Highly Selective Copper Catalyst Supported on Mesoporous Carbon for the Dehydrogenation of Ethanol to Acetaldehyde

Article abstract of DOI:10.1002/cctc.201500501

The dehydrogenation of ethanol to acetaldehyde is of great importance in synthetic chemistry and the fine chemical industry. In this study, we report that a mesoporous-carbon-supported Cu catalyst exhibited a high reaction rate and excellent product selectivity in ethanol dehydrogenation to acetaldehyde. Under the severe conditions of an ethanol concentration of 15 vol % and a gaseous hourly space velocity of 8 600 h^-1, the Cu-based carbon catalyst maintains a conversion of ≈73 % and an acetaldehyde selectivity of ≈94 % at 553 K. Meanwhile, a prominent space time yield (225 h^-1) of acetaldehyde is obtained, which is far higher than that (112 h^-1) on the mesoporous-SBA-15-supported Cu catalyst. Ethanol adsorption studies prove that the enrichment of the carbon support for reactants makes a contribution to the high reaction rate. Importantly, kinetic measurements indicate that the scarce surface groups of the mesoporous carbon support minimize the secondary reactions of acetaldehyde, which are generally catalyzed by -OH and/or -COOH on the surface of the support as exemplified by the mesoporous-SBA-15-supported Cu catalyst. This accounts for the excellent selectivity toward acetaldehyde on the mesoporous-carbon-supported Cu catalyst. Therefore, these surface characteristics of the carbon support show great advantages in this dehydrogenation reaction.

Full text of DOI:10.1002/cctc.201500501

DOI: 10.1002/cctc.201500501
Full Papers
Highly Selective Copper Catalyst Supported on
Mesoporous Carbon for the Dehydrogenation of Ethanol
to Acetaldehyde
Qing-Nan Wang, Lei Shi, and An-Hui Lu*[a]
The dehydrogenation of ethanol to acetaldehyde is of great
importance in synthetic chemistry and the fine chemical indus-
try. In this study, we report that a mesoporous-carbon-support-
ed Cu catalyst exhibited a high reaction rate and excellent
product selectivity in ethanol dehydrogenation to acetalde-
hyde. Under the severe conditions of an ethanol concentration
of 15 vol% and a gaseous hourly space velocity of 8600 hÀ1,
the Cu-based carbon catalyst maintains a conversion of ꢀ73%
and an acetaldehyde selectivity of ꢀ94% at 553 K. Meanwhile,
a prominent space time yield (225 hÀ1) of acetaldehyde is ob-
tained, which is far higher than that (112 hÀ1) on the mesopo-
rous-SBA-15-supported Cu catalyst. Ethanol adsorption studies
prove that the enrichment of the carbon support for reactants
makes a contribution to the high reaction rate. Importantly, ki-
netic measurements indicate that the scarce surface groups of
the mesoporous carbon support minimize the secondary reac-
tions of acetaldehyde, which are generally catalyzed by ÀOH
and/or ÀCOOH on the surface of the support as exemplified by
the mesoporous-SBA-15-supported Cu catalyst. This accounts
for the excellent selectivity toward acetaldehyde on the meso-
porous-carbon-supported Cu catalyst. Therefore, these surface
characteristics of the carbon support show great advantages in
this dehydrogenation reaction.
Introduction
The increase of alternative energy produced from renewable
biomass and its derivatives helps to alleviate our dependence
on fossil resources.[1,2] Ethanol, which is obtained easily from
biomass fermentation,[3] is particularly attractive because of its
increased annual production and low cost.[4,5] In synthetic
chemistry and the fine chemical industry, ethanol is utilized
widely for the production of hydrogen, acetic acid, 1-butanol,
ethyl acetate, 1,3-butadiene, acetaldehyde, and other chemi-
cals.[4,6–12] In particular, acetaldehyde, a significant bulk chemi-
cal, is valuable for the production of peracetic acid, pentaery-
thritol, pyridine bases, butanediol, and trichloroacetic alde-
hyde, with a worldwide production over 106 tons per year.[13,14]
As hydrogen is the sole byproduct, the dehydrogenation of
ethanol to acetaldehyde (DHEA)[11,12,15] represents an atom-eco-
nomic process,[8,9,16] which is a more favorable and green alter-
native to the current Wacker process of oxidation ethylene to
make acetaldehyde using PdCl2 and CuCl2 as catalysts in
strong acidic solutions.[4] Currently, in industrial production and
literature reports, SiO2-supported Cu catalysts are mainly used
in DHEA because of the nearly neutral surface of SiO2 com-
pared with Al2O3, ZnO, and ZrO2. However, the selectivity to-
wards acetaldehyde is not entirely satisfactory.[12,17–19] The low
selectivity is a fatal result of secondary reactions, that is, aldoli-
zation and ketonization,[20–22] which are generally catalyzed by
acid and/or base sites on the support surface.[10] Hence, there
is still a desire to develop new catalysts that have a notable se-
lectivity to acetaldehyde during DHEA.
Compared with conventional oxides, for example, SiO2 and
Al2O3, carbon materials possess a chemically inert nature[23,24]
and intrinsically hydrophobic properties.[25–27] The surface inert-
ness of the carbon material may inhibit the secondary reac-
tions of DHEA, and the hydrophobic surface probably exhibits
a potential enrichment ability for organic molecules, such as
ethanol. Moreover, mesoporous carbons (MCs) possess a fully
interconnected mesoporosity, which favors the dispersion of
active components and diffusion of reactants.[25,28] Consequent-
ly, it can be envisaged that if such an MC-supported Cu cata-
lyst is used for DHEA, excellent selectivity might be achieved.
To the best of our knowledge, there has been little research on
the use of MC-supported Cu catalysts for DHEA.
[a] Q.-N. Wang,+ Dr. L. Shi,+ Prof. A.-H. Lu
The State Key Laboratory of Fine Chemicals
School of Chemical Engineering
Dalian University of Technology
Dalian 116024 (P.R. China)
Fax: (+86)411-84986112
E-mail: anhuilu@dlut.edu.cn
[+] These authors contributed equally to this work.
In this paper, we selected MC as the support for the prepara-
tion of Cu catalysts and studied its catalytic performance for
DHEA. In comparison, a mesoporous silica SBA-15-supported
Cu catalyst (Cu/SBA-15) was evaluated for DHEA. It was found
that MC-supported catalysts exhibited excellent acetaldehyde
selectivity of 94.1% and a space time yield of 225 hÀ1 at 553 K
under a gaseous hourly space velocity (GHSV) of 8600 hÀ1. Fur-
thermore, the influence of the enrichment ability of the sup-
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cctc.201500501.
This publication is part of a Special Issue on “Carbon in the Catalysis
Community”. Once the full issue has been assembled, a link to its Table
of Contents will appear here.
ChemCatChem 2015, 7, 2846 – 2852
2846
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
port for the reactant on the reaction rate was investigated.
Combined with kinetic measurements, the origin of the high
selectivity to acetaldehyde on Cu/MC was elucidated.
Results and Discussion
Physicochemical properties of the catalysts
The precursors of the Cu/MC and Cu/SBA-15 catalysts were
prepared by the incipient wetness impregnation of Cu(NO3)2
aqueous solution (see details in the Experimental Section). The
physicochemical properties of the fresh catalysts are reported
in Table 1 and Figures 1 and 2. Highly dispersed metallic Cu
tends to be oxidized to high-valence oxide upon exposure to
the ambient atmosphere. Thus, in situ XRD characterization
(Figure 1a) was adopted to investigate the phase evolution of
Cu/MC catalysts under 5 vol% H2/Ar as a function of the reduc-
tion temperature. Diffraction peaks at 2q=43.3, 50.4, and
74.18 became observable because some Cu2+ species on the
Cu/MC catalysts precursor were reduced to metal Cu if the re-
duction temperature was ramped to 523 K. According to the
JCPDS card 04-0836, these diffraction peaks correspond to the
specific (111), (200), and (220) crystal planes of Cu, respective-
Figure 2. H2-TPR profiles of Cu/MC and Cu/SBA-15.
ly. In thermodynamic terms, there is little energy difference be-
tween the different low-index faces, so Cu metal is liable to
form multifaceted crystalline surfaces.[29] This result indicates
that only metallic Cu is visible after hydrogen treatment, which
is consistent with the result of Choi and Vannice.[29] With a fur-
ther increase of the reduction temperature to 823 K, the Cu
particle size increases slightly from 13.4 to 14.1 nm, which is
calculated by the Scherrer equa-
tion using the Cu (111) reflec-
tion. The particle size obtained
at the reduction temperature of
723 K for 2 h is presented in the
Table 1. For the Cu/SBA-15 cata-
lyst after reduction at 723 K, only
an amorphous diffraction peak
that ranges from approximately
15 to 358 is detected, therefore,
neither the peaks of metallic Cu
nor copper oxide appear (Fig-
ure S1a). This agrees with the re-
sults of Zhao et al. on Cu/SiO2.[30]
The crystallite size of the Cu on
this catalyst may be quite small
(below the detection limit of
XRD) or metallic Cu is oxidized
to a higher valence and, sponta-
neously, to a monolayer disper-
sion[31] upon exposure to air
during analysis.
Figure 1. a) In situ XRD patterns, b) N2 sorption isotherm, and c and d) TEM images of the Cu/MC catalyst.
Table 1. Structure parameters of the bare supports and the supported Cu catalysts after reduction.
Catalyst
Cu loading
[wt%]
SBET
[m2 gÀ1
Vtotal
[cm3 gÀ1
Dp
[nm]
Particle size [nm]
TEM
Dispersion of
Cu[a] [%]
Cu surface[a]
À1
]
]
XRD
N2O
[m2Cu gCu
]
MC
0
10
0
532
545
815
470
0.38
0.35
1.03
0.83
5.8
4.4
7.9
6.8
–
9.3
–
ND[b]
–
14.0
–
–
16.2
–
0
7.0
0
0
47.2
0
Cu/MC
SBA-15
Cu/SBA-15
10
ND
9.8
11.5
78.1
[a] Dispersion and surface of Cu were determined by N2O surface oxidation and H2-TPR titration. [b] ND, not detected.
ChemCatChem 2015, 7, 2846 – 2852
www.chemcatchem.org
2847
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
The porous structure and surface area of these samples
were analyzed by N2 sorption measurements at 77.4 K. The ni-
trogen sorption isotherm of the Cu/MC catalyst is of type IV in
shape, which represents a typical mesoporous structure (Fig-
ure 1b and Figure S1b). The Cu/SBA-15 sample also shows
a type IV isotherm, which is the same as that of the bare sup-
port.[32]
surface carboxyl groups on a carbon support were beneficial
to render a higher dispersion of Cu.[34] Although MC was ob-
tained by the pyrolysis of the carbon precursors at 1073 K,
a small amount of surface groups might remain.[35] Thus, the
first shoulder at 460 K is the reduction of highly dispersed Cu
particles.[36,37] The H2 consumption peak at 533 K can be as-
signed to the reduction of larger cupric oxide particles.[31] The
H2 consumption of these reduction processes are approximate-
ly 7.2, 32.6, and 5.3 mL/gcatalyst, which correspond to the areas
of the peaks at 460, 478, and 533 K, respectively. The quantity
of H2 consumed is higher than the theoretical value, approxi-
mately 35 mL/gcatalyst, which indicates the complete reduction
of Cu2+ to Cu. In addition, another H2 consumption peak ap-
pears gradually for this sample in a relatively high temperature
range of 713–950 K. This indicates the alkylation of carbon
with H2 catalyzed by Cu. The etching process enlarged the sur-
face area of the carbon support (Table 1) and favored the ex-
posure of active components to result in metal-porous/hollow
carbon catalysts.[38] This special structure might facilitate the
diffusion and reaction of reactants in the nanospace around
the Cu particles. A further increase of the reduction tempera-
ture leads to a decrease of this effect, which then becomes
absent as the particle size of Cu becomes larger. For Cu/SBA-
15, a sharp reduction peak appears at 496 K with a shoulder at
540 K. This main peak indicates the reduction of dispersed cup-
reous particles on the silica support.[31,37] The total H2 uptake of
this sample is approximately 41.5 mL/gcatalyst, which implies the
complete reduction of the CuO phase. Moreover, this result
also confirms that the active component on MC is reduced
easily to Cu under H2 treatment. The H2-TPR result is consistent
with the above discussion about the particle size distribution
and surface area of Cu/MC listed in Table 1.
The morphologies and structural details of the Cu/MC cata-
lyst were examined using TEM. The dark metal nanoparticles
assignable to Cu are highly dispersed in the mesopores of the
MC (Figure 1c and d and Figure S2). The inset statistical histo-
gram shows a narrow particle size distribution with an average
Cu nanoparticle size of approximately 9.3 nm, which is smaller
than that (ꢀ20 nm) of Cu/carbon nanofibers.[33] This is in line
with the results obtained from the in situ XRD patterns. It was
also reported that the reduction of CuO/C can partially remove
the anchoring groups from the carbon support to yield CO2 or
CO, which results in increased Cu particle sizes.[34] High-resolu-
tion transmission electron microscopy (HRTEM) shows that the
Cu particles have a d spacing of ꢀ0.209 nm, which corre-
sponds to the Cu (111) plane. This is consistent with the re-
sults of in situ XRD (Figure 1a), which includes only a small
proportion of (200) and (220) planes.
The structural parameters of Cu/MC and Cu/SBA-15 are sum-
marized in Table 1. The surface area and pore volume of Cu/
MC are almost the same as those of the bare support. Howev-
er, the pore diameter decreases from 5.8 to 4.4 nm. The surface
area, pore volume, and pore size of Cu/SBA-15 are dramatically
decreased compared to those of the support, for example, the
surface area is reduced from 815 to 470 m2gÀ1. This implies
that most of the deposited Cu component is filled into the
mesopores of the SBA-15 support. The average particle sizes of
the catalysts determined by N2O chemisorption, in situ XRD,
and TEM are also presented in Table 1. If we assume that the
Cu particles are spherical, the dispersion (D) of Cu on the Cu/
MC catalyst is 7.0%, which is lower than that (11.5%) of Cu/
SBA-15. Therefore, the Cu size of Cu/MC calculated by the
chemisorption of N2O is larger than that on Cu/SBA-15.
Catalytic performance of Cu/MC and Cu/SBA-15 catalysts
Ethanol dehydrogenation, catalyzed by Cu catalysts, is known
to produce acetaldehyde and hydrogen. However, byproducts
of ethyl acetate, methyl ethyl ketone, acetone, butanal, and
butanol[10,12,17–19] are also generated as a result of the further
reaction of acetaldehyde catalyzed by the acid and/or base
sites of the supports.[10,17] The results for the catalysis of DHEA
are compiled in Table 2 and Figures 3 and 4.
Hydrogen temperature-programmed reduction (H2-TPR;
Figure 2) was performed to reveal the type of metal species
and reducibility of Cu/MC and Cu/SBA-15. For the MC-support-
ed sample, there is a main reduction peak at 478 K, which rep-
resents the reduction of Cu2+ to Cu. In addition, two shoulder
peaks appear at 460 and 533 K. Dandekar et al. reported that
Under a GHSV of 26000 hÀ1 and 553 K (Table 2), there is
a slight difference between these two catalysts in terms of
Table 2. Catalytic results for ethanol dehydrogenation to acetaldehyde on Cu/MC and Cu/SBA-15.[a]
Catalyst
GHSV
[hÀ1
Conv.
[%]
Rate[b]
[sÀ1
Selectivity [%][c]
MEK
Yield[d]
[%]
STY[e]
[hÀ1
]
]
]
AC
EAC
ACE
Others
Cu/MC
Cu/SBA-15
Cu/MC
26000
26000
8600
83.0
90.1
72.7
82.8
0.110
0.070
0.096
0.064
95.1
82.1
94.0
68.1
1.7
8.2
3.0
0.7
1.3
1.4
2.0
0.7
1.0
0.6
0.7
1.8
7.4
1.0
79.0
73.9
68.4
56.4
258
147
225
112
Cu/SBA-15
8600
11.7
16.1
[a] Reaction conditions: catalyst 0.100 g, WHSVC H OH =2.4 hÀ1, N2 40 or 10 mLminÀ1, 553 K. [b] Rate represents the reaction rate of ethanol. [c] AC: acetalde-
2
5
hyde, EAC: ethyl acetate, MEK: methyl ethyl ketone, ACE: acetone, Others: 1-butanol, butanal, and C6+. [d] Yield is calculated as conversion”selectivity/
100. [e] STY represents the space time yield of acetaldehyde.
ChemCatChem 2015, 7, 2846 – 2852
www.chemcatchem.org
2848
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Cu dispersion of Cu/MC than that of Cu/SBA-15, at tempera-
tures below 473 K the reaction rate of Cu/MC is lower than
that of Cu/SBA-15. At temperatures above 473 K, with a de-
creased reactant concentration, the collision frequency be-
tween ethanol and active sites may be high, for which the en-
richment of the carbon support for the reactants contributes
to the high reaction rate. In view of the selectivity of acetalde-
hyde, Cu/MC is better than Cu/SBA-15, especially at a high re-
action temperature, for example, at 553 K the selectivity of Cu/
MC is 95.1%, whereas that of Cu/SBA-15 is 82.1% (Figure 3).
For catalytic reactions, the selectivity to the target products is
of importance. Although the Cu/MC catalyst has a lower Cu
dispersion than Cu/SBA-15, excellent acetaldehyde selectivity
and reaction rate are obtained. These exciting results show
that MC is a potential support for the preparation of active Cu-
based catalysts for DHEA. A comparative summary of the litera-
ture on ethanol dehydrogenation is listed in Table 3, which
supports the extraordinary activity of our carbon-supported
catalyst described here. This summary likewise presents what
is known about SiO2-based catalysts to aggravate side reac-
tions in ethanol dehydrogenation.
Figure 3. C2H5OH reaction rate and CH3CHO selectivity during dehydrogena-
tion over Cu/MC and Cu/SBA-15 as a function of the reaction temperature.
Reaction conditions: catalyst 0.1 g, weight hourly space velocity (WHSV) of
C2H5OH=2.4 hÀ1, N2 40 mLminÀ1, C2H5OH/N2 =5:95, GHSV=26000 hÀ1
.
Table 3. Comparison of the catalytic performance of Cu/MC and Cu cata-
lysts reported in the literature.
Catalyst
Cu
[wt%]
T
[K]
Conv.
[%]
Select.
[%]
Yield
[%]
Ref.
Cu/MC[a]
Cu/SiO2
Cu/SiO2
10
30
14.4
553
493
673
553
553
498
83.0
76.1
80.0
22.2
53.1
41.0
95.1
21.0
85.0
90.1
61.0
87.2
78.9
16.4
68.0
19.8
32.7
35.6
This work
[12]
[17]
Figure 4. Selectivity and space time yield of CH3CHO during dehydrogena-
tion over Cu/MC and Cu/SBA-15 as a function of the reaction temperature.
Reaction conditions: catalyst 0.1 g, WHSVC H OH =2.4 hÀ1, N2 10 mLminÀ1
,
2
5
C2H5OH/N2 =15:85.
Cu/SiO2
Cu/SiO2
10
10
[18]
[19]
[a] Reaction conditions: catalyst 0.1 g, WHSVC H OH =2.4 hÀ1
,
N2
2
5
conversion, 83.0% for Cu/MC and 90.1% for Cu/SBA-15. To re-
flect the real catalytic efficiency of these two catalysts, the re-
action rate is calculated as the amount of ethanol converted
[mmol] per the amount of surface Cu [mmol] obtained from
the metal dispersion [sÀ1] at each reaction temperature. The
comparison of reaction rates on these catalysts is shown in
Figure 3. Cu/MC shows a higher reaction rate than Cu/SBA-15
above 473 K. This superiority becomes clearer as the reaction
temperature increases, 0.11 versus 0.07 sÀ1 at 553 K. According
to literature reports,[12,17–19] Cu is the active center for this
DHEA. Although Cu/MC possesses a lower Cu dispersion than
Cu/SBA-15 (Table 1), it shows an excellent reaction rate for eth-
anol. So, to shed light on this phenomenon, the effect of the
support characteristics on the catalytic performance is consid-
ered.
40 mLminÀ1, C2H5OH/N2 =5:95.
Moreover, to explore the advantages of carbon-supported
Cu catalysts on product selectivity, more extreme reaction con-
ditions were investigated, for example, a lower GHSV of
8600 hÀ1 and a higher ethanol concentration of 15 vol%. The
selectivity and space time yield of acetaldehyde are shown in
Figure 4. The space time yield of acetaldehyde was defined as
the mass of acetaldehyde [g] formed per mass of surface Cu
[g] calculated from the Cu dispersion in 1 h [hÀ1] at each reac-
tion temperature. A selectivity of 94% retained almost con-
stant from 473 to 553 K. However, for Cu/SBA-15, it showed an
inverse proportional function between the selectivity and the
reaction temperature, and its selectivity was as low as approxi-
mately 68% at 553 K. Therefore, the space time yield of the
carbon-supported Cu catalyst is approximately twice as high as
that of Cu/SBA-15 at 553 K, which is accordingly 225 and
112 hÀ1. These results further demonstrate that MC is a favora-
ble alternative to silica to make highly active catalysts for the
dehydrogenation of ethanol.
If we consider the enrichment ability of hydrophobic carbon
for organic reactants, the adsorption experiments of ethanol
on MC and SBA-15 were conducted (Figure S3). The results
show that the carbon surface has a higher ethanol concentra-
tion than the silica surface. The reaction order of DHEA with re-
spect to ethanol is approximately 0.5 (Figure S4), therefore,
a high density distribution of ethanol molecules on the carbon
surface accelerates the reaction rate. As a result of the lower
ChemCatChem 2015, 7, 2846 – 2852
www.chemcatchem.org
2849
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Kinetic measurements
Generally, the characteristics of the active component have
a significant influence on the product distribution. To illumi-
nate the difference in product selectivity over these samples,
kinetic measurements were considered. An Arrhenius plot of
ethanol reaction rate (Figure 5) allows us to identify that the
Figure 6. Dependence of product selectivity and ethanol conversion on W/F
in the DHEA over a) Cu/MC and b) Cu/SBA-15. Reaction conditions:
WHSVC H OH =2.4–48 hÀ1, N2 40 mLminÀ1, C2H5OH/N2 =5:95, 533 K. Byprod-
Figure 5. Arrhenius plot of the reaction rate on the Cu/MC and Cu/SBA-15 at
453–533 K. Reaction conditions: catalyst 0.010 g for Cu/MC and 0.005 g for
Cu/SBA-15, WHSVC H OH =24 or 48 hÀ1, N2 40 mLminÀ1, C2H5OH/N2 =5:95.
2
5
2
5
ucts included ethyl acetate, methyl ethyl ketone, acetone, 1-butanol, buta-
nal, and C6+.
apparent active energy for the dehydrogenation reaction is
51.9 kJmolÀ1 for Cu/MC and 54.9 kJmolÀ1 for Cu/SBA-15. This
result implies that there is no difference in the nature of the
active sites between the two catalysts. It has been proven that
this reaction is structure insensitive.[39–41] Hence, it is apparent
that the properties of the supports are the dominant factors
that influence the acetaldehyde selectivity. Thus, we deduce
that the secondary reactions may be attributed to the selectivi-
ty difference between Cu supported on MC and SBA-15.
groups, ꢀ1.0Æ0.1 nm2,[43,44] which survive in isolated silanol
groups: SiÀOH.[45] However, the surface of the carbon material,
pyrolyzed at 1073 K, contains ꢀ0.04 ÀCOOH and ÀOH groups
per nm2 in addition to few other oxygenic units such as OÀC=
O.[35,46] Furthermore, the mass fraction of oxygen on mesopo-
rous silica SBA-15 is 53.3 wt%, which is much higher than that
on MC of 3.4 wt%.[47] Hence, on the basis of these results, we
consider that the different amounts of ÀOH and/or ÀCOOH
groups on support surfaces lead to the significant distinction
between Cu/MC and Cu/SBA-15 on acetaldehyde selectivity.
We combined our results with those in the literature to pro-
pose a possible schematic model of the reaction pathway of
ethanol dehydrogenation on MC- and silica-supported Cu cata-
lysts (Figure 7). To exhibit this difference vividly, the scarce sur-
face groups on the carbon support, ꢀ0.04 ÀCOOH and ÀOH
groups per nm2, are not shown. The secondary reactions are
To verify the above deduction, the effect of residence time
(W/F) on the product selectivity and ethanol conversion over
the two catalysts was investigated. With increasing W/F, the se-
lectivity to acetaldehyde decreases from 100 to 95.1%
(Figure 6), whereas the selectivity to byproducts increases from
0 to 3.1% on Cu/MC. Byproducts include ethyl acetate, methyl
ethyl ketone, acetone, 1-butanol, butanal, and C6+, which are
created by the cross- and self-aldol condensation of butanal
and acetaldehyde.[42] These results demonstrate clearly that
acetaldehyde appears as the primary product and the others
are obtained in the secondary reactions. On the basis of these
data, it was concluded that the acetaldehyde formed in a pri-
mary step was transformed successively to C3 or C4 oxygen-
containing components by aldolization and ketonization. This
tendency of Cu/SBA-15 is the same as that of the carbon-sup-
ported catalyst. However, the selectivity to acetaldehyde on
the SiO2-based catalyst is lower than that on Cu/MC. As the
active sites on both Cu-based MC and SBA-15 catalysts are
identical (Figure 5), the different product distributions are at-
tributed to the diverse surface nature of the supports. General-
ly, the difference between them is the amount of oxygenic
groups of the supports. On the surface of SBA-15 there are nu-
merous ÀOH groups, the only type of oxygen-containing
Figure 7. Scheme of C2H5OH conversion on Cu/MC and Cu/SBA-15.
ChemCatChem 2015, 7, 2846 – 2852
www.chemcatchem.org
2850
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
pregnation, the catalyst precursors were kept at 298 K for 30 min
followed by drying at 323 K for 12 h. Catalysts were reduced by
10 vol% H2 at 723 K for 2 h before reaction. For comparison,
10 wt% Cu/SBA-15 was also prepared using the same method.
catalyzed by SiÀOH on the surface of the SBA-15 support. As
a result, the selectivity to acetic aldehyde on Cu/MC is higher
than that of Cu/SBA-15.
Conclusions
Catalyst characterization
Cu nanoparticles supported on mesoporous carbon (MC) is an
active catalyst for the dehydrogenation of ethanol to acetalde-
hyde with excellent selectivity compared with Cu/SBA-15. The
reaction rate over Cu/MC was 1.7 times as high as that of Cu/
SBA-15 at 553 K with a C2H5OH/N2 molar ratio of 5:95 and
a gaseous hourly space velocity (GHSV) of 26000 hÀ1. A de-
crease of the GHSV and an increase of the reactant concentra-
tions led to a selectivity of 94% and a space time yield of
225 hÀ1 at 553 K and a GHSV of 8600 hÀ1 over the carbon-sup-
ported Cu catalyst, whereas a selectivity of 68% and an output
of 112 hÀ1 were obtained for Cu/SBA-15. A series of studies
were employed to explore the structure–activity relationship of
the catalysts. The high reaction rate of Cu/MC is attributed to
the strong enrichment ability of the carbon support for etha-
nol molecules. Kinetic measurements show that the essence of
the active sites is the same for the carbon- and silica-support-
ed catalysts, and the difference in acetaldehyde selectivity is
ascribed to the nature of the supports. The dependence of
product selectivity and ethanol conversion on residence time
suggests that abundant surface SiÀOH groups of SBA-15 pro-
mote the secondary reactions, which are largely responsible
for the low acetaldehyde selectivity of Cu/SBA-15. Conversely,
side reactions are minimized on the carbon-supported Cu cata-
lyst. Potentially, such a mesoporous carbon support with
a unique nanostructure can be a promising candidate for the
preparation of highly selective catalysts for ethanol dehydro-
genation.
The in situ XRD measurements of Cu/MC were performed by using
a Panalytical X’pert Pro Super X-ray diffractometer using CuKa radia-
tion (l=0.15418 nm) with a scanning angle (2q) of 30–808. The
tube voltage was 40 kV, and the current was 40 mA. Typically, Cu/
MC catalyst precursor was placed in a stainless-steel holder. Then
a
5 vol% H2/Ar mixture was introduced at a flow rate of
30 mLminÀ1. Temperature ramping programs were performed from
RT to 423, 523, 623, 723, and 823 K at a rate of 5 KminÀ1. The XRD
patterns were collected after samples reached the preset tempera-
tures for 30 min. The XRD pattern of Cu/SBA-15 after reduction at
723 K for 2 h was obtained by using a D/MAX-2400 diffractometer
using CuKa radiation (40 kV, 40 mA, l=0.154056 nm).
Nitrogen adsorption–desorption isotherms were measured by
using a TriStar 3000 adsorption analyzer (Micromeritics) at 77.4 K.
The supports were degassed at 473 K for 4 h before analysis, and
reduced samples were degassed at 363 K for 4 h. The BET method
was used to calculate the specific surface areas (SBET). Total pore
volumes (Vtotal) were calculated from the amount adsorbed at a rela-
tive pressure P/P0 of 0.99. Pore size distributions (PSDs) were deter-
mined from the adsorption branches of the isotherms using DFT.
TEM and HRTEM measurements were conducted by using a Tecnai
G2 20 S-Twin microscope with an acceleration voltage of 200 kV.
The reducibility of 10 wt% Cu/MC and 10 wt% Cu/SBA-15 was de-
termined by H2-TPR by using a Micromeritics Autochem II2920 in-
strument. Catalyst precursors (100 mg) were loaded into a quartz
tube and dried in an argon stream at 373 K for 30 min before re-
duction to drive off water adsorbed physically. After it was cooled
to 350 K, the sample was then heated in 30 mLminÀ1 of 8 vol% H2/
Ar at a rate of 10 KminÀ1 to 950 K. The amount of H2 consumed
during the run was detected by using a thermal conductivity de-
tector (TCD).
Experimental Section
A H2-N2O titration method was used to measure the Cu dispersion
and average particle diameter in three processes.[49] First, the
sample was reduced with 8 vol% H2/Ar at 723 K for 2 h. The first
H2 consumption X at 350–650 K was collected. This H2 consump-
tion provides the total number of Cu atoms. It was confirmed that
no further H2 consumption was observed at a higher temperature
by the reduction of Cu2+ to Cu. Second, the catalyst was cooled to
363 K, and 5 vol% N2O/He was purged to the sample for 30 min
for the oxidation of metallic Cu to cuprous oxide. Third, the
second H2-TPR was conducted from 313–573 K to reduce Cu+ to
Cu to obtain the second H2 consumption Y. The dispersion was cal-
culated by D [%]=2Y/X”100.[50]
Preparation of supports and catalysts
Resorcinol (99.5%) and formalin (37 wt% formaldehyde) were pur-
chased from Tianjin Kermel Chemical Reagent Co., Ltd. 1,6-Diami-
nohexane (DAH, 99.0%), Cu(II) nitrate trihydrate (>99.0%), and
ethanol (99.7%) were supplied by Sinopharm Chemical Reagent
Co., Ltd. Pluronic F127 was purchased from Fluka. All chemicals
were used as received. Mesoporous silica SBA-15 was synthesized
according to a previous report.[32]
MC was prepared using a procedure reported previously.[48] Typical-
ly, resorcinol (24 g) and Pluronic F127 (10 g) were dissolved in a sol-
vent mixture of ethanol (90 mL) and deionized water (72 mL) at
298 K. Afterward, 1,6-diaminohexane (1.247 g) was added to the
above solution, which was stirred for ꢀ30 min at 298 K. Subse-
quently, formalin (32.6 mL) was injected quickly into the solution.
After it was stirred at 298 K for another 10 min, the white homoge-
neous emulsion was then sealed in a flask and transferred to an
oven at 363 K for 4 h. The as-made polymer monolith was dried at
323 K for 24 h, followed by pyrolysis at 1073 K for 2 h under N2 at-
mosphere to obtain MC. Before catalyst preparation, MC was dried
at 393 K for 2 h in a clean vial to eliminate water. Cu/MC with
10 wt% of Cu was prepared by incipient wetness impregnation
with a Cu(NO3)2·3H2O aqueous solution (0.625 gmLÀ1). After im-
Catalytic tests
DHEA was performed in a fixed-bed quartz tube reactor (i.d. 8 mm)
under atmospheric pressure. Typically, catalyst (100 mg) was
placed in the central zone of the reactor. Before the reaction, the
catalyst was pretreated by in situ TPR from RT to 723 K for 2 h at
a rate of 5 KminÀ1 using 10 vol% H2 in N2 (44 mLminÀ1 in total).
When the temperature decreased to 373 K, ethanol (WHSV=
2.4 hÀ1) was introduced into the evaporator (373 K) by using a sy-
ringe pump and carried into the reactor by flowing nitrogen gas
ChemCatChem 2015, 7, 2846 – 2852
www.chemcatchem.org
2851
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
with a total flow rate of 40 mLminÀ1 (C2H5OH/N2 =5:95, molar
ratio). The product line was heated at 393 K before a cold trap to
remove the condensed liquid products. A GC7900 GC fitted with
a FFAP capillary column (30 m, 0.32 mm, 0.5 mm) and a flame ioni-
zation detector (FID) was connected to the line between the reac-
tor outlet and cold trap to analyze the products in the effluent
gas. After the cold trap, the dry gas was detected by a TCD
equipped with molecular sieves 5 Š and GDX-102 columns for the
analysis of gas products (e.g., C2H4, CH4, and others). CH4 was used
to bridge FID and TCD. The carbon balance of these reactions is
more than 96% [Eqs. (1) and (2)].
[15] V. L. Sushkevich, I. I. Ivanova, E. Taarning, ChemCatChem 2013, 5, 2367–
2373.
[16] J. Bauer, G. M. Veith, L. F. Allard, Y. Oyola, S. H. Overbury, S. Dai, ACS
Catal. 2012, 2, 2537–2546.
[17] J. Keuler, PhD thesis, University of Stellenbosch, South Africa, 2000.
[18] Y. Zhang, Master’s thesis, Tianjin University, China, 2007.
[19] A. G. Sato, D. P. Volanti, I. C. D. Freitas, E. Longo, J. Bueno, Catal.
Commun. 2012, 26, 122–126.
[20] J. Scalbert, F. Thibault-Starzyk, R. Jacquot, D. Morvan, F. Meunier, J.
Catal. 2014, 311, 28–32.
[21] T. Nishiguchi, T. Matsumoto, H. Kanai, K. Utani, Y. Matsumura, W.-J. Shen,
S. Imamura, Appl. Catal. A 2005, 279, 273–277.
[22] M. Iwamoto, M. Tanaka, S. Hirakawa, S. Mizuno, M. Kurosawa, ACS Catal.
2014, 4, 3463–3469.
[23] R. Lv, T. Cui, M.-S. Jun, Q. Zhang, A. Cao, D. S. Su, Z. Zhang, S.-H. Yoon, J.
Miyawaki, I. Mochida, F. Kang, Adv. Funct. Mater. 2011, 21, 999–1006.
[24] D. Su, S. Perathoner, G. Centi, Chem. Rev. 2013, 113, 5782–5816.
[25] S. Wang, Q. Zhao, H. Wei, J. Q. Wang, M. Cho, H. Cho, O. Terasaki, Y.
Wan, J. Am. Chem. Soc. 2013, 135, 11849–11860.
Ethanol conversion ¼ ðEthanolinÀEthanoloutÞ=Ethanolin  100 %
ð1Þ
Product selectivity ¼Carbon in given product=
ð2Þ
[26] Z. Y. Jin, A. H. Lu, Y. Y. Xu, J. T. Zhang, W. C. Li, Adv. Mater. 2014, 26,
3700–3705.
Carbon in all products  100 %
[27] G. P. Hao, Z. Y. Jin, Q. Sun, X. Q. Zhang, J. T. Zhang, A. H. Lu, Energy Envi-
ron. Sci. 2013, 6, 3740–3747.
[28] Y. Zhang, A. Wang, T. Zhang, Chem. Commun. 2010, 46, 862–864.
[29] K. I. Choi, M. Vannice, J. Catal. 1991, 131, 22–35.
[30] S. Zhao, H. Yue, Y. Zhao, B. Wang, Y. Geng, J. Lv, S. Wang, J. Gong, X.
Ma, J. Catal. 2013, 297, 142–150.
Acknowledgements
This project was financially supported by the National Natural
Science Foundation for Distinguished Young Scholars (No.
21225312), the National Natural Science Foundation of China
(No. 21403027), and the Fundamental Research Funds for the
Central Universities (DUT14ZD209).
[31] F. W. Chang, H. C. Yang, L. Roselin, W. Y. Kuo, Appl. Catal. A 2006, 304,
30–39.
[32] D. Zhao, J. Feng, Q. Huo, N. Melosh, G. H. Fredrickson, B. F. Chmelka,
G. D. Stucky, Science 1998, 279, 548–552.
[33] I. Kvande, D. Chen, M. Rønning, H. J. Venvik, A. Holmen, Catal. Today
2005, 100, 391–395.
[34] A. Dandekar, R. T. K. Baker, M. A. Vannice, J. Catal. 1999, 183, 131–154.
[35] D. C. Guo, J. Mi, G. P. Hao, W. Dong, G. Xiong, W. C. Li, A. H. Lu, Energy
Environ. Sci. 2013, 6, 652–659.
[36] K. H. Chuang, C. Y. Lu, M. Y. Wey, Y. N. Huang, Appl. Catal. A 2011, 397,
234–240.
Keywords: carbon · copper · dehydrogenation · mesoporous
materials · supported catalysts
[37] Z. He, H. Lin, P. He, Y. Yuan, J. Catal. 2011, 277, 54–63.
[38] D. H. Nam, J. H. Lee, N. R. Kim, Y. Y. Lee, H. W. Yeon, S. Y. Lee, Y. C. Joo,
Carbon 2015, 82, 273–281.
[39] N. Kanoun, M. P. Astier, G. M. Pajonk, Appl. Catal. 1991, 70, 225–236.
[40] N. Kanoun, M. P. Astier, G. M. Pajonk, J. Mol. Catal. 1993, 79, 217–228.
[41] M. J. L. Gines, E. Iglesia, J. Catal. 1998, 176, 155–172.
[42] V. L. Sushkevich, I. I. Ivanova, V. V. Ordomsky, E. Taarning, ChemSusChem
2014, 7, 2527–2536.
[43] S. W. Song, K. Hidajat, S. Kawi, Langmuir 2005, 21, 9568–9575.
[44] K. Fujdala, T. Tilley, J. Am. Chem. Soc. 2001, 123, 10133–10134.
[45] M. A. Natal-Santiago, J. A. Dumesic, J. Catal. 1998, 175, 252–268.
[46] G. P. Hao, F. Han, D. C. Guo, R. J. Fan, G. Xiong, W. C. Li, A. H. Lu, J. Phys.
Chem. C 2012, 116, 10303–10311.
[47] G. P. Hao, W. C. Li, D. Qian, A.-H. Lu, Adv. Mater. 2010, 22, 853–857.
[48] G. P. Hao, W.-C. Li, D. Qian, G.-H. Wang, W.-P. Zhang, T. Zhang, A.-Q.
Wang, F. Schüth, H.-J. Bongard, A.-H. Lu, J. Am. Chem. Soc. 2011, 133,
11378–11388.
[1] G. W. Huber, S. Iborra, A. Corma, Chem. Rev. 2006, 106, 4044–4098.
[2] P. Gallezot, Chem. Soc. Rev. 2012, 41, 1538–1558.
[3] J. Rass-Hansen, H. Falsig, B. Jørgensen, C. H. Christensen, J. Chem. Tech-
nol. Biotechnol. 2007, 82, 329–333.
[4] J. Mielby, J. Abildstrøm, F. Wang, T. Kasama, C. Weidenthaler, S. Kegnæs,
Angew. Chem. Int. Ed. 2014, 53, 12513–12516; Angew. Chem. Int. Ed.
2014, 126, 12721–12724.
[5] OECD-FAO Agricultural Outlook 2013, OECD Publishing, Paris, 2013.
[6] C. Angelici, B. M. Weckhuysen, P. C. Bruijnincx, ChemSusChem 2013, 6,
1595–1614.
[7] E. V. Makshina, M. Dusselier, W. Janssens, J. Degr›ve, P. A. Jacobs, B. F.
Sels, Chem. Soc. Rev. 2014, 43, 7917–7953.
[8] V. I. Sobolev, K. Koltunov, O. A. Simakova, A. R. Leino, D. Murzin, Appl.
Catal. A 2012, 433–434, 88–95.
[9] T. Takei, N. Iguchi, M. Haruta, Catal. Surv. Asia 2011, 15, 80–88.
[10] A. G. Sato, D. P. Volanti, D. M. Meira, S. Damyanova, E. Longo, J. M. C.
Bueno, J. Catal. 2013, 307, 1–17.
[11] R. Prasad, Mater. Lett. 2005, 59, 3945–3949.
[12] S.-i. Fujita, N. Iwasa, H. Tani, W. Nomura, M. Arai, N. Takezawa, React.
Kinet. Catal. Lett. 2001, 73, 367–372.
[49] A. Gervasini, S. Bennici, Appl. Catal. A 2005, 281, 199–205.
[50] E. D. Guerreiro, O. F. Gorriz, G. Larsen, L. Arrffla, Appl. Catal. A 2000, 204,
33–48.
[13] P. Liu, E. J. M. Hensen, J. Am. Chem. Soc. 2013, 135, 14032–14035.
[14] C. Caro, K. Thirunavukkarasu, M. Anilkumar, N. R. Shiju, G. Rothenberg,
Adv. Synth. Catal. 2012, 354, 1327–1336.
Received: May 5, 2015
Published online on August 6, 2015
ChemCatChem 2015, 7, 2846 – 2852
www.chemcatchem.org
2852
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim