Effect of feedstock solvent on the stability of Cu/SiO2 catalyst for vapor-phase hydrogenation of dimethyl oxalate to ethylene glycol

Article abstract of DOI:10.1039/c1cc15783c

The lifetime (200 h) of a Cu/SiO₂ catalyst under DMO-ethanol feedstock is 20 times longer than that (10 h) of the catalyst under DMO-methanol feedstock without any modification of the catalyst. The stabilization effect of ethanol on the active centers can effectively slow down the agglomeration of copper particles during the hydrogenation process. The Royal Society of Chemistry 2012.

Full text of DOI:10.1039/c1cc15783c

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2012, 48, 1177–1179
www.rsc.org/chemcomm
COMMUNICATION
Effect of feedstock solvent on the stability of Cu/SiO catalyst for
2
vapor-phase hydrogenation of dimethyl oxalate to ethylene glycolw
Jingdong Lin,* Xiaoqin Zhao, Yunhe Cui, Hongbin Zhang and Daiwei Liao
Received 19th September 2011, Accepted 17th November 2011
DOI: 10.1039/c1cc15783c
The lifetime (200 h) of a Cu/SiO catalyst under DMO–ethanol
2
stages of the production of EG. Nonetheless, the toxic chromium
contained in catalysts is dangerous to humans and the
environment and severely limits its practical applications.
Therefore, developing a high stability and environmentally-
friendly process is still the major area of interest in the selective
feedstock is 20 times longer than that (10 h) of the catalyst under
DMO–methanol feedstock without any modification of the
catalyst. The stabilization effect of ethanol on the active centers
can effectively slow down the agglomeration of copper particles
during the hydrogenation process.
9
,13
hydrogenation of DMO to EG.
In the past, DMO was usually dissolved in methanol as a
feedstock. Because dehydrogenation and oxidation of alcohols
easily occur on Cu-based catalyst, the Cu-based catalysts and
their relative catalytic reaction could be greatly affected by the
selected alcohols solvent. The effect of solvents on the stability
of the catalysts for the vapor-phase hydrogenation of DMO to
EG has not been reported yet. In present work, the DMO
dissolved in ethanol, labeled as DMO–ethanol, instead of the
DMO dissolved in methanol, labeled as DMO–methanol, was
used as a feedstock. The catalytic performance of the reduced
Ethylene glycol (EG) is a crucial chemical product, which is
widely used in various applications, such as polyester manu-
1
facture and antifreeze etc. At present, EG is mainly produced
2
by oxidation of petroleum-derived ethylene. However, with
the decrease of crude oil resources, synthesis of EG from
3
,4
syngas attracts more and more interest. The indirect synthesis
of EG from the syngas process includes two steps, namely the
catalytic coupling of CO with nitrite esters to dialkyl oxalates,
such as dimethyl oxalate (DMO), and the hydrogenation of
5
2
Cu/SiO catalysts prepared by ammonia-evaporation method
oxalates to EG. Since the 1970s, ruthenium-based homo-
was evaluated using vapor-phase hydrogenation of DMO. The
results show that the lifetime of the catalyst reacted under
DMO–ethanol feedstock, without any other modification, is
genous catalysts for hydrogenation of DMO have made some
6
representative results. However, it is difficult to separate the
products from catalysts in a homogenous reactor. Therefore,
more and more efforts are focused on vapor-phase hydrogenation
of oxalates. Extensive research on vapor-phase hydrogenation
2
0 times that of the catalyst reacted under DMO–methanol
feedstock.
4
,5,7–10
A comparison of catalytic activity as a function of the
reaction time for DMO–methanol and DMO–ethanol feed-
of oxalates to EG has been made by some groups.
It is
well-established that Cu/SiO2 catalysts have been of great
interest due to their good catalytic activities and selectivities
2
stock over 25 wt% Cu/SiO catalyst is shown in Fig. 1. In the
1
1
case of DMO–ethanol feedstock, the DMO conversion comes
in hydrogenation of DMO to EG. However, there are still
numerous technical problems to be solved. One of the most
fatal problems is the poor stability of Cu/SiO
vapor-phase hydrogenation of oxalates to EG.
2
catalysts for
Modification of the catalyst by doping is believed to be a
kind of effective method to improve the catalyst stability.
9
Doping with various elements, such as boric oxide, Cr,
5b,12
have shown good catalytic performance. The CuCr-based
catalysts with higher catalytic activity and longer lifetime have
been used as the preferred industrial catalysts in the early
Department of Chemistry, College of Chemistry and Chemical
Engineering, State Key Laboratory of Physical Chemistry for Solid
Surfaces, National Engineering Laboratory for Green Chemical
Productions of Alcohols–Ethers–Esters, Institute of Physical
Chemistry, Xiamen University, Xiamen 361005, China.
E-mail: jdlin@xmu.edu.cn; Fax: +86-592-2183043;
Tel: +86-592-2183045
w Electronic supplementary information (ESI) available: Experimental
details and characterization data. See DOI: 10.1039/c1cc15783c
Fig. 1 The performance of catalytic hydrogenation of DMO to EG
over 25 wt% Cu/SiO
2
catalyst. Reaction conditions: liquid hour space
À1
velocity (LHSV) = 0.6 h , P = 3.0 MPa, H
2
/DMO = 80 (mol/mol),
1
0 wt% DMO–ethanol or 10 wt% DMO–methanol, catalyst (0.5 g),
T = 473 K.
This journal is c The Royal Society of Chemistry 2012
Chem. Commun., 2012, 48, 1177–1179 1177

View Article Online
up to 100% and the selectivity of EG reaches 98% after 15 h
on the stream. The excellent catalytic performance is still
stable over 200 h. In contrast, the conversion of DMO and
the selectivity of EG decrease after reaction for 10 h under
DMO–methanol feedstock. Meanwhile, the selectivity of
methyl glycolate (MG) increases with decreasing of the selectivity
of EG. Obvious deactivation of catalyst using DMO–methanol
as feedstock is observed after 10 h.
XRD patterns of various Cu/SiO
Fig. 2. The feature at 2y around 221 is due to amorphous silica.
The peaks at 2y = 43.21, 50.41 and 74.11 in the Cu/SiO
2
catalysts are shown in
2
catalysts are assigned to copper. The sharp diffraction peaks
coming from copper indicate that large-size metallic crystallites
are present in the deactivated Cu/SiO2 catalysts. The mean
copper crystallite size of the fresh Cu/SiO catalyst is 4.7 nm,
2
Fig. 3 TEM images of 25 wt% Cu/SiO
2
catalysts. (A) Freshly
H
2
-reduced, (B) after reaction under DMO–methanol–H
h, (C) after reaction under DMO–methanol–H stream for 24 h, (D)
after reaction under DMO–ethanol–H stream for 5 h, (E) after
reaction under DMO–ethanol–H stream for 24 h, (F) after reaction
under DMO–ethanol–H stream for 200 h.
2
stream for
as calculated by the Scherrer equation. However, the mean sizes
5
2
of copper crystallite in the Cu/SiO catalyst after reaction under
2
2
DMO–methanol–H
2
stream for 5 h and 24 h are 6.7 nm and
2
17.7 nm, respectively. In contrast, in the case of the still-active
2
2
Cu/SiO on the stream after 24 h and 200 h under DMO–ethanol
feedstock, only small diffraction peaks of copper are observed. It
means that the particle size of copper may be hardly enlarged
after reaction under DMO–ethanol feedstock for 200 h.
As we know, two mechanisms, Ostwald ripening (OR) and
particle migration coalescence (PMC), for catalyst sintering
have been proposed, which are mainly based on the migration
of metal atoms or metal particles under a high reaction
2
TEM images of various Cu/SiO catalysts are shown in
14
Fig. 3. The light gray spherical particles are identified as silica
and the dark ones are assigned to copper. In the case of the
temperature. However, another key cause for catalyst sintering
is valence variation of metal species as a result of redox cycles
between surface metal species and reactants under the real
Cu/SiO catalyst reaction under DMO–methanol stream for
2
1
5
2
4 h, the copper particles are poorly dispersed. However, there
reaction conditions. In the present case, the possibility of OR
and PMC mechanisms may be disregarded due to the low
temperature (473 K) reaction. The copper particle size is not
significantly changed under hydrogen flow for 24 h at the same
temperature based on the XRD shown in Fig. 4.
is only slight aggregation of particles observed on the Cu/SiO
2
catalyst reaction under DMO–ethanol stream for 24 h or 200 h;
the active copper particles are still homogeneously dispersed on
the support.
According to the activities of vapor-phase hydrogenation of
DMO to EG, XRD and TEM characterizations, it is believed
that the catalyst with much a smaller particle size and much
higher dispersion of copper species is the main reason for the
The results presented in Fig. 4 also show that the copper
particle sizes are enlarged under H
streams. In contrast, the copper particle sizes are almost
unchanged under pure H . The above mentioned results
indicate that the sintering of the Cu/SiO catalyst is acceler-
2 2
–methanol and H –ethanol
2
higher activity and selectivity of EG over the Cu/SiO catalyst
2
2
(
also see Table S1, Fig. S1 and S2, ESIw). It also indicates that
ated by ethanol or methanol. It is well-known that copper is a
favorable catalyst for the oxidation of alcohols. Methanol and
ethanol plays an important role in slowing down the agglo-
meration of copper particles during the hydrogenation process.
ethanol oxidation over Cu/SiO
catalysts has been extensively
studied. In order to get further information about methanol
oxidation over a Cu/SiO catalyst, in situ FTIR characteriza-
tion was performed under He–methanol or He–ethanol stream
2
1
6
2
Fig. 2 XRD patterns of 25 wt% Cu/SiO
2
catalysts. (A) Freshly
-reduced, (B) after reaction under DMO–methanol–H stream for
h, (C) after reaction under DMO–methanol–H stream for 24 h,
stream for 5 h, (E) after
stream for 24 h, (F) after reaction
H
5
2
2
2
(
D) after reaction under DMO–ethanol–H
2
Fig. 4 XRD patterns of 25 wt% Cu/SiO
-reduced, (B) under H –methanol stream at 473 K for 24 h,
(C) under H –ethanol stream at 473 K for 24 h.
2
catalysts. (A) Freshly
reaction under DMO–ethanol–H
2
H
2
2
under DMO–ethanol–H
2
stream for 200 h.
2
1
178 Chem. Commun., 2012, 48, 1177–1179
This journal is c The Royal Society of Chemistry 2012

View Article Online
2
In summary, we have found that a Cu/SiO catalyst for
hydrogenation of DMO to EG using DMO–ethanol feedstock
was more stable than the DMO–methanol feedstock without
any modification of catalysts. The unstable active centers
caused by methanol, the CO splitting from methanol and the
oxidation of alcohol over Cu/SiO2 catalysts are the main
reasons for sintering of copper, which resulted in deactivation
of the catalysts. The stabilization effect of ethanol on the
active centers can effectively slow the increase in the size of the
copper particles.
This work was supported by the National Key Basic
Research Program of China (2011CBA00508).
Fig. 5 FTIR spectra of He–methanol and He–ethanol flow over 25 wt%
Cu/SiO catalyst at 473 K. (A) He–ethanol flow, (B) He–methanol flow.
Notes and references
2
1
(a) H. C. Fuller, Ind. Eng. Chem., 1924, 16, 624; (b) J. M. Hollis,
F. J. Lovas, P. R. Jewell and L. H. Coudert, Astrophys. J., 2002,
5
71, L59.
D. F. Othmer and M. S. Thakar, Ind. Eng. Chem., 1958, 50, 1235.
at 473 K. The results presented in Fig. 5 show that methyl
formate, formaldehyde, CO and CO are the main products of
methanol oxidation over the Cu/SiO catalyst under
He–methanol stream at 473 K. It is widely accepted that the
2
2
3 G. Xu, Y. Li, Z. Li and H. Wang, Ind. Eng. Chem. Res., 1995,
34, 2371.
2
4
5
6
Z. Li, Z. Qian, X. Zhao and W. Xiao, Chem. React. Eng. & Tech.,
004, 20, 121.
2
1
7
catalytic oxidation proceeds via the redox cycle. The redox
(a) L. R. Zehner, US Pat., 4005131, 1977; (b) L. R. Zehner and
W. R. Lenton, US Pat., 4112245, 1977.
(a) U. Matteoli, M. Bianchi, G. Menchi, P. Frediani and
F. Piacenti, J. Mol. Catal., 1985, 29, 269; (b) U. Matteoli,
G. Menchi, M. Bianchi and F. Piacenti, J. Organomet. Chem.,
cycle may result in the microscopic transport of surface atoms,
1
5
which accelerates the sintering. It is obvious that methanol
oxidation under reaction conditions is one of the causes of
copper sintering. For He–ethanol stream, the main products
are ethyl acetate and acetaldehyde, meanwhile the production
1986, 299, 233; (c) H. T. Teunissen and C. J. Elsevier, Chem.
Commun., 1997, 667.
7
(a) W. J. Bartley, US Pat., 4628128, 1986; (b) T. Susumu, F. Kozo,
N. Keigo, M. Masaoki and M. Katsuhiko, EU Pat., 0046983,
1987; (c) M. Haruhiko, H. Kouichi, U. Taizou, N. Yasuo,
I. Seizou, T. Takanori, JP Pat., 57123127, 1982.
A. Yin, X. Guo, W. Dai, H. Li and K. Fan, Appl. Catal., A, 2008,
of CO and CO is hardly observed. This fact indicates that,
2
under the reaction conditions, ethanol is more difficult to be
oxidized deeply into CO
2
than methanol and the ethanol
8
9
cannot be split into CO. We suggest that for the He–methanol
stream case the CO produced from methanol can intensify the
redox process of copper and the growing of Cu particles. Also,
the competition chemisorption between CO and DMO may
349, 91.
Z. He, H. Lin, P. He and Y. Yuan, J. Catal., 2010, 277, 54.
1
0 L. Chen, P. Guo, M. Qiao, S. Shen, H. Li, W. Shen, H. Xu and
K. Fan, J. Catal., 2008, 257, 172.
+
0
11 H. Miyazaki, T. Uda, K. Hirai, Y. Nakamura, H. Ikezawa and
T. Tsuchie, US Pat., 4585890, 1986.
disrupt the ratio of Cu /Cu of the active centers which
exacerbates the growing up of Cu particles.
1
2 D. Huang, Z. Chen, F. Chen, P. Lin, H. Lan and Y. He, Chin. Ind.
Catal., 1996, 4, 24.
The results of XRD presented in Fig. 2 and 4 also show that
there are some more aggregations of copper particles when
13 D. J. Thomas, J. T. Wehrli and M. S. Wainwright, Appl. Catal., A,
992, 86, 101.
1
DMO appears in H –methanol feedstock. However, when
2
1
4 (a) A. K. Datye, Q. Xu, K. C. Kharas and J. M. McCarty, Catal.
Today, 2006, 111, 59; (b) S. B. Simonsen, I. Chorkendorff, S. Dahl,
M. Skoglundh, J. Sehested and S. Helveg, J. Am. Chem. Soc., 2010,
DMO appears in H –ethanol feedstock, the copper particles
2
are still homogeneously dispersed on silica. It is suggested that
+
DMO, methanol and ethanol may prefer to adsorb on Cu
1
32, 7968.
1
8
15 (a) S. Komai, T. Hattori, A. Miyamoto and Y. Murakami, Chem.
Lett., 1983, 12, 1749; (b) F. Yang, M. S. Chen and
D. W. Goodman, J. Phys. Chem. C, 2009, 113, 254.
sites. Therefore, adsorption of DMO and methanol (or
ethanol) is competitive under the real reaction conditions.
On account of the weaker reducibility and chemisorption
ability of ethanol than methanol, ethanol oxidation may be
greatly reduced when DMO is present, which can effectively
slow down the agglomeration of active copper species.
On the other hand, the structure of glycol is more similar
with ethanol than methanol. This structural adaptability is in
favor of the stability of the catalytic active centers and the
lifetime of the catalysts.
16 (a) R. Zhang, Y. Sun and S. Peng, Fuel, 2002, 81, 1619;
(
b) S. D. Jackson, D. S. Anderson, G. J. Kelly, T. Lear,
D. Lennon and S. R. Watson, Top. Catal., 2003, 22, 173;
c) A. Y. Rozovskii and G. I. Lin, Top. Catal., 2003, 22, 137.
(
17 (a) D. Dadyburjor, S. Jewur and E. Ruckenstein, Catal. Rev. Sci.
Eng., 1979, 19, 293; (b) H. Kim, J. Yang and H. Jung, Appl. Catal.,
B, 2011, 101, 348.
1
8 (a) E. K. Poels and D. S. Brands, Appl. Catal., A, 2000, 191, 83;
b) A. Yin, X. Guo, W. Dai and K. Fan, J. Phys. Chem. C, 2009,
113, 11003.
(
This journal is c The Royal Society of Chemistry 2012
Chem. Commun., 2012, 48, 1177–1179 1179