Polyoxometalate-Gold(I)/H+ Complexes: Air-Stable, Efficient, Polyvalent, and Bifunctional Catalysts

Article abstract of DOI:10.1021/acs.organomet.5b00656

Gold(I)/H⁺-polyoxometalate hybrids, with general formula [POM]^x-[H]_x-1⁺[R₃PAu(MeCN)]⁺, were synthesized and fully characterized by ³¹P and ²⁹Si MAS or liquid NMR, FT-I

Full text of DOI:10.1021/acs.organomet.5b00656

Article
pubs.acs.org/Organometallics
+
Polyoxometalate-Gold(I)/H Complexes: Air-Stable, Efficient,
Polyvalent, and Bifunctional Catalysts
†
†
́ ́
Damien Hueber, Marie Hoffmann, Pierre de Fremont, Patrick Pale,* and Aurel
ien Blanc*,†
‡
,†
†
‡
Laboratoire de Synthes
Ligands a Architecture Contro
7070 Strasbourg, France
̀
e, Rea
́
ctivite
́
Organiques & Catalyse, Institut de Chimie de Strasbourg, and Laboratoire de Chimie des
de Strasbourg, 4 Rue Blaise Pascal,
̀
̂
́
lee, Institut de Chimie, UMR 7177−CNRS, Universite
́
6
*
S Supporting Information
+
ABSTRACT: Gold(I)/H -polyoxometalate hybrids, with
x−
+
+
general formula [POM] [H] [R PAu(MeCN)] , were
x−1
3
3
1
29
synthesized and fully characterized by P and Si MAS or
liquid NMR, FT-IR, and mass spectroscopy. All these
polyoxometalate gold(I) complexes exhibited excellent cata-
lytic activity and selectivity compared to standard homoge-
neous gold catalysts in a large range of reactions. The variation
of the ligand and the polyoxometalate nature around gold(I)
also highlighted the flexibility and the multifunctionality of
such catalysts, as well as their recyclability.
INTRODUCTION
Scheme 1. Different Types of Polyoxometalates-Based
Inorganic/Organic Hybrids
■
Polyoxometalates (POMs) refer to polyanionic clusters formed
by the assembly of early-transition-metal oxide building blocks
MO ], typically tungsten(IV) or molybdenum(VI). Since their
[
x
discovery two centuries ago, thousands of structurally different
1
POMs with various properties have been reported, impacting
2
all the areas of chemistry and notably catalysis. In 2010, this
large family of POM compounds have been categorized in three
3
distinct groups according to the polyoxoanion structure: clusters
without additional heteroatoms constitute the isopolyanion class,
clusters including heteroatoms such as P and Si as templates
form the heteropolyanion class, and the molybdenum-reduced
POM clusters represent a third class.
This first classification did not take into account the pre-
+
+
+
sence of inorganic (H , Na , ...) or organic cations (NR , ...)
types of organic−inorganic hybrids based on POMs are
emerging, and a third and new class has to be considered, in
which the organic part was associated with the inorganic POM8s
throughout coordination to metallic cations (type III, Scheme 1).
The combination of POMs with coinage-metal/organic
ligand complexes provides good examples of type III hybrid
class, and various structures have been formed via, notably,
three-component (cationic coinage metal, ligands, and POMs)
4
compensating the large negative charge of POMs. The nature
of the cationic counterpart is however a crucial factor of the
POMs’ physical or chemical properties. Among the numerous
possibilities, the combination of the inorganic polyanionic
POMs with organic entities as counterion provides a unique
and tremendous opportunity to fine-tune the POMs’ proper-
ties. As a result, the formation of multifunctional monomeric
or polymeric organic−inorganic hybrids can join the inherent
properties of both interdependent parts and go further toward
new physical or chemical properties. Inspired by the
classification of organic−inorganic materials, this latter class
of organic-POM-based material has been more recently
subdivided in two types. In the type I hybrids, the organic
and inorganic parts are associated by noncovalent interactions
electrostatic, hydrogen bonding, ...). Type II hybrids involve
4
9
hydrothermal synthesis. Interestingly, the charge-compensat-
ing coinage-metal cations could not only serve as structural
linkers between the organic ligands and inorganic POMs
5
10
6
but also bring their Lewis acid properties to the hybrids. In
this area, we have recently demonstrated that gold(I) cations
were effectively very good candidates in constructing type III
7
1
1,12
organic−inorganic complexes.
Moreover, depending on the
applied stoichiometry (x, Scheme 1), excellent and new gold
(
stronger bonding interactions with covalent bonds and usually
require the direct substitution of [MO ] of POMs (Scheme 1).
Received: July 31, 2015
x
However, in the continuously growing field of POMs, new
©
XXXX American Chemical Society
A
DOI: 10.1021/acs.organomet.5b00656
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
+
x−
reactivities have been obtained using such [(Ligand)Gold(I)-
formula for gold(I)/H -polyoxometalates 1−6: [POM] [H]+x−1-
3
+
x−
+
(
Solvent)] /POM hybrids compared to classical homoge-
[R PAu(MeCN)] (Table 1). This technique also showed the
x
neous gold catalysts (Scheme 2).
expected molecular peak for each anionic POM species [M −
+
yH ] (Table 1), as well as the presence of remaining protons on
Scheme 2. POM/Gold(I)-Catalyzed Stereoselective
these complexes. However, this analysis also allowed the
detection of minor amounts of higher order complexes
Rearrangement of Propargylic gem-Diesters into
12
x−
+
+
(E)-Enones
[POM] [H] [R PAu(MeCN)]2 (see the Supporting In-
x−2 3
formation).
NMR analysis in DMSO-d clearly proved the preservation
6
of the different POM structures (bolded chemical shifts in
Table 1) because the chemical shift of the central atoms of
2
9
these complexes exhibited the same value (−86.1 ppm Si
NMR for H SiW O ) for the Si-based POM and very similar
4
12 40
ones for the P-based POMs compared to the corresponding
In the present contribution, we reported the preparation
31
initial HPAs (−15.6 vs −15.3 ppm P NMR for H PW O ,
3
12 40
and characterization of new and stable polyoxometalate-gold(I)
complexes and their catalytic activities in various gold-catalyzed
reactions. By varying the ligand and polyoxometalate nature
around gold, we also demonstrated the flexibility and the multi-
functionality of such catalysts.
−
4.1 vs −3.8 ppm for H PMo O , −13.3 vs −12.8 ppm for
3
12 40
31
H P W O ). P NMR also allowed the identification of the
6
2
18 62
+
cationic gold(I) species [(Ph P)Au(MeCN)] , [(Me P)Au-
3
3
+
+
(
MeCN)] , and [(2,4-ditBuPhO) P)Au(MeCN)] with signals
3
at respectively 26.7 ± 0.1, −12.5, and 85.6 ppm, in close
15
analogy with known 31P NMR shifts from the literature. It is
RESULTS AND DISCUSSION
■
+
noteworthy to mention that traces of the inactive [(R P) Au]
3
2
Preparation and Characterization of Au(I)-POM
Complexes. In order to evaluate and compare the subsequent
catalytic activities of Au-POM complexes, typical POM
structures with different central atoms (Si, P) and metal oxides
31
species were also observed on each P spectra (1−4%).
+
POM-H /gold(I) 1−6 were further analyzed by ESI-MS in
both positive mode and FT-IR, confirming each complex
structure (see the Supporting Information).
(
W, Mo) and exhibiting different sizes and charges were
selected and combined with gold(I) carrying phosphine or
Before our first isolation and characterization of gold(I)-
10
polyoxometalates, the in situ combination of HPAs and
phosphite ligands (scheme in Table 1). Three Keggin- and one
+
Ph PAuMe in methanol has been successfully used in hydration
Dawson-type POM-H /PPh Au(I) complex, 1−4 (entries 1−4),
3
3
16
and/or hydroamination reactions and in the cyclization of
,6-enynes, but without characterization of the actual catalysts.
were obtained by simply mixing overnight in acetonitrile at
room temperature the corresponding heteropolyacids (HPAs,
conjugate acids of POM, respectively H SiW O , H PW O ,
17
1
To investigate the nature of the active species in such reactions,
we applied our methodology to prepare the complex possibly
responsible for these catalyses. This complex might be very
4
12 40
3
12 40
H PMo O , H P W O ) with an equimolar amount of the
3
12 40
6
2
18 62
commercially available methyl(triphenylphosphine)gold(I)
1
3
similar to or the same as complex 1, obtained from H SiW O
complex. To compare the influence of the ligands on the
4
12 40
+
and PPh Au(I)Me (see Table 1). We thus mixed these two
catalysis, two other SiW O -H /Au(I) complexes, 5 and 6
3
1
2
40
entities in methanol instead of acetonitrile and isolated a new
(
entries 5 and 6), were synthesized starting from respectively
methyl(trimethylphosphine)gold(I) and methyl[tris(2,4-di-
solid. The new complex 1
was then studied as previously.
MeOH
However, solid-state 31P NMR was used to avoid the possible
effect of deuterated solvent on the catalyst structure (Figure 1).
In contrast to 1 prepared in acetonitrile, for which no differences
between solid and liquid 31P NMR could be noticed, 1MeOH
tert-butylphenyl)phosphite]gold(I) complexes prepared from
14
reported procedures. All complexes 1−6 were easily obtained
in quantitative yields along with the formation of methane as the
sole byproduct. Such complexes remained highly insoluble in
classic organic solvents except in DMSO.
exhibited new signals, one at 27.6 ppm, which could be assigned
+
The unambiguous characterization of each complex was
to the [Ph
3
PAu(MeOH)] cation, and another one at 44.4 ppm,
3
1
29
+
18
achieved by combining liquid P and Si NMR and mass
spectroscopy using electrospray ionization technique (ESI-MS).
In its negative mode, the latter afforded the following general
which corresponded to the known [(Ph P) Au] cation.
3 2
The important amount of the latter suggested that a large
part of gold(I) was reduced to gold(0) during the reaction.
Table 1. Preparation of Various Gold(I)-Polyoxometalates 1−6 in Acetonitrile
a
NMR DMSO-d chemical shift
6
29Sia
31Pa
mass peaks M − yH
+
b
yield (%)
complex
[
[
[
[
[
[
H SiW O ][Ph PAu(MeCN)]
1
2
3
4
5
6
−86.1
26.8
1667.12 (y = 2)
1668.60 (y = 2)
1040.86 (y = 2)
1607.95 (y = 3)
1574.13 (y = 2)
1859.30 (y = 2)
99
96
96
98
99
99
3
12 40
3
H PW O ][Ph PAu(MeCN)]
26.6, −15.6
26.6, −4.1
26.7, −13.3
−12.9
85.6
2
12 40
3
H PMo O ][Ph PAu(MeCN)]
2
12 40
3
H P W O ][Ph PAu(MeCN)]
5
2
18 62
3
H SiW O ][Me PAu(MeCN)]
−86.1
−86.1
3
12 40
3
H SiW O ][(2,4-ditBuPhO) PAu(MeCN)]
3
12 40
3
a
b
In ppm. m/z observed on the ESI negative mode spectra.
B
DOI: 10.1021/acs.organomet.5b00656
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Figure 1. Comparison of 31P MAS NMR (rotation = 12 000 Hz) of 1 and 1MeOH, [H SiW O ][PPh Au(Solvent)], prepared respectively in
3
12 40
3
acetonitrile and methanol.
The triphenylphosphine liberated in this reduction process
could then form a relatively high amount of the stable but
filtration of the remaining residues and then letting the solutions
stand for 1 week to 6 months. The X-ray diffractions of the
so-obtained crystals furnished six structures, which do not
correspond to the expected structures (Figure 2). These
structures exhibit the expected POM cores corresponding to
the initial complexes 1−5, but their charges are fully balanced by
gold(I) cations, except for the Dawson POM-Au complex 4a,
+
inactive [(Ph P) Au] complex. Moreover, 1MeOH’s spectrum
3
2
exhibited an extra peak at 39.4 ppm that could presumably be
assigned to the formation of higher coordinate complexes
+
[
(Ph P) Au] (x = 3, 4) due to the release of an excess of
3
x
19
triphenylphosphine. These results, gained from the preparation
of complex 1MeOH in methanol, clearly questioned the nature and
the amount of active species in some previously reported gold-
+
which retains one proton under the form H O . They thus
5
2
represent new complexes. The complexes 1a, 2a, 3a, and 5a are
15,16
x−
catalyzed transformations.
saturated Keggin polyanions with the formula [POM] [R PAu-
3
+
X Ray Structures of Au(I)-POM Complexes. Due to the
high insolubility of complexes 1−6, various attempts to
crystallize them failed, even in DMSO, the solvent used for
liquid NMR spectroscopy. However, we were able to form
crystals by refluxing complexes 1−5 in acetonitrile, followed by
(MeCN)] . Surprisingly, another saturated complex, 1b, could
x
also be crystallized from complex 1. The X-ray diffraction
+
structure revealed the association of two [(Ph P) Au] along
3
2
+
with two [Ph PAu(MeCN)] around the anionic POM unit.
3
This unexpected structure could arise only from the formation
C
DOI: 10.1021/acs.organomet.5b00656
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Figure 2. Ellipsoid representations of XRD structures of saturated polyoxometalate-gold(I) complexes 1a, 1b, 2a, 3a, 4a, and 5a (hydrogen atoms
have been omitted for clarity).
Scheme 3. POM/Gold(I) Exchanges Leading to the More Soluble Hybrid Complex 1a from Hybrid 1
+
of trace amounts of [(Ph P) Au] (∼1%, see Figure 1) during
is fully counterbalanced by gold complex cations (Scheme 3).
In the particular case of the Dawson complex 4a coming from 4,
3
2
the preparation of complex 1 in acetonitrile. It is noteworthy
that in structures 1a and 1b, the central silicon atom holds a
special crystallographic position with eight half-oxygen atoms
the steric hindrance due to the accumulation of [Ph PAu-
3
+
6−
(MeCN)] cations around the polyoxoanion [P W O ] seemed
2
18 62
20
around it; such disorder is classical for Keggin anions.
to allow for only five of them and led to the presence of residual
+
These unexpected results revealed that upon heating for
recrystallization, these complexes underwent successive ex-
changes between POMs and gold cations, ultimately leading to
the more soluble hybrid complex in which the POM unit charge
acidity from [H O ] .
5
2
To provide more structural details, we have selected charac-
teristic bond lengths, interatomic distances, and angles from
the X-ray structures of the gold(I)-POM complexes (Table 2).
D
DOI: 10.1021/acs.organomet.5b00656
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Table 2. Selected Bond Lengths or Distances (Å) and Angles (deg) for Gold(I)-POM Complexes 1a,b−5a
a
b
complex
SiW O ] [Ph PAu(MeCN)]+ (1a)
Au−P
Au−N
P−Au−N
Au···Au
Au···O
3.14(6)
4
−
[
[
2.23(0)−2.23(7)
2.02(4)−2.08(3)
174.2−178.1
3.74(2)
8.16(6)
12
40
3
4
+
4−
+
2
c
c
SiW O ] [Ph PAu(MeCN)] [(Ph P) Au] (1b)
2.22(6), 2.30
2.03(2), −
175.0(2)
3.21(4)
5.17(6)
3.18(1)
12
40
3
2
3
2
c
c
1
72.8(9)
3
−
+
[
[
[
[
PW O ] [Ph PAu(MeCN)] (2a)
2.23(4)
2.03(9)
174.1(7)
6.45(7)
6.44(5)
8.02(8)
12
40
3
3
+
3
−
PMo O ] [Ph PAu(MeCN)] (3a)
2.23(8)
2.04(6)
174.5(9)
3.17 (2)
12
40
3
3
6
−
+
P W O ] [Ph PAu(MeCN)] , 4[MeCN], 2[O] (4a)
2.22(0)−2.23(6)
2.01(4)−2.08(1)
2.03(9)−2.06(7)
169.7(8)−176.9(3)
169.0(7)−170.4(7)
3.20(4)−3.49(3)
2.90(7)−3.30(8)
2
18 62
3
5
4
−
+
4
d
SiW O ] [Me PAu(MeCN)] , 2[MeCN] (5a)
2.22(7)−2.23(6)
3.06(9)
12
40
3
a
b
c
+
d
10
10
Shortest gold−gold distances. Shortest gold−oxygen distances. Data for [(Ph P) Au] . d −d interaction.
3
2
Concerning the hybrids 1a, 1b, 2a, 3a, and 4a, bearing
Table 3. Comparison of the Catalytic Activity
+
x−
+
+
[
Ph PAu(MeCN)] cations, the P−Au (2.22−2.23 Å) and
of [POM] [H] [R PAu(MeCN)] 1−6 vs the
3
x−1
3
Au−N (2.02−2.08 Å) distances were in agreement with various
Homogeneous Version in Two Representative
Gold-Catalyzed Transformations
21
gold(I)/acetonitrile complexes described in the literature.
The P−Au−N angles (174.1−176.9°) also followed the general
+
tendency for a standard [Ph PAu(MeCN)] cation, except in
3
the unsaturated Dawson complex 4a, where one of the angle
values was smaller than usual (169.0°). Such deviation could be
+
attributed to steric interactions between all [Ph PAu(MeCN)]
3
cations around the POM and was confirmed by the Au−N−C
angle of 148.6° on this cation, instead of 170−180° for the
others. The Au···Au distances in these complexes were between
3
.74 and 8.02 Å, indicating the absence of significant auro-
22
philic interactions. On the contrary, the X-ray structure of
complex 5a showed an effective d −d interaction between
two [Me PAu(MeCN)] (Au···Au = 3.06 Å), probably induced
1
0
10
+
3
by the smaller size of the PMe ligand compared to PPh .
3
3
In addition, the P−Au−N angle in this complex deviated
10
10
significantly from linearity (169−170°) due to this d −d
a
b
bonding.
Isolated yields. Starting material was recovered untouched.
c
1
Interestingly, the distances between the rising oxygen atoms
of POMs and gold atoms could be observed between 2.90 and
Estimated conversion based on the H NMR analysis of the crude
mixture.
3
.49 Å. These values clearly differ from those reported for
23
covalent Au−O bonds (e.g., Au−OH = 2.078 Å, Au−OAc =
entries 2−7 and 10−15), while a simple HPA did not promote
transformations (entries 8 and 16). For both transformations,
the influence of the POM unit was evaluated with the catalysts
1−4, containing the same triphenylphosphine ligand.
2
4
2.040 Å ) as well as from those without significant interaction
between Au and oxygen (e.g., ≥4.0 Å). They readily compared
to the distances observed for a POM-Au O cluster reported by
4
10
i
6
−
Nomiya (3.38−3.49 Å). Therefore, the present values probably
Although the basicity of the POM anions ([P W O ]
>
2
18 62
4−
3−
3−
28
4
−
3−
reflect the weakness of POMs ([SiW O ] , [PW O ] ,
1
2
40
12 40
[SiW O ] ≥ [PMo O ] > [PW O ] ) could be
linked to their catalytic activity, the obtained results showed
a more complex pattern. Indeed, the less basic gold counterion
1
2
40
12 40
12 40
3−
6−
[
PMo O ] , and [P W O ] ) as counterions associated
12 40 2 18 62
with gold atoms compared to those classically used in gold
−
14a
−
−
3
−
catalysis, such as OTf , SbF , and BF . Such low inter-
6
4
[PW O ] afforded excellent yields, 94% and 85%, in both
12 40
actions between the gold cation and its counterion foreshadow
reactions compared to the more basic silicotungstate anion
(respectively 86% and 75%; entries 3 and 11 vs 2 and 10).
Interestingly, catalyst 2 provided better yields than the homo-
geneous versions (Table 3, entries 3 vs 1 and 11 vs 9). The catalyst
potent catalytic activities of these hybrid polyoxometalate-gold(I)
25
complexes, 1−6.
Catalytic Activities. We evaluated the catalytic ability of
new POM-LAu /H complexes 1−6 in two representative
3−
n
x−n
[PMo O ] , 3, exhibited an ambivalent catalytic activity,
12 40
gold-catalyzed reactions both performed in CH Cl : the
affording the highest yield for the N-propargylcarboxamide
cyclization (96%) and the lowest amount of 8 (68%) in the
enyne ester rearrangement (Table 3, entries 4 and 12). In contrast,
the Wells−Dawson complex 4 exhibited an excellent activity,
disconnected to its relative basicity. Despite being the more basic
anion and thus considered as the less active catalyst, high yields
were nevertheless obtained, similar for both reactions and equal
to the homogeneous versions. These results may be attributed to
the larger size of the Dawson anion compared to the Keggin
complexes 1−3 (Table 3, entries 5 and 13 vs entries 1 and 9).
The electronic properties of ligands also played a crucial role
2
2
2
6
rearrangement of enyne esters to cyclopentenones (Table 3,
2
7
entries 1−8) and the cyclization of N-propargylcarboxamides
(
Table 3, entries 9−16). In homogeneous conditions, the
first transformation led to cyclopentenone 8 in 92% yield
starting from enyne acetate 7 with Ph PAuCl/AgSbF6 as
3
catalyst (Table 3, entry 1), while the cyclization of N-propargyl
benzamide 9 catalyzed by Ph PAuNTf afforded dihydroox-
3
2
azole 10 in 91% yield (Table 3, entry 9). These products and
yields were obtained using 2 mol % of gold catalyst. These data
served as reference values for our own investigations with our
heterogeneous catalysts.
29
in gold(I) catalysis. We thus studied the effect of the nature
4
−
With all of them, we were pleased to observe the forma-
of the phosphorus ligands associated with the [SiW O ]
12 40
tion of desired products in the same range of yields (Table 3,
anion (complexes 1, 5, and 6). Without surprise, the reactivity
E
DOI: 10.1021/acs.organomet.5b00656
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
increased with the electron-withdrawing characters of the
considered ligands (P(O-2,4-ditBuPh)3 > PPh3 > PMe3).
Complex 5, with a trimethylphosphine ligand, was clearly the
less reactive catalyst, requiring a long reaction time for both
transformations (Table 3, entries 6 and 14), while the catalyst 6
exhibited faster reaction rates and similar yields compared to 1
we successfully obtained the same ketone using the same
catalyst loading (2.5 mol %) under the same conditions. Very
interestingly, comparable and even better yields were achieved
with all our Au-POM catalysts (Table 4, entries 2−7, 70−89%).
A control experiment with a simple HPA did not afford the
ketone 13 but the N-tosylbuta-2,3-dienamine as a single pro-
duct in 60% yield coming from the acid-catalyzed aza-Cope
rearrangement (Table 4, entry 8). The latter set of results
clearly highlights the capacity of complexes 1−6 to act as
bifunctional catalysts, their gold counterpart ensuring the
cyclization process and the protic POM the hydrolysis of the
so-formed product.
(Table 3, entries 7 and 15 vs entries 2 and 10).
31
In an attempt to take benefit of residual Brønsted acidities
+
present on our gold(I)/H -polyoxometalate catalysts 1−6,
we applied them to a sequential gold/acid-catalyzed process.
We selected the gold-catalyzed aza-Prins cyclization/enol
ether hydrolysis described by Rhee as a benchmark reaction
3
0
(
Table 4). In the homogeneous version using the strongly
With these results in hand, we focused our attention on
complex 1 with the aim to evaluate its flexibility regarding
various gold transformations (Table 5). We first evaluated 1 as
Table 4. Bifunctional Catalytic Activity of
32
33
x−
+
+
[
POM] [H] [R PAu(MeCN)] 1−6 in Tandem
catalyst for alkynyloxirane and Meyer−Schuster rearrangements
of respectively compounds 14 and 16 (Table 5, entries 1−6).
Excellent yields were obtained in both rearrangements: furan
x−1
3
Gold-Catalyzed Aza-Prins Cyclization/Enol Ether
Hydrolysis of 11
15 was quantitatively isolated from alkynyloxirane 14, and
α,β-unsaturated ketone 17 was formed in 73% yield from
propargylic alcohol 16 (entries 2 and 5). Both heterogeneous
versions were once again as efficient and as rapid as the cor-
responding homogeneous versions (entry 2 vs 1 and entry 5 vs 4).
Acid catalysis was examined in both rearrangements and proved
very inefficient, as only minor amounts of products 15 and 17
were formed upon prolonged reaction time (entries 3 and 6).
We also engaged the diyne diether 18 in a double
hydroarylation−cationic cyclization process, reported to be
catalyzed by both gold and acidic medium and providing
a
entry
1
gold(I) catalyst (2.5 mol %)
time (h) yield (%)
b
26
(C F ) PAuSbF then p-TsOH/H O (10
2 then 24
80
6
5
3
6
2
mol %)
2
3
4
5
6
7
8
SiW-Ph PAu/H 1
1
70
75
86
67
68
89
3
3
PW-Ph PAu/H 2
0.5
0.5
1
3
2
PMo-Ph PAu/H 3
3
2
34
P W-Ph PAu/H 4
dioxafluoranthene. However, the reported homogeneous
(PPh AuSbF 5 mol %/triflic acid 30 mol %) as well as our
2
3
5
SiW-Me PAu/H 5
3
3
3
3
6
SiW-(ArO) PAu/H 6
0.5
24
heterogeneous version (POM-Au/H 1, 5 mol %) failed to
3
3
3
c
H SiW O ·25H O
−
furnish the expected compound. Only the product of double
hydroarylation (19) was obtained in both cases with similar
yields (Table 5, entries 7 and 8). A control experiment showed
that only gold cations were responsible for the hydroarylation
reaction (entry 9).
4
12 40
2
a
b
Isolated yields. Two-step yield using homogeneous conditions.
c
N-Tosylbuta-2,3-dienamine was isolated in 60% yield.
electron-withdrawing pentafluorotriphenylphosphine gold(I)
complex, the cyclic enol ether 12 was formed from N,O-acetal
1 and was subsequently hydrolyzed by addition of para-
toluenesulfonic acid and water to furnish the ketone 13 in
0% after 24 h (Table 4, entry 1). In a one-step procedure,
Finally, we investigated the cycloisomerization−hydroalkoxylation
35
1
of propargyl alcohols described by Krause. In this tandem
process, the 4-phenylbut-3-yn-1-ol 20 furnished the tetrahy-
drofuranyl ether 21 in 67% yield using a mixture of PPh AuBF
8
3
4
Table 5. Catalytic Activity of [H SiW O ][Ph PAu(MeCN)] (1) Relative to Various Gold-Catalyzed Transformations
3
12 40
3
a
Isolated yields.
F
DOI: 10.1021/acs.organomet.5b00656
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
and p-TsOH in ethanol (entry 10). With our bifunctional
catalyst 1, compound 20 also rearranged to ether 21 in 63%
yield, despite the longer reaction time (entry 11). This result
once again clearly highlighted the multifunctionality of our
catalyst 1, which was confirmed by the test with the silicotungstic
acid and gold catalyst alone (entries 12 and 13).
multivalent POM hybrids of type III. Moreover, the simple
and efficient synthesis of the catalysts described herein offered
new insights in the design of polyoxometalate-based inorganic/
organic hybrids.
EXPERIMENTAL SECTION
General Procedure for the Preparation of Polyoxometalate-
Gold Complexes (1−6). To a suspension of R PAuMe (1.0 equiv) in
■
Catalyst Recycling. The recyclability was examined by per-
forming several times the cyclization of N-propargylcarboxamide
3
acetonitrile (2.5 mL) was added at room temperature a solution of
heteropolyacid hydrate (1.0 equiv) in MeCN (2.5 mL). The resulting
mixture was stirred at room temperature for 16 h (formation of gray-
purple precipitate). The solvent was then evaporated and dried under
high vacuum for 24 h to give the desired catalyst.
9, which afforded the 5-methyl-2-phenyloxazole 10 at room
temperature with 2 mol % of our heterogeneous gold catalysts
3
6
(
see Table 3). The catalysts could be recovered from the
reaction mixture either by filtration over a nylon membrane or
decantation. To avoid possible ligand and metal oxide effects, we
[
Ph PAu(MeCN)][H SiW O ] (1): 99% (349.5 mg, 0.104 mmol
3 3 12 40
focused on the tungstic polyoxometalates 1, 2, and 4 associated
from 0.105 mmol of (PPh )AuMe and H SiW O ); gray-purple
3 4 12 40
+
29
31
with the [Ph PAu(MeCN)] cation for this study.
solid; Si MAS+1H DEC NMR (300 MHz, solid) δ −84.2 ppm; P
29
3
A clear influence of the POM nature was noticed (Figure 3).
Indeed, despite the excellent yield obtained in the first run
MAS NMR (400 MHz, solid) δ 44.4 (1%), 28.8 (99%) ppm; Si
31
NMR (600 MHz, DMSO-d ) δ −86.1; P NMR (300 MHz, DMSO-
6
d ) δ 42.9 (1%), 26.8 (99%) ppm; mp = 200−205 °C (dec); IR (neat)
6
−1
ν
(cm ) 686, 737, 877, 906, 967, 1017, 1102, 1435, 1480, 1682.
max
[
Ph PAu(MeCN)][H PW O ] (2): 99% (348.2 mg, 0.104 mmol
3 2 12 40
from 0.105 mmol of (PPh )AuMe and H PW O ); gray-purple
3
3
12 40
solid; 31P NMR (300 MHz, DMSO-d ) δ 42.9 (4%), 26.6 (96%),
6
−1
−
15.6 ppm; mp = 226−236 °C (dec); IR (neat) νmax (cm ) 686, 886,
73, 1076, 1104, 1436, 1481, 1715.
Ph PAu(MeCN)][H PMo O ] (3): 99% (244.0 mg, 0.107 mmol
9
[
3
2
12 40
from 0.108 mmol of (PPh )AuMe and H PMo O ); gray-purple
3
3
12 40
solid; 31P NMR (300 MHz, DMSO-d ) δ 42.9 (4%), 26.6 (96%),
6
−1
−
4.1 ppm; mp = 220−230 °C (dec); IR (neat) νmax (cm ) 687, 871,
54, 1059, 1104, 1435, 1482, 1610.
Ph PAu(MeCN)][H P W O ] (4): 98% (505.3 mg, 0.104 mmol
9
[
3
5
2
18 62
3
7
from 0.106 mmol of (PPh )AuMe and H P W O ); gray-white
3
6
2
18 62
solid; 31P NMR (300 MHz, DMSO-d ) δ 42.9 (1%), 26.7 (96%),
6
−1
−
13.5 ppm; mp = 217−227 °C (dec); IR (neat) νmax (cm ) 685, 736,
97, 957, 1025, 1088, 1436, 1481.
Me PAu(MeCN)][H SiW O ] (5): 99% (682.4 mg, 0.213 mmol
8
[
3
3
12 40
from 0.215 mmol of (PMe )AuMe and H SiW O ); dark gray solid;
3
4
12 40
31
29
x−
Figure 3. Recycling of [POM] [H]+ [Ph PAu(MeCN)] 1, 2, and
4
+
Si NMR (600 MHz, DMSO-d ) δ −86.1; P NMR (300 MHz,
6
x−1
3
DMSO-d ) δ 9.0 (14%), −12.9 (86%) ppm; mp = 215−225 °C (dec);
catalysts in the cyclization of N-(propargyl)benzamide 9.
6
−1
IR (neat) ν (cm ) 875, 906, 969, 1019, 1310, 1416, 1619.
max
3
−
[(O(2,4-ditBuPh)
3 3
PAu(MeCN)][H SiW12O40] (6): 99% (320.3 mg,
(
>90%), catalyst 2, with the [PW O ] polyanion, did not
12 40
0
.086 mmol from 0.087 mmol of (P(OR) )AuMe and H SiW O );
3
4
12 40
31
express any catalytic activity in the second run. Switching to
catalyst 1, having the [SiW O ] anion, allowed performing
three runs with correct yields (respectively 75%, 72%, 54%)
before being almost deactivated (10% for run 4). Finally, the
best recycling ability was obtained with the Dawson catalyst 4
[P W O ] ). In this case, the catalytic activity was maintained
during the fourth run, although the reaction temperature needed
29
gray-purple solid; Si NMR (600 MHz, DMSO-d ) δ −86.1;
P
6
4−
1
2
40
NMR (300 MHz, DMSO-d ) δ 85.6 ppm; mp = 203−208 °C (dec);
6
−1
IR (neat) ν (cm ) 739, 910, 966, 1017, 1074, 1209, 1489, 1615,
max
2868, 2904, 2956.
6−
(
ASSOCIATED CONTENT
Supporting Information
The Supporting Information is available free of charge on the ACS
Publications website at DOI: 10.1021/acs.organomet.5b00656.
2
18 62
■
*
S
to be slightly increased to 45 °C to reach full conversion (80, 78,
3
1
8
0, and 71 for runs 1−4). P NMR and mass analysis in positive
mode of catalyst 4 recovered after the fourth run showed almost
Complete characterization data and experimental proce-
dures of organic compounds (PDF)
+
exclusively the presence of the inactive cation [(Ph P) Au] .
Thus, the increase of polyoxometalate charge and size seems to
stabilize gold cations and be beneficial for the recycling.
3
2
Crystal data and CIF file for 1a (CCDC 1407459) (CIF)
Crystal data and CIF file for 1b (CCDC 1407460) (CIF)
Crystal data and CIF file for 2a (CCDC 1407461) (CIF)
Crystal data and CIF file for 3a (CCDC 1407462) (CIF)
Crystal data and CIF file for 4a (CCDC 1407463) (CIF)
Crystal data and CIF file for 5a (CCDC 1407464) (CIF)
CONCLUSION
■
In the present study, we disclosed a simple and efficient
synthesis of new gold-polyoxometalate hybrids. We were able
to fully characterize them. Furthermore, we also have shown
that these new gold-polyoxometalate hybrids are reliable and
efficient catalysts for various gold-catalyzed transformations,
such as rearrangement, cyclization, and hydroarylation reactions.
In all these reactions, the Au-POM catalysts behave as
effectively, and sometimes more effectively, as the corresponding
AUTHOR INFORMATION
Corresponding Authors
E-mail: ppale@unistra.fr.
E-mail: ablanc@unistra.fr.
■
*
*
+
+
homogeneous catalysts. Their bifunctionality (Au /H ) has also
been highlighted and opens the way toward more sophisticated
Notes
The authors declare no competing financial interest.
G
DOI: 10.1021/acs.organomet.5b00656
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
+
(
2
15) (a) For [Ph PAu(MeCN)] , see: Tang, Y.; Yu, B. RSC Adv.
ACKNOWLEDGMENTS
3
■
+
012, 2, 12686−12689. (b) For [Me PAu(MeCN)] , see: de Silva, E.
3
We gratefully acknowledge the CNRS and the French Ministry
of Research. A.B. and D.H. thank the Agence Nationale de la
Recherche (Grant ANR-11-JS07-001-01 SyntHetAu), for
respectively financial support and a Ph.D. fellowship. M.H.
thanks the French Ministry of Research for a Ph.D. fellowship.
N.; Bowmaker, G. A.; Healy, P. C. J. Mol. Struct. 2000, 516, 263−272.
+
(
c) For [(O(2,4-ditBuPh) )PAu(MeCN)] , see: Zhdanko, A.;
3
̈
Strobele, M.; Maier, M. E. Chem. - Eur. J. 2012, 18, 14732−14744.
(16) (a) Mizushima, E.; Sato, K.; Hayashi, T.; Tanaka, M. Angew.
Chem., Int. Ed. 2002, 41, 4563−4565. (b) Mizushima, E.; Hayashi, T.;
Tanaka, M. Org. Lett. 2003, 5, 3349−3352. (c) Zhang, C.; Cui, D.-M.;
Yao, L.-Y.; Wang, B.-S.; Hu, Y.-Z.; Hayashi, T. J. Org. Chem. 2008, 73,
REFERENCES
■
7
(
811−7813.
(
1) (a) Miras, H. N.; Vila-Nadal, L.; Cronin, L. Chem. Soc. Rev. 2014,
3, 5679−5699. (b) Ivanova, S. ISRN Chem. Eng. 2014, 2014, 963792.
c) Du, D.-Y.; Yan, L.-K.; Su, Z.-M.; Li, S.-L.; Lan, Y.-Q.; Wang, E.-B.
17) Nieto-Oberhuber, C.; Munoz, M. P.; Lopez, S.; Jimenez-Nunez,
E.; Nevado, C.; Herrero-Gomez, E.; Raducan, M.; Echavarren, A. M.
Chem. - Eur. J. 2006, 12, 1677−1693.
18) Brooner, R. E. M.; Brown, T. J.; Widenhoefer, R. A. Chem. - Eur.
4
(
Coord. Chem. Rev. 2013, 257, 702−717. (d) Song, Y.-F.; Tsunashima,
R. Chem. Soc. Rev. 2012, 41, 7384−7402. (e) Miras, H. N.; Yan, J.;
Long, D.-L.; Cronin, L. Chem. Soc. Rev. 2012, 41, 7403−7430. (f) Oms,
O.; Dolbecq, A.; Mialane, P. Chem. Soc. Rev. 2012, 41, 7497−7536.
2) Wang, S.-S.; Yang, G.-Y. Chem. Rev. 2015, 115, 4893−4962.
3) Long, D.-L.; Tsunashima, R.; Cronin, L. Angew. Chem., Int. Ed.
010, 49, 1736−1736.
4) For a clever example, see: Luo, S.; Li, J.; Xu, H.; Zhang, L.;
Cheng, J.-P. Org. Lett. 2007, 9, 3675−3678.
5) (a) Dolbecq, A.; Dumas, E.; Mayer, C. R.; Mialane, P. Chem. Rev.
(
J. 2013, 19, 8276−8284.
(
1
19) (a) Mays, M. J.; Vergnano, P. A. J. Chem. Soc., Dalton Trans.
979, 1112−1115. (b) Parish, R. V.; Parry, O.; McAuliffe, C. A. J.
(
(
2
(
Chem. Soc., Dalton Trans. 1981, 2098−2104. (c) Colton, R.; Harrison,
K. L.; Mah, Y. A.; Traeger, J. C. Inorg. Chim. Acta 1995, 231, 65−71.
(
(
20) Evans, H. T.; Popev, M. T. Inorg. Chem. 1984, 23, 501−504.
21) Herrero-Gomez, E.; Nieto-Oberhuber, C.; Lopez, S.; Benet-
Buchholz, J.; Echavarren, A. M. Angew. Chem., Int. Ed. 2006, 45, 5455−
459.
22) Schmidbaur, H.; Schier, A. Chem. Soc. Rev. 2008, 37, 1931−
951.
23) Gaillard, S.; Slawin, A. M. Z.; Nolan, S. P. Chem. Commun. 2010,
6, 2742−2744.
24) Schneider, S. K.; Herrmann, W. A.; Herdtweck, E. Z. Anorg. Allg.
(
2
5
(
1
(
4
(
010, 110, 6009−6048. (b) Santoni, M.-P.; Hanan, G. S.; Hasenknopf,
B. Coord. Chem. Rev. 2014, 281, 64−85. (c) Matt, B.; Moussa, J.;
Chamoreau, L.-M.; Afonso, C.; Proust, A.; Amouri, H.; Izzet, G.
Organometallics 2012, 31, 35−38.
(
(
1
(
8
6) Gomez-Romero, P. Adv. Mater. 2001, 13, 163−174.
7) Song, Y.-F.; Long, D.-L.; Ritchie, C.; Cronin, L. Chem. Rec. 2011,
Chem. 2003, 629, 2363−2370.
(
(
1, 158−171.
8) (a) Ren, Y.; Wang, M.; Chen, X.; Yue, B.; He, H. Materials 2015,
, 1545−1567. (b) Mirzaei, M.; Eshtiagh-Hosseini, H.; Alipour, M.;
Frontera, A. Coord. Chem. Rev. 2014, 275, 1−18.
9) For selected examples of Cu-POM of type III, see: (a) Wang, C.;
25) Jia, M.; Bandini, M. ACS Catal. 2015, 5, 1638−1652.
26) (a) Zhang, L.; Wang, S. J. Am. Chem. Soc. 2006, 128, 1442−
1
443. (b) Shi, F. Q.; Li, X.; Xia, Y.; Zhang, L.; Yu, Z. X. J. Am. Chem.
Soc. 2007, 129, 15503−15512.
(
(
27) (a) Hashmi, A. S. K.; Weyrauch, J. P.; Frey, W.; Bats, J. W. Org.
Ren, Y.; Feng, S.; Kong, Z.; Hu, Y.; Yue, B.; Deng, M.; He, H. J. Coord.
Chem. 2014, 67, 506−521. (b) Ren, Y.; Du, C.; Feng, S.; Wang, C.;
Kong, Z.; Yue, B.; He, H. CrystEngComm 2011, 13, 7143−7148.
Lett. 2004, 6, 4391−4394. (b) Hashmi, A. S. K.; Jaimes, M. C. B.;
Schuster, A. M.; Rominger, F. J. Org. Chem. 2012, 77, 6394−6408.
(
28) (a) Maksimov, G. M.; Timofeeva, M. N.; Likholobov, V. A. Russ.
(
c) Yu, F.; Kong, X.-J.; Zheng, Y.-Y.; Ren, Y.-P.; Long, L.-S.; Huang,
Chem. Bull. 2001, 50, 1529−1532. (b) Izumi, Y.; Matsuo, K.; Urabe, K.
R.-B.; Zheng, L.-S. Dalton Trans. 2009, 9503−9509. For selected
examples of Ag-POM of type III, see: (d) Wang, D.-d.; Peng, J.; Pang,
H.-j.; Zhang, P.-p.; Wang, X.; Zhu, M.; Chen, Y.; Liu, M.-g.; Meng, C.-
l. Inorg. Chim. Acta 2011, 379, 90−94. (e) Yang, H.; Gao, S.; Lu, J.; Xu,
B.; Lin, J.; Cao, R. Inorg. Chem. 2010, 49, 736−744. (f) Chen, J.-X.;
Lan, T.-Y.; Huang, Y.-B.; Wei, C.-X.; Li, Z.-S.; Zhang, Z.-C. J. Solid
State Chem. 2006, 179, 1904−1910. For selected examples of Au-
POM of type III, see: (g) Yoshida, T.; Nagashima, E.; Arai, H.;
Matsunaga, S.; Nomiya, K. Z. Anorg. Allg. Chem. 2015, 641, 1688.
J. Mol. Catal. 1983, 18, 299−314.
(
29) Gorin, D. J.; Sherry, B. D.; Toste, F. D. Chem. Rev. 2008, 108,
3
(
351−3378.
30) (a) Kim, C.; Bae, H. J.; Lee, J. H.; Jeong, W.; Kim, H.; Sampath,
V.; Rhee, Y. H. J. Am. Chem. Soc. 2009, 131, 14660−14661. (b) Kim,
C.; Lim, W.; Rhee, Y. H. Bull. Korean Chem. Soc. 2010, 31, 1465−1466.
(
31) For acid-catalyzed Aza-Cope rearrangement, see: Castelhano, A.
L.; Krantz, A. J. Am. Chem. Soc. 1984, 106, 1877−1879.
32) Blanc, A.; Tenbrink, K.; Weibel, J.-M.; Pale, P. J. Org. Chem.
009, 74, 5342−5348.
33) Pennell, M. N.; Turner, P. G.; Sheppard, T. D. Chem. - Eur. J.
012, 18, 4748−4758.
34) Mo, J.; Eom, D.; Lee, E.; Lee, P. H. Org. Lett. 2012, 14, 3684−
687.
35) Belting, V.; Krause, N. Org. Lett. 2006, 8, 4489−4492.
36) For recyclable gold catalysts, see: (a) Gross, E.; Liu, J. H.-C.;
Toste, F. D.; Somorjai, G. A. Nat. Chem. 2012, 4, 947−952.
b) Raducan, M.; Rodrigez-Escrich, C.; Cambeiro, X. C.; Escudero-
Adan, E. C.; Pericas, M. A.; Echavarren, A. M. Chem. Commun. 2011,
(
2
(
2
(
3
(
(
(
h) Yoshida, T.; Yasuda, Y.; Nagashima, E.; Arai, H.; Matsunaga, S.;
Nomiya, K. Inorganics 2014, 2, 660−673. (i) Nomiya, K.; Yoshida, T.;
Sakai, Y.; Nanba, A.; Tsuruta, S. Inorg. Chem. 2010, 49, 8247−8254.
(
j) Gruber, F.; Jansen, M. Z. Anorg. Allg. Chem. 2010, 636, 2352−2356.
(
10) For selected examples, see: (a) Han, Q.; Sun, X.; Li, J.; Ma, P.;
e, S.; Louis, B.;
Niu, J. Inorg. Chem. 2014, 53, 6107−6112. (b) Borghes
̀
Blanc, A.; Pale, P. Catal. Sci. Technol. 2011, 1, 981−986. (c) Egi, M.;
Umemura, M.; Kawai, T.; Akai, S. Angew. Chem., Int. Ed. 2011, 50,
1
2197−12200. (d) Yadav, J. S.; Subba Reddy, B. V.; Purnima, K. V.;
Jhansi, S.; Nagaiah, K.; Lingaiah, N. Catal. Commun. 2008, 9, 2361−
364.
11) Hueber, D.; Hoffmann, M.; Louis, B.; Pale, P.; Blanc, A. Chem. -
Eur. J. 2014, 20, 3903−3907.
12) For example of gold type II hybrids, see: Dupre,
Fensterbank, L.; Malacria, M.; Thorimbert, S.; Hasenknopf, B.; Laco
E. Chem. - Eur. J. 2012, 18, 12962−12965.
13) Complex 1 has been previously synthesized and fully
(
2
(
4
2
1
7, 4893−4895. (c) Egi, M.; Azechi, K.; Akai, S. Adv. Synth. Catal.
011, 353, 287−290. (d) Cao, W.; Yu, B. Adv. Synth. Catal. 2011, 353,
̂
903−1907. (e) Neatu, F.; Li, Z.; Richards, R.; Toullec, P. Y.; Genet,
(
́
N.; Brazel, C.;
te,
̈
̂
J.-P.; Dumbuya, K.; Gottfried, J. M.; Steinruck, H.-P.; Parvulescu, V. I.;
̂
Michelet, V. Chem. - Eur. J. 2008, 14, 9412−9418. (f) Fierro-Gonzalez,
J.; Gates, B. C. Chem. Soc. Rev. 2008, 37, 2127−2134. (g) Liu, X.; Pan,
Z.; Shu, X.; Duan, X.; Liang, Y. Synlett 2006, 2006, 1962−1964.
(
11
characterized and served as POM-gold(I) reference to compare the
structural and catalytic properties.
(
37) (a) Mbomekalle, I.-M.; Lu, Y. W.; Keita, B.; Nadjo, L. Inorg.
Chem. Commun. 2004, 7, 86−90. (b) Bennardi, D. O.; Romanelli, G.
P.; Jios, J. L.; Autino, J. C.; Baronetti, G. T.; Thomas, H. J. Arkivoc
(
14) (a) Battisti, A.; Bellina, O.; Diversi, P.; Losi, S.; Marchetti, F.;
Zanello, P. Eur. J. Inorg. Chem. 2007, 6, 865−875. (b) Yu, Z.; Ma, B.;
Chen, M.; Wu, H.-H.; Liu, L.; Zhang, J. J. Am. Chem. Soc. 2014, 136,
2
008, 123−130.
6904−6907.
H
DOI: 10.1021/acs.organomet.5b00656
Organometallics XXXX, XXX, XXX−XXX