X. Gao et al.
MolecularCatalysis462(2019)69–76
All the reactions were carried out in a 50 mL Teflon-lined stainless
around 334.7 and 339.9 eV were detected over Pd/CeO2, corresponding
to metallic Pd [26]. The Pd 3d peaks shifted to high binding energy
because of Au incorporation over AuPd bimetallic catalysts. There is
strong electronic interaction between Au and Pd owing to alloy effect,
wherein the electrons transfer from Pd to Au. The surface ratio of Au
and Pd was estimated to be 1.2 by XPS peak area, similar to that of bulk
phase in Au1Pd1/CeO2, indicating that Au and Pd atoms were homo-
genously distributed in AuPd alloy structure. As illustrated in Fig. 2c for
UV–vis spectra, Au/CeO2 exhibited characteristic surface plasmon re-
sonance (SPR) adsorption at 520 nm while Pd/CeO2 did not have any
SPR adsorption [39]. After the addition of Pd into Au/CeO2, the SPR
adsorption of Au declined sharply and even completely disappeared
owing to the alloy effect.
CO-DRIFTS were performed to probe electronic effect and the re-
sults are illustrated in Fig. 2d. Au/CeO2 exhibited only one carbonyl
stretching peak at 2114 cm−1, corresponding to linear adsorption of CO
on Au [40]. The intense two peaks were observed at around 2083 and
1923 cm−1, which were assigned to linear and bridged adsorption of
CO on Pd [40], respectively. Notably, the linear adsorption of CO lo-
cated at the corner of edges sites of undercoordinated Pd, while bridged
CO adsorption located on surface Pd. These two peaks presented a si-
milar change in intensity on AuPd/CeO2, strongly indicating that Au
and Pd species were homogenously dispersed and didn’t form core-shell
structure, consistent well with HRTEM results. Moreover, the adsorp-
tion peaks of CO displayed an obvious blue-shift with increasing Au
content, suggesting strong electronic interaction between Au and Pd
species. According to d-π model [40], Au electron density increases
from Pd donation during the formation of AuPd alloy.
steel autoclave. Typically, 0.2 mmol feed or 0.03 g organosolv lignin,
15 mL H2O, 2 mmol or 10 mmol formic acid and 0.1 g catalyst were
loaded into the reactor. The reaction was performed at preset tem-
perature for desired time under vigorous stirring. After the completion
of reaction, the reactor was quickly cooled to room temperature in an
ice-water bath. The products were collected by centrifugation, ex-
tracted by ethyl acetate, and analyzed by gas chromatography
(Shimadzu GC-2010) with a flame ionization detector using a DB-1
capillary column. Ethyl benzoate was used as an internal standard. The
assignments of products were determined by GC–MS. The quantitative
analysis was estimated by the following equations:
moles of initial reactant − moles of remained reactant
Conversion(%) =
Selectivity(%) =
moles of initial reactant
× 100
moles of one product
× 100
moles of all products
The concentrations of formic acid were analyzed by HPLC (Agilent
1100 Series) using UV–vis detector and an Eclipse XDB-C18 column.
The mobile phase was 0.005 mol/L H2SO4 flowing at a rate of 1.0 mL/
min. The column oven was set at 30 °C. For lignin depolymerization, the
product yield was defined as the mass of monophenol divided by the
mass of lignin.
3. Results and discussion
3.1. Catalyst characterization
H2-TPD was conducted to reflect the adsorption of H2 on active
metallic surface (Fig. 3). Au/CeO2 displayed two weak H2 desorption
peaks at around 80 and 425 °C, reflecting that H2 is not easily adsorbed
and activated on Au surface. Compared to Au/CeO2, AuPd alloy cata-
lysts obviously improve adsorption capacity of H2. Similar H2 deso-
rption peaks were observed over Pd/CeO2, suggesting that the ad-
sorption strength of H2 is similar with AuPd bimetallic catalysts.
Moreover, Pd/CeO2 showed two strong peaks at around 80 and 365 °C,
and possessed the highest adsorption capacity of H2.
The PVA capped AuPd bimetallic nanoparticles were prepared by
the co-reduction of PdCl2 and HAuCl4 by NaBH4, and subsequently
deposited on rod-shaped CeO2 support. The composition of AuPd na-
noparticles can be regulated by the initial molar ratio of their pre-
cursors and analyzed by ICP. The metal contents and textural properties
are listed in Table S1. All the samples displayed similar BET surface
area and pore distributions. As illustrated in Fig. S1, the typical diffuse
diffraction peaks around 28.8°, 33.3°, 47.8° and 56.5° were ascribed to
face-centered-cubic (fcc) structure of CeO2 [38]. Only weak Au dif-
fraction peak at 38.4° were detected over Au/CeO2 while no Pd-con-
taining peaks were observed, probably due to the high Pd dispersion or
low metal content. The representative TEM micrographs and particle
size distribution histograms for the AuPd/CeO2 catalysts are illustrated
in Fig. 1. In all the samples, the CeO2 primarily presented as disordered
rod shape, consistent with previous result [34]. Au nanoparticles pro-
ceeded to agglomeration and some of them were larger than 10 nm in
size over Au/CeO2. Pd nanoparticles were distributed uniformly on the
CeO2 surface with a 6.4 nm average particle size over Pd/CeO2. The
corresponding lattice fringes in HRTEM confirmed the presence of
crystalline Pd nanoparticles. All the bimetallic AuPd samples showed
highly dispersed nanoparticles and lower particle size compared to the
monometallic samples. In addition, the AuPd species form an extended
flat interface structure with rod-shaped crystalline ceria surface, sug-
gesting that AuPd nanoparticles are tightly attached on CeO2 surface.
The HRTEM image of bimetallic AuPd nanoparticles are indicative of
polycrystalline face-centered cubic (fcc) structure with exposed {1 1 1}
lattice fringes measured to be 0.227 nm, which is lower than that of fcc
Au (0.235 nm) and larger than the {1 1 1} spacing of fcc Pd (0.221)
[39]. This is probably related to the formation of alloy structure.
The formation of AuPd alloy structure for the bimetallic samples can
also be verified by XPS, UV–vis spectra and CO-DRIFTS. As displayed in
Fig. 2a of XPS spectra, Au/CeO2 exhibited spin-orbit split peaks cen-
tered at 83.7 and 87.3 eV, which were assigned to Au 4f7/2 and Au 4f5/2
of metallic Au, respectively [34]. Compared to Au/CeO2, the addition of
Pd induced the movement of Au 4f peaks towards low binding energy,
suggesting that Au species were in the electron-rich states. Two peaks at
3.2. Hydrogenolysis of benzyl phenyl ether (α-O-4) with formic acid
The catalytic performance was initially evaluated with model
compounds benzyl phenyl ether, 2-phenylethyl phenyl ether and di-
phenyl ether to mimic α-O-4, β-O-4 and 4-O-5 of lignin, respectively. As
listed in Table 1, only 3.5% conversion of benzyl phenyl ether and
40.1% conversion of formic acid were obtained over Au/CeO2 at 150 °C
for 1 h. Because the amount of formic acid far exceeds that of benzyl
phenyl ether, the low conversion of Au/CeO2 is not predominantly
ascribed to the lack of hydrogen donor. Pd/CeO2 gave 49.6% conver-
sion of benzyl phenyl ether and 76.3% conversion of formic acid.
Compared to Au/CeO2 and Pd/CeO2, all the bimetallic AuPd catalysts
displayed excellent activity of benzyl phenyl ether and complete con-
version of formic acid. Of them, Au1Pd1/CeO2 exhibited the highest
reactivity and reached 100% conversions of benzyl phenyl ether and
formic acid. Moreover, the conversion of benzyl phenyl ether over
Au1Pd1/CeO2 was much higher than that of mixed catalyst Au/CeO2+
Pd/CeO2, suggesting that there was remarkable synergistic effect be-
tween bimetallic species. Regarding the selectivity, all the samples gave
similar product distributions with good C–O bond cleavage ability.
41.8% phenol and 52.4% toluene were achieved over Au1Pd1/CeO2.
Kinetic analysis in Fig. S2 displays the correlation between reaction rate
with benzyl phenyl ether and formic acid concentration, respectively.
The reaction orders with respect to benzyl phenyl ether and formic acid
concentration were calculated to be 0.69 and 0.14, respectively. Thus,
the higher reaction order of benzyl phenyl ether indicates that C–O
bond cleavage is the rate-determining step.
Turnover frequency (TOF) can be employed to reflect the intrinsic
71