Chloride-free and water-soluble Au complex for preparation of supported small nanoparticles by impregnation method

Article abstract of DOI:10.1016/j.jcat.2017.07.002

A novel simple and easy impregnation method for preparation of small gold (Au) nanoparticle (NP) catalysts (<3?nm) deposited on various supports including silica, which is difficult to be applied for conventional methods, has been developed. Chloride-free and water-soluble precursor, Au complexes coordinated with β-alanine, were successful for the preparing Au NPs, which exhibited an average diameter less than 3?nm. Thermal behavior of the Au complex was investigated by TG-DTA and in situ XAFS. XAFS analyses and DFT calculations revealed a molecular structure of the Au complex to be square-planar coordination structure and mononuclear complex of Au³⁺. Lower decomposition and reduction temperature of the chloride-free Au complex prevented Au atoms from aggregating and from following growth of Au particles. The prepared Au/SiO₂ showed high selectivity for hydrogenation of 1-nitro-4-vinylbenzene into 1-ethyl-4-nitorobenzene and good performance for removal of unpalatable aroma by means of adsorption of polysulfide molecules.

Full text of DOI:10.1016/j.jcat.2017.07.002

Journal of Catalysis 353 (2017) 74–80
Contents lists available at ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
Chloride-free and water-soluble Au complex for preparation
of supported small nanoparticles by impregnation method
a,
⇑
a
a
a
a
Haruno Murayama , Takayuki Hasegawa , Yusuke Yamamoto , Misaki Tone , Moemi Kimura ,
b
c
d
e
e
a,
⇑
Tamao Ishida , Tetsuo Honma , Mitsutaka Okumura , Atsuko Isogai , Tsutomu Fujii , Makoto Tokunaga
a
Department of Chemistry, Kyushu University, Nishi-ku, Fukuoka 819-0395, Japan
Research Center for Gold Chemistry, Graduate School of Urban Environmental Sciences, Tokyo Metropolitan University, Hachioji, Tokyo 192-0397, Japan
Japan Synchrotron Radiation Research Institute, SPring-8, Hyogo 679-5198, Japan
Department of Chemistry, Osaka University, Toyonaka, Osaka 560-0043, Japan
b
c
d
e
National Research Institute of Brewing, Higashihiroshima, Hiroshima 739-0046, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
A novel simple and easy impregnation method for preparation of small gold (Au) nanoparticle (NP) cat-
alysts (<3 nm) deposited on various supports including silica, which is difficult to be applied for conven-
tional methods, has been developed. Chloride-free and water-soluble precursor, Au complexes
coordinated with b-alanine, were successful for the preparing Au NPs, which exhibited an average diam-
eter less than 3 nm. Thermal behavior of the Au complex was investigated by TG-DTA and in situ XAFS.
XAFS analyses and DFT calculations revealed a molecular structure of the Au complex to be square-planar
Received 12 May 2017
Revised 30 June 2017
Accepted 3 July 2017
Keywords:
3+
coordination structure and mononuclear complex of Au . Lower decomposition and reduction temper-
ature of the chloride-free Au complex prevented Au atoms from aggregating and from following growth
Gold catalysts
Supported gold
Silica
of Au particles. The prepared Au/SiO
2
showed high selectivity for hydrogenation of 1-nitro-4-
vinylbenzene into 1-ethyl-4-nitorobenzene and good performance for removal of unpalatable aroma
by means of adsorption of polysulfide molecules.
Impregnation
Chloride-free Au complex
Ó 2017 Elsevier Inc. All rights reserved.
À
1
. Introduction
studies have revealed that the adsorption of Cl ion onto an oxygen
2
defect site of TiO enhances the aggregation of Au atoms [13,14].
Gold (Au) nanoparticles (NPs) on supports such as metal oxides,
For practical use in the industry, the IP methods have advan-
tages that are suitable for large scale preparing of catalysts. Post-
impregnation treatments to obtain supported small Au NPs have
been reported. High-temperature hydrogen reduction is effective
active carbons, and polymers, have received considerable atten-
tions in recent years as catalysts [1,2], adsorbents [3], and optical
and electronic devices [4]. Generally, the size effect is critical to
the functions of Au NPs, i.e., smaller (<5 nm) particles are indis-
pensable for high catalytic activity [5–11]. Therefore, special tech-
niques have been established for preparing supported Au NPs,
including deposition-precipitation (DP) and co-precipitation (CP)
methods. These techniques are applicable to various metal oxides
for Au/TiO
2
and Au/C prepared by an IP method with AuCl
3
[15].
À
To remove Cl ions, washing with ammonia has been used as the
post-treatment of IP methods [16–20]. Delannoy et al. have
reported that small Au NPs (3–4 nm) on any type of oxides includ-
ing SiO
with HAuCl
2
were obtained by a modified incipient wetness IP method
O [20]. In the procedure, the IP using HAuCl
except for acidic materials such as SiO
2
, because of electrostatic
4
Á4H
2
4
À
repulsion between the Au(OH)
4
precursor and the support surface,
aqueous solution was followed by a post-treatment with an aque-
ous ammonia solution. The ammonia washing removes chloride
which is negatively charged. In contrast, the most conventional
impregnation (IP) method using the standard precursor chloroau-
3
+
ligands from Au precursor and forms an amino-hydroxo-aquo
cationic gold complex, which interacts with supports. When
ammonia contacts Au in solution, the formation of explosive Au–
ammonia complexes are provoked [21,22]. For the sake of safety,
the remains of no Au in the solution is a prerequisite.
ric acid (HAuCl
4
Á4H
2
O) is unsuccessful for the preparing small Au
NPs, because it results in the growth of particles via the aggrega-
tion of Au atoms [12]. Density functional theory (DFT) calculation
Another means to obtain supported small Au NPs has been
⇑
Corresponding authors.
E-mail addresses: haruno9@chem.kyushu-univ.jp (H. Murayama), mtok@chem.
reported that using of chloride-free precursors such as Au(OAc)
for the preparation [23,24]. Recently, Sakurai et al. have reported
3
kyushu-univ.jp (M. Tokunaga).
http://dx.doi.org/10.1016/j.jcat.2017.07.002
0
021-9517/Ó 2017 Elsevier Inc. All rights reserved.

H. Murayama et al. / Journal of Catalysis 353 (2017) 74–80
75
a detailed study of the IP method of preparing Au NPs with Au
OAc) as a precursor [24]. Au NPs supported on Al , TiO
CeO , and SiO have successfully been prepared with diameters
between 3 and 6 nm. However, because of the poor solubility of
Au(OAc) , a tedious alkaline solution treatment is required to pre-
pare the Au solution and is followed by evaporation of the dilute
aqueous solution for IP. In addition, the synthesis of Au(OAc)
requires two steps via unstable Au(OH)
One of the simplest approaches to realize direct IP method for
the preparing small Au NPs on various supports including SiO is
by total pore volume of the supports measured by nitrogen adsorp-
tion. For impregnation on 0.99 g of TiO , MCM-41, and Ketchen
(
3
2
O
3
2
,
2
2
2
black, the volume of Au solutions were 0.43 mL, 0.93 mL, and
1.51 mL, respectively. The Au loading amounts were 1 wt%
(10 mg of Au). After impregnation, the solid was calcined in air.
The heating rate was kept at 5 °C/min until 300 °C and then the
temperature was kept for 0.5 h.
3
3
3
.
2.3. Sample characterization
2
the use of chloride–free and water–soluble Au precursors. Rai
et al. have reported that the reaction between chloroauric acid
High–angle annular dark field scanning transmission electron
microscope (HAADF–STEM) images were acquired with a JEM-
ARM200F microscope (JEOL) operated at 200 kV. Powder X-ray
diffraction (XRD) data were collected on a MiniFlex600 (Rigaku
and -asparagine results in the formation of 1:1 complex [25]. On
L
the basis of the reaction, Kazachenko et al. have synthesized the
water–soluble Au complexes coordinated with glycine, histidine,
and tryptophan [26]. However the structures of the Au complexes
have scarcely been characterized. Exploiting the high solubility of
these complexes in water, we developed an incipient wetness
method involving only a very small amount of water. In the field
of preparing colloidal Au NPs, amino acids and amines have been
used as reducing and capping agents [27,28]. In our experiments,
amino acids act as ligands of Au complexes, which are completely
different from the roles of them in the synthesis of colloidal NPs.
In this paper, we describe a novel IP method followed by calci-
nation in air, which is fit for practical use, using readily accessible
Au complexes coordinated with b-alanine (Au–(b-ala)) as precur-
sors. Our method enables anchoring of the Au NPs onto acidic sup-
Co.) diffractometer using Cu Ka radiation. Au LIII-edge XAFS spectra
were collected at the BL14B2 beamline of SPring-8 (Hyogo, Japan).
Synchrotron radiation from the storage ring was monochromated
by a Si311 double-crystal. The experiments were conducted in the
transmission mode at room temperature. The XAFS spectra were
analyzed by using the program REX2000 (Rigaku Co.). During the
calcination process of preparing the supported Au NPs, in situ XAFS
spectra were also collected using the same optical system with the
XAFS measurements. The silica supported Au precursors were
placed in a quartz tube reactor and were heated by using the same
temperature program as that used in the preparation method. The
collection of a XAFS spectrum required approximately 3.5 min. TG-
DTA data were collected on a STA7300 (Hitachi High-Tech) thermal
analyzer. The samples were heated by using the same temperature
program as that used in the preparation method.
2
ports such as SiO , which are difficult to coat via DP and CP
methods. We determine the coordination structure and thermal
behavior of the Au–(b-ala) and oxidation state of Au in the complex
by analyses of Au LIII-edge X-ray absorption fine structure (XAFS)
spectra, thermogravimetry-differential thermal analysis (TG-
DTA), and DFT calculations. Catalytic performance of the prepared
2
.4. DFT calculations
Using DFT calculations, we optimized the geometry of the Au
Au/SiO
2
was investigated by hydrogenation of 1-nitro-4-
complex with b-alanine. The PBE0 functional was used for these
calculations. The SDD basis set was used for the Au atom, and
the 6-31+G(d,p) basis set was for N, C, O, and H atoms. In these cal-
culations, we assumed that the net charge of this complex was
neutral.
vinylbenzene, which has been a reaction studied using Au NPs on
various supports. In addition, adsorption of unpalatable aroma in
drinks, which is one of the possible applications of the Au/SiO
other than catalysts, was demonstrated.
2
2
. Experimental
.1. Synthesis of Au complexes
Au complexes coordinated with b-alanine (Au–(b-ala)) were
2.5. Hydrogenation of 1-nitro-4-vinylbenzene and adsorption of
dimethyl trisulfide (DMTS) experiments
2
The Au NPs prepared by our method, 1-nitro-4-vinylbenzene
(74.6 mg, 0.5 mmol), triethylene glycol dimethyl ether (30 lL) as
prepared using examples from synthetic procedures reported by
Kazachenko et al. [26]. Ethanol (3 mL) was added to 2 mL of a
NaOH aqueous solution (1.25 M) in which b-alanine (H N(CH ) -
2 2 2
COOH) (Tokyo Chemical Industry Co., Ltd.) (1.82 mmol) was
dissolved. A second solution was prepared by adding 1 mL of a
an internal standard, and ethyl acetate (2.0 mL) as solvent were
introduced into a glass inner tube of a stainless steel autoclave.
The autoclave was sealed, following the filling of hydrogen at
2.0 MPa. Catalytic reactions were then performed at 100 °C for
24 h. After the reactions, the mixture was filtered and analyzed
by a gas chromatography (GC) using an Agilent GC 6850 series II
equipped with an Agilent J&W HP-1 capillary column (0.32 mm i.
d., 30 m). For the adsorption experiments, the silica supported Au
HAuCl
4
2
Á4H O (Tanaka Kikinzoku Kogyo K.K.) (0.227 M) aqueous
solution to 4 mL of ethanol. The first solution containing
b-alanine and NaOH was then added to the second solution. The
mixed solution was left at À18 °C for 12 h. The precipitates were
collected by centrifugation and washed with water/ethanol (3/7
v/v%) solution, in which b-alanine was soluble, but Au–(b-ala)
was insoluble. The solid was dried under vacuum to obtain
Au–(b-ala) powder.
NPs (Au/SiO ) was added to 4 mL of ethanol solution of DMTS
2
(Tokyo Chemical Industry Co., Ltd.) (4.7 mg/L) and the mixture
was then kept at room temperature. The concentration of DMTS
was analyzed by GC. The adsorption amounts of DMTS were esti-
mated as differences from the initial concentration of DMTS.
2.2. Preparation of supported Au NPs
3. Results and discussion
Supported Au NPs were prepared by the incipient wetness IP
3.1. Sizes of Au NPs on various supports
method. In addition to the Au–(b-ala), HAuCl
a precursor. The precursor was dissolved in 1.20 mL of water,
and the solution was added to 0.99 g of SiO (CARiACT Q-15, Fuji
Silysia Chemical Ltd.). The volume of Au solutions were decided
4
Á4H
2
O was used as
The Au loading per obtained catalyst (Au/SiO
microwave plasma–atomic emission spectrometry to be 0.94 wt%.
Fig. 1a shows a portion of a HAADF–STEM image of the Au/SiO
2
) was analyzed by
2
2

7
6
H. Murayama et al. / Journal of Catalysis 353 (2017) 74–80
Fig. 1. HAADF–STEM image of the Au/SiO
using the Au–(b-ala) and HAuCl
Á4H
2
2
prepared by using the Au–(b-ala) (a). The inset shows size distribution of Au particles. XRD patterns for the Au/SiO prepared by
4
2
O (b). Schematic view of the preparation of Au/SiO
2
using the Au–(b-ala) and HAuCl
4
Á4H
2
O (c).
prepared by using the Au–(b-ala). The size distribution of the
observed spherical particles is displayed in a histogram in the inset
of Fig. 1a; their average diameter was 2.8 ± 0.8 nm. Moreover,
HAADF–STEM images of the Au NPs dispersed on various supports,
including TiO , MCM-41, and Ketchen black, using the Au–(b-ala)
2
as a precursor are displayed in the supporting information,
Fig. S1, which shows spherical NPs with the average diameters of
e.g., the decomposition temperature and reduction temperature, of
the precursors is important for preparing small metal NPs. The
thermal decomposition profile of the Au–(b-ala) was investigated
using TG-DTA. Three mass losses were observed at approximately
65 °C, 135 °C, and 300 °C, as shown in Fig. 2. The slow mass loss
from room temperature to 135 °C was identified as dehydration
of contained moisture, i.e., a drying process, on the basis of the very
small endothermic DTA peaks. Other mass losses observed at
approximately 135 °C and 300 °C with exothermic DTA peaks were
indicative of ligand dissociation of the Au–(b-ala).
2
.8 ± 0.7 nm, 2.5 ± 0.6 nm, and 2.9 ± 1.0 nm, respectively.
The Au–(b-ala) is thus applicable to the preparation of small Au
NPs on various supports. The Au/SiO were also characterized by
2
powder XRD. The XRD patterns (Fig. 1b) showed some sharp small
peaks and broad peaks. The sharp small peaks observed at 38.2°,
To determine the reduction temperature of the Au–(b-ala) and
chloroauric acid samples, in situ Au LIII-edge XAFS spectra collected
during the calcination process were analyzed. Before calcination, a
characteristic peak, the ‘white line’, which was attributed to the
excitation from the 2p3/2 to the 5 d or 6 s states, was observed at
11,922 eV in the spectra of Au–(b-ala) and chloroauric acid, as
shown in Fig.3a and b, respectively [29,30].
4
4.4°, 64.4°, and 77.6° in the pattern of the Au/SiO
2
prepared by
using HAuCl O were assigned to the (1 1 1), (2 0 0), (2 2 0),
4
Á4H
2
and (3 1 1) planes of Au metal, respectively. The small broad 111
peak of Au metal was also observed at the same angle in the
2
diffraction pattern for Au/SiO prepared by using Au–(b-ala). The
crystallite size was estimated to be 2.6 nm for the Au–(b-ala) sam-
ple according to the Scherrer equation and the FWHM value of the
Analyses of the white line intensity can be used to identify the
oxidation state [30]. The intensity of the white line for Au–(b-ala)
1
11 peak. The average diameter of the spherical particles observed
2 3
was approximately the same as that for Au O , thus indicating that
in the HAADF-STEM images was approximately the same as the
crystallite size estimated from the XRD pattern, thus indicating
that the Au NPs were supported on the SiO as small single Au
2
metal crystals. The crystallite size of Au NPs on the chloroauric acid
sample was also estimated to be 30.8 nm. Compared with the
preparation using the conventional precursor, HAuCl
Au–(b-ala) yielded smaller Au NPs that were highly dispersed on
the SiO supports. The relationships between the obtained particle
4 2
Á4H O, the
2
sizes and the procedures for preparation are summarized in Fig. 1c.
The broad peaks observed at approximately 20–30° were due to
amorphous SiO
2
.
3.2. Thermal analyses of the Au–(b-ala)
In general, heating enhances aggregation of metal atoms during
the reduction and calcination process. Therefore thermal behavior,
Fig. 2. TG-DTA curves of the Au–(b-ala).

H. Murayama et al. / Journal of Catalysis 353 (2017) 74–80
77
a
b
R.T.
R.T.
3
00°C
300°C
c
0
Fig. 3. Normalized X-ray absorption near-edge structure (XANES) spectra for silica supported Au–(b-ala) (a), HAuCl
fitting analyses for silica supported Au–(b-ala) and HAuCl during the calcination process as a function of temperature in flowing air (c). The inset shows the XANES spectrum
observed at 190 °C (open circle) and corresponding curve fitted (solid line) from initial state (long dashed short dashed line) and final state (dashed line).
4
(b), and the fraction of Au , x, determined by the pattern
4
the Au oxidation state of the Au–(b-ala) was +3. As a result of the
calcination, the white line intensity became weak and finally dis-
appeared. The samples obtained after the calcination were identi-
shown in the TG-DTA curves was attributed to a later decomposi-
tion of ligands that weakly interacted with the Au metal.
0
Given that the Au concentration of the prepared complex pow-
der was determined to be 39 wt% by microwave plasma atomic
emission spectrometry (MP-AES), the molar ratio of b-ala ligand/
Au was estimated to be 1.2. The details of this estimation are
described in Fig. S2 in the supporting information.
0
fied as Au by fingerprint analyses of their spectra. In addition,
two isosbestic points were observed at 11,928 eV and 11,954 eV
for Au–(b-ala) and at 11,925 eV and 11,953 eV for chloroauric acid.
These results indicate that the Au–(b-ala) and chloroauric acid
were reduced via a one-step reaction and that the oxidation state
of Au was changed from +3 to 0 without forming intermediates.
Hence, the XANES spectra were deconvoluted into the initial-
state and the final-state spectra. The inset in Fig. 3a shows the
XANES spectrum obtained at 190 °C and the corresponding curves
3.3. Structure of the Au–(b-ala)
À
We expected the Au–(b-ala) to not contain Cl ions because the
aggregation of Au atoms was inhibited. In Raman spectra of the Au/
0
fitted from the Au–(b-ala) complex (initial-sate) and the Au spec-
SiO
observed before and after calcination. On the other hand, Raman
spectrum of before calcined Au/SiO from HAuCl exhibits two
2
prepared by using the Au–(b-ala) (Fig. S3a), no peak is
tra (final-state) as a representative fitting result. The pattern fitting
converged, yielding an R–factor of 0.072%. This fitting result indi-
2
4
3
+
0
À1
À1
cates that the sample contained 22% Au and 78% Au . The frac-
peaks at 322.3 cm and 345.0 cm , which are characteristic of
Au–Cl vibrations [20]. The peaks disappeared after calcination.
These results indicate that there is no AuACl coordination in the
Au–(b-ala). The structure of the Au–(b-ala) was determined by
analyses of Au LIII-edge XAFS spectra as shown in Fig. 4.
0
tions of Au , x, at each temperature were determined from the
pattern fitting analyses and the following relationship: [100% = x
0
3+
%
Au + (100 À x)% Au ]. Fig. 3c shows a plot of the x values
obtained at each temperature, which indicates that the reduction
3
temperature (x = 50%) of the Au–(b-ala) and chloroauric acid were
The frequency and the relative amplitude of the k -weighted
1
70 °C and 200 °C, respectively. Compared with chloroauric acid,
2 3
oscillation for the Au–(b-ala) were similar to those for Au O , as
the Au–(b-ala) was reduced at a lower temperature.
is the dominant peak intensity of the Fourier transform function,
which appeared in the range from 0.13 to 0.20 nm. In contrast, they
The reduction temperature for the Au–(b-ala) determined by
the XANES analyses is in accordance with the temperature of main
mass loss observed via the TG-DTA analysis, as shown in Fig. 2. At
this temperature, the Au–(b-ala) decomposed via reduction from
differed from those for the AuCl
inant peak in the Fourier transform function for the AuCl
3
, as shown in Fig. 4a and b. A dom-
, which
3
was attributed to the scattering of photoelectrons ejected from
the Cl atoms located at the nearest distances from the absorber
the Au complex to Au0 metal. The third mass loss at 300 °C
3
+

7
8
H. Murayama et al. / Journal of Catalysis 353 (2017) 74–80
Table 2
a
The optimized geometries of the Au–(b-ala).
Au O3
2
Distance/nm
Angle and dihedral angle/°
R(10Au-9O)
R(10Au-11O)
R(10Au-13O)
R(10Au-1 N)
0.201999
0.196876
0.195472
0.208235
A(11O-10Au-13O)
A(13O-10Au-9O)
A(11O-10Au-1N)
A(9O-10Au-1N)
90.472
89.059
84.341
96.131
1.086
D(11O-13O-9O-1N)
Au–(β-ala)
AuCl3
b
AuCl3
Au O3
2
Au–(β-ala)
3
Fig. 4. k -weighted EXAFS oscillation (a) and Fourier transform (b) (without phase
correction) of the Au–(b-ala) pellet (black solid line), Au (red dashed line), and
AuCl k ꢀ 133.5 nm was used).
(blue long dashed short dashed line) (30.5 ꢀ
2 3
O
3
D
Table 1
Summary of curve-fitting analysis for the Au–(b-ala).
Na
.7 ± 0.7
rb/nm
dE /eV
c
d
r
d/nm
À0.001 ± 0.003
Fig. 5. Top-view (a) and side-view (b) of the optimized geometry of the Au–(b-ala).
3
0.204 ± 0.001
À0.13 ± 3.0
a
b
c
Coordination number.
Coordination distance.
Energy difference in the absorption threshold between the reference com-
, and Au–(b-ala).
Difference of Debye-Waller factor between Au
2 3
pound, Au O
d
2 3
O and Au–(b-ala).
Au atom, was observed in the range from 0.15 to 0.23 nm. These
results indicate that the coordination of the Au–(b-ala) can be
assigned to AuAO but not to AuACl. Here, we successfully prepared
a chloride-free precursor. Table 1 shows structure parameters for
the Au–(b-ala) obtained by curve-fitting analyses of its Fourier fil-
tered function. The peak discernible at 0.13–0.20 nm in the Fourier
transform function shown in the Fig. 4b was attributed to the scat-
tering of photoelectrons ejected from the O atoms located at the
nearest distances from the absorber Au atom. This scattering
occurred because the frequency and the relative amplitude of the
Fig. 6. Molecular structure of the Au–(b-ala).
XAFS spectrum of the Au O (coordination number, N = 4 and coor-
2 3
dination distance, r = 0.201 nm) [31]. Their Debye-Waller factor
was assumed to be 0.006 nm. The curve-fittings converged to give
R–factor of 0.26%. The coordination number of 3.7 and the coordi-
nation distance of 0.204 nm for the Au–(b-ala) were consistent
with those for Au O within the estimated range of errors, respec-
3
k -weighted oscillation for the Au–(b-ala) were similar to those
2
3
for the Au
2
O
3
previously described. The peak was subsequently
tively. In addition, the differences of the threshold energy and the
Debye-Waller factor were sufficiently small. The local structure
around the Au atoms in Au–(b-ala) was very similar to that around
the Au atoms in Au O . Because the molar ratio of b-ala ligand/Au
back-transformed into k space and fitted by considering the contri-
bution from the AuAO shells to determine the structural parame-
ters for the Au–(b-ala). The backscattering amplitude and phase
shift for the AuAO bond were obtained experimentally from the
2
3
was 1.2, and b-ala can be a bidentate ligand, the amino group and

H. Murayama et al. / Journal of Catalysis 353 (2017) 74–80
79
Table 3
2 2
Hydrogenation of 1-nitro-4-vinylbenzene by the Au/SiO and Au/TiO .
Catalyst
Conv./%
Yield/%
A
Precursor
B
C
Au/SiO
2
Au–(b-ala)
HAuCl
Au–(b-ala)
66
0
100
59
0
0
2
0
70
2
0
2
4
Au/TiO
2
carboxyl group of a b-ala can be assumed to be bound to Au3+
.
Because the scattering cross-section of an N atom is similar to that
of an O atom, the structure of the Au–(b-ala) given by the fitting
results shows good agreement with the structure of Au O .
2 3
Using DFT calculations, the geometry of the Au–(b-ala) were
optimized. The obtained structure is shown in Table 2 and Fig. 5.
These results indicate that the geometry around the Au ions in
this complex is square planar and the deformation from the proper
square planar one is small. Therefore, this result indicates that the
valence of Au ions in this complex is trivalent because the square-
planar structure is typical of the Au3 complex.
+
We propose a molecular structure of the Au–(b-ala), as shown
in Fig. 6. The mononuclear complex of Au3 has a square-planar
coordination structure. The amino group and carboxyl group of a
+
3+
b-ala molecule, which is bidentate, are bound to Au . The other
two coordination of Au3 are bound to hydroxyl groups,
respectively.
2
Fig. 7. Time dependence of the adsorption amounts of DMTS on Au/SiO .
+
for the unpalatable aroma of drinks such as Japanese sake [34,35],
beer [36], whiskey [37], and wine [38]. Silica is allowed to use for
Japanese Sake brewing by the National Tax Administration Agency
of Japan. Fig. 7 shows time dependence of the adsorption amount
of DMTS on the Au/SiO2 prepared by using the Au–(b-ala), on the
Au/SiO2 prepared by using HAuCl Á4H O, and on commercial Au/
3
.4. Application for hydrogenation reaction and for adsorbents
The catalytic performances, selectivity, of the Au/SiO
TiO for reduction of 1-nitro-4-vinylbenzene were investigated.
As shown in table 3, the Au/SiO prepared by the Au–(b-ala)
exhibits high selectivity for hydrogenation of vinyl group of
-nitro-4-vinylbenzene into 1-ethyl-4-nitrobenzene (A). On the
other hand, nitro group of 1-nitro-4-vinylbenzene was hydro-
genated into 4-vinylaniline (B) by the Au/TiO . Probably, polymer-
2
and Au/
2
2
4
2
SiO2 (Haruta Gold, Inc.) prepared by the deposition-reduction
method using Au(en) Cl as a precursor. The rate and quantity of
1
2
3
adsorbed DMTS on the Au/SiO2 prepared by using the Au–(b-ala)
2
were markedly higher than those on the Au/SiO prepared by using
2
ization of 4-vinylaniline can be occurred as a side reaction. The
over-reduced product, 4-ethylaniline (C) was formed at negligible
levels. Boronat et al. have reported that favorable adsorption of
HAuCl Á4H O.
4
2
Compared with the adsorbed amount on the commercial Au/
SiO , whose Au NP size was 7 nm, smaller Au NPs on SiO adsorbed
2
2
nitro group on TiO
nitro-3-vinylbenzene into 3-vinylaniline when Au/TiO
[
2
afforded nitro-selective hydrogenation of 1-
was used
32]. Meanwhile, poor interactions between the substrate and
a larger amount of DMTS, thus indicating that the DMTS molecules
2
adsorbed onto the surfaces of the Au NPs. The Au/SiO prepared by
2
using the Au–(b-ala) showed high performance for removal of
SiO
2
resulted in a low regioselectivity and yielded over-reduced
. The result that
selectively hydrogenates vinyl group is consistent with
the selectivity of Au NPs on silica previously reported. Turnover
frequency of Au/SiO for the hydrogenation of styrene has been
unpalatable aroma, although bulk Au particles, such as Au/SiO2
3
-ethylaniline as major product using Au/SiO
2
prepared by using HAuCl Á4H O, scarcely adsorbed any DMTS
4
2
Au/SiO
2
molecules. For further detailed discussion of these phenomena,
other kinds of studies, such as investigation for adsorption mecha-
nisms and spectroscopic measurements for adsorbates, are neces-
sary, which is a future research topic.
2
larger than that of nitrobenzene [32]. Au NPs on silica have been
highly selective for the hydrogenation of C@C bond in cinnamalde-
hyde to form hydrocinnamaldehyde [33]. We expect the nitro
group on the para-position withdraws electrons more efficiently
than meta-position which brought the higher reactivity of vinyl
group of 1-nitro-4-vinylbenzene compared to 1-nitro-3-
vinylbenzene. In addition, size-effect of Au NPs might be obtained
4. Conclusion
The Au NPs were highly dispersed on various supports,
especially on silica, through use of a novel incipient wetness IP
method. The average diameter of the Au NPs was determined to
be less than 3 nm on the basis of TEM observations. The prepara-
tion of smaller Au NPs on silica supports was achieved by using
2 4
because the Au/SiO prepared by HAuCl , having an average diam-
eter about 30 nm, showed no catalytic activity.
2
One of the possible practical application of the Au/SiO obtained
by our method is described. We subsequently applied the Au/SiO
to an adsorption test using DMTS in ethanol, which is responsible
2
Au–(b-ala) complexes, whose structure is similar to that of Au
(Fig. 6), as chloride-free precursors. The decomposition and
2 3
O

8
0
H. Murayama et al. / Journal of Catalysis 353 (2017) 74–80
reduction temperatures of the Au–(b-ala), which were investigated
by in situ XAFS and TG-DTA, were lower than those of HAuCl
O. These behaviors prevented Au atoms from aggregating. We
expect that Au complexes coordinated with another kind of amino
acid are also precursors for preparing supported small Au NPs. The
[8] T. Hayashi, K. Tanaka, M. Haruta, J. Catal. 178 (1998) 566–575.
9] S. Biella, L. Prati, M. Rossi, J. Catal. 206 (2002) 242–247.
10] T. Ishida, N. Kinoshita, H. Okatsu, T. Akita, T. Takei, M. Haruta, Angew. Chem.
Int. Ed. 47 (2008) 9265–9268.
[11] M. Okumura, S. Tsubota, M. Iwamoto, M. Haruta, Chem. Lett. (1998) 315–316.
12] M. Haruta, S. Tsubota, T. Kobayashi, H. Kageyama, M.J. Genet, B. Delmon, J.
Catal. 144 (1993) 175–192.
13] K. Tada, K. Sakata, Y. Kitagawa, T. Kawakami, S. Yamanaka, M. Okumura, Chem.
Phys. Lett. 579 (2013) 94–99.
14] K. Tada, K. Sakata, S. Yamada, K. Okazaki, Y. Kitagawa, T. Kawakami, S.
Yamanaka, M. Okumura, Mol. Phys. 112 (2014) 365–378.
15] H.-Y. Tsai, Y.-D. Lin, W.-T. Fu, S.D. Lin, Gold Bull. 40 (2007) 184–191.
16] Q. Xu, K.C.C. Kharas, A.K. Datye, Catal. Lett. 85 (2003) 229–235.
17] S. Ivanova, C. Petit, V. Pitchon, Appl. Catal. A 267 (2004) 191–201.
18] S. Ivanova, V. Pitchon, Y. Zimmermann, C. Petit, Appl. Catal. A 298 (2006) 57–
[
4 2
Á4H -
[
[
[
[
2
Au/SiO prepared by our method showed high activity and selec-
tivity for hydrogenation of 1-nitro-4-vinylbenzene into 1-ethyl-4-
nitrobenzene and good performance for removal of unpalatable
aroma.
[
[
[
[
Acknowledgement
6
4.
The HAADF-STEM observations were performed at the
Ultramicroscopy Research Center, Kyushu University. The XAFS
experiments were performed at the BL14B2 beamline of SPring-8
with the approval of the Japan Synchrotron Radiation Institute
[19] W.-C. Li, M. Comotti, F. Schüth, J. Catal. 237 (2006) 190–196.
20] L. Delannoy, N.E. Hassan, A. Musi, N.N.L. To, J.-M. Krafft, C. Louis, J. Phys. Chem.
B 110 (2006) 22471–22478.
21] J.M. Fisher, Gold Bull. 36 (2003) 155.
[22] G. Steinhauser, J. Evers, S. Jakob, T.M. Klapötke, G. Oehlinger, Gold Bull. 41
(2008) 305–317.
23] R. Renneke, S. McIntosh, V. Arunajatesan, M. Cruz, B. Chen, T. Tacke, H.
Lansink-Rotgerink, A. Geisselmann, R. Mayer, R. Hausmann, P. Schinke, U.
Rodemerck, M. Stoyanova, Top. Catal. 38 (2006) 279–287.
[
[
(
JASRI), proposal Nos. 2015A1702 and 2015B1781. Parts of the
XAFS experiments were performed at Kyushu University Beamline
SAGA-LS/BL06), proposal Nos. 2015IK007, 2015IIIK008, and
[
(
2
016IK004.
[24] H. Sakurai, K. Koga, Y. Iizuka, M. Kikuchi, Appl. Catal. A 462–463 (2013) 236–
46.
2
The authors gratefully acknowledge valuable support from UBE
[
25] S.J. Rai, P.C. Gupta, O.C. Saxena, Microchem. J. 17 (1972) 661–664.
industries, Ltd.
[
26] A.S. Kazachenko, E.V. Legler, O.V. Per’yanova, Y.A. Vstavskaya, Pharm. Chem. J.
3
3 (1999) 470–472.
27] M. Aslam, L. Fu, M. Su, K. Vijayamohanman, V.P. Dravid, J. Mater. Chem. 14
2004) 1795–1797.
[
[
[
[
[
[
Appendix A. Supplementary material
(
28] T. Maruyama, Y. Fujimoto, T. Maekawa, J. Colloid Interf. Sci. 447 (2015) 254–
HAADF–STEM images of the Au NPs supported on TiO
MCM-41, and Kethen black, details of estimation of the molar ratio
of b-ala ligands/Au in Au–(b-ala) complex on the basis of the
TG-DTA data, and Raman spectra for the Au/SiO
2
,
257.
29] G. Meitzner, G.H. Via, F.W. Lytle, J.H. Sinfelt, J. Phys. Chem. 96 (1992) 4960–
4964.
30] A. Pantelouris, G. Küper, J. Hormes, C. Feldmann, M. Jansen, J. Am. Chem. Soc.
117 (1995) 11749–11753.
31] P.G. Jones, H. Rumpel, E. Schwarzmann, G.M. Sheldrick, Acta Crystallogr. B 35
2
prepared by using
the Au–(b-ala) and HAuCl can be found in the online version.
4
(
1979) 1435–1437.
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.jcat.2017.07.002.
32] M. Boronat, P. Concepción, A. Corma, S. González, F. Illas, P. Serna, J. Am. Chem.
Soc. 129 (2007) 16230–16237.
[
[
33] H. Shi, N. Xu, D. Zhao, B.-Q. Xu, Catal. Commun. 9 (2008) 1949–1954.
34] A. Isogai, H. Utsunomiya, R. Kanda, H. Iwata, J. Agric. Food Chem. 53 (2005)
References
4118–4123.
[
35] A. Isogai, H. Utsunomiya, R. Kanda, H. Iwata, S. Nakano, J. Soc. Brew. Jpn. 101
(2006) 125–131 [In Japanese].
[
[
[
1] M. Haruta, T. Kobayashi, H. Sano, N. Yamada, Chem. Lett. (1987) 405–408.
2] M. Valden, X. Lai, D.W. Goodman, Science 281 (1998) 1647–1650.
3] S.-I. Lo, P.-C. Chen, C.-C. Huang, H.-T. Chang, Environ. Sci. Technol. 46 (2012)
[36] R.S. Williams, D.E.F. Gracey, J. Am. Soc. Brew. Chem. 40 (1982) 68–70.
[37] R.D.M. Prentice, G. McKernan, J.H. Bryce, J. Am. Soc. Brew. Chem. 56 (1998) 99–
103.
2
724–2730.
[
[
[
[
4] K.G. Thomas, P.V. Kamat, Acc. Chem. Res. 36 (2003) 888–898.
5] G.C. Bond, D.T. Thompson, Catal. Rev. Sci. Eng 41 (1999) 319–388.
6] M. Haruta, M. Daté, Appl. Catal. A 222 (2001) 427–437.
[38] N. Nishibori, A. Kuroda, O. Yamda, N. Goto-Yamamoto, Food Sci. Technol. Res.
23 (2017) 241–248.
7] R. Zanella, S. Giorgio, C.R. Henry, C. Louis, J. Phys. Chem. B 106 (2002) 7634–
7
642.