The synergistic role of the support surface and Au-Cu alloys in a plasmonic Au-Cu?LDH photocatalyst for the oxidative esterification of benzyl alcohol with methanol

Article abstract of DOI:10.1039/c9cp05992j

Layered double hydroxide-supported Au-Cu alloy nanoparticles (NPs) were found to be highly efficient catalysts for the oxidative esterification of benzyl alcohol with methanol in the presence of molecular oxygen under visible-light irradiation to prepare methyl benzoate. Here, we report that alloying small amounts of copper into gold nanoparticles can increase the ability to activate oxygen molecules to O₂^- radicals and display greater charge heterogeneity to promote the cleavage of the C-H bond of benzyl alcohol molecules by reinforcing the coordination of the intermediate with unsaturated metal active sites due to the LSPR effect of alloy NPs, which is the rate-limiting step of the reaction. Besides the Au-Cu alloy NPs, the support also played a pivotal role in the catalytic process. The support with the presence of acid-base pairs, in which the basic sites served as the reactant molecule adsorption sites to provoke the intermediate formation and the acidic sites promoted the recovery of the support surface, showed better performances by affecting the overall reaction rate completely. Moreover, applying this photocatalyst in the cross-esterification of aromatic alcohols and aliphatic alcohols displayed excellent yields.

Full text of DOI:10.1039/c9cp05992j

PCCP
View Article Online
View Journal
PAPER
The synergistic role of the support surface
and Au–Cu alloys in a plasmonic Au–Cu@LDH
photocatalyst for the oxidative esterification
of benzyl alcohol with methanol†
Cite this:DOI: 10.1039/c9cp05992j
ab
ab
Xiaoyu Wang,
Ruiyi Wang, *a Jie Wang,
*
Chaoyang Fanac and
ab
Zhanfeng Zheng
Layered double hydroxide-supported Au–Cu alloy nanoparticles (NPs) were found to be highly efficient
catalysts for the oxidative esterification of benzyl alcohol with methanol in the presence of molecular
oxygen under visible-light irradiation to prepare methyl benzoate. Here, we report that alloying small
À
ꢀ
amounts of copper into gold nanoparticles can increase the ability to activate oxygen molecules to O2
radicals and display greater charge heterogeneity to promote the cleavage of the C–H bond of benzyl
alcohol molecules by reinforcing the coordination of the intermediate with unsaturated metal active sites
due to the LSPR effect of alloy NPs, which is the rate-limiting step of the reaction. Besides the Au–Cu
alloy NPs, the support also played a pivotal role in the catalytic process. The support with the presence
of acid–base pairs, in which the basic sites served as the reactant molecule adsorption sites to provoke
the intermediate formation and the acidic sites promoted the recovery of the support surface, showed
better performances by affecting the overall reaction rate completely. Moreover, applying this photo-
catalyst in the cross-esterification of aromatic alcohols and aliphatic alcohols displayed excellent yields.
Received 4th November 2019,
Accepted 6th December 2019
DOI: 10.1039/c9cp05992j
rsc.li/pccp
cross-coupling reaction of halogenated aromatic hydrocarbons
with methanol is another direct route for the synthesis of
1. Introduction
Methyl benzoate is an important ester compound extensively methyl benzoate. However, the high temperature and pressure
applied as a valuable intermediate in organic synthesis, solvents, required for the reaction result in high energy consumption and
and edible and chemical essentials for daily use in chemical environmental pollution,4 which inspired us to explore greener
industries.1 The traditional synthesis method of methyl benzoate heterogeneous catalytic processes for the oxidative esterification
is via the esterification of benzoic acid with methanol using acids of benzyl alcohol to methyl benzoate. Supported Au catalysts
as catalysts,2 but the highly corrosive nature of acid catalysts to have been reported to be efficient for the direct oxidative
the equipment, the harsh reaction conditions and the serious esterification of benzyl alcohol and other aldehydes or alcohols
environmental pollution inhibit their applications. Besides, at with methanol to give the corresponding esters.5–9 However, the
present, reacting an acid halide or an acid anhydride with conversion or product selectivity is typically low and the reaction
methanol is also a major method to synthesize methyl benzoate, needs to be completed at a high temperature or in a long
but the additives required for the reaction lead to the low reaction time. Miyamura et al. reported that using polymer-
selectivity of the ester and a large amount of by-products.3 incarcerated gold nanoclusters (PI-Au) in catalyzing the aerobic
However, alcohol is more readily available and inexpensive than oxidation of alcohols to methyl esters needed nearly 24 h to
benzoic acid and acid halide as the material to synthesize achieve a TON of 540.8 In addition, Zhao et al. reported the use
methyl benzoate. In addition, the transition metal-catalyzed of an Au/meso-g-Al2O3 catalyst for the selective aerobic oxidation
of alcohols at a reaction temperature of 80 1C to achieve a
TOF of 145.8 h .
À1 10 The modification of Au catalysts with other
a State Key Laboratory of Coal Conversion, Institute of Coal Chemistry,
Chinese Academy of Sciences, Taiyuan, Shanxi 030001, P. R. China.
E-mail: wangruiyi@sxicc.ac.cn, zfzheng@sxicc.ac.cn
noble metals to form alloy NPs can improve the activity of the
reaction.11,12 For example, Wang et al. found that by alloying
small amounts of palladium into gold nanoparticles, the con-
version of benzyl alcohol greatly increased from 3.6% for
Au/LDH and 1.5% for Pd/LDH to 91.1% for Au–Pd/LDH.13
However, supported Au–Cu alloy catalysts applied in oxidation
b Center of Materials Science and Optoelectronics Engineering, University of Chinese
Academy of Sciences, Beijing 100049, P. R. China
c Taiyuan University of Technology, Taiyuan, 030024, P. R. China
† Electronic supplementary information (ESI) available: More experimental details
and catalyst characterization and reaction test results. See DOI: 10.1039/c9cp05992j
Phys. Chem. Chem. Phys.
This journal is © the Owner Societies 2019

View Article Online
Paper
PCCP
reactions are limited as the surface Cu atoms can be easily (heating rate 10 1C minÀ1) in dry air for 8 h, yielding mixed
14
oxidized by O2 and hence, the alloying effect is eliminated oxides denoted as MgAl-LDO.
considerably. Sugano et al. found that the visible-light irradiation
of Au–Cu alloy NPs during the reaction can suppress the oxidation
of surface Cu atoms and successfully maintain the alloying effect15
due to the localized surface plasmon resonance (LSPR) of noble
metal NPs, which can strongly absorb visible light. In particular,
alloying small amounts of copper into gold nanoparticles may
change the electronic state of metal nanoparticles and introduce
inter-metallic interactions, which can further increase the
ability to activate the reactants. As is reported, the synergistic
electron effect in Au–Pd alloys exists because of the electron
transfer from Pd to Au atoms, which can enhance the adsorption
ability of metal atoms for oxygen molecules and activate
2.2 Au–Cu@MgAl-LDH photocatalysts
The Au–Cu@MgAl-LDH catalyst was prepared using an
impregnation–reduction method. MgAl-LDO powder (0.5 g) was
placed in a beaker; then, HAuCl4 (6.6 ml, 0.01 M) and Cu(NO3)2
(3.3 ml, 0.01 M) aqueous solutions were added under magnetic
stirring. The aqueous solution of lysine (3 ml, 0.1 M) was then
added dropwise and vigorously stirred for 30 min and the pH was
8–9. Then, the freshly prepared aqueous solution of NaBH4 (2 ml,
0.35 M) was added dropwise. After stirring for 24 h at room
temperature, the solid was washed by deionized water (twice) and
ethanol (once). Finally, the photocatalyst was obtained by centrifuga-
tion (3000 rpm) and drying at 60 1C in a vacuum oven for 12 h. All
the other catalysts were prepared via the same method with different
quantities of HAuCl4 and Cu(NO3)2 aqueous solutions.
them to form O2 radicals;13 thus, the supported Au–Cu alloy
NP catalysts applied in the oxidative esterification of benzyl
alcohol under visible-light irradiation have attracted research
interest.
À
ꢀ
Moreover, the traditional supports of noble metal catalysts
generally involving Al2O3,5 TiO2,6 SiO2,7 polymers,8 MOFs16 and
hydrotalcite9,17–19 for the oxidative esterification of alcohols or
aldehydes have been reported. Layered double hydroxides (LDHs)
are synthetic or natural crystalline clays that consist of brucite-like
(Mg(OH)2) layers of edge-sharing MO6 octahedra, in which some
divalent cations such as Mg2+ are substituted by trivalent cations;
these are characterized by the presence of acid–base pairs, and
these pairs can facilitate the adsorption of reactants and inter-
mediate molecules. According to Feng and co-workers, in the
MgAl-LDH-supported Au–Pd alloy NP-catalyzed benzyl alcohol
oxidation reaction, the acid sites of the MgAl-LDH support were
responsive toward the improvement in catalytic activity, while
the basic sites give rise to high selectivity; however, this reaction
needs to be conducted at 140 1C.12 Thus, the use of milder
conditions for the oxidative esterification of benzyl alcohol is
rarely reported.
2.3 Characterization of catalysts
Powder X-ray diffraction (XRD) was performed on a Bruker D8
Advanced diffractometer operating with Cu Ka radiation
(l = 1.5405 Å). The morphology was studied using transmission
electron microscopy (TEM, Tecnai G2 F20 S-Twin) with an
accelerating voltage of 200 kV. X-ray photoelectron spectro-
scopy (XPS) data were obtained on USA Thermo ESCALAB
250 with a monochromatized Al Ka line source (200 W).
The binding energy correction was referenced to C 1s peaks
(284.8 eV) arising from surface hydrocarbons. The nitrogen
adsorption isotherms were measured at À196 1C on a TriStar II
3020 volumetric adsorption analyzer. Prior to the measurement,
all the samples were degassed under evacuation at 150 1C for
10 h. The surface area was calculated from the adsorption branch
by the Brunauer–Emmett–Teller (BET) method and the pore size
distribution was obtained by desorption isotherms using the
Barret–Joyner–Halender (BJH) method. Fourier transform infra-
red spectroscopy (FTIR) was performed using a Bruker Tensor II
spectrometer with KBr pellets. The UV-vis diffuse reflectance
spectra (DRS) were observed on a Hitachi UV-3900 UV-vis-NIR
spectrophotometer. The actual total metal loadings and the Au/
Cu molar ratios of various Au–Cu@MgAl-LDH samples were
determined by an inductively coupled plasma (ICP) spectrometer
on a Thermo iCAP6300 instrument.
Herein, we described a visible-light-driven process for the
oxidative esterification of benzyl alcohol with methanol, where
molecular oxygen was used as the oxidant and high conversion
and product selectivity were obtained under much milder conditions
over an LDH-supported Au–Cu NP photocatalyst. In this catalyst,
fused Au with Cu NPs increased the ability to activate the oxygen
molecules to O2 À radicals and showed greater charge heterogeneity
ꢀ
to promote the cleavage of the C–H bond of benzyl alcohol
molecules due to the LSPR effect of alloy NPs; this could drive
the rate-limiting step of the reaction under the premise that the
2.4 Catalytic reactions
support sites remained the same, and the acid–base pairs of Typically, the photocatalytic experiments were carried out in an
the support played pivotal roles in the adsorption of reactant oxygen atmosphere at the required temperature. The reaction
molecules to completely drive the overall reaction system.
system consisted of a catalyst (10 mg), benzyl alcohol (0.5 mmol)
in methanol (5 ml) and K2CO3 (0.2 mmol). The incident light
source used was a Xenon lamp at a UV cut-off wavelength below
420 nm. The liquid products were analyzed using Shimadzu
2014C GC equipped with a WondaCap 5 column after centri-
fugal separation of the catalyst. The identity of the products was
2. Experimental
2.1 Preparation of catalysts MgAl-LDHs
Layered double hydroxides were prepared using a sol–gel process established by the comparison of the retention times of authen-
following the established procedures and were denoted as tic samples and also by NMR spectroscopy. The single-coloured
MgAl-LDHs.20,21 Then, the LDHs were calcined at 450 1C LEDs (purple, blue, green and red) were also used as irradiation
Phys. Chem. Chem. Phys.
This journal is © the Owner Societies 2019

View Article Online
PCCP
Paper
sources to investigate the influence of light wavelength on the 3. Results and discussion
photocatalytic activity.
3.1 Photocatalytic performance
In order to strictly avoid the exposure of the reaction to light, the
LDH-supported Au–Cu alloy NP photocatalysts with various Au/
Cu ratios were synthesized (Au1Cu@Mg3Al-LDHs, Au2Cu@
Mg3Al-LDHs, and Au3Cu@Mg3Al-LDHs) (Table 1, entries 1–3)
to probe their catalytic activities in the oxidative esterification
of benzyl alcohol with methanol to give methyl benzoate in the
presence and absence of light. It was indicated that the
Au2Cu@Mg3Al-LDH catalyst (entry 1) exhibited optimal catalytic
activity with 97.0% conversion of benzyl alcohol, 99.4% selectivity
to methyl benzoate and a turnover frequency (TOF) of 172.8 hÀ1
at 30 1C for 3 h under visible-light irradiation, which were much
higher than those in the dark. The catalytic activities of the
monometallic Au catalyst and the mixture of monometallic Au
and Cu catalysts (entries 4 and 6) were obviously lower than that
of the alloy NP catalyst, and there was no catalytic activity while
using monometallic Cu and supports (Mg3Al-RE, Mg3Al-LDHs
and Mg3Al-LDO; entries 5, 7–9) as the catalysts. It was concluded
that alloying the two metals could further enhance the catalytic
activity. Due to the excellent performance, Au2Cu@Mg3Al-LDHs
quartz reaction bottle was covered with an aluminium foil. The
reaction temperature under irradiation was maintained the same as
that in the dark to make sure that the comparison is meaningful.
2.5 Calculation details of turnover frequency
Turnover frequency (TOF), the number of converted molecules
per active site and second, is calculated on the basis of the total
loadings of Au (there was no catalytic activity with monome-
tallic Cu), i.e., TOF = (mBA  XBA/MBA)/(mcat  wAu%  t/MAu),
where mBA is the mass of benzyl alcohol (in g) loaded in the
reactor, XBA is the conversion of benzyl alcohol (%), and MBA is
the molar mass of benzyl alcohol; mcat is the mass of the
catalyst loaded (g) in the reactor, wAu% is the loading mass of
Au per gram LDH-supported Au–Cu alloy NP photocatalyst (%),
MAu is the molar mass of Au and t is the reaction time (h).
2.6 Calculation details of apparent quantum yield (AQE)
AQE = [(Nlight À Ndark)/(the number of incident photons)] Â were chosen as the primary materials for the following study.
100%, where Nlight and Ndark are the numbers of products Then, the activation energies of light-driven and thermal-driven
formed under irradiation and in the dark, respectively.
reactions were calculated according to the Arrhenius equation
Table 1 Photocatalytic oxidative esterification of benzyl alcohol with methanola
Loading (wt%)b
Entry
1
Catalyst
Au
Cu
Light source
Conv. (%)
Sel.c (%)
TOFd (hÀ1
)
Au2Cu@Mg3Al-LDHs
1.89
0.27
Visible
Dark
Visible
Dark
Visible
Dark
Visible
Dark
Visible
Dark
Visible
Dark
Visible
Dark
97.0
61.3
17.4
14.5
74.8
42.7
43.8
25.8
0
99.4
97.0
73.2
63.1
98.6
94.2
91.3
81.3
0
172.8
104.4
93.6
75.6
100.8
57.6
79.2
46.8
0
2
Au1Cu@Mg3Al-LDHs
Au3Cu@Mg3Al-LDHs
Au@Mg3Al-LDHs
0.62
2.32
1.78
—
0.21
0.25
—
3
4
5
Cu@Mg3Al-LDHs
0.24
—
0
0
0
6
Au@Mg3Al-LDHs + Cu@Mg3Al-LDHse
Mg3Al-LDO-REf
—
7.9
9.5
0
38.2
32.8
0
—
—
—
—
7
—
—
0
0
8
Mg3Al-LDHs
—
—
Visible
Dark
0
0
0
0
—
—
9
Mg3Al-LDO
—
—
Visible
Dark
Visible
Dark
Visible
Dark
0
0
90.2
58.4
4.6
3.6
0
0
98.4
93.8
0
—
—
—
—
—
—
10
11
Au2Cu@Mg3Al-LDHsg
1.89
—
0.27
—
h
—
0
a
Reaction conditions: photocatalyst, 10 mg; benzyl alcohol, 0.5 mmol; methanol, 5 ml; K2CO3, 0.2 mmol; O2, atmospheric pressure; reaction temperature,
30 1C; reaction time, 3 h; light intensity (cut-off wavelength of 420 nm of Xe lamp), 0.5 W cmÀ2. The conversion and selectivity were calculated from the
reactant converted and the product formed, as measured by gas chromatography (GC). b ICP analysis results. c The 13C NMR and 1H NMR spectra of the
product (methyl benzoate) are shown in Fig. S3 and S4 (ESI), respectively; it was separated by the distillation of the reaction solution. d Turnover frequency
(TOF), the number of benzyl alcohol molecules converted to methyl benzoate per active site and second, is estimated on the basis of total Au atoms (ESI).
e Au@Mg3Al-LDHs and Cu@Mg3Al-LDHs were mechanically mixed in a mortar according to n(Au) : n(Cu) = 1 : 1. f Reconstruction of Mg3Al-LDHs. g Without
K2CO3, reaction temperature, 55 1C; reaction time, 18 h. h No catalyst, only K2CO3 was added; reaction temperature, 55 1C; reaction time, 18 h.
Phys. Chem. Chem. Phys.
This journal is © the Owner Societies 2019

View Article Online
Paper
PCCP
Fig. 1 (a) Spectrum showing photocatalytic action for the oxidative esterifica-
tion of benzyl alcohol using Au2Cu@Mg3Al-LDH photocatalyst under single-
coloured LED lights. The light-absorption spectrum (left axis) is the DR-UV/vis
spectrum of the Au2Cu@Mg3Al-LDH catalyst (black curve). (b) Dependence of
the catalytic activity of Au2Cu@Mg3Al-LDH photocatalyst for the oxidative
esterification of benzyl alcohol with methanol on the intensity of irradiation.
The light contribution = [(Ylight À Ydark)/Ylight] Â 100%, where Ylight and Ydark
are the product yields under irradiation and in the dark, respectively.
Reaction conditions: photocatalyst, 10 mg; benzyl alcohol, 0.5 mmol;
methanol, 5 ml; K2CO3, 0.2 mmol; O2, atmospheric pressure; reaction
temperature, 30 1C; reaction time, 3 h.
shown in Fig. S1 (ESI†); the values were found to be 105.9 and
115.1 kJ molÀ1, respectively. This means that light energy
can facilitate the reaction by reducing the activation energy.
The catalyst activity of Au2Cu@Mg3Al-LDHs remained stable
(Fig. S2a, ESI†) and a slight decline in activity during the cycling
experiment may be due to the oxidation of a small amount of Cu
during the catalytic cycle process (Fig. S2f, ESI†). No obvious
agglomeration of the alloy NPs appeared (Fig. S2b and c, ESI†);
the Cu content and the structure showed no obvious change
(Fig. S2d and e, ESI†) after being reused for five cycles.
In the control experiment, the dependence of photocatalytic
activity on wavelengths with the irradiation intensity main-
tained constant was studied (Fig. 1a). The apparent quantum
yield (AQE) of the reaction is closely related to the irradiation
wavelength. Moreover, the highest yield is achieved under
irradiation at the wavelength where alloy NPs have the most
intense absorption (527 nm). It can be deduced that the LSPR
effect of the alloy NPs makes a primary contribution to photo-
oxidative esterification. Moreover, the relationship between
the photocatalytic activity and the incident light intensity
was studied, as shown in Fig. 1b. When the irradiance was
0.2 W cmÀ2, the light contribution to the reaction was only
6.7% and when the irradiance increased to 0.5 W cmÀ2, a
35% yield of methyl benzoate was obtained. Hence, it was
concluded that the methyl benzoate yield increased with the
increase in the irradiation intensity, and the contribution of
irradiation to the overall reaction rate was greater at a higher
intensity; this indicates that a higher irradiation intensity can
provide more light-excited electrons to drive the reaction by
facilitating the cleavage of the C–H bond of the reactant
molecules in a photocatalytic process, which is the rate-limiting
step of the reaction.11,22–24
Fig. 2 (a) TEM image of the Au2Cu@Mg3Al-LDH sample. (b) Particle size
distribution of the NPs based on the statistical analysis of TEM images.
(c) HR-TEM image of an alloy particle. (d) XPS profiles of Au and Cu species
in Au2Cu@Mg3Al-LDH sample.
sample were obtained. It was shown that the alloy NPs were
distributed evenly on the LDH surface (Fig. 2a), and the mean
size of the alloy NPs was approximately 3–4 nm (Fig. 2b).
The high-resolution TEM (HR-TEM) images reveal the atomic
lattices of Au–Cu alloy NPs (Fig. 2c). The lattice fringe spacing
of 0.22 nm corresponds to the interplanar distance of the (111)
planes of the Au–Cu alloy.25 The characterization results con-
firmed the formation of Au–Cu alloy NPs. The light absorption
characterization of the Au2Cu@Mg3Al-LDH sample provided by
the UV-vis spectra (Fig. S5, ESI†) also proved alloying because a
red shift in the band at 527 nm due to the LSPR effect was
observed for the Au–Cu alloy NP sample compared to that for
the monometallic Au NP sample (Au2@Mg3Al-LDHs, 509 nm).26
Besides, the Mg3Al-LDH support itself could not contribute to
the photocatalytic activity because it exhibited weak absorption
at visible light wavelengths above 400 nm, as shown in Fig. S5
(ESI†). In contrast, the Au2Cu@Mg3Al-LDH sample displayed
strong absorption in the visible region of the spectrum, indicating
that Au2Cu@MgAl-LDHs could utilize most of the irradiation
energy delivered by the solar spectrum. To further characterize
the electronic state of Au–Cu alloy NPs on the support, X-ray
photoelectron spectroscopy (XPS) measurements of the Au2Cu@
Mg3Al-LDH sample were obtained (Fig. 2d), which also evinced the
incorporation of the Au–Cu alloy units onto the surface of the
Mg3Al-LDH support. It was revealed that the appearance of the Au
4f peaks at around 83.4 eV and 87.1 eV could be mainly attributed
to the Au0 state. The binding energies of Cu 2p1/2 at around
952.0 eV and Cu 2p3/2 at 932.5 eV could be attributed to the Cu0
3.2 Alloy effect
To confirm the formation of Au–Cu alloy NPs, transmission state.25 Although the challenge for the supported Cu NP catalysts
electron microscopy (TEM) images of the Au2Cu@Mg3Al-LDH in the aerobic oxidation reaction still existed, i.e., O2 could oxidize
Phys. Chem. Chem. Phys.
This journal is © the Owner Societies 2019

View Article Online
PCCP
Paper
surface Cu atoms14 and eliminate the alloying effects, the visible- Generally, there are negligible effects on the structural and textural
light irradiation of Au–Cu alloy NPs during the reaction could properties of Mg3Al-LDHs after loading Au–Cu alloy NPs.
suppress the oxidation of the surface Cu atoms and successfully
Molecular oxygen is activated by plasmonic metals. The ESR
maintain the alloying effect.15 Besides, as shown in Fig. S6a spectra of the catalyst dispersion solutions confirm the formation
À
ꢀ
(ESI†), the peaks of Au 4f7/2 show a positive shift from 83.1 eV of O2 radicals during the photocatalytic reactioÀn, as shown in
ꢀ
for Au@Mg3Al-LDHs to 83.4 eV for Au2Cu@Mg3Al-LDHs and Fig. S8 (ESI†). The typical signals of DMPO–O2 appeared for
the peaks of Au 4f5/2 shift from 86.6 eV for Au@Mg3Al-LDHs to both the Au2Cu@Mg3Al-LDH and Au@Mg3Al-LDH dispersion
87.1 eV for Au2Cu@Mg3Al-LDHs, suggesting that the electron solutions under irradiation in the presence of 5,5-dimethyl-1-
À
cloud density of Au species decreases in Au2Cu@Mg3Al-LDHs. pyrroline N-oxide (DMPO) as a trapping agent.13 No O2 signal
In contrast, as shown in Fig. S6b (ESI†), the Cu 2p3/2 peaks was detected in the dark and for the support Mg3Al-LDO-RE
show a negative shift from 932.8 eV for Cu@Mg3Al-LDHs to solution, which indicated that the plasmon-activated electrons on
932.5 eV for Au2Cu@Mg3Al-LDHs and the Cu 2p1/2 peaks shift the surface metaÀl sites activated the oxygen molecules (O2).
ꢀ
ꢀ
from 952.5 eV for Cu@Mg3Al-LDHs to 952.0 eV for Au2Cu@ Moreover, the O2 radical signal was significantly more intense
Mg3Al-LDHs, indicating that the electron cloud density of the for Au2Cu@Mg3Al-LDHs than for the monometallic Au catalyst
Cu species in Au2Cu@Mg3Al-LDHs increases. It was revealed (Au@Mg3Al-LDHs), indicating more efficient electron transfer to
that the fusion of Au with Cu led to electrons preferentially O2 in case of alloying the two metals.27 This is because the alloy NP
transferring from Au to Cu atoms, which also suggested that surface has greater charge heterogeneity than the pure metal NP
the eÀ transfer from Au atoms facilitated the successful surface, resulting in a longer lifetime of light-excited electrons,
reduction of the oxidized surface Cu. This electron transfer which increases the ability to activate oxygen.11
was also confirmed by Wang et al.;13 the peaks of Pd 3d5/2
3.3 Role of support
showed a positive shift and the Au 4f7/2 peaks showed a
negative shift compared to that for the corresponding mono- The crucial role of the support in supported metal catalysts was
metallic catalysts, which revealed that the fusion of Au with Pd considered. Au–Cu nanoparticles were immobilized on commercial
led to charge transfer from Pd to Au. Besides, the XRD patterns MgO and Al2O3 as comparable samples for the Au2Cu@Mg3Al-LDH
show that the Au–Cu@Mg3Al-LDH samples exhibit both a catalyst. As shown in Fig. 3a and Table S2 (ESI†), the support of
hydrotalcite-like structure and the corresponding oxide structure Au2Cu@Mg3Al-LDHs has a cooperative role for the acid–base pairs
after loading Au–Cu alloy NPs because of the memory effect to achieve excellent catalytic results (entry 1); because the sup-
of hydrotalcite (Fig. S7a, ESI†), which is in good agreement ports MgO and Mg3Al-LDHs contain basic surface sites and
with the FT-IR spectrum results (Fig. S7b, ESI†). The surface Al2O3 has acidic surface sites, all of them exhibit lower activity
area of the samples was maintained stable (Table S1, ESI†). under irradiation and in the dark than Au2Cu@Mg3Al-LDHs
Fig. 3 (a) Photocatalytic oxidative esterification of benzyl alcohol with methanol over Au–Cu alloy NPs with varied supports. Reaction conditions:
photocatalyst, 10 mg; benzyl alcohol, 0.5 mmol; methanol, 5 ml; K2CO3, 0.2 mmol; O2, atmospheric pressure; reaction temperature, 30 1C; reaction
time, 3 h; light intensity (cut-off wavelength of 420 nm of Xe lamp), 0.5 W cmÀ2. (*a: Au2Cu@MgO and Au2Cu@Al2O3 were mechanically mixed in a
mortar according to n(Mg) : n(Al) = 3 : 1; *b: direct loading of Au–Cu alloy NPs onto HT; *c: reaction time, 6 h and light source, full-spectrum Xe lamp
(0.5 W cmÀ2); *d: reaction time, 6 h; *e: the reaction was conducted in the dark using the catalyst, which has been irradiated under full-spectrum Xe lamp
for 6 h). (b) The amount of surface basic and acidic sites of varied supports determined by CO2-TPD and NH3-TPD profiles because the recovery of LDHs
by memory effect is not reliable in the process of TPD measurement due to the release of CO2 and H2O as the reaction temperature rises; so, we can only
compare the TPD profiles for different LDOs. The specific reasons are as follows: MgAl-LDHs were calcined at 450 1C in dry air for 8 h to release H2O,
CO32À, Al–OH and Mg–OH, yielding MgAl-LDO and then, the Au–Cu alloy was supported on the MgAl-LDO in the HAuCl4 (0.01 M) and Cu(NO3)2
(0.01 M) aqueous solutions by the reduction of NaBH4; meanwhile, the MgAl-LDO recovered the surface hydroxyl groups by water absorption and the
layered structure of hydrotalcite was restored. Thus, the trend of total acidity and alkalinity of MgAl-LDO can reflect the trend of total acidity and alkalinity
of layered hydrotalcite recovered by memory effect when the synthesis conditions are consistent.
Phys. Chem. Chem. Phys.
This journal is © the Owner Societies 2019

View Article Online
Paper
PCCP
(entries 2, 3, 5 and Fig. 3b). Besides, the mechanically mixed
Au2Cu@MgO and Au2Cu@Al2O3 photocatalysts also exhibited
low activity (entry 4), which also proved the crucial role of acid–
base pairs in the support of Au2Cu@Mg3Al-LDHs.28,29 As shown
in Fig. 3a and Table S2 (ESI†), the photocatalyst supports with
varied Mg/Al ratios were synthesized and various Mg/Al ratios
were determined by ICP, as shown in Table S3 (ESI†) (Au2Cu@
Mg3Al-LDHs, Au2Cu@Mg2Al-LDHs and Au2Cu@Mg1Al-LDHs);
the catalytic performance also agreed with the above results
(entries 1, 6, 7). The support of Au2Cu@Mg3Al-LDHs having the
maximum amount of basic and acidic sites exhibited the best
catalytic performance (Fig. 3b),30 as reflected by the TPD profile
trends of different LDOs (the details are given in the caption of
Fig. 3). Besides, the physical properties of the catalysts are shown
in Table S1 (ESI†). Obviously, the surface area is not the decisive
factor for the performance as the optimal catalyst has a relatively
small specific surface area. In addition, the particle size distributions
of the Au–Cu alloy NPs with varied support samples are shown in
Fig. S9a–e (ESI†). There was a little effect of supports on the alloy
particle sizes as the mean sizes of the alloy NPs were approximately
similar between 2 and 4 nm. It is concluded that the Mg3Al-LDH
support containing acid–base pairs plays an important role in this
catalytic system.
Fig. 4 (a) FT-IR spectra of Au2Cu@Zn2Al-LDHs before and after irradiation
under full-spectrum Xe lamp. (b) FT-IR subtraction spectra of Au2Cu@
Zn2Al-LDHs before and after irradiation under full-spectrum Xe lamp.
(c) In situ FT-IR spectra of Au2Cu@Mg3Al-LDHs after co-adsorption of
benzyl alcohol and methanol (*a: dropping the reaction solution of
0.5 mmol benzyl alcohol, methanol as solvent; *b: after argon purge for
30 min; *c: pass in oxygen and start to heat up). (d) In situ FT-IR spectra of
Au2Cu@Mg3Al-LDHs after co-adsorption of benzyl alcohol and methanol
(deduction of IR spectra of Au2Cu@Mg3Al-LDHs).
In addition, considering that illumination may affect the
surface functional groups of the support and then affect the reaction
eventually, another support, i.e., Zn2Al-LDHs (Zn2Al-LDO-RE)
that can exhibit light absorption in the spectrum range of a full-
spectrum Xe lamp (350–880 nm) (Fig. S10a and b, ESI†) was
chosen to further investigate the specific action of the support as shown in Fig. S11 (ESI†).31 The new band intensity of Zn–OQC
in the oxidative esterification of benzyl alcohol. This was (452 cmÀ1) was apparently stronger for the photocatalyst system
because the Mg3Al-LDH (Mg3Al-LDO-RE) support showed no without irradiation compared to that for the photocatalyst system
light absorption in the UV-vis spectrum (the mean size of alloy after irradiation under the full-spectrum Xe lamp. This indicated
NPs on the Zn2Al-LDH support was also approximately between that the increase in the surface hydroxyl groups after irradiation
2 and 4 nm, as shown in Fig. S9f (ESI†), which excluded the inhibited the adsorption of alcohol molecules due to the fact that
effect of alloy particle size). It was distinctively shown that the the increase in the hydroxyl groups suggests that the distance
yield of methyl benzoate declined to 17.7% under irradiation between the hydroxyl groups becomes smaller, which may inhibit
(full-spectrum Xe lamp) compared to the yield of 35.9% in the the adsorption of alcohol molecules because of the steric hindrance
dark (Fig. 3a and Table S2 entry 8, ESI†). When a cut-off effect, ultimately resulting in the decrease in activity (detailed
wavelength of 420 nm of the Xe lamp (420–880 nm) was used, explanation shown in the ESI†).
the yield of methyl benzoate was up to 53.3% (entry 9). Then,
Besides, as shown in Fig. 4c and Fig. S12 (ESI†), the intensity
the Au2Cu@Zn2Al-LDH photocatalyst was irradiated with the of the vibrational mode band of Mg–OH at 459 cmÀ1 decreases,
full-spectrum Xe lamp for 6 h before the reaction was con- indicating the dehydroxylation of Mg–OH of the Mg3Al-LDH
ducted in the dark; the yield of methyl benzoate (14.9%) was support32 during the reaction. Meanwhile, a new vibrational
almost equal to that of the reaction under irradiation (17.7%). mode band of Mg–O at 488 cmÀ1 appeared, and it exhibited
It was concluded that the Zn2Al-LDH support absorbed light in increase in intensity during the reaction and a blue frequency
the spectrum range of 350–420 nm to have a negative effect on shift compared to the IR spectra of Mg–OH in the Mg3Al-LDH
the catalytic process. In addition, as shown in the FT-IR spectra support; this indicated the formation of Mg–OQC after the
of Au2Cu@Zn2Al-LDHs before and after irradiation with the dehydroxylation of Mg–OH due to the adsorption of alcohol
full-spectrum Xe lamp (Fig. 4a and b), the intensity of the band molecules on the basic sites of the support by oxygen. Then, the
for hydroxyl groups at around 3460 cmÀ1 increases obviously intermediate aldehyde species was formed because of the
after irradiation under the Xe lamp. This indicated that the dehydrogenation of alcohols during the reaction. Moreover,
amount of hydroxyl groups on the surface of the Zn2Al-LDH as shown in Fig. S11 (ESI†), the same results are achieved in the
support changed greatly after irradiation, which ultimately in situ FT-IR spectra of Au2Cu@Zn2Al-LDHs after the co-adsorption
affected the adsorption of the reactant molecules on the support; of benzyl alcohol and methanol. In addition, the silane coupling
this was proven by the in situ FT-IR spectra of Au2Cu@Zn2Al- agent EPTMS (g-(2,3-epoxypropoxy)propytrimethoxysilane) that
LDHs after the co-adsorption of benzyl alcohol and methanol, has a negligible effect on the reaction was chosen, and the
Phys. Chem. Chem. Phys.
This journal is © the Owner Societies 2019

View Article Online
PCCP
Paper
Au2Cu@MgAl-LDH catalyst was silanized to cover the surface 3.5 Monitoring the surface reaction
hydroxyl groups of the support. It was found that the yield of
To understand the activation process of the reactants on the
catalyst surface, the evolution of the products during the time
course of the oxidative esterification of benzyl alcohol over the
Au2Cu@Mg3Al-LDH photocatalyst under irradiation and in the
dark was conducted (Fig. S13, ESI†). It is a remarkable fact that
formaldehyde was not detected by methanol oxidation, and
benzaldehyde was the only intermediate during the reaction
course both under light irradiation and in the dark. This
demonstrates that benzyl alcohol oxidation is easier than methanol
oxidation over this photocatalyst, with the implication that the
energy of the Au2Cu@Mg3Al-LDH photocatalyst is not enough to
activate the methanol molecules to dehydrogenate under this
reaction condition. It is concluded that benzyl alcohol is first
oxidized to benzaldehyde during the photocatalysis process; then,
benzaldehyde further reacts with methanol molecules to achieve
methyl benzoate on the surface of the catalyst.
In addition, in situ FT-IR spectra after the coadsorption of
benzyl alcohol and methanol on Au2Cu@Mg3Al-LDHs were
recorded (Fig. 4d and Fig. S14, ESI†) to further investigate the
activation of reactants (detailed explanation shown in the ESI†).
The decrease in the intensity of the O–H stretching vibration
band of benzyl alcohol and methanol at 3643 cmÀ1 indicates
that the hydroxyl groups of benzyl alcohol and methanol have
been dehydrogenated by the cleavage of the O–H bonds.33 The
gradual decrease in the intensity of the carbon–carbon double
bond and C–H bond deformation vibrations of the aromatic ring
of benzyl alcohol at 1620 cmÀ1, 1538 cmÀ1 and 1313 cmÀ1 34
indicates that some benzyl alcohol molecules have been desorbed
after the reaction process.35 The appearance and increase in
the intensity of the –CQO– bond stretching vibrations of
methyl benzoate decreased significantly from 96.4% to 43.9%
under visible-light irradiation and from 59.4% to 20.4% in the
dark (Fig. 3a), indicating that the alcohol molecules were
adsorbed on the surface of the support through hydroxyl groups
because the adsorption of alcohol molecules will be inhibited if
the surface hydroxyl groups of the support are covered.
On the basis of the above-mentioned results, it was concluded
that both the support and the metal contributed to high activity.
The support with acid–base pairs affected the adsorption of reactant
molecules, in which the alcohol molecules were adsorbed on
the basic sites of the support (Mg–OHdÀ); this could promote
the cleavage of the O–H bond to give an alkoxide intermediate
at the surface and form a water molecule during the reaction.
Then, the formed water molecule was adsorbed on the acidic
site of the support (Al–OHd+) to promote the recovery of hydro-
xyl groups on the catalyst surface to determine the overall
reaction rate completely.
3.4 Role of K2CO3
It was found that equivalent activities can be achieved when the
reaction temperature and time were increased to 55 1C and
18 h, respectively, in the absence of K2CO3 (Table 1, entry 10).
When only K2CO3 was added, slight activity was achieved both
under irradiation and in the dark (Table 1, entry 11), which
proved that the role of K2CO3 in this photocatalytic system is
that of a co-catalyst to facilitate the dehydrogenation of alcohol
because of its basic property as equivalent results were also
obtained by increasing the reaction temperature under alkaline-
free conditions.
Scheme 1 Schematic diagram of the proposed reaction mechanism.
Phys. Chem. Chem. Phys.
This journal is © the Owner Societies 2019

View Article Online
Paper
PCCP
benzaldehyde at 1715 cmÀ1 with the increase in reaction time from (IV). Third, the alloy site activates O2 and produces O2 À radicals;
0 min to 90 min indicate that the intermediate benzaldehyde is first the formed aldehyde can condense with methanol to form a
generated.36 The blue frequency shift of this band compared to the hemiacetal intermediate.38–40 Then, methyl benzoate and a water
IR spectra of neat benzaldehyde (1710–1630 cmÀ1) indicates that molecule are achieved by the oxidative dehydrogenation of the
the carbonyl oxygen atoms of benzaldehyde are associated hemiacetal intermediate (V). Finally, the rapid oxidation of metal
ꢀ
À
ꢀ
with the basic sites of the support because the coordination of hydride by the O2
radicals forms one molecule of water
C–H of benzaldehyde with unsaturated Au–Cu alloy NP active promoted by the acidic sites of the support while recovering
sites to form a metal–H bond can result in a decrease in the the initial metal active sites and hydroxyl groups on the support
CQO bond length of benzaldehyde. This provides another surface for another reaction cycle (VI, VII). Besides, it is noted
favourable evidence that the benzyl alcohol molecules are that the carbon balance of the reaction is satisfied within the
associated with the basic sites of the support of Au2Cu@M- experimental error (Table S4, ESI†).
g3Al-LDHs at the beginning of the catalytic reaction rather than
being associated with the Au–Cu alloy NP active sites; this is
because a red shift for the CQO bond is expected with respect Table
2
Photocatalytic selective oxidative esterification of aromatic
alcohols with aliphatic alcoholsa
to neat benzaldehyde when benzyl alcohol molecules are asso-
ciated with the Au–Cu alloy NP active sites as the coordination
of the carbonyl oxygen atoms with metal atoms can result in an
increase in the bond length. The proportional decrease in the
intensity of the C–H stretching vibration band at 2939 cmÀ1
and 2833 cmÀ1 indicates the cleavage of the C–H bond of
methylene of benzyl alcohol during the catalytic process. Then,
the appearance and increase in the intensity of the –C–O– bond
and –COO– bond stretching vibrations of methyl benzoate at
1749 cmÀ1 and 1422 cmÀ1, respectively,37 with the increase in
reaction time from 90 min to 400 min indicate that methyl
benzoate is then generated after nucleophilic attack at the
carbonyl carbon of intermediate benzaldehyde by the methoxy
group of methanol. The blue frequency shift of the –CQO–
bond compared to the IR spectra of neat methyl benzoate (1730–
1725 cmÀ1) indicates that the carbonyl oxygen atoms of methyl
benzoate are associated with the basic sites of the support. Mean-
while, the blue frequency shift of the C–H stretching vibration band
at 2939 cmÀ1 and 2833 cmÀ1 indicates that the inductive effect of
the C–OCH3 band is generated due to the nucleophilic addition of
the methoxy group of methanol with the carbonyl carbon, which
can result in a decrease in the bond length.
Entry Reactant
1b
Product
Yieldi (%) T (1C)/t (h)
96.4 (59.4) 30/3
2b
3b
4b
97.9 (78.8) 30/24
87.6 (68.5) 30/24
53.0 (26.3) 30/24
5c
6d
7e
8f
44.6 (23.9) 30/24
32.7 (10.4) 50/24
66.6 (25.2) 60/24
3.6 Proposed mechanism
On the basis of the above-mentioned analysis, we proposed a
possible mechanism for the oxidative esterification of benzyl
alcohol, illustrating that the Au–Cu alloy NPs as the light
absorption sites (Scheme 1) determined the rate-limiting step
of the reaction when the support sites were identical, and the
Mg3Al-LDH support as the reactant molecule adsorption sites
could drive the overall reaction rate. The reaction proceeds
through the synergistic effect between the Au–Cu alloy NPs and
the base–acid sites on the surface of the Mg3Al-LDH support.
First, benzyl alcohol adsorbed on the basic sites of the MgAl-LDH
support (Mg–OHdÀ) promotes the cleavage of the O–H bond of
benzyl alcohol to give an alkoxide intermediate at the surface and
form a water molecule (I, II). Second, the intermediate undergoes
coordination with the unsaturated metal active sites due to the
40.2 (14.0) 60/24
58.7 (30.7) 50/30
90.3 (51.7) 40/5
9g
10h
a
Reaction conditions: photocatalyst, 10 mg; aromatic alcohols, 0.5 mmol;
solvent, 5 ml; K2CO3, 0.2 mmol; O2, atmospheric pressure; light intensity
LSPR effect of alloy NPs to form a metal–H bond to afford (cut-off wavelength of 420 nm of Xe lamp), 0.5 W cmÀ2
unsteady metal-alcoholate-LDH species (III). Then, the light-
excited electrons can promote hydrogen abstraction from a-H
of the metal-alcoholate-LDH species to generate aldehyde species methanol, 50 ml. The values in parentheses are the results in the dark.
.
Methanol as
b
c
d
e
solvent. Ethanol as solvent. 1-Propanol as solvent. 1-Hexyl alcohol
f
g
h
as solvent. 1-Octanol as solvent. 1-Heptane as solvent. Methanol as
solvent, photocatalyst, 100 mg; benzyl alcohol, 5 mmol; K2CO3, 2 mmol;
i
Phys. Chem. Chem. Phys.
This journal is © the Owner Societies 2019

View Article Online
PCCP
Paper
3.7 Synthesis of different aromatic esters
overall reaction system completely. This simple, cost-effective, and
environmentally benign catalytic process exhibits a great potential
in the oxidative esterification of aromatic alcohols.
To investigate the universality of the Au2Cu@Mg3Al-LDH
photocatalyst in other oxidative esterifications (Table 2), the
chlorine and nitro groups were situated in the para-position of
benzyl alcohol; the ester yields of up to 97.9% and 87.6% were
achieved (entries 2 and 3) under visible-light irradiation. When
Conflicts of interest
the methoxy group was situated in the ortho-position of benzyl There are no conflicts to declare.
alcohol, the ester yield declined to 53.0% (entry 4) due to the
steric hindrance effect of the substituent. Besides, the oxidative
esterifications of benzyl alcohol with ethanol and 1-propanol
Acknowledgements
were investigated, which produced the corresponding aryl
This work was supported by the National Natural Science
esters in 44.6% and 32.7% yields by using the Au2Cu@Mg3Al-
LDH catalyst under visible-light irradiation (entries 5 and 6).
Here, it is worth noting that we could also achieve the yields of
66.6% and 40.2% by using visible-light irradiation in esterifica-
tion with long-chain aliphatic 1-hexyl alcohol and 1-octanol
(entries 7 and 8), respectively, which are more challengeable
Foundation of China (No. 21703276 and 21773284), the Shanxi
Science and Technology Department (No. 201801D221093), and
the Hundred Talents Program of the Chinese Academy of
Sciences and Shanxi Province.
because the uniform dispersion of the catalyst in the reaction Notes and references
solution will be inhibited and the dehydrogenation of aliphatic
alcohols will become harder; this is because the more the
1 J. Otera, Chem. Rev., 1993, 93, 1449–1470.
2 C. E. Rehbergand and C. H. Fisher, J. Am. Chem. Soc., 1944,
carbon atoms, the greater the viscosity and the greater the
ability to donate electrons. Furthermore, the self-esterification
of benzyl alcohol afforded the corresponding ester with an
excellent yield of up to 58.7% under visible-light irradiation
(entry 9). Although the selective oxidative esterifications always
require high reaction temperatures and high oxygen pressures,
as reported in the literature,41,42 the milder conditions in the
present study for oxidative esterification reactions are rarely
reported. In a word, excellent universality of this photocatalyst
in other oxidative esterifications was achieved. In addition, we
performed an expanded 10-fold reaction system (0.5 g scale) for
the oxidative esterification of benzyl alcohol with methanol to
demonstrate the catalytic utility of our catalyst, as shown in
Table 2, entry 10; the previous yield was verified up to 90.3%
under visible-light irradiation.
66, 1203–1207.
3 R. C. Larock, Comprehensive organic transformations: a guide
to functional group preparations, Wiley-VCH, New York, N.Y.,
1999.
4 A. Brennfu¨hrer, H. Neumann and M. Beller, Angew. Chem.,
Int. Ed., 2009, 48, 4114–4133.
5 T. Hayashi, T. Inagaki, N. Itayama and H. Baba, Catal.
Today, 2006, 117, 210–213.
6 E. Taarning, A. T. Madsen, J. M. Marchetti, K. Egeblad and
C. H. Christensen, Green Chem., 2008, 10, 408–414.
7 Y. Hao, Y. Chong, S. Li and H. Yang, J. Phys. Chem. C, 2012,
116, 6512–6519.
8 H. Miyamura, T. Yasukawa and S. Kobayashi, Green Chem.,
2010, 12, 776–778.
9 P. Liu, C. Li and E. J. M. Hensen, Chem. – Eur. J., 2012, 18,
12122–12129.
10 G. Zhao, M. Ling, H. Hu, M. Deng, Q. Xue and Y. Lu, Green
Chem., 2011, 13, 3088–3092.
4. Conclusions
In summary, Au2Cu@Mg3Al-LDHs were prepared by utilizing 11 Q. Xiao, Z. Liu, A. Bo, S. Zavahir, S. Sarina, S. Bottle, J. D.
the ‘‘memory effect’’ of LDHs. Also, the impregnation–reduction Riches and H. Zhu, J. Am. Chem. Soc., 2015, 137, 1956–1966.
of gold and copper salts was found to afford an efficient 12 J. Feng, C. Ma, P. J. Miedziak, J. K. Edwards, G. L. Brett,
photocatalyst for the oxidative esterification of benzyl alcohol
with methanol under visible-light irradiation. The excellent
D. Li, Y. Du, D. J. Morganb and G. J. Hutchings, Dalton
Trans., 2013, 42, 14498–14508.
photocatalytic performance could be attributed to the synergistic 13 Z. Wang, Y. Song, J. Zou, L. Li, Y. Yu and L. Wu, Catal. Sci.
effects of two factors: (1) the interaction of light-excited electrons Technol., 2018, 8, 268–275.
from the alloy NPs with the intermediate molecules afforded 14 X. Liu, A. Wang, L. Li, T. Zhang, C. Y. Mou and J. F. Lee,
metal-alcoholate-LDH species to promote the cleavage of the C–H J. Catal., 2011, 278, 288–296.
bond of the reactant molecules, which could drive the rate- 15 Y. Sugano, Y. Shiraishi, D. Tsukamoto, S. Ichikawa, S. Tanaka
limiting step of the reaction under the premise that the support and T. Hirai, Angew. Chem., Int. Ed., 2013, 52, 5295–5299.
sites appeared identical. (2) The acid–base pairs of Mg3Al-LDHs 16 X. Liu, B. Tang, J. Long, W. Zhan, X. Liu and Z. Mirza, Sci.
could facilitate the adsorption of the reactant molecules. The Bull., 2018, 63, 502–524.
basic sites adsorbed benzyl alcohol molecules to provoke the 17 G. Chen, R. Gao, Y. Zhao, Z. Li, G. I. N. Waterhouse, R. Shi,
cleavage of the O–H bond, the acidic sites promoted the formation
of water molecules and the recovery of hydroxyl groups on the
catalyst surface at the end of the reaction especially promoted the
J. Zhao, M. Zhang, L. Shang, G. Sheng, X. Zhang, X. Wen,
L. Z. Wu, C. H. Tung and T. Zhang, Adv. Mater., 2018,
30, 1704663.
Phys. Chem. Chem. Phys.
This journal is © the Owner Societies 2019

View Article Online
Paper
PCCP
18 Z. Lia, J. Liud, Y. Zhao, R. Shi, G. I. N. Waterhousef, Y. Wang, L. Z. 30 X. Wang, R. Wang, X. Gu, J. Jia and Z. Zheng, Catal. Sci.
Wu, C. H. Tunga and T. Zhang, Nano Energy, 2019, 60, 467–475. Technol., 2019, 9, 1774–1778.
19 Z. Li, J. Liu, Y. Zhao, G. I. N. Waterhouse, G. Chen, R. Shi, 31 R. Wang, K. Hashimoto, A. Fujishima, M. Chikuni, E. Kojima,
X. Zhang, X. Liu, Y. Wei, X. D. Wen, L. Z. Wu, C. H. Tung and
T. Zhang, Adv. Mater., 2018, 30, 1800527.
A. Kitamura, M. Shimohigoshiand and T. Watanabe, Adv.
Mater., 1998, 10, 135–138.
20 Y. Zhao, Y. Zhao, G. I. N. Waterhouse, L. Zheng, X. Z. Cao, 32 L. Hickey, J. T. Kloprogge and R. L. Frost, J. Mater. Sci., 2000,
F. Teng, L. Z. Wu, C. H. Tung, D. O’Hare and T. Zhang, Adv.
Mater., 2017, 29, 1703828.
21 J. B. Han, J. Lu, M. Wei, Z. L. Wang and X. Duan, Chem.
Commun., 2008, 5188–5190.
35, 4347–4355.
33 T. Ebata, T. Watanabe and N. Mikami, J. Phys. Chem., 1995,
99, 5761–5764.
34 E. B. WiLson, Phys. Rev., 1934, 45, 706–714.
22 R. Long, K. Mao, M. Gong, S. Zhou, J. Hu, M. Zhi, Y. You, 35 T. Beutel, J. Chem. Soc., Faraday Trans., 1998, 94, 985–993.
S. Bai, J. Jiang, Q. Zhang, X. Wu and Y. Xiong, Angew. Chem., 36 D. M. Meier, A. Urakawa and A. Baiker, J. Phys. Chem. C,
Int. Ed., 2014, 53, 3205–3209.
23 S. Sarina, E. R. Waclawik and H. Zhu, Green Chem., 2013, 15, 37 N. Sundaraganesan and B. D. Joshua, Spectrochim. Acta, Part
1814–1833. A, 2007, 68, 771–777.
24 S. Penner, P. Bera, S. Pedersen, L. T. Ngo, J. J. W. Harris and 38 A. Izumi, Y. Obora, S. Sakaguchi and Y. Ishii, Tetrahedron
C. T. Campbell, J. Phys. Chem. B, 2006, 110, 24577–24584. Lett., 2006, 47, 9199–9201.
25 Q. Xiao, S. Sarina, E. R. Waclawik, J. Jia, J. Chang, J. D. Riches, 39 M. Nielsen, H. Junge, A. Kammer and M. Beller, Angew.
H. Wu, Z. Zhengand and H. Zhu, ACS Catal., 2016, 6, 1744–1753. Chem., Int. Ed., 2012, 51, 5711–5713.
26 S. Pramanik, M. K. Mishra and G. De, CrystEngComm, 2014, 40 R. V. Jagadeesh, H. Junge, M. M. Pohl, J. Radnik,
2009, 113, 21849–21855.
16, 56–63.
27 N. Siemer, A. Lu¨ken, M. Zalibera, J. Frenzel, D. Munoz-
A. Bru¨ckner and M. Beller, J. Am. Chem. Soc., 2013, 135,
10776–10782.
˜
Santiburcio, A. Savitsky, W. Lubitz, M. Muhler, D. Marx and 41 S. Gowrisankar, H. Neumann and M. Beller, Angew. Chem.,
J. Strunk, J. Am. Chem. Soc., 2018, 140, 18082–18092.
28 B. M. Bhat, Orient. J. Chem., 2012, 28, 1751–1760.
29 A. H. Padmasri, A. Venugopal, V. D. Kumari, K. S. R. Rao
and P. K. Rao, J. Mol. Catal. A: Chem., 2002, 188, 255–265.
Int. Ed., 2011, 50, 5139–5143.
42 Y. Zhao, G. Chen, T. Bian, C. Zhou, G. I. N. Waterhouse,
L. Z. Wu, C. H. Tung, L. J. Smith, D. O’Hare and T. Zhang,
Adv. Mater., 2015, 27, 7824–7831.
Phys. Chem. Chem. Phys.
This journal is © the Owner Societies 2019