hetero-
geneous catalysts in Table 6, 16:1 is the highest molar ratio required by
heterogeneous catalysts whereas 12:1 is the highest molar ratio re-
quired by homogeneous catalysts. This observation tallied with the
statement that heterogeneous catalysts required slightly higher molar
ratio than homogeneous catalysts. From Table 6, it is also observed that
transesterification of WCO had been conducted by several groups of
researchers by employing homogeneous (KOH and NaOH) and hetero-
geneous (K3PO4 and CaDG) catalysts. The general optimum molar ratios
required by each type of catalyst were as followed: KOH and K3PO4
(6:1); NaOH and CaDG (9:1). Similar optimum molar ratio was reported
using KOH and K3PO4 to achieve maximum biodiesel yield or conver-
sion. This could be attributed to similar catalytic activity of K3PO4 and
Catalyst concentration Ultrasonic power Reaction time Biodiesel yield Ref.
(wt%) (W) (h) (%)
1 450 0.5 90.0 [25]
3% (w/v) 80 3 96.1 [81]
20 − 6 98.25 [83]
S.X. Tan et al. Fuel 235 (2019) 886–907

KOH [62].

[89]
[63]
[58]
[59]

[31]

[31]

[90]
[90]
[77]

[61]
[62]
Ref.

[4]
Yield (%)

4.2. Catalyst loading

88.4
92.0
93.5

96.8

94.1
81

98

98
−

−
−
−
The reaction of oils and alcohols is a slow process due to the im-
miscibility between oils and alcohols [7]. Hence, catalyst is required for
Biodiesel conversion

transesterification to accelerate the reaction and gives a better con-

version efficiency of oil to biodiesel [91]. Ultrasonication had been
proven by several researchers [46,85,92] that it could help in de-
creasing the catalyst loading required in either ultrasound-assisted
∼80

96.8
(%)

homogeneous or heterogeneous transesterification as compared to

99
99
−
−
−

−
−
conventional stirring. In the study conducted by Kumar et al. [46],
mechanical stirring required higher KOH concentration which was 1 wt
(min)
Time

% than ultrasonication to achieve the same conversion of 98.2% in

120
25
90
30

45

80

60
8

1
1
which 0.75 wt% was only needed by ultrasonication. Similar finding
was also reported by Chen et al. [92] in the transesterification of palm
Optimum molar

oil catalyzed by CaO. Lower catalyst amount (8 wt%) was able to

achieve maximum biodiesel yield (92.7%) in the case of ultrasonic
cavitation than magnetic stirring (55% biodiesel yield with 10 wt%
ratio

16:1

12:1

12:1
12:1
6:1

6:1
9:1

9:1

9:1

5:1
7:1
9:1

CaO). Based on their results, it could be deduced that ultrasonication

could achieve 20–25% reduction in catalyst loading. In the esterifica-
Temperature (K)

tion of Karanja oil conducted by Mohod et al. [85], same catalyst

loading (2 wt% H2SO4) was required by ultrasound and mechanical
stirring approaches to reduce the acid value of Karanja oil from
14.15 mg KOH/g oil. Nonetheless, the final acid value achieved by ul-
333
323
333

318

323
323
323

333
323
−

trasound approach was 50% lower than that of mechanical stirring

Catalyst type; loading

method. Final acid values obtained by ultrasound and mechanical

KHC4H4O6/TiO2; 5.0
Effects of methanol to oil molar ratio on ultrasound-assisted homogeneous and heterogeneous base-catalyzed transesterification.

stirring approaches were 2.7 and 4.1 mg KOH/g oil, respectively.

In homogeneous transesterification, the beneficial effect was due to
Ba(OH)2; 3.0
NaOH; 1.25

NaOH; 1.25
K3PO4; 3.0

K3PO4; 1.0
NaOH; 1.0

NaOH; 1.0
CaDG; 1.0

higher chemical activity in the presence of ultrasound cavitations [46].

KOH; 1.0

CaO; 5.0
KOH; 1

In heterogeneous transesterification, the grinding effect of ultrasound

(wt%)

on the catalyst could help to expose new catalytic active sites and in-
crease the interfacial area between catalyst and reactant [37,77].
Karanj and soybean oil (40:

Therefore, emulsion droplets of alcohol and oil produced could contact

with the catalyst particles easily, thus achieving effective mixing be-
Silybum marianum oil

Schleichera triguga oil

tween liquid reactants and solid catalyst [93]. In addition, micro-

streaming effect of ultrasound could also help to maintain the catalytic
Jatropha oil
60) (w/w)

activity by dislodging the inactive surface formed and attached on the

Karabi oil
Feedstock

denotes for ultrasound-assisted homogeneous alkaline-catalyzed transesterification.

catalyst, thus providing more active sites for the catalytic reaction [94].
WCO

WCO
WCO

WCO

WCO

WCO
WCO

Inadequate catalyst loading can lead to incomplete conversion of TG

into FAME. As the catalyst loading increases, the conversion of TG and
Power (W)

the yield of biodiesel will also increase until an optimum catalyst

loading is attained. Further increment in the catalyst loading will cause
240
250
375
120

150

150

500

500
500
250

600
50

a decrement in the yield and increase the production cost. With ex-
cessive addition of alkaline catalyst beyond the optimum loading, al-
kaline catalyst can cause more TGs to react with the catalyst through
Frequency

saponification [95]. During washing, the soap present in the ester phase
20–30
(kHz)

has the tendency to accumulate at the interfacial region between the

20
40
22
22

20

20

20

20
20
20

20

liquids. Due to its hydrophilic and hydrophobic structure, the soap

molecules are oriented perpendicular to the interfacial region between
Direct sonication (horn) (Pulse

the two immiscible phases: water and FAME. The soap molecules inside
Indirect sonication (bath)

Indirect sonication (bath)

Direct sonication (probe)
Direct sonication (horn)

Direct sonication (horn)

Direct sonication (horn)

Direct sonication (horn)

Direct sonication (horn)

Direct sonication (horn)

Direct sonication (horn)

Direct sonication (horn)

FAME and water molecules will form emulsion, thus hindering the
Ultrasonic reactor type

purification of FAME [96]. In addition, large amount of catalyst in-

(Continuous mode)

creases the solubility of FAME in the glycerol, causing entrainment of

FAME in the glycerol phase after phase separation [97]. Decrement in
the FAME yield with excessive catalyst loading was encountered by
mode)

Hingu et al. [98] and Kesgin et al. [99] in their studies.

In heterogeneous transesterification, the number of catalyst parti-
cles and active sites increase with the catalyst loading and thus in-
4.5:1, 6:1, 9:1,

4.5:1, 6:1, 9:1,

Various molar

6:1, 9:1, 12:1,

creasing the accessibility of TG and alcohol to the catalyst surface.

6:1, 9:1, 12:1

6:1, 9:1, 12:1

3:1, 6:1, 9:1,
4:1, 6:1, 8:1
8:1 to 18:1

9:1 to 17:1

Further increment of catalyst loading beyond the optimum value will

4:1 to 7:1

4:1 to 7:1
4:1 to 7:1

cause either a decrement or negligible effect on the biodiesel conversion

13.5:1

13.5:1
ratio

14:1

12:1

or yield due to the reduction of interspatial space between TG and al-

cohol molecules [77]. Moreover, increasing catalyst beyond the op-
Table 6

2014*

2015*

2015*

2016*
2016*

*
2014
2015
2015

2016

2016

2016
2016
Year

timum loading will cause formation of emulsion layer which makes the
Note:

biodiesel separation to be difficult [62]. In Jatropha biodiesel synthesis

895
S.X. Tan et al. Fuel 235 (2019) 886–907

assisted by ultrasonic irradiation, excessive magnesia-alumina (Mg/Al)

[89]
[63]
[31]

[31]

[58]
[59]

[90]
[90]
[77]
[61]
Ref.

[4]
(> 1.0 wt%) as catalyst was found to cause a decrease in biodiesel yield

Yield (%)
in addition to the formation of emulsion layer [60]. It should be noted
that inadequate catalyst loading will give rise to unsatisfactory bio-

88.4
96.8

92.0
93.5

94.1
81

98

−
diesel yield and incur higher production costs. Therefore, optimum

–
–
amount of catalyst is required to effectively convert the TG into FAME.
It is noticed from Table 7 that the range of catalyst loading to achieve

conversion (%)
maximum yield or conversion was between 0.5 and 1.5 wt% for ultra-

Biodiesel
sound-assisted homogeneous alkaline-catalyzed transesterification
while higher range of catalyst loading (1.0 to 5.0 wt%) was generally

∼85

96.8
100
99
−

−
−

−
required for ultrasound-assisted heterogeneous base-catalyzed transes-

–
terification.

(min)
Time

120

100
120
25

40
45
4.3. Reaction time

1
1
Alcohol type; alcohol to oil
It had been documented that shorter reaction time was required by
ultrasonication compared to the conventional stirring. Many re-
searchers had compared the reaction time needed by both approaches

Methanol;16:1

Methanol;12:1
Methanol; 6:1

Methanol; 9:1

Methanol; 9:1

Methanol; 6:1

Methanol; 6:1

Methanol; 6:1
Methanol; 6:1
Methanol; 9:1
Methanol;9:1
in their studies which are shown in Table 8. Ultrasonication can reduce

molar ratio
the reaction time by at least 25% and as high as 95.8% compared to the
conventional stirring. The reactions under ultrasonication are more
reactive compared to mechanical stirring process. This could be at-

Temperature (K)
tributed to the larger interfacial area of reactant droplets due to the
physical effects of cavitation phenomena in terms of intense turbulence
and mixing generated in the reactor which then led to the enhancement
in the biodiesel yield or conversion [8,100,101]. The average interfacial

333

333
333
318

323
323
323
333
area generated by ultrasonication was found to be 67 times higher than

–
that generated by mechanical agitation. The droplet sizes for ultrasonic
Catalyst type; optimum

KHC4H4O6/TiO2; 5.0
emulsification were in the narrow range of 0.82–44.6 µm. For emulsi-
fication by mechanical agitation, a much wider droplet size distribution
loading (wt%)
Effects of catalyst loading on ultrasound-assisted homogeneous and heterogeneous base-catalyzed transesterification.

was obtained, ranging from 8.1 to 610 µm [102].

Ba(OH)2; 3.0
NaOH; 1.25

NaOH; 0.75
K3PO4; 3.0
NaOH; 0.5

NaOH; 1.5

KOH; 1.25
CaDG; 1.0
Theoretically, the conversion of fatty acid esters will increase with
KOH; 1.0

CaO; 5.0
reaction time [103]. The reaction is slow initially due to the mixing and
dispersion of alcohol into the oil. After some time, the reaction will
proceed rapidly. However, prolonged reaction time will cause a de-
Karanj and soybean oil
Silybum marianum oil

Schleichera triguga oil

crement in the product yield due to the backward reaction of transes-
terification, causing more FFAs to form soaps [44,104]. The transes- (40: 60) (w/w)
terification reaction was found to be dependent on the reaction time in

denotes for ultrasound-assisted homogeneous alkaline-catalyzed transesterification.

the ultrasonic transesterification of Silybum marianum oil catalyzed by
Karabi oil
Feedstock

KHC4H4O6/TiO2 [63]. Low reaction rate in the first 10 min could be due
WCO

WCO

WCO

WCO
WCO

WCO
WCO

to inadequate agitation to enhance proper mixing and dispersion of

methanol and catalyst onto the oil. The maximum biodiesel yield
Power (W)

(90.1%) was attained in 30 min. However, excess reaction time (after

30 min) resulted in slight reduction in biodiesel yield due to reversible
240
250
150

150

375
120
500

500
500
250
50

reaction.
The influence of reaction time (from 0.5 to 2.5 min, with increments
of 0.5 min) on the transesterification of WCO was investigated by
Frequency

Martinez-Guerra and Gude [31] for both pulse and continuous sonica-
20–30
(kHz)

tion conditions. The highest yield of 98% was achieved by pulse soni-
20
40
20

20

22
22
20

20
20
20

cation at 2.5 min and 91% followed by continuous sonication at 1 min.

During pulse sonication, the biodiesel yield increased with reaction
Indirect sonication (bath)

(probe)
Direct sonication (horn)

Direct sonication (horn)

Direct sonication (horn)

Direct sonication (horn)

Direct sonication (horn)
Direct sonication (horn)

(horn)
(horn)

(horn)
Ultrasonic reactor type

time. On the other hand, the biodiesel yield began to drop upon
(Continuous mode)

reaching 1.5 min reaction time for continuous sonication. The drop in
sonication
sonication
sonication
sonication

the biodiesel yield with increasing reaction time by the usage of con-
(Pulse mode)

tinuous sonication had validated the recommendation given by Chand

et al. [105]. Continuous ultrasound irradiation was not recommended
Direct
Direct
Direct
Direct

for long reaction time due to local temperature increase (hot spots)
which could degrade the oil and cause undesired reactions, leading to a
decrement in the biodiesel yield [106].
0.5, 0.75, 1, 1.25

0.5, 0.75, 1, 1.25

Various catalyst
loading (wt%)

0.5, 1.25, 2

0.5, 1.25, 2

0.5, 0.75, 1
0.5 to 1.25
0.5, 1, 1.5

4.4. Reaction temperature

1.0 to 9.0

1.0 to 4.0

1.0 to 4.0
3.0 to 5.0

For reactions with conventional stirring methods, external heating is

paramount and reaction rate will heavily depend on the reaction tem-
Table 7

2014*

2015*

2015*

2016*

2016*
2016*

*
2014

2015
2015

2016
2016
Year

perature with vigorous mechanical mixing [38]. On the other hand,

Note:

ultrasonic mixing eliminates the need of external heating. The

896
S.X. Tan et al. Fuel 235 (2019) 886–907

Table 8
Reaction time needed by employing ultrasonic cavitation and conventional stirring.
Year Feedstock Reaction conditions Ultrasonic cavitation Conventional stirring Reduction Ref.
percentage (%)
Reaction Yield (%) Conversion (%) Reaction Yield (%) Conversion (%)
time (min) time (min)

2010 WCO • Methanol to oil (9:1) 5 − 94.6 120 89.3 95.8 [100]
• 1Temperature:
wt% NaOH (mechanical)
• Methanol to oil343(9:1)K
a

2014 Palm oil • 1 wt% KOH 15 98.5 − 120 89.5 87.5 [8]
• Temperature: 333 K (mechanical)
• Methanol to oil (3:1)
Palm oil • 1 wt% KOH 15 93 − 150 75 90
• Temperature: 333 K (mechanical)
• Methanol to oil ratio (9:1)
2014 Palm oil • 8 wt% CaO 40 70 − 100 70 (magnetic) − 60 [92]
• Temperature: 333 K
• Methanol to oil (8:1)
2014 Silybum • 1 wt% KOH 15 82.46 − 50 82.35 − 70 [107]
marianum oil • Temperature: 333 K (mechanical)
• Methanol to oil (6:1)
2015 Waste fish oil • 1 wt% KOH 30 79.6 − 60 78 − 50 [108]
• Temperature: 328 K (mechanical)
• Methanol to oil (7:1)
2015 Cynara • 1 wt% NaOH 20 97 − 60 95.8 − 40 [109]
cardunculus L. • Temperature: 333 K (mechanical)
seed oil • Methanol to oil (12:1)
2016 Jatropha oil • 1 wt% K PO 45 98 − 60 92* − 25 [62]
• Temperature: 323 K
• Ethanol to oil ratio (6:1)
3 4
b

2016 Citrullus vulgaris • 1.5 wt% crystalline 120 87 − 480 78* − 75 [110]
seed oil • manganese carbonate
Citrullus vulgaris
seed oil
• Isopropanol
(6:1)
to oil ratio 180 86 − 720 74* − 75

• 1.5 wt% crystalline

manganese carbonate
2016 Nerium oleander • 1st step: Methanol to oil
molar ratio (0.4 vol%),
1st step:
10–15
97 − 1st step: 60
2nd step: 60
92 (magnetic) − 75–83 [76]

1 vol% H2SO4, 328 K 2nd step:

• 2nd step: Methanol to oil
molar ratio (0.2 vol%),
10–15

1v/w% KOH, 328 K

2017 Jatropha oil • Methanol to oil (10:1) 30 95.4 − 180 89.3 ≈ 83 [64]
• 6Temperature:
wt% SrO-CaO (magnetic)
• 338 K

Note: * denotes for stirring type is not mentioned in the study; a

denotes for temperature used in conventional stirring only; b
denotes for temperature used in
ultrasonication only.

capability of ultrasound to increase the temperature inherently can be
proven through the experiments performed by Boffito et al. [111]. In
the ultrasound experiments at 336 K, the thermostat was set at 313 K
and the ultrasound provided the heat to achieve the desired bath
temperature. In fact, the bath temperature reached 336 K within sec-
onds of initiating sonication which was a lot faster compared to con-
ventional heating. The efficiency of sonication over magnetic stirring at
room temperature had been studied by Ragavan and Roy [112].
Transesterification of rubber seed oil was completed at 305 K with an
optimum yield of 80.7% which could not be achieved by magnetic
stirring method. No visible separation of methyl esters was observed
after 15 min of magnetic stirring with methanol/oil/KOH molar ratio
6:1:0.13 at room temperature. Higher temperature was required by
magnetic stirring method to initiate the transesterification process.
Since ultrasound-assisted transesterification can take place at room
temperature (303 K), which is much lower than the boiling point of
alcohol, it could reduce external heat energy effectively [113].
The reaction rate and yield of the biodiesel product are clearly in-
fluenced by reaction temperature [2]. At lower reaction temperature,
the oil has higher viscosity and impedes the bubble formation, resulting
in poor mixing between the oil and alcohol-catalyst phases [38,107]. As
reaction temperature rises, the viscosity of oil will decrease and it be-
comes susceptible to cavitation. Consequently, the miscibility of alcohol
in oil increases and subsequently promotes biodiesel formation [36]. In
addition, the kinetic rate constant of this endothermic chemical reac-
tion also increases since there is higher energy input for the reaction to
take place [2,36]. Thus, shorter reaction time can also be achieved at
higher temperature. However, it is important to avoid vaporization of
alcohol which will reduce the contact area with the reactant [2]. In
heterogeneous transesterification reaction, the effect of reaction tem-
perature is very important as the system consists of three phases (oil-
alcohol-catalyst). The reaction can be slow due to the presence of dif-
fusion barrier among the different phases [63]. The diffusion resistance
between different phases will be minimized with increasing tempera-
ture as the viscosity of the reaction mixture is reduced. Therefore, so-
lubility of alcohol in oil phase will increase and the catalyst will have a
better contact with the reactants [58].
Nonetheless, beyond the optimum reaction temperature, the bio-
diesel yield will decrease because a higher reaction temperature will
also accelerate side reactions such as saponification [44,104]. Another
reason can be ascribed to the damping effect of the cavitation effects at
higher operating temperatures [4,98]. At higher reaction temperature,
the equilibrium vapour pressure will increase which leads to easier
bubble formation. However, the cavitation bubbles formed contain
more vapours. The release of solvent vapours interferes with the cavi-
tation effects, causing cushioning of collapse. Therefore, bubbles

897
S.X. Tan et al.
implode with less intensity and reduces the mixing effect of ultrasound
on the transesterification reaction, resulting in reduced mass transfer
and biodiesel yield [31,38,59].
4.5. Energy consumption
In terms of energy consumption, mechanical stirring usually re-
quires higher energy consumption than ultrasonication. It was reported
that ultrasonication only required one-third to a half of the energy that
was consumed by mechanical agitation [30]. Yin et al. [49] had re-
corded the energy consumption for both mechanical stirring and ul-
trasonic probe irradiation methods to reach biodiesel conversion of
95% from sunflower oil. It was observed that mechanical stirring re-
quired 0.31 kWh whereas ultrasonic probe irradiation only needed
0.18 kWh. Ultrasonication was also discovered to be more energy effi-
cient than conventional stirring by other researchers such as Gupta
et al. [59], Yadav et al. [61] and Brasil et al. [114]. More energy was
consumed by conventional stirring as continuous heating was required
to maintain the temperature of water bath or solution [49,61]. External
heating source is unnecessary in biodiesel synthesis aided by ultrasonic
irradiation. This is because cavitation from ultrasound will result in
localized increment in temperature at the phase boundary and subse-
quently enhanced the reaction. Therefore, less energy is consumed in
the production of biodiesel in the long run [49].
Delivery of the ultrasonic power to the reaction mixture determines
the degree of cavitation and thus directly affecting the biodiesel yield
[59]. Optimization of ultrasonic power allows achieving the highest
biodiesel conversion and yield at the lowest possible power, which re-
duce the production cost of biodiesel. Generally, biodiesel conversion
increases with an increment in ultrasonic power until optimum ultra-
sonic power is reached. When the ultrasonic power is higher, the col-
lapse of the cavitation bubbles will be harsher due to higher jet velocity,
thus enhancing micromixing between the oil and methanol phases at
the phase boundary. Finer emulsion is then formed, increasing mass-
transfer coefficient and biodiesel conversion. On the other hand, bio-
diesel conversion is reduced when further increasing the ultrasonic
power beyond the optimum value. Cushioning effect at higher ultra-
sonic power reduces the energy transfer into the system, thus providing
lower cavitation activity [36]. Furthermore, huge amount of cavitation
bubbles will be generated when high ultrasonic power is introduced
into the reaction mixture. Excessive bubbles will likely merge and form
larger and more stable (long-lived) bubbles and create a barrier to
acoustic energy transfer. This phenomenon of efficiency loss in the
power transfer is referred as decoupling effect [31].
Fig. 3. Wave profiles for (a) continuous sonication (b) pulse sonication. Adapted with permission from (Martinez-Guerra E, Gude VG) [31]. Continuous and pulse
sonication effects on transesterification of used vegetable oil. Energy Convers Manage. 2015;96:268–76. Copyright (2015) Elsevier Ltd.
898
Fuel 235 (2019) 886–907
4.6. Phase separation time
Generally, the separation of the reaction mixture containing FAME,
glycerol and excess methanol requires several hours through gravity
settling. The separation time depends on the amount of excess methanol
used and the temperature of the reaction mixture. Separation will
generally occur faster in the reaction mixture containing less excess of
methanol. This fact could be explained by the differences in the density
of the glycerol phase due to the varying excess volume of methanol.
Glycerol has much higher density (1.26 g/cm3) than methanol (0.79 g/
cm3), and both liquids are polar compounds which will form a uniform
phase at any mixing ratio. Therefore, the glycerol layer with smaller
excess of methanol has a higher density, resulting in a faster phase
separation between the FAME layer and the glycerol layer due to the
larger difference in the density of both layers. Lower temperature of the
reaction mixture will contribute to lower dissolution of glycerol and
methanol in the FAME layer, thus leading to shorter separation time
[5].
As mentioned previously in Sections 4.1 and 4.4, ultrasound ap-
proach requires lower alcohol to oil molar ratio and lower reaction
temperature compared to conventional stirring method. Therefore, it
can be assumed that lower excess of alcohol is usually present through
ultrasound approach. The assumption could be further affirmed since
ultrasonic transesterification requires lower reaction temperature than
conventional transesterification using either mechanical or magnetic
stirring. Hence, the overall temperature of the reaction mixture will be
lower for ultrasonic agitation.
In the transesterification of WVOs carried out by Refaat et al. [115],
biodiesel yield of 96.15% was achieved under optimum conditions
(methanol/oil molar ratio of 6:1, 1 wt% KOH and 338 K) after reaction
time of 1 h with the aid of mechanical stirring. These optimum condi-
tions were then implemented again using ultrasonication (100 W,
20 kHz) in another study by Refaat and El-Sheltawy [116]. It was found
out that the separation time was reduced remarkably from 8 h to 25 min
in addition to shorter reaction time (reduced from 1 h to 5 min) and
higher yields (98–99%) were achieved. According to Kumar et al.
[117], ultrasonication reduced the separation time from 5 to 10 h to less
than 30 min in the production of biodiesel from Jatropha oil at me-
thanol: ethanol: oil molar ratio of 3:3:1 with 0.75 wt% catalyst. The
ease of separation of glycerol and catalyst was also found to be higher
in the case of ultrasonication compared to mechanical stirring for the
biodiesel synthesis from Nagchampa oil by Gole and Gogate [70], which
reduced the purification time and energy.
4.7. Ultrasonic pulse mode
Ultrasound can be applied in pulse mode or continuous mode.
S.X. Tan et al. Fuel 235 (2019) 886–907

Examples of wave patterns for continuous and pulse sonication (5 s on/
1 s off) are shown in Fig. 3(a) and (b), respectively. For continuous
sonication (Fig. 3(a)), ultrasound waves are delivered continuously and
relaxation intervals are impossible for the liquid phase in the reaction
medium, thus increasing thermal energy of the reaction medium. As a
result, energy input will be partially lost as they are being transformed
into vibrational or cavitation energy. Transesterification reaction may
be energized by the thermal effect of the continuous sonication but
prolong exposure is not recommended. This is because unfavourable
emulsification of the reactants may complicate the product separation
owing to over excitation of the reactants. On the contrary, relaxation
intervals are possible for pulse sonication (Fig. 3(b)), rendering ex-
citation gap to the reaction mixture to form good emulsions without
increasing thermal energy of the reaction medium significantly.
Nevertheless, longer sonication time is required by pulse sonication
than continuous sonication as lower excitation energy is imparted [31].
Chand et al. [105] had proven that pulse mode gave a greater yield of
esters than continuous mode. In their study, pulse mode (5 s on/25 s
off) and continuous mode (for 15 s) were compared. 96% biodiesel
yield was obtained in less than 90 s in the pulse mode while 86% yield
was achieved in 15 s for continuous mode. These results are in agree-
ment with the study carried out by Martinez-Guerra and Gude [31]. In
addition to lower the yield produced, application of ultrasound in
continuous mode was also found to cause tip erosion and was less en-
ergy efficient than pulsed mode [118].
Pulse is also known as the ratio of ultrasound working time to its
idling time [87]. It could also be expressed in duty cycle, which is the
percentage of ultrasound working time to the total ultrasound working
time and idling time. At lower pulse, lower biodiesel conversion is
obtained due to the macro-stirring effect of the ultrasound which is too
mild to mix the immiscible reactants evenly. By increasing the pulse to
a certain extent, the biodiesel conversion increases as better emulsifi-
cation of the two immiscible layers is achieved [98]. The effect of using
pulsed ultrasound was investigated by Hingu et al. [98] in the biodiesel
synthesis from WCO for 40 min reaction time. For the pulse 2 s on/2 s
off (duty cycle of 50%), the extent of conversion was 62% and 5 s on/1 s
off (duty cycle of ≈83%) increased the conversion to 65.5%. For pulse
duration of 1 min on/5 s off (≈92%), conversion of 89.5% was acquired
and the yield was further increased to 94.5% after dry washing the
esters. In the ultrasonic transesterification of WCO using 1 wt% CaDG as
a catalyst, methanol to oil molar ratio of 9:1 and 333 K, biodiesel yield
increased when the duty cycle increased from 30% (3 s on/7 s off) to
50% (7 s on/3 s off). However, no significant increment in the yield was
detected for further enhancement in the duty cycle to 70% (7 s on/3 s
off) [59]. Therefore, it could be concluded that if the pulse (duty cycle)
exceeded the optimum range, the biodiesel conversion or yield would
increase with a lower rate. This was because in such condition, the
effect of initial vibrational shock applied to the reactants by ultrasound
waves would become identical with uniform waves [87].
4.8. Biodiesel conversion or yield
In order to compare the biodiesel conversion or yield attained be-
tween conventional stirring and ultrasonication, several researchers
had conducted the experiments under similar reaction conditions in
which the results are tabulated in Table 9. From Table 9, it can be
observed that ultrasonication can increase the biodiesel yield or con-
version by 4.0%–27.9% in comparison to conventional stirring. Since
the transesterification reaction is mass controlled due to the im-
miscibility of methanol and oil, the micro-turbulence generated due to
cavitation bubbles can result in larger interfacial area and higher
temperature and pressure. For conventional stirring, agitation intensity
plays an important role in the transesterification reaction. The mass
transfer of TGs from oil phase towards the methanol–oil interface can
be the rate limiting step and hence poor mass transfer between the two
phases results in slow reaction rate [72]. In short, it can be concluded
that ultrasonication can achieve higher biodiesel conversion or yield
than conventional stirring under comparable reaction conditions.
5. Kinetic study on ultrasound-assisted transesterification
There are ample of studies which had been published on the ul-
trasound-assisted transesterification by utilizing non-edible feedstocks.
However, the fundamental info such as kinetic studies on ultrasound-
assisted transesterification were rarely investigated [119]. From the
kinetic study, information on reaction order, reaction rate and Ea can be
obtained which is vital for process scale-up.
For the determination of reaction order, different assumptions of the
reaction order are applied. Pseudo first-order reaction is assumed when
the amount of methanol in the reactants is large enough for the me-
thanol concentration to remain constant throughout the reaction
[113,120]. For transesterification of TG due to the presence of excess
methanol, the reaction order can be considered as an irreversible
pseudo second-order reaction [58,59]. Then the reaction rate constant
(k) can be found using Eq. (4) for first-order reaction and Eq. (5) for
second-order reaction [58,121].

Table 9
Comparison of biodiesel yield or conversion of different feedstocks using ultrasound and conventional stirring approaches.
Year Feedstock Type of ultrasound/stirring Ultrasonic cavitation Conventional stirring Increase percentage Ref.
(%)
Yield (%) Conversion (%) Yield (%) Conversion (%)

2010 WCO • Ultrasonic horn (200 W, 20 kHz) − 89.5 − 57.5 32 [98]

• Mechanical stirring (1000 rpm)
2010 Oreochromis niloticus • Ultrasonic bath (60 W, 40 kHz) 98.2 − ∼85 − 14.2 [50]
oil • Mechanical stirring
2011 Rubber seed oil • Ultrasonic bath (33 ± 3 Hz) 91 − 87 − 4 [112]
• Magnetic stirring (> 300 rpm)
2012 Jatropha oil • Ultrasonic horn (400 W, 40 kHz) − 98 − 79 19 [72]
• Mechanical stirring (600 rpm)
2013 WCO • Stirrer (6-blade turbine, 1000 rpm) 90 − 70 − 20 [25]
2014 WCO • Ultrasonic horn (22 kHz, 375 W) 92 − 59 − 33 [58]
• Overhead stirrer (1000 rpm)
2015 WCO • Ultrasonic horn (120 W, 22 kHz) 93.5 − 65.6 − 27.9 [59]
• Overhead stirrer (800 rpm)
2015 Mahua oil • Ultrasonic horn (1000 W, 20–36 kHz) 97.4 − 93 − 4.4 [71]
• Magnetic stirring (500 rpm)
2016 Schleichera triguga oil • Ultrasonic probe (250 W, 20 kHz) − 96.8 − 63.29 33.51 [77]
• Conventional stirring (did not specify the
condition in the paper)

899
S.X. Tan et al. Fuel 235 (2019) 886–907

k = −ln(1−X )/ t
kCTG0 = [X /(1−X )]/ t
Several researchers had performed kinetic studies on ultrasound-
assisted transesterification in which the rate constant (k), Ea and fre-
quency factor (A) for the transesterification at optimum operating
conditions are tabulated in Table 10. The lower the value of Ea and the
higher the value of kinetic constant will indicate a high reaction rate
[149]. It can be noticed from Table 7 that Ea is the highest during es-
terification of Jatropha oil catalyzed by H2SO4 which is 167.419 kJ/
mol. Such high Ea would result in lower reaction kinetics [52]. High
temperature was required for high Ea in order to increase the solubility
of methanol in the ester phase which in turn increased the reaction rate
(reaction kinetics). However, cavitation effects of ultrasound at higher
temperature would also dampen due to cushioned collapsing of cavities
[94]. As observed from Table 10, the optimum temperature required by
the esterification of Jatropha oil catalyzed by H2SO4 was the highest
(343 K). Although bubble formation was easier at higher temperature,
more vapours were found in the cavitational bubbles. Interference of
the solvent vapours with the cavitation effect will affect the mixing
effect of the ultrasound on the esterification reaction, resulting in a
decrement in solubility of methanol in ester phase and lower reaction
[31,38]. Hence, chlorosulfonic acid was employed instead of H2SO4 in
another study carried out by Choudhury et al. [33]. The Ea for the
process (57 kJ/mol) was at least 3 times lower than the energy for
H2SO4 catalyzed transesterification. The huge difference of the Ea was
likely due to the difference in the active sites of transesterification re-
action with H2SO4 and chlorosulfonic acid catalyst.
It can also be observed from Table 10 that there are three studies
which synthesized WCO using different types of alcohol (methanol and
methyl acetate) and catalysts (potassium methoxide, K3PO4 and CaDG).
By rearranging the Ea required by each catalyst in ascending order,
potassium methoxide required the lowest Ea (56.97 kJ/mol), followed
by K3PO4 (64.24 kJ/mol) and CaDG (119.23 kJ/mol). The optimum
temperatures to achieve optimum biodiesel yield for potassium meth-
oxide, K3PO4 and CaDG were 313 K, 323 K and 333 K, respectively,
which affirmed that higher Ea required higher temperature. Lowest Ea
for potassium methoxide could be attributed to its two-phase homo-
geneity as compared to the triphasic system (liquid/liquid/solid) cata-
lyzed by K3PO4 and CaDG. Triphasic system often encountered higher
mass transfer limitations which resulted in higher Ea. It was noticed that
the Ea required by CaDG was almost double compared to the Ea re-
quired by both potassium methoxide and K3PO4. The possible reason
could be due to the high impurity of the WCO which hindered the ac-
tivity of the catalyst [54].
The kinetic behaviours of the ultrasonic single step (chlorosulfonic
acid) and conventional double-step processes (H2SO4 and chlor-
osulfonic acid) were compared by Choudhury et al. [33]. The Ea for
ultrasonic single step process was 57.33 kJ/mol which showed a re-
duction of approximately 40% to 31.29 kJ/mol for the conventional
(4) double-step process. Higher Ea of the ultrasonic single step process
could be attributed to catalyst inhibition due to water traces in the
(5) reaction mixture. On the other hand, absence of water molecules in the
reaction mixture in conventional double-step process contributed to
reduction in Ea. In the transesterification of WCO using CaO (conven-
tional) and CaDG (ultrasound), CaO (conventional) only required
around 2/3 of the Ea required by CaDG (ultrasound). The Ea for CaO
(conventional) and CaDG (ultrasound) were 79 and 119.23 kJ/mol,
respectively. It was estimated that CaDG (ultrasound) should require
lower Ea than CaO (conventional) due to the presence of ultrasound in
enhancing the mass transfer limitation. In addition, catalytic activity of
CaDG was higher than CaO due to the presence of basic oxygen anion,
which could easily attract protons from OH group of methanol to form
methoxide ions [59]. Therefore, the only possible reason to explain this
phenomena could be due to the high amount of impurities present in
the WCO which poisoned the catalyst through absorption [54]. In the
esterification of Karanja oil conducted by Mohod et al. [85], the kinetic
rate constants at optimum catalyst loading of 2 wt% H2SO4 were
0.275 × 10−3 and 0.262 × 10−3 L·mol−1·min−1 using ultrasonication
and mechanical stirring, respectively. Similarly, in the transesterifica-
tion of Silybum marianum oil, Takase et al. [107] discovered that the
reaction rate constants via ultrasonication with methanol and ethanol
were higher than via mechanical stirring. Higher kinetic rate constant
attained by ultrasonication could be ascribed to better efficacy of ul-
trasonication than mechanical stirring to generate microemulsion from
the intense mixing of methanol and oil by cavitation effects.
6. Pilot reactors for ultrasound-assisted biodiesel production
The utilization of ultrasound in biodiesel production has presented a
huge possibility for implementation in pilot or industrial scales.
However, only a few examples of the relevant case studies were re-
ported, which were tabulated in Table 11. From Table 11, it was no-
ticed that the biodiesel yields were quite high which were mostly above
90%. However, lower biodiesel yield was observed when the reactor
volume was higher than 6 L. According to the ultrasound-assisted bio-
diesel synthesis from canola oil conducted by Stavarache et al. [122],
the biodiesel yield decreased from 95% to 50% when the reactor vo-
lume increased from 2.62 L to 6.35 L. The reduction in the biodiesel
yield was attributed to a decrement in ultrasonic power and bubble
density, which reduced the contact area between the reactants and
induced mass transfer limitation. Another important observation was
that when the residence time increased from 20 to 30 min while
maintaining the reactor volume (6.35 L), the biodiesel yield decreased
from approximately 90% to 50%. It was suspected that accumulation of
glycerol at the bottom of the reactor would trap more methanol mo-
lecules into the glycerol layer especially when longer residence time
was employed. This could occur since both the glycerol and methanol
were polar compounds; hence they could dissolve with each other at
any ratio. Therefore, there was insufficient methanol to promote a

Table 10
Rate constant, activation energy and frequency factor for the transesterification at optimum operating conditions.
Year Feedstock Alcohol Catalyst Reaction Rate constants, k Activation energy, Ea Frequency factor, A Ref.
order (kJ/mol)

2014 Silybum marianum Methanol KOH First 2.3 × 10−2 s−1 (333 K) − − [107]
oil
Ethanol KOH First 7.0 × 10−3 s−1 (353 K) − −
2014 Jatropha oil Methanol Chlorosulfonic acid First 1.33 × 10−2 min−1 (333 K) 57.33 1.57 × 107 min−1 [33]
2014 Jatropha oil Methanol CaO Third 1.46 × 10−3 L2∙mol−2∙s−1 133.5 4.32 × 1017 L2∙mol−2∙s−1 [54]
(338 K)
2015 WCO Methanol K3PO4 Second 0.14 L∙mol−1∙min−1 (323 K) 64.24 2.95 × 109 L∙mol−1∙min−1 [58]
2015 WCO Methanol CaDG Second 0.198 L∙mol−1∙min−1 (333 K) 119.23 9.07 × 1017 L∙mol−1∙min−1 [59]
2016 Schleichera triguga Methanol H2SO4 First 0.236 min−1 (313 K) 21.75 1075.99 min−1 [77]
oil
Methanol Ba(OH)2 Second 0.18 L∙mol−1∙min−1 (323 K) 53.26 6.1 × 107 L∙mol−1∙min−1

900
 S.X. Tan et al.

Table 11
Existing pilot reactors for ultrasonic biodiesel production.
Year Process Feedstock Reactor Flowrate Frequency (kHz); Residence time Methanol-to-oil Temperature (K) Catalyst; Yield (%); time (min) Ref.
volume (L) (mL/min) Power (W) (min) molar ratio (mol/ loading (wt%)
mol)

2007 Continuous process Commercial 2.62 − 45; 600 10 6:1 311–313 KOH; − > 90; − [122]
edible oil
20 > 90; −
30 > 90; −
2.62 − 45; 600 10 7.5:1 311–313 KOH > 95; −
20 > 95; −
30 > 95; −
6.35 − 45; 600 20 7.5:1 311–313 KOH less than90; −
30 50; 360
2010 Circulation process Canola 0.8 8000 20; 1000 40 5:1 Room KOH; 0.7 99; 50 [5]
temperature

901
2010 Two-stage continuous process WCO 0.8 1st stage: 1.5 20; 1000 1st stage: 0.53 1st stage: 2.5:1 293–298 KOH; 1st 1st stage: 81; 25 [123]
2nd stage:2.0 2nd stage: 0.40 2nd stage: 1.5:1 stage: 0.7 2nd stage: 99; 20
2nd stage: 0.3
2010 Two-step process (conventional heating under Soybean oil 5 55 21.5; 600 9 6:1 318 KOH; 0.15 − [131]
mechanical stirring for 30 min at 318 K followed
by ultrasonication for 35 min)
2015 Circulation process Soybean oil 5 2000 19; 600 60 6:1 333 MeONa; 0.6 97; 15 [114]
2016 Three-step process (esterification and WCO 6 16 30; − − 8:1 − KOH; − 78; 40 [132]
neutralization in batch mode followed by
transesterification in continuous mode)
2017 Batch mode (longitudinal horn equipped with Karanja oil 4 − 36; 150 − 5:1 Room H2SO4; 2.0 80.9 (in term of acid [85]
two mechanical stirrers) temperature value reduction); 70
Fuel 235 (2019) 886–907
S.X. Tan et al.
constant conversion [122,123]. According to Thanh et al. [123], lower
flow rate which corresponded to longer residence time would enhance
the emulsion efficiency of the reactants, resulting to an increment in
biodiesel yield. Nonetheless, one important point to note is that bio-
diesel yield or conversion is affected by the combined effect of flow rate
or residence time and ultrasonic amplitude in the ultrasonic transes-
terification processes [124]. Employing low ultrasonic amplitude at
higher flow rate decreases the ultrasonic power dissipated into the re-
actor, thus lowers the biodiesel yield due to low cavitation and acoustic
jets. On the other hand, employing high ultrasonic amplitude at lower
flow rate increases the ultrasonic power which then promotes the
backward reaction and therefore decreases biodiesel yield. More studies
should be carried out to investigate the effect of residence time or flow
rate on the biodiesel yield by taking ultrasonic amplitude into con-
sideration.
The reason of lacking industrial sonochemical processes might be
due to the complicated design of efficient cavitation reactors, given the
lack of data linking bubble dynamics with experimental predictions for
large-scale processes. Biodiesel from Cedrus deodara oil had been syn-
thesized by Mohan et al. [125] using four different ultrasonic reactors
which were triple-frequency flow cell, double-frequency flow cell, bath-
type and horn-type. Energy consumptions by using these four types of
ultrasounds were compared and they found that triple-frequency ul-
trasonic reactor was the most energy efficient. The order of energy
consumption was triple-frequency flow cell < double-frequency flow
cell < bath-type < horn-type. Their results were in well agreement
with the results reported by Gogate et al. [32]. Reaction times to
achieve optimum biodiesel yields for triple-frequency, double-fre-
quency, bath-type and horn-type ultrasonic reactors were 99% in
15 min, 99% in 30 min, 99% in 45 min and 98.5% in 50 min, respec-
tively, at 60% of the maximum rated power conditions. From their
studies, it was revealed that triple-frequency ultrasonic reactor was able
to attain the optimum biodiesel yield under the shortest reaction time
with the lowest energy consumption. Recently, effects of different ul-
trasound modes (single-frequency, dual-frequency of sequential mode
(SQM) and simultaneous mode (SMM)) on biodiesel conversion were
studied by Yin et al. [126] with the reactants in a counter-current state.
It was revealed that biodiesel conversion enhanced by single-frequency
was lower than that enhanced by dual-frequency. This was because
single-frequency reactor suffered from non-uniform volumetric energy
dissipation in the reaction medium (due to directional sensitivity of the
ultrasound field) that limited the yield of the chemical transformation.
On the other hand, multi-frequency ultrasound had better efficiency
than the single frequency in terms of higher pressure and temperature
produced from the implosive collapse of cavitation bubbles which led to
higher biodiesel yield. Moreover, multi-frequency ultrasound generated
strong interference pattern between two ultrasound waves. This inter-
ference reduced the directional sensitivity of single-frequency ultra-
sound and aided in achieving more uniform spatial energy dissipation
in the reaction medium, which resulted in overall volumetric efficiency.
For dual-frequency, biodiesel conversion of SMM was higher than that
of SQM. It was ascribed to enhanced mechanical disturbance of the
reaction medium by simultaneous dual-frequency irradiation which
generated more cavitation bubbles than sequential dual-frequency ir-
radiation. Moreover, types of reactors used during ultrasound-assisted
production of biodiesel will also affect the biodiesel yield. Murillo et al.
[127] had synthesized biodiesel from WVO with the assistance of ul-
trasound and Burkholderia cepacia lipase as catalyst. They evaluated the
kinematic viscosity and the yield of biodiesel produced with the utili-
zation of two different reactors which were packed bed reactor (PBR)
and batch reactor. Based on their findings, PBR achieved better bio-
diesel yield (67.30% versus 32.70%) and lower kinematic viscosity
(7.88 mm2∙s−1 versus 25.44 mm2∙s−1 at 293 K) than batch reactor. For
the case of batch reactor, discrete quantities of volume were used for
the synthesis of biodiesel. Lack of sufficient flow current resulted in
limited contact of the interfacial surfaces between the particles of WVO
902
Fuel 235 (2019) 886–907
and methanol with the immobilized enzymes. For the case of PBR,
continuous flow rate allowed the reaction mixture to get as much
contact as possible with the immobilized enzymes and to move with the
generated water from esterification reactions while avoiding the accu-
mulation of the traces of glycerol inside the bed. Therefore, biodiesel
yield achieved by PBR was twice of that by batch reactor. Since the
extent of biodiesel conversion was higher for the case of PBR than batch
reactor, kinematic viscosity also followed the same trend.
Furthermore, the reactor characteristics have to be tailored in terms
of the rheological properties of the reaction mixture (solution or sus-
pension, viscosity, surface tension, etc.). This is because the behaviour
of micro bubbles will be affected by the rheological properties of re-
action mixture [128]. For example, the rheological properties of reac-
tion mixture will change when the concentration of alcohol decreases
while the concentration of glycerol and biodiesel increase at the same
time. Such changes can easily affect the bubbles dynamics, which
eventually influence the effectiveness of ultrasonic irradiation in the
reaction mixture. When the concentration of alcohol decreases, the
viscosity of the reaction media will increase which will impede the
growth of the micro bubbles. Its radius size will reduce along with a
drop in the bubble temperature and pressure as well as a decrement in
oscillatory velocity. This will subsequently result in reduced effect on
enhancing the quality of the micro emulsion as the reaction proceeds
[128]. Apart from rheological properties of reaction mixture, operating
conditions of the reaction mixture will also influence the behaviour of
micro bubbles. According to the study performed by Sajjadi et al. [128],
intensity of ultrasound irradiation played the most important role
compared to reaction temperature and alcohol concentration. Radius,
internal temperature and pressure of bubbles significantly increased
with power amplitude due to stronger expansion and greater energy
accumulation. At the same time, increase in reaction temperature and
alcohol concentration made the bubbles characteristics rather mod-
erate. It was also found that the growth of bubbles radius was reduced
by about 4 times as the reaction progressed and the reaction mixture
became more viscous and dense. In order to validate their findings,
more studies should be conducted as the available data is limited. In
addition, the effect of other parameters (catalyst loading, reactor vo-
lume, reactor design and flow rate) on micro bubbles’ behaviour in the
ultrasound-assisted biodiesel production at pilot or industrial scale
should be examined by employing different catalysts and other non-
edible feedstocks.
Ultrasound can also be synergised in tandem with supercritical
method for continuous biodiesel production in pilot reactor. As men-
tioned in Section 1, high operating temperature, pressure and me-
thanol-to-triglycerides molar ratio were required by supercritical
method. All these factors will impair the desired profitability of in-
dustrial production and render supercritical method becomes less
competitive with conventional biodiesel production methods [129]. In
the pilot scale of continuous rapeseed biodiesel production assisted by
supercritical methanol alone, poor mixing of alcohol and oil led to low
reaction rate and increased reaction time, which contributed to higher
energy consumption for the process. With the synergistic effect of ul-
trasound, reaction rate and biodiesel conversion were increased
through increasing the solubility of oil in alcohol and the area of con-
tact surface of the immiscible phases (vigorous mixing) [130]. In the
continuous production of biodiesel from rapeseed oil by ultrasound-
assisted transesterification in supercritical ethanol conducted by Ma-
zanov et al. [129], the effects of ethanol to rapeseed oil molar ratio
(12:1–20:1), temperature (623 K–653 K), catalyst loading and types
(ZnO/Al2O3, MgO/Al2O3 and SrO/Al2O3) on fatty acid ethyl ester
(FAAE) yield were investigated. Biodiesel yield of 97.46% was achieved
under optimum conditions of 12:1 ethanol to oil molar ratio, SrO/Al2O3
(2 wt% SrO) at 623 K. However, there were only few publications re-
porting on the continuous production of biodiesel from edible feedstock
by ultrasound-assisted transesterification combined with supercritical
method. There was also lack of study pertaining to the case of using
S.X. Tan et al.
non-edible feedstock. Therefore, more studies should be carried out to
investigate the feasibility of continuous biodiesel production assisted by
ultrasound and supercritical method.
7. Ultrasonic reactive extraction for biodiesel production
Reactive extraction (RE) is also known as in situ transesterification
in which the oil bearing materials such as oil seeds or seed cakes contact
directly with an alcohol solution at ambient temperature and pressure.
In this method, alcohol acts as an extracting solvent and an esterifica-
tion reagent, so both oil extraction and transesterification of oil into
biodiesel proceed in one step [30]. In other words, once the oil is ex-
tracted out from seeds, it is subsequently converted to esters [133].
Expensive costs employed for the oil extraction and degumming in the
biodiesel production can be reduced as the isolation and refining of oil
seeds are not needed. In addition, the biodiesel yield could be max-
imized up to 98% [27,30]. It was reported that RE can reduce biodiesel
production cost by integrating multiple biodiesel processing stages
which constitute over 70% of the overall cost, even when using refined
oil as feedstock [134]. In order to reduce the alcohol requirement for
high efficiency of RE, the oil seeds need to be dried prior to the reaction
taking place [135].
Integration of ultrasonication in the RE process can provide suitable
pre-treatment condition to change the physical and chemical structure
of the lignocellulosic seed. Ultrasound used in the process could induce
physical pre-treatment through the formation of cavitation bubbles in
the liquid phase that grow and violently collapse. Furthermore, addi-
tion of ultrasonication could induce mechanical treatment for dis-
rupting biological structure and increase the surface area while en-
hancing the saccharification of cellulose for maximum cavitation effect
at 323 K. Moreover, the thermal temperature employed in the process
could deactivate the toxic phorbol ester while producing crude bio-
diesel, glycerol by-product and completely de-oiled seed residues
[136].
However, there are only few studies conducted on ultrasound-as-
sisted RE directly from non-edible oil seeds. Chadha et al. [137]
achieved 94.1% of biodiesel yield with the assistance of ultrasonic
probe (22 kHz; 750 W) under the optimum conditions of methanol to
seed ratio (w/w) of 10:1, 2–3 mm seed size, 1 wt% NaOH and 80 min
reaction time. Zahari et al. [138] obtained 62.21% biodiesel yield with
0.15 N NaOH and < 1 mm seed size. The details of methanol to seed
ratio, reaction temperature and reaction time in achieving biodiesel
yield of 62.21% were not mentioned in their studies. Koutsouki et al.
[109] synthesized biodiesel from Cynara cardunculus L. seeds with the
assistance of ultrasonic horn (24 kHz, 400 W, 80% of ultrasonic am-
plitude and cycle at 0.7). Optimum biodiesel yield of 96% was achieved
at methanol to oil molar ratio of 550:1, 9.5 wt% NaOH and reaction
Fig. 4. Brief overview of ultrasonic extraction mechanism. Adapted with permission from (Jadhav AJ, Holkar CR, Goswami AD, Pandit AB, Pinjari DV) [140].
Acoustic cavitation as a novel approach for extraction of oil from waste date seeds. ACS Sustainable Chemistry & Engineering. 2016;4:4256–63. Copyright (2016)
American Chemical Society.
903
Fuel 235 (2019) 886–907
time of 20 min. Kumar et al. [139] attained the maximum biodiesel
conversion of 92% within 20 min by using methanol to seed ratio (w/w)
of 100:1, > 1− < 2 mm seed size, 1.5 wt% KOH, 50% of ultrasonic
amplitude and 0.3 s cycle.
7.1. Mechanism of ultrasonic reactive extraction on oil seed for biodiesel
production
The suggested ultrasonic extraction mechanism by Jadhav et al.
[140] is shown in Fig. 4. In the first step of the extraction processes, the
irregular dry seed particles are surrounded by a solvent layer. Uptake of
the solvent by the irregular dry seed particles causes them to begin
swelling, expanding and acquiring a smoother shape. A dynamic in-
teraction is present at the surface of the seeds in the course of swelling
and decreases successively upon reaching the maximum limit of swel-
ling. Direct diffusion is then being interfered by the formation of the
stagnant layer surrounding the seed particle which possibly causes the
extraction process to cease. In the second step, solvated compound
(seed oil) moves towards the stagnant layer and diffuses into the bulk
solvent. The extraction efficiency will be hampered due to the presence
of stagnant layer as a diffusion barrier. Diffusion process will become
harder if the stagnant layer has high adhesiveness and the extraction
yield will be lower. Moreover, blockage of crevices by the stagnant
layer could hamper the diffusion of seed oil [140].
By employing ultrasonication, larger amplitude ultrasonic wave
passes through the liquid media (solvent). This creates and collapses the
cavities (bubbles) in a very short time, which contain very high tem-
perature and pressure. The asymmetric collapse of these cavities results
in the generation of violent shock waves and high-speed jets. These
shock waves then strike the surface of the seed, which break the stag-
nant layer as well as create cracks, crevices, and microfractures. At the
same time, impingement by high-speed jets results in surface peeling,
erosion and particle breakdown. Due to these physical effects, the sol-
vent can easily penetrate the plant’s cell wall, diffuse through the seed,
reach the cell tissues, accelerate the extraction of intracellular products
(i.e. oil bodies) with solvent and permit intracellular products release
into the extraction environment (solvent) [140,141]. Once the oil
bodies have been extracted, they will be converted to biodiesel si-
multaneously [134].
The proposed mechanism was confirmed by Jadhav et al. [140]
using the field emission scanning electron microscopy (FE-SEM) ana-
lysis of raw and ultrasonically treated date seed. Raw date seed had a
regular, intact, smooth surface (as shown in Fig. 5(a) and (c)) whereas
ultrasonically treated date seed had irregular shape with formation
microfractures and crevices on its surface (as shown in Fig. 5(b) and
(d)). The breakdown of seed due to the physical effect of cavitation had
also been confirmed from the particle size distribution of the fresh and
S.X. Tan et al. Fuel 235 (2019) 886–907

Fig. 5. FE-SEM images of raw date seed at (a) 2000 X and (c) 10,000 X and ultrasonically treated date seed at (b) 2000 X and (d) 10,000 X. Adapted with permission
from (Jadhav AJ, Holkar CR, Goswami AD, Pandit AB, Pinjari DV) [140]. Acoustic cavitation as a novel approach for extraction of oil from waste date seeds. ACS
Sustainable Chemistry & Engineering. 2016;4:4256–63. Copyright (2016) American Chemical Society.

ultrasonically treated date seed powder (as shown in Fig. 6). Ultra-
sonically treated seed had narrower particle size distribution (in the
range of 500 nm) as compared to raw seed (in the range of 1600 nm)
due to the shrinkage of seed size.
8. Challenges and feasibility for future development
Based on the review discussed in this work, there are several im-
minent developments which can be conducted in the future. Since there
is still a knowledge gap on the fundamental studies dealing with reac-
tion pathway of TGs on solid acid catalysts, more in-depth studies
should be performed. Besides, more parameter studies can be con-
ducted on the two-step heterogeneous ultrasound-assisted transester-
ification by utilizing non-edible feedstocks as the literature information
is still relatively scarce. For ultrasound-assisted interesterification,
parameter studies should be also conducted with different types of
catalyst (chemical and enzyme catalysts) and different non-edible
feedstocks. This can gain a better insight on the behaviour of the pro-
cess under different optimum conditions.
Ultrasonic RE is a promising approach to synthesize biodiesel from
seed directly. However, majority of the studies had performed ultra-
sonic RE from liquid oil directly instead of the solid oil-bearing seeds.
The optimum parameters obtained such as alcohol to oil ratio, catalyst
loading, ultrasonic power and ultrasonication mode by employing ul-
trasonic RE from oil might not be suitable to achieve maximum bio-
diesel yield for ultrasonic RE from oil seeds directly. Till now, the sci-
entific data on the ultrasonic RE from seed directly is very limited.
Future research could be directed more towards the effect of parameters
on biodiesel yield or conversion, especially the effect of alcohol to seed
ratio instead of alcohol to oil ratio.
The effect of various parameters such as catalyst loading, reactor
volume, reactor design and flow rate on the micro bubble’s behaviour in
ultrasound-assisted biodiesel production at pilot or industrial scale also
requires further investigation. The reason of lacking industrial

904
S.X. Tan et al.
Fig. 6. Particle size distribution of fresh date seed powder and ultrasonically
treated date seed powder. Adapted with permission from (Jadhav AJ, Holkar
CR, Goswami AD, Pandit AB, Pinjari DV) [140]. Acoustic cavitation as a novel
approach for extraction of oil from waste date seeds. ACS Sustainable Chemistry
& Engineering. 2016;4:4256–63. Copyright (2016) American Chemical Society.
sonochemical processes might be due to the complicated design of ef-
ficient cavitation reactors, given the lack of sufficient info linking the
bubble dynamics with experimental predictions for large-scale pro-
cesses. Furthermore, the reactor characteristics have to be tailored in
terms of the rheological properties of the reaction mixture (solution or
suspension, viscosity, surface tension, etc.).
In addition, it has been observed that majority of the studies have
employed simple designs of sonochemical reactors such as ultrasonic
horn and ultrasonic bath which may not be effective at commercial
scale operation [141]. Besides, many experiments were performed in a
batch reactor which could not provide a closer basis for the role of
ultrasound-assisted reactor in an industrial scale [142]. Therefore, more
comprehensive theoretical work on the pilot scale with continuous
processing should be carried out before commercialization of ultra-
sound-assisted biodiesel production plant can be realized. For the
combination of ultrasound and supercritical methods for continuous
production of biodiesel in pilot scale, more studies should be performed
for both edible and non-edible feedstocks with varying compositions.
Further researches may be directed in terms of the development of
reactor design based on multi-frequency operation [141]. This finding
would be beneficial in the design of an industrial scale of ultrasound-
assisted biodiesel production plant. However, no substantial work is
found on energy consumption analysis and comparison among various
types of ultrasonic reactors for biodiesel production [125]. Hence, more
studies should be performed to support and verify their claims. Another
important point is the economic analysis which also needs to be per-
formed based on the pilot scale demonstration plant’s performance data
in order to eliminate the need for unavoidable assumptions in the cost
estimations for the process units [38].
9. Conclusion
Biodiesel is a strong candidate to replace fossil fuel as it is en-
vironmental friendly, economical competitive, technically feasible and
the sources for biodiesel synthesis are easily available. Ultrasonication
is a promising method to replace mechanical or magnetic stirring in the
biodiesel production process. It is not only capable to eliminate the
mass transfer limitation encountered in the transesterification reaction,
but also eliminates the needs of separate heating and agitation due to
the localized temperature increment and the formation of micro jets.
905
Fuel 235 (2019) 886–907
This review has demonstrated that remarkable intensification in terms
of reduction in alcohol to oil ratio (16.67%−25% reduction), catalyst
amount (20%−25% reduction), reaction time (25%−95.8% reduc-
tion), reaction temperature, phase separation time and energy con-
sumption (33%−50% reduction) as well as improved biodiesel con-
version and yield (4%−27.9% increment) can be obtained with the
application of ultrasound in transesterification. Overall, ultrasonication
is advantageous to be used for process intensification in biodiesel pro-
duction and possesses huge potential to prepare biodiesel for industrial
production if the pilot reactor of ultrasound-assisted biodiesel produc-
tion is found to be economically feasible and the biodiesel produced is
of comparable quality and yield to fossil fuels.
Acknowledgements
The authors wish to acknowledge the University of Malaya, Kuala
Lumpur, Malaysia, for funding support of this work under the FRGS-
MRSA (MO014-2016) and University Tunku Abdul Rahman Research
Fund (UTARRF).
References
[1] Dharma S, Ong HC, Masjuki H, Sebayang A, Silitonga A. An overview of engine
durability and compatibility using biodiesel–bioethanol–diesel blends in com-
pression-ignition engines. Energy Convers Manage 2016;128:66–81.
[2] Koh MY, Ghazi TIM. A review of biodiesel production from Jatropha curcas L. oil.
Renewable Sustainable Energy Rev 2011;15(5):2240–51.
[3] Su C-H. Recoverable and reusable hydrochloric acid used as a homogeneous cat-
alyst for biodiesel production. Appl Energy 2013;104:503–9.
[4] Parida S, Sahu DK, Misra PK. A rapid ultrasound-assisted production of biodiesel
from a mixture of Karanj and soybean oil. Energy Sources Part A
2016;38(8):1110–6.
[5] Thanh LT, Okitsu K, Sadanaga Y, Takenaka N, Maeda Y, Bandow H. Ultrasound-
assisted production of biodiesel fuel from vegetable oils in a small scale circulation
process. Bioresour Technol 2010;101(2):639–45.
[6] Lee KT, Lim S, Pang YL, Ong HC, Chong WT. Integration of reactive extraction with
supercritical fluids for process intensification of biodiesel production: prospects
and recent advances. Prog Energy Combust Sci 2014;45:54–78.
[7] Talebian-Kiakalaieh A, Amin NAS, Mazaheri H. A review on novel processes of
biodiesel production from waste cooking oil. Appl Energy 2013;104:683–710.
[8] Manickam S, Arigela VND, Gogate PR. Intensification of synthesis of biodiesel from
palm oil using multiple frequency ultrasonic flow cell. Fuel Process Technol
2014;128:388–93.
[9] Singh B, Bux F, Sharma Y. Comparison of homogeneous and heterogeneous cata-
lysis for synthesis of biodiesel from Madhuca indica oil. Chem Ind Chem Eng Q/
CICEQ 2011;17(2):117–24.
[10] Chuah LF, Bokhari A, Yusup S, Klemeš JJ, Akbar MM, Saminathan S. Optimisation
on pretreatment of kapok seed (Ceiba pentandra) oil via esterification reaction in an
ultrasonic cavitation reactor. Biomass Convers Biorefin 2016:1–9.
[11] Aransiola E, Ojumu T, Oyekola O, Madzimbamuto T, Ikhu-Omoregbe D. A review
of current technology for biodiesel production: state of the art. Biomass Bioenergy
2014;61:276–97.
[12] Monisha J, Harish A, Sushma R, Krishna Murthy T, Blessy BM, Ananda S. Biodiesel:
a review.
[13] Meira M, Quintella CM, Ribeiro EMO, Silva HRG, Guimarães AK. Overview of the
challenges in the production of biodiesel. Biomass Convers Biorefin
2014;5(3):321–9.
[14] Badday AS, Abdullah AZ, Lee KT, Khayoon MS. Intensification of biodiesel pro-
duction via ultrasonic-assisted process: a critical review on fundamentals and re-
cent development. Renewable Sustainable Energy Rev 2012;16(7):4574–87.
[15] Abbasi S, Diwekar UM. Characterization and stochastic modeling of uncertainties
in the biodiesel production. Clean Technol Environ Policy 2014;16(1):79–94.
[16] Singh B. Production of Biodiesel from Plant Oils-An Overview. J Biotechnol Bioinf
Bioeng 2014;1(2):33–42.
[17] Vyas AP, Verma JL, Subrahmanyam N. A review on FAME production processes.
Fuel 2010;89(1):1–9.
[18] Saleh J, Tremblay AY, Dubé MA. Glycerol removal from biodiesel using membrane
separation technology. Fuel 2010;89(9):2260–6.
[19] Oh PP, Lau HLN, Chen J, Chong MF, Choo YM. A review on conventional tech-
nologies and emerging process intensification (PI) methods for biodiesel produc-
tion. Renewable Sustainable Energy Rev 2012;16(7):5131–45.
[20] Mohapatra S, Das P, Swain D, Satapathy S, Sahu SR. A review on rejuvenated
techniques in biodiesel production from vegetable oils. Int J Curr Eng Technol
2016;6:100–11.
[21] Kiss AA. Novel process for biodiesel by reactive absorption. Sep Purif Technol
2009;69(3):280–7.
[22] Maddikeri GL, Pandit AB, Gogate PR. Intensification approaches for biodiesel
synthesis from waste cooking oil: a review. Ind Eng Chem Res
2012;51(45):14610–28.
S.X. Tan et al.
[23] D’Alessandro B, Bidini G, Zampilli M, Laranci P, Bartocci P, Fantozzi F. Straight
and waste vegetable oil in engines: Review and experimental measurement of
emissions, fuel consumption and injector fouling on a turbocharged commercial
engine. Fuel 2016;182:198–209.
[24] Helwani Z, Othman M, Aziz N, Fernando W, Kim J. Technologies for production of
biodiesel focusing on green catalytic techniques: a review. Fuel Process Technol
2009;90(12):1502–14.
[25] Maddikeri GL, Pandit AB, Gogate PR. Ultrasound assisted interesterification of
waste cooking oil and methyl acetate for biodiesel and triacetin production. Fuel
Process Technol 2013;116:241–9.
[26] Islam A, Taufiq-Yap YH, Chan E-S, Moniruzzaman M, Islam S, Nabi MN. Advances
in solid-catalytic and non-catalytic technologies for biodiesel production. Energy
Convers Manage 2014;88:1200–18.
[27] Baskar G, Aiswarya R. Trends in catalytic production of biodiesel from various
feedstocks. Renewable Sustainable Energy Rev 2016;57:496–504.
[28] Kiss AA, Bildea CS. A review of biodiesel production by integrated reactive se-
paration technologies. J Chem Technol Biotechnol 2012;87(7):861–79.
[29] El-Ibiari N, El-Enin SA, Gadalla A, El-Ardi O. El-Diwani G. Biodiesel production
from castor seeds by reactive extraction conventionally and via ultra-sound using
response surface methodology. IJISET-International Journal of Innovative Science.
Eng Technol 2014;1(6)..
[30] Veljković VB, Avramović JM, Stamenković OS. Biodiesel production by ultra-
sound-assisted transesterification: state of the art and the perspectives. Renewable
Sustainable Energy Rev 2012;16(2):1193–209.
[31] Martinez-Guerra E, Gude VG. Continuous and pulse sonication effects on trans-
esterification of used vegetable oil. Energy Convers Manage 2015;96:268–76.
[32] Gogate PR, Tayal RK, Pandit AB. Cavitation: a technology on the horizon. Curr Sci
2006;91(1):35–46.
[33] Choudhury HA, Srivastava P, Moholkar VS. Single-step ultrasonic synthesis of
biodiesel from crude Jatropha curcas oil. AlChE J 2014;60(5):1572–81.
[34] He B, Van Gerpen JH. Application of ultrasonication in transesterification pro-
cesses for biodiesel production. Biofuels 2012;3(4):479–88.
[35] Hajinezhad A, Abedi S, Ghobadian B, Noorollahi Y. Biodiesel production from
Norouzak (Salvia lerifolia) seeds as an indigenous source of bio fuel in Iran using
ultrasound. Energy Convers Manage 2015;99:132–40.
[36] Mahamuni NN, Adewuyi YG. Optimization of the synthesis of biodiesel via ul-
trasound-enhanced base-catalyzed transesterification of soybean oil using a mul-
tifrequency ultrasonic reactor. Energy Fuels 2009;23(5):2757–66.
[37] Kumar D, Kumar G, Singh C. Ultrasonic-assisted transesterification of Jatropha
curcus oil using solid catalyst, Na/SiO 2. Ultrason Sonochem 2010;17(5):839–44.
[38] Gude VG, Grant GE. Biodiesel from waste cooking oils via direct sonication. Appl
Energy 2013;109:135–44.
[39] Hong IK, Jeon H, Lee SB. Effect of mixed alcohol reactants on ultrasonic alcoho-
lysis of canola oil. J Ind Eng Chem 2014;20(3):911–5.
[40] Ho WWS, Ng HK, Gan S. Advances in ultrasound-assisted transesterification for
biodiesel production. Appl Therm Eng 2016;100:553–63.
[41] Ejikeme P, Anyaogu I, Ejikeme C, Nwafor N, Egbuonu C, Ukogu K, et al. Catalysis
in biodiesel production by transesterification processes-an insight. J Chem
2010;7(4):1120–32.
[42] Hara M. Environmentally benign production of biodiesel using heterogeneous
catalysts. ChemSusChem 2009;2(2):129–35.
[43] Encinar J, Gonzalez J, Rodriguez J, Tejedor A. Biodiesel fuels from vegetable oils:
transesterification of Cynara cardunculus L. oils with ethanol. Energy Fuels
2002;16(2):443–50.
[44] Leung DY, Wu X, Leung M. A review on biodiesel production using catalyzed
transesterification. Appl Energy 2010;87(4):1083–95.
[45] Na Y, Youyong S, Hua W, Wu Z, Xiaohong H. Experimental study on preparation of
biodiesel through transesterification with ultrasonic assistant. Materials for
Renewable Energy & Environment (ICMREE), 2011 International Conference on.
1. IEEE; 2011:284-7.
[46] Kumar D, Kumar G, Singh C. Fast, easy ethanolysis of coconut oil for biodiesel
production assisted by ultrasonication. Ultrason Sonochem 2010;17(3):555–9.
[47] Lam MK, Lee KT, Mohamed AR. Homogeneous, heterogeneous and enzymatic
catalysis for transesterification of high free fatty acid oil (waste cooking oil) to
biodiesel: a review. Biotechnol Adv 2010;28(4):500–18.
[48] Ramadhas AS, Jayaraj S, Muraleedharan C. Biodiesel production from high FFA
rubber seed oil. Fuel 2005;84(4):335–40.
[49] Yin X, Ma H, You Q, Wang Z, Chang J. Comparison of four different enhancing
methods for preparing biodiesel through transesterification of sunflower oil. Appl
Energy 2012;91(1):320–5.
[50] Santos FF, Malveira JQ, Cruz MG, Fernandes FA. Production of biodiesel by ul-
trasound assisted esterification of Oreochromis niloticus oil. Fuel 2010;89(2):275–9.
[51] Parkar PA, Choudhary HA, Moholkar VS. Mechanistic and kinetic investigations in
ultrasound assisted acid catalyzed biodiesel synthesis. Chem Eng J
2012;187:248–60.
[52] Choudhury HA, Malani RS, Moholkar VS. Acid catalyzed biodiesel synthesis from
Jatropha oil: mechanistic aspects of ultrasonic intensification. Chem Eng J
2013;231:262–72.
[53] Mathiarasi R, Partha N. Optimization, kinetics and thermodynamic studies on oil
extraction from Daturametel Linn oil seed for biodiesel production. Renewable
Energy 2016;96:583–90.
[54] Choudhury HA, Goswami PP, Malani RS, Moholkar VS. Ultrasonic biodiesel
synthesis from crude Jatropha curcas oil with heterogeneous base catalyst: me-
chanistic insight and statistical optimization. Ultrason Sonochem
2014;21(3):1050–64.
[55] Ali SD, Javed IN, Rana UA, Nazar MF, Ahmed W, Junaid A, et al. SrO-CaO mixed
906
Fuel 235 (2019) 886–907
metal oxides catalyst for ultrasonic-assisted transesterification of Jatropha oil into
biodiesel. Aust J Chem 2016.
[56] Ramachandran K, Suganya T, Gandhi NN, Renganathan S. Recent developments
for biodiesel production by ultrasonic assist transesterification using different
heterogeneous catalyst: a review. Renewable Sustainable Energy Rev
2013;22:410–8.
[57] Choedkiatsakul I, Ngaosuwan K, Assabumrungrat S. Application of heterogeneous
catalysts for transesterification of refined palm oil in ultrasound-assisted reactor.
Fuel Process Technol 2013;111:22–8.
[58] Pukale DD, Maddikeri GL, Gogate PR, Pandit AB, Pratap AP. Ultrasound assisted
transesterification of waste cooking oil using heterogeneous solid catalyst.
Ultrason Sonochem 2015;22:278–86.
[59] Gupta AR, Yadav SV, Rathod VK. Enhancement in biodiesel production using
waste cooking oil and calcium diglyceroxide as a heterogeneous catalyst in pre-
sence of ultrasound. Fuel 2015;158:800–6.
[60] Deng X, Fang Z, Liu Y-H, Yu C-L. Production of biodiesel from Jatropha oil cata-
lyzed by nanosized solid basic catalyst. Energy 2011;36(2):777–84.
[61] Yadav AK, Khan ME, Pal A, Singh B. Ultrasonic-assisted optimization of biodiesel
production from Karabi oil using heterogeneous catalyst. Biofuels 2016:1–12.
[62] Jogi R, Murthy YS, Satyanarayana M, Rao TN, Javed S. Biodiesel production from
degummed Jatropha curcas oil using constant-temperature ultrasonic water bath.
Energy Sources Part A 2016;38(17):2610–6.
[63] Takase M, Chen Y, Liu H, Zhao T, Yang L, Wu X. Biodiesel production from non-
edible Silybum marianum oil using heterogeneous solid base catalyst under ultra-
sonication. Ultrason Sonochem 2014;21(5):1752–62.
[64] Ali SD, Javed IN, Rana UA, Nazar MF, Ahmed W, Junaid A, et al. Novel SrO-CaO
mixed metal oxides catalyst for ultrasonic-assisted transesterification of Jatropha
oil into biodiesel. Aust J Chem 2017;70(3):258.
[65] Badday AS, Abdullah AZ, Lee K-T. Ultrasound-assisted transesterification of crude
Jatropha oil using alumina-supported heteropolyacid catalyst. Appl Energy
2013;105:380–8.
[66] Badday AS, Abdullah AZ, Lee K-T. Optimization of biodiesel production process
from Jatropha oil using supported heteropolyacid catalyst and assisted by ultra-
sonic energy. Renewable Energy 2013;50:427–32.
[67] Badday AS, Abdullah AZ, Lee K-T. Ultrasound-assisted transesterification of crude
Jatropha oil using cesium doped heteropolyacid catalyst: interactions between
process variables. Energy 2013;60:283–91.
[68] Maneechakr P, Samerjit J, Karnjanakom S. Ultrasonic-assisted biodiesel produc-
tion from waste cooking oil over novel sulfonic functionalized carbon spheres
derived from cyclodextrin via one-step: a way to produce biodiesel at short reac-
tion time. RSC Adv 2015;5(68):55252–61.
[69] Nikseresht A, Daniyali A, Ali-Mohammadi M, Afzalinia A, Mirzaie A. Ultrasound-
assisted biodiesel production by a novel composite of Fe(III)-based MOF and
phosphotangestic acid as efficient and reusable catalyst. Ultrason Sonochem
2017;37:203–7.
[70] Gole VL, Gogate PR. Intensification of synthesis of biodiesel from nonedible oils
using sonochemical reactors. Ind Eng Chem Res 2012;51(37):11866–74.
[71] Bahadur S, Goyal P, Sudhakar K, Prakash Bijarniya J. A comparative study of ul-
trasonic and conventional methods of biodiesel production from mahua oil.
Biofuels 2015;6(1–2):107–13.
[72] Worapun I, Pianthong K, Thaiyasuit P. Two-step biodiesel production from crude
Jatropha curcas L. oil using ultrasonic irradiation assisted. J Oleo Sci
2012;61(4):165–72.
[73] Saha R, Goud VV. Ultrasound assisted transesterification of high free fatty acids
karanja oil using heterogeneous base catalysts. Biomass Convers Biorefin
2014;5(2):195–207.
[74] Bahadur S, Goyal P, Sudhakar K. Ultrasonic assisted transesterification of neem oil
for biodiesel production. Energy Sources Part A 2015;37(17):1921–7.
[75] Prakash Maran J, Priya B. Modeling of ultrasound assisted intensification of bio-
diesel production from neem (Azadirachta indica) oil using response surface
methodology and artificial neural network. Fuel 2015;143:262–7.
[76] Yadav AK, Khan ME, Pal A, Dubey AM. Biodiesel production from Nerium oleander
(Thevetia peruviana) oil through conventional and ultrasonic irradiation methods.
Energy Sources Part A 2016;38(23):3447–52.
[77] Sarve AN, Varma MN, Sonawane SS. Ultrasound assisted two-stage biodiesel
synthesis from non-edible Schleichera triguga oil using heterogeneous catalyst:
Kinetics and thermodynamic analysis. Ultrason Sonochem 2016;29:288–98.
[78] Reshad AS, Panjiara D, Tiwari P, Goud VV. Two-step process for production of
methyl ester from rubber seed oil using barium hydroxide octahydrate catalyst:
process optimization. J Cleaner Prod 2017;142:3490–9.
[79] Tian Y, Xiang J, Verni CC, Soh L. Fatty acid methyl ester production via ferric
sulfate catalyzed interesterification. Biomass Bioenergy 2018;115:82–7.
[80] dos Santos Ribeiro J, Celante D, Simões SS, Bassaco MM, da Silva C, de Castilhos F.
Efficiency of heterogeneous catalysts in interesterification reaction from macaw oil
(Acrocomia aculeata) and methyl acetate. Fuel 2017;200:499–505.
[81] Subhedar PB, Gogate PR. Ultrasound assisted intensification of biodiesel produc-
tion using enzymatic interesterification. Ultrason Sonochem 2016;29:67–75.
[82] Tian Y, Xiang J, Verni CC, Soh L, Brahma B, Nath AJ, et al. Fatty acid methyl ester
production via ferric sulfate catalyzed interesterification. Biomass Bioenergy
2018;115:1–266.
[83] Tavares GR, Gonçalves JE, dos Santos WD, da Silva C. Enzymatic interesterifica-
tion of crambe oil assisted by ultrasound. Ind Crops Prod 2017;97:218–23.
[84] Mohan S, Pal A, Singh R. The production of semal oil methyl esters through a
combined process reactor. Energy Sources Part A 2017:1–8.
[85] Mohod AV, Subudhi AS, Gogate PR. Intensification of esterification of non edible
oil as sustainable feedstock using cavitational reactors. Ultrason Sonochem
S.X. Tan et al.
2017;36:309–18.
[86] Kumar D, Ali A. Transesterification of low-quality triglycerides over a Zn/CaO
heterogeneous catalyst: kinetics and reusability studies. Energy Fuels
2013;27(7):3758–68.
[87] Samani BH, Zareiforoush H, Lorigooini Z, Ghobadian B, Rostami S, Fayyazi E.
Ultrasonic-assisted production of biodiesel from Pistacia atlantica Desf. oil. Fuel
2016;168:22–6.
[88] Mootabadi H, Salamatinia B, Bhatia S, Abdullah AZ. Ultrasonic-assisted biodiesel
production process from palm oil using alkaline earth metal oxides as the het-
erogeneous catalysts. Fuel 2010;89(8):1818–25.
[89] Fayyazi E, Ghobadian B, Najafi G, Hosseinzadeh B. Genetic algorithm approach to
optimize biodiesel production by ultrasonic system. Chem Prod Process Model
2014;9(1):59–70.
[90] Khosravi E, Shariati A, Nikou MRK. Instant biodiesel production from waste
cooking oil under industrial ultrasonic irradiation. Int J Oil Gas Coal Technol
2016;11(3):308–17.
[91] Wang Y-Y, Đăng TnHp, Chen B-H, Lee,. D-J. Transesterification of triolein to
biodiesel using sodium-loaded catalysts prepared from zeolites. Ind Eng Chem Res
2012;51(30):9959–65.
[92] Chen G, Shan R, Shi J, Yan B. Ultrasonic-assisted production of biodiesel from
transesterification of palm oil over ostrich eggshell-derived CaO catalysts.
Bioresour Technol 2014;171:428–32.
[93] Zhang F, Fang Z, Wang Y-T. Biodiesel production directly from oils with high acid
value by magnetic Na2SiO3@Fe3O4/C catalyst and ultrasound. Fuel
2015;150:370–7.
[94] Gaikwad ND, Gogate PR. Synthesis and application of carbon based heterogeneous
catalysts for ultrasound assisted biodiesel production. Green Process Synth
2015;4(1):17–30.
[95] Encinar J, Pardal A, Martínez G. Transesterification of rapeseed oil in subcritical
methanol conditions. Fuel Process Technol 2012;94(1):40–6.
[96] Encinar J, González J, Pardal A. Transesterification of castor oil under ultrasonic
irradiation conditions. Preliminary results. Fuel Process Technol 2012;103:9–15.
[97] Noureddini H, Harkey D, Medikonduru V. A continuous process for the conversion
of vegetable oils into methyl esters of fatty acids. J Am Oil Chem Soc
1998;75(12):1775–83.
[98] Hingu SM, Gogate PR, Rathod VK. Synthesis of biodiesel from waste cooking oil
using sonochemical reactors. Ultrason Sonochem 2010;17(5):827–32.
[99] Kesgin C, Yücel S, Özçimen D, Terzioğlu P, Attar A. Transesterification of hazelnut
oil by ultrasonic irradiation. Int J Green Energy 2014;13(3):328–33.
[100] Sebayang D, Agustian E, Praptijanto A. Transesterification of biodiesel from waste
cooking oil using ultrasonic technique. 2010.
[101] Guldhe A, Singh B, Rawat I, Bux F. Synthesis of biodiesel from Scenedesmus sp. by
microwave and ultrasound assisted in situ transesterification using tungstated
zirconia as a solid acid catalyst. Chem Eng Res Des 2014;92(8):1503–11.
[102] Ramachandran K, Al-Zuhair S, Fong C, Gak C. Kinetic study on hydrolysis of oils
by lipase with ultrasonic emulsification. Biochem Eng J 2006;32(1):19–24.
[103] Freedman B, Pryde E, Mounts T. Variables affecting the yields of fatty esters from
transesterified vegetable oils. J Am Oil Chem Soc 1984;61(10):1638–43.
[104] Eevera T, Rajendran K, Saradha S. Biodiesel production process optimization and
characterization to assess the suitability of the product for varied environmental
conditions. Renewable Energy 2009;34(3):762–5.
[105] Chand P, Reddy CV, Verkade JG, Grewell D. Enhancing biodiesel production from
soybean oil using ultrasonics. Am Soc Agric Biol Eng 2008.
[106] Salamatinia B, Mootabadi H, Hashemizadeh I, Abdullah AZ. Intensification of
biodiesel production from vegetable oils using ultrasonic-assisted process: opti-
mization and kinetic. Chem Eng Process Process Intensif 2013;73:135–43.
[107] Takase M, Feng W, Wang W, Gu X, Zhu Y, Li T, et al. Silybum marianum oil as a
new potential non-edible feedstock for biodiesel: a comparison of its production
using conventional and ultrasonic assisted method. Fuel Process Technol
2014;123:19–26.
[108] Maghami M, Sadrameli S, Ghobadian B. Production of biodiesel from fishmeal
plant waste oil using ultrasonic and conventional methods. Appl Therm Eng
2015;75:575–9.
[109] Koutsouki A, Tegou E, Kontakos S, Kontominas M, Pomonis P, Manos G. In situ
transesterification of Cynara cardunculus L. seed oil via direct ultrasonication for
the production of biodiesel. Fuel Process Technol 2015;134:122–9.
[110] Krishnaiah G, Pasnoori S, Santhoshi P, Rajanna K, Rao YR, Patnaik KR. Ultrasonic
and microwave effects on crystalline Mn (II) carbonate catalyzed biodiesel pro-
duction using watermelon (Citrullus vulgaris) seed oil and alcohol (fibrous flesh) as
exclusive green feedstock. Biofuels 2016:1–7.
[111] Boffito D, Galli F, Pirola C, Bianchi C, Patience G. Ultrasonic free fatty acids es-
terification in tobacco and canola oil. Ultrason Sonochem 2014;21(6):1969–75.
[112] Ragavan SN, Roy DV. Transesterification of rubber seed oil by sonication tech-
nique for the production of methyl esters. Biomass Convers Biorefin
2011;1(2):105–10.
[113] Zou H, Lei M. Optimum process and kinetic study of Jatropha curcas oil pre-es-
terification in ultrasonical field. J Taiwan Inst Chem Eng 2012;43(5):730–5.
[114] Brasil AN, Oliveira LS, Franca AS. Circulation flow reactor with ultrasound irra-
diation for the transesterification of vegetable oils. Renewable Energy
2015;83:1059–65.
[115] Refaat A, Attia N, Sibak HA, El Sheltawy S, ElDiwani G. Production optimization
and quality assessment of biodiesel from waste vegetable oil. Int J Environ Sci
Technol 2008;5(1):75–82.
907
Fuel 235 (2019) 886–907
[116] Refaat A, El Sheltawy S. Comparing three options for biodiesel production from
waste vegetable oil. WIT Trans Ecol Environ 2008;109:133–40.
[117] Kumar D, Kumar G, Johari R, Kumar P. Fast, easy ethanomethanolysis of Jatropha
curcus oil for biodiesel production due to the better solubility of oil with ethanol in
reaction mixture assisted by ultrasonication. Ultrason Sonochem
2012;19(4):816–22.
[118] Subhedar PB, Gogate PR. Alkaline and ultrasound assisted alkaline pretreatment
for intensification of delignification process from sustainable raw-material.
Ultrason Sonochem 2014;21(1):216–25.
[119] Okitsu K, Maeda Y, Bandow H. Ultrasound assisted production of fatty acid methyl
esters from transesterification of triglycerides with methanol in the presence of
KOH catalyst: optimization, mechanism and kinetics. Ultrason Sonochem
2014;21(2):467–71.
[120] Birla A, Singh B, Upadhyay S, Sharma Y. Kinetics studies of synthesis of biodiesel
from waste frying oil using a heterogeneous catalyst derived from snail shell.
Bioresour Technol 2012;106:95–100.
[121] Vyas AP, Verma JL, Subrahmanyam N. Effects of molar ratio, alkali catalyst
concentration and temperature on transesterification of Jatropha oil with me-
thanol under ultrasonic irradiation. Adv Chem Eng Sci 2011;1(02):45.
[122] Stavarache C, Vinatoru M, Maeda Y, Bandow H. Ultrasonically driven continuous
process for vegetable oil transesterification. Ultrason Sonochem
2007;14(4):413–7.
[123] Thanh LT, Okitsu K, Sadanaga Y, Takenaka N, Maeda Y, Bandow H. A two-step
continuous ultrasound assisted production of biodiesel fuel from waste cooking
oils: a practical and economical approach to produce high quality biodiesel fuel.
Bioresour Technol 2010;101(14):5394.
[124] Mostafaei M, Ghobadian B, Barzegar M, Banakar A. Optimization of ultrasonic
assisted continuous production of biodiesel using response surface methodology.
Ultrason Sonochem 2015;27:54–61.
[125] Mohan S, Pal A, Singh R. Biodiesel production from Cedrus deodara oil in different
types of ultrasonic reactors and energy analysis. Energy Sources Part A
2016;38(24):3709–15.
[126] Yin X, Zhang X, Wan M, Duan X, You Q, Zhang J, et al. Intensification of biodiesel
production using dual-frequency counter-current pulsed ultrasound. Ultrason
Sonochem 2017;37:136–43.
[127] Murillo G, Sun J, Ali SS, Yan Y, Bartocci P, He Y. Evaluation of the kinematic
viscosity in biodiesel production with waste vegetable oil, ultrasonic irradiation
and enzymatic catalysis: a comparative study in two-reactors. Fuel
2018;227:448–56.
[128] Sajjadi B, Asaithambi P, Abdul Aziz AR, Ibrahim S. Mathematical analysis of the
effects of operating conditions and rheological behaviour of reaction medium on
biodiesel synthesis under ultrasound irradiation. Fuel 2016;184:637–47.
[129] Mazanov SV, Gabitova AR, Usmanov RA, Gumerov FM, Labidi S, Amar MB, et al.
Continuous production of biodiesel from rapeseed oil by ultrasonic assist trans-
esterification in supercritical ethanol. J Supercrit Fluids 2016;118:107–18.
[130] Biktashev SA, Usmanov R, Gabitov R, Gazizov R, Gumerov F, Gabitov F, et al.
Transesterification of rapeseed and palm oils in supercritical methanol and
ethanol. Biomass Bioenergy 2011;35(7):2999–3011.
[131] Cintas P, Mantegna S, Gaudino EC, Cravotto G. A new pilot flow reactor for high-
intensity ultrasound irradiation. Application to the synthesis of biodiesel. Ultrason
Sonochem 2010;17(6):985–9.
[132] Widayat, Satriadi H, Choirudin F, Fitriana A, Kiono BFT. Biodiesel production with
continuous processing and direct ultrasonic assisted. 2016 International
Conference on Sustainable Energy Engineering and Application (ICSEEA). IEEE;
2016:122-6.
[133] Banković-Ilić IB, Stamenković OS, Veljković VB. Biodiesel production from non-
edible plant oils. Renewable Sustainable Energy Rev 2012;16(6):3621–47.
[134] Shuit SH, Lee KT, Kamaruddin AH, Yusup S. Reactive extraction and in situ es-
terification of Jatropha curcas L. seeds for the production of biodiesel. Fuel
2010;89(2):527–30.
[135] Haas MJ, Scott KM. Moisture removal substantially improves the efficiency of in
situ biodiesel production from soybeans. J Am Oil Chem Soc 2007;84(2):197–204.
[136] Zahari MSM, Ismail S, Ibrahim MZ, Lam SS, Mat R. Ultrasonicated Jatropha curcas
seed residual as potential biofuel feedstock. Jurnal Teknologi 2015;77(1).
[137] Chadha P, Arora AK, Prakash S, Jha MK, Puri SK, Tuli DK, et al. Ultrasonic assisted
in situ transesterification of Jatropha seed to biodiesel. J Sci Ind Res
2012;71(4):290.
[138] Zahari MSM, Ismail S, Ibrahim MZ, Lam SS, Mat R. Study of enhanced reactive
extraction process using ultrasonication for Jatropha curcas seed. Appl Mech Mater
2014;699.
[139] Kumar G. Ultrasonic-assisted reactive-extraction is a fast and easy method for
biodiesel production from Jatropha curcas oilseeds. Ultrason Sonochem
2017;37:634–9.
[140] Jadhav AJ, Holkar CR, Goswami AD, Pandit AB, Pinjari DV. Acoustic cavitation as
a novel approach for extraction of oil from waste date seeds. ACS Sustainable
Chem Eng 2016;4(8):4256–63.
[141] Shirsath S, Sonawane S, Gogate P. Intensification of extraction of natural products
using ultrasonic irradiations—a review of current status. Chem Eng Process
Process Intensif 2012;53:10–23.
[142] Poosumas J, Ngaosuwan K, Quitain AT, Assabumrungrat S. Role of ultrasonic ir-
radiation on transesterification of palm oil using calcium oxide as a solid base
catalyst. Energy Convers Manage 2016;120:62–70.