Study on preparation of Ca/Al/Fe

3
O
4
magnetic composite solid catalyst and its
application in biodiesel transesterication
Shaokun Tang , Liping Wang, Yi Zhang, Shufen Li, Songjiang Tian, Boyang Wang
Key Laboratory for Green Chemical Technology of Ministry of Education, School of Chemical Engineering & Technology, Tianjin University, Tianjin, 300072, China
a b s t r a c t a r t i c l e i n f o
Article history:
Received 19 July 2011
Received in revised form 12 November 2011
Accepted 26 November 2011
Available online 24 December 2011
Keywords:
Magnetic
Fe
3
O
4
nanoparticles
Solid base catalyst
Biodiesel
Transesterication
A magnetic composite solid catalyst was prepared by loading calcium aluminate onto Fe
3
O
4
nanoparticles via
a chemical synthesis method. The optimum conditions for the catalyst preparation were investigated. The in-
uences of the molar ratio of Ca to Fe, calcining temperature, calcining time on the catalytic performance
were studied. The catalyst with the highest activity was obtained when the molar ratio of Ca to Fe was 5:1;
calcining temperature was 600 C and calcining time was 6 h. The catalyst was characterized by thermogravi-
metric analyses (TGA), X-ray diffraction(XRD), scanning electronic microscope (SEM), BrunauerEmmettTeller
method (BET) and vibrating sample magnetometer (VSM). Furthermore, the magnetic composite solid catalyst
showed high catalytic activity for transesterication reaction for preparing biodiesel and the biodiesel yield
reached 98.71%under the optimumconditions. The activity and recovery rate of this magnetic composite catalyst
can be well maintained after 5 cycles of catalysis. This catalyst showed magnetism and can be easily separated
magnetically. Both the catalytic activity and the recovery rate of the magnetic composite solid catalyst were
much higher than those of pure calcium aluminate catalyst.
2011 Elsevier B.V. All rights reserved.
1. Introduction
Biodiesel has attracted attention in recent years as a renewable
biofuel with less pollutant emissions compared to mineral diesel on
its combustion [1,2]. Transesterication is the most common method
for biodiesel preparation. Most biodiesel today is produced in the
presence of homogeneous catalysts such as sodiummethoxide, sodium
or potassiumhydroxide [3]. Freedman et al. [4] reported that NaOHcan
transformvegetable oil to biodiesel completely in anhour. But their dis-
advantages are the complicated processes of post treatments and
pollution.
In this case, researches have focused on nding a suitable hetero-
geneous catalyst that can be easily separated and give a high yield
and conversion without compromise [5]. Since the catalytic activity
of basic catalysts is higher than that of acid solids, they have been
preferably studied. Heterogeneous solid base catalysts have been
presently reported such as metal oxides CaO [6-9], SrO [10] and so
on; alkali-doped metal oxides KF/MgO [11], CaO/Al
2
O
3
[12], MgO/
Al
2
O
3
[12], Li/CaO [13], CaO/ZnO [14] and so on; supported ones,
such as Na/NaOH/-Al
2
O
3
[15] and KF/-Al
2
O
3
[16]; hydrotalcites
MgAl [1719], KF/hydrotalcite [20], KF/CaMgAl hydrotalcite [21]
and so on; sodium aluminate [22], SnCl
2
[23], calcium ethoxide [24]
and mayenite [25]. Although these heterogeneous catalysts have
advantages in catalyst separation and pollution reduction, most of
them have some limitations in catalytic activity and stability.
The heterogeneous base catalyst for biodiesel preparation has
been developed in our group. Heterogeneous base catalyst K/KOH/
-Al
2
O
3
was prepared and used in the transesterication of rapeseed
oil with methanol to produce biodiesel [26]. The result showed that
the catalyst K/KOH/-Al
2
O
3
had high catalytic activity and the yield
of biodiesel could reach as high as 84.52% after 1 h reaction at 60 C,
with a 9:1 molar ratio of methanol to oil, a catalyst amount of
4 wt.%. However, after two-times use, it was found that the biodiesel
yield decreased to 37.6%, and the catalyst lost weight of 10.8%, which
meant the relatively low stability and recovery rate of the catalyst.
Therefore, the development of a heterogeneous catalyst with high
activity and stability is still in high demand.
Nanometer magnetic solid base catalyst can be separated easily
from the reagents by an external magnetic eld, which can effectively
prevent catalyst loss and improve its recovery rate during separation
process. Furthermore, magnetic nanoparticlessupported solid catalyst
show high dispersion so as to contact reactants more sufciently based
on the high surface area of the supported magnetic nanoparticles.
Nanometer magnetic catalyst has been attracting more and more in-
creasing attention in recent years. Fe
3
O
4
magnetic particles were used
to immobilize lipase as catalyst for biodiesel production. The results
showed not only high catalytic activity but also the advantages of easy
separation and reuse [27,28]. Hu et al. [29] developed a nano-
magnetic solid base catalyst KF/CaOFe
3
O
4
based on Fe
3
O
4
magnetic
core by impregnation method. The catalyst recovery was more than
90%. When the reaction was carried out at 65 C with a methanol/oil
Fuel Processing Technology 95 (2012) 8489
Corresponding author. Tel./fax: +86 22 27408578.
E-mail address: shktang@tju.edu.cn (S.K. Tang).
0378-3820/$ see front matter 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.fuproc.2011.11.022
Contents lists available at SciVerse ScienceDirect
Fuel Processing Technology
j our nal homepage: www. el sevi er . com/ l ocat e/ f upr oc
molar ratio of 12:1 and a catalyst concentration of 4 wt.%, the biodiesel
yield exceeded 95% at 3 h of reaction time. Liu et al. [30] prepared nano-
meter magnetic solid base catalysts for transesterication reaction by
loading CaO on Fe
3
O
4
with Na
2
CO
3
and NaOH as precipitator, respec-
tively. The conversion rate of transesterication reaction catalyzed by
nanometer magnetic solid base catalysts under optimum conditions
could reach 95% in 80 min, and to 99% in 4 h. Fatty acid methyl esters
(FAME) yield was more than 90% after 5-times use and still more than
70% after 10-times use. The recovery rate of the Ca(OH)
2
Fe
3
O
4
catalyst
after reaction reached up to 91.45%. According to Scherer formula
D=0.89/cos, the average particle diameter was calculated to be
49.76 nm. But the BET specic surface area of this catalyst was only
3.72 m
2
/g and the saturation magnetization of magnetic catalysts was
smaller than 1 eum/g.
In our work, calcium aluminate catalyst has been developed via a
chemical synthesis method for transesterication reaction of biodiesel
production. Furthermore, the magnetic composite solid catalyst has
been prepared by loading calcium aluminate onto the nanometer
Fe
3
O
4
to improve the catalytic performance including the catalytic
activity and recovery in application to biodiesel preparation.
2. Experimental section
2.1. Materials
Rapeseed oil was purchased from Ningbo zhengda grain and oil
Company. Sodium hydroxide, methanol, Al, CaCO
3
, absolute alcohol
and isopropanol were analytical grade, obtained from Tianjin Jiangtian
Chemical Reagent Company, China. Fe
3
O
4
(99.5%, 20 nm spherical par-
ticles) was purchased from Shanghai Yutian Chemical Company.
2.2. Preparation of Ca/Al/Fe
3
O
4
magnetic composite catalysts
NaOH was solved into distilled water, and then aluminum sheet
was added into NaOH solution. After reaction, sodium meta-
aluminate solution was obtained. CaO obtained by calcining of
CaCO
3
at 1000 C for 10 h, and CaO was added into distilled water,
then calcium hydroxide suspensions formed. The molar ratio of
NaOH, Al and CaO was 2:2:3. A certain amount of Fe
3
O
4
nanoparticles
was added into the mixed solution of NaOHand Ca(OH)
2
with vigorous
stirring at 80 C for 5 h. The as-prepared powders were thoroughly
washed with deionized water, dehydrated using alcohol, and then
dried under vacuum for 12 h. At last, the dried powder was calcined
to produce magnetic composite solid catalysts.
The inuences of the molar ratio of Ca to Fe, calcining tempera-
ture and calcining time on FAME yield were investigated in this
work. Besides, in order to compare the catalytic activity of the mag-
netic composite solid catalyst with that of pure calcium aluminate
catalysts, the calcium aluminate catalysts were prepared in the
same way.
2.3. Transesterication for biodiesel
The transesterication of 67.5 g rapeseed oil was carried out in a
250 ml round-bottom ask, provided with a thermostatic and mag-
netic stirring system. In our previous study, the optimum transester-
ication conditions catalyzed by calcium aluminate were obtained as
follows: methanol/oil molar ratio 15:1, catalyst dosage 6 wt.% of rape-
seed oil, reaction temperature 65 C, the stirring rate of 270 rpm and
reaction time of 3 h.
All the transesterication reactions in this paper were performed
under the obtained optimum conditions to determine the catalytic
activity of the catalysts. After reaction, the magnetic composite solid
catalysts were recycled by magnetic separation. A small amount of
the product mixtures was taken and washed with deionized water
at 70 C, centrifuged for 30 min. Then the sample taken from the oil
phase was analyzed by gas chromatography.
2.4. GC and GC-MS analysis of FAME
The sample was taken from the oil phase and determined by
GC (SP-2100) with capillary column of H.J.PEG-20M
(30 m0.32 mm0.5 m). To detect the yield of biodiesel, methyl
salicylate was used as internal standard, ethyl acetate as solvent. The
column temperature was 130 C, the temperatures of the injector and
detector were respectably 180 Cand 280 C. Programmedtemperature
was that initial temperature was 130 C remaining 3 min, the tempera-
ture was elevated by 30 C/min to 190 C, then elevated by 5 C/min to
230 C remaining 5 min.
The components of fatty acid methyl esters after transesterica-
tion reaction were determined by GC-MS (6890-5973N, Aglient).
The ion source of MS was electron ionization at 230 C with capillary
column of HP-INNOwax (30 m0.25 mm0.25 m). The interface
and quadrupole temperatures were 250 C and 150 C respectively.
The initial column temperature was 160 C remaining 2 min, and
then was programmed elevated by 5 C/min to 240 C remaining
10 min. There were mainly seven components including methyl
palmiate, methyl stearate, methyl oleate, methyl linoleate, methyl
linolenate, methyl cis-11-eicosenoate and methyl erucate
respectively.
The FAME yield was calculated by internal standard method using
the following equation:
Yield
FAME

f
ester
A
ester
A
internal

m
internal
m
esters
100%: 1
Informula (1), A
ester
is the peak area of fatty acidmethyl esters, A
internal
is the peak area of internal standard (methyl salicylate), m
internal
is the
mass of internal standard (methyl salicylate), m
esters
is the mass of fatty
acid methyl esters andf
ester
is the correction factor of fatty acid methyl es-
ters. The correction factors of methyl palmiate, methyl stearate, methyl
oleate, methyl linoleate, methyl linolenate, methyl cis-11-eicosenoate
and methyl erucate were determined as 0.8250, 1.1789, 0.8121, 0.8154,
0.9556, 1.9434, 3.0437 respectively.
2.5. Characterization of the solid catalysts
The morphologies of the as-prepared calcium aluminate catalysts
and magnetic composite solid catalysts were observed by scanning
electronic microscopy (SEM, Nanosem 430, FEI, USA). The thermal
stability of catalyst was examined using thermogravimetric analyzer
(TGA, TA-50, Shimadzu, Japan) from room temperature to 750 C
under an inert nitrogen atmosphere and a heating rate of
10 C min
1
. The magnetic hysteresis loops of the Fe
3
O
4
particles
were measured with a vibrating sample magnetometer (VSM, LDJ
9600, LDJ Electronics, USA) at roomtemperature under applied mag-
netic eld of 1T. X-ray diffraction analysis was carried out by using
an X'Pert PRD X-ray diffraction system (Philips X' pert, Cu K radia-
tion k=1.54056 , USA). The specic surface areas of the catalysts
were detected by the BET nitrogen adsorption method at 196 C
(Tristar3000, Micromeritics, USA).
3. Results and discussion
3.1. Catalytic activities of magnetic composite solid catalysts
To explore the optimal conditions for catalyst preparation, different
molar ratios (10:1, 9:1, 8:1, 7:1, 6:1, 5:1, 4:1, 3:1 and 2:1) of Ca to Fe,
calcining temperature (450 C, 500 C, 550 C, 600 C, 650 C, 700 C,
750 C and 800 C), calcining time (2 h, 3 h, 4 h, 5 h, 6 h, 7 h, 8 h and
9 h) were studied.
85 S.K. Tang et al. / Fuel Processing Technology 95 (2012) 8489
3.1.1. Effect of molar ratio of Ca to Fe
The effect of n (Ca): n (Fe) on the FAME yield was studied when
the calcining temperature was 600 C and calcining time was 6 h.
Fig. 1 indicated that the FAME yield increased with the increase of
the molar ratio of Ca to Fe from 2:1 to 5:1. However, with the further
increase of the molar ratio of Ca to Fe from5:1 to 10:1, the FAME yield
decreased. The FAME yield at the molar ratio of Ca to Fe 5:1 was the
highest, and also higher than that with pure calcium aluminate cata-
lysts (89.87%).
XRD determination showed that the magnetic composite catalyst
after calcination was the composite of Ca
12
Al
14
O
33
, CaO and Fe
3
O
4
(seen in Section 3.4.2). On one hand, with the increase of molar
ratio of Fe to Ca, the relative content of Fe
3
O
4
increases and the catalytic
active components Ca
12
Al
14
O
33
and CaO for transesterication de-
crease, and thus reduces the catalytic activity. On the other hand, due
to high specic area of the Fe
3
O
4
nanoparticles supporter, the Ca/Al/
Fe
3
O
4
magnetic composite solid catalyst has larger specic surface
area and thus shows higher dispersion in the reactant with the addition
of Fe
3
O
4
particles. The data of the specic surface areas of the catalysts
can be seen in Section 3.4.4. As a result, the catalytic activity and the reac-
tion yield will be promoted. Based on the above two competitive factors,
the optimum molar ratio of Ca to Fe is 5:1.
3.1.2. Effect of calcining temperature
In this work, the calcining temperature varied within a range from
450 C to 800 C at the molar ratio of Ca to Fe of 5:1 and calcining time
of 6 h. With the increase of calcining temperature, the catalyst gradu-
ally develops into crystal and its specic area increases. The specic
surface area of the magnetic composite catalyst at 600 C was the
highest (25.89 m
2
/g) by BET determination. Correspondingly, the bio-
diesel yield initially increased and reached the highest 98.71% when
the calcining temperature was 600 C as seen in Fig. 2. However, too
high temperature (>600 C) results in the surface sintering and the
reduction of specic surface area of the composite catalyst, which
leads to the decrease of the catalytic activities and the biodiesel
yield with the further increase of the calcining temperature. There-
fore, the optimum calcining temperature is 600 C.
3.1.3. Effect of calcining time
Fig. 3 shows the effect of the calcining time on the yield of methyl
ester when the molar ratio of Ca to Fe is 5:1 and the calcining temper-
ature is 600 C. It can be found that the yield of methyl ester was the
highest 98.71% when magnetic catalysts were calcined for 6 h at
600 C.
With the increase of the calcining time from2 h to 6 h, the catalyst
gradually develops into stable crystal and the specic area of catalyst
increases. As a result, the FAME yield increases and reaches the highest
when the magnetic catalyst is calcined for 6 h. With the further increase
of the calcining time, part of active component of the catalysts may
be lost and catalyst agglomeration will happen, which results in the
decrease of the FAME yields. In our work, the calcining time of 6 h
is considered as the optimum.
In summary, Ca/Al/Fe
3
O
4
magnetic composite catalyst with the
highest catalytic activity has been obtained when the molar ratio of
Ca to Fe was 5:1, calcining temperature was 600 C and calcining
time was 6 h. All the as-calcined magnetic composite catalysts men-
tioned in the following text were prepared under the optimumcondi-
tions. This magnetic composite catalyst can lead to a higher product
yield 98.71% in the transesterication of biodiesel.
3.2. Determination of recovery rate of catalysts
After the transesterication reaction, the product mixtures with
pure calcium aluminate catalyst or magnetic composite catalyst
were respectively put in 2200 gauss magnetic eld for 0.5 h. Then
the upper layer liquid was taken out and the remaining magnetic
mixture was washed with isopropanol three times, alcohol two
Fig. 1. Effect of n (Ca): n (Fe) on FAME yield. Calcining temperature 600 C; calcining
time 6 h.
Fig. 2. Effect of calcining temperature on FAME yield. Molar ratio of Ca to Fe 5:1; calcining
time 6 h.
Fig. 3. Effect of calcining time on FAME yield. Molar ratio of Ca to Fe 5:1 and calcining
temperature 600 C.
86 S.K. Tang et al. / Fuel Processing Technology 95 (2012) 8489
times and then dried. The results showed that the recovery rate of
pure calcium aluminate catalyst was 80.81% while that of Ca/Al/
Fe
3
O
4
magnetic composite catalyst was higher to 93.80%, which was
also higher than that of Ca(OH)
2
Fe
3
O
4
catalyst (91.45%) reported
in Ref. [30].
3.3. Reusability tests of catalysts
Catalyzed by Ca/Al/Fe
3
O
4
magnetic composite catalyst or pure cal-
cium aluminate catalyst, transesterication of rapeseed oil with
methanol was respectively conducted several times under the same
reaction conditions. The experiment continued for 3 h for every run
and the tests repeated 5-times. Fig. 4 showed that after 5-times use,
the FAME yield reached to 93.53% and the recovery rate kept about
87% for Ca/Al/Fe
3
O
4
magnetic composite catalyst. In comparison, the
FAME yield decreased to 86.02% after 5-times and the recovery rate
for pure calcium aluminate catalyst was less than 70%. It indicated
that the Ca/Al/Fe
3
O
4
magnetic composite catalyst can be easily
recycled with a little loss by magnetic eld and can maintain higher
catalytic activity and higher recovery even after being used 5 times
than pure calcium aluminate catalyst.
The change of reaction rate with reaction time at the ve recycles
is given in Fig. 5. It can be found that the FAME yield decreased with
recycling times due to the loss of catalytic active components during
the reaction and separation processes. Furthermore, the reaction rate
kept fast in cycles 1 and 2 and high FAME yield can reach after 2 h.
However, the reaction rate gradually decreased with the further recycle
and high yield can be achieved after 3 h.
3.4. Catalyst characterization
3.4.1. Thermogravimetric analysis of catalyst precursor
The TG proles of the catalyst precursor with/without magnetic
nanoparticles are shown in Fig. 6. Compared the thermogravimetric
analysis of pure calcium aluminate catalyst precursor (curve a) with
Ca/Al/Fe
3
O
4
magnetic composite catalyst precursor (curve b), there
both existed four mass losses at 60120 C, 190280 C, 350450 C
and 610690 C respectively. The rst mass loss corresponded to
the loss of absorbed water at precursor surface. The second mass
loss at 190280 C was extremely clear, which assigned to the loss
of crystal water of calcium aluminate. The mass loss of constitutional
water with the form of OH

at 350450 C resulted in the change of

the crystal structure from Ca
3
Al
2
(OH)
12
to Ca
12
Al
14
O
33
and CaO,
and the mass loss at 610690 C was due to the catalyst sintering.
The results of the thermal analysis suggested that the proper cal-
cining temperature of magnetic solid base catalyst was in the
range of 450610 C. In this work, the calcining temperature was
studied from 450 C to 800 C at the molar ratio of Ca to Fe of
5:1 and calcining time of 6 h. And the optimum calcining temper-
ature was 600 C.
3.4.2. XRD
The XRD patterns of pure calcium aluminate catalyst, as-
synthesized Ca/Al/Fe
3
O
4
catalyst, as-calcined Ca/Al/Fe
3
O
4
catalyst
and commercial grade Fe
3
O
4
are shown in Fig. 7. By X-ray diffraction
analysis, as-synthesized Ca/Al/Fe
3
O
4
catalyst had the characteristic
peaks of Ca
3
Al
2
(OH)
12
cubic crystal because the position and rela-
tive intensity of the main peaks matched those from the JCPDS card
(24-0217) for Ca
3
Al
2
(OH)
12
well. After calcined at 600 C, new
phases were observed in the corresponding XRD pattern. The peaks
of 2 values at 18.1, 27.8, 29.7, 33.3, 35.0, 41.1, 46.5, 57.1, 66.8, 71.8
were the characteristic peaks of mayenite (Ca
12
Al
14
O
33
) crystal
(JCPDS Card No. 48-1882) and peaks of 2 values at 32.2, 37.3, 53.9
were indexed as cubic CaO (JCPDS Card No. 37-1497). The main
phase of as-calcined magnetic composite catalyst was mayenite
(Ca
12
Al
14
O
33
). Seen from Fig. 7, the as-calcined magnetic composite
catalyst had the characteristic peaks of Ca
12
Al
14
O
33
, CaO and Fe
3
O
4
(marked with arrows in Fig. 7).
Fig. 4. Recovery rate and lifetime comparison of pure calcium aluminate catalyst and
Ca/Al/Fe
3
O
4
magnetic composite catalyst.
Fig. 5. FAME yield with reaction time at ve recycles catalyzed by Ca/Al/Fe
3
O
4
magnetic
composite catalyst.
Fig. 6. The TG curves of pure calcium aluminate catalyst precursor (a) and Ca/Al/Fe
3
O
4
magnetic composite catalyst precursor (b).
87 S.K. Tang et al. / Fuel Processing Technology 95 (2012) 8489
In XRD patterns of as-synthesized magnetic catalyst, the ve dif-
fraction peaks have been assigned to the characteristic reections of
Fe
3
O
4
. It can be concluded that Fe
3
O
4
nanoparticles have been em-
bedded in/on the as-synthesized catalyst. The pattern of the as-
calcined magnetic catalyst not only had the characteristic peaks of
Fe
3
O
4
, Ca
12
Al
14
O
33
and CaO, but also appeared new characteristic
peaks of -Fe
2
O
3
at 2=38.1, 54.2. It suggested that parts of
Fe
3
O
4
have been oxidized to -Fe
2
O
3
after calcination at 600 C for
6 h.
Moreover, according to Scherer formula D=0.89/cos, the av-
erage particle diameter of pure calcium aluminate catalyst and as-
calcined Ca/Al/Fe
3
O
4
magnetic catalyst were calculated to be
47.42 nm and 31.51 nm respectively. It suggested that the composite
catalyst supported by Fe
3
O
4
nanoparticles had smaller grain size.
3.4.3. SEM
The SEM images of the calcined catalysts with/without magnetic
nanoparticles are shownin Fig. 8. Seenin Fig. 8a), the calciumaluminate
catalyst was formed by large particles because of the catalyst agglomer-
ation at high temperature. However, the magnetic composite catalyst
showed some differences in particle size and morphology with calcium
aluminate catalyst as observed in Fig. 8b). It was obvious that the
particles of the magnetic composite catalyst were smaller andmore reg-
ular than those of calcium aluminate catalyst and less agglomeration
happened in the magnetic catalyst. It's speculated that the addition of
Fe
3
O
4
magnetic nanoparticles has effects in decreasing the agglomera-
tion in some degree.
3.4.4. BET
By BET measurement, the specic surface area of the magnetite
was 62.21 m
2
/g and that of pure calcium aluminate catalyst was
21.29 m
2
/g. Furthermore, the BET specic surface area of magnetic
composite catalyst prepared under the optimum condition was
25.89 m
2
/g, which was larger than that of pure catalyst attributed to
the addition of Fe
3
O
4
nanoparticles with large specic surface area.
The larger specic surface area of the magnetic composite catalyst
can give a reasonable explanation why the higher catalytic activity
and product yield in the transesterication reaction of biodiesel
have been achieved with this magnetic catalyst mentioned in
Section 3.1.
3.4.5. EDS
EDS spectra analysis for the magnetic composite catalyst showed
that the catalyst consists of Ca, O, Al and Fe elements. According to
the EDS analysis (seen in Table 1), the molar ratio of Ca to Fe was
15:1, which was higher than the actual ratio of Ca to Fe (5:1) in the
preparation. It indicated that not all the Fe
3
O
4
nanoparticles used
have been embedded in/on the catalyst. Furthermore, not uniform
distribution of Fe
3
O
4
nanoparticles in the catalyst maybe another
reason for the discordance between the analysis data and experi-
mental data.
3.4.6. VSM
Fig. 9 shows magnetic hysteresis loops of the Fe
3
O
4
particles and
magnetic composite catalyst measured at room temperature by
VSM. The Fe
3
O
4
particles exhibited superparamagnetic behavior and
the saturation magnetization was about 57.85 emu/g. In comparison,
the saturation magnetization of the Ca/Al/Fe
3
O
4
composite catalyst
was about 6.34 emu/g, which was higher than that of Ca(OH)
2
Fe
3
O
4
catalyst (b1 emu/g) reported in Ref. [30]. This was just the reason
why the recovery rate of our catalyst was higher than the data in this
reference. With the aid of the forcing magnetic eld, for example, a
solid magnet, this magnetic composite catalyst developed in our study
can be separated more easily and completely.
Fig. 7. XRD patterns of pure calcium aluminate catalyst, as-synthesized Ca/Al/Fe
3
O
4
magnetic catalyst, as-calcined Ca/Al/Fe
3
O
4
magnetic catalyst and commercial grade
Fe
3
O
4
.
Fig. 8. SEM image of pure calcium aluminate catalyst (a) and Ca/Al/Fe
3
O
4
magnetic composite catalyst (b).
88 S.K. Tang et al. / Fuel Processing Technology 95 (2012) 8489
4. Conclusion
In summary, calcium aluminate solid base catalyst was developed
and furthermore a magnetic composite catalyst was prepared by
loading calcium aluminate onto Fe
3
O
4
nanoparticles with a chemical
synthesis method under the proper conditions of the molar ratio of
Ca to Fe 5:1, calcining temperature 600 C and calcining time 6 h.
XRD showed the composite catalyst has the characteristic peaks of
Ca
12
Al
14
O
33
, CaO and Fe
3
O
4
. Comparing the two catalysts without/
with magnetic nanoparticles, the Ca/Al/Fe
3
O
4
magnetic composite
catalyst had smaller size and higher specic surface area. As a result,
the magnetic composite catalyst had higher catalytic activity and
can lead to a higher product yield 98.71% in the transesterication
of biodiesel. Furthermore, it was more convenient and thorough to
separate the prepared Ca/Al/Fe
3
O
4
magnetic composite catalyst
from the reactant mixture with a solid magnet, so the recovery rate
of the catalyst was higher than that of the calcium aluminate catalyst.
After 5-times use, the biodiesel yield was more than 90% and the re-
covery rate was about 87% for the Ca/Al/Fe
3
O
4
magnetic composite
catalyst.
The addition of magnetic nanoparticles helps the dispersion of the
solid catalyst and supplies more sufcient contact area between the
reactants and catalyst. Besides, the magnetism of the catalyst helps
the separation from the reactant mixture. This work gives a signi-
cant approach to the application of magnetic nanoparticles and can
be attempted to extend to other catalyst systems.
Acknowledgment
This project was nancially sponsored by the Scientic Research
Foundation for the Returned Overseas Chinese Scholars, State Education
Ministry, by the National Natural Science Foundation of China
(20776107) and Innovation funding of Tianjin University.
References
[1] F. Ma, M.A. Hanna, Biodiesel production: a review, Bioresource Technology 70
(1999) 115.
[2] Y.C. Sharma, B. Singh, S.N. Upadhyay, Advancements in development and charac-
terization of biodiesel: a review, Fuel 87 (2008) 23552373.
[3] Y.C. Bak, J.H. Choi, S.B. Kim, D.W. Kang, Production of bio-diesel fuels by transes-
terication of rice bran oil, Korean Journal of Chemical Engineering 13 (1996)
242245.
[4] B. Freedman, E.H. Pryde, T.L. Mounts, Variables affecting the yields of fatty esters
from transesteried vegetable oils, Journal of the American Oil Chemists' Society
61 (1984) 3843.
[5] A. Brito, M.E. Borges, N. Otero, Zeolite Y as a heterogeneous catalyst in biodiesel
fuel production from used vegetable oil, Energy & Fuels 21 (2007) 32803283.
[6] M.L. Granados, M.D.Z. Poves, D.M. Alonso, R. Mariscal, F.C. Galisteo, R.M. Tost, J.
Santamaria, J.L.G. Fierro, Biodiesel from sunower oil by using activated calcium
oxide, Applied Catalysis B: Environmental 73 (2007) 317326.
[7] M. Kouzu, T. Kasuno, M. Tajika, Y. Sugimoto, S. Yamanaka, J. Hidaka, Calciumoxide
as a solid base catalyst for transesterication of soybean oil and its application to
biodiesel production, Fuel 87 (2008) 27982806.
[8] S.M. Son, K. Kusakabe, Transesterication of sunower oil in a countercurrent
trickle-bedreactor packedwith a CaOcatalyst, Chemical Engineering and Processing
50 (2011) 650654.
[9] A. Kawashima, K. Matsubara, K. Honda, Acceleration of catalytic activity of calci-
um oxide for biodiesel production, Bioresource Technology 100 (2009)
696700.
[10] X.J. Liu, H.Y. He, Y.J. Wang, S.L. Zhu, Transesterication of soybean oil to biodiesel
using SrO as a solid base catalyst, Catalysis Communications 8 (2007)
11071111.
[11] T. Wan, P. Yu, S.K. Gong, Q. Li, Y.B. Luo, Application of KF/MgO as a heterogeneous
catalyst in the production of biodiesel fromrapeseed off, Korean Journal of Chemical
Engineering 25 (2008) 9981003.
[12] E.S. Umdu, M. Tuncer, E. Seker, Transesterication of Nannochloropsis oculata
microalga's lipid to biodiesel on Al
2
O
3
supported CaO and MgO catalysts, Biore-
source Technology 100 (2009) 28282831.
[13] L.C. Meher, M.G. Kulkarni, A.K. Dalai, S.N. Naik, Transesterication of karanja
(Pongamia pinnata) oil by solid basic catalysts, European Journal of Lipid Science
and Technology 108 (2006) 389397.
[14] A.C. Alba-Rubio, J. Santamaria-Gonzalez, J.M. Merida-Robles, R. Moreno-Tost, D.
Martin-Alonso, A. Jimenez-Lopez, P. Maireles-Torres, Heterogeneous transesterica-
tion processes by using CaO supported on zinc oxide as basic catalysts, Catalysis
Today 149 (2010) 281287.
[15] H.J. Kim, B.S. Kang, M.J. Kim, Y.M. Park, D.K. Kim, J.S. Lee, K.Y. Lee, Transesterica-
tion of vegetable oil to biodiesel using heterogeneous base catalyst, Catalysis
Today 93 (2004) 315320.
[16] L.F. Cui, G.M. Xiao, B. Xu, G.Y. Teng, Transesterication of cottonseed oil to bio-
diesel by using heterogeneous solid basic catalysts, Energy & Fuels 21 (2007)
37403743.
[17] D.G. Cantrell, L.J. Gillie, A.F. Lee, K. Wilson, Structure reactivity correlations in
MgAl hydrotalcite catalysts for biodiesel synthesis, Applied Catalysis A: General
287 (2005) 183190.
[18] W. Xie, H. Peng, L. Chen, Calcined MgAl hydrotalcites as solid base catalysts for
methanolysis of soybean oil, Journal of Molecular Catalysis A: Chemical 246
(2006) 2432.
[19] C.C.C.M. Silva, N.F.P. Ribeiro, M.M.V.M. Souza, D.A.G. Aranda, Biodiesel production
from soybean oil and methanol using hydrotalcites as catalyst, Fuel Processing
Technology 91 (2010) 205210.
[20] L. Gao, B. Xu, G. Xiao, J. Lv, Transesterication of palmoil with methanol to biodiesel
over a KF/hydrotalcite solid catalyst, Energy & Fuels 22 (2008) 35313535.
[21] L.J. Gao, G.Y. Teng, J.H. Lv, G.M. Xiao, Biodiesel synthesis catalyzedby the KF/CaMgAl
hydrotalcite base catalyst, Energy & Fuels 24 (2010) 646651.
[22] T. Wan, P. Yu, S.G. Wang, Y.B. Luo, Application of sodium aluminate as a heteroge-
neous base catalyst for biodiesel production from soybean oil, Energy & Fuels 23
(2009) 10891092.
[23] A.L. Cardoso, S.C.G. Neves, M.J. da Silva, Kinetic study of alcoholysis of the fatty
acids catalyzed by tin chloride(II): an alternative catalyst for biodiesel produc-
tion, Energy & Fuels 23 (2009) 17181722.
[24] X. Liu, X. Piao, Y. Wang, S. Zhu, Calcium ethoxide as a solid base catalyst for the
transesterication of soybean oil to biodiesel, Energy & Fuels 22 (2008)
13131317.
[25] M.J. Campos-Molina, J. Santamaria-Gonzalez, J. Merida-Robles, R. Moreno-Tost,
M.C.G. Albuquerque, S. Bruque-Gamez, E. Rodriguez-Castellon, A. Jimenez-Lopez, P.
Maireles-Torres, Base catalysts derived from hydrocalumite for the transesterica-
tion of sunower oil, Energy & Fuels 24 (2010) 979984.
[26] H.B. Ma, S.F. Li, B.Y. Wang, R.H. Wang, S.J. Tian, Transesterication of rapeseed oil
for synthesizing biodiesel by K/KOH/-Al
2
O
3
as heterogeneous base catalyst,
Journal of the American Oil Chemists' Society 85 (2008) 263270.
[27] M. Ying, G.Y. Chen, Study on the production of biodiesel by magnetic cell biocatalyst
based on lipase-producing Bacillus subtilis, Applied Biochemistry and Biotechnology
137 (2007) 793803.
[28] W.L. Xie, N. Ma, Immobilized lipase on Fe
3
O
4
nanoparticles as biocatalyst for
biodiesel production, Energy & Fuels 23 (2009) 13471353.
[29] S.Y. Hu, Y.P. Guan, Y. Wang, H.Y. Han, Nano-magnetic catalyst KF/CaOFe
3
O
4
for
biodiesel production, Applied Energy 88 (2011) 26852690.
[30] C. Liu, P.M. Lv, Z.H. Yuan, F. Yan, W. Luo, The nanometer magnetic solid base catalyst
for production of biodiesel, Renewable Energy 35 (2010) 15311536.
Table 1
EDS analysis of Ca/Al/Fe
3
O
4
magnetic composite catalyst.
Element Wt.% At%
OK 24.75 41.89
AlK 24.64 24.73
CaK 46.26 31.26
FeK 4.35 2.11
Matrix Correction ZAF
Fig. 9. Magnetic hysteresis loops of commercial grade Fe
3
O
4
and Ca/Al/Fe
3
O
4
magnetic
composite catalyst.
89 S.K. Tang et al. / Fuel Processing Technology 95 (2012) 8489