One-Pot Synthesis of Cyclodextrin-Doped Cu-SiO2 Catalysts for Efficient Hydrogenation of Dimethyl Oxalate to Ethylene Glycol

Article abstract of DOI:10.1002/cctc.201701246

The β-cyclodextrin (β-CD) was successfully applied to synthesize Cu-SiO₂ catalysts by the sol-gel method. The first and most important question addressed was increasing the copper loading with higher dispersion in the silica support. Another task was to elucidate the structure–performance relationship. Furthermore, catalysts with various β-CD doping levels were systematically characterized and employed in the hydrogenation of dimethyl oxalate (DMO) to ethylene glycol (EG). The 0.1CD-Cu-SiO₂ catalyst with 17.39 wt % copper showed 99.9 % DMO conversion and the highest EG selectivity of 95.0 % at 210 °C, as well a lifetime of over 240 h. The reaction results disclosed that the catalytic performance could be tuned facilely by changing the amount of β-CD doping, which effectively influenced the Cu metallic surface area and dispersion.

Full text of DOI:10.1002/cctc.201701246

Accepted Article
Title: One-pot Synthesis of Cyclodextrin Doped Cu-SiO2 Catalysts for
Efficient Hydrogenation of Dimethyl Oxalate to Ethylene Glycol
Authors: Yuan-Gen Yao, Run-Ping Ye, Ling Lin, Chang-Qing Liu, and
Chong-Chong Chen
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemCatChem 10.1002/cctc.201701246
Link to VoR: http://dx.doi.org/10.1002/cctc.201701246
A Journal of
www.chemcatchem.org

10.1002/cctc.201701246
ChemCatChem
FULL PAPER
One-pot Synthesis of Cyclodextrin Doped Cu-SiO2 Catalysts for
Efficient Hydrogenation of Dimethyl Oxalate to Ethylene Glycol
Run-Ping Ye,[a,b] Ling Lin,[a] Chang-Qing Liu, [a,b] Chong-Chong Chen,[a,b] and Yuan-Gen Yao*[a]
performance owing to smaller Cu particles size, larger BET
surface area, and higher dispersion in silica.[7] Nevertheless,
this excellent method has a shortcoming, namely low copper
loading. In our previous work, pure Cu-SiO2 catalysts with
different copper loadings prepared by sol-gel method
exhibited good catalytic performance, but the deficiency was
that agglomeration among the copper species was serious
when copper loading was higher than 10 wt%.[8]
Karakassides et al. prepared Cu-MCM-41 containing up to
4.7 wt% copper, resulting in the collapse of the ordered
mesoporous framework with Cu content higher than 5 wt%.[9]
We have been seeking ways to reduce the amount of
agglomeration and enhance copper loading. Fortunately, it
was accepted that binuclear copper complex with β-
cyclodextrin (β-CD) could be readily synthesized.[10] Gupta et
al. have employed the Cu(II)-β-CD as water soluble
green catalyst for the one-pot conversion benzyl halide to
benzyl azide and 1,2,3-trizoles at ambient temperature.[11]
Kaifer et al. have reported that CD-capped Pd nanoparticles
were effective heterogeneous catalysts for the hydrogenation
of water-soluble alkenes,[12] as well as for the for the Suzuki
reaction.[13] Meanwhile, Zheng et al. have successfully
prepared more ordered mesoporous SiO2 and TiO2 materials
by introducing β-CD as template with sol-gel method,[14]
which were mainly attributed to the special cylinder-shaped
structure of β-CD. Recently, new kind of non-noble Metal@C
nanoparticles were prepared with EDTA or glucose by Liu et
al. to selective hydrogenation of substituted nitroarenes to
corresponding anilines under mild conditions.[15] Thus, these
similar organic matters have been widely used in the area of
heterogeneous catalysis. Prior to this work, we have also
explored the applications of β-CD to Cu-SiO2 catalysts
prepared by ammonia evaporation method. However, β-CD
was washed away by distilled water after ammonia
evaporation. Furthermore, this method also has some
disadvantages, such as more procedures, sensitive to
evaporation temperature, and emission of high concentration
of ammonia.[16]
Abstract: The β-cyclodextrin (β-CD) was successfully
applied to synthesize Cu-SiO2 catalysts by the sol-gel
method. The first and most important question addressed
was increasing copper loading with higher dispersion in silica
support. Another task was to elucidate the structure-
performance relationship. Furthermore, the catalysts with
various β-CD doping were systematically characterized and
employed in the hydrogenation of dimethyl oxalate (DMO) to
ethylene glycol (EG). The 0.1CD-Cu-SiO2 catalyst with 17.39
wt% copper showed 99.9% DMO conversion and the highest
EG selectivity of 95.0% at 210 °C, as well a lifetime of over
240 h. The reaction results disclosed that the catalytic
performance could be tuned facilely by changing the amount
of β-CD doping, which effectively influenced Cu metallic
surface area and dispersion.
Introduction
Hydrogenation of esters has become an increasingly
significant process for the production of alcohols,[1] especially
ethylene glycol (EG) that possesses attractive applications in
synthesis of polyester, lubricant and antifreeze.[2]
Traditionally, EG is mainly produced from petroleum-derived
ethylene oxide in the industrial approach.[3] With increasing
use of EG and petroleum resource shortage in China,
production of EG is currently interested in dimethyl oxalate
(DMO) hydrogenation and successful industrialization with
an annual capacity of 20 million tons in China.[4] Generally,
versatile Cu-based catalysts are the most extensively used in
vapor-phase hydrogenation of oxalates to EG.[5]
As we all know, there was a direct approach, namely
sol-gel method, to synthesize mesoporous copper silicates
based on suitable organic functional silicon alkoxide.[6]
Zhang et al. prepared Cu-SiO2 catalysts by ammonia
evaporation and sol-gel methods, indicating that the catalyst
synthesized by sol-gel method exhibited better catalytic
Herein, a series of β-CD doped Cu-SiO2 catalysts with
higher copper loading were prepared via a one-pot synthesis
approach based on sol-gel method. Copper-based catalysts
with introduction of β-CD were highly efficient in the vapor-
phase hydrogenation of DMO to EG due to the enhanced
dispersion of copper species and formation of defects. A
probable formation mechanism of β-CD doped Cu-SiO2
[a] Key Laboratory of Coal to Ethylene Glycol and Its Related Technology,
Fujian Institute of Research on the Structure of Matter, Chinese
Academy of Sciences, Fuzhou, Fujian 350002, China.
Dr. Yuan-Gen Yao Fax: +86-591-8371-4946; Tel: +86-591-6317-3138;
E-mail: yyg@fjirsm.ac.cn
[b] University of Chinese Academy of Sciences, 100049, Beijing, P. R.
China
Supporting information for this article is given via a link at the end of the
document.
1
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701246
ChemCatChem
FULL PAPER
catalysts was obtained, as well as deep insight into the
relationship between copper species and catalytic
performance.
more alcohol byproducts under high temperature (~
240 °C).[18] As shown in Figure 1A, the 0.1CD-Cu-SiO2
catalyst exhibited 99.9% DMO conversion and the highest
EG selectivity of 95.0% at 210 °C. It should be pointed out
that the Selec.EG was only 58.4% with 95.0% Conv.DMO for
pure Cu-SiO2 catalyst at 210 °C and the optimal reaction
temperature for it was 240 °C. Catalytic performance (T=
210 °C) over xCD-Cu-SiO2 catalysts with different mass
fraction of β-CD introduction are displayed in Figure 1B.
Along with the increase of doping β-CD, the selectivity of EG
showed volcanic variation, and the highest selectivity of EG
was obtained over the 0.1CD-Cu-SiO2 sample. More details
about DMO conversion and product selectivity as a function
of reaction temperature over other xCD-Cu-SiO2 catalysts
can be found in Figure S1. This result indicated that the
introduction of β-CD had significant effect on the catalytic
performance.
Results and Discussion
Catalytic activity and stability
The gas-phase hydrogenation of DMO was performed on a
fixed-bed reactor at different temperature to discover the
catalytic activities of xCD-Cu-SiO2 catalysts. Many excellent
studies describing DMO hydrogenation comprises several
continuous reactions (Scheme 1), also producing several
byproducts including 1,2-butanediol (1,2-BDO) and 1,2-
propanediol (1,2-PDO).[16b, 17] An obvious feature of the DMO
hydrogenation on Cu-based catalyst is that the product can
be significantly influenced by reaction temperature, generally
producing more MG under low temperature (~ 190 °C) and
Figure 1. (A) DMO conversion and product selectivity as a function of reaction temperature over the 0.1CD-Cu-SiO2 catalyst.
(B) Catalytic performance over xCD-Cu-SiO2 catalysts with different mass fraction of β-CD introduction at 210 °C. Reaction
conditions: P(H2)= 2.0 MPa, WHSVDMO= 0.72 g g-catal-1 h-1, H2/DMO= 50.
Table 1. Physicochemical properties of the xCD-Cu-SiO2 catalysts.
Cu loading C loading N loading
SBET
(m2 g-1)c
319.0
275.6
324.5
294.8
227.8
Vp
(cm3 g-1)d
0.17
Dp
SCu
Cu dispersion
Catalysts
(wt%)a
17.14
17.78
17.39
16.03
15.56
(wt%)b
(wt%)b
(nm)e (m2 g-1)f
(%) f
12.2
13.3
16.5
14.9
8.6
Cu-SiO2
0.85, <0.3 4.44, <0.3
1.52, <0.3 3.39, <0.3
2.72, <0.3 2.04, <0.3
6.65, <0.3 0.63, <0.3
8.99, 5.53 0.41, 0.36
2.7
3.2
2.7
2.2
2.5
13.5
15.2
18.4
15.4
8.7
0.05CD-Cu-SiO2
0.1CD-Cu-SiO2
0.2CD-Cu-SiO2
0.3CD-Cu-SiO2
0.19
0.18
0.12
0.11
a Determined by ICP-OES.
b Determined by EA for the catalyst precursors: the former for the dried samples and the latter for the calcined ones.
c BET specific surface area for the calcined samples.d Total pore volume. e BJH desorption average pore diameter.
f Cu metallic surface area and dispersion determined by N2O titration.
2
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701246
ChemCatChem
FULL PAPER
completely decomposed after calcination. The C content for
the calcined 0.3CD-Cu-SiO2 sample was 5.53 wt%, which
was much higher than other catalysts. Notably, the pure
dried Cu-SiO2 sample showed a 0.85 wt% C loading,
indicating that some TEOS was not completely hydrolyzed
and was coated by the gel. When the β-CD increased, the C
content was effectively increased while N content was
decreased from 4.44 wt% to 0.41 wt%. The only nitrogen
Scheme 1. Reaction pathway for the hydrogenation of DMO
to MG, EG, and EtOH.
-
source was from NO3 group, thus the dried catalyst
precursors exhibited some NO3- group.
SBET and SCu
Table 1 also lists the textual properties of xCD-Cu-SiO2
catalysts with different β-CD amount. It was found that the
Cu-SiO2 catalyst prepared by the sol-gel method possessed
a relatively high SBET of 319.0 m2 g-1 and a small Vp of 0.17
cm3 g-1. The SBET and Vp of xCD-Cu-SiO2 catalysts varied
from the β-CD doping. With the increment of β-CD
introduction from 0.1 to 0.3, both the SBET and Vp decreased
from 324.5 to 227.8 cm3 g-1, 0.18 to 0.11 cm3 g-1,
respectively. The excrescent β-CD loading probably made
the copper species cover the surface and plug some pores,
resulting in a decrease of SBET and Vp. The N2 adsorption-
desorption isotherms of calcined samples was illustrated in
Figure S2, as well as their pore size distribution curves. The
samples presented Langmuir type IV isotherms with H1-type
hysteresis loops, suggesting that they were mesoporous
materials.[19] This phenomenon was in agreement with other
mesoporous Cu-SiO2 catalysts prepared by ammonia-
evaporation method, independent of promoters.[4a, 16b, 20] All
the Dp were about 2.7 nm, which indicated that Cu-SiO2
catalyst doping with β-CD would not influence the pore
shape. In addition, Cu surface area and Cu dispersion of
xCD-Cu-SiO2 catalysts were summarized in Table 1.
Chemisorption of N2O further showed slightly higher Cu
dispersion (16.5%) and more active Cu surface area (18.4
m2 g-1) for 0.1CD-Cu-SiO2 catalyst. However, when excess
amount of β-CD was introduced, the metallic copper would
be agglomerated.
Figure 2. DMO conversion and EG selectivity vs time on
stream over the Cu-SiO2 and 0.1CD-Cu-SiO2 catalysts.
Reaction conditions: T= 210 °C, P(H2)= 2.0 MPa,
WHSVDMO= 0.72 g g-catal-1 h-1, H2/DMO= 50.
To investigate the long-term stability of the 0.1CD-Cu-
SiO2 catalyst, comparison of catalytic activity and selectivity
to EG as a function of time on stream for the pure Cu-SiO2
catalyst are displayed in Figure 2. The β-CD-free Cu-SiO2
catalyst showed an obvious deactivation within 80 h. On the
contrary, the 0.1CD-Cu-SiO2 catalyst displayed excellent
performance in both DMO conversion (~ 99.9%) and
selectivity to EG (~ 95.0%), which was stable for 240 h under
identical reaction conditions. This suggests its promising
applications in industrial catalyst and the inherent reasons
will be further discussed below.
Physiochemical properties of the catalysts
Chemical compositions
TEM images
The real copper loading of the flesh catalysts were
determined by ICP-OES, as shown in Table 1. The preset
values were calculated as the mass of Cu divided by the total
mass of CuO and SiO2. It is clear that the copper loading
was about 17 wt%, which was a little less than the theoretical
copper loading (22 wt%). This indicated that the catalyst
precursors still contained a certain amount of hydroxyl
groups after calcination (300 °C).[8] The 0.3CD-Cu-SiO2
catalyst exhibited the lowest copper loading (15.56 wt%),
which was probably because that some β-CD was not
Figure 3 shows the TEM images of xCD-Cu-SiO2 catalysts
to clarify their morphology. In the calcined samples (Figure
3A-D), no lamellar structure of copper phyllosilicate could be
observed. Inversely, some bulk copper species unevenly
dispersed on the silica support over pure Cu-SiO2 catalyst
(Figure 3A). As the 0.1CD-Cu-SiO2 precursor doped proper
amount of β-CD, which was decomposed after calcination.
Some stacking fault was found in the calcined 0.1CD-Cu-
SiO2 sample (Figure 3B-C), as well as in the calcined 0.2CD-
Cu-SiO2 catalyst (Figure S3A-B). With more β-CD
3
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701246
ChemCatChem
FULL PAPER
introduction, we cannot observe this stacking fault in 0.3CD-
Cu-SiO2 sample (Figure 3D). After reduction at 350 °C,
highly dispersed Cu particles with a mean size of 4.7 nm and
3.3 nm were observed for Cu-SiO2 and 0.1CD-Cu-SiO2
catalysts, respectively. For the reduced 0.3CD-Cu-SiO2
dispersion of copper species with larger particle sizes (Figure
S3C-D). This confirmed again that copper species were
more uniformly dispersed on the SiO2 matrix for 0.1CD-Cu-
SiO2 sample. An excess amount of β-CD loading (x≥ 0.2)
would lead to serious agglomeration, which was in
agreement with the results from N2O chemisorption.
sample, it apparently exhibited
a
non-homogenous
Figure 3. TEM images of (A) calcined Cu-SiO2, (B) calcined 0.1CD-Cu-SiO2, (C) enlargement of ellipse region in (B), (D)
calcined 0.3CD-Cu-SiO2, (E) Cu-SiO2-Reduced, and (F) 0.1CD-Cu-SiO2-Reduced.
Figure 4. XRD patterns of (A) the pure CD, (B) the dried catalyst precursors, (C) the calcined catalyst precursors, and (D) the
fresh reduced catalysts: (a) Cu-SiO2; (b) 0.05CD-Cu-SiO2; (c) 0.1CD-Cu-SiO2; (d) 0.2CD-Cu-SiO2; (e) 0.3CD-Cu-SiO2.
4
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701246
ChemCatChem
FULL PAPER
The H2-TPR measurements were performed to study the
reducibility of the copper species in xCD-Cu-SiO2 catalyst
precursors. It is commonly known that different copper
species probably possess different electron density around
copper, resulting in distinct reduction temperatures.[18] As
can be seen in Figure 5A, each calcined sample exhibited at
least one hydrogen consumption peak with different shapes
at various temperatures. Due to the XRD patterns have
already shown the presence bulk CuO in the calcined xCD-
Cu-SiO2 (x≤ 0.2) samples, the peaks at lower temperature (~
210 °C) and higher temperature (~ 230 °C) can be assigned
to the reduction of well dispersed CuO including ion-
exchanged Cu-O-Si units and the reduction of small CuO
clusters, respectively.[5a, 22-23] The peak centered at 250 °C
for pure Cu-SiO2 catalyst should be attributed to the
reduction of larger CuO clusters.[24] Furthermore, the Tmax
position of second peak was shifting to lower temperature
region with the increase of β-CD doping. When the β-CD
loading was 0.3, the second reduction peak disappeared and
only a peak at 215 °C was observed, which was attributed to
reduction of Cu2O to Cu. The reduction peaks became
narrowing and shifted toward lower temperatures probably
because of the decrease in crystallininty of CuO with
increase in β-CD introduction as evidenced by XRD results.
Besiedes, the calcined 0.3CD-Cu-SiO2 sample had a 29.4%
loss of H2 consumption compared with Cu-SiO2 sample
(Figure 5A). Thus the self-reduction behavior was happened
to it, which was in accordance with the XRD results.
Copper species on the catalysts
XRD analysis
The XRD patterns of the pure β-CD and all the catalyst
samples are shown in Figure 4. The β-CD template
displayed strong characteristic peaks from 2θ of 5° to 45°
(Figure 4A), which were not observed on the dried catalyst
precursors. As clearly depicted in Figure 4B, all the samples
exhibited a broad and diffuse peak at 22°, which was
attributed to the feature peak of amorphous SiO2.[2a,
21]
Besides, only the dried Cu-SiO2 and 0.05CD-Cu-SiO2
samples were composed of a strong phase of Cu2(OH)3NO3
as evidenced by obvious reflection at 2θ of 12.8° and 25.8°
(JCPDS 75-1779),[22] indicating the poor dispersion of copper
species in Cu-SiO2 catalyst. But luckily for more β-CD doped
Cu-SiO2 samples, no crystalline characteristic peaks were
observed, suggesting better dispersion of β-CD and copper
species. Interestingly, an characteristic diffraction peak of
Cu2O at 2θ of 36.4°(JCPDS 05-0667) was observed for the
dried 0.3CD-Cu-SiO2 (Figure 4B-e).
Figure 4C presents the XRD patterns of the xCD-Cu-
SiO2 catalysts before reduction. Except for the characteristic
peak of silica, more and more weak diffraction peaks of CuO
at 2θ of 35.5° and 38.7°(JCPDS 05-0661) were observed
when x≤ 0.2. A weakly diffraction peak of some residual
Cu2(OH)3NO3 at 2θ of 12.8° could still be found over the
calcined 0.05CD-Cu-SiO2 (Figure 4C-b). However, with the
continuous increment in β-CD doping, the peaks of CuO
were disappeared while the peaks of Cu2O and Cu were
arisen. It was interesting that the auto-reduction was
occurred on 0.3CD-Cu-SiO2 catalyst. Ai et al. reported that
the auto-reduction of copper oxides with carbon nanotubes
could directly realize, owing to the interaction of copper
species with electron deficient interior surface of CNTs.[3]
After reduction (Figure 4D), the strong diffraction peak at 2θ
of 43.3° along with two small ones at 50.4° and 74.1° were
assigned to Cu species (JCPDS 04-0836), while the weak
diffraction peak at 36.4° was characteristic of Cu2O.[18]
Therefore, the reduced xCD-Cu-SiO2 catalysts exhibited
crystalline Cu and well dispersed Cu+ species, especially for
0.1CD-Cu-SiO2 catalyst.
The NH3-TPD characterization was performed to
evaluate the strength and quantity of acid sites on the xCD-
Cu-SiO2 catalysts. It was observed that a sharp peak
centered at 123 °C (weak acid site), a shoulder peak
centered at 255 °C (moderate acid site) and a broad weak
peak appeared centered at 525 °C (strong acid site)
appeared over the xCD-Cu-SiO2 (x≤ 0.1) catalyst surfaces
(Figure 5B). However, the middle peak intensity gradually
decreased while the strong acid site improved much with
more β-CD introduction (x≥ 0.2). Meanwhile, the catalytic
activity was increased first and then decreased much. Thus it
is inferred that the addition of β-CD was responsible for the
surface
acidity
and
EG
selectivity.
H2-TPR and NH3-TPD analysis
Figure 5. (A) H2-TPR and (B) NH3-TPD profiles of the calcined xCD-Cu-SiO2 catalysts.
5
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701246
ChemCatChem
FULL PAPER
Figure 6. FT-IR spectra of (A) pure β-CD, (B) the dried samples, (C) the calcined precursors of catalysts with different β-CD
loading, and (D) the I962/I800 intensity ratio from picture (C).
FT-IR analysis
Cu2Si2O5(OH)2, thus indicating the absence of copper
hydroxide and copper phyllosilicate in calcined xCD-Cu-SiO2
catalysts [25]. Notwithstanding, copper phyllosilicate was often
The FT-IR spectrum of pure β-CD is given in Figure 6A. A
typical peak at 2923 cm-1 was attributed to the antisymmetric
C-H vibration of -CH2, while it was not shown in the dried
catalyst precursors with β-CD doping (Figure 6B), as well as
other characteristic peaks of β-CD. As shown in Figure 6B,
obvious bending absorption of Cu-O-H bond at 679 cm-1 and
a sharp absorption at 1382 cm-1 related to NO3- groups were
first saw for Cu-SiO2 and 0.05CD-Cu-SiO2 catalysts. It
indicated the presence of Cu2(OH)3NO3 as also proved by
XRD results. The adsorption band at 945 cm-1 was assigned
to the bond vibration of Cu-O-Si units, which was shifted to
962 cm-1 after calcination at 300 °C (Figure 6C). The Si-O
bonds of amorphous SiO2 support also showed different
adsorption bands at ~ 1095, 800, and 465 cm-1. The broad
absorption band in the range 3650-3250 cm-1, especially for
the previous two catalysts, was owing to the overlapping of
the OH stretching vibration of adsorbed water, ethanol and
silanols (Figure 6B).[24] This further indicated that some
TEOS still remained in the dried Cu-SiO2 precursor. Similarly,
the band at ~ 1664 cm-1 corresponds to the bending mode of
OH groups of adsorbed H2O.[21] There were no frequencies
of the δOH bands at 694 cm-1 for Cu(OH)2 and 670 cm-1 for
formed in the Cu-Si catalysts prepared by ion-exchanged
26]
and ammonia evaporation methods.[16b,
In addition, the
relative amount of Cu-O-Si units in calcined samples could
be calculated by the ratio of band intensities at 962 and 800
cm-1. Figure 6D clearly shows that the relative amount of Cu-
O-Si units largely increased with the β-CD doping. Therefore,
the dominating copper species were Cu-O-Si layers in the
calcined catalysts.
TGA analysis
More detailed information on copper species could be further
investigated by TGA analysis in the N2 atmosphere given in
Figure 7. It was clear that the decomposition temperature of
pure β-CD was ~ 300 °C in N2 or air conditions (Figure 7A
and Figure S4), in keeping with the reported results [27]. As
the EA analysis showed that C contents were less than 0.3
wt% in the calcined xCD-Cu-SiO2 (x≤ 0.2) samples, β-CD
doping in catalyst was nearly decomposed before 300 °C. As
shown in Figure 7B, all the samples continuously lost weight
6
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701246
ChemCatChem
FULL PAPER
from RT to 300 °C, which was probably attributed to the
evaporation of the absorbed water (< 120 °C) and the loss of
crystal water, as well as well dispersed β-CD (120-300 °C).
With a further elevating the temperature, the following weight
loss was mainly due to the transformation of bulk CuO to into
Cu2O or the collapse of silica skeleton [3, 5a]. The inset picture
of Figure 7B indicated that Cu-SiO2 catalyst exhibited the
least weight loss before 300 °C and continued to slightly loss
weight, while the 0.1CD-Cu-SiO2 sample was more stable in
high temperature.
Figure 7. TGA analysis of (A) pure β-CD and (B) the dried xCD-Cu-SiO2 catalysts in N2 atmosphere (inset picture was
enlargement from 250-350 °C): (a) Cu-SiO2; (b) 0.05CD-Cu-SiO2; (c) 0.1CD-Cu-SiO2; (d) 0.2CD-Cu-SiO2; (e) 0.3CD-Cu-SiO2.
Figure 8. XPS spectra of the xCD-Cu-SiO2 catalysts. (A) Cu 2p photoelectron spectra of the calcined samples, (B) Cu 2p
photoelectron spectra of the reduced samples, (C) Cu LMM XAES spectra of the reduced samples: (a) Cu-SiO2; (b) 0.05CD-
Cu-SiO2; (c) 0.1CD-Cu-SiO2; (d) 0.2CD-Cu-SiO2; (e) 0.3CD-Cu-SiO2.
935.4 eV along with the shakeup satellite peaks indicated
Surface chemical states of the catalysts
that in calcined xCD-Cu-SiO2 (x≤ 0.2) samples copper
existed in the oxidation state of Cu2+. Meanwhile, the surface
In DMO hydrogenation reaction, it is significant to investigate
of calcined 0.3CD-Cu-SiO2 catalyst also possessed some
Cu2+ species and the major copper species were Cu2O, as
evidenced by XRD analysis. Generally, the Cu 2p3/2 BE of
the surface chemical states of the catalysts due to a
synergetic effect between Cu0 and Cu+.[18] As can be seen
from part (A) of Figure 8, the Cu 2p3/2 peak centered at ~
7
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701246
ChemCatChem
FULL PAPER
CuO is detected at 933.5 eV[16b, 28] and copper phyllosilicate
is 934.9 eV,[16b, 26] which are relatively lower than 935.4 eV. It
was maybe a charge transfer from copper ions toward the
electron-rich hydrophobic cavity and exterior hydrophilic
surface (Scheme 2A).[27] Above all, β-CD molecules would
cross-link with each other through the hydroxyls under
33]
silica matrix that lead to the big positive Cu 2p3/2 BE shift,
namely, a strong interaction between the Cu2+ and SiO2
ultrasonication (Scheme 2B).[27,
Figure S5 showed that
[21]
.
only 0.2CD could not be well dissolved in 21.6 g water, while
adding with 10.6 Cu(NO3)2·3H2O could obtain a
Thus the XPS results explained that in the calcined
precursors exhibited highly dispersed Cu-O-Si units, which
were comparable to copper phyllosilicate.[16b, 29]
g
homogeneous and blue transparent solution, even for the
0.3CD. We first assumed that the acid condition would be
favorable for β-CD dissolution and then added 3 mL 1M
HNO3, but it had no effect on β-CD dissolution. The UV-vis
spectra of these examples indicated that the absorbance of
NO3- and d-d transfer have been decreased with an increase
in β-CD concentration (Figure S5). These results indicated
that there was interaction between Cu2+ and β-CD, giving
rise to well dispersed β-CD and copper species. Moreover,
the characteristic peaks of β-CD disappeared in both XRD
(Figure 4B) and FT-IR files (Figure 6B) after drying, as well
as decomposed earlier in TGA curves (< 300 °C, Figure 7B).
On the contrary, the dried pure Cu-SiO2 catalyst without
doing with any amount of β-CD displayed obvious copper
species of Cu2(OH)3NO3 instead of Cu-O-Si units, as
evidenced by EA analysis (Table 1), XRD (Figure 4B) and
FT-IR files (Figure 6B). Huang et al. reported that the main
copper species of CuO-SiO2 catalyst prepared by the
impregnation method was Cu2(OH)3NO3, which were easily
transformed into bulk CuO after calcination. As a result, it
showed a much poor activity and short stability in glycerol
reaction.[22] Meanwhile, the higher mass fraction of β-CD (x>
0.1) would obviously led to bigger spheres (Figure S3C-D),
thus the SBET decreased from 324.5 to 227.8 cm3 g-1 and
copper dispersion also decreased from 16.5 to 8.6% (Table
1).
The XPS spectra and X-ray induced Auger spectra
(XAES) of the reduced xCD-Cu-SiO2 catalysts were also
illustrated in Figure 8B-C. The BE of Cu 2p3/2 in reduced
samples were changed to 933.1 eV and showed no any
corresponding satellite peaks, which was responsible for
complete reduction of Cu2+ to Cu+ or / and Cu0.[19] The Cu
LMM spectra were further used to discriminate between Cu+
and Cu0 species,[30] as shown in Figure 8C and Table S1. All
the Cu LMM spectra displayed a broad and asymmetrical
peak, declaring clearly that Cu+ and Cu0 species were stable
coexistence on the surface of the reduced xCD-Cu-SiO2
catalysts. Furthermore, the original Cu LMM peaks were
deconvoluted to two symmetrical peaks centered at ~ 913.0
and 916.5 eV, corresponding to Cu+ and Cu0 species,
respectively.[31] The results listed in Table S1 showed that
the ratio of Cu+ / (Cu+ + Cu0) did not change much with the
β-CD doping mass fraction lower than 0.1. However, the
ratio was decreased from 62.1% to 57.5% when the β-CD
loading was 0.3, which was probably because that the main
copper species were already Cu2O and its weekly interaction
with support. Hence, the high ratio of Cu+ for the reduced
catalysts implies the strong interaction between copper and
support.
After calcination, the doing β-CD molecules were
removed and left some stacking fault, as supported by EA
analysis (Table 1) and TEM images (Figure 3B-C). Yao’s
group recently has reported similar defect mechanism for
enhancing oxygen reduction reaction activity by using metal-
organic frameworks as hard templates or precursors.[35] As
we know, some loosely and well dispersed CuO were easily
aggregated to form large CuO particles as identified by XRD
(Figure 4C) and H2-TPR (Figure 5A) results. Fortunately,
these results further showed that CuO particle sizes became
smaller and fewer when increasing β-CD introduction.
Furthermore, it was surprised that a auto-reduction reaction
was happened to 0.3CD-Cu-SiO2 catalysts as illustrated by
XRD (Figure 4C) and XPS analysis (Figure 8A). Regrettably,
it displayed a poor DMO hydrogenation activity. From the FT-
IR result (Figure 7C-D), we concluded that the main copper
species were Cu-O-Si units rather than copper phyllosilicate
formed during sol-gel process. Owing to the strong
interaction between silica and copper species, the copper
oxidation state of most metallic copper was +1 in reduced
Structure-performance relationship of the
catalysts
Formation mechanism and structure evolution during
the synthesis process
It is commonly known that the synthesis processes of
heterogeneous catalysts are comprised of multi steps, and
the compositions would go through complicated changes.[32]
In this work, the xCD-Cu-SiO2 catalysts prepared by sol-gel
method, containing three steps: sol-gel generation,
calcination and reduction. The whole structure evolution
processes with a possible mechanism for formation of β-CD
doped Cu-SiO2 catalysts were systemically discussed below.
For the purpose of investigating the role of β-CD during
nucleation and growth of solid gel, different mass fraction (x=
0-0.3) of β-CD were doped to Cu-SiO2 catalyst. Many articles
proves that there is interaction between various metal ions
27]
such as Cu2+,[10b,
Ca2+,[33] Fe3+,[34] and β-CD or its
derivatives. It is attributed to the special structure of β-CD
that possesses a segment of a hollow cone with an interior
8
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701246
ChemCatChem
FULL PAPER
xCD-Cu-SiO2 catalysts, as well as some Cu0 species (Table
S1 and Figure 8C).
ensured the uniform distribution of the copper precursor in
silica channels. After calcination, the generated defects
promoted the dispersion of copper species and exposed
more active sites to enhance its catalytic performance.
Therefore, doping proper amount of β-CD is an efficient
strategy to synthesize Cu-SiO2 catalyst with small particle
size and high dispersion.
Given our results and the literature,[27, 33, 36] a possible
schematic mechanism for formation of β-CD doped Cu-SiO2
catalysts is proposed and shown in Scheme 2. The Cu2+
would be absorbed on β-CD to form a homogenous complex
spheres. Then the presence of ordered Cu-β-CD structure
Scheme 2. A possible schematic mechanism for formation of β-CD doped Cu-SiO2 catalysts.
Insight into correlation between copper species and
catalytic performance
hence more difficult to be reduced before calcination (Figure
S6). The dried 0.1CD-Cu-SiO2 catalyst produced a very
small SBET of 86.2 m2 g-1 and a low Vp of 0.03 cm3 g-1(Figure
S7). Thus EG yield at 210 °C over dried 0.1CD-Cu-SiO2
catalyst was much poor, namely 0.05 g g-catal-1 h-1. As shown
in Figure S4, β-CD was nearly decomposed at 550 °C . Thus
the dried 0.3CD-Cu-SiO2 sample was further calcined at
550 °C for 5 h. As listed in Table S3, after removing the
carbon- and nitrogen-containing residue species to expose
more active sites, the 0.3CD-Cu-SiO2-550 catalyst delivered
a much better performance. However, it was still lower than
the 0.1CD-Cu-SiO2 sample. The reason that why the 0.3CD-
Cu-SiO2 sample achieved a lower catalytic performance was
complicated. First, from the XRD and H2-TPR results, we
know that the auto-reduction behaviours was happened to
the 0.3CD-Cu-SiO2 sample alone. Then, the NH3-TPD result
indicated that its moderate acid site decreased much while
the strong acid site improved. Ai et al. reported that suitable
surface acidity could promote the hydrogenation of DMO.[38]
Furthermore, as mentioned above, it possessed a large Cu
NPs, a low dispersion and a low copper surface area, as well
as a low ratio of Cu+ / (Cu+ + Cu0). Taken together, the
As mentioned above, three kinds of copper species in the
calcined precursors, namely Cu-O-Si units, high dispersed
CuO and bulk CuO clusters. The amount of different kind of
copper species and Cu dispersion varied much with
increasing β-CD loading. To clearly show the effect of β-CD
loading on the EG yield and the metallic specific area, Figure
9 was illustrated their relations. The 0.1CD-Cu-SiO2 catalyst
exhibited the highest copper surface area and thus showed a
perfect DMO hydrogenation activity with a TOF value of 11.4
h-1, which was much higher than pure Cu-SiO2 catalyst (4.8
h-1, Table S2). But excessive β-CD loading would lead to
bigger spheres and agglomeration. Consequently, the EG
-1
yield was decreased from 0.36 to 0.12 g g-catal h-1. This
finding suggests that the copper surface was mainly
responsible for the catalytic activity of Cu-SiO2 in the DMO
hydrogenation, which is consistent with Yin’s conclusion in
[21]
Cu-HMS catalysts with variation in copper loading
.
Without doubt, other factors, such as Cu particle sizes, SBET
,
ratio of Cu+ / (Cu+ + Cu0), pore structural properties, metal-
support interaction and stacking fault, all influence the
catalytic activity over Cu-SiO2 catalysts.[21, 36-37] For example,
the copper species were not well dispersed and
0.3CD-Cu-SiO2 sample did not deliver
a comparable
performance.Therefore, a proper mass fraction of β-CD (x=
0.1) with highly dispersed Cu-O-Si units and copper surface
9
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701246
ChemCatChem
FULL PAPER
area were needed to achieve optimum performance in DMO
hydrogenation to EG.
O-Si units was more stable than the Cu+ species reduced
[5a]
from copper phyllosilicate in the Cu-SiO2 catalyst
.
Therefore, we can speculate that the stable catalytic
behavior of 0.1CD-Cu-SiO2 catalyst was attributed to the
highly dispersed copper species immobilized by the support.
In addition, it was surprised to find that the ratio of Cu+ / (Cu+
+ Cu0) clearly increased in both Cu-SiO2-Used and 0.1CD-
Cu-SiO2-Used catalysts under H2 atmosphere (Figure S12),
which was opposite to the results of La-Cu-SiO2 catalysts.[30]
Similar phenomenon was happened to the spent Cu-SiO2
catalyst used for hydrogenation of methyl acetate and the
authors attributed this finding to the surface oxidation of
copper species at Cu-SiO2 interface by methyl acetate.[41] In
fact, the stable surface chemical environment plays
important roles in maintaining catalytic stability.[40] From this
point, there is room for improvement in stability of 0.1CD-Cu-
SiO2 sample.
Figure 9. EG yield and copper metal surface area as a
function of β-CD loading. Reaction conditions: P(H2)= 2.0
MPa, T= 210 °C, WHSVDMO= 0.72 g g-catal-1 h-1, H2/DMO= 50.
Conclusions
Explanation on the long-term stability
In this work, the effect of introduced β-CD on the structure
and properties of Cu-SiO2 catalysts was systematically
investigated. β-CD doping first inhibited and then accelerated
the coagulation of copper species and even reduced them as
doping amount increased, and the optimized doping amount
for the outstanding stability and activity was observed for the
0.1CD-Cu-SiO2 catalyst. This was attributed to highly
dispersed copper species with smaller particles sizes, more
Cu-O-Si units, larger SCu and SBET. However, the XPS
analysis of the used catalysts indicated that the ratio of Cu+
and Cu0 was still slowly unbalanced. Therefore, it should
take more hopeful techniques to immobilize copper species
in our future work.
Figure 2 indicated that 0.1CD-Cu-SiO2 catalyst showed
much higher activity and remarkable stability than pure Cu-
SiO2 catalyst in time on-stream. Besides the same solvent
deactivation effect on the Cu-based catalysts,[39] we need to
discuss many other reasons. Firstly, carbon deposition was
excluded since both of the used catalysts exhibited carbon
content of 3.92% and 4.77% in Cu-SiO2-Used and 0.1CD-
Cu-SiO2-Used catalysts, respectively. Zheng et al. also did
not consider carbon deposition as the main reason for Cu-
SiO2 catalyst deactivation.[30] Then SBET decreasing and Cu
sintering were believed to be the main cause of Cu-SiO2
catalyst deactivation. As shown in Figure S8, the SBET of Cu-
SiO2-used catalyst was reduced from 319.0 to 245.1 cm3 g-1
(23.2%), while the 0.1CD-Cu-SiO2-used catalyst had only a
8.4% loss of the initial value of specific surface area. Popa et
al. prepared a new kind of stable Cu-SiO2 catalyst by using
(NH4)2CO3 as precipitant, showing a smaller loss of SBET as
well.[16a] Both the XRD patterns and TEM images indicated
that Cu nanoparticles of the used Cu-SiO2 catalyst
aggregated more seriously (Figure S9-10). Third, the
deactivation was related to hydrogenation product. As Cu-
SiO2 catalyst showed a MG selectivity of ~ 35.0%, the
selectivity of EG was decreased much. Generally, once MG
was generated more than 10%, we should increase the
reaction temperature to prevent the deactivation owing to
MG would accelerate copper species sintering.[31] Yin et al.
presumed that MG adsorbed via the carbonyl oxygen atoms
on the catalyst surfaces under reaction conditions.[40]
Furthermore, the 0.1CD-Cu-SiO2 catalyst possessed more
Cu-O-Si units and defects, which could stabilize the active
copper species during DMO hydrogenation. As shown in
Figure S11, Ding et al. reported that Cu2O reduced from Cu-
Experimental Section
Catalyst preparations
The β-CD doped catalysts were synthesized by the sol-gel method, as
reported in our previous paper.[8] A typical preparation procedure of these
catalysts is illustrated in Scheme 3 and described as follows: 10.6 g
Cu(NO3)2·3H2O and a certain amount of β-CD (Mass fraction x= 0-0.3)
were dissolved with 21.6 g distilled water under ultrasonication (25 kHz,
40 °C) for 15 min in a beaker. A homogeneous and blue transparent
solution was obtained. Then other regents including 31.8
g
tetraethylorthosilicate (TEOS) and 41.4 g ethanol (EtOH) were added into
the prepared solutions under vigorous stirring. The mixed solution was
then stirred at 65 °C for 1~2 h until generating solid gel. After being aged
at room temperature (RT) for 24 h, the obtained xerogel was dried at
65 °C for 7 h, 70 °C for 7 h, 100 °C for 40 h and 120 °C for 4 h. Finally,
the β-CD doped Cu-SiO2 composite catalyst precursors were calcined at
300 °C for 5 h in air and noted as xCD-Cu-SiO2 catalysts, where x
represents mass fraction of β-CD.
This article is protected by1co0pyright. All rights reserved.

10.1002/cctc.201701246
ChemCatChem
FULL PAPER
PerkinElmer Lambda 365. The surface species were detected by X-ray
photoelectron spectroscopy (XPS) using a ESCALAB 250Xi equipped
with an Al Kα X-ray radiation source (hν = 1486.6 eV). The binding
energy (BE) values were referenced to the Si 2p peak at 103.7 eV.
Acknowledgement
This work was supported by the “Strategic Priority Research
Program” of the Chinese Academy of Sciences
(XDA07070200, XDA09030102), the Science Foundation of
Fujian Province (2006l2005) and Fujian industrial guide
project (2016H0048). The authors also thank all researchers
of test center, FJIRSM, CAS.
Scheme 3. Schematic illustration of the xCD-Cu-SiO2 catalyst
preparation.
Catalytic performance
The five catalysts were conducted on a stainless-steel fixed-bed reactor
with a continuous flow unit to measure their catalytic activity and
durability. Typically, 3.5 g of the catalyst precursor (10-20 meshes) was
packed into the center of the reactor with the thermocouple inserted into
the catalyst bed to better control of reaction temperature. The sample
was first activated under pure H2 (99.99% purity) flow (100 ml min-1) at
350 °C for 5h and then cooled to the desired reaction temperature (170-
280 °C). A 20 wt% DMO (99% purity) in methanol was pumped into the
reactor using a high pressure pump (Lab Alliance Series I pump). It
should be pointed out that the system pressure was 2.0 MPa, H2/DMO
molar ratio is 50, and the weight hour space velocity (WHSVDMO) was
Keywords: sol-gel method
·
hydrogenation
·
β-
cyclodextrin · copper species · ethylene glycol
[1] a) S. Chakraborty, H. G. Dai, P. Bhattacharya, N. T. Fairweather, M.
S. Gibson, J. A. Krause, H. R. Guan, J. Am. Chem. Soc. 2014, 136,
7869-7872; b) C. C. Zhang, D. H. Wang, M. Y. Zhu, F. Yu, B. Dai,
Catalysts 2017, 7, 75-87.
-1
0.72 g g-catal h-1. The products were condensed and analyzed on a gas
chromatograph (GC-900C, Shanghai TianPu analysis instrument Co.,
LTD) equipped with a flame ionization detector (FID) and a DB-WAX 60m
capillary column. The conversion of DMO was calculated as the input
mole amount of DMO subtracts the output mole amount of DMO and
then divided by the input mole amount of DMO. The selectivity of a
product was calculated as the mole amount of one product divided by the
total mole amount of all products formed.
[2] a) Y. J. Zhao, Y. Q. Zhang, Y. Wang, J. Zhang, Y. Xu, S. P. Wang, X.
B. Ma, Appl. Catal. A : Gen. 2017, 539, 59-69; b) C. C. Zhang, D. H.
Wang, M. Y. Zhu, F. Yu, B. Dai, ChemistrySelect 2016, 1, 2857-2863;
c) H. L. Yue, Y. J. Zhao, X. B. Ma, J. L. Gong, Chem. Soc. Rev.
2012, 41, 4218-4244.
[3] P. P. Ai, M. H. Tan, Y. Ishikuro, Y. Hosoi, G. H. Yang, Y. Yoneyama, N.
Tsubaki, ChemCatChem 2017, 9, 1067-1075.
[4] a) Z. He, H. Q. Lin, P. He, Y. Z. Yuan, J. Catal. 2011, 277, 54-63; b)
L. Y. Qiao, Q. H. Li, Z. F. Zhou, R. Si, Y. G. Yao, ChemCatChem
2016, 8, 1909-1914.
Catalyst characterization
The copper loading was measured by inductively coupled plasma optical
emission spectrometer (ICP-OES) on a Jobin Yvon Ultima2. Elemental
analysis (EA) was obtained by Elementar Vario EL-Cube elemental
analyzer. N2 adsorption-desorption isotherms were conducted on a
Micromeritics ASAP 2020 at 77 K. Before measurement, the catalyst was
degassed in vacuum at 250 °C for 5 h. Then the specific surface area
(SBET) was estimated by the Brunauer-Emmett-Teller (BET) method. In
the meantime, the specific surface area of metallic copper (SCu) and
copper dispersion were performed on the TP-5080 instrument by the
adsorption and decomposition of N2O with the pulse titration method.
Thermogravimetric analysis (TGA) data were collected on a NETSCHZ
STA-449C thermal analyzer in the temperature range 30-900 °C with a
heating rate of 10 °C min-1 under N2 or air atmosphere. X-ray diffraction
patterns (XRD) were obtained from a Rigaku MiniFlex II diffractometer
using Cu-Kα radiation source at a scan rate of 2° per min with a scanning
angle (2θ) ranging from 10° to 85°. Transmission electron microscopy
[5] a) J. Ding, T. Popa, J. K. Tang, K. A. M. Gasem, M. H. Fan, Q.
Zhong, Appl. Catal. B: Environ. 2017, 209, 530-542; b) H. R. Yue, Y.
J. Zhao, S. Zhao, B. Wang, X. B. Ma, J. L. Gong, Nat. Commun.
2013, 4, 2339-2346; c) C. Wen, F. Q. Li, Y. Y. Cui, W. L. Dai, K. N.
Fan, Catal. Today 2014, 233, 117-126.
[6] M. A. Karakassides, K. G. Fournaris, A. Travlos, D. Petridis, Adv.
Mater. 1998, 10, 483-486.
[7] Y. J. Zhang, N. Zheng, K. J. Wang, S. J. Zhang, J. Wu, J.
Nanomater. 2013, 2013, 1-6.
[8] L. Lin, P. B. Pan, Z. F. Zhou, Z. J. Li, J. X. Yang, M. L. Sun, Y. G.
Yao, Chin. J. Catal. 2011, 32, 957-969.
[9] M. A. Karakassides , A. Bourlinos, D. Petridis, L. Coche-Guerènte, P.
Labbè, J. Mater. Chem. 2000, 10, 403-408.
[10] a) Y. Matsui, D. Suemitsu, Bull. Chem. Soc. Jpn. 1985, 58, 1658-
1662; b) B. Kaboudin, Y. Abedi, T. Yokomatsu, Org. Biomol. Chem.
2012, 10, 4543-4548.
(TEM) images were obtained on
a Tecnai F20 apparatus at an
[11] D. Gupta, A. Mishra, S. Kundu, ChemistrySelect 2017, 2, 2997-3008.
[12] J. Alvarez, J. Liu, E. Román, A. E. Kaifer, Chem. Commun. 2000, 13,
1151-1152.
acceleration voltage of 200 kV. Temperature-programmed reduction (H2-
TPR) was carried out on a Micromeritics Autochem II 2920 instrument
connected with a Hidden Qic-20 mass spectrometry (MS). The H2
consumption of calcined catalysts was calculated from H2-TPR results
using CuO as reference material. It was first integrated by the software of
Origin 9.0 (100~400 °C ) and then divided by the sample mass (g). The
temperature-programmed desorption profiles (NH3-TPD) were obtained
on a TP-5080 instrument. Fourier-transform infrared (FT-IR) spectra were
recorded on a Bruker Vertex 70 FT-IR with a spectral resolution of 2 cm-1
in the range 4000-400 cm-1. The UV-vis spectra were obtained on
[13] L. Strimbu, J. Liu, A. E. Kaifer, Langmuir 2003, 19, 483-485.
[14] a) J. Y. Zheng, J. B. Pang, K. Y. Qiu, Y. Wei, J. Sol-Sel Sci. Techn.
2002, 24, 81-88; b) J. Y. Zheng, J. B. Pang, K. Y. Qiu, Y. Wei, J.
Mater. Chem. 2001, 11, 3367-3372.
[15] a) L. C. Liu, F. Gao, P. Concepción, A. Corma, J. Catal. 2017, 350,
218-225; b) L. C. Liu, P. Concepción, A. Corma, J. Catal. 2016, 340,
1-9.
This article is protected by1co1pyright. All rights reserved.

10.1002/cctc.201701246
ChemCatChem
FULL PAPER
[16] a) T. Popa, Y. L. Zhang, E. L. Jin, M. H. Fan, Appl. Catal. A : Gen.
2015, 505, 52-61; b) L. F. Chen, P. J. Guo, M. H. Qiao, S. R. Yan, H.
X. Li, W. Shen, H. L. Xu, K. N. Fan, J. Catal. 2008, 257, 172-180.
[17] Y. Huang, H. Ariga, X. L. Zheng, X. P. Duan, S. Takakusagi, K.
Asakura, Y. Z. Yuan, J. Catal. 2013, 307, 74-83.
Received:
Published online:
[18] J. Gong, H. Yue, Y. Zhao, S. Zhao, L. Zhao, J. Lv, S. Wang, X. Ma, J.
Am. Chem. Soc. 2012, 134, 13922-13925.
[19] Y. J. Zhao, S. M. Li, Y. Wang, B. Shan, J. Zhang, S. P. Wang, X. B.
Ma, Chem. Eng. J. 2017, 313, 759-768.
[20] A. Y. Yin, J. W. Qu, X. Y. Guo, W. L. Dai, K. N. Fan, Appl. Catal. A :
Gen. 2011, 400, 39-47.
[21] A. Y. Yin, X. Y. Guo, W. L. Dai, K. N. Fan, J. Phys. Chem. C 2009,
113, 11003-11013.
[22] Z. W. Huang, F. Cui, H. X. Kang, J. Chen, X. Z. Zhang, C. G. Xia,
Chem. Mater. 2008, 20, 5090-5099.
[23] K. V. R. Chary, K. K. Seela, G. V. Sagar, B. Sreedhar, J. Phys.
Chem. B 2004, 108, 658-663.
[24] Z. l. Wang, Q. S. Liu, J. F. Yu, T. H. Wu, G. J. Wang, Appl. Catal. A :
Gen. 2003, 239, 87-94.
[25] T. Toupance, M. Kermarec, J. F. Lambert, C. Louis, J. Phys. Chem. B
2002, 106, 2277-2286.
[26] C. J. Van, P. A. Elberse, A. Mulder, J. W. Geus, Appl. Catal. 1990,
59, 275–289.
[27] X. F. Xiao, R. F. Liu, C. F. Qiu, D. C. Zhu, F. Liu, Mater. Sci. Eng. C
2009, 29, 785-790.
[28] A. Gervasini, M. Manzoli, G. Martra, A. Ponti, N. Ravasio, L. Sordelli,
F. Zaccheria, J. Phys. Chem. B 2006, 110, 7851-7861.
[29] R. P. Ye, L. Lin, J. X. Yang, M. L. Sun, F. Li, B. Li, Y. G. Yao, J. Catal.
2017, 350, 122-132.
[30] X. L. Zheng, H. Q. Lin, J. W. Zheng, X. P. Duan, Y. Z. Yuan, ACS
Catal. 2013, 3, 2738-2749.
[31] C. Wen, A. Y. Yin, Y. Y. Cui, X. L. Yang, W. L. Dai, K. N. Fan, Appl.
Catal. A : Gen. 2013, 458, 82-89.
[32] a) Z. Q. Wang, Z. N. Xu, M. J. Zhang, Q. S. Chen, Y. M. Chen, G. C.
Guo, RSC Adv. 2016, 6, 25185-25190; b) R. Schlogl, Angew. Chem.
Int. Ed. 2015, 54, 3465-3520.
[33] J. Z. Xu, S. Xu, J. Geng, G. X. Li, J. J. Zhu, Ultrason. Sonochem.
2006, 13, 451-454.
[34] B. Kaboudin, R. Mostafalu, T. Yokomatsu, Green Chem. 2013, 15,
2266.
[35] a) X. J. Zhao, X. Q. Zou, X. C. Yan, C. L. Brown, Z. G. Chen, G. S.
Zhu, X. D. Yao, Inorg. Chem. Front. 2016, 3, 417-421; b) H. Y. Zhao,
C. H. Sun, Z. Jin, D. W. Wang, X. C. Yan, Z. G. Chen, G. S. Zhu, X.
D. Yao, J. Mater. Chem. A 2015, 3, 11736-11739; c) X. J. Zhao, H. Y.
Zhao, T. T. Zhang, X. C. Yan, Y. Yuan, H. M. Zhang, H. J. Zhao, D.
M. Zhang, G. S. Zhu, X. D. Yao, J. Mater. Chem. A 2014, 2, 11666-
11671.
[36] S. Zhao, H. R. Yue, Y. J. Zhao, B. Wang, Y. C. Geng, J. Lv, S. P.
Wang, J. L. Gong, X. B. Ma, J. Catal. 2013, 297, 142-150.
[37] a) Y. J. Zhao, S. Zhao, Y. C. Geng, Y. L. Shen, H. R. Yue, J. Lv, S. P.
Wang, X. B. Ma, Catal. Today 2016, 276, 28-35; b) Y. Y. Cui, B.
Wang, C. Wen, X. Chen, W. L. Dai, ChemCatChem 2016, 8, 527-
531.
[38] P. P. Ai, M. H. Tan, N. Yamane, G. G. Liu, R. G. Fan, G. H. Yang, Y.
Yoneyama, R. Q. Yang, N. Tsubaki, Chem. Eur. J. 2017, 23, 8252-
8261.
[39] C. Wen, Y. Y. Cui, W. L. Dai, S. H. Xie, K. N. Fan, Chem. Commun.
2013, 49, 5195-5197.
[40] A. Y. Yin, C. Wen, X. Y. Guo, W. L. Dai, K. N. Fan, J. Catal. 2011,
280, 77-88.
[41] X. M. Huang, M. Ma, S. Miao, Y. P. Zheng, M. S. Chen, W. J. Shen,
Appl. Catal. A : Gen. 2017, 531, 79-88.
This article is protected by1co2pyright. All rights reserved.

10.1002/cctc.201701246
ChemCatChem
FULL PAPER
Entry for the Table of Contents (Please choose one layout)
Layout 2:
FULL PAPER
Run-Ping Ye, Ling Lin, Chang-Qing Liu,
Chong-Chong Chen, and Yuan-Gen Yao*
Page No. – Page No.
One-pot Synthesis of Cyclodextrin
Doped Cu-SiO2 Catalysts for Efficient
Hydrogenation of Dimethyl Oxalate to
Ethylene Glycol
Here we report a series of cyclodextrin doped Cu-SiO2 catalysts prepared via a
one-pot synthesis method, which largely enhance the selectivity of ethylene glycol
and long-term durability.
This article is protected by1co3pyright. All rights reserved.