Au?Pd Selectivity-switchable Alcohol-oxidation Catalyst: Controlling the Duality of the Mechanism using a Multivariate Approach

Article abstract of DOI:10.1002/cctc.201900512

Supported gold-palladium nanoparticles are highly selective catalysts for the oxidation of alcohols. However, little is known about how integrated reaction conditions can affect the chemoselectivity of a specific catalytic system. Herein, a novel Au?Pd selectivity-switchable catalyst supported on SrCO₃ is reported; and a multivariate optimization was suggested as the key process to better understand the formation of the products. The optimization approach considered temperature, pressure, time of reaction, and Au : Pd molar ratio, and settled that the temperature and Au : Pd molar ratio showed an important effect on the ester yield, while just the metal molar ratio significantly influenced the selectivity for the aldehyde. Thus, taking into consideration the experimental data and optimized conditions, we were able to efficiently switch the selectivity by just changing the pressure of the system in a benzyl alcohol oxidation reaction. In addition, we proposed that the presence of O₂ implies that there are two catalytic pathways, which leads to different selectivity, allowing us to bring some mechanistic insights dealing with the duality of the mechanism. Such outcomes are based on dense experimental results and characterizations (FT-IR, Rietveld refinement, XPS, H₂-TPR, and EDS-STEM). The catalyst was also very stable, presenting activity up to 6 runs without loss of activity and selectivity, under certain reaction conditions.

Full text of DOI:10.1002/cctc.201900512

Accepted Article
Title: Au-Pd selectivity-switchable alcohol-oxidation catalyst: controlling
the duality of the mechanism using a multivariate approach
Authors: Itaciara E. M. da S. Melo, Samuel A. A. de Sousa, Laíse N.
dos S. Pereira, Jefferson M. Oliveira, Karla P. R. Castro, Jean
C. S. Costa, Edmilson M. de Moura, Carla V. R. de Moura,
and Marco Aurelio Suller Garcia
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemCatChem 10.1002/cctc.201900512
Link to VoR: http://dx.doi.org/10.1002/cctc.201900512
A Journal of
www.chemcatchem.org

ChemCatChem
10.1002/cctc.201900512
Au-Pd selectivity-switchable alcohol-oxidation catalyst: controlling the
duality of the mechanism using a multivariate approach
Ms. Itaciara E. M. da S. Melo, Dr. Samuel A. A. de Sousa, Ms. Laíse N. dos S.
Pereira, Jefferson M. Oliveira, Karla P. R. Castro, Dr. Jean C. S. Costa, Dr. Edmilson
M. de Moura, Dr. Carla V. R. de Moura, Dr. Marco A. S. Garcia*
Chemistry Department, Federal University of Piauí, Campus Universitário Ministro
Petrônio Portella, 64049-550 Teresina PI (Brazil)
*
Corresponding author. E-mail: marcoasuller@gmail.com
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201900512
Abstract
Supported gold-palladium nanoparticles are highly selective catalysts for the
oxidation of alcohols. However, little is known about how integrated reaction conditions
can affect the chemoselectivity of a specific catalytic system. Herein, a novel Au-Pd
selectivity-switchable catalyst supported on SrCO is reported; and a multivariate
3
optimization was suggested as the key process to understand the formation of the
products better. The optimization approach considered temperature, pressure, time of
reaction, and Au:Pd molar ratio, and settled that the temperature and Au:Pd molar ratio
showed an important effect on the ester yield, while just the metal molar ratio
significantly influenced the selectivity for the aldehyde. Thus, taking into consideration
the experimental data and optimized conditions, we were able to efficiently switch the
selectivity by just changing the pressure of the system in a benzyl alcohol oxidation
reaction. Also, we proposed that the presence of O implies that there are two catalytic
2
pathways, which leads to different selectivity, allowing us to bring some mechanistic
insights dealing with the duality of the mechanism. Such outcomes are based on dense
experimental results and characterizations (FT-IR, Rietveld refinement, XPS, H -TPR,
2
and EDS-STEM). The catalyst was also very stable, presenting activity up to 6 runs
without loss of activity and selectivity, under certain reaction conditions.
Keywords: SrCO ; Au-Pd nanoparticles; alcohol oxidation; factorial design
3
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201900512
Introduction
In recent years, major research efforts have been given to the development of
nanoparticles (NPs) and nanostructured materials with potential application in the
chemical industry.[1] Following the discoveries of Haruta[2] and Hutchings[3] in the use
of gold catalysts as environmentally benign processes, the catalysis by supported gold
and gold-palladium NPs has established new methodologies for the achievement of
organic transformations under milder reaction conditions,[4] such as propene
epoxidation,[5] oxidation of several compounds[6] and other chemical matters.[7] In this
scenario, oxidation reactions of alcohols can be highlighted since they are considered as
fundamental functional-group transformations in organic synthesis due to the possibility
of improving access to compounds like aldehydes, ketones, esters, and carboxylic
acids.[8] However, the need for toxic stoichiometric reagents as manganese[9] and
chromium[10] is a great disadvantage, which prompts the use of metal-based catalysts.
For that, one of the reasons for the choice of a bimetallic system is the electronic
coupling between the metals, which produces nanocrystals with improved physical-
chemistry properties and, consequently, superior catalytic performances than single
metals.[11] Moreover, the use of Au and Pd is interesting since they are miscible in a
wide variety of compositions, easily forming Au-Pd alloys NPs.[12]
Catalytic systems are a long-standing challenge since their chemoselectivity is
limited, requiring modulation of NPs or support surfaces[13], ligands,[14] changing on the
reaction conditions[15] or the use of additives as external base[6f, 6g], which hampers their
wide applications on more complex organic synthetic pathways and total synthesis.[16]
Thus, selectivity and mechanisms are important features that need to be understood to
assist catalysts enhancements for wide applications. To achieve high selectivity, the
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201900512
production of by-products via consecutive and parallel reactions must be considered for
the design of improved catalytic methods.[17] For that matter, benzyl alcohol is one of
the most studied model molecules for selective oxidation reactions[7c, 18] since it allows
simpler elemental considerations regarding reaction conditions, the nature of the
catalyst, and mechanisms insights. Therefore, constant efforts have been made by
researchers to address such correlations using this reaction model.
In addition to the bimetallic and substrate choices for specific studies, the support
used as NPs carrier is essential because support effects and metal-support interactions
can be considered selectivity-driving processes.[19] Strontium-based or -derived supports
have been investigated in the literature, and their effect on the catalysis is quite
motivating. Titanate nanotubes were modified with alkali and alkaline earth metal ions
and supported with Au NPs for benzyl alcohol oxidation. The authors observed that the
catalytic activity had a direct relationship with the basicity of the support without external
base addition, suggesting a strong interaction between the support and the NPs.[20] Castro et
al. observed a similar effect by using a mixture of Sr(OH)
2
, Sr(OH)
2
. H
2
O, and SrCO ;
3
however, K
2
CO
3
addition was essential for the catalyst stability.[6f] Considering that a
catalytic process as a whole is a complex issue, an available tool for the correlation among
all the variables mentioned above has been increasingly used to help researches: a
multivariate analysis. Its efficiency in building models, optimize multifactor experiments,
and evaluate the effects of plentiful aspects provides the possibility of gathering all the
information and predict how a system response will be under certain conditions.[21]
Herein, we propose the synthesis of a catalyst comprised of Au-Pd NPs
impregnated by sol-immobilization in SrCO . We have demonstrated that the
3
combination of Au-Pd NPs over such support can provide a pronounced performance
for the solvent-free oxidation of benzyl alcohol without the need for external base
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201900512
addition. Moreover, the study of temperature, pressure, time of reaction, and Au:Pd
molar ratio were optimized for attaining conversion and selectivity to benzaldehyde or
benzyl benzoate using multivariate processes, aiming a selectivity-switchable system by
only changing one reaction condition. We also performed a full characterization of the
best catalyst for our proposes (Au:Pd molar ratio of 1:1.5), which associated with the
experimental data, provided mechanistic insights. The stability of the catalyst was
satisfactory, with six runs without loss of activity.
Experimental
Materials
All chemicals used in the experiments were of analytical grade, bought from
Sigma-Aldrich, and used as received, without further purification.
Catalyst preparation
The catalyst was synthesized by using a method described elsewhere with
modifications.[22] The preparation of a catalyst with Au:Pd molar ratio of 1:1.5 is
described as an example: in a typical procedure, a previously prepared aqueous solution
containing 7.1 mg of palladium(II) chloride (≥ 99.9%) in 10 wt.% HCl was added in a
3
9
0-mL aqueous solution containing 31.7 mg of gold(III) chloride solution (HAuCl
4
,
9.99% trace metals basis, 30 wt.% in dilute HCl). The same procedure was used for the
preparation of other Au:Pd molar ratios, changing the quantities of metal precursors.
Such ratios will be further shown. After a 5-minute magnetic stirring at room
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201900512
temperature, 0.6 mL of a 2.0 wt.% aqueous solution of polyvinyl alcohol (PVA, 80%)
was added to the solution, and the system was further stirred for 5 min. Subsequently, a
freshly prepared 0.1% sodium borohydride (NaBH , powder, ≥ 98.0%) aqueous solution
was added dropwise under magnetic stirring. A 5-fold excess of NaBH (mol/mol),
related to the mol of metals, was added to the main mixture, changing its quantity for
the different Au:Pd molar ratios. The solution turned darkish instantly after the NaBH
addition, but it was kept under stirring for 30 min. The catalyst support consists of
commercial strontium carbonate (SrCO , ≥ 98%).
4
4
4
3
After the NPs synthesis, 500 mg of the support was added to the prepared sol and
stirred for 2 h, at room temperature. Finally, the suspension was centrifuged at 2000
rpm for 2 min. The catalyst was dried in an oven at 50°C for 5 h before storage in an
amber bottle. The blackish powder was designated as Au-Pd/SrCO .
3
Catalyst characterization
The scanning transmission electron microscopy (STEM) images of the as-
th
prepared and spent (after the 6 run) Au-Pd/SrCO
3
catalysts were obtained with an FEI
Tecnai G2 F20 transmission electron microscope (Thermo Fisher Scientific,
Massachusetts, EUA) operating at 200 kV with an energy dispersive spectrometer
(Bruker XFlash 6130T Silicon Drift Detector). The element distribution maps were
recorded using energy-dispersive X-Ray spectroscopy (EDS). Samples for microscopy
were prepared by drop casting an isopropanol suspension of the materials over a grid
comprised of carbon-coated copper, followed by drying under ambient conditions. The
nanoparticles’ size was determined by using the ImageJ software. For that, 200 particles
were considered. The metal content in the catalysts (before and after usage) was
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201900512
measured by Flame Atomic Absorption Spectroscopy (FAAS), using an AA-6300
Atomic Absorption Spectrophotometer (Shimadzu Corp, Kyoto, Japan). The digestion
procedure of the samples was performed using concentrated nitric and hydrochloric
acids in the ratio of 1:3 (HNO
3
:HCl) at the heating of 115°C for 2 h. X-ray diffraction
(XRD) pattern of the catalyst was recorded on an XRD-6000 diffractometer (Shimadzu
Corp, Kyoto, Japan) with Cu Kα radiation (1.5418 Å), operating at 40 kV and 40 mA.
Rietveld refinement of the as-prepared catalyst was performed using Rex 0.8.2 software.
The X-ray photoemission spectra (XPS) were obtained with ESCA + spectrometer
system equipped with an EA 125 hemispherical analyzer and XM 1000 monochromated
X-ray source (Scientia Omicron, Uppsala, Sweden) in Al K (1486.7 eV). The X-ray
source was used with a power of 280 W, as the spectrometer worked in a constant pass
energy mode of 50 eV. The calibration of the XPS spectra for the charge accumulation
was performed using C 1s peak (BE = 284.8 eV). The Fourier Transform Infrared
Spectroscopy (FT-IR) spectra were acquired by using a Spectrum 100 (Perkin Elmer,
-1
Massachusetts, EUA), set to measure 32 cumulative scans at 4 cm . The samples were
prepared as KBr pellets. Temperature-programmed reduction with hydrogen (H -TPR)
2
was carried out in a ChemBET-Pulsar instrument (Quantachrome Instruments, Boynton
Beach, United States) equipped with a thermal conductivity detector. Typically, 0.05 g
of a catalyst was dried under a He flow at 120°C for 1 h and cooled to room
temperature. The H
0% H /N , with a linear temperature increasing rate of 10°C min . The unit cell
representation of SrCO was modeled by the VESTA (version 3.4.0, JP-Minerals,
2
-TPR profiles were obtained between 50 and 1100°C in a flow of
-1
1
2
2
3
Ibaraki, Japan) and Diamond (version 3.2g, Crystal Impact GbR, Bonn, Germany).
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201900512
Catalytic procedure
All the experiments used the oxidation of benzyl alcohol as a model reaction.
pressure.[6g] The
They were carried out in a glass Fisher-Porter reactor under O
2
reactions were performed at different temperatures under different pressures on a
magnetic stirrer coupled to a heating plate with temperature control. In a typical
reaction, 1 mL (9.6 mmol) of benzyl alcohol and 43.5 mg of the catalyst (2.0 wt.%
metal) were loaded in the glass tube. Then, the system was purged 3 times with a stream
of oxygen, and filled with the gas at a pressure between 1 and 5 bar. The resulting
mixture was magnetically stirred at 500 rpm at a selected reaction time and temperature,
which will be specified. After the reaction, the catalyst was separated from the products
by centrifugation. Then, 20 μL of the reaction solution was mixed with 1 mL of
methylene chloride (CH
2
Cl ) to determine the oxidation product yields by GC. P-xylene
2
was added to the product as an internal standard. Gas chromatography (GC) analyses
were performed by using a GC-2010 Plus equipment (Shimadzu, Kyoto, Japan) using a
Carbowax capillary column. The activity of the catalyst was measured by testing the
conversion of the benzyl alcohol and the selectivity and yield of benzaldehyde and
benzyl benzoate (also, the selectivity of benzoic acid was considered and is in shown in
Table S1). The conversion of benzyl alcohol, product selectivity, and product yield were
calculated as follow:
퐴푙푐ꢁℎꢁ푙푏ꢂ푔푖ꢃ,푚ꢄꢅ−퐴푙푐ꢁℎꢁ푙푎푓푡ꢂꢆ.푚ꢄꢅ
퐶표푛푣푒푟푠ꢀ표푛 =
푥 100%
Equation 1
Equation 2
퐴푙푐ꢁℎꢁ푙푏ꢂ푔푖ꢃ,푚ꢄꢅ
푃ꢊꢁ푑푢푐ꢋ ꢁꢌ ꢍꢎꢋꢏꢊꢏꢐꢋ푚ꢄꢅ
푆푒ꢇ푒ꢈꢉꢀ푣ꢀꢉ푦 =
푥 100%
퐴푙푙 푝ꢊꢁ푑푢푐ꢋꢐ푚ꢄꢅ
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201900512
푌ꢀ푒ꢇꢑ = 푆푒ꢇ푒ꢈꢉꢀ푣ꢀꢉ푦 표ꢒ ꢉꢓ푒 ꢔ푟표ꢑꢕꢈꢉ 표ꢒ ꢀ푛ꢉ푒푟푒푠ꢉ 푥 퐶표푛푣푒푟푠ꢀ표푛
Equation 3
before each
For catalyst recycling, the Au-Pd/SrCO
cycle and dried in an oven at 50°C.
3
was washed with CH
2
Cl
2
Experimental design for catalyst performance
4
Two full factorial designs with four independent variables (2 ) were performed.
Such a process was carried out to screen the statistically significant factors for the
catalyst performance in the benzyl alcohol oxidation reaction, in terms of benzyl
benzoate and benzaldehyde yields. The investigated variables and their levels (uncoded)
were: pressure (1 and 5 bar), temperature (80 and 120°C), time of reaction (0.5 and 2.5
h) and Au:Pd molar ratio (1:0.1 and 1:1.5).
Regarding the benzyl benzoate yield, the significant factors were the
temperature, the Au:Pd molar ratio, and their interaction. To obtain a suitable model for
this product, experiments were performed with additional levels (temperatures: 71.7 and
1
28.3 °C; Au:Pd molar ratios: 1:0 and 1:1.8) maintaining all the others variables in
proper levels (pressure: 5 bar and reaction time: 2.5 h), which were pointed out by the
4
results from the 2 full factorial experimental design. Also, to evaluate the experimental
variation for the validation of the model by Analysis of Variance (ANOVA) replicates
(three) at the central point were carried out. After this step, the Response Surface
Methodology (RSM) was used to reach a better visualization provided by the
mathematical model. Finally, replicated experiments regarding only the Au:Pd molar
ratios were performed to find the optimum value of this variable, maximizing the benzyl
benzoate yield. Only the Au:Pd molar ratio was pointed out as significant for the
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201900512
benzaldehyde yield; then, an optimum value for the Au:Pd molar ratio was searched to
maximize the response, maintaining the other factors in proper levels (pressure: 1 bar,
reaction time: 2.5 h and temperature: 120 °C). One-way ANOVA was used to evaluate
the significance of the Au:Pd molar ratio on the yields, considering its levels. Table 1
summarizes all levels (coded and uncoded) of all the studied variables. The experiments
were performed randomly at different days to avoid systematic errors. The statistical
software used for calculations was STATISTICA® (Version 10.0, StatSoft Inc., Tulsa,
USA). For statistically significant differences between means, a post hoc pairwise
comparison with Tukey’s test was used. Statistical significance was set at P < 0.05.
Results and discussion
Catalyst characterization
Our investigations started with the catalyst synthesis using a simple PVA-
stabilized Au-Pd NPs preparation method, with reduction of the metals using NaBH
.[22]
4
Our approach enabled the synthesis of pre-formed NPs, which were immobilized onto
the support. The choice of support was based on a previous study published by some of
us.[6f] The strategy used at that moment was to prepare SrO upon the calcination of the
product of the reaction between strontium nitrate and sodium hydroxide; however, a
mixture of Sr(OH)
2
, Sr(OH)
2
. H
2
O and SrCO was obtained, yet it presented remarkable
3
activity for the oxidation of benzyl alcohol. Since strontium compounds have not been
fully studied in the literature as supports for gold, palladium, and gold-palladium NPs,
and that SrO is not stable for our purposes, we have decided to explore the application
of SrCO
3
as catalytic support. All the following characterizations were performed with
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201900512
Au:Pd molar ratio of 1:1.5, due to experimental results presented forward. However, it
is worth mentioning that all the Au:Pd molar ratios used herein were considered as the
total amount of metal detected by FAAS (not just the moles of surface Au and Pd
atoms).
We have used a commercial SrCO material; however, as the immobilization step
3
can promote some support modification, characterizations were necessary to discover
the specific material identity fully. We aimed to have just one strontium phase to have
specific metal-support interactions[23] with well-known compounds. To focus on our
efforts on the accomplishment of having just one phase of the strontium compound, we
have used some complementary techniques to guarantee the composition of the support,
as well as the catalyst itself as a whole. FT-IR spectra were recorded for SrCO
Pd/SrCO materials, which are shown in Fig. S1A. Both samples present the peaks that
correspond to CO
3
and Au-
3
2
-
-1
3
: 698 (706), 857 (855), 1071, 1454 (1476), 1774 (1772) cm , being
the parenthesized values for Au-Pd/SrCO
3
catalyst, when it differs from the SrCO
3
itself.[24] Such differences can be associated with the interaction between the support
defects and metal NPs, changing the energy of the SrCO surface, causing slight
3
shiftings.[25] The results suggest the achievement of a unique phase for the support used,
which had to be confirmed.
To elucidate such feature, XRD patterns of the SrCO containing Au-Pd NPs were
3
recorded to analyze the specific crystal planes that comprise the sample. The
diffractogram shows lines that correspond to the carbonate itself, once the low
concentration of the Au-Pd NPs hampered the observation of their diffraction peaks in
the XRD. The observed main diffraction peaks of the SrCO
3
(Fig. S1B) are positioned
at 2θ angles of 25.21°, 25.89°, 29.61°, 31.51°, 34.48°, 35.09°, 36.23°, 36.60°, 41.32°,
4
4.05°, 45.65°, 46.7°, 47.7°, 49.98°, which correspond to the planes of (111), (021),
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201900512
(002), (012), (102), (200), (112), (190), (220), (221), (041), (202), (132), (113),
respectively. Such data can be attributed to the orthorhombic phase of SrCO
3
.[24] In
order to rely on the information of the phase purity and crystallinity, we have performed
a Rietveld refinement. Table 2 shows the lattice parameters, unit structural volume, site
occupancy and atomic positions determined from the refinement. The SrCO
isostructural with a point-group symmetry D2h(mmm) and lattice constants as follows: a
5.0923(2) Å; b = 8.3893(8), c = 6.0100(9) Å. The Rietveld refinement fitted well to
the experimental data, with values for the Goodness of Fit (GoF), Rprofile (R ) and
weighted profile (Rwp) of 3.5075%, 0.2058%, and 0.2550%, respectively. All the results
3
phase is
=
p
R
confirmed a unique phase comprised of SrCO
3
in an orthorhombic phase (ICSD
is shown in
2
02793). A schematic representation of the coordination assembly of SrCO
3
Fig. S1C. The structure belongs to the crystal space group D2h16(Pmcn) with Z = 4. The
FT-IR spectrum formerly shown validates the spatial group observed in the structural
unit for the strontium carbonate. It should be noted that the number of observed bands
for normally degenerate vibrations can also be used as an indication of the group of
molecular points. [26] The structural unit of SrCO
anions of CO
fundamental vibration modes (ν1 mode is IR inactive; ν2 bend, ν3 asymmetric stretch
contains four cations of Sr and four
2+
3
2−. The linear molecule of O=C=O with the D1h symmetry has four IR
3
-1
and ν4 planar bend around are located at 857, 1454 and 698 cm , respectively, as
shown in Fig. S1A).[27]
The composition of the catalyst was also investigated in details through elemental
analysis studies using line scan and mapping in STEM mode. The catalysts were
synthesized to have 2.0 wt.% of metal in different ratios, which were confirmed by
FAAS. The elemental mapping shows NPs distributed all over the support, which is
following the synthesis procedure (as shown in the spectrum image scanning in Fig.
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201900512
1
A). Fig. 1B-D illustrates the STEM-EDS images for the as-synthesized catalyst Au-
Pd/SrCO (1:1.5), where Au and Pd spatial distributions are almost overlapped with
3
each other since the Au and Pd are intermixed in the Au-Pd NPs in accordance to the
atomic interdiffusion presented by the material.[28] Such strong intermixing provides a
large body of metals interfaces, favoring the catalytic activity of the system,[29]
suggesting the formation of an alloy between the metals. The mean diameter for metal
NPs was 4.55 ± 1.22 nm (Fig. S2). Also, the Sr spatial distribution is reliably uniform,
as established in Fig 1D.
TPR-H
impregnated with the Au-Pd NPs (Au-Pd/SrCO
presented an H consumption shoulder centered at 890°C and an intense H
2
analyses were performed for the naked SrCO
). As displayed in Fig. 2A, SrCO
peak
3
and the support
3
3
2
2
consumption centered at 1024°C, which can be assigned to the reduction of the
compound due to desorption of carbonate species[30] and reduction of carbonate to CO
or, eventually, CH
were detected, confirming that the reduction of Au and Pd species was efficient upon
the NaBH usage; however, the shifting of the reduction temperatures to lower values
814 and 1009°C) after the metal impregnation can be attributed to some sort of
4
.[30b]After the immobilization of the metals, no new reduction peaks
4
(
interaction between the metals and the support. The observation can be explained by H
2
absorption on the Pd[31] or Au-Pd[32] surfaces, which weaken the H-H bond, decreasing
the H activation barrier. Thus, it is more readily available for the strontium compound
2
reduction. Such a feature is very important for catalysis reactions since strong
interactions between the metals and the support usually enhance the catalytic activity of
a given system.[33] In addition, electronic structure modulation by electron transfer
among the metals also can be very expressive on the enhancement of a given catalytic
system performance.[7d]
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201900512
XPS analyses were performed to clarify the composition of the material and
confirm the obtained information from the H -TPR analyses. The survey scan of the
2
catalyst (Fig. S3) revealed the presence of Au, Pd, Sr, O, and C, as expected. The
carbon content can be related to adventitious C and surface carbonates. Fig. 2B presents
0
the high-resolution Au 4f with binding energies suggesting Au species (Au 4f7/2, 82.9
eV; Au 4f5/2, 86.5 eV). Such values are following the literature[34] and with our designed
process since a 5-fold excess of NaBH was employed in a clear attempt to reduce the
4
3
+
metals fully. No obvious peaks for Au were detected, but they cannot be completely
excluded. The deconvolution of the high-resolution Pd 3d XPS spectra presented (Fig.
0
2
C) Pd as the only chemical state (Pd 3d5/2, 334.8 eV; Pd 3d3/2, 340.0 eV). One may
notice that any slight shift on the binding energy can be related to the chemical
environment since a bimetallic system was under consideration[35] and, although not
2
+
[36]
observed, a minor contribution from Pd cannot be rejected. The binding energy of
34.8 eV for Sr 3d1/2 (Fig. 2D) corresponds to SrCO , while the other two binding
1
3
energies are attributed to Sr–O bonds (Sr d3/2, 132.5 eV; Sr d3/2, 134.0 eV).[37] The O 1s
peaks tend to be broad due to the overlapping of components (Fig. 2E). The best fitting
was achieved with two individual peaks at energies of 530.7 and 532.1 eV, which are
assigned as C–OH and C–O–C species, respectively.[38] The C 1s peaks are
characteristic of the carbonate support, corroborating to the other results presented
herein (Fig. 2F). The binding energies of 284.4 and 286.3 eV are typically peaks of
adventitious carbon. The energy of 288.8 eV can be associated with carbonate, while the
2
90.3 eV can be assigned to C 1s component for hydrocarbons.[39]
Experimental design for catalyst performance
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201900512
Previous catalytic tests using the proposed material have shown that the main
products observed in the oxidation reaction were benzaldehyde and benzyl benzoate,
depending on the reaction conditions, with inexpressive benzoic acid content. Through a
univariate approach, a screening of Au:Pd molar ratio was performed, which was used
to choose the parameters for the multivariate methodology used in the proposed
experimental design. At first sight, an effect on the catalyst selectivity was quite
surprising since we just modulated the reaction conditions and the Au:Pd ratios, without
any additive or changing on the solvent, as proposed before in the literature.[40] Then, a
multivariate approach was proposed to evaluate how some previously chosen variable
may affect the selectivity for the main products of benzyl alcohol oxidation for the
system herein studied. The tests were performed individually for each product
formation, in an attempt to deeply show how the reaction conditions would affect the
reaction pathway.
Full factorial designs consisting of 16 experimental runs, based on responses of
the benzyl alcohol oxidation in terms of benzyl benzoate and benzaldehyde yields, are
shown in Table 3. As a matter of comparison, conversion and selectivity results are
2
shown in Table S1. The coefficient of determination (R ) of the model for benzyl
benzoate yield was equal to 0.91, indicating that the independent variables were able to
explain 91% of the variation on the dependent variable around the global mean. Among
the investigated variables, the temperature and Au:Pd molar ratio showed a significant
effect on the benzyl benzoate yield, as pointed out by the Pareto Chart of standardized
effects in Fig. 3A. Such chart shows both the magnitude and importance of the
parameters, and the horizontal dashed line corresponds to the t value from the student
distribution, with 95% of confidence (p = 0.05), and proper degrees of freedom. The
positive value of the significant effect of interaction between the temperature and Au:Pd
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201900512
molar ratio indicates that the maximization of the response is obtained when both
factors are simultaneously moved to higher levels.
Time of reaction and pressure showed no statistical significance on the benzyl
benzoate yield, i.e., their modifications between the studied levels were not able to
promote a significant change in the response. However, for further investigations, as the
model suggests that the yield is increased when the parameters are set in higher levels, it
is reasonable to maintain the pressure at 5 bar and the reaction time at 2.5 h. In this
context, a central composite design was proposed to obtain a quadratic model, relating
the significant factors and the benzyl benzoate yield. For this, experiments were
performed with additional levels for temperature (71.7 and 128.3 °C) and Au:Pd molar
ratio (1:0 and 1:1.6), consisting of axial points and three replicates at the central point
(see Table 3). The replicates at the central points provide a new level and are useful to
access the experimental variation used in the validation of the quadratic model by
ANOVA. The factorial part of the central composite design corresponds to the
experiments #8, #7, #15 and #16 in Table 3. The additional data for the central
composite design are presented in Table 4.
As can be seen in Table 4, the levels for the Au:Pd molar ratio are slightly
different from ±√2 (±1.41), which consists with the value for rotatability and
orthogonality (designated as alpha or α), when the factorial part of the design is a full
one, and it has two factors. An experimental design is rotatable if the variance of its
estimations depends only on the distance from the central points, i.e., if the precision of
the estimated response is equal for all points situated on a circumference centered at the
center of the design (for designs with more factors, it could be a sphere or a
hypersphere). Also, for experimental designs, it can be said that two factors are
orthogonal to each other when they vary independently. In other words, the factor level
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201900512
settings for two factors are uncorrelated. Thus, the effects are estimated unconfounded
with the other ones. However, rotatable or orthogonal criteria may not be strictly
followed due to practical constraints on the design region, i.e., a particular value may
not be feasible, and a neighboring value may have to be chosen[41]. The set values
(±1.14) correspond to the Au:Pd molar ratio equal to 1:0 and 1:1.6. These values were
defined since initially the Au:Pd molar ratio of 1:0.1 was set at the level equal to -1.
Thus, the code equal to ꢖ√2 would bring the new ratio to a logically impossible
negative value. Therefore, the proposed experimental design, using coded levels for
Au:Pd molar ratio as ±1.14, is near-rotatable and orthogonal, owing to the small
difference of the value to reach the conditions of rotability and orthogonality.
The central composite design resulted in the quadratic model presented in
Equation 4,
2
1
Y= 4.08(± 1.65)+16.36(± 1.03)X +9.66(± 1.19) X +18.91(± 1.13)X
1
4
2
4
+
9.02(± 1.60) X +17.78(± 1.46) X X
Equation 4
1
4
where Y stands for the benzyl benzoate yield, X
1
and X stand for the Au:Pd molar ratio
4
and temperature (as shown in Table 1), X
1
4
X stands for the interaction between the
factors.
In Equation 4, the intervals of confidence at 95% of probability are presented
between parentheses for each parameter. The quadratic model presented a coefficient of
2
2
determination (R ) and adjusted coefficient of determination (adj-R ) of 0.96 and 0.92,
respectively, indicating that the factors can explain a large part of the variation around
the global mean and, also, that the model does not overfit the data set. ANOVA of the
model is shown in Table 5, where the significance of the parameters is evaluated. All
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201900512
parameters are significant at 95% of confidence level, with p-values lower than 0.05.
Therefore, the variation explained by the model is significantly higher than the residual
variation; i.e., the regression is significant. On another hand, the quadratic model
presented a small lack of fit, since the F value obtained was 192.3 (although close to the
critical F value). Despite this, only slight variations were observed between predicted
and experimental values, as one can see in Fig. 3B, which shows the plot for predicted
versus experimental benzyl benzoate yields. Furthermore, it was observed that the
distribution of the residuals is closely normal, as presented in the residuals plot of Fig.
3
C and the normal probability plot in Fig. 3D.
Through the quadratic model, the response surface was obtained, and it is shown
in Fig. 4A. The response surface shows an increase in the benzyl benzoate yields when
both factors, temperature and Au-Pd molar ratio, are moved simultaneously to the
higher levels. Thus, it would be reasonable to perform additional optimization around
the higher levels of the factors. Nevertheless, the increasing of the temperature, linked
to the pressure of 5 bar, could have damaged the glass Fisher-Porter reactor, since it is
used for small-scale reactions, which was the case; however, the bottle can stand much
lower pressures than that in a metal reactor. Hence, it hampered the employed higher
level for such a variable. Therefore, it was possible to assess the conditions for the study
considering the maintenance of 5 bar of pressure, achieving 120°C, to prevent critical
issues. In this scenario, the last optimization was performed, under the mentioned
conditions, evaluating the increasing of the Au:Pd molar ratio for the benzyl benzoate
yield. For this, replicated experiments were performed for Au-Pd molar ratios of 1:1.5,
1
:2 and 1:4. The results are shown in Table S2.
One-way ANOVA indicated that the means of the benzyl benzoate yields at the
Au:Pd molar ratio levels (groups) did not present significant difference at 95%
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201900512
confidence level. The calculated F-value was 1.30, and the p-value was 0.34; thus the
between-level variation is statistically equal to within-level variation. Basically, there is
no difference by using the catalyst in the proportion of 1:1.5 and 1:2 or 1:1.5 and 1:4.
As a consequence, we decided to use the 1:1.5 molar ratio. Therefore, under the
experimental range studied here, the best conditions for obtaining benzyl benzoate
yields were: 5 bar of O , 2.5h of reaction, Au:Pd molar ratio of 1:1.5 and 120°C,
2
ensuring integrity conditions for the system and the possible highest yield.
The full factorial designs consisting of 16 experimental runs (Table 3) for
2
benzaldehyde yields as response showed a coefficient of determination (R ) of 0.72, i.e.,
the independent variables (factors) were able to explain 72% of the variation on the
dependent variable around the global mean. Among the investigated variables, only the
Au:Pd molar ratio showed a marginally significant effect on the benzaldehyde yield,
accordingly to the Pareto Chart of standardized effects in Fig. 4B.
The positive value of the significant effect of the Au:Pd molar ratio indicates that
the maximization of the response is obtained when this factor is moved to higher levels.
The factors – reaction time, pressure and temperature – showed no statistical
significance on the benzaldehyde yield. On the other hand, the model suggested that the
yield is increased when these parameters are set in the higher, lower and higher levels,
respectively. Also, through the Pareto chart in Fig. 4B, it can be seen that the time of
reaction is more important than the pressure and temperature for the response.
To access the significance of the Au:Pd molar ratio on the cited yield, one-way
ANOVA was performed on replicated experiments with Au:Pd molar ratios of 1:1.5,
1
1
:2 and 1:4. As pointed out by the full factorial design, the remain variables were set as
bar, 2.5 h, and 120 °C. The results are shown in Table S3.
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201900512
ANOVA indicated that the means of the benzaldehyde yield at the Au:Pd molar
ratio levels (groups) presented significant difference at 95% confidence level. The
calculated F-value was 40.01, and the p-value was 0.00034; as a result, the between-
level variation is statistically different from within-level variation. Post hoc test of
significant difference between means revealed that the mean of the benzaldehyde yield
at the Au:Pd molar ratio of 1:1.5 was different from the other means at the Au:Pd molar
ratio of 1:2 and 1:4, while these last two means are equal between them (Tukey’s test, p
<
0.05). In other words, there is no difference on choosing Au:Pd molar ratio of 1:2 or
1
:4, as a condition for the production of benzaldehyde, but there is a difference on
choosing Au:Pd molar ratio of 1:1.5. Consequently, the benzaldehyde yield mean was
higher at Au:Pd molar ratio of 1:1.5, the chosen optimal molar ratio for benzaldehyde
production. Considering the statistical analyses performed herein, the Au:Pd molar ratio
of 1:1.5 is efficient in terms of reaction yield and selectivity changing by just adjusting
the pressure of the system.
Mechanistic considerations
Inferences about the mechanism of benzyl alcohol oxidation into benzaldehyde
and benzyl benzoate were based on the present reaction conditions used for the
optimization of our catalyst. Such a system is complex since it encompasses gas, liquid
(the substrate) and interaction between the NPs and the catalyst support; due to the
possibility of multiple reaction pathways and some uncertainty, the observations are
somehow limited. However, the dense experimental data herein presented and the
theoretical background from the literature made us believe that we were able to present
some insights into the reaction mechanism for the proposed catalyst.
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201900512
The multivariate study enabled the analysis of the interaction among the chosen
variables for the system. Considering the limitations of the glass reactor, the suggested
model indicated that the significant variables for benzyl benzoate production were
temperature and Au:Pd molar ratio, while just the metals molar ratio presented a
significant effect on the selectivity for benzaldehyde. Thus, the optimized model
indicated that maintaining unchanged all the variables of the system (120°C, 2.5h, and
Au:Pd molar ratio of 1:1.5), except one (pressure), the catalyst would be classified as
selectivity-switchable one. Such a feature is highly desirable once chemoselectivity is a
challenge for catalytic systems. The model showed that the unique feature for a
selectivity-switchable system is pressure: 1 bar is enough for the benzaldehyde
obtaining; while 5 bar of O pressure were proved to have the best effect on the benzyl
2
benzoate production. Such findings were essential for mechanism comments.
Nepak and Srinivas have shown that the basicity of the support for Au-based
catalysts influences the electronic properties and oxidation activity of the catalyst, and
presented a strontium-modified titanate nanotube as one of the most prominent systems
among alkali and alkaline earth metal ions,[42] showing the importance on the support
choice. Our outcome related to the Au:Pd molar ratio of 1:1.5 is similar to the observed
by Sarina et al., which found that the Au:Pd molar ratio of 1:1.86 was the optimal
charge heterogeneity for the oxidation of benzyl alcohol.[43] Such difference may be
explained by the study-type approach: we are performing multivariate analysis, and the
cited study applies a univariate methodology, without considering the possible
interaction among the variables of the system.
Blank experiments have shown that negligible activity was observed over the
support itself, which confirms that the catalytic activity is due to the Au-Pd NPs, which
is in accordance to the DFT studies performed by Cui et al.[7c] They have shown that the
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201900512
electronegativity difference between Pd and Au bears electron-rich and slightly
positively charged sites. Such heterogeneity is the reason for the improved catalytic
activity observed for Au-Pd systems,[44] which can explain why our catalyst does not
present the necessity of having additives, such as external base addition, to present
remarkable performances. However, the intrinsic basicity of the support needs to be
considered since it plays an important role for the catalytic performance, being the
reason for the pursuing of a one-phase strontium compound from the initial synthesis of
the catalyst. Our system is following the literature since the Au-Pd systems limit the
formation of toluene, hindering the disproportionation pathway needed for its
formation.[22] Also, by similarity with carbocation chemistry, the toluene formation is
likely not to be formed under basic conditions, where O–H cleavage is expected to be
preferred instead of the cleavage of the C–O bond of benzyl alcohol. Consequently, it
appears feasible that the appropriate catalytic sites are located close to the edge of the
metal particles where there are metal-support interactions. However, considering that
high pressure favors the benzyl benzoate production and that low pressures produce
benzaldehyde, we believe that the O concentration may lead to different reaction
2
mechanism for the formation of both products. It has been proposed that the differences
in the selectivity for Au-Pd systems can be explained by a decrease in oxygen
adsorption on their surfaces.[45]
In light of the foregoing, we can comment on the reaction mechanism based on
the experimental data we have presented. Fig. 5 shows the reaction mechanism for the
benzaldehyde formation, illustrating that the reaction proceeds with the synergy
between the Au-Pd NPs and the basic sites on the surface of the SrCO (I). In the first
3
2
+
step, benzyl alcohol interacts with the basic site Sr of the support (II), causing an
-
abstraction of the proton by the O of the carbonate, which provides an alkoxide
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201900512
intermediate (III) in the interface of the Au-Pd alloy and the surface of the support. In
the second step (IV), we propose that the intermediate goes through coordination with
the Au-Pd alloy to form an unstable metal-H bond (metal-alcoholate). This intermediate
undergoes β-hydride elimination, which forms metal-hydride species and releases the
product (benzaldehyde). Then, the electron-rich surface of the alloy activates the
molecular oxygen and produce activated oxygen species, which removes the H from the
alloy surface, producing water and oxygen as a by-product and restoring the catalyst for
a new catalytic cycle (V).[20]
Fig. 6 shows the reaction mechanism for the benzyl benzoate formation, which
applies the same idea from the previous mechanism, i.e., the synergy between the Au-
Pd NPs and the basic sites of the support (I). However, it appears that the basic sites
have little impact on the selectivity once the pressure increasing led to a selectivity
change. Thus, the proposed mechanism for benzyl benzoate formation involves oxygen
species that are far from just an H-abstraction process, as seen in the benzaldehyde
2
+
production mechanism. To start, benzyl alcohol interacts with the basic site Sr of the
support (II), forming an alkoxide intermediate (III), where the oxygen species
interacting with alloy directly attacking the –CH –. Such a mechanism provides
2
coordination of the substrate moiety with the alloy and metal-H bond formations,
eliminating water. The coordinated substrate forms a carbonyloxyl intermediate (IV).
The next step counts with the interaction of a new benzyl alcohol molecule with a
neighbor basic site, producing the benzoate (V), and restoring the catalyst.[45]
Recycling experiments
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201900512
The previous section presents the versatility of the catalysts; however, as the
exploitation of the support was an important feature to be pursued, recycling
experiments were proposed to evaluate the stability of the catalyst. As shown in Fig. 7,
the catalyst kept the high activity in 5 runs, leading to conversions of more than 90% in
just 2.5 h, using the optimized conditions for benzaldehyde formation (the results are
presented in yield for benzaldehyde). The catalyst did not need base addition to keep its
performance; in addition, from one cycle to the other, it was just washed with CH
2
Cl
2
and dried in an oven for 2 h. This result suggests that no deactivating processes took
place and that the catalyst can be used in additional runs. The yield presented
maintenance as well, which could be switched to benzoate, following the data presented
before.
th
The elemental mapping for the spent catalyst (Fig. S4), after its 5 run, presented
mean diameter for metal NPs of 4.64 ± 1.42 nm (Fig. S2), i.e., no aggregation was
observed when compared to the as-prepared catalyst. Also, FAAS analyses presented no
metal leaching in the successive cycles. These results suggest that the catalyst can have
industrial potential considering the straightforward catalyst usage.
Conclusions
Au-Pd NPs were supported effectively on SrCO . The catalysts were
3
characterized by different techniques to guarantee the support phase composition and
the oxidation of the metals. The catalytic activity of the supported catalyst was studied
in the oxidation of benzyl alcohol using oxygen as oxidant under solvent-free
conditions. The experimental designs suggested that the temperature and Au:Pd molar
ratio were the most significant effects for the benzyl benzoate production, while just the
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201900512
metal molar ratio presented significance for the selectivity of benzaldehyde. Using the
optimum conditions of 120°C, 2.5 h, and Au:Pd molar ratio of 1:1.5, only the pressure
was changed for a selectivity switch. The optimization methodology combined with the
characterization techniques allowed some mechanisms insights, suggesting a dual
mechanism depending on the O
without loss of performance.
2
pressure. The catalyst was stable and used 6 times
Acknowledgments
The authors acknowledge financial support from FAPEPI, Capes, and CNPq.
Thanks to CCMC/CDMF (FAPESP No. 2013/07296-2, Brazil) for the XPS
measurements. The authors also acknowledge Instituto de Pesquisas Energéticas e
Nucleares, IPEN-CNEN, for the TPR analyses and Dra. Liane M. Rossi for the atomic
absorption analyses.
References
[
1]
a) W. J. Stark, P. R. Stoessel, W. Wohlleben, A. Hafner, Chem. Soc. Rev. 2015, 44, 5793-
805; b) L. M. Dias Ribeiro de Sousa Martins, S. A. C. Carabineiro, J. Wang, B. G. M.
Rocha, F. J. Maldonado‐Hódar, A. J. Latourrette de Oliveira Pombeiro, ChemCatChem
017, 9, 1211-1221.
5
2
[
[
[
2]
3]
4]
M. Haruta, T. Kobayashi, H. Sano, N. Yamada, Chem. Lett. 1987, 16, 405-408.
G. J. Hutchings, J. Catal. 1985, 96, 292-295.
a) H. Alshammari, M. Alhumaimess, M. H. Alotaibi, A. S. Alshammari, Journal of King
Saud University-Science 2017, 29, 561-566; b) C.-H. Tsai, M. Xu, P. Kunal, B. G. Trewyn,
Catal. Today 2018, 306, 81-88.
[
[
5]
6]
a) B. Dragoi, I. Mazilu, A. Chirieac, C. Ciotonea, A. Ungureanu, E. Marceau, E. Dumitriu,
S. Royer, Catal. Sci. Technol. 2017, 7, 5376-5385; b) X. Feng, N. Sheng, Y. Liu, X. Chen,
D. Chen, C. Yang, X. Zhou, ACS Catal. 2017, 7, 2668-2675; c) Z. Lu, M. Piernavieja-
Hermida, C. H. Turner, Z. Wu, Y. Lei, J. Phys. Chem. C 2018, 122, 1688-1698.
a) R. Güttel, M. Paul, F. Schüth, Catal. Sci. Technol. 2011, 1, 65-68; b) F. Kettemann, S.
Witte, A. Birnbaum, B. Paul, G. Clavel, N. Pinna, K. Rademann, R. Kraehnert, J. r. Polte,
ACS Catal. 2017, 7, 8247-8254; c) J. C. Costa, R. V. Gonçalves, É. Teixeira‐Neto, L. M.
Rossi, ChemistrySelect 2017, 2, 660-664; d) A. Teixeira-Neto, R. V. Gonçalves, C.
Rodella, L. M. Rossi, E. Teixeira-Neto, Catal. Sci. Technol. 2017, 7, 1679-1689; e) G. M.
Mullen, L. Zhang, E. J. Evans, T. Yan, G. Henkelman, C. B. Mullins, PCCP 2015, 17, 4730-
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201900512
4
738; f) P. R. Castro, A. S. Garcia, W. C de Abreu, A. A. de Sousa, V. R. de Moura, C. S.
Costa, E. M de Moura, Catalysts 2018, 8, 83; g) W. C. de Abreu, M. A. Garcia, S.
Nicolodi, C. V. de Moura, E. M. de Moura, RSC Adv. 2018, 8, 3903-3909.
a) K. Luo, L. Zhang, H. Jiang, L. Chen, S. Zhu, Chem. Commun. 2018, 54, 1893-1896; b) S.
A. Shahzad, M. A. Sajid, Z. A. Khan, D. Canseco-Gonzalez, Synth. Commun. 2017, 47,
[
7]
8]
7
2
35-755; c) W. Cui, Q. Xiao, S. Sarina, W. Ao, M. Xie, H. Zhu, Z. Bao, Catal. Today 2014,
35, 152-159; d) J. Zhao, R. Jin, Nano Today 2018, 18, 86-102.
[
a) D. Shi, J. Liu, R. Sun, S. Ji, S. M. Rogers, B. M. Connolly, N. Dimitratos, A. E. Wheatley,
Catal. Today 2018, 316, 206-213; b) D. Nepak, S. Darbha, Catal. Commun. 2015, 58,
1
49-153; c) Y. Zhang, F. Gao, M.-L. Fu, Chem. Phys. Lett. 2018, 691, 61-67; d) M.
Khawaji, D. Chadwick, ChemCatChem 2017, 9, 4353-4363; e) R. Tan, W. Zhang, Z. Xiao,
J. Wang, W. Fu, D. Yin, ChemCatChem.
[
[
[
9]
10]
11]
R. J. Gritter, T. J. Wallace, J. Org. Chem. 1959, 24, 1051-1056.
J. Muzart, Chem. Rev. 1992, 92, 113-140.
a) X. Yang, L. T. Roling, M. Vara, A. O. Elnabawy, M. Zhao, Z. D. Hood, S. Bao, M.
Mavrikakis, Y. Xia, Nano Lett. 2016, 16, 6644-6649; b) C. J. Serpell, J. Cookson, D.
Ozkaya, P. D. Beer, Nat. Chem. 2011, 3, 478-483; c) X. Huang, Z. Zhao, L. Cao, Y. Chen,
E. Zhu, Z. Lin, M. Li, A. Yan, A. Zettl, Y. M. Wang, Science 2015, 348, 1230-1234.
a) G. J. Hutchings, Chem. Commun. 2008, 1148-1164; b) J. A. Lopez-Sanchez, N.
Dimitratos, N. Glanville, L. Kesavan, C. Hammond, J. K. Edwards, A. F. Carley, C. J. Kiely,
G. J. Hutchings, App. Catal. A 2011, 391, 400-406.
[
[
[
[
12]
13]
14]
15]
a) T. A. Silva, E. Teixeira-Neto, N. López, L. M. Rossi, Sci. Rep. 2014, 4, 5766; b) E. M. de
Moura, M. A. Garcia, R. V. Gonçalves, P. K. Kiyohara, R. F. Jardim, L. M. Rossi, RSC Adv.
2
015, 5, 15035-15041.
a) Y.-M. Wang, S. L. Buchwald, J. Am. Chem. Soc. 2016, 138, 5024-5027; b) C.
Weatherly, J. M. Alderson, J. F. Berry, J. E. Hein, J. M. Schomaker, Organometallics
2
017, 36, 1649-1661.
a) S. Karanjit, A. Jinasan, E. Samsook, R. N. Dhital, K. Motomiya, Y. Sato, K. Tohji, H.
Sakurai, Chem. Commun. 2015, 51, 12724-12727; b) C. Shen, Z. Wei, H. Jiao, X. F. Wu,
Chem-Eur J 2017, 23, 13369-13378.
[
[
[
[
16]
17]
18]
19]
a) J. W. Fyfe, A. J. Watson, Synlett 2015, 26, 1139-1144; b) M. Marigo, P. Melchiorre,
ChemCatChem 2010, 2, 621-623.
a) C. Parmeggiani, C. Matassini, F. Cardona, Green Chem. 2017, 19, 2030-2050; b) T.
Mallat, A. Baiker, Chem. Rev. 2004, 104, 3037-3058.
C. M. Olmos, L. E. Chinchilla, A. Villa, J. J. Delgado, H. Pan, A. B. Hungría, G. Blanco, J. J.
Calvino, L. Prati, X. Chen, App. Catal. A 2016, 525, 145-157.
a) J. A. Gualteros, M. A. Garcia, A. G. da Silva, T. S. Rodrigues, E. G. Cândido, F. A. e
Silva, F. C. Fonseca, J. Quiroz, D. C. de Oliveira, S. I. C. de Torresi, J. Mater. Sci. 2019, 54,
2
8
38-251; b) S. Oh, Y. K. Kim, C. H. Jung, W. H. Doh, J. Y. Park, Chem. Commun. 2018, 54,
139-8282.
[
[
20]
21]
D. Nepak, D. Srinivas, RSC Adv. 2015, 5, 47740-47748.
L. Wang, H. Guo, X. Chen, Q. Chen, X. Wei, Y. Dinga, B. Zhua, Catal. Sci. Technol. 2015,
5
, 3340-3351.
[
22]
P. J. Miedziak, Q. He, J. K. Edwards, S. H. Taylor, D. W. Knight, B. Tarbit, C. J. Kiely, G. J.
Hutchings, Catal. Today 2011, 163, 47-54.
[
[
23]
24]
S. Tauster, Acc. Chem. Res. 1987, 20, 389-394.
P. Lu, X. Hu, Y. Li, M. Zhang, X. Liu, Y. He, F. Dong, M. Fu, Z. Zhang, RSC Adv. 2018, 8,
6
315-6325.
[
[
25]
26]
G. Blaise, C. Le Gressus, AIP Adv. 2018, 8, 095228.
P. Ptáček, E. Bartoníčková, J. Švec, T. Opravil, F. Šoukal, F. Frajkorová, Ceram. Int. 2015,
4
1, 115-126.
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201900512
[
[
27]
28]
a) M. A. Alavi, A. Morsali, Ultrason. Sonochem. 2010, 17, 132-138; b) B. E. ScHrrrz, W.
B. WHtrE, Am. Mineral. 1977, 62, 36-50; c) B. Sreedhar, M. Sulochana, C. S. Vani, D. K.
Devi, N. S. Naidu, Eur. J. Org. Chem. 2014, 3, 234-239.
a) J. Xu, T. White, P. Li, C. He, J. Yu, W. Yuan, Y.-F. Han, J. Am. Chem. Soc. 2010, 132,
1
0398-10406; b) J. K. Edwards, A. F. Carley, A. A. Herzing, C. J. Kiely, G. J. Hutchings,
Faraday Discuss. 2008, 138, 225-239; c) Y. Mizukoshi, K. Sato, T. J. Konno, N.
Masahashi, Appl. Catal., B 2010, 94, 248-253; d) T. Pasini, M. Piccinini, M. Blosi, R.
Bonelli, S. Albonetti, N. Dimitratos, J. A. Lopez-Sanchez, M. Sankar, Q. He, C. J. Kiely,
Green Chem. 2011, 13, 2091-2099; e) Q. Wang, X. Cui, W. Guan, X. Zhang, C. Liu, T.
Xue, H. Wang, W. Zheng, Microchim. Acta 2014, 181, 373-380.
[
[
29]
30]
C.-H. Cui, J.-W. Yu, H.-H. Li, M.-R. Gao, H.-W. Liang, S.-H. Yu, ACS nano 2011, 5, 4211-
4
218.
a) V. Rico-Pérez, E. Aneggi, A. Trovarelli, Catalysts 2017, 7, 28; b) M. Ghelamallah, S.
Kacimi, P. Granger, Arab. J. Geosci. 2018, 11, 221.
[
[
31]
32]
I. A. Pašti, N. M. Gavrilov, S. V. Mentus, Adv. Phys. Chem. 2011, 2011.
M. G. Sandoval, R. Luna, G. Brizuela, A. O. Pereira, C. Miranda, P. Jasen, J. Phys. Chem.
C 2017, 121, 8613-8622.
[
[
[
33]
34]
35]
Y.-H. Ke, X.-X. Qin, C.-L. Liu, R.-Z. Yang, W.-S. Dong, Catal. Sci. Technol. 2014, 4, 3141-
3
150.
Z.-L. Wang, J.-M. Yan, H.-L. Wang, Y. Ping, Q. Jiang, J. Mater. Chem. A 2013, 1, 12721-
2725.
T. C. Taucher, I. Hehn, O. T. Hofmann, M. Zharnikov, E. Zojer, J. Phys. Chem. C 2016,
20, 3428-3437.
1
1
[
[
36]
37]
S. Dutta, C. Ray, S. Mallick, S. Sarkar, A. Roy, T. Pal, RSC Adv. 2015, 5, 51690-51700.
M. Nerantzaki, M. Filippousi, G. Van Tendeloo, Z. Terzopoulou, D. Bikiaris, O. Goudouri,
R. Detsch, A. Grüenewald, A. Boccaccini, Express Polym. Lett. 2015, 9, 773-789.
Y. Wang, M. K. Bayazit, S. J. Moniz, Q. Ruan, C. C. Lau, N. Martsinovich, J. Tang, Energ.
Environ. Sci. 2017, 10, 1643-1651.
[
38]
[
[
[
39]
40]
41]
A. Shchukarev, D. Korolkov, Open Chem. 2004, 2, 347-362.
J. Sun, X. Tong, Z. Liu, S. Liao, X. Zhuang, S. Xue, Catal. Commun. 2016, 85, 70-74.
L. V. Candioti, M. M. De Zan, M. S. Camara, H. C. Goicoechea, Talanta 2014, 124, 123-
1
38.
[
[
42]
43]
D. Nepak, D. Srinivas, RSC Advances 2015, 5, 47740-47748.
S. Sarina, H. Zhu, E. Jaatinen, Q. Xiao, H. Liu, J. Jia, C. Chen, J. Zhao, J. Am. Chem. Soc.
2
013, 135, 5793-5801.
[
[
44]
45]
W. Tang, G. Henkelman, J. Chem. Phys. 2009, 130, 194504.
A. Savara, C. E. Chan‐Thaw, J. E. Sutton, D. Wang, L. Prati, A. Villa, ChemCatChem 2017,
9
, 253-257.
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201900512
Fig. 1 The morphology of the as-prepared catalyst in the spectrum image scanning (A),
and the STEM-EDS elemental map images of Sr, Au, Pd, and (B, C, D, respectively).
This article is protected by copyright. All rights reserved.

ChemCatChem
10.1002/cctc.201900512
Fig. 2. (A) H
f, (C) Pd 3d, (D), Sr 3d, (E) O 1s, (F) C1s levels. The red line represents the fitted
spectrum, and the black dots are the background.
2
-TPR profiles of SrCO
3
and Au-Pd/SrCO catalyst. XPS spectra of (B) Au
3
4
This article is protected by copyright. All rights reserved.