Stabilization and recovery of gold catalysts in the cyclopropanation of alkenes within ionic liquids

Article abstract of DOI:10.1016/j.jcat.2008.07.009

Au(I) and Au(III) salts (NaAuCl₄ and KAuCN₂ respectively) were found to be active catalysts for the cyclopropanation of alkenes with ethyldiazoacetate, in many cases affording high yields of cyclopropanecarboxylates. But these gold s

Full text of DOI:10.1016/j.jcat.2008.07.009

Journal of Catalysis 259 (2008) 26–35
Contents lists available at ScienceDirect
Journal of Catalysis
www.elsevier.com/locate/jcat
Stabilization and recovery of gold catalysts in the cyclopropanation of alkenes
within ionic liquids
∗
∗
Avelino Corma , Irene Domínguez, Tania Ródenas, María J. Sabater
Instituto de Tecnología Química, UPV-CSIC, Universidad Politécnica de Valencia, Avenida los Naranjos s/n, 46022 Valencia, Spain
a r t i c l e i n f o
a b s t r a c t
Article history:
Au(I) and Au(III) salts (NaAuCl4 and KAuCN2 respectively) were found to be active catalysts for the
cyclopropanation of alkenes with ethyldiazoacetate, in many cases affording high yields of cyclopropane-
carboxylates. But these gold salts rapidly decomposed and agglomerated when working with organic
solvents, resulting in less active and less selective gold metal particles. With ionic liquids (ILs) as
solvents, the gold catalysts were stabilized, especially Au(I), and products and catalyst separation and
recycling could be achieved. In both cases, NaAuCl4 and KAuCN2 underwent reductive transformation to
Au(0) to afford gold nanoparticles, which remained stabilized in the IL, behaving as an authentic metal
nanoparticle reservoir. It was found that cyclopropanation and aromatic addition were sensitive to gold
particle size, whereas the formation of fumarate and maleate esters was insensitive to gold particle size.
© 2008 Elsevier Inc. All rights reserved.
Received 19 May 2008
Revised 16 July 2008
Accepted 16 July 2008
Available online 23 August 2008
Keywords:
Gold catalysis in ionic liquids
Cyclopropanation with gold catalysts
1. Introduction
reaction media. We show that when the same gold salt catalysts
are used in the presence of the appropriated IL, the Au(III) and
Au(I) salts can be stabilized toward the formation of gold metal
agglomerates, thereby resulting in higher yields of cyclopropanes
than can be achieved using conventional solvents. Moreover, the
separation of products and catalyst recovery and recycling is much
easier when using ILs.
Gold has been shown to be an active and selective catalyst for
numerous reactions, including C–C [1–5], C–N [6–10], C–O [11,12],
C–H [13–15], and C–Si [16] bond-forming reactions; oxidation [17];
and enantioselective hydrogenation of olefins and imines [18,19].
In many cases, gold salts have good activity and selectivity, but
have drawbacks associated with catalyst recovery and recycling
and, even more importantly, in many cases can decompose dur-
ing the reaction, leading to the formation of large particles of gold
metal that are not active or unselective. Thus, it is of interest to
find a procedure to stabilize the gold salts and avoid decomposi-
tion while achieving easy separation of the reaction products and
catalyst recycling.
The transition metal-catalyzed transfer of carbene units from
diazo compounds to alkenes provides a powerful organic synthesis
tool for obtaining cyclopropane derivatives, which are important
intermediates for the construction of relatively complex molecules
[20–27]. Cyclopropanation of alkenes from diazo compounds can
be catalyzed by Cu(I) [23]. Because the external electronic structure
of Cu(I) is similar to that of Au(I), gold possibly could be a suit-
able cyclopropanation catalyst. Indeed, the literature reports that a
gold(I)-based catalyst also can be an efficient and chemoselective
catalyst for this reaction [28]. Here we demonstrate that Au(I) as
well as Au(III) salts (KAuCN2 and NaAuCl4) can catalyze the cyclo-
propanation of alkenes by ethyldiazoacetate in ionic liquids (ILs) as
2. Experimental
2.1. Materials and starting reagents
The starting reagents (alkenes and ethyldiazoacetate) and cat-
alysts NaAuCl4·2H2O and KAuCN2 were purchased from Aldrich.
[MOIM] PF6, with purity of 99.9% and water content of 182.3 ppm,
was supplied by Fluka. Halides and sulfates ions were not detected.
[BMIM] PF6, with purity >98.5%, water content <0.02%, and halo-
gen content (as chloride) <10 mg/kg, was supplied by Solvent In-
novation. [BMIM] BF4, with purity >98.5%, water content <0.02%,
and anion traces (bromide ꢀ25 mg/kg; chloride ꢀ25 mg/kg; ni-
trate ꢀ25 mg/kg and phosphate ꢀ10 mg/kg), was supplied by
Solvent Innovation. [BMIM] Br, with purity >97.0% and water con-
tent <1%, was supplied by Solvent Innovation.
2.2. General techniques
Room temperature UV–vis spectrum of a chloroform solution of
catalyst containing IL [BMIM] PF6 was recorded with a Shimadzu
UV–vis scanning spectrophotometer. The size distribution profile
of small gold nanoparticles was determined using a Nano Zeta-
sizer Malvern (ZEN3600) MPT-2 analyzer (with a red laser wave-
Corresponding authors. Fax: +34 96 3877809.
E-mail addresses: acorma@itq.upv.es (A. Corma), mjsabate@itq.upv.es
*
(M.J. Sabater).
0021-9517/$ – see front matter © 2008 Elsevier Inc. All rights reserved.
doi:10.1016/j.jcat.2008.07.009

A. Corma et al. / Journal of Catalysis 259 (2008) 26–35
27
length of 633 nm). At least two measurements were obtained for
each sample, and the final Z-average is given as an average value
(see Supplementary material). Raman spectra were obtained with
an “in via” Renishaw spectrometer, equipped with a microscope
(Olympus). The exciting wavelength was 785 nm from a Renishaw
HPNIR laser with a power of ca. 15 mW on the sample. Solution
NMR spectra were recorded with a Bruker Avance 300 spectrom-
eter at 300.13 MHz (1H), 75.47 MHz (13C), and 121.49 MHz (13P).
1H and 13C NMR spectra of isolated cyclopropanecarboxylates were
recorded in Cl3CD. 13P NMR spectra of gold-containing ILs were
obtained in DMSO-d6. The gold content of ILs was determined
by inductively coupled plasma optical emission spectrometry (ICP-
OES) using a Varian 715-ES spectrometer.
The spectral data of all products were identical to those of authen-
tic samples.
3. Results and discussion
Gold(III)-catalyzed carbene transfer reaction from ethyldiazoac-
etate (EDA) to styrene was initially carried out in the presence of
NaAuCl4 but in the absence of solvent. Under these experimental
conditions, the cis/trans-cyclopropanecarboxylates were obtained
with low selectivity (see entry 1 in Table 1) due to the formation
of relatively large amounts of EDA dimerization products (diethyl
fumarate and maleate), together with o-, m-, p-aromatic addition
products (Scheme 1).
The formation of these three isomers can be accounted for by a
formal insertion of the :CH2COOEt carbene into the sp2 C–H aro-
matic bond [28].
2.3. Synthesis of thiol-stabilized gold nanoparticles (1–3 nm)
The influence of solvent on activity and chemoselectivity of cy-
clopropanation was studied using solvents with different polarities.
The findings, given in Table 1, show that independent of the sol-
vent, high conversions could be achieved in all cases, whereas
chemoselectivity to cis/trans-cyclopropanecarboxylates clearly was
much higher for the less-polar toluene (entries 2–6, Table 1). Re-
gardless of the solvent used, NaAuCl4 decomposed slowly during
the addition of EDA, and metallic gold formed as a thin golden
“mirror” that could be recovered at the end of reaction.
Because the gold particles formed by catalyst decomposition
may very well also be catalytically active but less selective toward
the formation of cyclopropanes, we carried out a control exper-
iment in which styrene and EDA were allowed to react in the
presence of metal gold nanoparticles (1–3 nm) freshly prepared as
described previously [29]. Despite the fact that gold nanoparticles
stabilized by surface halide ions are likely to be the best model for
studying the activity of metal nanoparticles, we prepared colloidal
gold nanoparticles capped with thiols, given that gold nanoparti-
cles are usually surrounded with monolayers of organic molecules
when prepared in organic solvents, whereas nanoparticles sta-
The thiol-derived gold nanoparticles were prepared as de-
scribed previously [29]. The resulting material was handled as a
simple chemical compound.
2.4. General procedure for the cyclopropanation reactions
To a stirred solution of alkene (3 mmol), 8% mol of the NaAuCl4
catalyst (or 5% mol KAuCN2 or 8% mol CuCl2) and the IL (3 ml),
ethyldiazoacetate (1 mmol), was added either completely before
starting the reaction or slowly in a dropwise manner over 2 h. The
resulting mixture was stirred at room temperature for 24 h. After
the reaction was completed, the reaction mixture was extracted
with diethyl ether (1 × 10 ml). The organic solvent was dried with
anhydrous MgSO4 and concentrated under vacuum. The samples
were analyzed by gas chromatography.
Cyclopropanecarboxylates were isolated by column gas chro-
matography on silica gel (Merck, 100–200 mesh; ethyl acetate–
hexane 1:9). They were purified and identified by comparison of
their NMR-1H, NMR-13C and mass spectra with literature values.
Table 1
Cyclopropanation reaction of styrene catalyzed by NaAuCl4 with ethyldiazoacetate (EDA) in different solvents at room temperature
Entry
Solvent
Conv. (%)a
Yield (%)
TONe
cis/transb
o-, m-, p-c
Dimersd
1
2
3
4
5
6
7
–
97
93
91
85
96
91
87
21/28
16/29
16/32
22/29
25/39
32/40
3/28
30
12
16
21
25
26
18
20
–
1
17.3
11.9
7.9
10.5
12.5
14.5
16
CH3CN
CH3NO2
1,2-Dichloroethane
CH2Cl2
Toluene
Toluenef
2
10
2
g
–
a
Based in EDA consumption and determined by GC after total consumption of EDA with n-dodecane as internal standard.
Cyclopropanecarboxylate isomers (cis + trans).
b
c
d
e
f
Addition aromatic products at the ortho-, meta- and para-position (o-, m-, p-).
Diethyl fumarate and maleate.
Calculated as mmol of converted substrate/mmol of catalyst.
n-Dodecanethiol stabilized Au(0) nanoparticles (1–3 nm) as catalyst in toluene as solvent reaction [29].
Unidentified products of high molecular weight were detected by GC (instead of fumarate and maleate esters).
g
Scheme 1. Carbene transfer reaction from EDA to styrene catalyzed by NaAuCl4.

28
A. Corma et al. / Journal of Catalysis 259 (2008) 26–35
Fig. 1. Structure of ionic liquids used in this study.
Table 2
Cyclopropanation reaction of alkenes with ethyldiazoacetate (EDA) catalyzed by NaAuCl4 in ionic liquids at room temperature
Entry
Alkene
Solvent
Conv.
(%)a
Yield (%)
cis/transb
TONe
o-, m-, p-c
Dimersd
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
Styrenef
[BMIM] PF6
[BMIM] PF6
[BMIM] PF6
[BMIM] P F6
[BMIM] PF6
[BMIM] PF6
[BMIM] PF6
[BMIM] PF6
[BMIM] PF6
[BMIM] PF6
[BMIM] PF6
[OMIM] PF6
[OMIM] PF6
[OMIM] PF6
[OMIM] PF6
[OMIM] PF6
98 (99)
85
29
91
99
99
90
99 (94)
82
100 (98)
78
86 (99)
83 (95)
83
75
59 (99)
32/62 (30/63)
20/29
2/0
32/49
29/52
34/38
8/19
2/3 (37/40)
67m
– (41/43)
27m
1/83 (35/59)
32/40 (40/42)
79m
10/22
15/19 (44/44)
2 (1)
12
9
7
11
6
3
1 (4)
–
– (4)
6
2 (2)
– (3)
4
3
1 (4)
2
10 (13)
11.4
6
11
23.5
21
7.2
9 (12)
10
– (13)
12.3
7.4 (11)
7.4 (13)
5
Styreneg
2 (24)h
Styrenei
–
–
7
21
4
–
–
–
Styrenej
Styrenek
Styrenel
p-Methoxystyrene
p-Trifluoromethylstyrenef
β-Methylstyrene
α-Methylstyrenef
Cyclooctene
5
Styrenef
– (2)
– (3)
–
–
– (3)
p-Trifluoromethylstyrenef
β-Methylstyrene
p-Methoxystyrene
α-Methylstyrenef
5
8.3 (12)
a
Based in EDA consumption and determined by GC after total consumption of EDA with n-dodecane as internal standard.
Cyclopropanecarboxylates (cis + trans).
b
c
Addition aromatic products at the ortho-, meta- and para-position (o-, m-, p-).
Diethyl fumarate and maleate.
d
e
f
Calculated as mmol of converted substrate/mmol of catalyst.
Results in parenthesis refer to experiments in which EDA was completely added before starting the reaction.
g
h
i
◦
Reaction carried out at 60 C.
Oligomerization products in parenthesis.
HCl as catalyst.
j
◦
Ionic liquid treated at 70 C under reduced pressure for 8 h.
k
l
NaAuCl4·2H2O + CuCl2.
CuCl2.
m
Σ stereoisomers.
bilized with halides are usually prepared in aqueous medium.
glomeration [40]. If this is so, then it also may be possible to
stabilize Au(III) species in an IL better than can be done with
organic solvents. To begin, we chose [BMIM] PF6 and the even
more hydrophobic 1-methyl-3-octylimidazolium hexafluorophos-
phate [MOIM] PF6 to carry out the cyclopropanation reaction (see
Fig. 1).1
Results from Table 2 (entry 1) show that with [BMIM] PF6, it is
possible to obtain better selectivities and yields to cyclopropana-
tion products than with any of the organic solvents used before.
Moreover, the yields of cyclopropanecarboxylates obtained from
aromatic alkenes, such as styrene, are much higher than those
reported in the literature for a known gold(I) catalyst [28]. The
evolution of the products during reaction was followed by taking
samples at different time intervals (Fig. 2). The IL [BMIM] PF6 was
extracted with diethyl ether.2
Thus, a catalytic amount of gold nanoparticles (stabilized with n-
dodecanethiol) was added to fresh toluene and styrene. EDA was
slowly added, and the reaction was monitored by GC.
The results obtained with gold nanoparticles after 24 h demon-
strate lower conversion and yield of cyclopropanecarboxylates than
those achieved using NaAuCl4 as catalyst in the same solvent,
toluene (entry 7, Table 1). This suggests that the low selectivity
toward cyclopropanes obtained in organic solvents may not be due
solely to the Au(III) catalyst, but also may be related to the forma-
tion of metallic gold by decomposition of the catalyst that directs
the reaction toward the unwanted products. If this indeed is the
case, then it would be necessary to increase the stability of the
salt during the reaction to increase the selectivity of the process.
We return to this point later in the paper.
Fig. 2 shows that the yield of cis+trans-cyclopropanes increased
rapidly up to ∼80% during the first 8 h, then continued to increase
up to 94% over 24 h. In contrast, the yield of aromatic addition
products remained constant up to the completion of the reaction.
3.1. Cyclopropanation reactions catalyzed by NaAuCl4 and KAuCN2 in
ILs as solvents
ILs are being used as potential green replacements for con-
ventional organic solvents due to their negligible vapor pres-
sure and the possibility of tuning their physicochemical properties
through structural modification of the cation and choice of an-
ion [30–39]. More recently, the IL 1-butyl-3-methylimidazolium
hexafluorophosphate (commonly abbreviated as [BMIM] PF6) has
shown the ability to stabilize palladium nanoparticles without
the need to add stabilizing agents to avoid the typical metal ag-
−
1
PF6 provides hydrophobicity to the imidazolium based IL with respect to other
−
−
counter anions, such as BF4 , Cl , etc. Similarly the use of long tail substituents at
the cation may contribute to increase even further this property.
2
[BMIM] PF6 is fully miscible with polar organic solvents such as ethanol,
dimethylsulfoxide, acetonitrile, acetone, etc., being at the same time fully or par-
tially immiscible in less polar organic solvents such as toluene, hexane, diethyl
ether, etc.

A. Corma et al. / Journal of Catalysis 259 (2008) 26–35
29
Fig. 4. Particle size for gold nanoparticles measured as a function of time in the
cyclopropanation reaction of styrene catalyzed by NaAuCl4 upon (a) slow addition
of EDA and (b) immediate complete addition of EDA in [BMIM] PF6 as solvent.
Fig. 2. Evolution of different products with time in the cyclopropanation reaction
of styrene with EDA catalyzed by NaAuCl4 in the ionic liquid [BMIM] PF6: (a) cis +
trans-cyclopropanes; (b) addition aromatic products (o-, m-, p-) and (c) dimers (en-
try 1, Table 2).
of adding all of the EDA at the beginning of the reaction on the cy-
clopropanation reaction of styrene catalyzed by NaAuCl4 in [BMIM]
PF6, with the idea that the high viscosity of ILs might slow dimer
formation (see Fig. 3). In this experiment, the reaction was almost
completed within 6 h, and, more interestingly, the yields of the
EDA coupling products (fumarate and maleate) were undetectable
(Fig. 3). Another interesting finding was that nearly identical yields
of cis + trans-cyclopropanes were obtained with respect to the ex-
periment in which EDA was added slowly during the first 2 h of
the reaction period. This finding is very important, because the re-
action time for cyclopropanation was shortened considerably and
also the catalyst working time was extended considerably, because
the catalyst could be reused more times before unactive metal ag-
glomerates formed.
Closely connected to that experiment, we also compared the
possible effect of either an initial complete addition or slow ad-
dition of EDA on the kinetics of gold nanoparticle formation in
[BMIM] PF6 as solvent. We used dynamic light scattering (DLS)3
[41] to evaluate the size of gold particles that could form with
NaAuCl4 in the IL [BMIM] PF6 (see Section 2 and Fig. 4). Fig. 4
shows that the slow addition of diazocompound not only sup-
pressed the dimerization reaction of EDA (although this may not
be necessarily the case with [BMIM] PF6 as solvent), but also con-
tributed to slowing the growth of gold nanoparticles during the
reaction.
Taking into account a previous study in which the reaction be-
tween alcohols and alkenes was catalyzed by AuCl3 and the rate
of decomposition and metal formation was slowed by introduction
of CuCl2 [11], we performed the cyclopropanation reaction in the
presence of NaAuCl4 and CuCl2 as co-catalysts in the IL [BMIM] PF6
as solvent (entry 5, Table 2). Contrary to our expectations, the se-
lectivity to cyclopropanation was lower here than in the absence
of CuCl2. Nevertheless, when we carried out the reaction in the
presence of CuCl2 (without NaAuCl4), CuCl2 clearly was also active
but less selective than Au(III) for performing the cyclopropanation
reaction in the IL, accounting for the loss of selectivity observed
(entry 6, Table 2).
Fig. 3. Plot showing the yields of reaction products versus time in the cyclopropa-
nation reaction of styrene with EDA (immediate complete addition) catalyzed by
NaAuCl4 in the ionic liquid [BMIM] PF6: (a) cis + trans-cyclopropanes; (b) addition
aromatic products (o-, m-, p-) and (c) others.
A control reaction between styrene and EDA was performed in the
◦
absence of catalyst in the IL [BMIM] PF6 at 20 C and no cyclo-
propanation was observed.
The addition time of EDA (i.e., concentration of EDA in the
reaction mixture) should be considered as a possibly important
variable in cyclopropanation reactions, because the metal carbene
intermediate can interact with another diazoester molecule (rather
than with the olefin) to form a dimer. This is the reason why de-
creasing the rate of EDA addition should, in principle, result in bet-
ter cyclopropane yields and explains why numerous studies in the
literature use long addition times at the start of the reaction [23].
If this were the case, then the selectivity toward cyclopropanes
would be expected to increase with an increasing styrene:EDA ra-
tio, because the greater this ratio, the more likely that the metal
carbene intermediate will interact with styrene rather than with
another carbene to form a dimer. In most cases (as in our case),
both parameters can be adjusted simultaneously; indeed, we used
a long duration of EDA addition (2 h) and an excess of olefin to
carry out the reactions (see Section 2). We also studied the effect
To evaluate the scope of this gold-catalyzed reaction, we stud-
ied the reactivity of other electron-rich and electron-deficient
3
Fundamentals of this technique (also known as Photon Correlation Spec-
troscopy) are based on the fact that particles, emulsions and molecules in suspen-
sion undergo Brownian motion. If the particles or molecules are illuminated with a
laser, the intensity of the scattered light fluctuates at a rate that is dependent upon
the size of the particles. Analysis of these intensity fluctuations yields the veloc-
ity of the Brownian motion and hence the particle size using the Stokes–Einstein
relationship.

30
A. Corma et al. / Journal of Catalysis 259 (2008) 26–35
Fig. 5. Evolution of different products with time in the cyclopropanation reaction of
styrene with EDA catalyzed by KAuCN2 in the ionic liquid BMIM PF6: (a) cis + trans-
cyclopropanes; (b) addition aromatic products (o-, m-, p-), (c) others.
Fig. 6. Plot showing the yields of reaction products versus time in the cyclopropana-
tion reaction of styrene with EDA (immediate addition) catalyzed by KAuCN2 in the
ionic liquid BMIM PF6: (a) cis + trans-cyclopropanes; (b) addition aromatic products
(o-, m-, p-), (c) others and (d) dimers (fumarate and maleate esters).
styrene derivatives (p-methoxystyrene and p-trifluoromethylsty-
rene, respectively). In this case, both olefins converted almost com-
pletely, although giving very low yields of cyclopropanecarboxy-
lates. Although the substituents at this para-position (methoxy and
trifluoromethyl) hindered the insertion of the CH2COOEt unit into
the aromatic ring, this did not contribute to an increased yield
of the desired cyclopropanecarboxylates (see entries 7 and 8, Ta-
ble 2).
With β-methylstyrene, good conversion values could be
achieved after 24 h of reaction, and neither addition aromatic
products nor fumarate and maleate esters were detected by GC
(entry 9, Table 2). α-Methylstyrene converted almost completely,
giving a complex mixture of unidentified products (entry 10, Ta-
ble 2). Finally, we also screened the carbene transfer reaction from
EDA into the aliphatic olefin cis-cyclooctene (entry 11, Table 2),
and found that in this case, low yields of the corresponding cyclo-
propanecarboxylates could be achieved.
We also studied the effect of an initial complete addition of
EDA on the cyclopropanation reaction of alkenes α-methylstyrene
and p-trifluromethylstyrene catalyzed by NaAuCl4 in [BMIM] PF6
(see entries 8 and 10 in Table 2). The reaction was much faster
(completed in 1 h in both cases), and the cyclopropane yields were
improved significantly with respect to the experiments in which
EDA was added slowly over 2 h (see Table 2). Even more impor-
tantly, no dimer formation was detected.
The cyclopropanation reaction of styrene derivatives when cat-
alyzed by NaAuCl4 in the IL [MOIM] PF6 resulted in much bet-
ter yields of cyclopropanecarboxylates than those achieved when
these reactions were carried out in [BMIM] PF6 (see entries 12–16,
Table 2). The most significant feature with [MOIM] PF6 was the al-
most exclusive formation of the trans-cyclopropanecarboxylate in
the case of styrene (entry 12, Table 2).
Again, we evaluated the possible influence of an immediate
complete addition of EDA in [MOIM] PF6 IL. To do so, we used the
cyclopropanation of α-methylstyrene and p-trifluoromethylstyrene
with NaAuCl4. As in the study with [BMIM] PF6, the reaction was
completed rather rapidly, affording high yields of cyclopropanecar-
boxylates in both cases (4 and 2 h for α-methylstyrene and p-tri-
fluoromethylstyrene, respectively). Minor amounts of fumarate an
maleate dimers were detected by GC analysis in both cases (see
entries 13 and 16 in Table 2). Similarly, with KAuCN2 as catalyst
and [BMIM] PF6 as solvent, the yields of cyclopropanecarboxylates
derived from styrene were again higher than those reported in the
literature for a related gold(I) complex (entry 1, Table 3) [28].
Fig. 5 shows the evolution of different reaction products over
time for the cyclopropanation reaction of styrene with EDA cat-
alyzed by KAuCN2 in [BMIM] PF6. The figure indicates that the
yield of cis + trans-cyclopropanes increased to approximately 70%
during the first 10 h, then it increased slightly to 78% after
24 h. Similarly, the yield of aromatic addition products increased
smoothly, reaching a maximum after 24 h.
We also studied the effect of adding the total amount of EDA
when starting the reaction on the cyclopropanation reaction of
styrene in [BMIM] PF6 with KAuCN2 as the catalyst (Fig. 6). The re-
sults show that the reaction was almost completed in 8 h, demon-
strating that the yields of the EDA-coupling products (fumarate and
maleate) also were very low (Fig. 6). As in the case of NaAuCl4,
nearly identical yields of cis + trans-cyclopropanes were obtained
with respect to the experiment in which EDA was added in 2 h at
the start of the reaction period (cf. Fig. 5).
We have seen that when water (10% mol) was added in the
cyclopropanation of styrene with KAuCN2 as catalyst, the yield of
cyclopropanes decreased slightly with respect to the experiment
in the absence of water (see entries 1 and 2 in Table 3), indicating
that possible differences in the water present within the reaction
media accompanying the catalysts, reactants, or ILs should have a
small influence on reactivity.
Similarly, the cyclopropane yields were significantly higher for
the alkenes α-methylstyrene, p-trifluoromethylstyrene, and cy-
clooctene compared with those obtained with NaAuCl4 as catalyst
in the same IL (entries 2, 3, and 5, Table 3). In striking contrast,
p-methoxystyrene afforded poorer results in cyclopropanation (see
entry 4, Table 3).
With the exception of β-methylstyrene, the results obtained
in cyclopropanation reactions with KAuCN2 as catalyst improved
when the catalyzed reactions were carried out in [MOIM] PF6 as
solvent (see entries 6–10, Table 3). In this case, the most remark-
able features with KAuCN2 were the absence of typical EDA dimer-
ization products (especially when the diazocompound was added
slowly) and the higher yields of aromatic addition products with
respect to the Au(III) catalyst NaAuCl4 (cf. Tables 2 and 3).
We studied the effect of adding all of the EDA when start-
ing the reaction for reactions catalyzed by KAuCN2 in both ILs
[BMIM] PF6 and [MOIM] PF6 with α-methylstyrene and p-tri-
fluoromethylstyrene as model olefins. As in previous cases, the
reactions were completed in a few hours (6–12 h) and afforded
high yields of cyclopropanecarboxylates in all cases (see entries 3,
4, 8 and 10 in Table 3).
3.2. Stability of ILs in the presence of NaAuCl4 and KAuCN2
As stated earlier, the catalyst evaluation revealed enhanced ac-
tivity and chemoselectivity of [AuCl4]− for the cyclopropanation

A. Corma et al. / Journal of Catalysis 259 (2008) 26–35
31
Table 3
Cyclopropanation reaction of alkenes catalyzed by K(CN)2Au with ethyl diazoacetate (EDA) in ionic liquids at room temperature
Entry
Alkene
ILs
Conv. (%)a
Yield (%)
TONd
cis/transb
o-, m-, p-c
1
2
3
4
5
6
7
8
9
Styrenee
[BMIM] PF6
[BMIM] PF6
[BMIM] PF6
[BMIM] PF6
[BMIM] PF6
[BMIM] PF6
[OMIM] PF6
[OMIM] PF6
[OMIM] PF6
[OMIM] PF6
[OMIM] PF6
100 (98)
95
97 (96)
99 (93)
98
100
98
100 (96)
100
100 (93)
100
32/46 (30/43)
29/32
22 (16)
15
28 (7)
17 (2)
–
28
22
17 (5)
15
27 (4)
4
8 (11)
8
7 (14)
16 (13)
10
Styrenef
α-Methylstyrenee
p-Trifluoromethylstyrenee
p-Methoxystyrene
Cyclooctene
15/53 (40/43)
3/27 (31/39)
1/2
68g
34/44
6.3
Styrenee
5.8(12)
7.7 (13)
8.9
7.7 (15)
9.1
α-Methylstyrenee,h
p-Methoxystyrene
p-Trifluoromethylstyrenee,f
β-Methylstyrene
51/18 (40/47)
15/68
10/12 (36/40)
55e
10
11
a
Based in EDA consumption and determined by GC after total consumption of EDA with n-dodecane as internal standard.
Cyclopropanecarboxylate diastereomers (cis + trans).
b
c
d
e
f
Addition aromatic products at the ortho-, meta- and para-position (o-, m-, p-).
mmol converted substrate/mmol catalyst.
Results in parenthesis refer to experiments in which EDA was completely added before starting the reaction.
10% mol water was added.
Σ stereoisomers.
g
h
Minor amounts of dimers were detected (<2%).
−
reaction in the IL [BMIM] PF6 compared with other conventional
In parallel, and to avoid the undesired PF6 hydrolysis, we tried
◦
to exclude moisture from [BMIM] PF6 by heating the IL at 70 C
solvents. Nevertheless, we note that [BMIM] PF6 became acidic
−
−3
under reduced pressure (10
mbar) for 8 h [47]. Results for the
shortly after AuCl4 was added, suggesting that either the Au(III)
−
cyclopropanation of styrene with EDA with NaAuCl4·2H2O as the
catalyst are given in Table 2 (see entry 4). As can be deduced from
these results, the yields of cyclopropanecarboxylates in [BMIM] PF6
after 24 h did not improve significantly after drying compared with
the results obtained without the exclusion of water. Nonetheless,
the most remarkable feature in this case is that GC did not detect
the formation of fumarate and maleate esters (see entries 1 and 4
in Table 2). Contrary to our expectations, the more hydrophobic
nature of [MOIM] PF6 did not completely prevent PF6 hydrolysis
in the presence of some water. The same species, PO2F , PO3F2−
and H3PO4 were detected by 31P NMR when the spectrum of a
mixture of NaAuCl4 and [MOIM] PF6 was recorded under identical
experimental conditions (see Fig. 7B).
salt or PF6 anions may hydrolyze under these experimental con-
ditions [42–44]. Hydrolysis of AuCl4 can effectively account for
the existence of AuCl4 in pH- and strength-dependent equilibrium
with gold species (vs AuCl3OH and AuCl3(H2O), among others)
−
−
−
with simultaneous HCl formation [42].
−
Similarly, PF6 anions can undergo hydrolysis to give phosphates
and HF [43,44]. The formation of these compounds then may con-
tribute to acid-catalyzed reactions or initiate catalyst degradation.
In connection with this, we carried out a control reaction in the
presence of HCl (but in the absence of tetrachloroaurate) to evalu-
ate the influence of mineral acids in the cyclopropanation reaction.
We found that the olefin (styrene) converted rather sluggishly (up
to 29%) after 24 h of reaction and that the yield of cyclopropane
products was very low (2%) (entry 3, Table 2). In this case, a com-
plex mixture of unidentified products of high molecular weight
was detected by gas chromatography (entry 3, Table 2).
−
−
,
2
We note that a possible reaction between imidazolium cations
(in the IL) and AuCl4 may occur simultaneously to form a gold
−
N-heterocyclic carbene complex, as has been described previously
for palladium [48]. But we discounted this possibility, because for-
mation of a similar gold carbene complex was not detected by
1H NMR, despite the fact that equimolar amounts of tetrachloroau-
rate(III) salt and [BMIM] PF6 were used to record the 1H NMR
spectrum.
The acidity of the reaction medium sometimes may cause side
reactions. In the case of cyclopropanation of styrene derivatives,
the selectivity toward cyclopropanes depends in part on the acid-
ity of the reaction medium, given that acid catalysts are able to
promote the cationic polymerization of styrene derivatives. Exam-
ples of this effect are described in the literature. For instance,
in cyclopropanation reactions catalyzed by solid copper catalysts,
the lowest selectivities toward cyclopropanes were achieved when
acidic K-montmorillonite was used as the support [45].
−
It is worth pointing out to that the counteranion PF6 was
stable when the gold(I) complex KAuCN2 was added to the ILs
[BMIM] PF6 and [MOIM] PF6 separately. In fact, the typical hy-
−
drolytic species derived from PF6 were not detected by 31P NMR
spectroscopy in any case. This suggests that Au(III) may be acting
as a catalyst in the hydrolysis of PF6 , competing with cylopropa-
nation. In this respect, the possibility that gold(III) may catalyze
PF6 degradation with simultaneous formation of transition metal
We used 31P NMR spectroscopy to detect the formation of prod-
−
−
ucts derived from a plausible PF6 hydrolytic reaction that could
generate acid molecules. In this experiment, an equimolar mix-
ture of [BMIM] PF6 and NaAuCl4 was submitted to vigorous stir-
ring at room temperature under nitrogen for 2 h, after which
the 31P NMR spectrum of this mixture was recorded. Through
analysis of this 31P spectrum (Fig. 7A), three different species de-
−
fluorides cannot be discarded [43].
On the other hand, and to check whether a plausible release of
HCN from KAuCN2 may decrease the selectivity in the cyclopropa-
nation reaction, we recorded the 1H, 31P, and 13C NMR spectra of
the IL before and after the cyclopropanation reaction of styrene
was completed (see entry 1, Table 3). Again, the 1H, 31P, and
13C NMR spectra confirmed that the IL did not change after the
reaction, suggesting that [BMIM] PF6 was stable under these reac-
tion conditions.
−
rived from the hydrolysis of PF6 could be detected: a triplet with
chemical shift at δ31P = −15.77 ppm [J(P, F) = 953.4 Hz], a dou-
blet with δ31P = −6.89 ppm [J(P − F) = 911.82 Hz], and a singlet
−
at 0 ppm. The first two signals were tentatively assigned to PO2F2
and PO3F2−, respectively [46], whereas the singlet was assigned to
H3PO4 (an acid that could be formed in the final stage of the hy-
drolysis). The septet with δ31P = −143.8 ppm [J(P−F) = 711.33 Hz]
We conducted an additional experiment in which the cyclo-
propanation reaction of styrene with EDA was carried out with
HCN as the catalyst (generated in situ from equimolar amounts
−
belongs to PF6 [46].

32
A. Corma et al. / Journal of Catalysis 259 (2008) 26–35
Fig. 7. 31 P NMR spectra of two equimolar mixtures of NaAuCl4 and [BMIM] PF6 (A) and NaAuCl4 and [OMIM] PF6 respectively (B). Signals a, b, c and d are tentatively assigned
−
−
to PF6 , PO2F , PO3F2− and H3PO4 respectively.
2
Table 4
Results obtained in cyclopropanation reaction of styrene with EDA catalyzed by NaAuCl4 in different BMIM based ionic liquids
Entry
Solvent
C (%)a
Y (%)
TONe
cis/transb
o-, m-, p-c
Dimersd
Other
1
2
3
BMIM PF6
BMIM Br
BMIM BF4
98
98
86
32/62
21/7
13/33
2
6
4
2
–
–
–
64
36
10
11
11
a
Based in EDA consumption and determined by GC after total consumption of EDA with n-dodecane as internal standard.
Cyclopropanecarboxylates (cis + trans).
b
c
Addition aromatic products at the ortho-, meta- and para-position (o-, m-, p-).
Diethyl fumarate and maleate.
d
e
Calculated as mmol of converted substrate/mmol of catalyst.
of KCN and water) in [BMIM] PF6 but in the absence of KAuCN2.
3.3. Reusability and recycling of NaAuCl4 and KAuCN2 catalysts in
[BMIM] PF6
In this case, no reaction was observed after 24 h, confirming that
HCN did not catalyze the cyclopropanation reaction or any sec-
ondary reaction. HCN likely is not sufficiently acidic to promote
the cationic polymerization of the olefin styrene under these ex-
perimental conditions.
Reusability experiments conducted with NaAuCl4 as the catalyst
in [BMIM] PF6 demonstrated a gradual decrease of chemoselectiv-
−
ity. In this case, the [BMIM] PF6 containing catalyst was recycled in
−
a series of consecutive experiments. Fig. 8 shows the comparative
conversions and yields of cis- and trans-cyclopropanation products
derived from styrene with repeated cycles. The results indicate that
although the rate of salt decomposition may have slowed signifi-
cantly in [BMIM] PF6, the gold nanoparticles formed under these
experimental conditions grew slowly over time, thereby decreasing
the chemoselectivity toward the production of cyclopropanecar-
boxylates. In this respect, attempts to detect gold nanoparticle for-
Because the PF6 based IL can be partially hydrolyzed in the
presence of water, we performed the cyclopropanation reaction in
other ILs containing the 1-methyl-3-butyl imidazolium cation but
with different anions. Table 4 gives the results obtained for the
cyclopropanation reaction of styrene with EDA in [BMIM] Br and
[BMIM] BF4 ILs at room temperature, along with results obtained
with BMIM PF6 for comparison. As can be deduced from this table,
the best performance was obtained with [BMIM] PF6. The more
hydrophilic bromide and tetrafluoroborate ILs afforded poorer re-
sults on cyclopropanation. It is interesting to note that despite the
mation by Raman spectroscopy were unsuccessful, because the Ra-
−1
man band centered at ca. 265 cm
and assigned to gold nanopar-
ticles [49] (associated with chloride ions bound to gold nanoparti-
cles) overlapped with a Raman band vibration characteristic of the
IL at approximately the same wavenumber.
−
fact that halide anion (Br ) tends to stabilize the formation of
gold particles, avoiding metal agglomeration, this positive effect
has not been reflected in the reaction by increased yields of cy-
lopropanecarboxylates.
Nonetheless, the absence of the Au(III)–Cl Raman vibration at
−1
355 cm
(characteristic of the original complex NaAuCl4) after

A. Corma et al. / Journal of Catalysis 259 (2008) 26–35
33
(A)
Fig. 9. UV–vis spectrum of NaAuCl4 containing [BMIM] PF6 after cyclopropanation
reaction of styrene with EDA (first use) recorded in CHCl3.
and characteristic absorption band associated with the formation
of Au nanoparticles at 520 nm in the UV–vis spectrum (the typi-
cal plasmon absorption resonance band for gold nanoparticles) was
seen [50].
It appears that the increasing nanoparticles size with succes-
sive catalyst recyclings may be related to the observed decrease
in selectivity toward cyclopropanes. To check this hypothesis, we
prepared gold nanoparticles of different sizes (i.e., 8, 22, 79, and
102 nm) following the method by seeding described by Haiss
et al. [51] with some modifications. The IL BMIM PF6 was used as
solvent, and NaBH4 was used as reducing agent to form the gold
nanoparticles, albeit in the absence of a stabilizing agent. In each
case, the [BMIM] PF6 containing the gold nanoparticles was used
in the cyclopropanation reaction of styrene with EDA. The results,
shown in Fig. 10, demonstrate a clear dependency of product se-
lectivity on gold particle size for cis + trans-cyclopropanes and o-,
m-, p-aromatic addition products (Figs. 10A and 10B), with the for-
mation of fumarate and maleate esters apparently independent of
this parameter (Fig. 10C). The highest yields of cyclopropanes were
obtained with the smallest nanoparticles (Fig. 10A), with selectiv-
ity remaining high for particle sizes of ∼22 nm and decreasing to
approximately 45% and 30% for particle sizes ∼79 and 102 nm, re-
spectively.
(B)
Fig. 8. (A) Comparative conversions (C%) and yields (Y%) of cis- and trans-
cyclopropanocarboxylates against number of runs in the carbene transfer reac-
tion from EDA to styrene catalyzed with NaAuCl4 in the ionic liquid [BMIM] PF6;
(B) comparative Z-average of gold nanoparticles obtained by DLS technique after
successive reuses in the cyclopropanation reactions of styrene catalyzed by NaAuCl4
in [BMIM] PF6.
initial use strongly suggests that the anionic AuCl4 salt likely un-
derwent a reductive transformation in the IL [BMIM] PF6 during
the reaction. UV–vis spectra of the gold-containing [BMIM] PF6
confirms this fact, with a broad and characteristic absorption band
associated with the formation of Au nanoparticles appearing at
520 nm (the typical plasmon absorption resonance band for gold
nanoparticles) after initial use (Fig. 9) [50].
We used DLS to evaluate the size of gold particles that formed
with NaAuCl4 in the IL [BMIM] PF6. Analysis of particle size and
particle size distribution was carried out on a Malvern Nano
Zetasizer analyzer (see Section 2 and Supplementary material).
The data obtained reveal that the Z-average particle size for the
nanoparticle entities produced after the first two uses was still
<10 nm (7.5 and 9.0 nm for gold particles after first and second
use, respectively); however, this Z-average increased to 23 nm after
a third use (Fig. 8; also see Section 2 and Supplementary material).
At this point, we attempted to identify the chemical species
responsible for gold salt reduction by detecting the formation of
gold nanoparticles on the basis of DLS. In this case, two different
BMIM PF6 solutions containing NaAuCl4 were treated separately
with styrene and ethyldiazoacetate at room temperature under vig-
orous stirring. The particle size and particle size distribution of
these mixtures were analyzed using a Malvern Nano Zetasizer an-
alyzer at different time intervals. These experiments revealed that
formation of nanoparticles did not occur after olefin was added
(even after 24 h), but gold nanoparticles were formed 2 h after
EDA was added (Z-average particle size, 100 nm). Fumarate and
maleate esters were detected by GC analysis, whereas a broad
Interestingly, aromatics formed by addition reactions follow
the opposite trend with gold particle size, because the selectiv-
ity to these products increased with increasing gold particle size
(Fig. 10B). This agrees with the fact that the greatest yields of aro-
matic addition products were obtained in conventional organic sol-
vents (see Table 1), in which the catalyst was rapidly transformed
into metal agglomerates during the initial stages of the reaction.
At this point, we reiterate that the instability observed for the
−
IL containing NaAuCl4 (hydrolysis of, e.g., PF6 ) may facilitate the
formation of different gold species that could catalyze the reac-
tion in a nonselective way; making it difficult to correlate the
cyclopropane yields with the nanoparticle size in reactions with
NaAuCl4 as the catalyst. Finally, the analysis of gold content on the
catalyst containing IL ([BMIM] PF6) revealed no loss of metal from
this IL after three reuses.
In striking contrast, the greater stability and permanency of
[AuCN2]− in the IL [BMIM] PF6 is reflected in the fact that neither
loss of activity nor loss of selectivity was seen after three consec-
utive uses (Fig. 11). In this case, the DLS technique was again used
to assess the stability of gold particles that may form with KAuCN2

34
A. Corma et al. / Journal of Catalysis 259 (2008) 26–35
(A)
(A) Yield of cis + trans-cyclopropanation products versus time
(B)
Fig. 11. (A) Comparative conversions (C%) and yields (Y%) of cis/trans-cyclo-
propanocarboxylates against number of runs in the carbene transfer reaction from
EDA to styrene catalyzed with KAuCN2 in the ionic liquid [BMIM] PF6; (B) com-
parative Z-average of gold nanoparticles obtained by DLS technique after succes-
sive reuses in the cyclopropanation reactions of styrene catalyzed by KAuCN2 in
[BMIM] PF6.
(B) Yield of o-, m-, p-aromatic addition products versus time
tent revealed that no leaching occurred from the [BMIM] PF6 IL
after three uses.
4. Conclusion
In the present study, ionic gold complexes in different imida-
zolium ILs were used as recoverable and reusable homogeneous
catalysts in the cyclopropanation reaction of alkenes with ethyl-
diazoacetate to give cis- and trans-cyclopropanes. The following
trends were found: With few exceptions, the trans-isomer was the
major product of the reaction, and the stability and performance
of these gold salts is much higher in ILs than in typical solvents
used in this reaction (e.g., 1,2-dichloroethane, CH2Cl2, acetonitrile).
Nonetheless, salt decomposition also occurred in the IL [BMIM]
PF6, and gold nanoparticles were formed that could be stabilized
by the IL, avoiding rapid agglomeration, in contrast to what occurs
in conventional solvents, in which large gold metal agglomerates
are formed in the first run. This findings has two important impli-
cations: Small gold nanoparticles should be active for performing
the cyclopropanation reaction, and stable gold metal nanoparticles
can be obtained by decomposing the gold salts in determined ILs.
The activity of the small gold nanoparticles is demonstrated by
the fact that after three uses, the yields to cyclopropanocarboxy-
lates were much larger than those achieved when working with
organic solvents. Indeed, the easy workup protocol for conducting
gold-catalyzed cyclopropanations facilitates the reaction. Dissolving
ionic gold complexes in imidazolium ILs is a very simple strategy
(C) Yield of fumarate and maleate esters versus time
Fig. 10. Plots showing the yields of reaction products versus time in the cyclopropa-
nation reaction of styrene with EDA catalyzed by gold (0) nanoparticles in the ionic
liquid BMIM PF6. Z-average particle size: (a)
(d) 102 nm.
8 nm; (b) 22 nm; (c) 79 nm and
in the IL [BMIM] PF6. The Z-average of the nanoparticle entities
produced after initial use was 5.6 nm. These small particles re-
tained almost all of their original size after a second use (6.3 nm)
and, in contrast to the previous study, were smaller than 20 nm
after a third use (Fig. 11B). The good activity of these small gold
nanoparticles is demonstrated by the fact that after three uses the
yields to cyclopropanocarboxylates still exceeded those obtained
when working with organic solvents. Finally, analysis of gold con-

A. Corma et al. / Journal of Catalysis 259 (2008) 26–35
35
for converting a nonrecoverable catalytic system into a reusable
and recoverable catalytic system.
[18] (a) Y. Ito, M. Sawamura, T. Hayashi, J. Am. Chem. Soc. 108 (1986) 6405;
(b) C. González-Arellano, A. Corma, M. Iglesias, F. Sánchez, Chem. Commun.
(2005) 3451.
[19] A.S.K. Hashmi, Angew. Chem. Int. Ed. 44 (2005) 6990;
L. Zhang, S. Wang, J. Am. Chem. Soc. 128 (2006) 1442.
Acknowledgments
[20] M.P. Doyle, D.C. Forbes, Chem. Rev. 98 (1998) 911–933.
[21] H.N.C. Wong, M.Y. Hon, C.W. Tse, Y.C. Yip, J. Tanko, T. Hudlicky, Chem. Rev. 89
(1989) 165.
[22] M.P. Doyle, McKervey, T. Ye, Modern Catalytic Methods for Organic Synthesis
with Diazo Compounds: From Cyclopropanes to Ylides, Wiley, New York, 1998.
[23] M.R. Fructos, T.R. Belderrain, M.C. Nicasio, S.P. Nolan, H. Kaur, M.M. Diaz-
Requejo, P.J. Pérez, J. Am. Chem. Soc. 126 (2004) 10846.
[24] A.J. DelMonte, E.D. Dowdy, D.J. Watson, in: Organometallics in Process Chem-
istry, in: Topics in Organometallic Chemistry, vol. 6, Springer-Verlag, 2004,
pp. 97–122.
Financial support by the Dirección General de Investigación
Científica y Técnica of Spain (Project MAT2006-14274-C02-01) is
gratefully acknowledged. I.D. and T.R. thank the Consejo Supe-
rior de Investigaciones Científicas for I3-P fellowships. The authors
thank the referees for their useful comments.
Supplementary material
[25] G. Maas, Chem. Soc. Rev. 33 (2004) 183.
Supplementary material for this article may be found on Sci-
enceDirect, in the online version.
[26] K. Miki, K. Ohe, S. Uemura, J. Org. Chem. 68 (2003) 8505.
[27] J.S. Yadav, B.V.S. Reddy, P. Narayana Reddy, Adv. Synth. Catal. 346 (2004) 53.
[28] M.R. Fructos, T.R. Belderrain, P. de Fremont, N.M. Scott, S.P. Nolan, M.M. Diaz-
Requejo, P.J. Pérez, Angew. Chem. Int. Ed. 44 (2005) 5284.
[29] M. Brust, M. Walker, D. Bethell, D.J. Schiffrin, R. Whyman, J. Chem. Soc. Chem.
Commun. (1994) 801.
Please visit DOI: 10.1016/j.jcat.2008.07.009.
References
[30] T. Welton, Chem. Rev. 99 (1999) 2071.
[31] P. Wasserscheid, W. Keim, Angew. Chem. Int. Ed. 39 (2000) 3772.
[32] R. Sheldon, Chem. Commun. (2001) 2399.
[1] (a) C. Gonzalez-Arellano, A. Abad, A. Corma, H. García, M. Iglesias, F. Sanchez,
Angew. Chem. Int. Ed. 46 (2007) 1536;
[33] C.M. Gordon, Appl. Catal. A 222 (2001) 101.
[34] R. Giernoth, Top. Curr. Chem. 276 (2007) 1.
(b) A. Corma, C. Gonzalez-Arellano, M. Iglesias, S. Perez-Ferreras, F. Sanchez,
Synlett 11 (2007) 1771.
[35] R.D. Rogers, K.R. Sheldon, J. Am. Chem. Soc. 125 (2003) 7480.
[36] N.V. Plechkova, K.R. Seddon, in: P. Tundo, A. Perosa, F. Zecchini (Eds.), Methods
and Reagents for Green Chemistry, Wiley, Hoboken, 2007.
[37] M. Maase, in: P. Wasserscheid, T. Welton (Eds.), Ionic Liquids in Synthesis, sec-
ond ed., Wiley–VCH, Weinheim, 2008, pp. 663–687.
[2] (a) S. Ma, S. Yu, Z. Gu, Angew. Chem. Int. Ed. 45 (2006) 200;
(b) V. Mamane, T. Gress, H. Krause, A. Fürstner, J. Am. Chem. Soc. 126 (2004)
8654.
[3] M.H. Suhre, M. Reif, S.F. Kirsch, Org. Lett. 7 (2005) 3925.
[4] A. Corma, C. Gonzalez-Arellano, M. Iglesias, F. Sanchez, J. Catal. 238 (2006) 497;
A. Corma, E. Gutierrez-Puebla, M. Iglesias, A. Monge, S. Perez-Ferreras,
F. Sanchez, Adv. Synth. Catal. 348 (2006) 1899.
[38] N.V. Plechkova, K.R. Seddon, Chem. Soc. Rev. 37 (2008) 123.
[39] H. Zao, S.V. Malhotra, Aldrichim. Acta 35 (2002) 75.
[40] (a) J. Durand, E. Teuma, F. Malbosc, Y. Kihn, M. Gómez, Catal. Commun.
(2008) 273;
9
[5] S. Carrettin, M.C. Blanco, A. Corma, A.S.K. Hashmi, Adv. Synth. Catal. 348 (2006)
1283.
(b) J. Kraemer, E. Redel, R. Thomann, C. Janiak, Organometallics 27 (2008) 1976;
(c) X. Yang, Z. Fei, D. Zhao, W.H. Ang, Y. Li, P.J. Dison, Inorg. Chem. 47 (2008)
3292;
(d) G. Machado, J.D. Scholten, T. da Vargas, S.R. Teixeira, L.H. Ronchi, J. Dupont,
Int. J. Nanotechnol. 4 (2007) 541;
[6] (a) R.A. Widenhoefer, X. Han, Eur. J. Org. Chem. (2006) 4555;
(b) N. Nishima, Y. Yamamoto, Angew. Chem. Int. Ed. 45 (2006) 3314.
[7] (a) Y. Fukuda, K. Utimoto, H. Nozaki, Heterocycles 25 (1987) 297;
(b) T.E. Müller, Tetrahedron Lett. 39 (1998) 5961;
(c) M.R. Luzung, J.P. Markham, F.D. Toste, J. Am. Chem. Soc. 126 (2004) 10858;
(d) D. Kadzimirsz, D. Hildebrandt, K. Merz, G. Dyker, Chem. Commun. (2006)
661.
(e) C. Vincenzo, N. Angelo, A. Monopoli, A. Damascelli, I. Eliana, N. Cioffi, J.
Organomet. Chem. 692 (2007) 4397;
[8] (a) For recent reviews see: A.S.K. Hashmi, Gold Bull. 37 (2004) 51;
(b) A. Höffmann-Roder, N. Krause, Org. Biomol. Chem. 3 (2005) 387;
(c) A. Arcadi, S. Di Giuseppe, F. Marinelli, E. Rossi, Curr. Org. Chem. 8 (2004)
795.
(f) M. Pedro, J. Dupont, Chem. Eur. J. 13 (2006) 32;
(g) D. Astruc, F. Lu, J.R. Aranzaes, Angew. Chem. Int. Ed. 44 (2005) 7852;
(h) M. Antonietti, D. Kuang, B. Smarsly, Y. Zhou, Angew. Chem. Int. Ed. 43
(2004) 4988;
[9] E. Mizushima, T. Hayashi, M. Tanaka, Org. Lett. 5 (2003) 3349.
[10] A. Corma, P. Concepción, I. Domínguez, V. Fornés, María J. Sabater, J. Catal. 251
(2007) 39.
(i) H. Itoh, K. Naka, Y. Chujo, J. Am. Chem. Soc. 126 (2004) 3026.
[41] (a) D.E. Koppel, J. Chem. Phys. 57 (1972) 4814;
(b) B.J. Frisken, Appl. Opt. 40 (2001) 4087;
[11] (a) X. Zhang, A. Corma, Chem. Commun. (2007) 3080;
(b) X. Zhang, A. Corma, Dalton Trans. 3 (2008) 397.
[12] E. Genin, P.-Y. Toullec, S. Antoniotti, C. Brancour, J.-P. Genet, V. Michelet, J. Am.
Chem. Soc. 128 (2006) 3112.
[13] A. Comas-Vives, C. Gonzalez-Arellano, A. Corma, M. Iglesias, F. Sánchez, J. Am.
Chem. Soc. 128 (2006) 4756.
[14] A. Corma, M. Boronat, S. Gonzalez, F. Illas, Chem. Commun. (2007) 3371.
[15] A. Corma, C. Gonzalez-Arellano, M. Iglesias, F. Sanchez, Chem. Commun. (2005)
3451.
[16] (a) A. Corma, C. Gonzalez-Arellano, M. Iglesias, F. Sánchez, Angew. Chem. Int.
Ed. 46 (2007) 7820;
(c) P.A. Hassan, S.K. Kulshreshtha, J. Colloid Interface Sci. 300 (2006) 744;
(d) E.R. Sadiku, R.D. Anderson, J. Appl. Polym. Sci. 102 (2006) 166.
[42] G.C. Bond, C. Louis, D.T. Thompson, in: G. Hutchings (Ed.), Catalysis by Gold, in:
Catalytic Science Series, vol. 6, Imperial College Press, London, 2006.
[43] J.R. van Wazer, Phosphorous and Its Compounds, vol. 1, Wiley, New York, 1958.
[44] G.S. Fonseca, A.P. Umpierre, P.F.P. Fichtner, S.R. Teixeira, J. Dupont, Chem. Eur.
J. 9 (2003) 3263.
[45] X. Liu, et al., Appl. Catal. A 239 (2003) 279–286;
H.U. Blaser, A. Baiker, R. Prins, in: J.M. Fraile, B. García, J.I. García, J.A. Mayoral,
F. Figueras (Eds.), Heterogeneous Catalysis and Fine Chemicals, IV, in: Studies
in Surface Science and Catalysis, 1997, p. 571.
(b) A.M. Caporusso, L. Antonella, E. Schiavi, G. Martra, G. Vitulli, P. Salvadori,
J. Organomet. Chem. 690 (2005) 1063.
[17] (a) M. Haruta, Catal. Today 36 (1997) 153;
[46] A.V. Plakhonyk, L. Ernt, R. Schmutzler, J. Fluorine Chem. 126 (2005) 27.
[47] K.R. Seddon, A. Stark, M.J. Torres, Pure Appl. Chem. 72 (12) (2000) 2275.
[48] L. Xu, W. Chen, J. Xiao, Organometallics 19 (2000) 1123.
[49] Z.L. Li, T.H. Wu, Z.J. Niu, W. Huang, H.D. Nie, Electrochem. Commun. 6 (2004)
44.
(b) J. Guzman, B.C. Gates, J. Am. Chem. Soc. 126 (2004) 2672;
(c) M. Valden, X. Lai, D.W. Goodman, Science 281 (1998) 1647;
(d) J. Guzman, S. Carrettin, J.C. González-Fierro, Y. Hao, B.C. Gates, A. Corma,
Angew. Chem. Int. Ed. 44 (2005) 4778;
[50] (a) U. Kreibig, M. Vomer, Optical Properties of Metal Clusters, Springer-Verlag,
Heidelberg, 1995;
(e) S. Biella, M. Rossi, Chem. Commun. (2003) 378;
(f) S. Carrettin, P. McMorn, P. Johnston, K. Griffin, G.J. Hutchings, Chem. Com-
mun. (2002) 696;
(b) G.P. Mitchell, C.A. Mirkin, R.L. Letsinger, J. Am. Chem. Soc. 121 (1999) 8122;
(c) Y. Voloktitin, J. Sinzing, L.J. Jong, G. Schmid, M.N. Vargaftik, I.I. Moiseev,
Nature 384 (1996) 621;
M.D. Hughes, Y.-J. Xu, P. Jenkins, P. McMorn, P. Landon, D.I. Enache, A.F. Carley,
G.A. Attard, G.J. Hutchings, F. King, E. Hugh Stitt, P. Johnston, K. Griffin, C.J.
Kiely, Nature 437 (2005) 1132.
(d) G. Schmid, Chem. Rev. 92 (1992) 1709;
(e) M. Kanehara, E. Kodzuka, T. Tenarishi, J. Am. Chem. Soc. 128 (2006) 13084.
[51] W. Haiss, N.T.K. Thanh, J. Aveyard, D.G. Fernig, Anal. Chem. 79 (2007) 4215.