Synergistic palladium-catalyzed C(sp3)-H Activation/C(sp 3)-O bond formation: A direct, step-economical route to benzolactones

Article abstract of DOI:10.1002/anie.201105894

Simplified access: Substituted benzolactones can be obtained in one step by a Pd-catalyzed ligand-accelerated C(sp³)-H bond-activation/C(sp ³)-O bond-formation protocol. This step-economical approach enables the preparation of benzolactones with a wide variety of functional groups and different substitution patterns. The method is characterized by its simplicity and the avoidance of protecting groups.

Full text of DOI:10.1002/anie.201105894

Communications
DOI: 10.1002/anie.201105894
À
C H Activation
3
3
À
À
Synergistic Palladium-Catalyzed C(sp ) H Activation/C(sp ) O Bond
Formation: A Direct, Step-Economical Route to Benzolactones**
Petr Novꢀk, Arkaitz Correa, Joan Gallardo-Donaire, and Ruben Martin*
À
In the last decade, C H bond-activation protocols have
profoundly changed the landscape of organic synthesis
through unconventional bond-disconnection strategies for
the assembly of complex organic molecules.[1] Ideally, direct-
toward benzolactones 2 would imply a direct catalytic
conversion of benzoic acids 1, thus drastically reducing the
overall number of synthetic steps (Scheme 1, path c). Despite
encouraging precedents,[9] particularly the pioneering and
elegant work reported by Sames and co-workers when using
Pt catalysts,[9a] this seemingly routine transformation is not yet
efficient because of the low functional group tolerance, the
high price of Pt,[10] and the limited substitution patterns that
can be accessed, thus enforcing a change in strategy. As part
of our ongoing interest in the synthesis of benzoic acids by Pd-
catalyzed CO2 activation,[11] it was anticipated that 1 might
À
ing groups that are commonly employed in C H activation
processes should have a dual role by assisting chelation and
subsequently promoting further functionalization. The use of
benzoic acids holds great promise in this regard,[2] as
illustrated by the recent work of Yu and co-workers,[3] and
other research groups.[4] Despite the available background
knowledge, the development of catalytic methods for activat-
ing C(sp ) H bonds is still in its infancy.[5] Indeed, it is highly
undergo a Pd -catalyzed activation of a proximal C(sp ) H
3
II
3
À
À
3
À
desirable to design new synthetic pathways based on C(sp )
bond, followed by a virtually unexplored reductive elimina-
H activation[6] in order to dramatically increase molecular
complexity while avoiding tedious functional group manipu-
lation.
tion of a PdII intermediate to form a C(sp ) O bond
3
[12]
À
(Scheme 1, path c).[13,14] This step would be rather challenging
because of the large energy gap between the highest occupied
À
Benzolactones are a prominent structural motif of many
bioactive natural products and pharmaceutically important
compounds.[7] Classical methods for the synthesis of benzo-
lactones include the cyclization of hydroxy acids or halolac-
tonization processes (Scheme 1, path b).[8] Unfortunately,
these methods require prefunctionalization steps, which
limit the applicability because of the need for additional
synthetic steps (Scheme 1, path a). The most attractive route
molecular orbital (HOMO) of the Pd O bond and the lowest
À
unoccupied molecular orbital (LUMO) of the Pd C bond,
[15,16]
À
and the substantial ionic character of the Pd O bond.
Overall, path c (Scheme 1) would constitute a direct, step-
economical approach toward benzolactones 2.[17] We hypothe-
sized that the judicious choice of a supporting ligand and its
appropriate fine-tuning might play an important, if not
critical, role in the synthesis of 2. Herein, we demonstrate
that these two mechanistically distinct Pd-catalyzed processes
can be drastically accelerated by the employment of an N-
protected amino acid as the supporting ligand in the synthesis
of highly functionalized benzolactones with a diverse set of
substitution patterns that are beyond reach otherwise.
We began our study with 1a as the model substrate. After
considerable optimization,[18] we found that the use of Pd-
(OAc)2, K2HPO4, and Ag2CO3 as the oxidant in chloroben-
zene as the solvent afforded a remarkable level of activity
(Table 1). As expected from our previous work in inert-bond
activation,[11,19] a minor modification in the ligand backbone
had a detrimental impact on the reaction outcome. Among
the ligands examined, we noticed that the use of L2 and L3
was highly beneficial (Table 1, entries 2–3).[20] Subsequently,
we found that commercially available N-protected amino
acids L4–L9 could be successfully employed as ligands
Scheme 1. Synthetic approaches to benzolactones.
[*] Dr. P. Novꢀk, Dr. A. Correa, Dr. J. Gallardo-Donaire, Dr. R. Martin
Institute of Chemical Research of Catalonia (ICIQ)
Av. Paꢁsos Catalans, 16, 43007, Tarragona (Spain)
E-mail: rmartinromo@iciq.es
(Table 1, entries 4–10). The beneficial effect of using N-
2
À
protected amino acids for the activation of various C(sp ) H
bonds has already been demonstrated by the pioneering work
of Yu and co-workers;[21] however, the use of N-protected
[**] Financial support from the ICIQ foundation, Consolider Ingenio
2010 (CSD2006-0003), and MICINN (CTQ2009-13840) is gratefully
acknowledged. Johnson Matthey, Umicore, and Nippon Chemical
Industrial are acknowledged for a gift of metal salts and ligands.
A.C., J.G., and R.M. thank MICINN for J.C., FPI, and RyC
fellowships.
3
À
amino acids for the functionalization of C(sp ) H bonds has
received much less attention.[22] To the best of our knowledge,
N-protected amino acids have not been employed as ligands
3
À
in Pd-catalyzed C(sp ) O bond-forming reactions. Ligand L7
was particularly active; its use drastically reduced the yield of
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201105894.
3a while the yield of 2a was increased up to 95% (Table 1,
12236
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 12236 –12239

Table 1: Optimization of reaction conditions.[a]
Entry
Ligand
Yield [%][b]
2a
3a
1
2
3
4
5
6
7
8
L1
L2
L3
L4
L5
L6
L7
L7
L8
L9
–
22
17
1
13
3
12
11
1
63[c]
71[c]
7
74
43
95[c]
80[c,d]
36
1
9
32
37
14
10
42
22
11[e]
[a] 1a (0.50 mmol, 0.25m in PhCl as the solvent), Pd(OAc)2 (10 mol%),
L (30 mol%), Ag2CO3 (3 equiv), K2HPO4 (2.50 equiv), 1408C. [b] Yield
determined by gas chromatography using dodecane as the internal
standard. [c] Yield of isolated product. [d] Pd(OAc)2 (5 mol%) was used.
[e] Conditions reported in Ref. [9] were used. The entry in bold represents
the optimized reaction conditions. Bn=benzyl.
entry 7). Interestingly, this transformation could also be
carried out at a lower catalyst loading with 80% yield
(Table 1, entry 8). Intriguingly, the favourable profile when
using L7 can not be explained simply by electronic or steric
effects, as L8 or L9 (Table 1, entries 9 and 10, respectively) are
not particularly effective, thus showing the subtleties of our
protocol. It is worth mentioning that, in contrast to the
previous use of AgI salts for promoting decarboxylation,[23]
our AgI-mediated protocol only afforded 2a. In order to put
these results into perspective, we compared our optimized
reaction with the Pt-catalyzed process[9] (Table 1, entry 11)
and demonstrated the superior activity of our catalytic system
based on L7.
We next turned our attention to the scope of this
transformation. As shown in Scheme 2, the chemoselectivity
of our methodology was nicely illustrated by the fact that
ethers (2d,q,z), silyl ethers (2g,h), amides (2i), ketones
(2m,za), acetals (2j), esters (2x), alkenes (2y), and trifluoro-
methyl groups (2k) were all well accommodated, thus giving a
direct access to benzolactones that are inaccessible by
classical routes.[8,24] Moreover, the reaction could be con-
ducted in the presence of aryl halides, thus allowing for
further manipulation by conventional cross-coupling reac-
tions (2 f,v). Remarkably, compounds bearing nitrogen-con-
taining heterocycles (2l) or keto groups with a-acidic protons
(2m) remained unaffected, thus showing the potential of this
transformation. More interestingly, our methodology allowed
Scheme 2. Preparation of benzolactones. Reaction conditions as in
Table 1 with ligand L7 and, unless stated otherwise, with PhCl as
solvent. Yields are those of isolated products (average of two
independent runs). [a] A mixture of PhCl/NMP=4:1 was used as
solvent. [b] NMP was used as solvent. [c] Conditions reported in
Ref. [9] were used. NMP=N-methylpyrrolidin-2-one, TIPS=triiso-
propylsilyl, TMS=trimethylsilyl.
benzoic acid unit were employed (2s–2z). In these particular
3
[25,26]
À
cases, selective C(sp ) H bond activation was observed.
These results are quite remarkable in light of the unsuccessful
preparation of 2s–2v under the same reaction conditions
based on Pt catalysts.[9,27] The successful preparation of 2r
(albeit in lower yields) is equally instructive as it indicates that
À
for the discrimination between different C H bonds when
substrates with substituents at the meta position of the
Angew. Chem. Int. Ed. 2011, 50, 12236 –12239
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 12237

Communications
our reaction is not limited to ortho-methyl benzoic acids as
amounts of Pd(OAc)2 were used (Scheme 5, steps a and b,
respectively).[18] Similarly, 1a–Pd was shown to be both
chemically and kinetically competent as an intermediate in
the absence of Ag2CO3, and afforded 2a in 84% yield
(Scheme 5, step c).[18] Interestingly, the absence of K2HPO4
had a deleterious effect and led exclusively to 3a. At present,
we also cannot exclude the intermediacy of k2-PdII com-
plexes.[29] As expected, 1a–Pd could also be used as a
precatalyst and afforded 2a in a comparable 83% yield
(Scheme 5, step c).[18] Even though these experiments cannot
be used as an ultimate proof, they provide evidence that a
PdII/PdIV catalytic cycle[30] is highly unlikely, and support the
notion that the AgI salts act with a dual role by recovering the
catalytic species and participating in an early transmetalation
step (Scheme 4).
substrates. Notably, subtle solvent modification results in
different 4/5 ratios, thus suggesting the intermediacy of PdII/p-
allyl species[25] or styrene derivatives, respectively (Scheme 3).
Scheme 3. Preparation of benzolactones 4 and 5. Reaction conditions
as in Table 1 with ligand L7, and a) NMP (4/5=3.6:1) and b) PhCl/
NMP=4:1 (5/4=5:1) as solvent.
A tentative mechanism for the synthesis of benzolactones
is shown in Scheme 4. We propose a pathway consisting of a
transmetalation of an initially generated silver salt 1a–Ag
We also conducted experiments with isotopically labeled
substrates (Scheme 6). No significant kinetic isotope effect
was observed in the competitive reaction of 1a and [D11]-1a;
Scheme 6. Kinetic isotope effects.
similar results were also found in the intramolecular cycliza-
tion of [D3]-1b. The observed isotope effects suggest that the
3
Scheme 4. Mechanistic hypothesis for the synthesis of benzolactones.
TM=transmetallation.
À
C(sp ) H bond cleavage might not be rate-determining,
which is a rather peculiar observation as the vast majority
of C H activation processes possess high kH/kD values.[31]
À
Although further investigations need to be conducted,[32] at
3
3
3
À
À
À
followed by C(sp ) H activation to give II. C(sp ) O bond-
forming reductive elimination then delivers 2a and Pd0, which
upon oxidation by AgI regenerates the catalytically active
species. At present, we cannot rule out a mechanistic pathway
consisting of the dissociation of the carboxylate ligand in II to
present we support the hypothesis that the C(sp ) O bond-
forming reductive elimination[12,13] is rate-limiting in our
protocol. Overall, we believe our results are highly instruc-
tive, and advocate the notion that L7 might play an important
3
À
role in both the activation of the C(sp ) H bond and the
3
À
generate a p-benzylic Pd intermediate, followed by an
reductive elimination to form the C(sp ) O bond.
[28]
À
intramolecular Pd O bond-forming reaction. To support
In conclusion, we have developed a direct and operation-
ally simple protocol to prepare benzolactones with a wide
variety of functional groups and a diverse set of substitution
patterns. We believe that our represents a significant step
forward for truly applicable, yet active catalysts for a step-
economical route to benzolactones. Future investigations are
aimed at gaining further mechanistic evidence and the
development of an asymmetric version of this route.
our mechanistic hypothesis, we decided to study the reactivity
of some putative intermediates within the catalytic cycle
(Scheme 5). Consistent with the proposed mechanism, 1a–Ag
was cleanly converted to 2a when stoichiometric or catalytic
Received: August 20, 2011
Published online: October 31, 2011
À
Keywords: amino acids · benzolactones · C H activation ·
homogeneous catalysis · palladium
.
Scheme 5. Reactivity of the putative intermediates. a) Pd(OAc)2
(1 equiv), L7 (3 equiv), K2HPO4 (2.50 equiv); b) Pd(OAc)2 (10 mol%),
L7 (30 mol%), Ag2CO3 (3 equiv), K2HPO4 (2.50 equiv); c) L7 (3 equiv),
K2HPO4 (2.50 equiv); d) 1a–Pd (10 mol%), L7 (30 mol%), Ag2CO3
(3 equiv), K2HPO4 (2.50 equiv).
[1] For selected reviews, see: a) C. S. Yeung, M. V. Dong, Chem.
Rev. 2011, 111, 1215 – 1292; b) J. F. Hartwig, Chem. Soc. Rev.
12238 www.angewandte.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 12236 –12239

2011, 40, 1992 – 2002; c) L. McMurray, F. OꢀHara, M. Gaunt,
Chem. Soc. Rev. 2011, 40, 1885 – 1898; d) T. W. Lyons, M.
Sanford, Chem. Rev. 2010, 110, 1147 – 1169; e) L. Ackermann,
R. Vicente, A. R. Kapdi, Angew. Chem. 2009, 121, 9976 – 10011;
Angew. Chem. Int. Ed. 2009, 48, 9792 – 9826011; f) X. Chen,
K. M. Engle, D. H. Wang, J.-Q. Yu, Angew. Chem. 2009, 121,
5196 – 5217; Angew. Chem. Int. Ed. 2009, 48, 5094 – 5115.
[2] T. Satoh, M. Miura, Synthesis 2010, 3395 – 3409.
[3] For selected references, see: a) D. H. Wang, J. Q. Yu, J. Am.
Chem. Soc. 2011, 133, 5767 – 5769; b) D. H. Wang, K. M. Engle,
B. F. Shi, J. Q. Yu, Science 2010, 327, 315 – 319; c) Y.-H. Zhang,
J.-Q. Yu, J. Am. Chem. Soc. 2009, 131, 14654 – 14655.
[4] For selected references, see: a) C. L. Sun, J. Liu, Y. Wang, X.
Zhou, B.-J. Li, Z.-J. Shi, Synlett 2011, 883 – 886; b) H. A. Chiong,
Q.-N. Pham, O. Daugulis, J. Am. Chem. Soc. 2007, 129, 9879 –
9884; c) K. Ueura, T. Satoh, M. Miura, Org. Lett. 2007, 9, 1407 –
1409; d) K. J. Fraunhoffer, N. Prabagaran, L. E. Sirois, M. C.
White, J. Am. Chem. Soc. 2006, 128, 9032 – 9033.
[5] a) H. Li, B.-J. Li, Z.-J. Shi, Catal. Sci. Technol. 2011, 1, 191 – 206;
b) R. Jazzar, J. Hitce, A. Renaudat, J. Sofack-Kreutzer, O.
Baudoin, Chem. Eur. J. 2010, 16, 2654 – 2672.
[6] For selected references, see: a) S. Rousseaux, M. Davi, J. Sofack-
Kreutzer, C. Pierre, C. E. Ketalidis, E. Clot, K. Fagnou, O.
Baudoin, J. Am. Chem. Soc. 2010, 132, 10706 – 10716; b) S.
Rousseaux, S. I. Gorelsky, B. K. W. Chung, K. Fagnou, J. Am.
Chem. Soc. 2010, 132, 10692 – 10705; c) J. J. Neumann, S.
Rakshit, T. Drçge, F. Glorius, Angew. Chem. 2009, 121, 7024 –
7027; Angew. Chem. Int. Ed. 2009, 48, 6892 – 6895; d) M.
Lafrance, S. I. Gorelsky, K. Fagnou, J. Am. Chem. Soc. 2007,
129, 14570 – 14571.
[7] a) K. Yoganathan, C. Rossant, Y. Huang, M. S. Butler, A. D.
Buss, J. Nat. Prod. 2003, 66, 1116 – 1117; b) A. Arnone, G.
Assante, G. Nasini, S. Strada, A. Vercesi, J. Nat. Prod. 2002, 65,
48 – 50.
[15] J. F. Hartwig, Inorg. Chem. 2007, 46, 1936.
IV
3
À
[16] For a related Pd -catalyzed oxygenation of unactivated C(sp )
H bonds, see: a) A. R. Dick, K. L. Hull, M. S. Sanford, J. Am.
Chem. Soc. 2004, 126, 2300 – 2301; b) L. V. Desai, K. L. Hull,
M. S. Sanford, J. Am. Chem. Soc. 2004, 126, 9542 – 9543.
[17] For a recent synthesis of benzolactones by C(sp ) H activation
2
À
with alkyl halides or alkenes, see: a) Y.-H. Zhang, B.-F. Shi, J.-Q.
Yu, Angew. Chem. 2009, 121, 6213 – 6216; Angew. Chem. Int. Ed.
2009, 48, 6097 – 6100; b) L. Ackermann, J. Pospech, Org. Lett.
2011, 13, 4153 – 4155.
[18] See the Supporting Information for details.
[19] a) A. Flores-Gaspar, R. Martin, Adv. Synth. Catal. 2011, 353,
1223 – 1228; b) P. ꢁlvarez-Bercedo, A. Flores-Gaspar, A.
Correa, R. Martin, J. Am. Chem. Soc. 2010, 132, 466 – 467; c) P.
ꢁlvarez-Bercedo, R. Martin, J. Am. Chem. Soc. 2010, 132,
17352 – 17353.
[20] L. Ackermann, Chem. Rev. 2011, 111, 1315 – 1345.
[21] a) H.-X. Dai, A. F. Stepan, M. S. Plummer, Y.-H. Zhang, J.-Q.
Yu, J. Am. Chem. Soc. 2011, 133, 7222 – 7228; b) Y. Lu, D. Leow,
X. Wang, K. M. Engle, J.-Q. Yu, Chem. Sci. 2011, 2, 967 – 971;
c) K. M. Engle, D.-H. Wang, J.-Q. Yu, J. Am. Chem. Soc. 2010,
132, 14137 – 14151; d) Y. Lu, D.-H. Wang, K. M. Engle, J.-Q. Yu,
J. Am. Chem. Soc. 2010, 132, 5916 – 5921, and references therein.
[22] B.-F. Shi, N. Maugel, Y.-H. Zhang, J.-Q. Yu, Angew. Chem. 2008,
120, 4960 – 4964; Angew. Chem. Int. Ed. 2008, 47, 4882 – 4886.
[23] a) L. J. Goossen, C. Linder, N. Rodrꢂguez, P. P. Lange, A.
Fromm, Chem. Commun. 2009, 7173 – 7175; b) J. Cornella, C.
Sanchez, D. Banawa, I. Larrosa, Chem. Commun. 2009, 7176 –
7178.
[24] Future investigations are aimed at improving catalyst perfor-
mance as unreacted starting material was recovered from
reactions with 50–70% yield.
À
[25] For an elegant macrolactonization method initiated by C H
oxidation by p-allyl intermediates, see Ref. [4d].
À
[8] A. K. Yudin, Catalyzed Carbon Heteroatom Bond Formation,
[26] A substituent at the 5-position was required for selectively
À
Wiley-VCH, Weinheim, 2011, pp. 35 – 68.
[9] a) B. Dangel, J. Johnson, D. Sames, J. Am. Chem. Soc. 2001, 123,
8149 – 8150; b) L. Min, S. Chang, Tetrahedron Lett. 2006, 47,
1375 – 1379.
activating the benzylic C H bond. Indeed, the reaction of the
corresponding ortho-methylbenzoic acid did not afford any
benzolactone at all.
[27] No conversion to products was observed.
[10] Platinum is more expensive than palladium (10.000$/mol versus
1500$/mol, respectively).
[11] A. Correa, R. Martin, J. Am. Chem. Soc. 2009, 131, 15974 –
15975.
[28] For related processes, see: a) R. Kuwano, H. Kusano, Org. Lett.
2008, 10, 1979; b) H. Grennberg, J. E. Bꢃckvall, Chem. Eur. J.
1998, 4, 1083.
[29] R. Giri, J. Q. Yu, J. Am. Chem. Soc. 2008, 130, 14082 – 14083.
[30] For selected reviews, see: a) P. Sehnal, R. J. K. Taylor, I. J. S.
Fairlamb, Chem. Rev. 2010, 110, 824 – 889; b) L. M. Xu, B. J. Li,
Z. Yang, Z.-J. Shi, Chem. Soc. Rev. 2010, 39, 712 – 733; c) K.
MuÇiz, Angew. Chem. 2009, 121, 9576 – 9588; Angew. Chem. Int.
Ed. 2009, 48, 9412 – 9423.
[12] While this manuscript was in preparation, an elegant Pd-
3
À
promoted reductive elimination toward C(sp ) O bonds was
reported: S. L. Marquard, J. F. Hartwig, Angew. Chem. 2011, 123,
7257 – 7261; Angew. Chem. Int. Ed. 2011, 50, 7119 – 7123.
[13] For a review on reductive elmininations of high-valent species to
À
C O bonds, see: A. N. Vedernikov, Top. Organomet. Chem.
[31] M. Gꢄmez-Gallego, M. A. Sierra, Chem. Rev. 2011, 111, 4857 –
2010, 31, 101.
4963.
2
3
À
À
[14] For a recent report of C(sp ) O bond formation initiated by
[32] These results do not allow us to rule out a C(sp ) H activation
2
À
phenol-directed C(sp ) H activation, see: B. Xiao, T.-J. Gong,
Z.-J. Liu, J.-H. Liu, D.-F. Luo, J. Xu, L. Liu, J. Am. Chem. Soc.
2011, 133, 9250 – 9253.
via benzylic radicals.
Angew. Chem. Int. Ed. 2011, 50, 12236 –12239
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 12239