the desired 2-substituted indoline 2a in moderate yield
(Scheme 3).6 Attempts to optimize the cyclization reac-
tion by varying the initiator and by using Bu3SnH failed
to afford a higher yield.
Scheme 1. Radical Cascade onto ortho-Azidoallylbenzenes
Scheme 3. Initial Cascade Reaction Attempt
a two-step process consisting of the addition of ethyl
iodoacetate 4 under iodine atom transfer conditions
followed by an intramolecular nucleophilic substitution
(Scheme 2).10 However, the radical process did not take
place since N-alkylation of the aniline by the highly
reactive iodoacetate rapidly took place affording a mix-
ture of N-alkylated products. This simple approach was
therefore abandoned in favor of a radical cascade ap-
proach starting from 2-allyl-1-azidobenzene.
The use of triethylborane was investigated next. In the
presence of oxygen, Et3B efficiently produces ethyl radi-
cals that initiate radical chain processes, even at low tem-
perature.14 Moreover, it can also act as a chain transfer
reagent since it undergoes homolytic substitution with
aminyl radicals15 (such an intermediate aminyl radical is
expected in our proposed cascade; see Scheme 1). Based
on the tin-free conditions developed in our group for the
radical carboazidation,16 we first tried to prepare indoline
2a from azide 1a and iodide 4 in the presence of an excess
of Et3B in EtOH/H2O. Under these conditions, only the
iodine atom transfer product 5a was obtained (Table 1,
entry 1). With an open to air system the radical cascade
was still unsuccessful in EtOH/H2O. However, indoline
2a was obtained in 63% yield along with a small amount
of the corresponding indole 6a when benzene and 4 equiv
of Et3B were used (entry 2). Increasing the amount of
borane did not give a higher yield, and using less than 4
equiv resulted in the incomplete conversion of 5a to 2a.
Indeed, 5a was always formed after addition of the first
equivalent of Et3B, and its complete conversion to the
cyclized product occurred only after further addition 3
equiv of Et3B. Working with a slight excess of aryl azide
(ratio 1.5:1) afforded 2a in 88% yield contaminated by
9% of 6a (entry 3). The influence of the ratios of starting
materials on the yield is explained by minor side reactions
involving the azide.17 A rapid screening of solvents
(Table 1, entries 4À7) showed a significant reduction of
the formation of 6a when the reaction was carried out in
Scheme 2. Preliminary Experiment onto 2-Allylaniline
Several substituted 2-allyl-1-azidobenzenes 1 were read-
ily prepared according to literature procedures from com-
mercially available anilines via either a sequence involving
N-alkylation, aza-Claisen rearrangement, and mild con-
version into aryl azides or a two-step procedure using Stille
coupling.11 The tandem radical addition/cyclization was
then investigated using the model reaction between 1a
and ethyl iodoacetate 4 as a radical precursor. The reac-
tion temperature was critical for this transformation since
ortho-azidoallylbenzene derivatives are known to under-
go intramolecular cycloaddition at temperatures above
80 °C.9,12 In an initial attempt, the use of hexabutylditin
as a chain transfer reagent was tested (Scheme 3).13 This
reaction afforded exclusively the product of iodine atom
transfer 5a without cyclization. Treatment of the iodide
5a with (Me3Si)3SiH and AIBN as an initiator afforded
(14) Darmency, V.; Renaud, P. Top. Curr. Chem. 2006, 263, 71–106.
(15) For selected examples, see: (a) Bertrand, M. P.; Feray, L.;
Nouguier, R.; Stella, L. Synlett 1998, 780–782. (b) Miyabe, H.; Ueda,
M.; Naito, T. J. Org. Chem. 2000, 65, 5043–5047. (c) Ueda, M.; Miyabe,
H.; Miyata, O.; Naito, T. Tetrahedron 2009, 65, 1321–1326. (d) Valpuesta,
(10) For Et3B-induced halogen atom transfer reactions, see: (a)
Yorimitsu, H.; Nakamura, T.; Shinokubo, H.; Oshima, K.; Omoto,
K.; Fujimoto, H. J. Am. Chem. Soc. 2000, 122, 11041–11047. (b)
Yorimitsu, H.; Shinokubo, H.; Matsubara, S.; Oshima, K.; Omoto,
K.; Fujimoto, H. J. Org. Chem. 2001, 66, 7776–7785.
(11) For detailledpreparation ofazides 1, see SupportingInformation.
(12) Smith, P. A. S.; Chou, S. P. J. Org. Chem. 1981, 46, 3970–3977.
(13) (a) Ollivier, C.; Renaud, P. J. Am. Chem. Soc. 2001, 123, 4717–
4727. (b) Panchaud, P.; Ollivier, C.; Renaud, P.; Zigmantas, S. J. Org.
Chem. 2004, 69, 2755–2759.
~
M.; Munoz, C.; Dı
´
az, A.; Torres, G.; Suau, R. Eur. J. Org. Chem. 2010,
1934–1942.
(16) Panchaud, P.; Renaud, P. J. Org. Chem. 2004, 69, 3205–3207.
(17) Trace amounts of 2-propylindoline resulting from the addition
of an ethyl radical to the double bond of 1a followed by cyclization were
observed. By running the reaction in the absence of iodide 4, 2-propy-
lindoline could be isolated in 5À10% yield. Products resulting from an
intramolecular dipolar cycloaddition of the azidoalkene 1a could not be
identified, but this process could also be responsible for a partial
decomposition of 1a.
B
Org. Lett., Vol. XX, No. XX, XXXX