Angewandte
Chemie
Table 2: Direct transformation of diaryl alkynes 1 to amides 2.[a]
porting Information, Table S1, entry 8). However, consider-
ing the cost, we choose to use Ag2CO3 as co-catalyst. Control
experiments indicated that [PPh3AuCl] is essential for this
transformation (Table 1, entry 6). In contrast to the Au
catalyst, when the Ag catalyst was used alone this amide
formation did not occur. Besides, only a trace amount of
desired product was detected when the reaction was per-
formed in the absence of TFA (Table 1, entry 9). Reactions in
which either HOTf, HN(Tf)2, or HOAc were used as an acid
additive did not give good results (Table 1, entries 10–12). We
also investigated the use of PhCl and HOAc as the solvent.
However, this resulted in low yields (Table 1, entries 13–15).
The reactivity of other gold catalysts with different ligands
(XPhos, JohnPhos, IPr, IMes, (4-CF3C6H4)3P) were also
investigated. Unfortunately, these catalysts gave poor results
(entries 18–22). After extensive screening on other parame-
ters (see the Supporting Information, Table S1), the optimum
reaction conditions were determined as: alkynes (0.5 mmol,
1.0 equiv), [PPh3AuCl] (10 mol%), Ag2CO3 (10 mol%),
TMSN3 (2.0 equiv, added in two portions), H2O (2.0 equiv),
TFA (200 mL), DCE (2 mL), 608C.
Entry Alkynes 1
Amides 2
Yield
[%][b]
1
2
3
4
72
74
42
56
The scope of the substrates were investigated under the
optimized conditions (Table 2). A variety of 1,2-diarylethynes
were found to be compatible with this protocol. Substituted
N,2-diphenylacetamides were generated through the cleavage
5
67
35
43
6[c]
À
of Csp Csp2 bonds of diarylethynes. Reactions of diarylethynes
bearing electron-donating substituents (Me, OMe, and nBu)
on the aryl ring afforded the desired products 2b, 2c, and 2d,
respectively, in moderate yields (74%, 42%, and 56%,
respectively). When diarylethynes with both electron-rich
and electron-poor aromatic substituents were employed, the
azide anion preferred to add at the electron-rich side of the
alkynes, thus leading to a low regioselectivity (see the
Supporting Information). Significantly, in addition to diaryl-
ethynes, alkyl-substituted phenylethynes, such as but-1-yn-1-
ylbenzene (1 f), 1-methoxy-4-(oct-1-yn-1-yl)benzene (1g), 1-
(cyclohexylethynyl)-4-methoxybenzene (1h), and 1-methoxy-
4-(5-phenylpent-1-yn-1-yl)benzene (1i) also gave the desired
7
8[d]
9[d]
10
44
47
0
[a] Reaction conditions: 1 (0.5 mmol), [PPh3AuCl] (0.05 mmol), Ag2CO3
(0.05 mmol), TMSN3 (1.0 mmol), H2O (1.0 mmol), TFA (200 mL) in DCE
(2 mL), stirred at 608C. [b] Yield of the isolated product. [c] HOTf
(5.2 mmol) was used instead of TFA. [d] 0.5 mmol TMSN3 was used.
À
amide products through highly selective Csp2 Csp bond
cleavage (Table 2, entries 6–9). The possible amide products
À
from Csp3 Csp bond cleavage were not detected in these cases.
Unfortunately, electron-deficient internal alkyne ethyl 3-
phenylpropiolate (1j) did not react under these conditions.
When the reactions of terminal alkynes were tested under
the optimized conditions, the reaction of 3a in TFA produced
the desired amide 4a in 70% yield (Table 3, entry 1). Based
on this result, TFA was chosen as the solvent for the reaction
of terminal alkynes. A variety of phenylethynes were con-
verted into the desired products in modest to good yields (up
to 93%, Table 3). Various substituents, including electron-
donating groups, such as methyl (4c and 4k), methoxy (4b, 4j
and 4l), tert-butyl (4g), and phenyl (4h), and electron-
withdrawing groups, such as fluoride (4d and 4m), carboxyl
(4i), were tolerated at the meta, ortho, and para positions of
the aromatic ring. Halogens were tolerated on the aromatic
ring (4d, 4e, 4 f, 4j, and 4m), thus offering an opportunity for
further cross-coupling, and facilitating the expedient synthesis
of complex compounds. Notably, the carboxylic acid group of
N-phenylacetamide 4i was also tolerated in this transforma-
tion (Table 3, entry 9).
To elucidate the mechanism, possible intermediates were
investigated. As ketones were detected as a by-product in this
reaction, we assumed that one possible pathway is a tandem
process involving gold-catalyzed hydration of alkynes to
ketones[10] and a subsequent Schmidt reaction.[11] To verify
this possibility, the reaction of 1a in the absence of TMSN3
was studied, and 1,2-diphenylethanone (5) was indeed
obtained in 73% yield (Scheme 2a). However, 5 failed to
be converted into 2a under the standard reaction conditions
(Scheme 2b). Thus, ketones are probably not a key inter-
mediate in this transformation.
Gold catalysts are one of the most effective catalysts for
electrophilic activation of alkynes toward a variety of
nucleophiles.[12] In a simplified form, nucleophilic attack on
Angew. Chem. Int. Ed. 2013, 52, 7850 –7854
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
7851