Iodine recycling via 1,3-migration in iodoindoles under metal catalysis

Article abstract of DOI:10.1039/c3cc44073g

3-Substituted (indol-2-yl)-α-allenols show divergent patterns of reactivity under metal catalysis. An unprecedented intramolecular 1,3-iodine migration is described.

Full text of DOI:10.1039/c3cc44073g

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
Iodine recycling via 1,3-migration in iodoindoles under
metal catalysis†
Cite this: Chem. Commun., 2013,
49, 7779
a
b
a
a
´
´
Benito Alcaide,* Pedro Almendros,* Jose M. Alonso, Sara Cembellın,
Received 31st May 2013,
Accepted 4th July 2013
c
a
d
´
´
Israel Fernandez, Teresa Martınez del Campo and M. Rosario Torres
DOI: 10.1039/c3cc44073g
www.rsc.org/chemcomm
3-Substituted (indol-2-yl)-a-allenols show divergent patterns of
reactivity under metal catalysis. An unprecedented intramolecular
1,3-iodine migration is described.
Scheme 1 Metal-catalysed reactions of 3-chloro/bromo (indol-2-yl)-a-allenols 1
and 2.
Despite the fact that aryl halides are used in many metal-catalysed
synthetic developments,1 low atom economy is a disadvantage
because the heteroatom is usually eliminated. A great challenge to
be accomplished is the conversion of readily available aryl halides
into halogenated products in which the heteroatom is not elimi-
nated but reintegrated into the reaction product.2 Recently, we
have successfully reported metal-catalysed carbocyclizations of
3-unsubstituted (indol-2-yl)-a-allenols for the direct preparation
of the relevant carbazole nucleus.3 We envisioned that a different
behaviour of indole-tethered allenols might be achieved if the
reactive C3-indole position was substituted with an activating
group. Herein, we report our findings starting from 3-halo- and
3-phenoxy-(indol-2-yl)-a-allenols 1–4.
To explore the possibility of a 1,3-heteroatom migration, chloro-
and bromoallenes 1 and 2 were initially chosen. Unfortunately,
2,5-dihydrofurans 5, formed through the usual palladium-
catalysed oxycyclization reaction,4 and dienes 6, formed via
gold-catalysed rearrangement, were the only products formed
(Scheme 1). The above experiments suggested that the halide
recycling is troublesome.
halogen recycling reaction possible.2 We first investigated the
reactions of allenols 3a–e bearing a C3-iodosubstituent at the indole
nucleus under our previously optimized gold-catalysed conditions.
Interestingly, a separable mixture of carbazoles 7a–e and iodo-
carbazoles 8a–e was obtained (Scheme 2). The iodocyclization of
allenol 3a afforded the corresponding 3-iodocarbazole 8a in 69% yield
and carbazole 7a in 7% yield. Diminished iodocarbazole/carbazole
selectivity of ethyl- and phenyl-substituted reactants 3b and 3c, were
observed with respect to methyl-substituted allenes 3a, 3d and 3e.
In addition to the expected carbazole 7e and iodocarbazole 8e,
1-hydroxy-3-iododihydrocarbazole 9e was also formed from the
chloroderivative 3e. It should be noted, that in our previous work
on metal-catalyzed carbocyclization of 3-unsubstituted (indol-2-yl)-
a-allenols, we were not able to obtain iodocarbazoles 8 by trapping
the postulated organometallic intermediate with halogenated
reagents.3a Considering the versatility of organic iodides in chemical
transformations, iodinated carbazoles 8 are potentially interesting
building blocks for further manipulation.5 The structure of
We thought that the use of an iodo-alkenyl rather than a Cl(Br)-
species to initiate the allene functionalization could make the
a
´
´
Grupo de Lactamas y Heterociclos Bioactivos, Departamento de Quımica Organica,
´
Unidad Asociada al CSIC, Facultad de Quımica, Universidad Complutense de
Madrid, 28040-Madrid, Spain. E-mail: alcaideb@quim.ucm.es;
Fax: +34 91-3944103
b
c
´
´
Instituto de Quımica Organica General, IQOG, CSIC, Juan de la Cierva 3,
28006-Madrid, Spain. E-mail: Palmendros@iqog.csic.es; Fax: +34 91-5644853
´
´
´
Departamento de Quımica Organica, Facultad de Quımica,
Universidad Complutense de Madrid, 28040-Madrid, Spain
d
´
´
CAI Difraccion de Rayos X, Facultad de Quımica,
Universidad Complutense de Madrid, 28040-Madrid, Spain
† Electronic supplementary information (ESI) available: Experimental procedures,
characterization data of new compounds, computational details, and copies of
NMR spectra. CCDC 926119. For ESI and crystallographic data in CIF or other
electronic format see DOI: 10.1039/c3cc44073g
Scheme 2 Synthesis of carbazoles 7, 3-iodocarbazoles 8, and 3-iododihydro-
carbazole 9e through carbocyclization/halogen recycling reactions of iodoallenols
3 under gold catalysis.
c
This journal is The Royal Society of Chemistry 2013
Chem. Commun., 2013, 49, 7779--7781 7779

View Article Online
Communication
ChemComm
Scheme 4 Synthesis of 1-oxygenated carbazoles 10 through carbocyclization/
hydrodephenoxylation reaction of phenoxyallenols 4 under gold catalysis.
Fig. 1 ORTEP drawing of 3-iodocarbazole 8d.
1-methoxycarbazole 10b in fair yield in the presence of Gagosz’s
catalyst.8 In contrast, phenyl-substituted allene 4c could not lead to
the formation of the corresponding 1-hydroxycarbazole, affording
instead the 2,5-dihydrofuran 5c. Hence, the hydroxy group in phenyl-
substituted 3-phenoxy-(indol-2-yl)-a-allenol 4c exclusively undergoes
the 5-endo oxycyclization reaction, instead of 6-endo carbocyclization.
A possible pathway9 for the gold-catalysed generation of
1-oxygenated carbazoles 10 is outlined in Scheme 5. Initially,
the formation of a complex 4-Au(L) through coordination of the
gold salt to the distal allenic double bond may be involved.
Species 4-Au(L) undergoes an intramolecular chemo- and regio-
selective 6-endo-trig carbocyclization reaction to produce the
auratetrahydrocarbazole 11. This nucleophilic attack from the
C3-indole site occurs as a result of the stability of the intermediate
iminium type cation 11. Next, a phenol elimination10 step occurs
in tricycle 11 through C3–OPh bond cleavage to generate the
dihydrocarbazolium 12. Aromatization by loss of proton generates
neutral species 13, which followed by protonolysis of the carbon–
gold bond afforded 1-oxygenated carbazoles 10 with concurrent
regeneration of the gold catalyst (Scheme 5).
3-iodocarbazole 8d was unambiguously confirmed with the
help of X-ray diffraction analysis on suitable crystals of this
compound (Fig. 1).6
In an attempt to improve the iodocarbocyclization efficiency under
related metal-catalysed conditions, we screened a different catalytic
system such as PdCl2(PPh3)2 in the reaction with 3-iodo-(indol-2-yl)-
buta-2,3-dienol 3a. While the Pd-catalysed reaction proceeded with an
optimal product distribution (a 100 : 0 ratio of the desired 1,3-iodine
migration product to the non-iodinated carbazole), the isolated yield
of 3-iodocarbazole 8a was poor (38%). Therefore, we moved to a
different catalytic system. Finally, compound 8a was prepared in
acceptable yield (61%) via the reaction of 3a in the presence of a
Pd–Cu bimetallic system in DMF. Nicely, indoles 3a, 3e, 3f and 3h,
bearing a methyl substituent on the allene moiety, exclusively
furnished 3-iodocarbazoles 8a, 8e, 8f and 8h (Scheme 3). Unfortu-
nately, attempts to use phenyl-substituted substrates 3c and 3g
proved to be unsuccessful for the construction of the corresponding
iodocarbazoles, possibly because of both unfavourable steric factors
as well as a direct interaction of the p-aromatic system with the
metal center from the catalyst. In addition to atom economy and
bond-forming efficiency, the above metal-catalysed cases, shown in
Schemes 2 and 3, may be considered as examples of the rare
recycling of halogen groups via 1,3-halogen migration.7
Next, the annulation of 3-phenoxy-(indol-2-yl)-a-allenols 4 was
examined (Scheme 4). To test the reactivity of allenes 4, we started
the initial investigation on the gold-catalysed reaction of allene 4a
under otherwise identical reaction conditions used for its iodo-
counterpart 3a. Interestingly, it was found that substrate 4a was
exclusively transformed into 1-hydroxycarbazole 10a (Scheme 4).
This interesting transformation can be explained through a gold-
catalysed allenic carbocyclization with concomitant hydro-
dephenoxylation (see below). Thus, it was found that the synthesis
of structurally interesting 1-oxygenated carbazoles could be con-
trolled by the C3-substituent on the indole ring in allenes of type
1–4. Next, 3-phenoxy-(indol-2-yl) allenes 4b and 4c were examined in
this reaction (Scheme 4). Allene 4b was successfully converted to
Density functional theory (DFT) calculations have been
carried out at the PCM-M06/def2-SVP//B3LYP/def2-SVP level11
to gain more insight into the reaction mechanism of the above
discussed transition metal-catalysed carbocyclization/halogen
recycling reactions of iodoallenols 3. Thus, the corresponding
computed reaction profile of the reaction of allenol 3a and the
+
model catalyst AuPMe3 is shown in Fig. 2, which shows the
respective free energies, DG298, in dichloroethane solution.
The process begins with the exergonic coordination of the
catalyst to the distal allenic double bond of 3a to form intermediate
INT1 (DGR,298 = ꢀ11.9 kcal molꢀ1). Then, the nucleophilic attack of
Scheme 3 Synthesis of 3-iodocarbazoles 8 through carbocyclization/halogen
recycling reactions of iodoallenols 3 under palladium catalysis.
Scheme 5 Mechanistic explanation for the Au(I)-catalysed synthesis of 1-oxy-
genated carbazoles 10 from phenoxyallenols 4.
c
7780 Chem. Commun., 2013, 49, 7779--7781
This journal is The Royal Society of Chemistry 2013

View Article Online
ChemComm
Communication
Fig. 3 Comparison of the migratory aptitude of halogen atoms in the proposed
1,3-shift. Relative free energies are given in kcal molꢀ1. All data have been
computed at the PCM(dichloroethane)-M06/def2-SVP//B3LYP/def2-SVP level.
involving chlorine and bromine atoms are much higher than the
barrier associated with the migration of iodine (DGa298 = 29.3 and
23.8 kcal molꢀ1 for Cl and Br, respectively). Therefore, our calcula-
tions suggest that the migratory aptitude of halogen atoms in this
transition metal-mediated process follows the order I c Br > Cl,
which is in nice agreement with the experimental findings.14
In conclusion, in salient contrast to the reaction of 3-phenoxy-
(indol-2-yl) allenes, which were transformed into 1-oxygenated
carbazoles, 3-iodo-(indol-2-yl) allenes afforded 3-iodocarbazoles
through rare recycling of halogen groups via 1,3-halogen migration.
Besides, a computational study suggested the intermediacy of an
iodonium cation species formed through an unprecedented intra-
molecular iodine cation addition to a metal-activated double bond.
Support for this work from MINECO [Projects CTQ2012-33664-
C02-01, CTQ2012-33664-C02-02, CTQ2010-20714-C02-01, and
Consolider-Ingenio 2010 (CSD2007-00006)], and CAM (Projects
S2009/PPQ-1752 and S2009/PPQ-1634) is gratefully acknowledged.
S. C. thanks MEC for a predoctoral grant. J. M. A. thanks
Fig. 2 Computed reaction profile (PCM(dichloroethane)-M06/def2-SVP//B3LYP/
def2-SVP level) for the reaction between 3a and AuPMe3+. Relative free energies
are given in kcal molꢀ1 and bond distances in the transition states in angstroms.
the C3-indole position on the gold(I)-activated double-bond delivers
auratetrahydrocarbazole INT2. This carbocyclization reaction
occurs through the transition state TS1 with an activation barrier of
DGa298 = 14.0 kcal molꢀ1 in an exergonic transformation (DGR,298
=
ꢀ6.6 kcal molꢀ1), which is compatible with the process at room
temperature. Alternatively, it has been recently suggested that
species related to INT2 may be formed from spiranic species INT20
through a 1,2-migration reaction.12 However, our calculations
indicate that the initial formation of INT20 via TS10, a saddle point
associated with the C2-indole nucleophilic attack, is kinetically and
thermodynamically less favoured than the process involving TS1,
which makes the alternative pathway non-competitive. The origins
of this behaviour are found in the well-known activation of the
C3-carbon atom by the nitrogen atom of the indole.13 Once INT2 is
formed, it is transformed into the iodonium species INT3 through
´
Comunidad Autonoma de Madrid and Fondo Social Europeo
for a postdoctoral contract.
Notes and references
1 Chem. Soc. Rev., 2011, 40, themed issue 10, Cross coupling reactions
in organic synthesis.
2 For an overview, see: J. M. Schomaker and R. D. Grigg, Synlett, 2013,
401.
TS2 (an activation barrier of DGa
= 16.8 kcal molꢀ1) in an
´
3 (a) B. Alcaide, P. Almendros, J. M. Alonso, M. T. Quiros and
298
´
P. Gadzinski, Adv. Synth. Catal., 2011, 353, 1871; (b) B. Alcaide,
exergonic process (DGR,298 = ꢀ2.7 kcal molꢀ1). As shown in Fig. 2,
TS2 is associated with the 1,3-migration of the iodine atom to the
endocyclic double bond of the adjacent six-membered ring. This
step resembles that of the typical electrophilic halogen addition to
alkenes. Indeed, the computed positive NBO-charge at the iodine
atom in INT3 (q = +0.35e) clearly confirms the cyclic-iodonium
cation nature of this species. Therefore, this step can be viewed as
an unprecedented intramolecular iodine cation addition to a metal-
activated double bond. The next step of the transformation involves
the liberation of the metal catalyst through formation of the
´
P. Almendros, J. M. Alonso and I. Fernandez, Chem. Commun., 2012,
48, 6604.
4 S. Ma, Chem. Rev., 2005, 105, 2829.
5 (a) J. Li and A. C. Grimsdale, Chem. Soc. Rev., 2010, 39, 2399;
¨
(b) A. W. Schmidt, K. R. Reddy and H.-J. Knolker, Chem. Rev.,
2012, 112, 3193.
6 CCDC 926119.
7 (a) R. D. Grigg, R. Van Hoveln and J. M. Schomaker, J. Am. Chem.
Soc., 2012, 134, 16131; (b) S. G. Newman and M. Lautens, J. Am.
¨
Chem. Soc., 2011, 133, 1778; (c) P. Nosel, T. Lauterbach, M. Rudolph,
F. Rominger and A. S. K. Hashmi, Chem.–Eur. J., 2013, 19, 8634.
´
8 N. Mezailles, L. Ricard and F. Gagosz, Org. Lett., 2005, 7, 4133.
9 A. S. K. Hashmi, Angew. Chem., Int. Ed., 2010, 49, 5232.
corresponding iododihydrocarbazoles 9 from INT4. Subsequent 10 (a) A. S. K. Hashmi, W. Yang and F. Rominger, Angew. Chem., Int. Ed.,
¨
2011, 50, 5762; (b) A. S. K. Hashmi and M. Wolfle, Tetrahedron, 2009,
65, 9021.
11 See computational details in the ESI†.
aromatization by dehydration would produce the observed 3-iodo-
carbazoles 8. Although the isolation of tricycle 9e from the reaction
of 3e as outlined in Scheme 2 was fortuitous, the result argues 12 B. Cheng, G. Huang, L. Xu and Y. Xia, Org. Biomol. Chem., 2012, 10, 4417.
´
13 (a) J. Barluenga, E. Tudela, A. Ballesteros and M. Tomas, J. Am.
in favour of the suggested reaction mechanism, because an
observable intermediate of type 9 was formed.
Chem. Soc., 2009, 131, 2096; (b) T. Cao, J. Deitch, E. C. Linton and
M. C. Kozlowski, Angew. Chem., Int. Ed., 2012, 51, 2448.
Finally, we have also investigated why chlorine or bromine 14 A similar trend has been observed in 1,2-dyotropic reactions. See:
´
´
(a) I. Fernandez, F. M. Bickelhaupt and F. P. Cossıo, Chem. Eur. J.,
2012, 18, 12395. For a recent review on dyotropic reactions, see:
substituted allenols 1 and 2 do not undergo a similar 1,3-migration
to that found for iodoallenols 3. As clearly shown in Fig. 3, the
computed activation barriers associated with the 1,3-halogen shifts
´
´
(b) I. Fernandez, F. P. Cossıo and M. A. Sierra, Chem. Rev., 2009,
109, 6687.
c
This journal is The Royal Society of Chemistry 2013
Chem. Commun., 2013, 49, 7779--7781 7781