Mild ArI-catalyzed C(sp2)-H or C(sp3)-H functionalization/C-O formation: An intriguing catalyst-controlled selectivity switch

Article abstract of DOI:10.1002/anie.201407011

A tandem C(sp²)H and C(sp³)H functionalization/CO bond formation catalyzed by iodine(III) reagents generated in situ has been developed. The method shows wide scope under mild conditions and exhibits an unprecedented selectivity profile that can be switched depending on the catalyst employed.

Full text of DOI:10.1002/anie.201407011

.
Angewandte
Communications
DOI: 10.1002/anie.201407011
Cyclization
2
3
À
À
À
Mild ArI-Catalyzed C(sp ) H or C(sp ) H Functionalization/C O
Formation: An Intriguing Catalyst-Controlled Selectivity Switch**
Xueqiang Wang, Joan Gallardo-Donaire, and Ruben Martin*
2
3
Abstract: A tandem C(sp )ÀH and C(sp )ÀH functionaliza-
tion/CÀO bond formation catalyzed by iodine(III) reagents
generated in situ has been developed. The method shows wide
scope under mild conditions and exhibits an unprecedented
selectivity profile that can be switched depending on the
catalyst employed.
D
ramatic progress in the field of CÀH functionalization has
[1]
enabled the design of unconventional synthetic strategies.
Whereas a myriad of catalytic CÀC and CÀN bond-forming
reactions have been developed, CÀO bond formation has
Scheme 1. Synthesis of benzolactones by catalytic CÀH functionaliza-
[
2]
tion.
received much less attention, partly owing to the large
energy gap between MÀO HOMO and MÀC LUMO frontier
[
3]
[12]
orbitals. Thus, the development of an innovative, yet
suitable oxidants could be equally effective. Despite further
advances,
[
13]
practical, CÀH functionalization/CÀO bond-forming process
at the outset of our investigations, a CÀH
[
4,5]
has become a goal for synthetic chemists.
The preparation
functionalization assisted by carboxylic acids and ArI cata-
lysts en route to benzolactones did not have any literature
of benzolactones is an ideal target, since compounds such as
the lamellarins and cytosporones, among others, show attrac-
[
14]
precedents.
Herein, we describe a benign ArI-catalyzed
[6]
2
3
tive biological properties.
process for C(sp )/C(sp )ÀH functionalization and CÀO bond
formation (Scheme 1, paths b and c). The method is user-
friendly, operates with a wide substrate scope, occurs under
mild conditions, and can be carried out in air. Initial studies
showed an unprecedented and intriguing selectivity profile
that can be easily modulated by the nature of the catalyst
[
7]
[8]
We and others have reported the synthesis of benzo-
lactones through platinum- or palladium-catalyzed CÀH
functionalization/CÀO bond formation assisted by carboxylic
[
9]
acids (Scheme 1, path a). However, these reactions are often
I
air- or moisture-sensitive, and stoichiometric Ag , PhI(OR) ,
2
II
[15]
or Cu reagents are usually required. Recently, copper-
employed.
catalyzed oxidative protocols have shown to be powerful
We began our investigations by examining the reactivity
[
10]
[16]
synthetic alternatives to these reactions. As global demand
and prices for noble metals continue to rise, however,
chemists are being challenged to design metal-free processes.
of 1a. After some experimentation,
(20 mol%) in HFIP with ACOOH as the oxidant enabled
we found that I8
[17]
the synthesis of 2a in nearly quantitative yield and in air at
III
[18,19]
To this end, the use of well-defined I reagents has recently
room temperature.
This finding is particularly remark-
[
11]
gained considerable momentum. Although the generation
of substantial aryl iodide (ArI) residues constitutes a signifi-
cant barrier for the implementation of such protocols, early
studies by the research groups of Kita, Ochiai, and Togo
showed that catalytic amounts of ArI in the presence of
able and constitutes an additional bonus, practicality aside,
[
7,8,10]
when compared with related metal-catalyzed protocols.
Encouraged by these findings, we set out to explore the
preparative scope of our surprisingly facile I8-catalyzed
2
C(sp )ÀH functionalization/CÀO bond-forming reaction
(
Scheme 2). As shown, the outcome was largely insensitive
to changes in the electronic nature of the substrates. Thus,
acetals (product 2c), nitro groups (product 2e), esters
(products 2g, 2j, and 2l), and aryl fluorides (product 2m)
were well accommodated. Similarly, non-aromatic carboxylic
[
*] X. Wang, J. Gallardo-Donaire, Prof. R. Martin
Institute of Chemical Research of Catalonia (ICIQ)
Av. Paꢀsos Catalans 16, 43007 Tarragona (Spain)
E-mail: rmartinromo@iciq.es
3
Prof. R. Martin
acids (product 2i) or substrates prone to C(sp )ÀH function-
Catalan Institution for Research and Advanced Studies (ICREA)
Passeig Lluꢀs Companys, 23, 08010 Barcelona (Spain)
alization (product 2h) posed no problems. To put these results
into perspective, we observed little conversion, if any, of 1c,
[
**] We thank the ICIQ Foundation, the European Research Council
1
d, or substrates bearing electron-withdrawing groups, such
(
ERC-277883), and MINECO (CTQ2012-34054) for support. John-
as 1e, 1 f, 1j, and 1l, under previously developed copper-
son Matthey, Umicore, and Nippon Chemical Industrial are
acknowledged for a gift of metal and ligand sources. J.G.-D. thanks
MICINN for a FPI fellowship. Dr. A. Correa, A. Shafir, and Prof. K.
MuÇiz are gratefully acknowledged for insightful discussions.
[10]
catalyzed conditions (Scheme 2). Products 2k–n were iso-
lated as single regioisomers, and 2m was unambiguously
[20]
characterized by X-ray crystallographic analysis.
These
results are in contrast with those of copper-catalyzed proto-
cols, which invariably provide regioisomeric mixtures of these
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201407011.
1
1084
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 11084 –11087

Angewandte
Chemie
3
Scheme 3. C(sp )ÀH functionalization/CÀO formation. Reaction condi-
tions: as for Scheme 2, but with I6 (20 mol%), at 408C, open to air.
Yields given are for the isolated product and an average over two
independent experiments. [c] The reaction was carried out with PIFA
2
Scheme 2. C(sp )ÀH functionalization/CÀO formation. Reaction condi-
tions: 1 (0.20 mmol), I8 (20 mol%), AcOOH (2.20 equiv), HFIP
1 mL), room temperature, 12 h. Yields given are for the isolated
(
(2.5 equiv).
product and an average over two reactions. [a] The reaction was
carried out with I11 (20 mol%). [b] The reaction was carried out with
AcOOH (4.40 equiv) at 808C. [c] The reaction was carried out with
AcOOH (1.10 equiv). [d] The product was obtained as a single regio-
isomer. [e] The product was obtained as a 4:1 mixture of regioisomers
under the conditions described in Ref. [10a]. [f] The reaction was
carried out with PIFA (1.0 equiv). [g] The product was obtained as
a 6:1 mixture of regioisomers under the conditions described in
Ref. [10a]. HFIP=1,1,1,3,3,3-hexafluoro-2-propanol, PIFA=[bis(tri-
fluoroacetoxy)iodo]benzene.
The preparation of 4j–l indicates that five-membered rings
other than pyrrolidinone can be equally effective. Similarly,
pyridobenzoxazinones (4i) are also within reach. Interest-
ingly, 4l was obtained as a single diastereoisomer, the
configuration of which was confirmed by X-ray crystallog-
[24,25]
raphy.
The observed influence of the ArI backbone suggested
that site selectivity could be possible under appropriate
conditions. We found an unprecedented catalyst-controlled
selectivity switch when using 1o (Scheme 4). Before process
optimization, we found that the use of I11 resulted in the
formation of the expected product 2o in 37% yield together
with a minor regioisomer, which was initially assigned as 2o’’.
Careful NMR spectroscopic analysis, however, revealed that
[
10]
compounds. We believe these results nicely illustrate the
complementarity of I8- and copper-catalyzed oxidative pro-
cesses.
A closer look at the literature indicates that harsh
conditions are typically required to prepare benzolactones
3
by metal-catalyzed C(sp )ÀH functionalization with carbox-
[
7,8d]
ylic acids as directing groups.
Challenged by this finding,
we wondered whether the mild protocol in Scheme 2 could be
3
[21]
adapted to a C(sp )ÀH functionalization scenario.
We
identified 5H-pyrrolo[1,2-a][3,1]benzoxazinones (4) as ideal
targets to demonstrate the feasibility of such a concept, given
the potential of these frameworks as lifespan-altering com-
3
pounds and the ease of C(sp )ÀH functionalization a to the
[
22]
nitrogen atom (Scheme 3). Among the catalysts tested, we
found that I6 was particularly competent for our pur-
[
18,23]
poses.
The method tolerated a variety of substitution
patterns without significantly influencing the reaction out-
come (Scheme 3). Interestingly, carboxylic acids (product 4d)
and aryl halides were tolerated (products 4b,e–h), thus
providing ample opportunity for cross-coupling techniques.
Scheme 4. Catalyst-controlled selectivity switch for the transformation
of 1o–r. Bn=benzyl.
Angew. Chem. Int. Ed. 2014, 53, 11084 –11087
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 11085

.
Angewandte
Communications
our assignment was premature, and this minor regioisomer
turned out to be 2o’, a species not expected on the basis of
simple selectivity between the two ortho CÀH sites. Intrigu-
ingly, the use of structurally related I6 resulted in a selectivity
[26–28]
switch to give predominantly 2o’ in 70% yield.
This
result might suggest that an incipient positive charge is
developed on the electron-rich aromatic ring, thus triggering
[
29]
a [1,2]-aryl shift. The structures of both 2o and 2o’ were
Scheme 6. Catalyst-controlled selectivity switch for the transformation
of 1s.
unequivocally established by X-ray crystallographic analysis
[30]
(
2
Scheme 4, top). Importantly, the successful preparation of
p’, 2q’, and 2r’ demonstrates the generality of this intriguing
selectivity switch (Scheme 4, bottom).
alization event. Gratifyingly, we found that the transforma-
Next, we focused our attention on unraveling the origin of
the selectivity switch shown in Scheme 4 by studying the
reaction of 1o with I reagents derived from I6 and I11.
tion of 1s followed a distinctive pattern when I6 or I8 were
[
31,32]
used as catalysts (Scheme 6).
While one might have
III
2
anticipated that 1s would trigger a C(sp )ÀH functionaliza-
tion en route to 2s, this was not the case, and 4a or 4a’ were
obtained exclusively, as ultimately confirmed by X-ray
Interestingly, we found that both I6(OAc) and I11(OAc)2
2
provided preferentially 2o’ (Scheme 5, entries 1 and 5),
[
33,34]
crystallographic analysis.
Whether these observations
indicate a general trend for other substrate combinations or
there are other mechanistic implications is the subject of
ongoing studies in our laboratories.
2
3
In summary, we have developed a C(sp )ÀH or C(sp )ÀH
III
functionalization/CÀO bond-forming process catalyzed by I
reagents generated in situ. The reaction occurs under mild
conditions and with an intriguing selectivity switch depending
on the catalyst used. This air-insensitive method is a cheap
and practical alternative to related metal-catalyzed protocols.
Further investigations into related processes are currently
ongoing.
Received: July 8, 2014
Revised: July 28, 2014
Published online: August 25, 2014
Scheme 5. Unraveling of the observed selectivity profile for 1o.
Keywords: CÀH functionalization · domino reactions ·
.
homogeneous catalysis · metal-free reactions · site selectivity
whereby I6(OAc) provided a slightly better selectivity profile
2
(
entry 1). These findings were rather surprising, since the I6-
or I11-catalyzed event with AcOOH as the oxidant resulted in
much lower 2o/2o’ ratios (entries 4, 7 and Scheme 4). We
speculated that AcOOH might not be a mere spectator and
could have dual roles, both as an oxidant and as a modulator
at the I center. In line with this notion, the inclusion of
AcOOH significantly eroded the selectivity profile with
[1] a) J. Wencel-Delord, T. Drçge, F. Liu, F. Glorius, Chem. Soc. Rev.
2011, 40, 4740; b) L. McMurray, F. OꢀHara, M. Gaunt, Chem.
Soc. Rev. 2011, 40, 1885; c) X. Chen, K. M. Engle, D. H. Wang,
J. Q. Yu, Angew. Chem. 2009, 121, 5196; Angew. Chem. Int. Ed.
2009, 48, 5094.
III
[
2] A. K. Yudin, Catalyzed Carbon-Heteroatom Bond Formation,
Wiley-VCH, Weinheim, 2011, pp. 35 – 68.
I6(OAc)2 (entries 2 and 3) or even caused a selectivity
[3] A. N. Vedernikov, Top. Organomet. Chem. 2010, 31, 101.
2
switch with I11(OAc) (entry 6). We believe these results
[4] For selected metal-catalyzed C(sp )ÀH/CÀO formation, see:
2
a) A. K. Cook, M. H. Emmert, M. S. Sanford, Org. Lett. 2013, 15,
tacitly suggest that the selectivity pattern is dictated by both
the nature of the employed aryl iodide and the concentration
of AcOOH. The role of the latter was nicely illustrated by
careful analysis of the 2o’/2o ratio as a function of time when
5428; b) C. L. Sun, J. Liu, Y. Wang, X. Zhou, B. J. Li, Z. J. Shi,
Synlett 2011, 883; c) X. Wang, Y. Lu, H. D. Sai, J. Q. Yu, J. Am.
Chem. Soc. 2010, 132, 12203.
3
[
5] For selected C(sp )ÀH/CÀO formation, see: a) K. J. Stowers, A.
[
16]
using catalytic amounts of I6,
concentration of AcOOH had a profound influence on the
o’/2o ratio. A similar behavior was observed for PhI(1o)2,
with which the presence of AcOOH had a deleterious effect
clearly showing that the
Kubota, M. S. Sanford, Chem. Sci. 2012, 3, 3192; b) S. Zhang, G.
He, Y. Zhao, K. Wright, W. A. Nack, G. Chen, J. Am. Chem. Soc.
2012, 134, 7313; c) M. A. Bigi, S. A. Reed, M. C. White, J. Am.
Chem. Soc. 2012, 134, 9721.
2
[16]
[6] K. Yoganathan, C. Rossant, Y. Huang, M. S. Butler, A. D. Buss,
on selectivity.
Overall, we believe that the results in
J. Nat. Prod. 2003, 66, 1116.
Scheme 5 reveal an intimate interplay between the aryl
[
[
7] P. Novꢁk, A. Correa, J. Gallardo-Donaire, R. Martin, Angew.
Chem. 2011, 123, 12444; Angew. Chem. Int. Ed. 2011, 50, 12236.
8] a) X. F. Cheng, Y. Li, Y. M. Su, F. Yin, J. Y. Wang, J. Sheng, H. U.
Vora, X. S. Wang, J. Q. Yu, J. Am. Chem. Soc. 2013, 135, 1236;
[
31]
iodide motif and the inclusion of noninnocent additives.
In light of these results, we wondered whether a related
selectivity switch could be applied in another CÀH function-
1
1086 www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 11084 –11087

Angewandte
Chemie
3
III
b) M. Yang, X. Jiang, W. J. Shi, Q. L. Zhu, Z. J. Shi, Org. Lett.
013, 15, 690; c) Y. Li, Y. J. Ding, J. Y. Wang, Y. M. Su, X. S.
Wang, Org. Lett. 2013, 15, 2574; d) L. Min, S. Chang, Tetrahedron
Lett. 2006, 47, 1375.
[21] For another C(sp )ÀH activation assisted by RCO H and I , see:
2
2
T. Dohi, N. Takenaga, A. Goto, A. Maruyama, Y. Kita, Org. Lett.
2007, 9, 3129.
[22] “Method using lifespan-altering compounds for altering the
lifespan of eukaryotic organisms, and screening for such com-
pounds”: D. S. Goldfarb, US20090163545, June 25, 2009.
[23] No reaction took place when N-phenylpyrrolidinone was
exposed to benzoic acid in the presence of a catalytic amount
of I6. A similar result was observed when the methyl ester of 3a
(3a-Me) was used.
[
9] For an excellent review, see: K. M. Engle, T. S. Mei, M. Wasa,
J. Q. Yu, Acc. Chem. Res. 2012, 45, 788.
10] a) J. Gallardo-Donaire, R. Martin, J. Am. Chem. Soc. 2013, 135,
[
9350; b) Y. Wang, A. Gulevich, V. Gevorgyan, Chem. Eur. J.
2
013, 19, 15836.
III
[
11] For selected reviews on the use of I reagents, see: a) R.
Samanta, K. Matcha, A. P. Antonchick, Eur. J. Org. Chem. 2013,
[24] CCDC 1012791 (4l) contains the supplementary crystallo-
graphic data for this paper. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif.
5769; b) M. Uyanik, K. Ishihara, Chem. Commun. 2009, 2086;
c) V. V. Zhdankin, P. J. Stang, Chem. Rev. 2008, 108, 5299; d) T.
Wirth, Angew. Chem. 2005, 117, 3722; Angew. Chem. Int. Ed.
2
3
005, 44, 3656; e) H. Tohma, Y. Kita, Adv. Synth. Catal. 2004,
46, 111.
[25] The formation of 4l is consistent with the formation of iminium
ions: P. Magnus, C. Hulme, W. Weber, J. Am. Chem. Soc. 1994,
116, 4501.
[26] For a mechanistic hypothesis, see the Supporting Information.
[27] The reaction of 1o with I8 or PhI (I1) as catalysts gave rise to
a mixture of 2o and 2o’ in 67 and 56% yield (I8: 1:1.7 ratio; I1:
1:1.2 ratio), thus clearly demonstrating the influence of the aryl
backbone on selectivity. For the use of other ArI catalysts, see
the Supporting Information.
[
12] a) Y. Yamamoto, H. Togo, Synlett 2006, 798; b) M. Ochiai, Y.
Takeuchi, T. Katayama, T. Sueda, K. Miyamoto, J. Am. Chem.
Soc. 2005, 127, 12244; c) T. Dohi, A. Maruyama, M. Yoshimura,
K. Morimoto, H. Tohma, Y. Kita, Angew. Chem. 2005, 117, 6349;
Angew. Chem. Int. Ed. 2005, 44, 6193.
[
13] For selected ArI-catalyzed CÀO bond-forming reactions, see:
a) T. Dohi, N. Takenaga, T. Nakae, Y. Toyoda, M. Yamasaki, M.
Shiro, H. Fujioka, A. Maruyama, Y. Kita, J. Am. Chem. Soc.
[28] Control experiments showed that 2o and 2o’ cannot be
interconverted.
2
013, 135, 4558; b) M. Uyanik, D. Suzuki, T. Yasui, K. Ishihara,
Angew. Chem. 2011, 123, 5443; Angew. Chem. Int. Ed. 2011, 50,
331.
[29] For selected [1,2]-carbon-atom shifts promoted by PhI(OR)2,
see: a) U. Farid, F. Malmedy, R. Claveau, L. Albers, T. Wirth,
Angew. Chem. 2013, 125, 7156; Angew. Chem. Int. Ed. 2013, 52,
7018; b) K. Hata, H. Hamamoto, Y. Shiozaki, S. B. Cꢂmmerer, Y.
Kita, Tetrahedron 2007, 63, 4052. In these examples, the nature
5
2
[
[
14] For a remarkable C(sp )ÀH/CÀO bond-forming process with
K S O , see Ref. [10b].
2
2
8
15] For early studies with stoichiometric PhI(OR)2 under UV
irradiation, see: a) S. Furuyama, H. Togo, Synlett 2010, 2325;
b) T. Muraki, H. Togo, M. Yokoyama, J. Chem. Soc. Perkin
Trans. 1 1999, 1713.
16] See the Supporting Information for details.
17] T. Dohi, M. Ito, K. Morimoto, Y. Minamitsuji, N. Takenaga, Y.
Kita, Chem. Commun. 2007, 4152.
18] Little conversion was observed in the absence of a catalyst.
Additionally, the inclusion of TEMPO or BHT inhibited the
reaction, thus suggesting single-electron-transfer processes.
19] Low yields of 2a were found with 4-iodobenzaldehyde (29%), 4-
of the ArI(OR) reagent employed had minimal influence on
2
selectivity.
[30] CCDC 1012788 (2o) and 1012789 (2o’) contain the supplemen-
tary crystallographic data for this paper. These data can be
obtained free of charge from The Cambridge Crystallographic
Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
[31] The influence of the ArI backbone can be attributed to a greater
[
[
[
[
III
electrophilicity of the I species generated from I6 in compar-
ison to those generated from I11 or I8.
[32] The exposure of 1s to previous copper-catalyzed oxidative
[
10]
iodobenzoic acid (14%), I (40%), and KI (22%). These results
conditions gave neither conversion into 4a, 4a’ nor into 2s.
[33] Compound 1s was cleanly converted into either 4a or 4a’ in an
Ar atmosphere, thus suggesting that atmospheric oxygen does
not come into play.
2
suggest that the reactivity of I8 cannot be merely attributed to
the oxidation of the benzylic CÀH bond in I8 or degradation into
inorganic iodine compounds.
[
20] CCDC 1012792 (2m) contains the supplementary crystallo-
graphic data for this paper. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif.
[34] CCDC 1012793 (4a) and 1012790 (4a’) contain the supplemen-
tary crystallographic data for this paper. These data can be
obtained free of charge from The Cambridge Crystallographic
Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
Angew. Chem. Int. Ed. 2014, 53, 11084 –11087
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 11087