Characterization of KF/γ-Al2O3 catalyst for the synthesis of diethyl carbonate by transesterification of ethylene carbonate

Article abstract of DOI:10.1007/s10562-010-0354-8

KF/γ-Al₂O₃ catalysts were prepared by impregnation method and investigated for the transesterification of ethylene carbonate (EC) with ethanol to synthesize diethyl carbonate (DEC). The KF/γ-Al₂O₃ catalysts were characterized by nitrogen physisorption, XRD and FT-IR techniques, and three new species: K ₃AlF₆, KOH and K₂CO₃ were found on the catalysts. Experimental results indicate that KOH and K₂CO ₃ are the major active species and K₃AlF₆ is inactive for DEC synthesis. Increasing the KF loading favors the formation of K₂CO₃ and consequently menhances the activity of the KF/γ-Al₂O₃ catalysts. However, when KF loading exceeded 50 mmol/g, the activity of the KF/γ-Al₂O₃ catalysts decreased. This may be due to the presence of intact KF on the catalyst, which may dilute the content of active species in the catalyst and cover the active species. The KF/γ-Al₂O₃ (50 mmol/g) catalyst exhibits the best catalytic performance. With this catalyst, a 72 mol% yield of DEC (based on EC) was obtained at 298 K. Springer Science+Business Media, LLC 2010.

Full text of DOI:10.1007/s10562-010-0354-8

Catal Lett (2010) 137:232–238
DOI 10.1007/s10562-010-0354-8
Characterization of KF/c-Al O Catalyst for the Synthesis
2
3
of Diethyl Carbonate by Transesterification of Ethylene
Carbonate
Peng Qiu
Suitao Qi
•
Bolun Yang
•
Chunhai Yi
•
Received: 22 January 2010 / Accepted: 19 April 2010 / Published online: 1 May 2010
Ó Springer Science+Business Media, LLC 2010
Abstract KF/c-Al O catalysts were prepared by impreg-
(40.6 vs. 18.2%) and it has more favorable gasoline/water
2
3
nation method and investigated for the transesterification of
ethylene carbonate (EC) with ethanol to synthesize diethyl
carbonate (DEC). The KF/c-Al O catalysts were charac-
distribution coefficients than dimethyl carbonate (DMC)
and ethanol, DEC has been considered as a promising fuel
additive [1, 2]. As a reactive intermediate, DEC can be
used to produce pharmaceuticals, urethanes, ureas, and so
on [3]. In addition, DEC is also an excellent solvent and an
electrolyte for lithium ion batteries [4]. Thus, much
attention has been paid to the synthesis of DEC.
2
3
terized by nitrogen physisorption, XRD and FT-IR tech-
niques, and three new species: K AlF , KOH and K CO
3
3
6
2
were found on the catalysts. Experimental results indicate
that KOH and K CO are the major active species and
2
3
K AlF is inactive for DEC synthesis. Increasing the KF
3
Several chemical routes have been reported for the
preparation of DEC: the phosgene route [5], oxidative
carbonylation of ethanol [1], reaction between ethyl nitrite
and carbon monoxide [1], and alcoholysis of urea [6]. The
major drawback of the first three routes is the use of poi-
sonous gases, such as phosgene and carbon monoxide.
Alcoholysis of urea suffers from a low yield of DEC, only a
14% yield of DEC was obtained under the optimum reac-
tion conditions [6]. One promising method to produce DEC
is the transesterification of ethylene carbonate (EC) with
ethanol (Scheme 1). This method possesses two major
advantages: firstly, both EC and ethanol are low toxic
substance, which means a safe operation; secondly, EC is
produced industrially by the reaction between carbon
dioxide and oxirane. As a result, carbon dioxide, which is a
greenhouse gas, is indirectly utilized to produce DEC.
It is well-known that the transesterification of EC with
alcohols can be catalyzed by both acid and base catalysts.
However, Knifton and Duranleau [7] discovered that base
catalysts are more active than acid catalysts. Many
homogenous and heterogenous base catalysts have been
reported for the reaction of EC with methanol or ethanol to
synthesize DMC or DEC, such as alkali metal alkoxides
and hydroxides [8], thallium compounds [9], poly-4-vinyl
pyridine [10], MgO-CeO2 [11], amino-functionalized
MCM-41 [12], Mg–Al hydrotalcite [13], organic phos-
phines/amine supported on partially cross-linked polymers
6
loading favors the formation of K CO and consequently
2
3
enhances the activity of the KF/c-Al O catalysts. However,
2
3
when KF loading exceeded 50 mmol/g, the activity of
the KF/c-Al O catalysts decreased. This may be due to the
2
3
presence of intact KF on the catalyst, which may dilute the
content of active species in the catalyst and cover the active
species. The KF/c-Al O (50 mmol/g) catalyst exhibits the
2
3
best catalytic performance. With this catalyst, a 72 mol%
yield of DEC (based on EC) was obtained at 298 K.
Keywords Diethyl carbonate ꢀ Transesterification ꢀ
Potassium fluoride ꢀ Alumina
1
Introduction
Diethyl carbonate (DEC) is a versatile chemical due to its
excellent chemical and physical properties. Since its oxy-
gen content is higher than that of methyl tert-butyl ether
P. Qiu ꢀ B. Yang (&) ꢀ C. Yi ꢀ S. Qi
Department of Chemical Engineering, State Key Laboratory of
Multiphase Flow in Power Engineering, Xi’an Jiaotong
University, Xi’an, Shaanxi 710049, People’s Republic of China
e-mail: blunyang@mail.xjtu.edu.cn
1
23

Characterization of KF/c-Al
2
O
3
Catalyst for the Synthesis of Diethyl Carbonate
233
Ltd., China. All the chemicals were of analytical grade and
used without further purification.
2.2 Preparation of Catalysts
Several KF catalysts with different supports were pre-
pared by an impregnation method that is analogous to
that reported by Bo et al. [21]. Firstly, an appropriate
Scheme 1 Diethyl carbonate synthesis reaction
amount of KFꢀ2H O was dissolved in 20 mL distilled
[
7], Na PO [14], tungstate-based solid catalysts [15].
3
2
4
water to prepare an aqueous solution of KF. After that, a
However, homogeneous catalysts have a drawback that
they are difficult to separate from the reaction mixture. And
most of the heterogenous catalysts are reported requiring
high temperature, usually above 373 K, to obtain a mod-
erate yield of DMC. Considering the fact that the reaction
of ethanol with EC is more difficult than that of methanol
with EC, synthesis of DEC from EC and ethanol usually
requires higher temperatures. For instance, Krimm et al. [9]
used thallium hydroxide as catalyst to obtain a 58% yield
of DEC at 423 K. Dhuri et al. [16] reported that around
1
.0 g solid support (c-Al O , ZnO, TiO , ZrO , La O ,
2
3
2
2
2 3
AC) was added to this solution. The mixture was aged
overnight, dried at 383 K for 24 h, and then calcined at
6
73 K in the air for 3 h (AC supported catalysts were
calcined under the protection of nitrogen). KOH and
K CO were supported on c-Al O following the same
2
3
2 3
procedure. KF was prepared by evacuating KFꢀ2H O at
2
3
2
13 K for 24 h.
.3 Apparatus and Procedure
2
5% DEC yield was obtained using Amberlyst A-21 cat-
alyst at 393 K. Therefore, it would be desirable to find a
new heterogenous base catalyst that can catalyze the
transesterification of EC with ethanol under mild condi-
tions (low temperature and atmospheric pressure).
The reaction was carried out in a 100 mL round bottom
flask equipped with a thermometer, a magnetic stirrer, and
a water-cooled condenser. The flask was charged with
40 mmol EC, 400 mmol ethanol, and 0.4 g normal octane
Potassium fluoride (KF) supported on alumina
(internal standard), the charged flask was then placed in a
(
KF/Al O ) has been used as a heterogenous base catalyst
2 3
thermostatic bath and heated to a desired temperature with
stirring at 800 rpm. When the reaction mixture reached the
desired temperature, a 0.5 g catalyst was added and the
reaction started. The reaction mixture was sampled peri-
odically. The samples were analyzed by gas chromatog-
raphy [Fuli Analytical Instrument Co. Ltd., Zhejing,
China, GC9790, equipped with a flame-ionization detector
and a KB-5 capillary column (30 m 9 0.32 mm 9
in a number of base-catalyzed reactions such as Michael
addition, double bond isomerization and aldol condensa-
tion [17–19]. The KF/Al O catalyst shows high activity
2
3
for these reactions due to its strongly basic nature [20].
Although the use of KF/Al O as catalyst first appeared in
2
3
the literature 30 years ago, the exact nature of active spe-
cies on it is still controversial. In this study, KF/c-Al O
2
3
catalysts were prepared and employed for DEC synthesis
0.25 lm)].
from EC and ethanol. The KF/c-Al O3 catalysts were
2
characterized by XRD and FT-IR techniques to identify the
active species and to make the mechanisms of transesteri-
fication clearly on the catalyst.
2.4 Catalyst Characterization
The BET specific surface area, pore volume and pore size
were measured by the nitrogen adsorption method at 77 K
using a JW-K nitrogen adsorption analyzer (made by
JWGB Corporation of China). FT-IR spectra were
2
Experimental
-
1
recorded in the range 4000–400 cm
using a Nicolet
2
.1 Materials
Avatar 360 spectrometer. All the samples were pretreated
under vacuum (2 Torr) at 383 K for 3 h to remove
moisture. After that, the samples were mixed with KBr at
a weight ratio of 1:50. Powder X-ray diffraction patterns
of various samples were recorded on a PANalytical
X-Pert PRO diffractometer (made by PANalytical Cor-
Normal octane, c-Al O , ZrO , La O , and DEC were
2
3
2
2 3
purchased from Sinopharm Chemical Reagent Co., Ltd.
Ethanol 99.7% anhydrous and ZnO were obtained from
Xi’an Chemical Reagent of China. Potassium fluoride
dihydrate (KFꢀ2H O) was purchased from Tianjing
2
poration of Netherlands) with nickel-filtered Cu-K radi-
a
Hengxing Chemical Reagent of China. Ethylene carbonate
ation at 40 kV and 40 mA. All the measurements were
was purchased from Aldrich. TiO and activated carbon
2
performed over a 2h range of 10–90°.
(
AC) were purchased from Tianjing Jingbei Chemical Co.,
123

2
34
P. Qiu et al.
3
Results and Discussion
Table 2 Effect of different supports on catalytic performance
Catalyst
KF loading
mol/g)
Reaction time
(min)
DEC yield
(mol%)
3
.1 Effects of Supports
(
KF/La
2
O
3
20
20
20
20
20
30
20
30
–
100
100
100
100
100
200
100
200
100
100
22.1
45.2
6.5
Six different materials: c-Al O , ZnO, TiO , ZrO , La O
2
3
2
2
2 3
KF/c-Al
2
O
3
and AC, were used as support to prepare KF supported
catalysts. The BET specific surface areas, pore volumes
and pore sizes of the supports and catalysts are presented in
Table 1. It can be seen from this table that the specific
surface areas, pore volumes and pore sizes of c-Al O and
KF/TiO
KF/ZrO
KF/AC
KF/AC
2
2
5.0
0.7
0.9
2
3
KF/ZnO
KF/ZnO
KF
5.8
AC decreased markedly after they were supported with KF.
This suggests that large structural changes occur during
impregnation of these two supports. However, when other
solid supports were impregnated with KF, only a little
change can be observed in textural properties.
8.2
Trace
Trace
c-Al
2
O
3
–
Reaction conditions: EC, 40 mmol; ethanol, 400 mmol; normal
octane (internal standard), 0.4 g; catalyst, 0.5 g; reaction temperature,
The catalytic activity of different KF supported catalysts
are shown in Table 2. It can be seen form this table that KF/
c-Al O shows the highest catalytic activity among the
3
23 K; atmospheric pressure
2
3
several catalysts tested. A 45 mol% yield of DEC (based on
3.2 Effect of KF Loading
EC) was obtained in the presence of the KF/c-Al O catalyst.
2
3
However, the DEC yield was decreased nearly by half when
To investigate the effect of KF loading on the activity of
c-Al O was replaced by La O . The other catalysts: KF/
the KF/c-Al
2
O
3
catalysts, several KF/c-Al O catalysts
2 3
2
3
2 3
ZnO, KF/TiO and KF/ZrO are almost inactive for DEC
2
with different KF loading were prepared and their cata-
lytic performance was tested. As can be seen in Fig. 1,
the DEC yield increases significantly with an increase in
KF loading when the KF loading is below 50 mmol/g.
However, a further increase in KF loading decreased the
2
synthesis. To our surprise, the KF/AC catalyst performed
much worse than the KF/c-Al O catalyst. The yield of DEC
2
3
obtained on KF/c-Al O is almost two orders of magnitude
2
3
higher than that obtained on KF/AC. This phenomenon can
not be explained by the specific surface area of the supports,
since the specific surface area of AC is two times higher than
that of c-Al O . Moreover, both unsupported KF and pure c-
activity of the KF/c-Al
decreased from 63 to 27 mol% when the KF loading
increased from 50 to 120 mmol/g. The KF/c-Al cata-
O catalysts. The DEC yield
2 3
O
2 3
2
3
Al O exhibit little catalytic activity for DEC synthesis, also
2
lyst with a KF loading equal to 50 mmol/g shows the
3
shown in Table 2. These results suggest that there is a strong
highest catalytic activity.
interaction between KF and c-Al O .
2
3
3.3 Effect of Catalyst Loading
The effect of catalyst loading on the DEC yield was
Table 1 Textural properties of different supports and catalysts
investigated. Since the KF/c-Al O (50 mmol/g) catalyst
2
3
Support
BET surface
2
area (m /g)
Average pore
diameter (nm)
Pore volume
3
(cm /g)
c-Al
2
O
3
230
483
5.6
11.8
3.6
0.69
AC
0.32
La
TiO
ZrO
ZnO
2
O
3
21
0.023
0.057
0.076
0.015
0.47
2
16.8
16.1
4.3
13.6
17.1
18.6
9.8
2
KF/c-Al
2
O
3
132
371
4.9
KF/AC
2.8
0.29
KF/La
KF/TiO
KF/ZrO
KF/ZnO
2
O
3
19
0.021
0.052
0.063
0.013
2
14.2
13
12.5
16.4
17.3
2
Fig. 1 Effect of KF loading on catalytic activity. Reaction condi-
tions: EC, 40 mmol; ethanol, 400 mmol; normal octane (internal
standard), 0.4 g; catalyst, 0.5 g; reaction temperature, 323 K; reaction
time, 200 min; atmospheric pressure
3.8
The catalysts tested have a KF loading of 20 mmol/g
1
23

Characterization of KF/c-Al
2
O
3
Catalyst for the Synthesis of Diethyl Carbonate
235
Fig. 2 Effect of catalyst loading on the DEC yield. Reaction
conditions: EC, 40 mmol; ethanol, 400 mmol; normal octane (inter-
Fig. 3 Influence of reaction temperature on the yield of DEC.
Reaction conditions: EC, 40 mmol; ethanol, 400 mmol; normal
octane (internal standard), 0.4 g; catalyst, 0.5 g; atmospheric pressure
2 3
nal standard), 0.4 g; catalyst, KF/c-Al O (50 mmol/g); reaction
temperature, 323 K; reaction time, 130 min; atmospheric pressure
ethanol is a reversible exothermic reaction. An increase in
reaction temperature diminishes the equilibrium constant
and consequently decreases the equilibrium yield.
exhibits the best catalytic performance, the experiments
were performed at 323 K with different amounts of the KF/
c-Al O (50 mmol/g) catalyst. The results are shown in
2
3
3
.5 Active Species on KF/c-Al O
Fig. 2.
2 3
It can be seen from Fig. 2 that when the catalyst loading
is between 0.1 and 0.5 g, an increase in catalyst loading
significantly increases the DEC yield. However, a further
increase in catalyst loading from 0.5 to 0.8 g leads to a
slow increase in DEC yield from 59 to 64 mol%. When the
catalyst loading exceeds 0.8 g, the DEC yield is nearly
independent of catalyst loading. This indicates that a
chemical equilibrium has been reached. Therefore, 0.5 g is
a suitable catalyst loading amount in this experimental
system.
To find out the origin of high activity of KF/c-Al O cat-
2
3
alyst, several KF/c-Al O catalysts with different KF
2
3
loading were characterized by XRD. As can be seen in
Fig. 4, XRD analysis confirms the presence of K AlF ,
3
6
KOH and K CO on the KF/c-Al O catalysts. The dif-
2
3
2 3
fraction peaks at 2h = 18.2°, 29.9°, 36.9°, 43°, 53.2° and
6
2° are ascribed to K AlF , the diffraction peak at
h = 30.9° is ascribed to KOH, as supported by reference
3
6
2
[
22], and the peaks at 2h = 12.8° and 32.5° are ascribed to
K CO . K AlF , KOH and K CO exist on all catalysts
2
3
3
6
2
3
tested. The above results indicate that the KF loaded can
3
.4 Influence of Reaction Temperature
react with c-Al O and H O to form K AlF and KOH
2
3
2
3
6
(
reaction 1). This conclusion is consistent with that
The influence of reaction temperature on the DEC yield is
summarized in Fig. 3. The reaction temperature was varied
between 298 and 351 K. The catalyst used was KF/c-Al O
reported by Weinstock et al. [23]. Moreover, it can be seen
from Fig. 4 that the diffraction peaks ascribed to KF
appears when KF loading reaches 50 mmol/g. It indicates
2
3
with a KF loading equal to 50 mmol/g.
that for the KF/c-Al O (50 mmol/g) catalyst, some of the
2 3
As can be seen in Fig. 3, the KF/c-Al O3 catalyst
2
exhibited high activity for DEC synthesis, even at room
temperature (298 K), a 60 mol% DEC yield (based on EC)
was obtained in 300 min. When the reaction temperature
increased, the DEC formation rate was increased markedly,
especially in the initial 150 min. At 351 K, the reaction
reached equilibrium in 120 min. It can be expected that the
reaction could proceed more rapidly at a temperature
higher than 351 K. However, because of the low boiling
point of ethanol, the reaction mixture cannot be heated to a
temperature exceeding 351 K under atmospheric pressure.
In addition, Fig. 3 also shows that the equilibrium yield of
DEC decreases with reaction temperature. The DEC
equilibrium yield was 72 mol% at 298 K, 67 mol% at
3
13 K, 65 mol% at 323 K and 63 mol% at 351 K. This
2 3
Fig. 4 XRD patterns of KF/c-Al O catalysts with different KF
indicates that the transesterification between EC and
loading
123

2
36
P. Qiu et al.
KF loaded, but not all, is reacted with c-Al O to form
2
3
KOH.
1
2KF þ Al O þ 3H O ! 2K AlF þ 6KOH
ð1Þ
In addition, IR spectroscopy also confirms the formation of
2
3
2
3
6
K AlF and K CO on the KF/c-Al O catalysts. As shown
3
6
2
3
2 3
-
in Fig. 5, the infrared band at 586 cm is assigned to the
1
3
AlF6 stretching frequency of K AlF and the bands at
-
3
6
-
1
2-
7
06, 884, 1,378 and 1,450 cm are assigned to the CO3
stretching frequency of K CO3 [24, 25]. However, the
2
-
1
characteristic band of KOH at 3,502 cm
was not
observed on all catalyst samples. The reason for this can
be considered to be the lower content of KOH on the KF/c-
Al O catalysts, because even the pure KOH shows a weak
Fig. 6 Comparison of DEC yields obtained with different catalysts.
Reaction conditions: EC, 40 mmol; ethanol, 400 mmol; normal
octane (internal standard), 0.4 g; catalyst, 0.5 g; reaction temperature,
2
3
-
1
band at 3,502 cm (Fig. 5). According to a relative report
26], K CO may be formed by the reaction of KOH with
323 K; reaction time, 200 min; atmospheric pressure. 1, K
KF/c-Al (50 mmol/g); 3, KOH/c-Al (15 mmol/g); 4, K
Al (10 mmol/g)
3
AlF
CO
6
; 2,
/c-
2
O
3
O
2 3
2
3
[
2
3
2 3
O
CO2 that is in the air during preparation of catalyst
reaction 2). To further confirm this fact, an additional KF/
c-Al O (50 mmol/g) catalyst was prepared by drying and
(
2
3
catalyst. This result is somewhat different from that pro-
posed by Feng et al. [27], in which they propose that
K AlF can behave as strong basic center and that K AlF
6
calcination under protection of N and then characterized
2
by IR spectroscopy (Fig. 5). The infrared bands for K CO
2
3
3
6
3
of this catalyst weakened sharply compared with that of the
is responsible for high activity of KF/c-Al O catalyst.
2
3
KF/c-Al O (50 mmol/g) catalyst that is prepared in the
To investigate whether KOH and K CO are the major
2 3
2
3
air. This indicates that atmospheric CO plays an important
2
active species on KF/c-Al O catalyst, KOH and K CO
2
3
2
3
role in forming K CO . Additionally, it should be noted
3
were supported on c-Al O respectively and used as cata-
2
2 3
that the infrared bands assigned to K CO and K AlF
6
lyst to synthesize DEC. However, according to XRD
analysis (Fig. 4), there are still KF peaks on the XRD
pattern of the KF/c-Al O (50 mmol/g) catalyst. This fact
2
3
3
strengthened with KF loading when KF loading is below
0 mmol/g. This indicates that increasing the KF loading in
this region favors the formation of K CO and K AlF .
5
2
3
2
3
3
6
indicates that some, but not all, of the KF loaded is con-
verted into KOH. Thus, 15 mmol of KOH and 10 mmol
K CO were supported on 1 g c-Al O separately. As can
2
KOH þ CO ! K CO þ H O
ð2Þ
2
2
3
2
2
3
2 3
be seen in Fig. 6, the KOH/c-Al O (15 mmol/g) and
2
3
To determine whether K AlF is the active species on
3
6
K CO /c-Al O (10 mmol/g) catalysts show similar reac-
2
3
2 3
KF/c-Al O catalyst, pure K AlF was used as catalyst to
2
3
3
6
tivity to the KF/c-Al O (50 mmol/g) catalyst. These
2
3
synthesize DEC. As shown in Fig. 6, K AlF exhibits little
3
6
results suggest that KOH and K CO may be the major
2
3
catalytic activity for DEC synthesis. Therefore, K AlF is
6
3
active species on KF/c-Al O catalyst.
2
3
not likely to be the major active species on KF/c-Al O
2
3
Furthermore, the formation of KOH can explain the
effect of KF loading on the activity of the KF/c-Al O
2
3
3
catalysts. When KF loading is below 50 mmol/g, c-Al O
2
is present in large excess. Most of the KF loaded is reacted
with c-Al O and H O to form KOH and K AlF . As
2
3
2
3
6
shown in Fig. 4, the XRD patterns of the KF/c-Al O
2
3
(
10 mmol/g) and KF/c-Al O (20 mmol/g) catalysts do not
2
3
show any clear diffraction peaks due to KF species and the
intensity of diffraction peaks ascribed to K AlF strength-
3
6
ens. This suggests that the majority of the KF loaded may
be converted into K AlF and KOH at a low KF loading.
3
6
Increasing the KF loading in this region increases the
amount of KOH produced and consequently enhances the
activity of the catalyst. It seems strange that the activity of
the KF/c-Al O catalysts decreased when KF loading
2
3
increased from 50 to 120 mmol/g, because the KF/c-Al O
2
3
Fig. 5 IR spectra of different KF/c-Al
O
2 3
catalysts
(120 mmol/g) catalyst should contain more KOH than the
1
23

Characterization of KF/c-Al
2
O
3
Catalyst for the Synthesis of Diethyl Carbonate
237
KF/c-Al O (50 mmol/g) catalyst according to reaction 1.
2
3
Two possible reasons can explain this phenomenon. First,
when the KF loading is higher than 50 mmol/g, some of the
KF loaded was not converted into KOH and K CO . As can
2
3
be seen in Fig. 4, the diffraction peaks ascribed to KF
strengthens dramatically when KF loading increases from
5
0 to 120 mmol/g. The presence of intact KF could dilute
the content of active components (KOH and K CO ) in the
2
3
catalysts and consequently decrease the catalyst activity.
As supported by FT-IR analysis (Fig. 5), the intensity of
infrared bands for K CO decreased with KF loading when
2
3
KF loading exceeded 50 mmol/g. Second, the intact KF
may cover the active components on the catalysts. Ando
et al. [28] suggested that theoretical monolayer coverage of
KF on the c-Al O support can be achieved when KF
Fig. 8 IR spectra of KF/c-Al
after reused
2
O
3
(50 mmol/g) catalyst before and
2
3
loading reached 0.6 mmol/g. Therefore, it is reasonable to
infer that, with a KF loading higher than 50 mmol/g, most
of the KOH formed are covered by intact KF. As a result,
the activity of the KF/c-Al O catalysts decreased when KF
2
3
loading exceeded 50 mmol/g.
3
.6 Reusability of KF/c-Al O
2
3
The reusability of catalyst is important for commercializing
a new catalytic system. Thus, an attempt was made to
reuse KF/c-Al O catalyst. After each experiment, the KF/
2
3
c-Al O (50 mmol/g) solid catalyst was removed from the
2
3
reaction mixture by filtration, washed with ethanol and then
dried at 383 K for 24 h. As can be seen in Fig. 7, the
activity of the KF/c-Al O catalyst decreased with number
2 3
Fig. 9 XRD patterns of KF/c-Al O (50 mmol/g) catalyst before and
after reused
2
3
of reaction cycles. The DEC yields obtained on fresh cat-
alyst (the first run) and the catalyst that has been reused
three times (the forth run) are 62 and 42 mol%, respec-
tively, under the experimental conditions employed. The
decrease in activity for the KF/c-Al O catalyst is most
bands assigned to K CO weakened with number of reac-
2 3
tion cycles. Whereas, the intensity of the band assigned to
AlF did not change much. Moreover, the loss of K CO
was also confirmed by XRD analysis. As shown in Fig. 9,
the diffraction peaks ascribed to K CO disappeared after
the catalyst was reused three times and the peaks due to
AlF still exit. This indicates that K AlF is not likely to
be the major active species on the KF/c-Al O catalysts. In
K
3
6
2
3
2
3
2
3
likely due to the leach out of KOH and K CO into the
2
3
reaction mixture. As can be seen from Fig. 8, the infrared
K
3
6
3
6
2
3
addition, it can be seen from Fig. 9 that some KOH still
exist on the reused KF/c-Al O catalyst. The remaining
2
3
KOH should be responsible for the 42 mol% yield of DEC.
These results further confirm that KOH and K CO , rather
2
3
than K AlF , are the major active species on the KF/c-
3
6
Al O catalysts.
2
3
4
Conclusions
Diethyl carbonate can be prepared via transesterification of
EC with ethanol over KF/c-Al O catalyst at lower tem-
2
3
2 3
Fig. 7 Reusability of KF/c-Al O (50 mmol/g) catalyst. Reaction
perature and atmospheric pressure. IR and XRD analyses
show that three new species: K AlF , KOH and K CO
3
conditions: EC, 40 mmol; ethanol, 400 mmol; normal octane (inter-
nal standard), 0.4 g; catalyst, 0.5 g; reaction temperature, 323 K;
reaction time, 200 min; atmospheric pressure
3
6
2
were formed on the KF/c-Al O catalysts. Among them,
2
3
123

2
38
P. Qiu et al.
KOH and K CO are the major active species for DEC
2
6. Wang D, Yang B, Zhai X, Zhou L (2007) Fuel Process Technol
8
7
8
3
8:807–812
. Knifton JF, Duranleau RG (1991) J Mol Catal 67:389–399
. Frevel LK, Gilpin JA (1972) US Patent 3,642,858
synthesis. Increasing the KF loading increases the amount
of K CO produced and consequently enhances the activity
2
3
of the KF/c-Al O catalysts. However, when KF loading
9. Krimm H, Buysch HJ, Rudolph H (1981) US Patent 4,307,032
10. Jagtap SR, Bhor MD, Bhanage BM (2008) Catal Commun
2
3
exceeded 50 mmol/g, the activity of the KF/c-Al O cata-
2
3
9
:1928–1931
1. Abimanyu H, Kim CS, Ahn BS, Yoo KS (2007) Catal Lett
18:30–35
lysts decreased significantly. The KF/c-Al O (50 mmol/g)
2
3
1
catalyst shows the best catalytic performance. With this
1
catalyst, the DEC synthesis reaction reached equilibrium in
12. Feng XJ, Lu XB, He R (2004) Appl Catal A 272:347–352
13. Watanabe Y, Tatsumi T (1998) Microporous Mesoporous Mater
5
50 min at 298 K and a maximum DEC yield of 72 mol%
2
2:399–407
was obtained. The activity of the KF/c-Al O (50 mmol/g)
2
3
1
4. De Filippis P, Scarsella M, Borgianni C, Pochetti F (2005)
Energy Fuels 20:17–20
catalyst decreased with number of reaction cycles. The
decrease in activity is attributed to the leach out of KOH
and K CO into reaction mixture. In addition, the equi-
15. Sankar M, Nair CM, Murty KVGK, Manikandan P (2006) Appl
Catal A 312:108–114
2
3
1
6. Dhuri SM, Mahajani VV (2006) J Chem Technol Biotechnol
1:62–69
librium yield of DEC decreased from 72 to 63 mol% when
reaction temperature increased from 298 to 351 K. This
fact indicates that the transesterification of EC with ethanol
is a reversible exothermic reaction.
8
1
7. Tsuji H, Kabashima H, Kita H, Hattori H (1995) React Kinet
Catal Lett 56:363–369
18. Kabashima H, Tsuji H, Nakata S, Tanaka Y, Hattori H (2000)
Appl Catal A 194–195:227–240
1
9. Handa H, Baba T, Sugisawa H, Ono Y (1998) J Mol Catal A
34:171–177
0. Blass BE (2002) Tetrahedron 58:9301–9320
1. Bo X, Guomin X, Lingfeng C, Ruiping W, Lijing G (2007)
Energy Fuels 21:3109–3112
2. Mohamad AA, Mohamed NS, Yahya MZA, Othman R, Ramesh
S, Alias Y, Arof AK (2003) Solid State Ionics 156:171–177
3. Weinstock LM, Stevenson JM, Tomellini SA, Pan SH, Utne T,
Jobson RB, Reinhold DF (1986) Tetrahedron Lett 27:3845–3848
4. Duke CVA, Miller JM, Clark JH, Kybett AP (1990) J Mol Catal
Acknowledgements Financial support for this work from the
National Basic Research Program of China (973 Program,
1
2
2
2
(
009CB219906), National Natural Science Foundation of China
20976144, 20776117) and Specialized Research Fund for the Doc-
toral Program of Higher Education of China (20070698037) are
gratefully acknowledged.
2
2
2
References
6
2:233–242
1
. Dunn BC, Guenneau C, Hilton SA, Pahnke J, Eyring EM,
Dworzanski J, Meuzelaar HLC, Hu JZ, Solum MS, Pugmire RJ
25. Evans JV, Whateley TL (1967) Trans Faraday Soc 63:2769–2777
26. Lillo-Rodenas MA, Cazoria-Amoros D, Linares-Solano A (2003)
Carbon 41:267–275
(2001) Energy Fuels 16:177–181
2
3
4
. Pacheco MA, Marshall CL (1997) Energy Fuels 11:2–29
. Shaikh AAG, Sivaram S (1996) Chem Rev 96:951–976
. Yoshida H, Fukunaga T, Hazama T, Terasaki M, Mizutani M,
Yamachi M (1997) J Power Sources 68:311–315
27. Feng YX, Yin N, Li QF, Wang JW, Kang MQ, Wang XK (2008)
Catal Lett 121:97–102
28. Ando T, Clark JH, Cork DG, Hanafusa T, Ichihara J, Kimura T
(1987) Tetrahedron Lett 28:1421–1424
5
. Muskat IE, Strain F (1945) US Patent 2,379,250
1
23