Effect of the Cu/SBA-15 catalyst preparation method on methyl acetate hydrogenation for ethanol production

Article abstract of DOI:10.1039/c4nj00134f

Methyl acetate (MA) hydrogenation to ethanol was investigated over SBA-15 supported copper (Cu/SBA-15) catalysts prepared by different methods, including impregnation, deposition precipitation, and homogeneous deposition precipitation. The catalysts were characterized by means of the Brunauer-Emmett-Teller method, X-ray diffraction, transmission electron microscopy, temperature-programmed reduction and X-ray photoelectron spectroscopy. Of the three catalysts, it was found that the Cu/SBA-15 catalyst prepared by the homogeneous deposition precipitation method has the smallest particle size, and the appropriate Cu⁺/(Cu⁺ + Cu ⁰) molar ratio of 53.5%. The high dispersion of the active components, and the synergetic effect between Cu⁰ and Cu⁺ contributed to the high activity of Cu/SBA-15 prepared via the homogeneous deposition precipitation method. A MA conversion of 93.6% with an ethanol selectivity of 98.5% was obtained for MA hydrogenation in methanol solution under mild reaction conditions. Both of these values were much higher than those obtained over catalysts prepared by the other two methods. The catalyst still maintained good performance in the co-hydrogenation of acetic acid and MA in methanol solution. A high MA conversion of 99.5% and an ethanol selectivity of 99.5% were obtained for pure MA hydrogenation over the best catalyst. It was found that methanol solution inhibited the conversion of MA, and suppressed the side reaction of MA trans-esterification with the product ethanol at higher LHSV. the Partner Organisations 2014.

Full text of DOI:10.1039/c4nj00134f

NJC
View Article Online
View Journal | View Issue
PAPER
Effect of the Cu/SBA-15 catalyst preparation
method on methyl acetate hydrogenation for
ethanol production
Cite this: New J. Chem., 2014,
3
8, 2792
Shurong Wang, Wenwen Guo, Haixia Wang, Lingjun Zhu, Shi Yin and Kunzan Qiu*
Methyl acetate (MA) hydrogenation to ethanol was investigated over SBA-15 supported copper (Cu/SBA-15)
catalysts prepared by different methods, including impregnation, deposition precipitation, and homogeneous
deposition precipitation. The catalysts were characterized by means of the Brunauer–Emmett–Teller
method, X-ray diffraction, transmission electron microscopy, temperature-programmed reduction and X-ray
photoelectron spectroscopy. Of the three catalysts, it was found that the Cu/SBA-15 catalyst prepared by
+
the homogeneous deposition precipitation method has the smallest particle size, and the appropriate Cu /
+
0
(
Cu + Cu ) molar ratio of 53.5%. The high dispersion of the active components, and the synergetic effect
0
+
between Cu and Cu contributed to the high activity of Cu/SBA-15 prepared via the homogeneous
deposition precipitation method. A MA conversion of 93.6% with an ethanol selectivity of 98.5% was
obtained for MA hydrogenation in methanol solution under mild reaction conditions. Both of these values
were much higher than those obtained over catalysts prepared by the other two methods. The catalyst still
maintained good performance in the co-hydrogenation of acetic acid and MA in methanol solution. A high
MA conversion of 99.5% and an ethanol selectivity of 99.5% were obtained for pure MA hydrogenation over
the best catalyst. It was found that methanol solution inhibited the conversion of MA, and suppressed the
side reaction of MA trans-esterification with the product ethanol at higher LHSV.
Received (in Porto Alegre, Brazil)
7th January 2014,
Accepted 24th March 2014
2
DOI: 10.1039/c4nj00134f
www.rsc.org/njc
Therefore, there is a great demand for the development of
new processes for ethanol production.
1
. Introduction
Ethanol, a promising alternative to petroleum-derived fuels,
The global supply of acetic acid (AcOH) is reported to have
6
has attracted increasing attention recently. It can be blended an overcapacity of more than 40% in 2010. Consequently, the
with gasoline, in certain proportions, to form ethanol gasoline. hydrogenation of AcOH to ethanol can simultaneously solve the
In addition, the use of ethanol–gasoline blended fuels can problems of ethanol shortage and AcOH overcapacity. However,
simultaneously reduce air pollution and the depletion of petro- there are many problems in AcOH hydrogenation, such as the
1
leum fuel resources. Currently, there are two main methods high cost of the catalyst, the low conversion rate, and the
7
,8
for ethanol production: fermentation and ethylene hydration. requirement for expensive equipment. The process of AcOH
In fermentation technology, a large part of the fermentation esterification to methyl acetate (MA), followed by the hydro-
feedstock consists of edible grain crops, which has a negative genation of MA to ethanol, can avoid these problems.
2
effect on global food supplies. An alternative fermentation
feedstock is the abundant lignocellulosic biomass. However,
the cellulosic ethanol cost is still high, due to the high cost of
enzyme production and the high enzymatic dosages required to
3
hydrolyse pretreated biomass. In addition, the cost and pro-
MA hydrogenation is an important step in this synthetic
duction capacity of ethylene-derived ethanol depend mainly on route and thus the design of the hydrogenation catalyst,
the oil supply. An alternative route for ethanol production is suitable for the MA hydrogenation reaction, is of great impor-
ethanol synthesis from syngas. This route still faces the obsta- tance. Ethyl acetate (EtOAc) is one of the main by-products
4
,5
cles of low CO conversion and low ethanol selectivity.
generated from the trans-esterification of MA with the product
9
–11
ethanol.
Wang et al. studied MA hydrogenation over Cu-S-
1
0,11
973 and Cu0.4–Zn0.6–O catalysts.
MA conversions of 22.5%
State Key Laboratory of Clean Energy Utilization, Zhejiang University,
and 72.6% were obtained accordingly, along with an ethanol
selectivity of 27.6% and 38%, i.e., 55.2% and 76.0% of the
Hangzhou 310027, P. R. China. E-mail: qiukz@zju.edu.cn; Fax: +86 571 87951616;
Tel: +86 571 87952066
2792 | New J. Chem., 2014, 38, 2792--2800
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2014

View Article Online
Paper
NJC
theoretical maximum, respectively. It was also found that the to obtain the aqueous solution, and then 5.0 g of SBA-15 was
formation of EtOAc was the main reason for the reduction of added to form a suspension. The suspension was then kept
ethanol selectivity. Hence, it is of great interest to develop an at 363 K for 7 h. The precipitate obtained was separated by
efficient catalyst to improve MA conversion and suppress EtOAc filtration, washed with deionized water, dried at 383 K over-
formation. Recently, mesoporous silica SBA-15 materials have night, and calcined at 723 K for 4 h. The prepared catalyst was
been widely used as catalytic supports due to the two-dimensional denoted as Cu/SBA-15-H.
hexagonal symmetry of their cylindrical pores and their high
12,13
2.2. Catalyst characterization
hydrothermal stability.
for the hydrogenation of esters to alcohols.
catalysts were prepared for MA hydrogenation.
Copper-based catalysts are efficient
14,15
Here, Cu/SBA-15 The actual copper loadings of the prepared catalysts were
determined by inductively coupled plasma atomic emission
Preparation methods will affect the distribution of the active spectroscopy (ICP-AES, Thermo iCAP 6000). The catalysts were
species, the interaction between the active species and the sup- firstly digested with the help of microwave in the inverse aqua
port, and the chemical states of the elements. All of these may regia (HNO : HCl = 3 : 1), and then analyzed by the ICP. The
3
16,17
eventually affect the catalytic activity.
Thus, we prepared pore texture and structure of the catalysts were examined by the
. N adsorption–desorption isotherms
Cu/SBA-15 catalysts by different methods including the impreg- physical adsorption of N
2
2
nation (IM) method, the deposition precipitation (DP) method, were recorded at 77 K using a Quadrasorb SI apparatus. Before
and the homogeneous deposition precipitation (HDP) method, the test, the sample was degassed at 573 K for 3 h. The specific
and compared their catalytic performance in the gas phase surface area was calculated according to the multipoint
hydrogenation of MA. In the AcOH esterification with methanol, Brunauer–Emmett–Teller (BET) method. The pore-size distribu-
alcohol is usually added in excess to realize high acid conver- tion was determined by the Barrett–Joyner–Halenda (BJH)
18,19
sion.
Accordingly, we carried out MA hydrogenation in metha- method from the desorption branches of the adsorption iso-
nol solution, without separation. Since it is difficult to convert therms. Powder X-ray diffraction (XRD) analysis of the prepared
AcOH completely in the first step, the product was a mixed catalysts was carried out on a D/max-Ra X-ray diffractometer
solution of methanol, MA and a small amount of AcOH. There- with a Cu Ka radiation source operating at 40 kV and 30 mA, a
fore, we also studied the co-hydrogenation behavior of MA and step of 0.0161 and a collecting time of 10 s was used. The
AcOH in methanol solution over the optimized catalyst. Finally, particle sizes and their distribution were determined by trans-
pure MA hydrogenation was also investigated for comparison.
mission electron microscopy (TEM, Philips-FEI, Tecnai G2F30)
at 300 kV. Temperature programmed reduction (TPR) was
carried out on an AutoChem II 2920 instrument. The catalysts
were pretreated with He at 523 K for 60 min and then cooled to
2. Experimental
3
23 K. The reduction was carried out with a mixture of 10%
2.1. Catalyst preparation
À1
H
2
/He at a ramp rate of 10 K min
up to 973 K. X-ray
The Cu/SBA-15 catalysts were prepared by three methods, photoelectron spectroscopy (XPS) and X-ray Auger electron
namely the IM, DP, and HDP methods. The nominal Cu loading spectroscopy (XAES) were carried out using a PHI 5000C
amount of the three catalysts was 20 wt%. The SBA-15 zeolite ESCA system equipped with an MgKa X-ray radiation source
was purchased from Nanjing XFnano Materials Tech Co., Ltd, (hn = 1253.6 eV). Binding energies were calibrated using the C
China. It was used as a support for the preparation of all the 1s peak at 284.6 eV as reference, with an uncertainty of Æ0.2 eV.
catalysts.
2.3. Test of catalytic activity
IM: First, 3.8 g of Cu(NO
3
)
2
Á3H
2
O was dissolved in 60 ml of
deionized water at ambient temperature to obtain the aqueous The reactions were carried out in a continuous-flow fixed-bed
solution, and then 5.0 g of SBA-15 was added to form a reactor. In each experiment, 2 g of the catalyst was sandwiched
suspension. After stirring for 8 h, the suspension was heated with quartz sand and packed into a steel tube reactor with an
to 363 K to evaporate water completely, and then dried at 383 K inner diameter of 8 mm. A thermocouple was inserted into the
overnight. The final sample was calcined at 723 K for 4 h. The center of the catalytic bed to monitor the reaction temperature.
prepared catalyst was denoted as Cu/SBA-15-I.
Before the reaction, the catalysts were activated with pure H at
2
2
0
À1
DP: First, 3.8 g of Cu(NO
3
)
2
Á3H
2
O was dissolved in 112 ml 633 K for 3 h at 5 K min , and then cooled to the reaction
of deionized water at ambient temperature to obtain the aqu- temperature. Afterwards, MA–methanol solution, MA–AcOH–
eous solution, and then 5.0 g of SBA-15 was added to form a methanol solution or liquid MA was fed into the reaction
À1
suspension. And 112 ml of 0.16 mol L
2 3
Na CO aqueous preheating zone using a micropump, evaporated to gas, and
solution was added dropwise to the suspension under stirring carried by the hydrogen flow into the reactor. The flow rate of
at 333 K. After aging at 333 K for 2 h, the mixture was filtered H was controlled using a flow meter (Brooks, 5850E). For each
2
off, and then washed with deionized water, dried at 383 K catalyst, the tests at different temperature were all performed
overnight, and calcined at 723 K for 4 h. The prepared catalyst against fresh catalysts. And the result was obtained after 5 h on
was denoted as Cu/SBA-15-D.
stream. The flow rate of the gas effluent was measured using an
O and 13.4 g of urea were electronic flow meter (Agilent, AMD1000), and the composition
2
1
HDP: First, 3.8 g of Cu(NO
3
)
2
Á3H
2
dissolved in 223 ml of deionized water at ambient temperature of the gas was analyzed using an online gas chromatograph
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2014
New J. Chem., 2014, 38, 2792--2800 | 2793

View Article Online
NJC
Paper
(GC, Agilent 7890A) equipped with a 6-port valve and a 10-port
valve. CO, H and CO were separated using a Porapak Q and
2
2
carbon sieve packed column and detected using a thermal
conductivity detector. The gaseous hydrocarbons were sepa-
rated using a HP-Plot Q capillary column and detected using a
flame ionization detector (FID). Liquid products were con-
densed and analyzed using an Agilent 7890A GC equipped with
an Innowax capillary column and FID. The MA conversion
(XMA), AcOH conversion (XAcOH) and product selectivity (S ) were
i
calculated using the following equations, with the carbon
balance held at 100 Æ 5%.
XMA(%) = (nMAin À nMAout)/nMAin  100%
(1)
(2)
(3)
X
AcOH(%) = (nAcOHin À nAcOHout)/nAcOHin  100%
P
S
i
(%) = (m
i
 n
i
)/ (m
i
 n )  100%
i
where nMAin, nAcOHin and nMAout, nAcOHout represent the number of
moles of MA and AcOH in the feed and outlet, respectively. n
i
is
the number of moles of product i, and m is the carbon number of
i
product i. Since methanol is an inevitable byproduct generated
along with ethanol and not the competitive reaction product in
the MA hydrogenation, the product selectivity calculations do not
include methanol. In the tests, no gaseous product was detected.
3. Results and discussion
3.1. Catalyst characterization
3.1.1. Textural properties of catalysts. The actual Cu con-
tents in the calcined Cu/SBA-15 catalysts are summarized in
Table 1. It is 17.2 wt% for Cu/SBA-15-I and Cu/SBA-15-H catalysts,
which is somewhat lower than that for the Cu/SBA-15-D catalyst
Fig. 1 (a) N
tion curves of different catalysts.
2
adsorption–desorption isotherms and (b) pore size distribu-
(
18.7 wt%). The textural properties of the support and catalysts are
2
À1
shown in Table 1. The specific surface area of SBA-15 is 574 m g
with an average pore size of 7.5 nm. The BET surface area of the
loaded catalysts is clearly lower than that of SBA-15. Fig. 1(a) shows supported catalysts, another kind of pore of 3.5 nm appears and
the N adsorption–desorption isotherms of the catalysts. They the pore diameter of 6.8 nm decreases in comparison to the bare
,
2
exhibit type IV adsorption isotherms, and a typical H2 type hyster- SBA-15 material. It may be ascribed to the fact that partial copper
esis loop, indicating that the catalysts consist of cylindrical pores loading into the cylindrical pores of SBA-15 decreased the pore size
24
with some additional pore structure. The isotherms of Cu/SBA-15-H correspondingly. A similar result was observed by Guo et al. For
show a less sharp inflection and a more deformed H2 hysteresis Cu/SBA-15-H, a wider pore size distribution is found, between 3–
loop, which are due to the partial destruction of the ordered 8 nm, compared with those of the Cu/SBA-15-I and Cu/SBA-15-D
22
structure. It was found that alkaline conditions had a negative catalysts. This may be attributed to the destruction of some
effect on the hydrothermal stability of mesoporous SBA-15 to a channels of SBA-15 during the catalyst preparation, and the
23
certain extent, which may be the reason for the partial destruction deposition of copper species in the pores of SBA-15.
of the silica wall during catalyst preparation. Fig. 1(b) displays the 3.1.2. XRD. Fig. 2 shows the XRD patterns of the Cu/SBA-15
pore size distribution of the support and the catalysts. For the catalysts. Small-angle XRD patterns are shown in Fig. 2(a). The
Table 1 Textural properties of the prepared catalysts
a
2
À1
3
À1
b
b
c
d
+
Support/catalysts Cu loading (wt%)
S
BET (m g
)
V
Pore (cm g
)
Dpore (nm)
d
Cu (nm)
d
Cu O (nm)
d
copper species (nm)
X
(Cu
)
(%)
2
SBA-15
—
574
398
295
378
1.1
7.5
5.7
7.9
5.7
—
—
—
—
3.5
—
85
54
—
Cu/SBA-15-I
Cu/SBA-15-D
Cu/SBA-15-H
17.2
18.7
17.2
0.56
0.58
0.54
41.5
31.2
—
24.6
32.7
53.5
4.5
a
b
c
d
+
+
0
Determined by ICP-AES. Crystallite sizes calculated by the Scherrer formula. Determined by TEM. Intensity ratio of Cu to (Cu + Cu )
obtained by deconvolution of Cu LMM XAES spectra.
2794 | New J. Chem., 2014, 38, 2792--2800
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2014

View Article Online
Paper
NJC
reference (Fig. 2(c)). For Cu/SBA-15-I and Cu/SBA-15-D, sharp
diffraction peaks at 2y = 33.41, 35.61, 38.81, 48.81, 61.51, 66.21
and 68.11 are assigned to copper oxide indicating that a large
metallic crystallite size was present in the catalysts. However,
no diffraction peaks for any crystalline phases of copper are
observed for Cu/SBA-15-H, implying that the active components
were well dispersed in this preparation process. After
reduction, the diffraction peaks detected at 2y = 43.31, 50.41
are assigned to Cu for Cu/SBA-15-I and Cu/SBA-15-D (Fig. 2(c)).
However, no visible diffraction peaks for Cu can be detected for
Cu/SBA-15-H, indicating that Cu was highly dispersed and the
particle sizes were too small to be detected. And a new crystal
phase assigned to Cu O with a 2y of 36.51 and 42.41 was found
2
for Cu/SBA-15-H. The crystallite sizes of the active components
were calculated according to the Scherrer equation and the
results are summarized in Table 1. The copper crystallite sizes
are 41.5 and 31.2 nm for Cu/SBA-15-I and Cu/SBA-15-D, respec-
tively. However, the crystallite size of Cu
about 3.5 nm.
2
O for Cu/SBA-15-H is
3.1.3. TEM. Fig. 3 displays TEM images of the catalysts
after reduction. From Fig. 3(A1) and (B1), it can be seen that the
support preserves its hexagonal pore structure, and the large Cu
particles are distributed on the outside surface of the SBA-15.
As shown in Fig. 3(A2) and (B2), the active components of
Cu/SBA-15-I and Cu/SBA-15-D are large and poorly dispersed,
with average particle sizes of 85 and 54 nm, respectively. The
particle size was calculated based on the statistics of about
150 particles from the TEM images in different areas. The
diameter of the hexagonal pores of SBA-15 is 7.5 nm, estimated
from Fig. 3(B1). This value is consistent with the average pore
size determined by BJH. TEM images of Cu/SBA-15-H (Fig. 3(C1)
and (C2)) indicated the partial destruction of the ordered
structure. The active components are well-dispersed, with
average particle sizes calculated to be 4.5 nm. The average
particle size of the three catalysts (high to low) is: dCu/SBA-15-I
4
dCu/SBA-15-D 4 dCu/SBA-15-H. The crystal planes for Cu and Cu O
2
can be found in Fig. 3(A3), (B3) and (C3), which indicate that
the active components of these catalysts are Cu and Cu
2
O.The
w d ,
i i
P
P
3
2
average particle size was calculated using d =
where w
diameter d
.1.4. TPR. The H
Fig. 4. The TPR profile of pure CuO is also included for
SBA-15 sample exhibits three diffraction peaks, at 2y = 0.81 comparison. A broad peak at the temperature of B550–750 K
100), 1.61 (110) and 1.81 (200), which are typical of SBA-15 XRD is observed. A single peak centered at B665 K was found by
w
i
d
i
/
i
is the number of particles having a characteristic
Fig. 2 XRD patterns of the support and the prepared catalysts: (a) small-
angle XRD patterns, (b) after calcination and (c) after reduction.
26
i
(within a given diameter range).
3
2
-TPR profiles of samples are shown in
(
2
7
patterns. After the loading of the active components, a shift of Jiang et al. for pure CuO reduction. The Cu/SBA-15-H sample
the diffraction peaks at 0.81 to higher angles is observed for displays a sharp reduction peak centered at 503 K, much lower
Cu/SBA-15-I and Cu/SBA-15-D catalysts, indicating the contrac- than bulk CuO reduction. Combining this with the XRD results,
tion of the framework due to the loading of copper species and it indicates that the peak centered at 503 K should be assigned
2
5
0
the calcination treatment. However, no diffraction peaks were to the reduction of the small well dispersed CuO to Cu and
+
found for 2y between 01 and 31 for the Cu/SBA-15-H catalyst. In Cu . For the Cu species with a high dispersion and a narrow
2
+
combination with the TEM images shown below, it indicated particle size distribution, the two step reduction (Cu
-
+
0
14,15
the partial destruction of the ordered pore structure of SBA-15, Cu - Cu ) may take place at a similar temperature,
due to the dissolution of the support during the HDP process. then only one peak is observed. For the H
and
2
-TPR profiles of
XRD patterns of the prepared catalysts after calcination are Cu/SBA-15-D and Cu/SBA-15-I, the peak at lower temperature
shown in Fig. 2(b). The XRD pattern of SBA-15 is also given for (B500–510 K) may be attributed to the reduction of surface
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2014
New J. Chem., 2014, 38, 2792--2800 | 2795

View Article Online
NJC
Paper
Fig. 3 TEM images of the prepared catalysts after reduction. (A1)–(A3) Cu/SBA-15-I; (B1)–(B3) Cu/SBA-15-D; (C1)–(C3) Cu/SBA-15-H.
CuO particles and the strong interaction between copper spe-
2
8,29
cies and the support.
This result illustrates that the IM and
DP methods result in poorer dispersion of the copper species,
bigger particle size, and higher reduction temperature, com-
pared with the HDP method.
3.1.5. XPS. The XPS and XAES spectra of the reduced
catalysts are illustrated in Fig. 5. In Fig. 5(A), Cu2p3/2 and
Cu2p1/2 peaks centered at binding energies (BE) of 933.2 and
+
0
30
9
53.0 eV were conventionally assigned to Cu or Cu species.
The slightly higher BE of Cu2p3/2 was observed, which was
attributed to the interaction of the copper species with the
0
+
support. Generally, the binding energies of the Cu and Cu
Fig. 4 TPR profiles of the prepared catalysts.
species are almost the same, and it depends on the XAES
spectra to distinguish them. The broad and asymmetric peak
copper species (Cu - Cu ). The main peak (B593 K) and the of the Cu LMM Auger spectra indicates that Cu and Cu may
shoulder at B630 K may mainly belong to the reduction of bulk coexist on the surface of Cu/SBA-15 catalysts. Two peaks,
2
+
+
+
0
2796 | New J. Chem., 2014, 38, 2792--2800
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2014

View Article Online
Paper
NJC
Fig. 6 Influence of reaction temperature on (a) MA conversion and (b) the
selectivity of products over Cu/SBA-15 catalysts for MA hydrogenation in
À1
methanol solution (P = 2 MPa, n(H
2
)/n(MA) = 60, LHSV = 0.8 h (12.5 wt%
MA, 87.5 wt% methanol)).
Fig. 5 (A) XPS spectra and (B) Cu LMM Auger spectra of the reduced Cu/
SBA-15 catalysts: (a) Cu/SBA-15-I, (b) Cu/SBA-15-D, (c) Cu/SBA-15-H.
5
8
13 K and 523 K, respectively, with an MA conversion of 35.8%,
8.4%, and an ethanol selectivity of 74.7%, 93.3%, respectively.
+
0
Cu/SBA-15-H showed the best catalytic performance, since its MA
conversion reached 93.6%, with an ethanol selectivity of 98.5%,
at a relatively low temperature of 493 K.
centered at 914.2 and 917.7 eV, correspond to the Cu and Cu
species, respectively. The reduced catalysts showed different
+
+
0
ratios of Cu /(Cu + Cu ) (Table 1); the Cu/SBA-15-H catalyst has
+
+
0
Of the three catalysts, Cu/SBA-15-H has the smallest particle
the highest Cu /(Cu + Cu ) ratio.
+
+
0
size, and the highest Cu /(Cu + Cu ) ratio of 53.5%. The
catalytic activity is inversely correlated with the particle size.
In MA hydrogenation, it was found that Cu dissociatively adsorbs
3.2. Catalytic activities
0
+
32
3.2.1. MA hydrogenation in different reaction media over
H
2
, and Cu stabilizes the acyl species and the methoxy.
+
Cu/SBA-15 catalysts. Gas-phase hydrogenation of MA in methanol Moreover, Cu may function as electrophilic or Lewis acid sites
solution to ethanol was carried out over the Cu/SBA-15 catalysts to polarize the CQO group via the electron lone pair on its
prepared using different methods. As shown in Fig. 6, ethanol oxygen, thus improving the ester group reactivity. Therefore,
3
3
+
was the main product of MA hydrogenation, and methyl Cu plays an indispensable role in MA hydrogenation. Previous
formate was the main by-product, which can be ascribed to studies of Cu-based catalysts for ester hydrogenation have
3
1
+
0
the methanol dehydrogenation. Under reaction conditions of shown that a Cu /Cu ratio of about 1 or slightly higher than
P = 2 MPa, n(H )/n(MA) = 60, a liquid hourly space velocity 1 is more appropriate for obtaining better catalytic perfor-
2
LHSV means the volume of liquid input per volume of the mance. He et al. reported that the 1B-Cu–SiO catalyst
2
1
5,34,35
15
(
À1
catalyst per hour) of 0.8 h , all three catalysts showed low displayed the best catalytic performance and activity for vapor-
catalytic activity at low temperature; when the temperature phase hydrogenation of dimethyl oxalate to ethylene glycol; this
reached their respective transition temperature, the MA conver- catalyst has a Cu /(Cu + Cu ) ratio of 64.1%. Chen et al. obtained
sion increased significantly. The ethanol selectivity decreased with the highest hydrogenation activity with a Cu /(Cu + Cu ) ratio of
temperature increase over the three catalysts. The catalytic perfor- 54.9% over the CuSi-363 catalyst, which also had the largest Cu
+
+
0
34
+
+
0
+
mance of the three catalysts from high to low is Cu/SBA-15-H 4 surface area. In the present work, the Cu/SBA-15-H catalyst with
+
+
0
Cu/SBA-15-D 4 Cu/SBA-15-I. The optimum reaction temperature a Cu /(Cu + Cu ) ratio of 53.5% showed the best performance.
for MA hydrogenation over Cu/SBA-15-I and Cu-SBA-15-D was In conclusion, the optimal catalytic activity of Cu/SBA-15-H
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2014
New J. Chem., 2014, 38, 2792--2800 | 2797

View Article Online
NJC
Paper
Table
catalyst
2
MA and AcOH co-hydrogenation over the Cu/SBA-15-H contained a small amount of unreacted AcOH. Therefore, we
a
further investigated AcOH and MA co-hydrogenation in metha-
nol solution over the Cu/SBA-15-H catalyst. As listed in Table 2,
Selectivity (%)
the AcOH, being a small amount in the reactant, can be totally
MA
AcOH
Methyl
converted. When the content of AcOH increased, the MA
conversion decreased while the ethanol selectivity remained
unchanged. It is speculated that a large amount of methanol
first reacted with AcOH to produce MA, and then ethanol
was formed by MA hydrogenation. This demonstrates that
conversion (%) conversion (%) Ethanol formate
b
MA
93.6
—
98.5
98.5
98.6
99.5
1.5
1.5
1.4
0.5
c
MA + 1/8AcOH 92.5
MA + 1/4AcOH 91.9
MA + 1/2AcOH 86.6
100
100
100
a
Reaction conditions: P = 2 MPa, T = 493 K, n(H
2
)/n(MA + AcOH) = 60, co-hydrogenation of MA and a small amount of AcOH can be
À1
b
LHSV = 0.8 h
.
MA–methanol solution: 12.5 wt% MA, 87.5 wt%
achieved in methanol solution, and it was possible to realize
efficient indirect transformation from AcOH to ethanol.
In order to exclude the influence of methanol medium on
MA hydrogenation, pure MA hydrogenation was also studied
c
methanol. MA–AcOH–methanol solution: AcOH was added into the
MA–methanol solution (b) at the AcOH/MA molar ratio of 1/8.
is ascribed to the small particle size, the fine copper disper- over Cu/SBA-15-H, the catalyst which exhibited the highest
0
+
sion, and the synergistic effect between Cu and Cu , enhan- hydrogenation activity. At n(H
cing the dissociation of hydrogen and the activation of MA, sion was 99.5%, higher than that obtained in methanol solution.
respectively. This may be due to the presence of a large amount of methanol
2
)/n(MA) = 60, T = 493 K, MA conver-
In the AcOH esterification with methanol, alcohol is usually solution, which inhibited MA hydrogenation in the forward reac-
added in excess to realize high acid conversion. But it is tion, resulting in the decrease of MA conversion. The ethanol
1
8,19
difficult to convert AcOH completely,
and the products selectivity was 99.5%, and the main byproduct was EtOAc with
Fig. 7 Influence of n(H
2
)/n(MA) on MA hydrogenation in methanol solution over the catalysts at their respective optimum reaction temperatures: (a) MA
À1
conversion, (b) ethanol selectivity (P = 2 MPa, LHSV = 0.8 h (12.5 wt% MA, 87.5 wt% methanol)), (c) influence of n(H
over the Cu/SBA-15-H catalyst (P = 2 MPa, T = 493 K, LHSV = 0.4 h ).
2
)/n(MA) on pure MA hydrogenation
À1
2798 | New J. Chem., 2014, 38, 2792--2800
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2014

View Article Online
Paper
NJC
the selectivity of 0.4%, generated from the trans-esterification
of MA with the product ethanol. This result is different from
that of MA hydrogenation in methanol solution, where methyl
formate is the main by-product. As there is a large amount of
methanol in the feed, the side reaction of the above trans-
esterification reaction was suppressed significantly. We also
conducted EtOAc hydrogenation in methanol solution and
found that EtOAc was completely converted to ethanol and
MA. Therefore, no EtOAc was detected in the product of MA
hydrogenation in methanol solution.
3
.2.2. Influence of n(H )/n(MA). The effect of n(H )/n(MA)
2
2
on the performance of MA hydrogenation in different media at
their respective optimum reaction temperatures is shown in
Fig. 7. Fig. 7(a) shows that the MA conversion increased with
the increase of n(H
solution. Higher values of n(H
2
)/n(MA) for MA hydrogenation in methanol
)/n(MA) enhanced MA hydroge-
2
nation and suppressed the side reactions of MA dehydrogena-
tion; ethanol selectivity increased correspondingly(Fig. 7(b)).
With the same value of n(H )/n(MA), the order of catalytic activity
2
is also Cu/SBA-15-H 4 Cu/SBA-15-D 4 Cu/SBA-15-I. At n(H2)/
n(MA) = 15, the MA conversion was as low as 23.5% and the
ethanol selectivity was 26.7% over the Cu-SBA-15-I catalyst. In
contrast, the Cu/SBA-15-H catalyst performed better, with an MA
conversion of 72.3% and an ethanol selectivity of 89.0% at its
optimum reaction temperature. For pure MA hydrogenation over
the Cu/SBA-15-H catalyst (Fig. 7(c)), MA conversion and ethanol
selectivity present the same trend, higher values of n(H )/n(MA)
resulted in a higher MA conversion and ethanol selectivity. At
2
Fig. 8 Influence of LHSV on the performance of the Cu/SBA-15-H
catalyst in (a) MA hydrogenation in methanol solution (P = 2 MPa, T =
n(H )/n(MA) = 15, the MA conversion was as high as 95.0% and 493 K, n(H )/n(MA) = 60, 12.5 wt% MA, 87.5 wt% methanol) and (b) pure MA
2
2
the ethanol selectivity was 94.1%, both higher than that obtained hydrogenation (P = 2 MPa, T = 493 K, n(H
2
)/n(MA) = 15).
in methanol solution. This indicated that it was more favorable
for pure MA hydrogenation than MA hydrogenation in methanol
tested at 493 K, 2 MPa and H /MA = 15, as displayed in Fig. 9. In
2
solution without separation.
.2.3. Influence of LHSV. We further investigated the effect
the whole stability test process, experiments at high LHSV and
3
À1
low LHSV were carried out alternately. At an LHSV of 0.4 h , the
of LHSV on MA hydrogenation performance in different media
over Cu/SBA-15-H. As shown in Fig. 8(a), under conditions of
Cu/SBA-15-H catalyst maintained its initial high activity with
9
5.0%–96.4% MA conversion and 94.0%–95.5% ethanol selectiv-
P = 2 MPa, T = 493 K, n(H )/n(MA) = 60, when the LHSV
À1
2
ity. And at an LHSV of 2 h , the CO conversion remained steady
at 53.7%–54.6% during the 160 h test. This indicated that the
catalyst possessed high stability for MA hydrogenation.
À1
À1
increased from 0.8 h to 6.0 h for the MA hydrogenation
in methanol solution, the MA conversion decreased slightly
from 93.6% to 90.1%, and the ethanol selectivity decreased
from 98.5% to 96.8%. As Cu/SBA-15-H shows good catalytic
2
performance at an n(H )/n(MA) of 15 for the pure MA hydro-
genation, the influence of LHSV on pure MA hydrogenation was
studied at an n(H )/n(MA) of 15. As illustrated in Fig. 8(b), when
2
À1
À1
the LHSV increased from 0.4 h to 2.3 h , the MA conversion
decreased sharply from 95.0% to 45.1%, and the ethanol
selectivity decreased from 94.1% to 47.3%, with EtOAc selectiv-
ity increasing from 5.6% to 52.4%. This implied that in the case
of higher MA LHSV, side reaction of MA trans-esterification
with the product ethanol became relatively intense without
methanol solution, thus the selectivity for the EtOAc byproduct
increased.
3.2.4. Catalyst stability. In the practical applications of MA
hydrogenation, long-term stability is an important parameter for
a promising catalyst besides activity and selectivity. The stability
Fig. 9 Stability of Cu/SBA-15-H in the hydrogenation of MA (P = 2 MPa,
2
of the catalyst with the best performance (Cu/SBA-15-H) was T = 493 K, n(H )/n(MA) = 15).
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2014
New J. Chem., 2014, 38, 2792--2800 | 2799

View Article Online
NJC
Paper
4
. Conclusions
13 G. Luo, S. R. Yan, M. H. Qiao and K. N. Fan, Appl. Catal., A,
2007, 332, 79.
Cu/SBA-15 catalysts were prepared by three different methods,
namely IM, DP, and HDP. Cu/SBA-15-H was found to be the
most efficient catalyst for MA hydrogenation. The MA conver-
sion and ethanol selectivity reached 99.5% and 99.5% respec-
tively for pure MA hydrogenation, under the mild reaction
conditions. Cu/SBA-15-H also displayed good catalytic perfor-
mance in the co-hydrogenation of MA and AcOH in methanol
solution, achieving efficient indirect transformation from
AcOH to ethanol. And it was found that methanol solution
inhibited the conversion of MA, and suppressed the side
reaction of MA trans-esterification with the product ethanol
at higher LHSV. Characterization of the catalysts reveals that
the optimal catalytic activity of the Cu/SBA-15-H catalyst is a
result of highly dispersed copper particles, small particle size
1
1
1
1
4 S. R. Wang, X. B. Li, Q. Q. Yin, L. J. Zhu and Z. Y. Luo, Catal.
Commun., 2011, 12, 1246.
5 Z. He, H. Q. Lin, P. He and Y. Z. Yuan, J. Catal., 2011,
277, 54.
6 E. D. Guerreiro, O. F. Gorrizz, J. B. Rivarola and L. A. Arr u´ a,
Appl. Catal., A, 1997, 165, 259.
7 X. Zhanga, Y. p. Li, S. b. Qiu, T. J. Wang, M. Y. Ding,
Q. Zhang, L. L. Ma and Y. X. Yu, Chin. J. Chem. Phys.,
2013, 26, 77.
1
1
2
2
2
2
2
2
8 Z. G. Guo, S. R. Wang, G. H. Xu, Q. J. Cai and L. J. Zhu,
BioResources, 2011, 6, 2539.
9 Y. J. Liu, E. Lotero and J. G. Goodwin Jr, J. Catal., 2006,
242, 278.
0 M. P. Kapoor, Y. Ichihashi, W. J. Shen and Y. Matsumura,
Catal. Lett., 2001, 76, 3.
1 C. J. G. Van Der Grift, P. A. Elberse, A. Mulder and J. W.
Geus, Appl. Catal., A, 1990, 59, 275.
2 G. Lapisardi, F. Chiker, F. Launay, J.-P. Nogier and J.-L.
Bonardet, Catal. Commun., 2004, 5, 277.
3 S. Wu, Y. Han, Y. C. Zou, J. W. Song, L. Zhao, Y. Di, S. Z. Liu
and F. S. Xiao, Chem. Mater., 2004, 16, 486.
0
+
and the synergetic effect between the Cu and Cu species.
Acknowledgements
We acknowledge financial support from the National Natural
Science Foundation of China (51276166), the National Science and
Technology Supporting Plan Through Contract (2011BAD22B06),
and Zhejiang Provincial Natural Science Foundation of China
4 X. Guo, A. Yin, W. L. Dai and K. Fan, Catal. Lett., 2009,
(R1110089).
132, 22.
5 F. J. Brieler, M. Fr o¨ ba, L. M. Chen, P. J. Klar, W. Heimbrodt,
H.-A. K. v. Nidda and A. Loidl, Chem. – Eur. J., 2002, 8,
References
185.
1
M. A. R. Sadiq, Y. K. Ali and A. R. Noor, Eng. & Tech. Journal, 26 S. R. Wang, X. B. Li, L. Guo and Z. Y. Luo, Int. J. Hydrogen
2011, 29, 1438.
Energy, 2012, 37, 11122.
2
3
M. Balat and H. Balat, Appl. Energy, 2009, 86, 2273.
M. E. Himmel, S. Y. Ding, D. K. Johnson, W. S. Adney,
27 X. Y. Jiang, G. L. Lu, R. X. Zhou, J. X. Mao, Y. Chen and
X. M. Zheng, Appl. Surf. Sci., 2001, 173, 8.
M. R. Nimlos, J. W. Brady and T. D. Foust, Science, 2007, 28 L. Chen, T. Horiuchi, T. Osaki and T. Mori, Appl. Catal., B,
15, 804. 1999, 23, 259.
X. G. Song, Y. J. Ding, W. M. Chen, W. D. Dong, Y. P. Pei, 29 S. Bennici, P. Carniti and A. Gervasini, Catal. Lett., 2004,
J. Zang, L. Yan and Y. Lu, Catal. Commun., 2012, 19, 100.
98, 187.
R. Xu, C. Yang, W. Wei, W. H. Li, Y. H. Sun and T. D. Hu, 30 C. D. Wagner, W. M. Riggs, L. E. Davis, J. F. Moulder and
3
4
5
6
J. Mol. Catal. A: Chem., 2004, 221, 51.
G. E. Muilenerg, Handbook of X-ray Photoelectron Spectro-
scopy, Perkin-Elmer Corporation, Physical Electronics
Division, Minnesota, 1979.
Global Data, Acetic Acid: World Supply Outweighs Demand,
31 January, 2011.
7
8
W. Rachmady and M. A. Vannice, J. Catal., 2000, 192, 322.
31 R. Zhang, Y. H. Sun and S. Y. Peng, J. Nat. Gas Chem., 2000,
9, 110.
G. Onyesty ´a k, S. Harnos, A. Kaszonyi, M. Stolcova and
ˇ
´
D. Kall o´ , Catal. Commun., 2012, 27, 159.
32 E. K. Poels and D. S. Brands, Appl. Catal., A, 2000, 191, 83.
9
P. S. Wehner and B. L. Gustafson, J. Catal., 1992, 135, 420. 33 M. Bart o´ k and A. Moln ´a r, in The Chemistry of Double-Bonded
1
1
1
0 D. Wang, G. H. Yang, Q. X. Ma, M. B. Wu, Y. S. Tan,
Y. Yoneyama and N. Tsubaki, ACS Catal., 2012, 2, 1958.
1 D. Wang, G. H. Yang, Q. X. Ma, Y. Yoneyama, Y. S. Tan, 34 L. F. Chen, P. J. Guo, M. H. Qiao, S. R. Yan, H. X. Li,
Y. Z. Han and N. Tsubaki, Fuel, 2013, 109, 54. W. Shen, H. L. Xu and K. N. Fan, J. Catal., 2008, 257, 172.
2 H. Friedrich, J. R. A. Sietsma, P. E. D. Jongh, A. J. Verkleij 35 H. Q. Lin, X. L. Zheng, Z. He, J. W. Zheng, X. P. Duan and
Functional Groups (Suppl. A3), ed. S. Patai, Wiley, New York,
1997, p. 843.
and K. P. D. Jong, J. Am. Chem. Soc., 2007, 129, 10249.
Y. Z. Yuan, Appl. Catal., A, 2012, 287, 445.
2
800 | New J. Chem., 2014, 38, 2792--2800
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2014