Communications
explored methods in indole synthesis,[2f,11] to our knowledge
the analogous 5-exo-dig process and dehydrative cyclizations
of substituted propargylic alcohols of the type 1 have
remained sparse.[2c,f,h]
side product. More importantly, the contrasting catalytic
activities observed for the reactions mediated by AgOTf and
TfOH, respectively, provided evidence that the cationic AuI
complex is the active species (see Table S1 in the Supporting
Information).
Initially, we chose to focus our attentions on the cyclo-
isomerization/Friedel–Crafts alkylation of 1a by a variety of
Lewis and Brønsted acid catalysts to establish the reaction
conditions (Table 1 and Table S1 in the Supporting Informa-
tion). This initial study revealed that treatment of 1a with
5 mol% of AuCl and 5 mol% of AgOTf in toluene at reflux
gave the indenyl-fused indole 2a in 74% yield along with the
side product 6[12,13] in 25% yield (Table 1, entry 1). The indole
To define the scope of the present procedure, we next
turned our attention to the reactions of a variety of 1-phenyl-
1-(2-(tosylamino)phenyl)prop-2-yn-1-ol derivatives (Table 2).
These experiments showed that with AuCl/AgOTf as the
catalyst, starting alcohols 1b–k and 1m efficiently underwent
the tandem cycloisomerization/Friedel–Crafts alkylation pro-
cess and gave the corresponding products 2b–k and 2m in
good to excellent yields (Table 2, entries 1–10 and 12). The
structure of 2 f was also confirmed by X-ray crystal structure
analysis.[14] Under the standard reaction conditions, o,p-
dimethylaniline alcohol 1l was the only example that gave a
mixture of decomposition products based on TLC and
1H NMR analysis of the crude mixture (Table 2, entry 11).
Having established the reaction conditions that gave
indenyl-fused indoles, we next examined the scope of this new
methodology for the synthesis of 2,3-disubstituted indoles
(Table 3). As mentioned earlier, we anticipated such control
of product chemoselectivity could be achieved by varying the
nature of the R1 group on the vinyl gold moiety in species B
shown in Scheme 1. With this in mind, we first tested the
reaction of the terminal acetylenic alcohols 1n and 1o with
5 mol% of AuCl and AgOTf under the standard conditions
and found 3b and 3c could be obtained as the sole product in
77 and 86% yield, respectively (Table 3, entries 1 and 2).
Under similar conditions, repeating the reaction of 1n in the
1
product was confirmed by H NMR analysis (see the Sup-
porting Information for NMR analysis) and X-ray crystal
structure analysis of two closely related products.[14,15] Our
studies subsequently showed that the introduction of HMPA
resulted in an increase in product yield and formation of 3a as
a side product in 6% yield (Table 1, entry 2). By using CaSO4
to remove water from the reaction and in combination with
HMPA, formation of this latter byproduct could be sup-
pressed to give 2a as the sole product in 94% yield (Table 1,
entry 3). In contrast, an inspection of entries 4–10 in Table 1
shows that repeating the reaction with DMI or DMAP in
place of HMPA or in other solvents was markedly less
effective. Similarly, a survey of other Lewis and Brønsted acid
catalysts did not provide improved results (Table 1,
entries 11–24 and Table S1 in the Supporting Information).
In these latter reactions, a number of the conditions examined
also lead to either the recovery of the starting alcohol in yields
of up to 64% or gave the Z-indolin-3-ol 7[16] as an additional
Table 1: Optimization of the reaction conditions.[a]
Entry
Catalyst
Solvent
Yield [%][b]
Entry
Catalyst
Solvent
Yield [%][b]
2a
3a
6
7
2a
3a
6
7
1
AuCl/AgOTf
AuCl/AgOTf
AuCl/AgOTf
AuCl/AgOTf
AuCl/AgOTf
AuCl/AgOTf
AuCl/AgOTf
AuCl/AgOTf
AuCl/AgOTf
AuCl/AgOTf
AuCl/AgSbF6
AuCl/AgPF6
toluene
toluene
toluene
toluene
toluene
(CH2Cl)2
THF
74
87
94
20
–
–
6
–
–
–
–
–
–
–
25
–
–
67
87
40
–
80
55
–
–
–
–
–
–
–
–
–
–
–
–
–
13
14
15
16
17
18
19
20
21
22
23
24
AuCl3/AgOTf
[Au(PPh3)]Cl/AgOTf
[AuL1]Cl/AgOTf
[AuL2]Cl/AgOTf
[AuL3]Cl/AgOTf
[AuL4]Cl/AgOTf
AuCl
toluene
toluene
toluene
toluene
toluene
toluene
toluene
toluene
toluene
toluene
toluene
toluene
61
32[g]
38[g]
5
12
3
42
32
17[g]
14
12[g]
8
24
11
26
–
10
–
20
24
12
–
–
–
–
–
76
72
–
–
–
–
–
–
–
–
52
–
–
–
–
–
2[c]
3[d]
4[e]
5[f]
6
44
–
–
[e]
7
8
9
10
11
12
MeCN
AuCl3
1,4-dioxane
DMSO
toluene
toluene
25
–
[Au(PPh3)]Cl
[Au(PPh3)]NTf2
[AuL3MeCN]SbF6
[AuL4]NTf2
[e]
–
32
–
47
–
–
65
38
–
–
7
57
–
[a] All reactions were performed with 0.2 mmol of 1a in the presence of 5 mol% of catalyst at reflux for 17 hours. [b] Yield of isolated product.
[c] Reaction performed with 20 mol% of HMPA for 2 hours. [d] Reaction performed with 20 mol% of HMPA and CaSO4 (35 mg) for 2 hours.
[e] Reaction performed with 20 mol% of DMI and CaSO4 (35 mg) for 2 hours. [f] Reaction performed with 20 mol% of DMAP and CaSO4 (35 mg) for
2 hours. [g] Recovery of 1a in 18–64% yield. DMAP=4-dimethylaminopyridine, DMI=1,3-dimethyl-2-imidazolidinone, DMSO=dimethyl sulfoxide,
HMPA=hexamethylphosphoramide, L1 =tris(4-trifluoromethylphenyl)phosphine; L2 =1,3-bis(2,6-diisopropylphenyl)imidazol-2-ylidene; L3 =(1,1’-
biphenyl-2-yl)-di-tert-butylphosphine; L4 =2-dicyclohexyl(2’,4’,6’-triisopropylbiphenyl)phosphine, Tf=trifluoromethanesulfonyl, THF=tetrahydro-
furan.
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2010, 49, 4619 –4623