Efficient imine synthesis from oxidative coupling of alcohols and amines under air atmosphere catalysed by Zn-doped Al2O3 supported Au nanoparticles

Article abstract of DOI:10.1016/j.jcat.2019.07.027

Direct oxidative coupling of alcohols and amines is regarded as an effective and green approach for imine synthesis under mild conditions. In this work, Zn-doped γ-Al₂O₃ supported Au nanoparticles was demonstrated as highly active and selective heterogeneous catalyst for a series of imine productions with good to excellent yields, from alcohols and amines via direct oxidative coupling under air atmosphere without extra base additives. Various physicochemical techniques, including ICP-MS, XRD, N₂ physisorption, TEM, XPS and CO₂-/NH₃-TPD, were used to study the properties of the catalysts. Well-dispersed Au⁰ nanoparticles with a mean size of ca. 2.9 nm were found highly effective in activating alcohols in the presence of reactive amines. The amount of Zn²⁺ dopant and the calcination temperature of support during catalyst preparation showed crucial impact on tuning the intrinsic activity for oxidation of benzyl alcohol to benzaldehyde (i.e., the rate-determining step for the model reaction), which was disclosed to be related with the active surface oxygen species and the acidic-basic property of support. The 0.4% Au/Zn_0.02Al₂O₃ catalyst calcined at 400 °C exhibited the highest TOF (39.1 h^?1) at 60 °C based on a >99% yield to benzylideneaniline among all the ever-reported Au-based catalysts. Moreover, this catalyst could afford 98% yield to benzylideneaniline at only 30 °C and work effectively for the gram-scale synthesis. It also showed considerable stability after five consecutive recycling.

Full text of DOI:10.1016/j.jcat.2019.07.027

Journal of Catalysis 377 (2019) 110–121
Contents lists available at ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
Efficient imine synthesis from oxidative coupling of alcohols and amines
under air atmosphere catalysed by Zn-doped Al O supported Au
2
3
nanoparticles
Shipeng Wu a,1, Weixiao Sun a,1, Junjie Chen , Jinghan Zhao , Qiue Cao a,b, Wenhao Fang a, , Qihua Zhao a,
a
a
⇑
⇑
a
School of Chemical Science and Technology, Key Laboratory of Medicinal Chemistry for Natural Resource – Ministry of Education, Functional Molecules Analysis
and Biotransformation Key Laboratory of Universities in Yunnan Province, Yunnan University, 2 North Cuihu Road, 650091 Kunming, China
b
National Demonstration Center for Experimental Chemistry and Chemical Engineering Education, Yunnan University, China
a r t i c l e i n f o
a b s t r a c t
Article history:
Direct oxidative coupling of alcohols and amines is regarded as an effective and green approach for imine
synthesis under mild conditions. In this work, Zn-doped -Al supported Au nanoparticles was demon-
strated as highly active and selective heterogeneous catalyst for a series of imine productions with good
Received 9 June 2019
Revised 13 July 2019
Accepted 13 July 2019
c
2 3
O
to excellent yields, from alcohols and amines via direct oxidative coupling under air atmosphere without
extra base additives. Various physicochemical techniques, including ICP-MS, XRD, N
XPS and CO -/NH -TPD, were used to study the properties of the catalysts. Well-dispersed Au nanopar-
ticles with a mean size of ca. 2.9 nm were found highly effective in activating alcohols in the presence of
reactive amines. The amount of Zn dopant and the calcination temperature of support during catalyst
preparation showed crucial impact on tuning the intrinsic activity for oxidation of benzyl alcohol to
benzaldehyde (i.e., the rate-determining step for the model reaction), which was disclosed to be related
2
physisorption, TEM,
0
2
3
Keywords:
Imine
2
+
Oxidative coupling
Au nanoparticles
Acidity-basicity
Surface oxygen species
Heterogeneous catalysis
with the active surface oxygen species and the acidic-basic property of support. The 0.4% Au/Zn0.02Al
2
O
3
À1
catalyst calcined at 400 °C exhibited the highest TOF (39.1 h ) at 60 °C based on a >99% yield to benzyli-
deneaniline among all the ever-reported Au-based catalysts. Moreover, this catalyst could afford 98%
yield to benzylideneaniline at only 30 °C and work effectively for the gram-scale synthesis. It also showed
considerable stability after five consecutive recycling.
Ó 2019 Elsevier Inc. All rights reserved.
1
. Introduction
Imines stand as a class of important N-containing organic inter-
often receive criticisms from a lack of selectivity, resulting conse-
quently in complexity and high cost for extra separation and purifi-
cation. Among others, direct coupling of alcohols and amines under
oxidative conditions is regarded as a very ideal strategy toward
imine synthesis owning to its friendly environmental impact, high
atom economy and fast response [18–23]. Moreover, this green
and practical method allows transforming a wide spectrum of reac-
tive, low-cost and easily available alcohols to diverse imines at
lower temperature and pressure but with only water formed as
the by-product.
During the last decade, a few supported Pd [24,25], Pt [26] and
Ru [27,28] catalysts have been reported for direct oxidative cou-
pling of alcohols and amines due to the high reactivity of noble-
metal nanoparticles in activating alcohols. Nevertheless, most of
mediates and have been extensively used in manufacture of phar-
maceuticals, bioactive compounds and agrochemicals [1–8].
Traditional synthesis of imines relies on the condensation of car-
bonyl compounds with amines. But such a process is usually
assisted with acid medium, leading to several environmental
issues due to use of hazardous regents and/or formation of toxic
by-products. On light of that, alternative methodologies have been
dedicated to direct synthesis of imines mainly through oxidation of
secondary amines [9–11], dimerization of primary amines [12–14],
and other approaches [15–17]. Nonetheless, those approaches
2
those catalytic systems have to work effectively in pure O atmo-
⇑
Corresponding authors.
E-mail addresses: wenhao.fang@ynu.edu.cn (W. Fang), qhzhao@ynu.edu.cn
sphere and at high temperatures. And moreover, excessive base
additives are most often required. Therefore, from the point view
of high atom economy and positive environmental impact, it is of
(
Q. Zhao).
1
S. Wu and W. Sun contributed equally to this work.
https://doi.org/10.1016/j.jcat.2019.07.027
021-9517/Ó 2019 Elsevier Inc. All rights reserved.
0

S. Wu et al. / Journal of Catalysis 377 (2019) 110–121
111
reaction pathway and plausible reaction mechanism remain
unclear in the point view of molecular catalysis.
Catalyst
H O
H O
2
2
Herein, in this work we develop for the first time Zn-doped
c
-Al
new supported Au catalyst for imine synthesis through direct
oxidative coupling of alcohol and amine. The Au/Zn Al catalyst
2 3 X 2 3
O supported Au nanoparticles (Au/Zn Al O ), which is a
Step 1: Rate-determining
Step 2: Fast, without catalyst
2 3
O
Scheme 1. Schematic illustration of the reaction pathway of oxidative coupling of
X
benzyl alcohol and aniline.
shows superior activity, selectivity, stability and scope towards
imine productions by using very mild temperature and oxidant,
offering a highest TOF ever reported for Au-based catalysts in liter-
ature. We will demonstrate the role of Au nanoparticles, the unique
effect of Zn-doping and calcination conditions on support, investi-
gate reaction parameters and reusability, and finally elucidate the
importance to develop new noble metal based catalysts and
technologies for coupling alcohols with amines under oxidative
conditions.
As described in Scheme 1, it has been well illustrated for the
model reaction that coupling benzyl alcohol with aniline under-
goes a tandem reaction pathway [18–23], in which oxidation of
alcohol to aldehyde is a slow and catalysed step (i.e., identified
as the rate-determining step), while the subsequent condensation
of aldehyde with amine is a fast and sometimes uncatalysed step.
In such a context, design and fabrication of active sites to selec-
tively oxidize alcohols to aldehydes remains no doubt critical for
obtaining high yield to imines. As expected, the catalyst property
would be decisive not only for conversion of alcohols but also
selectivity to imines.
Supported Au nanoparticles have been widely recognised as
promising catalysts for diverse mild and efficient selective aerobic
oxidations in liquid phase [29–34]. As Au-based catalysts can offer
better resistance to water and oxygen, it is generally accepted that
Au catalysts are more selective and stable for aerobic oxidation of
many different alcohols rather than the conventional Pd and Pt oxi-
dation catalysts [35–38]. To the best of our knowledge, there are
only a few Au-based catalysts reported for oxidative coupling of
alcohol and amine, as summarised in Table 1. All the catalysts
showed quite satisfying selectivity (95–99%) to benzylideneaniline
but rather different conversion of benzyl alcohol. Au/HAP pre-
sented the optimal performance among the monometallic Au cat-
X 2 3
structure-activity relationship of the Au/Zn Al O catalysts.
2
. Experimental
2.1. Catalyst preparation
Zn-doped Al
i.e., Zn/Al ratio in mol.) was prepared by co-precipitation method.
A mixed solution of Zn(NO O and Al(NO Á9H O was dis-
Á6H
solved in deionized water, and then co-precipitated by dropwise
addition of NH OH solution (28 wt%) under stirring (600 rmp) till
2 3 x 2 3
O donated as Zn Al O (x = 0.01, 0.02 and 0.04,
3
)
2
2
3
)
3
2
4
pH value of the mixture reached 9. The obtained slurry was stirred
at room temperature for 2 h. The solids were recovered by filtra-
tion and washed by excessive deionized water when the filtrate
presented a neutral pH value. Finally, the solids were oven-dried
at 110 °C for 12 h followed by a calcination at (T = 300, 400, 500,
6
50 and 800 °C, with a ramp of 2 °C/min) under air flow for 5 h.
Al and ZnO were also prepared by this protocol.
Supported Au nanocatalysts with a loading of 0.4 wt% were pre-
2 3
O
pared by immobilization of the PVA-stabilized colloidal Au
nanoparticles onto various supports. Typically, a poly(vinyl alco-
hol) (MW = 10,000) aqueous solution was added into a HAuCl
4
solution (PVA/Au = 1.2/1, in w/w) under stirring. After 15 min, a
suspension of powdery support was added and the mixture was
vigorously stirred at room temperature for 15 min. Then a freshly
alysts at mild temperatures using O
2
as oxidant [39]. A high TOF
À1
(
33.0 h ) with 99% yield to benzylideneaniline was reported. It
was demonstrated that a delicate cooperation between metallic
Au and the acidic-basic sites of the HAP surface may play a key role
in determining the efficiency and compatibility of the Au/HAP cat-
alyst for not only the rate-determining alcohol oxidation but also
for the subsequent condensation step of the tandem reaction.
4 4
made NaBH solution (NaBH /Au = 4/1, in w/w) was slowly intro-
duced into the mixture to yield a dark-purple colloidal solution.
After it was stirred for another 2 h, the solid catalyst was recovered
by filtration and washed repeatedly with excessive hot deionized
water (90 °C) to remove PVA and with excessive deionized water
2
Among the bimetallic Au-Pd catalysts, Au-Pd@ZrO exhibited the
+
À
to remove Na and Cl ions. Finally, the resultant Au nanocatalyst
optimal performance via a one-pot, two-step process for tandem
imine synthesis followed by the benzyl alcohol oxidation, affording
was dried in vacuum at 50 °C for 24 h.
À1
a comparable TOF (35.2 h ) but with a lower yield (94%) to imine
[
40]. It was disclosed that the superior catalytic properties of alloy
2.2. Catalytic reaction
NP catalyst was because the surface of alloy NPs showed higher
charge heterogeneity than that of pure metal NPs. However, it
can be concluded that most of those Au catalysts still suffer from
addition of base additives, insufficient conversion of alcohol and
uncertainty in reusability. Hence those above would further ham-
per the green footprint of imine production. Furthermore, the
Oxidative coupling of alcohols and amines was performed in a
TM
WP-TEC-1020HC parallel reactor provided by WATTCAS (WAT-
TECS LAB EQUIPMENT CO., LTD.). In a typical protocol, 0.5 mmol
benzyl alcohol, 0.75 mmol aniline, 5 mL toluene and 40 mg catalyst
were added successively. And then the reaction mixture was
Table 1
Summary of supported Au-based catalysts for benzylideneaniline synthesis from oxidative coupling of benzyl alcohol and aniline.
À1
Reference
Catalyst
Au/HAP
T (°C)
Oxidant (1 bar)
Base
Time (h)
Conv. (%)
Yield (%)
TOF (h
)
Reuse times
Stability
[39]
[41]
[42]
[43]
[44]
[40]
60
20
60
40
60
40
60
O
O
O
O
O
2
2
2
2
2
None
KOCH
KOCH
NaOH
NaOH
None
None
3
99
7
>99
7
33.0
14.4
0.9
5
Yes
Au/TiO
Au/ZrO
AuPd /resin
2
3
3
24
24
12
12
8
Not done
Not done
6
Not done
3
5
Unknown
Unknown
No
Unknown
Yes
2
32
80
74
94
99
32
80
70
94
>99
b
4
2.2
b
PICB-Au/Pd
Au-Pd@ZrO
Au/Zn0.02Al
3.9
a
35.2b
39.1
2
Air
Air
This work
2
O
3
8
Yes
a
Two stepped reaction of oxidation of benzyl alcohol to benzaldehyde (7 h) and condensation of benzaldehyde with aniline (1 h).
Based on the total amount of Au and Pd.
b

1
12
S. Wu et al. / Journal of Catalysis 377 (2019) 110–121
stirred (900 rpm) at 60 °C under air (1 atm) for 8 h. The tube was
immediately cooled in ice water and the reaction mixture was
analysed using a Thermo Scientific TRACE 1310 gas chromatograph
microscopy (TEM) was conducted on a Philips-FEITECNAI F30 field
emission electron microscope with an acceleration voltage of
300 kV.
(
GC) equipped with a flame ionization detector (FID) and a TR-5
column (30 m  0.32 mm  0.25 m). The quantification was car-
l
3
. Results and discussion
ried out using the external standard method. A series of toluene
solutions in different concentrations of benzyl alcohol, benzalde-
hyde or benzylideneaniline were used as external standard, respec-
tively. Each reaction was at least repeated twice to guarantee a
reproducible result. The conversion of benzyl alcohol, the selectiv-
ity to alkylamines and the yield of imine were calculated, on the
basis of a carbon balance. For the reusability test, the catalyst
was separated by centrifugation (8000 rpm for 10 min), washed
three times in turn with ethanol (100 mL) and water (100 mL),
dried at 50 °C for 24 h under vacuum, and was used for the next
run.
0
2+
3
.1. Importance of Au nanoparticles and Zn doping
N-benzylideneaniline is selected as a model imine from direct
oxidative coupling of benzyl alcohol and aniline (Eq. 1), and a ser-
ies of supported Au catalysts are evaluated (Table 2). The supports
2 3
were lab-made and calcined under air flow at 400 °C. Au/Al O and
Au/ZnO give 6% and 12% conversion of alcohol, respectively. While
selectivity of imine is found at 99% over both catalysts with alde-
hyde formed as the sole by-product. Adding a small amount of
Zn in Al
icant increase of alcohol conversion and excellent retention of
imine selectivity (>99%). The Au/Zn0.02Al catalyst shows the
2 3
O support (Zn/Al, 0.01–0.04 in mol.) brings about signif-
2.3. Characterization methods
2 3
O
optimal yield at 73% to benzylideneaniline at 60 °C under air atmo-
sphere. To be noted, the bare supports are found inactive to imine
synthesis (Table S1), showing that Au nanoparticles are essential
and work as the active sites. Interestingly, adding the same amount
of Zn (0.02 mol.%) to Au rather than Al O leads to a much lower
2 3
alcohol conversion (30%). This clearly confirms that the enhance-
ment on catalytic activity can be only realized by incorporating a
The textural property of support materials was studied by N
2
adsorption-desorption at 77 K using BET and BJH methods on a
Micromeritics TriStar II 3020 Surface-Area and Porosimetry ana-
lyzer. The metal loadings for Au, Zn and Al were precisely mea-
sured by inductively coupled plasma mass spectrometry (ICP-MS)
on an Agilent 7500a apparatus. X-ray diffraction (XRD) was con-
ducted using a Rigaku TTR III Diffractometer provided with Cu
2 3
suitable Zn to Al O , probably associated with the unique materials
Ka radiation (k = 1.5418 Å) and a beam voltage of 40 kV. The pat-
property.
terns were registered in the 2h domain (5–90°) with a measured
step of 0.02° and a time integration of 0.2 s. X-ray photoelectron
spectroscopy (XPS) was performed under ultra-high vacuum on a
To optimise conversion of benzyl alcohol, the catalyst mass is
increased. As shown in Table S2, conversion gradually goes up with
increasing catalyst mass from 40 mg to 80 mg. A total conversion is
obtained when the catalyst mass is doubled to 80 mg. This suggests
the difficulty in activating benzyl alcohol by using metal catalysts.
In addition, the reactant composition, i.e., the ratio of benzyl alco-
hol to aniline is believed to affect the yield of imine. Though the
stoichiometry of alcohol to amine for oxidative coupling reaction
is of 1 as used in some work [23,39], different stoichiometry are
also employed, i.e., alcohol/amine molar ratio is of 1.5 [4,25] or
0.5 [18,19]. Given that, the amount of aniline is reduced to 1.25
and 1.0 equiv. to investigate the influence. As shown in Table 2,
the decrease in the amount of aniline leads mainly to lowering
selectivity of imine, as well as conversion of benzyl alcohol. In
return, selectivity of benzaldehyde is increased. This implies a pro-
motion effect on yield to imine by an excessive use of aniline in the
Thermo Scientific Escalab 250Xi system provided with Al K
ation. The binding energy shift due to the surface charging was
adjusted using a reference to the C 1s line at 284.6 eV. CO and
NH temperature-programmed desorption (TPD) was carried out
a radi-
2
3
on a Micromeritics Autochem II Chemisorption analyzer. 50 mg
sample loaded in a quartz reactor was pre-treated with He at
2
00 °C for 1 h. After cooling to 50 °C, CO
performed by switching He to a 10% CO
mixed gas and then maintaining for 1 h. The gas-phase CO
2
or NH
/NH -90% He (in vol.)
/NH
3
adsorption was
2
3
2
3
was purged by He at the same temperature. TPD was then per-
formed in He flow by raising the temperature to 800 °C at a rate
2 3
of 10 °C/min. The desorbed CO or NH was detected on a
ThermoStar GSD 301 T2 mass spectrometer. Transmission electron
Table 2
2 3 X 2 3
Oxidative coupling of benzyl alcohol and aniline over Al O -, ZnO- and Zn Al O -supported Au catalysts.
(
Eq. 1)
Entry
Catalyst
Au/Al
Au/ZnO
Au/Zn0.01Al
Au/Zn0.02Al
Au/Zn0.04Al
Conv. 1 (%)
Select. 3 (%)
Select. 4 (%)
Yield 3 (%)
1
2
3
4
5
6
7
8
9
2
O
3
6
99
99
1
1
5
12
42
73
14
30
100
100
97
33
11
41
72
13
29
>99
97
90
32
2
2
2
O
O
O
3
3
3
>99
>99
>99
>99
>99
97
<1
<1
<1
<1
<1
3
Au-Zn0.02/Al
2
O
3
a
Au/Zn0.02Al
Au/Zn0.02Al
Au/Zn0.02Al
Au/Zn0.02Al
2
O
2
O
2
O
2
O
3
3
3
3
b
c
93
>99
7
<1
d
1
0
Reaction conditions: benzyl alcohol, 0.5 mmol; aniline, 0.75 mmol; catalyst, 40 mg; toluene, 5 mL; air, 1 atm; temperature, 60 °C; time, 8 h.
a
Catalyst, 80 mg.
Benzyl alcohol, 0.5 mmol; aniline, 0.625 mmol.
Benzyl alcohol, 0.5 mmol; aniline, 0.5 mmol.
Ar, 1 atm.
b
c
d

S. Wu et al. / Journal of Catalysis 377 (2019) 110–121
113
Table 3
2 3
Oxidative coupling of benzyl alcohol and aniline over Zn0.02Al O -supported noble metal catalysts.
Formation rate 3 (g g h )b
À1
Entry
Catalyst
Conv. 1 (%)
Select. 3 (%)
Select. 4 (%)
1
2
3
4
Au/Zn0.01Al
2
O
3
100
55
34
2
>99
92
99
<1
0
<1
<1
35.0
17.9
11.9
<1
a
Pd/Zn0.02Al
Ru/Zn0.02Al
Pt/Zn0.02Al
2
O
3
2
O
3
2
O
3
99
Reaction conditions: benzyl alcohol, 0.5 mmol; aniline, 0.75 mmol; catalyst, 80 mg; toluene, 5 mL; air, 1 atm; temperature, 60 °C; time, 8 h.
a
8
% selectivity to benzylaniline.
b
Based on the amount of noble metals.
present catalytic system, as supported by literature [20–22]. There-
after, some other noble metals (i.e., typical oxidation catalysts)
including Pd, Ru, Pt are loaded onto the optimal Zn0.02Al O sup-
2 3
port with the same metal loading (0.4 wt%) and further tested for
benzylideneaniline formation. As summarised in Table 3,
displays a relatively higher surface area probably due to the rela-
tively low crystallinity and abundant crystal defects of the
ZnAAlAO solid solution. It is usually believed that such morphol-
ogy could probably provide more reactive sites and be benefit to
effectively dispersing active Au nanoparticles [35]. ZnO displays a
2
Au/Zn0.02Al
2
O
3
shows the best catalytic performance with >99%
type-III isotherm and a low specific surface area (5 m /g).
yield to imine. Supported Ru and Pt could offer excellent selectivity
of imine but alcohol conversion is found much lower, i.e., 34% and
To further rationalise the role of Au nanoparticles and Zn dop-
0
ing, XPS, TEM, and CO
2
-/NH
3
-TPD are carried out. Metallic Au spe-
2
%, respectively. To be noted, Pd/Zn0.02Al
ate alcohol conversion at 55% but yield the secondary amine (i.e.,
% selectivity to benzylaniline) as by-product, in agreement with
2
O
3
may afford a moder-
cies are identified on surface of all the catalysts (Fig. S1, Table 5)
[29,30]. This confirms that Au nanoparticles in metallic state are
catalytic active sites. The surface Au loadings from XPS (ca.
0.35 wt%) are slightly lower than the bulk values from ICP-MS
8
Pd-based catalysts [25,45]. A highest formation rate of benzylide-
neaniline (35.0 g g h ) can be obtained over the Au/Zn0.02Al
catalyst taking into account the efficiency of active metal toward
imine formation, among different noble metals catalysts.
To understand the superior catalytic performances of the Au/
À1
2+
3+
2
O
3
(Table 4). Then, typical Zn (Fig. 3) and Al species (Fig. S1) are
clearly observed [49], revealing co-presence of ZnO and Al on
catalyst surface. And moreover, ZnO is found segregated onto the
surface (Table 4). In addition, the best Au/Zn0.02Al catalyst exhi-
bits a lower B.E. for the Au 4f7/2 peak, which is indicative of the
strong interaction between Au nanoparticles and the Zn0.02Al
2 3
O
2 3
O
Zn
X 2 3
Al O catalysts, some preliminary information on their compo-
sition and structure are analysed by ICP-MS, XRD and
N
2
2 3
O
physisorption. As listed in Table 4, Au is effectively loaded by
adsorption method onto different supports. ICP-MS shows the pre-
cise Au loadings nearly the same as the theoretical values (0.4 wt
support due to the electron transfer from support to active metal.
The O 1s peak can be deconvoluted into two bands (Fig. 3). The
band at a lower B.E. of ca. 530.6 eV is characteristic of lattice O2
À
%
). And a very small amount of Zn is confirmed to be successfully
added to Al by co-precipitation method. Al and ZnO present
typical diffraction peaks due to -Al (JCPDS #29-0063) and
hexagonal ZnO (JCPDS #36-1451), respectively (Fig. 1). While Zn-
doped Al display patterns identical to -Al without presence
of ZnO phase, probably due to the quite low loading of Zn. The
incorporation of ZnO into Al is confirmed by the shift of Al
species (i.e., O
a 2 3
species) in Al O or doped-Al oxides [50], while
2
O
3
2 3
O
the band at a higher B.E. of ca. 532.1 eV can be attributed to the
À
2À
c
2
O
3
surface oxygen species (i.e., O
b
species) mainly due to OH , O
À
and/or O ions [51]. To be noted, the B.E. of O 1s peak for
Au/ZnO is found lower but in consistent with the data from typical
2
O
3
c
2 3
O
b
ZnO in literature [52]. The O species are most often associated
2
O
3
2
O
3
with active oxygen vacancy sites, and have been shown as highly
active species for oxidative coupling of benzyl alcohol and aniline
[21,22,25]. A volcano-like relationship can be correlated in this
work between conversion of benzyl alcohol and the proportion of
peaks to lower 2h values, which is caused by the doping of Zn ions
in lattice (atomic radius: Zn, 0.74 Å > Al, 0.54 Å), resulting in an
expansion of the mean size of Al
shown in Fig. 2, both Al and Zn
tive type-IV isothermal profiles with an obvious H
located at a relative pressure of 0.4–0.9. This is indicative of meso-
porous materials. Moreover, pore size distribution obtained from
BJH isotherms confirms the relatively uniform mesoporous struc-
2
O
3
crystals (Table 4) [46,47]. As
Al supports exhibit distinc-
hysteresis loop
2
O
3
X
O
2 3
O
b
species on surface of the Au/Zn
version is obtained on Au/Zn0.02Al
of active O species.
Al
X 2
O
3
catalysts. An optimal con-
2
2 3
O due to its highest proportion
b
For this representative catalyst, TEM analysis shows small and
uniform Au nanoparticles highly dispersed on the support that
ture that can be attributed to accumulation of oxide nanoparticles
2 3
displays a typical thin-layered morphology due to doped Al O
2
[
48]. Then, high specific surface areas between 259 and 304 m /g
(Fig. 4-a) [53]. The size of Au nanoparticles is found narrowly dis-
tributed and the mean particle size is measured of 2.9 ± 0.2 nm.
HRTEM image (Fig. 4-b) reveals an interplanar spacing
(0.234 nm) identical to the (1 1 1) plane of Au. A geometric mor-
phology of truncated cubo-octahedral may be recognised. As a
matter of fact, such small and highly-dispersed Au nanoparticles
are obtained (Table 4), and Zn
with respect to Al , suggesting the influence of Zn doping in a
small amount. The BET surface area of Zn Al O generally decreases
compared to Al . This can be associated with large lattice distor-
X 2 3
Al O oxides show relevant values
2 3
O
X
2
2 3
O
2 3
tions generated from incorporation of Zn. The Zn0.02Al O oxide
Table 4
Composition and textural property of the Au/Zn
X
Al
2 3
O catalysts.
(nm)a
S
BET (m /g)b
2
V
(cm /g)b
3
D
P
(nm)b
Entry
Catalyst
Au loading (wt.%)
Zn/Al molar ratio
d,Al
2
O
3
P
ICP-MS
XPS
ICP-MS
XPS
1
2
3
4
5
Au/Al
2
O
3
0.39
0.40
0.40
0.39
0.36
0.34
0.35
0.36
0.34
0.33
À
À
4.2
5.0
5.2
6.0
À
298
259
304
281
5
0.49
0.31
0.27
0.31
0.02
5.3
4.2
3.7
4.2
Au/Zn0.01Al
Au/Zn0.02Al
Au/Zn0.04Al
Au/ZnO
2
O
2
O
2
O
3
3
3
0.01
0.02
0.04
À
0.02
0.03
0.05
À
19.6
a
Mean crystal size of Al
Data obtained from the bare supports.
2
O
3
calculated by Scherrer equation from the (4 4 0) plane of XRD patterns.
b

1
14
S. Wu et al. / Journal of Catalysis 377 (2019) 110–121
(
(
): Al O
2
3
): ZnO
a
a
b
c
b
c
d
d
shift
e
10
20
30
40
50
Theta ( )
60
70
80
90
62 64 66 68 70 72
2 Theta ( )
2
Fig. 1. XRD patterns with a zoomed region for the (a) Al
2
O
3
, (b) Zn0.01Al
2
O
3
, (c) Zn0.02Al
2
O
3
, (d) Zn0.04Al
2
O
3
, and (e) ZnO supports.
3
e
(0.02 cm /g, 19.6 nm)
2
5
m /g
e
3
d
c
(0.31 cm /g, 4.2 nm)
2
2
3
81 m /g
d
c
3
(0.27 cm /g, 3.7 nm)
2
04 m /g
2
2
59 m /g
b
a
b
(0.31 cm
3
/g, 4.2 nm)
2
2
98 m /g
3
a
(0.49 cm /g, 5.3 nm)
30
0
10
20
40
0
0.2
0.4
0.6
0.8
1
Pressure (P/P )
Pore width (nm)
0
Fig. 2. N
2
adsorption/desorption isotherms and the corresponding pore size distribution for (a) Al
2
O
3
, (b) Zn0.01Al
2
O
3
, (c) Zn0.02Al
2
O
3
, (d) Zn0.04Al
2
O
3
, and (e) ZnO supports.
Table 5
XPS parameters of Au 4f7/2, Al 2p3/2, Zn 2p3/2 and O 1 s lines for the Au/Zn
X
Al
2 3
O catalysts.
Entry
Catalyst
Bonding energy (eV)
b a b
O /(O + O ) (%)
Au 4f7/2
Al 2p3/2
Zn 2p3/2
a b
O 1s (O , O )
1
2
3
4
5
Au/Al
2
O
3
83.5
83.6
83.4
83.6
82.9
74.3
74.5
74.4
74.4
À
À
530.6, 532.1
530.7, 532.2
530.6, 532.1
530.7, 532.2
64.4
65.5
68.7
56.5
53.9
Au/Zn0.01Al
Au/Zn0.02Al
Au/Zn0.04Al
Au/ZnO
2
O
2
O
2
O
3
3
3
1021.9
1021.8
1021.9
1020.9
529.7, 531.2
À
bearing this morphology have been found highly efficient toward
aerobic oxidation of simple-to-complex alcohols [4,38,54].
The acidic-basic properties of a series of supports are examined
O2 ions [55]. While the broad NH
3
desorption peak located at
ca. 350–750 °C can generally be attributed to the medium-to-
strong Lewis acid sites related to Al or Zn cations [56]. With
increasing the Zn dopant both CO - and NH -desorption peaks
2 3
obviously become more intensive and shift to higher temperatures.
This suggests that the strength and number of acidic or basic sites
3
+
2+
2
+
by NH
acidic or basic sites can be measured, which could be related with
interactions between surface OH group and NH or CO probe
molecule, respectively. Zn-doped Al exhibits similar acidic
and basic sites to those of Al . The broad CO desorption peak
can be assigned to the contribution of the medium basic sites cen-
3
- and CO
2
-TPD. As displayed in Fig. 5, no apparent weak
3
2
2
O
3
of Zn
X 2 3
Al O supports may be enhanced. The moderate-strength
2
O
3
2
acidic and basic sites can be measured on the optimal Zn0.02Al O
2 3
support (Table S3). As already demonstrated in literature for
supported-Au catalysts, those supports bearing both acidic
and basic sites in an appropriate strength play a crucial role in
n+
2À
3+
2+
tred at ca. 350 °C due to the M – O pairs (M = Al or Zn ) and
the strong basic sites centred at ca. 440 °C due to the isolated

S. Wu et al. / Journal of Catalysis 377 (2019) 110–121
115
Zn 2p3/2
O 1s
Zn 2p
Zn 2p1/2
e
d
O
e
d
c
O
O
O
c
b
b
a
5
35 534 533 532 531 530 529 528
1048 1043 1038 1033 1028 1023 1018
B.E. (eV)
B.E. (eV)
2 3 2 3 2 3 2 3
Fig. 3. O 1s and Zn 2p XPS spectra of the (a) Au/Al O , (b) Au/Zn0.01Al O , (c) Au/Zn0.02Al O , (d) Au/Zn0.04Al O , and (e) Au/ZnO catalysts.
2 3
Fig. 4. (a) TEM image with insertion of size distribution of Au nanoparticles, and (b) HRTEM image of Au nanoparticles for the Au/Zn0.02Al O catalyst.
CO -TPD
NH -TPD
3
2
e
e
d
c
d
c
b
b
a
a
100
200
300
400
500
600 150
250
350
450
550
650
750
Temperature ( C)
Temperature ( C)
Fig. 5. CO
2
-TPD and NH
3
-TPD profiles of the (a) Al
2
O
3
, (b) Zn0.01Al
2
O
3 2 3 2 3
, (c) Zn0.02Al O , (d) Zn0.04Al O , and (e) ZnO supports.

1
16
S. Wu et al. / Journal of Catalysis 377 (2019) 110–121
activating the OAH bond in various alcohols [57–59]. Moreover, it
was revealed that the acidic-basic sites of the hydroxyapatite sur-
face may play an important role in determining the efficiency and
compatibility of the Au/HAP catalyst for imine synthesis from alco-
hols and amines [39].
2 3
calcination temperature used for the Zn0.02Al O support has an
effect mainly on conversion of benzyl alcohol. Between 300 °C
and 800 °C a volcano-like tendency can be observed and the opti-
mal conversion is obtained over the support calcined at 400 °C.
To rationalise this effect, XRD, N
and XPS are conducted. When calcined at 300 °C the Zn0.02Al
support exhibits AlO(OH) phase (JCPDS #49-0133) due to the dried
at 110 °C) solid (Fig. S2). If calcined at higher than 400 °C only
Al phase is observed. The mean size of Al crystals slightly
2 2 3
physisorption, CO -/NH -TPD,
2 3
O
3.2. Influence of calcination temperature on supports
(
c-
2
O
3
2 3
O
The calcination temperature is usually important and decisive
for the final structural, morphological, acidic-basic, redox and
other properties of oxide materials. As shown in Table 6, the
grows bigger with increasing calcination temperature (Table S4).
A series of typical type-IV isotherms and the mesoporosity are well
Table 6
2 3
Oxidative coupling of benzyl alcohol and aniline over Au nanoparticles loaded on the Zn0.02Al O supports calcined at different temperatures.
Entry
T, Calcination (°C)
Conv. 1 (%)
Select. 3 (%)
Select. 4 (%)
Yield 3 (%)
1
2
3
4
5
300
400
500
650
800
62
73
67
59
52
>99
>99
>99
99
<1
<1
<1
1
61
72
66
58
51
99
1
2 3
Reaction conditions: benzyl alcohol, 0.5 mmol; aniline, 0.75 mmol; catalyst, Au/Zn0.02Al O , 40 mg; toluene, 5 mL; air, 1 atm; temperature, 60 °C; time, 8 h.
100
(
a)
30 (b)
Hollow symbols: with aniline
Solid symbols: without aniline
: Au/Zn0.02Al O
with aniline
80
60
40
20
0
,
2
3
without aniline
20
10
0
,
: Au/Zn0.01Al O3
2
,
: Au/Zn0.04Al O3
2
,
: Au/ZnO
:
Conv. (PhCH OH)
2
,
: Au/Al O3
2
:
Yield (PhC=NPh)
Yield (PhCHO)
:
with aniline
0
2
4
6
8
10
12
0
20
40
60
Reaction time (h)
Reaction time (min)
100 (c)
(d)
100
80
60
40
20
0
Ar
Air
80
60
40
20
0
:
Select. (PhC=NPh)
:
with catal. at 80 C
without catal. at 80 C
with catal. at 60 C
without catal. at 60 C
:
Conv. (PhCH OH)
2
:
:
:
0
0.5
1
1.5
2
0
1
2
3
4
5
6
7
Reaction time (h)
Reaction time (h)
Fig. 6. (a) Time course and (b) kinetic curve for oxidative coupling of benzyl alcohol with aniline and aerobic oxidation of benzyl alcohol. (c) Time course for condensation of
benzaldehyde and aniline in the presence of Ar with or without catalyst. (d) Time course for oxidative coupling of benzyl alcohol with aniline in the presence of Ar or air.
Reaction conditions: benzyl alcohol, 0.5 mmol; aniline, 0.75 mmol; benzaldehyde, 0.5 mmol; Au/Zn0.02Al O , 80 mg; toluene, 5 mL; air, 1 atm; A, 1 atm; temperature, 60 °C.
2 3

S. Wu et al. / Journal of Catalysis 377 (2019) 110–121
117
maintained (Fig. S3), however, the specific surface areas obviously
decrease from 304 m /g to 145 m /g due to the formation of highly
crystalized oxides and the irregular distribution of pore size
2
2
H
O
O
H
Air(O2)
H
H
O2
−
VO
(
Fig. S3, Table S4). Thus the reactive surface sites of the support
would be reduced upon the higher calcination temperature. The
number and strength of the basic sites of Zn0.02Al are reinforced
H O H O
O2−
2
2
n+
M
O2 , O−
−
O2−
2
O
3
O2
−
Mn+
when increasing calcination temperature (300 ? 800 °C), whereas
the number of the acidic sites is found clearly declined even if the
acid strength becomes slightly higher (Table S3, Fig. S4). Besides,
the valence states of Au, Zn, Al and O are independent with the cal-
cination temperature (Fig. S5, Table S5), despite the proportion of
N
Scheme 2. Plausible reaction mechanism for oxidative coupling of benzyl alcohol
and aniline catalysed by Au/Zn Al
X
2 3
O .
reactive surface O
at higher temperatures. Therefore, it can be disclosed that calcina-
tion of the Zn Al supports is a crucial factor that may influence
b
species descends when the support is calcined
X
2
O
3
of catalyst or not (Fig. 6-c). Moreover, >99% yield of benzylide-
neaniline can be achieved after only 1 h at 80 °C. This is in agree-
ment with the previous reports [18,21]. Previously, S. Gao et al.
verified the first order for bimolecular condensation of benzalde-
hyde and aniline in the homogeneous liquid phase in absence of
catalyst [18]. Therefore, as soon as benzaldehyde is produced from
benzyl alcohol, it can be immediately condensed with aniline to
yield imine. Hence, the aerobic oxidation of benzyl alcohol to ben-
zaldehyde can be considered as the rate-determining step for the
direct coupling of benzyl alcohol with aniline. As expected, the ini-
tial conversion rates of benzyl alcohol over a series of Au/Zn Al O
the catalytic performances mainly by modification on crystalline
structure, acidity-basicity and surface oxygen species for the final
catalysts.
3.3. Studies on reaction pathway and mechanism
To deeply understand the reaction pathway and the plausible
reaction mechanism, the time-dependent performance of the
Au/Zn0.02Al catalyst is measured for oxidative coupling of ben-
2 3
O
X
2 3
zyl alcohol with aniline and aerobic oxidation of benzyl alcohol in
the absence of aniline, respectively (Fig. 6-a). It can be clearly seen
that the catalyst is highly selective to imine formation. Only traces
of benzaldehyde (select. <0.5%) as by-product is formed during the
oxidative coupling whole reaction. Conversion of benzyl alcohol
increases linearly during the first 2 h and already approaches
catalysts depending on the Zn/Al molar ratio are in consistent with
their performances for coupling alcohol and aniline (Fig. 6-b,
Table 2). Metallic Au nanoparticles play a predominant role and
allow significantly improving the yield of imine by accelerating
the rate-determining step, because Au-based catalysts have been
highly recognized for their superior behaviours for diverse mild
and efficient selective aerobic oxidation of alcohols. On the other
hand, molecular O2 and the surface Ob oxygen species play an
important role in the rete-determining step. A time-online plot
experiment shows that conversion of benzyl alcohol to benzalde-
hyde ’stops’ under inert gas after 4 h (Fig. 6-d), i.e., conversion of
benzyl alcohol stabilises at 33%, and then adding air into the reac-
tor triggers the oxidation process and leads to an obvious increase
in the conversion. This may confirm the contribution of the surface
O species in the Zn Al O supports, because those reactive oxygen
5
0%. By further extending reaction time the increase of alcohol
conversion becomes relatively gentle, reaching 100% after 8 h.
Compared to aerobic oxidation of benzyl alcohol which needs
1
1 h to achieve a full conversion, the presence of aniline enables
to facilitate conversion of benzyl alcohol by accelerating the initial
rate, as clearly shown in Fig. 6-b and Table 7. At the same time, the
catalytic oxidation of benzyl alcohol to benzaldehyde is an equilib-
rium or reversible reaction that can be promoted by the addition of
aniline probably related with its weak basicity. The observation
here is in agreement with the previous investigation on molar ratio
of aniline to benzyl alcohol (Entry 7–9, Table 2). To be recalled,
both oxidative coupling and aerobic oxidation cannot be driven
b
X
2 3
species could be progressively consumed and then, after full
depletion, the reaction would probably stop. When fresh air is
re-provided, O could be dissociated at the vacancy site to generate
2
by the Au-absent Zn
X
2
Al O
3
oxides. And moreover, it is confirmed
active surface O species.
b
in this work that the condensation of benzaldehyde with aniline
proceeds very fast, a stoichiometric yield (>99%) of benzylideneani-
line can be obtained at 60 °C after 2 h in the presence of Ar
It is widely accepted that the OH group in benzyl alcohol is
prone to adsorb onto the metallic Au nanoparticles to form a metal
alkoxide intermediate [31,35], as shown in Scheme 2. Several com-
putational studies demonstrated that the OH group or the formed
alkoxide species can rapidly diffuse over the Au (1 1 1) plane,
owing to the quite low energy barrier [60]. On the other hand, it
is generally accepted that additional basic additives or basic sites
could promote the activation of the OAH bond in alcohol [58,59].
(Fig. 6-c). To be mentioned, the yield of imine in absence of catalyst
is found nearly the same as the value with catalyst under either O
2
or Ar. To further identify the role of the catalyst in condensation
step, the reaction is performed with a higher temperature at
8
0 °C under the same conditions. Interestingly, almost the same
n+
2À
data during time course are also obtained whatever the presence
Thus, it can be speculated that the basic sites (i.e., M –O pairs
2À
as medium basic sites and O ions as strong basic sites) from
the Zn Al support should be benefit to activating the OAH bond
X
2 3
O
Table 7
to form the alkoxide intermediate. The subsequent formation of
benzaldehyde, as well as the AuAH hydride species, via
b-hydrogen elimination is the rate-controlled step and requires
small Au nanoparticles in metallic state [37,58]. Hence the reactive
Comparison on the initial conversion rates of benzyl alcohol in oxidative coupling
with aniline and in aerobic oxidation over the Au/Zn0.02Al O catalyst.
2 3
Entry
Catalyst
Initial conv. rate
Initial conv. rate
À1 À1
À1 À1
(
mmolÁgcat.Áh ),
(mmolÁgcat.Áh ),
b
surface O species may participate in the reaction by abstracting
with aniline
without aniline
the b-H proton from the alkoxide intermediate, thus forming ben-
1
2
3
4
5
Au/Al
2
O
3
0.25
0.98
1.56
0.53
0.36
0.18
0.83
1.19
0.43
0.29
zaldehyde. In addition, the strong interaction between Au0 and
Au/Zn0.01Al
Au/Zn0.02Al
Au/Zn0.04Al
Au/ZnO
2
2
2
O
O
O
3
3
3
Zn
could further promote the catalytic activity at low temperatures.
Moreover, the acidic sites on the Zn Al support might partici-
pate in the transformation of AuAH hydride which is believed to
be oxidized by O to yield water [39]. The final step is the rapid
X 2 3
Al O may lower the formation energy of oxygen vacancy and
X
2 3
O
Reaction conditions: benzyl alcohol, 0.5 mmol; aniline, 0.75 mmol; Au/Zn0.02Al
0 mg; toluene, 5 mL; air, 1 atm; temperature, 60 °C; time, 1 h.
2 3
O ,
8
2

1
18
S. Wu et al. / Journal of Catalysis 377 (2019) 110–121
and uncatalysed condensation of the as-formed benzaldehyde with
aniline to give benzylideneaniline. In this work, the optimal
nanoparticles (Fig. S6). In addition, the used catalyst shows no
change on the valence state of active components. Moreover, rela-
Au/Zn0.02Al
and proportion of reactive surface O
2
O
3
catalyst shows the highest specific surface area
species, a moderate basicity
tively similar acidic-basic property due to Zn
be still observed after recycling use.
X 2 3
Al O support could
b
and acidity, and also a strong interaction with highly dispersed
metallic Au nanoparticles. These unique features could be respon-
sible for the excellent catalytic performance for direct oxidative
coupling of alcohol and amine.
In the point view of potential industrial application, an ideal
heterogeneous catalyst is expected to work efficiently at mild con-
ditions and ought to be applicable to large-scale production. To our
2 3
satisfaction, the Au/Zn0.02Al O catalyst allows effectively coupling
benzyl alcohol and aniline at as low as only 30 °C. An excellent
yield (98%) to benzylideneaniline can be obtained by simply
extending reaction time to 72 h and involving a higher catalyst
mass (Table 8). To the best of our knowledge, this is the best data
ever reported at a near room temperature for Au-based catalysts
3
.4. Reusability, low-temperature test, scaling-up test and comparison
study
The stability and reusability of a heterogeneous catalyst are
2 3
important for sustainable chemistry. The used Au/Zn0.02Al O
2 3
(Table 1). The developed Au/Zn0.02Al O catalyst is superior to
other supported-Au and Au-Pd catalysts in terms of reaction condi-
tions and catalytic efficiency. Our catalyst can work effectively
without any base additives at a very mild temperature (30 °C) by
catalyst after the post-treatment (see Section 2) is tested for the
recycling reactions under the same conditions. During a consecu-
tive use of five cycles, as shown in Fig. 7, the catalyst still remains
a very stable yield to benzylideneaniline (>99%), behaving as good
as the fresh catalyst. To avoid any possible saturation of catalytic
activity due to the used optimised reaction conditions, conversion
of benzyl alcohol during the kinetic-controlled region, i.e., the ini-
tial stage after 1 h, is examined. The catalyst indeed shows a con-
siderable stability. Moreover, no further conversion of alcohol
can be observed during the leaching test by hot filtration to remove
the solid catalyst (Fig. 7), which suggests the good retention of cat-
alytic active components. ICP-MS analysis shows that the amount
of Au, Al and Zn species present in the filtrate is below the detec-
tion limit (2 ppb), confirming no leaching of Au or support in the
used reaction mixture. This may be attributed to the existence of
À1
using air as oxidant. Moreover, a highest TOF (39.1 h ) is obtained
at 60 °C based on a >99% yield to benzylideneaniline. Finally, this
catalyst shows satisfying stability and reusability. In addition, the
catalytic performance at low temperature is found also one of
2
the best compared with CeO -based catalysts recently reported
in the literature [19,21,23]. Furthermore, the developed catalyst
is tested in the concentrated reactant mixture (Table 8). It enables
to synthesise gram-scale benzylideneaniline (1.08 g), and an excel-
lent yield up to 99% is achieved at 60 °C after 24 h. Those above
performances suggest a sustainable and practical catalytic system
2 3
based on the Au/Zn0.02Al O catalyst for imine synthesis.
0
strong interaction between the active Au species and the robust
3.5. Substrate scope
Zn
X
2 3
Al O support. XRD and TEM measurements on the used cata-
lyst clearly reveal the integrity of the characteristic structure of
Table 9 reports direct oxidative coupling of aniline with various
Zn
X
2
Al O
3
support and the similarity in size distribution of Au
primary alcohols over the Au/Zn0.02Al
2
O
3
catalyst at 60 °C. To be
1
00
5
4
3
2
1
80
60
40
20
0
With catalyst
Leaching test
0
%
20%
40%
60%
80%
100%
0
2
4
6
8
Conv. 1 after 1 hr Yield 3 after 8 hrs
Time (h)
2 3 2 3
Fig. 7. Recycling test of the Au/Zn0.02Al O catalyst within kinetic-controlled region and under optimised conditions, and leaching test of metals in the Au/Zn0.02Al O
catalyst. Reaction conditions: benzyl alcohol, 0.5 mmol; aniline, 0.75 mmol; catalyst, 80 mg; toluene, 5 mL; air, 1 atm; temperature, 60 °C.
Table 8
Low-temperature test and scaling-up test for benzylideneaniline synthesis from oxidative coupling of benzyl alcohol and aniline over the Au/Zn0.02Al
2
O
3
catalyst.
Yield 3 (%)
Entry
1a
Reactants: 1, 2
Catal. mass (g)
T (°C)
Time (h)
0.5 mmol, 0.75 mmol
0.16
0.72
30
60
72
24
98
b
2
6 mmol (0.65 g), 9 mmol (0.84 g)
99 (1.08 g)
Reaction conditions:
a
toluene, 5 mL; air, 1 atm. The by-product is benzaldehyde.
toluene, 10 mL; air, 1 atm.
b

S. Wu et al. / Journal of Catalysis 377 (2019) 110–121
119
Table 9
Oxidative coupling of aniline with different primary alcohols over the Au/Zn0.02Al
2 3
O catalyst.
Reaction conditions: alcohol, 0.5 mmol; aniline, 0.75 mmol; catalyst, 80 mg; toluene, 5 mL; air, 1 atm; temperature, 60 °C. Reaction time, conversion of alcohol and yield of
imine were reported, respectively.
recalled, selectivities to the produced imines in Table 9 are found
higher than 99%. The substituted benzyl alcohols are first tested.
Benzyl alcohol with both electron-donating (MeO, Me and tBu)
and electron-withdrawing (NO and Cl) substituents can be effec-
2
tively converted, and high yield of imines (3a–3f, 93–99%) are
obtained. Still, some differences in the yield and the required reac-
tion time may be related with the effect of property and location of
substituents. The electron-donating groups would reduce the elec-
trophilicity of benzylic carbon by enriching its electron density,
leading consequently to a less reactivity with the nucleophilic
Table 10
2 3
Oxidative coupling of benzyl alcohol with different primary amines over the Au/Zn0.02Al O catalyst.
Reaction conditions: benzyl alcohol, 0.5 mmol; amine, 0.75 mmol; catalyst, 80 mg; toluene, 5 mL; air, 1 atm; temperature, 60 °C. Reaction time, conversion of benzyl alcohol
and yield of imine were reported, respectively.

1
20
S. Wu et al. / Journal of Catalysis 377 (2019) 110–121
amine. The electron-withdrawing groups could facilitate removing
hydrogen from hydroxyl group. The deprotonated alcohol or alkox-
ide would abstract hydride from benzylic carbon via oxidative
dehydrogenation to generate carbonyl compound that reacts
immediately with aniline to form imine. Besides, the effect of loca-
tion (3b and 3c) and steric hindrance (3d) of substituents cannot
be disregarded. Cyclohexylmethanol and pyridin-4-methanol are
examined in this reaction, to respectively afford 95% (3g) and
synthesis from direct oxidative coupling of alcohol and amine.
n+
2À
2À
The basic sites (M –O pairs and O ions) on the Zn
port could be benefit to activating the OAH bond to form the alkox-
ide intermediate. And then the reactive surface O species may
abstract the b-H proton from the alkoxide intermediate to thus
X 2 3
Al O sup-
b
n+
form an aldehyde. Finally the acidic sites (M ions) on the Zn
X 2
Al -
O
3
support may participate in the transformation of AuAH hydride
2
which was believed to be oxidized by O to yield water.
9
3% (3h) yield to the desired imines, respectively. The
aliphatic 1-octanol shows as expected a much lower yield to the
imine 3i (30%), which is well recognised by the difficulty in
activating the non-branched fatty alcohols. Furthermore, several
complex alcohols including 3,4-(methylenedioxy)benzyl alcohol,
Acknowledgements
This work was supported by Natural Science Foundation of
China (21603187, 21763031), National Special Funds of China
(C6176102), Program for Excellent Young Talents of Yunnan
University, Postgraduate Foundation of Yunnan University
(2018185), and Program for Innovative Research Team (in Science
and Technology) in University of Yunnan Province. Authors
sincerely thank Mr. G.-H. Zhu from Advanced Analysis and
Measurement Center (YNU) for ICP-MS analysis.
2
-naphthalenemethanol and 2-thiophenemethanol can be also
activated and converted to the corresponding imines (3j–3l) with
a moderate to low yield.
Table 10 reports direct oxidative coupling of benzyl alcohol
2 3
with various primary amines over the Au/Zn0.02Al O catalyst at
6
0 °C. To be noted, selectivities to the produced imines in Table 10
are found higher than 99%. As is previously disclosed the selective
oxidation of benzyl alcohol is the key step in the tandem coupling
process. Thereby, the nature of amines may probably have an
Appendix A. Supplementary material
2 3
impact on the conversion and product yield. The Au/Zn0.02Al O
catalyst shows a high imine yield of 3m to 3q (94–99%) from the
substituted anilines. As expected, electron-withdrawing group
Supplementary data to this article can be found online at
https://doi.org/10.1016/j.jcat.2019.07.027.
(
Cl) and electron-donating groups (MeO and Me) could enhance
References
the nucleophilicity of aniline and thus promote the nucleophilic
attack, despite some minor variations could be observed due to
the property and the location of substituents. Compared with ani-
line, coupling benzyl alcohol with benzyl amine also yields a high
imine 3r (96%), which may be due to the high basicity of benzyl
amine and/or the possibility of its self-coupling reaction to form
imine. The linear aliphatic 1-octanamine would afford a high
desired imine production (3s, 92%). Thereby, it is evident again of
that, when comparing to the yield of 3i, the conversion of an alco-
hol to an aldehyde is the rate- and yield-determining step in the
cross-coupling of alcohols with amines to imines. t-Butylamine
and cyclohexylamine could also react with benzyl alcohol to yield
the desired imines (3t, 3u) in a moderate to low yield. In addition,
pyridin-4-amine offers a low imine yield to 3v (28%), which is
probably influenced by the basicity of pyridine group. In summary,
[1] L. He, X.-B. Lou, J. Ni, Y.-M. Liu, Y. Cao, H.-Y. He, K.-N. Fan, Chem. Eur. J. 16
(
2010) 13965–13969.
2] S. Kobayashi, Y. Mori, J.S. Fossey, M.M. Salter, Chem. Rev. 111 (2011) 2626–
704.
[3] L. He, Y. Qian, R.-S. Ding, Y.-M. Liu, H.-Y. He, K.-N. Fan, Y. Cao, ChemSusChem 5
2012) 621–624.
[
2
(
[
4] Q. Wang, Y. Deng, F. Shi, Catal. Sci. Technol. 4 (2014) 1710–1715.
[
5] L. Zhang, Y. Zhang, Y. Deng, F. Shi, Catal. Sci. Technol. 5 (2015) 3226–3234.
[6] D. Chandra Mohan, A. Sadhukha, G. Maayan, J. Catal. 355 (2017) 139–144.
[
[
[
7] J. Jin, C. Yang, B. Zhang, K. Deng, J. Catal. 361 (2018) 33–39.
8] D. Zhang, X. Han, T. Dong, X. Guo, C. Song, Z. Zhao, J. Catal. 366 (2018) 237–244.
9] K. Yamaguchi, N. Mizuno, Angew. Chem. Int. Ed. 42 (2003) 1480–1483.
[10] H. Yuan, W.-J. Yoo, H. Miyamura, S. Kobayashi, J. Am. Chem. Soc. 134 (2012)
3970–13973.
1
[
[
11] T. Sonobe, K. Oisaki, M. Kanai, Chem. Sci. 3 (2012) 3249–3255.
12] S. Furukawa, Y. Ohno, T. Shishido, K. Teramura, T. Tanaka, ACS Catal. 1 (2011)
1150–1153.
[13] M. Largeron, M.-B. Fleury, Angew. Chem. Int. Ed. 51 (2012) 5409–5412.
[
[
[
14] L. Al-Hmoud, C.W. Jones, J. Catal. 301 (2013) 116–124.
15] Y. Xiang, Q. Meng, X. Li, J. Wang, Chem. Commun. 46 (2010) 5918–5920.
16] J. Huang, L. Yu, L. He, Y.-M. Liu, Y. Cao, K.-N. Fan, Green Chem. 13 (2011) 2672–
2677.
17] X. Cui, F. Shi, Y. Deng, Chem. Commun. 48 (2012) 7586–7588.
18] B. Chen, J. Li, W. Dai, L. Wang, S. Gao, Green Chem. 16 (2014) 3328–3334.
19] M. Tamura, K. Tomishige, Angew. Chem. Int. Ed. 54 (2015) 864–867.
2 3
the Au/Zn0.02Al O catalyst shows a wide scope in activating vari-
ous primary alcohols and amines for imine synthesis via direct
oxidative coupling. The electron effect of substituents, the nucle-
ophilicity and/or electrophilicity of substrates, and even the
acidic-basic properties of substrates and catalysts, are worth
studying comprehensively in a future work.
[
[
[
[20] X. Huang, L. Liu, H. Gao, W. Dong, M. Yang, G. Wang, Green Chem. 19 (2017)
69–777.
21] H. Zhang, C. Wu, W. Wang, J. Bu, F. Zhou, B. Zhang, Q. Zhang, Appl. Catal. B 227
2018) 209–217.
7
[
(
4
. Conclusions
[22] J. Bu, W. Wang, H. Zhang, C. Wu, B. Zhang, Y. Cao, F. Zhou, Q. Zhang, X. Zhang,
Mol. Catal. 459 (2018) 46–54.
[
[
23] J. Qin, Y. Long, W. Wu, W. Zhang, Z. Gao, J. Ma, J. Catal. 371 (2019) 161–174.
24] M.S. Kwon, S. Kim, S. Park, W. Bosco, R.K. Chidrala, J. Park, J. Org. Chem. 74
(2009) 2877–2879.
In summary, we developed a novel, efficient and sustainable
Au-based catalytic system for imine synthesis from the direct
oxidative coupling of alcohol and amine under air atmosphere.
[
[
25] M. Zhang, S. Wu, L. Bian, Q. Cao, W. Fang, Catal. Sci. Technol. 9 (2019) 286–301.
26] W. He, L. Wang, C. Sun, K. Wu, S. He, J. Chen, P. Wu, Z. Yu Chem, Eur. J. 17
Au nanoparticles loaded on Zn-doped
2 3
c-Al O was reported for
(
2011) 13308–13317.
the first time as active, selective and stable heterogeneous catalyst
for a wide scope of imine productions from the primary alcohols
and amines at very mild conditions. It was demonstrated that
the uniform and small (2.9 nm) Au nanoparticles in metallic state
were active sites and responsible for activating and converting an
alcohol to an aldehyde, i.e., the rate-determining step. The subse-
quent condensation of aldehyde with amine was found fast and
uncatalysed. The variation of Zn2 doping and calcination temper-
ature would modify the textural property, acidity-basicity and sur-
face oxygen species for support materials, which were decisive
[27] J.W. Kim, J. He, K. Yamaguchi, N. Mizuno, Chem. Lett. 38 (2009) 920–921.
[28] Y. Zhang, F. Lu, H.-Y. Zhang, J. Zhao, Catal. Lett. 147 (2017) 20–28.
[29] A.S.K. Hashmi, G.J. Hutchings, Angew. Chem. Int. Ed. 45 (2006) 7896–7936.
[30] G.J. Hutchings, M. Brust, H. Schmidbaur, Chem. Soc. Rev. 37 (2008) 1759–1765.
[31] C. Della Pina, E. Falletta, L. Prati, M. Rossi, Chem. Soc. Rev. 37 (2008) 2077–
2095.
[
32] A. Lackmann, C. Mahr, M. Schowalter, L. Fitzek, J. Weissmüller, A. Rosenauer, A.
Wittstock, J. Catal. 353 (2017) 99–106.
[
33] W. Hu, D. Li, Y. Yang, T. Li, H. Chen, P. Liu, J. Catal. 357 (2018) 108–117.
+
[34] A. Jouve, G. Nagy, F. Somodi, C. Tiozzo, A. Villa, A. Balerna, A. Beck, C.
Evangelisti, L. Prati, J. Catal. 368 (2018) 324–335.
[
[
35] A. Corma, H. Garcia, Chem. Soc. Rev. 37 (2008) 2096–2126.
36] W. Fang, Q. Zhang, J. Chen, W. Deng, Y. Wang, Chem. Commun. 46 (2010)
1547–1549.
X 2 3
for the excellent performance of Au/Zn Al O catalyst for imine

S. Wu et al. / Journal of Catalysis 377 (2019) 110–121
121
[
37] J. Chen, W. Fang, Q. Zhang, W. Deng, Y. Wang, Chem. Asian. J. 9 (2014) 2187–
196.
[49] Y. Geng, L. Guo, S.-S. Xu, Q.-Q. Sun, S.-J. Ding, H.-L. Lu, D.W. Zhang, J. Phys.
Chem. C 115 (2011) 12317–12321.
2
[
[
38] T. Gao, T. Gao, W. Fang, Q. Cao, Mol. Catal. 439 (2017) 171–179.
39] H. Sun, F.-Z. Su, J. Ni, Y. Cao, H.-Y. He, K.-N. Fan, Angew. Chem. Int. Ed. 48
[50] W. Fang, C. Pirez, S. Paul, M. Jiménez-Ruiz, H. Jobic, F. Dumeignil, L. Jalowiecki-
Duhamel, Int. J. Hydrogen Energy 41 (2016) 15443–15452.
[51] Y. Dou, M. Zhang, S. Zhou, C. Oldani, W. Fang, Q. Cao, Eur. J. Inorg. Chem. 2018
(2018) 3706–3716.
[52] R. Al-Gaashani, S. Radiman, A.R. Daud, N. Tabet, Y. Al-Douri, Ceram. Int. 39
(2013) 2283–2292.
[53] K. Yan, Y. Liu, Y. Lu, J. Chai, L. Sun, Catal. Sci. Technol. 7 (2017) 1622–1645.
[54] J. Ma, Z. Liu, J. Song, L. Zhong, D. Xiao, H. Xi, X. Li, R. Sun, X. Peng, Green Chem.
20 (2018) 5188–5195.
[55] Q. Liu, L. Wang, C. Wang, W. Qu, Z. Tian, H. Ma, D. Wang, B. Wang, Z. Xu, Appl.
Catal. B 136–137 (2013) 210–217.
(
2009) 4390–4393.
40] W. Cui, Q. Xiao, S. Sarina, W. Ao, M. Xie, H. Zhu, Z. Bao, Catal. Today 235 (2014)
52–159.
41] S. Kegn
2 (2010) 1437–1441.
42] W. Cui, H. Zhu, M. Jia, W. Ao, Y. Zhang, B. Zhaorigetu, React. Kinet. Mech. Catal.
09 (2013) 551–562.
[
[
[
[
1
æs, J. Mielby, U.V. Mentzel, C.H. Christensen, A. Riisager, Green Chem.
1
1
43] L. Zhang, W. Wang, A. Wang, Y. Cui, X. Yang, Y. Huang, X. Liu, W. Liu, J.-Y. Son,
H. Oji, T. Zhang, Green Chem. 15 (2013) 2680–2684.
[
[
[
[
44] J.-F. Soulé, H. Miyamura, S. Kobayashi, Chem. Commun. 49 (2013) 355–357.
45] S. Furukawa, R. Suzuki, T. Komatsu, ACS Catal. 6 (2016) 5946–5953.
46] T. Gao, J. Chen, W. Fang, Q. Cao, W. Su, F. Dumeignil, J. Catal. 368 (2018) 53–68.
47] T. Gao, Y. Yin, G. Zhu, Q. Cao, W. Fang, Catal. Today (2019), https://doi.org/
[56] M. Turco, G. Bagnasco, C. Cammarano, P. Senese, U. Costantino, M. Sisani, Appl.
Catal. B 77 (2007) 46–57.
[57] T. Gao, Y. Yin, W. Fang, Q. Cao, Mol. Catal. 450 (2018) 55–64.
[58] W. Fang, J. Chen, Q. Zhang, W. Deng, Y. Wang, Chem. Eur. J. 17 (2011) 1247–
1256.
1
0.1016/j.cattod.2019.03.065.
[
48] E.M. Seftel, E. Popovici, M. Mertens, K.D. Witte, G.V. Tendeloo, P. Cool, E.F.
Vansant, Microporous Mesoporous Mater. 113 (2008) 296–304.
[59] J. Chen, Q. Zhang, W. Fang, Y. Wang, Chin. J. Catal. 31 (2010) 1061–1070.
[60] J. Gong, C.B. Mullins, Acc. Chem. Res. 42 (2009) 1063–1073.