Pd-catalyzed aryl C-H imidation with arene as the limiting reagent

Article abstract of DOI:10.1021/ja4064926

An amine-N-oxide-ligated palladium complex, in conjunction with a silver cocatalyst, catalyzes imidation of arenes by the reagent N- fluorobenzenesulfonimide. The reaction enables imidation of a variety of arenes at or below room temperature, requires no coordinating directing group on the substrate, and gives synthetically useful yields with only 1 equiv of arene. Mechanistic data implicate an unusual mechanism devoid of commonly invoked organometallic intermediates: oxidation of the Pd catalyst occurs as the turnover-limiting step, while C-H bond functionalization occurs subsequently at a high oxidation state of the catalyst.

Full text of DOI:10.1021/ja4064926

Communication
pubs.acs.org/JACS
Pd-Catalyzed Aryl C−H Imidation with Arene as the Limiting Reagent
Gregory B. Boursalian,‡ Ming-Yu Ngai,‡ Katarzyna N. Hojczyk, and Tobias Ritter*
Department of Chemistry and Chemical Biology, Harvard University, Cambridge, Massachusetts 02138, United States
S
* Supporting Information
ABSTRACT: An amine-N-oxide-ligated palladium com-
plex, in conjunction with a silver cocatalyst, catalyzes
imidation of arenes by the reagent N-fluorobenzenesulfon-
imide. The reaction enables imidation of a variety of arenes
at or below room temperature, requires no coordinating
directing group on the substrate, and gives synthetically
useful yields with only 1 equiv of arene. Mechanistic data
implicate an unusual mechanism devoid of commonly
invoked organometallic intermediates: oxidation of the Pd
catalyst occurs as the turnover-limiting step, while C−H
bond functionalization occurs subsequently at a high
oxidation state of the catalyst.
ransition metal (TM) catalysis holds promise as a strategy
for the direct functionalization of otherwise unreactive C−
Figure 1. C−H imidation catalyzed by 1 and Ag(bipy)2ClO4, and X-ray
T
structure of the cation of 1 (ellipsoids drawn at 50% probability).
H bonds. Many reported TM-catalyzed aryl C−H functionaliza-
tion approaches rely on metalation of the arene followed by
functionalization of the aryl−metal σ-bond.1 To accelerate an
otherwise sluggish metalation step, coordinating directing groups
are commonly employed to accelerate carbon−metal bond
formation; without such directing groups, often multiple
equivalents or even solvent quantities of arene are required to
drive the metalation.2 With few exceptions,3 catalytic C−H
functionalization without the aid of directing groups remains an
unmet challenge. Here we report a catalytic, intermolecular C−H
imidation of arenes enabled by a departure from the conventional
metalation−functionalization sequence. The new palladium
catalyst 1 enables group transfer without the formation of
conventionally targeted organometallic intermediates. Our
transformation affords synthetically useful yields with the arene
substrate as the limiting reagent.
Imidation of arenes by the reagent N-fluorobenzenesulfon-
imide (NFBS) is catalyzed by 1 and Ag(bipy)2ClO4 as shown in
Figure 1. Both catalyst 1 and the Ag cocatalyst are required in the
reaction; control experiments in which either is omitted gave
<10% of imidated product, although the Ag cocatalyst can be
replaced with Ru(bipy)3(PF6)2 with similar results (Table 1, 2a,
2b). The reaction proceeds equally well in the presence or
absence of light. Catalyst 1 was readily prepared from
tetrakis(acetonitrile)palladium(II) triflate and the N-(2-
pyridylmethyl)pyrrolidine-N-oxide ligand, which itself is avail-
able in two steps from commercial starting materials. The catalyst
can also be generated in situ by mixing Pd(NCMe)4(OTf)2 and
the ligand, with results nearly identical to those obtained with
isolated and purified 1. Because catalyst 1 is easily prepared and
stored, we have used it directly in our investigations. The unusual
pyridine-N-oxide ligand motif is exceptionally effective for the
catalytic imidation; several other Pd-based catalysts supported by
ligands such as bipyridine afforded product in at most 3% yield
(Supporting Information). A rationale for the distinct ability of
the amine-N-oxide to effect arene functionalization is presented
below; we propose that the reactivity of well-defined catalyst 1
induced by the amine-N-oxide ligands is also responsible for the
distinct reactivity when compared to other Pd-catalyzed
amination reactions, such as benzylic amination with NFBS.8
A variety of arenes, including N- and S-heteroarenes, could be
efficiently imidated (Table 1). Selectivity is substrate-intrinsic;
resonance donors, such as alkoxy and halogen groups, direct
imidation ortho/para, similar to electrophilic aromatic sub-
Historically, perhaps the most important method of
introducing an amino group into arenes has been electrophilic
aromatic nitration followed by reduction of the nitro group.4
However, nitration of arenes typically requires strongly acidic or
oxidizing reaction conditions, which limits its use on substrates
with sensitive functional groups. Alternatively, several aniline
derivatives can be reliably prepared by TM-catalyzed directed
C−H amidation,5 but the requirement for a coordinating
directing group limits the potential substrate scope. Non-
chelation-assisted TM-catalyzed C−H amination reactions,6 as
well as metal-free approaches involving the use of an oxidant and
amine to introduce the C−N bond, have also been reported.7
However, unless the arene has particularly acidic C−H bonds,
such as in perfluoroarenes or benzoxazole derivatives,6c,d,7b these
methods require a minimum of 1.5 equiv, and not uncommonly
solvent quantities of arene substrate, and yields are often based
on the amidating reagent, not the arene.
Received: June 26, 2013
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ja4064926 | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
Table 1. Substrate Scope of C−H Imidation Catalyzed by 1 and Ag(bipy)2ClO4
a
b
Ru(bipy)3(PF6)2 (2.5 mol%) used in place of Ag(bipy)2ClO4, 50 °C reaction temperature, 0.2 M. Along with arene imidation, 4% benzylic
c
imidation was observed. Reaction performed at 4 °C. * denotes site of amidation of other constitutional isomer.
stitution. Inductive donors do not direct as effectively (e.g., 2c,
2j), and substrates that lack a strong directing bias (e.g., 2n−2p)
afford mixtures of constitutional isomers. Functional groups that
can cause problems for nitration such as esters and silanes are
tolerated. Arenes more electron-poor than those shown in Table
1 show diminished reactivity and give lower yields. Competitive
C−H fluorination and double imidation are significant side
reactions for more electron-rich arenes, though performing the
reaction at lower temperature (4 °C) can substantially reduce
both side reactions for some substrates (2r−2t, 2v, 2w, 2y).
Potential coordinating directing groups do not influence
regioselectivity as they do in directed catalytic C−H function-
alization reactions5 (2b, 2e, 2l, 2q). The reactions for which the
data are shown in Table 1 were performed rigorously dry and air-
free, but similar results were obtained for reactions performed in
air with reagent-quality solvents (see Supporting Information).
B
dx.doi.org/10.1021/ja4064926 | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
Product 2b, synthesized on 12 g scale with this methodology, was
chosen to demonstrate the removal of the sulfonyl protecting
groups. Treatment of 2b with magnesium in methanol under
sonication produced aniline 3b in 85% yield (eq 1).
Mechanistic data, detailed below, unambiguously implicate a
mechanism with the following noteworthy features: (1)
oxidation of the bis-cationic Pd(II) complex 1 enabled by the
amine-N-oxide ligands, (2) involvement of the cocatalyst in
single-electron redox chemistry, (3) irreversible substrate
binding prior to C−H bond functionalization, and (4) C−N
bond formation occurring without C−H palladation. A plausible
mechanistic proposal including these features is depicted in
Scheme 1. Turnover-limiting oxidation of catalyst 1 yields
Figure 2. Valence orbital diagram of 1.
Turnover-limiting oxidation of 1 by NFBS is supported by the
measured rate law of the reaction, which is first-order with
respect to both 1 and NFBS, and zero-order with respect to arene
and Ag(bipy)2ClO4. When only NFBS and 1 are combined in
acetonitrile, catalytic reduction of NFBS to HN(SO2Ph)2 and
HF is observed, with the reducing equivalents evidently derived
from the solvent. The rate of the catalytic reduction of NFBS is
identical to that of NFBS consumption in the imidation reaction
(Scheme 2). The rate law and the identical rate for both reactions
shown in Scheme 2 establish turnover-limiting oxidation of 1 by
NFBS to yield a common, short-lived high-valent Pd
intermediate such as II.
+
Pd(IV) complex II. Single-electron reduction of II by Ag(bipy)2
and substrate association follow to form Pd(III) intermediate III.
Intermediate III formally transfers sulfonimidyl radical to the
bound substrate, which expels a delocalized radical and
regenerates 1. Oxidation of the radical intermediate by
Ag(bipy)22+, followed by deprotonation, yields the sulfon-
imidated product.
Oxidation of Pd(II) complexes typically requires strongly σ-
donating anionic ligands such as hydrocarbyl ligands, which are
absent from 1.9 DFT calculations suggest that the HOMO of 1 is
an extended M−L π-antibonding orbital of dxz parentage from
Scheme 1. Proposed Catalytic Cycle
2
Pd, instead of the dz -based orbital more typical of square-planar
d8 complexes. The oxygen lone pairs of the amine-N-oxide ligand
interact strongly with the Pd-based dxz orbital, driving it higher in
2
energy than the dz -based orbital (Figure 2). This interaction may
explain how complex 1 can be oxidized by NFBS, despite its two
formal positive charges.
We propose that the Ag cocatalyst serves to provide access to
intermediate III, which is the putative intermediate responsible
2+
for C−N bond formation. Ag(bipy)2 was observed by EPR
spectroscopy during catalysis, which implicates the cocatalyst in
redox reactivity. Inner-sphere reactivity of the cocatalyst can be
ruled out because coordinatively saturated Ru(bipy)3(PF6)2 is
also an effective cocatalyst. Furthermore, the oxidation of
Ru(bipy)32+ to Ru(bipy)33+ by NFBS is substantially accelerated
in the presence of 1, consistent with oxidation of 1 to II by NFBS
2+
followed by single-electron oxidation of Ru(bipy)3 by II (see
Supporting Information for details). Reduction of II by the
cocatalyst to yield the C−N bond-forming species III explains
why, for most arenes, substrate consumption is not observed in
the absence of the cocatalyst.
Because oxidation of 1 by NFBS is turnover-limiting, the C−N
bond-forming step cannot be studied kinetically. We therefore
employed competition kinetic isotope effect (KIE) experiments
to probe this step (Scheme 3). An inverse secondary intra-
Scheme 2. Kinetic Study of Imidation and NFBS Reduction
Catalyzed by 1
molecular kH/kD = 0.80
0.01 for imidation of 1,3,5-
trideuteriobenzene was measured. The measured KIE implicates
rehybridization of the C−H bond from sp2 to sp3 in the product-
determining transition state, consistent with C−N bond
formation via inner-sphere addition of dibenzenesulfonimidyl
radical to the bound arene in intermediate III. Inverse secondary
KIEs are unusual for Pd-catalyzed C−H functionalization
reactions; C−H palladations at Pd(II)10 and Pd(IV)11 typically
display primary KIE values. Electrophilic addition would also be
consistent with the observed intramolecular isotope effect.
However, inverse secondary KIE values for electrophilic
C
dx.doi.org/10.1021/ja4064926 | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
Campbell for helpful discussions, M. G. Campbell for X-ray
analysis, and F. Bastuck for 5 g scale synthesis of catalyst 1.
Scheme 3. Intra- and Intermolecular Kinetic Isotope Effects
REFERENCES
■
(1) (a) Chen, X.; Engle, K. M.; Wang, D.-H.; Yu, J.-Q. Angew. Chem.,
Int. Ed. 2009, 48, 5094. (b) Lyons, T. W.; Sanford, M. S. Chem. Rev.
2010, 110, 1147. (c) Engle, K. M.; Mei, T.-S.; Wasa, M.; Yu, J.-Q. Acc.
Chem. Res. 2012, 45, 788. (d) Neufeldt, S. R.; Sanford, M. S. Acc. Chem.
Res. 2012, 45, 936.
(2) Kuhl, N.; Hopkinson, M. N.; Wencel-Delord, J.; Glorius, F. Angew.
Chem., Int. Ed. 2012, 51, 10236.
(3) (a) Mkhalid, I. A. I.; Barnard, J. H.; Marder, T. B.; Murphy, J. M.;
Hartwig, J. F. Chem. Rev. 2010, 110, 890. (b) Nagib, D. A.; MacMillan,
D. W. C. Nature 2011, 480, 224.
(4) Olah, G. A.; Ripudaman, M.; Narang, S. C. Nitration: Methods and
Mechanisms; VCH Publishers, Inc.: New York, 1989.
(5) (a) Thu, H.-Y.; Yu, W.-Y.; Che, C.-M. J. Am. Chem. Soc. 2006, 128,
9048. (b) Ng, K.-H.; Chan, A. S. C.; Yu, W.-Y. J. Am. Chem. Soc. 2010,
132, 12862. (c) Xiao, B.; Gong, T.-J.; Xu, J.; Liu, Z.-J.; Liu, L. J. Am.
Chem. Soc. 2011, 133, 1466. (d) Sun, K.; Li, Y.; Xiong, T.; Zhang, J.;
Zhang, Q. J. Am. Chem. Soc. 2011, 133, 1694. (e) Yoo, E. J.; Ma, S.; Mei,
T.-S.; Chan, K. S. L.; Yu, J.-Q. J. Am. Chem. Soc. 2011, 133, 7652. (f) Ng,
K.-H.; Zhou, Z.; Yu, W.-Y. Org. Lett. 2012, 14, 272. (g) Grohmann, C.;
Wang, H.; Glorius, F. Org. Lett. 2012, 14, 656.
substitution are rare;12 most commonly, KIEs close to unity are
observed for electrophilic processes due to the opposing effects
of rehybridization and hyperconjugation on the zero-point
vibrational energy of the affected C−H bond.13 Addition of
nitrogen-based radicals to arenes is well-precedented, including
catalytic imidations with NFBS which are proposed to proceed
through high-valent transition metal imidyl radical species.14
Inner-sphere attack from an intermediate such as III to form the
putative aryl radical is supported by an intermolecular
competition kH/kD = 1.03 0.02. The absence of a KIE in this
case is consistent with irreversible substrate binding prior to C−
N bond formation.10,11b
We have described the first general aryl C−H imidation
reaction which gives synthetically useful yields with only 1 equiv
of arene and does not require coordinating directing groups. The
transformation is made possible by a departure from the most
common strategies: we have developed a catalyst supported by
amine-N-oxide ligands, which enable oxidation of the doubly
cationic Pd(II) complex prior to substrate activation. C−H
functionalization proceeds from a high oxidation state complex
without the formation of conventional organometallic inter-
mediates. We anticipate that our distinct approach may find
utility in other C−H functionalization reactions.
(6) (a) Díaz-Requejo, M. M.; Belderraín, T. R.; Nicasio, M. C.;
Trofimenko, S.; Perez, P. J. J. Am. Chem. Soc. 2003, 125, 12078. (b) Li,
́
Z.; Capretto, D. A.; Rahaman, R. O.; He, C. J. Am. Chem. Soc. 2007, 129,
12058. (c) Zhao, H.; Wang, M.; Su, W.; Hong, M. Adv. Synth. Catal.
2010, 352, 1301. (d) Guo, S.; Qian, B.; Xie, Y.; Xia, C.; Huang, H. Org.
Lett. 2011, 13, 522. (e) Froehr, T.; Sindlinger, C. P.; Kloeckner, U.;
Finkbeiner, P.; Nachtsheim, B. J. Org. Lett. 2011, 13, 3754. (f) Shrestha,
R.; Mukherjee, P.; Tan, Y.; Litman, Z. C.; Hartwig, J. F. J. Am. Chem. Soc.
2013, 135, 8480.
(7) (a) Antonchick, A. P.; Samanta, R.; Kulikov, K.; Lategahn, J. Angew.
Chem., Int. Ed. 2011, 50, 8605. (b) Froehr, T.; Sindlinger, C. P.;
Kloeckner, U.; Finkbeiner, P.; Nachtsheim, B. J. Org. Lett. 2011, 13,
3754. (c) Kim, H. J.; Kim, J.; Cho, S. H.; Chang, S. J. Am. Chem. Soc.
2011, 133, 16382. (d) Samanta, R.; Bauer, J. O.; Strohmann, C.;
Antonchick, A. P. Org. Lett. 2012, 14, 5518.
(8) Iglesias, A.; Alvarez, R.; de Lera, A.; Muniz, K. Angew. Chem., Int. Ed.
̃
2012, 51, 2225.
(9) (a) Oloo, W.; Zavalij, P. Y.; Zhang, J.; Khaskin, E.; Vedernikov, A.
N. J. Am. Chem. Soc. 2010, 132, 14400. (b) McCall, A. S.; Wang, H.;
Desper, J. M.; Kraft, S. J. Am. Chem. Soc. 2011, 133, 1832. (c) McCall, A.
S.; Kraft, S. Organometallics 2012, 31, 3527.
(10) Simmons, E. M.; Hartwig, J. F. Angew. Chem., Int. Ed. 2012, 51,
3066.
ASSOCIATED CONTENT
* Supporting Information
Detailed experimental procedures, spectroscopic characteriza-
tion for all new compounds, mechanistic data, and crystallo-
graphic information files. This material is available free of charge
via the Internet at http://pubs.acs.org.
■
S
(11) (a) Rosewall, C. F.; Sibbald, P. A.; Liskin, D. V.; Michael, F. E. J.
Am. Chem. Soc. 2009, 131, 9488. (b) Sibbald, P. A.; Rosewall, C. F.;
Swartz, R. D.; Michael, F. E. J. Am. Chem. Soc. 2009, 131, 15945.
(c) Racowski, J. M.; Ball, N. D.; Sanford, M. S. J. Am. Chem. Soc. 2011,
133, 18022.
(12) (a) Olah, G. A.; Kuhn, S. J.; Flood, S. H. J. Am. Chem. Soc. 1961,
83, 4571. (b) Olah, G. A. Acc. Chem. Res. 1971, 4, 240. (c) Nakane, R.;
Kurihara, O.; Takematsu, A. J. Org. Chem. 1971, 36, 2753.
(13) (a) Streitwieser, A.; Jagow, R. H.; Fahey, R. C.; Suzuki, S. J. Am.
Chem. Soc. 1958, 80, 2326. (b) Melander, L. Isotope Effects on Reaction
Rates; The Ronald Press Co.: New York, 1960; p 112. (c) Taylor, R.
Electrophilic Aromatic Substitution; John Wiley & Sons: West Sussex,
England, 1990; p 33.
AUTHOR INFORMATION
Corresponding Author
ritter@chemistry.harvard.edu
■
Author Contributions
‡G.B.B. and M.-Y.N. contributed equally.
Notes
(14) (a) Stella, L. Angew. Chem., Int. Ed. Engl. 1983, 22, 337. (b) Ni, Z.;
Zhang, Q.; Xiong, T.; Zheng, Y.; Li, Y.; Zhang, H.; Zhang, J.; Liu, Q.
Angew. Chem., Int. Ed. 2012, 51, 1244. (c) Zhang, H.; Pu, W.; Xiong, T.;
Li, Y.; Zhou, X.; Sun, K.; Liu, Q.; Zhang, Q. Angew. Chem., Int. Ed. 2013,
52, 2529. (d) Kaneko, K.; Yoshino, T.; Matsunaga, S.; Kanai, M. Org.
Lett. 2013, 15, 2502.
The authors declare no competing financial interest. A
provisional patent application on catalysts and methods
presented herein has been filed through Harvard.
ACKNOWLEDGMENTS
■
Funding was provided by NIH-NIGMS (GM088237) NIH-
NIBIB (EB013042). G.B.B. thanks NSF GRFP (DGE0644491)
for a graduate fellowship. We thank S. E. Parker and M. G.
D
dx.doi.org/10.1021/ja4064926 | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX