222
F.S.H. Simanjuntak et al. / Applied Catalysis A: General 401 (2011) 220–225
100
100
80
60
40
20
80
60
DMC/glycerol = 2:1
DMC/glycerol = 4:1
DMC/glycerol = 6:1
Yield (%)
Conversion (%)
Selectivity (%)
40
20
0
0.5
1.0
1.5
2.0
2.5
3.0
0
10
20
30
40
50
60
70
80
90
catalyst/glycerol (mol%)
reaction time (min)
Fig. 1. Effect of the amount of CaO on the transesterification of dimethyl carbon-
ate with glycerol. Reaction condition: Glycerol (20 mmol), DMC (40 mmol), 75 ◦C,
30 min.
Fig. 2. Effect of the reaction time (CaO uncalcined) on the yield of glycerol carbonate.
Reaction condition: catalyst/glycerol = 1 mol%, T = 75 ◦C.
than 1 mol%, the solution was completely transparent.
considerably high activity almost comparable to Na2O although the
basicity of CaO is much weaker than that of Na2O. When reagent
increased further to 94.0% when CaO was used after calcination at
900 ◦C for 3 h. One interesting point observed with the use of cal-
cined CaO is that the dissolution of CaO is completed in 5 min (see
Supporting information, Fig. S1). In contrast, the reaction mixture
with uncalcined CaO showed a little turbidity even after the com-
pletion of the reaction. The reason for this turbidity is not clear at
the moment, but it seems that H2O and or CO2 adsorbed on the
surface of the uncalcined CaO retards the dissolution. The possible
effect of water on the dissolution of CaO is somewhat supported by
the turbidity of the solution after the exposure of the calcined CaO
in air for 3 h. As a whole, the activity of a metal oxide catalyst is
greatly affected by the dissolution time and the degree of dissolu-
tion: the more soluble the catalyst, the higher the catalytic activity,
suggesting that the transesterification proceeds mostly in a homo-
geneous way. Therefore, the significantly lower activity of MgO and
ZnO can be ascribed in part to their extremely poor solubilities in
the reaction mixtures. In the metal oxide-catalyzed transesterifi-
cation between glycerol and DMC, it is easily conceivable that the
first step for the generation of an active species is the acid-base
interaction between a hydroxyl group of glycerol and metal oxide.
In this context, the generation of the active species from the less
basic MgO would be much more difficult than that from the more
basic CaO. In other word, the basicity of MgO is not sufficiently
high enough to generate the similar homogeneous active species
to that observed in the CaO-catalyzed transesterification [20,21].
Considering the extremely low activity and solubility of MgO, the
transesterification in the presence of MgO seems to proceed in a
heterogeneous way.
The effect of reaction time was also investigated at 75 ◦C at
varying molar ratio of DMC/glycerol. The catalyst loading was main-
tained at 1 mol% with respect to glycerol. Fig. 2 reveals that, at the
early stages of the reaction up to 2–3 min, the formation of glycerol
carbonate was negligible. However, the yield of glycerol carbonate
started to increase rapidly thereafter, implying that there exists an
induction period for the formation of an active calcium species from
CaO. It is interesting to notice that the reaction time to reach equi-
librium increases with the increasing molar ratio of DMC/glycerol
from 2 to 6. This result may imply that, above the molar ratio of
2, heterogeneous catalysis is also functioning in the CaO-catalyzed
transesterification (vide infra).
The effect of molar ratio of DMC to glycerol was also investigated
at 75 ◦C for 15 min. As shown in Fig. 3, the yield of glycerol carbon-
ate reaches a maximum of 85.4 at the molar ratio of 2 and then
decreases steadily with the increase of the molar ratio up to 10,
again suggesting that, above the molar ratio of 2, heterogeneous
catalysis is also functioning in the CaO-catalyzed transesterifica-
tion. In fact, the turbidity of the reaction mixture increased with
the increase of DMC concentration. It is likely that the increased
of CaO with glycerol, thereby preventing the formation of soluble
To see the correlation between the amount of dissolved Ca
species and the catalytic activity, the amount of dissolved Ca
species was analyzed (Fig. 3). When the molar ratio of DMC/glycerol
increased from 1 to 2, the dissolved Ca content decreased from 15.6
to 11.5 mg. Considering that the initial Ca content in CaO was 17 mg,
approximately 91.8% and 67.6% of CaO transformed into soluble
species at the DMC/glycerol molar ratios of 1 and 2, respectively.
It is worth to note that, in spite of decrease in the amount of dis-
solved Ca species as going from 15.6 to 11.5 mg, the yield of glycerol
carbonate increased from 68.0 to 85.3%. On further increase of the
molar ratio to 4, the dissolved Ca content drastically dropped down
to 0.3 mg, which is only 2.6% of that measured at the molar ratio
of 2. On the contrary, the reduction of glycerol carbonate yield was
much smaller from 85.3 to 63.9% as the molar ratio was decreased
from 2 to 4. It is surprising to observe that the transesterification
proceeds smoothly even in the presence of such a low dissolved Ca
content in the solution. These results may suggest that DMC plays a
pivotal role to shift the transesterification toward the formation of
glycerol carbonate is strongly affected by the concentration of DMC.
The effect of catalyst loading on the yield of glycerol carbonate
also supports the idea that CaO-catalyzed transesterification take
places in a homogeneous way. As illustrated in Fig. 1, the yield and
the selectivity of the glycerol carbonate did not vary much when the
catalyst loading was decreased from 3 to 0.5 mol% with respect to
glycerol. Likewise, the yield of glycerol carbonate remained almost
constant with the decrease of catalyst loading from 1 to 0.5 mol%.
However, the yield of glycerol carbonate was reduced to 62.9%
when the catalyst loading was further decreased to 0.25 mol%. The
turbidity of the reaction mixture was affected by the catalyst con-
centration. When 1 mol% CaO (uncalcined) was used, the solution