Binary Mg-Fe oxide as a highly active and magnetically separable catalyst for the synthesis of ethyl methyl carbonate

Article abstract of DOI:10.1039/c5ra02894a

Magnetic binary Mg-Fe oxides were prepared by a co-precipitation method, characterized and tested in the synthesis of ethyl methyl carbonate (EMC) from di methyl carbonate (DMC) and diethyl carbonate (DEC). The obtained results showed that the Mg/Fe oxide catalyst with a 1:1 molar ratio and calcined at 400 °C exhibited superior catalytic activity. The yield of EMC could reach 66% (at 100 °C for 1.5 h) with a TOF of 220 mmol h^-1 g_cat^-1. The prepared catalysts could be magnetically separated, and reused for ten runs without noticeable deactivation. XRD and M?ssbauer spectra revealed that there was a synergistic effect between Mg and Fe oxides in the catalysts, which was consistent with the results of TPR, i.e. the introduction of the Mg component favored the reduction of the Fe₂O₃. XPS and IR characterizations indicated that there were a large number of accessible Fe-OHs on the surface of MgFe-400, and combining the Fe-OHs with the basic MgO may be related to the highly catalytic performance.

Full text of DOI:10.1039/c5ra02894a

RSC Advances
View Article Online
View Journal | View Issue
PAPER
Binary Mg–Fe oxide as a highly active and
magnetically separable catalyst for the synthesis of
ethyl methyl carbonate
Cite this: RSC Adv., 2015, 5, 25849
Peixue Wang,ab Shimin Liu,a Xiangyuan Ma,a Yude He,a Ahmad S. Alshammaric
a
*
and Youquan Deng
Magnetic binary Mg–Fe oxides were prepared by a co-precipitation method, characterized and tested in
the synthesis of ethyl methyl carbonate (EMC) from di methyl carbonate (DMC) and diethyl carbonate
(DEC). The obtained results showed that the Mg/Fe oxide catalyst with a 1 : 1 molar ratio and calcined
at 400 ꢀC exhibited superior catalytic activity. The yield of EMC could reach 66% (at 100 ꢀC for 1.5 h)
with a TOF of 220 mmol hꢁ1 gcatꢁ1. The prepared catalysts could be magnetically separated, and
reused for ten runs without noticeable deactivation. XRD and Mossbauer spectra revealed that there
¨
was a synergistic effect between Mg and Fe oxides in the catalysts, which was consistent with the
results of TPR, i.e. the introduction of the Mg component favored the reduction of the Fe2O3. XPS and
IR characterizations indicated that there were a large number of accessible Fe-OHs on the surface of
MgFe-400, and combining the Fe-OHs with the basic MgO may be related to the highly catalytic
performance.
Received 14th February 2015
Accepted 6th March 2015
DOI: 10.1039/c5ra02894a
www.rsc.org/advances
DEC, such as Ti(OBu)4, Bu2SnO, and BuSnCl3,6 the separation
problems between homogeneous catalysts and EMC as well as
1. Introduction
the recovery of catalysts still remained in such processes. Gan
et al. reported lithium diethylamide and lithiated carbon cata-
lysts for EMC, which performed high activity for the reaction,7
whereas these catalysts are very expensive limiting their prac-
tical application. Other heterogeneous catalytic systems
including the amorphous mesoporous aluminophosphate,8
basic metal oxides,9 Al-MCM-41 or Al-Zn-MCM-41,10 MOFs,11
ZIF-8 (ref. 12) and ZIF-67 (ref. 13) were also found to be active
catalysts for this reaction, but only <110 mmol hꢁ1 gcatꢁ1 TOFs
were obtained (Table 1). In this context, more effective catalystic
system is still highly desired in view of the lower cost and time-/
energy-saving, thereby increasing the yield and simplifying the
production separation process.
Since magnetic materials have emerged as efficient supports
for catalysts, which can facilitate the separation from the reac-
tion media using an external magnetic eld,14,15 in this study, we
prepared the magnetic binary Mg–Fe oxides as recoverable
catalysts for the synthesis of EMC via the transcarbonatation of
DMC with DEC. The resulting catalytic system exhibited supe-
rior activity and magnetical separability. XRD, BET, XPS, TPR,
Ethyl methyl carbonate (EMC) could be used as an important
methylation, ethylization or carbonylation agent for the
syntheses of numerous organic products. More importantly,
EMC could be employed as a co-solvent in nonaqueous elec-
trolytes to improve the energy density and discharge capacity of
lithium ion cells.1–3 Traditionally, EMC could be produced from
the esterication of methyl chloroformate with alcohol;4
however, this route generally involves environmentally
hazardous chemicals. Another method for the synthesis of EMC
is the transesterication of dimethyl carbonate (DMC) with
ethanol, but the separation of EMC is very difficult due to the
formation of three binary azeotropes.5 Alternatively, the trans-
carbonatation of DMC with diethyl carbonate (DEC) is consid-
ered as a promising approach to EMC, because all the reaction
mixtures (DEC, DMC and EMC) can be used as solvents in
nonaqueous electrolytes avoiding the separation problem.
Moreover, the composition of products can be adjusted through
controlling the conversion and the molar ratio of raw materials.
Although several homogenous catalytic systems have been
reported previously for the transcarbonatation of DMC with
¨
Mossbauer spectra, CO2-TPD and FT-IR studies were conducted
to explore the relationship between structure and performance,
and the corresponding mechanism was preliminarily proposed.
Moreover, the inuences of the catalysts with different
components, optimal catalyst activation, and reaction variables
(such as temperature, catalyst amount) were also investigated in
a batch reactor system.
aCentre for Green Chemistry and Catalysis, Lanzhou Institute of Chemical Physics,
Chinese Academy of Sciences, Lanzhou, 730000, China. E-mail: ydeng@licp.cas.cn;
Fax: +86-931-4968116; Tel: +86-931-4968116
bGraduate School of the Chinese Academy of Sciences, Beijing, 100039, China
cNational Nanotechnology Research Center, King Abdulaziz City for Science and
Technology, P.O. Box 6086, Riyadh 11442, Saudi Arabia
This journal is © The Royal Society of Chemistry 2015
RSC Adv., 2015, 5, 25849–25856 | 25849

View Article Online
RSC Advances
Paper
Table 1 Summary of the work reported on synthesis of EMC from DMC and DEC over heterogeneous catalystsa
TOF (mmol hꢁ1 gcat
)
ꢁ1
Catalyst
Operation conditions (temp., time, cat. amount)
DEC con. (%)
Lithiated diethylamide
Alumino-phosphate
MgO
Al-Zn-MCM-41
MOFs
25 ꢀC, 48 h, 0.2 wt%
50
50
20
25
22
—
41
81
17
93 ꢀC, 2 h, 5 wt%
103 ꢀC, 4 h, 2.4 wt%
44.2
85
50.1
50.7
83.4
175 ꢀC, 5 h, 6 wt%, the ow rate 1.5 mL hꢁ1
100 ꢀC, 3 h, 2 wt%
ZIF-8
ZIF-67
100 ꢀC, 3 h, 1 wt%
100 ꢀC, 24 h, 2 wt%
a
TOF: based on the mole of DEC converted per gram catalyst per hour.
A
57Co/Pd (activity y 25 m Ci) source and a-Fe standard were
2. Experimental
2.1 Catalysts preparation and characterization
used. The experimentally obtained spectra were tted to
mathematical processing according to the least squares
method.
The surface base properties of the catalysts were measured by
temperature programmed desorption (TPD) of CO2 and carried
out on TPD ow system equipped with an MS detector (DM300,
AMETEK, USA). In a typical experiment, the solid sample (200 mg
Chemicals used in this study including Mg(NO3)2$6H2O,
FeSO4$7H2O and Na2C2O4 obtained from Tianjin Chemical
Reagent Factory (analytical reagent, >99%).
The binary Mg–Fe oxides were prepared by co-precipitation
according to the literature procedures.16 A mixture of Mg(NO3)2-
$6H2O and FeSO4$7H2O in designated ratios was dissolved into
100 mL distilled water. Subsequently, 40 mmol Na2C2O4 was
dissolved in to 100 mL water at 80 ꢀC and then was added
dropwise into the former solution under stirring. The resulting
precipitate was aged, ltrated, washed, and calcined in static air
for 3 h. The resulted catalysts are denoted as xMgFe-T, x is molar
ratios of Mg and Fe, T is the calcination temperature.
ꢀ
with particle size 160–200 mm) was pretreated at 400 C for 1 h
under argon gas ow (50 mL minꢁ1) and then cooled to room
temperature. The sample was subsequently exposed to CO2
stream (50 mL minꢁ1) at room temperature for 1 h and ushed
again with argon gas for 1 h to remove any physico-adsorbed CO2.
ꢀ
The desorption prole was recorded at a heating rate of 10 C
minꢁ1 from room temperature to 400 C and maintained until
ꢀ
the MS signal of CO2 returned to the baseline. The quantitative
analysis for CO2 desorption is calculated based on the integration
of the corresponding TPD traces, preliminarily calibrated by the
injection of pure CO2 pulses.
Fourier transform infrared spectroscopy (FT-IR) transmission
data were collected from pressed catalyst disk made with KBr in
the range of 4000–400 cmꢁ1 with a Nicolet 5700 FT-IR.
The BET surface areas, pore volumes and average pore
diameters of catalysts were obtained with physisorption of N2
using a Micromeritics ASAP 2010.
X-ray diffraction (XRD) was measured on a Siemens D/max-
RB powder X-ray diffract meter. Diffraction patterns were
recorded with Cu Ka radiation (40 mA, 40 kV) over a 2q range of
20ꢀ to 80ꢀ and a position-sensitive detector using a step size of
0.01ꢀ and a step time of 0.15 s.
Surface analysis of the catalysts was performed by X-ray
photoelectron spectroscopy (XPS) on VG ESCALAB210 using a
Mg Ka radiation at a pass energy of 20 eV. The energy scale was
calibrated and corrected for charging using the C 1s (285.0 eV)
line as the binding energy reference.
2.2 EMC synthesis from DMC and DEC
Typical procedure for the synthesis of EMC: 50 mmol DMC,
50 mmol DEC and 0.104 g (1 wt%, based on the total mass of
reactants) catalyst were charged into a 50 mL ask equipped
with a magnetic stirring and a reux condenser. The reaction
mixture was then heated to 100 ꢀC with continuous stirring for
1.5 h. Aer the reaction, the mixture was cooled to room
temperature. The qualitative and quantitative analyses of the
liquid mixture were conducted with GC-MS (Agilent 6890/5973)
and GC (Agilent 7890) equipped with a SE-54 capillary column
and a FID detector. In the experiment to test the reusability of
catalyst, binary Mg–Fe catalyst was recovered using an external
magnetic eld and reused for the next time without further
processing.
Temperature-programmed reduction (TPR) of H2 was carried
out on TPR/TPD ow system equipped with a TCD detector. TPR
analysis was conducted with 10% H2/Ar (50 mL minꢁ1). In a
typical experiment, the solid sample (60 mg with particle size
160–200 mm) was pretreated at 400 ꢀC for 1 h under air ow
(50 mL minꢁ1) and then cooled to room temperature under
argon gas ow (50 mL mꢁin1ꢁ1). The prole was recorded at a
ꢀ
ꢀ
heating rate of 10 C min from room temperature to 850 C
and maintained at this temperature until the TCD signal of H2
returned to the baseline. Quantitative hydrogen consumptions
were obtained by use of TPR calibration runs using a standard
copper oxide sample (Micromeritics) in place of the catalyst.
3. Results and discussion
3.1 Results of the catalysts characterization
¨
The Mossbauer spectra were obtained at room temperature
with a WissEL (Bench-MB500, Germany) electromechanical
The physico-chemical properties of MgO, Fe2O3 and binary Mg–
Fe oxides are summarized in Table 2. Obviously, the BET
spectrometer working in
a constant acceleration mode.
25850 | RSC Adv., 2015, 5, 25849–25856
This journal is © The Royal Society of Chemistry 2015

View Article Online
Paper
RSC Advances
surface areas of pure MgO and Fe2O3 are much smaller than
that of binary Mg–Fe oxides. For the binary Mg–Fe oxides, the
porous volumes and BET surface area decreased with the
increase of the Fe content, which might be due to the formation
of different bulk Fe species. Meanwhile, the BET surface areas
of Mg–Fe oxides decreased monotonically from 90 to 23 m2 gꢁ1
with the increase of calcination temperatures from 400 to
600 ꢀC, which could be attributed to the sintering effect. The
obtained results suggested that both Fe content and calcination
temperatures have strong impacts on the catalyst textures.
The X-ray diffractions of the binary Mg–Fe oxides are shown
in Fig. 1. The characteristic diffraction peaks at 29.0, 33.7, 48.5
and 57.6ꢀ of Fe2O3 (JCPDS 24-0072) were detected for all the
samples demonstrating the formation of Fe2O3. Besides, the
peaks at 30.1, 35.5, 43.1 and 62.6ꢀ were observed in 0.5MgFe-400
(Fig. 1a), which could be well indexed to spinel MgFe2O4 (JCPDS
17-0464). Compared to the 0.5MgFe-400, the diffraction peaks
of MgO at 43 and 63ꢀ occurred for 2MgFe-400 (Fig. 1b) and
MgFe-400 (Fig. 1c), and the intensity of such diffraction peaks
become stronger with higher Mg content, while the character-
istic peaks of spinel MgFe2O4 disappeared. Moreover, the
intensity of diffraction peaks of the MgFe2O4 in MgFe-600
(Fig. 1e) was much higher than that of MgFe-500 (Fig. 1d),
indicating that the higher calcined temperature favored the
formation of spinel MgFe2O4 from MgO and Fe2O3.16
Fig. 1 XRD patterns of (a) 0.5MgFe-400 (b) 2MgFe-400 (c) MgFe-400
(d) MgFe-500 and (e) MgFe-600.
surface of catalysts and the Fe amounts were obviously
impacted by calcined temperature.
TPR results of MgO, Fe2O3 and MgFe-400 are described in
Fig. 3. The pure MgO does not show any hydrogen consumption
under the experimental conditions (Fig. 3a). As a reference, a
The X-ray photoelectron spectroscopy (XPS) analyses of
binary Mg–Fe oxides were also performed in this work. The
binding energy of Mg 2p and Fe 2p varied slightly around
49.4 ꢂ 0.1 eV and 710.9 ꢂ 0.2 eV, respectively, suggesting that
the chemical states of Mg and Fe species on the catalyst surface
were mainly Mg2+ and Fe3+. From the high resolution O 1s XPS
spectra (Fig. 2), two peaks could be observed in the prepared
catalysts with binding energy of 529.9 and 531.6 eV, which could
be assigned to lattice oxygen in oxides and hydroxyl groups
derived from surface-adsorbed oxygen, respectively.17,18 It was
found that the ratio of hydroxyl and oxide groups decreased
with the calcined temperature elevating from 400 to 600 ꢀC,
implying that the hydroxyl partly converted to lattice oxygen at
higher temperature. Furthermore, according to the XPS results,
the surface atom ratios of Fe3+ and Mg2+ over MgFe-400, -500
and -600 are 3.6, 7.4 and 5.4, respectively, which are higher than
the theoretical value 1, meaning that Fe species enriched on the
Fig. 2 O 1s XPS spectra of (a) MgFe-400 (b) MgFe-500 and (c) MgFe-
600.
Table 2 Physico-chemical properties of the binary Mg–Fe oxides
H2 consumptionc
Total basic amountd
a
b
SBET
dp
vp
Catalyst
(m2 gꢁ1
)
(nm)
(cm3 gꢁ1
)
(mmol gꢁ1
)
(mmol gꢁ1
)
2MgFe-400
MgFe-400
0.5MgFe-400
MgFe-500
MgO
156
90
88
62
29
23
10
9.4
13.6
14.3
16.1
26.3
24.4
25.6
0.54
0.45
0.41
0.32
0.21
0.15
0.05
3.6
5.5
5.9
—
0
—
249
205
93
113
69
MgFe-600
Fe2O3
52
32
5.8
a
b
c
Average pore size. Average pore volume. H2 consumption calculated with respect to standard CuO-TPR run under similar conditions.
Determined by TPD.
d
This journal is © The Royal Society of Chemistry 2015
RSC Adv., 2015, 5, 25849–25856 | 25851

View Article Online
RSC Advances
Paper
of Fe3O4, ultimately involving the formation of metallic iron.19 In
comparison with TPR proles of pure Fe2O3, two reduction peaks
at 422 and 668 ꢀC were found over MgFe-400 (Fig. 3c). Based on
the hydrogen consumed at the rst peak (1.46 mmol gꢁ1), it could
be determine that the actual ratio of Mg/Fe in the MgFe-400 was
1.17 : 1. However, the hydrogen consumed at the second peak
(4.07 mmol gꢁ1) was about 9 times lesser than the corresponding
consumption of Fe3O4 to Fe (35 mmol gꢁ1) in this catalyst,
indicating that only part of Fe3O4 was reduced to Fe. Additionally,
ꢀ
the shi in the reduction temperature from 442 to 422 C indi-
cated that some sort of interaction occurred between Mg species
and Fe oxide, and the introduction of Mg component may favor
the reduction of the Fe oxide.
¨
The Mossbauer spectra for the Fe2O3 and binary Mg–Fe
oxides at room temperature are shown in Fig. 4, and the
Fig. 3 TPR-H2 profiles of (a) MgO (b) Fe2O3 and (c) Mg–Fe-400.
¨
Mossbauer parameters are listed in Table 3. The formation of
Fe2O3 was revealed by the sextet with the isomer shi (IS) of
0.367 mm sꢁ1 and quadrupole splitting (QS) ꢁ0.178 mm sꢁ1 in
sharp peak centered at 442 ꢀC and a broad peak at 727 ꢀC along
with a shoulder at 800 ꢀC could be observed in pure Fe2O3
(Fig. 3b). Generally, the peak at 442 ꢀC is assigned to the
reduction of Fe2O3 to Fe3O4, and the hydrogen consumed at this
peak (2.05 mmol gꢁ1) corresponded well to the total consump-
tion of Fe2O3 to Fe3O4 (2.08 mmol gꢁ1). Besides, the higher
temperature peak could be ascribed to the subsequent reduction
¨
the Mossbauer spectra as shown in Fig. 4a. As Mg was intro-
duced into the iron oxide (Fig. 4b), one sextet and two para-
magnetic doublets were observed. In the two doublets, the
doublet with IS ¼ 0.331 mm sꢁ1 and QS ¼ 0.653 mm sꢁ1 is
typical of superparamagnetic Fe3+ species on the octahedral
sites, denoted as Fe3+ (A) in Table 3, which may be assigned to
MgFe2O4; another doublet peak with IS ¼ 0.377 mm sꢁ1 and
Fig. 4 Mossbauer spectra of Fe2O3 and binary Mg–Fe oxides recorded at room temperature.
¨
25852 | RSC Adv., 2015, 5, 25849–25856
This journal is © The Royal Society of Chemistry 2015

View Article Online
Paper
RSC Advances
Table 3 Mossbauer parameters of Fe2O3 and binary Mg–Fe oxides at room temperaturea
¨
IS
QS
Hhf
(T) ꢂ 0.01
G
Area
(%)
Sample
Component
(mm sꢁ1) ꢂ 0.002
(mm sꢁ1) ꢂ 0.01
(mm sꢁ1) ꢂ 0.006
Fe2O3
Fe2O3
0.367
0.315
0.331
0.377
0.315
0.317
ꢁ0.178
0.003
0.653
2.612
0.011
0.002
51.33
49.43
—
—
49.92
49.91
0.37
0.57
—
—
0.53
0.61
100
Fe2O3
89.19
9.19
1.62
100
0.5MgFe-400
Fe3+ (A)
Fe3+ (B)
Fe2O3
MgFe-400
2MgFe-400
Fe2O3
100
a
IS: isomer shi, QS: quadrupole splitting, Hhf: hyper ne magnetic elds, G: line width.
QS ¼ 2.612 mm sꢁ1 could be ascribed to high spin Fe3+ on the in the region of 50–350 ꢀC, which were derived from the weak
tetrahedral sites, denoted as Fe3+ (B). For MgFe-400 (Fig. 4c) and medium strength basic sites.16 From the quantitative
and 2MgFe-400 (Fig. 4d), the raw spectra can also be tted to results of CO2-TPD (Table 2), it can be seen that the total basic
one sextet but the relative intensity decreased compared to the amount decreased from 249 to 93 mmol gꢁ1 with the decrease
pure Fe2O3. The reason may be due to the formation of MgO, of the Mg content, suggesting that the basic site was mainly
which could be also conrmed by XRD. At the same time, the due to the MgO phase in the binary Mg–Fe oxides. Moreover,
line width of the sextuplet in MgFe-400 was substantially the total basic amount sharply decreased from 205 to 52 mmol
broadened in comparison to the corresponding sextuplet of gꢁ1 with the calcination temperature elevating from 400 to 600
Fe2O3, indicating the high dispersion of Fe2O3 in the sample, ꢀC, which probably because of the decrease of content of MgO
which may result from the occurrence of a random distribu- in catalyst.
tion of constituent Fe3+ cations in the presence of Mg2+
.
In
Fig. 6 displays the FT-IR spectra of the fresh (a) and reused
20
addition, the value of Hhf (Hhf ¼ 49.92 T) for MgFe-400 was MgFe-400 (b). The signal at around 3400 cmꢁ1 could be attrib-
lower than that of the pure Fe2O3 (Hhf ¼ 51.33 T), which uted to the O–H stretch derived from the surface Fe-OH, and the
demonstrated that the coordination environment of the Fe intensity of this peak decreased aer used 10 times because
atom changed. The obtained results were consistent with the Fe-OHs were partially replaced by Fe–OCH3 bonds. Meanwhile,
TPR analyses, i.e. the addition of Mg component favored the the absorptions at 1632, 1100 and 816 cmꢁ1 were detected for
reduction of the Fe2O3.
fresh catalyst, which could be assigned to the H–O–H bending,
The base properties of these oxide catalysts were studied by Fe–O bending and Fe–OH–Fe bending, respectively.22
TPD of adsorbed CO2. The CO2-TPD results of Fe2O3, MgO and Compared to the fresh one, a new band at 2961 cmꢁ1 corre-
binary Mg–Fe oxide samples are exhibited in Fig. 5. For the sponding to the C–H stretching vibration was observed, which
MgO sample (Fig. 5a), a broad peak centered at 250 ꢀC was may be due to trace of methoxide species residing on the used
observed, which was ascribed to medium strength basic sites catalyst surface;23 and the peak 1024 cmꢁ1 owing to C–O
of O2ꢁ
.
In comparison with the MgO, only a small peak vibration of methoxide group further conrmed that the
21
meaning the less basic sites was examined on the CO2-TPD methoxide group adsorbed on the catalyst surface. In addition,
curves of pure Fe2O3 (Fig. 5b). It was found that all binary Mg– bicarbonate species formation involving surface hydroxyl
Fe oxide samples possessed broadened CO2 desorption peaks groups emerged on reused catalyst, which characterized a C–OH
Fig. 5 CO2-TPD profiles of (a) MgO (b) Fe2O3 (c) MgFe-600 (d) MgFe-
500 (e) 0.5MgFe-400 (f) MgFe-400 and (g) 2MgFe-400.
Fig. 6 FT-IR patterns of (a) fresh and (b) reused MgFe-400.
This journal is © The Royal Society of Chemistry 2015
RSC Adv., 2015, 5, 25849–25856 | 25853

View Article Online
RSC Advances
Paper
Table 4 Results of transcarbonatation of DMC and DEC over various catalystsa
Yieldc (%)
TOF (mmol hꢁ1 gcat
)
ꢁ1
Entry
Catalyst
n(DMC)/n(DEC)
1
2
3
4
5
6
7
8
9
10
MgO
Fe2O3
1 : 1
1 : 1
1 : 1
1 : 1
1 : 1
1 : 1
1 : 1
1 : 1
2 : 1
1 : 2
7
9
4
46
51
39
16
2
22
30
13
153
170
130
55
8
220
226
b
MgO + Fe2O3
0.5MgFe-400
MgFe-400
2MgFe-400
MgFe-500
MgFe-600
MgFe-400
MgFe-400
66
34
a
Reaction conditions: 50 mmol DMC, 50 mmol DEC, 0.104 g catalyst, 100 ꢀC, 1.5 h. b Physically mixed MgO and Fe2O3 (molar ratio of Mg/Fe ¼ 1 : 1).
Determined by GC, calculated in terms of DEC (moles of EMC/2 ꢃ moles of DEC). TOF: calculated based on the mole of DEC converted per gram
c
catalyst per hour.
bending mode at 1260 cmꢁ1 as well as symmetric O–C–O EMC was sharply decreased with the calcination temperature
stretching bands at 1480 cmꢁ1
.
from 400 to 600 C (entries 5, 7 and 8), indicating that 400 C
was the proper activation temperature for better catalytic
activity. Generally, it was widely accepted that trans-
esterication reaction was efficiently catalyzed by acid–base
catalysts.25 Therefore, it is reasonable to infer that the sharply
decrease of catalytic activity might be due to the decrease of
total basic sites (Table 2) with the calcination temperature
24
ꢀ
ꢀ
3.2 Results for synthesis of EMC
As shown in Table 4, the transcarbonatation of DMC and DEC
for EMC was investigated over various catalysts with different
proportions and activation temperatures. Only <10% of EMC
yield were obtained over MgO, Fe2O3 and physical mixed oxide,
indicating that those catalysts performed low activity for such
transcarbonatation process (entries 1–3). As the binary Mg–Fe
oxides with different proportions were used as catalysts,
(entries 4–6), the yield of EMC was greatly improved. Above all,
when the ratio of Mg and Fe was 1, 51% yield of EMC was
achieved. It's worth mentioning that the TOF could be up to
ꢀ
elevating from 400 to 600 C. Besides, the ratios of DMC and
DEC from 2 : 1 to 1 : 2 were also studied (entries 9 and 10), and
the yield of EMC could reach 67.8%. In this way, the molar
ratio of the products mixture could be changed through
adjusting the molar ratio of raw material.
ꢁ1
3.3 Effect of reaction conditions
170 mmol hꢁ1 gcat over MgFe-400, which was better than
previously reported results to the best of our knowledge.6–13 If In order to optimize reaction conditions, the inuence of
further increased the ratio of Mg and Fe to 2, the yield of EMC various reaction parameters was investigated. The impact of the
decreased to 39%. Obviously, the catalytic activity of binary amount of catalyst charged on the EMC synthesis was rst
Mg–Fe oxides was higher than that of the physical mixture, examined varying from 0 to 3 wt% (respect to the total mass of
demonstrating that the integration of Mg and Fe components raw materials) under 100 ꢀC. As seen in Fig. 7A, EMC yield was
could remarkably enhance the catalytic activity and some very low without any catalyst. As 1 wt% catalyst was charged,
synergistic effect probably occurred. Additionally, the yield of 51% yield of EMC with >99% selectivity was achieved, however,
Fig. 7 Effect of reaction conditions on the catalytic performance of MgFe-400 influence of (A) the amount of catalyst charged; (B) the
temperature; reaction conditions: 50 mmol DMC, 50 mmol DEC, 100 ꢀC, 1.5 h.
25854 | RSC Adv., 2015, 5, 25849–25856
This journal is © The Royal Society of Chemistry 2015

View Article Online
Paper
RSC Advances
chemisorbed on the Lewis acid sites and base sites of the
catalyst to form intermediate. In this case, MgO activates the
carbonyls of reactants, and the surface –OH groups play an
important role in formation of reaction intermediate simulta-
neously. A possible mechanism for the transcarbonatation
between DMC and DEC was proposed over the binary Mg–Fe
oxides system (Scheme 1). First, chemisorption of DEC and
DMC could occur on the Mg2+ of MgFe-400. Then two adjacent
OH groups attack DEC and DMC carbonyl carbon to form the
corresponding alkoxy groups and bicarbonate intermediate.
Once the intermediate is produced, it would migrate to different
regions where the alkoxy groups are available to yield mixtures
of dialkyl products.
4. Conclusions
Fig.
8 The reusability of MgFe-400 catalyst for EMC synthesis.
Reaction conditions: 50 mmol DMC, 50 mmol DEC, 0.104 g catalyst,
100 ꢀC, 1.5 h.
In summary, magnetic binary Mg–Fe oxides with different
proportions were prepared by co-precipitation method, and
MgFe-400 displayed high catalytic performance for the
synthesis EMC from DMC and DEC (the highest TOF can reach
226 mmol hꢁ1 gcatꢁ1). It was found that the presence of Fe-OHs
and basic sites of the MgO played a critical role on the activation
of the reactants in the transcarbonatation reaction. Moreover,
the catalyst could be easily isolated using an external magnetic
eld and recovered for 10 runs without signicant loss of
catalytic activity.
as 2 wt% catalyst was used, no remarkable increase of EMC yield
was observed, which suggested that 1 wt% catalyst might be
sufficient for this process. In addition, the inuence of
temperature on EMC synthesis was also tested (Fig. 7B). It was
found that the yield of EMC increased with the raising of
reaction temperature, and the maximum yield (51%) was
ꢀ
ꢀ
obtained at 100 C. Therefore, 100 C was the proper reaction
temperature for the transcarbonatation process.
Acknowledgements
3.4 Recycling of the catalyst
This work was supported by National Natural Science Founda-
tion of China (no. 21002107).
The reusability of solid catalysts is a key parameter in deter-
mining their suitability in practical applications. Therefore,
attempts were made to reuse the MgFe-400 catalyst 10 times by
simple magnetic separation, and reused without any process-
ing. The obtained results (Fig. 8) indicated that the catalyst
could retain its activity performance during the recycles. The
ICP-AES analyses showed that only 3 ppm of Mg and 7 ppm of
Fe content were leached aer 10 recycles, which indicated that
MgFe-400 catalyst displayed excellent stability in this process.
Notes and references
1 H. Okuno and H. Koshina, US. Pat. 5,484,669, 1996.
2 E. J. Pichta and W. K. Behl, J. Power Sources, 2000, 88, 192–
196.
3 H. Okuno, H. Koshina and A. Morta, J. Power Sources, 1997,
67, 341.
4 M. J. D'Souza, J. A. Knapp, G. A. Fernandez-Bueno and
D. N. Kevill, Int. J. Mol. Sci., 2012, 13, 665–682.
3.5 The reaction mechanism
´
5 T. Keller, J. Holtbruegge and A. Gorak, Chem. Eng. J., 2012,
180, 309–322.
The mechanism for the transcarbonatation of DMC and DEC
has been reported previously,8 in which DMC and DEC were
6 G. Zhuo, Z. Shen and X. Jiang, Chin. J. Catal., 2004, 25, 171–
172.
7 H. Gan, M. Palazzo and E. S. Takeuchi, US. Pat. 5,962,720,
1999.
8 J. Shi, G. Liu, Z. Fan, L. Nie, Z. Zhang, W. Zhang, Q. Huo,
W. Yan and M. Jia, Catal. Commun., 2011, 12, 721–725.
9 Z. Shen, X. Jiang and W. Zhao, Catal. Lett., 2003, 91, 63–67.
10 A. Palani, N. Gokulakrishnan, M. Palanichamy and
A. Pandurangan, Appl. Catal., A, 2006, 304, 152–158.
11 Y. Zhou, J. Song, S. Liang, S. Hu, H. Liu, T. Jiang and B. Han,
J. Mol. Catal. A: Chem., 2009, 308, 68–72.
12 X. Zhou, H. Zhang, G. Wang, Z. Yao, Y. Tang and S. Zheng, J.
Mol. Catal. A: Chem., 2013, 366, 43–47.
Scheme 1 A possible mechanism for the transcarbonatation between
DMC with DEC.
This journal is © The Royal Society of Chemistry 2015
RSC Adv., 2015, 5, 25849–25856 | 25855

View Article Online
RSC Advances
Paper
13 L. Yang, L. Yu, M. Sun and C. Gao, Catal. Commun., 2014, 54, 20 H. Haneda, H. Yamamura, A. Watanabe and S. Shirasaka, J.
86–90. Solid State Chem., 1987, 68, 273–284.
14 S. Luo, X. Zheng and J. Cheng, Chem. Commun., 2008, 5719– 21 V. R. Chintareddy and M. L. Kantam, Catal. Surv. Asia, 2011,
5721. 15, 89–110.
15 A. Lu, W. Schmidt, N. Matoussevitch, H. Bonnemann, 22 A. Baldermann, R. Dohrmann, S. Kauold, C. Nickel,
B. Spliethoff, B. Tesche, E. Bill, W. Kiefer and F. Schuth,
Angew. Chem., Int. Ed., 2004, 43, 4303–4306.
I. Letofsky-papst and M. Dietzel, Clay Miner., 2014, 49,
391–415.
16 J. Shang, S. Liu, L. Lu, X. Ma, Y. He and Y. Deng, Catal. 23 M. Sankar, C. M. Nair, K. V. G. K. Murty and P. Manikandan,
Commun., 2012, 28, 13–17. Appl. Catal., A, 2006, 312, 108–114.
17 S. Kaliaguine, A. Van Neste, V. Szabo, J. E. Gallot, M. Bassir 24 V. K. Dıez, C. A. Ferretti, P. A. Torresi, C. R. Apesteguıa and
´
´
and R. Muzychuck, Appl. Catal., A, 2001, 209, 345–358.
18 J. P. Dacquin and C. Dujardin, J. Catal., 2008, 253, 37–49.
19 A. Venugopal and M. S. Scurrell, Appl. Catal., A, 2004, 258,
241–249.
J. I. Di Cosimo, Catal. Today, 2011, 173, 21–27.
25 J. Otera, Chem. Rev., 1993, 93, 1449–1470.
25856 | RSC Adv., 2015, 5, 25849–25856
This journal is © The Royal Society of Chemistry 2015