Homocoupling of Phenylboronic Acid using Atomically Dispersed Gold on Carbon Catalysts: Catalyst Evolution Before Reaction

Article abstract of DOI:10.1002/cctc.201701840

Coupling reactions to form new C?C bonds are extensively used in industrial synthetic processes. Gold has been shown to be an active catalyst for such reactions; however, conflicting reports exist as to whether cationic Au or metallic Au acts as the active species. We prepared a heterogeneous catalyst consisting of atomically dispersed Au–Cl_x supported on carbon and showed this to be active in the homocoupling of phenylboronic acid to biphenyl. However; characterisation of the catalyst materials, even after just a short exposure time to the reactants, revealed rapid reduction and sintering of the Au species into larger metallic nanoparticles, which we propose to be the true active species in this instance. This study suggests that if cationic Au is an active catalyst, it must be stabilised against reduction and agglomeration by either forming complexes which are more stable than common chlorides or by strongly anchoring them firmly onto alternative support materials; as in this case the carbon supported Au–Cl species were easily reduced.

Full text of DOI:10.1002/cctc.201701840

Accepted Article
Title: Homocoupling of Phenylboronic Acid using Atomically Dispersed
Gold on Carbon Catalysts: Catalyst Evolution Before Reaction
Authors: Tanja Parmentier, Simon R Dawson, Grazia Malta, Li Lu,
Thomas E Davies, Simon A. Kondrat, Simon J Freakley,
Christopher J Kiely, and Graham John Hutchings
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.201701840
Link to VoR: http://dx.doi.org/10.1002/cctc.201701840
A Journal of
www.chemcatchem.org

10.1002/cctc.201701840
ChemCatChem
Homocoupling of Phenylboronic Acid using Atomically Dispersed
Gold on Carbon Catalysts: Catalyst Evolution Before Reaction
Tanja E. Parmentiera, Simon R. Dawsona, Grazia Maltaa, Li Lub, Thomas E. Daviesa, Simon
A. Kondrata,c, Simon J. Freakleya, Christopher J. Kielya,b and Graham J. Hutchingsa*
a Cardiff Catalysis Institute, School of Chemistry, Cardiff University, Main Building, Park
Place, Cardiff, CF10 3AT, UK.
b Department of Materials Science and Engineering, Lehigh University, 5 East Packer Avenue,
Bethlehem, Pennsylvania, 18015, USA
c
Department of Chemistry, Loughborough University, Loughborough, Leicestershire, LE11
3TU, UK
* hutch@cardiff.ac.uk
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701840
ChemCatChem
Abstract
Coupling reactions to form new C-C bonds are extensively used in industrial synthetic
processes. Gold has been shown to be an active catalyst for such reactions however, conflicting
reports exist as to whether cationic Au or metallic Au is acting as the active species. We
prepared a heterogeneous catalyst consisting of atomically dispersed Au-Clx supported on
carbon and showed this to be active in the homocoupling of phenylboronic acid to biphenyl.
However; characterisation of the catalyst materials, even after just a short exposure time to the
reactants, revealed rapid reduction and sintering of the Au species into larger metallic
nanoparticles which we propose to be the true active species in this instance. This study
suggests that if cationic Au is an active catalyst, it must be stabilised against reduction and
agglomeration by either forming complexes which are more stable than common chlorides or
by strongly anchoring them firmly onto alternative support materials; as in this case the carbon
supported Au-Cl species were easily reduced.
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701840
ChemCatChem
Coupling reactions to form new C-C bonds are widely used in the pharmaceutical,
agrochemical and fine chemical industries.[1] Suzuki coupling catalysts are typically
homogenous palladium-based catalysts that are used to couple organo-boron compounds with
organic halides or triflates, forming new C-C bonds.[2] One of the most widely used coupling
reactions in an industrial context is the synthesis of the fungicide Boscalid, where Pd-based
catalysts are used in a Suzuki coupling reaction to produce a biaryl intermediate in a one-step
reaction process, rather than the five synthetic steps previously utilised.[1a] The substrate
tolerance and general applicability of these transformations make such coupling catalysts a key
synthetic tool in the production of many high value materials.
Typically the catalytic cycle of the palladium catalysed reaction under basic conditions
involves the oxidative addition of the aryl halide to a Pd(0) centre followed by a transmetalation
of an aryl substituent from a boronate species to generate a di(aryl) Pd(II) species. Reductive
elimination of the di(aryl) species then regenerates the initial Pd(0) centre.[3] Extensive studies
have been conducted on the development of homogeneous Pd catalysts for numerous organic
transformations, however, the final separation of such catalysts from the reaction mixtures
remains a significant challenge. Heterogeneous-type Suzuki coupling catalysts have also been
synthesised by immobilising palladium complexes onto high surface area supports (e.g., silica,
zeolites and activated carbon) thus allowing the catalysts to be more easily separated from the
product by filtration.[4]
Besides palladium-based materials, several other metals have been proposed to be effective
catalysts for these reactions. Corma et al. postulated and subsequently demonstrated that
Au(III) could be active as a Suzuki coupling catalyst due to (i) Au(III) being isoelectronic with
Pd(II) and (ii) the redox couple of Au(III) – Au(I) being isoelectronic with the Pd(II) – Pd(0)
species needed to effect the reaction.[5] Carbon – carbon coupling has also been observed by
Au(I) complexes via transmetallation of Au(I) to Pd(II).[6] Corma and co-workers prepared
gold supported on nanocrystalline ceria and demonstrated that this material was able to catalyse
the homocoupling of phenylboronic acid in the presence of a base. In subsequent FT-IR studies
they were able to demonstrate that the activity of the Au/CeO2 catalyst is directly proportional
to either (i) the concentration of Au(III) species present on the gold nanoparticles / support
interface or (ii) the concentration of highly dispersed cationic Au species stabilised on the CeO2
support material.[7] To demonstrate the catalytic activity of cationic-species, a Au(III) Schiff
base complex was synthesised and shown to be active in this reaction.[8] Au(III) has also been
able to form carbon – carbon bonds by oxidative coupling of vinyl groups in vinylgold
intermediates.[9] In addition, simple gold-containing salts like AuCl and NaAuCl4 have also
reported to be active in the homocoupling of phenylboronic acids.[10] Recently a
photochemical, visible light induced Suzuki-type coupling of arylboronic acid and
aryldiazonium satls was developed.[11] Guzman et al. also showed that metallic gold
nanoparticles supported on both CeO2 and ZrO2 are also active in the homocoupling reaction
[12]
of phenylboronic acid.
The activity of these catalysts has also been correlated to the
crystallite size of the support material, with smaller crystallites being better able to stabilise the
Au(III) species. Hasmi and co-workers also observed homocoupling of boronic acids when
they prepared gold (III) complexes from boronic acids.[13]
These studies collectively suggest the Au(III) species are the active entities for this reaction.
However, more recently multiple studies have claimed otherwise and demonstrated the activity
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701840
ChemCatChem
of gold nanoparticles in the reaction. For instance, PVP-stabilised Au colloids in aqueous
solution have been shown to be active, as have base-free metallic Au na noparticles supported
on Al-Mg oxides.[6b, 14]
Malta et. al. recently demonstrated using in-situ X-ray absorption spectroscopy (XAS) studies
that the active form of the Au/C catalyst used for acetylene hydrochlorination is comprised of
atomically dispersed AuClx species which are thermally stable under high temperature
conditions (i.e., 200 °C).[15] The acetylene hydrochlorination reaction is proposed to occur via
the Au(I) – Au(III) redox cycle as predicted through correlations with standard electrode
potentials, which suggested to us that this catalyst could also be active as a homocoupling
catalyst.[16] In this study, we have investigated this Au/C single site catalyst for the
homocoupling of phenylboronic acid and compared it’s performance to that of a sol-
immobilised Au/C nanoparticle catalyst and a simple gold salt (HAuCl4).
Results & Discussion
The recent discovery that gold cations dispersed on high surface area activated carbon supports
can catalyse the gas phase acetylene hydrochlorination reaction at high temperatures suggested
that not only can single atom catalysts operate and retain an atomic dispersion under harsh
reaction conditions, but that they could become technologically significant in some industrial
processes.[15a] Homocoupling to create new C-C bonds is widely used in the synthesis of many
chemical materials, and cationic Au species have been postulated to be for active for such
reactions, either at the interfacial periphery of supported metallic nanoparticles or stabilised as
isolated species on oxide materials. We have prepared 1wt% Au/C catalysts by the
immobilisation of colloidal metal Au nanoparticles and through the impregnation of Au from
aqua regia (to generate atomically dispersed AuClx species) using a high surface area activated
carbon material as catalyst support.
X-ray diffraction (XRD) patterns of the catalysts prepared by sol immobilisation, (1% Au/C
(sol)) and with aqua regia (1% Au/C (AR)), are shown in Figure 1(a). In the case of the 1%
Au/C (sol) catalyst, where polymer stabilised metallic nanoparticles existed in solution prior to
immobilisation, a clear diffraction pattern consistent with f.c.c. Au (ICCD 01-071-4164) was
obtained, with the principal reflection at 38 °(2Q), corresponding to the set of Au {111} planes.
Analysis using the Scherrer equation suggested an average crystallite size of 6 nm in this (1%
Au/C (sol)) sample. In contrast, the fresh 1% Au/C (AR) sample showed no features consistent
with the presence of Au nanocrystals implying a high dispersion of Au existed in this catalyst.
Figure 1(b) however demonstrated that heat treatment of the 1% Au/C (AR) catalyst, under a
nitrogen atmosphere (in order to prevent combustion of the carbon support material), during
an in-situ XRD experiment eventually caused sintering of the Au species into discrete
detectable metallic nanoparticles. Figure 1(b) demonstrates that no diffraction peaks
characteristic of f.c.c Au were observed until the temperature of the reaction cell reached 310
°C. This implies that the Au species in the 1% Au/C (AR) sample have a tendency to remain
in a highly dispersed state even at elevated temperatures up to ~300 °C. When the temperature
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701840
ChemCatChem
was increased above 310 °C, the evolution of more distinct diffraction patterns showed the
growth of these Au crystallites through thermally activated mobility of the Au species.
Transmission electron microscopy (TEM) analysis of the 1% Au/C (sol), Figure 2(a), revealed
homogeneous dispersion of Au nanoparticles on the activated carbon support. Analysis of the
particle size distribution from such micrographs (Figure 2(b)) gave a mean particle size of 6.6
nm with a spread of about ± 3nm, which is consistent with the value derived from XRD
measurements. This particle size distribution is comparable with the previously reported Au
nanoparticle catalysts prepared for this reaction which were supported on CeO2 and ZrO2.[12]
However, in direct contrast to these materials which were reported to contain substantial
populations of Aud+ stabilised by the oxide support, X-ray photoemission spectroscopy (XPS,
Figure 2(c)) of our 1% Au/C (sol) catalyst showed it to consist of exclusively metallic Au with
a characteristic binding energy of 84 eV. This difference in Au speciation is probably due to
the catalyst being prepared by a sol-immobilisation method in the presence of a strong
reductant, NaBH4, which facilitates complete reduction of the HAuCl4 precursor. STEM-
HAADF analysis of the fresh 1% Au/C (AR) catalyst, (Figure 3), revealed the presence of a
near complete atomic dispersion of Au species on the carbon support, once again consistent
with our deductions from XRD analysis. XPS analysis of this 1% Au/C (AR) sample also
suggested the presence of metallic Au. However, it is now well recognised that Au-Clx species
supported on carbon can be susceptible to reduction by the X-ray flux used in XPS analysis
resulting in the generation of Au clusters and nanoparticles from highly dispersed cationic
species.[15a, 16-17]
X-ray absorption spectroscopy (XAS) was carried out on the fresh 1% Au/C (AR) catalyst to
verify that the fresh catalyst contained only highly dispersed cationic Au species rather than
the metallic Au species detected by XPS analysis. The normalised Au L3- edge X-ray
absorption near edge structure (XANES), shown in Figure 4(a), supported the HAADF-STEM
analysis and confirmed that this catalyst consisted of highly dispersed cationic Au species. The
presence of a white line absorption feature (indicative of partial 5d band occupancy) having a
normalised intensity value of 0.76 when compared to appropriate standards (protocol reported
in [15a] and used for the analysis presented in this study) suggested that the catalyst comprised
of a mixture Au(I)/Au(III) species with no features of the post-edge spectrum being consistent
with bulk Au. In addition, significant intensity in the Fourier transform (FT) corresponding to
references for Au-Cl type species can be seen in the extended X-ray absorption fine structure
(EXAFS) presented in Figure 4(b). These findings are consistent with our previous reports on
the preparation of active atomically dispersed Au/C catalysts for the acetylene
hydrochlorination reaction. [15a]
Using these catalysts, we conducted homocoupling reactions using phenylboronic acid in the
presence of base to produce biphenyl as the major coupling product. These reactions were
performed in air, the oxygen present in the air is might be a stoichiometric oxidant during the
reaction. It should be noted that phenol was produced as the minor product in all reactions as a
result of oxidation of the aryl boronate species in solution. Figure 5(a) shows the conversion
versus time-on-line profiles for the 1% Au/C (AR) and 1% Au/C (sol) heterogeneous reactions
and the HAuCl4 homogeneous reaction. Each reaction was performed with an equal amount of
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701840
ChemCatChem
gold (0.005 mmol). The 1% Au/C (sol) catalyst showed minimal conversion (2-3%) of
phenylboronic acid throughout the 6 h reaction period. While many reports claim that oxide
supported nanoparticles are active for this reaction, in this particular case the PVA stabilised
Au particles dispersed on C were inactive despite having a mean particle size comparable to
those in these other studies.[12, 14a, 14b, 18] As in many examples of catalysis using polymer
stabilised nanoparticles, it is known that the presence of the ligand can hinder reactivity through
blocking of the active sites. In addition, other studies utilised wet impregnation methods to
produce nanoparticulate Au catalysts that were free of polymer stabilisers. In this case, with
PVA being sparingly soluble in the reaction solvent, it is highly likely that the catalytic surface
is completely blocked by the PVA polymer residing on the nanoparticles after preparation,
especially as there is precedent in the literature for activity in the case of PVP-stabilised
colloidal Au which demonstrates that Au nanoparticles in this size range can in some instance
be active.[18]
The catalyst prepared using aqua regia (1% Au/C (AR)), which is initially atomically dispersed,
shows good activity for the homocoupling of phenylboronic acid reaction, reaching 25%
conversion after 6 h. The progressive increase of conversion value with increasing reaction
time also suggests that there is no catalyst deactivation occurring. In common with the 1%
Au/C (AR) catalyst, the homogeneous reaction carried out using HAuCl4 also reached 24%
conversion after 6 h of reaction. However, it should be noted that the conversion tends to
plateau over the course of the reaction in this latter case. Figure 5(b) summarises the selectivity
values obtained for each of the catalysts system. The 1% Au/C (sol) catalyst produced only
traces of biphenyl at the low conversions achieved, and displayed a meagre 4% selectivity,
confirming the proposed mechanism in the literature that phenol production is not a consecutive
reaction of biphenyl. Reactions using biphenyl as the reactant resulted in no significant phenol
production. The 1% Au/C (AR) catalyst produced biphenyl as the major product at 60.5 %
selectivity irrespective of conversion levels, along with a significant amount of phenol,
typically 39.5%. Due to the limited scope of this study, which was purely to investigate the
effectiveness of this class of catalysts for homocoupling reactions as compared to sol-
immobilised catalysts, no additional experiments were conducted to optimise the potential
yield of biphenyl over phenol.
It has been reported that simple Au-chloride salts are active catalysts for this reaction. In
addition many studies show that metallic Au nanoparticles, even after additional treatment with
NaBH4 to remove any cationic species, can still be active catalysts, especially when dispersed
on basic support materials.[10, 14a] Due to the similar activity levels displayed by the HAuCl4
and the 1% Au/C (AR) catalyst, that latter of which is known to consist of supported Au-Cl
species in the fresh catalyst, we further investigated the stability of our catalyst.[15a]
Characterisation of the 1% Au/C (AR) catalyst after 6 h of reaction by XANES, as shown in
Figure 6(a), revealed a significant evolution of the catalyst structure, which now much more
closely resembled the metallic Au standard. Furthermore, EXAFS analysis shown in Figure
6(b) revealed significant Au-Au interactions when comparing against the observed intensity of
the Fourier transform with that of the metallic Au standard. The Au-Cl FT intensity was greatly
reduced compared to the fresh catalyst (Figure 4(a)) suggesting that the atomic dispersion had
become unstable and undergone significant sintering under reaction conditions. A linear
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701840
ChemCatChem
combination fitting analysis (using KAuCl4, AuCl and metallic Au standards) of the fresh and
used 1% Au/C (AR) samples shown in Figure 7(a) confirmed that the Au was cationic in nature
in the fresh sample and had roughly a 1:1 Au(I) to Au (III) chloride species. TEM analysis,
Figure 7(b), of the 1% Au/C (AR) sample after reaction showed that the catalyst now consisted
of large nanoparticles with a mean particle size of between 30-40 nm (representative image
shown in Figure 7(c)) supporting the XAS analysis. Despite this high degree of sintering, even
after filtering of the reaction solution, only very low amounts of Au (9 ppb) were measured by
ICP-MS in the reaction solution.
A sample of the 1% Au/C (AR) catalyst was recovered from the reaction after 6 h, washed with
EtOH and then dried before being retested. The activity of this sample was re-tested for
durations of 1 and 2 h due to the sample showing no deactivation during the initial reaction
time-on-line. Figure 8 shows that this catalyst showed similar activity upon re-use when
compared to the fresh sample, which suggests that the catalyst nanostructure did not change
significantly during the reaction itself and the transformation from an atomically dispersed to
a nanocrystallline form occurred immediately on exposure to the reactants. It was also observed
during the reaction using the homogeneous HAuCl4 catalyst, that the colour of the reaction
solution changed from yellow to purple suggesting Au reduction from a cationic to metallic
form under the reaction conditions. To confirm if the similar reduction was occurring in the
supported 1% Au/C (AR) catalyst, this catalyst was suspended in a solution containing all the
ingredients for reaction and was filtered after only 5 min of contact at room temperature. SEM
analysis (Figure S1) revealed the presence of Au nanoparticles in the size range observed
previously by TEM in the sample that had been used for 6 h. This coupled with the observation
that the catalyst displayed a constant reaction rate and is re-usable, supports the hypothesis that
it is the initial exposure to the reactants that causes the significant rapid reduction and
aggregation of the atomically dispersed Au species into particulate form.
We propose that previous reports which concerned the use of HAuCl4 and AuCl as catalysts
for the homocoupling of phenylboronic acid, could in fact undergo a similar significant
reduction to metallic nanoparticles either under reaction conditions or even just on mere
exposure to the reactants. Stable Au cations in a system which uses polar solvents to allow
complete dissolution of the base and reactants, will require strong interaction with the support
material. The use of supports such as CeO2, ZrO2 or TiO2 which can stabilise Au cations at the
nanoparticle interface periphery or at surface vacancies sites has been previously reported. So
has the use of Au cationic complexes with high stability constants, such as the Schiff bases, [7,
12, 19]. Such stabilisation strategies are important as the stability of Au-Cl species towards the
reactants is minimal in the case of our 1% Au/C (AR) catalyst. The reduction of the Au-Cl
entities in our case presumably occurs by the oxidation of the aryl boronate species generated
in-situ through reaction of the phenylboronic acid with the added base, which results in phenol
production and the reduction of the Au species. The activity levels of the non-ligand protected
gold nanoparticles produced in this study compare favourably with the activities reported for
the Au/ CeO2 and Au/ZrO2 systems taking into account the differences in particle size and
reaction conditions. To better understand the role of cationic Au in these coupling reactions,
more stable cationic Au complexes are required which display a better stability to the reducing
nature of the boronate species. However, this enhanced stability may come at a cost in terms
of the overall reactivity. Our future studies on this reaction will focus on the use of sulfur-
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701840
ChemCatChem
stabilised Au/C catalysts which have been shown to be especially stable in the acetylene
hydrochlorination reaction.[20]
Conclusions
We prepared an atomically dispersed Au/C catalyst using an aqua regia impregnation method
as a model for supported cationic Au complexes which have been previously used as catalysts
for the homocoupling of phenylboronic acid. This catalyst showed activity in the homocoupling
of phenylboronic acid towards biphenyl and was re-usable. However, detailed structural
characterisation revealed that exposure to the reactants resulted in rapid reduction and
agglomeration of the Au species into nanoparticles. The activity of the reduced Au species was
comparable to that described in previous reports, and our experiments demonstrated that
supported cationic Au-Clx is not stable to the reducing species present in the reaction solution.
To fully elucidate the role of cationic Au more stable complexes need to be produced and
investigated. However, this may reduce the overall activity of the catalysts by restricting the
action of the redox cycle required for the reaction to proceed.
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701840
ChemCatChem
Experimental
The 1 wt% Au/C (AR) catalyst was prepared by a wet impregnation with aqua regia. Powdered
activated carbon (Norit ROX 0.8, 100 - 140 mesh) was used as support. HAuCl4.xH2O (Alfa
Aaesar, 99.9 %) was dissolved in 5.4 mL aqua regia (3 parts by volume [HCl (Fisher, 32
wt.%)]: 1 part by volume [HNO3 (Fisher, 70 wt. %)]) and added dropwise to the activated
carbon while being stirred continuously. The mixture was left to stir for 60 min at room
temperature after which it was placed in a furnace at 140 °C for 16 h under a N2 flow. Sol
immobilisation was used to prepare another Au/C catalyst (designated Au/C (sol)). For this
preparation aqueous solutions of HAuCl4.3H2O ((Alfa Aesar, 99.9 %) and polyvinyl alcohol
(PVA) (1wt%, Aldrich, MW 9000-10000, 80% hydrolysed) were prepared. The desired amount
of these solutions (to make 1wt% Au and Au/PVA w/w=1.2) was added to 800 mL of water.
A fresh solution of NaBH4 (0.1 M, Aldrich, 99.99%) was prepared and 6.6 ml added to the
Au+PVA solution (NaBH4 : Au =5:1). It was then stirred for 30 mins to allow the sol to
generate. The carbon support (Norit ROX 0.8, 100 - 140 mesh, desired amount for 1wt% Au
loaded catalyst) was then added to the sol mixture, together with a few drops of concentrated
H2SO4 (Fisher Laboratory reagent grade, >95 %). After 1 h of stirring the mixture was filtered,
washed thoroughly with distilled water, and left to dry for 16 h at 110 °C.
In a typical catalytic test, methanol (5 mL, Fisher Laboratory reagent grade) was added to a
flask (50 mL) containing phenylboronic acid (0.3 mmol, Sigma Aldrich > 97.0 % HPLC-
grade), potassium carbonate (0.4 mmol) and catalyst (100 mg of 1wt % Au/C). Catalytic testing
was performed using a Radleys carousel magnetic stirrer at 45 °C and 800 rpm. O-xylene was
used as an internal standard. The catalytic activity of gold (III) chloride was tested using a
solution of HAuCl4.3H2O (Alfa Aesar, 99.9 %). The reaction mixture was analysed using
Agilent Technologies 1200 series HPLC equipped with 2 reverse phase columns (Agilent
Poroshell 120, SB-C18, 4.6x150 mm, 2.7 µm) connected in series and a DAD detector
(G1315D - 1260 DAD VL). Acetonitrile (50 %, Fischer, HPL gradient grade) and water (50
%, Fischer, HPL gradient grade) was used at the mobile phase (0.5 mL/min).
Powder X-ray diffraction (XRD) spectra were acquired between 5° and 80° using an X’Pert
Pro PAN Analytical powder diffractometer employing a Cu Kα radiation source operating at
40 keV and 40 mA. Analysis of the spectra obtained was carried out using X’Pert High Score
Plus software. In-situ XRD was performed on a Panalytical X’Pert diffractometer with an
Anton Paar 900K in-situ cell. The XRD spectra were collected at temperatures between 25°C
and 600 °C under an N2 flow (10 mL/min). The sample was heated at a rate of 10 °C/min and
was kept for 5 min at a certain temperature before the spectra were collected.
Materials for examination by scanning transmission electron microscopy (STEM) were dry
dispersed onto a holey carbon TEM grid. These supported fragments were examined using the
HAADF-STEM imaging mode in an aberration corrected JEOL ARM-200CF scanning STEM
operating at 200kV. A JEOL 2100 TEM operating at 200 kV was used to study the particle size
distribution of the Au/C (sol) and spent Au/C (AR) catalyst. These samples were dispersed in
ethanol and a drop of the solution applied onto a holey carbon TEM grid and allowed to
evaporate. ImageJ was used to measure the particle sizes.
XAFS data was collected at the B18 beamline at the Diamond Light Source in Harwell, UK.
The measurements were performed using a QEXAFS set-up with a fast-scanning Si (111)
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701840
ChemCatChem
double crystal monochromator and a 36 element Ge fluorescence detector. Data has been
processed using Athena software.
XPS was carried out using a Thermo Scientific K-alpha photoelectron spectrometer with
monochromatic AlKα radiation. The resulting spectra were processed in CasaXPS and
calibrated against the C(1s) line at 284.7 eV.
SEM imaging on the used catalysts was performed on a Tescan MAIA3 field emission gun
scanning electron microscope (FEG-SEM). Samples were dry dispersed onto 300 mesh Cu
grid supported holey carbon film and imaged using the backscattered electron detector at 15
kV. Samples were uncoated
Acknowledgements
We would like to thank MaxNet Energy for financial support. We would also like to thank the
Cardiff University TEM facility for the electron microscopy.
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701840
ChemCatChem
Figure 1 – X-ray diffraction patterns of a) 1% Au/C prepared by sol-immobilisation (█) and
aqua regia (AR) treatment (█) b) in-situ X-ray diffraction analysis of the 1% Au / C (AR)
catalyst at various elevated temperatures under an inert nitrogen atmosphere.
a)
b)
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701840
ChemCatChem
Figure 2 – a) Bright-field TEM micrograph of the 1% Au/C (sol) catalyst, b) its
corresponding particle size distribution and c) the Au4f region of its XPS spectrum.
a)
b)
c)
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701840
ChemCatChem
Figure 3 - Representative HAADF-STEM images of atomically dispersed Au species on an
aqua regia treated carbon support.
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701840
ChemCatChem
Figure 4 a) Ex-situ Au L3 edge–normalized XANES spectra of the fresh 1% Au/C (AR)
catalyst (█) and a gold-foil reference material (à); b) Fourier transform of k3-weighted χ
EXAFS ex-situ data of the fresh 1% Au/C (AR) catalyst (█) and a gold-foil reference (à).
Variation in magnitude of Fourier transform is plotted as function of distance R from the Au
absorber.
a)
b)
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701840
ChemCatChem
Figure 5 - a) Catalytic activity and b) selectivity data of 1%Au/C (AR) (█), 1%Au/C (sol)
(p) and HAuCl4 (l) for the homocoupling of phenylboronic acid.
a)
b)
Reaction conditions: Phenylboronic acid (0.3 mmol), K2CO3 (0.4 mmol) and catalyst (100
mg of 1wt % Au/C) in 5 mL MeOH. Stirred at 800 rpm at 45 °C. O-xylene was used as an
internal standard.
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701840
ChemCatChem
Figure 6 - a) Ex-situ Au L3 edge–normalized XANES spectra of a used 1% Au/C (AR)
catalyst (█) and a gold-foil reference material (à); b) Fourier transform of k3-weighted χ
EXAFS ex-situ data of the used 1% Au/C (AR) catalyst (█) and a gold-foil reference (à).
Variation in magnitude of Fourier transform is plotted as a function of distance R from the
Au absorber.
a)
b)
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701840
ChemCatChem
Figure 7 - a) Results from linear combination fitting of the signals of the fresh (left) and used
(right) 1% Au/C (AR) catalyst. The different gold phases in this plot are differentiated by the
following keys: █ Au(0), █ Au(I) and █ Au(III). b) BF-TEM micrograph of the 1% Au/C
(AR) catalyst after 6 h reaction and c) its corresponding particle size distribution.
a)
b)
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701840
ChemCatChem
c)
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701840
ChemCatChem
Figure 8 – Conversion versus time-on-line curves for the fresh 1% Au/C (AR) catalyst (█)
and a used 1% Au/C (AR) catalyst (à).
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701840
ChemCatChem
[1]
a) T. J. Colacot, Platinum Metals Review 2011, 55, 84-90; b) V. L. Budarin, P. S. Shuttleworth,
J. H. Clark, R. Luque, Current Organic Synthesis 2010, 7, 614-627; c) C. Torborg, M. Beller,
Advanced Synthesis & Catalysis 2009, 351, 3027-3043; d) X. F. Wu, P. Anbarasan, H.
Neumann, M. Beller, Angew Chem Int Ed Engl 2010, 49, 9047-9050.
A. Suzuki, Journal of Organometallic Chemistry 1999, 576, 147–168.
George B. Smith, G. C. Dezeny, D. L. Hughes, A. King, T. R. Verhoeven, Journal of
Organometallic Chemistry 1994, 59, 8151-8156.
[2]
[3]
[4]
[5]
[6]
M. Mora, C. Jimenez-Sanchidrian, J. R. Ruiz, J Colloid Interface Sci 2006, 302, 568-575.
S. Carrettin, J. Guzman, A. Corma, Angew Chem Int Ed Engl 2005, 44, 2242-2245.
a) A. S. Hashmi, A. M. Schuster, F. Rominger, Angew Chem Int Ed Engl 2009, 48, 8247-8249;
b) A. S. K. Hashmi, R. Döpp, C. Lothschütz, M. Rudolph, D. Riedel, F. Rominger, Advanced
Synthesis & Catalysis 2010, 352, 1307-1314; c) A. S. K. Hashmi, C. Lothschütz, R. Döpp, M.
Ackermann, J. De Buck Becker, M. Rudolph, C. Scholz, F. Rominger, Advanced Synthesis &
Catalysis 2012, 354, 133-147.
[7]
[8]
[9]
S. Carrettin, A. Corma, M. Iglesias, F. Sánchez, Applied Catalysis A: General 2005, 291, 247-
252.
C. Gonzalez-Arellano, A. Corma, M. Iglesias, F. Sanchez, Chem Commun (Camb) 2005, 1990-
1992.
A. S. K. Hashmi, M. C. Blanco, D. Fischer, J. W. Bats, European Journal of Organic Chemistry
2006, 2006, 1387-1389.
[10]
[11]
T. Matsuda, T. Asai, S. Shiose, K. Kato, Tetrahedron Letters 2011, 52, 4779-4781.
a) S. Witzel, J. Xie, M. Rudolph, A. S. K. Hashmi, Advanced Synthesis & Catalysis 2017, 359,
1522-1528; b) J. Xie, M. Rudolph, F. Rominger, A. S. K. Hashmi, Angew Chem Int Ed Engl
2017, 56, 7266-7270; c) J. Xie, J. Li, V. Weingand, M. Rudolph, A. S. Hashmi, Chemistry A
European Journal 2016, 22, 12646-12650.
[12]
[13]
N. G. Willis, J. Guzman, Applied Catalysis A: General 2008, 339, 68-75.
a) A. S. K. Hashmi, T. D. Ramamurthi, F. Rominger, Journal of Organometallic Chemistry 2009,
694, 592-597; b) A. S. K. Hashmi, T. D. Ramamurthi, M. H. Todd, B. A. S.-K. Tsang, K. Graf,
Australian Journal of Chemistry 2010, 63, 1619-1626.
[14]
[15]
a) L. Wang, W. Zhang, D. Sheng Su, X. Meng, F. S. Xiao, Chem Commun (Camb) 2012, 48,
5476-5478; b) C.-H. Liu, C.-Y. Lin, J.-L. Chen, N.-C. Lai, C.-M. Yang, J.-M. Chen, K.-T. Lu, Journal
of Catalysis 2016, 336, 49-57; c) S. K. Movahed, M. Shariatipour, M. Dabiri, RSC Adv. 2015, 5,
33423-33431.
a) G. Malta, S. A. Kondrat, S. J. Freakley, C. J. Davies, L. Lu, S. Dawson, A. Thetford, E. K.
Gibson, D. J. Morgan, W. Jones, P. P. Wells, P. Johnston, C. R. A. Catlow, C. J. Kiely, G. J.
Hutchings, Science 2017, 355, 1399–1403; b) XAS has even been used to characterise the
oxidation state of gold in homogeneous gold-catalysed oxidative esterfication reactions, see
: A. S. Hashmi, C. Lothschutz, M. Ackermann, R. Doepp, S. Anantharaman, B. Marchetti, H.
Bertagnolli, F. Rominger, Chemistry A European Journal 2010, 16, 8012-8019.
G. Malta, S. J. Freakley, S. A. Kondrat, G. J. Hutchings, Chem Commun (Camb) 2017, 53,
11733-11746.
a) Y. Y. Fong, B. R. Visser, J. R. Gascooke, B. C. Cowie, L. Thomsen, G. F. Metha, M. A. Buntine,
H. H. Harris, Langmuir 2011, 27, 8099-8104; b) X. Liu, M. Conte, D. Elias, L. Lu, D. J. Morgan,
S. J. Freakley, P. Johnston, C. J. Kiely, G. J. Hutchings, Catal. Sci. Technol. 2016, 6, 5144-5153.
Hironori Tsunoyama, Hidehiro Sakurai, Nobuyuki Ichikuni, Yuichi Negishi, T. Tsukuda,
Langmuir 2004, 20, 11293-11296.
[16]
[17]
[18]
[19]
[20]
C. Gonzalezarellano, A. Corma, M. Iglesias, F. Sanchez, Journal of Catalysis 2006, 238, 497-
501.
P. Johnston, N. Carthey, G. J. Hutchings, J Am Chem Soc 2015, 137, 14548-14557.
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701840
ChemCatChem
This article is protected by copyright. All rights reserved.