NiO-doped Au/Ti-powder: A catalyst with dramatic improvement in activity for gas-phase oxidation of alcohols

Article abstract of DOI:10.1016/j.apcata.2013.07.005

An active and stable NiO@Au/Ti-powder catalyst, which can be obtained by doping Au/Ti-powder (Au: 35-40 nm) with NiO, has been developed for the gas-phase oxidation of alcohols. Gold particles are found to be partially covered with tiny NiO segments to form specific NiO@Au ensembles thereby leading to a dramatic conversion improvement from only ～5% (without NiO doping) to ～94% for the benzyl alcohol oxidation at 280 C. Additionally, the selective oxidation of cyclopropyl carbinol can proceed over this catalyst at 280 C with a selectivity of 94% and a conversion of 72-80% throughout the entire 300 h test. The hybrid active-sites, Ni₂O₃-Au⁺, on the NiO@Au ensemble are identified using X-ray photoelectron spectroscopy (XPS) and X-ray absorption spectroscopy (XAS).

Full text of DOI:10.1016/j.apcata.2013.07.005

Accepted Manuscript
Title: NiO-doped Au/Ti-powder: A Catalyst with Dramatic
Improvement in Activity for Gas-phase Oxidation of Alcohols
Author: Guofeng Zhao Huanyun Hu Zheng Jiang Shuo Zhang
Yong Lu
PII:
S0926-860X(13)00403-1
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.apcata.2013.07.005
APCATA 14321
To appear in:
Applied Catalysis A: General
Received date:
Revised date:
Accepted date:
13-3-2013
29-6-2013
3-7-2013
Please cite this article as: G. Zhao, H. Hu, Z. Jiang, S. Zhang, Y. Lu,
NiO-doped Au/Ti-powder: A Catalyst with Dramatic Improvement in Activity
for Gas-phase Oxidation of Alcohols, Applied Catalysis A, General (2013),
http://dx.doi.org/10.1016/j.apcata.2013.07.005
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

NiO-doped Au/Ti-powder: A Catalyst with Dramatic Improvement in Activity
for Gas-phase Oxidation of Alcohols
Guofeng Zhaoa, Huanyun Hua, Zheng Jiangb, Shuo Zhangb and Yong Lu*a
a Shanghai Key Laboratory of Green Chemistry and Chemical Processes, Department of Chemistry,
East China Normal University, Shanghai 200062, China.
b Shanghai Synchrotron Radiation Facility, Shanghai Institute of Applied Physics, Chinese
Academy of Sciences, Shanghai 201204, China.
* Corresponding author: Yong Lu,
Shanghai Key Laboratory of Green Chemistry and Chemical Processes, Department of Chemistry,
East China Normal University, Shanghai 200062, China.
Tel.: +86-21-62233424; Fax: +86-21-62233424;
Email address: ylu@chem.ecnu.edu.cn (Y. Lu).
1
Page 1 of 34

Abstract
An active and stable NiO@Au/Ti-powder catalyst, which can be obtained by doping
Au/Ti-powder (Au: 35-40 nm) with NiO, has been developed for the gas-phase oxidation of
alcohols. Gold particles are found to be partially covered with tiny NiO segments to form specific
NiO@Au ensembles thereby leading to a dramatic conversion improvement from only ~5%
(without NiO doping) to ~94% for the benzyl alcohol oxidation at 280 °C. Additionally, the
selective oxidation of cyclopropyl carbinol can proceed over this catalyst at 280 °C with a
selectivity of 94% and a conversion of 72-80% throughout the entire 300 h test. The hybrid
active-sites, Ni2O3-Au+, on the NiO@Au ensemble are identified using X-ray photoelectron
spectroscopy (XPS) and X-ray absorption spectroscopy (XAS).
Keywords: gold catalyst, NiO, interface, active site, alcohol oxidation, aldehyde
2
Page 2 of 34

1. Introduction
Selective oxidations of alcohols to carbonyl compounds are some of the most important
processes in organic chemistry, both in laboratory and in industrial manufacturing [1-9]. However,
the use of various toxic and expensive stoichiometric inorganic oxidants (notably Cr reagents) and
volatile organic solvents in the existing oxidation processes causes serious problems in product
separation and environmental issues, obviously conflicting the atom-economical concept of green
chemistry [1-9]. Therefore, there should be a paradigm shift away from traditional approaches,
aimed at using atmospheric air and recyclable heterogeneous catalysts under solvent-free conditions
[1-9].
Supported gold catalysts have been extensively studied for the aerobic oxidation of alcohols
[1-9]. Generally, their activity is related to the size of gold nanoparticles (Au NPs) [2,4,10-13], the
chemical state of the gold [14-16], the support properties [17-21], and the gold-support interactions
[22]. The small Au NPs (<5 nm) has been widely accepted to be essential for good catalytic activity
[12], and the other factors are all discussed based on the use of the small Au NPs. Recently,
however, some reports disclosed the occurrence of the opposite behavior, which warrants the
reconsideration of this fundamental belief. For example, catalysts comprising a transition metal
oxide deposited on a single crystal of gold, such as CeO2/Au(111) [23,24] and TiO2/Au(111) [25],
are not even nano-scale but still show higher catalytic activity than Au NPs supported on oxides.
Therefore, particle size is not the only parameter that strongly affects the activity of these catalysts.
There should be another model to explain the excellent activity displayed by the large gold particles
in some reactions.
More recently, NiO@Au ensembles (i.e., partial coverage of 20-30 nm Au particles with NiO
3
Page 3 of 34

segments) have been reported by our group, which are highly active and selective for the gas-phase
oxidation of alcohols [26-28]. Guided by our previous results, we attempted to place NiO@Au
ensembles on a suitable support for obtaining a practical and versatile catalyst with high
activity/selectivity, good stability and high heat conductivity. Because gas-phase alcohol oxidation
reactions are highly exothermic, it is desirable to endow the catalyst with high heat conductivity for
rapidly dissipating reaction heat from the catalytic reactor bed. The oxide-supported catalysts
demonstrated good low-temperature activity, but their poor heat conductivity causes hotspots to
form in the reaction bed [3]. Therefore, metal support is a reasonable alternative. Compared to the
metallic Ni [26-28] and Cu [29,30] microfibrous supports, Ti powder was a good support candidate
because of its low price, low coking activity, high oxidation-/acid-corrosion resistance, inert
behavior toward the alcohol oxidation, and relatively high heat conductivity. In order to further
elucidate the activity origin of the NiO@Au/Ti-powder catalyst, X-ray photoelectron spectroscopy
(XPS) and X-ray absorption spectroscopy (XAS) were employed to reveal the active-sites on the
NiO@Au ensembles.
2. Experimental
2.1 Catalyst preparation
The Au/Ti-powder catalyst (5 wt% Au, denoted as Au-5/Ti) was prepared by impregnating
Ti-powder (200-300 mesh, Alfa Aesar) with HAuCl4 solution, followed by calcination at 300 °C in
air for 4 h. Au-5/Ti catalyst was then impregnated with aqueous NiCl2 solution (to 1 wt% NiO) and
calcined at 400 °C in air for 4 h, which transformed the NiCl2 into NiO. By doing so, a NiO-doped
Au-5/Ti catalyst was obtained and denoted as NiO-1@Au-5/Ti-Cl. NiO-1@Au-5/Ti-N and
NiO-1@Au-5/Ti-C catalysts were also prepared by using this method except that Ni(NO3)2 and
4
Page 4 of 34

Ni(Ac)2 were used as precursors, respectively. Catalysts with different Au and NiO loadings were
prepared by following the same procedure and simply tuning the amount of HAuCl4 and nickel salts
in the corresponding solutions.
Additionally, the NiO-1/Ti-Cl catalyst was prepared by impregnating Ti-powder with a NiCl2
solution to NiO loading of 1 wt%, followed by calcination at 400 ºC in air for 4 h. The
Au-5@NiO-1/Ti-Cl catalyst was prepared by impregnating the NiO-1/Ti-Cl catalyst with a solution
of HAuCl4 to Au loading of 5 wt%, followed by calcination at 300 °C in air for 4 h.
2.2 Catalyst characterization
The catalysts were characterized with X-ray diffraction (XRD, Rigaku Uitima IV diffractometer
(Cu K)), scanning electron microscopy (SEM, Hitachi S-4800) equipped with an energy
dispersive X-ray analysis (EDX) unit (Oxford, UK), and transmission electron microscopy (TEM,
JEOL-JEM-2010 instrument at 200 kV). XPS were recorded on a VG EscaLab 220i spectrometer,
which used a standard Al K X-ray source (300 W) and an analyzer pass energy of 20 eV. All
binding energies are referenced to the C1s line at 284.9 eV. XAS data of the Au LIII-edge (11,919
eV) and Ni K-edge (8350 eV) were collected in transmission mode on the BL14W1 beam line of
the Shanghai Synchrotron Radiation Facility (SSRF). The typical electron beam energy was 3.5
GeV, and the current was 300 mA. A cryogenically cooled double-crystal Si (111) monochromator
was used to minimize the harmonics. The extended X-ray absorption fine structure (EXAFS)
spectra were analyzed with the ATHENA software. The loading of gold for all of the samples was
determined by inductively coupled plasma atomic emission spectrometry (ICP-AES) on a Thermo
Scientific iCAP 6300 ICP spectrometer.
5
Page 5 of 34

2.3 Reactivity tests
The gas-phase oxidation of alcohols was performed on a fixed-bed quartz tube reactor (700 mm
length by 16 mm inner diameter) under atmospheric pressure as described previously [31,32].
Catalyst used in each test was 0.3 g (200-300 mesh). Alcohols were fed continuously via a
high-performance liquid pump in parallel with O2 (oxidant) and N2 (diluted gas of 100 ml/min)
feeding, which used calibrated mass flow controllers, into the reactor under steady-state temperature
conditions. The weight hourly space velocity (WHSV) was calculated by dividing the mass flow
rate of the alcohol feedstock by the catalyst mass. The effluent was cooled using an ice-salt bath
(-15 ºC) to condense the vapors for analysis using a Shimadzu-2014 gas chromatography-flame
ionization detector (GC-FID) with a 60 m HP-5ms capillary column. The gas-phase products, such
as H2, COx, and C1-C3 hydrocarbons, were analyzed using an HP-5890 GC with a thermal
conductivity detector (TCD) and a 30 m AT-plot 300 capillary column. The reproducibility of the
conversion and selectivity of the product was high and the error range was ± 0.2%. A carbon mass
balance close to 100% was achieved.
3 Results and discussion
3.1 Highly active NiO@Au ensembles
HAuCl4 and NiCl2 were chosen as the preferred precursors in the fabrication of the NiO@Au
ensembles because of the previously reported galvanic exchange reaction that yielded metallic gold
particles and NiCl2 from HAuCl4 and Ni fiber [26,27]. First, we prepared an Au/Ti-powder catalyst
(5 wt% Au; denoted as Au-5/Ti) by impregnation of Ti-powder with a HAuCl4 solution and
subsequent calcination at 300 °C in air for 4 h. Specific surface area of the Au-5/Ti catalyst is only
0.2 m2/g, close to that (0.1 m2/g) for the Ti-powder support itself (Table S1). SEM (Fig. 1A), XRD
6
Page 6 of 34

(Fig. 1B) and TEM (Fig. 1C) measurement results clearly showed that the HAuCl4 decomposed to
form metallic gold particles on Ti-powder surface. As shown in Table 1 and Fig. S1, gold particles
of the Au-5/Ti catalyst, with size around 35-40 nm, are much larger than the small Au particles (<5
nm) desired for achieving good catalytic activity [12]. When subjected to oxidation of benzyl
alcohol, not surprisingly, the Au-5/Ti catalyst delivered very low benzyl alcohol of <6%,
comparable to the Ti-powder support itself (entries 1,2 in Table 1). Subsequently, the Au-5/Ti
catalyst was impregnated with NiCl2 solution to a NiO loading of 1 wt%, followed by calcination at
400 ºC. The XRD patterns demonstrated that NiCl2 was transformed into NiO, which co-existed
with Au particles to produce a NiO-doped Au/Ti catalyst (denoted as NiO-1@Au-5/Ti-Cl; Fig. 2,
Scheme S1). The EDX results indicated that the Ni was on the surface of the Au particles (Fig. S2).
Most interestingly, NiO-doping resulted in a dramatic improvement in activity for the gas-phase
oxidation of benzyl alcohol. Benzyl alcohol conversion of only 5.2% was observed over the Au-5/Ti
(entry 2 in Table 1), but a high conversion of 93.7% was obtained over the NiO-1@Au-5/Ti-Cl
(entry 3 in Table 1). Note that the specific surface area of the catalysts almost remained unchanged
(i.e., 0.2 m2/g; Table S1) with or without NiO-doping. Thus, the dramatic conversion promotion
assignment to the surface area change can be reasonably ruled out. For reference, NiO-1/Ti-Cl was
prepared and, when subjected to the benzyl alcohol oxidation, yielded a conversion of only 2.3%
(entry 4 in Table 1). Clearly, the singular use of Au or NiO particles was ineffective for the
gas-phase oxidation of alcohols.
Fig. 1D and Fig. S3 show the TEM images of the NiO-1@Au-5/Ti-Cl catalyst. Clearly, large Au
particles (~40 nm) were partially covered with small NiO segments to form NiO@Au ensembles.
Additionally, when we placed 5 wt% Au onto the NiO-1/Ti-Cl catalyst, the resultant
7
Page 7 of 34

Au-5@NiO-1/Ti-Cl catalyst did not show dramatic activity improvement yet (entry 5 in Table 1).
This catalyst delivered a benzyl alcohol conversion of only 21.1%, much lower than that (93.7%)
for the NiO-1@Au-5/Ti-Cl. Note that the Au-5@NiO-1/Ti-Cl catalyst was obtained by placing NiO
(1wt%) firstly onto the Ti-powder support to form NiO-1/Ti-Cl and subsequently depositing gold
(5wt%) on the NiO-1/Ti-Cl. By using this method, it is difficult to form the NiO@Au ensembles
when compared to the method for the NiO-1@Au-5/Ti-Cl catalyst preparation. The above results
indicated that the formation of the NiO@Au ensembles is essential for the high activity of the
Au-5@NiO-1/Ti-Cl catalyst. This observation is in agreement with our previous report for the
Au/Ni-fiber catalyst system [26,27].
3.2 Influences of nickel precursors, loadings of gold and nickel
However, we wonder whether it is only method for forming the NiO@Au ensembles use NiCl2
as the precursor. To address this question, two additional catalysts were prepared by impregnating
Au-5/Ti catalysts with Ni(Ac)2 and Ni(NO3)2 solution (to 1 wt% NiO) and subsequently calcining at
400 °C in air for 4 h. These two new catalysts, denoted as NiO-1@Au-5/Ti-C and
NiO-1@Au-5/Ti-N, were examined in the gas-phase oxidation of benzyl alcohol. As expected, both
catalysts showed activity comparable to that of the NiO-1@Au-5/Ti-Cl catalyst, yielding benzyl
alcohol conversions of 92.1% and 94.6% (entries 6, 7 in Table 1). These results indicated that upon
fabricating the special NiO@Au ensembles, the corresponding catalysts always show excellent
activity in the reaction, regardless of the identity of the nickel precursors (Fig. S4). It should be
noticeable that our catalysts were still active at lower reaction temperatures. Over the
NiO-1@Au-5/Ti-N catalyst, for instance, the benzyl alcohol conversion declined from 94.6% to
86.0% as the temperature was decreased from 280 to 230 °C (entries 7-10 in Table 1). An
8
Page 8 of 34

acceptable conversion of 63.5% could still be obtained when the temperature was decreased to 220
°C (entries 11 in Table 1).
In addition, the Au and NiO loadings of the NiO@Au/Ti-Cl catalysts could be tuned within a
large range without risking the degradation of their catalytic performance (Fig. 3). The influence of
NiO loading on the catalyst performance was investigated from 0.3 to 7 wt%, with the Au loading
maintained at 5 wt% (Fig. 3A). Benzyl alcohol conversion was increased from 54.2 to 93.7% along
with the increase of NiO loading from 0.3% to 1 wt%, and then remained almost unchanged until to
4 wt%. Further increasing NiO loading to 5 wt% and then up to 7 wt% led to a decrease in the
benzyl alcohol conversion to 79.1% and then 77.1%. By setting the Au/NiO weight ratio of 5, the
effect of gold loading was investigated from 1 to 5 wt%, with the results as shown in Fig. 3B.
Clearly, the catalytic activity almost showed no alteration until the gold loading was reduced to 3
wt% (corresponding NiO loading: 0.6 wt%). Note that the reduction of the Au loading is
particularly desirable for practical applications.
3.3 Selectivity oxidation reactivity for other alcohols
The NiO-1@Au-5/Ti-Cl catalyst was also tested for the oxidation of a range of straight-chain
(1-octanol, 2-octanol), benzylic (1-phenylethanol, 2-phenylethanol), cyclic (cyclohexanol) and
polynary (1,2-propanediol) alcohols, with the results as summarized in Table 2 and Table S2.
Conjugated aromatic alcohols could be oxidized to the corresponding conjugated aldehydes or
ketones faster than the aliphatic alcohols. At equivalent selectivity, for example, a high conversion
of 95.3% could be obtained for the 1-phenylethanol oxidation at 300 °C but it was only 51.2% for
the 2-phenylethanol oxidation even at 340 °C. Similarly, the straight-chain octanol was less reactive
than the 1-phenylethanol. Interestingly, the primary linear aliphatic octanol was more reactive than
9
Page 9 of 34

the secondary counterparts: 51.9% conversion for 1-octanol at 300 °C vs. 23.4% conversion for
2-octanol at 360 °C. Cyclohexanol could be selectively oxidized to cyclohexanone (a key raw
material in the synthesis of many useful chemicals, such as caprolactam fornylon 6 and adipic acid
for nylon 66) at a conversion of 67.8% with a selectivity of 97.4% at 340 °C. For the 1,2-propylene
glycol oxidation, it is in general favorable to form hydroxylketone rather than methyl glyoxal.
Notably, our NiO-1@Au-5/Ti-Cl catalyst delivered a good methyl glyoxal selectivity of 71.0% and
a high conversion of 91.6%.
3.4 Reaction stability tests
Stability is a very important practical consideration for a heterogeneous catalyst. For the benzyl
alcohol oxidation at 280 °C, the NiO-1@Au-5/Ti-Cl catalyst delivered a single-run lifetime as long
as 130 h with a very steady catalytic reactivity (conversion: 88-94%, selectivity: 99%; Fig. 4A),
which showed much better activity and selectivity over time than the other reported catalysts [1,3].
This single run lifetime over the NiO-1@Au-5/Ti-Cl catalyst is comparable to the 140 h lifetime
demonstrated by our previously reported Au-4/Ni-fiber-300 catalyst [26,27]. In this 130-h single run,
the equilibrium inversion rate of benzyl alcohol was 18.4 g per hour per gram of
NiO-1@Au-5/Ti-Cl and a benzaldehyde yield of 2.3 kg per gram NiO-1@Au-5/Ti-Cl (or 46 kg per
gram of Au) could be achieved. Additionally, the liquid effluent could be automatically separated
into a water phase and a clear pale yellow organic phase, which contained predominantly
benzaldehyde (CO, CO2, benzene, toluene and benzoic acid were formed in trace amounts).
As previously noted, the lifetime of Au/SiO2 catalyst is only 70 h using a lower WHSV of 10 h-1
at 315 °C, with the selectivity of 90-100% and the conversion in the wider range of 50-75% [1,3].
The decay of the catalytic behavior was ascribed to the formation of a pitch-dark deposition caused
10
Page 10 of 34

by the weak heat transfer ability. Whereas the K-Cu-TiO2 shows a very high conversion of 99% at
210 °C (bed temperature: 227 °C) using a very low WHSV of 0.6 h-1, the reported lifetime of it is
only 50 h and its throughput is extremely low [33].
We then conducted a longer-term test using cyclopropyl carbinol over the NiO-1@Au-5/Ti-Cl
catalyst. At 280 °C, the conversion of cyclopropyl carbinol was ~80% and remained unchanged
within the first 230 h run, using a WHSV of 20 h-1 and an O2/hydroxyl molar ratio of 0.6. After that,
the conversion was decreased slowly to ~72% within another 70 h time-on-stream. Note that the
selectivity to cyclopropyl formaldehyde was always ~94% throughout the entire 300-h test (Fig.
4B). In this 300-h single run, the average conversion of cyclopropyl carbinol was approximately
77%. Cyclopropyl formaldehyde could be obtained at a rate of 15.4 g per hour over per gram
NiO-1@Au-5/Ti-Cl, and a yield of 4.6 kg per gram NiO-1@Au-5/Ti-Cl (or 92 kg per gram Au)
could be achievable. After 300-h oxidation of cyclopropyl carbinol, the size of Au particles and the
NiO@Au ensemble structure were retained well (Fig. S5), which indicated that the catalyst is stable
and robust.
3.5 Identification of active sites
3.5.1 XPS studies
As is well known, a reaction takes place on the active sites at the catalyst surface. Therefore,
revealing the active sites is particularly desirable for elucidation of the activity improvement
associated with the formation of NiO@Au ensembles. To accomplish this goal, surface-sensitive
XPS technique was used to track the evolution behavior of the surface chemical composition for the
NiO-1@Au-5/Ti-Cl catalysts after undergoing a specifically designed three-stage reaction
procedure in the tubular reactor at 280 °C: the reaction was carried out with alcohol/O2 mixture
11
Page 11 of 34

feedstock, subsequently with sole alcohol by cutting-off O2 and then re-feeding O. Table 3 and Fig.
5 summarized the XPS results for the NiO-1@Au-5/Ti-Cl samples collected at the each stage of the
above reaction treatment procedure. On the sample a (Table 3, undergoing reaction with alcohol/O2
mixture feedstock for 0.5 h), apart from Au0 (Au4f (7/2): 84.0 eV) and NiO (Ni2p: 854.4 eV; O1s:
529.7 eV), not only were a large amount of Ni2O3 specimens (Ni2p: 856.2 eV; O1s: 531.8 eV)
formed but a considerable number of Au+ specimens were also clearly detected with the Au4f (7/2)
peak at 84.8 eV (Fig. 5-a). On such surface, the Au+ surface content to total surface gold atoms was
37.0% and surface Ni2O3 content to total surface nickel oxides was 77.0% with the surface Au+/Ni3+
ratio of 1/14 (Table 3, Table S3).
Interestingly, when the sample a reacted with benzyl alcohol alone at 280 ºC for 15 min, which
was achieved by cutting off O2 from the stream (Fig. 5-b, named sample b), the Au+ surface content
was decreased significantly from 37.0% to less than 4% while the surface Ni2O3 content was
decreased from 77.0% to 60.0% by correspondingly transferring to NiO (surface NiO content was
increased from 23.0% to 40.0%; Table 3). As a result, the benzyl alcohol conversion showed a sharp
reduction from 94% to 2%. After the sample b reacted with mixture of benzyl alcohol and O2 by
retrieving the O2 feeding at 280 °C for 10 min (i.e., sample c), the surface Au+ and Ni2O3 contents
were recovered, nearly reaching the level of the sample a, and this was accompanied by a
significant increase in benzyl alcohol conversion to 88% (Fig. 5-c, Table 3). Unlike the
high-low-high evolution behavior of the surface Au+ and Ni2O3 content, the surface TiO2 content
remained at 36.0% in the whole process of with-without-with O2 reaction procedure (Fig. S6, Table
S3).
12
Page 12 of 34

3.5.2 XAS studies
As a sensitive technology to probe the local geometry and identify the oxidation states of
metallic particles [34-36], the X-ray absorption near-edge structure (XANES) was also employed to
study the three catalyst samples (same as in above XPS studies) at the Au L3 and Ni K edge, with
the results as shown in Fig. 6 and Fig. 7. The characteristic high-low-high evolution behavior for
the Au+ and Ni2O3 specimens was observed again along with the respective with-without-with O2
presence in the reaction feed gas, which is in good agreement with the XPS results. The sample a
showed a higher intensity of whiteline than that of the Au foil (Fig. 6A-a,b) and a similar XANES
with the reported mixture of Au+ complexes and Au0 particles [36]. This observation definitely
indicated that Au+ formed in the NiO@Au ensembles. Interestingly, as the sample a underwent the
reaction for 15 min with only benzyl alcohol in the absence of O2, the whiteline intensity exhibited
a slight decrease and the resulting XANES was similar with that of Au foil. This indicated that the
Au+ disappeared almost completely (Fig. 6A-b,c). After feeding O2 again for 10 min (i.e., sample c),
the whiteline intensity increased slightly while an XANES similar with that for the sample a
appeared again. This result indicated that the amount of Au+ was re-increased nearly close to that
for the sample a (Fig. 6A-c,d). This trend could also be observed in the changes in Au-O
coordination over the course of these three samples (Fig. 6B, Table S4). An Au-O coordination was
present at a distance of 1.7 Å in the sample a (Fig. 6B-a), being consistent with the previously
reported results [37]. In the sample b, the Au-O coordination almost disappeared and the data were
similar to those of the Au foil (Fig. 6B-b,c). In the sample c, not surprisingly, the Au-O coordination
increased once again to be close to that for the sample a (Fig. 6B-a,d).
Additionally, the whiteline intensity of the XANES at the Ni K edge also showed high-low-high
13
Page 13 of 34

changes along with the respective with-without-with O2 reaction procedure (Fig. 7A). For the
sample a, the whiteline intensity at 8350 eV was slightly lower than that of the Ni2O3 (Fig. 7A-a,b)
because the Ni species in the sample a was a mixture of NiO and Ni2O3. Along with the O2 cutting
off and subsequently re-feeding, the whiteline intensity at 8350 eV became slightly lower for the
sample b (Fig. 7A-c) than that for the sample a (Fig. 7A-b) and then retrieved nearly close to that
for the sample a (Fig. 7A-d). For reference, the XANES of the NiO was also included and it
exhibited a much lower whiteline intensity than Ni2O3 (Fig. 7A-a,e). In addition, the changes in
Ni-O coordination across these three samples also demonstrated the same high-low-high trend (Fig.
7B). The Ni-O coordination was presented at a distance of 1.6 Å in the sample a, which is similar to
the Ni-O coordination of Ni2O3 [36] (Fig. 7B-a,b). In comparison with the sample a, the Ni-O
coordination was decreased slightly (Fig. 7B-c) in the sample b due to the O2 cutting off. However,
after re-feeding O2 the Ni-O coordination in the sample c was increased again close to the level of
the sample a (Fig. 7B-b,d). Combining the XANES results with the information from the XPS
studies, it was rational to infer that the high-low-high changes of the absorption intensity at 8350 eV
was attributable to the high-low-high change of Ni2O3 content.
3.5.3 Ni2O3-Au+ hybrid active sites
The Au-5/Ti and NiO-1/Ti-Cl catalysts delivered a very low activity for the benzyl alcohol
oxidation (<5%) and, even when they were physically mixed, the benzyl alcohol conversion still
remained very low (only 4.3%, entry 12 in Table 1). However, the NiO-1@Au-5/Ti-Cl (-N, and -A)
catalysts with the formation of NiO@Au ensembles all delivered much higher benzyl alcohol
conversion of >90% at 280 °C (Table 1, Fig. 1D, Fig. S4). These results indicated that the
co-existence of NiO and Au in the form of NiO@Au ensembles, in contrast to being physically
14
Page 14 of 34

mixed, was essential for the high activity for the alcohol oxidation. Additionally, the NiO@Au
ensembles were analyzed by TEM and the NiO-Au interface was clearly detected (Fig. S7A-B). The
EDX results (Fig. S7C) indicated the presence of elemental Au and Ni in the cycled parts in Fig.
S7A-B. As noted previously, the interface between NiO and Au is essential for high alcohol
oxidation activity [38]. As shown in Fig. S7A-B, only the lattice fringe of NiO was detected while
the lattice fringe of Au could not be detected because the gold particle was so large. Moreover, the
lattice fringe of the NiO segments was slightly bent rather than the linear lattice fringe of bulk NiO,
which might be a result of the formation of NiO@Au ensemble structure.
The Au-5/Ti, NiO-1/Ti-Cl, and NiO-1@Au-5/Ti-Cl catalysts were characterized by XPS, with
the results as summarized in Table S5. Interestingly, the Au+ content in the Au-5/Ti and the Ni2O3
content in the NiO-1/Ti-Cl were very low, while both was increased greatly in the
NiO-1@Au-5/Ti-Cl catalyst. The large difference in the Au+ and Ni2O3 contents in these catalysts
was associated with the pronounced difference in the benzyl alcohol conversion (Au-5/Ti: 5.2%,
NiO-1/Ti-Cl: 2.3%, and NiO-1@Au-5/Ti-Cl: 93.7%; Table 1). This observation indicated that the
Au+ and Ni2O3 were important to the catalyst performance. Additionally, in the with-without-with
O2 reaction procedure, over the NiO-1@Au-5/Ti-Cl catalyst the benzyl alcohol conversion also
exhibited high-low-high evolution behavior (e.g., 94% vs. 2% vs. 88%; Table 3).
The above XPS and XANES results clearly showed the high-low-high surface contents of Au+
and Ni2O3 but not NiO (Fig. 6 and Fig. 7). This observation was clearly associated with the
high-low-high evolution behavior of the benzyl alcohol conversion in the with-without-with O2
reaction procedure. Accordingly, hybrid active-sites of Ni2O3-Au+ was defined, which is the origin
of the dramatic activity improvement associated with the formation of the NiO@Au ensembles.
15
Page 15 of 34

With the sharp high-low-high evolution behavior of the benzyl alcohol conversion, the Au+ surface
content also showed the same sharp high-low-high change (e.g., 37.0% vs. 3.4% vs. 33.0%; Table
3). Meanwhile, the surface Ni2O3 exhibited a similar high-low-high content shift (e.g., 77.0% vs.
60.0% vs. 73.0%; Table 3), but the change was smaller when compared with the benzyl alcohol
conversion and the Au+ surface content. Clearly, there was a stronger relationship between the
benzyl alcohol conversion and the Au+ surface content, indicating that the Au+ might work as the
main active species for the alcohol oxidation with the Ni2O3 as an assisting species. It has been
reported in some studies that the Ni2O3 specimens play a key role not only in promoting the
formation of Au+ species and stabilizing them but also in acting as a surface-active-oxygen supplier
[39,40].
By nature, it could be proposed that Ni2O3 could not only increase and stabilize the Au+ located
at the NiO-Au interface but also could adsorb oxygen molecules into its oxygen vacancies and
activate the adsorbed molecular oxygen, which resulted in the active oxygen species. Such oxygen
species could be transferred onto the Au+ sites to react with alcohol while opening a new oxygen
vacancy for the next catalytic cycle.
4 Conclusions
The interesting date-cake-like active NiO@Au ensembles, which are characteristic of the partial
coverage of the large Au particles with small NiO segments, are successfully fabricated from
various precursors and are found to be highly active in the gas-phase oxidation of alcohols. In
addition, the Au and NiO loadings can be tuned within a large range without any degradation of the
catalytic performance of the NiO@Au ensembles. Notably, with the high-low-high changes in the
benzyl alcohol conversion following an operation with-without-with O2 in the reactant stream, the
16
Page 16 of 34

high-low-high surface content of Au+ is associated with an analogous change with Ni2O3 but not
NiO. The hybrid active-sites of Ni2O3-Au+ are defined accordingly, which is the origin for the
low-temperature activity. The NiO@Au ensembles show synergistic effect to promote the
generation of high density Ni2O3-Au+ sites at the interface between NiO and Au particles. By nature,
the Ni2O3 specimens play a key role not only in promoting the formation of Au+ and stabilizing
them but also in acting as surface-active-oxygen species supplier.
Acknowledgements
We thank the NSF of China (20973063, 21076083, 21273075), the MOST of China
(2011CB201403), the Fundamental Research Funds for the Central Universities, the Specialized
Research Fund for the Doctoral Program of Higher Education (20090076110006), the Shanghai
Rising-Star Program (10QH1400800) and the Shanghai Leading Academic Discipline Project
(B409) for their financial support, as well as the Electron Spectroscope Center of the East China
Normal University for providing the TEM measurements.
References
[1] S. Biella, M. Rossi, Chem. Commun. 2003, 378-380.
[2] A.S.K. Hashmi, G.J. Hutchings, Angew. Chem. Int. Ed. 45 (2006) 7896-7936.
[3] C.D. Pina, E. Falletta, M. Rossi, J. Catal. 260 (2008) 384-386.
[4] A. Corma, H. Garcia, Chem. Soc. Rev. 37 (2008) 2096-2126.
[5] J. P. Mao‚ M. M. Deng‚ Q. S. Xue‚ L. Chen‚ Y. Lu, Catal. Commnun. 10 (2009) 1376-1379.
[6] L. Prati, F. Porta, Appl. Catal. A-Gen. 291 (2005) 199-203.
[7] A. Villa, N. Janjic, P. Spontoni, D. Wang, D.S. Su, L. Prati, Appl. Catal. A-Gen. 364 (2009)
221-228.
17
Page 17 of 34

[8] F. Li, Q. Zhang, Y. Wang, Appl. Catal. A-Gen. 334 (2008) 217-226.
[9] X. Wang, H. Kawanami, S.E. Dapurkar, N.S. Venkataramanan, M. Chatterjee, T. Yokoyama, Y.
Ikushima, Appl. Catal. A-Gen. 349 (2008) 86-90.
[10] M. Haruta, T. Kobayashi, H. Sano, N. Yamada, Chem. Lett. 2 (1987) 405-408.
[11] R. Meyer, C. Lemire, S.K. Shaikhutdinov, H.J. Freund, Gold Bull. 37 (2004) 72-124.
[12] M.S. Chen, D.G. Goodman, Science 306 (2004) 252-255.
[13] A.A. Herzing, C.J. Kiely, A.F. Carley, P. Landon, G.J. Hutchings, Science 321 (2008)
1331-1335
[14] G.C. Bond, D.T. Thompson, Catal. Rev. Sci. Eng. 41 (1999) 319-388.
[15] B.K. Min, C.M. Friend, Chem. Rev. 107 (2007) 2709-2724.
[16] J.C. Fierro-Gonzalez, B.C. Gates, Chem. Soc. Rev. 37 (2008) 2127-2134
[17] P.X. Huang, F. Wu, B.L. Zhu, X.P. Gao, H.Y. Zhu, T.Y. Yan, W.P. Huang, S.H. Wu, D.Y. Song,
J. Phys. Chem. B 109 (2005) 19169-19174.
[18] J. Guzman, S. Carrettin, A. Corma, J. Am. Chem. Soc. 127 (2005) 3286-3287.
[19] M. Comotti, W.C. Li, B. Spliethoff, F. Schüth, J. Am. Chem. Soc. 128 (2006) 917-924.
[20] J. Li, N. Ta, Y. Li, W.J. Shen, Chin. J. Catal., 29 (2008) 823.
[21] R. Si, M. Flytzani-Stephanopoulos, Angew. Chem. Int. Ed., 47 (2008) 2884-2887.
[22] A. Abad, P. Concepción, A. Corma, H. García, Angew. Chem. Int. Ed. 44 (2005) 4066-4069.
[23] J.A. Rodriguez, S. Ma, P. Liu, J. Hrbek, J. Evans, M. Pérez, Science 318 (2007) 1757-1760.
[24] C.M.Y. Yeung, K.M.K. Yu, Q.J. Fu, D. Thompsett, M.I. Petch, S.C. Tsang, J. Am. Chem. Soc.
127 (2005) 18010-18011.
[25] T. Fujitani, I. Nakamura, T. Akita, M. Okumura, M. Haruta, Angew. Chem. Int. Ed. 48 (2009)
18
Page 18 of 34

9515-9518.
[26] G.F. Zhao, H.Y. Hu, M.M. Deng, Y. Lu, Chem. Commun. 47 (2011) 9642-9644.
[27] G.F. Zhao, H.Y. Hu, W. Chen, Z. Jiang, S. Zhang, J. Huang, Y. Lu, Catal. Sci. Technol. 3 (2013)
404-408.
[28] G.F. Zhao, M.M. Deng, Y.F. Jiang, H.Y. Hu, J. Huang, Y. Lu, J. Catal. 301 (2013) 46-53.
[29] G.F. Zhao, H.Y. Hu, M.M. Deng, M. Ling, Y. Lu, Green Chem. 13 (2011) 55-58.
[30] G.F. Zhao, H.Y. Hu, M.M. Deng, M. Ling, Y. Lu, ChemCatChem 3 (2011) 1629-1636.
[31] M.M. Deng, G.F. Zhao, Q.S. Xue, L. Chen, Y. Lu, Appl. Catal. B-Environ. 99 (2010) 222-228.
[32] J.P. Mao, M.M. Deng, L. Chen, Y. Liu, Y. Lu, AIChE J. 56 (2010) 1545-1556.
[33] J. Fan, Y.H. Dai, Y.L. Li, N.F. Zheng, J.F. Guo, X.Q. Yan, G.D. Stucky, J. Am. Chem. Soc. 131
(2009) 15568-15569.
[34] J. Guzman, B.C. Gates, J. Am. Chem. Soc. 126 (2004) 2672-2673.
[35] M.K. Oudenhuijzen, J.A. van Bokhoven, J.T. Miller, D.E. Ramaker, D.C. Koningsberger, J.
Am. Chem. Soc. 127 (2005) 1530-1540.
[36] J.A. van Bokhoven, C. Louis, J.T. Miller, M. Tromp, O.V. Safonova, P. Glatzel, Angew. Chem.
Int. Ed. 45 (2006) 4651-4654.
[37] J.T. Miller, A.J. Kropf, Y. Zha, J.R. Regalbuto, L. Delannoy, C. Louis, E. Bus, J.A. van
Bokhoven, J. Catal. 240 (2006) 222-234.
[38] A. Villa, G.M. Veith, D. Ferri, A. Weidenkaff, K.A. Perry, S. Campisia, L. Prati, Catal. Sci.
Technol. 3 (2013) 394-399.
[39]A. Trovarelli, Catal. Rev. Sci. Eng. 38 (1996) 439-520.
[40] G. F. Zhao, J. Huang, Z. Jiang, S. Zhang, L. Chen, Y. Lu, Appl. Catal. B-Environ. 140-141
19
Page 19 of 34

(2013) 249-257
20
Page 20 of 34

List of Figure Captions:
Fig. 1 (A) SEM image of the Au-5/Ti catalyst; (B) XRD patterns of the Au-5/Ti catalyst and Ti
powder; (C) TEM image of the Au-5/Ti catalyst; (D) TEM image of the NiO-1@Au-5/Ti-Cl
catalyst.
Fig. 2 XRD patterns of the NiO-1@Au-5/Ti-Cl catalyst with the formation of NiO.
Fig. 3 Influences of NiO and Au loadings on the catalytic performance of the NiO@Au/Ti-Cl
catalyst for the gas-phase oxidation of benzyl alcohol at 280 °C using O2/hydroxyl = 0.6 and
WHSV = 20 h-1. (A) The influence of NiO loading (setting Au loading constant at 5 wt%); (B) The
influence of Au loading (setting Au/NiO weight ratio constant at 5). Pre-activated process: the
catalysts underwent 1 h reaction in the gas-phase oxidation of benzyl alcohol at 380 °C with
O2/hydroxyl = 0.6 and WHSV = 20 h-1.
Fig. 4 (A) Longer-term test using a reaction/regeneration cycle for the NiO-1@Au-5/Ti-Cl catalyst
in the gas-phase oxidation of benzyl alcohol. Reaction conditions: 280 °C, O2/hydroxyl = 0.6,
WHSV = 20 h-1. Regeneration conditions: oxidation at 400 ºC in air for 3 h to burn off carbon
deposits; (B) Longer-term test for the NiO-1@Au-5/Ti-Cl catalyst in the gas-phase oxidation of
cyclopropyl carbinol. Reaction conditions: 280 ºC, O2/hydroxyl = 0.6, WHSV = 20 h-1.
Fig. 5 XPS spectra in O1s, Ni2p, and Au4f regions. Sample a: NiO-1@Au-5/Ti-Cl reacting with
benzyl alcohol at 280 ºC for 0.5 h in the present of O2. Benzyl alcohol conversion = 94%; Sample b:
Sample a after switching off the O2 feed followed by a 15 min reaction with benzyl alcohol alone at
280 ºC. Benzyl alcohol conversion = 2%; Sample c: Sample b catalyzing benzyl alcohol oxidation
by re-feeding O2 again under the same conditions as for the sample a for 10 min. Benzyl alcohol
conversion = 88%.
21
Page 21 of 34

Fig. 6 XANES of Au L3-edge (A) and Modulus of the Fourier-transform Au L3-edge signal (B) of
several catalysts: (a) sample a as in Fig. 5-a; (b) Au foil; (c) sample b as in Fig. 5-b; (d) sample c as
in Fig. 5-c.
Fig. 7 Ni K-edge XANES (A) and modulus of the Fourier-transform of Ni K edge signal (B) of
several catalysts. (a) Ni2O3; (b) sample a as in Fig. 5-a; (c) sample b as in Fig. 5-b; (d) sample c as
in Fig. 5-c; (e) NiO.
22
Page 22 of 34

Table 1 Gas-phase oxidation of benzyl alcohol catalyzed by various designed catalysts.
d
DAu
(nm)
-
TReact.
(oC)
280
280
280
280
280
280
280
260
240
230
220
280
Conversion
(%)
Selectivityf
(%)
Entry
Catalyst a,b,c
1
Ti-powder
0
0
2
Au-5/Ti
35
33
-
5.0
99.7
98.5
99.8
98.9
98.4
98.5
98.6
98.6
98.9
99.2
99.6
3
NiO-1@Au-5/Ti-Cl
NiO-1/Ti-Cl
93.7
2.3
4
5
Au-5@NiO-1/Ti-Cl
NiO-1@Au-5/Ti-C
NiO-1@Au-5/Ti-N
NiO-1@Au-5/Ti-N
NiO-1@Au-5/Ti-N
NiO-1@Au-5/Ti-N
NiO-1@Au-5/Ti-N
Au-5/Ti&NiO-1/Ti-Cle
33
34
34
34
34
34
34
34
21.1
92.1
94.6
89.1
87.2
86.0
63.5
4.3
6
7
8
9
10
11
12
a The Au loading was 5 wt% and NiO loading (if has) was 1 wt%; see experimental section for detailed catalyst
preparation; b The catalysts were all pre-activated over the gas-phase oxidation of benzyl alcohol at 380 ºC for 1 h
using the molar ratio of O2 to alcoholic hydroxyl (O2/hydroxyl) = 0.6 and weight hourly space velocity (WHSV) =
20 h-1, and then evaluated at 220-280 °C with other reaction parameters unchanged; c To represent different nickel
d
precursors: Cl-NiCl2, C-Ni(Ac)2, N-Ni(NO3)2; The particle size was estimated from the XRD patterns using
f
Scherrer’s equation. e The physical mixture of the catalysts Au-5/Ti and NiO-1/Ti-Cl (mass ratio, 50:50). The
byproducts are CO, CO2, benzene, toluene and benzoic acid in traces, whose total selectivity is 1.0-1.5%.
23
Page 23 of 34

Table 2. Oxidation of various alcohols over the NiO-1@Au-5/Ti-Cl catalyst using WHSV of 20 h-1.
Substrate
O2/hydroxyl
TReact.
(oC)
300
340
300
360
340
360
Conversion
(%)
Selectivity
(%)
(mol/mol)
0.7
1-phenylethanol
2-phenylethanol
1-octanol
95.3
98.3
0.7
51.2
98.1
0.6
51.9
92.5
2-octanol
0.6
23.4
66.9
Cyclohexanol
1,2-propanediol
0.6
67.8
97.4
0.7
91.6
71.0
24
Page 24 of 34

Table 3. The XPS results of the NiO-1@Au-5/Ti-Cl samples.
Fraction (%)
Elements
Au
B.E. (eV)
Sample aa
63.0
Sample bb
96.6
Sample cc
67.0
Au0
84.0
Au+
84.7
37.0
3.4
33.0
Ni
O
NiO
854.4
856.2
529.2
532.1
530.4
23.0
40.0
27.0
Ni2O3
NiO
77.0
60.0
73.0
11.0
20.0
13.0
Ni2O3
TiO2
53.0
44.0
51.0
36.0
36.0
36.0
a
Sample a: NiO-1@Au-5/Ti-Cl reacting with benzyl alcohol at 280 ºC in the presence of O2 for 0.5 h. Benzyl
b
alcohol conversion = 94%; Sample b: Sample a after switching off the O2 feed followed by a 15 min reaction
c
with benzyl alcohol alone at 280 ºC. Benzyl alcohol conversion = 2%; Sample c: Sample b catalyzing benzyl
alcohol oxidation by re-feeding O2 again under the same conditions as for sample a for 10 min. Benzyl alcohol
conversion = 88%.
25
Page 25 of 34

200 nm
A
2.5 m
C
D
NiO
Au
Au
Particle
Particle
10 nm
10 nm
Fig. 1 (A) SEM image of the Au-5/Ti catalyst; (B) XRD patterns of the Au-5/Ti catalyst and Ti
powder; (C) TEM image of the Au-5/Ti catalyst; (D) TEM image of the NiO-1@Au-5/Ti-Cl
catalyst.
26
Page 26 of 34

30
40
50
60
70
2-theta(degree)
Fig. 2 XRD patterns of the NiO-1@Au-5/Ti-Cl catalyst with the formation of NiO.
27
Page 27 of 34

100
80
60
40
20
0
Conversion(%)
Selectivity(%)
A
0
1
2
3
4
5
6
7
NiO loading (wt%)
100
80
60
40
20
0
Conversion
Selectivity
B
1
2
3
4
5
Au loading (wt%)
Fig. 3 Influences of NiO and Au loadings on the catalytic performance of the NiO@Au/Ti-Cl
catalyst for the gas-phase oxidation of benzyl alcohol at 280 °C using O2/hydroxyl = 0.6 and
WHSV = 20 h-1. (A) The influence of NiO loading (setting Au loading constant at 5 wt%); (B) The
influence of Au loading (setting Au/NiO weight ratio constant at 5). Pre-activated process: the
catalysts underwent 1 h reaction in the gas-phase oxidation of benzyl alcohol at 380 °C with
O2/hydroxyl = 0.6 and WHSV = 20 h-1.
28
Page 28 of 34

100
80
60
40
20
0
1st
2st
regeneration
regeneration
Conversion(%)
Selectivity(%)
A
0
40
80
120
160
Time on stream(h)
100
80
60
Conversion
Selectivity
40
B
0
50
100 150 200 250 300
Time on stream(h)
Fig. 4 (A) Longer-term test using a reaction/regeneration cycle for the NiO-1@Au-5/Ti-Cl catalyst
in the gas-phase oxidation of benzyl alcohol. Reaction conditions: 280 °C, O2/hydroxyl = 0.6,
WHSV = 20 h-1. Regeneration conditions: oxidation at 400 ºC in air for 3 h to burn off carbon
deposits; (B) Longer-term test for the NiO-1@Au-5/Ti-Cl catalyst in the gas-phase oxidation of
cyclopropyl carbinol. Reaction conditions: 280 ºC, O2/hydroxyl = 0.6, WHSV = 20 h-1.
29
Page 29 of 34