Synthesis of Carbazoles, Indoles, and Dibenzofurans
TABLE 1. Optimization of the Carbazole Synthesis
SCHEME 1. Synthesis of Substituted Carbazoles via Vinylic
to Aryl Palladium Migration
% yield
entry
R
base
ligand
additive time (h)
(1a:1b)a
1
2
3
4
5
6
7
8
9
H
2NaOAc 10% PPh3
2Na2CO3 10% PPh3
24
24
24
24
12
6
6
12
12
trace
0
0
20 (10:1)
60b (10:1)
73 (10:1)
71b (10:1)
trace
H
H
H
H
H
H
2NEt3
10% PPh3
2NaOAc 10% PPh3 1TBAC
2CsPiv
2CsPiv
2CsPiv
10% PPh3
5% dppm
5% dppm
5% dppm
5% dppm
1TBAC
process. The arylpalladium intermediate generated subsequently
undergoes intramolecular arylation to afford the carbazole
products. Herein, we wish to report a complete account on this
nitrogen-directed palladium migration, an extension of this
methodology to the synthesis of biologically interesting diben-
zofurans,7 and the synthesis of indoles8 in which the arylpal-
ladium intermediate is trapped by an intramolecular Mizoroki-
Heck reaction. Furthermore, substrates labeled with deuterium
have also been prepared and employed in this process to explore
the mechanistic details of this rearrangement.
Me 2CsPiv
Ph 2CsPiv
0
a All reactions were conducted on a 0.25 mmol scale at 100 °C in 4 mL
of DMF, and the ratio of aryl iodide to alkyne was 1:1 (sealed vial, under
Ar). The ratio of 1a to 1b, as determined by 1H NMR spectroscopy, is
reported in parentheses. b GC yield.
mol % of Pd(OAc)2, 5 mol % of bis(diphenylphosphino)methane
(dppm), and 2 equiv of CsO2CCMe3 (CsPiv) in DMF at 100
°C.
2. Synthesis of Carbazoles by Nitrogen-Directed Vinylic
to Aryl Palladium Migration. We next examined the reaction
using various internal alkynes to determine the scope and
limitations of this process. The results are shown in Table 2.
Theoretically, when 4,4-dimethyl-2-pentyne was allowed to react
with N-phenyl-3-iodoaniline, the previously reported consecutive
vinylic to aryl to allylic palladium migration could also occur,
and a π-allylpalladium complex I would be generated as shown
in Scheme 2.5 However, as determined by GC-MS analysis, only
the expected carbazole product was found, and a 44% yield of
one regioisomer 2a was isolated (Table 2, entry 2). When
4-octyne was employed as the starting material, the reaction
was very messy, and only a 35% yield of the carbazole 3 was
obtained (entry 3). In this system, the vinylpalladium intermedi-
ate generated from carbopalladation may undergo â-H elimina-
tion to afford an allene, which may account for the low yield
of carbazole in this reaction. To avoid loss of the volatile alkyne
(the boiling point of 4,4-dimethylpentyne is only 70 °C) or
possible â-H elimination, 2,2-dimethyl-3-octyne was prepared
and allowed to react with our diarylamine. However, only a
48% yield of the desired product 4a was isolated (entry 4).
1-Phenyl-1-propyne afforded a 65% yield of two regioisomers
5a and 5b in a 12:1 ratio (entry 5). In the case of dipheny-
lacetylene, the arylpalladium intermediate formed by vinylic to
arylpalladium migration might be expected to undergo direct
arylation of one of the phenyl groups of the diphenylacetylene,
affording phenylamino-substituted benzylidenefluorenes II or
III (Scheme 2).2 Surprisingly, a 69% yield of a single isomer
6 was isolated from this reaction (entry 6), and no benzylidene-
fluorene products were observed. We have also examined the
reaction of this aniline with a couple of other aryl acetylenes
bearing diverse functionalities on the arene. When 1-(4-
nitrophenyl)-1-butyne was employed in our carbazole synthesis,
a very messy reaction was observed and none of the desired
product was evident by GC-MS analysis (entry 7). However,
when a moderate electron-withdrawing ester group (CO2Et) was
present on the phenyl ring of the alkyne, a 71% yield of a single
regioisomer 7a was isolated by flash chromatography (entry 8).
Results and Discussion
1. Optimization of Reaction Conditions. In our initial work
on this carbazole synthesis, we treated N-phenyl-3-iodoaniline
and 1 equiv of 1-phenyl-1-butyne with 5 mol % of Pd(OAc)2,
10 mol % of PPh3, and 2 equiv of NaOAc in N,N-dimethyl-
formamide (DMF) at 100 °C for 24 h (Table 1, entry 1).
Unfortunately, only a trace amount of the desired carbazole
product 1a was observed by GC-MS analysis. This reaction was
subsequently carried out using both an inorganic base Na2CO3
(entry 2) and an organic base NEt3 (entry 3), but none of the
desired carbazole product was observed. When 1 equiv of n-Bu4-
NCl (TBAC) was added to the NaOAc reaction, a 20% yield
of a 10:1 ratio of isomeric carbazoles 1a and 1b was obtained
(entry 4). We next conducted the model reaction in the presence
of 2 equiv of CsO2CCMe3 (CsPiv) because of its superior
solubility in DMF. To our delight, a 60% yield of the desired
products was obtained (entry 5). By simply replacing PPh3 with
a bidentate ligand bis(diphenylphosphino)methane (dppm), we
isolated a 73% yield of the two regioisomers by flash chroma-
tography (entry 6). We then repeated the same reaction in the
presence of 1 equiv of TBAC, but it appears that the presence
of a chloride source is unnecessary for this transformation (entry
7). The lack of a substituent on the aniline nitrogen is also crucial
because the corresponding amines with Me and Ph substituents
produced none of the anticipated carbazoles (entries 8 and 9).
In conclusion, the “optimal” reaction conditions for this
nitrogen-directed vinylic to aryl palladium migration utilize 5
(7) (a) Kikumoto, R.; Tamao, Y.; Ohkubo, K.; Tezuka, T.; Tonomura,
S.; Okamoto, S.; Hijikata, A. J. Med. Chem. 1980, 23, 1293. (b) Robertson,
D. W.; Lacefield, W. B.; Bloomquist, W.; Pfeifer, W.; Simon, R. L.; Cohen,
M. L. J. Med. Chem. 1992, 35, 310. (c) Oliveira, A. A. G.; Raposo, M. M.;
Oliveira-Campos, A. F.; Griffiths, J.; Machado, A. H. HelV. Chim. Acta
2003, 86, 2900. (d) Crich, D.; Grant, D. J. Org. Chem. 2005, 70, 2384.
(8) For recent reviews, see: (a) Nussbaum, F. Angew. Chem., Int. Ed.
2003, 42, 3068. (b) Battistuzzi, G.; Cacchi, S.; Fabrizi, G. Eur. J. Org.
Chem. 2002, 16, 2671. (c) Black, D. S. C. Synlett 1993, 610. (d) Saxton, J.
E. Nat. Prod. Rep. 1994, 11, 493.
J. Org. Chem, Vol. 71, No. 14, 2006 5341