The role of gold acetylides as a selectivity trigger and the importance of gem-diaurated species in the gold-catalyzed hydroarylating-aromatization of arene-diynes

Article abstract of DOI:10.1021/om200946m

Terminal 1,2-dialkynylarenes undergo an unexpected cyclization hydroarylation reaction toward naphthalene derivatives in benzene as the solvent. The regioselectivity of the reaction can be controlled by careful catalyst tuning. Also, the preparation of a bench-stable cationic amine complex or simple heterogenization of the catalyst on neutral aluminum oxide, which enables efficient catalyst recycling, was possible. Intensive mechanistic investigations were undertaken, giving new insights into the so-far underestimated role of acetylides in gold chemistry. The gold plays a fascinating dual role serving to both catalyze the reaction and activate the substrate by AuC-bond formation. Evidence of gem-diaurated compounds playing an important part for gold catalysis is also reported.

Full text of DOI:10.1021/om200946m

Article
pubs.acs.org/Organometallics
The Role of Gold Acetylides as a Selectivity Trigger and the
Importance of gem-Diaurated Species in the Gold-Catalyzed
Hydroarylating-Aromatization of Arene-Diynes
A. Stephen K. Hashmi,* Ingo Braun, Matthias Rudolph, and Frank Rominger
Organisch-Chemisches Institut, Ruprecht-Karls-Universitat Heidelberg, Im Neuenheimer Feld 270, 69120 Heidelberg, Germany
̈
S
* Supporting Information
ABSTRACT: Terminal 1,2-dialkynylarenes undergo an un-
expected cyclization hydroarylation reaction toward naphtha-
lene derivatives in benzene as the solvent. The regioselectivity
of the reaction can be controlled by careful catalyst tuning.
Also, the preparation of a bench-stable cationic amine complex
or simple heterogenization of the catalyst on neutral aluminum
oxide, which enables efficient catalyst recycling, was possible.
Intensive mechanistic investigations were undertaken, giving new insights into the so-far underestimated role of acetylides in gold
chemistry. The gold plays a fascinating dual role serving to both catalyze the reaction and activate the substrate by Au−C-σ bond
formation. Evidence of gem-diaurated compounds playing an important part for gold catalysis is also reported.
Scheme 1. Intermolecular Domino Reaction of 1,5-Diynes
with Methanol
1. INTRODUCTION
In the last 11 years, the use of homogeneous gold complexes
for a broad array of organic transformations has become an
established field in modern organic chemistry.1 Because of their
remarkable ability to activate alkynes, gold complexes are frequently
used as catalysts for the nucleophilic addition of heteroatoms to
multiple bonds.2
Furthermore, enynes are often applied as substrates that give
access to multifaceted reactions with incredible gain of
molecular complexity.3 Recently, diynes have attracted gold
chemists as well, opening the field for new domino reactions by
initial attack of a nucleophile to the triple bond, followed by
subsequent reactions with the so-formed reactive intermediates
(for example, an enol ether).4
Because of their easy accessibility via 2-fold Sonogashira
coupling, we considered 1,2-dialkynyl benzenes as suitable
substrates for possible domino reactions with external
nucleophiles.5 In the course of our investigation, we discovered
an unexpected cyclization arylation reaction with benzene. The
ability to influence the selectivity of this reaction and detailed
mechanistic studies are the topics of this contribution.
as well as phosphites and isonitrile ligands were less competent
catalysts. A subsequent solvent screening revealed that the best
results were obtained with nonpolar solvents, such as benzene
or toluene, while an increase in polarity of the solvent, for
example, dichloromethane or dioxane, led to a significant
decrease in yields.6 While methyl-substituted diyne 1b delivered
comparable results under the optimized conditions (Scheme 1),
no reaction or only slow and unselective conversions were
observed with sterically hindered alkynes, such as tert-
butylacetylene, TMS-protected alkynes, and phenyl-substituted
alkynes. Because of their high reactivity, terminal alkynes led to
complex mixtures of inseparable products. In addition to the
problems resulting from bulky substituents on the alkynes,
another drawback of this strategy was the impossibility of
controlling the regio- and positional selectivity of the
nucleophilic attack in the case of the intermolecular addition
of the nucleophile to unsymmetrical substrates.
2. RESULTS
2.1. The Initial Observation: Unexpected Reaction
Course and Atypical Regioselectivities. Our initial studies
were focused on the intermolecular addition of alcohols to
symmetrically substituted 1,2-dialkynyl benzenes. After an
intensive catalyst screening for the reaction of n-butyl-
substituted starting diyne 1a and methanol,6 we were able
to obtain 65% of the desired α-naphthol derivative 2a at room
temperature (Scheme 1).7 Besides the SPhos-ligand,8 other
Buchwald-type ligands, NHC ligands,9 and a MeO-KITPHOS
ligand10 showed comparable results, while simpler phosphanes
When screening other nucleophiles with terminal diyne 1c in
benzene, instead of the expected products arising from the
incorporation of the nucleophile, we observed the incorporation
Received: October 12, 2011
Published: January 11, 2012
© 2012 American Chemical Society
644
dx.doi.org/10.1021/om200946m | Organometallics 2012, 31, 644−661

Organometallics
Article
Scheme 2. Intermolecular Addition of Benzene to 1c
of benzene. A mixture of two isomers was isolated; their
ratio depended on the nucleophile used in the reaction.
Therefore, we performed the reaction in benzene in the
absence of other nucleophiles. Two products were gained in an
overall yield of 85% and in a 2:1 ratio (Scheme 2). Fortunately,
we were able to obtain crystals of both products. The results of
the X-ray crystal structure analyses unambiguously proved the
formation of α- and β-phenylnaphthalenes α-2c and β-2c
(Figure 1).
of the in situ activation led to a slightly different product
distribution (α:β = 67:33 vs 51:49; entries 1 and 2). The reason
for this difference in ratios might be the incomplete activation
process. This was further confirmed through the experiments
discussed in section 2.2.5 and indicated by the results of the use
of IPrAuCl without any further activation (entry 3). In this
case, a high preference for the β-isomer (α:β = 7:93) was
found, albeit after prolonged reaction times.
To exclude a formation of the β product by a rearrangement
after an initial Markovnikov addition, the α product was
subjected to the gold catalyst under the reaction conditions. No
conversion to the β product could be observed (Scheme 3,
upper part). Furthermore, model substrate 3 was used to verify
if a possible alkene formation (from the alkyne and benzene,
Markovnikov hydroarylation) in the first reaction step and a
subsequent rearrangement to the thermodynamically more
stable stilbene derivative 4 are possible (Scheme 3, lower part).
Like in the previous case, no isomerization was observed under
the reaction conditions. The enyne corresponding to 3 has an
entirely different reactivity; see section 2.3.3.
Figure 1. Solid-state molecular structures of α-2c (left) and β-2c
(right).
Scheme 3. Control Experiments Excluding a Formation of
β-2c from α-2c
2.2. Screenings of Different Parameters and Their
Influence on the Selectivity. 2.2.1. Nature of the
Catalyst. Encouraged by these findings, a new catalyst
screening was performed for the benzene addition reaction as
well (Table 1 and Figure 2). Control experiments without
catalyst, with p-TsOH, as well as with AgNTf2 did not show
promising results. The use of simple complexes, such as AuCl
and AuCl3, also led to poor results, even if silver salt11 was
added.6 In contrast to the addition of methanol, SPhos and
KITPHOS complexes only showed moderate performance for
the reaction with benzene (entries 12 and 13). Of all the
screened gold complexes, only one phosphite complex (entry 15)
and the IPrAuNTf2 NHC complex (entry 1) showed good to
excellent yields, while structurally related complexes possessed
only poor reactivity (entries 7, 8, 9, 14). While most of the
complexes led predominantly or exclusively to the formation of
the expected α product, significant amounts of the unexpected
β-naphthalene were observed with the IPr−NHC complex.
Gold-catalyzed intermolecular nucleophilic addition reactions
with terminal alkynes usually form the Markovnikov-type
products, and only a few examples are reported for the anti-
Markovnikov case.12 A remarkable counterion effect on the
selectivity was observed during the screening of different silver
salts. While silver hexafluoroantimonate (entry 4) delivered α-
naphthalene α-2c as the main product, silver triflate delivered
mainly the β regioisomer (entry 5). Silver triflimide (entry 2)
led to an equal mixture of the two isomers. Interestingly, the
use of an isolated preactivated gold−triflimide complex instead
2.2.2. Influence of the Catalyst Loading at Room
Temperature. Different sets of experiments were performed
to understand the influence of the catalyst on the selectivity.
Interestingly, if the reaction was performed not at 80 °C, but at
room temperature, a shift in selectivity toward α product α-2c
was detected. However, only low conversions were observed
using the identical catalyst loading of 5 mol % (Table 2, entry 1).
Therefore, we increased the amount of added catalyst, and complete
conversion was possible with 15 mol % catalyst with an excellent
α-selectivity (Table 2, entries 2 and 3), albeit after prolonged
reaction times.
645
dx.doi.org/10.1021/om200946m | Organometallics 2012, 31, 644−661

Organometallics
Article
a
Table 1. Screening of Different Catalysts
entry
catalyst
IPrAuNTf2
cocatalyst
time (h)
conv (%)
yield α-2c (%)
yield β-2c (%)
α-2c:β-2c
total yield (%)
1
2
6
100
100
95
60
44
5
30
42
62
20
48
7
67:33
51:49
7:93
90
86
67
78
79
69
13
23
49
19
34
24
32
17
60
IPrAuCl
AgNTf2
24
10 days
24
24
6
3
IPrAuCl
4
IPrAuCl
AgSbF6
AgOTf
AgNTf2
AgNTf2
AgNTf2
AgNTf2
AgNTf2
AgNTf2
100
100
100
44
58
31
62
12
21
44
18
31
13
13
17
60
74:26
39:61
90:10
92:8
5
IPrAuCl
6
IPrAuIIICl3
NACAuCl
IBioxAuCl
SIPriPrAuCl
Ph3PAuCl
(o-CF3-Ph)3PAuCl
SPhosAuNTf2
7
24
24
24
24
24
24
24
24
24
1
8
61
2
91:9
9
88
5
90:10
95:5
10
11
12
13
14
15
59
1
b
83
3
91:9
95
11
19
0
54:46
41:59
100:0
100:0
MeO-KITPHOS-AuNTf2
(F3CCH2O)3PAuCl
100
66
AgNTf2
AgNTf2
(2,4-di-tBuPhO)3PAuCl
100
0
a
Unless stated otherwise, reaction of diyne 1c (12 mg) was carried out with 5 mol % catalyst and 15 mol % cocatalyst in benzene (0.5 mL) in a small
b
vial at 80 °C. Conversions and yields were determined after 24 h by GC using 1 equiv of hexamethylbenzene as an internal standard. Synthetic
procedure and crystallographic data are given in the Supporting Information.
Table 2. Influence of the Catalyst Loading on the
Conversion and Yield for IPrAuNTf2 at Room Temperature
a
catalyst
loading
entry (mol %)
yield yield
total
yield
(%)
temp
(°C)
time conv α-2c β-2c
α-2c:
β-2c
(h) (%) (%)
(%)
1
2
3
5
∼20 (rt) 48
∼20 (rt) 48
∼20 (rt) 48
43
92
31
70
80
1
1
1
97:3
99:1
99:1
32
71
81
10
15
100
a
Conversions and yields were determined by GC using 1 equiv of
hexamethylbenzene as an internal standard.
(Figure 4). Similar observations could be made from the other
experiments at elevated temperatures.
2.2.4. Catalyst Loading at Elevated Temperature. In
addition to the experiments at room temperature from
Table 2, we then tested different catalyst loadings, but at
elevated temperatures where a competing β pathway should be
possible.6 Figure 5 illustrates the high dependency of the
reaction selectivity on the catalyst loading. High catalyst
loadings lead to the favored formation of the α isomer,
whereas the β product was mainly formed when using low
catalyst loadings.
To check if the active catalyst was changed under the
reaction conditions at elevated temperatures, the IPrAuNTf2
was heated in benzene for 48 h at 80 °C prior to the addition of
the starting material. No effect on the conversion rates or
influence on the selectivity was observed.
2.2.5. Additives. Inspired by the finding that the in situ
activation led to a different selectivity than the preactivated
catalyst (see Table 1), we then focused on the use of different
additives (Table 3). An increasing amount of nonactivated
IPrAuCl as an additive led to higher amounts of the β product
(entries 1−3). Whereas silica gel as an acidic support did not
Figure 2. Gold catalysts used for the screening.
2.2.3. Influence of the Temperature on the Selectivity. To
investigate the effect of the temperature on the selectivity,
reaction temperatures were varied (room temperature; 40, 60,
and 80 °C), and the product selectivity was determined via GC
techniques after different conversion times.6 For all the
monitored temperatures, a fast conversion to the α product
was observed in the beginning of the reaction and almost no β
product was formed at this stage. As visible from Figure 3
(reaction at 60 °C) after an initiation phase, a linear formation
of the β product was observed and the rate of formation of the
α product was slower than before (but also linear). As the rate
of the β product formation is higher than that of the α product
formation after the initiation phase, the selectivity of the
reaction shifts to higher β:α ratios as the reaction progresses
646
dx.doi.org/10.1021/om200946m | Organometallics 2012, 31, 644−661

Organometallics
Article
Figure 3. Conversion and yields for different reaction times at 60 °C (5 mol % IPrAuNTf2).
Figure 4. β:α ratios at different conversions (5 mol % IPrAuNTf2, 60 °C).
Figure 5. Dependency of the selectivity on the catalyst loading (IPrAuNTf2, 80 °C).
have a great impact on the reaction course (entry 4), the use of
both basic and neutral aluminum oxide greatly enhanced the
formation of the β isomer (entries 5−8). Even if only 1 equiv of
aluminum oxide was used, a significantly higher amount of the
β product was formed (entry 5). Furthermore, even at room
temperature, a high β-selectivity was possible by the use of
additional aluminum oxide, but reaction rates turned out to be
low in this case (entry 8). Triethylamine as a basic additive was
647
dx.doi.org/10.1021/om200946m | Organometallics 2012, 31, 644−661

Organometallics
Article
the gold catalyst with triethylamine in benzene in a larger scale,
and we were able to isolate a quantitative amount of the amine
complex 5 (Scheme 4). The assignment of the structure could
also be confirmed by an X-ray crystal structure analysis (Figure 6).
Table 3. Influences of Additives on the Selectivity (5 mol %
IPrAuNTf2)
a
yield
α-2c
(%)
yield
β-2c
(%)
total
yield
(%)
time
(h)
α-2c:
β-2c
entry
additive
equiv
b
1
6
6
38
22
18
52
15
52
67
70
33
70
42:58
25:75
20:80
61:39
18:82
90
89
88
85
85
Scheme 4. Preparation of Amine Complex 5
b
2
IPrAuCl
IPrAuCl
SiO2
0.1
b
3
0.25
1.0
6
c
4
6
c
5
Al2O3
(neutral)
1.0
24
d
6
Al2O3
(neutral)
1
24
3
3
9
84
74
80
3:97
4:96
87
77
89
d
7
Al2O3
(basic)
de
,
8
Al2O3
(neutral)
192
10:90
9
NEt3
1.0
0.1
24
24
2
2
86
89
2:98
2:98
88
91
10 NEt3
a
Conversions and yields were determined by GC using 1 equiv of
hexamethylbenzene as an internal standard. In all cases, full conversion
b
c
was observed. 2.5 mol % IPrAuNTf2. The additive was heated with
the catalyst for 1 h at 80 °C before the standard solution of the
d
substrate and the internal standard were added. The catalyst was
loaded on top of a pipet with 500 mg of Al2O3. Benzene (2 mL) was
flashed through the pipet, and the dried ALOX/catalyst mixture was
e
used in the reaction. Performed at room temperature.
also efficient at 80 °C. With 1 equiv, a nearly perfect
β-selectivity could be achieved (α:β = 2:98; entry 9), and
even catalytic amounts of triethylamine (10 mol %) delivered
an excellent selectivity in comparable reaction times (entry 10).
Different pyridine bases were also tested as additives. Here,
only the bulky 2,6-di-tert-Bu-pyridine led to complete
conversions and the selectivity enhancement was less effective
(α:β = 10:90, 5 mol %, 80 °C) than with triethylamine.6
Repeated reaction runs were possible for adsorbed gold
catalyst on neutral aluminum oxide (Table 4). Up to five runs
Figure 6. Solid-state molecular structure of 5 (Au−N = 2.092(7) pm;
Au−C = 1.949(8) pm; C−Au−N = 179.5(3)°).
To investigate if this new catalyst would be a bench-stable
catalyst for the selective formation of the β-naphthalene
derivative β-2c, we performed the transformation without any
additional additives. To our delight, the desired product was
formed in excellent selectivity (α:β = 2:98, Scheme 5).
Table 4. Adsorption on Neutral Al2O3 (5 mol % IPrAuNTf2,
80 °C)
Scheme 5. β-Selective Reaction with Catalyst 5
a
conv
(%)
time
(h)
yield α-2c
(%)
yield β-2c
(%)
α-2c:
β-2c
total yield
(%)
run
1
2
3
4
5
100
100
100
100
100
1
4.5
10
18
26
3
4
4
4
3
84
86
80
82
70
3:97
4:96
5:95
5:95
4:96
87
90
84
86
73
2.3. Mechanistic Investigation. 2.3.1. Reactivity of
Gold Acetylides Formed under the Reaction Conditions.
On the basis of the high influence of basic additives on the
reaction, we concentrated on gaining further insight into the
origin of the change in selectivity and the mechanism of this
reaction. As the formation of gold acetylides from terminal
alkynes is a common reaction under basic conditions,13 we
considered gold acetylides as intermediates in the catalytic
cycle.14
a
The catalyst was loaded on top of a pipet with 500 mg of Al2O3.
Benzene (2 mL ) was flashed through the pipet, and the dried ALOX/
catalyst mixture was used in the reaction. Conversions and yields were
determined by GC using 1 equiv of hexamethylbenzene as an internal
standard. After one run, the ALOX/catalyst system was filtered,
washed with benzene, and reused in the next run.
were possible without a significant decrease in the turnover
number. After each run, the catalyst could be recovered easily
by simple filtration, but the catalytic activity (turnover
frequency) decreased; the reaction time increased from 1 to
26 h. Still, overall, a remarkable 100 turnovers were achieved.
Attempts to run the reaction with triethylamine at room
temperature failed. Here, a fast formation of a crystalline solid
was observed after the addition of the base to the catalyst
solution even in the absence of the substrate. Hence, we mixed
To verify if a gold acetylide may participate, we synthesized
monogold alkynyl complex 6 (Figure 7).6
Heating the isolated gold acetylide 6 in benzene and in the
absence of additional active catalyst led to no conversion even
after 24 h at 80 °C. Additionally, no conversion was observed
when a catalytic amount of acetylide 6 (5 mol %) was added to
diyne 1c under the same conditions. Hence, we added
additional active catalyst (2.5 mol % IPrAuNTf2) to a mixture
648
dx.doi.org/10.1021/om200946m | Organometallics 2012, 31, 644−661

Organometallics
Article
Scheme 6. Equilibrium between Free Catalyst and Activated Substrate 6
Table 5. Investigation of a Possible Equilibrium between Diyne 1c (5 mol % IPrAuNTf2, 80 °C) and 6 under Different
a
Conditions
b
entry
catalyst
additive
time (h) conv (%) yield α-2c (%) yield β-2c (%) α-2c:β-2c total yield (%)
1
2
3
4
gold acetylide 6
gold acetylide 6
gold acetylide 6
IPrAuNTf2
1 equiv NEt3
6
6
9
100
100
100
0
60
2
2
24
88
30
2
84
90
90
10 mol % HNTf2
72:28
2:98
1 equiv NEt3 and 10 mol % HNTf2
24
6
60
67:33
a
b
Conversions and yields were determined by GC using 1 equiv of hexamethylbenzene as an internal standard. Added after 24 h.
of diyne 1c and 5 mol % acetylide 6 at 80 °C. In this case, the β
product was detected almost exclusively (α:β = 3:97 instead of
42:58 without additive) in 85% yield after 12 h! Thus, the
reaction of the active catalyst with acetylide 6 must be
considerably faster than that with diyne 1c, if one considers the
20-fold excess of diyne 1c. Furthermore, diyne 1c must be
transformed to more reactive acetylide 6 during the catalytic
cycle as otherwise an unselective reaction course would be detected
after all of the acetylide is consumed. The lack of catalytic activity of
the gold acetylide without additional catalyst seems to be paradox;
as in the triethylamine case (as described above, the reaction even
works with 100 mol % of triethylamine, i.e., 20 equiv of base with
respect to the catalyst!), a fast formation of this species should
inhibit the further reaction. In this case, a possible equilibrium of
Figure 7. Molecular structure of acetylide 6 in the solid state
(Au−Calkyne = 1.984(3) pm; Au−Ccarbene = 2.010(2) pm; C−Au−C =
177.10(9)°).
free catalyst and acetylide can be considered. This equilibrium
should be influenced by the base and the corresponding acid that is
formed during acetylide formation (Scheme 6).
To prove this assumption, diyne 1c was subjected to catalytic
amounts of gold acetylide 6 in the presence of different acidic
and basic additives (Table 5). As expected, the control
experiment with additional triethylamine did not show any
catalytic activity (entry 1). Addition of bistrifluoromethane-
sulfonimide (10 mol %, entry 2) indeed led to catalytic activity;
in this case, an even higher α-selectivity was observed than for
the reaction without any additive (entry 4). Under the original
reaction conditions in the presence of triethylamine, the
corresponding triethylammonium triflimide should lead to
buffered reaction conditions. Thus, we considered a mixture of
these two species to simulate the correct acid-to-base ratio that
should exist during the reaction (entry 3). In fact, addition of a
premixed solution of triethylamine and bistrifluoromethane-
sulfonimide to the starting material led to catalytic activity, and
excellent β-selectivity was observed (α:β = 2:98). In summary,
the selectivity can nicely be explained by the above equilibrium.
For an efficient formation of the β product, the acetylide must
coexist with active catalyst in solution, which is only possible in
a buffered medium. Addition of acid shifts the equilibrium from
Scheme 6 to the left side. Therefore, only the diyne 1c and the
649
dx.doi.org/10.1021/om200946m | Organometallics 2012, 31, 644−661

Organometallics
Article
Scheme 7. Bestmann−Ohira Chain Extension of Gold Acetylide 7
Scheme 8. Influence of a Gold Acetylide on the Positional Selectivity
active catalyst coexist, causing the reaction to select for the α
product. If no conjugated acid is present, the whole reaction is
inhibited as no active catalyst for the β-selective conversion of
the acetylide complex 6 is formed. Regarding the frequent use
of terminal alkynes as substrates for gold-catalyzed reactions, it
seems likely that gold acetylides take part as reactive
Figure 9. NMR comparison of the reaction of diyne 1d (top) and the
reaction of gold acetylide 8 (bottom).
positional selectivity that should occur by the use of diyne 1d
as the starting material. The assignment of monoacetylide 8
was proven by the results of an X-ray structure analysis
(Figure 8).
Figure 8. Solid-state molecular structure of 8 (Au−Calkyne = 1.985(5)
pm; Au−Ccarbene = 2.024(5) pm; C−Au−C = 175.50(18)°).
Both of the substrates were subjected to the gold catalyst,
and the selectivity was analyzed by 1H NMR integration
(Figure 9, Scheme 8).6 In the case of gold acetylide 8, aqueous
hydrochloric acid was added after complete conversion to
ensure complete protodemetalation. Remarkably, with mono-
gold acetylide 8, a high selectivity was observed. In this case, the
benzene was almost exclusively added to the alkyne that had
been bearing the NHC−gold unit before! Not surprisingly, the
control experiment with substrate 1d, which contains two
terminal alkynes, led to poor selectivity. Furthermore, in this
case, a majority of the regioisomer β-2d-b was formed. Here,
the selectivity is most probably determined by the difference in
pKa values of the two alkynes based on the mesomeric donor
effect of the methoxy group, which only influences the para
position. The gold acetylide forms at the more acidic alkyne
position meta to the methoxy group, leading to the formation
of β-2d-b as the main product.
intermediates in many of these reactions. To the best of our
knowledge, there are no mechanistic investigations on the effect
of these species concerning selectivity as well as reactivity
properties. Therefore, we focused our investigations on obtaining
more information about the reaction mechanism.
2.3.2. Comparing the Positional Selectivity of Gold
Acetylides and Terminal Alkynes. To explore whether the
terminal alkyne or the gold acetylide is favored for the benzene
addition, we used asymmetrically substituted diyne 1d as a
substrate for the gold catalysis. Furthermore, the corresponding
gold acetylide 8 was prepared from aldehyde 7. A chain
extension using an excess of Bestmann−Ohira reagent15 was
possible even in the presence of the gold acetylide moiety
(Scheme 7). This strategy was used to avoid problems with the
650
dx.doi.org/10.1021/om200946m | Organometallics 2012, 31, 644−661

Organometallics
Article
Scheme 9. Mechanistic Consideration of a Benzene Addition Prior to the Cycloisomerization
Table 6. Product Distribution for the Conversion of the Possible Intermediates 9 and 10
2.3.3. Exclusion of the Participation of Enynes Generated
by Initial Hydroarylation. One explanation for the difference
between terminal alkynes and gold acetylides could be the
formation of two regioisomeric enyne systems 9 and 10, which
undergo selective intramolecular cyclization to the naphthalene
isomers (Scheme 9). In the case of a terminal alkyne, a
Markovnikov-type addition would be expected, which should
lead to enyne 9. This would be consistent with previously
reported intermolecular hydroarylations of terminal alkynes.16
If the gold caused an umpolung of the alkyne in the gold
acetylide, the anti-Markovnikov-type addition could lead to the
β isomer.
Scheme 10. Crossover Labeling Experiments with Additional
Water
Thus, we synthesized the possible enyne compounds 9
and 10 and subjected them to the catalyst (Table 6). The
conversion of enyne 9 in benzene led to an inseparable mixture
of products derived from an unselective 5-exo-dig and 6-endo-dig
cyclization process (entry 1). As there were reports with similar
systems using dichloromethane as a solvent,17 we used these
published conditions as well (entry 1; parentheses), but no
significant change in selectivity was observed. On the basis of
these findings, a mechanism that involves an initial
intermolecular formation of an enyne system can be excluded
for the formation of the α product. This was also confirmed by
the fact that, in all the aforementioned catalyses, we never
observed any fulvene derivatives like 11! The catalysis with
regioisomeric stilbene derivative 10 only showed very slow
conversion (entry 2). Only a complex mixture of inseparable
compounds was observed. To exclude an electronic effect from
the methyl groups in 1c, unsubstituted diyne 1e was converted
under the optimized conditions (entry 3). In contrast to the
above-mentioned enyne systems, an exclusive formation of
651
dx.doi.org/10.1021/om200946m | Organometallics 2012, 31, 644−661

Organometallics
Article
Figure 10. Isotopic labeling studies: (A) 1c + C6H6; (B) 1c + C6D6; (C) 1c-d2 + C6H6; (D) 1c-d2 + C6D6.
naphthalene derivates α-2e and β-2e could be determined in
excellent β-selectivity (entry 3). For this reason, a participation
of an enyne of type 10 can also be excluded as intermediates
during the formation of the β products.
of H2O and D2O were performed. To avoid a base-catalyzed
H/D exchange in the presence of water, gold acetylide 6 was
used as an additive. As visible from Scheme 11, only the
Hb-position is affected by additional water in the solvent. Here,
a remarkable isotopic effect can be monitored, whereas in the
case of D2O, only 31% of D is incorporated in the Hb-position,
and in the case of H2O, 90% of H was embedded (Scheme 11).
Furthermore, no water addition was observed even if an excess
of water was present.
In combination with the previous experiments, it is obvious
that the proton in the Hb-position must originate from the
alkyne proton itself and that traces of water in the solvent only
play a minor role.
The most plausible explanation is that a gold atom is located
at the Hb-position during the last stage of the reaction cascade.
As no exchange was visible with benzene at this position and,
therefore, also protodemetalation via the benzene is impossible,
we elucidated if a possible catalyst transfer from an aryl gold
intermediate to diyne 1c is possible.18
2.3.5. Proof of the Catalyst Transfer. Indeed, we could prove
a quantitative ligand exchange in a stoichiometric experiment with
phenyl gold precursor 12 and starting diyne 1c (Scheme 12 and
Figure 11). As during most of the conversion (with the exception
of the final phase), a great excess of diyne 1c is present under the
reaction conditions, it seems very likely that a catalyst transfer of
an aryl gold intermediate to starting diyne 1c occurs during the
reaction.
2.3.4. Isotopic Labeling Experiments. For further mecha-
nistic insights, 1c-d2 was prepared. The undeuterated 1c as well
as deuterated 1c-d2 were both converted in deuterated and
undeuterated solvent with 10 mol % triethylamine as an
additive (Scheme 10 and Figure 10).6 To exclude exchange
processes, 1c-d2 was heated for 16 h in deuterated benzene.
Even when H2O was added, no H/D exchange was observed.
The experiment with undeuterated 1c in deuterated benzene
(under β-promoting conditions) led to a 90% incorporation of
the deuterium at the α-position (Figure 10B). The other
positions in the product are not affected. The crossover
experiment with 1c-d2 in undeuterated benzene also showed
a nearly quantitative incorporation of a former benzene
proton in the α-position. Furthermore, Hc was not affected,
and 75% of deuterium was still present in the Hb-position. As
there are no other deuterium sources, the high level of
deuteration at the Hb-position must originate from a former
alkyne (Figure 10C).
Finally, 1c-d2 was converted in deuterated benzene. In this
case, Ha and Hc were not affected. Interestingly, a moderate H/
D exchange was observed for the Hb-position. In this case,
traces of water in the solvent are responsible. To prove this
assumption, crossover experiments in the presence of an excess
652
dx.doi.org/10.1021/om200946m | Organometallics 2012, 31, 644−661

Organometallics
Article
Figure 11. Ligand-exchange experiment.
Catalytic amounts of various organo-gold additives provided
further proof for this hypothesis. Even if only 10 mol % of the
additive was added at the same time with the catalyst, extremely
high β-selectivity was observed for all of the tested species
(Table 7). This indicates that a very fast exchange process must
take place initiating the β-cycle by formation of the gold
acetylide 6.
Having proven the catalyst transfer mechanism, we wanted to
verify the kinetics/product selectivity experiments from Figure
3 under these optimized conditions. The results are shown in
Figure 12. We were delighted to see that, indeed, no initiation
phase is observed. The β product is formed from the very
beginning, and the product formation is almost linear for most
of the conversion. Under these optimized conditions, the selec-
tivity for the β product is high throughout the whole conversion;
even at the end, a β:α ratio of 98:2 is detected.
because of the formation of amine complex 5. Considering the
above-mentioned possibility of adding organo-gold compounds
for the initiation of the β-cycle, we were curious if a β-selective
reaction was possible at room temperature by the use of gold
acetylide 6 as an additive. After the addition of the gold catalyst
to the starting diyne and catalytic amounts of acetylide, an
immediate precipitation of bronze-colored tiny needles took
place and no formation of a naphthalene was monitored. An
experiment with stoichiometric amounts of gold acetylide and
IPrAuNTf2 gave the same result. By simple filtration of the
residue and washings with pentane, we were able to obtain a
solid that could be fully characterized (Scheme 11).
Fortunately, we were able to obtain crystals suitable for X-ray
structure analysis as well. The result of this analysis
unambiguously proves the formation of a gem-diaurated aryl
species 16 (Figure 13). While several reports of related species
exist,14a,19 including an exciting β-naphthyl digold complex
of Osawa et al.,19j only one other publication reported the
isolation of such species from a catalysis reaction. In that
2.3.6. Isolation of Stable Organo-Gold Intermediates. As
mentioned earlier, our efforts to perform a β-selective reaction
at room temperature in the presence of triethylamine failed
a
Table 7. Influences of Various Organo-Gold Additives (10 mol %) on the Selectivity (5 mol % IPrAuNTf2, 80 °C)
entry
additive
time (h)
yield α-2c (%)
yield β-2c (%)
α-2c:β-2c
total yield (%)
1
2
3
4
12
13
14
15
6
6
2
3
2
3
85
84
93
88
2:98
3:97
2:98
3:97
87
87
95
91
13.5
6
a
Conversions and yields were determined by GC using 1 equiv of hexamethylbenzene as an internal standard. Additive and catalyst were added at
the same time to a standard solution of substrate and internal standard.
653
dx.doi.org/10.1021/om200946m | Organometallics 2012, 31, 644−661

Organometallics
Article
Scheme 11. Stoichiometric Preparation of gem-Diaurated Species 16
Scheme 12. Isolation of Monogold Aryl Intermediate 17
shows a strong aurophilic interaction,22 which is consistent
with reported data for diaurated compounds.23 Interestingly,
the Au−C1 bond distances (212.5 and 213.8 pm) are also in
the range of other known compounds bearing phosphane
ligands at the gold atom. This indicates that the influence of the
ligand at the metal center plays a minor role. In comparison to
the Au−C1 bond length of monogold compound 17 (203.5
pm), a significant elongation in bond length can be observed
for the diaurated species. Regarding the naphthalene skeleton,
there is an elongation of the C1−C3 bond as well as the C1−
C2 bond by the addition of each of the gold atoms. The most
case, an allene starting material was applied.20 Furthermore, no
gem-diaurated species containing carbene ligands at the metal
center can be found in the literature.
Interestingly, no precipitation of complex 16 and no change
in color of the reaction media were observed in the presence of
triethylamine. However, a complete conversion was visible with
thin-layer chromatography, and a new compound could be
detected that had a different Rf value than the usual
naphthalene products. In a substoichiometric experiment
(lower catalyst loadings led to slow conversions in the presence
of excess amounts of coordinating triethylamine), we were able
to isolate this compound by column chromatography on
neutral aluminum oxide (Scheme 12). Again, it was possible to
obtain a crystal structure. The results of the analysis are
depicted in Figure 14. In this case, a monogold aryl compound
17 was isolated.2a,b,21 The gold atom is located at the position
were the H/D exchange was visible in the deuteration
experiments.
A short summary of the relevant data gained from the X-ray
crystallographic analyses of the organo-gold intermediates 16
and 17 as well as β product β-2c is depicted in Table 8. The
Figure 13. Solid-state molecular structure of 16 (Au−Au = 2.7607(3)
pm; Au−Ccarbene = 2.018(6), 2.008(7) pm; Au−Carene = 2.135(6) pm;
Au−Au−Ccarbene = 152.66(18)°, 149.63(19)°; Carene−Au−Au =
49.54(18)°, 49.74(17)°).
Au−Au bond length of gem-diaurated compound 16 of 276.1 pm
Figure 12. Conversion and yields for different reaction times with the optimized reaction conditions analogue to Table 7, entry 1 (benzene, 80 °C,
5 mol % IPrAuNTf2, 10 mol % IPrAuPh as the additive to induce gold acetylide formation).
654
dx.doi.org/10.1021/om200946m | Organometallics 2012, 31, 644−661

Organometallics
Article
a steric interaction of the phenyl group and the bulky carbene
ligand at the metal center.
2.3.7. Catalytic Activity of the Isolated Intermediates. With
the isolated gold intermediates in hand, we tested their
catalytic activity for the naphthalene synthesis (Table 9). Not
unexpectedly, monogold compound 17 alone did not show
any activity. The 5% of product that was obtained most
probably stems from protodemetalation via diyne 1c. As
mentioned earlier, the monogold acetylide 6 does not react in
the absence of a weak acid as no active cationic catalyst exists
(see also Scheme 6). Addition of active catalyst IPrAuNTf2 to
this mixture led to a selective conversion to the β product
(entry 2). This gives further proof that it is the formation of a
gold acetylide from organo-gold compound 17 that initiates the
β-selective cycle. Last, but not least, the gem-diaurated species
16 was checked for its catalytic activity (entry 3). In this case,
even in the absence of an additional catalyst, a fast conversion
to the β product was monitored. We assume that, at the
elevated temperature, there is an equilibrium between the gem-
diaurated compound 16, the monogold aryl 17, and the active
IPrAuNTf2 catalyst.24
Figure 14. Solid-state molecular structure of 17 (Au−Carene = 2.035(6)
pm; Au−Ccarbene = 2.013(6) pm; C−Au−C = 178.0(2)°).
significant change can be monitored for the C1−C3 length,
between the naphthalene β-2c (138.9 pm) and the gem-diaurated
compound 16 (145.5 pm).
The triangle in the diauryl compound, which is formed by
the C1−Au1−Au2 atoms, shows a significant slippage of 10.45°
with respect to the naphthalene plane. This might be caused by
2.4. Discussion of the Possible Reaction Mechanism.
Schemes 13 and 14 depict our mechanistic hypothesis of this
complex reaction in the presence or in the absence of a base,
Table 8. Comparison of Structural Parameters of β-2c, 17, and 16
Table 9. Catalytic Activity of Intermediates 16 and 17
entry
catalyst system
aryl-AuIPr 17 (5 mol %)
time (h)
conv (%)
yield α-2c (%)
yield β-2c (%)
α-2c:β-2c
total yield (%)
1
2
3
24
6
13
100
100
0
2
5
5
95
83
5
97
88
aryl-AuIPr 17 (10 mol %) + IPrAuNTf2 (5 mol %)
[aryl-(AuIPr)2][NTf2] 16 (2.5 mol %)
2:98
6:94
6
655
dx.doi.org/10.1021/om200946m | Organometallics 2012, 31, 644−661

Organometallics
Article
Scheme 13. Mechanistic Consideration for the Catalytic Cycle in the Presence of Triethylamine as Additive
respectively. To avoid a highly complex mechanistic scheme,
they are drawn separately.
naphthalene gold intermediate I. The proton, which is released
during the benzene addition step, then protodemetalates the
catalyst, forming the α-product α-2c. Competing with simple
protodemetalation, a catalyst transfer from intermediate I to
diyne is also possible. This initiates the β-cycle by formation of
the key intermediate 6. A fast cyclization of 6 then takes place,
but, at room temperature, all of the free catalyst is trapped via
the precipitation of diaurated compound 16. This also inhibits
the α-cycle, which explains the incomplete conversions at room
temperature for low catalyst loadings.
Scheme 13 is our mechanistic rationale for the reaction
involving a base, such as triethylamine, as an initiator of the
β-cycle. The upper cycle represents the aforementioned
equilibrium between the free catalyst and gold acetylide 6
(from Scheme 6). At room temperature, no product β-2c is
formed. In this case, only gold acetylide 6 can be monitored via
thin-layer chromatography, indicating that not enough free
catalyst is available at room temperature.
However, at higher temperatures, a balanced equilibrium
between the free catalyst and activated substrate 6 exists.
Because of the higher reactivity of the gold acetylide 6, only
traces of the α product are now formed. The β-selective
catalytic cycle now starts with gold acetylide 6, which reacts to
form the monogold species 17. A subsequent catalyst transfer
from 17 to starting diyne 1c leads to regeneration of the gold
acetylide 6 after release of the β product.
At higher temperatures, an equilibrium between diaurated
compound 16 and monogold aryl 17 exists, and therefore, free
catalyst is available for the reaction. As the amount of free
catalyst is reduced, the reaction rate for the α product slows
down. The rate-limiting step for the formation of β-2c is most
probably the ligand exchange between monogold compound 17
and diyne 1c, which regenerates gold acetylide 6. This
assumption is based on the linear reaction course even at the
late state of the reaction (otherwise, the formation of β-2c
should accelerate at this stage). The initiation of the β-cycle is
also possible by the addition of an external organo-gold
A different mechanism must be considered in the absence
of a base (Scheme 14). At the beginning of the reaction,
diyne 1c reacts with free catalyst to produce α-phenyl
656
dx.doi.org/10.1021/om200946m | Organometallics 2012, 31, 644−661

Organometallics
Article
Scheme 14. Mechanistic Consideration without Base
compound. Here, even organo-gold compounds that are not
intermediates of the β-cycle can be applied as long as they are
able to perform a ligand exchange with diyne 1c. As always, an
excess of additive in regard to the activated catalyst was applied,
and a constant amount of gold acetylide 6 is available at any time.
Thus, the rate-limiting step is no longer the catalyst transfer, and
the selectivity is controlled efficiently by the different cyclization
rates of free alkyne 1c and monoacetylide 6.
657
dx.doi.org/10.1021/om200946m | Organometallics 2012, 31, 644−661

Organometallics
Article
Our mechanistic insight for the selectivity-determining step is
still highly speculative. For the α case, we can exclude an initial
formation of an enyne via Markovnikov-type addition of
benzene onto diyne 1c. Therefore, an initial cyclization seems
likely, which forms a species resembling a naphthyl cation II in
a benzene solvent cage. Nucleophilic attack of benzene then
delivers the final product α-2c after protodemetalation
(Scheme 15).
mesomeric contribution of a bent allene carbenoid structure
IVb.25 Finally, nucleophilic attack by the benzene onto the
highly electrophilic β-position delivers monogold organyl 17
after protodemetalation (Scheme 16).
An alternative mechanistic suggestion is shown in Scheme 17.
This mechanism also potentially could explain the formation of
the β product. An activation at the aurated alkyne V activates for
the reaction with the benzene to deliver VI. A [1,2]-shift of the
proton transfer via transition state VII would deliver the gold
carbenoid VIII. Elimination of [Au]+ to intermediate IX, which
then is activated by the gold catalyst at the other alkyne to deliver
X, finally provides β-2c by a [1,2]-shift of a proton.
Scheme 15. Mechanistic Hypothesis for the Formation of
α-2c
A number of experiments clearly disprove this mechanism. All
efforts to add benzene to a preformed gold acetylide failed. Even
more important, our labeling studies reported above always show
that the proton set free from the benzene always with high
selectivity ends up in the α-position of the product (Figure 10B),
which cannot be explained by the mechanism from Scheme 17.
The most convincing mechanism that is in accordance with
the experiments is depicted in Scheme 18. First, the acetylide 6
is formed by the catalyst transfer or by a basic additive (in the
absence of an additive, maybe even directly from the diyne and
IPrAuNTf2). The dual activation (intermediate XI) then causes
a dual role of gold: one gold center activates the triple bond as
an electrophile by π-coordination; the gold acetylide reacts as a
nucleophile at the β-carbon. Thus, a five-membered ring is
closed (XII). The latter possesses a fulvene and a gold
vinylidene substructure.26 With benzene XIII is formed, a [1,3]-
shift of the proton then provides the gold(I) carbenoid XIV.
One conceivable mechanistic rationale for the formation of
the β product starts with the activation of the terminal alkyne
by a second gold atom. A fast cyclization based on the increased
nucleophilicity of the gold acetylide then delivers a naphthyl
cation III. A related yne-yne cyclization where a nucleophilic
attack of a gold acetylide onto a gold-activated alkyne takes
place has previously been proposed by the Gagosz group during
their synthesis of interesting cycloalkynes.4q
The cationic species III then isomerizes in a fast process via a
1,2-shift of the gold atom to β-naphthyl cation IVa. The latter
might be additionally stabilized by the second gold atom via a
Scheme 16. Mechanistic Hypothesis for the Formation of Monogold Aryl 17
Scheme 17. Mechanistic Model Disproved by the Experimental Results
658
dx.doi.org/10.1021/om200946m | Organometallics 2012, 31, 644−661

Organometallics
Article
Scheme 18. Best Mechanistic Model to Explain the Observed Experimental Facts
A ring expansion by a [1,2]-shift of the reactive allylic group
delivers XV.26f,27 After eliminiation of [Au]+, the observed
arylgold(I) complex28 17 is formed, the latter in equilibrium
with the observed digold species 18.
picture of this highly complex reaction was obtained. There is
evidence for gold vinylidene complexes being involved.
Considering the importance of terminal alkyne substrates in
gold catalysis, these results may influence future progress in
gold chemistry either by achieving new selectivities for known
reactions or by increasing the scope of new reactions based on
the high reactivity of gold acetylides.
3. SUMMARY AND CONCLUSIONS
The results in this paper clearly demonstrate the importance of
gold acetylides in reactions bearing terminal alkynes. The
results also clearly indicate that these species are highly
activated substrates for gold catalysis. To avoid inhibiting the
reaction and to ensure efficient conversion, an equilibrium
between the activated substrate and the free catalyst is
necessary. The generation of gold acetylides can even be
achieved without the addition of an external base. In these
cases, a catalyst transfer from organo-gold intermediates to the
starting alkynes completes the catalytic cycle. Therefore, an
induction phase and a shift in the selectivity of the reaction are
observed. To circumvent the initiation phase, different additives
can be used to preform the acetylides. Even if the acetylide
concentrations are low in regard to the free alkynes, their
enormous reactivity controls the reaction outcome and perfect
selectivitiy can be achieved. For the new cyclization hydro-
arylation sequence with nonactivated benzene, both α- and
β-naphthalene derivatives can be synthesized with high selectivity.
Conveniently, both a simple bench-stable cationic amine
complex and a recyclable, heterogeneous catalyst can efficiently
convert the starting material to the desired β product.
Furthermore, the isolation and unambiguous characterization
via X-ray crystal structure analysis of the first gem-diaurated aryl
gold compound stemming from a prior cyclization reaction
were possible. This is also the only example for a gem-diaurated
compound bearing carbene ligands. The high influence of these
species on gold catalysis has been demonstrated. Together with
other isolated intermediates and labeling experiments, a precise
ASSOCIATED CONTENT
* Supporting Information
Detailed experimental procedures and characterization of new
compounds. This material is available free of charge via the
Internet at http://pubs.acs.org.
■
S
AUTHOR INFORMATION
Corresponding Author
*Fax: (+49)6221-544205. E-mail: hashmi@hashmi.de.
■
ACKNOWLEDGMENTS
Gold salts were generously donated by Umicore AG & Co. KG.
■
REFERENCES
■
(1) (a) Dyker, G. Angew. Chem. 2000, 112, 4407−4409; (b) Angew.
Chem., Int. Ed. 2000, 39, 4237−4239. (c) Hashmi, A. S. K Chem. Rev.
2007, 107, 3180−3211. (d) Jimen
́
ez-Nunez, E.; Echavarren, A. M.
́
Chem. Commun. 2007, 333−346. (e) Furstner, A.; Davis, P. W. Angew.
̈
Chem. 2007, 119, 3478−3519; (f) Angew. Chem., Int. Ed. 2007, 46,
3410−3449. (g) Skouta, R.; Li, C.-J. Tetrahedron 2008, 64, 4917−
4938. (h) Li, Z.; Brouwer, C.; He, C. Chem. Rev. 2008, 108, 3239−
3265. (i) Arcadi, A. Chem. Rev. 2008, 108, 3266−3325. (j) Gorin, D. J.;
Sherry, B. D.; Toste, F. D. Chem. Rev. 2008, 108, 3351−3378.
(k) Hashmi, A. S. K.; Rudolph, M. Chem. Soc. Rev. 2008, 37, 1766−
1775. (l) Sengupta, S.; Shi, X. ChemCatChem 2010, 2, 609−619.
(m) Nevado, C. Chimia 2010, 64, 247−251. (n) Corma, A.; Leyva-
Per
́
ez, A.; Sabater, M. J. Chem. Rev. 2011, 111, 1657−1712. (o) Aubert,
659
dx.doi.org/10.1021/om200946m | Organometallics 2012, 31, 644−661

Organometallics
Article
C.; Fensterbank, L.; Garcia, P.; Malacria, M.; Simonneau, A. Chem. Rev.
2011, 111, 1954−1993.
(11) (a) Yang, C.-G.; He, C. J. Am. Chem. Soc. 2005, 127, 6966−
6967. (b) Nguyen, R.-V.; Yao, X.; Li, C.-J. Org. Lett. 2006, 8, 2397−
2399.
(2) (a) Muzart, J. Tetrahedron 2008, 64, 5815−5849. (b) Hashmi, A.
S. K.; Buhrle, M. Aldrichimica Acta 2010, 43, 27−33. (c) Hashmi, A. S.
(12) (a) Reetz, M. T.; Sommer, K. Eur. J. Org. Chem. 2003, 3485−
3496. (b) Shi, Z.; He, C. J. Org. Chem. 2004, 69, 3669−3671.
(c) Chary, B. C.; Kim, S. J. Org. Chem. 2010, 75, 7928−7931.
(13) Examples for the synthesis of gold acetylides by using
triethylamine as a base: (a) Lu, X.-X.; Li, C.-K.; Chung-Chin
Cheng, E.; Zhu, N.; Wing-Wah Yam, V. Inorg. Chem. 2004, 43, 2225−
2227. (b) Hu, Q. Y.; Lu, W. X.; Tang, H. D.; Sung, H. H. Y.; Wen, T.
B.; Williams, I. D.; Wong, G. K. L.; Lin, Z.; Jia, G. Organometallics
2005, 24, 3966−3973. (c) Tang, H.-S.; Zhu, N.; Wing-Wah Yam, V.
Organometallics 2007, 26, 22−25. (d) He, X.; Zhu, N.; Wing-Wah
Yam, V. Organometallics 2009, 28, 3621−3624. (e) Fora new method
using NHC−Au−OH complexes, see: Gaillard, S.; Slawin, A. M. Z.;
Nolan, S. P. Chem. Commun. 2010, 46, 2742−2744.
(14) For hints of the participation of gold acetylides in gold catalysis,
see also: (a) Cheong, P. H.-Y.; Morganelli, P.; Luzung, M. R.; Houk,
K. N.; Toste, F. D. J. Am. Chem. Soc. 2008, 130, 4517−4526. (b)
Reference 4q. Gold acetylides as “Grignard-Type” reactants: (c) Wei,
C.; Li, C.-J. J. Am. Chem. Soc. 2003, 125, 9584−9585. (d) Yao, X.; Li,
C.-J. Org. Lett. 2006, 8, 1953−1955. (e) Lo, V. K.-Y.; Liu, Y.; Wong,
M.-K.; Che, C.-M. Org. Lett. 2006, 8, 1529−1532. (f) Huang, B.; Yao,
X.; Li, C.-J. Adv. Synth. Catal. 2006, 348, 1528−1532. (g) Yan, B.; Liu,
Y. Org. Lett. 2007, 9, 4323−4326. (h) Cheng, M.; Zhang, Q.; Hu,
X.-Y.; Li, B.-G.; Ji, J.-X.; Chan, A. S. C. Adv. Synth. Catal. 2011, 353,
1274−1278.
̈
K.; Rudolph, M. Chem. Commun. 2011, 47, 6536−6544.
(3) (a) Ma, S.; Yu, S.; Gu, Z. Angew. Chem. 2006, 118, 206−209;
(b) Angew. Chem., Int. Ed. 2006, 45, 200−203. (c) Zhang, L.; Sun, J.;
́
Kozmin, S. A. Adv. Synth. Catal. 2006, 348, 2271−2296. (d) Jimenez-
Nunez, E.; Echavarren, A. M. Chem. Rev. 2008, 108, 3326−3350.
́
(e) Echavarren, A. M.; Jimen
́
́
ez-Nunez, E. Top. Catal. 2010, 53,
924−930.
(4) (a) Shibata, T.; Fujiwara, R.; Takano, D. Synlett 2005, 13, 2062−
2066. (b) Odedra, A.; Wu, C.-J.; Pratap, T. B.; Huang, C.-W.; Ran, Y.-
F.; Liu, R.-S. J. Am. Chem. Soc. 2005, 127, 3406−3412. (c) Oh, C. H.;
Kim, A.; Park, W.; Park, D. I.; Kim, N. Synlett 2006, 17, 2781−2784.
(d) Zhao, J.; Hughes, C. O.; Toste, F. D. J. Am. Chem. Soc. 2006, 128,
7436−7437. (e) Lian, J.-J.; Chen, P.-C.; Lin, Y.-P.; Ting, H.-C.; Liu,
R.-S. J. Am. Chem. Soc. 2006, 128, 11372−11373. (f) Oh, C. H.; Kim,
A. New J. Chem. 2007, 31, 1719−1721. (g) Lian, J.-J.; Liu, R.-S. Chem.
Commun. 2007, 1337−1339. (h) Das, A.; Chang, H.-K.; Yang, C.-H.;
Liu, R.-S. Org. Lett. 2008, 10, 4061−4064. (i) Yeom, H.-S.; Lee, J.-E.;
Shin, S. Angew. Chem. 2008, 120, 7148−7151; (j) Angew. Chem., Int.
Ed. 2008, 47, 7040−7043. (k) Zhang, C.; Cui, D.-M.; Yao, L.-Y.;
Wang, B.-S.; Hu, Y.-Z.; Hayashi, T. J. Org. Chem. 2008, 73, 7811−
7813. (l) Tang, J.-M.; Liu, T.-A.; Liu, R.-S. J. Org. Chem. 2008, 73,
8479−8483. (m) Oh, C. H.; Kim, A. Synlett 2008, 5, 777−781.
(n) Sohel, S.; Md., A.; Liu, R.-S. Chem. Soc. Rev. 2009, 38, 2269−2281.
(o) Sperger, C.; Fiksdahl, A. Org. Lett. 2009, 11, 2449−2452.
(p) Meng, J.; Zhao, Y.-L.; Ren, C.-Q.; Li, Y.; Li, Z.; Liu, Q. Chem. Eur.
J. 2009, 15, 1830−1834. (q) Odabachian, Y.; Le Goff, X. F.; Gagosz, F.
Chem. Eur. J. 2009, 15, 8966−8970. (r) Oh, C. H.; Karmakar, S. J. Org.
Chem. 2009, 74, 370−374. (s) Kramer, S.; Madsen, J. L. H.;
(15) (a) Ohira, S. Synth. Commun. 1989, 19, 561−564. (b) Pietruszka,
J.; Witt, A. Synthesis 2006, 24, 4266−4268.
(16) Related examples, see refs 12a and 12b.
(17) Shibata, T.; Ueno, Y.; Kanda, K. Synlett 2006, 3, 411−414.
(18) For evidence of a catalyst transfer, see also ref 14a.
(19) (a) Nesmeyanov, A. N.; Perevalova, E. G.; Grandberg, K. I.;
Lemenovskii, D. A.; Baukova, T. V.; Afanassova, O. B. J. Organomet.
Chem. 1974, 65, 131−144. (b) Andrianov, V. G.; Struchkov, Y. T.;
Rossinskaya, E. R. J. Struct. Chem. 1974, 65−72. (c) Schmidbaur, H.;
Rottlander, M.; Skrydstrup, T. Org. Lett. 2010, 12, 2758−2761.
̈
(t) Sperger, C. A.; Fiksdahl, A. J. Org. Chem. 2010, 75, 4542−4553.
(u) Cui, D.-M.; Ke, Y.-N.; Zhuang, D.-W.; Wang, Q.; Zhang, C.
Tetrahedron Lett. 2010, 51, 980−982. (v) Hashmi, A. S. K.; Buhrle, M.;
̈
Inoguchi, Y. Chem. Ber. 1980, 113, 1646−1653. (d) Uson
A.; Brun, P. J. Organomet. Chem. 1980, 197, 369−373. (e) Uson
Laguna, A.; Fernandez, E. J.; Mendia, A.; Jones, P. G. J. Organomet.
Chem. 1988, 350, 129−138. (f) Uson, R.; Laguna, A.; Fernandez, E. J.;
́
, R.; Laguna,
Wolfle, M.; Rudolph, M.; Wieteck, M.; Rominger, F.; Frey, W. Chem.
̈
́
, R.;
Eur. J. 2010, 16, 9846−9854. (w) Hirano, K.; Inaba, Y.; Watanabe, T.;
Oishi, S.; Fujii, N.; Ohno, H. Adv. Synth. Catal. 2010, 352, 368−372.
(x) Sperger, C.; Strand, L. H. S.; Fiksdahl, A. Tetrahedron 2010, 66,
7749−7754. (y) Chen, Y.; Li, G.; Liu, Y. Adv. Synth. Catal. 2011, 353,
392−400. (z) Nun, P.; Dupuy, S.; Gaillard, S.; Poater, A.; Cavallo, L.;
Nolan, S. P. Catal. Sci. Technol. 2011, 1, 58−61. (aa) Hashmi, A. S. K.;
́
́
́
Ruiz-Romero, M. E.; Jones, P. G.; Lautner, J. J. Chem. Soc., Dalton
Trans. 1989, 2127−2131. (g) Baukova, T. V.; Kuz’mina, L. G.;
Oleinikova, N. A.; Lemenovskii, D. A.; Blumenfel’d, A. L. J. Organomet.
Chem. 1997, 530, 27−38. (h) Schuster, O.; Schier, A.; Schmidbaur, H.
Organometallics 2003, 22, 4079−4083. (i) Porter, K. A.; Schier, A.;
Schmidbaur, H. Organometallics 2003, 22, 4922−4927. (j) Osawa, M.;
Hoshino, M.; Hashizume, D. J. Chem. Soc., Dalton Trans. 2008, 2248−
Haffner, T.; Rudolph, M.; Rominger, F. Chem. Eur. J. 2011, 17, 8195−
̈
8201. (ab) Hirano, K.; Inaba, Y.; Takahashi, N.; Shimano, M.; Oishi,
S.; Fujii, N.; Ohno, H. J. Org. Chem. 2011, 76, 1212−1227. (ac) Zhang,
D.-H.; Yao, L.-F.; Wie, Y.; Shi, M. Angew. Chem. 2011, 123, 2631−
2635; (ad) Angew. Chem., Int. Ed. 2011, 50, 2583−2587.
(5) For the use of intramolecular nucleophiles, see refs 4w and 4ab.
(6) All details are available in extra tables in the Supporting
Information.
2252. (k) Seidel, G.; Lehmann, C. W.; Furstner, A. Angew. Chem.
̈
2010, 122, 8644−8648; (l) Angew. Chem., Int. Ed. 2010, 49, 8466−
8470. (m) Hooper, T. N.; Green, M.; Russell, C. A. Chem. Commun.
2010, 46, 2313−2315. (n) Blanco, M. C.; Cam
Jones, P. G.; Laguna, A.; Lopez-de-Luzuriaga, J. M.; Olmos, M. E.;
Villacampa, M. D. Organometallics 2011. DOI: 10.1021/om200397t.
(20) (a) Weber, D.; Tarselli, M. A.; Gagne, M. R. Angew. Chem. 2009,
121, 5843−5846; (b) Angew. Chem., Int. Ed. 2009, 48, 5733−5736.
(c) Weber, D.; Gagne, M. R. Org. Lett. 2009, 11, 4962−4965.
́
ara, J.; Gimeno, M. C.;
(7) For related studies with ruthenium and platinum catalysts at high
temperature, see: (a) Lo, C.-Y.; Kumar, M. P.; Chang, H.-K.; Lush,
S.-F.; Liu, R.-S. J. Org. Chem. 2005, 70, 10482−10487. (b) Taduri, B. P.;
Ran, Y.-F.; Huang, C.-W.; Liu, R.-S. Org. Lett. 2006, 8, 883−886.
(c) Taduri, B. P.; Odedra, A.; Lung, C.-Y.; Liu, R.-S. Synthesis 2007, 13,
2050−2054 . See also refs 4b and 4n.
́
́
́
(21) For isolation of stable intermediates using base to avoid
protodemetalation, see also: (a) Hashmi, A. S. K.; Schuster, A. M.;
Rominger, F. Angew. Chem. 2009, 121, 8396−8398; (b) Angew. Chem.,
Int. Ed. 2009, 48, 8247−8249. (c) Hashmi, A. S. K. Angew. Chem.
2010, 122, 5360−5369; (d) Angew. Chem., Int. Ed. 2010, 49, 5232−
5241. (e) Hashmi, A. S. K.; Ramamurthi, T. D.; Rominger, F. Adv.
Synth. Catal. 2010, 352, 971−975. (f) LaLonde, R. L.; Brenzovich, W.
E. Jr.; Benitez, D.; Tkatchouk, E.; Kelley, K.; Goddard, W. A. III;
Toste, F. D. Chem. Sci. 2010, 1, 226−233.
(8) Nieto-Oberhuber, C.; Paz MuC
́ ́
́
̧oz, M.; Lopez, S.; Jimenez-Nunez,
̃
E.; Nevado, C.; Herrero-Gomez, E.; Raducan, M.; Echavarren, A. M.
́
Chem. Eur. J. 2006, 12, 1677−1693.
(9) Nolan, S. P. Acc. Chem. Res. 2011, 44, 91−100.
(10) (a) Hashmi, A. S. K.; Loos, A.; Littmann, A.; Braun, I.; Knight,
J.; Doherty, S.; Rominger, F. Adv. Synth. Catal. 2009, 351, 576−582.
(b) Doherty, S.; Knight, J. G.; Hashmi, A. S. K.; Smyth, C. H.; Ward,
N. A. B.; Robson, K. J.; Tweedley, S.; Harrington, R. W.; Clegg, W.
Organometallics 2010, 29, 4139−4147. (c) Hashmi, A. S. K.; Loos, A.;
Doherty, S.; Knight, J. G.; Robson, K. J.; Rominger, F. Adv. Synth.
Catal. 2011, 353, 749−759.
(22) Selected examples about “aurophilicity”: (a) Schmidbaur, H.
Gold Bull. 1990, 23, 11−21. (b) Schmidbaur, H. Chem. Soc. Rev. 1995,
391−400.
660
dx.doi.org/10.1021/om200946m | Organometallics 2012, 31, 644−661

Organometallics
Article
(23) Au−Au distances for gem-diaurated compounds found in the
literature (refs 19a−c, 19e, 19i−k): 272.7−284.6 pm. Average value
(7 structures): 277.8 pm.
(24) For studies on thermal effects on the equilibrium of diaurated
compounds, see also ref 19j.
(25) Selected examples for strained allenes as intermediates:
(a) Burrell, R. C.; Daoust, K. J.; Bradley, A. Z.; DiRico, K. J.;
Johnson, R. P. J. Am. Chem. Soc. 1996, 118, 4218−4219. (b) Rodríguez,
D.; Navarro, A.; Castedo, L.; Domínguez, D.; Saa,
2, 1497−1500. (c) Rodríguez, D.; Martínez-Esperon
Vazquez, A.; Castedo, L.; Domínguez, D.; Saa, C. J. Org. Chem. 2004,
́
C. Org. Lett. 2000,
́
, M. F.; Navarro-
́
́
69, 3842−3848. (d) Dunetz, J. R.; Danheiser, R. L. J. Am. Chem. Soc.
2005, 127, 5776−5777. (e) Hayes, M. E.; Shinokubo, H.; Danheiser,
R. L. Org. Lett. 2005, 7, 3917−3920. (f) Pena, D.; Iglesias, B.;
̃
́ ́
Quintana, I.; Perez, D.; Guitian, E.; Castedo, L. Pure Appl. Chem. 2006,
78, 451−455. Selected examples for “Carbene or Carbocation?”:
(g) Furstner, A.; Morency, L. Angew. Chem. 2008, 120, 5108−5111;
̈
(h) Angew. Chem., Int. Ed. 2008, 47, 5030−5033. (i) Hashmi, A. S. K.
Angew. Chem. 2008, 120, 6856−6858; (j) Angew. Chem., Int. Ed. 2008,
47, 6754−6756. (k) Benitez, D.; Shapiro, N. D.; Tkatchouk, E.; Wang,
Y.; Goddard, W. A. III; Toste, F. D. Nat. Chem. 2009, 1, 482−486.
(26) Such gold(I) vinylidenes have been suggested before; see:
(a) Mamane, V; Hannen, P.; Furstner, A. Chem. Eur. J. 2004, 10,
̈
4556−4575. (b) Soriano, E.; Contelles, M. Organometallics 2006, 25,
4542−4553. (c) Seregin, I. V.; Gevorgyan, V. J. Am. Chem. Soc. 2006,
128, 12050−12051. (d) Lavallo, V.; Frey, G. D.; Kouzar, S.;
Donnadieu, B.; Bertrand, G. Proc. Natl. Acad. Sci. U.S.A. 2007, 104,
13569−13573. (e) Seregin, I. V.; Schammel, A. W.; Gevorgyan, V.
Tetrahedron 2008, 64, 6876−6883. (f) Xia, Y.; Dudnik, A. S.; Li, Y.;
Gevorgyan, V. Org. Lett. 2010, 12, 5538−5541.
(27) For other examples of such [1,2]-shifts, see: (a) Cabello, N.;
Jimen
J. Org. Chem. 2007, 4217−4223. (b) Nieto-Oberhuber, C.; Per
Galan, P.; Herrero-Gomez, E.; Lauterbach, T.; Rodríguez, C.; Lop
S.; Bour, C.; Rosellon, A.; Cardenas, D. J.; Echavarren, A. M. J. Am.
́
ez-Nunez, E.; Bunuel, E.; Car
́
denas, D. J.; Echavarren, A. M. Eur.
ez-
ez,
́
̃
̃
́
́
́
́
́
́
Chem. Soc. 2008, 130, 269−279. (c) Dudnik, A. S.; Xia, Y.; Li, Y.;
Gevorgyan, V. J. Am. Chem. Soc. 2010, 132, 7645−7655.
(28) Hashmi, A. S. K.; Ramamurthi, T. D.; Rominger, F. Adv. Synth.
Catal. 2010, 352, 971−975.
661
dx.doi.org/10.1021/om200946m | Organometallics 2012, 31, 644−661