Gold nanoparticles supported on magnesium oxide as catalysts for the aerobic oxidation of alcohols under alkali-free conditions

Article abstract of DOI:10.1016/j.jcat.2012.05.009

Gold nanoparticles supported on magnesium oxide were shown to be efficient heterogeneous catalysts for the liquid-phase oxidation of a wide range of alcohols using molecular oxygen as a sole oxidant in the absence of co-catalysts or additives. The Au/MgO material was prepared through the deposition- precipitation method and characterized by XRD, XPS, XAS, HRTEM, UV-Vis spectroscopy, and N₂ adsorption techniques. The formation of gold nanoparticles on the MgO surface under temperature programmed reduction was monitored by in situ UV-Vis-Mass spectroscopy. Various carbonylic monoterpenoids important for fragrance and pharmaceutical industries were obtained in good to excellent yields starting from biomass-based monoterpenic alcohols, such as isoborneol, perillyl alcohol, and carveol. The Au/MgO catalyst also performs a selective one-pot oxidative esterification of benzyl alcohol in alkali-free methanol solutions in the absence of any additive to give methyl benzoate in a virtually quantitative yield.

Full text of DOI:10.1016/j.jcat.2012.05.009

Journal of Catalysis 292 (2012) 148–156
Contents lists available at SciVerse ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
Gold nanoparticles supported on magnesium oxide as catalysts for the aerobic
oxidation of alcohols under alkali-free conditions
a
b
c
c
c
Vinícius V. Costa , Miguel Estrada , Yulia Demidova , Igor Prosvirin , Vladimir Kriventsov ,
a
d
d
a,
⇑
Rafaela F. Cotta , Sergio Fuentes , Andrey Simakov , Elena V. Gusevskaya
a
Departamento de Química, Universidade Federal de Minas Gerais, 31270-901 Belo Horizonte, MG, Brazil
Posgrado en Física de Materiales, Centro de Investigación Científico y de Educación Superior de Ensenada, C.P. 22860 Ensenada, B.C., Mexico
Boreskov Institute of Catalysis, Lavrentieva 5, Novosibirsk 630090, Russia
b
c
d
Universidad Nacional Autónoma de México, Centro de Nanociencias y Nanotecnología, Km. 107 Car. Tijuana a Ensenada, C.P. 22860 Ensenada, B.C., Mexico
a r t i c l e i n f o
a b s t r a c t
Article history:
Gold nanoparticles supported on magnesium oxide were shown to be efficient heterogeneous catalysts
for the liquid-phase oxidation of a wide range of alcohols using molecular oxygen as a sole oxidant in
the absence of co-catalysts or additives. The Au/MgO material was prepared through the deposition–
Received 27 March 2012
Revised 8 May 2012
Accepted 9 May 2012
Available online 14 June 2012
precipitation method and characterized by XRD, XPS, XAS, HRTEM, UV–Vis spectroscopy, and N
2
adsorption techniques. The formation of gold nanoparticles on the MgO surface under temperature
programmed reduction was monitored by in situ UV–Vis–Mass spectroscopy. Various carbonylic
monoterpenoids important for fragrance and pharmaceutical industries were obtained in good to excel-
lent yields starting from biomass-based monoterpenic alcohols, such as isoborneol, perillyl alcohol, and
carveol. The Au/MgO catalyst also performs a selective one-pot oxidative esterification of benzyl alcohol
in alkali-free methanol solutions in the absence of any additive to give methyl benzoate in a virtually
quantitative yield.
Keywords:
Gold catalysts
Selective oxidation
Oxygen
Benzyl alcohol
Monoterpenic alcohols
Ó 2012 Elsevier Inc. All rights reserved.
1
. Introduction
The oxidation of alcohols is one of the most challenging
in stoichiometric amounts due to the formation carboxylic acid
salts is performing the reaction in methanol solutions [24–30]. In
these systems, carboxylic acids are trapped by methanol, which
itself is rather inert to aerobic oxidation, so that methyl esters
are formed as final products.
reactions in synthetic organic chemistry. The resulting carbonyl
compounds are widely employed in various fields of chemical
industry; however, many commonly used processes are still based
on stoichiometric oxidation reactions which produce large
amounts of wastes [1]. For these reasons, metal catalyzed oxida-
tions of alcohols, especially those that involve molecular oxygen
as a final oxidant, currently attract much attention. Various
homogeneous [2,3] and heterogeneous [4–9] transition metal cat-
alysts have been reported for the aerobic oxidation of alcohols,
with gold nanoparticles (NPs) being among the most promising
ones [10–23].
The oxidation of alcohols over gold catalysts often requires the
presence of base to promote the abstraction of hydrogen from the
substrate and to capture free carboxylic acid formed at the reac-
tion, which could poison the catalyst [12]. Nevertheless, several
gold-containing catalysts have been developed for the aerobic oxi-
dation of alcohols under base-free conditions [14–23]. One of the
recently suggested approaches to prevent the consumption of base
Methyl esters find a commercial use as solvents, diluents,
extractants, and flavoring agents. Their syntheses commonly
consist in two-step procedures that include the esterification of
carboxylic acids or their derivatives obtained in a separate step.
Only a few catalysts capable to promote direct transformations of
alcohols into methyl esters using molecular oxygen as a final
oxidant have been reported, with all of them being based on gold
NPs, as far as we know [24–30]. In a previous work, we found that
gold NPs supported on Ce- or Ti-modified hexagonal mesoporous
silica (HMS) efficiently performed the one-pot aerobic oxidative
esterification of benzyl alcohol [30]. However, the systems were
active only in the presence of the base co-catalyst (potassium
carbonate), likewise most of other gold systems reported for the
oxidative esterification of alcohols. One of a few examples of the
synthesis of carboxylic esters from alcohols under neutral condi-
tions involves the use of gold NPs supported on nanocrystalline
2 3 2 3
b-Ga O [28]. It has been suggested that the Au/b-Ga O acts as a
bifunctional catalyst due to the increased surface Lewis acidity of
the support.
⇑
Corresponding author. Fax: +55 31 34095700.
E-mail address: elena@ufmg.br (E.V. Gusevskaya).
0
021-9517/$ - see front matter Ó 2012 Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.jcat.2012.05.009

V.V. Costa et al. / Journal of Catalysis 292 (2012) 148–156
149
In the present work, we have prepared the material containing
gold NPs supported on the magnesium oxide and studied its
activity in the aerobic oxidation of various alcohols. The formation
of NPs on the MgO surface has been monitored by in situ
UV–Vis–Mass spectroscopy as it is well established that catalytic
properties of gold NPs are strongly related to the appropriate
preparation procedure [13].
The obtained Au/MgO sample showed high catalytic activity in
the aerobic oxidative esterification of benzyl alcohol in the absence
of base or any other additive. The oxidation of alcohols under base-
free conditions catalyzed by gold NPs supported on magnesium
oxide has been described by several groups [14,17–21]. However,
to the best of our knowledge, no data on the applications of these
materials for the oxidative methoxylation of alcohols have been
previously reported.
of gold particles, more than 200 particles were chosen. The mean
diameter (d
m
) of particles was calculated using the following for-
P
ðx
i
d
i
Þ
P
i
mula: d
m
¼
, where x is the number of particles with diam-
i
ix
i
eter d
i
.
In situ UV–Vis–Mass analysis of the transformation of gold spe-
cies under temperature programmed reduction (TPR) was carried
out in a lab-made set-up [39] with simultaneous analysis of gas
phase components and recording of the UV–Vis spectra of the
sample. UV–Vis spectra were collected using an AVANTES Ava-
Spec-2048 UV–Visible spectrometer equipped with an AvaLight-
DHS light source and a high temperature optic fiber reflection
probe located close to the external wall of the quartz reactor
(i.d. 4 mm, o.d. 6 mm). UV–Vis spectra recorded each 15 s (time
of spectrum recording was about 5 ms) were obtained by the sub-
traction of the initial spectrum recorded at room temperature
from those recorded at elevated temperatures. The reactor packed
The catalyst was also tested in the aerobic oxidation of natural
monoterpenic alcohols. Terpenic compounds, in general, are bio-
mass-based renewable substrates widely used in the flavor and
fragrance industry, which is essentially based on the chemistry of
terpenes. For several years, we have been interested in catalytic
oxidations of natural products, including terpenes, over ruthenium
2 3
with Al O was used as the reference. The analysis of gas phase
components was done in-line with an HPR20 mass spectrometer
(Hiden). TPR was carried out under linear sample heating with a
ꢁ1
ramp rate of 20 °C min up to 350 °C using a gas mixture con-
sisting of H (5 vol.%), Ar (5 vol.%), and He (balance). The relative
[
9], cobalt [31,32], palladium [33–35], and chromium [36] cata-
2
lysts. In continuation of our ongoing project aimed at adding value
to natural ingredients of essential oils, we report herein the addi-
tive-free aerobic oxidation of various monoterpenic alcohols cata-
lyzed by Au/MgO to give valuable fragrance aldehydes and ketones.
content of the desorbed products was estimated considering all
possible mass-fragments according to the Hiden mass-library,
while hydrogen uptake was calculated using a reference gas
mixture.
The extended X-ray absorption fine structure (EXAFS) and X-
ray absorption near edge structure (XANES) spectra (transition
and fluorescent modes) of the Au–L edges for all the studied
3
2
. Experimental
samples were obtained at the EXAFS Station of the Siberian Syn-
chrotron Radiation Center in Novosibirsk (SSRC). The storage ring
VEPP-3 with the electron beam energy of 2 GeV and the average
stored current of 90 mA was used as the source of radiation. The
X-ray energy was monitored with a channel cut Si(111) mono-
chromator. The harmonic rejection was performed by using a
SiO mirror for all measurements. All EXAFS and XANES spectra
2
were recorded with the steps of ca.1.5 eV and ca. 0.3 eV,
respectively.
2
.1. Catalyst preparation
Gold (3 wt.%) was supported by deposition–precipitation
using urea as a precipitation agent as in [37]. MgO (4.0 g, Mal-
linckrodt) was added to an aqueous solution (400 mL) of HAuCl
4
ꢁ3
(
1.6 ꢀ 10 M) and urea (0.42 M). The initial pH of the solution
was ca. 2. The suspension was vigorously stirred at 80 °C for
h. Then, the solid material was filtered and washed with
4
ammonium hydroxide (25.0 M) for 30 min. The last procedure
developed in [38] is quite effective to stabilize small gold NPs.
After stirring with ammonium hydroxide, the pH of the solution
was ca. 10. Finally, the sample was washed with water until the
pH of the solution reached the value of 7, then filtered, and
dried at room temperature for 24 h. This sample, denoted as
the freshly prepared Au/MgO catalyst, was then reduced in the
The EXAFS spectra were treated using the standard procedures
40,41]. The background was removed by extrapolating the pre-
edge region onto the EXAFS region in the forms of polynomials.
Three cubic splines were used to construct the smooth part of
the absorption coefficient. The inflection point of the edge of the
X-ray absorption spectrum was used as initial point (k = 0) of the
[
EXAFS spectrum. The radial distribution of the atoms function
ꢁ1
flow of H
2
(5 vol.%), Ar (5 vol.%), and He (balance) (50 mL min
)
3
(
RDF) was calculated from the EXAFS spectra in k
v
(k) by using
ꢁ1
at a heating rate of 20 °C min up to 350 °C, and the resulting
material, denoted as the reduced Au/MgO catalyst, was used in
catalytic tests.
ꢁ1
Fourier transform in the wave number interval of 3.0–12.0 Å . A
curve fitting procedure with the EXCURV92 code [42] was em-
ployed to determine the distances and coordination numbers
3
(
CNs). It was realized for k
v(k) in similar wave number intervals
2.2. Catalyst characterization
after preliminary Fourier filtering using the known XRD literature
data for the bulk compounds. The Debye–Waller factors were
2
2
Gold content (2.6 wt.%) was determined by inductively coupled
fixed: 2
Photoelectron spectra were recorded using a SPECS spectrome-
ter with a PHOIBOS-150 hemispherical energy analyzer and AlK
irradiation (h = 1486.61 eV, 200 W). A binding energy (BE) scale
r = 0.009–0.013 Å .
plasma atomic emission spectroscopy (ICP-AES) on a Varian
Liberty 110 instrument. Solid samples pretreated at 350 °C in Ar
flow for 30 min were first digested at room temperature in HF
a
m
for 12 h and then in a mixture of HCl and HNO
3
for 30 min. The
was preliminarily calibrated by the position of the peaks of Au
4f7/2 (84.0 eV) and Cu 2p3/2 (932.67 eV) core levels. The catalyst
was mounted on a sample holder using a nickel grid to be able
to heat a sample up to 500 °C. The BE of peaks was corrected to
take into account a sample charging by referencing to the BE of
the Mg 2s peak (88.9 eV) (internal standard). In addition to the sur-
vey, photoelectron spectra, more narrow spectral regions of Au
4f + Mg 2s, were recorded. For the survey spectra, the pass energy
of the analyzer was 50 eV, while that for the narrow spectral
regions was 20 eV.
solution obtained was diluted with deionized water and analyzed.
The specific surface area was determined using the BET method
by nitrogen thermal adsorption measurements in a Gemini 2600
Micromeritics device. Before analysis, the sample was heated in
Ar flow at 300 °C for 1 h.
High-resolution transmission electron microscopy (HRTEM)
was performed with a JEOL 2010 microscope. The sample was
dispersed by ultrasonic in isopropanol and supported on a copper
grid covered with carbon film. To determine the mean diameter

1
50
V.V. Costa et al. / Journal of Catalysis 292 (2012) 148–156
2.3. Catalytic oxidation experiments
the agglomerates of the precursor contain not only gold cations
but also anionic species, they should be larger than the resulting
particles of the metallic gold. Thus, we suggest that the size of
the gold NPs formed on the surface of the freshly prepared Au/
MgO catalysts during the TEM measurements is determined by
the size of the precursor agglomerates.
TRP profiles and in situ UV–Vis–Mass spectra for the freshly pre-
pared Au/MgO sample at TPR are presented in Fig. 3. Within the
temperature range studied (up to 350 °C), a small hydrogen uptake
The reactions were carried out in a stainless steel reactor
equipped with a magnetic stirrer. In a typical run, a mixture of
the substrate (0.4–1.2 mmol), solvent (if any) (2 mL), and the
catalyst (10–40 mg; ca. 0.6–2.4 wt.%; Au: 0.01–0.75 mol%) was
transferred in the reactor. The reactor was pressurized with oxygen
to the total pressure of 10 atm and placed in an oil bath; then, the
solution was intensively stirred at 80–130 °C for the reported time.
The reactions were followed by gas chromatography (GC)
2 2 3
as well as the desorption of H O, CO and NH was observed. In the
(
Shimadzu 17 instrument, Carbowax 20 M capillary column). At
UV–Vis spectra of the sample, the plasmon peak of metallic gold
NPs centered in the range of 500–560 nm appeared under the
TPR conditions, with the intensity and position of this peak being
strongly depended on the temperature. The application of the
in situ UV–Vis spectroscopy allowed monitoring the process of
the formation of gold NPs in the course of TPR. The intensity of
the plasmon peak increased significantly with the temperature rise
indicating the increase in the amounts of gold NPs. The changes in
the relative plasmon intensity, water desorption, and hydrogen up-
take during the TPR are shown in Fig. 4. The profile of the plasmon
intensity changes suggests that there are three distinguishable
steps in the process of gold NPs formation: the first slow step at
appropriate time intervals, stirring was stopped and after catalyst
settling aliquots were taken and analyzed by GC. To ensure correct
GC results in solvent-free reactions and in those with high initial
substrate concentrations, the aliquots were diluted with methanol
before the analysis. The structures of the products were confirmed
by GC/MS (Shimadzu QP2010-PLUS instrument, 70 eV).
3
. Results and discussion
3
.1. Activation and characterization of the catalyst
5
0–180 °C, the second fast step at 180–225 °C, and the third slow
The UV–Vis spectrum of the freshly prepared Au/MgO catalyst
step at 225–350 °C.
is depicted in Fig. 1. The spectrum is characterized with the
adsorption band at 350 nm (shown with arrow) which can be
assigned to gold hydroxide species [43]. It is well established that
the presence of metallic gold NPs could be easily detected with the
band at ca. 520 nm, which is characteristic for the optical absorp-
tion of light excited oscillating conductivity electrons of metallic
gold NPs: the so-called ‘‘plasmon resonance’’ [44]. The absence of
this band in the spectrum shown in Fig. 1 provides a strong
evidence for the absence of metal gold NPs in the freshly prepared
Au/MgO sample.
However, metallic gold NPs of a hexagonal shape with a rela-
tively small diameter (shown with white arrow in Fig. 2) were ob-
served in TEM micrographs of the freshly prepared Au/MgO
sample. These particles could be formed due to the decomposition
of the gold precursor in a vacuum chamber of the electron micro-
scope as it was observed in [13]. The size distribution of gold NPs in
the freshly prepared Au/MgO sample is shown in Fig. 2. The esti-
mated value of 2.3 nm seems to reflect the average size of the gold
precursor agglomerates formed on the sample surface during the
preparation procedures. No significant migration of the gold parti-
cles along the support surface is expected in the course of the
reduction in the microscope due to relatively mild conditions. As
The first step proceeds with intense water evolving, albeit with-
out any hydrogen consumption. Therefore, the formation of gold
NPs at this step is determined by the thermal decomposition of
3 2 3
the gold precursor (Au(OH) ? Au O ? Au) only. At this step,
the formation of gold NPS is accompanied with a significant shift
of the plasmon resonance (Fig. 4) which indicates the changes in
the particle size. The peak position is shifted toward the red region
of the spectrum which could be explained by the suggestion that
growing gold NPs are coated with a layer of compounds character-
ized with a high value of dielectric function. Such an explanation is
based on the theoretical estimations presented in [45]. Therefore, it
can be suggested that gold NPs are formed in this temperature re-
gion mainly inside the primary particles of gold hydroxide, which
have been supported on the MgO surface during the sample
preparation.
The second step of the process, i.e., the fast formation of gold
NPs within the narrow temperature interval of 180–225 °C, is
accompanied by the hydrogen consumption (Fig. 4). However,
the hydrogen uptake corresponds to only 0.05% of the value re-
quired for the complete reduction of the gold precursor. Therefore,
the formation of gold NPs at this step mainly occurs also due to the
thermal decomposition of the dehydrated gold deposit. Similarly to
the previous step, significant changes in the plasmon position were
observed suggesting that the formation of gold particles occurred
with their agglomeration. The red shift of the plasmon position
indicates that gold NPs formed at this step are still covered by a
3
Au(OH) layer.
In the course of the third step, the formation of gold NPs is being
concluded. The low rate of this step is determined by the decompo-
sition of the residual part of the gold precursor. The reduction with
hydrogen of the external surface of the gold deposit particles at
2
25 °C provokes the desorption of ammonia. The latter indicates
that at this point, the Au(OH) film, which covers the gold nanopar-
3
0
ticles, is completely reduced to Au causing the desorption of
ammonia because of its lower affinity toward the metallic gold
compared to gold hydroxide. A further temperature increase up
to 350 °C does not affect the position of plasmon indicating stabil-
ity of the gold NPs toward sintering under these conditions.
TEM images of the reduced Au/MgO sample are presented in
Fig. 5. The gold NPs formed under the sample reduction described
above are characterized with a quasi spherical shape. The compar-
ison of the TEM images for the freshly prepared and reduced Au/
Fig. 1. UV–Visible spectrum of the freshly prepared Au/MgO sample.

V.V. Costa et al. / Journal of Catalysis 292 (2012) 148–156
151
Fig. 2. TEM data for the freshly prepared Au/MgO sample: the TEM micrograph (on the left), the histogram of the gold NPs distribution in size (on the right).
Fig. 3. In situ-UV–Vis–Mass analysis of the freshly prepared Au/MgO sample at TPR (on the left: UV–Vis spectra; on the right: TPR profiles).
Fig. 4. TPR analysis of the freshly prepared Au/MgO. Hydrogen uptake, water desorption, and relative plasmon intensity vs. temperature (on the left). Relative plasmon
intensity and plasmon position vs. temperature (on the right).
MgO samples showed that the sample treatment at high tempera-
ture (up to 350 °C) resulted in the agglomeration of the particles.
Their average diameter was found to be equal to 3.4 nm (Fig. 5).
The surface analysis of the reduced Au/MgO sample by X-ray
photoelectron spectroscopy (XPS) was used to clarify the chemical
state of the gold species formed on the catalyst surface. The XP

1
52
V.V. Costa et al. / Journal of Catalysis 292 (2012) 148–156
Fig. 5. TEM data for the reduced Au/MgO sample: the TEM micrograph (on the left), the histogram of the gold NPs distribution in size (on the right).
Fig. 6. The experimental XP spectrum (dots) and the results of fitting of the Mg 2s
dashed line) and Au 4f region (solid lines) for the reduced Au/MgO catalyst.
(
spectrum for Mg 2s and Au 4f regions is presented in Fig. 6. The
spectrum for Au 4f is characterized only by one doublet of two
spin–orbit components separated by 3.67 eV, i.e., Au 4f7/2 and Au
3
Fig. 7. The XANES spectra (Au–L ) of the samples studied: (a) the freshly prepared
Au/MgO; (b) the reduced Au/MgO; (c) the reference Au foil; (d) the reference oxide,
ꢃ
2
Au O
3
[34]. Au–L
3
edge, – white line.
4
f
5/2, which can be attributed to metallic gold species according
to [46].
The evolution of the chemical and coordination state of the gold
species at the catalyst preparation was verified by X-ray absorption
spectroscopy (XAS). The normalized XANES spectra for the freshly
prepared and reduced Au/MgO samples, the reference Au-foil and
2.87 Å, CN = 12) [49]. Fitting performed as described in Experimen-
tal gives very close values for RAu–Au (ca. 2.86 Å) and CN (ca. 12.0).
The main peak in the RDF for the freshly prepared Au/MgO sam-
ple, which is located at ca. 1.2–2.2 Å, can be attributed to the Au–O
distance of ca. 2.0 Å (Fig. 8a) [50,51]. Fitting gives the following set
Au
freshly prepared Au/MgO sample (Fig. 7a) is similar to that of
Au (Fig. 7d), with only minor differences being observed. The
2 3
O from [47], are shown in Fig. 7. The XANES spectrum of the
3+
of values: RAu–O ꢂ 1.99 Å and CN ꢂ 5.9–6.0. Thus, only Au cations
present in this sample in nearly octahedral oxygen surrounding.
This Au hydroxide–oxide structure is different from the bulk
2 3
O
high amplitude of the white line clearly indicates the presence of
the oxidized phase [48]; thus, the main part of the gold species
3
+
2 3
Au O oxide where Au cations are in four-coordinated positions
in both freshly prepared Au/MgO and Au
2
O
3
samples presents as
[50]. No metallic gold species were detected in the freshly pre-
pared Au/MgO sample within the method limit. The results ob-
tained are in a good agreement with the UV–Vis data for this
sample.
The shape of the RDF for the reduced Au/MgO sample (Fig. 8d) is
in general similar to that for the Au-foil with the fcc structure;
however, the amplitude of the first Au–Au peak located at ca.
1.7–3.5 Å is slightly lower. Fitting gives the values of ca. 2.84 Å
for RAu–Au and 8.2–8.5 for CN, which are fully consistent with the
correlation between the Au–Au coordination number and the
3
+
Au cations in oxygen surrounding. The XANES spectrum of the re-
duced Au/MgO sample (Fig. 7b) is similar to that of the reference
Au-foil (Fig. 7c). The amplitude of the white line is quite low in
both materials. Thus, gold mainly presents in the reduced Au/
0
MgO sample as the Au metallic phase.
The RDFs describing the Au local arrangements for the samples
are shown in Fig. 8. The RDF for the reference Au-foil (Fig. 8c) is
typical for the face-centered cubic (fcc) structure, with the main
peak corresponding to the shortest Au–Au distance (RAu–Au
=

V.V. Costa et al. / Journal of Catalysis 292 (2012) 148–156
153
The reactions occurred with very low catalyst loadings
0.6 wt.%) so that turnover numbers (TONs) reached high values
up to ca. 30.000) reflecting high stability of the catalyst. The cata-
(
(
lyst is a solid material that is insoluble in the reaction mixture; it
can be separated from the products by simple centrifugation or fil-
tration. These features represent the important technological
advantages of the process.
The oxidation of benzyl alcohol in the presence of Au/MgO read-
ily occurred at 100 °C resulting in a complete conversion in 3 h to
give almost exclusively methyl ester 1c (92%) along with aldehyde
1
b (8%, Table 1, run 1). Keeping the mixture under stirring up to 6 h
resulted in the almost complete conversion of aldehyde 1b to give
ester 1c, which was obtained at the end of the reaction in a virtu-
ally quantitative yield.
To increase the catalyst efficiency in terms of turnover numbers
(
TONs), the amount of the substrate was doubled (Table 1, runs 2
and 3). In these runs, it can be clearly seen the increase in ester
selectivity with the reaction time. In run 2, a virtually complete
conversion of the substrate was observed for 6 h; however, benzal-
dehyde 1b accounted for ca. 30% of the products. At longer reaction
time, most of the aldehyde was converted in the methyl ester,
which is consistent with a two-step mechanism involving the oxi-
dation of alcohol to aldehyde and further oxidation of aldehyde to
ester (see below). In run 3, ester 1c was obtained in a virtually
quantitative yield.
Fig. 8. The curves of the radial distribution function of atoms (RDF) describing the
The results of runs 2 and 3 (Table 1) correspond to the TON of
nearly 20.000 with respect to the surface amounts of gold; i.e.,
the fraction of the gold atoms that are located on the surface of
the gold particles and are, therefore, accessible for the substrate.
The surface amounts of Au (8.47% of the total amounts) can be cal-
culated based on the average particle diameter (3.4 nm) deter-
mined by TEM as the XAS study has confirmed that the reduced
Au/MgO material contains no appreciable amounts of the Au parti-
cles smaller than 1 nm.
gold local arrangement for the samples studied: (a) the freshly prepared Au/MgO;
(
b) the reduced Au/MgO; (c) the reference Au foil.
interatomic Au–Au bond length determined by van Bokhoven et al.
52]. The decrease in the coordination number compared to Au foil
[
may be explained by the distortions of the metal fcc structure and
nano-size effects [53]. No Au3 species were detected in the re-
duced Au/MgO sample within the method limit.
+
The XAS (XANES/EXAFS) results obtained for the reduced Au/
MgO sample are in a good agreement with the XPS and TEM data.
The size of the Au particle may be estimated from the previously
reported correlation between the coordination number and the
particle size [54]. The obtained value correlates well with the
TEM data (Fig. 5) implying that the reduced Au/MgO sample con-
tains no significant amount of small gold particles (less than
The performance of the Au/MgO catalyst in the oxidative ester-
ification of benzyl alcohol is comparable with that of the Au/HMS
catalysts reported in our previous work [30]. However, the Au/
HMS systems required a base co-catalyst (potassium carbonate):
in the absence of base, no conversion of the alcohol was observed.
The important advantage of the Au/MgO catalyst is no need in
base, the catalyst is active in the absence of any base co-catalyst
added. In run 4, which was performed under the same conditions
used in [30], however, without the base, the reaction showed the
1
nm) which cannot be observed by the TEM techniques. In other
words, the histogram presented in Fig. 5 represents well the size
distribution of most of the gold particles present on the surface
of the reduced Au/MgO sample.
ꢁ1
initial turnover frequency (TOF) of 155 min and TON of nearly
2.000. These results compare well with most of the reported cat-
3
alysts and put the prepared material among the best gold catalysts
developed for the oxidative methoxylation of benzyl alcohol
3.2. Catalytic studies
[
25,27–29]. The catalyst after run 3 in Table 1 was reused convert-
The catalytic activity of the reduced Au/MgO sample in the oxi-
ing another portion of the substrate without a significant loss of
activity and selectivity.
dation of benzyl alcohol (1a) was tested in methanol solutions un-
der the atmosphere of molecular oxygen. In all experiments, this
material was applied as a sole catalyst in the absence of any co-cat-
alysts or additives. The results are presented in Table 1.
It was found that Au/MgO effectively catalyzed the oxidative
esterification of benzyl alcohol showing high activity, whereas in
the presence of pure support or without any catalyst added, the
conversion of the substrate was negligible. The reaction resulted
in two major products: benzaldehyde 1b and methyl benzoate 1c
Although the reaction occurs smoothly with Au/MgO as the sole
catalyst, the addition of potassium carbonate (0.16 equiv.) in-
creases the rate of both reaction steps: the oxidation of benzyl
alcohol to benzaldehyde and further oxidation of the latter into
the ester (cf. run 6 and 5 in Table 1).
It should be emphasized that in the presence of Au/MgO, benzyl
alcohol can be oxidized by molecular oxygen without any solvent,
i.e., in a solvent-free system (Table 2, runs 1 and 2). However, high
selectivity for benzaldehyde has been obtained only at low sub-
strate conversions. Then, a significant difference in a mass balance
was observed due to the formation of oligomers, which were not
CG determinable.
(
9
Scheme 1). A combined selectivity for these two products was
0–100% in all the runs, with only small amounts of benzoic acid
and benzyl benzoate being detected in reaction solutions. Methyl
benzoate and benzyl benzoate are formally the products of the
esterification of benzoic acid with methanol and benzyl alcohol,
respectively. The relative amounts of the products 1b and 1c
depended on the reaction time, with the ester 1c being a predom-
inant product at the end of the reaction.
The mechanism proposed for the oxidative esterification of pri-
mary alcohols involves as the first step the oxidation of the alcohol
into the aldehyde followed by the condensation reaction between
the aldehyde and alcohol resulting in hemiacetal [26,28]. The lat-

1
54
V.V. Costa et al. / Journal of Catalysis 292 (2012) 148–156
Table 1
Oxidation of benzyl alcohol (1a) catalyzed by Au/MgO in methanol solutions.
a
Run
Substrate (mmol)
Temperature (°C)
Time (h)
Conversion (%)
Product selectivity (%)
TONb
TOFc (minꢁ1
)
Aldehyde 1b
Ester 1c
1
0.4
1.2
1.2
2.5
2.5
2.5
100
110
120
110
100
100
3
6
100
100
8
2
92
98
7.084
n.d.
n.d.
n.d.
155
68
2
6
98
100
31
12
64
83
1
0
19.415
21.595
32.494
23.843
38.751
3d
4
1
8
64
100
57
2
42
98
1
0
38
96
67
33
27
58
1
5
1
9
16
75
58
44
28
48
6e
1
6
40
95
45
3
46
87
148
a
Conditions: catalyst 2.6 wt.% Au/MgO (10 mg, ca. 0.6 wt.%); methanol (2 mL); 10 atm (O
Turnover number (TON) is calculated as a ratio between the amounts of 1b and 1c formed and the surface amounts of Au, considering that 1c is formed from 1a via 1b and
2
). Conversion and selectivity were determined by GC; n.d. – not determined.
b
both steps are catalyzed by Au: TON = [n(1b)+2n(1c)]/n (Au), where n represents the amount of the indicated compound or Au in moles. The surface amounts of Au
correspond to 8.47% of the total amounts (calculated based on the average particle diameter of 3.4 nm).
c
Initial rate of the substrate conversion per mol of Au (initial turnover frequency).
The catalyst was reused after this run without a significant loss of activity and selectivity.
K CO – 0.4 mmol.
2 3
d
e
Au/MgO catalyst
O2, methanol
Ph-CH OH
PhCHO
+
PhCO(OCH3)
Au/MgO
2
R' CH(OH)R''
a - 9a
+
1/2 O2
R' C(O)R'' +
H O
2
1a
1b
1c
1
Au: 0.02 - 0.75 mol%
1b - 9b
Scheme 1. Oxidation of benzyl alcohol 1a into benzaldehyde 1b and methyl
benzoate 1c.
Scheme 2. Oxidation of natural alcohols into corresponding aldehydes or ketones.
ter, which is considered as a key intermediate of the process,
undergoes then dehydrogenation to give the ester. Small amounts
of benzoic acid detected in the reaction solutions could be the
result of the dehydrogenation of a hydrate formed by the interac-
tion of the aldehyde with trace water instead of methanol. The
formation of the hemiacetal as a reaction intermediate has been
also suggested to explain the formation of benzyl benzoate at the
solvent-free oxidation of benzyl alcohol over gold catalysts [17].
The activity of the Au/MgO material was also examined in the
aerobic oxidation of a wide variety of primary and secondary
alcohols, including sterically hindered cyclic alcohols. In all exper-
iments, this material was also applied as a sole catalyst in the
absence of base co-catalysts. Corresponding aldehyde and ketones
were detected as major, often exclusive, products of these reac-
tions (Scheme 2). Representative results are collected in Table 2.
A secondary benzylic alcohol, 1-phenylpropanol (2a), can be
oxidized over Au/MgO to give corresponding ketone 2b (Scheme
3) in a solvent-free system (Table 2, run 3) as well as in methanol
or toluene solutions (Table 2, runs 4–6). It should be mentioned
that the reaction occurs much faster in methanol than in toluene
Table 2
a
Oxidation of alcohols catalyzed by Au/MgO.
Run
1c
Substrate (mmol)
Solvent
Catalyst (mg)
Temperature (°C)
Time (h)
Conversion (%)
Product
Selectivity (%)
TONb
Benzyl alcohol (1a)
Benzyl alcohol (1a)
1-Phenylpropanol (2a)
1-Phenylpropanol (2a)
1-Phenylpropanol (2a)
None
None
None
Methanol
Methanol
20
20
10
10
10
100
120
120
130
110
4
4
8
2
2
8
2
8
8
7
10
8
8
8
15
44
30
97
73
90
40
50
25
90
98
86
76
12
34
20
1b
1b
2b
2b
2b
2b
2b
3b
3b
4b
5b
6b
6b
7b
8b
9b
100
65
95
100
100
100
100
90
12.262
35.970
20.436
3.543
c
2
3
c
4
5
d
9.809
3.778
1.812
900
4.905
2.725
1.567
1.362
408
6d
7
1-Phenylpropanol (2a)
Nerol (3a)
Nerol (3a)
Perillyl alcohol (4a)
Isoborneol (5a)
Carveol (6a)
Carveol (6a)
Menthol (7a)
Isopulegol (8a)
Citronellol (9a)
Toluene
Methanol
Toluene
Methanol
Methanol
Methanol
Toluene
Methanol
Methanol
Methanol
10
10
10
20
40
20
20
10
20
20
110
80
80
110
110
120
120
130
120
120
8
9
85
97
d
1
1
1
1
1
1
0d
1
2
3
4
100
100
100
100
80
8
8
613
340
5
46
a
Conditions: substrate (0.4 mmol), solvent (2 mL), 10 atm (O
TON – moles of the substrate converted/moles of Au. TON was calculated with respect to the surface amounts of gold (8.47% of the total amounts as calculated from the
2
), catalyst contains 2.6 wt.% of Au. Conversion and selectivity were determined by GC.
b
average particle diameter of 3.4 nm).
c
Substrate – 2 mL (18 mmol of 1a or 15 mmol of 2a).
Substrate – 1.2 mmol.
d

V.V. Costa et al. / Journal of Catalysis 292 (2012) 148–156
155
OH
OH
OH
OH
OH
OH
OH
OH
2
a
3a
3b
4a
4b
5a
6
a
7a
8a
9a
O
O
O
O
O
O
O
O
2
b
5b
6
b
7b
8b
9b
Scheme 3. Structures of alcoholic substrates 2a–5a and corresponding products
b–5b.
2
Scheme 4. Structures of alcoholic substrates 6a–9a and corresponding products
b–9b.
6
(Table 2, cf. runs 5 and 6), a tendency that has been later confirmed
by the study with other alcohols.
Non-activated monoterpenic alcohols, menthol (7a), isopulegol
8a), and citronellol (9a) (Scheme 4) were much less reactive to-
In a further work, the Au/MgO catalyst was tested for the aero-
bic oxidation of various natural monoterpenic alcohols in order to
clarify the substrate scope and to synthesize carbonylic terpenoids
important for fragrance and pharmaceutical industries. With most
of the alcohols, the reaction conditions were varied to achieve bet-
ter yields for corresponding aldehydes or ketones. The results of
the runs with the best yields are presented in Table 2.
Nerol (3a), a monoterpenic primary allylic alcohol available
from many essential oils, was oxidized over the Au/MgO catalyst
giving as a major product (Z)-citral (3b), an important aroma com-
pound with a strong lemon odor (Table 2, runs 7 and 8; Scheme 3).
Citral was obtained with 90% selectivity at 50% conversion of nerol.
The reaction was much slower in toluene solutions, like it was ob-
served with 1-phenylpropanol (Table 2, cf. runs 7 and 8).
The oxidation of another monoterpenic primary allylic alcohol,
perillyl alcohol (4a), occurred with an excellent selectivity to give
perillyl aldehyde (4b) in nearly quantitative yield (Scheme 2; Table
(
ward oxidation under similar conditions than their activated coun-
terparts, carveol, and nerol. Although the reactions with 7a and 8a
were selective to corresponding ketones (menthone 7b and isop-
ulegone 8b), low conversions were attained under relatively dras-
tic conditions (120–130 °C, 8 h, Table 2, runs 13–15). It should be
mentioned that no formation of the corresponding methyl esters
has been observed at the oxidation of primary alcohols 3a, 4a,
and 9a, whereas the oxidation of benzyl alcohol under similar con-
ditions gives mainly methyl benzoate. The reasons for such a dif-
ferent behavior of these substrates are not clear yet and need
further investigation.
4
. Conclusions
Monitoring the formation of gold nanoparticles on the surface
2, run 9). Perillyl aldehyde, which is much more expensive than
of magnesium oxide has revealed, for the first time as far as we
know, that the process occurs through three distinguishable steps
of the thermal decomposition of gold hydroxide, which are accom-
panied by negligible hydrogen consumption. Stabilization of gold
hydroxide particles with ammonia hydroxide at the Au/MgO prep-
aration resulted in the formation of truly uniform gold nanoparti-
cles. The material is an effective heterogeneous catalyst for the
liquid-phase oxidation of a wide range of alcohols with environ-
mentally benign molecular oxygen, in the absence of any co-cata-
lyst or additive. In addition, the Au/MgO catalyst performs the one-
pot base-free aerobic oxidative esterification of benzyl alcohol
exhibiting high activity, selectivity to methyl benzoate and stabil-
ity. The oxidation of various biomass-based monoterpenic alcohols
over the Au/MgO catalyst resulted in carbonylic terpenoids, useful
for fragrance and pharmaceutical industries, in good to excellent
yields.
original alcohol 4a, is used as food additive for flavoring and in per-
fumery to add spiciness.
The oxidation of sterically hindered isoborneol (5a) over Au/
MgO material gives almost quantitatively camphor (5b)
(Scheme 3), a compound with a strong, aromatic odor used as a lo-
cal analgesic, respiratory stimulant, plasticizer and an antimicro-
bial agent (Table 2, run 10).
The Au/MgO catalyst is also very effective for the oxidation of
the commercial (Sigma–Aldrich) mixture of trans and cis isomers
of carveol (6a) resulting in carvone (6b) with ca. 100% selectivity
(Scheme 4, Table 2, runs 11 and 12). Similarly to what has been ob-
served with other substrates, the reaction is faster in methanol
solutions than in toluene (Table 2, cf. runs 11 and 12). Thus, we
can conclude that methanol is a better solvent than toluene to per-
form the oxidation of alcohols over the Au/MgO catalyst. Carveol
6
a and perillyl alcohol 4a are both allylic alcohols with a para-
menthenic skeleton; however, carveol is a secondary alcohol. A
similar reaction rate was attained for carveol at higher tempera-
ture and higher catalyst loading as compared to perillyl alcohol,
probably, due to the steric hindrance of the hydroxyl group (Table
Acknowledgments
2
6
, cf. runs 9 and 11). The results obtained for substrates 3a, 4a and
suggest that the Au/MgO catalyst promotes selective
,b-unsaturated alcohols into corresponding
,b-unsaturated aldehydes or ketones without intramolecular
hydrogen transfer and isomerization.
The authors thank E. Flores, A. Diaz, P. Casillas, V. Garcia, M.
Sainz, F. Ruiz, G. Vilchis, E. Aparicio, J. Peralta, J. Palomares, and
M. Vega for technical support. This research project was partly sup-
ported by DGAPA-PAPIIT (UNAM, México) through Grant IN
224510 and by CNPq, CAPES, FAPEMIG and INCT-Catálise (Brazil).
a
transformations of
a
a

1
56
V.V. Costa et al. / Journal of Catalysis 292 (2012) 148–156
[
[
29] H. Miyamura, T. Yasukawa, S. Kobayashi, Green Chem. 12 (2010) 776.
30] L.A. Parreira, N. Bogdanchikova, A. Pestryakov, T.A. Zepeda, I. Tuzovskaya, M.H.
Farías, E.V. Gusevskaya, Appl. Catal. A 397 (2011) 145.
References
[
[
[
[
1] R.A. Sheldon, I.W.C.E. Arends, A. Dijksman, Catal. Today 57 (2000) 157.
2] J. Muzart, Tetrahedron 59 (2003) 5789.
3] M.J. Schultz, M.S. Sigman, Tetrahedron 62 (2006) 8227.
4] K. Mori, T. Hara, T. Mizugaki, K. Ebitani, K. Kaneda, J. Am. Chem. Soc. 126 (2004)
[
[
[
[
[
[
31] L. Menini, M.C. Pereira, L.A. Parreira, J.D. Fabris, E.V. Gusevskaya, J. Catal. 254
(
2008) 355.
32] P.A. Robles-Dutenhefner, K.A. da Silva Rocha, E.M.B. Sousa, E.V. Gusevskaya,
J. Catal. 265 (2009) 72.
33] M.G. Speziali, P.A. Robles-Dutenhefner, E.V. Gusevskaya, Organometallics 26
1
0657.
[
[
[
[
[
5] T. Mallat, A. Baiker, Chem. Rev. 104 (2004) 3037.
(
2007) 4003.
6] T. Nishimura, N. Kakiuchi, M. Inoue, S. Uemura, Chem. Commun. (2000) 1245.
7] K. Yamaguchi, N. Mizuno, Angew. Chem., Int. Ed. 41 (2002) 4538.
8] T. Matsushita, K. Ebitani, K. Kaneda, Chem. Commun. (1999) 265.
9] V.V. Costa, M.J. Jacinto, L.M. Rossi, R. Landers, E.V. Gusevskaya, J. Catal. 282
34] M.G. Speziali, V.V. Costa, P.A. Robles-Dutenhefner, E.V. Gusevskaya,
Organometallics 28 (2009) 3186.
35] L.A. Parreira, L. Menini, J.C. da Cruz Santos, E.V. Gusevskaya, Adv. Synth. Catal.
352 (2010) 1533.
(
2011) 209.
36] P.A. Robles-Dutenhefner, B.N.S. Brandao, L.F. de Sousa Liniker, E.V. Gusevskaya,
Appl. Catal. A 399 (2011) 172.
37] R. Zanella, S. Giorgio, C.R. Henry, C. Louis, J. Phys. Chem. 106 (2002) 7634.
38] S.I. Ivanova, C. Petit, V. Pitchon, Catal. Today 113 (2006) 182.
39] Yu.S. Solkina, S.I. Reshetnikov, M. Estrada, A. Simakov, D.Yu. Murzin,
I.L. Simakova, Chem. Eng. J. 176–177 (2011) 42.
40] D.I. Kochubey, EXAFS Spectroscopy of Catalysts, Nauka, Novosibirsk, 1992.
41] K.V. Klementiev, code VIPER for Windows, freeware. <www.desy.de/~klmn/
viper.html>.
42] N. Binsted, J.V. Campbell, S.J. Gurman, P.C. Stephenson, EXCURV92 Program
Code, SERC Daresbury Laboratory, UK, 1991.
43] E. Smolentseva, A. Simakov, S. Beloshapkin, M. Estrada, E. Vargas, V. Sobolev, R.
Kenzhin, S. Fuentes, Appl. Catal. B 115–116 (2012) 117.
44] D.L. Feldheim, C.A. Foss, Metal Nanoparticles: Synthesis, Characterization and
Applications, Basel Marsel Dekker Inc., New York, 2002.
45] I. Tuzovskaya, N. Bogdanchikova, A. Simakov, V. Gurin, A. Pestryakov,
M. Avalos, M. Farias, Chem. Phys. 338 (2007) 23.
46] J.F. Moulder, W.F. Stickle, P.E. Sobol, K.D. Bomben, Handbook of X-ray
Photoelectron Spectroscopy, Perkin-Elmer Corporation, Physical Electronics
Division, Eden Prairie, 1992. p. 179.
47] N. Weiher, E. Bus, L. Delannoy, C. Louis, D.E. Ramaker, J.T. Miller, J.A. van
Bokhoven, J. Catal. 240 (2006) 100.
48] H. Yoshida, S. Nonoyama, Y. Yazawa, T. Hattori, Phys. Scr. T115 (2005) 813.
49] ICSD Database Code 64701-Au0.
50] ICSD Database Code 8014-Au2O3.
51] ICSD Database Code 64701-AuAlO2.
52] 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.
53] R.E. Benfield, J. Chem. Soc. Faraday Trans. 88 (1992) 1107.
54] A.I. Frenkel, C.W. Hills, R.G. Nuzzo, J. Phys. Chem. 105 (2001) 12689.
[
[
[
[
[
[
10] L. Prati, M. Rossi, J. Catal. 176 (1998) 552.
11] C. Della Pina, E. Falletta, L. Prati, M. Rossi, Chem. Soc. Rev. 37 (2008) 2077.
12] G.J. Hutchings, Chem. Commun. (2008) 1148.
[
[
[
13] A. Corma, H. Garcia, Chem. Soc. Rev. 37 (2008) 2096.
14] V.R. Choudhary, D.K. Dumbre, Catal. Commun. 13 (2011) 82.
15] S. Meenakshisundaram, E. Nowicka, P.J. Miedziak, G.L. Brett, R.L. Jenkins,
N. Dimitratos, S.H. Taylor, D.W. Knight, D. Bethell, G.J. Hutchings, Faraday
Discuss. 145 (2010) 341.
[
[
[
16] D.I. Enache, J.K. Edwards, P. Landon, B. Solsona-Espriu, A.F. Carley, A.A. Herzing,
M. Watanabe, C.J. Kiely, D.W. Knight, G.J. Hutchings, Science 311 (2006) 362.
17] V.R. Choudhary, D.K. Dumbre, Top. Catal. 52 (2009) 1677.
18] M. Sankar, E. Nowicka, R. Tiruvalam, Q. He, S.H. Taylor, C.J. Kiely, D. Bethell,
D.W. Knight, G.J. Hutchings, Chem. Eur. J. 17 (2011) 6524.
[
[
[
[
[
[
[
[
19] M. Boronat, A. Corma, F. Illas, J. Radilla, T. Ródenas, M.J. Sabater, J. Catal. 278
(
2011) 50.
[
[
20] P. Haider, J.-D. Grunwaldt, A. Baiker, Catal. Today 141 (2009) 349.
21] G.L. Brett, Q. He, C. Hammond, P.J. Miedziak, N. Dimitratos, M. Sankar,
A.A. Herzing, M. Conte, J.A. Lopez-Sanchez, C.J. Kiely, D.W. Knight, S.H. Taylor,
G.J. Hutchings, Angew. Chem. Int. Ed. 50 (2011) 10136.
[
[
22] A. Villa, G.M. Veith, L. Prati, Angew. Chem. Int. Ed. 49 (2010) 4499.
23] M. Alhumaimess, Z. Lin, W. Weng, N. Dimitratos, N.F. Dummer, S.H. Taylor,
J.K. Bartley, C.J. Kiely, G.J. Hutchings, ChemSusChem 5 (2012) 125.
24] S.K. Klitgaard, A.T. DeLa Riva, S. Helveg, R.M. Werchmeister, C.H. Christensen,
Catal. Lett. 126 (2008) 213.
[
[
[
[
[
[
[
[
[
25] T. Hayashi, T. Inagaki, N. Itayama, H. Baba, Catal. Today 117 (2006) 210.
26] I.S. Nielsen, E. Taarning, K. Egeblad, R. Madsen, C.H. Christensen, Catal. Lett.
1
16 (2007) 35.
[
[
27] E. Taarning, A.T. Madsen, J.M. Marchetti, K. Egeblad, C.H. Christensen, Green
Chem. 10 (2008) 408.
28] F.-Z. Su, J. Ni, H. Sun, Y. Cao, H.-Y. He, K.-N. Fan, Chem. Eur. J. 14 (2008) 7131.
[
[