Synthesis of gem-diaurated species from alkynols

Article abstract of DOI:10.1002/chem.201204491

A number of enol ether-derived diaurated species were synthesized directly from different alkynols and cationic gold complexes in the presence of a non-nucleophilic base (proton sponge). The reaction can be easily applied for in situ generation of diaurated species from all common types of hydroalkoxylation substrates: 5-endo, 5-exo/6-endo, 6-exo/7-endo and intermolecular types. Six examples were also synthesized in individual state as stable hexafluoroantimonate salts. Whereas diaurated species are obtained reliably from all conventional mononuclear gold catalysts, application of binuclear ones often gave diaurated species with unusual properties. The preliminary results point to complexities of behavior of binuclear gold catalysts and would require more research in future for this subclass. The formation of diaurated species from various gold-oxo compounds (LAu)₂OH⁺, (LAu) ₃O⁺, and LAuOH (L=phosphine ligand) was also studied. Of these three types, only (LAu)₂OH⁺ is reactive, whereas (LAu)₃O⁺ and LAuOH are not reactive alone but require acidic promoters to enable the reaction. These differences in reactivity were explained by ability of these compounds to generate the necessary acetylene π-complex intermediate. Catch the gold fish: A number of enol ether derived diaurated species were synthesized directly from different alkynols and cationic gold complexes in the presence of a non-nucleophilic base (proton sponge; see scheme). The reaction can be easily applied for the in situ generation of diaurated species from all common types of hydroalkoxylation substrates: 5-endo, 5-exo/6-endo, 6-exo/7-endo and even intermolecular types. The reactivity of various gold oxo compounds (LAu)₂OH⁺, (LAu) ₃O⁺, and LAuOH (L=phosphine ligand) towards alkynols was also examined. Copyright

Full text of DOI:10.1002/chem.201204491

DOI: 10.1002/chem.201204491
Synthesis of gem-Diaurated Species from Alkynols
Alexander Zhdanko and Martin E. Maier*[a]
Abstract: A number of enol ether-de-
salts. Whereas diaurated species are
obtained reliably from all conventional
mononuclear gold catalysts, application
of binuclear ones often gave diaurated
species with unusual properties. The
preliminary results point to complexi-
ties of behavior of binuclear gold cata-
lysts and would require more research
in future for this subclass. The forma-
rived diaurated species were synthe-
sized directly from different alkynols
and cationic gold complexes in the
presence of a non-nucleophilic base
(proton sponge). The reaction can be
easily applied for in situ generation of
diaurated species from all common
types of hydroalkoxylation substrates:
5-endo, 5-exo/6-endo, 6-exo/7-endo and
intermolecular types. Six examples
were also synthesized in individual
state as stable hexafluoroantimonate
tion of diaurated species from various
gold-oxo compounds (LAu)2OH+,
(LAu)3O+, and LAuOH (L=phos-
phine ligand) was also studied. Of
these three types, only (LAu)2OH+ is
reactive, whereas (LAu)3O+ and
LAuOH are not reactive alone but re-
quire acidic promoters to enable the
reaction. These differences in reactivity
were explained by ability of these com-
pounds to generate the necessary ace-
tylene p-complex intermediate.
Keywords: alkynes · diaurated spe-
cies · enol ethers · gold · gold inter-
mediates
Introduction
Molecules containing an alkyne functionality allow
a
number of useful transformations.[1] A classical reaction is
the hydration of alkynes to ketones, which typically is per-
formed in presence of HgII salts and acid. Related to this is
the hydroalkoxylation of alkynes. In recent years, addition
reactions to alkynes have benefited a lot from gold catalysis.
For example, alkynes with internal hydroxyl functions can
be converted to cyclic enol ethers by using an AuI catalyst.
However, knowledge about the mechanism of gold-cata-
lyzed transformations involving alkynes remains much more
limited.[2] This also holds true for the hydroalkoxylation of
alkynes. The current understanding of this mechanism still
largely relies on computational methods. The addition is
thought to proceed through the following steps: 1) coordina-
tion of gold onto an alkyne to give a p-complex A,[3] 2) nu-
cleophilic attack of an alcohol to give a vinyl gold inter-
mediate B, and 3) subsequent protodeauration to give the
primary product C (Scheme 1).[4] The last event liberates the
gold catalyst, ready to start the next catalytic cycle.
Scheme 1. Key steps in the gold-catalyzed hydroalkoxylation and concur-
rent formation of gem-diaurated species D.
it was somehow overlooked by the gold catalysis community
until 2009 when Gagnꢀ and co-workers observed the gem-
diaurated species for the first time in a catalytic allene hy-
droarylation.[6] Later, diaurated species were detected in
other catalytic processes.[7] Since diaurated species do not
undergo protodeauration as easily as vinyl gold species, it
was soon realized that they should be inhibitory for cataly-
sis. Therefore, the knowledge about formation of these spe-
cies and their behavior within a catalytic cycle is of funda-
mental importance for practical applications of gold cataly-
sis, because avoiding this species should decrease the cata-
lyst loadings and benefit the catalytic cycle. The current
knowledge about this problem still primarily relies on model
However, it is known by now that vinyl gold complexes
exhibit strong aurophilic properties that may result in for-
mation of gem-diaurated species D. Although the phenom-
enon of gem-diauration has long been known (since 1974),[5]
[a] A. Zhdanko, Prof. Dr. M. E. Maier
Institut fꢁr Organische Chemie, Universitꢂt Tꢁbingen
Auf der Morgenstelle 18, 72076 Tꢁbingen (Germany)
Fax : (+49)7071-295137
+
[8]
systems, that is, the aryl diaurated species Ar
(AuL)2
.
In
contrast, direct experimental elucidation of the role of diau-
rated species in a catalytic process has only begun to
emerge, as exemplified by mechanistic investigations of
E-mail: martin.e.maier@uni-tuebingen.de
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201204491.
3932
ꢃ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 3932 – 3942

FULL PAPER
gold-catalyzed hydroalkoxylation of allenes recently publish-
ed by Wiedenhoefer, Gagnꢀ, and co-workers.[9] Nevertheless,
catalytic intermediates of the gold-catalyzed hydroalkoxyla-
tion of alkynes still remain to be experimentally explored.[10]
In connection with the gold-catalyzed spiroacetal forma-
tion of certain alkynediols, we discovered that enol ether-de-
rived diaurated species D are formed directly if an alkynol
is allowed to react with a gold catalyst in the presence of a
base. The intermediancy of this kind of species in a catalytic
hydroalkoxylation process was first suggested by Fꢁrstner
and co-workers in 2010.[11] In that report, a boron-to-gold
exchange reaction, rather than direct synthesis from an
alkyne, was used for the synthesis of diaurated species and
only one example of enol ether-derived diaurated species
was described, which still remains the only example up to
now.
Prompted by the fact that nothing else is known about
chemical properties of this kind of diaurated species, the
properties of the enol ether-derived vinyl gold species, the
ligand effects during their formation, or the exact pathways
of transformations of the diaurated species within the cata-
lytic cycle, we embarked on a project with the aim to under-
stand the role of diaurated species in more detail, at least
for catalytic hydroalkoxylation. Subsequent investigation on
the behavior of this species should provide important in-
sights on their role in AuI-catalyzed hydroalkoxylation reac-
tions. In the current paper we describe the formation of
enol ether-derived diaurated species from alkynols by using
ten gold complexes bearing different ligands (Figure 1).
Since only methodologies towards the synthesis of model
+
diaurated species Ar
A
our strategy can be considered as the first to target diaurat-
ed species that are real intermediates in gold catalysis.
Results and Discussion
Formation of diaurated species from mononuclear catalysts:
Thus, simple addition of a gold catalyst (substoichiometric
amount) to a solution of 3-heptyn-1-ol and proton sponge in
CDCl3 triggered an immediate reaction resulting in forma-
tion of diaurated species D as a single organogold product
(Scheme 2).[12] Under such conditions, the transformation
was accompanied by competitive enol ether formation.
Analogously, the diaurated species could be directly gener-
ated in [D8]THF by using MeLi or another soluble strong
base to stoichiometrically generate the alkoxide, which
would then react with two equivalents of gold in a strictly
stoichiometric manner. An important requirement for the
successful formation of diaurated species is that the base
should be well soluble in the reaction media, otherwise
simple gold-catalyzed hydroalkoxylation will predominantly
occur with (almost) complete regeneration of the initial cat-
alyst. After some experimentation, we concluded that the
use of proton sponge as a base in CDCl3 or CD2Cl2 as sol-
Figure 1. Gold catalysts used in this study.
Scheme 2. Synthesis of diaurated species from alkynols.
Chem. Eur. J. 2013, 19, 3932 – 3942
ꢃ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3933

M. E. Maier and A. Zhdanko
vents provided the most practically simple and reliable
route for generation of diaurated species not only for char-
acterization purposes, but also for the subsequent direct
study in solution. This is associated with the following bene-
fits: mixtures of proton sponge and its salt (bearing weakly
alysts 1–3, 5, and 7 the reaction is essentially clean, generat-
ing the diaurated species, enol ethers, PrSpH+, and free
MeCN as the only new components of the mixture (also
shown in Figure 2).[13]
In a completely similar manner as was performed with 3-
alkyne-1-ols, we conducted reactions with other alkynols
with the aim of collecting characterization data for the cor-
responding diaurated species D16–D23 (Scheme 2). Interest-
ingly, 4-alkynols reacted with catalysts 2 and 7 selectively
providing 5-exo diaurated species D18, D19, and D21 (how-
ever the corresponding enol ethers are still obtained as mix-
tures of 5-exo/6-endo isomers with low selectivity). This indi-
cates that even if the cyclization event is not selective, the
corresponding 5-exo/6-endo isomers of vinyl gold B bearing
the bulky ligands differ enough in reactivity so that the 5-
exo diaurated species is formed preferentially or even exclu-
sively (Scheme 3). In contrast, catalyst 1 normally provided
a mixture 5-exo/6-endo species D16 and D17 with low selec-
tivity (1:1.6 ratio). The following general reactivity series
was observed for substrates: 3-yne-1-ols>4-yne-1-ols@
5-yne-1-ols. The former two types reacted rather instantly
(ca. 5 min), whereas the latter required prolonged time and/
or heating.
ꢀ
and non-nucleophilic counterions such as OTfꢀ or SbF6 ),
which are present as byproducts in solution, form separate,
sharp, easily integrated signals that enable accurate quanti-
tative analysis of the NMR spectra (a representative NMR
spectrum of the reaction mixture is given on Figure 2). Next,
PrSpH+ can be easily neutralized by using a base that is not
soluble in the organic solvent. This provided, after filtration,
clear solutions of untouched diaurated species free of active
protons, which is necessary for the subsequent generation of
vinyl gold species and for study of the equilibrium between
vinyl gold and diaurated species. For this purpose, LiOH or
K2CO3 were successfully used. Further, in case of volatile
enol ethers, they can be completely evaporated from the re-
action mixture in vacuo to provide simplification of the mix-
ture upon subsequent dissolution. Since diaurated species
are formed in competitive conditions, an excess of alkynol
should be used, typically 2 equiv. It can be said that the use
of proton sponge simply ensures strong inhibition of the cat-
alytic reaction, which virtually stops as soon as all the gold
is bound in the form of diaurated species.
The described approach is not only applicable for the effi-
cient generation of diaurated species in situ, but also for
synthesis of these compounds in their individual state. In
this study, six examples of diaurated species (D1–3, D5,
D19, and D22) were isolated as stable hexafluoroantimonate
solid salts in 44–90% yields (Scheme 2). Even though com-
plete conversion of the starting gold complexes to diaurated
We applied these conditions to a series of 3-alkyne-1-ols
and a series of mononuclear gold catalysts 1–7 and found
that 5-endo diaurated species D1–D10 were successfully
formed as the only organogold products in all the cases
without exclusion in 99% in situ yield (Scheme 2). With cat-
Figure 2. Representative 1H NMR spectrum of the reaction mixture generated from pentynol (pen), gold catalyst 2, and the proton sponge (PrSp) in
CDCl3. Assignments of all peaks are given. The spectrum shows the presence of enol ether C5 and diaurated species D5 in a ratio of about 1.05:1. The
ꢀ
CH2 OR peaks appear as apparent triplets at d=4.30 and 4.21 ppm, respectively. The tert-butyl groups in each of the phosphine ligands are diastereotop-
ic and additionally are split up due to the coupling with the phosphorus atom. They integrate well with regard to the protons in D5. The 31P NMR (inset)
confirms the formation of D5 as a single organogold product.
3934
www.chemeurj.org
ꢃ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 3932 – 3942

Synthesis of gem-Diaurated Species
FULL PAPER
ꢀ
Several attempts to measure the 19F resonance of SbF6 ,
either in diaurated species or in starting catalysts, were all
unsuccessful as no signal appeared. 19F NMR data is applica-
ble only for L5 derivatives; giving signals of the CF3 group
around d=ꢀ63 ppm.
In four cases (D1–3 and D19) 13C NMR spectra were re-
corded. Here, the observation of the CAu2 ipso-carbon
signal is very intriguing since this signal is often not found,
especially in case of known aryl diaurated species Ar-
AHCTUNGTRENNUNG
(AuL)2+. In our case, a small triplet, almost equal to noise,
was observed around d=100 ppm in all the four carbon
spectra. In the HMBC spectra of D1 and D2 this signal
shows no cross-peaks, but in the other two cases clear cross-
peaks were observed, proving that this weak signal is not
2
noise, but the necessary carbon signal (Figure 4). The JCꢀP
Scheme 3. Kinetic resolution concept to explain predominant formation
of 5-exo diaurated species.
species was always reached in situ, the yield of the isolated
species was naturally dependent on the success in recrystalli-
zation, which was performed on a 20–100 mg reaction scale.
All new diaurated species were characterized by various
NMR spectroscopic methods and ESI mass spectra (see the
Experimental Section and the Supporting Information). In
the ESI mass spectra, the species show a strong, easily de-
tectable signal of the cation in the positive mode and the
corresponding anion signal in the negative mode. Their
common feature in 1H NMR spectra is always the simple
pattern of enol ether signals (e.g., always one triplet for
CH2O unit), whereas the pattern of ligand signals may be
different depending on molecular dynamics and symmetry.
For example, the diaurated species derived from L2 always
exhibit two doublets for tBu residues instead of only one in
the catalyst. This is a consequence of the symmetrical struc-
ture of diaurated species, which
Figure 4. Key HMBC correlations (arrows) and chemical shifts of a- and
ipso-carbon signals (dashed arrows) in diaurated species D1–3 and 19.
coupling constant was found to be 58–65 Hz, the same as
was earlier reported by the group of Fꢁrstner.[11] It can be
concluded, that in contrast to aryl-based diaurated species
Ar
generally exhibit this signal in NMR spectroscopic analyses.
(AuL)2+, enol ether-derived diaurated species should
ACHTUNGTRENNUNG
+
possesses only one symmetry
Likely in ArACTHGUNETRNNU(G AuL)2 this signal is also not absent, but diffi-
element: the symmetry plane
passing through the enol ether
unit, as exemplified in Figure 3.
This symmetry plain is now per-
cult to recognize because it might be hidden in the aromatic
region behind other stronger signals. Another important fea-
ture in the 13C NMR spectra is the appearance of the a-C
atom at relatively low fields at d=191–207 ppm. Interesting-
ly, in HMBC spectra this carbon gives strong cross-peaks to
all the hydrogen atoms located up to three bonds away
(Figure 4).
pendicular to the symmetry
Figure 3. Symmetry plain pass-
planes of the ligands, meaning
ing through the enol ether unit
that the elements of the ligands
accounts for the NMR spectro-
being previously in equivalent
environments are now no
longer equivalent and exhibit
different chemical shifts, that is,
In addition to the spectroscopic methods, in one case the
structure of diaurated species was fully confirmed by X-ray
analysis (D1, Figure 5). In the X-ray structure an apparent
scopical features observed.
ꢀ
aurophilic interaction (Au Au 2.95 ꢄ) is still present despite
they are diastereotopic. The enol core/ligand 1:2 ratio is
simply established in all cases by integration and is indica-
tive to the overall molecular composition of the species.
In the 31P NMR spectra, all diaurated species (derived
from mononuclear catalysts 1–5) exhibit a single resonance,
which is in accordance with the symmetry of the species.
the large size of the ligand. The C=C (136.3 pm) bond is
ꢀ
lengthened and the C O to the enol ether carbon
(134.0 pm) is shortened compared with normal enol ether
geometry, in complete accordance with previous observa-
tions.[11]
Chem. Eur. J. 2013, 19, 3932 – 3942
ꢃ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3935

M. E. Maier and A. Zhdanko
Scheme 4. Formation of the dppf-derived diaurated species D11.
tion could be obtained also from 31P NMR spectroscopy,
which showed a number of weak broad resonances. This
precluded us from establishing the exact composition of this
material. However, ESI-HRMS analysis of the reaction mix-
ture exhibited a signal at m/z 1183.27406, which is consistent
with the presence of diaurated species D12 with the molecu-
lar composition as shown in Scheme 5; their monomeric
monocationic composition was evident from the isotope pat-
tern. However, it cannot be excluded that this is a result of
the HRMS conditions and that in the normal reaction mix-
ture at room temperature, oligomeric diaurated species of ir-
regular or unknown structure might also be present. There-
fore, although a species with the exact structure as shown
for D12 might exist, we may not confirm if this is the only
possible component in the mixture. Anyway, whatever the
exact structure is, the conclusion that this substance has to
consist of diaurated units follows from the experimental ob-
servations performed on this material in situ (Scheme 5).
We found that it reacted with excess PPh3 to give a clear
mixture of PPh3-derived vinyl gold B2 and (S)-Tol-BINAP-
Figure 5. X-ray structure of the diaurated complex D1.
Formation of diaurated species from binuclear catalysts: Be-
sides the mononuclear catalysts, we also examined three bi-
nuclear ones (8–10) in reaction with 3-heptyn-1-ol under
standard conditions. Theoretically, from a binuclear gold cat-
alyst one may expect formation of a cyclic diaurated species.
This was indeed found to be true in the case of catalyst 8
giving cyclic diaurated species D11, whose composition was
established by NMR spectroscopy showing a single set of
signals of the enol ether core, a single phosphorus resonance
and a single molecular ion peak at 1059.11573 in the ESI-
HRMS spectrum (Scheme 4).
+
AuPPh3 as a byproduct.14 In another experiment, it was
treated with Et4N+Clꢀ. NMR spectra of the resulting mix-
ture clearly indicated formation of two kinds of vinyl gold
species (assigned to B3a and B3b) and the corresponding
In contrast to catalyst 8, binuclear catalysts 9 and 10 dis-
played some peculiarities that deserve special attention.
Thus, diaurated species obtained from 9 exhibited a very
complicated inconclusive 1H NMR spectrum. No informa-
(S)-Tol-BINAP
ACHTUNGTERNN(UNG AuCl)2 complex. This also proved that the
initial chaotic spectrum for the diaurated species with the
Scheme 5. Observation of diaurated species D12 with unknown structure and subsequent transformations to confirm the diaurated nature of the com-
pound. Key phosphorus resonances are shown (dashed arrows).
3936
www.chemeurj.org
ꢃ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 3932 – 3942

Synthesis of gem-Diaurated Species
FULL PAPER
simplest structure D12 is not the sign of decomposition or
some undesired process, but indicates the unusual complexi-
ty of the system, which cannot be conveniently analyzed by
room-temperature NMR spectroscopy.
terns, whereas in the 31P NMR spectrum, each isomer exhib-
its a single resonance due to the equivalence of both phos-
phorus atoms in each isomer. The remaining signal at d=
32.0 ppm was tentatively assigned to monomeric diaurated
species D15 temporarily arising in the reaction mixture.
These conclusions are supported by ESI-HRMS of the reac-
tion mixture, which exhibits the following ions: m/z
In case of catalyst 10, interesting chemistry was also ob-
served. The diaurated species generated from 3-heptyn-1-ol,
the proton sponge, and the catalyst in CD2Cl2 rather unex-
pectedly displayed two triplets for the CH2O units. In the
31P NMR spectrum, two clear resonances at d=32.4 and
30.3 ppm were observed. The H,H-COSY spectrum indicat-
ed that the CH2O triplets belong to two different enol ether
units. Even more surprisingly, monitoring the reaction mix-
ture revealed the evolution of this diaurated species; in the
31P NMR spectrum, the two initial resonances gradually dis-
appeared and four new signals arose at d=41.4, 35.9, 35.7,
and 32.0 ppm, respectively. With a longer reaction time, the
signal at d=32.0 ppm also disappeared so that at the end
1225.21283 (D14), 917.16416 (D15), and 609.153 for
2+
[dpppAu]2
. The isotope pattern of the ion at m/z
917.16416 corresponded exclusively to a single charged
cation. Unfortunately, the formation of dimeric dicationic
species D13 could not be confirmed by ESI-HRMS, but
their existence follows from the aforementioned features of
the NMR spectra.
Synthesis of diaurated species from intermolecular hydroal-
koxylation: The possibility to synthesize diaurated species
from simple internal alkynes and alcohols in an intermolecu-
lar manner was also briefly investigated. Thus, when 3-
hexyne was treated with complex 1 or 2, the proton sponge
or tBu2Py, and an excess of MeOH in CDCl3 no reaction
with the organic substrate was observed (aurated oxonium
ion was the only product). But when methanol was used as
a solvent, diaurated species D24 could be quickly generated,
as evidenced by the appearance of the corresponding signals
in the NMR spectra (Scheme 7). However, conversion of 2
1
only the three signals remained. At this point, the H NMR
spectrum showed the CH2O units as two multiplets, indica-
tive of an asymmetric environment. The resonance at d=
41.4 ppm was assigned to [dpppAu]22+, which exists as di-
meric dication. Accordingly, the two remaining signals were
therefore assigned to new gold complexes. These interesting
observations are explained as follows (Scheme 6). Initial ob-
servation of the two enol ether residues and the two phos-
phorus resonances indicates that the dimeric diaurated spe-
cies D13 is formed, which naturally exists as a mixture of
two geometric isomers, giving duplication of spectra. Due to
symmetry reasons, each isomer exhibits a single phosphorus
resonance and a single CH2O triplet (homotopic groups).
2+
With time, this compound expels a [dpppAu]2 unit and
gives rise to a triaurated species D14, which features a sand-
wich-type complexed gold and exists also as a pair of geo-
metric cis/trans isomers. Due to the lack of a symmetry
plane passing through each enol ether ring, the CH2O pro-
tons are now diastereotopic and give more complicated pat-
Scheme 7. Formation of diaurated species D24 from hexyne, MeOH, and
complex 2 in presence of tBu2Py.
to D24 reached only about a 60% maximum, with the re-
mainder of material being the unbound form of the catalyst.
Further monitoring of the reaction mixture after a pro-
longed time revealed a decrease in diaurated species as con-
sumption of 3-hexyne progressed. This implies that D24 is
not entirely stable under the reaction conditions, eventually
slowly turning back to the free catalyst when 3-hexyne is no
longer available.
The ability of the reaction to occur only in the presence
of a huge excess of an alcohol (taken as a solvent) under-
scores the high reversibility of the primary interaction be-
tween the alcohol and an alkyne p-complex, which is now
intermolecular and more strongly shifted to the left com-
pared with the intramolecular process in alkynols, slowing
down the overall process. The same phenomenon explains
the aforementioned reactivity series for alkynols
(Scheme 8).
Scheme 6. Evolution of (dppp)Au22+-derived diaurated species.
Chem. Eur. J. 2013, 19, 3932 – 3942
ꢃ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3937

M. E. Maier and A. Zhdanko
The different reactivity of these oxo-compounds is in line
with their different ability to undergo ligand exchange with
an alkyne to generate alkyne p-complex A. The formation
of this common active intermediate of gold-catalyzed hydro-
alkoxylation is the initial step of the whole process. Natural-
ly, the more stable the gold compound is towards the ligand
exchange, the less reactive it is. Thus, the (LAu)2OH+ I spe-
cies appears to be the most active because it is able to inter-
act already with weak nucleophiles to give reversibly
LAuNu+ and LAuOH, which was already demonstrated in
our previous work. According to our ligand strength
series,[15] the Keq for reaction of (LAu)2OH+ with an alkyne
is expected to be approximately 10ꢀ4 to 10ꢀ5. This value ap-
pears to be high enough to trigger fast formation of diaurat-
ed species ([Eq. (1)] in Scheme 10). In contrast to I,
Scheme 8. Reversibility of the initial interaction between an alcohol and
an alkyne p-complex.
Use of aurated oxonium salts and gold hydroxide as gold
sources: Besides the LAu+/PrSp combination, we also con-
sidered the possibility to use aurated oxonium salts as pre-
cursors for diaurated species. Recently diaurated oxonium
salts (LAu)2OH+ have been described.[15,16] These com-
pounds can be viewed as being constructed from two LAu+
electrophilic units and one basic OHꢀ unit, which represents
a gold/base 2:1 stoichiometry that is exactly required for the
synthesis of diaurated species. Accordingly, one might antici-
pate the direct synthesis of diaurated species without using
additional base, according to Equation (1) in Scheme 9. This
Scheme 9. Reactions of alkynols with oxonium salts (hypothetical bal-
anced equations, see main text for the experimental results).
possibility was verified and was found to be true. Indeed,
(L2Au)2OH+ was smoothly converted to diaurated species
D1 upon simple treatment with heptynol. However, the
process was again accompanied with competitive enol ether
formation so that the overall result is similar to the case of
the cat. 2/PrSp combination.
Correspondingly, triaurated oxonium salts would be ex-
pected to react with an alkynol to give a mixture of vinyl
gold and diaurated species according to Equation (2) in
Scheme 9. Surprisingly, (Ph3PAu)3O+ reacted only very slug-
gishly with heptynol to give the diaurated species and
Ph3PAuCl as the only gold products.[17] Notably, if
(Ph3PAu)3O+ was allowed to react with heptynol in the
presence of proton sponge salt PrSpH+, diaurated species
D4 was immediately formed according to general Equa-
tion (3) in Scheme 9.
Finally, gold hydroxides LAuOH, a new type of useful
gold compounds, can be envisaged as potential precursors
for direct vinyl gold synthesis, avoiding diaurated species
formation, according to Equation (4) in Scheme 9. However,
treatment of L2AuOH with heptynol in C6D6 demonstrated
that this is hardly applicable in practice; monitoring the re-
action mixture at room temperature revealed gold alkoxide
L2AuOR formation as the only process. At elevated tem-
peratures, some quantity of vinyl gold could be generated,
but the process was largely accompanied by decomposition
and therefore cannot be considered of practical value.
Scheme 10. Initial steps in reaction of aurated oxonium salts and LAuOH
with an alkyne. L=phosphine ligand.
(LAu)3O+ II appears to be a more stable complex with the
charge stabilized on all three gold units and being addition-
ally stabilized with more aurophilic interactions (three vs.
one in I). Not surprisingly, the extremely low reactivity of II
alone is explained by its virtual inability to interact with an
alkyne to give A. Indeed, the aurophilic interactions would
be broken and currently non-existing (LAu)2O III would be
formed in such a process ([Eq. (2)] in Scheme 10). This also
explains why (LAu)3O+ would be inefficient as a catalyst
for a catalytic process.[18] However, the reactivity of II gets
drastically increased in the presence of an acidic promoter.
This acceleration effect may be explained in two ways. First,
the acid would allow a fast and irreversible protonation of
III to give I, pushing the initially disfavored equilibrium to
the right ([Eq. (2)] in Scheme 10). Second, the acid would
3938
www.chemeurj.org
ꢃ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 3932 – 3942

Synthesis of gem-Diaurated Species
FULL PAPER
allow protonation of II itself to reversibly form a hypotheti-
cal dication IV as intermediate or transition state ([Eq. (3)]
in Scheme 10). This species should be highly reactive to-
wards even weak nucleophiles such as alkyne, making the
formation of A easy. Of these two hypotheses, the former is
less feasible because if the formation of III would be possi-
ble at all, it would definitely not hesitate to deprotonate any
other molecules in solution (H2O, CDCl3, alkynol) that are
already acidic enough.[19] So there would be no need to use
any additional stronger proton source to enhance the reac-
tivity of gold. Rather, the latter hypothesis gains strong sup-
port from the fact, that (LAu)3O+ can be aurated to form
the isolable dicationic (LAu)4O2+, as was established by
Schmidbaur and co-workers almost 20 years ago.[20] It can be
concluded, that despite being cationic, cation II still has the
ability to add electrophiles and undergo electrophilic substi-
tution at oxygen through these reactive dicationic adducts.[21]
Therefore, we conclude that protonation of (Ph3PAu)3O+ to
form IV is indeed necessary prior to formation of the active
species A.[22,23] This was previously not recognized. Being
formed, dication IV would either undergo ligand exchange
to give A or would dissociate spontaneously to active gold
electrophile LAu+ eventually also leading to A.
Similarly as was described for II, the extremely low reac-
tivity of gold hydroxide LAuOH in forming vinyl gold spe-
cies is also explained by its inability to form A in the corre-
sponding ligand-exchange reaction ([Eq. (4)] in Scheme 8).
As can be understood from our ligand strength series,[15] the
equilibrium between LAuOH and an alkyne can be estimat-
ed to be far beyond 10ꢀ10, virtually making alkyne activation
impossible. Naturally, an alkyne is unable to displace OHꢀ
from gold! Likewise it was described for II, LAuOH also re-
quires an acidic promoter to initiate the reaction (being a
strong base it will react with H+ to release the active LAu+
catalyst and H2O).
indeed proven by direct observation of the species in situ
during catalytic runs.[24] Extensive studies on the role of
these species as catalytic intermediates in gold catalysis are
underway in our laboratory.
Experimental Section
General: Since the exact outcome of the competition between auration/
protodeauration was a priori not known, the following syntheses were
performed under NMR control. All manipulations were performed with-
out the use of an inert atmosphere.
Synthesis of D1: In one shot, complex
2 (78.9 mg, 0.102 mmol,
2.00 equiv) was added to solution of 3-heptynol-1 (17.07 mg,
a
0.152 mmol, 2.98 equiv) and the proton sponge (15.1 mg, 0.0706 mmol,
1.38 equiv) in CDCl3 (1.4 mL), with the reaction flask being shaken by
hand. A white crystalline precipitate appeared soon (PrSpH+·SbF6ꢀ).
After being shaken for 1 min, a part of the solution was transferred into
an NMR tube. 1H and 31P NMR analysis revealed complete consumption
of the starting complex and formation of diaurated species as a sole gold-
containing product. The molar ratio of the components in the reaction
mixture at this moment was enol ether/D1/heptynol/PrSp=
1:1.08:1.22:0.44. TLC analysis indicated the presence of unreacted hepty-
nol and the remainder of the material simply stayed on the starting line
(petroleum ether/EtOAC, 2:1). Since the starting heptynol and proton
sponge were still available in the mixture, the content of the NMR tube
was combined with the rest of the material and an additional amount of
2 (24.9 mg 0.032 mmol, 0.63 equiv) was added and the mixture was ana-
lyzed again after 5 min. 1H and 31P NMR analysis now indicated a ratio
enol ether/D1/heptynol/PrSp=1:1.17:0.49:0.10. At this moment, the reac-
ꢀ
tion mixture was filtered to separate most of the PrSpH+·SbF6 from the
system (the filter cake was washed with CDCl3). Since the final product
will be obtained by crystallization from benzene, the rest of proton
sponge salt has to be neutralized because it is also insoluble in benzene.
For this purpose, freshly powdered LiOH·H2O (43 mg) was added to the
clear filtrate and the suspension was ultrasonicated for 3 min (Note:
NMR analysis indicated that still some quantity of PrSpH+ was present
in solution and more LiOH·H2O powder (56 mg) was added). At this
moment, NMR indicated complete disappearance of PrSpH+. Then, the
suspension was filtered through celite and the filter cake washed with
CH2Cl2. The clear filtrate was transferred to a weighted receiver flask
and evaporated in vacuo until dry (the volatile enol ether also disap-
peared at this stage). The residue was redissolved in a minimum volume
of CH2Cl2 and layered with benzene; tiny crystal needles soon appeared.
As most of the material had crystallized, the mixture was concentrated to
further remove CH2Cl2 and a fresh portion of benzene (about 1 mL) was
added. The final product is insoluble in benzene and the supernatant sol-
ution was easily removed by filtration performed by using a Pasteur pip-
ette with a piece of cotton at the end. The crystalline material in the
flask was washed twice with benzene and dried in vacuo to yield the dia-
urated species (87.2 mg, 90%) as very small white needles. NMR analysis
confirmed that the compound crystallized as benzene solvate (1 molecule
of benzene per 1 molecule of the cation). These crystals were unsuitable
for X-ray analysis; the benzene-free material obtained by crystallization
from CH2Cl2/pentane was also unsuitable. Finally, we found that the com-
pound was sparingly soluble in methanol and simple crystallization from
this solvent provided very good crystals for X-ray analysis, containing no
incapsulated solvent molecules. 1H NMR (400 MHz, CDCl3): d=7.87–
7.93 (m, 2H), 7.48–7.52 (m, 4H), 7.32–7.43 (m, 6H), 7.26 (d, 2H), 7.08–
7.16 (m, 4H), 4.20 (t, J=9.3 Hz, 2H), 2.28 (m, 2H), 1.67 (t, J=9.3 Hz,
2H), 1.74 (overlapped sextet, J=7.4 Hz, 2H), 1.51 (d, J=15.3 Hz, 18H),
1.32 ppm (d, J=15.2 Hz, 18H), 0.88 (t, J=7.4 Hz, 3H); 31P NMR
(162 MHz, CDCl3): d=63.94 ppm; 13C NMR (100 MHz, CDCl3): d=
196.4 (t, J=2.9 Hz), 149.0 (d, J=14.6 Hz), 142.7 (d, J=5.9 Hz), 134.47,
133.9 (d, J=7.3 Hz), 130.9, 130.1, 129.6, 128.4, 127.9, 127.1 (d, J=5.9 Hz),
125.2 (d, J=39.5 Hz), 103.6 (t, J=60 Hz), 74.4, 38.01 (d, J=22.7 Hz),
37.97 (d, J=21.2 Hz), 36.3, 35.9, 31.5 (d, J=7.3 Hz), 30.7 (d, J=6.6 Hz),
Conclusion
We have demonstrated that enol ether-derived diaurated
species can be directly obtained from cationic gold(I) cata-
lysts and alkynols by using a proton sponge as a base or by
using diaurated oxonium salt (LAu)2OH+ as the sole reac-
tant. Other oxo compounds of gold (LAu)3O+ and LAuOH
were found to be almost inactive towards alkynols, which
ꢁ
was explained by their inability to interact with the C C
bond due to increased stability of these complexes towards
nucleophilic attack. This study demonstrates that the forma-
tion of diaurated species is generally not precluded even by
bulky ligands at gold. This has led to the development of a
practical general route for the synthesis of diaurated species
in situ or in an individual state. The described synthesis
occurs in catalytically relevant manner, that is, it must
follow the same mechanism as shown in Scheme 1, but the
catalytic turnover is blocked by trapping active protons. This
suggests the involvement of diaurated species also in a
standard catalytic hydroalkoxylation process, which was
Chem. Eur. J. 2013, 19, 3932 – 3942
ꢃ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3939

M. E. Maier and A. Zhdanko
22.6, 14.2 ppm; HRMS (ESI) m/z calcd for C47H65Au2OP2+: 1101.3836
[M+]; found 1101.3832.
traces of (Ph3PAu)3O+ formed, so even this short ultrasonication was not
necessary. Then the suspension was filtered through celite and the filter
cake washed with CH2Cl2 directly to a weighted receiver flask and evapo-
rated in vacuo until dry (volatile enol ether also disappeared at this
stage). As judged by NMR spectroscopy, the oily residue was already
quite pure diaurated species, the mole ratio of the components in the
mixture was diaurated/pentynol/PrSp=1:0.09:0.20 (which could be
ꢂ94% mass of pure material). The subsequent crystallization of the ma-
terial proved more difficult than in other cases. Attempted crystallization
from CH2Cl2/pentane was unsuccessful (the concentrated oily residue re-
mained uncrystallized even if cooled to ꢀ308C). The use of CH2Cl2/
MeOH proved more successful: the product crystallized soon as a solid
powder. Obviously, crystal growth rate is rather slow. Nevertheless, the
pure product was obtained as creamy, very small crystals after several at-
tempts of crystallizations from CH2Cl2/MeOH. Cold MeOH should be
always used because of the moderate solubility of the product. Bigger
crystals arose from slow evaporation of the mother liquor. The product is
well soluble in CH2Cl2, chloroform, sparingly soluble in MeOH. With
benzene it forms an insoluble oil (so crystallization from benzene is prac-
tically impossible). 1H NMR (400 MHz, CDCl3): d=7.46–7.55 (m, 6H),
7.36–7.42 (m, 24H), 4.69 (t, J=9.1 Hz, 2H), 2.87 (t, J=9.1 Hz, 2H),
2.52 ppm (s, 3H); 31P NMR (162 MHz, CDCl3): d=37.07 ppm; 13C NMR
(100 MHz, CDCl3): d=197.7 (t, J=4 Hz), 133.9 (d, J=13.9 Hz), 132.2,
129.6 (d, J=11.7 Hz), 128.7 (d, J=57.1 Hz), 109.0 (t, J=65 Hz), 76.0,
37.5, 20.2 ppm.
ꢀ
Note: Later we found that PrSpH+·SbF6 in chloroform solutions is
much more readily neutralized by freshly powdered K2CO3 (by shaking
the reaction mixture for 10 seconds, without ultrasound). This does not
ꢀ
apply however for CH2Cl2, in which PrSpH+·SbF6 is much more soluble.
Also, the rate of neutralization is dependent on the amount or residual
water in solution; it occurs more sluggishly in a highly dry (freshly
opened) CDCl3.
Synthesis of D2: Since the corresponding fast-forming oxonium species
(L5Au)3O+ is not soluble in CDCl3, thus preventing the formation of dia-
urated species, this synthesis should be performed in CH2Cl2. In one shot,
complex 5 (81.8 mg, 0.0715 mmol, 2.00 equiv) was added to a solution of
3-heptynol-1 (8.03 mg, 0.0706 mmol, 1.97 equiv) and the proton sponge
(12 mg, 0.0561 mmol, 1.57 equiv) in CD2Cl2 (0.7 mL), with the reaction
flask being shaken by hand. The resulting solution was filtered by using
cotton and a Pasteur pipette (to remove traces of Ag2O, coming from
Ag+ impurity in the catalyst) and transferred into an NMR tube. 1H and
31P NMR analysis revealed complete consumption of the starting complex
and formation of diaurated species as a sole gold-containing product. The
molar ratio of the components in the reaction mixture at this moment
was enol ether/D2/heptynol=0.59:1:0.60. Then, freshly powdered K2CO3
was added and the reaction mixture was shaken and analyzed by NMR
from time to time to observe sluggish neutralization of PrSpH+. This way
approximately 80% of the proton sponge was neutralized. Since we plan-
ned to obtain the final product by crystallization from MeOH, we hy-
pothesized that minor quantities of PrSpH+ would not cause problems.
Also we knew that this diaurated species will slowly decompose if excess
of base is present in solution. Therefore, without waiting for complete
neutralization the suspension was filtered through celite (the filter cake
was washed with CH2Cl2) and the clear filtrate was transferred to a
weighted receiver flask and evaporated in vacuo until almost dry (the
diaurated species had crystallized already during the time of evapora-
tion). The wet residue (containing tiny crystal needles) was mixed with
methanol and allowed to stay in the cold for about 30 min. The superna-
tant solution was easily removed by filtration performed by using a Pas-
teur pipette with a piece of cotton at the end, the crystals washed with
cold MeOH and dried in vacuo to yield diaurated species (34.9 mg, 47%)
as white needles. The material additionally obtained from the mother
liquor was of bad quality and therefore was disregarded. Surprisingly, the
compound was found to be insoluble in CDCl3. 1H NMR (400 MHz,
CD2Cl2): d=8.20 (s, 6H), 7.98 (d, J=12.4 Hz, 12H), 4.77 (t, J=9.1 Hz,
2H), 2.97 (t, J=9.1 Hz, 2H), 2.82 (t, J=7.4 Hz, 2H), 1.74 (sextet, J=
7.4 Hz, 2H), 0.91 ppm (t, J=7.4 Hz, 3H); 31P NMR (162 MHz, CD2Cl2):
d=40.77 ppm; 19F NMR (376 MHz, CD2Cl2): d=ꢀ63.69 ppm; 13C NMR
(100 MHz, CD2Cl2): d=206.5 (t, J=4 Hz), 134.3 (d, J=16 Hz), 134.0 (dq,
JP=12 Hz, JF =35 Hz), 130.2 (d, J=55.6 Hz), 128.0, 122.7 (d, JF =
273 Hz), 106.8 (t, J=66 Hz), 77.7, 36.8, 24.0, 13.9.
Synthesis of D5: Complex 2 (70.0 mg, 0.0907 mmol, 2.00 equiv) was
added in one shot to a solution of 3-pentynol-1 (7.79 mg, 0.0926 mmol,
2.04 equiv) and a proton sponge (10.2 mg, 0.0477 mmol, 1.05 equiv) in
CDCl3 (0.7 mL), with shaking in hand. A white crystalline precipitate
soon appeared (PrSpH+ SbF6ꢀ). At this moment, the reaction mixture
ꢀ
was filtered to separate the most of PrSpH+ SbF6 from the system (the
1
filter cake was washed with CDCl3). H and 31P NMR spectroscopic anal-
yses revealed complete consumption of the starting complex and forma-
tion of diaurated species as a sole gold-containing product. The molar
ratio of the components in the reaction mixture at this moment was enol
ether/D5/pentynol/PrSp=0.98:1:0.02:0.06, indicating that the initial
amounts of the reagents were chosen well and no corrections were re-
quired. Then freshly powdered K2CO3 (20 mg) was added for complete
neutralization of the remaining PrSpH+. The suspension was filtered
through celite (the filter cake was washed with CH2Cl2) and the clear fil-
trate was transferred to a weighted receiver flask and evaporated in
vacuo until almost dry. The residue was layered with methanol and al-
lowed to remain at room temperature for 30 min and then cooled for
about 30 min. Crystallization occurred smoothly and the cold supernatant
solution was easily removed by using filtration performed with a Pasteur
pipette with a piece of cotton at the end. The crystals were washed with
cold MeOH and dried in vacuo to yield diaurated species (52 mg, 88%)
as white needles. The compound was characterized by 1H, and 31P NMR
spectra. 1H NMR (400 MHz, CDCl3): d=7.88–7.92 (m, 2H), 7.48–7.52
(m, 4H), 7.31–7.39 (m, 6H), 7.21 (d, 2H), 7.10–7.13 (m, 4H), 4.21 (t, J=
9.1 Hz, 2H), 1.98 (s, 3H), 1.74 (t, J=9.1 Hz, 2H), 1.52 (d, J=15.3 Hz,
18H), 1.32 (d, J=15.2 Hz, 18H); 31P NMR (162 MHz, CDCl3): d=
63.91 ppm; HRMS (ESI): m/z calcd for C45H61Au2OP2+: 1073.3523 [M+];
found 1073.3531.
Synthesis of D3: Complex 1 (58.5 mg, 0.0795 mmol, 2.00 equiv) was
added in one shot to a solution of 3-pentynol-1 (11.5 mg, 0.137 mmol,
3.4 equiv) and the proton sponge (16.5 mg, 0.0771 mmol, 1.9 equiv) in
CDCl3 (1.4 mL), with the reaction flask being shaken by hand. A white
crystalline precipitate soon appeared (PrSpH+·SbF6ꢀ). After being
shaken for 1 min, part of the solution was transferred into an NMR tube.
1H and 31P NMR analysis revealed complete consumption of the starting
complex and formation of diaurated species as a sole gold-containing
product. The mole ratio of the components in the reaction mixture at this
moment was enol ether/D3/pentynol/PrSp=0.67:1:1.03:0.43. Since the
starting pentynol and proton sponge were still available in the mixture,
the content of the NMR tube was combined with the rest of the material
and additional 1 (28.7 mg) was added and the mixture analyzed again
after 5 min; however, that was still not enough and more 1 (5.07 mg) was
added. At this moment the reaction mixture was filtered to separate
Synthesis of D19: Complex 2 (39.0 mg, 50.5 mmol) was added in one shot
to a solution of 4-hexynol-1 (3.2 mg, 32.6 mmol), the proton sponge
(5.93 mg, 27.7 mmol) in CDCl3 (0.6 mL), with the reaction flask being
shaken by hand. A white crystalline precipitate soon appeared (PrSpH+
SbF6ꢀ). After 2 min, the supernatant solution was filtered into an NMR
tube by using a piece of cotton and a Pasteur pipette. NMR examination
revealed complete formation of diaurated species. The molar ratio of the
components in the reaction mixture at this moment was 5-exo diaurated/
6-endo diaurated/5-exo enol ether/6-endo enol ether/hexynol/PrSp=
1:0.08:0.06:0.07:0.07:0.05. From this spectrum it became clear that the ini-
tial amounts of the reagents were chosen well and no corrections were
required. Next, the reaction mixture was treated with freshly powdered
K2CO3 (ca. 15 mg) to achieve complete neutralization of PrSpH+. Then
the suspension was filtered through celite directly into a weighted receiv-
er flask (the filter cake washed with CH2Cl2). The solution was evaporat-
ꢀ
most of PrSpH+·SbF6 from the system (the filter cake was washed with
CDCl3). Freshly powdered K2CO3 (19 mg) was added to the clear filtrate
and the suspension was briefly shaken (the crystals changed their appear-
ance almost instantly) and then ultrasonicated for 5 more seconds. NMR
analysis indicated complete disappearance of PrSpH+ and even that
3940
www.chemeurj.org
ꢃ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 3932 – 3942

Synthesis of gem-Diaurated Species
FULL PAPER
ed in vacuo until dry (the volatile enol ethers also disappeared at this
stage). The residue was diluted with few drops of CH2Cl2 and layered
with benzene. Crystallization did occur, but the precipitate was too diffi-
cult to separate from the supernatant solution, therefore crystallization
was performed with methanol (about 1 mL) at ꢀ308C. The product was
washed with cold methanol (ꢀ30 to ꢀ608C). The solid residue was dried
in vacuo to yield diaurated species (23 mg, 69%) as white needles. From
the mother liquor, a mixture of 5-exo diaurated/6-endo diaurated species
(1:0.13) could be obtained (6.5 mg). 1H NMR (400 MHz, CDCl3): d=
7.87–7.91 (m, 2H), 7.45–7.50 (m, 4H), 7.25–7.38 (m, 6H), 7.08–7.17 (m,
6H), 4.46 (t, J=6.7 Hz, 2H), 2.71 (t, J=7.3 Hz, 2H), 2.04 (app pent,
2H), 1.49 (d, J=15.2 Hz, 18H), 1.32 (d, J=15.2 Hz, 18H), 1.05 ppm (s,
3H); 31P NMR (162 MHz, CDCl3): d=63.66; 13C NMR (100 MHz,
CDCl3): d=190.8 (t, J=3.3 Hz), 149.3 (d, J=14.6 Hz), 142.6 (d, J=
5.9 Hz), 134.4, 133.8 (d, J=8.0 Hz), 130.7, 129.5, 129.2, 128.4, 128.3,
128.1, 126.9 (d, J=6.6 Hz), 125.4 (d, J=38.8 Hz), 97.9 (t, J=58 Hz), 73.8,
Gorin, B. D. Sherry, F. D. Toste, Chem. Rev. 2008, 108, 3351–3378;
i) E. Jimꢀnez-NfflÇez, A. M. Echavarren, Chem. Rev. 2008, 108,
3326–3350; j) Z. Li, C. Brouwer, C. He, Chem. Rev. 2008, 108,
3239–3265; k) A. Arcadi, Chem. Rev. 2008, 108, 3266–3325; l) J.
Muzart, Tetrahedron 2008, 64, 5815–5849; m) M. Rudolph, A. S. K.
Hashmi, Chem. Soc. Rev. 2012, 41, 2448–2462; n) H. C. Shen, Tetra-
hedron 2008, 64, 7847–7870; o) R. A. Widenhoefer, Chem. Eur. J.
2008, 14, 5382–5391; p) D. J. Gorin, F. D. Toste, Nature 2007, 446,
395–403; q) A. Fꢁrstner, P. W. Davies, Angew. Chem. 2007, 119,
3478–3519; Angew. Chem. Int. Ed. 2007, 46, 3410–3449; r) E. Jimꢀ-
nez-NfflÇez, A. M. Echavarren, Chem. Commun. 2007, 333–346;
s) A. S. K. Hashmi, Chem. Rev. 2007, 107, 3180–3211; t) A. S. K.
Hashmi, G. J. Hutchings, Angew. Chem. 2006, 118, 8064–8105;
Angew. Chem. Int. Ed. 2006, 45, 7896–7936.
[2] a) A. S. K. Hashmi, Angew. Chem. 2010, 122, 5360–5369; Angew.
Chem. Int. Ed. 2010, 49, 5232–5241; b) L.-P. Liu, G. B. Hammond,
Chem. Soc. Rev. 2012, 41, 3129–3139.
[3] a) M. Pernpointner, A. S. K. Hashmi, J. Chem. Theory Comput.
2009, 5, 2717–2725; b) M. Lein, M. Rudolph, S. K. Hashmi, P.
Schwerdtfeger, Organometallics 2010, 29, 2206–2210.
38.4 (d, J=22.0 Hz), 38.3 (d, J=20.5 Hz), 36.7, 31.5 (d, J=7.3 Hz), 30.8
+
(d, J=6.6 Hz), 24.5, 22.5; HRMS (ESI): m/z calcd for C46H63Au2OP2
:
1087.3680 [M+]; found 1087.3682.
Synthesis of D22: Complex 2 (24.0 mg, 31.1 mmol, 2.00 equiv) was added
in one shot to a solution of 5-octynol-1 (2.22 mg, 17.6 mmol), the proton
sponge (3.55 mg, 16.7 mmol, 1.07 equiv) in CDCl3 (0.6 mL), with the reac-
tion flask being shaken by hand. A white crystalline precipitate appeared
soon (PrSpH+ SbF6ꢀ). Because of insufficient rate of the 6-exo cycliza-
tion, the resulting rate of diaurated species formation is slow, therefore at
this moment only oxonium salt (L2Au)2OH+ was generated. After shak-
ing for 1 min, the solution was filtered into an NMR tube using a piece
of cotton and a Pasteur pipette. The reaction mixture was heated at 408C
for 30 min. After this time, NMR examination revealed complete con-
sumption of octynol and formation of diaurated species, but the oxonium
salt was still present. Therefore, more octynol (about 1 mg) was added
and the reaction mixture was brought to a condition when only small
traces of oxonium salt were present. Then, the reaction mixture was fil-
tered once again and treated with freshly powdered K2CO3 (about
20 mg). NMR examination showed complete neutralization of PrSpH+.
At this moment the reaction mixture was filtered through celite into a
weighted receiver flask and evaporated in vacuo until dry (volatile enol
ethers also disappeared at this stage). Then MeOH (about 0.8 mL) was
added to the residue and the solution was kept at ꢀ308C for 30 min. The
product precipitated as a white solid. The supernatant solution was re-
moved by using a Pasteur pipette with cotton. The solid was washed with
small portions of cold (ꢀ30 to ꢀ608C) MeOH and dried in vacuo to
afford the first crop of the title compound as a white solid, 9.3 mg
[4] a) C. M. Krauter, A. S. K. Hashmi, M. Pernpointner, ChemCatChem
2010, 2, 1226–1230; b) G. Mazzone, N. Russo, E. Sicilia, Organome-
tallics 2012, 31, 3074–3080.
[5] A. N. Nesmeyanov, E. G. Perevalova, K. I. Grandberg, D. A. Leme-
novskii, T. V. Baukova, O. B. Afanassova, J. Organomet. Chem.
1974, 65, 131–144.
[6] a) D. Weber, M. A. Tarselli, M. R. Gagnꢀ, Angew. Chem. 2009, 121,
5843–5846; Angew. Chem. Int. Ed. 2009, 48, 5733–5736; b) Ag+ en-
gages in a similar dimetalation reaction to afford the corresponding
mixed-metal species with an Au···Ag interaction, see: D. Weber,
M. R. Gagnꢀ, Org. Lett. 2009, 11, 4962–4965.
[7] a) A. S. K. Hashmi, I. Braun, P. Nçsel, J. Schꢂdlich, M. Wieteck, M.
Rudolph, F. Rominger, Angew. Chem. 2012, 124, 4532–4536;
Angew. Chem. Int. Ed. 2012, 51, 4456–4460; b) A. S. K. Hashmi, I.
Braun, M. Rudolph, F. Rominger, Organometallics 2012, 31, 644–
661; c) Y. Chen, M. Chen, Y. Liu, Angew. Chem. 2012, 124, 6285–
6290; Angew. Chem. Int. Ed. 2012, 51, 6181–6186; d) A. S. K.
Hashmi, M. Wieteck, I. Braun, P. Nçsel, L. Jongbloed, M. Rudolph,
F. Rominger, Adv. Synth. Catal. 2012, 354, 555–562.
[8] a) D. Weber, T. D. Jones, L. L. Adduci, M. R. Gagnꢀ, Angew. Chem.
2012, 124, 2502–2506; Angew. Chem. Int. Ed. 2012, 51, 2452–2456;
b) J. E. Heckler, M. Zeller, A. D. Hunter, T. G. Gray, Angew. Chem.
2012, 124, 6026–6030; Angew. Chem. Int. Ed. 2012, 51, 5924–5928.
[9] T. J. Brown, D. Weber, M. R. Gagnꢀ, R. A. Widenhoefer, J. Am.
Chem. Soc. 2012, 134, 9134–9137.
1
(44%). H NMR (400 MHz, CDCl3): d=7.88–7.92 (m, 2H), 7.44–7.48 (m,
4H), 7.29–7.39 (m, 6H), 7.03–7.17 (m, 6H), 4.32 (t, J=6.0 Hz, 2H), 2.49
(t, J=6.8 Hz, 2H), 1.95 (app. pent, 2H), 1.71 (app pent, 2H), 1.51–1.58
(m, 2H), 1.469 (d, J=15.2 Hz, 18H), 1.35 (d, J=15.3 Hz, 18H), 0.60 (t,
J=7.4 Hz, 3H) ppm; 31P NMR (162 MHz, CDCl3): d=64.51 ppm;
HRMS (ESI): m/z calcd for C48H67Au2OP2+: 1115.3993 [M+]; found
1115.3995.
[10] Quite recently, observation of diaurated species in the gold-cata-
lyzed hydroalkoxylation of 1-phenylpropyne by ESI-MS was report-
ˇ
ˇ
ˇ
ed: J. Roithovµ, S. Jankovµ, L. JaSíkovµ, J. Vµna, S. Hybelbauerovµ,
Angew. Chem. 2012, 124, 8503–8507; Angew. Chem. Int. Ed. 2012,
51, 8378–8382.
[11] G. Seidel, C. W. Lehmann, A. Fꢁrstner, Angew. Chem. 2010, 122,
8644–8648; Angew. Chem. Int. Ed. 2010, 49, 8466–8470.
[12] For the use of a base to systematically obtain vinyl gold species
from catalysis reactions, see: a) A. S. K. Hashmi, A. M. Schuster, F.
Rominger, Angew. Chem. 2009, 121, 8396–8398; Angew. Chem. Int.
Ed. 2009, 48, 8247–8249; b) A. S. K. Hashmi, T. D. Ramamurthi, F.
Rominger, Adv. Synth. Catal. 2010, 352, 971–975; c) A. S. K.
Hashmi, A. M. Schuster, S. Gaillard, L. Cavallo, A. Poater, S. P.
Nolan, Organometallics 2011, 30, 6328–6337; d) A. S. K. Hashmi,
Gold Bull. 2009, 42, 275–279.
Acknowledgements
We thank Dr. M. Strçbele for the X-ray analyses, Dr. D. Wistuba for
HRMS analyses, Dr. K. Eichele and the Institut fꢁr Anorganische
Chemie for allowing us to use their NMR spectrometer.
[13] If the reactants are taken in such a ratio that the substrate (alkynol)
finishes sooner than all the gold is converted to diaurated species,
and excess of PrSp is still available in the mixture, then the remain-
ing gold will be present as aurated oxonium ion (LAu)2OH+ or
(LAu)3O+ depending on the catalyst initially used. This oxonium
ion is further converted into diaurated species if more substrate is
added. If it is present as impurity in a starting gold catalyst, Ag+
will form a black precipitate.
[1] For reviews on gold catalysis, see: a) A. Corma, A. Leyva-Perꢀz,
M. J. Sabater, Chem. Rev. 2011, 111, 1657–1712; b) M. Bandini,
Chem. Soc. Rev. 2011, 40, 1358–1367; c) T. C. Boorman, I. Larrosa,
Chem. Soc. Rev. 2011, 40, 1910–1925; d) A. S. K. Hashmi, M.
Bꢁhrle, Aldrichimica Acta 2010, 43, 27–33; e) N. D. Shapiro, F. D.
Toste, Synlett 2010, 675–691; f) S. Sengupta, X. Shi, ChemCatChem
2010, 2, 609–619; g) N. Bongers, N. Krause, Angew. Chem. 2008,
120, 2208–2211; Angew. Chem. Int. Ed. 2008, 47, 2178–2181; h) D. J.
Chem. Eur. J. 2013, 19, 3932 – 3942
ꢃ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3941

M. E. Maier and A. Zhdanko
[14] Prior to the generation of vinyl gold, the initial PrSpH+ has to be
neutralized to prevent protodeauration, occurring otherwise imme-
diately (LiOH was used as indicated on the Scheme). The identity
of B2 follows from the independent synthesis from D2. The complex
(S)-Tol-BINAPAuPPh3 is known from ref. [15].
[15] A. Zhdanko, M. Strçbele, M. E. Maier, Chem. Eur. J. 2012, 18,
14732–14744.
Ph3PAuACTHNGUTERNNUG
(Sol)+ +2 PrSp+H2O is strongly shifted to the left (rather
(Ph3PAu)3O+ is stoichiometrically obtained by this equation). But
there is no doubt that the highly reactive dicationic species IV has
to be involved in the equilibrium in an undetectable amount. If
strong acid is used instead of PrSpH+ (for example TfOH),
(Ph3PAu)3O+ will simply undergo electrophilic substitution at
oxygen to give Ph3PAuOTf and H2O as final products. More discus-
sion about mechanisms of formation and interconversion between
+
[16] S. Gaillard, J. Bosson, R. S. Ramón, P. Nun, A. M. Z. Slawin, S. P.
Nolan, Chem. Eur. J. 2010, 16, 13729–13740.
various aurated oxonium ions, LAuOH and LAu
ref. [15].
(Sol)+ was given in
ACHTUNGTRENNUNG
[17] A related intramolecular aminoauration of alkenes by (Ph3PAu)3O+
was reported. This process also required a prolonged reaction time.
See: R. L. LaLonde, J. Brenzovich, W. E. D. Benitez, E. Tkatchouk,
K. Kelley, W. A. Goddard III, F. D. Toste, Chem. Sci. 2010, 1, 226–
233.
[23] Although sometimes (Ph3PAu)3O+ is used alone as a sole catalyst,
its activity is naturally much less than the activity of normal cationic
LAuACHTNUTGRNEUNG
(Sol)+ catalysts. This is evidenced by prolonged reaction times
or diminished yields of the catalytic reactions. However, the milder
[18] The inability of (Ph3PAu)3O+ to give diaurated species nicely dem-
onstrates that any catalytic process with this salt taken as a catalyst
will occur sluggishly (or not at all), not because of slow protodeaura-
tion, but simply because of inability to form the necessary organo-
gold intermediates.
[19] Since LAuOH is known to be a strongly basic compound, it is quite
sure to assume that if (LAu)2O existed it would be even more basic;
it would be considered as an organic soluble analogue of K2O (simi-
larly as LAuOH can be considered as organic soluble analogue of
KOH).
[20] H. Schmidbaur, S. Hofreiter, M. Paul, Nature 1995, 377, 503–504.
[21] Similarly, protonation of (LAu)2OH+ can be considered in addition
to the simple ligand exchange possible with this complex.
[22] PrSpH+ naturally is a very weak acid to react with (Ph3PAu)3O+
alone and the total equilibrium (Ph3PAu)3O+ +2 PrSpH+ +3 Sol$3
reactivity of (Ph3PAu)3O+ becomes advantageous in the case of sen-
sitive substrates in which the active LAuACTHNUTRGNEUNG
(Sol)+ catalysts caused de-
composition or an unselective reaction. This can be understood
from the reports of Toste et al. concerning cycloisomerizations of
enynes: a) P. H.-Y. Cheong, P. Morganelli, M. R. Luzung, K. N.
Houk, F. D. Toste, J. Am. Chem. Soc. 2008, 130, 4517–4526; b) B. D.
Sherry, F. D. Toste, J. Am. Chem. Soc. 2004, 126, 15978–15979. Ac-
cording to our data, activity of (Ph3PAu)3O+ can be greatly en-
hanced (while still maintaining the mild conditions) if it is used in
combination with a suitable mild acid (PrSpH+ and stronger) or an-
other suitable oxophilic electrophile.
[24] A. Zhdanko, unpublished results.
Received: December 17, 2012
Published online: February 11, 2013
3942
www.chemeurj.org
ꢃ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 3932 – 3942