Ultrasonics 44 (2006) e411–e414

www.elsevier.com/locate/ultras

Preparation of biodiesel with the help of ultrasonic

and hydrodynamic cavitation
Jianbing Ji, Jianli Wang *, Yongchao Li, Yunliang Yu, Zhichao Xu
College of Chemical Engineering and Materials Science, Zhejiang Key Laboratory of Green Chemistry Synthesis Technology,
Zhejiang University of Technology, Hangzhou 310014, China

Available online 2 June 2006

Abstract

An alkali-catalyzed biodiesel production method with power ultrasonic (19.7 kHz) has been developed that allows a short reaction
time and high yield because of emulsification and cavitation of the liquid–liquid immiscible system. Orthogonality experiments were
employed to evaluate the effects of synthesis parameters. Furthermore, hydrodynamic cavitation was used for biodiesel production in
comparison to ultrasonic method. Both methods were proved to be efficient, and time and energy saving for the preparation of biodiesel
by transesterification of soybean oil.
Ó 2006 Published by Elsevier B.V.

Keywords: Biodiesel; Ultrasonic; Hydrodynamic cavitation; Energy consumption

1. Introduction introduced into the reactants and the methanolysis was

Biodiesel fuel, chemically consisted of fatty acid methyl
esters (FAMEs) produced by methanolysis of natural tri-
glycerides, such as animal fats and vegetable oils, is a kind
of biomass energy [1]. It is known as an alternative of con-
ventional petrodiesel due to its renewability and better
combustion performance [2,3]. The conventional biodiesel
technology involves the use of a base or acid catalyst at
or near the boiling temperatures of the triglyceride/metha-
nol mixture. In recent years, the researches on enzymatic
and supercritical transesterification are active [4,5]. David
et al. [6] reported that the methoxide base-catalyzed meth-
anolysis of soybean oil at 40 °C (methanol:oil = 6:1) to
synthesize FAMEs proceeded approximately 15 times more
slowly than butanolysis at 30 °C. It was interpreted to be
the result of a two-phase reaction in which methanolysis
occurred only in the methanol phase. So a cosolvent was
*
Corresponding author. Tel.: +86 571 88320917; fax: +86 571
88320053.
E-mail address: wangjl@zjut.edu.cn (J. Wang).
improved obviously. Freedam et al. [7] studied the kinetics
of sodium methoxide-catalyzed methanolysis of soybean
oil and found that the reaction had a low activation energy
of 33–83.7 kJ mol 1. The interaction of reagents was con-
sidered as the determined step of the above process.
It is known that power ultrasonic is a useful tool for
strengthening mass transfer of liquid–liquid heterogeneous
system [8]. In this process, cavities would be created by the
irradiation of power ultrasonic with sufficient energy in the
immiscible liquids. As a result, micro fine bubbles are
formed. The asymmetric collapse of the cavitation bubbles
disrupts the phase boundary and impinging of the liquids
creates micro jets, leading to intensive emulsification of
the system. Carmen et al. [9] had studied the effect of power
ultrasonic of different frequencies on the transesterification
reaction of vegetable oils with short-chain alcohols. Hydro-
dynamic cavitation [10] is another technology developed
recently and allows the generation of cavity collapse condi-
tions similar to acoustic cavitation, thereby enabling differ-
ent applications requiring different cavitational intensities,
which have been successfully carried out using acoustic
cavitation phenomena but at much lower energy inputs

0041-624X/$ - see front matter Ó 2006 Published by Elsevier B.V.

doi:10.1016/j.ultras.2006.05.020
e412 J. Ji et al. / Ultrasonics 44 (2006) e411–e414

as compared to sonochemical reactors. In the present Table 2

work, power ultrasonic and hydrodynamic cavitation reac- Orthogonality experiments design
tors were applied to prepare biodiesel through sodium Factor A, B, molar C, pulse D,
hydroxide-catalyzed methanolysis of soybean oil. Finally, power/W ratio frequency temperature/°C
energy consumption of mechanical stirring (MS), hydrody- I 100 3:1 0.4 25
namic cavitation (HC) and power ultrasonic (PU) was II 150 4.5:1 0.7 35
III 200 6:1 1.0 45
estimated.

2. Methods

2.1. Materials

The molecular weight (MW) of soybean oil was calcu-

lated from its saponification value (SV) and acid value
(AV), and is given in Table 1, Methanol, NaOH, potassium
dihydrogen phosphate and other reagents were used as
purchased.

2.2. Reaction and optimization

Fig. 2. Schematic diagram of hydrodynamic cavitation system setup. 1–
The reactions were carried out in an ultrasonic reactor tank; 2 – cold water; 3 –pump; 4 – orifice plate.
with a jacket, as presented in Fig. 1. The horn of the trans-
ducer was submerged 2 cm in the reactive mixture. The
temperature of the reaction mixture was controlled by a 2.3. Analysis
water bath. Vegetable oil (100 g) was poured into the reac-
tor at the beginning of each reaction for warm-up. The Samples were analyzed by a capillary gas chromato-
reaction started when a quantitative amount of methanol graph (GC) with a FID detector.
liquor dissolved in NaOH was poured into the heated reac-
tor. Orthogonality experiments were designed with 4 fac- 3. Results and discussion
tors and 3 levels, which are listed in Table 2.
Hydrodynamic cavitation system setup is presented in 3.1. Orthogonality experiments
Fig. 2. For cavitation, the process was carried out similar
to that described in the literature [11]. FAMEs yields in 30 min and orthogonality analysis are
listed in Table 3. According to range magnitude of the fac-
tors surveyed, substrate molar ratio was the primary factor,
Table 1 while temperature and pulse frequency of ultrasonic were
Properties of the vegetable oil secondary. The order of the effect on FAMEs yield of the
Properties AV SV IV MW factors was substrate molar ratio > temperature > pulse
(mgKOH/g) (mgKOH/g) (gI2/100 g) (g/mol) frequency > ultrasonic power. In this study, the optimal
Soybean oil 0.2 193.9 129.2 868.9 reaction conditions might be 6:1 substrate molar ratio,
45 °C, continuous ultrasonic and 150 W ultrasonic power.
The stoichiometric substrate molar ratio was 3:1. But in
this case, FAMEs yield was low. To obtain a high yield, the
substrate molar ratio must be increased [12]. Continuous
ultrasonic (pulse frequency = 1) could induce strong emul-
sification of the methanol–oil phases in a short time. During
the experiments, it was observed that higher the ultrasonic
power, shorter was the entire mixing time. As presented in
Table 3, continuous ultrasonic could give a short reaction
equilibrium time of 10–20 min and almost 100% yield under
6:1 substrate molar ratio. Pulse ultrasonic (pulse fre-
quency < 1) was proved to be an efficient way to save energy
and enhance mass transfer [13]. But as pulse frequency was
lower than 0.7, macro-stirring effect of the ultrasonic was
Fig. 1. Schematic diagram of ultrasonic system setup. 1– condensator; 2 – too mild to mix the immiscible reactants well. However,
transducer; 3 –ultrasonic reactor; 4 – stand support; 5 – thermometer; 6 – in the case of ultrasonic power of 200 W, the yield of
ultrasonic generator. FAMEs decreased. The possible reason was that methanol
 J. Ji et al. / Ultrasonics 44 (2006) e411–e414 e413

Table 3 Table 4
Orthogonality experiments results Energy consumption for the transesterification of 1 Kg soybean oil by MS,
PU and HC
No. A B C D 30 min
yield/% MS PU HC
1 I I I I 69.45 Energy consumption(Wh/Kg) 500 250 183
2 I II II II 89.33
3 I III III III 100.00
4 II I II III 86.18
5 II II III I 90.20 together with mechanical stirring for improving the transe-
6 II III I II 96.60 sterification reaction, are presented in Fig. 3. The equilib-
7 III I III II 85.45
rium reaction time was shortened in the order of PU, HC
8 III II I III 90.52
9 III III II I 96.78 and MS. Power ultrasonic gave the shortest reaction time
and the highest yield. Mechanical stirring offered the slow-
M1 258.78 241.08 256.57 256.43 804.51
est reaction rate. PU and HC methods reduced the reaction
M2 272.98 270.05 272.29 271.38
M3 272.75 293.38 275.65 276.70 equilibrium time to 10–30 min. In the case of ultrasonic
Rj 14.20 52.30 19.08 20.27 cavitation, it agreed well with the results in the literature
Optimization 150 6:1 1.0 45 [15], the reaction rate constants were three to five times
higher than those reported in the literature for mechanical
agitation. In the case of hydrodynamic cavitation, we sus-
gasification induced by high power ultrasonic produced a pect that the observed mass transfer and kinetic rate
mass of bubbles in the reaction system. The methanol filled enhancements were also due to the increase in the interfa-
bubbles reduced the methanol content in the liquid phase cial area. However, no apparent evidence indicated that
reaction system and interface area. So, larger ultrasonic these cavitations could change the reactions equilibrium.
power does not always give better results. It can be con- It means that all reaction achieve similar equilibrium yields
cluded that the overall reaction rate depends on the emulsi- regardless of the reaction time.
fication degree of the reaction system.
3.3. Energy consumption
3.2. Hydrodynamic cavitation
The respective energy consumption of transesterification
It was reported that cavitating conditions identical to by MS, PU and HC was calculated according to 1 kg bio-
acoustic cavitation could be generated in hydrodynamic diesel formation regardless of heating. From Table 4, it can
cavitation (HC), which even had a better effect on mixing be seen that the PU and HC processes required approxi-
immiscible liquids [14]. Furthermore, scale-up of hydrody- mately a half of the energy that was consumed by the
namic cavitation to meet industrial-scale operations had MS method. It is no doubt that PC and HC were efficient,
better opportunities than the ultrasonic reactor by reason concerning with time or energy.
of its easier generating and less sensitivity to the geometric
details of the reactor. The results of these two methods, 4. Conclusions

100 Power ultrasonic was proved to be an efficient method

for the preparation of biodiesel by the transesterification
of soybean oil. Hydrodynamic cavitation had similar
80
enhancement effect on the transesterification reaction. Both
methods gave shorter reaction time and less energy con-
FAMEs yield (%)

60 sumption than the conventional mechanical stirring

method. Hydrodynamic cavitation was a potential method
that could be used for biodiesel production at industrial
M.S.
40 scale due to its easy scale-up property.
P.U.
H.C.
Acknowledgements
20

The authors would like to thank for the financial sup-

0 port from ‘‘Technology Keystone Items of Hangzhou City,
0 10 20 30 40 50 60 China’’ (Contract No. 200432131).
Reaction time (min.)
Fig. 3. Effect of different methods on FAMEs yield. Reaction conditions: References
substrate molar ratio 6:1; temperature 45 °C; KOH amount 1 wt.%; MS:
speed 900 rpm; PU: frequency 19.7 kHz; power 150 W; HC: operation [1] Taichi Samukawa, Masatu Kaieda, Takeshi Matsumoto, et al.,
pressure 0.7 MPa; single orifice. Journal of Bioscience and Bioengineering 90 (2) (2000) 180–183.
e414 J. Ji et al. / Ultrasonics 44 (2006) e411–e414
[2] B.K. Barnwal, M.P. Sharma, Renewable and Sustainable Energy
Reviews (2004) 1–16.
[3] Yongchao Li, Jianli Wang, Jianbing Ji, et al., China Oils and Fats 30
(5) (2005) 59–64.
[4] Mohamed M. Soumanou, Uwe T. Bornscheuer, European Journal of
Lipid Science and Technology 105 (2003) 656–660.
[5] Giridhar Madras, Chandana Kolluru, Rajnish Kumar, Fuel 83 (2004)
2029–2033.
[6] David G.B. Boocock, Samir K. Konar, Vinnie Mao, et al., Biomass
and Bioenergy 11 (1) (1993) 43–50.
[7] B. Freedman, R.O. Butterfield, E.H. Pryde, JAOCS 63 (1986) 1375–
1380.
[8] Ruo Feng, Ultrasonics Handbook, Publishing Company of Nanjing
University, Nanjing, 1999.
[9] C. Stavarache, M. Vinatoru, R. Nishimura, et al., Ultrasonics
Sonochemistry 12 (2005) 367–372.
[10] ParagR. Gogate, AniruddhaB. Pandit, Ultrasonics Sonochemistry 12
(2005) 21–27.
[11] G.V. Ambulgekar, S.D. Samant, A.B. Pandit, Ultrasonics Sono-
chemistry 11 (2004) 191–196.
[12] Mei Sheng, Dengfeng Guo, Dahua Zhang, China Oils and Fats 27 (1)
(2002) 70–72.
[13] Ruo Feng, Yiyun Zhao, Changping Zhu, et al., Ultrasonics Sono-
chemistry 9 (2002) 231–236.
[14] V.S. Moholkar, P. Senthil Kumar, A.B. Pandit, Ultrasonics Sono-
chemistry 6 (1999) 53–65.
[15] J.A. Colucci, E.E. Borrero, F. Alape, Journal of the American Oil
Chemists Society 82 (7) (2005) 525–530.