Heterogeneous amination of bromobenzene over titania-supported gold catalysts

Article abstract of DOI:10.1016/j.jcat.2012.09.002

Amination of bromobenzene with aniline was carried out on heterogeneous Au-TiO₂ catalysts prepared using the deposition-precipitation method. The deposition of gold in loadings in the range 0.1-0.9 wt.% on titania occurred with the formation of gold particles with controlled size and oxidation state as demonstrated from TEM and XPS measurements. In addition, Raman and DR-UV-Vis investigation of these catalysts showed that the interaction of Au particles with the support stabilized the metallic state of gold. Catalytic results demonstrated that the conversion and selectivity depended on the gold-particle size and dispersion, while the mechanism of reaction was in direct relation with the reaction medium. Working in DMC, the solvent was playing the role of base. The advantage of working in DMC was avoiding the formation of KBr which is very difficult to be separated from the reaction products. HBr resulted as a by-product formed adducts with the coupling product, which was separated as crystals at room temperature by a simple filtration. However, in DMC, the reaction occurred with low selectivities. In dioxane, the presence of potassium ethoxide as base was mandatory. The reaction occurred with very high selectivities for moderate conversions. The conversion was limited by KBr released from the reaction. However, the use of the heterogeneous catalyst has the advantage of scavenger, since KBr becomes chemisorbed on the surface and could be simple removed by percolation and the catalyst can be reused.

Full text of DOI:10.1016/j.jcat.2012.09.002

Journal of Catalysis 296 (2012) 43–54
Contents lists available at SciVerse ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
Heterogeneous amination of bromobenzene over titania-supported gold catalysts
a
a
b
b
a
Madalina Ciobanu , Bogdan Cojocaru , Cristian Teodorescu , Florin Vasiliu , Simona M. Coman ,
c,
⇑
a,
⇑
Walter Leitner , Vasile I. Parvulescu
a
Department of Organic Chemistry, Biochemistry and Catalysis, Faculty of Chemistry, University of Bucharest, Bdul Regina Elisabeta, 4-12, Bucharest 030016, Romania
National Institute of Materials Physics, Atomistilor 105b, 077125 Magurele-Ilfov, Romania
RWTH Aachen, Lehrstuhl fur Technische Chemie und Petrolchemie, Worringerweg 1, 52056 Aachen, Germany
b
c
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 18 July 2012
Revised 5 September 2012
Accepted 5 September 2012
Available online 23 October 2012
Amination of bromobenzene with aniline was carried out on heterogeneous Au–TiO2 catalysts prepared
using the deposition–precipitation method. The deposition of gold in loadings in the range 0.1–0.9 wt.%
on titania occurred with the formation of gold particles with controlled size and oxidation state as dem-
onstrated from TEM and XPS measurements. In addition, Raman and DR–UV–Vis investigation of these
catalysts showed that the interaction of Au particles with the support stabilized the metallic state of gold.
Catalytic results demonstrated that the conversion and selectivity depended on the gold-particle size and
dispersion, while the mechanism of reaction was in direct relation with the reaction medium. Working in
DMC, the solvent was playing the role of base. The advantage of working in DMC was avoiding the for-
mation of KBr which is very difficult to be separated from the reaction products. HBr resulted as a by-
product formed adducts with the coupling product, which was separated as crystals at room temperature
by a simple filtration. However, in DMC, the reaction occurred with low selectivities. In dioxane, the pres-
ence of potassium ethoxide as base was mandatory. The reaction occurred with very high selectivities for
moderate conversions. The conversion was limited by KBr released from the reaction. However, the use of
the heterogeneous catalyst has the advantage of scavenger, since KBr becomes chemisorbed on the sur-
face and could be simple removed by percolation and the catalyst can be reused.
Keywords:
Heterogeneous catalysis
Gold particles
Titania
Precipitation–deposition technique
Amination of bromobenzene
Aniline
Ó 2012 Elsevier Inc. All rights reserved.
1
. Introduction
C–N coupling also known as Buchwald–Hartwig amination is al-
More recently, literature reported non-precious metals as cata-
lysts able to replace noble metals for cross-coupling processes.
Among the metals in competition with palladium were mentioned
copper [18–23], nickel [24], and more recently, iron [25–28]. As an
ready recognized as a reaction with multiple applications in organ-
ic syntheses [1–4]. Versatile organic building blocks and natural
products were synthesized using this route [5], and some of these
products were implemented at a production scale [6,7].
This reaction was pioneered by Kosugi et al. [8] using homoge-
neous Pd-based catalysts and tin amides as N-nucleophiles. Then,
an example of palladium(0)-mediated carbon–nitrogen bond for-
mation using stoichiometric quantities of palladium(0) was re-
ported by Boger [9]. Afterward, Buchwald [10,11] and Hartwig
0
example, iron chloride in the presence of N,N -dimethylethyl-
enediamine was affording moderate to good yields toward the
N1-arylindoles under relatively mild reaction conditions [29]. Re-
cently, the role of iron in cross-coupling reactions was questioned,
considering that in fact the reactions were catalyzed by copper
impurities of the iron catalyst [30]. However, in spite of the
reduced price, palladium catalysts often possess a higher activity
than their non-noble metal alternatives, in addition, enabling the
conversion of less reactive substrates [31]. Silver was indicated
to promote the activity of Pd-catalyzed amination reactions [32].
Working with such homogeneous catalysts is the subject of
other severe critics. Even in concentrations of the order of few
ppm, they typically led to final compounds containing residual pal-
ladium catalysts [33]. Another disadvantage of these procedures is
the short lifetime of the catalyst tending to further aggregate and
form inactive Pd-black nano-particles [34]. Simple deposition of
these catalysts on supports considering adsorption bonds is not
improving the behavior. The higher the Pd loading is, the faster
the deactivation relative to the catalytic process occurs. Monguchi
[
12,13] independently improved the catalyst and reaction condi-
tions working under of free amines conditions. Then, the develop-
ment of the palladium catalysts for this reaction known large
achievements, and in this moment, we dispose on a large number
of ligands [14]. Improvements in the yield were found by using
bidentate phosphine ligands [15,16], although other discoveries
of Wolfe and Buchwald [17] indicated that bidentate binding is
unnecessary.
⇑
Corresponding authors. Fax: +40 21 410 02 41.
E-mail address: vasile.parvulescu@g.unibuc.ro (V.I. Parvulescu).
0
021-9517/$ - see front matter Ó 2012 Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.jcat.2012.09.002

4
4
M. Ciobanu et al. / Journal of Catalysis 296 (2012) 43–54
et al. [35] published the first palladium on charcoal-catalyzed aryl-
amination of aryl bromides and chlorides showing that the phos-
phane ligand controls the success of the coupling, bidentate
ligands being the most effective for this reaction. Other examples
were provided latter [36].
Under these conditions, heterogeneization of the active palla-
dium catalysts via stronger bonds appeared as a first approach.
In addition, the heterogeneization of the catalysts may allow an
easy separation and recycling of expensive palladium catalysts.
The reported procedures include heterogeneization of ligands on
polymeric supports followed by complexation with palladium
Based on these achievements, Au nanoparticles have been ap-
plied as powerful catalysts for various other oxidative transforma-
tions, such as in the oxidation of alcohols, aldehydes [68–70],
amines [71,72], hydrocarbons [73], and in the epoxidation of al-
kenes [74] utilizing oxygen or air as oxidant. The active species,
0
that is, small metallic clusters of Au , partly oxidized gold particles
d+
(Au ), partly reduced gold particles resulted by interacting with
dꢀ
electron-rich metal defects of the support (Au ), undercoordinat-
ed gold atoms, or cationic Au(I) and/or Au(III) species are in direct
dependence with the nature of the reaction [75–81]. In addition,
for oxidative catalytic reactions, it has been shown that gold nano-
[
37–40] or the fixation homogeneous palladium catalysts onto
2
particles promote the catalytic activity of TiO [82,83].
nanoparticles to combine thus the advantages of homogeneous
and heterogeneous catalysis [41,42]. On this basis, it is also possi-
ble the development of liquid–liquid biphasic catalysts [43]. The
synthesis of cationic imidazolium-based phosphanes is part of
the same approaches, exhibiting high stability against air and
water [44]. Their complexes with palladium can be immobilized
following the supported ionic liquid catalyst concept (SILP). Simple
deposition of palladium on typical supports has also been reported
This study reports the results obtained in amination of bromo-
benzene using titania-supported gold catalysts prepared through
the precipitation deposition technique.
2
. Experimental
2.1. Catalyst preparation
[
45–49,42,50].
The catalysts were prepared by precipitation–deposition tech-
For the conversion of simple, active substrates, ligandless ap-
nique using an already reported protocol [77]. Thus, 1 g of TiO
gussa P25 was added to the corresponding volume for the desired
amount of gold of an aqueous solution of HAuCl O (0.2 M). The
ꢁ3H
pH of the solution was stabilized at 8–9 by dropwise addition of a
2
De-
proaches are more attractive for industry by reducing costs dra-
matically, especially when considering that most of the state of
the art ligands are patented [51]. Correa and Bolm [52] reported
a ligand-free procedure for the N1-arylation of heterocycles cata-
lyzed by Cu(I) oxide while Taillefer and co-workers [53] indicated
4
2
0
1
.2 M NaOH solution. Then, the mixture was stirred at 80 °C for
2 h. After that, the suspension was filtered, washed with distilled
2
an improved behavior for a Cu(II) oxide when using [Fe(acac) ] as a
ꢀ
water until no Cl was detected, and the solid was dried under vac-
co-catalyst. Copper and nickel oxide particles supported within
charcoal have also been developed for Buchwald–Hartwig amina-
tions. High-loading metal catalysts (5 wt.% of Cu, 5 wt.% of Ni) were
active for copper-catalyzed reactions, while relatively inactive
insofar as nickel catalysis was concerned [54].
uum at 80 °C for 48 h. Following this procedure, samples with 0.3,
.5, 0.7 and 0.9 wt.% Au were obtained.
0
2.1.1. Catalyst characterization
Recent reports indicated that such a reaction can be performed
with good results using organocatalysts [55]. In spite of the very
good yields, the separation of the catalyst raised typical questions
addressed to homogeneous catalysis.
The reaction conditions are very important for Buchwald–Har-
twig amination. All the reports about this reaction identified the
presence of an alkaline alcoxide as crucial for the success of amina-
tion. The effect of solvents on the product distribution and reaction
rate was also demonstrated by several authors [56]. Working un-
der microwave activation was not reducing the reaction tempera-
ture but shortened the reaction time [57].
The scope of this study was to investigate amination of bromo-
benzene using another concept. More specifically, we looked for a
metallic species able to subtract a proton from the amine followed
by its reaction with the aryl-halide, and we selected for this pur-
pose gold.
The obtained catalysts were characterized using different tech-
niques. Textural characteristics (surface area and pore diameter)
were determined from the adsorption–desorption isotherms of
nitrogen at ꢀ196 °C using a Micromeritics ASAP 2020 Surface Area
and Porosity Analyzer. DRUV–Vis spectra were collected under
ambient conditions with a Specord 250 (Analytic Jena) equipment.
The equipment is provided with a measuring device in the reflec-
tance mode using Spectralon as reference material. The collected
spectra were transformed using the Kubelka–Munk F(R) function.
The band gap energy was calculated from the dependence of
2
[F(R)ꢁh
m
] as a function of h
m
, where h
m
is the energy of the incident
photons. DRIFT spectra were collected using a Nicolet 4700 spec-
trometer from Thermo Electron. The samples were placed inside
the analyzing chamber attached to a Smart Collector. The spectra
were registered at the room temperature. The final spectra were
averaged from 400 s scans at a resolution of 4 cm . Raman spectra
were collected with a Horiba Jobin Yvon – Labram HR UV–Visible–
NIR (200–1600 nm) Raman Microscope Spectrometer, using a laser
with the wavelength of 632 nm. The spectra were collected from
ꢀ1
Gold catalysis has attracted great interest in the last decade
[
58]. Although Au is not a good catalyst in bulk or in the form of
large particles, when prepared as nanoparticles on an oxide sup-
port, it becomes one of the most active catalysts [59–63]. Experi-
mental evidences corroborated with theoretical calculations
revealed that the nanoparticles are anchored to oxygen vacancies
on the surfaces, which allows the formation of a high areal density
of nanoparticles. The smaller the particle, the greater is the fraction
of atoms directly in contact with the support and therefore influ-
enced by it, while at the same time, the fraction of coordinately
unsaturated surface atoms also increases, and this changes the
physical properties of the whole particle [64]. Reciprocally, the
presence of gold particles may influence the properties of the sup-
ꢀ1
10 scans at a resolution of 2 cm . Powder X-ray Diffraction pat-
terns were collected at room temperature using a Shimadzu
XRD-7000 apparatus and the Cu K
k = 1.5406 A, 40 kV, 40 mA) with a scanning rate of 0.1 2
in the 2 range of 5–80. XPS spectra were recorded at room tem-
a monochromatic radiation
ꢀ
1
H
min ,
H
perature using a SSX-100 spectrometer, Model 206 from Surface
Science Instrument. The pressure in the analysis chamber during
the analysis was 1.33 mPa. Monochromatized Al K
a radiation
(h = 1486.6 eV) generated by bombarding the Al anode with an
m
electron gun operated with a beam current of 12 mA and acceler-
ation voltage of 10 kV has been used. The spectrometer energy
scale was calibrated using the Au 4f7/2 peak centered at 83.98 eV.
Charge correction was made with the C1s signal of adventitious car-
bon (C–C or C–H bonds) located at 284.8 eV. The atomic surface
2 2 2 3
port. The reduction of reducible supports (TiO , CeO , Fe O ) by
hydrogen is catalyzed by gold; this could be caused by hydrogen
spillover from the metal or by changes induced in the electronic
properties of the support [65–67].

M. Ciobanu et al. / Journal of Catalysis 296 (2012) 43–54
45
Table 2
Optical band-gap of Degussa P25 support and Au/TiO
2
samples.
Sample
Calculated band-gap (eV)
0
0
0
0
.3%Au/TiO
.5%Au/TiO
.7%Au/TiO
.9%Au/TiO
2
2
2
2
3.06
3.07
3.05
3.08
2.99
Degussa P25
Fig. 1. Adsorption (h)–desorption (O) isotherms and pore size distribution for the
.9 wt.% Au/TiO catalyst.
0
2
Table 1
The surface area and the pore size for the Au–TiO
2
catalysts.
3
Sample
Surface area BET
Average pore
size (nm)
Pore volume (cm /g)
(BJH desorption)
2
(m /g)
Degussa P25
52.0
56.3
53.1
51.2
39.2
23.0
33.6
33.8
31.2
27.3
0.32
0.51
0.55
0.54
0.57
0
0
0
0
.3%Au/TiO
.5%Au/TiO
.7%Au/TiO
.9%Au/TiO
2
2
2
2
Fig. 3. DRIFT spectra for: (a) 0.3 wt.%Au/TiO
2 2
; (b) 0.5 wt.%Au/TiO ; (c) 0.7 wt.%Au/
2 2
TiO ; (d) 0.9 wt.% Au/TiO .
Fig. 2. DR–UV–Vis spectra for: (a) Degussa P25; (b) 0.3 wt.%Au/TiO
TiO ; (d) 0.7 wt.% Au/TiO ; (e) 0.9 wt.% Au/TiO
2
; (c) 0.5 wt.%Au/
2
2
2
.
Fig. 4. Raman spectra for: (a) TiO
2 2
Degussa P25; (b) 0.3 wt.% Au/TiO ; (c) 0.5 wt.%
Au/TiO ; (d) 0.7 wt.% Au/TiO ; (e) 0.9 wt.% Au/TiO .
2
2
2
compositions were calculated using the sensitivity factors pro-
vided with the ESCA 8.3 D software, applied to the surface below
the corresponding fitted XPS signals. An estimated error of
catalyst, 93 mg aniline, and 188 mg bromobenzene (molar ratio of
1:1.2). The experiments were performed under atmospheric pres-
sure, temperatures of 150–220 °C, and stirring rate of 1.100 rpm.
Two kinds of solvents were used: (i) basic solvents such as di-
methyl carbonate (DMC), diphenyl carbonate (DPC), dimethyl-
formamide (DMF) and pyridine; (ii) nonpolar aprotic solvents
such as 1,4-dioxane and toluene. In the last case, 0.3–2 mmols
potassium ethoxide was also added. Reactants and products were
analyzed by CG-MS using a Trace GC 2000 coupled with DSQ MS
from Thermo Electron Corporation. The structure of the resulted
±0.1 eV can be assumed for all measurements. Specimens for elec-
tron microscopy were prepared by suspending them in acetone
and transferring to a copper grid coated with an amorphous carbon
support. TEM and HRTEM images were recorded on a JEOL JEM
ARM 200F electron microscope operated at 300 keV. The gold
and titania crystallite sizes were established from the measure-
ment of 500–1000 particles for each sample.
2.2. Catalytic tests
1
13
products was confirmed by H- and C–NMR spectroscopy using
a Bruker AV 400 spectrometer, in DMSO solvent and Me Si as inter-
nal standard. The leaching of Au in the reaction products was
The catalytic tests were carried out in a stainless steel magnet-
ically stirred autoclave with Teflon-lined bar loaded with 50 mg of
4

4
6
M. Ciobanu et al. / Journal of Catalysis 296 (2012) 43–54
Table 3
m
Average particle size (d ) and standard deviation (Std. Dev.) values for 0.1 wt.% Au/
TiO
2
, 0.7 wt.% Au/TiO
2
, and 0.9 wt.% Au/TiO
2
catalysts.
ꢂ
ꢂ
Sample
d
m
Au (nm)
Std. Dev. Au
d
m
TiO
2
(nm)
Std. Dev. TiO
2
0
0
0
.1 Au/TiO
.7 Au/TiO
.9 Au/TiO
2
2
2
8.2
14.3
17.1
4.49
3.00
2.44
47.4
46.9
47.0
10.72
9.81
10.55
nomenon of superficial plasmon excitations arises from an impor-
tant growing of surface nanoparticles, which depends on the
resonant wavelength, dimension and shape of particles, and local
dielectric media [85]. Compared to the parent TiO , for the investi-
2
gated gold samples, it was observed that a slight increase in the
band gap at around 3.05–3.08 eV (Table 2).
DRIFT spectra of gold samples presented the characteristic
bands of TiO
2
(Fig. 3) associated with vibrations of Ti–O–Ti and
Fig. 5. X-ray diffractograms for: (a) 0.3 wt.% Au/TiO
.7 wt.% Au/TiO ; (d) 0.9 wt.% Au/TiO
2
; (b) 0.5 wt.% Au/TiO
2
; (c)
ꢀ1
Ti–O bonds, at 800 and 1400 cm . There was no evidence of
Ti–O–Ti and Ti–O bonds disturbance after the deposition of gold.
The Raman spectra (Fig. 4) display three bands at 399, 514,
0
2
2
.
ꢀ
1
checked using ICP–OES (inductively coupled plasma optical emis-
sion spectrometry) with an Agilent 720/730 Series apparatus.
and 639 cm
TiO -anatase and four Raman active modes with A1g + B1g + B2g + E
symmetries specific for TiO -rutile. Differently to DRIFT, the inten-
g
corresponding to A1g + 2B1g + 3E symmetries of
2
g
2
sities of the Raman modes arising from the extension vibration of
the anatase structure were drastically reduced by the presence of
gold. Similar effects were observed by Scurrell and co-workers
[86] who associated these to a slight perturbation of the anatase
3
. Results and discussions
3.1. Catalysts characterization
phase by the interaction of gold ions with the TiO
was suggested to indicate the migration of gold from the exterior
layer of solid TiO into the bulk.
In concordance to the spectroscopic evidences, the XRD pat-
terns indicated the preservation of the crystalline state of TiO
for all samples (Fig. 5). In spite of the observed differences in sur-
face areas, the preparation procedure and the thermal treatment
induced no essential disturbances in the initial crystalline state
2
surface. This
Fig. 1 displays a typical adsorption–desorption isotherm of
nitrogen at ꢀ196 °C for 0.9% Au/TiO
2
catalyst and the pore size dis-
2
tribution. All samples were characterized by a dominant macropo-
rous texture with bimodal pore size distribution. Micropores
corresponded to pores smaller that 1 nm (not accessible to the
investigated reactants) while the dominant meso/macropores
were distributed in a range from 20 to 70 nm. Such pores ensure
an easy diffusion of both reactant and product molecules without
2
of TiO
terns indicated only very small reflections at 2
and 77°, respectively. The diffraction lines from 2
2
. In accordance with the small gold loading, the XRD pat-
of 38°, 44°, 64°,
= 25.3°, 36.9°,
any mass transfer limitation. Except the 0.9%Au/TiO
2
sample, the
H
deposition of gold resulted without any essential change in the
surface area, but with a major change in the pore volume (Table 1).
H
48.1°, 53.9°, 55.1°, 68.9°, and 75.1° were attributed to (101),
These changes in the texture, except the 0.9%Au/TiO
2
sample, seem
(004), (200), (105), (211), (116), and (215) reflections of
to be not influenced by the gold loading but by the precipitation–
deposition process itself.
2
TiO -anatase, while the diffraction lines from 2H = 27.4°, 36.1°,
38.5°, 41.2°, 53.7°, 56.6°, 62.6°, and 70.3° were attributed to
DR–UV–Vis spectra (Fig. 2) exhibits absorption bands around
(110), (101), (200), (111), (211), (220), (200), and (112) reflec-
4
15 nm (2.99 eV) (typical for the transitions between valence band
and conduction band of commercial TiO (P25) (3.0–3.2 eV)) and
50 nm (due to the vibrations of superficial plasmons). The absorp-
2
tions of TiO -rutile phase.
Fig. 6 show typical TEM images obtained with 0.1, 0.7, and
0.9 wt.% Au samples that were examined in detail to obtain metal
particle size distributions. The results are summarized in Table 3
from which it can be seen that the average gold-particle size varied
from 8 nm (0.1%) to 17 nm (0.9%). However, for each concentration,
the particle size distribution was rather narrow. At the lowest Au
loading, the density of gold particles was very low in comparison
2
5
tion band in the visible region is responsible for the pink–violet
color of the samples. It was suggested that the resonance of the
superficial plasmons caused by the collective oscillations of the
electrons from the surface of gold nanoparticles is the key factor
2
of the promoting effect of gold on the TiO activity [84]. The phe-
2 2 2
Fig. 6. TEM pictures of 0.1 wt.% Au/TiO (a), 0.7 wt.% Au/TiO (b), and 0.9 wt.% Au/TiO (c) catalysts.

M. Ciobanu et al. / Journal of Catalysis 296 (2012) 43–54
47
2 2
Fig. 7. HRTEM images for 0.1 wt.% Au/TiO (a) and 0.7 wt.% Au/TiO (b) catalysts.
while the O1s band at around 529 eV was assigned to the frame-
work oxygen (Fig. 9).
XPS Au/Ti ratios correlate very well with the TEM results. Low
loadings of gold corresponded to low XPS Au/Ti ratios that indicate
a low coverage of the support surface.
The XPS spectra of the catalysts investigated in the amination of
bromobenzene with aniline showed different binding energies.
Thus, in the Au4f region, a shift of about 0.6 eV toward higher val-
ues was measured. A shift of the O1s band of about 1.2 eV (to
5
30.2 eV) was also detected (Fig. 10, Table 4). New bands also ap-
peared in the XPS spectra of the spent catalysts: one band located
at a binding energy of 531.7 eV, which was assigned to acidic sur-
face hydroxyl groups, another at 68.5 eV, assigned to Br3d5/2 level
[
89] and two bands at 295.5 eV and 292.8 eV assigned to K2p1/2
and K2p3/2 levels, respectively (Fig. 10). The K/Ti ratio was 0.870.
Recycling the catalyst under inert atmosphere also caused a
4
2
þ
shift of the XPS Ti2p level. Typically, Ti
exhibits a peak around
p3=2
3
2
þ
4
58.8 eV, while the Ti
level is located around 457 eV [90]. The
p3=2
spectrum of the Ti2p level depicted in Fig. 10 presented a peak lo-
cated at 457.2 eV that, by comparison with the spectrum in Fig. 9
4
+
3+
(
458.5 eV), suggested a reduction of Ti to Ti during the reaction.
According to the literature [91,92], such a value of binding en-
2
Fig. 8. XPS spectra of Au4f in fresh Au/TiO samples.
ergy for the K2p level suggested that the counter anion of potas-
2ꢀ
sium ion is an oxide anion (O ), which further indicated that
potassium ions strongly interacted with the lattice oxygen of
TiO . The same authors [91] shown that the K2p3/2 band is slightly
2
with the cases of 0.7% Au and 0.9% specimens. The low Au loading
resulted in a very low gold-particle density and consequently in
larger error bars on the average particle-size values.
The morphology of Au particles was generally spherical if the
particle size not exceeds 10 nm. For increased sizes, the particles
become irregular exhibiting flat facets. In the case of the highest
Au loading, many Au particles were sitting on the interface be-
shifted toward higher binding energies when the K loading
increases. This result suggested that the surface hydroxide is not
silent in this process and that the presence of free OH groups on
the support surface is mandatory.
The gold/titanium ratios calculated from XPS measurements did
not indicate important gold agglomerations on titania surface nor
in fresh samples neither after the catalytic reaction of aniline with
bromobenzene (spent samples) (Table 4).
tween TiO
ticles were supported on the surface of TiO
TiO had a similar size and shape in all the samples.
2
particles, whereas for low loadings, most of the Au par-
2
particles. As expected,
2
Fig. 7a shows a HRTEM image where the (111) lattice planes
are clearly seen in two Au particles of size between 5 and 10 nm.
Other isolated gold particle is presented in Fig. 7b where the
3.2. Catalytic tests
(
112) lattice planes of various TiO
2
particles can be also visualized,
but no orientation relationship is observed between the Au(111)
Literature provided clear evidences on pH importance in Buch-
wald–Hartwig amination. Strong basic conditions are necessary,
and they are generally ensured by adding to the reaction system
an alkaline hydroxide, carbonate, or alcoxide [19]. On the other
hand, the formation of the alkaline salt makes difficult the separa-
tion of both the product and catalyst at the end of the reaction.
However, the use of the basic reagents is strongly correlated with
the nature of the solvent (i.e., to the solubility conditions) and in
order to check the effect of these factors the amination of bromo-
benzene with aniline (Eq. (1)) has been carried out in different sol-
vents. Experiments without the addition of alkaline compounds
were carried out in basic solvents.
and TiO
and TiO
2
(112) planes, although the lattice spacings of Au(111)
(112) are very close to each other (2.35 Å and 2.33 Å,
2
respectively).
The presence of cationic Au species is usually detected in XPS
spectra by a band located at values higher than 84 eV [87]. The
XPS spectra of the fresh prepared Au/TiO samples in the region
2
of the Au 4f7/2 level display a band in the region 82.6–83.3 eV, indi-
0
cating that gold exists as Au species. A negative charge transfer
from the framework oxygen of TiO
2
to the gold species [88]
4+
(Fig. 8) is, however, not excluded. The presence of Ti from the
2
TiO framework was characterized by the Ti2p band at 458 eV,

4
8
M. Ciobanu et al. / Journal of Catalysis 296 (2012) 43–54
2
Fig. 9. XPS spectra of O1s and Ti2p of the fresh Au/TiO samples.
Table 5 shows the evolution of the selectivity in N-methylani-
line and N,N-dimethylaniline as a function of the Au loading for a
conversion of ca 20%. N-methylaniline was the main reaction prod-
uct for the reactions carried out in DMC. For low loadings of gold,
the reaction was quite selective to N-methylaniline (80%). The in-
crease in the gold loading and of the conversion led to a decrease
in the selectivity to this product. This was partially covered by an
increase in the selectivity to diphenylamine and partially to a fur-
ther methylation of N-methylaniline to N,N-dimethylaniline.
Conversion and especially the selectivity were also influenced
by the stirring rate. The selectivity in diphenylamine increased
with the stirring rate until 1.100 rpm, while no further changes
were determined for higher stirring rates. To avoid such effects,
all the experiments were carried out at a stirring rate of 1.100 rpm.
The optimization of this reaction (change of the temperature,
modification of the aniline-to-bromobenzene ratio) was not
accompanied by any additionally increase in the selectivity to
diphenylamine. Although aniline conversion was relatively high,
its reaction with DMC led to large amounts of by-products. The in-
crease in the conversion with the gold loading was a good indica-
tion of the fact that gold is catalyzing this reaction. Nevertheless,
ð1Þ
3
(
.2.1. Catalytic tests in organic basic solvents: dimethyl carbonate
DMC), diphenyl carbonate (DPC), dimethylformamide (DMF), and
pyridine
Working in DMC as solvent, in the absence of catalyst (e.g.,
blank reaction), at 150 °C, and using an aniline/bromobenzene mo-
lar ratio of 1/1.2, the aniline was converted in a proportion of 20.7%
after 24 h. However, no diphenylamine has been formed under
these conditions. The GC–MS analysis in line with the NMR results
shown only by-products of aniline with the solvent (e.g., N-methy-
laniline and N,N-dimethylaniline). This result was expectable as
long as it is well known that DMC methylates anilines, phenols,
and carboxylic acids in conditions close to those investigated in
this study [93]. Bromobenzene remained completely untrans-
formed under these conditions.
despite moderate selectivities, the use of heterogeneous Au/TiO
2
catalysts brings the advantage of easy separation of diphenylamine
from the reaction medium. At room temperature, HBr released
from reaction formed a crystalline adduct with the coupling prod-
uct, which at room temperature can be separated by a simple fil-
tration. The catalysts were separated at the reaction temperature
by filtration and can be easily recycled. No loss of the activity
was determined after four successive cycles (Fig. 11). Also, XPS
analysis of the spent catalysts indicated no changes compared with
spectra presented in Fig. 8, while the ICP–OES analysis indicated at
the level of ppm no Au leached in the product solution.
Pure TiO
conversion and products distribution. In the presence of the Au/
TiO catalysts, the target diphenylamine product was formed
2
led to very similar results, in terms of both the aniline
2
+
+
+
(
1
1
GC–MS, (m/z): 180 (M , 100), 181 (M +1, 84), 182 (M +2, 12),
83 (M +3), 51 (7), 63 (2), 76(5), 77 (35), 78 (8), 104(5), 152 (4),
78 (2); H NMR (DMSO, d (ppm)): 7.2 (3H, m), 7.4 (5H, m), 7.9
2H, m), 8.46 (1H, s)). Both the conversion of aniline and bromo-
benzene and the selectivity to diphenylamine were found to de-
pend on the gold loading (Fig. 9). Therefore, the maximum in
conversion (78.8%) and selectivity (37%) was achieved on 0.9 wt%
Au. Further increase in the gold loading to 1.0% led to the same le-
vel of selectivity in diphenylamine and conversion of aniline
+
1
(
Using DPC led to a partial decomposition of the solvent to phe-
nol, while in DMF or pyridine as reaction solvents and Au/TiO as
2
catalysts, neither aniline nor bromobenzene were converted, indi-
cating that such compounds are less reactive than DMC. Further-
(
Fig. 11, Table 5). This result correlates very well with TEM results
that demonstrated that loadings higher than 0.7 wt.% are leading to
interface sited large Au particles. Similar by-products with those
from the reaction of aniline with DMC were identified.
2 3
more, the addition of an inorganic base as NaOH, KOH, Na CO ,
or CsF to the reaction medium had no effect on the catalytic
performances.

M. Ciobanu et al. / Journal of Catalysis 296 (2012) 43–54
49
2
Fig. 10. XPS spectra of the 0.9%Au/TiO spent sample (Au4f, O1s, Br1s, and K2p band).

5
0
M. Ciobanu et al. / Journal of Catalysis 296 (2012) 43–54
Table 4
The Au/Ti ratio determined from XPS measurements.
Sample
Ti 2p3/2 energy
level (eV)
Au 4f energy
level (eV)
O 1s energy
(eV)
XPS Au/Ti
ratio (%)
4f5/2/4f7/2
0
0
0
0
0
.3%Au/TiO
.5%Au/TiO
.7%Au/TiO
.9%Au/TiO
.9%Au/TiO
2
2
2
2
2
458.0
458.2
458.2
458.5
458.5
87.1/83.1
87.2/83.3
87.1/83.3
86.4/82.6
87.0/83.1
529.7
529.4
529.5
529.9
0.31
0.42
0.52
0.90
0.88
a
530.2/531.7
a
Spent sample.
Fig. 12. Catalytic behavior of Au/TiO
2
catalysts in toluene (aniline/bromobenzene:
1/1.2 M ratio, catalyst: 50 mg, solvent: 5 mL, temperature: 200 °C, potassium
ethoxide: 0.3 mmol, 1.100 rpm, reaction time: 24 h).
Table 6
Selectivity in diphenylamine, phenylbenzene and N-benzylidene-aniline as function
of the gold loading (50 mg catalyst, 1/1.2 aniline/bromobenzene molar ratio,
0.3 mmol potassium ethoxide, 5 mL toluene, 200 °C, 1.100 rpm, conversion ca 20%).
Loading of
gold (wt.%)
Selectivity in
diphenylamine
Selectivity in
phenylbenzene
(%)
Selectivity in N-
benzylidene-aniline
(%)
(
%)
0.3
0.5
0.7
0.9
1.0
0
0
18
22
22
16
20
24
38
40
84
80
58
40
38
3.2.2. Catalytic tests in non-polar aprotic solvents in the presence of an
inorganic base
The limitation in selectivity using DMC determined a change of
strategy. The classic solvents employed in the Buchwald–Hartwig
amination were either non-polar or aprotic in nature, such as m-
xylene or dioxane [56]. Based on this state of the art, we decided
to investigate the Au/TiO
,4-dioxane and toluene) and in the presence of potassium ethox-
ide, as a base.
2
catalysts using similar solvents (e.g.,
1
The variation of the aniline conversion and selectivity in diphe-
nylamine as a function of gold loading, in toluene, are displayed in
Fig. 12. Under these conditions, both conversions and selectivities
in diphenylamine were lower than those obtained in DMC. How-
ever, the nature of by-products was different. For low loadings of
gold (0.3–0.5%), no diphenylamine resulted while the main prod-
ucts were N-benzylideneaniline (Schiff base) (84%) and phenylben-
Fig. 11. (a) Conversion of aniline and selectivity to diphenylamine as a function of
the gold loading (catalyst = 50 mg, aniline/bromobenzene molar ratio = 1/1.2,
DMC = 5 mL, T = 150 °C, 1.100 rpm, reaction time: 24 h); (b) conversion of aniline
and selectivity to diphenylamine for 0.9 wt.% Au catalysts for four runs.
Table 5
+
+
+
Selectivity in diphenylamine, N-methylaniline and N,N-dimethylaniline as function of
the gold loading (50 mg catalyst, 1/1.2 aniline/bromobenzene molar ratio, 5 mL DMC,
zene (GC–MS, (m/z): 154 (M , 100), 155 (M +1, 12), 156 (M +2, 1),
153(44), 152(32), 151(11), 128(5), 115(3), 89(1), 77(7), 76(12),
1
50 °C, 1.100 rpm, conversion ca 20%).
7
5(2)) (Table 6) (Scheme 1). For loadings higher than 0.7 wt.%
Loading of
gold (wt.%)
Selectivity in
diphenylamine
Selectivity in N-
methylaniline (%)
Selectivity in N,N-
dimethylaniline (%)
diphenylamine resulted in a small extent and the selectivity to
phenylbenzene increased.
(%)
The formation of the first two by-products (e.g., phenylbenzene
and N-benzylideneaniline) was due to an unproductive side reac-
tion (involving molecules of toluene from the reaction medium),
which can compete with reductive elimination wherein the amide
undergoes beta hydride elimination to yield the hydrodehalogen-
ated arene (which forward led to phenylbenzene by a coupling
0.3
0.5
0.7
0.9
1.0
20
22
25
37
37
80
70
62
45
43
0
8
13
18
20

M. Ciobanu et al. / Journal of Catalysis 296 (2012) 43–54
51
Scheme 1. Detected by-products in Buchwald–Hardwig reaction in toluene, as a function of the loading of gold.
reaction) and the N-benzylideneaniline product. The origin of
Table 7
phenylbenzene is thus not related to the presence of aniline, and
under the reaction conditions, it was also formed in the absence
of aniline as a parallel reaction in competition with the main reac-
tion leading to diphenylamine. Separate experiments carried out
Selectivity in diphenylamine, phenylbenzene, and carbazole as a function of the gold
loading (50 mg catalyst, 1/1.2 aniline/bromobenzene molar ratio, 0.3 mmol potassium
ethoxide, 5 mL dioxane, 200 °C, 1.100 rpm, 24 h).
Loading of
gold (wt.%)
Selectivity in
diphenylamine (%)
Selectivity in
phenylbenzene (%)
Selectivity in
carbazole (%)
2
without aniline (50 mg of 0.7 wt%Au/TiO catalyst, 188 mg bromo-
benzene, 0.3 mmoles potassium ethoxide) led to only phenylben-
zene. However, the conversion was around 20% after 24 h.
0.3
0.5
47
69
100
84
43
17
0
12
13
10
14
0
4
3
0.7
0.9
1.0
On the other hand, the formation of the latter compound (e.g.,
N-benzylideneaniline) is most probably due to the toluene oxida-
tion to benzaldehyde in the presence of the catalyst (that can lib-
erate small amounts of network oxygen). Such a behavior was
proved by recycling experiments. Under an inert atmosphere, the
recycled catalyst exhausted the removable surface oxygen and N-
benzylideneaniline was not any more formed. On the contrary, in
air, the N-benzylideneaniline was favored and the selectivity in
this compound increased up to 14%. Fig. 10 provided additional
84
for the 0.7%Au catalyst (Fig. 13) when for conversions lower than
0%, the analysis of the reaction products indicated only the forma-
2
tion of the desired product. For higher conversions (>30%), like for
the other catalysts, by-products started to be also formed (e.g.,
phenylbenzene), thus decreasing the selectivity in diphenylamine.
The selectivity was also depended on the gold loadings (Table 7).
Except for the 0.7%Au catalyst, phenylbenzene and in a small ex-
tent carbazole were the by-products. At low Au loadings
3
2
þ
arguments to support this behavior. The presence of the Ti
le-
p3=2
vel in the spectra of the spent catalysts suggested, indeed, the re-
lease of oxygen during this reaction.
Working in dioxane led to a complete different behavior. The
selectivity was much improved. The best results were obtained
(
(
0.3 wt.% Au), the selectivity in phenylbenzene was rather high
43%).
As in the case of using DMC solvent, the optimization of the
reaction (change of the temperature, modification of the aniline-
to-bromobenzene ratio) was not accompanied by any additionally
increase in the selectivity to diphenylamine.
The catalysts can be easily separated by filtration, but the sim-
ple recycling was not possible. XPS shown, in the case of the cata-
lysts tested in dioxane and potassium ethoxide, the existence of
KBr that most probably blocked the active surface of the catalysts.
For the catalysts tested in DMC, no bromine has been evidenced
because this was released as an adduct of diphenylamine. In addi-
2
tion, Au/TiO catalysts ex-situ poisoned with KBr also exhibited no
activity. As the catalytic tests shown, the selectivity to diphenyl-
amine was well correlated with the content in potassium ethoxide
in the reaction medium. Thus, the increase in the potassium ethox-
ide content from 0.3 mmol to 2.0 mmol led to a linear increase in
the aniline conversion until around 50% with a maximum of selec-
tivity in diphenylamine of 89% for 1.1 mmol potassium ethoxide
(Fig. 14). The increase in the potassium ethoxide content over this
value favored the further substitution of diphenylamine to
triphenylamine.
The coverage of the surface by KBr follows the conversion of
aniline. Separation of the catalyst after certain reaction times fol-
lowed by the XPS analysis and corroboration with the GC–MS ana-
lytic data led to the dependences depicted in Fig. 15. Once the
Fig. 13. Catalytic performances as a function of the gold loadings in dioxane
(
aniline/bromobenzene: 1/1.2 M ratio, catalyst: 50 mg, solvent: 5 mL, temperature:
00 °C, potassium ethoxide: 0.3 mmol, 1.100 rpm, reaction time: 24 h).
2

5
2
M. Ciobanu et al. / Journal of Catalysis 296 (2012) 43–54
Fig. 14. The variation of aniline conversion and diphenylamine selectivity with the
amount of potassium ethoxide, in the presence of 0.9 wt.%Au/TiO catalyst
DPA = diphenylamine; TPA = triphenylamine) (aniline/bromobenzene: 1/1.2 M
ratio, catalyst: 50 mg, solvent: 5 mL, temperature: 200 °C, 1.100 rpm, reaction
time: 24 h).
Fig. 17. Catalytic behavior of Au/TiO
amount in the presence of 1.1 mmol of potassium ethoxide (aniline/bromobenzene:
/1.2 M ratio, catalyst: 50 mg, solvent: 5 mL, temperature: 200 °C, pottasium
2
catalysts in dioxane as a function of the gold
2
(
1
ethoxide: 1.1 mmol, 1.100 rpm, reaction time: 24 h).
surface becomes blocked by halide, no further conversion of aniline
has been determined.
Interestingly, washing the catalysts with dioxane removed KBr
in an enough extent to recycle the catalysts several times with
the same catalytic performances.
Indeed, 1.1 mmol of potassium ethoxide correlated very well
with the aniline/bromobenzene molar ratio of 1/1.2 and repre-
sented the optimal composition to achieve high selectivities. Thus,
irrespective of the solvent, both the aniline conversion and diphe-
nylamine selectivity reached the highest values (Figs. 16 and 17).
However, it is important to notice again that the level of both con-
version and selectivity was strongly influenced by the loading of
gold onto titania.
Due to the Lewis basicity of the oxygen atoms, 1,4-dioxane was
able to solvate many inorganic compounds. This property together
with the basicity character explained the better results obtained in
this solvent comparing with toluene.
3.3. General assessment
Fig. 15. Conversion of aniline and XPS K/Ti ratios as a function of the reaction time.
DR–UV–vis spectra revealed bands due to the vibrations of
superficial plasmons and the presence of metallic Au [94]. In addi-
tion, Raman characterization confirmed the existence of the inter-
action between gold and the support, interaction that, in fact, could
be responsible for the above plasmons. The presence of metallic
gold in these catalysts was also confirmed by XPS that also indi-
cated for the fresh catalysts the existence of titanium only as Ti4+
species. TEM analysis indicated that gold was deposed with a
rather narrow size distribution, but the average size and the den-
sity of the gold particles increased with the gold loading. XPS char-
acterization of the investigated catalysts revealed the existence of
potassium and bromine on the catalysts surface that was assigned
to a strong interaction of the alkaline species with the oxygen an-
ion on the surface.
Based on the catalysts characterization and catalytic perfor-
mances, it can be concluded that the amination of bromobenzene
to diphenylamine is influenced in terms of conversion and selectiv-
ity by the size of the gold particles and their density. Thus, the in-
crease in the gold loading generated an increase in the conversion
and selectivity up to 0.7 wt.% Au, while higher loading led to larger
Au particles with a lower interaction with the support, generating
the observed catalytic behavior.
Fig. 16. Catalytic behavior of Au/TiO
amount in the presence of 1.1 mmol of potassium ethoxide (aniline/bromobenzene:
/1.2 M ratio, catalyst: 50 mg, solvent: 5 mL, temperature: 200 °C, potassium
ethoxide: 1.1 mmol, 1.100 rpm, reaction time: 24 h).
2
catalysts in toluene as a function of the gold
The Buchwald–Hartwig amination is generally initiated by the
oxidative addition of the zero-valent metal into halo benzene
1

M. Ciobanu et al. / Journal of Catalysis 296 (2012) 43–54
53
the formation of diphenylamine. In the same manner, the reaction
can go until the formation of triphenylamine. The reaction stops
when the surface is blocked with the resulted KBr, as shown by
XPS.
The formation of phenylbenzene seems to follow a different
route. As far as this reaction occurs also in the absence of aniline,
it should be the result of the abstraction of bromide by potassium
and dimerization of the resulted phenyl entities.
The work with these heterogeneous catalysts brought the
advantage that KBr rested chemisorbed on the solid surface and
on the fact that it can be removed by simple percolation with diox-
ane. Then, the catalyst was easily recycled for four times with very
similar performances.
In none of the experiments, the ICP–OES analysis of the reaction
products does indicated the presence of leached gold. In agreement
with the mechanisms proposed in Schemes 2a and 2b, these data
confirm that the mechanism of activation of the reaction of gold
is different on that occurring on palladium. Recent reports indi-
cated that using immobilized palladium in this reaction led to a
small concentration of leached free-ligand noble metals in solu-
tion, which is able to catalyze the amination reaction [95].
Scheme 2a. Proposed mechanism for the heterogeneous amination in organic
solvents with basic properties and Au/TiO catalysts.
2
4
. Conclusion
[
10–13]. Working with gold such a route is less probable. However,
The deposition of gold on titania occurred with the formation of
it is more probable that Au would participate in this reaction via an
electron transfer. In addition, the reaction mechanism was con-
trolled by the reaction medium. When working in DMC, aniline is
most probably activated at gold species, followed by the nucleo-
phylic attack of bromobenzene and the formation of the crystalline
adduct of the resulted diphenylamine with HBr (Scheme 2a). This
is an important advantage of this reaction because at the end of
the reaction, the catalyst was not poisoned. In this case, however,
the selectivity was limited by the faster reaction of DMC with ani-
line which does not require the presence of a catalyst. On the other
side, surprisingly, working in DPC led to only a partial decomposi-
tion to phenol. In the presence of an inorganic base (e.g., potassium
ethoxide), the reaction pathway was different (Scheme 2b). In this
case, the presence of the strong base determined the extraction of a
hydrogen atom from amine leading to new OH groups observed in
the Raman spectra of the support, followed by the blockage of the
surface oxygen anions by potassium, as observed in the XPS spec-
tra. Then, the dehydrogenated amine attacked bromobenzene with
gold particles with controlled size and oxidation state. The interac-
tion of these particles with the support stabilized the metallic state
of gold. Amination of bromobenzene with aniline on these cata-
lysts was found to depend on both the properties induced by the
gold loading and the reaction medium. As literature unanimously
reported, the presence of a base is mandatory in this reaction.
Working in DMC, the catalyst was playing the role of base. In
this way, the formation of KBr, which is very difficult to be sepa-
rated from the reaction products, was avoided. Moreover, HBr re-
sulted as a by-product form adducts with the coupling product,
which were separated as crystals at room temperature by a simple
filtration. In spite of low selectivities, comparing with literature
data, this is the first report on the synthesis of diphenylamine
through amination of bromobenzene without using an inorganic
base. DMC is a modern green solvent and requires low tempera-
tures compared to temperatures used in heterogeneous catalysis.
Working in dioxane and in the presence of potassium ethoxide,
as base, led to very high selectivities for moderate conversions. The
2
Scheme 2b. Proposed mechanism for the heterogeneous amination in molecular organic solvents in the presence of potassium ethoxide and Au/TiO catalysts.

5
4
M. Ciobanu et al. / Journal of Catalysis 296 (2012) 43–54
[43] J.R. Naber, S.L. Buchwald, Angew. Chem., Int. Ed. 49 (2010) 9469.
conversion was limited by KBr released from the reaction. How-
ever, the use of the heterogeneous catalyst had the advantage of
scavenger, since KBr was chemisorbed on the surface. In addition,
it can be easily removed from the surface by simple percolation
and the catalyst can be reused. Differently to recent reports on pal-
ladium, no leaching of gold has been detected in this reaction.
The transfer of these reactions in flow systems is presently
examined in our laboratories, and further research will focus on
the development on the increase in the conversion.
[
44] A. Dumrath, X.-F. Wu, H. Neumann, A. Spannenberg, R. Jackstell, M. Beller,
Angew. Chem., Int. Ed. 49 (2010) 8988.
[
45] S.S. Prockl, W. Kleist, K. Kohler, Tetrahedron 61 (2005) 9855.
[46] K. Kohler, R.G. Heidenreich, S.S. Soomro, S. Prockl, Adv. Synth. Catal. 350 (2008)
930.
2
[
[
47] W. Kleist, J.-K. Lee, K. Kohler, Eur. J. Inorg. Chem. (2009) 261.
48] M. Lamblin, L. Nassar-Hardy, J.-C. Hierso, E. Fouquet, F.-X. Felpin, Adv. Synth.
Catal. 352 (2010) 33.
[
[
49] Y. Monguchi, Y. Fujita, K. Endo, S. Takao, M. Yoshimura, Y. Takagi, T. Maegawa,
H. Sajiki, Chem. Eur. J. 15 (2009) 834.
50] S. Wittmann, A. Schatz, R.N. Grass, W.J. Stark, O. Reiser, Angew. Chem., Int. Ed.
49 (2010) 1867.
[
[
[
51] C. Torborg, M. Beller, Adv. Synth. Catal. 351 (2009) 3027.
52] A. Correa, C. Bolm, Adv. Synth. Catal. 349 (2007) 2673.
53] M. Taillefer, N. Xia, A. Ouali, Angew. Chem., Int. Ed. 46 (2007) 934.
Acknowledgments
The authors kindly acknowledge to the European Commission –
Research: The Seventh Framework Program for the financial sup-
port through FP7 Project No. 246461 and CNCS for the Project PNII
[54] B.H. Lipshutz, D.M. Nihan, E. Vinogradova, B.R. Taft, Z.V. Boskovic, Org. Lett. 10
2008) 4279.
[
(
55] C.-L. Sun, H. Li, D.-G. Yu, M. Yu, X. Zhou, X.-Y. Lu, K. Huang, S.-F. Zheng, B.-J. Li,
Z.-J. Shi, Nature Chem. 2 (2010) 1044.
2
75. Part of the work was supported by the strategic grant POS-
[56] H. Christensen, S. Kiil, K. Dam-Johansen, O. Nielsen, M.B. Sommer, Org. Proc.
Res. Dev. 10 (2006) 762.
[
[
DRU/88/1.5/S/56668, Project ‘‘Process Improvement Training in
Doctoral Studies in Science’’ co-financed by the European Social
Foundation within the Sectorial Operational Program Human Re-
sources Development 2007–2013.
57] M. Koley, M. Schnurch, M.D. Mihovilovic, Tetrahedron 67 (2011) 4169.
58] G.C. Bond, C. Louis, D.T. Thompson, in: G. Hutchings (Ed.), Catalysis by Gold,
Catalytic Science Series, vol. 6, Springer, Berlin, 2006.
[59] G.J. Hutchings, J. Catal. 96 (1985) 292.
[
[
60] M. Haruta, T. Kobayashi, H. Sano, N. Yamada, N. Chem. Lett. 16 (1987) 405.
61] M. Haruta, H. Kageyama, N. Kamijo, T. Kobayashi, F. Delannay, Stud. Surf. Sci.
Catal. 44 (1989) 33.
References
[
62] M. Haruta, N. Yamada, T. Kobayashi, S. Iijima, J. Catal. 115 (1989) 301.
[
[
1] A.R. Muci, S.L. Buchwald, Top. Curr. Chem. 219 (2002) 131.
2] J.F. Hartwig, in: C. Austruc (Ed.), Modern Arene Chemistry, Wiley-VCH,
Weinheim, 2002, p. 107.
[63] B. Chowdhury, J.J. Bravo-Suarez, M. Date, S. Tsubota, M. Haruta, Angew. Chem.,
Int. Ed. 45 (2006) 412.
[64] A. Corma, P. Serna, Science 313 (2006) 332.
[
[
[
3] J.F. Hartwig, in: E. Negishi (Ed.), Handbook of Organopalladium Chemistry for
Organic Synthesis, vol. 1, John Wiley & Sons, New York, 2002, p. 1051.
4] L. Jiang, S.L. Buchwald, in: A. deMeijere, F. Diederich (Eds.), Metal Catalyzed
Cross-Coupling Reactions, second ed., Wiley-VCH, Weinheim, 2004.
5] K.C. Nicolaou, S.A. Snyder, Classics in Total Synthesis II, Willey-VCH,
Weinheim, 2003.
[65] J. Guzman, S. Carrettin, J.C. Fierro-Gonzalez, Y. Hao, B.C. Gates, A. Corma,
Angew. Chem., Int. Ed. 44 (2005) 4778.
[66] S. Carrettin, P. Concepcion, A. Corma, J.M. Lopez-Nieto, V.F. Puntes, Angew.
Chem., Int. Ed. 43 (2004) 2538.
[67] T. Risse, S. Shaikhutdinov, N. Nilius, M. Sterrer, H.-J. Freund, Acc. Chem. Res. 41
(2008) 949.
[
[
6] B. Schlummer, U. Scholz, Adv. Synth. Catal. 346 (2004) 1599.
7] D.B. Damon, R.W. Dugger, S.E. Hubbs, J.M. Scott, R.W. Scott, Org. Process Res.
Dev. 10 (2006) 472.
8] M. Kosugi, M. Kameyama, T. Migita, Chem. Lett. (1983) 927.
9] D.L. Boger, J.S. Panek, Tetrahedron Lett. 25 (1984) 3175.
[68] A. Abad, P. Concepcion, A. Corma, H. García, Angew. Chem., Int. Ed. 117 (2005)
4134.
[69] D.I. Enache, J.K. Edwards, P. Landon, B. Solsona-Espriu, A.F. Carley, A.A. Herzing,
M. Watanabe, C.J. Kiely, D.W. Knight, G.J. Hutchings, Science 311 (2006) 362.
[70] A. Dhakshinamoorthy, H. Garcia, Chem. Soc. Rev. 41 (2012) 5262.
[71] L. Aschwanden, T. Mallat, M. Maciejewski, F. Krumeich, A. Baiker,
ChemCatChem 2 (2010) 666.
[72] A. Grirrane, A. Corma, H. Garcia, Nat. Protoc. 5 (2010) 429.
[73] M.I. bin Saiman, G.L. Brett, R. Tiruvalam, M.M. Forde, K. Sharples, A. Thetford,
R.L. Jenkins, N. Dimitratos, J.A. Lopez-Sanchez, D.M. Murphy, D. Bethell, D.J.
Willock, S.H. Taylor, D.W. Knight, C.J. Kiely, G.J. Hutchings, Angew. Chem., Int.
Ed. 51 (2012) 5981.
[
[
[
[
10] A.S. Guram, R.A. Rennels, S.L. Buchwald, Angew. Chem. Int. Ed. Engl. 34 (1995)
348.
11] J.P. Wolfe, S. Wagaw, J.-F. Marcoux, S.L. Buchwald, Acc. Chem. Res. 31 (1998)
05.
12] J. Louie, J.F. Hartwig, Tetrahedron Lett. 36 (1995) 3609.
13] J. Hartwig, Acc. Chem. Res. 31 (1998) 852.
14] R. Martin, S.L. Buchwald, Acc. Chem. Res. 41 (2008) 1461.
15] J. Wolfe, S. Wagaw, S. Buchwald, J. Am. Chem. Soc. 118 (1996) 7215.
16] M. Driver, J. Hartwig, J. Am. Chem. Soc. 118 (1996) 7217.
17] J. Wolfe, S. Buchwald, ACIEE 38 (1999) 2413.
1
8
[
[
[
[
[
[
[
[
[
[
[
[
[
[74] B.K. Min, C.M. Friend, Chem. Rev. 107 (2007) 2709.
[75] M.S. Chen, D.W. Goodman, Science 306 (2004) 252.
[76] D. Matthey, J.G. Wang, S. Wendt, J. Matthiesen, R. Schaub, E. Laegsgaard, B.
Hammer, F. Besenbacher, Science 315 (2007) 1692.
18] C.-J. Li, Acc. Chem. Res. 35 (2002) 533.
19] S.R. Chemler, P.H. Fuller, Chem. Soc. Rev. 36 (2007) 1153.
20] H. Plenio, Angew. Chem., Int. Ed. 47 (2008) 6954.
21] F. Monnier, M. Taillefer, Angew. Chem., Int. Ed. 47 (2008) 3096.
22] D. Ma, Q. Cai, Acc. Chem. Res. 41 (2008) 1450.
23] M. Carril, R. SanMartin, E. Dominguez, Chem. Soc. Rev. 37 (2008) 639.
24] (a) S. Ikeda, Acc. Chem. Res. 33 (2000) 511;
[77] F. Neatu, Z. Li, R. Richards, P.Y. Toullec, J.-P. Genet, K. Dumbuya, J.M. Gottfried,
H.P. Steinruck, V.I. Parvulescu, V. Michelet, Chem. Eur. J. 14 (2008) 9412.
[78] M. Chen, D.W. Goodman, Chem. Soc. Rev. 37 (2008) 1860.
[79] J.C. Fierro-Gonzaleza, B.C. Gates, Chem. Soc. Rev. 37 (2008) 2127.
[80] T. Fujitani, I. Nakamura, Angew. Chem., Int. Ed. 50 (2011) 10144.
[81] H. Zhang, T. Watanabe, M. Okumura, M. Haruta, N. Toshima, Nat. Mater. 11
(2012) 49.
(
b) J. Montgomery, Acc. Chem. Res. 33 (2000) 467.
[
[
[
[
25] A. Correa, O.G. Mancheno, C. Bolm, Chem. Soc. Rev. 37 (2008) 1108.
26] A.B.D. Sherry, A. Furstner, Acc. Chem. Res. 41 (2008) 1500.
27] A. Furstner, Angew. Chem., Int. Ed. 48 (2009) 1364.
[82] Z. Dawson, P.V.J. Kamat, J. Phys. Chem. B 105 (2001) 960.
[83] V. Subramanian, E.E. Wolf, P.V. Kamat, J. Am. Chem. Soc. 126 (2004) 4943.
[84] Y. Tian, T. Tatsuma, J. Am. Chem. Soc. 127 (2005) 7632.
[85] D. Lahiri, V. Subramanian, T. Shibata, E.E. Wolf, B.A. Bunker, P.V. Kamat, J. Appl.
Phys. 93 (2003) 2575.
28] W.M. Czaplik, M. Mayer, J. Cvengros, A. Jacobi von Wangelin, ChemSusChem 2
(
2009) 396.
[
[
[
[
[
[
29] A. Correa, C. Bolm, Angew. Chem., Int. Ed. 46 (2007) 8862.
30] S.L. Buchwald, C. Bolm, Angew. Chem., Int. Ed. 48 (2009) 5586.
31] A. Frisch, M. Beller, Angew. Chem., Int. Ed. 44 (2005) 674.
32] Ch.S. Bryan, M. Lautens, Org. Lett. 10 (2008) 4633.
33] C.C. Mauger, G. Mignani, Org. Process Res. Dev. 8 (2004) 1065.
34] M. Tromp, J.R.A. Sietsma, J.A. van Bokhoven, G.P.F. Van Strijdonck, R.J. van
Haaren, A.M.J. van der Eerden, P.W.N.M. van Leeuwen, D.C. Konigsberger,
Chem. Commun. (2003) 128.
[86] K. Mallick, M.J. Witcomb, M.S. Scurrell, Appl. Catal. A: Gen. 259 (2004) 163.
[87] M.P. Casaletto, A. Longo, A. Martorana, A. Prestianni, A.M. Venezia, Surf. Interf.
Anal. 38 (2006) 215.
[88] S. Arrii, F. Morfin, A.J. Renouprez, J.L. Rousset, J. Am. Chem. Soc. 126 (2004)
1199.
[89] A.Yu. Tarasova, L.I. Isaenko, V.G. Kesler, V.M. Pashkov, A.P. Yelisseyev, N.M.
Denysyuk, O.Yu. Khyzhun, J. Phys. Chem. Solids 73 (2012) 674.
[90] Y. Fu, H. Du, S. Zhang, W. Huang, Mater. Sci. Eng. A 403 (2005) 25.
[91] K. Asano, C. Ohnishi, S. Iwamoto, Y. Shioya, M. Inoue, Appl. Catal. B: Environ. 78
(2008) 242.
[
35] Y. Monguchi, K. Kitamoto, T. Ikawa, T. Maegawa, H. Sajiki, Adv. Synth. Catal.
3
50 (2008) 2767.
[
[
[
[
36] A. Komaromia, Z. Novak, Adv. Synth. Catal. 352 (2010) 1523.
37] C.A. Parrish, S.L. Buchwald, J. Org. Chem. 66 (2001) 3820.
38] Y.-C. Yang, T.-Y. Luh, J. Org. Chem. 68 (2003) 9870.
[92] J.M. Moggia, V.G. Milt, M.A. Ulla, L.M. Cornaglia, Surf. Interf. Anal. 35 (2003)
216.
[93] W.-C.h. Shieh, S. Dell, O. Repi, J. Org. Chem. 67 (2002) 2188.
[94] M. Magureanu, N.B. Mandache, J. Hu, R. Richards, M. Florea, V.I. Parvulescu,
Appl. Catal. B: Environ. 76 (2007) 275.
39] A. Gupta, T.R. Long, D.G. Rethwisch, N.B. Bowden, Chem. Commun. 47 (2011)
1
0236.
[
[
40] H. Christensen, S. Kiil, K. Dam-Johansen, Org. Proc. Res. Dev. 11 (2007) 956.
41] M. Beller, H. Fischer, K. Kuhlein, C.-P. Reisinger, W.A. Herrmann, J. Organomet.
Chem. 520 (1996) 257.
[95] M. Al-Amin, T. Honma, N. Hoshiya, S. Shuto, M. Arisawa, Adv. Synth. Catal. 354
(2012) 1061.
[
42] M.-J. Jin, D.-H. Lee, Angew. Chem., Int. Ed. 49 (2010) 1119.