Preparation, characterization and application of heterogeneous solid base catalyst for biodiesel production from soybean oil

Preparation, characterization and application of
heterogeneous solid base catalyst for biodiesel production
from soybean oil
Yihuai Li
a
, Fengxian Qiu
a,
*, Dongya Yang
a
, Xiaohua Li
b
, Ping Sun
b
a
School of Chemistry and Chemical Engineering, Jiangsu University, Xuefu Road 301, 212013 Zhenjiang, PR China
b
Jiangsu Provincial Key Laboratory of Power Machinery and Application of Clean Energy, 212013 Zhenjiang, PR China
article info
Article history:
Received 24 September 2010
Received in revised form
22 February 2011
Accepted 4 March 2011
Available online 24 March 2011
Keywords:
Heterogeneous catalyst
Transesterification
Biodiesel
Potassium hydroxide
Neodymium oxide
abstract

A solid base catalyst was prepared by neodymium oxide loaded with potassium hydroxide
and investigated for transesterification of soybean oil with methanol to biodiesel. After
loading KOH of 30 wt.% on neodymium oxide followed by calcination at 600

C, the catalyst
gave the highest basicity and the best catalytic activity for this reaction. The obtained
catalyst was characterized by means of X-ray diffraction (XRD), Fourier transform infrared
spectroscopy (FTIR), Scanning electron microscopy (SEM), Thermogravimetric analysis
(TGA), N
2
adsorptionedesorption measurements and the Hammett indicator method. The
catalyst has longer lifetime and maintained sustained activity after being used for five
times, and were noncorrosive and environmentally benign. The separate effects of the
molar ratio of methanol to oil, reaction temperature, mass ratio of catalyst to oil and
reaction time were investigated. The experimental results showed that a 14:1 M ratio of
methanol to oil, addition of 6.0% catalyst, 60

C reaction temperature and 1.5 h reaction
time gave the best results and the biodiesel yield of 92.41% was achieved. The properties of
obtained biodiesel are close to commercial diesel fuel and is rated as a realistic fuel as an
alternative to diesel.
ª 2011 Elsevier Ltd. All rights reserved.
1. Introduction
As conventional fuels are diminishing and environmental
pollution is aggravating, alternative fuels have gained signifi-
cant attention [1]. Biodiesel fuel, as a promising alternative
diesel fuel to conventional fossil diesel produced by a catalytic
transesterification of vegetable oils, animal fats and waste
cooking oils with short chain alcohol, is becoming a favorable
biofuel in many regions of the world [2,3], Compared to

conventionaldiesel from petroleum, biodiesel is technicallyand
economically more competitive because of its renewability,
biodegradability, low emission profiles, high Flash point,
excellent lubricity and superior cetane number [4]. In addition,
the use of biodiesel has the potential to reduce both the levels of
pollutants and potential or probable carcinogens [5].
Biodiesel can be produced through transesterification of
vegetable oils and fats with methanol in the presence of a
suitable catalyst. In conventional homogeneous method of
fatty acid methyl ester (FAME) synthesis, the removal of cata-
lysts after reaction is unwanted step of biodiesel synthesis,
where a large amount of wastewater is produced during
neutralization the catalyst (NaOH or KOH) and FAME washing
during separation from side products (glycerol, salt). Acid-
catalyzed process often uses sulfonic acid and hydrochloric
* Corresponding author. Tel.: þ86 51188791800.
E-mail address: (F. Qiu).
Available at www.sciencedirect.com
/>biomass and bioenergy 35 (2011) 2787e2795
0961-9534/$ e see front matter ª 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.biombioe.2011.03.009
acid as catalysts, however, the reaction time is very long
(48e96 h) even at reflux of methanol, and a high molar ratio of
methanol to oil is needed (30e150:1) [6].
Compared with homogeneous catalysts, heterogeneous
catalysts can provide green and recyclable catalytic systems
[7,8]. The advantage of heterogeneous catalyst usage is its fast
and easy separation from the reaction mixture without
requiring the use of neutralization agent. There are many solid
heterogeneous acid- andalkali-catalysts for biodiesel synthesis.

Tungstated zirconia (WO
3
/ZrO
2
) was prepared by method of
impregnation was a promising heterogeneous acid catalyst [9].
Various carbohydrate-derived and a carbon-based solid acid
catalyst [10,11] have good catalytic activity to high free fatty
acid-containing waste oils. Unfortunately, the performances of
these acid catalysts are still inferior compared with the base
catalysts. For thisreason, a widevariety of solid bases havebeen
examined for transesterification reactions for biodiesel
synthesis. Example include CaO [12],SrO[13],KNO
3
loaded on
flyash [14],ZnOeLa
2
O
3
[15] and zinc aluminate [16].But,these
heterogeneous catalysts require a high temperature to achieve
the high conversion. Other heterogeneous base catalysts like
CaMnO
3
[17],KNO
3
/Al
2
O
3

[18],andMgeAlhydrotalcites[19]have
also demonstrated some potential for activity in production of
biodiesel. However, these catalysts need more time (more than
3 h) to reach the higher biodiesel yield. The result will increase
the production cost due to the requirements for high tempera-
tures and a long time operation.
Neodymium oxide (Nd
2
O
3
) or rare earth sesquoxides is
widely used in various applications such as photonic, lumi-
nescent materials, catalyst for automotive industry, UV
absorbent, glass-polishing materials, and protective coatings.
However, in this work, a new type of catalyst for biodiesel
synthesis withKOH asactive componenton neodymium oxide
support was synthesized using the way of impregnation, and
reported the activity and selectivity of the basic solids for the
transesterification of soybean oil with methanol. A screening
of the reaction conditions has been carried out by examining
the effectof theconcentration ofcatalyst, theinitial methanol/
oil, catalyst/oil molar ratio, reaction temperature and time.
2. Experimental
2.1. Materials
Soybean oil was purchased from Jinlongyu Company (Fujian,
China). Methanol, zirconium dioxide (ZrO
2
), titanium dioxide
(TiO
2

), alumina (Al
2
O
3
), neodymium oxide (Nd
2
O
3
), potassium
hydroxide (KOH), potassium iodide (KI), potassium bromate
(KBrO
3
), potassium hydrogen phthalate (C
8
H
5
O
4
K) and potas-
sium nitrate (KNO
3
) were obtained from Sinopharm Chemical
Reagent Co. Ltd., (Shanghai, China). All solvents were AR
grade and were used without purification.
2.2. Preparation of catalyst
All the catalysts were prepared by incipient wetness impreg-
nation of different porous medium supports with solution of
potassium compounds. For this purpose, the required amount
of aqueous KOH solution was slowly added to the support and
kept 24 h. The catalytic carrier was previously calcined in

a mufflefor12hat600

C.Afterimpregnation,thecatalysts were
dried for12hat 100

C andthenthe solid was calcinedinamuffle
furnace at designed temperature for 12 h before use for the
reaction.
2.3. Characterization of the catalyst
FT-IR spectra of the samples were obtained between 4000 and
400 cm
À1
on a KBr powder with an FTIR spectrometer (AVATAR
360, Nicolet, Madison, USA). A minimum of 32 scans was signal-
averaged with a resolution of 2 cm
À1
in the 4000e400-cm
À1
range.
Scanning electron microscopy (SEM) images were obtained
with 20-kV accelerating voltage with a field emission scanning
electron microscope (S-4800, HITACHI Corp., Tokyo, Japan).
X-ray diffraction (XRD) patterns of selected samples were
obtained were recorded by the reflection scan with nickel-
filtered Cu Ka radiation (D8, Bruker-AXS, Germany). The X-ray
generator was run at 40 kV and 70 mA. All the XRD measure-
ments were performed at 2q values between 10 and 80

.
Thermogravimetric analysis (TGA) was performed on a Netz-

sch instrument ( STA 449C, Netzsch, S eligensta dt, Germa ny). The
programmed heating range was from room temperature to
1300

C, at a heating rate o f 10

C/min under a nitrogen atmo-
sphere. The measurement was taken using 6e10 mg samples.
The nitrogen adsorption and desorption isotherms were
measured at À196

C using a NDVA2000e analytical system
made by Quntachrome Corporation (USA). The specific surface
area was calculated by Brunauer-Emmett-Teller (BET) method.
Pore size distribution and pore volume were calculated by
Barrett-Joyner-Halenda (BJH) method.
Hammett indicator experiments were conducted to deter-
mine the basic strength of each catalyst. The Hammett indica-
tors used were bromothymol blue (pKa ¼ 7.2), phenolphthalein
(pKa ¼ 9.8), 2,4-dinitroaniline (pKa ¼ 15), and 4-nitroaniline
(pKa ¼ 18.4). Typically, 300 mg of the catalyst was mixed with
1 mL of a solution of Hammett indicators diluted in 10 mL
methanol and allowed to sit for at least 2 h. After the equili-
bration,thecolor of thecatalystwasnoted. The basicstrength of
the catalyst was taken to be higher than the weakest indicator
that underwent a color change and lower than the strongest
indicator that underwent no color change. To measure the
basicity of solid bases, the method of Hammett indicator-
benzene carboxylic acid (0.02 mol/L anhydrous ethanol solu-
tion) titration was used.

2.4. Transesterification of soybean oil and chemical
analyses
The transesterification reactions were performed at 60

Cin
a 125 ml three-neck reaction flask equipped with a condenser
by refluxing 10 mL of methanol (247 mmol) with 15.82 g of
soybean oil (commercial edible grade, acid value ¼ 0.976 mg
KOH/g, saponification index ¼ 188.6 mg KOH/g, and average
molecular weight ¼ 896.88 g/mol)and 0.95g of catalyst(6 wt.%).
The catalyst was activating at 773 K for 12 h before use for the
reaction. After the reaction completion, the samples were
separated from catalyst and glycerol by centrifuge. The glyc-
erol could be separated because it was insoluble in the esters
biomass and bioenergy 35 (2011) 2787e27952788
and had a much higher density. Then methanol was removed
using rotary evaporation and the obtained product was
analyzed by gas chromatography (GC) to determine the bio-
diesel yield (fatty acid methyl ester, FAME).
Reference materials and samples were analyzed by a 7890A
gas chromatograph (Agilent Technology Inc. USA), equipped
with a flame-ionization detector (FID) and a HP-5 capillary
column (30 m  0.32 mm  0.25 mm). Helium was used as the
carrier gas. The oven temperature ramp program was 135

Cfor
10 min, 170

Cat10


C/min, and held for 10 min. 250

Cat25

C/
min, and held 2 min. The flow rate of hydrogen was 30 ml/min
and the flow rate of air was about 400 ml/min. Temperatures of
the injector and detector were 280 and 300

C. The injection was
performed in splitmode with a split ratioof100:1. Biodieselyield
was quantified in the presence of tricaprylin as an internal
standard. The analysis of biodiesel for each sample was carried
out by dissolving 1 ml of biodiesel sample into 5 ml of n-hexane
and injecting 0.5 ml of this solution in GC, in the same condition
described as above. The biodiesel yield was calculated from the
content of methyl esters analyzed by GC with the following
equation:
Biodiesel yield ¼
m
tricaprylin
A
B
f
tricaprylin
A
tricaprylin
m
s
 100% (1)

where m
tricaprylin
¼ weight of the internal standard, A
B
¼ peak
area of FAME, f
tricaprylin
¼ response factor, A
tricaprylin
¼ peak
area of the internal standard, and m
s
¼ weight of the sample.
Determination of sulfur content of biodiesel was measured
by Inductively Coupled Plasma Emission Spectrometer (ICP)
using Intrepid XP Radial ICP-OES (VISTA-MPX, Varian, USA)
with a concentric nebulizer and CCD detectors technology.
Flash point was determined by a closed-cup tester (BF-02,
Dalian NorthAnalytical InstrumentsCo., Ltd.),using ASTM D 93.
3. Results and discussion
3.1. Screening of catalyst
The catalytic activity screening of Nd
2
O
3
loaded with different
potassium compounds (KOH,KI, KBrO
3
,C
8

H
5
O
4
KorKNO
3
)inthe
soybean oil transesterification was performed. The results were
summarized Table 1. To make direct comparisons, the same
reaction conditions, as shown in Table 1,wereemployedfor
each catalyst in all experiments. The reaction conditions were
not optimized for the highest reaction yield; however, they
provided a way to compare the activities of the catalysts. Obvi-
ously, it is observed from Table 1 that pure Nd
2
O
3
and KOH
catalyst exhibited no activity and serious saponification
phenomenon due to excess KOH amount [20], respectively.
However, when potassium compounds were loaded on Nd
2
O
3
and activated at high temperatures, the supported catalysts
except KNO
3
/Nd
2
O

3
showed catalytic activities. Thus, it is
essential to support potassium compounds on Nd
2
O
3
to
generate the catalytic activities for the transesterification reac-
tion. Among the catalysts tested, Nd
2
O
3
loaded with KOH, KI,
KBrO
3
or C
8
H
5
O
4
K exhibited comparatively high activities,
giving biodiesel yields higher than 80%. Especially, KOH/Nd
2
O
3
demonstrated the superior catalytic activity compared to the
other catalysts. When the transesterification was conducted
overtheKOH/Nd
2

O
3
catalyst,the highest biodieselyield of 89.7%
was achieved. Over KBrO
3
/Nd
2
O
3
, KI/Nd
2
O
3
and C
8
H
5
O
4
K/Nd
2
O
3
catalysts, however, the lower biodiesel yields in the range of
17.05e74.06% were obtained, attributable to their relatively low
catalytic activities. Based on these results, the catalytic activity
is in the following order: KOH/Nd
2
O
3

> KBrO
3
/Nd
2
O
3
> C
8
H
5
O
4
K/
Nd
2
O
3
> KI/Nd
2
O
3
> KNO
3
/Nd
2
O
3
.
The base strengths of Nd
2

O
3
modified with different potas-
sium compounds were measured by using Hammett indica-
tors. As evident in Table 1, loading of KBrO
3
,orC
8
H
5
O
4
Konthe
surface of Nd
2
O
3
generated the weaker basic sites with pK
BHþ
in
the range of 7.2e9.8. Taking both the base strength and the
catalytic activity into account, we can conclude that the
observed activities of Nd
2
O
3
-supported catalysts seem to be
related to their base strengths, i.e. the higher base strengths of
the catalysts result in the higher conversions. In particular, the
KI/Nd

2
O
3
or KNO
3
/Nd
2
O
3
sample possessed the weakest base
strength in the range of pK
BHþ
<7.2, consequently exhibiting
weak or no catalytic activity. As for the catalytic sites on KOH/
Nd
2
O
3
sample, it can also be proposed that the K
2
Ospecies,
which was possibly formed by dehydroxylation of the OH
groups, was at least a part of catalytically active sites. As
remarked above, it seems that the transesterification reaction
needs strongly basic sites.
The effect of supports on the activity of the catalyst was
listed in Table 2. Obviously, when KOH was supported on
Table 1 e Catalytic activity and base strength of Nd
2
O

3
loaded with different potassium compounds.
Catalyst Basic strength
(pK
BHþ
)
Biodiesel
yield (%)
Nd
2
O
3
<7.2 No reaction
KOH 15e18.4 Serious saponification
phenomenon
KOH/Nd
2
O
3
9.8e15 89.7
KBrO
3
/Nd
2
O
3
7.2e9.8 74.06
KI/Nd
2
O

3
<7.2 17.05
C
8
H
5
O
4
K/Nd
2
O
3
7.2e9.8 32.03
KNO
3
/Nd
2
O
3
<7.2 No reaction
Transesterification condition: methanol/oil molar ratio, 12:1; cata-
lyst amount 6 wt.%; reaction time, 3 h; reaction temperature, 60

C.
All catalysts were activating at 600

C for 12 h before use for the
reaction.
Table 2 e Catalytic activities and base strengths of KOH
supported on the different carriers.

Catalyst Basic strength (pK
BHþ
) Biodiesel yield (%)
KOH/TiO
2
9.8 < H < 15 86.47
KOH/ZrO
2
9.8 < H < 15 85.43
KOH/Al
2
O
3
9.8 < H < 15 88.87
KOH/Nd
2
O
3
9.8 < H < 15 89.70
Transesterification condition: methanol/oil molar ratio, 12:1; cata-
lyst amount, 6 wt.%; reaction time, 3 h; reaction temperature, 60

C.
All catalysts were activating at 600

C for 12 h before use for the
reaction.
biomass and bioenergy 35 (2011) 2787e2795 2789
different carriers, the base strengths were almost equality, but
the activity of the catalyst was different greatly. KOH/Nd

2
O
3
was the most active catalyst for the transesterification reac-
tion, giving a conversion of 89.70%. Over KOH/ZrO
2
, KOH/
Al
2
O
3
and KOH/TiO
2
catalysts, even though they possessed
different centers of a base strength, the high biodiesel yields of
85.43%, 88.87% and 86.47% were also achieved, respectively.
Thus, Nd
2
O
3
can be regarded as the best support. From these
discussions, KOH/Nd
2
O
3
showed the best catalytic activity. On
account of the high activity of the catalysts in the trans-
esterification reaction, KOH/Nd
2
O

3
was, therefore, selected for
further investigation and its properties were studied in more
detail.
A 14 wt.% (KOH to Nd
2
O
3
weight ratio) KOH on Nd
2
O
3
support was prepared by an impregnation method and the
following procedure: a solid Nd
2
O
3
support (25 g) was mixed
with KOH (3.5 g) in 15 mL of water, and the resulting solid was
dried in an oven at 90

C for 24 h. The solid was then crushed
and calcined in air at 600

C for 12 h. Similarly, 17, 25, 30 and
32 wt.% KOH loaded Nd
2
O
3
were prepared. The effect of KOH

loading amounts on the biodiesel yield was shown in Table 3.It
can be seen from Table 3 that when the loading amount of KOH
increased from 14 wt.% to 30 wt.%, the biodiesel yield increased
from 80.47% to 89.52%. Then, the biodiesel yield decreased with
the loading amount of KOH. This is because the base strength
of catalyst increases with the loading amount of KOH. On the
other hand, the catalytic activity and activity sites also increase
with the loading amount of KOH. But, with further increase in
the amount of loaded KOH, the basicity may decrease the
surface basic sites, which resulted in a drop of the catalytic
activity towards the reaction. This is presumably due to the
coverage of surface basic sites by the excessive KOH. These
sites are inaccessible to incoming reactants when the amount
of loaded KOH exceeded 30 wt.%. Therefore, catalytic activity
and biodiesel yield decreased. On the basis of the results, the
optimum loading amount of KOH was 30 wt.%.
Moreover, the biodiesel yield of 30 wt.% KOH/Nd
2
O
3
sample
calcined at different temperatures was measured by the same
method, and the results are presented in Table 4. From the
Table 4, it can be observed that the maximum biodiesel yield,
reaching 90.02%, was obtained at a calcination temperature of
600

C. But, a low level of biodiesel yield was observed below
512 and above 700


C. Obviously, the biodiesel yield changes
with calcination temperatures parallel the changes in the
catalytic activity for the transesterification reaction.
Based on these results, the optimal preparation conditions
of the catalyst are load KOH, support Nd
2
O
3
, loading amount
30% and calcination temperature of 600

C. Therefore, 30 wt.%
KOH/Nd
2
O
3
catalyst was selected for further investigation of
transesterification of soybean oil.
3.2. Catalyst characterizations
A series of catalysts were synthesized by incipient wetness
impregnation method and calcined at 600

C. In this high
temperature, KOH decomposed into K
2
O. The powder X-ray
diffraction patterns of KOH/Nd
2
O
3

samples with various
loading amounts of KOH were presented in Fig. 1. As can be
seen, when the loading amount of KOH was 14 wt.% (curve a),
diffraction peaks (2q ¼ 27.4

, 30.9

, 40.5

, 47.6

and 57.2

)
assigned to the amorphous Nd
2
O
3
support were registered on
the XRD patterns, and only a specie such as K
2
O(2q ¼ 29.6

)
was observed, indicating the good dispersion of K
2
OonNd
2
O
3

in the form of a monolayer due to the interaction between K
2
O
and the surface of the support at a low loading of KOH. And
when the loading amount of KOH was further increased to
25 wt. % (curve b), the new phase of K
2
O can be observed at
32.1

and 51.6

. But, when the loading amount of KOH is
further increased to over 30 wt. % (curves c and d), a new
phase of a compound containing potassium and neodymium
elements could be observed at 2q ¼ 25.8

, 38.8

and 41.2

[21].
The phenomena can be a result from the incorporation of K
þ
ions into the vacancies in the structure of the neodymium
oxide, or K
þ
ions may react with hydroxyl groups to form
NdeOeK on the surface during heat treatment. The result will
confirm by FT-IR spectrum of catalyst. Moreover, the intensi-

ties of some diffraction peaks (2q ¼ 29.6

, 32.1

, 38.8

, 41.2

and
51.6

) increased with increase of the loading amount of KOH.
On the other hand, the characteristic peaks of Nd
2
O
3
(27.4

,
30.9

, 40.5

, 47.6

and 57.2

) were almost unchanged on the XRD
patterns regardless of the loading amount of KOH. It is note-
worthy that the solidestate reaction between the guest

compound and the surface of the support in the activation
process is favorable for the catalyst to get a high catalytic
Table 3 e The effect of KOH loading amounts on the biodiesel yield.
KOH loading amount (%) 14 17 25 30 32
Basic strength (pK
BHþ
) 9.8 < H < 15 9.8 < H < 15 9.8 < H < 15 9.8 < H < 15 9.8 < H < 15
Biodiesel yield (%) 80.47 81.52 85.72 89.52 73.45
Transesterification condition: methanol/oil molar ratio, 15:1; catalyst amount, 6 wt.%; reaction time, 3.0 h; reaction temperature, 60

C.
Table 4 e The effect of calcination temperature on the biodiesel yield.
Calcination temperature (

C) 311 422 512 600 700 790
Basic strength (pK
BHþ
) 9.8 < H < 15 9.8 < H < 15 9.8 < H < 15 9.8 < H < 15 9.8 < H < 15 9.8 < H < 15
Biodiesel yield (%) 78.80 81.23 81.00 90.02 77.83 73.50
Transesterification condition: methanol/oil molar ratio, 12:1; catalyst amount, 6 wt.%; reaction time, 3.0 h; reaction temperature, 60

C.
biomass and bioenergy 35 (2011) 2787e27952790
activity. In the case of KOH/Nd
2
O
3
, the K
þ
ion of KOH could

insert in the vacant sites of Nd
2
O
3
, accelerating dissociative
dispersion and decomposition of KOH to form basic sites in
the activation process. The more potassium compounds are
loaded on the Nd
2
O
3
, the more free vacancies decrease, which
results in the surface enrichment of potassium species that is
probably considered to be the active sites for base-catalyzed
reactions. When the amount of potassium cations loaded on
Nd
2
O
3
was below the saturation uptake of K
þ
, it could be well
dispersed. As a result, the number of basic sites together with
the activities of the catalysts would increase with the potas-
sium contents. However, if Nd
2
O
3
was loaded with too much
KOH, the KOH could not be well dispersed and, for this reason,

not all but only a part of the loaded KOH could be decomposed.
Moreover, as mentioned in the preceding sections, the excess
KOH would cover the basic sites on the surface of the catalysts
resulting in a lowered catalytic activity (Table 3).
SEM images of Nd
2
O
3
and 30 wt.% KOH/Nd
2
O
3
catalyst were
shown in Fig. 2. The SEM photographs of Nd
2
O
3
and KOH/
Nd
2
O
3
samples showed the crystallites of 0.2e1 mm size.
Evidently, as shown in Fig. 2, no important difference was
observed between Nd
2
O
3
and KOH/Nd
2

O
3
samples, thus sug-
gesting a good dispersion of KOH on the surface of Nd
2
O
3
.
Based on these results, after loading of KOH, Nd
2
O
3
retained
its structure that was important for catalysis and therefore the
potassium species was found highly distributed upon the
surface of the support.
FTIR spectra of Nd
2
O
3
and KOHeNd
2
O
3
catalyst were
recorded and shown in Fig. 3. The spectrum of the support
shows sharp peak at 1475 cm
À1
. From the spectrum of catalyst,
the new peaks at 880 and 706 cm

À1
are attributed to the KeO
and KeOeNd bonds, respectively. Furthermore, the broad
band at around 3100e3300 cm
À1
region could be partly
assigned to the stretching vibration of NdeOeK groups [22]
which K
þ
ions could replace the protons of isolated hydroxy
groups to form NdeOeK groups in the activation process and
were probably considered to be the active species of this cata-
lyst. This achieves the same results as the XRD analysis.
However, this vibration is well overlapped with the broad
vibration band of OH groups which is ascribed to OH stretching
vibration of thehydroxylgroupsattached to thectalystsurface,
in addition to water molecules absorbed from the atmosphere.
The BET surface area, pore volume, and pore diameter of
Nd
2
O
3
and30 wt.%KOH/Nd
2
O
3
catalystwere measured.TheBET
surface area as well as the pore volume decreased with loading
potassium hydroxide, and this tendency was more outstanding
in the case of potassium. The thermal behavior of 30 wt. % KOH/

Nd
2
O
3
sample was shown in Fig. 4. This figure showed that the
first weight loss at lower temperature (<200

C) corresponds to
the water loss from internal and external surfaces of the
samples. The second weight loss (200e400

C), is due to the
decomposition of the KOH and K
2
CO
3
. The last weight loss at
above 400

C is attributed to the decomposition of the K
2
Oand
the residual hydroxyl groups bonded to the oxide lattice. The
decomposition products of KOH, probably forming both K
2
O
speciesandNdeOeK groupsinthecomposite,werepossiblythe
main active sites for the transesterification reaction. In this
study, effect of repeated use of KOH/Nd
2

O
3
catalyst onbiodiesel
yield was investigated in the optimal transesterification reac-
tionconditions. The resultin Table 5indicated that the biodiesel
Fig. 1 e XRD patterns of (a) 14 wt.% KOH/Nd
2
O
3
, (b) 25 wt.%
KOH/Nd
2
O
3
, (c) 30 wt.% KOH/Nd
2
O
3
, (d) 32 wt.% KOH/Nd
2
O
3
.
Fig. 2 e SEM images of (a) Nd
2
O
3
and (b) 30 wt.% KOH/Nd
2
O

3
.
biomass and bioenergy 35 (2011) 2787e2795 2791
yields had no significant changes and were in excess of 90%
during the repeated experiments. It maintained sustained
activity even after being used for five times and the biodiesel
yield was only slightly decreased from 90.11% to 90.05%. This
was because neodymium oxide compounds are dissolvable in
methanol.
3.3. Influence of the transesterification reaction
conditions
The transesterification process consists of a sequence of three
consecutive reversible reactions where the triglyceride is
successively transformed into diglyceride, monoglyceride, and
finally intoglycerin andthe FAME. The molar ratio of methanol
to soybean oil is one of the important factors that affect the
conversion to methyl esters. Stoichiometrically, 3 mol of
methanol are required for each mole of triglyceride, but in
practice a higher molar ratio is employed in order to drive the
reaction towards completion and produce more methyl esters
as products. This is because that the biodiesel yield could be
improved by introducing excess amounts of methanol to shift
the equilibrium to the right-hand side. As represented in Fig. 5,
the biodiesel yields grew as the methanol-loading molar ratio
increased, and the biodiesel yield was increased considerably.
The maximum biodiesel yield (90.59%) was obtained when the
molar ratio was very close to 14:1. In comparison, the biodiesel
yield increased from 77.49% to 90.59% when the molar ratio
was increased from 6:1 to 14:1. However, beyond the molar
ratio of14:1,the excessivelyaddedmethanol had nosignificant

effect on the production yield and the biodiesel yield was
90.12% at 16:1. The reason is that the catalyst content
decreased with increase of methanol content. Therefore, we
could concludethatto elevatethe biodiesel productionyield an
excess methanol feed was effective to a certain extent and the
optimum molar ratio of methanol to oil was 14:1.
The dependence of the biodiesel yield on the reaction time
was investigated. The reaction time was varied in the range
0.5e8h.Fig. 6 revealed that the transesterification reaction
was strongly dependent on reaction time, at the beginning
(<0.5 h), the reaction was slow due to the mixing and the
dispersion of methanol into oil, and the biodiesel yield was
increased very fast in the reaction time range between 0.5 and
1.5 h. Moreover, excess reaction time leaded to a bit reduction
in the product yield due to the backward reaction trans-
esterification, resulting in a loss of esters as well as causing
more fatty acids to form soaps [23,24]. So the optimum reac-
tion time was obtained at 1.5 h.
Fig. 3 e FT-IR spectra of Nd
2
O
3
and catalyst.
Fig. 4 e Thermogravimetric analysis of 30 wt. % KOH/
Nd
2
O
3
.
Fig. 5 e Effect of different molar of methanol to oil on the

biodiesel yield (catalyst amount, 6 wt.%; reaction time,
3.0 h; reaction temperature, 60

C).
Table 5 e Effect of repeated use of KOH/Nd
2
O
3
catalyst on
biodiesel yield.
Repeated times 12345
Biodiesel yield (%) 90.11 90.07 90.06 90.04 90.05
Transesterification condition: methanol/oil molar ratio, 12:1; cata-
lyst amount, 6 wt.%; reaction time, 1.5 h; reaction temperature,
60

C.
biomass and bioenergy 35 (2011) 2787e27952792
In the presence of heterogeneous catalysts, the reaction
mixture constitutes a three-phase system, oilemethanol-cata-
lyst, inwhich the reactionwouldbe sloweddown because of the
diffusion resistance between different phases. However, the
reaction rate can be accelerated at higher reaction tempera-
tures. In this paper, the synthesis of biodiesel from soybean oil
was conducted atvarious temperatures (40

C, 50

C, 60


C, 65

C
and 70

C). As shown in Fig. 7, the reaction rate was slow at low
temperatures, but the biodiesel yield first increased and then
decreased with the increase of the reaction temperature.
Generally, a more rapid reaction rate could be obtained at high
temperatures, which is due to the endothermic nature of
transesterification reaction [25], but at high temperatures,
methanol was vaporized and formed a large number of bubbles,
which inhibited the reaction on the three-phase interface. The
optimum reaction temperature was 60

C and biodiesel yield
arrived at 92.41%.
When increasing the amount of loading catalyst, the slurry
(mixture of catalyst and reactants) was become too viscous
giving rise to a problem of mixing and a demand of higher
power consumption for adequate stirring. On the other hand,
when the catalyst loading amount was not enough, maximum
biodiesel yield could not be reached. To avoid this kind of
problem, an optimum amount of catalyst loading had to be
investigated. The influence of the catalyst amounts was
studied at a 14:1 M ratio of methanol to soybean oil at reflux of
methanol for 2 h. The catalyst amount was varied in the range
of 1.0% and 9.0%. These percentages were weight fractions of
the oil supplied for this reaction. The reaction profile of Fig. 8
indicated that the transesterification reaction was strongly

dependent upon the catalyst applied. As is evident from Fig. 8,
when the catalyst amount increased from 1.0% to 6.0%, the
biodiesel production yield was increased. However, with
further increase in the catalyst amount the biodiesel yield was
Fig. 6 e Effect of reaction time on the biodiesel yield
(methanol/oil molar ratio, 14:1; catalyst amount, 6 wt.%;
reaction temperature, 60

C).
Fig. 7 e Effect of reaction temperature on the biodiesel yield
(methanol/oil molar ratio, 14:1; catalyst amount, 6 wt.%;
reaction time, 1.5 h).
Fig. 8 e Effect of catalyst amount on the biodiesel yield
(methanol/oil molar ratio, 14:1; reaction time, 2.0 h;
reaction temperature, 60

C).
Table 6 e The various absorption peaks of biodiesel.
Wavenumber
(cm
À1
)
Group
attribution
Vibration type Absorption
intensity
3462.27 eOH Stretching Weak
3008.91 ¼ CeH Stretching Strong
2925.76 eCH
2

Asymmetric
stretching vibration
Strong
2855.00 eCH
2
Symmetric stretching
vibration
Strong
1743.54 eC]O Stretching Strong
1461.48 eCH
2
Shear-type vibration Middling
1360.75 eCH
3
Bending vibration Middling
1017.78 CeOeC Anti-symmetric
stretching vibration
Weak
1171.41 CeOeC Symmetric stretching
vibration
Middling
722.75 eCH
2
Plane rocking
vibration
Weak
biomass and bioenergy 35 (2011) 2787e2795 2793
decreased, which was possibly due to a mixing problem
involving reactants, products and solid catalyst. The optimum
catalyst loading amount was found to be 6.0% in this system

and the maximum biodiesel yield reached to 92.40%.
From the above results, the reaction does not require too
much time to dispose of the products, for example, neutrali-
zation, washing and drying. If the catalyst can be used
commercially, filtration is a possible way to recycle the cata-
lyst and decrease the cost. As a heterogeneous solid base
catalyst, the prepared KOH/Nd
2
O
3
catalyst has a longer cata-
lyst lifetime and better stability than current homogeneous
catalysts. It is noncorrosive and environmentally benign. It
can be applied to produce biodiesel commercially.
3.4. Characterization and properties of biodiesel
FT-IR spectrum of the obtained biodiesel was listed in Table 6.
From the analysis of the Table 6, we could get the sample
including all groups which we needed. At the same time, it
proved that the compound was the kind of structures having
long-chain fatty acid esters.
The content of sulfur and its proper determination play an
important role regarding fuels and products of petrochemical
industry. The problem of appropriate determination of sulfur is
important both from environmental and analytical aspects,
because some specifications order to the compulsion decrease
of the concentration of sulfur (e.g. from 2005 their maximum
concentration is 50 mg/kg in fuels in the countries of European
Union). Over the past few decades, there are numerous spec-
troscopic techniques to analyze the qualitative and quantita-
tive elemental composition of fuels. For example, inductively

coupled plasma atomic emission spectroscopy (ICPeAES),
inductively coupled plasma mass spectroscopy (ICPeMS) and
flame or graphic furnace atomic absorption spectroscopy (AAS)
were adopted. Each technique has advantageous properties in
terms of analytical figures of merit. The atomic absorption and
emission techniques are typically used for analysis of the
products of hydrocarbon industry. The ICP technique is a fast
analytical method, but needs preliminary sample preparation
(e.g. digestion). In this work, the sulfur content of biodiesel was
carried out by Inductively Coupled Plasma Emission Spec-
trometer (ICP) using Intrepid XP Radial ICP-OES (VISTA-MPX,
Varian, USA) with a concentric nebulizer and CCD detectors
technology. After performing the background equivalent
concentration experiment to test the instrument sensitivity,
the ICP operating conditions applied are presented in Table 7.
In order to determine the sulfur content, the biodiesel was
carried out by nitrification firstly. A certain amount of biodiesel
was added to concentrated hydrochloric acid (8 mL) and
concentrated nitric acid (2 mL), and placed overnight. Subse-
quently, the mixture was gently filtered and got clear liquid,
then the clear liquid was evaporated in the fuming cupboard
about 20 min. Calibration standards were made up from some
standard solutions of sulfur. The ranges of the calibration
curves (5 points) were selected to match the expected different
concentrations standard solution for the sulfur element of the
sample investigated. Linearity was checked in the range of
0e40 mg/g. From the calibration curve, the sulfur content of
biodiesel was obtained. The sulfur content of the obtained
biodiesel was listed in Table 8.
The properties of biodiesel, density, cetane number, flash

point, cold filter plugging point, acid number, water content,
ash content and total glycerol content, were determined and
listed in the Table 8. Table 8 also showed comparisons of the
obtained biodiesel and the standards of biodiesel in china,
Europe and the United States. The properties of the obtained
biodiesel, in general, show many similarities, and therefore,
Table 8 e Comparison of properties of the obtained biodiesel and the standards of biodiesel in china, Europe and the
United States.
Item Obtained biodiesel China GB/T 20828e2007 USA ASTM D 6751e03 Europe EN 14214
Density (kg L
À1
) 0.896 (20

C) 0.82e0.90 (20

C) 0.82e0.90 (20

C) 0.86e0.90 (15

C)
Flash point (

C) 168 !130 ＞130 ＞120
Cold filter plugging point (

C) À5.0 eeSpring:0
Summer:À10
Autumn:À20
Sulfur content (w/w,%) 0.0065 0.05 0.0015 ＜0.001
Cetane value 56 !49 !47 ＞51

Acid value (KOH) (mg g
À1
) 0.6 0.8 ＜0.8 ＜0.5
Water content (w/w,%) 0.04 0.05 0.05 0.05
Ash content (w/w,%) 0.018 0.05 0.02 0.02
Total glycerol content (w/w,%) 0.020 0.024 0.024 0.025
Table 7 e The properties of ICP-OES spectrometer.
Type Intrepid XP Radial ICP-OES
Nebulizer Concentric, with a cyclonic spray chamber
RF-generator 40.68 MHz crystal-controlled
Power 1200 W
Reflected power 20 W
Observation high 6 mm above the coil
Optical system Czerny-Turner vacuum-monochromator
Grating Holographic, 1800 groves/mm
Focal length 0.75m
Optical range 160e800 nm
Resolution 1st order: 0.018 nm
Detector Photomultiplier
Emission lines used l
1
¼ 180.731 nm
For the S analysis l
2
¼ 181.970 nm
biomass and bioenergy 35 (2011) 2787e27952794
the properties of obtained biodiesel from the soybean oil is
rated as a realistic fuel as an alternative to diesel.
4. Conclusions
Nd

2
O
3
loaded with KOH, which was prepared by impregnation
of powdered Nd
2
O
3
with an aqueous solution of KOH followed
by calcination at a high temperature, showed high catalytic
activities for the transesterification reaction. Both the K
2
O
species formed by the thermal decomposition of loaded KOH,
and the surface KeOeNd groups formed by saltesupport
interactions, were probably the main reasons for the catalytic
activity towards the reaction. The activities of the heteroge-
neous base catalysts correlated with their corresponding basic
properties. The catalyst with 30 wt.% KOH loading on Nd
2
O
3
and calcined at 600

C for 12 h was found to be the optimum
catalyst, which gave the best catalytic activity. When the
reaction was carried out at reflux of methanol, with a molar
ratio of methanol to oil of 14:1, a reaction time 1.5 h, a reaction
temperature 60


C and a catalyst amount 6.0%, the highest
biodiesel yield reached 92.41%. The properties of obtained
biodiesel from soybean oil are close to commercial diesel fuel
and is rated as a realistic fuel as an alternative to diesel.
Acknowledgment
This project was supported by the Natural Science of Jiangsu
Province (BK2008247), Jiangsu Provincial Key Laboratory of
Power Machinery and Application of Clean Energy Foundation
(QK08007).
references
[1] Fukuda H, Kondo A, Noda H. Biodiesel fuel production by
transesterification of oils. J Biosci Bioeng 2001;92:405e16.
[2] Marchetti JM, Miguel VU, Errazu AF. Possible methods for
biodiesel production. Renew Sust Energ Rev 2007;11:1300e11.
[3] Qiu FX, Li YH, Yang DY, LiXH, Sun P. Biodiesel production from
mixed soybean oil and rapeseed oil. Appl Energ 2011;88:2050e5.
[4] Meher LC, Kulkarni MG, Dalai AK, Naik SN.
Transesterification of karanja (Pongamia pinnata) oil by solid
basic catalysts. Eur J Lipid Sci Tech 2006;108:389e97.
[5] Singh SP, Singh D. Biodiesel production through the use of
different sources and characterization of oils and their esters
as the substitute of diesel: a review. Renew Sust Energ Rev
2010;14:200e16.
[6] Lotero E, Liu Y, Lopez DE, Suwannakarn K, Bruce DA,
Goodwin Jr JG. Synthesis of biodiesel via acid catalysis. Ind
Eng Chem Res 2005;44:5353e63.
[7] Qiu FX, Li YH, Yang DY, Li XH, Sun P. Heterogeneous solid
base nanocatalyst: preparation, characterization and
application in biodiesel production. Bioresour Technol 2011;
102:4150e6.

[8] Sakai T, Kawashima A, Koshikawa T. Economic assessment
of batch biodiesel production processes using homogeneous
and heterogeneous alkali catalysts. Bioresour Technol 2009;
100:3268e76.
[9] Park YM, Lee JY, Chung SH, Park IS, Lee SY, Kim DK, et al.
Esterification of used vegetable oils using the heterogeneous
WO
3
/ZrO
2
catalyst for production of biodiesel. Bioresour
Technol 2010;101:S59e61.
[10] Lou WY, Zong MH, Duan ZQ. Efficient production of biodiesel
from high free fatty acid-containing waste oils using various
carbohydrate-derived solid acid catalysts. Bioresour Technol
2008;99:8752e8.
[11] Shu Q, Gao J, Nawaz Z, Liao Y, Wang D, Wang J. Synthesis of
biodiesel from waste vegetable oil with large amounts of free
fatty acids using a carbon-based solid acid catalyst. Appl
Energ 2010;87:2589e96.
[12] Liu X, He H, Wang Y, Zhu S, Piao X. Transesterification of
soybean oil to biodiesel using CaO as a solid base catalyst.
Fuel 2008;87:216e21.
[13] Liu X, He H, Wang Y, Zhu S. Transesterification of soybean oil
to biodiesel using SrO as a solid base catalyst. Catal Commun
2007;8:1107e11.
[14] Kotwal MS, Niphadkar PS, Deshpande SS, Bokade VV,
Joshi PN. Transesterification of sunflower oil catalyzed by
flyash-based solid catalysts. Fuel 2009;88:1773e8.
[15] Yan S, Salley SO, Ng KYS. Simultaneous transesterification

and esterification of unrefined or waste oils over ZnO-La
2
O
3
catalysts. Appl Catal A-Gen 2009;353:203e12.
[16] Bournay L, Casanave D, Delfort B, Hillion G, Chodorge JA.
New heterogeneous process for biodiesel production: a way
to improve the quality and the value of the crude glycerin
produced by biodiesel plants. Catal Today 2005;106:190e2.
[17] Kawashima A, Matsubara K, Honda K. Development of
heterogeneous base catalysts for biodiesel production.
Bioresour Technol 2008;99:3439e43.
[18] Xie W, Peng H, Chen L. Transesterification of soybean oil
catalyzed by potassium loaded on alumina as a solid-base
catalyst. Appl Catal A-Gen 2006;300:67e74.
[19] Xie W, Peng H, Chen L. Calcined Mge Al hydrotalcites as solid
base catalysts for methanolysis of soybean oil. J Molecul
Catal A-Chem 2006;246:24e32.
[20] Helwani Z, Othman MR, Aziz N, Fernando WJN, Kim J.
Technologies for production of biodiesel focusing on green
catalytic techniques: a review. Fuel Process Technol 2009;90:
1502e14.
[21] Noiroj K, Intarapong P, Luengnar uemitchai A, Jai-In S. A
comparative study of KOH/Al
2
O
3
and KOH/NaY catalysts for
biodiesel production via transesterification from palm oil.
Renew Energ 2009;34:1145e50.

[22] Krupay BW, Amenomiya Y. Al-kali-promoted alumina
catalysts: I. Chemisorption and oxygen exchange of carbon
monoxide and carbon diox-ide on potassium-promoted
alumina catalysts. J Catal 1981;67:362e70.
[23] Eevera T, Rajendran K, Saradha S. Biodiesel production
process optimization and characterization to assess the
suitability of the product for varied environmental
conditions. Renew Energ 2009;34:762e5.
[24] Ma F, Clement LD, Hanna MA. The effects of catalyst, free
fatty acids, and water on transesterification of beef tallow.
Trans Am Soc Agric Eng 1998;41:1261e4.
[25] Wen ZZ, Yu XH, Tu ST, Yan JY, Dahlquist E. Biodiesel
production from waste cooking oil catalyzed by TiO
2
eMgO
mixed oxides. Bioresour Technol 2010;101:9570e6.
biomass and bioenergy 35 (2011) 2787e2795 2795