La-modified mesoporous Mg-Al mixed oxides: Effective and stable base catalysts for the synthesis of dimethyl carbonate from methyl carbamate and methanol

Article abstract of DOI:10.1039/c5cy01712b

A series of La-containing Mg-Al hydrotalcite-like (HTl) precursors with different La contents (Mg²⁺:Al³⁺:La³⁺ = 3:1:x, where x varies from 0 to 1.0) were synthesized using a co-precipitation method followed by hydrothermal treatment. X-ray diffraction and thermogravimetric measurements demonstrated that the yield of the HTl phase decreased with increasing La content. The La-modified Mg-Al mixed oxides (HTC-La) were then obtained by thermal decomposition of the corresponding HTl precursors, and the mesoporous structure was formed during calcination. It was demonstrated that the structure and surface basic properties of the HTC-La samples strongly depended on the amount of La additive. Simultaneously, the resulting HTC-La materials were used as solid base catalysts for the synthesis of dimethyl carbonate (DMC) from methyl carbamate (MC) and methanol. Then, the correlation between their basic properties and catalytic performance was studied in detail. The incorporation of a suitable amount of La into HTC-La catalysts was beneficial for the production of DMC, and a DMC yield of 54.3% with a high DMC selectivity of 80.9% could be achieved when x was tuned to 0.5 under the optimized reaction conditions. In addition, the HTC-La catalyst could be readily recycled while maintaining high catalytic activity and selectivity for DMC. Furthermore, in situ FTIR experiments were carried out to elucidate the adsorption behaviours of the reactants. On the basis of the experimental results, a plausible basic catalytic mechanism wherein MC and methanol were activated simultaneously on the basic sites of the catalyst was proposed for this catalytic reaction.

Full text of DOI:10.1039/c5cy01712b

Catalysis
Science &
Technology
View Article Online
View Journal
PAPER
La-modified mesoporous Mg–Al mixed oxides:
effective and stable base catalysts for the
synthesis of dimethyl carbonate from methyl
carbamate and methanol†
Cite this: DOI: 10.1039/c5cy01712b
a
ab
Jie Ma, a Haiwen Yu,a Jingzhu Shena
*
*
Dengfeng Wang, Xuelan Zhang,
and Wei Weib
A series of La-containing Mg–Al hydrotalcite-like (HTl) precursors with different La contents (Mg2+ : Al3+
:
La3+ = 3 : 1 : x, where x varies from 0 to 1.0) were synthesized using a co-precipitation method followed by
hydrothermal treatment. X-ray diffraction and thermogravimetric measurements demonstrated that the
yield of the HTl phase decreased with increasing La content. The La-modified Mg–Al mixed oxides (HTC-
La) were then obtained by thermal decomposition of the corresponding HTl precursors, and the meso-
porous structure was formed during calcination. It was demonstrated that the structure and surface basic
properties of the HTC-La samples strongly depended on the amount of La additive. Simultaneously, the
resulting HTC-La materials were used as solid base catalysts for the synthesis of dimethyl carbonate (DMC)
from methyl carbamate (MC) and methanol. Then, the correlation between their basic properties and cata-
lytic performance was studied in detail. The incorporation of a suitable amount of La into HTC-La catalysts
was beneficial for the production of DMC, and a DMC yield of 54.3% with a high DMC selectivity of 80.9%
could be achieved when x was tuned to 0.5 under the optimized reaction conditions. In addition, the HTC-
La catalyst could be readily recycled while maintaining high catalytic activity and selectivity for DMC. Fur-
thermore, in situ FTIR experiments were carried out to elucidate the adsorption behaviours of the reac-
tants. On the basis of the experimental results, a plausible basic catalytic mechanism wherein MC and
methanol were activated simultaneously on the basic sites of the catalyst was proposed for this catalytic
reaction.
Received 8th October 2015,
Accepted 30th October 2015
DOI: 10.1039/c5cy01712b
www.rsc.org/catalysis
toxicity. More important, it is extensively applied as an
oxygen-containing fuel additive in place of toxic and less bio-
1. Introduction
Recently, much effort has been devoted to the development of
new synthesis methods based on CO2 as a feedstock for
organic carbonates owing to their excellent properties and
commercial applications. As an important homologue of the
dialkyl carbonate family, dimethyl carbonate (DMC) has
received much attention. As an environmentally benign build-
ing block, DMC can replace phosgene, dimethyl sulphate,
chloromethane, and methyl chloroformate as a carbonylation,
methylation, esterification, or ester interchange reagent for
organic synthesis without pollution. DMC is also used as a
green solvent and flavoring agent of foodstuff due to its low
degradable MTBE additives. Moreover, DMC has been used as
an electrolyte due to its high dielectric constant.1–4 Up to
now, a number of synthetic processes are known for the pro-
duction of DMC including phosgenation of methanol, oxida-
tion/carbonation of methanol, and transesterification of
organic carbonates. However, each of the aforementioned pro-
cesses suffers from the corresponding shortcomings such as
being poisonous, being easily explosive, and having high
investment and production costs.5–7 Recently, CO2, a readily
available, inexpensive and environmentally acceptable mate-
rial, has been used as a raw material for DMC synthesis. Nev-
ertheless, the direct synthesis of DMC from CO2 and metha-
nol is still far from practical application because of the
difficulty in activation of carbon dioxide, reactivation of the
catalyst and thermodynamic limitation.8–10 In order to over-
come these shortcomings, an alternative approach is to pre-
pare DMC from urea and methanol, which is widely consid-
ered as an indirect route for utilization of CO2 because the
a College of Chemistry, Chemical Engineering and Materials Science, Zaozhuang
University, Zaozhuang, 277160, PR China. E-mail: bitwdf@163.com
b State Key Laboratory of Coal Conversion, Institute of Coal Chemistry, Chinese
Academy of Sciences, Taiyuan, 030001, PR China
† Electronic supplementary information (ESI) available. See DOI: 10.1039/
c5cy01712b
This journal is © The Royal Society of Chemistry 2015
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
released NH3 can be easily separated, recycled and further uti-
lized to produce urea by reaction with CO2. Besides, no water
is formed during this process.11 Therefore, the subsequent
separation and purification would be simple and the cost
could be reduced markedly. It is well known that urea
methanolysis is a typical stepwise reaction: urea firstly reacts
with methanol to produce the intermediate methyl carbamate
(MC), which is further converted to DMC by reaction with
methanol. It is generally accepted that the first step is fast
and highly selective even without catalysts, but the DMC syn-
thesis from MC and methanol is more difficult.12,13 Hence,
the second reaction is usually considered to be the rate-
control step, and dividing this process into two isolated steps
would be a more promising method. Therefore, the key to
improve the DMC synthesis via methanolysis of urea is to
develop effective catalysts towards the reaction of MC and
methanol.
Several kinds of catalysts including organic tin, inorganic
salts, ionic liquids, pure metal oxides and Fe2O3 supported
catalysts have been tested towards the direct reaction of urea
and methanol or its analogues.14–18 Among them, zinc-
containing catalysts such as ZnO, Zn–Ce, Zn–Ca and Zn–Al
mixed oxides presented high catalytic activity.19–22 Especially,
ZnO–CeO2–La2O3 could improve DMC synthesis remarkably
and give 50.4% DMC yield.23 However, the main problem
associated with zinc-containing catalysts might be the easy
leaching of the active component, which has been proved by
several researchers. For instance, Zhao et al. reported that
ZnO was converted to the homogeneous species when it cata-
lyzed urea alcoholysis.24 An et al. discovered that the ZnO
phase of the ZnO–PbO catalyst was seriously leached out
when it promoted the reaction of ethyl carbamate and etha-
nol.25 Recently, Fujita and co-workers further pointed out
that synthesis of glycerol carbonate from urea glycerolysis
using Zn-containing solid catalysts took place homogeneously
but not heterogeneously.26 With respect to the isolated sec-
ond reaction, solid bases were tested in a batch reactor, but
the DMC yield was below 20%.12 Our group reported that
using ZnCl2 and LaCl3 as catalysts, the DMC yield could
reach 33.6% and 53.7%, respectively.27–29 However, it is well
known that catalyst recovery is a great inconvenience for
industrial production using homogeneous catalysts. In order
to develop heterogeneous catalysts, we have synthesized a
Zn–Fe mixed oxide that gave a DMC yield of 30.7%.30 Unfor-
tunately, it also had a ZnO phase, which might not resist the
flushing of the reactants in the continuous reaction, and its
catalytic activity was still relatively low. Thus, it is still a chal-
lenging task to develop new heterogeneous catalysts
exhibiting good performance for the synthesis of DMC from
MC and methanol.
isomorphously substituted by Al3+ or other trivalent metal
cations, resulting in the formation of positively charged
layers, which in turn are charge compensated by the anions
2−
(An−) such as CO3 or OH− present in the interlayer.31 More
important, the mixed oxides obtained by the thermal decom-
position of these compounds exhibit homogeneous disper-
sion of M2+ and M3+ at the atomic level, high stability against
sintering, high surface area and tunable basic properties.32–34
Therefore, HTlcs with different compositions are among the
most investigated catalyst precursors because of the remark-
able properties of the final catalysts, which have been used in
many reactions such as aldol condensation, Michael addi-
tion, transesterification and so on.35–38 In our previous work,
we have reported that Mg–Al mixed oxide catalysts were effec-
tive solid base catalysts for the reaction of urea and phenol.39
Very recently, we have discovered the promotion effect of
transition metal on the basicity and activity of calcined
hydrotalcite catalysts for diethyl carbonate synthesis.40 How-
ever, the effect of rare earth element doping on the basicity
of calcined hydrotalcite catalysts and their catalytic perfor-
mance in DMC synthesis remains unknown. As part of our
continuous research work, here, we wish to show that La-
modified Mg–Al mixed oxides (HTC-La) obtained by calcina-
tion of the corresponding La-containing hydrotalcite-like
(HTl) precursors are efficient solid base catalysts for DMC
synthesis from MC and methanol, and the DMC yield could
reach 54.3% under optimal reaction conditions. Besides,
their activity could be tailored by simple adjustment of the
amount of La additive. The correlations between the textural,
chemical, structural, and basic properties of the HTC-La cata-
lysts and their catalytic performances in MC methanolysis
were investigated. At the same time, the catalytic activities
and recyclability of the catalysts were systematically studied.
Finally, we proposed a plausible reaction mechanism wherein
MC and methanol were simultaneously activated on the sur-
face basic sites of the HTC-La catalysts based on in situ FTIR
experimental results.
2. Experimental
2.1 Catalyst preparation
La-containing hydrotalcite-like compounds with various
Mg2+ : Al3+ : La3+ atomic ratios were synthesized using a co-
precipitation method followed by hydrothermal treatment.41
The Mg2+ : Al3+ : La3+ atomic ratio in the starting solution was
kept at 3 : 1 : x, where x varied from 0 to 1.0. In a typical pro-
cedure, two aqueous solutions, a 100 mL solution of Mg2+
,
Al3+, and La3+ nitrates and a 100 mL mixed solution of NaOH
and Na2CO3 precipitants, were added dropwise to 100 mL of
deionized water under vigorous stirring. The addition was
performed over 1.0–1.5 h and the pH value was maintained
close to 10 by addition of suitable amounts of 2 M NaOH
solution. The resulting gel-like slurry was transferred into
autoclaves and hydrothermally treated at 80 °C for 24 h.
Then, the as-prepared product was collected after being fil-
tered and washed with deionized water until the pH of the
Hydrotalcite (HT) or hydrotalcite-like compounds (HTlcs)
are layered double hydroxides belonging to anionic clay with
the general formula [M(1−x)2+Mx3+(OH)2]x+·(An−)x/n·mH2O, where
M2+ and M3+ are the divalent and trivalent cations, respec-
tively. Their structure is very similar to that of brucite, in
which some of the divalent metal ions (M2+ = Mg2+) could be
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Catalysis Science & Technology
Paper
filter liquor became 7.0. Finally, the precipitate was dried at
100 °C for 12 h to obtain HTl precursors.
cell for scanning at suitable temperatures, and the spectra
were recorded with a Nicolet Magna 550 FTIR spectrometer
in the region of 4000–400 cm−1 at a resolution of 4 cm−1. In
the present work, four types of experiments were performed,
namely: (a) MC thermal decomposition: 15 mg of MC was
ground and pressed to a self-supporting pellet, and then the
disk was put into the closed cell. (b) MC adsorption on the
catalyst: the sample was obtained by mixing 14 mg of the cat-
alyst and 1 mg of MC, which were then ground and pressed
to a self-supporting pellet, and then the pellet was put into
the closed cell. (c) Methanol adsorption on the catalyst: 15
mg of the catalyst was finely ground and pressed to a self-
supporting pellet. After pretreatment under vacuum for 3 h,
10 μL of methanol was injected into the closed cell. (d) Co-
adsorption of MC and methanol on the catalyst: the sample
was obtained by mixing 14 mg of the catalyst and 1 mg of
The as-prepared HTlcs were further calcined in air at
500 °C for 7 h at a heating-up rate of 5 °C min−1 to obtain the
final mixed oxide catalysts. The synthesized hydrotalcite-like
precursors and corresponding mixed oxides were denoted as
HT-NLa before calcination and HTC-NLa after calcination,
respectively, where N/10 represented the La3+ : Al3+ atomic ratio
in the starting solution. For comparison, the La2O2CO3 catalyst
was prepared under the same conditions using LaIJNO3)3·5H2O
as the La source and Na2CO3 as the precipitant, followed by
calcination at 500 °C for 7 h under a CO2 atmosphere.
2.2 Characterization
Thermogravimetric analysis (TGA) was performed on
a
NETZSCH STA-409 thermal analyzer under nitrogen (50 mL
min−1) with a heating rate of 10 °C min−1 from room temper-
ature to 800 °C to study the thermal decomposition behavior
of the catalyst precursors. The powder X-ray diffraction (XRD)
patterns of the samples were recorded on a Rigaku Miniflex
diffractometer using a Cu target with a Ni filter in the 2θ
range of 5–80°. Elemental chemical analysis of the samples
was performed using inductively coupled plasma-optical
(ICP) emission spectroscopy (Thermo iCAP 6300). Nitrogen
adsorption–desorption isotherms were obtained at liquid
nitrogen temperature, −196 °C, with a Micromeritics ASAP-
2020 instrument (Norcross, GA), using static adsorption pro-
cedures. The samples were degassed at 200 °C for 5 h prior
to the measurement. The BET isotherm and BJH pore size
distribution were used to obtain the specific surface area and
pore volume, respectively. The morphology of the samples
was investigated using a Hitachi S4800 scanning electron
microscope (SEM) with an accelerating voltage of 20.00 kV.
X-ray photoelectron spectroscopy (XPS) measurements were
performed on a Thermo Scientific k-Alpha spectrometer
equipped with an Al anode (Al Kα = 1486.6 eV) operating at
72 W and a spot size of 400 μm under ultrahigh vacuum
(10−7 Pa). The binding energies were calibrated internally for
surface charging by adventitious carbon deposit C (1s) with
Eb = 284.5 eV.
The basic properties of the catalysts were determined by
CO2 temperature-programmed desorption (CO2-TPD). The
catalysts (0.1 g, 40–60 mesh) were placed in a quartz reactor
bed. After 5 h of pretreatment in Ar flow at 500 °C, the cata-
lysts were cooled to room temperature. Then, they were satu-
rated with pure CO2 (30 L min−1) using a six-way valve for 3 h
and then flushed with Ar (40 L min−1) flow to remove all
physically adsorbed molecules. Afterward, the CO2-TPD exper-
iments were started from 30 to 500 °C with a heating rate of
10 °C min−1 under Ar flow (40 L min−1), and the desorbed
CO2 was detected by using an AMETEK mass spectrometer.
The CO2 peak area was quantitatively calibrated by injecting
CO2 pulses.
MC, which were then ground and pressed to a self-
supporting pellet. After pretreatment under vacuum for 3 h,
various amounts of methanol (0, 10, 20, 40, 60, 80, and 100
μL) were introduced into the closed cell, and the cell was rap-
idly heated to 180 °C. Then, the FTIR spectra of co-
adsorption of reactants on the catalyst were collected at this
temperature. With respect to (a), (b) and (c) experiments, the
FTIR cell was heated from room temperature to 200 °C, and
the spectra were collected once every 2 min in this work.
2.3 Catalytic test
The catalytic activities of the catalysts for the synthesis of
DMC from MC and methanol were evaluated in a 100 mL
stainless steel autoclave reactor equipped with electric
heating, a reflux column and a magnetic stirrer under the
assigned conditions. In a typical process, 0.5 g of the freshly
prepared catalysts, 0.05 mol (3.75 g) of MC and 1 mol (32 g)
of methanol were placed in the reactor. Then, it was rapidly
heated to 200 °C and kept for 6 h with magnetic stirring.
During the reaction, the temperature error was less than 2
°C. After the reaction, the autoclave was cooled to room tem-
perature, and the product mixture in the autoclave was
weighed, clarified and determined using a gas chromato-
graph with a PEG-20M capillary column and a flame ioniza-
tion detector (FID). The column was temperature-
programmed from 80 to 180 °C.
3. Results and discussion
3.1 Textural and structural properties of the prepared
materials
The XRD patterns of the HTl precursors with Mg2+ : La3+ : Al3+
from 3 : 1 : 0 to 3 : 1 : 1 are shown in Fig. 1(a). Obviously, HT-
0La and HT-2La precursors exhibited profiles similar to the
pure layered double hydroxide structure with sharp and sym-
metric peaks for the (003), (006), (110) and (113) planes and
broad asymmetric peaks for the (009), (015), and (018) planes
(JCPDS 70-2151).31 For HT-5La, HT-8La and HT-10La sam-
ples, the peaks of a mixture of LaIJOH)3 (JCPDS 06-0585) and
La2IJCO3)2IJOH)2 (JCPDS 70-1774) were also detected.42–44
In situ Fourier transform infrared spectroscopic (FTIR)
experiments were performed in a closed stainless FTIR steel
This journal is © The Royal Society of Chemistry 2015
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
Fig. 1 XRD patterns of HT-La precursors with different La contents (a); HTC-La mixed oxides with different La contents and La2O2CO3 catalysts (b).
According to previous reports, La favored the formation of
lanthanum carbonate species in the very early stages of the
co-precipitation because it possessed the lowest electronega-
tivity compared to other rare earth elements.42,43,45 Mean-
while, La preferred to be located in the interlayer gallery of
HT as separate hydroxide and hydroxyl carbonate phases due
to its relatively large ionic radius.42 On the other hand, the
peak intensities and sharpness of the (003) and (006) planes,
which were directly proportional to the crystallinity of the
material, were observed to decrease with increasing La con-
tent of the precursors. This might be ascribed to the follow-
ing reasons. Firstly, La3+ possessed a higher ionic radius
(0.103 nm) than Mg2+ (0.072 nm) and Al3+ (0.053 nm),
resulting in large distortions when it was incorporated into
the HTl layers. Moreover, it was suggested that a high Al
content was favorable for the formation of HTl precursors.46
In this study, the La3+ : Al3+ atomic ratios in the starting solu-
tion was increased from 0 : 1 to 1 : 1, so the Al content
decreased relatively with increasing La content. Conse-
quently, the yields of the HTl phase in the precursors
decreased and such an HTl structure almost disappeared
when x reached 1.0.
It was well known that HTlcs belonged to the hexagonal
crystal system, and they could be described by two important
values, a and c.31 As seen in Table 1, both parameters a and c
of La-containing HTl precursors increased with the increase
in La content, and the values of these parameters were in
good agreement with those reported by others.42,43 Parameter
a (a = 2d110) was a function of the average radius of the metal
cations in the layers and reflected the density of metal ions
Table 1 Textural and chemical properties as well as weight loss (occurring in the second and third steps of thermal decomposition) of the HTl
precursors
Mg2+
:
Lattice parametersb (Å)
Weight
Al3+ : La3+
atomic
ratioa
loss (%)
(second :
third step)
Pore
SBET
volume
Samples
a
c
(m2 g−1
)
(cm3 g−1
)
HT-0La
HT-2La
HT-5La
HT-8La
HT-10La
3 : 1 : 0
3.061
3.063
3.066
3.068
3.072
23.598
23.624
23.937
24.253
24.409
18.5 : —
13.9 : 7.5
12.6 : 8.1
12.3 : 9.7
11.9 : .8
21
33
41
39
36
0.12
0.16
0.24
0.21
0.19
3 : 1 : 0.2
3 : 1 : 0.5
3 : 1 : 0.8
3 : 1 : 1
a
The nominal atomic ratio in the synthesis mixture. b Determined by XRD, a = 2d110, c = 3d003
.
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Catalysis Science & Technology
Paper
in the (110) plane. The increase in parameter a might be due
to the isomorphous substitution of Mg2+ or Al3+ by La3+ in
the HTl layers. La3+ exhibited a larger ionic radius in compar-
ison with Mg2+ and Al3+; thus more La3+ insertion into
brucite sheets led to an increase in the average radius of the
cations in the layers.44 Parameter c (c = 3d003) was a measure
of the thickness of the HTl layer and the interlayer distance,
which further relied on the electrostatic force in the HTl
interlayer. Due to its larger ionic radius, La3+ exhibited a
lower charge density than Mg2+ and Al3+. Thus, the insertion
of La3+ into the HTlc layer would cause a decrease in electro-
static forces between the layer and the interlayer. In addition,
La3+ preferred to be located in the interlayer gallery of HT as
separate hydroxide and hydroxyl carbonate phases, which
resulted in the decrease of the electrostatic attraction
between the layer and interlayer anions, leading to the
increase of the layer spacing.42
The XRD patterns of the samples after calcination are
shown in Fig. 1(b). Clearly, heat treatment of the HTlcs at
500 °C destroyed the HTl structure since no characteristic
reflection peaks of HTlcs were found in the XRD patterns. All
of these calcined materials presented reflections at 43.3° and
62.8° owing to the (200) and (220) planes, respectively, of the
MgO periclase (JCPDS 45-0946) phase. Nevertheless, the
intensity of the peaks due to MgO periclase decreased and
almost vanished when the La additive amount reached x =
1.0. Diffraction peaks assigned to Al2O3 were not found in
the calcined catalysts, implying that Al2O3 might exist in an
amorphous state. As x increased to 0.5, the diffraction peaks
ascribed to monoclinic La2O2CO3 (JCPDS 48-1113) formed by
re-adsorption of gaseous CO2, coming from HTlc decomposi-
tion or air contamination during the sample manipulation,
were detected.42–44 As for HTC-8La and HTC-10La materials,
the diffraction peaks of hexagonal La2O2CO3 (JCPDS 37-804)
also appeared at 2θ of 11.0°, 22.2°, 25.2°, 27.6°, 30.5°, 33.6°,
44.4°, 47.4° and 50.3°, which indicated that a higher content
of La was favorable for La2O2CO3 crystal transformation from
monoclinic to hexagonal. In addition, the formation of
La2O2CO3 led to the shift of the peaks of the (200) plane of
MgO to lower angles, which was probably because of the fact
that some La3+ ions had been incorporated into the MgO
lattice.
Fig. 2 Thermogravimetry and differential thermogravimetry (TG-DTG)
profiles of HTl precursors.
Fig. 2 shows the weight loss and the weight loss rates of
the HTlcs during calcination, which reveals the transforma-
tion of HTl samples into the corresponding mixed oxides.
Obviously, the TG profile of the pure Mg–Al hydrotalcite (HT-
0La) precursor consisted of two major steps, while the La-
containing HTlcs showed four typical weight losses. The first
loss for all the samples was associated with the peak at 175–
218 °C with a shoulder peak at 67–85 °C. According to previ-
ous reports, it was due to the removal of physically absorbed
water, interlayer water molecules and small amounts of
weakly bound OH groups. The second weight loss at 360–410
°C is due to the elimination of both the hydroxyl groups from
the HTl network and carbonate anions from the interlayer
anion, resulting in the collapse of the layered structure.47
With regard to La-containing precursors, the third weight
loss at 420–440 °C might be ascribed to the dehydroxylation
of La–OH, resulting in the transformation of La hydroxide
and hydroxyl carbonate phases into La2O2CO3.48 Further-
more, another weight loss was found at around 710 °C in La-
containing HTlcs, which might be because of the thermal
decomposition of La2O2CO3.49 Obviously, this weight loss
became more and more intense with increasing La content.
This phenomenon was consistent with the result of XRD,
which showed that the number of La2O2CO3 peaks increased
from HTC-2La to HTC-10La samples.
The weight losses for the second and third steps are
compared in Table 1. It could be seen that with increasing
La content, the second weight loss step due to the
This journal is © The Royal Society of Chemistry 2015
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
decomposition of the HTl species decreased, while the third
weight loss, owing to La hydroxide and hydroxyl carbonate
phases, increased. This suggested that the higher the La3+
:
Al3+ atomic ratio in the mother liquor, the less HTl phases
and the more separate La hydroxide and hydroxyl carbonate
phases existed in the precursors, which was in accordance
with the XRD patterns of these samples. In addition, the TG
profiles demonstrated that the introduction of La led to a
decrease in the decomposition temperatures, especially for
the third weight loss in La-containing HTlcs (for example,
HT-2La: 440 °C; HT-5La: 435 °C; HT-8La: 429 °C; HT-10La:
420 °C). Furthermore, compared with the HT-0La sample, the
second step position in La-containing precursors also shifted
toward lower temperature with the increase in La content.
These results implied a gradual decrease in the thermal sta-
bility of the La-containing HTlcs when x increased from 0.2
to 1.0. It might be ascribed to the presence of separate lan-
thanum hydroxide and hydroxyl carbonate phases in the
interlayer space, which led to a decrease in the electrostatic
attraction between the layer and the interlayer anions that
resulted in the increase of the interlayer space and lower
thermal stability.42,49
In the present work, the surface compositions of the cal-
cined samples, as determined by XPS, are compared with the
chemical compositions of the catalysts measured by ICP in
Table 2. It was clearly observed that the surface was consider-
ably depleted of La, while the same surface was enriched with
Mg and Al for La-modified Mg–Al mixed oxides. This was in
accordance with the results in a previous report, where La :
Zn atomic ratios in the surface decreased significantly when
Zn–Al–La HTlcs were prepared.49
The textural properties of HTC-La catalysts were deter-
mined by the N2 adsorption–desorption technique. The N2
adsorption–desorption isotherms and the corresponding
pore size distribution of the calcined samples are shown in
Fig. 3. Apparently, all the samples showed typical type-IV
adsorption isotherms with clear hysteresis loops at higher
relative pressure, characteristic of a mesoporous solid, and
their pore sizes were widely distributed in the range of 5–45
nm. These hysteresis loops were narrow, indicating that the
mesoporous pores were regular. Besides, the hysteresis loops
Fig.
3
N2 adsorption–desorption isotherms (a) and pore-size
distributions (b) of HTC-La catalysts.
exhibited the features of H3 type (IUPAC classification). It is
well known that this type of hysteresis is usually related to
materials consisting of plate-shaped particles. This was con-
sistent with the traditional view of these types of mixed
oxides derived from HTlcs.50 The specific surface area and
pore volume of the calcined samples are listed in Table 2.
Table 2 Textural, chemical and basic properties of various catalysts
Amounts of basic sites (μmol g−1) and
Pore
Atomic ratio in solidsa
Mg2+ : Al3+ : La3+
Surface atomic ratiob
Mg2+ : Al3+ : La3+
contributionc
SBET
volume
Sample
(m2 g−1
)
(cm3 g−1
)
Total
W
M
S
HTC-0La
HTC-2La
HTC-5La
HTC-8La
HTC-10La
La2O2CO3
124
106
95
84
71
0.33
0.52
0.49
0.45
0.41
0.16
2.69 : 1 : —
2.72 : 1 : 0.21
2.83 : 1 : 0.54
2.85 : 1 : 0.88
2.93 : 1 : 1.12
—
2.47 : 1 : —
2.41 : 1 : 0.09
2.64 : 1 : 0.16
2.66 : 1 : 0.28
2.75 : 1 : 0.34
—
193.5
223.8
264.5
242.4
211.6
107.7
68.1 (35.2)
60.6 (27.1)
51.8 (19.6)
42.0 (17.3)
22.0 (10.4)
52.6 (48.9)
114.2 (59.0)
141.5 (63.2)
180.1 (68.1)
152.2 (62.8)
120.4 (56.9)
31.8 (29.5)
11.2 (5.8)
21.7 (9.7)
32.6 (12.3)
48.2 (19.9)
69.2 (32.7)
23.3 (21.6)
28
a
b
c
Values measured by ICP. Evaluated by XPS. The value in parentheses was the contribution of single basic sites to the number of total
basic sites.
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Catalysis Science & Technology
Paper
It could be seen that the calcined samples had a higher sur-
face area and pore volume than the corresponding HTls (see
Table 1), which could be ascribed to the fact that a large
amount of CO2 was released from HTlcs during calcina-
tion.41 This might be favorable for them to exhibit high cata-
lytic activity. Prescott et al. and Meher et al. have reported
that a high content of Al2O3 in the catalysts was favorable to
increase the specific surface area.51,52 Besides, as discovered
to the formation of large sintered particles (see the SEM
profiles).49
The morphologies of the HTl precursors are shown in
Fig. 4. All the HTl samples consisted mainly of well-
developed and plate-shaped crystals, which suggested the for-
mation of a layered structure. For the pure Mg–Al HT (HT-
0La) material, it exhibited mainly hexagonal plate-shaped
crystals, which was the same as the result in the previous lit-
erature.38 However, the surface of the plate-shaped crystals
became rough and indistinct with the increase in La additive
amount. This also implied that the introduction of lantha-
num was not favorable for the synthesis of the HTl structure,
which was in agreement with the XRD results and previous
reports.42,43,49 The particles of the samples calcined after 500
°C maintained a similar morphology to that of the precur-
sors, though some crystals were destroyed (see Fig. 5). Mean-
while, their particles seemed to be sintered and the particle
size of the calcined samples increased as the La additive
amount increased. This might be because of the formation
2−
by Cantrell et al., Al favored higher CO3 contents in the
interlayer; therefore more open porous networks were
formed after thermal treatment.53 In this work, as the La
content increased, correspondingly, the content of Al
decreased relatively in these La-containing mixed oxides. At
the same time, La might be incorporated into the pore sys-
tem of HTC-La catalysts during the preparation process. So,
their surface area and pore volume became lower as x
increased from 0 to 1.0 (see Table 2). Furthermore, the low-
est surface area of HTC-10La, which was not favorable for
the exposure of more active sites, could also be attributed
Fig. 4 SEM images of HTl precursors.
This journal is © The Royal Society of Chemistry 2015
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
Fig. 5 SEM images of mixed oxides.
the La2O2CO3 phase in the catalysts, which was proved by the
XRD patterns in Fig. 1(b). Considering that their surface area
decreased with the increase in La additive amount, it was rea-
sonable to suggest that less La2O2CO3 phase and smaller par-
ticle size favor the formation of mixed oxide catalysts with
higher specific surface area.
the TPD profiles of these samples could be split into three
Gaussian peaks, indicating that three types of basic sites with
different strengths were present on each surface. Depending
on the temperature of desorption, these desorption peaks
could be assigned to weak (CO2 desorption below 200 °C),
moderate (CO2 desorption between 200 and 400 °C) and
strong (CO2 desorption higher than 400 °C) basic sites.
According to the description of reported studies, the weak
basic sites were related to OH− groups, the moderate basic
sites were associated with metal–oxygen pairs (such as Mg–O,
Al–O, and La–O), and the strong basic sites were assigned to
low-coordination unsaturated oxygen atoms.38,41,55 And from
the CO2-TPD profiles of the HTC-La samples, both the mod-
erate and strong basic sites gradually shifted to higher tem-
perature with the rise of La content, which suggested that a
greater amount of La content resulted in the increase in mod-
erate and strong basic strength. Moreover, it should be men-
tioned that among all the samples, La2O2CO3 showed the
highest desorption temperature of the strong basic sites at
about 437 °C. Thus, based on CO2-TPD characterization, the
basic strength of these catalysts followed this order:
3.2 The surface basicity of catalysts
For solid base catalysts, two factors including basic strength
and total basicity (basic site amount per unit weight of cata-
lyst) were selected to evaluate their surface basic properties.54
Here, the surface basic properties of the catalysts were exam-
ined by CO2-TPD and the basic species could be assigned
according to the temperature at which peaks appeared.
Higher desorption temperature pointed to stronger basic
strength and higher CO2 uptake indicated higher total basic-
ity. Fig. 6 shows the CO2 desorption profiles for the prepared
La-modified Mg–Al mixed oxides. For comparison, the pro-
duced La2O2CO3 sample was also investigated. Apparently,
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Catalysis Science & Technology
Paper
generation of coordinatively unsaturated O2− ions, and the
lower stability of the layered structure was beneficial to the
formation of unsaturated O2− ions.56 According to the above
analysis based on TGA, a gradual decrease in the thermal sta-
bility of the La-containing HTlcs was found when x increased
from 0 to 1.0. Consequently, the metal–oxygen pairs in the
mixed oxides would gradually transform to unsaturated O2−
ions as the La content increased. Therefore, it seemed that
both the BET surface area and La content could affect the
number of basic sites and surface basic site distribution of
HTC-La catalysts. Considering the amounts of both moderate
and strong basic sites, their basic densities decreased in the
following sequence: HTC-5La (212.7 μmol g−1) > HTC-8La
(200.4 μmol g−1) > HTC-10La (189.6 μmol g−1) > HTC-2La
(163.2 μmol g−1) > HTC-0La (125.4 μmol g−1) > La2O2CO3
(55.1 μmol g−1). So, for HTC-La catalysts, the basicity of the
moderate and strong basic sites first increased until x = 0.5,
then decreased with further increase in La content.
3.3 Catalytic performance
3.3.1 The reaction system of MC and methanol to synthe-
size DMC. It is well known that the reaction system of MC
and methanol could be described by reactions 1–3 in
Scheme 1 no matter what catalyst is used.
At first, DMC could be synthesized via the substitution of
the amino group of MC by the methoxy group in methanol.
The product DMC, which was a favorable methylating agent
for the amino group, could further react with MC to yield
N-methyl methyl carbamate (NMMC).12 Furthermore, at a rel-
atively high reaction temperature, DMC was easily thermally
decomposed into dimethyl ether (DME) and carbon dioxide
in the presence of solid basic and acidic oxides.57,58 It should
be mentioned that DME could not be detected by GC because
it was easily volatilized when the autoclave was opened.18–24
Thus, the catalysts were evaluated on the basis of MC conver-
sion, DMC yield and selectivity, and NMMC yield in the pres-
ent case.
3.3.2 Effect of La content on the DMC synthesis. The cata-
lytic performances of various catalysts were evaluated for the
methanolysis of MC to DMC. The results are summarized in
Table 3. Notably, the reaction of MC and methanol hardly
occurred in the absence of a catalyst (entry 1). However, HTC-
0La exhibited high activity towards DMC synthesis, yielding
28.2% DMC. The catalytic performances of the La-containing
mixed oxide catalysts are presented in entries 3–6. The four
catalysts all gave higher MC conversion compared to the
HTC-0La sample, indicating that they had higher catalytic
activities. Nevertheless, the DMC yield and selectivity differed
significantly among them. When the HTC-2La catalyst was
used, a 34.7% yield of DMC was achieved, which was a bit
higher than that of HTC-0La. Under the same reaction condi-
tions, HTC-5La exhibited excellent catalytic performance with
a high DMC yield of 54.3% and a good DMC selectivity of
80.9%. As the La additive amount further increased, both the
MC conversion and the DMC selectivity decreased
Fig. 6 CO2-TPD of HTC-La catalysts and La2O2CO3 sample.
La2O2CO3 > HTC-10La > HTC-8La > HTC-5La > HTC-2La >
HTC-0La.
On the other hand, the basicity of all the samples is listed
in Table 2. One can see that the La2O2CO3 catalyst showed
the least total basicity, which might be due to its low specific
surface area resulting in the decrease of its exposed basic
sites. Moreover, the contribution of the weak basic sites to
the total basic sites was about 50%. For the La-containing
mixed oxides, this proportion decreased with increasing La
additive amount, and their moderate and strong basic peaks
were more intense (72.9–89.6% of the total number of basic
sites). Therefore, the presence of La not only improved the
basic strength, but also promoted the amounts of moderate
and strong basic sites of HTC-La catalysts. Notably, although
the number of strong basic sites of HTC-La increased as the
La content increased, the amount of moderate basic sites
increased first and reached the maximum when x was 0.5.
Then, it gradually decreased with further increase in La con-
tent. These results revealed that apart from the BET surface
area, there was some other cause that was responsible for the
change in the number of moderate and strong basic sites. In
theory, introduction of La could increase the amount of
metal–oxygen pairs (La–O) on the catalyst surface, and led to
the rise in the number of moderate basic sites for metal
oxides. Nevertheless, as reported by Wu et al., the metal–oxy-
gen pairs could partially break up, which led to the
This journal is © The Royal Society of Chemistry 2015
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
Scheme 1 The reaction system for the reaction of MC and methanol.
Table 3 Catalytic performance of various catalysts in the present worka
Entry
Catalyst
MC conversion (%)
DMC selectivity (%)
DMC yield (%)
NMMC yield (%)
1
2
3
4
5
6
7
—
4.1
32.3
41.2
67.1
59.5
52.4
21.9
63.4
87.3
84.2
80.9
72.6
64.5
51.6
2.6
28.2
34.7
54.3
43.2
33.8
11.3
0
HTC-0La
HTC-2La
HTC-5La
HTC-8La
HTC-10La
La2O2CO3
1.2
2.0
4.1
5.7
8.9
0.3
a
Reaction conditions: reaction temperature, 200 °C; reaction time, 6 h; catalyst amount, 0.5 g; MC, 0.05 mol; methanol, 1.0 mol. The product
selectivity and yields were based on MC.
remarkably. For instance, HTC-8La and HTC-10La catalysts
gave 43.2% and 33.8% DMC yields, respectively. At the same
time, their selectivity decreased to 72.6% and 64.5%, respec-
tively. Besides, a further investigation into the performance
of the prepared La2O2CO3 was also conducted. Obviously, it
showed inferior activity compared with the aforementioned
mixed oxide catalysts (entry 7). More important, it is notewor-
thy that for these catalysts, their catalytic activities decreased
in the sequence of HTC-5La > HTC-8La > HTC-10La > HTC-
2La > HTC-0La > La2O2CO3. Meanwhile, the DMC selectivity
achieved over them followed this order: HTC-0La > HTC-2La
> HTC-5La > HTC-8La > HTC-10La > La2O2CO3. So, the cur-
rent catalytic performance tests suggested that the insertion
of La into Mg–Al mixed oxides could remarkably improve the
DMC synthesis from MC and methanol.
3.3.3 Effect of basic properties on the DMC synthesis. For
catalysts, an increase in their BET surface areas was favorable
for achieving higher catalytic activity. The surface areas of
the catalysts are listed in Table 2. It could be seen that their
BET surface areas decreased in the order of HTC-0La > HTC-
2La > HTC-5La > HTC-8La > HTC-10La > La2O2CO3. How-
ever, their catalytic activity followed this sequence: HTC-5La
> HTC-8La > HTC-10La > HTC-2La > HTC-0La > La2O2CO3.
Thus, the BET surface area of the catalysts was not the sole
factor for their catalytic activity in the present work. As men-
tioned in the discussion above, the moderate and strong
basic sites of the HTC-La samples were enhanced remarkably;
therefore, their catalytic activities might be closely related to
the amounts of moderate and strong basic sites. In addition,
it should be noted that the DMC selectivity decreased as the
Fig. 7 The correlation between MC conversion and the amounts of both moderate and strong basic sites (a); the relationship between DMC
selectivity and desorption temperature of strong basic sites (basic strength) (b).
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Catalysis Science & Technology
Paper
basic strength of strong basic sites increased. In order to clar-
ify well the correlation between the MC conversion and the
amounts of both moderate and strong basic sites, as well as
the relationship between their DMC selectivity and basic
strength, we illustrated the catalytic performance at 2 h when
the reaction was at the beginning in Fig. 7. It could be seen
that the sequence of MC conversion was in good agreement
with the trend of basicity of both moderate and strong basic
sites (Fig. 7(a)), whereas the trend for DMC selectivity was
observed to be in contrast with the order of their basic
strength (Fig. 7(b)). These phenomena further suggested that
the amounts of moderate and strong basic sites played a cru-
cial role in MC conversion. However, a too strong basic
strength was not beneficial to obtaining a high DMC selectiv-
ity, which might be because of the fact that side reactions
such as thermal decomposition of DMC and NMMC synthe-
sis would aggravate over strong basic catalysts.57,58 This was
consistent with the results in previous work, in which the
DMC selectivity also decreased with rise of basic strength
when basic oxides such as CaO, La2O3 and MgO were used as
catalysts.12 Therefore, it was reasonable to draw the conclu-
sion that the enhancement of the catalytic performance of
La-modified Mg–Al mixed oxide catalysts towards DMC syn-
thesis from MC and methanol was due to the adjustment of
the basic properties by incorporation of a suitable amount of
La additive.
3.3.4 Effect of reaction conditions over HTC-5La catalyst.
The reaction conditions were further optimized over the
HTC-5La catalyst. The effect of reaction temperature on the
DMC synthesis is illustrated in Fig. 8(a). Obviously, the reac-
tion temperature exhibited a significant impact on the reac-
tion. The conversion of MC consistently increased with the
increase in temperature. However, the DMC yield and selec-
tivity increased sharply in the temperature range of 180–200
°C, and then decreased when the reaction temperature
exceeded 200 °C. It could be seen that the maximum DMC
yield of 54.3% with the highest DMC selectivity of 80.9% was
obtained when the reaction temperature was 200 °C. Theoret-
ically, the DMC synthesis was an endothermic reaction and
the temperature had a positive effect on DMC formation.19
However, higher temperature also shortened the time needed
to reach the maximum DMC concentration and accelerated
the rate of side reactions including N-methylation of MC. As a
result, the yield of the by-product NMMC increased signifi-
cantly to 6.4% at the cost of DMC consumption when the
reaction temperature was 210 °C. Therefore, the suitable reac-
tion temperature should be 200 °C.
The effect of reaction time on the DMC synthesis is dem-
onstrated in Fig. 8(b). Clearly, as the reaction proceeded, the
conversion of MC increased. The DMC yield and selectivity
reached their maximal values at 6 h, and then, the consump-
tion of DMC in the further reaction surpassed gradually the
Fig. 8 Effect of reaction conditions on the DMC synthesis over the HTC-5La catalyst. Effect of reaction temperature (a). Reaction conditions:
reaction time, 6 h; catalyst amount, 0.5 g; molar ratio of methanol/MC, 20 : 1. Effect of reaction time (b). Reaction conditions: reaction tempera-
ture, 200 °C; catalyst amount, 0.5 g; molar ratio of methanol/MC, 20 : 1. Effect of catalyst amount (c). Reaction conditions: reaction temperature,
200 °C; reaction time, 6 h; molar ratio of methanol/MC, 20 : 1. Effect of the molar ratio of methanol/MC (d). Reaction conditions: reaction temper-
ature, 200 °C; reaction time, 6 h; catalyst amount, 0.5 g.
This journal is © The Royal Society of Chemistry 2015
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
formation of DMC from MC. Thus, the suitable reaction time
was chosen as 6 h.
stable solid catalyst, the reusability of HTC-5La was also
investigated. After the first run, the used HTC-5La sample
was separated by filtration, washed with methanol several
times, dried in an oven at 80 °C and reused without any pre-
treatment. Its catalytic performance was evaluated and the
results are shown in Fig. 9. To our delight, HTC-5La could be
recycled and reused at least four times without obvious loss
of activity. At the same time, the recovered HTC-5La catalyst
after reused four times (HTC-5La-fourth) was analyzed using
SEM and XRD techniques. SEM presented that the plate-like
morphology was unchanged after the catalyst was used four
times (see Fig. 5). Additionally, the XRD pattern of HTC-5La-
fourth is shown in Fig. 1(b). Obviously, its diffraction peak
number was the same as the fresh one, and the peak inten-
sity didn't show any significant change. Therefore, the HTC-
5La catalyst exhibited higher stability than previously
reported zinc-based catalysts for this reaction, and a hetero-
geneous solid catalyst with excellent catalytic performance
and stability was successfully developed.
The effect of the catalyst amount was also studied and the
results are shown in Fig. 8(c). With the increase of the cata-
lyst amount, the conversion of MC always monotonically
increased, but the DMC yield and selectivity reached the max-
imum values of 54.3% and 80.9% at 0.5 g and then
decreased. It was ascribed to the fact that the yield of further
reaction product NMMC increased gradually when the cata-
lyst amount was more than 0.5 g. Therefore, the suitable cata-
lyst amount was chosen as 0.5 g in the present case.
The effect of the methanol/MC molar ratio on DMC syn-
thesis is presented in Fig. 8(d). The results demonstrate
clearly that the DMC yield and selectivity increased first in
the range of 10–20 and then decreased in the range of 20–25.
When the methanol/MC molar ratio was 20, the DMC yield
and selectivity reached the maximum 54.3% and 80.9%,
respectively. When the methanol/MC molar ratio was lower
than 20, the MC concentration was higher and the side reac-
tion of N-methylation of MC with DMC would occur easily,
resulting in a lower DMC yield and selectivity. In theory, a
higher methanol/MC molar ratio could promote the reaction
equilibrium towards DMC synthesis. Nevertheless, a too high
methanol/MC molar ratio would lead to the reduction of MC
concentration. Besides, the following in situ FTIR experi-
ments revealed that MC and methanol molecules were co-
absorbed on the surface basic sites of the HTC-5La catalyst.
Consequently, some active sites of the catalyst for activating
MC would be occupied by a larger amount of methanol.
Thus, the reaction rate might decrease, resulting in a lower
DMC yield when the methanol/MC molar ratio exceeded 20.
In the present case, from a practical view, the suitable metha-
nol/MC molar ratio was selected as 20.
3.4 FTIR spectra between reactants and HTC-5La catalyst
The adsorption behaviours on the solid catalyst surface play
an important role in the reaction pathways. In order to
3.3.5 Reusability and stability of HTC-5La catalyst. The
reusability of catalysts played a crucial role in the DMC syn-
thesis via MC methanolysis in practical industrial produc-
tion. Therefore, in order to explore a highly effective and
Fig. 10 FTIR spectra evolution of MC (a); MC absorbed on HTC-5La
(b); methanol absorbed on HTC-5La (c) at different temperatures; and
MC with various amounts of methanol co-adsorbed on HTC-5La at
180 °C (d).
Fig. 9 Reusability of the HTC-5La catalyst for the DMC synthesis from
MC and methanol. Reaction conditions: 200 °C, 6 h, 0.5 g of catalyst,
0.05 mol of MC, 1.0 mol of methanol.
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Catalysis Science & Technology
Paper
investigate the possible mechanism of the reaction between
MC and methanol, in situ FTIR experiments under different
conditions were conducted.
5La catalyst, resulting in the formation of methoxide species
coordinated to metal atoms. Moreover, the two bands
became more and more obvious at higher temperature,
implying that methanol was prone to deprotonation, forming
methoxy groups on the catalyst surface at higher
temperature.
3.4.1 MC decomposition. Fig. 10(a) demonstrates the FTIR
spectra evolution of MC with increasing temperature. At
room temperature, the characteristic bands of MC ascribed
to the CO stretching vibration (amide I band) and NH2
bending vibration (amide II band) were observed at 1700 and
1610 cm−1, respectively.59 It is worth noting that a new band
appeared at 2198 cm−1, which was assigned to the NCO
asymmetric stretching vibration of isocyanic acid when the
temperature surpassed 140 °C.17,60 Apparently, the intensity
of the bands attributed to MC decreased, accompanying that
of the band at 2198 cm−1 that increased rapidly with further
increase in temperature. This implied that the formation of
isocyanic acid was due to the decomposition of MC, which
was in accordance with the results discovered by Li et al.60
3.4.2 MC adsorption on HTC-5La catalyst. Then, the
adsorption of MC over the HTC-5La catalyst was studied at
given temperature regions, and the FTIR spectra are shown
in Fig. 10(b). Clearly, the intensity of both amide I and amide
II bands of MC decreased rapidly as the temperature
increased. At the same time, the amide I band showed a blue
shift gradually from 1700 to 1728 cm−1, accompanying the
red shift of the amide II band from 1610 to 1598 cm−1 before
it disappeared, which did not occur when pure MC
decomposed. These alternations pointed out the interaction
of nitrogen in MC with unsaturated metal cation sites on the
HTC-5La catalyst, resulting in the amino group of MC bound
to the metal oxide surface (M–NH2).61 More important, a new
band assigned to the stretching vibration of NCO bound
to the metal oxide surface (M–NCO, MMg, Al or La)
was found at around 2190 cm−1. It could be seen that the
intensity of NCO increased first before the temperature
increased to 160 °C, implying that MC was swiftly activated
and decomposed into isocyanic acid over the catalyst. How-
ever, its intensity decreased with further increase in tempera-
ture, suggesting the enhanced desorption of isocyanate-
containing species. In addition, compared to the blank test
without the catalyst, the characteristic band of NCO was
clearly observed at a lower temperature of 80 °C on the HTC-
5La catalyst surface. Thus, all these results proved that MC
could be activated over the HTC-5La catalyst and the charac-
teristic band of NCO still existed when the temperature
increased to 200 °C.
3.4.4 Co-adsorption of MC and methanol on HTC-5La cat-
alyst. In order to study the reaction procedure on the HTC-
5La catalyst more clearly, we carried out MC and methanol
co-adsorption experiments (see Fig. 10(d)). In the present
case, we followed the change of the band at 2190 cm−1 corre-
sponding to the stretching vibration of NCO bound to
the metal oxide surface (M–NCO), the band at 1750 cm−1
assigned to the CO stretching vibration of DMC,63 as well
as the amide I and II bands of MC. Without methanol, MC
was activated and decomposed into isocyanic acid species by
the basic sites of the catalyst. As the amount of methanol
increased, the intensity of M–NCO decreased quickly and
disappeared when the amount of methanol increased to 40
μL. During this process, the characteristic bands of MC were
found at about 1700 and 1610 cm−1. This revealed that the
decomposition of MC might be a reversible reaction, and MC
would be re-generated via the reaction between methanol
and isocyanic acid species when excessive methanol existed
in the reaction system. At the same time, it should be noted
that for the MC molecule, its amide I band had a blue shift,
while the amide II band showed a red shift before they
disappeared. This also implied that MC was adsorbed on the
catalyst surface and its amino group was coordinated with
the cationic metal surface atom (M–NH2).61 Besides, a new
band at 1750 cm−1 ascribed to DMC was detected when the
methanol amount was 60 μL. Apparently, the amide I and II
bands of MC decreased rapidly with the further increase in
methanol amount, and they vanished as the amount of meth-
anol reached 100 μL. Meanwhile, the intensity of the charac-
teristic band of DMC consistently increased during this pro-
cess, revealing the production of DMC from MC and
methanol over the HTC-5La catalyst.
3.4.5 Plausible reaction mechanism for DMC synthesis
from MC and methanol. According to the FTIR experimental
results and previous work,60 a plausible basic catalytic mech-
anism wherein MC and methanol were activated simulta-
neously on the surface basic sites of the HTC-5La catalyst
was proposed. As shown in Scheme 2, MC was adsorbed on
the catalyst surface through its amino group bound to the
metal oxide surface. Consequently, MC was easily
decomposed into isocyanic acid. As mentioned in the discus-
sion above, the decomposition of MC into isocyanic acid and
methanol might be a reversible reaction. In this work,
because the molar ratio of methanol/MC was selected as 20,
the reaction equilibrium would shift to MC formation in the
presence of excessive methanol in a short time. Then, the re-
synthesized MC was coordinated with the cationic surface
metal (Mg, Al or La) atom of the mixed oxide through the N
atom of the amino group. As a result, the electrons of the
C–N group of MC would be redistributed, and then, the
3.4.3 Methanol adsorption on HTC-5La catalyst. Fig. 10(c)
illustrates the FTIR spectra evolution of methanol adsorbed
on the HTC-5La catalyst at different temperatures. Two new
bands at 2395 and 2823 cm−1 corresponding to C–H asym-
metric and symmetrical stretching vibrations of deprotonated
methanol or methoxide species were clearly visible, which
indicated the adsorption of the methoxy group on the surface
of the HTC-5La catalyst. This was in accordance with previ-
ously reported studies of adsorbed methanol on metal oxide
catalysts.22,62 Thus, these experimental results suggested that
methanol could be activated and deprotonated over the HTC-
This journal is © The Royal Society of Chemistry 2015
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
Scheme 2 Possible reaction mechanism for the DMC synthesis from MC and methanol over HTC-5La catalyst.
proton moved to the carbon atom with the appearance of a
carbocation and a nitrogen anion. At the same time, the
HTC-5La catalyst could also activate methanol through the
abstraction of Hδ+ by the basic sites, giving a strong nucleo-
philic methoxy group, which formed a complex with the
unsaturated metal cations of the catalyst. Subsequently, a
lone pair of electrons of the oxygen atom in the methoxy
group formed a bond with the electrophilic carbonyl carbon
of the activated MC to generate DMC, while the hydrogen
proton breaking away from the methanol molecule combined
with the amino group leaving from the activated MC mole-
cule to form NH3.
Al,22 Zn/Fe,30 and ZnO–CeO2–La2O3 mixed oxides23 were often
used to catalyze the DMC synthesis. Even though these cata-
lysts were highly active, their reusability might be limited.
The present HTC-5La catalyst showed higher activity com-
pared to the other reported solid catalysts. In addition, the
present catalyst was reusable, and its preparation method
was simple and repeatable. Based on the above results, we
can conclude that the HTC-5La sample is an active, stable
and practical catalyst for the DMC synthesis from MC and
methanol.
4. Conclusions
La-modified mesoporous Mg–Al mixed oxides derived from
hydrotalcite-like compounds containing La (Mg2+ : Al3+ = 3)
with La3+ : Al3+ varying from 0 to 1.0 were synthesized via a
simple co-precipitation method followed by hydrothermal
treatment. The La content had a significant influence on the
physicochemical and catalytic properties of the HTC-La cata-
lysts for the synthesis of DMC from MC and methanol. The
following conclusions might be drawn:
(1) The pure and crystalline Mg–Al–La hydrotalcite-like
compounds could be obtained by adding suitable amounts of
La into pure Mg–Al hydrotalcite, and the yields of
hydrotalcite-like compounds decreased with increasing La
content.
3.5 Activity comparison of HTC-5La catalyst with other refer-
ence catalysts
In this work, the catalytic activity of the HTC-5La catalyst is
compared with several representative catalysts, and the com-
parative results are listed in Table S1 in the ESI.† Homoge-
neous catalysts such as ZnCl2 and LaCl3 exhibited good cata-
lytic performance,28,29 but they were inconvenient for catalyst
recovery in practical production. Although ZnO exhibited
high catalytic activity towards the reaction between urea and
methanol,19 it was almost inactive for this reaction.24 In fact,
it was a precursor of the homogeneous catalyst for the reac-
tion of urea and methanol.24 Compared to pure oxide CaO,12
Fe2O3 supported on MCM-49 (Fe2O3/HMCM-49) showed
higher catalytic activity with 97.3% DMC selectivity.18 Unfor-
tunately, the DMC yield was lower than 35%. Besides, Zn/
(2) La could be used to fine-tune the structural, textural
and chemical properties of the pure Mg–Al mixed oxides.
With the increased La3+ : Al3+ atomic ratio, the basic strength
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Catalysis Science & Technology
Paper
of HTC-La monotonically increased. However, the basicity of
12 M. H. Wang, H. Wang, N. Zhao, W. Wei and Y. H. Sun,
Catal. Commun., 2006, 7, 6–10.
moderate and strong basic sites first increased until La3+
:
Al3+ = 0.5, then decreased with further increase in La content.
(3) Both the basic strength and basicity of the HTC-La cat-
alysts were demonstrated to have a significant impact on the
DMC synthesis. It was found that the catalysts that possessed
more moderate and strong basic sites could give high MC
conversion. In contrast, the DMC selectivity decreased as the
basic strength increased. A suitable amount of La was favor-
able for the DMC synthesis, and an excellent DMC yield of
54.3% with a high DMC selectivity of 80.9% could be
achieved over the HTC-5La catalyst. Simultaneously, this cata-
lyst exhibited high stability on the basis of the reusability test
and SEM, XRD and N2 adsorption–desorption characterization.
(4) The in situ FTIR experiments revealed that MC and
methanol were co-adsorbed on the surface of the HTC-5La
sample. Further study indicated that MC was decomposed
into a metal isocyanato group, and methanol was activated to
produce a strong nucleophilic methoxy group by the basic
sites. Finally, a plausible basic catalytic mechanism was pro-
posed for this catalytic reaction.
13 J. J. Sun, B. L. Yang and H. Y. Lin, Chem. Eng. Technol.,
2004, 27, 435–439.
14 T. Cho, T. Tamura, T. Cho and K. Suzuki, US Pat., 5534649,
1996.
15 B. L. Yang, D. P. Wang, H. Y. Lin, J. J. Sun and X. P. Wang,
Catal. Commun., 2006, 7, 472–477.
16 H. Wang, B. Lu, X. G. Wang, J. W. Zhang and Q. H. Cai, Fuel
Process. Technol., 2009, 90, 1198–1201.
17 Q. B. Li, N. Zhao, W. Wei and Y. H. Sun, J. Mol. Catal. A:
Chem., 2007, 270, 44–49.
18 Ch. Zhang, B. Lu, X. G. Wang, J. X. Zhao and Q. H. Cai,
Catal. Sci. Technol., 2012, 2, 305–309.
19 M. H. Wang, N. Zhao, W. Wei and Y. H. Sun, Ind. Eng.
Chem. Res., 2005, 44, 7596–7599.
20 W. Joe, H. J. Lee, U. G. Hong, Y. S. Ahn, C. J. Song, R. J.
Lwon and I. K. Song, J. Ind. Eng. Chem., 2012, 18, 1018–1022.
21 X. M. Wu, M. Kang, N. Zhao, W. Wei and Y. H. Sun, Catal.
Commun., 2014, 46, 46–50.
22 X. M. Wu, M. Kang, Y. L. Yin, F. Wang, N. Zhao, F. K. Xiao,
W. Wei and Y. H. Sun, Appl. Catal., A, 2014, 473, 13–20.
23 W. Joe, H. J. Lee, U. G. Hong, Y. S. Ahn, C. J. Song, R. J.
Lwon and I. K. Song, J. Ind. Eng. Chem., 2012, 18,
1730–1735.
24 W. B. Zhao, W. C. Peng, D. F. Wang, N. Zhao, J. P. Li, F. K.
Xiao, W. Wei and Y. H. Sun, Catal. Commun., 2009, 10,
655–658.
25 H. An, X. Zhao, L. Guo, C. Jia, B. Yuan and Y. Wang, Appl.
Catal., A, 2012, 433–434, 229–235.
26 S. Fujita, Y. Yamanishi and M. Arai, J. Catal., 2013, 297,
137–141.
Acknowledgements
This work is financially supported by the “Strategic Priority
Research Program—Climate Change: Carbon Budget and
Related Issues” of the Chinese Academy of Sciences
(XDA05010108), the National Natural Science Foundation of
China (21401164), the Natural Science Foundation of Shan-
dong Province, China (ZR2012BL07, ZR2013BL019), and the
State Key Laboratory of Coal Conversion Open Foundation
(J12-13-609).
27 W. B. Zhao, X. Y. Qin, Y. H. Li, Z. S. Zhang and W. Wei, Ind.
Eng. Chem. Res., 2012, 51, 16580–16589.
28 W. B. Zhao, F. Wang, W. C. Peng, N. Zhao, J. P. Li, F. K.
Xiao, W. Wei and Y. H. Sun, Ind. Eng. Chem. Res., 2008, 47,
5913–5917.
29 D. F. Wang, X. L. Zhang, Y. Y. Gao, F. K. Xiao, W. Wei and
Y. H. Sun, Fuel Process. Technol., 2010, 91, 1081–1086.
30 D. F. Wang, X. L. Zhang, Y. Y. Gao, F. K. Xiao, W. Wei and
Y. H. Sun, Catal. Commun., 2010, 11, 430–433.
31 F. Cavani, F. Trifirò and A. Vaccari, Catal. Today, 1991, 11,
173–301.
Notes and references
1 Y. Ono, Appl. Catal., A, 1997, 155, 133–166.
2 A. Dibenedetto and A. Angelini, Adv. Inorg. Chem., 2014, 66,
25–81.
3 P. Rounce, A. Tsolakis, P. Leung and A. York, Energy Fuels,
2010, 24, 4812–4819.
4 S. Megahed and W. Ebner, J. Power Sources, 1995, 54,
155–162.
5 H. Hood and H. R. Mordock, J. Phys. Chem., 1919, 23,
498–512.
6 Sh. Wang, L. Wei, Y. Dong, Y. Zhao and X. Ma, Catal.
Commun., 2015, 72, 43–48.
7 P. Kumar, V. C. Srivastava and I. M. Mishra, Catal. Commun.,
2015, 60, 27–31.
8 B. A. Santos, V. M. Silva, J. M. Loureiro and A. E. Rodrigues,
ChemBioEng Rev., 2014, 1, 214–229.
32 L. H. Zhang, C. Zheng, F. Li, D. G. Evans and X. Duan,
J. Mater. Sci., 2008, 43, 237–243.
33 P. Gao, F. Li, H. J. Zhan, N. Zhao, F. K. Xiao, W. Wei, L. S.
Zhong, H. Wang and Y. H. Sun, J. Catal., 2013, 298, 51–60.
34 G. Busca, U. Costantino, F. Marmottini, T. Montanari, P.
Patrono, F. Pinzari and G. Ramis, Appl. Catal., A, 2006, 310,
70–78.
9 M. Zhang, M. Xiao, Sh. Wang, D. Han, Y. Lu and Y. Meng,
J. Cleaner Prod., 2015, 103, 847–853.
35 L. Faba, E. Díaz and S. Ordóñez, Appl. Catal., B, 2012, 113–
114, 201–211.
10 D. C. Stoian, E. Taboada, J. Llorca, E. Molins, F. Medina and
A. M. Segarra, Chem. Commun., 2013, 49, 5489–5491.
11 P. Ball, H. Fuellmann and W. Heitz, Angew. Chem., Int. Ed.
Engl., 1980, 19, 718–720.
36 D. Tichit, D. Lutic, B. Coq, R. Durand and R. Teissier,
J. Catal., 2003, 219, 167–175.
37 M. Mokhtar, T. S. Saleh and S. N. Basahel, J. Mol. Catal. A:
Chem., 2012, 353–354, 122–131.
This journal is © The Royal Society of Chemistry 2015
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
38 P. Liu, M. Derchi and E. J. M. Hensen, Appl. Catal., B,
2014, 144, 135–143.
39 D. F. Wang, X. L. Zhang, W. Wei and Y. H. Sun, Catal.
Commun., 2012, 28, 159–162.
40 D. F. Wang, X. L. Zhang, Ch. L. Liu, T. T. Cheng, W. Wei
and Y. H. Sun, Appl. Catal., A, 2015, 505, 478–486.
41 G. D. Wu, X. Wang, B. Chen, J. P. Li, N. Zhao, W. Wei and
Y. H. Sun, Appl. Catal., A, 2007, 329, 106–111.
50 Q. X. Ma, T. Sh. Zhao, D. Wang, W. Q. Niu, P. Lv and N.
Tsubaki, Appl. Catal., A, 2013, 464–465, 142–148.
51 H. A. Prescott, Z. Li, E. Kemnitz, A. Trunschke, J. Deutsch,
H. Lieske and A. Auroux, J. Catal., 2005, 234, 119–130.
52 L. C. Meher, R. Gopinath, S. N. Naik and A. K. Dalai, Ind.
Eng. Chem. Res., 2009, 48, 1840–1846.
53 D. G. Cantrell, L. J. Gillie, A. F. Lee and K. Wilson, Appl.
Catal., A, 2005, 287, 183–190.
42 P. Unnikrishnan and D. Srinivas, Ind. Eng. Chem. Res.,
2012, 51, 6356–6363.
54 K. Tanabe, M. Misono, Y. Ono and H. Hattori, New solid
acids and bases studies in the surface science and catalysis,
Elsevier, Amsterdam, 1989, pp. 14–16.
55 Z. Liu, J. A. Cortés-Concepción, M. Mustian and M. D.
Amiridis, Appl. Catal., A, 2006, 302, 232–236.
56 G. D. Wu, X. L. Wang, W. Wei and Y. H. Sun, Appl. Catal., A,
2010, 377, 107–113.
57 Y. Ch. Fu, H. Y. Zhu and J. Y. Shen, Thermochim. Acta,
2005, 434, 88–92.
43 R. Bîrjega, O. D. Pavel, G. Costentin, M. Che and E.
Angelescu, Appl. Catal., A, 2005, 288, 185–193.
44 E. Rodrigues, P. Pereira, T. Martins, F. Vargas, T. Scheller, J.
Correa, J. Del Nero, S. G. C. Moreira, W. Ertel-Ingrisch,
C. P. D. Campos and A. Gigler, Mater. Lett., 2012, 78,
195–198.
45 R. D. Shannon, Acta Crystallogr., Sect. A: Cryst. Phys., Diffr.,
Theor. Gen. Crystallogr., 1976, 32, 751–767.
46 M. Behrens, I. Kasatkin, S. Kuhl and G. Weinberg, Chem.
Mater., 2010, 22, 386–397.
58 M. Selva, M. Fabris and A. Perosa, Green Chem., 2011, 13, 863–872.
59 J. C. Carter and J. E. Devia, Spectrochim. Acta, Part A,
1973, 29, 623–632.
47 W. Yang, Y. Kim, P. K. T. Liu, M. Sahimi and T. T. Tsotsis,
Chem. Eng. Sci., 2002, 57, 2945–2953.
60 F. J. Li, H. Q. Li, L. G. Wang, P. He and Y. Cao, Catal. Sci.
Technol., 2015, 5, 1021–1034.
48 J. P. Ramirez, J. Overeijnder, F. Kapteijin and J. A. Moulijn,
Appl. Catal., B, 1999, 23, 59–72.
49 H. G. Li, X. Jiao, L. Li, N. Zhao, F. K. Xiao, W. Wei, Y. H. Sun
and B. Sh. Zhang, Catal. Sci. Technol., 2015, 5, 989–1005.
61 S. K. Sahni, Transition Met. Chem., 1979, 4, 73–76.
62 W. Zhang, H. Wang, W. Wei and Y. H. Sun, J. Mol. Catal. A:
Chem., 2005, 231, 83–88.
63 T. Beutel, J. Chem. Soc., Faraday Trans., 1998, 94, 985–993.
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2015