Communications
Cu+/Cu0 value of 0.64. The Cu/AC-673 catalyst had the highest
surface Cu+ content with a molar ratio of Cu+/Cu0 =1.05. Fur-
ther increase in the calcination temperature decreased the Cu+
/Cu0 distribution in the catalyst. As shown in Figure S7 in the
Supporting Information, the intensity of Cu+–CO on the Cu/
AC-673 catalyst in the in-situ CO adsorption FTIR spectra is
clearly enhanced compared with that on the Cu/AC-non cata-
lyst, which illustrates that more Cu+ species exist on Cu/AC-
673. This result is in line with that from XAES.[8]
The unique catalytic performance of Cu/AC-673, which
showed relatively high DMO conversion and much higher se-
lectivity to MG compared with other Cu/AC-x catalysts, is at-
tributed to the balance between the Cu+/Cu0 species. Further-
more, the relatively repressed hydrogenation activity caused
by the large copper particles also plays a significant role in pre-
venting further hydrogenation of MG to EG. The Cu/AC-673
catalyst can run stably at 493 K with a DMO conversion of 83%
and yield of MG higher than 75% without any deactivation
even after 120 h time on stream (see Figure S8 in the Support-
ing Information), which indicates analogous stability with the
reported Ag/SiO2 catalyst.[3] XPS examination of the spent Cu/
AC-673 catalyst revealed that the chemical state of the copper
species did not change compared with the freshly reduced
one, confirming the stabilization effect caused by the calcina-
tion process under an N2 atmosphere (see Figure S9 in the
Supporting Information).
It is known that the strong interaction between the metal
and the supports can stabilize the Cu+ species in the reduced
catalyst. The FTIR results indicated that the metal–support in-
teractions in the catalyst increased with increasing calcination
temperature; thus, Cu/AC-673 had higher Cu+ content com-
pared with the Cu/AC-573 catalyst. However, both the heat
treatment under an N2 atmosphere and pretreatment under
H2/Ar (vol%) at 573 K could cause the reduction of the copper
species in the catalysts. With the increase in the calcination
temperature, the carbon reduction effect has a greater influ-
ence on the copper reduction and further cause the deep re-
duction of the copper species to the Cu0 state. Thus, the value
of Cu+/Cu0 decreased in the Cu/AC-773 catalyst. In addition,
Cu/CNT-673 also had the molar ratio of Cu+/Cu0 =0.92, similar
to the Cu/AC-673 ratio, which confirmed the calcination effect
on the surface Cu+/Cu0 distribution.
In summary, unlike the conventional Cu/SiO2 catalyst, the
Cu/AC-673 catalyst displayed a unique catalytic performance
with high selectivity to MG under relatively high DMO conver-
sion. The copper species in the catalyst were reduced by the
AC support during the calcination process, which also played
an important role in stabilizing the copper particles. The high
yield to MG is attributed to the moderate hydrogenation activi-
ty and high molar ratio of Cu+/Cu0. The ammonia evapora-
tion–impregnation method, which can generate homogene-
ously dispersed copper-based catalysts, could also be used to
synthesize other highly active supported catalysts.
It has been reported that hydrogenation of DMO takes place
on the copper surface and the balanced Cu+/Cu0 sites are the
main catalytic centers of the catalyst. The metallic Cu0 species
play important roles in the H2 activation and the Cu+ species
could be considered as the stabilizer of the methoxy and acyl
species, which are intermediates in DMO hydrogenation.[9] In
addition, Cu+ sites could function as electrophilic or Lewis
acidic sites to polarize the C=O bond through the electron
lone pair in oxygen; thus, improving the reactivity of the ester
group in DMO. Based on Gong’s study, the turnover frequency
(TOF) of the catalysts in the DMO hydrogenation process de-
creased with increasing Cu0/(Cu0 +Cu+) value in the range of
0.33–1.00.[10] The Cu0/(Cu0 +Cu+) values of the Cu/AC-x cata-
lysts in the present work are in the range of 0.49–0.62. Cu/AC-
673, with the lowest Cu0/(Cu0 +Cu+) distribution, displayed
the highest catalytic activity, strongly confirming the crucial
effect of the Cu+/Cu0 balance. Our results are also consistent
with Huang’s study in which the MG yield reached the maxi-
mum when Cu+/(Cu0 +Cu+) reached its summit.[11]
Cu/CNT-673, which has a similar Cu0/(Cu0 +Cu+) value, ex-
hibited higher DMO conversion and selectivity to EG compared
with Cu/AC-673. The XRD calculation and TEM images revealed
the much smaller copper particle size in the reduced Cu/CNT-
673 catalyst. It is known that smaller metal particles exhibit
higher hydrogenation activities compared with larger ones
owing to the large specific areas and low activation ener-
gies.[12] Thus, the copper particles in Cu/CNT-673 are expected
to exhibit higher catalytic activities compared with those in
Cu/AC-673. As a result, the higher hydrogenation activity of
Cu/CNT-673, which displayed higher DMO conversion, can fur-
ther convert the MG intermediate into EG.
Experimental Section
Catalyst preparation
All the carbon supports were treated with concentrated nitric acid
at 353 K for 24 h to functionalize the supports with carboxyl and
nitro groups before loading with copper species. Detailed charac-
terizations of the functionalized supports are shown in the Sup-
porting Information.
The Cu/AC-x catalysts with 20 wt% copper loading were prepared
by a facile ammonia evaporation–impregnation method. Firstly,
Cu(CH3COO)2·H2O (3.75 g) was dissolved in deionized water
(200 mL). Then, approximately 20 mL of aqueous ammonia
(25 wt%) was added to the above solution and the pH was adjust-
ed to 11.0. After that, the pretreated AC support (4.8 g) was added
into the solution. The as-obtained suspensions was stirred for 4 h
at room temperature, and then the temperature was increased to
363 K until the solvent had evaporated completely to decompose
the cuprammonia species. Finally, the catalyst precursors were cal-
cined under an N2 flow at different temperatures. The catalysts
were named Cu/AC-x where x stands for the calcination tempera-
ture. The catalyst without calcination was named Cu/AC-non. Cu/
CNT-673 catalyst with 20 wt% copper loading and CNT support
was synthesized by the same method as Cu/AC-673. All the cata-
lyst were pelletized, ground to 40–60 meshes, and went through
further hydrogen treatment at 573 K for 4 h with a ramping rate of
2 KminÀ1 from room temperature for pre-activation before the cat-
alytic tests.
ChemCatChem 2016, 8, 527 – 531
530
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim