1
 CHARACTERIZATION AND CATALYTIC PERFORMANCE
2
ON TRANSESTERIFICATION OF PALM OLEIN OF
3
POTASSIUM OXIDE SUPPORTED ON RH-MCM-41 FROM
4
RICE HUSK SILICA
5
Surachai Artkla1, Nurak Grisdanurak2, Suthasinee Neramittagapong3,
6
Jatuporn Wittayakun1*
7
1
School of Chemistry, Institute of Science, Suranaree University of Technology,
Nakhon Ratchasima, 30000, Thailand, Tel.:0-4422-4256; Fax.: 0-4422-4185;
Email: jatuporn@sut.ac.th
2
Department of Chemical Engineering, Faculty of Engineering, Thammasat
University, Pathumthani, 12120, Thailand
3
Department of Chemical Engineering, Faculty of Engineering, King Mongkut’s
Institute of Technology Ladkrabang, Bangkok, 10520, Thailand
*
Corresponding author
1
ABSTRACT
2
Mesoporous material, RH-MCM-41, was synthesized with rice husk silica source
3
by hydrothermal method.
4
diffraction peaks of the (100), (110) and (200) planes at 2.3, 4.0 and 4.7 degree,
5
respectively; and possessed high specific surface area of 1,231 m2/g and narrow
6
pore size distribution in the range of 1.8–4.2 nm. It was used as a catalytic support
7
for
8
K2O/RH-MCM-41 with K2O loading of 4, 8 and 12%wt. Upon loading with K2O,
9
RH-MCM-41 surface area decreased significantly indicating the collapse of
10
mesoporeous structure and the tendency to collapse increased with K2O loading.
11
The of K2O/RH-MCM-41 catalysts were tested for biodiesel production from palm
12
olein oil and methanol via transesterification at 50, 75 and 100°C. The catalyst with
13
8%K2O loading gave the highest conversion at all tested temperature. At this
14
loading, the activity increased with temperature and the highest conversion was
15
84% at 100°C. Products from transesterification were mainly methyl palmitate
16
(C16) and unsaturated methyl ester (oleate linoleate and linolanate, C18).
potassium
oxide
The obtained material showed characteristic X-ray
(K2O)
from
CH3COOK
precursor
to
17
18
Keywords: RH-MCM-41, MCM-41, K2O, Transesterification, Biodiesel,
19
Heterogeneous catalysis.
20
21
produce
22
INTRODUCTION
23
MCM-41, one of the M41S families has been widely interested because of its
24
excellent properties such as large surface area (1000-1400 m2/g), well defined
25
regular pore diameter (20-100 Å), narrow pore size distribution and thermal
26
stability. Normally, MCM-41 can be synthesized by mixing silica source such as
27
sodium silicate, fumed silica or tetraethylorthosilicate with a template solution and
28
crystallized under hydrothermal conditions (Kumer et al., 2001; Bailek et al.,
29
1991). After separation of the solid crystals from solution, the organic template can
30
be removed by calcination at 540 °C to result MCM-41 (Grisdanurak et at., 2003).
31
Because of its properties, MCM-41 can be used as a support for immobilization
32
of metal catalysts. In this work, it was synthesized with rice husk silica, assigned as
33
RH-MCM-41, and further used as a support for K2O and the catalysts would be
34
referred to as K2O/RH-MCM-41. This catalyst was basic and suitable for catalysis
35
that requires basicity such as transesterification. Thus, this research focused on
36
catalysis of K2O/RH-MCM-41 for transesterification that converts triglyceride
37
from oil/fat and alcohol to alkyl esters to produce glycerol and methyl ester or
38
biodiesel (equation 1).
O
O
H 2C
O
C
H 2C
R1
O
R 'O
C
O
R1
R 'O
C
R2
b ase or
O
HC
OH
R2 +
C
3R 'O H
acid
HC
OH
+
O
CH2
O
C
O
CH2
R3
OH
R 'O
C
R3
(1)
39
40
41
Triglyceride
alcohol
Glycerol
methyl esters
42
Conventionally, the homogeneous base catalyst such as NaOH, KOH and
43
NaOCH3 are preferred because they produce high yield of the alkyl ester product,
44
the reaction time is short (about 2 hour) and the cost of raw materials is low.
45
Unfortunately, the use of base catalysts is limited only on the well-refined vegetable
46
oil with less than 2.0% free fatty acid (FFA) in order to avoid the formation of soap,
47
undesired by-product. In addition, it is difficult to separate ester products from
48
glycerol by-products in homogeneous catalysis and the yield of methyl esters is low
49
(Ma and Hanna, 1999). Heterogeneous catalysis could overcome these drawbacks.
50
Here we report the preparation of RH-MCM-41 and K2O/RH-MCM-41 and
51
characterization by XRD and BET. The catalysis of K2O/RH-MCM-41 for the
52
biodiesel production was also studied to understand the effects of reaction
53
temperature and potassium oxide content.
54
EXPERIMENTAL
55
Silica Extraction
56
Rice husk silica (RHS) was prepared by acid leaching and calcinations as
57
described by Wittayakun et al. (2007). The element analysis of white rice husk
58
silica (RH-SiO2) by X-ray fluorescence showed that the purity of silica was 98%
59
along with trace amount of alumina, potassium oxide and calcium oxide. The X-ray
60
diffraction (XRD) pattern of RH- SiO2 exhibited a broad 2 peak at 22˚ which is a
61
characteristic of amorphous silica. Its specific surface area from BET surface area
62
analysis was 234.5 m2/g.
63
Preparation of RH-MCM-41
64
RH-MCM-41 was synthesized by a procedure modified from literature (Srinivas
65
et al., 2004) with the gel molar ratio of 1.0SiO2 : 3.0NaOH : 0.25CTABr : 180H2O.
66
Briefly, cetyltrimethylammonium bromide (CTABr) and RH-SiO2 (in 3.33M
67
NaOH solution) were dissolved in distilled water and stirred for 4 hours. Its pH was
68
adjusted to 11.5 by H2SO4 and the gel mixture was crystallized in a teflon-linen
69
autoclave and annealed hydrothermically in an oven at 100˚C for 72 hours. After
70
filtration and drying at 100˚C, the powder was washed and calcined at 540˚C to
71
remove the template and characterized by XRD and BET.
72
Preparation of K2O/RH-MCM-41
73
K2O/RH-MCM-41 was prepared by impregnation method adapted from
74
literature with K2O loading of 4, 8 and 12%wt (Xie et al., 2006). The dried
75
RH-MCM-41 was mixed with CH3COOK in methanol solution and vigorously
76
stirred for 3 hours. Finally, the mixture was washed with distilled water, calcined at
77
500˚C for 5 hours and characterized by XRD and BET.
78
Characterization or RH-MCM-41 and K2O/RH-MCM-41
79
Powder X-Ray Diffraction (XRD) patterns were obtained using Cu K radiation
80
on a Bruker axs D5005 diffractometer in which the samples were scanned from 1.5
81
to 15.0 degrees (2).
82
Nitrogen adsorption-desorption isotherms were determined at -196˚C from a
83
relative pressure of 0.001 - 0.99 on an ASAP 2010 analyzer. Before measurement,
84
each sample was degassed at 300˚C for 12 hours. The pore size and pore volumes
85
were calculated from the desorption branches of the isotherm using
86
Barrett-Joyner-Halenda (BJH) method.
87
Catalytic testing for transesterification
88
Triglyceride (palm olein oil) used as a raw material for reaction consisted of
89
39.8% wt. palmitic acid (C16:0), 54.0% wt. unsaturated acids [oleic acid (C18:1),
90
linoleic acid (C18:2) and linolenic acid (C18:3)] and 4.4% wt. stearic acid (C18:0)
91
(Department of Thailand Agriculture, 2007). Palm olein oil (4.0 cm3) was
92
preheated to reaction temperature and added to a mixture between methanol (10.0
93
cm3) and catalyst (0.3 g). The mixture was stirred for 7 hours at 50, 75 or 100 ˚C.
94
During the reaction, a pale yellowish liquid was formed and the viscosity of the
95
mixture decreased significantly. The pale yellowish solution was evaporated to
96
remove the excess methanol and the resulting liquid was separated from the catalyst
97
by gravity under refrigeration. The obtained yellowish solution which contained
98
fatty acid methyl ester (FAME) was analyzed by gas chromatography on a
99
Shimadzu GC14-A equipped with an FID detector.
100
101
Results and discussion
102
Characterization of RH-MCM-41 and K2O/RH-MCM-41 by XRD and BET
103
XRD patterns of RH-MCM-41 and K2O/RH-MCM-41 with K2O loading of 4, 8,
104
and 12% wt. are shown in Figure 1. Spectrum of RH-MCM-41 showed the
105
characteristic reflections of the (100), (110) and (200) planes of hexagonal structure
106
at 2.3, 4.0 and 4.7 degree, respectively, similar to that of MCM-41 from literature
107
(Papp et al,. 2005). When RH-MCM-41 was loaded with potassium oxide, the only
108
peak observed was that of the (100) plane and the intensity decreased with the
109
amount of K2O. This indicated that mesoporeous structure of RH-MCM-41
110
collapsed upon K2O addition. The position of the (100) peak also shifted to higher
111
value with the K2O amount indicating the decrease of d spacing. The position of the
112
(100) peak, unit cell width (a*), calculated d100 are showed in Table 1. Thus, the
113
K2O loading affected the RH-MCM-41 hexagonal structure and the surface area of
114
RH-MCM-41
115
12%K2O/RH-MCM-41, the (100) peak was barely observable indicating that its
116
mesoporeous structure collapsed almost completely. Although it was still unclear
117
about
118
12%K2O/RH-MCM-41
119
12%K2O/RH-SiO2 which was less porous to observe the influence of support.
120
The
was
structure
nitrogen
expected
of
to
decrease
the
collapsed
was
investigated
adsorption-desorption
are
displayed
in
upon
K2 O
RH-MCM-41,
and
compared
isotherms
figure
the
2.
of
loading.
catalysis
with
that
For
of
of
RH-MCM-41
and
RH-MCM-41(2a)
and
121
K2O/RH-MCM-41s
122
4%K2O/RH-MCM-41(2b) gave Type-IV isotherm with three well-defined stages.
123
In the first step, the adsorption at relative pressure around 0.0-0.2, concaved to the
124
P/P0 axis due to monolayer adsorption in external surface which were large pores.
125
The lower adsorption volume on 4%K2O/RH-MCM-41 indicated lower surface
126
area. The adsorption at relative pressure of 0.2-0.4 for RH-MCM-41 and 0.4-0.8
127
for K2O/RH-MCM-41 were an adsorption in mesopores.
128
4%K2O/RH-MCM-41 were smaller than those in RH-MCM-41 and thus, required
129
higher pressure. The last step was a plateau until the relative pressure approached
130
one and the adsorption volume increased again to form condensation on the surface.
131
The isotherms of 8%K2O/RH-MCM-41(2c) and 12%K2O/RH-MCM-41(2d)
132
were different from that of 4%K2O/RH-MCM-41 in which the adsorption in
133
mesopores at relative pressure 0.4-0.8 disappeared indicating more collapse of the
134
MCM-41 mesoporeous structure. In addition, the adsorption amount at low relative
The mesopores of
135
pressure on 8%K2O/RH-MCM-41 and 12%K2O/RH-MCM-41 compared to that on
136
4%K2O/RH-MCM-41 indicating lower surface area. These results confirmed the
137
XRD results that the increase of K2O content on RH-MCM-41 ruined mesoporeous
138
structure.
139
The
pore
size
distribution
of
mesopore
in
RH-MCM-41
and
140
K2O/RH-MCM-41s were presented in Fig. 3. The pore diameter of RH-MCM-41
141
was centered at 20.4 Å. A small portion of micropores was still presented in all
142
K2O/RH-MCM-41 samples with diameter of 31.0 Å.
143
The total pore volume and specific surface area of RH-MCM-41 and
144
K2O/RH-MCM-41s are presented in Table 2. There was a dramatic decrease of
145
surface area and pore size with the addition of K2O to RH-MCM-41. In
146
K2O/RH-MCM-41 samples, the pore size and surface area decreased with K2O
147
content. The table also included BET results of RH-SiO2 and 8% K2O/RH-SiO2.
148
These data were used for comparison in the section below.
149
Results from catalytic testing of K2O/RH-MCM-41 and K2O/RH-SiO2
150
The conversions of fatty acids were displayed in Fig. 4a-4c in which the
151
formation of methyl palmitate (C16:0), unsaturated methyl esters, including methyl
152
oleate (C18:1), methyl linoleate (C18:2) and methyl linolenate (C18:3), and methyl
153
stearate (C18:0) were plotted versus catalysts with different K2O loading,
154
respectively. The formation of methyl esters occurred depending on the amount of
155
fatty acids in the raw material and the most active catalyst for transesterification
156
was 8%K2O/RH-MCM-41. In addition, the conversion 8%K2O/RH-MCM-41 also
157
depended on temperature. However, the conversion at 100˚C was not significantly
158
higher than that at 75˚C (namely, 84% VS. 82%). As a result, the temperature at
159
75C was considered to be the more suitable condition with regards to energy
160
saving.
161
To compare catalysts on different supports, the catalytic activity of
162
8%K2O/RH-MCM-41 and that of 8%K2O/RH-SiO2 were determined at at 100 °C
163
and the results are compared in Fig. 5. The 8%K2O/RH-MCM-41 gave higher
164
conversion of both C-16 and C-18 than 8%K2O/RH-SiO2. This might be
165
contributed to the difference in the surface area because the first one had higher
166
surface area (56 versus 6 m2/g).
167
CONCLUSION
168
Loading K2O onto RH-MCM-41 significantly changed the support mesoporeous
169
structure resulting the structure collapsed and lower surface area.
The
170
K2O/RH-MCM-41 was active for the transesterification of palm olein oil with
171
methanol. The performance depended with the K2O loading and temperature. The
172
highest conversion was observed on 8%K2O/RH-MCM-41 at 100˚C. Both
173
saturated methyl ester (C16:0) and unsaturated methyl esters (C18:1, C18:2 and
174
C18:3) were obtained with approximately 70-80% yield.
175
ACKNOWLEDGMENT
176
Research funding, scholarship for outstanding student and graduate thesis fund was
177
from Suranaree University of Technology.
178
supported by the National Synchrotron Research Center (Grant 2-2548/PS01).
179
REFERENCES
180
Kumer, D., Schumucher, K., de Fresne von Hoheneschen, C.C., and Unger, K-M.
181
K. (2001). RH-MCM-41, MCM-48 and related mesoporous adsorbents: their
This research was also partially
182
synthesis and characterization. Colloids and Surfaces A: Physicochemical
183
and Engineering Aspects., 187: 109.
184
185
Bialek, R., Meier, W.M. and Davis, M.E. (1991). The synthesis and structure of
SSZ-24, the silica analog of AIPO4-5. Zeolite., 11(5): 438.
186
Khemthong, P., Prayoonpokarach, S., Wittayakun, J. (2007). Synthesis and
187
Characterization of Zeolite LSX from Rice Husk Silica, Suranaree Journal
188
of Science and Technology, 12(4), in press.
189
Grisdanurak, N., Chiarakorn, S. and Wittayakun, J. (2003). Utilization of
190
mesoporous molecular sieves synthesized from natural source rice husk
191
silica to chlorinated volatile organic compounds (CVOC) adsorption.
192
Korean Journal of Chemical Engineering., 20(5): 950.
193
194
195
196
Ma, F. and Hanna, M.A. (1999). Biodiesel production: a review. Bioresource
Technology., 70(1): 1.
Srinivas, D., Srivastava, R. and Ratnasamy, P. (2004). Transesterifications over
titanosilicate molecular sieves. Catalysis Today., 96(3): 127.
197
Xie, W., Huang, X. and Li, H. (2006). Soybean oil methyl ester preparation using
198
NaX Zeolite loaded with KOH as a heterogeneous catalyst. Bioresource
199
Technology., 98(4):936.
200
Park, S-H., Kim, B-H., Selvaraj, M. and Lee,T-G. (2007). Synthesis and
201
characterization of mesoporous Ce-Mn-MCM-41 molecular sieve. Journal
202
of Industrial and Engineering Chemistry., 13(4): 637.
203
Papp, A., Molnár, Á. and Mastalir, Á. (2005). Catalytic investigation of Pd
204
particles supported on MCM-41 for the selective hydrogenations of terminal
205
and internal alkynes. Applied Catalysis A: General., 289(2): 256.
206
Department of Thailand Agriculture. (2007). Oil palm research center, Available
207
from: http://www.doa.go.th/palm/linkTechnical/processOilpalm.html.
208
Accessed Dec 11, 2007.
209
210
211
List of tables
212
Table 1. Structure properties data of RH-MCM-41, K2O/RH-MCM-41 and
213
214
215
216
K2O/SiO2
Table 2. Pore volumes and surface areas of RH-SiO2, RH-MCM-41, K2O/RH-SiO2
and K2O/RH-MCM-41
217
List of figures
218
Figure 1. XRD patterns of (a) MCM-41 (b) 4%K2O/MCM-41 (c)
219
8%K2O/MCM-41 and (d) 12%K2O/MCM-41
220
Figure
2.
N2
adsorption-desorption
isotherm
of
RH-MCM-41
and
221
K2O/RH-MCM-41; (a) RH-MCM-41 (b) 4%K2O/RH-MCM-41 (c)
222
8%K2O/RH-MCM-41 (d) 12%K2O/RH-MCM-41.
223
Figure 3. Pore size distribution of RH-MCM-41 and K2O/RH-MCM-41; (a)
224
RH-MCM-41 (b) 4%K2O/RH-MCM-41 (c) 8%K2O/RH-MCM-41 (d)
225
12%K2O/RH-MCM-41.
226
Figure 4a. Formation of methyl palmitate on K2O/RH-MCM-41at various
227
temperatures, b. Formation of unsaturated methyl esters (methyl oleate,
228
C18:1; methyl linoleate, C18:2 and methyl linolenate, C18:3) on
229
K2O/RH-MCM-41at various temperatures, c. Formation of methyl
230
stearate (C18:0) on K2O/RH-MCM-41at various temperatures
231
Figure 5. Catalytic activity of 8%K2O supported on RH-MCM-41 and RH-SiO2 at
232
233
234
100 °C
235
236
Table 1. Structure properties data of RH-MCM-41, K2O/RH-MCM-41 and
K2O/SiO2
237
Components
2
a* (Å)
d100 (Å)
RH-MCM-41
2.3
43.6
37.8
4%K2O/RH-MCM-41
2.5
39.4
34.1
8%K2O/RH-MCM-41
2.6
38.6
33.4
12%K2O/RH-MCM-41
2.7
37.5
32.5
238
239
* Unit cell parameter of RH-MCM-41 and K2O/RH-MCM-41 of the 100 plan,
240
calculated form a100 
241
2
d100
3
242
243
Table 2. Pore volumes and surface areas of RH-SiO2, RH-MCM-41,
K2O/RH-SiO2 and K2O/RH-MCM-41
244
Vp(cm3/g)*
SBET (m2/g)
RH-SiO2
0.11
234.2
RH-MCM-41
0.97
1231.4
4%K2O/RH-MCM-41
0.19
118.5
8%K2O/RH-MCM-41
0.11
55.8
8%K2O/RH-SiO2
0.003
6.11
12%K2O/RH-MCM-41
0.06
44.7
Components
245
246
247
*Total pore volumes, calculated from N2 desorption by BJH method.
248
Intensity (a.u.)
12%K2O/RH-MCM-41
8%K2O/RH-MCM-41
4%K2O/RH-MCM-41
RH-MCM-41
0
2
4
6
8
10
12
14
2
249
250
Figure 1. XRD patterns of (a) MCM-41 (b) 4%K2O/MCM-41 (c)
251
8%K2O/MCM-41 and (d) 12%K2O/MCM-41
252
16
700
a
3
Volume Adsorbed (cm )
600
x5
500
400
b
300
c
x8
x8
200
d
100
0
0.0
0.2
0.4
0.6
0.8
1.0
P/P0
253
254
Figure
255
K2O/RH-MCM-41;
256
8%K2O/RH-MCM-41 (d) 12%K2O/RH-MCM-41.
257
2.
N2
adsorption-desorption
(a)
RH-MCM-41
isotherm
(b)
of
RH-MCM-41
and
4%K2O/RH-MCM-41
(c)
3
Volume adsorbed (cm )
0.20
0.18
0.16
0.14
0.12
a
0.10
x40
b
0.08
0.06
x80
c
0.04
d
0.02
x40
0.00
10
20
30
40
50
60
Poresize distribution (Angstrom)
258
259
Figure 3. Pore size distribution of RH-MCM-41 and K2O/RH-MCM-41; (a)
260
RH-MCM-41 (b) 4%K2O/RH-MCM-41 (c) 8%K2O/RH-MCM-41 (d)
261
12%K2O/RH-MCM-41.
262
263
%Conversion…..
80
70
60
50
40
30
20
10
0
50
75
100
Temperature (Celcius)
4%K2O/MCM-41
8%K2O/MCM-41
12%K2O/MCM-41
90
80
70
60
50
40
30
20
10
0
70
%Conversion…..
%Conversion…..
264
60
50
40
30
20
10
0
50
75
100
50
Temperature (celcius)
4%K2O/MCM-41
75
100
Temperature (Celcius)
8%K2O/MCM-41
4%K2O/MCM-41
12%K2O/MCM-41
8%K2O/MCM-41
12%K2O/MCM-41
265
266
267
Figure 4 a. Formation of methyl palmitate on K2O/RH-MCM-41at various
268
temperatures, b. Formation of unsaturated methyl esters (methyl oleate, C18:1;
269
methyl linoleate, C18:2 and methyl linolenate, C18:3) on K2O/RH-MCM-41at
270
various
271
K2O/RH-MCM-41at various temperatures
272
temperatures,
c.
Formation
of
methyl
stearate
(C18:0)
on
90
%Conversion…..
80
70
60
50
40
30
20
10
0
C16:0
C18:1,2,3
C18:0
FAMEs
8%K2O/MCM-41
8%K2O/SiO2
273
274
275
Figure 5. Catalytic activity of 8%K2O supported on RH-MCM-41 and RH-SiO2 at
100 °C