Influence of Microwave Irradiation on the Structural Properties of Carbon-Supported Hollow Copper Nanoparticles and Their Effect on the Synthesis of Dimethyl Carbonate

Article abstract of DOI:10.1002/cctc.201501182

Novel activated carbon (AC)-supported highly dispersed hollow Cu nanoparticles (NPs) (Cu/AC) with exposed {1 1 1} facets have been prepared by microwave irradiation for the synthesis of dimethyl carbonate (DMC). In particular, Cu NPs with a large cavity diameter of 35 nm are formed after irradiation from room temperature to 360 C within a mere 8 min without additional irradiation, thus benefiting from the rapid heating of the microwave procedure. In this study, an Ostwald ripening mechanism is proposed. DFT calculations are consistent with the analysis of CO temperature-programmed desorption, which found that Cu(1 1 1) facets are more favorable for the weak adsorption of CO, which supports the formation of DMC. The as-prepared catalysts exhibit the highest DMC formation rate in terms of turnover frequency and 100 % selectivity for DMC can be achieved. The large surface area of the hollow Cu NPs and the exposed {1 1 1} crystal planes are highlighted as being responsible for the excellent catalytic rate and superior selectivity, respectively.

Full text of DOI:10.1002/cctc.201501182

DOI: 10.1002/cctc.201501182
Full Papers
Influence of Microwave Irradiation on the Structural
Properties of Carbon-Supported Hollow Copper
Nanoparticles and Their Effect on the Synthesis of
Dimethyl Carbonate
Meijiao Ren, Jun Ren,* Panpan Hao, Jinzhou Yang, Donglei Wang, Yongli Pei, Jian-Ying Lin,
[a]
and Zhong Li
Novel activated carbon (AC)-supported highly dispersed
hollow Cu nanoparticles (NPs) (Cu/AC) with exposed {111}
facets have been prepared by microwave irradiation for the
synthesis of dimethyl carbonate (DMC). In particular, Cu NPs
with a large cavity diameter of 35 nm are formed after irradia-
tion from room temperature to 3608C within a mere 8 min
without additional irradiation, thus benefiting from the rapid
heating of the microwave procedure. In this study, an Ostwald
ripening mechanism is proposed. DFT calculations are consis-
tent with the analysis of CO temperature-programmed desorp-
tion, which found that Cu(111) facets are more favorable for
the weak adsorption of CO, which supports the formation of
DMC. The as-prepared catalysts exhibit the highest DMC for-
mation rate in terms of turnover frequency and 100% selectivi-
ty for DMC can be achieved. The large surface area of the
hollow Cu NPs and the exposed {111} crystal planes are high-
lighted as being responsible for the excellent catalytic rate and
superior selectivity, respectively.
Introduction
A revolution in the synthesis of nanomaterials has provided
the opportunity to produce catalysts with controlled size and
shape; this has led to the discovery that the catalytic selectivity
and turnover frequency correlate with the size and shape of
cause the process generally involves several complicated
[
3–5]
steps
and partial collapse of the hollow structures, which
are the building blocks for the construction of microsized
hollow structures on removing the templates. In contrast,
others depend on the use of facile template-free methods, in-
cluding mutual diffusion processes (known as the Kirkendall
[
1]
nanoparticles (NPs). This is referred to as “nanocatalysis”. In
the last decade, research on hollow metal nanostructures has
become prevalent because of their large surface areas, low
densities, and high catalytic rates, which are superior to their
[
9]
[10]
effect),
galvanic replacement,
the intermediate-crystal
[
11]
[12]
method, hydrolysis and condensation, and a reduction re-
[
2]
[13]
solid counterparts.
action between the core materials and reactants.
For in-
Despite intense efforts, it remains a significant challenge to
develop facile and reliable methods for the controlled synthe-
sis of hollow structures in terms of their size, morphology, and
components. General approaches for the preparation of
hollow structures have involved the utilization of various re-
movable or sacrificial templates, including hard templates
stance, hollow ceria nanospheres synthesized by template-free
and microwave-assisted aqueous hydrothermal methods
showed an excellent adsorption capacity for heavy metal
[
14]
ions. In CO oxidation, a 100% CO conversion is achieved at
[
15]
room temperature. Wang et al. reported a facile and cost-ef-
fective method without using any external templates to fabri-
cate porous hollow microstructures by a solid state chemistry
based controlled decomposition–dissolution of single metal
salt sources, which is particularly useful for practical industrial
applications.
[
3]
[4]
[5]
(
(
polymers, silica, and metallic cores ) and soft templates
[
6]
[7]
[8]
vesicles, micelles, and organic surfactants ). New function-
alized hollow CuO nanospheres on charcoal (CuO/C) employ-
ing poly(N-vinyl-2-pyrrolidone) as a sacrificial template have
been found to be effective catalysts under microwave irradia-
Dimethyl carbonate (DMC) has been utilized as an environ-
mentally green chemical, a potential alternative to certain toxic
and corrosive reagents including phosgene and dimethyl sul-
[8g]
tion because they have exhibited high reaction rates. Never-
theless, templating methods are quite tedious and costly be-
[
16]
fide, and an oxygen-containing additive for gasoline. Among
the synthesis methods of DMC, there has been great interest
in the vapor-phase oxidative carbonylation of methanol over
[
a] M. Ren, Prof. J. Ren, P. Hao, J. Yang, D. Wang, Y. Pei, J.-Y. Lin, Z. Li
Taiyuan University of Technology
[
16,17]
Key Laboratory of Coal Science and Technology
No. 79 Yingze West Street, Taiyuan 030024 (P.R. China)
E-mail: renjun@tyut.edu.cn
supported copper-based catalysts.
Among the previously
developed catalyst systems, chloride-containing Cu-based cata-
lysts often suffer from severe deactivation because of the loss
Supporting Information and ORCID(s) from the author(s) for this article
are available on the WWW under http://dx.doi.org/10.1002/
cctc.201501182.
[
18]
of chloride during the reaction. Cu-exchanged zeolites dis-
play good efficiency for DMC synthesis with strongly enhanced
ChemCatChem 2016, 8, 861 – 871
861
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
[
19]
stability. However, the preparation procedure is tedious be-
cause it involves high temperatures and is time consuming.
Copper(I) bromide has been dissolved in various ionic liquids
for the continuous gas-phase oxycarbonylation of methanol,
which has been found to be an active catalyst that increased
[20]
the reaction rate and stability, but halide loss through the
gas stream was identified as the main reason for catalyst deac-
tivation.
Recently, chloride-free Cu/activated carbon (AC) catalysts
have been an area of significant research because the deacti-
[
21]
vation caused by chloride loss can be ultimately avoided.
Our previous studies have provided strong evidence for the
successful introduction of highly active Cu/AC catalysts into
DMC synthesis by using density functional theory (DFT) calcu-
Figure 1. DMC formation rate over Cu/AC catalysts at different reaction tem-
peratures (T=353–383 K).
[
22]
[21]
lations combined with experimental research, which con-
0
firmed that it is Cu that acts as the active center for DMC for-
mation (see the Supporting Information, Figures S1 and S2). In
addition, Hua et al. reported that the morphology of NPs de-
termines the exposed crystal planes and thus significantly in-
[23]
fluences the catalytic performance. Considering the excellent
properties of hollow NPs, the replacement of copper solid par-
ticles by the corresponding hollow particles is suggested to
provide a larger specific surface area and unprecedented cata-
lytic performance for DMC synthesis. In addition, it remains
a significant challenge to develop new catalysts with high se-
lectivity and stability. Inspired by the thoughts of all the above,
a significant improvement in the fabrication of analogous cata-
lysts for DMC synthesis is proposed.
Herein, we have successfully produced a new series of AC-
supported hollow Cu NPs catalysts by a template-free, micro-
wave-assisted irradiation–carbothermic reduction, which makes
it quite attractive for practical applications because of the
rapid facile synthesis, low cost, and high selectivity. Moreover,
we have studied the effects of different microwave irradiation
conditions on the fine structures of the hollow Cu NPs, as well
as their catalytic performance. Furthermore, the formation
mechanism of such architectures and experimental results to-
gether with the corresponding theoretical calculations on the
adsorption of CO on different surfaces of Cu in the rate-limit-
ing step are also studied in detail.
Figure 2. Arrhenius plots for the oxidative carbonylation of methanol over
Cu/AC catalysts (T=353–383 K).
À7
À2 À1
a maximum of 2.7”10 molm
s
when the temperature
was 383 K, which can be attributed to the large active surface
area of the hollow structures with more active sites. The corre-
sponding apparent activation energy (E ) and pre-exponential
a
factor (A) were obtained from the Arrhenius plots and are
shown in Table 1, with Cat-360 having a minimum E =41.5Æ
a
À1
3.8 kJmol , which was less than the previously reported value
À1 [24]
of 64 kJmol . In addition, the value of E was found to in-
a
Results and Discussion
Effects of reaction temperature on DMC formation rate
Table 1. Kinetic parameters for the oxidative carbonylation of methanol
over Cu/AC catalysts.
To assess the effect of reaction temperature on the vapor-
phase oxidative carbonylation of methanol, DMC production
rates were measured as a function of temperature in the tem-
perature range 353–383 K, as shown in Figure 1 and Figure 2.
The temperature was found to affect the DMC synthesis reac-
tion very favorably; reaction rates with all materials increased
with increasing temperature. In addition, a smaller difference
between reaction rates is observed at lower temperatures of
the temperature regime, whereas at higher temperatures, the
difference in reactivity becomes larger. Notably, reaction rates
with Cat-360 (NPs irradiated from room temperature to 3608C)
change more rapidly compared with the others and reached
[
a]
[b]
Catalyst
E
[
a
A
À1
À1
kJmol
]
[s
]
Cat-100
Cat-250
Cat-300
Cat-360
389.0Æ40.2
158.3Æ14.7
89.4Æ9.1
41.5Æ3.8
91.7Æ9.5
166.1Æ17.3
74.2Æ7.1
79.2Æ7.4
65.8Æ5.9
0.05
0.6
3.7
34.4
1.5
0.1
10.8
5.6
33.1
Cat-400
Cat-450
Cat-360-3
Cat-360-10
Cat-TF-60
[a] Apparent activation energy. [b] Pre-exponential factor.
ChemCatChem 2016, 8, 861 – 871
www.chemcatchem.org
862
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
crease with an increase in irradiation temperature and irradia-
tion time. Such an increase in E indicates that Cat-360 without
a
additional irradiation is more favorable for the formation of
DMC.
Cu/AC catalysts prepared by irradiation at different temper-
atures
XRD analysis
The XRD patterns of Cu/AC catalysts prepared by irradiation at
different temperatures varying from 100 to 4508C are shown
in Figure 3. It is clear that when the sample is irradiated to
Figure 4. TEM images (insets are HR-TEM images) of Cu/AC catalysts pre-
pared at different irradiation temperatures. The inset image in the lower
right corner of (d) is the corresponding SAED pattern.
Figure 3. XRD patterns of Cu/AC catalysts obtained at different microwave
irradiation temperatures.
1
008C, only diffraction peaks related to Cu (OH) NO are de-
TEM analysis
2
3
3
tected, which implies that the precursor has not decomposed
at this low temperature. As the irradiation temperature increas-
es above 2508C, additional peaks corresponding to CuO
appear, whereas the Cu (OH) NO peaks disappear, which can
As shown in Figure 4(a–f), the TEM images reveal the interest-
ing hollowing process of the Cu/AC catalysts prepared by irra-
diation to different temperatures, which are vital in illuminat-
ing the formation mechanism of the hollow NPs. On the basis
of former reports, it is speculated that the Ostwald ripening
process includes three stages: particles aggregates, core–shell
2
3
3
be attributed to the thermal decomposition of Cu (OH) NO to
2
3
3
CuO, HNO , and H O, and then, HNO resolves into NO , O ,
3
2
3
2
2
[
25]
and H O. This means solid and amorphous spheres are ob-
2
[14,26]
[28]
tained in the initial stage.
The observed main diffraction
structures, and hollow spheres. It is difficult to identify clear
peaks at 43.38, 50.58, and 74.28, which corresponded to the
characteristic facets of (111), (200), and (220) of active crystal-
line Cu (JCPDS 65-9026), and the additional diffraction peaks of
particles in Figure 4(a) when the irradiation temperature is
1008C because of the undecomposed precursor. At 2508C,
owing to surface energy minimization, redistribution occurs in
the process of particles mixing randomly and generating inter-
nal voids; thus, irregular particles emerge, which is the CuO
phase (Figure 4(b)) as determined by analyzing the lattice
spacing. These are ascribed to the thermal decomposition of
the precursors depending on the irradiation temperature.
When the temperature is 3008C, the inner core of the crystal-
lites, which have a higher surface energy and a nanoscaled di-
ameter, dissolves and transfers to the outside of the shells to
redeposit on the better crystallized nanoparticles of the
Cu O (JCPDS 05-0667) at 36.48, 52.58, and 61.38 are prominent
2
when the temperature is elevated to 3008C. These arise from
[27]
the microwave-induced carbothermic reduction.
The XRD
patterns at this temperature indicate that the products are
mixtures of Cu and Cu O. Regarding the synthesis of hollow in-
2
teriors, it is recognized that the ripening process may involve
[
28]
the mass transfer between the solid core and outer particles.
Note that the full width at half maximum of the Cu peaks is
decreased at 3608C, illustrating that Cu particles with smaller
[
29]
[31]
sizes are well dispersed on the surface of the supports. Fur-
thermore, the intensity of the diffraction peaks of Cu tend to
strengthen when increasing the temperature continually, indi-
cating that high-temperature roasting (>3608C) may lead to
shells. The sharp contrast between the dark edges and pale
centers in the TEM micrograph (Figure 4(c)) reveals the forma-
tion of hollow structures with small interior cavities, and the
particles were determined to be formed of Cu O and Cu. When
2
[30]
active metal growth and even cause them to unite.
the temperature is increased to 3608C, the solid cores are con-
ChemCatChem 2016, 8, 861 – 871
www.chemcatchem.org
863
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
sumed and disappear, forming hollow cavities inside the
spheres. The surface of the AC is coated with highly dispersed
and well-defined hollow structures with a large cavity, which
can be confirmed as mainly Cu with exposed {111} crystal
planes on determining a lattice spacing of 0.21 nm (Fig-
ure 4(d)). In its corresponding selected-area electron diffraction
(
SAED) pattern, the spots with a circular symmetry indicate the
0
formation of the single-crystalline nanocopper with Cu sur-
faces, in agreement with the XRD analysis. For Cat-400, the
shell thickness increased gradually, but the {111} crystal planes
were still exposed, as shown in Figure 4(e). In particular, owing
to the exceptional advantage of the uniform heating of micro-
wave irradiation, which leads to homogeneous structures of
[
14]
NPs,
the typical exposed Cu(111) crystal planes probably
Figure 5. CO-TPD patterns of Cu/AC catalysts prepared under different irradi-
ation temperatures.
[
23]
benefit to a great extent. Interestingly, there is a study re-
porting that active centers that have a uniform morphology
and selectively expose a single type of crystal plane are prom-
ising as highly selective catalysts. Hence, we determined that
the uniform hollow Cu NPs with a high density of Cu(111)
facets should be favorable for high selectivity for DMC forma-
tion.
Table 2. CO-TPD data for Cu/AC catalysts prepared at different irradiation
temperatures.
Catalyst
T
[8C]
Total area
[a.u.]
Peak area ratio
[%]
On further raising the irradiation temperature, however, it is
found (Figure 4(f)) that the cavities become irregular in shape,
and some are even unclear compared with others, which is at-
tributed to the agglomeration of nanocrystals, resulting in the
distortion of the cavities owing to their extrusion into each
other. Automatically, the specific active surface areas of the
hollow Cu NPs and the ideal Cu(111) facets are undoubtedly
reduced. This phenomenon can be clearly seen in the inset of
Figure 4(f). Consequently, the conclusion can be drawn that
the formation of cavities depends significantly on the irradia-
tion temperature, and that cavities evolve initially by growing
larger, then undergo irreversible distortion before finally disap-
pearing. The optimum irradiation temperature for this process
is 3608C, which can prevent high-temperature-induced mor-
phology changes. This conclusion is also supported by the
XRD results.
Cat-AC
Cat-100
Cat-250
Cat-300
Cat-360
Cat-400
Cat-450
–
–
235.9
265.5
234.4
–
–
–
–
144.1
145.4
145.3
146.4
125.1
120.5
249.5
387.7
314.1
308.3
183.4
156.7
45.4
51.6
62.8
–
–
89.7
54.6
48.4
37.2
–
–
10.3
–
250.5
a well-resolved peak around 1408C, which turns out to be the
best desorption capability of weakly adsorbed CO (Table 2).
Previous theoretical investigations indicated that CO adsorp-
tion on the Cu(111) surface is both thermodynamically favored
[
33]
and is a non-dissociative adsorption.
Thus, a preliminary
summary is that the weak CO desorption capability is probably
related to the selectively exposed surfaces of the Cu(111) crys-
tal planes. However, for Cat-450, in addition to a weak desorp-
tion peak around 1408C, a particularly weak peak around
2
508C is observed, which can be attributed to the small quan-
CO temperature-programmed desorption (TPD) analysis
tity of Cu O supported on the AC as the interaction between
2
+
0
The CO-TPD profiles of the Cu/AC catalysts prepared at various
irradiation temperatures are presented in Figure 5, and the
Gauss–Newton curve-fitting results are given in Table 2. As de-
picted in Figure 5, because AC does not show any desorption
properties in a control TPD spectra, the two peaks around
Cu and CO is much stronger than that between Cu and CO,
as indicated in Figure S3 in the Supporting Information.
Structures and energetics of reactants absorbed on Cu(111),
Cu(220), and Cu(200)
1
408C and 2408C that are observed for the samples Cat-100,
Cat-250, and Cat-300 can be assigned to the desorption of CO
molecules interacting with the catalysts weakly and strongly,
To analyze the details of the experimental results by using
a theoretical approach, various possible adsorptions of CO on
the Cu(111), Cu(200), and Cu(220) planes have been consid-
ered. The proposed reaction mechanism is illustrated in
[
32]
respectively. When increasing the irradiation temperature to
008C, the intensity of the CO desorption at 2408C decreases
3
[
22]
sharply and even becomes negligible. Correspondingly, the
ratio of these two peaks increases slowly (45.4/54.6, 51.6/48.4,
Scheme 1. In the chemical equations, the active site is de-
noted by an asterisk. By analogy, the notation (Z)* indicates an
active site interacting with species Z. Studies on the mecha-
nism of DMC synthesis have shown that CO insertion into the
adsorbed CH O to produce the adsorbed CH OCO species is
6
2.8/37.2), and the weak CO desorption capability increases
from 113.3 to 197.3. Subsequently, at 3608C and 4008C, only
the weak desorption peak around 1408C is observed, and the
total amount of the weakly adsorbed CO on Cu decreases
from 308.3 to 183.4. Of the catalysts, only Cat-360 exhibits
3
3
[
22,34]
proposed to be the rate-determining step.
Therefore, in
this section, the adsorption energy (Eads), activation barriers,
ChemCatChem 2016, 8, 861 – 871
www.chemcatchem.org
864
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Scheme 1. The proposed mechanism of DMC formation by oxidative carbon-
ylation of methanol.
and reaction energies of the CO insertion reaction on different
crystal planes of Cu are discussed, and the corresponding re-
sults are summarized in Figure 6 and Table 3. Here, the adsorp-
Figure 7. Optimized co-adsorbed structures and transition states on different
0
Cu surfaces of Cu /AC involved from Equation (4). Dark orange balls repre-
sent Cu atoms, red balls represent O atoms, gray balls represent C atoms,
and white balls represent H atoms throughout this study.
Figure 6. Energy profiles with the energy values involved in the reaction of
CO insertion to methoxide species on different Cu surfaces.
À1
mic and follows the order Cu(111) (37.9 kJmol )<Cu(220)
À1
À1
(
38.5 kJmol )<Cu(200) (50.7 kJmol ), indicating that
Cu(111) is the most favorable for the process. In other words,
the geometries of the weak interaction modes for CO on the
Cu(111) facets are thermodynamically favored, which is re-
sponsible for the formation of DMC. These results fit quite well
with the previous experimental observations.
Table 3. The adsorption energies of reactants and intermediates ad-
sorbed on different surfaces.
À1
Species
E
ads [kJmol
Cu(220)
]
Cu(111)
Cu(200)
CO
À81.6
À256.2
À175.6
À95.0
À268.0
À198.7
À101.3
À267.5
À203.2
OCH
CH OCO
3
Catalytic test results
3
The catalytic results with the catalysts synthesized by irradia-
tion to different temperatures are presented in Table 4. It was
observed that as the irradiation temperature increases from
tion energy is used to characterize the strength of CO bonding
on the Cu surfaces and is defined as Eads =E(CO/Cu)ÀE(Cu)À
E(CO), where E(CO/Cu) is the total energy for Cu with the ad-
sorbed CO on the surface, E(CO) is the total energy of free CO,
and E(Cu) is the total energy of the bare slab of the surface.
The selected optimized configurations of the reactants, transi-
tion states (TSs), and products involved in Equation (4) are
shown schematically in Figure 7.
1008C to 3608C, the conversion of CH OH (CCH3OH), the selectiv-
3
ity (SDMC), turnover frequency (TOF), and rate for DMC forma-
tion increase remarkably (i.e., from 2.0 to 5.8%, 78.4 to 100%,
À5 À1
À7
À2 À1
0.9 to 8.4”10 s , and 0.1 to 2.7”10 molm s , respec-
tively) and reach a maximum value at 3608C, then decline dra-
matically beyond that point. As expected, Cat-360 shows the
À5 À1
highest TOF value of 8.4”10 s . To justify the good catalytic
It was found that the crystal plane is very sensitive to these
thermodynamic parameters. By comparison, the binding of the
performances of Cat-360, the observed rate was compared
with reported values; Li et al. reported a production rate for
À1 À1
[35]
[36]
CO molecule to the Cu(111) surface is found to be distinctly
DMC of 229 mgg
h
over Cu/AC at 393 K, Ma et al. ach-
À1
À1 À1
weaker (Eads =À81.6 kJmol ) than to Cu(220) (Eads
=
ieved the best production rate of 117 mgg
h
on Cu O/AC
2
À1
À1
À5 À1
+
À95.0 kJmol ) or Cu(200) (E =À101.3 kJmol ), as shown in
at 413 K, and 3.8”10
s
was reported over a Cu Y zeolite
ads
[
34]
Table 3. Nevertheless, the weakly adsorbed CO can decrease
by Bell and Zhang at 403 K.
Importantly, all these values
the activation energy for further reaction. Interestingly, Cu(111)
were obtained with more severe reaction conditions than
those used with Cat-360. Thus, it appears that the hollow
structure NPs in this study are more conducive to the reaction
than solid ones. Additionally, the SDMC of Cat-360 is up to
À1
(
93.9 kJmol ) alone can reduce the activation barrier com-
À1
pared
with
Cu(220)
(140.8 kJmol )
and
Cu(200)
À1
(
181.0 kJmol ), giving an insertion reaction that is endother-
ChemCatChem 2016, 8, 861 – 871
www.chemcatchem.org
865
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Table 4. Results of chemisorption and intrinsic catalytic synthesis over various Cu/AC catalysts prepared at different irradiation temperatures.
7
5
Catalyst
d
Cu
d
[nm]
Cu
d
[nm]
Cu
Cu dispersion
(DCu) [%]
S
Cu
2
C
[%]
CH3 OH
S
[%]
DMC
STY
[mgg
Rate(”10 )
[molm
TOF(”10 )
[
a]
[b]
[c]
[d]
À1 [e]
À1 À1
À2 À1 [f]
À1
[nm]
[m g
]
h
]
s
]
[s
]
Cat-100
Cat-250
Cat-300
Cat-360
Cat-400
Cat-450
–
–
–
–
–
–
0.5
1.2
7.8
10.5
6.1
3.6
8.9
–
0.7
1.3
9.2
15.7
7.4
5.3
2.4
–
2.0
2.6
3.5
5.8
2.6
2.2
9.2
9.2
–
78.4
82.2
87.1
100.0
100.0
80.9
59.3
61.0
85.0
81.1
280.2
333.1
596.9
303.6
253.3
229
0.1
0.7
1.1
2.7
0.9
0.5
–
0.9
2.2
3.9
8.4
3.8
2.1
–
35.7
51.4
55.1
67.6
12.0
–
20.5
34.6
39.2
49.3
8.0
–
34.1
50.0
53.9
65.3
11.0
–
[35]
Cu/AC
Cu O/AC
[36]
2
117
–
–
–
–
3.8
+
[34]
Cu Y zeolite
–
–
–
–
–
[
a] Average particle size determined by TEM. [b] Determined by N
Scherrer equation and XRD data. [d] DCu =the amount of exposed Cu on the surface of the catalysts from Equation (9). [e] The specific copper surface area
determined by N O chemisorption calculated from Equation (8). [f] DMC formation rate.
2
O chemisorption from Equation (10). [c] Calculated from the Cu(111) plane by using
2
1
3
00%; moreover, no byproducts are detected. In addition, Cat-
Figure 8. For this, the catalysts were heated from room tem-
perature to the desired irradiation temperature, which was
then held for 0, 3, or 10 mins. The diffraction peaks centered at
2q=43.38, 50.58, and 74.28 are attributed to crystalline Cu
(JCPDS 65-9026), whereas the weak and broad diffraction
peaks located at 2q=36.38, 42.28, and 61.28 (JCPDS 05-0667)
60 shows a much higher CCH3OH value than the other five sam-
ples, which are all less than 4%. It is believed that C
is de-
CH3OH
[37]
pendent on the amount of surface Cu atoms. Interestingly,
the 100% selectivity is remarkably higher than any previously
reported values, with the exception of that obtained on the
[38]
CuCl/MCM-41 catalyst by Li et al.
can be assigned to crystalline Cu O, indicating that the copper
2
Combining the results of XRD, TEM, CO-TPD, and the DFT
calculations stated above, it is clear that the significant varia-
tions of TOF (and rate) for DMC formation depend strongly on
the increased specific surface area of the hollow Cu structures
possessing the largest cavity, which provides more available
active centers for the reactants. Regarding the high selectivity
for DMC, this can be attributed to the adequately exposed
Cu(111) crystal planes that are identified as the catalytically
active sites for the production of DMC under the optimal con-
species have undergone the transformation from precursor to
copper oxide and then to metallic Cu and Cu O under the mi-
2
[
25a,27]
crowave-induced carbothermic reduction at 3608C.
In ad-
dition, the intensity of the Cu species becomes sharper with in-
creasing reaction time, indicating that the metal particles grow
and even aggregate in the form of micro-crystals in this situa-
[
27c]
tion.
Hence, the decreasing specific surface area is partially
[
19c]
attributed to the increase in crystalline Cu.
The average out-
side diameters of Cat-360-0, Cat-360-3, and Cat-360-10 are
50.0, 53.6, and 57.6 nm, respectively, calculated by using the
Debye–Scherrer equation. It is clear that the sizes of the cata-
lysts show a considerable increase on increasing reaction time,
implying the coalescence of partial particles. This indicates that
the retention time has a significant influence on controlling
the crystallite growth, which further affects the textural struc-
ture and specific surface area.
[22,39]
ditions of microwave irradiation.
In addition, it is suggested that the Cu(111) crystal planes
are dominant for the weak adsorption of CO around 1408C
rather than the intense peak around 2408C, which is probably
related to Cu O species, hindering the formation of the meth-
2
[32a]
oxide species and thus the formation of DMC;
this further
inhibits the side reactions and hence improves the selectivity
for DMC. Interestingly, note that the sequence of increasing
surface concentration of weakly adsorbed CO on the function-
ing catalysts is in the order Cat-360>Cat-300>Cat-400ꢀCat-
2
50>Cat-450ꢀCat-100, in line with the observed sequences
for TOF and selectivity of DMC. Therefore, we strongly believe
that the exposed Cu(111) facets can be identified as the best
adsorption sites for the weak adsorption of CO. Coincidentally,
the preferential adsorption further determines the higher se-
lectivity of the catalyst for the oxidative carbonylation of meth-
anol, and the TOFDMC also is enhanced accordingly.
Cu/AC catalysts prepared at different retention times
XRD analysis
XRD patterns of Cat-360 synthesized with different retention
times, Cat-360-0, Cat-360-3, and Cat-360-10, are displayed in
Figure 8. XRD patterns of Cu/AC catalysts prepared at different irradiation
times.
ChemCatChem 2016, 8, 861 – 871
www.chemcatchem.org
866
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
TEM analysis
Catalytic test results
The microstructures of Cat-360 at different reaction times were
further studied by TEM, as shown in Figure 9(a–c). For all three
samples treated by microwave irradiation, the surface of the
catalysts is covered with hollow metal particles majorly com-
As shown in Figure S4 and Table S1 (in the Supporting Infor-
mation), the AC was dominated by micropores, implying the
micropore structure is not blocked by Cu nanoparticles. There-
fore, the as-prepared supported catalysts with porous channels
can ensure high permeation and mass-transfer rates for species
involved in a catalytic reaction. It is clearly observed that for
Cat-360, increasing the irradiation time leads to a remarkable
prising Cu(111) and to a lesser extent Cu O(111). In the case
2
where the temperature was not held at 360 C for any signifi-
cant time (Cat-360-0), hollow structures with the largest cavity
of approximately 35 nm diameter were formed; moreover, cer-
tain “open-mouth hollow spheres” are also present, as indicat-
ed by the black arrows in Figure 9(a), and in the inset, a high-
magnification image of an individual open-mouth sphere clear-
ly shows that the opened sphere is hollow, which increases the
specific surface area for reactants to some extent. When the
catalysts are collected after 3 min, in addition to some hollow
metal particles with a smaller cavity of 15 nm, solid particles
start to emerge. As the irradiation duration is extended to
decrease of more than 47% of the TOFDMC, from 8.4 to 4.4”
À5 À1
10 s ; simultaneously, the values of both C
and SDMC
CH3OH
substantially decrease from 5.8 to 3.2% and from 100 to
80.2%, respectively, as demonstrated in Table 5. This behavior
corresponds well with the morphology-dependent catalytic se-
[
23]
lectivity of the hollow nanocrystals. On the basis of the re-
sults of XRD and TEM, the time taken for microwave irradiation
has a tremendous effect on the development of hollow struc-
tures and volume sizes, which in turn affect the active surface
areas, further determining the catalytic performance. Mean-
while, the amount of exposed Cu(111) crystal planes is also af-
fected, which relatively sensitively affects the catalytic selectivi-
1
0 min, the quantities of solid particles increase significantly,
whereas the hollow structures become almost ambiguous, and
noticeable distortion occurred unexpectedly (Figure 9(c)) be-
cause of serious aggregation and sintering of Cu particles. In
contrast, in the catalysts roasted by the conductive heating
method, Cat-TF-60, it is very difficult to find any hollow parti-
cles (Figure 9(d)), and the irregular solid particles are approxi-
mately 42.5 nm in size based on 50 particles randomly select-
ed. The corresponding SAED pattern shows concentric diffrac-
tion rings, indicating the polycrystalline nature of the sphere
[
22]
ty. Moreover, the catalysts roasted by the tubular furnace
barely have any hollow particles, so express lower reaction
rates and selectivity for DMC than the optimal catalyst irradiat-
ed by a microwave oven; this further highlights the advan-
tages of the microwave oven, especially the characteristic of
the quick rundown of irradiation temperature.
(
Figure 9(d)). All rings can be indexed to the diffractions from
The formation mechanism of hollow copper nanostructures
0
Cu O and Cu . These results agree well with the XRD patterns
2
mentioned above. In summary, the size of the cavity and spe-
cific surface area of hollow Cu NPs can be effectively controlled
by varying the irradiation time.
On the basis of the analytical results of XRD and TEM, a possi-
ble mechanism for the formation of hollow Cu nanostructures
is proposed, as illustrated in Scheme 2.
Primarily, the outer layer of Cu(NO ) is decomposed in situ
3
2
to produce CuO nanostructures under vacuum without any
significant collapse. Subsequently, almost all the CuO particles
are immediately reduced in situ to produce a high concentra-
tion of Cu and are surrounded by Cu(111) crystal planes be-
cause of the microwave-induced carbothermic reduction when
[
27a,b]
the temperature is above 3008C.
These are deposited on
the surface of the precursor, forming a thin discontinuous
metal shell, which allows the out-diffusion of the inner parti-
cles. The precursor inside failed to decompose because of the
large temperature difference between the inside and outside,
thus leading to the formation of Cu(NO ) @Cu core–shell struc-
3
2
tures. The core begins to continually pyrolyze depending on
[
40]
thermal conduction, wherein the released gas
and excess
Gibbs free energy stored in the amorphous structure work as
the driving force for the outward diffusion of the core spe-
[
41]
cies, leading to a continual decrease in the core size, and
eventual depletion accompanied by the well-known inside-out
[
42]
Ostwald ripening process. Therefore, the internal precursors
acted as the cores and are considered as sacrificial templates
in the formation of the hollow structure. As the reduction and
decomposition proceeded, the decomposed materials originat-
ing from the cores constantly attached to the previously
formed shells and are subsequently reduced to low-valent
Figure 9. TEM images (insets are HR-TEM images) of Cu/AC catalysts pre-
pared with different irradiation times and Cu/AC catalyst roasted in a tubular
furnace (d; inset is the HR-TEM image and corresponding SAED pattern).
ChemCatChem 2016, 8, 861 – 871
www.chemcatchem.org
867
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Table 5. Results of chemisorption and intrinsic catalytic synthesis over various Cu/AC catalysts prepared with different irradiation times.
7
5
Catalyst
d
Cu
d
[nm]
Cu
d
[nm]
Cu
Cu dispersion
(DCu) [%]
S
Cu
2
C
[%]
CH3OH
S
[%]
DMC
Rate(”10 )
[molm
TOF(”10 )
[a]
[b]
[c]
[d]
À1 [e]
À2 À1 [f]
À1
[nm]
[m g
]
s
]
[s
]
Cat-360-0
Cat-360-3
Cat-360-10
Cat-TF-60
51.4
54.3
60.1
42.5
34.6
38.7
43.1
23.9
50.0
53.6
57.8
40.7
10.5
6.3
2.9
15.7
10.4
9.9
5.8
3.3
3.2
3.6
100
80.2
81.5
79.3
2.7
1.5
1.2
1.7
8.4
4.7
4.4
5.3
8.4
11.5
[
a] Average particle size determined by TEM. [b] Determined by N
Scherrer equation and XRD data. [d] DCu =the amount of exposed Cu on the surface of the catalysts from Equation (9). [e] The specific copper surface area
determined by N O chemisorption calculated from Equation (8). [f] DMC formation rate.
2
O chemisorption from Equation (10). [c] Calculated from the Cu(111) plane by using
2
Scheme 2. Schematic of the Ostwald-ripening-coupled self-assembly process
of the hollow Cu structures.
Figure 10. Microwave calibration curve.
[
14]
metal species; finally, a compact metal shell is fabricated. In
this manner, various metal nanostructures with hollow interiors
can be perfectly synthesized.
Notably, such stable, hole-like microspheres display a perfect
catalytic rate of DMC production compared with other sam-
ples, ascribed in part to the large specific surface area and the
exclusive exposed Cu(111) crystal planes, which provide more
access for the catalytic reaction. Therefore, to some extent,
one can envisage that such structural characteristics as high-
energy facets and high surface area are propitious to the for-
mation rate and selectivity of the target product.
Interestingly, the features of the “open-mouth hollow
spheres” in the inset of Figure 9(a) can be observed for some
copper NPs, further indicating that the NPs are actually hollow
spherical structures. As for the formation of fresh holes in the
shell, conventionally, it can be interpreted as follows. Because
oxide grows over the perimeter of metal precursors up to a cer-
tain critical thickness and then expands in size through a grad-
ual process, the shape of the void is related to the initial shape
[43]
of the metal precursors, which acts as a so-called template
Conclusions
[15]
for creating the shape of the void. Note that the voids of
the hollow copper structures are generally not spherical. Thus,
compressive stress builds up because of the internal thermal
decomposition of the precursors, and the released gas ex-
pands, which causes strong propulsion, leading to size expan-
sion. Furthermore, the hollow interior, particularly with a non-
spherical void and thin shell, is neither energetically favorable
nor has sufficient mechanical strength to sustain the internal
pressure compared with the external vacuum environment.
These reasons might contribute to the instability of hollow
The development of catalysts with high selectivity and forma-
tion rate for DMC synthesis has been achieved for the first
time. The optimal sample characterized for the hollow copper
NPs has large cavities of approximately 35 nm diameter and
50 nm outside diameter, it is well dispersed on the AC support,
and is rapidly fabricated by heating the precursor from room
temperature to 3608C within a mere 8 min in a special micro-
À5 À1
wave heating system. A high TOFDMC value of 8.4”10
s
with CCH3OH of 5.8% and 100% SDMC are obtained, which are
much better than the previously reported Cu–zeolites or AC-
supported copper catalysts. Notably, the attractive catalytic
performance of the Cu/AC catalysts is attributed to the high
surface area of the hollow copper NPs, whereas the excellent
selectivity for DMC can be attributed to the completely ex-
posed Cu(111) crystal planes. DFT calculations show that the
[44]
copper NPs and the breakage of cavities. Specifically, if an-
nealing is performed immediately when the temperature
reaches the desired value of 3608C, the morphology of the
cracked hollow structures would be retained to form stable
holes, utilizing the advantage of the microwave not having
thermal inertia, which is not possible in the conductive heating
method.
À1
adsorption energy of CO on Cu(111) (À81.6 kJmol ) is weaker
ChemCatChem 2016, 8, 861 – 871
www.chemcatchem.org
868
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
than that on Cu(220) and Cu(200), which contributes to the
weak adsorption of CO and further formation of DMC.
Characterization technologies
X-ray diffraction (XRD) data were recorded by using an X-ray dif-
fractometer (D/Max 2500, Rigaku Industrial Co., Tokyo, Japan) with
The formation mechanism of the novel hollow copper mi-
crostructures can be elucidated by Ostwald ripening, wherein
the initial stage is the aggregation of primary NPs collected by
the decomposition of Cu(NO ) , followed by the hollowing of
a CuK radiation source (l=1.54056 Š) and a graphite monochro-
a
mator (operating at 40 kV and 100 mA). The range of the spectral
scan (2q) was 108–808 and was obtained at a rate of 88min .
À1
3
2
the core and the progressive redistribution of matter from the
interior to exterior. Finally, a hollow architecture is obtained by
self-assembly under microwave-induced carbothermic reduc-
tion. Unexpectedly, the volume of the cavity is reduced and
disappears completely at an elevated irradiation temperature
or a prolonged reaction time, which gives rise to a correspond-
ing decline in the formation rate and selectivity of DMC.
Transmission electron microscopy (TEM) and selected-area electron
diffraction (SAED) were performed by using a Hitachi H-800 trans-
mission electron microscope operating at 200 kV. The samples for
TEM measurements were prepared by dispersing the products in
ethanol by using an ultrasonic bath for 10 min, and then, a drop of
the suspension was placed on a carbon-coated copper grid at
room temperature.
CO temperature-programmed desorption (CO-TPD) was performed
by using a Finsore-3010 apparatus (Zhejiang FINETEC). A sample
(
50 mg, >200 mesh) was placed in a U-shaped quartz tube and
À1
then heated at 108Cmin to 3008C under a stream of He
Experimental Section
À1
(
20 mLmin ) for 1.5 h. Subsequently, the sample was cooled to
Materials
ambient temperature. To saturate the catalyst, a CO stream
À1
(
20 mLmin ) was injected into the reactor for 60 min. Thereafter,
Copper nitrate (Cu(NO ) ·3H O) was purchased from Sinopharm
3
2
2
the sample was swept with helium to remove the physically ab-
sorbed carbon monoxide. After purging the sample, the furnace
Chemical Reagent Co., Ltd. AC is a commercial product obtained
from Xinhua Chemical Plant (Taiyuan, China). Quartz sand used for
all experiments was obtained from Tianjin Guangfu Technology De-
velopment Co., Ltd. All chemicals were of analytical grade and
used without any further purification.
temperature was increased to 6008C at
a heating rate of
À1
1
08Cmin under a helium flow.
[45]
N O pulse chemisorption
2
of the catalysts was performed by
using a Micrometrics AutoChem II 2920. First, the sample (100 mg)
was placed in a U-shape quartz reactor and each catalyst was re-
duced at 4508C in a 10% H /Ar mixture for 60 min. Then, the reac-
2
tor was purged with N at 508C. Subsequently, N O was used to
2
2
Preparation of Cu/AC catalysts
oxidize surface copper atoms to Cu O (s). The specific surface area
2
of Cu was calculated by dissociative N O adsorption on the surface
of copper with the pulse titration method based on Equation (7):
In a typical synthesis, Cu(NO ) ·3H O (2.4 g) was dissolved in deion-
2
3
2
2
ized water (4.5 mL). The homogeneous solution was impregnated
dropwise into AC (3.2 g, >200 mesh) and then placed in a 708C
water bath for 12 h. Subsequently, the mixture was kept in air at
2 CuðsÞ þ N O ! N þ Cu OðsÞ
ð7Þ
2
2
2
[21]
7
08C for 12 h and then at 908C for 12 h. Finally, the precursor
was placed in an NJZ4-3 microwave experimental setup (Nanjing
Jiequan Microwave Equipment Co., Ltd., Nanjing, China) for irradia-
tion at different temperatures and times under a vacuum of
À0.05 MPa, and then, the Cu/AC catalysts with a Cu loading of
Additionally, a filter filled with soda lime for trapping acidic gas
N O was employed The evolution of N was analyzed by using
2
.
2
a thermal conductivity detector (TCD) to detect the consumption
19
À2
of N O (X) with 1”10 atomsm . The value of d in nanome-
average
1
6.7 wt% were obtained.
2
ters was calculated by assuming a spherical particle shape. The
The temperature in the microwave oven was measured by an infra-
red radiation thermometer and then corrected by a K-type thermo-
couple. The detailed temperature linear calibration procedure was
as follows. The precursor prepared herein was irradiated at a heat-
2
À1
active copper surface areas (m g ), copper metal dispersion (%),
Cu
and average copper particle size (nm) were calculated from Equa-
tions (8)–(10), respectively:
À1
ing rate of approximately 488Cmin to 4508C under a relatively
2
19
XNAa
low power of approximately 320 W, and, simultaneously, the tem-
perature was recorded at an interval of 108C based on the surface
temperature; the surface temperature of the catalyst was mea-
sured by an infrared thermometer. The inside temperature was de-
termined by using a K-type thermocouple and is denoted as the
inner temperature. The temperature calibration curve is shown in
Figure 10. The Cu/AC catalysts irradiated at 100, 250, 300, 360, 400,
and 4508C corresponded to sample inner temperatures in the mi-
crowave oven of 183, 455, 545, 654, 727, and 8178C, respectively,
and were denoted as Cat-100, Cat-250, Cat-300, Cat-360, Cat-400,
and Cat-450. The catalysts irradiated at 3608C for 0, 3, or 10 min
were denoted as Cat-360-0, Cat-360-3, and Cat-360-10, respectively.
Thus, Cat-360 is the same sample as Cat-360-0. For comparison,
the catalyst heated in a GSL-1500X tubular furnace (Hefei Kejing
Materials Technology Co., Ltd., Hefei, China) at 3608C in a 10% H2/
SCu ¼ 1
ð8Þ
ð9Þ
 10  ð1 þ aÞ
2Xð1 þ aÞM
Cu
DCu
¼
a
6
000
daverage ¼ 1
ð10Þ
Cu
 SCu
Where, 2 is a stoichiometric factor, X is the number of moles of
À1
N O experimentally consumed per unit mass of catalyst (molg
),
2
cat
23
À1
19
N is Avogadro’s number (6.022”10 mol ), 1”10 is number of
surface Cu atoms per unit surface area (atomsm ), a is the copper
A
À2
weight percentage in the sample determined by AAS=m /m
(%), MCu is the relative atomic mass of copper (63.46 gmol ).
Cu
AC
À1
N mixture for 60 min was designated as Cat-TF-60.
2
ChemCatChem 2016, 8, 861 – 871
www.chemcatchem.org
869
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Catalytic tests
to validate the optimized TS structures, and TS confirmation was
performed on every TS structure to confirm that they led to the
desired reactants and products.
The catalytic reaction was conducted in a fixed-bed flow system
described in previous studies.
[21,30a]
The experimental conditions
were as follows: reaction pressure P=1.7 MPa, feed gas
The equilibrium lattice constant of Cu is aCu =0.3685 nm, com-
pared with an experimental value of aCu =0.3615 nm. Good agree-
ment was observed between our computed data and the experi-
mental data reported by Kittel. The substrates are modeled as four
atomic layers of metal separated by a vacuum region of 15 Š thick-
ness. The two uppermost substrate layers and reactant molecules
are allowed to relax, maintaining the volume constant. A p(3”3)
unit cell is adopted for the Cu(111), Cu(220), and Cu(200) sur-
faces.
À1
n(CO):n(O )=11/1, CH OH flow rate=0.056 mLmin , gas hourly
2
3
À1
space velocity (GHSV)=4600 h , and the mass of catalyst (m )=
cat
0
.25 g. A given mass of catalyst (0.25 g) and quartz sand (40~60
mesh, 3 g) were transferred into a stainless steel tube reactor that
had an inner diameter of 6 mm and a length of 450 mm. The addi-
tion of quartz sand was to prevent the generation of hotspots in
the catalyst bed. The thermocouple was directly inserted into the
[46]
center of the catalyst bed. Subsequently, methanol (CH OH) was
3
pumped through a Series III microfeeder (America Series Co., Ltd.)
into a preheater, where it was evaporated and mixed with carbon
monoxide (CO) and oxygen (O ), then entered into the reactor and
passed over the catalyst. Flowrates of oxygen and carbon monox-
ide were set by using mass flow controllers (Beijing Sevenstar In-
2
Acknowledgements
This work has been supported by grants from the National Natural
Science Foundation of China (21376159 and 21276169).
struments, Inc). Finally, the feed streams of DMC, CH OH, dime-
3
thoxy methane (DMM), and methyl formate (MF) were injected
into the automatic gas sampler (GS-2, Shanghai Sida, China) and
were analyzed on-line by using a gas chromatograph (GC-950,
Shanghai Haixin, China) equipped with a FID detector. The space
Keywords: Cu(111) · dimethyl carbonate · hollow metal
nanoparticles · microwave irradiation · oxidative carbonylation
À1 À1
time yield (STY, mgg h ), the turnover frequency of DMC (TOF,
À1
s ), and formation rate (rate per unit of Cu surface area,
À2 À1
[1] a) H. Lee, S. E. Habas, S. Kweskin, D. Butcher, G. A. Somorjai, P. Yang,
molm
s ) of DMC, were calculated by using Equations (11)–(13):
Angew. Chem Int. Ed. 2006, 45, 7824–7828; Angew. Chem. 2006, 118,
7
988–7992; b) M. A. Mahmoud, C. E. Tabor, M. A. El-Sayed, Y. Ding, Z. L.
mDMC
STY ¼ m
VCH3OH Â 1CH3OH Â C
 SDMC  MDMC
1
CH3OH
¼
 m
Wang, J. Am. Chem. Soc. 2008, 130, 4590–4591.
2] R. J. Newhouse, H. Wang, J. K. Hensel, D. A. Wheeler, S. Zou, J. Z. Zhang,
J. Phys. Chem. Lett. 2011, 2, 228–235.
3] a) J.-Y. Gong, S.-H. Yu, H.-S. Qian, L.-B. Luo, X.-M. Liu, Chem. Mater. 2006,
18, 2012–2015; b) L. Wang, Y. Ebina, K. Takada, T. Sasaki, Chem.
Commun. 2004, 1074–1075.
[4] a) K. Kamata, Y. Lu, Y. Xia, J. Am. Chem. Soc. 2003, 125, 2384–2385;
b) D. J. Monk, D. R. Walt, J. Am. Chem. Soc. 2004, 126, 11416–11417;
c) S. Anandhakumar, M. Sasidharan, C.-W. Tsao, A. M. Raichur, ACS Appl.
Mater. Interfaces 2014, 6, 3275–3281.
 t
2 Â M
 t
cat
CH3OH
cat
[
[
ð11Þ
mDMC
 DCu  t
V
CH OH Â 1CH OH Â CCH OH Â SDMC Â MDMC
3
3
3
TOF ¼ m
¼
2 Â M
cat
CH3OH
1
Â
mcat  DCu  t
ð12Þ
ð13Þ
nDMC
V
CH OH Â 1CH OH Â CCH OH Â SDMC
1
[
5] Z. Bai, L. Yang, L. Li, J. Lv, K. Wang, J. Zhang, J. Phys. Chem. C 2009, 113,
0568–10573.
6] H. Djojoputro, X. Zhou, S. Qiao, L. Wang, C. Yu, G. Lu, J. Am. Chem. Soc.
006, 128, 6320–6321.
3
3
3
rate ¼ S
¼
 S
1
 t
2 Â M
 t
Cu
CH3OH
Cu
[
2
Where, CCH3OH is the conversion of CH OH (%), SDMC is the selectivity
[7] a) L. Wang, T. Sasaki, Y. Ebina, K. Kurashima, M. Watanabe, Chem. Mater.
2002, 14, 4827–4832; b) F. Caruso, Chem. Eur. J. 2000, 6, 413–419.
3
of DMC (%), DCu is the copper metal dispersion (%), S is the active
Cu
2
À1
[
8] a) Y. Wan, D. Zhao, Chem. Rev. 2007, 107, 2821–2860; b) X. Zhao, X. Y.
Bao, W. Guo, F. Y. Lee, Mater. Today 2006, 9, 32–39; c) Y. Liu, Y. Chu, Y.
Zhuo, L. Dong, L. Li, M. Li, Adv. Funct. Mater. 2007, 17, 933–938; d) T.
He, D. Chen, X. Jiao, Y. Wang, Adv. Mater. 2006, 18, 1078–1082; e) X.
Wang, X. L. Wu, Y. G. Guo, Y. Zhong, X. Cao, Y. Ma, J. Yao, Adv. Funct.
Mater. 2010, 20, 1680–1686; f) Y. Sun, Y. Xia, Science 2002, 298, 2176–
copper surface area (m g ), VCH3OH is the flowrate of CH OH
Cu
3
À1
(
(
0.056 mLmin ), M
is the relative atomic mass of CH OH
is the density of CH OH (0.7918 gmL ).
CH3OH
3
À1
À1
32.04 gmol ), 1
CH3OH
3
Computational details
2179; g) H. Woo, B. Mohan, E. Heo, J. C. Park, H. Song, K. H. Park, Nano-
3
scale Res. Lett. 2013, 8, 390.
All calculations were performed with the Dmol program imple-
mented in the Materials Studio 5.5 package (Accelrys Ltd.). General-
ized gradient approximation functionals by Perdew and Wang
[
9] J. J. Teo, Y. Chang, H. C. Zeng, Langmuir 2006, 22, 7369–7377.
[
10] Y. Xiong, B. Wiley, J. Chen, Z. Y. Li, Y. Yin, Y. Xia, Angew. Chem. Int. Ed.
005, 44, 7913–7917; Angew. Chem. 2005, 117, 8127–8131.
11] a) J. Fei, Y. Cui, X. Yan, W. Qi, Y. Yang, K. Wang, Q. He, J. Li, Adv. Mater.
008, 20, 452–456; b) H. Cçlfen, M. Antonietti, Angew. Chem. Int. Ed.
2005, 44, 5576–5591; Angew. Chem. 2005, 117, 5714–5730.
[12] C. E. Fowler, D. Khushalani, S. Mann, Chem. Commun. 2001, 2028–2029.
[13] a) D. Kim, J. Park, K. An, N.-K. Yang, J.-G. Park, T. Hyeon, J. Am. Chem.
Soc. 2007, 129, 5812–5813; b) X. W. Lou, C. Yuan, Q. Zhang, L. A. Archer,
Angew. Chem. Int. Ed. 2006, 45, 3825–3829; Angew. Chem. 2006, 118,
3909–3913.
2
(
PW91) were used to account for the electronic energy of ex-
[
change correlation. A polarized split-valence basis set, termed as
double numeric polarized basis set, was used. This basis set is com-
parable in size and quality to the Gaussian G** basis set. The inner
electrons of the Cu atoms were kept frozen and were replaced
with an effective core potential. Convergence criteria for the opti-
mization of energy, maximum force, and maximum displacement
2
À5
À1
consisted of threshold values of 2”10 Ha, 0.004 HaŠ , and
[
14] C.-Y. Cao, Z.-M. Cui, C.-Q. Chen, W.-G. Song, W. Cai, J. Phys. Chem. C
010, 114, 9865–9870.
15] L. Wang, F. Tang, K. Ozawa, Z. G. Chen, A. Mukherj, Y. Zhu, J. Zou, H. M.
0
.005 Š, respectively. Hence, we set the threshold value for self-
2
À5
consistent field density convergence to 1.0”10 Ha. To determine
accurate activation barriers, the transition states (TSs) for reactions
were explored by complete linear synchronous transit/quadratic
synchronous transit. In addition, frequency analysis has been used
[
Cheng, G. Lu, Angew. Chem. 2009, 121, 7182–7185.
[16] M. A. Pacheco, C. L. Marshall, Energy Fuels 1997, 11, 2–29.
[17] N. Keller, G. Rebmann, V. Keller, J. Mol. Catal. A 2010, 317, 1–18.
ChemCatChem 2016, 8, 861 – 871
www.chemcatchem.org
870
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
[
[
18] a) K. Tomishige, T. Sakaihori, S.-i. Sakai, K. Fujimoto, Appl. Catal. A 1999,
81, 95–102; b) R. Jiang, Y. Wang, X. Zhao, S. Wang, C. Jin, C. Zhang, J.
Mol. Catal. A 2002, 185, 159–166.
19] a) S. King, J. Catal. 1996, 161, 530–538; b) S. King, Catal. Today 1997,
[33] Z.-J. Zuo, W. Huang, P.-D. Han, Z.-H. Li, Acta Phys. Chim. Sin. 2009, 25,
2507–2512.
[34] Y. Zhang, A. T. Bell, J. Catal. 2008, 255, 153–161.
[35] G. Zhang, Z. Li, H. Zheng, T. Fu, Y. Ju, Y. Wang, Appl. Catal. B 2015, 179,
95–105.
1
3
3, 173–182; c) X. Ding, X. Dong, D. Kuang, S. Wang, X. Zhao, Y. Wang,
Chem. Eng. J. 2014, 240, 221–227.
20] M. J. Schneider, M. Haumann, M. Stricker, J. Sundermeyer, P. Wasser-
scheid, J. Catal. 2014, 309, 71–78.
[36] B. Yan, S. Huang, S. Wang, X. Ma, ChemCatChem 2014, 6, 2671–2679.
[37] G. Merza, B. Lµszló, A. Oszkó, G. Pótµri, K. Baµn, A. Erd o˝ helyi, J. Mol.
Catal. A 2014, 393, 117–124.
[
[
[
[
21] J. Ren, D. Wang, Y. Pei, Z. Qin, J. Lin, Z. Li, Chem. J. Chinese U. 2013, 34,
[38] Z. Li, K. Xie, R. C. T. Slade, Appl. Catal. A 2001, 205, 85–92.
[39] M. Komarneni, J. Shan, A. Chakradhar, E. Kadossov, S. Cabrini, U. Bur-
ghaus, J. Phys. Chem. C 2012, 116, 5792–5801.
[40] J. Zhou, H. Song, X. Chen, L. Zhi, J. Huo, B. Cheng, Chem. Mater. 2009,
21, 3730–3737.
2
594–2600.
22] J. Ren, W. Wang, D. Wang, Z. Zuo, J. Lin, Z. Li, Appl. Catal. A 2014, 472,
7–52.
4
23] Q. Hua, T. Cao, X. K. Gu, J. Lu, Z. Jiang, X. Pan, L. Luo, W. X. Li, W. Huang,
Angew. Chem. Int. Ed. 2014, 53, 4856–4861; Angew. Chem. 2014, 126,
[41] Y. Han, H. Wang, M. Zhang, M. Su, W. Li, K. Tao, Inorg. Chem. 2008, 47,
1261–1263.
4
956–4961.
[
[
24] S. A. Anderson, T. W. Root, J. Catal. 2003, 217, 396–405.
25] a) B. Wang, H. B. Wu, L. Zhang, X. W. D. Lou, Angew. Chem. Int. Ed. 2013,
[42] a) X. W. D. Lou, L. A. Archer, Z. Yang, Adv. Mater. 2008, 20, 3987–4019;
b) X. W. Lou, Y. Wang, C. Yuan, J. Y. Lee, L. A. Archer, Adv. Mater. 2006,
18, 2325–2329; c) M. Choi, K. Na, J. Kim, Y. Sakamoto, O. Terasaki, R.
Ryoo, Nature 2009, 461, 246–249; d) R. Qiao, X. L. Zhang, R. Qiu, J. C.
Kim, Y. S. Kang, Chem. Mater. 2007, 19, 6485–6491; e) J. H. Pan, X.
Zhang, A. J. Du, D. D. Sun, J. O. Leckie, J. Am. Chem. Soc. 2008, 130,
11256–11257.
[43] a) A. Mahapatra, U. Bhatta, T. Som, J. Phys. D 2012, 45, 415303; b) Y.
Deng, H. Wang, L. Xu, Y. Wu, C. Zhong, W. Hu, RSC Adv. 2013, 3, 4666–
4672.
5
2, 4165–4168; Angew. Chem. 2013, 125, 4259–4262; b) J. Ren, M. Ren,
D. Wang, J. Lin, Z. Li, J. Therm. Anal. Calorim. 2015, 120, 1929–1939.
[
[
26] F. Wang, A. Kalam, L. Chang, D. Xie, A. S. Al-Shihri, G. Du, Mater. Lett.
2
013, 92, 96–99.
27] a) W. Yu, B. Yang, X. Chen, W. Jiang, Q. Yu, B. Xu, Vacuum 2012, 86,
005–2009; b) D. Vennerberg, R. Quirino, M. R. Kessler, Adv. Eng. Mater.
013, 15, 366–372; c) J. Bian, X. Wei, Y. Jin, L. Wang, D. Luan, Z. Guan,
2
2
Chem. Eng. J. 2010, 165, 686–692.
[
[
[
28] H. G. Yang, H. C. Zeng, J. Phys. Chem. B 2004, 108, 3492–3495.
29] R. L. Moss, D. Pope, B. Davis, D. Edwards, J. Catal. 1979, 58, 206–219.
30] a) J. Ren, C. Guo, L. Yang, Z. Li, Chin. J. Catal. 2013, 34, 1734–1744; b) D.
Liang, J. Gao, J. Wang, P. Chen, Z. Hou, X. Zheng, Catal. Commun. 2009,
[44] J. Liu, D. Xue, Adv. Mater. 2008, 20, 2622–2627.
[45] P. Munnik, M. Wolters, A. Gabrielsson, S. Pollington, G. Headdock, J.
Bitter, P. De Jongh, K. De Jong, J. Phys. Chem. C 2011, 115, 14698–
14706.
1
0, 1586–1590.
[46] K. Takehira, T. Shishido, Catal. Surv. Asia 2007, 11, 1–30.
[
[
31] A.-X. Wang, D.-Q. Chu, L.-M. Wang, B.-G. Mao, H.-M. Sun, Z.-C. Ma, RSC
Adv. 2014, 4, 7545–7548.
32] a) Y. Zhang, D. N. Briggs, E. de Smit, A. T. Bell, J. Catal. 2007, 251, 443–
4
52; b) M. S. Han, B. G. Lee, B. S. Ahn, D. J. Moon, S. I. Hong, Appl. Surf.
Received: October 29, 2015
Sci. 2003, 211, 76–81.
Published online on January 28, 2016
ChemCatChem 2016, 8, 861 – 871
www.chemcatchem.org
871
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim