Copper-Fiber-Structured Pd-Au-CuOx: Preparation and Catalytic Performance in the Vapor-Phase Hydrogenation of Dimethyl Oxalate to Ethylene Glycol

Article abstract of DOI:10.1002/cctc.201501415

Cu-fiber-structured ternary Pd-Au-CuO_x catalysts engineered from nano- to macro-scales have been developed for the vapor-phase dimethyl oxalate (DMO) hydrogenation to ethylene glycol (EG), with the aid of galvanic deposition of Pd and Au onto a thin-sheet microfibrous structure using 8 μm Cu fiber. Effects of Pd and Au loadings and their ratio have been investigated on the catalyst performance as well as the reaction conditions including reaction temperature and pressure, liquid weight hourly space velocity, and H₂/DMO ratio. The promising 0.1 Pd-0.5 Au-CuO_x/Cu-fiber catalyst is capable of converting 97-99 % DMO into EG product at a selectivity of 90-93 %. This catalyst is stable for at least 200 h. The Pd-Au-Cu₂O synergistically promotes the hydrogenation activity and stabilizes Cu⁺ sites to suppress deep reduction deactivation.

Full text of DOI:10.1002/cctc.201501415

DOI: 10.1002/cctc.201501415
Full Papers
Copper-Fiber-Structured Pd–Au–CuOx: Preparation and
Catalytic Performance in the Vapor-Phase Hydrogenation
of Dimethyl Oxalate to Ethylene Glycol
Lupeng Han, Li Zhang, Guofeng Zhao,* Yanfei Chen, Qiaofei Zhang, Ruijuan Chai, Ye Liu,
and Yong Lu*[a]
Cu-fiber-structured ternary Pd–Au–CuOx catalysts engineered
from nano- to macro-scales have been developed for the
vapor-phase dimethyl oxalate (DMO) hydrogenation to ethyl-
ene glycol (EG), with the aid of galvanic deposition of Pd and
Au onto a thin-sheet microfibrous structure using 8 mm Cu
fiber. Effects of Pd and Au loadings and their ratio have been
investigated on the catalyst performance as well as the reac-
tion conditions including reaction temperature and pressure,
liquid weight hourly space velocity, and H2/DMO ratio. The
promising 0.1Pd–0.5Au–CuOx/Cu-fiber catalyst is capable of
converting 97–99% DMO into EG product at a selectivity of
90–93%. This catalyst is stable for at least 200 h. The Pd–Au–
Cu2O synergistically promotes the hydrogenation activity and
stabilizes Cu+ sites to suppress deep reduction deactivation.
Introduction
Ethylene glycol (EG) is an important intermediate chemical
thanks to its wide applications in polyester manufacture, anti-
freeze compounds, and solvents.[1–3] The current commercial
production of EG mainly depends on the oxidation of petrole-
um-derived ethylene. However, with the petroleum dwindling
and the ever-increasing demand for EG, the indirect synthesis
of EG from syngas attracts extensive attention, in which syngas
can be readily derived from biomass, coalbed methane, natural
gas, and coal. This indirect process could be realized through
a two-step process consisting of the coupling of CO with ni-
trite esters to form dimethyl oxalate (DMO) and the subse-
quent hydrogenation of DMO to EG.[4,5] The first step has been
scaled up to commercial production with a capacity of
10000 tons per year in 2010,[6] and the following DMO hydro-
genation process has been considerably investigated in both
homogeneous and heterogeneous systems. Ruthenium-based
homogeneous catalysts were previously used for liquid-phase
DMO hydrogenation with a high EG yield, but greatly suffered
from the problems of corrosion and separation.[7] From an in-
dustrial point of view, the vapor-phase hydrogenation process
is more attractive because of the convenience of catalyst sepa-
ration and higher production efficiency especially for the EG
production in a bulk form. It has been well established that
Cu-based heterogeneous catalysts are highly active and selec-
tive for the vapor-phase hydrogenation of DMO to EG. Howev-
er, the easy deactivation and inadequate lifespan of Cu-based
catalysts severely hinder their practical industrialization. Gener-
ally, modification of the catalysts is believed to be an effective
method to improve the catalyst stability. CuCr catalyst was re-
ported to be highly active with a long lifetime of 1134 h, how-
ever, the toxicity of Cr greatly hampers its application.[8] Conse-
quently, research efforts have been focused on the develop-
ment of Cr-free Cu-based catalysts for this reaction. It is report-
ed that the lifetime of Cu/SiO2 can be dramatically prolonged
from only 30 h to 200 h by La2O3 modification,[2] to 300 h by
boron modification,[9] and to 600 h by ZrO2 modification[3]
under identical reaction conditions. AuCu/SBA-15 has also
been reported to be effective for the hydrogenation of DMO
to EG, with a prolonged lifetime of 240 h.[10] More recently, hy-
droxyapatite-supported Cu catalyst has been fabricated by
a facile ammonia-assisted one-pot synthesis method, which
owns prolonged lifetime to 120 h because of the stabilizing-
effect of the phosphate species on the Cu particles and Cu+
species.[11]
Although numerous promising results have been obtained,
there are many technical problems to be solved for industrial
application. Most of all, the usage of silica support is consid-
ered to be problematic because of the severe loss of the silica
from the Cu/SiO2 in the form of tetramethoxysilane if using
methanol as solvent.[12] This silica loss is considered to be a cru-
cial factor promoting copper aggregation, and likely results in
low-quality rather than polymerization-grade EG. On the other
hand, the DMO hydrogenation to EG is an exothermic reaction
(DH=À58.73 kJmolÀ1),[13] normally generating hotspots that
expedites catalyst sintering deactivation. Most current efforts
have been devoted to improving the dispersion of Cu species
or adding promoters to alleviate the agglomeration of Cu
nanoparticles (NPs).[9–11,14] In contrast, less attention has been
[a] L. Han, L. Zhang, Dr. G. Zhao, Y. Chen, Q. Zhang, R. Chai, Prof. Dr. Y. Liu,
Prof. Dr. Y. Lu
Department Shanghai Key Laboratory of Green Chemistry and Chemical
Processes
School of Chemistry and Molecular Engineering
East China Normal University
Shanghai 200062 (China)
E-mail: gfzhao@chem.ecnu.edu.cn
ylu@chem.ecnu.edu.cn
Supporting information for this article can be found under http://
dx.doi.org/10.1002/cctc.201501415.
ChemCatChem 2016, 8, 1065 – 1073
1065
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
paid to the catalyst’s heat-transfer ability, although the cata-
lysts usually need to operate in adiabatic packed-bed reactors
with high conversion, for which a good heat transfer is manda-
tory. Hence, from both academic and industrial points of view,
it is particularly urgent to develop a novel SiO2-free catalyst
with unique combination of high catalytic activity and selectiv-
ity, structural robustness, and excellent thermal conductivity.
Consequently, metal microfibrous-structured materials with
much higher heat-transfer ability than oxides have attracted in-
creasing interests as catalyst supports in strongly endothermic
and exothermic reactions.[15] We have demonstrated successful
applications of microfibrous matrixes and foams in the devel-
opment of many structured catalysts for the methanol-to-ole-
fins process,[16] vapor-phase oxidation of alcohols,[17] oxidative
coupling of methanol to methyl formate,[18] Fischer–Tropsch
synthesis to lower olefins,[19] coalbed methane deoxygena-
tion,[20] NH3 cracking,[21] syngas methanation,[22] and methanol
steam reforming.[23] In the light of the above success, we
aimed to utilize metal fibers with microfibrous structure to pre-
pare a novel SiO2-free Cu-based catalyst for DMO hydrogena-
tion. We expected that the sintering of the Cu NPs would be
relieved because of the higher thermal conductivity of the
metal fiber than that of the oxide supports and the SiO2 leach-
ing encountered for the traditional Cu/SiO2 during the DMO
hydrogenation would be avoided.[12]
Figure 1. Structural features from nano- to macro-scales for the structured
0.5Pd–2.5Au–CuOx/Cu-fiber catalyst. A) Photograph of the macroscopic
sample; B) SEM image, showing its 3D porous network structure; C) XRD
patterns of a) CuOx/Cu-fiber, b) 0.5Pd–CuOx/Cu-fiber, c) 2.5Au–CuOx/Cu-fiber,
and d) 0.5Pd–2.5Au–CuOx/Cu-fiber catalysts; D) high-magnification SEM
image of 0.5Pd–2.5Au–CuOx/Cu-fiber catalyst.
work structure with entirely open but irregular submillimeter-
scaled macropores. Satisfactorily, Pd and Au were firmly em-
bedded onto the Cu-fiber surface with the aid of galvanic dep-
osition method, and the actual loadings account for about
70% of theoretical ones for all samples (see the inductively
coupled plasma atomic-emission spectroscopy, ICP–AES, results
in the Supporting Information, Table S1). Such galvanic deposi-
tion can proceed spontaneously if wetting the Cu-fiber surface
with the aqueous solution containing the appointed amounts
of Pd and Au cations, owing to the distinct electrode potential
differences between the Cu2+/Cu0 (0.34 V; or Cu+/Cu0, 0.52 V)
and Au3+/Au0 (1.5 V; or Pd2+/Pd0, 0.95 V) pairs. XRD patterns of
the fresh catalysts are shown in Figure 1C. Only one XRD peak
centered at 36.38 could be clearly detected on the CuOx/Cu-
fiber and the 0.5Pd–CuOx/Cu-fiber catalyst, which was attribut-
ed to Cu2O (JCPDS-77-0199). Except for the Cu2O diffraction
peak, peaks at 35.48 and 38.68 attributed to CuO (JCPDS-80-
1916) were also detected on the 2.5Au–CuOx/Cu-fiber and
0.5Pd–2.5Au–CuOx/Cu-fiber catalysts. Note that no XRD peaks
of Au, Pd, or PdO phase could be detected on all samples, indi-
cating high dispersion of the Au and Pd species. Indeed, the
TEM images shown in Figure 2 and Supporting Information
Figure S1 also confirmed the highly dispersed Au and Pd parti-
cles with sizes of approximately 4 nm over all samples. In addi-
tion, high magnification SEM image of 0.5Pd–2.5Au–CuOx/Cu
fiber revealed the large CuOx crystalline grains (Figure 1D). The
above results indicated that our novel galvanic deposition
method was working effectively and efficiently to create such
structured 0.5Pd–CuOx/Cu-fiber, 2.5Au–CuOx/Cu-fiber, and
0.5Pd–2.5Au–CuOx/Cu-fiber catalysts, which were expected to
be competent in DMO hydrogenation reaction.
Herein we report on a high-performance monolithic struc-
tured Pd–Au–CuOx/Cu-fiber catalyst by galvanically depositing
Pd and Au nanoparticles (NPs) onto the Cu-fiber surfaces. The
promising Pd–Au–CuOx/Cu-fiber catalyst with 0.1 wt% Pd and
0.5 wt% Au could deliver DMO conversion of 97–99% with EG
selectivity of 90–93% and could assure activity/selectivity
maintenance for at least 200 h at 2708C and 2.5 MPa, using
a liquid weight hourly space velocity (LWHSV) of 5.3 hÀ1 and
a H2/DMO molar ratio of 180. The synergistic effect of the ter-
nary Pd–Au–CuOx composite is tentatively discussed. Note that
some parts of results concerning this work were previously re-
ported as a short communication,[24] which focused on the
novel microstructured catalyst system rather than the effects
of Pd and Au loadings on the catalyst performance as well as
the reaction conditions including reaction temperature/pres-
sure, LWHSV, and H2/DMO ratio.
Results and Discussion
Pd and/or Au galvanic deposition, structure, and morpholo-
gy of the catalyst
Initially, the monolithic sinter-locked microfibrous-structured
8 mm Cu fibers were prepared by regular wet layup papermak-
ing with a subsequent sintering process,[25] which were then
tailored into desired circular chips (8 mm diameter, 2 mm
thickness). Subsequently, 2.5Au–CuOx/Cu-fiber, 0.5Pd–CuOx/
Cu-fiber, and 0.5Pd–2.5Au–CuOx/Cu-fiber catalysts were pre-
pared by the galvanic deposition method following by drying
at 1008C overnight and calcination at 3008C for 2 h in air. As
shown in Figure 1A,B, our representative circular chip-like
0.5Pd–2.5Au–CuOx/Cu-fiber sample has a 3D microfibrous net-
Catalytic performance
The as-prepared 0.5Pd–CuOx/Cu-fiber, 2.5Au–CuOx/Cu-fiber,
0.5Pd–2.5Au–CuOx/Cu-fiber, and CuOx/Cu-fiber catalysts were
ChemCatChem 2016, 8, 1065 – 1073
www.chemcatchem.org
1066
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Moreover, Wang and co-workers have reported that Au-modi-
fied 6Cu1.9Au/SBA-15 catalyst exhibits remarkable enhanced
catalytic performance for the hydrogenation of DMO to EG
compared with 6Cu/SBA-15, as indicated by the increased
turnover frequency (TOF from 23 to 121 hÀ1.[10] Excitingly, the
0.1Pd–0.5Au–CuOx/Cu-fiber catalyst with co-introduction of Pd
and Au in this study could further enhance the catalytic per-
formance, delivering the highest DMO conversion of 99% and
the optimal selectivity of 91% toward EG. Clearly, the above re-
sults suggested the existence of a synergistic promotion effect
of Pd and Au on the catalyst activity, which will be further dis-
cussed below.
Figure 2. A) TEM image and B) particle size distribution of 0.5Pd-2.5Au–
CuOx/Cu-fiber catalyst.
tested in the DMO hydrogenation reaction at 2708C and
2.5 MPa, using a LWHSV of 5.3 hÀ1 and H2/DMO molar ratio of
180, with the results as shown in Table 1. The reference catalyst
of CuOx/Cu-fiber, with Cu2O phase formation (Figure 1C) that
provides the crucial catalytic active sites of Cu+ for the hydro-
genation of DMO,[3,9,26–29] showed mere DMO conversion of
40% and poor EG selectivity of only 1%. If Pd or Au was intro-
duced into CuOx/Cu-fiber, interestingly, either Pd or Au dis-
played promotion effect, and as a result, the catalytic hydroge-
nation of DMO was significantly increased. As-obtained 0.5Pd–
CuOx/Cu-fiber and 2.5Au–CuOx/Cu-fiber were capable of con-
verting DMO at conversions of 98% and 94% into EG with se-
lectivities of 87% and 60%, respectively. Analogously, it has
been reported that the assistant role of Pd in dissociating mo-
lecular H2 contributes to the enhancement of the hydrogena-
tion performance of Cu-based catalysts in numerous reactions,
such as methanol synthesis and alkynes hydrogenation.[30,31]
Effects of Pd/Au ratio and Pd (Au) loadings
It has been reported that the composition of Pd–Au bimetallic
catalysts has a significant effect on their catalytic performan-
ces, such as Pd (Au) loadings and Pd/Au ratio.[32] Herein, the
effect of Pd/Au weight ratio on the Pd–Au–CuOx/Cu-fiber cata-
lysts for the DMO hydrogenation was investigated at a fixed
Pd loading of 0.5 wt% and a fixed Au loading of 0.5 wt%, with
the results displayed in Figure 3A,B. The actual loadings of Au
and Pd measured by ICP–AES are listed in Table S1. As shown
in Figure 3A, at a fixed Pd loading of 0.5 wt%, the DMO con-
version and EG selectivity slightly increased and 99% and 91%,
respectively, along with the increase of the Au loading to a Pd/
Au weight ratio up to 1:1 (corresponding to 0.5 wt% Pd and
0.5 wt% Au), but it showed a gradual decline with further in-
creasing the Pd/Au weight ratio. Similar evolution behavior of
the conversion/selectivity against the Pd/Au weight ratio was
observed also for keeping the Au loading constant at 0.5 wt%.
As shown in Figure 3B, the 0.1Pd–0.5Au–CuOx/Cu-fiber cata-
lyst with the Pd/Au weight ratio of 1:5 delivered the highest
activity and selectivity (99% DMO conversion and 93% EG se-
lectivity). Consequently, the Pd/Au weight ratio of 1:5 was con-
sidered to be optimal for our Pd–Au–CuOx/Cu-fiber catalyst for
the DMO hydrogenation and the 0.1Pd–0.5Au–CuOx/Cu fiber
was the most promising catalyst with combination of high-per-
formance and acceptably low precious metal usage. At a fixed
Pd/Au weight ratio of 1:5, Pd–Au–CuOx/Cu-fiber catalysts with
further lowering the loadings of Au and Pd were prepared and
tested in the DMO hydrogenation, but no conversion and se-
Table 1. DMO hydrogenation performance over the various catalysts.[a]
Catalyst
DMO conversion
[%]
Selectivity [%]
MG
EG
EtOH
CuOx/Cu-fiber
40
98
94
99
1
87
60
91
96
10
39
5
3
3
1
4
0.5Pd–CuOx/Cu-fiber
2.5Au–CuOx/Cu-fiber
0.5Pd–CuOx/Cu-fiber
[a] Reaction conditions: T=2708C, LWHSV=5.3 hÀ1
methanol, P(H2)=2.5 MPa, n(H2)/n(DMO)=180. MG=methyl glycolate.
, 13 wt% DMO in
Figure 3. Effect of the Pd/Au ratio on the DMO hydrogenation performance of Pd–Au–CuOx/Cu-fiber catalysts with A) fixed Pd loading of 0.5 wt% and
B) fixed Au loading of 0.5 wt%; C) Effects of Pd and Au loadings on the DMO hydrogenation performance of Pd–Au–CuOx/Cu-fiber catalysts with fixed Pd/Au
weight ratio of 1:5. Reaction conditions: T=2708C, LWHSV=5.3 hÀ1, 13 wt% DMO in methanol, P(H2)=2.5 MPa, n(H2)/n(DMO)=180. MG=methyl glycolate.
ChemCatChem 2016, 8, 1065 – 1073
www.chemcatchem.org
1067
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
lectivity comparable to those for the 0.1Pd–0.5Au–CuOx/Cu-
fiber catalyst could be yielded (Figure 3C). For example, the
0.02Pd–0.1Au–CuOx/Cu-fiber with 0.02 wt% Pd and 0.1 wt%
Au loadings showed a DMO conversion of 89% with a relatively
low EG selectivity of only 43% under identical reaction condi-
tions.
sion increased from 45% to 99% and the EG selectivity in-
creased from 12% to 93%. With further increasing temperature
to 2808C, the DMO conversion remained at 99%, however, the
EG selectivity decreased to 91% while the ethanol selectivity
increased from 3% at 2708C to 6%. Clearly, relatively high re-
action temperature was favorable for our Au–Pd–CuOx/Cu-fiber
catalyst for hydrogenation of DMO into EG. However, excessive
hydrogenation would proceed to form ethanol at too high
temperature thereby leading to a decline of EG selectivity.
In Figure 4B, the DMO conversion and product selectivities
for the DMO hydrogenation using 0.1Pd-0.5Au–CuOx/Cu-fiber
catalyst are plotted against the reaction pressure at 2708C
using a H2/DMO ratio of 180 and a LWHSV of 5.3 hÀ1. Clearly,
catalytic activity and products distribution were sensitive to re-
action pressure. DMO conversion and EG selectivity showed
volcano evolution behavior with increasing the reaction pres-
sure from 0.5 to 3.0 MPa. The DMO conversion increased from
24% to 99% and the EG selectivity increased from 23% to
93% with increasing the H2 pressure from 0.5 to 2.5 MPa,
which is a common promotion effect in hydrogenation reac-
tions.[33] If further increasing the H2 pressure up to 3.0 MPa, the
DMO conversion still remained unchanged at 99%, however,
the EG selectivity fell down to 90% associated with an increase
of ethanol selectivity from 3% (2.5 MPa) to 6% as a result of
the excessive hydrogenation.
Effects of reaction conditions
It is known that DMO hydrogenation comprises several contin-
uous reactions (Scheme 1), including DMO hydrogenation to
methyl glycolate (MG), MG hydrogenation to EG, and deep hy-
drogenation of EG to ethanol. The catalyst performance for the
DMO hydrogenation is distinctly influenced by the reaction
conditions including temperature, pressure, LWHSV, and molar
ratio of H2/DMO. Thus the effects of the reaction conditions on
the catalytic performance of our 0.1Pd–0.5Au–CuOx/Cu-fiber
catalyst for the DMO hydrogenation were investigated, with
the results shown in Figure 4A–D.
In Figure 4A, the temperature-dependent conversion and se-
lectivities for the DMO hydrogenation using 0.1Pd–0.5Au–
CuOx/Cu-fiber catalyst are displayed, at 2.5 MPa using a H2/
DMO ratio of 180 and LWHSV of 5.3 hÀ1. With the increase of
the reaction temperature from 210 to 2708C, the DMO conver-
In Figure 4C, the DMO conversion and EG selectivity for the
DMO hydrogenation using 0.1Pd–0.5Au–CuOx/Cu-fiber catalyst
is displayed as a function of the LWHSV at 2708C and 2.5 MPa
using a H2/DMO of 180. Clearly, the DMO conversion was re-
tained at 99% or above with the increase of the LWHSV up to
5.3 hÀ1 and gradually declined from 99% to 85% with the
LWHSV from 5.3 to 14.2 hÀ1 as a result of inadequate residen-
tial time that was unfavorable to the DMO hydrogenation. Re-
garding the EG selectivity, it exhibited a volcano-shaped distri-
bution within the LWHSV range studied. The EG selectivity
mounted from 38% to 93% with the LWHSV from 2.7 to
5.3 hÀ1 and then continuously declined from the climax to
52% with further increase of the LWHSV up to 14.2 hÀ1. It is
logical that a proper residential time might exist for DMO hy-
drogenation to EG, because much longer residential time led
to excessive hydrogenation of EG into ethanol while quite
shorter residential time was not enough for intermediate prod-
uct MG hydrogenation to EG.
Scheme 1. Hydrogenation of DMO to MG, EG, and ethanol.
The plot in Figure 4D shows a volcano-shaped curve for the
DMO conversion and EG selectivity against the H2/DMO molar
ratio ranging from 60 to 240. At 2708C, 2.5 MPa and 5.3 hÀ1,
the DMO conversion and EG selectivity both reached a climax
at the H2/DMO molar ratio of 180. The use of low H2/DMO
molar ratio of <180 made the 0.1Pd–0.5Au–CuOx/Cu-fiber cat-
alyst not active and selective enough, while the high H2/DMO
molar ratio of 240 impaired the EG selectivity likely because
the MG residence time was not long enough to be hydrogen-
ated into EG.
Figure 4. Effects of certain reaction parameters on DMO hydrogenation per-
formance of 0.1Pd–0.5Au–CuOx/Cu-fiber: A) reaction temperature, at reac-
tion conditions: LWHSV=5.3 hÀ1, 13 wt% DMO in methanol, P=2.5 MPa,
n(H2):n(DMO)=180; B) pressure, at reaction conditions: 2708C,
On the basis of the above results, the optimal reaction con-
ditions were: reaction temperature of 2708C, H2 pressure of
2.5 MPa, total LWHSV of 5.3 hÀ1 and H2/DMO molar ratio of
180, for the DMO hydrogenation over the 0.1Pd–0.5Au–CuOx/
LWHSV=5.3 hÀ1, 13 wt% DMO in methanol, n(H2):n(DMO)=180; C) LWHSV,
at reaction conditions: 2708C, 13 wt% DMO in methanol, P=2.5 MPa,
n(H2):n(DMO)=180; (D) molar ratio of H2/DMO, at reaction conditions:
2708C, LWHSV=5.3 hÀ1, 13 wt% DMO in methanol, P=2.5 MPa.
ChemCatChem 2016, 8, 1065 – 1073
www.chemcatchem.org
1068
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Cu-fiber catalyst using a feed of 13 wt% DMO in methanol. As
a result, a high DMO conversion of>99% was achievable with
a high EG selectivity of 93%.
vidually promoted ones. To reveal the synergistic promotion
nature of Pd–Au, TOF (the number of converted DMO mole-
cules per surface Cu+ site per hour) measurements on the
0.1Pd–0.5Au–CuOx/Cu-fiber, 0.1Pd– CuOx/Cu-fiber, 0.5Au–
CuOx/Cu-fiber as well as CuOx/Cu-fiber were firstly conducted
at 2108C and at controlled DMO conversion below 30%, with
the results shown in Table 2. The CuOx/Cu-fiber delivered a low
Stability
It is well-known that the durability of Cu-based catalysts is the
tightest bottleneck that restricts its industrial applications. A
long-term test of the 0.1Pd–0.5Au–CuOx/Cu-fiber catalyst was
performed under the optimal reaction conditions, with the re-
sults as shown in Figure 5. For comparison, stability testing of
Table 2. DMO hydrogenation conversion and the corresponding TOF
over the various catalysts.
Catalyst
DMO conversion
[%][a]
Cu+ sites
[molgcat
TOF
[hÀ1
]
À1 [b]
[c]
]
0.1Pd–0.5Au–CuOx/Cu-fiber
CuOx/Cu-fiber
0.1Pd–CuOx/Cu-fiber
0.5Au–CuOx/Cu-fiber
32.0
4.2
6.4
1.80”10À6
8.46”10À6
2.73”10À6
3.60”10À6
612
17
80
18.3
174
[a] To obtain the intrinsic activity of all the catalysts, DMO conversion was
kept below 30% under the following reaction conditions: catalyst 0.5 g,
2108C, LWHSV of total liquid feed of 9 hÀ1 with 13 wt% DMO dissolved in
methanol, H2 pressure of 2.5 MPa, molar ratio of H2 to DMO of 180.
[b] The calculation of Cu+ site is exemplified for 0.1Pd–0.5Au–CuOx/Cu-
fiber catalyst: Cu+ sites=surface area of 0.1Pd–0.5Au–CuOx/Cu-fiber
(0.48 m2gÀ1)/area occupied by one Cu+ site (0.444 nm2, see the equation
in TOF calculations)=1.08”1018 =1.80”10À6 mol/g. [c] The TOF was cal-
culated as TOF=converting DMO rate/Cu+ sites, for the detailed calcula-
tion method see the Experimental Section.
Figure 5. Conversion and product selectivity for the 200 h test of DMO hy-
drogenation over 0.1Pd–0.5Au–CuOx/Cu-fiber. Reaction conditions:
T=2708C, LWHSV=5.3 hÀ1, 13 wt% DMO in methanol, P(H2)=2.5 MPa,
n(H2)/n(DMO)=180.
TOF of 17 hÀ1, but the TOF was increased to 80 or 174 hÀ1 with
Pd or Au adding (i.e., 0.1Pd–CuOx/Cu-fiber, or 0.5Au–CuOx/Cu-
fiber). This observation clearly confirmed the activity-promot-
ing effect of Pd or Au alone on the CuOx/Cu-fiber. Most nota-
bly, co-adding Pd and Au provided an activity-promoting
effect much higher than the Pd or Au adding solely, and the
corresponding 0.1Pd-0.5 Au–CuOx/Cu-fiber catalyst delivered
a high TOF of 612 hÀ1. The above results clearly revealed the
existence of synergistic interaction among ternary Au–Pd–CuOx
composite, which was responsible for the significantly en-
hanced intrinsic reactivity.
the CuOx/Cu-fiber, 0.1Pd–CuOx/Cu-fiber and 0.5Au–CuOx/Cu-
fiber catalysts were also performed, with the results shown in
Figure S2A–C. Clearly, the CuOx/Cu-fiber catalyst showed
a poor stability with a DMO conversion decline from 40% to
12% within only 27 h (Figure S2A). Whereas the 0.1Pd–CuOx/
Cu-fiber catalyst delivered a high initial DMO conversion of
98% and EG selectivity of 90%, the poor stability was its tech-
nique shortcoming, showing a fast decline of DMO conversion
from 98% to 42% within 45 h (Figure S2B). Notably, the
0.5Au–CuOx/Cu-fiber catalyst exhibited significantly enhanced
stability with a good maintenance of the DMO conversion of
90–92% throughout entire 80 h test but suffered from a low
EG selectivity of only 13–58% (Figure S2C). Very interestingly,
the 0.1Pd-0.5Au–CuOx/Cu-fiber catalyst provided a unique
combination of high activity/selectivity of the Pd–CuOx/Cu
fiber and promising stability for the Au–CuOx/Cu fiber; such
catalyst not only delivered a very high initial conversion of
99% but also significantly prolonged the catalyst lifetime at
such high conversion to at least 200 h without any signs of de-
activation (Figure 5). A possible explanation is that synergistic
interaction among ternary Au–Pd–CuOx composite was gener-
ated, which was paramount to the activity-promoting and sta-
bility-enhancing effects.
H2 temperature-programmed reduction (TPR), a very useful
tool for revealing metal–metal and metal–support strong inter-
actions, was used to study the interaction of the ternary Au–
Pd–CuOx composite as well as the binary Pd–CuOx and Au–
CuOx, with the results shown in Figure 6. As shown in Fig-
ure 6A, the CuOx/Cu-fiber catalyst provided a single-peak H2
TPR profile with a peak temperature of 3658C, assigned to
Cu2O (by XRD in Figure 1C) reduction. The 0.1Pd–CuOx/Cu-
fiber catalyst also provided a single-peak H2-TPR profile but
the peak temperature shifted down to 2408C. This observation
indicated that the reduction of Cu2O in this catalyst (by XRD in
Figure 1C) was catalyzed by Pd, probably with the H2 spillover
of Pd that promoted the Cu2O reduction. Logically, the pro-
moted activity of 0.1Pd–CuOx/Cu-fiber was attributed to the
assistant dissociation of H2 on Pd sites with spillover of active
H species onto the neighboring Cu+ sites at which DMO hy-
drogenation took place. However, such H2-assistant role of Pd
could simultaneously facilitate the reduction of Cu+ species,
the crucial catalytic active sites for the hydrogenation of
Insight into the synergistic interaction of Au–Pd–CuOx
Pd–Au copromoted CuOx/Cu-fiber catalyst performed better in
the catalytic hydrogenation of DMO to EG than Pd or Au indi-
ChemCatChem 2016, 8, 1065 – 1073
www.chemcatchem.org
1069
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Figure 6. TPR profiles of A) the different catalysts and B) the Pd–Au–CuOx/Cu-fiber catalysts with Pd loading of 0.5 wt% but varied Au/Pd weight ratio.
DMO,[3,9,26–29] thereby leading to a very short lifetime (Fig-
ure S2B). Interestingly, for the 0.5Au–CuOx/Cu-fiber catalyst the
reduction peak of Cu2O–CuO (by XRD in Figure 1C) became
broadened and asymmetric while the reduction temperature
was shifted to 4958C, which is much higher than that for
either the 0.1Pd–CuOx/Cu-fiber or the CuOx/Cu-fiber. The re-
tarded reduction of cationic Cu in the 0.5Au–CuOx/Cu-fiber
could be attributed to the stabilizing-effect of Au on cationic
Cu.[10,34] Accordingly, Au adding was believed to be helpful for
tuning the Cu+/Cu0 proportion to a proper level and assuring
its maintenance, which not only improved the catalyst activi-
Figure 7. Cu2p XPS spectra of the samples after different reaction times.
ty[10] but also enhanced the catalyst stability (Figure S2C) in
a) CuOx/Cu-fiber after 10 h; b) CuOx/Cu-fiber after 30 h; c) 0.1Pd–CuOx/Cu-
fiber after 10 h; d) 0.1 Pd–CuOx/Cu-fiber after 50 h; e) 0.5Au–CuOx/Cu-fiber
comparison with the CuOx/Cu-fiber base catalyst. For the
0.1Pd–0.5Au–CuOx/Cu-fiber, interestingly, Pd-catalyzed reduc-
tion of the Cu2O–CuO (by XRD in Figure 1C) was not observed
along with the Pd adding into the 0.5Au–CuOx/Cu-fiber, solidly
evidenced by broad asymmetric hydrogen consumption peak
with a high reduction temperature of 5358C. Clearly, adding
Pd would not counteract the stabilizing effect of Au on the
cationic Cu, and consequently, the 0.1Pd–0.5Au–CuOx/Cu-fiber
showed good stability for at least 200 h (Figure 5). On the
other hand, the 0.1Pd–0.5Au–CuOx/Cu-fiber delivered the EG
selectivity comparable to the initial EG selectivity (at 5 h) for
the 0.1Pd–CuOx/Cu-fiber but higher than that (at 5 h) for the
0.5Au–CuOx/Cu-fiber (Figure 5 and Figure S2B and C), indicat-
ing that such catalyst effectively and efficiently integrated high
activity/selectivity of the 0.1Pd–CuOx/Cu-fiber with the good
stability of the 0.5Au–CuOx/Cu-fiber.
after 10 h; f) 0.5Au–CuOx/Cu-fiber after 60 h; g) 0.1Pd–0.5Au–CuOx/Cu-fiber
after 10 h; h) 0.1Pd-0.5Au–CuOx/Cu-fiber after 60 h.
matter if in active or deactivated state, show XPS peaks at
binding energies of 932.7 and 952.5 eV, which are the charac-
teristic peaks of Cu2p3/2 and Cu2p1/2 for Cu+ or Cu0 species. In
addition, no shakeup satellite peaks of Cu2+ at 942.3 eV could
be detected, which indicates that most of the Cu2+ species
were reduced to Cu0 and/or Cu+ under the reaction condition.
As the XPS Cu2p3/2 peaks cannot differentiate between Cu+
and Cu0, Cu LMM Auger spectra were recorded to distinguish
Cu+ and Cu0 by their different binding energies in the Cu LMM
spectra: Cu+ at 570.0 eV and Cu0 at 567.7 eV.[35–37] As shown in
Figure 8A,B, asymmetric Auger binding-energy peaks could be
detected over all samples, demonstrating the simultaneous ex-
istence of Cu0 and Cu+ on the surface of the catalysts, and the
molar percentage of Cu+ to the total Cu0 +Cu+ content after
deconvolution are listed in Table 3. The Cu+ fraction of the
CuOx/Cu-fiber catalyst decreased from 74.3% (10 h running on
stream) to 43.0% (30 h running on stream) with DMO conver-
sion decreasing from 40% to 12% (Figure 8A, Table 3, and Fig-
ure S2A). For the 0.1Pd–CuOx/Cu-fiber catalyst, the Cu+ frac-
tion decreased from 54.8% (10 h running on stream) to 34.0%
(50 h running on stream) and DMO conversion decreased from
98% to 42% (Figure 8A, Table 3, and Figure S2B). Combined
with the results of H2 TPR, it is further confirmed that the intro-
duction of Pd could not only promote the catalytic activity
(TOF=80 hÀ1 for 0.1Pd–CuOx/Cu-fiber vs. 17 hÀ1 for CuOx/Cu-
fiber, in Table 2) but also catalyze the reduction of Cu+ species,
Furthermore, the H2 TPR measurements for the Pd–Au–
CuOx/Cu-fiber catalysts with more Pd loading of 0.5 wt% at
varying Au/Pd weight ratio were also performed, with the re-
sults as shown in Figure 6B. Clearly, the reduction temperature
of the cationic Cu could be shifted to 4608C even adding
0.1 wt% Au into the 0.5Pd–CuOx/Cu-fiber, and progressively to
5758C along with increasing the Au loading up to 2.5 wt%
(corresponding to Au/Pd ratio of 5:1). Such results further con-
firmed the strong stabilizing effect of Au on the cationic Cu in
the catalyst.
To further reveal the nature of synergistic interaction among
ternary Au–Pd–CuOx composite, the valence states of surface
Cu species over every sample at different stages during the
stability testing were characterized by X-ray photoelectron
spectroscopy (XPS). As shown in Figure 7, all the samples, no
ChemCatChem 2016, 8, 1065 – 1073
www.chemcatchem.org
1070
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Figure 8. Cu LMM Auger spectra of the samples after different reaction time. A1) CuOx/Cu-fiber after 10 h; A2) CuOx/Cu-fiber after 30 h; A3) 0.1Pd–CuOx/Cu-
fiber after 10 h; A4) 0.1Pd–CuOx/Cu-fiber after 50 h; B1) 0.5Au–CuOx/Cu-fiber after 10 h; B2) 0.5Au–CuOx/Cu-fiber after 60 h; B3) 0.1Pd–0.5Au–CuOx/Cu-fiber
after 10 h; B4) 0.1Pd–0.5Au–CuOx/Cu-fiber after 60 h.
Table 3. Molar ratio of Cu+ to Cu0 +Cu+ on the catalyst surfaces after
DMO hydrogenation.
Cu+/(Cu0 +Cu+) [%]
Catalyst
Stage A[a]
Stage B[b]
Stage C[c]
CuOx/Cu-fiber
74.3
54.8
64.0
62.7
43.0
34.0
n.d.
n.d.
n.d.[d]
n.d.
66.6
62.5
0.1Pd–CuOx/Cu-fiber
0.5Au–CuOx/Cu-fiber
0.1Pd-0.5Au–CuOx/Cu-fiber
[a] All the catalysts after the reaction for 10 h. [b] The CuOx/Cu-fiber after
the reaction for 30 h and the deactivated 0.1Pd–CuOx/Cu-fiber after the
reaction for 50 h. [c] The catalysts after the reaction for 60 h. [d] Not de-
tected (n.d.). Reaction conditions: T=2708C, LWHSV=5.3 hÀ1, 13 wt%
DMO in methanol, P(H2)=2.5 MPa, n(H2)/n(DMO)=180.
Figure 9. A,B) HAADF–STEM images and C,D) the corresponding elemental
maps of 0.1Pd–0.5Au–CuOx/Cu-fiber after reaction for 10 h.
leading to a short lifetime. However, the Cu+ fraction of the
0.5Au–CuOx/Cu-fiber catalyst remained at 64–67% in 60 h run-
ning on-stream with a steady DMO conversion of 90–92% (Fig-
ure 8B, Table 3, and Figure S2C), which further evidenced that
Au not only provided an activity-promoting effect (TOF=
174 hÀ1 for 0.5Au–CuOx/Cu-fiber vs. 17 hÀ1 for CuOx/Cu-fiber),
but also a stabilizing-effect of Au on the cationic Cu. Interest-
ingly, the Cu+ fraction of the 0.1Pd-0.5Au–CuOx/Cu-fiber cata-
lyst also remained at approximately 62–63% after the 60 h run-
ning (Figure 8B and Table 3). Clearly, with the addition of Pd
into 0.5Au–CuOx/Cu-fiber, the catalyst activity (TOF=612 hÀ1
for 0.1Pd-0.5Au–CuOx/Cu-fiber vs. 174 hÀ1 for 0.5Au–CuOx/Cu-
fiber, in Table 2) is further significantly improved but the cata-
lyst stability is not impaired. These results convincingly indicat-
ed the specific synergistic effect between Pd and Au that sig-
nificantly improved the activity of Cu+ but not attenuated the
ability of Au for stabilizing Cu+ sites.
mogeneously with Au in the mapped part (Figure 9C,D), indi-
cating the formation of Au–Pd alloy. It has been reported that
strong electron interaction could occur over Au–Pd alloy as
well as in the Au@Pd core–shell structures.[38–40] Therefore, the
electron interaction for Au and Pd in the Au–Pd–CuOx compo-
site was further investigated by XPS. The spectra in Figure 10
clearly reveal the evolution of Pd3d3/2 and Au4f7/2 binding en-
ergies for 0.1Pd–0.5Au/Cu-fiber compared to 0.1Pd–CuOx/Cu-
fiber and 0.5Au–CuOx/Cu-fiber (10 h running on stream): the
metallic Pd3d3/2 binding energy showed a clear upshifting
from 335.0 eV for 0.1Pd–CuOx/Cu-fiber to 335.6 eV for 0.1Pd–
0.5Au–CuOx/Cu-fiber (Figure 10A), while the metallic Au4f7/2
binding energies showed a slight downshifting from 84.2 eV
for 0.5Au–CuOx/Cu-fiber to 84.0 eV for 0.1Pd–0.5Au–CuOx/Cu-
fiber, likely resulting from the low Pd/Au weight ratio of only
1:5 (Figure 10B). The results indeed confirmed the strong Pd–
Au electron interaction, in which Pd species for the 0.1Pd–
0.5Au–CuOx/Cu-fiber was electron-deficient to some extent be-
cause of the Au electron-drawing effect. Such electron interac-
tion between Pd and Au contributed to the improvement of
catalytic performance, which has been evidenced for another
important reaction of the direct synthesis of H2O2 from H2 and
O2.[38–40] As previous reported, much higher H2 conversion was
obtained on an Au@Pd core–shell composite and Au–Pd alloy
than either Pd or Au because of the improved ability for H2 ac-
tivation to generate more active hydrogen species.[38–40] It thus
One question is not answered yet: why the Pd-catalyzed re-
duction of the cationic Cu was suppressed in the presence of
Au? We believe that the existence of a Pd–Au electron interac-
tion is paramount to the enhanced activity and prolonged sta-
bility. To check this speculation, we firstly probed the nano-
structure of Au and Pd in the Au–Pd–CuOx composite. The
0.1Pd-0.5Au–CuOx/Cu-fiber catalyst after 10 h running on
stream was characterized by the elemental maps of high-angle
annular dark-field scanning transmission electron microscopy
(HAADF–STEM). As shown in Figure 9, Pd was distributed ho-
ChemCatChem 2016, 8, 1065 – 1073
www.chemcatchem.org
1071
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Figure 10. A) XPS spectra of Pd3d of 0.1Pd–CuOx/Cu-fiber and 0.1Pd–0.5Au–CuOx/Cu-fiber after 10 h run; (B) XPS spectra of Au4f of 0.5Au–CuOx/Cu-fiber
and 0.1Pd–0.5Au–CuOx/Cu-fiber after 10 h run.
to obtain the catalyst products, referred as to nPd–CuOx/Cu-fiber,
mAu–CuOx/Cu-fiber, and nPd–mAu–CuOx/Cu-fiber (n and m are the
theoretical loadings of Pd and Au, respectively). For reference, the
pure Cu-fiber substrate directly calcined in air at 3008C for 2 h was
also prepared, denoted as CuOx/Cu-fiber.
can be inferred that the significant improvement of the DMO
hydrogenation activity for the 0.1Pd–0.5Au–CuOx/Cu-fiber cat-
alyst benefited from the Pd–Au electron interaction in Au–Pd
alloy, in analogy with Au@Pd core–shell composite and Au–Pd
alloy for direct H2O2 synthesis.[38–40] Furthermore, such electron
interaction among Au–Pd alloy also played a significant role
possibly in suppressing Pd-catalyzed reduction of the cationic
Cu species thereby leading to the improved catalyst stability.
Catalyst characterization
The catalysts were characterized by SEM (Hitachi S-4800, Japan),
XRD (Rigaku Uitima IV with CuKa radiation at 35 kV and 25 mA,
Japan), ICP–AES (Thermo Scientific iCAP 6300, USA). Specific sur-
face area was calculated using the standard BET theory based on
the N2 adsorption isotherm obtained on a Quantachrome appara-
tus (Autosorb-3B, USA) at À1968C. H2 TPR was performed on
a Quantachrome chemisorption apparatus (ChemBET 3000, USA)
with a thermal conductivity detector and an online Mass Spec-
trometer (Proline Dycor, AMETEK Process Instrument, USA). The
XPS and Auger electron spectroscopy were recorded on an Escalab
250xi spectrometer, using a standard AlKa X-ray source (300 W) and
an analyzer pass energy of 20 eV. All binding energies were refer-
enced to the adventitious C1s line at 284.6 eV.
Conclusions
We have established a promising Cu-fiber-structured ternary
Pd–Au–CuOx catalysts engineered from nano- to macro-scales
for the vapor-phase dimethyl oxalate (DMO) hydrogenation to
ethylene glycol, with the aid of galvanic deposition of Pd and
Au onto a thin-sheet microfibrous structure using 8 mm Cu
fiber. Effects of Pd and Au loadings and their ratio have been
investigated on the catalyst performance as well as the reac-
tion conditions including reaction temperature and pressure,
liquid weight hourly space velocity, and H2/DMO ratio. The op-
timized 0.1Pd–0.5Au–CuOx/Cu-fiber catalyst is capable of 97–
99% DMO conversion with ethylene glycol selectivity of 90–
93% and stable for at least 200 h. The Pd–Au–Cu2O synergisti-
cally promotes the hydrogenation activity and stabilizes Cu+
sites to suppress deep reduction deactivation. It is believed
that our finding opens up a new path to develop a highly effi-
cient microstructured catalyst for DMO hydrogenation, which
exhibits great potential for industrial applications.
Reactivity tests
The hydrogenation of DMO to EG was performed in a fixed-bed
stainless-steel reactor with 8 mm inner diameter. Firstly the catalyst
(0.5 g) was loaded into the reactor and heated to 2708C, and then
pure hydrogen was introduced to the reactor until the pressure
was raised to 2.5 MPa, and DMO methanol solution (13 wt%) was
simultaneously introduced to the reactor. The DMO methanol solu-
tion was pumped with a high-pressure advection pump and the
liquid products were collected and analyzed by using a Shimadzu
2014C gas chromatography-flame ionization detector with a HP-IN-
NOWax column. The DMO conversion and product selectivity were
calculated as follows:
Experimental Section
Catalyst preparation
The thin-sheet sinter-locked microfibrous structure consisting of
15 vol% 8 mm Cu fibers and 85 vol% voidage, with entirely open
3D porous network, were prepared through wet layup paper
making/sintering processes[25] to use as substrates. The Pd, Au, and
Pd–Au were then galvanically deposited onto the Cu-fiber surface
of the microfibrous structure by incipiently wetting the substrates
using an aqueous solution containing the appointed amounts of
palladium acetate, chloroauric acid, and both of them (Sinopharm
Chemical Reagent Co., Ltd., China) at RT. The as-prepared samples
were dried overnight at 1008C and calcined in air at 3008C for 2 h
DMO conversion ð%Þ ¼ ð1ÀADMO
Si ð%Þ ¼ ðAifi=SAifiÞ Â 100
f
DMO=SAifiÞ Â 100
ð1Þ
ð2Þ
where Ai is the peak area of the individual component i and fi is
the molar correction factor of the individual component i. The
turnover frequency (TOF) of all the catalysts was defined as the
number of DMO converted per Cu+ site per time, and was calcu-
lated based on the following equation:
ChemCatChem 2016, 8, 1065 – 1073
www.chemcatchem.org
1072
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
DMO converting rate ðmol gcatÀ1 hÀ1
Þ
[11] C. Wen, Y. Cui, X. Chen, B. Zong, W. L. Dai, Appl. Catal. B 2015, 162,
483–493.
[12] C. Wen, Y. Cui, W. L. Dai, S. Xie, K. N. Fan, Chem. Commun. 2013, 49,
5195–5197.
[13] H. Yue, Y. Zhao, L. Zhao, J. Lv, S. Wang, J. Gong, X. Ma, AIChE J. 2012,
58, 2798–2809.
ð3Þ
TOF ¼
À1
numbers of Cuþ sites ðmol gcat
Þ
The number of surface Cu+ sites was calculated based on the fol-
lowing equation:
ꢀ
ꢁ
[14] A. Y. Yin, X. Y. Guo, W. L. Dai, H. X. Li, K. N. Fan, Appl. Catal. A 2008, 349,
91–99.
[15] E. Reichelt, M. P. Heddrich, M. Jahn, A. Michaelis, Appl. Catal. A 2014,
476, 78–90.
X
Xi
XCuþ
nðCuþÞ
pri2 ¼ SBET
ð4Þ
[16] X. Y. Wang, M. Wen, C. Z. Wang, J. Ding, Y. Sun, Y. Liu, Y. Lu, Chem.
Commun. 2014, 50, 6343–6345.
[17] G. F. Zhao, Y. K. Li, Q. F. Zhang, M. M. Deng, F. H. Cao, Y. Lu, AIChE J.
2014, 60, 1045–1053.
[18] Q. F. Zhang, Y. K. Li, L. Zhang, L. Chen, Y. Liu, Y. Lu, J. Catal. 2014, 317,
54–61.
where n(Cu+) is the number of Cu+ sites, Xi is the molar fraction of
individual atom i analyzed by XPS (i contains Cu0, Cu+, Au, Pd, C,
and O), SBET is the specific surface area of catalysts. To obtain the in-
trinsic activity of catalysts, the DMO conversion was kept below
30% at 2108C and high LWHSV of 9 hÀ1
.
[19] L. P. Han, C. Z. Wang, G. F. Zhao, Y. Liu, Y. Lu, AIChE J. 2016, DOI:
10.1002/aic.15061.
[20] Q. F. Zhang, X. P. Wu, G. F. Zhao, Y. K. Li, C. Z. Wang, Y. Liu, X. Q. Gong, Y.
Lu, Chem. Commun. 2015, 51, 12613–12616.
[21] Y. Liu, H. Wang, J. F. Li, Y. Lu, Q. S. Xue, J. C. Chen, AIChE J. 2007, 53,
1845–1849.
The Xi (i contains Cu0 and Cu+, Au, Pd, C and O) was calculated on
the basis of XPS spectra. As exemplified for the 0.1Pd–0.5Au–
CuOx/Cu-fiber catalyst, the surface molar ratio of Cu:Au:Pd:C:O is
provided from the XPS survey spectra. Firstly, the surface molar
ratio of Cu:Au: Pd:C:O is supposed to be XCu:XAu:XPd:XC:XO (XCu
+
[22] Y. K. Li, Q. F. Zhang, R. J. Chai, G. F. Zhao, F. H. Cao, Y. Liu, Y. Lu, Chem-
CatChem 2015, 7, 1427–1431.
[23] M. Ling, G. F. Zhao, W. Chen, M. M. Wang, Q. S. Xue, Y. Lu, Int. J. Hydro-
gen Energy 2011, 36, 12833–12842.
[24] L. Zhang, L. P. Han, G. F. Zhao, R. J. Chai, Q. F. Zhang, Y. Liu, Y. Lu, Chem.
Commun. 2015, 51, 10547–10550.
[25] G. F. Zhao, H. Y. Hu, M. M. Deng, M. Ling, Y. Lu, Green Chem. 2011, 13,
55–58.
X
Au +XPd +XC +XO =1). As Cu2p3/2 XPS spectra cannot differentiate
between Cu+ and Cu0, Cu LMM Auger spectra were used to differ-
entiate Cu+ and Cu0. Considering the extremely low Cu2+ percent-
age in all the Cu states, Cu2+ was omitted from the total Cu states
to calculate the “Xi”. Subsequently, the Cu+/Cu0 molar ratio could
be determined using Cu LMM Auger spectra and was supposed to
be cCu+:cCu0 (cCu+ +cCu0 =1). Finally, the surface molar ratio of Cu0/
Cu+/Au/Pd/C/O could be determined to be (XCu ”cCu0):(XCu ”cCu+):
XAu:XPd:XC:XO.
[26] Y. Wang, Y. Shen, Y. Zhao, J. Lv, S. Wang, X. Ma, ACS Catal. 2015, 5,
6200–6208.
[27] D. S. Brands, E. K. Poels, A. Bliek, Appl. Catal. A 1999, 184, 279–289.
[28] E. K. Poels, D. S. Brands, Appl. Catal. A 2000, 191, 83–96.
[29] J. Gong, H. Yue, Y. Zhao, S. Zhao, L. Zhao, J. Lv, S. Wang, X. Ma, J. Am.
Chem. Soc. 2012, 134, 13922–13925.
Acknowledgements
[30] I. Meliµn-Cabrera, M. L. Granados, J. L. G. Fierro, J. Catal. 2002, 210, 285–
294.
[31] A. J. McCue, C. J. McRitchie, A. M. Shepherd, J. A. Anderson, J. Catal.
2014, 319, 127–135.
This work was funded by the NSF of China (21473057, 21273075,
U1462129, 21076083) and the “973 program” (2011CB201403)
from the MOST of China.
[32] E. C. Corbos, P. R. Ellis, J. Cookson, V. Briois, T. I. Hyde, G. Sankarc, P. T.
Bishop, Catal. Sci. Technol. 2013, 3, 2934–2943.
[33] S. Li, Y. Wang, J. Zhang, S. Wang, Y. Xu, Y. Zhao, X. Ma, Ind. Eng. Chem.
Res. 2015, 54, 1243–1250.
Keywords: alloys · copper · hydrogenation · gold · palladium
[34] P. Liu, E. J. M. Hensen, J. Am. Chem. Soc. 2013, 135, 14032–14035.
[35] K. L. Deutsch, B. H. Shanks, J. Catal. 2012, 285, 235–241.
[36] F. Severino, J. L. Brito, J. Laine, J. L. G. Fierro, A. L. Agudo, J. Catal. 1998,
177, 82–95.
[1] Y. C. Li, S. R. Yan, W. M. Yang, Z. K. Xie, Q. L. Chen, B. Yue, H. Y. He, J. Mol.
Catal. A 2005, 226, 285–290.
[2] X. Zheng, H. Lin, J. Zheng, X. Duan, Y. Yuan, ACS Catal. 2013, 3, 2738–
2749.
[37] I. Platzman, R. Brener, H. Haick, R. Tannenbaum, J. Phys. Chem. C 2008,
112, 1101–1108.
[3] Y. Zhu, X. Kong, D. B. Cao, J. Cui, Y. Zhu, Y. W. Li, ACS Catal. 2014, 4,
3675–3681.
[38] J. K. Edwards, S. J. Freakley, A. F. Carley, C. J. Kiely, G. J. Hutchings, Acc.
Chem. Res. 2014, 47, 845–854.
[4] C. Wen, F. Li, Y. Cui, W.-L. Dai, K. N. Fan, Catal. Today 2014, 233, 117–
126.
[39] J. K. Edwards, B. E. Solsona, P. Landon, A. F. Carley, A. Herzing, C. J. Kiely,
G. J. Hutchings, J. Catal. 2005, 236, 69–79.
[5] Z. N. Xu, J. Sun, C. S. Lin, X. M. Jiang, Q. S. Chen, S. Y. Peng, M. S. Wang,
G. C. Guo, ACS Catal. 2013, 3, 118–122.
[40] J. K. Edwards, A. Thomas, A. F. Carley, A. A. Herzing, C. J. Kiely, G. J.
Hutchings, Green Chem. 2008, 10, 388–394.
[6] X. Gao, Y. Zhao, S. Wang, Y. Yin, B. Wang, X. Ma, Chem. Eng. Sci. 2011,
66, 3513–3522.
[7] H. T. Teunissen, C. J. Elsevier, Chem. Commun. 1997, 667–668.
[8] Y. Zhu, S. Wang, L. Zhu, X. Ge, X. Li, Z. Luo, Catal. Lett. 2010, 135, 275–
281.
[9] Z. He, H. Q. Lin, P. He, Y. Z. Yuan, J. Catal. 2011, 277, 54–63.
[10] Y. N. Wang, X. P. Duan, J. W. Zheng, H. Q. Lin, Y. Z. Yuan, H. Ariga, S. Taka-
kusagi, K. Asakura, Catal. Sci. Technol. 2012, 2, 1637–1639.
Received: December 28, 2015
Published online on February 17, 2016
ChemCatChem 2016, 8, 1065 – 1073
www.chemcatchem.org
1073
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim