Orthogonal reactivity of acyl azides in C-H activation: Dichotomy between C-C and C-N amidations based on catalyst systems

Article abstract of DOI:10.1021/ol500602b

The dual reactivity of acyl azides was utilized successfully in C-H activation by the choice of catalyst systems: while selective C-C amidation was achieved under thermal Rh catalysis, a Ru catalyst was found to mediate direct C-N amidation also highly se

Full text of DOI:10.1021/ol500602b

Letter
pubs.acs.org/OrgLett
Orthogonal Reactivity of Acyl Azides in C−H Activation: Dichotomy
between C−C and C−N Amidations Based on Catalyst Systems
Kwangmin Shin,‡,† Jaeyune Ryu,‡,† and Sukbok Chang*,†,‡
†Center for Catalytic Hydrocarbon Functionalization, Institute of Basic Science (IBS), Daejeon 305-701, Korea
‡Department of Chemistry, Korea Advanced Institute of Science & Technology (KAIST), Daejeon 305-701, Korea
S
* Supporting Information
ABSTRACT: The dual reactivity of acyl azides was utilized
successfully in C−H activation by the choice of catalyst
systems: while selective C−C amidation was achieved under
thermal Rh catalysis, a Ru catalyst was found to mediate direct
C−N amidation also highly selectively. Investigations of the mechanistic dichotomy between two catalytic systems are also
presented.
Scheme 1
ransition-metal-catalyzed direct C−H bond functionaliza-
Ttion has emerged as a straightforward tool for the
construction of carbon−carbon or carbon−heteroatom bonds.1
Along with the notable advances in the C−H bond activation
methods, metal-mediated C−H additions to carbon−carbon
multiple bonds have been extensively investigated.2 On the
other hand, C−H addition across unsaturated carbon−nitrogen
bonds is still in its infancy.3,4 In particular, procedures allowing
for the direct insertion of C−H bonds into isocyanates are
highly demanding since it can effectively provide synthetically
valuable amide moieties.4 While Re-catalyzed C−H addition of
(hetero)arenes to isocyanates were reported earlier,4a,b Rh-4c,e
and Ru-catalyzed4d procedures were disclosed more recently.
Acyl azides have been widely used in organic synthesis, and
among those examples, the most notable utility is to employ
them as precursors for isocyanates via the ‘Curtius rearrange-
ment’ which can be induced most often thermally (Scheme
1a).5 Nucleophiles react with the carbon center of in situ
generated isocyanates to afford amide products.6 However, to
the best of our knowledge, acyl azides have never been applied
to the direct C−H functionalization as precursors of
isocyanates mainly due to the difficulty in controlling the
dual reactivity of acyl azides, thus leading to a mixture of C−C
and C−N amidated products (Scheme 1b).
In the context of our studies to utilize organic azides as
efficient amino sources in the C−H functionalizations,7 we have
developed the Rh-8 and Ru-catalyzed9 C−H amination
protocols using sulfonyl, aryl, and alkyl azides. Acyl azides
were also found to work as an amido source in the Ir-catalyzed
amidation under mild conditions.10 In these studies, we were
curious about whether acyl azides can be employed in a
selective manner, thus serving not only as an amino source but
also as a carbon donor. Described herein is the first example of
controlling the dual reactivity of acyl azides between C−C and
C−N amidations depending on catalyst systems (Scheme 1c).
In addition, mechanistic studies on this orthogonal selectivity
are also presented.
We initially tried to find optimal conditions for the selective
C−C amidation of 2-phenylpyridine (1a) with benzoyl azide
(2a) (Table 1). A combination of [RhCp*Cl2]2 with AgSbF6, a
catalyst system widely employed in C−H functionalizations,11
resulted in both C−C (3a) and C−N (4a) amidations in high
yield, but with low selectivity (entry 1). Employing other silver
additives possessing weakly coordinating counterions did not
improve the selectivity in 1,2-dichloroethane (entries 2−4).
Whereas the use of other solvents resulted in only marginal
effects (entries 5−7), an additive of acetonitrile gave rise to a
significant increase of selectivity. For instance, the C−C
amidation product (3a) was formed highly favorably (17:1)
over the C−N product (4a) with the addition of 1.0 equiv of
CH3CN (relative to 1a, entry 8). However, an excessive
amount was detrimental (entry 9). When a cationic Rh species
Received: February 25, 2014
Published: March 14, 2014
© 2014 American Chemical Society
2022
dx.doi.org/10.1021/ol500602b | Org. Lett. 2014, 16, 2022−2025

Organic Letters
Letter
with various sensitive functional groups such as ketone, ester, or
aldehyde in addition to a halide (3g−3i).
Table 1. Optimization Table of the Rh-Catalyzed C−C
a
Amidation Reaction
The scope of acyl azides was next examined by varying
substituents. Benzoyl azides substituted with electron-donating
or -neutral groups underwent the C−C amidation smoothly,
providing the desired products in good yields with high
chemoselectivity (3j−3l). Similarly, reactants bearing electron-
withdrawing groups were viable for the present amidation
(3m−3n), indicating that electronic variation of azides did not
much affect the reaction efficiency and selectivity. Azides
substituted with bromo and chloro groups also participated in
the reaction (3o−3p). The preference for the C−C amidation
over C−N amidation was observed to be over 10:1 (measured
yield
b
entry
catalyst system (mol %)
solvent
3a/4a
(%)
1
2
3
4
5
6
7
[RhCp*Cl2]2 (4) + AgSbF6 (16) 1,2-DCE
[RhCp*Cl2]2 (4) + AgNTf2 (16) 1,2-DCE
[RhCp*Cl2]2 (4) + AgBF4 (16) 1,2-DCE
1.7:1
1.2:1
1.5:1
1.3:1
1.3:1
2.1:1
3.8:1
17:1
17:1
80
68
90
90
79
90
82
73
36
91
[RhCp*Cl2]2 (4) + AgPF6 (16)
1,2-DCE
[RhCp*Cl2]2 (4) + AgSbF6 (16) THF
[RhCp*Cl2]2 (4) + AgSbF6 (16) 1,4-dioxane
[RhCp*Cl2]2 (4) + AgSbF6 (16) toluene
[RhCp*Cl2]2 (4) + AgSbF6 (16) 1,2-DCE
[RhCp*Cl2]2 (4) + AgSbF6 (16) 1,2-DCE
1
by H NMR of crude reaction mixture) in all cases examined
(except 3j where it was 8:1).
Having established the Rh-catalyzed selective C−C amida-
tion procedure, we next tried to develop a new system allowing
for selective C−N amidation using acyl azides as amino sources.
While we were unable to switch the chemoselectivity favoring
C−N amidation by using rhodium catalysts even under various
conditions,13 we were pleased to see that a ruthenium(II)
catalyst14 could accomplish this goal (Table 2). While a cationic
c
8
d
9
10 [RhCp*(MeCN)3][SbF6]2 (8)
1,2-DCE
8.1:1
a
b
1a (0.2 mmol) and 2a (0.36 mmol) in solvent (0.5 mL). Total yield
of 3a and 4a, determined by H NMR. CH3CN (100 mol %) was
c
1
d
added. CH3CN (0.2 mL) was added.
bound to CH3CN, prepared separately according to the
literature,12 was used as a catalyst, a slightly lower selectivity,
but with a still synthetically satisfactory ratio, was observed
(entry 10).
a
Table 2. Additive Effects in the Ru-Catalyzed Amidation
With the optimized conditions in hand, we then investigated
the generality of the selective C−C amidation reaction by using
a range of substrates and acyl azides (Scheme 2). Aryls bearing
entry
additives
none
temp (°C) yield (3a, %) yield (4a, %)
a
1
2
3
4
5
6
70
50
50
50
50
25
<5
<5
<5
<5
<5
<5
17
72
50
68
Scheme 2. Scope of the Rh-Catalyzed C−C Amidation
NaOAc
1-AdCO2H
MesCO2H
b
o-NO2C6H4CO2H
o-NO2C6H4CO2H
85(80 )
70
a
1a (0.2 mmol) and 2a (0.36 mmol) in 1,2-DCE (0.5 mL). Yields are
b
1
determined by H NMR. Isolated yields are given.
species, generated in situ from [RuCl2(p-cymene)]2 and
AgSbF6, catalyzed the C−N amidation at 70 °C with high
selectivity but in low efficiency (entry 1), the product yield was
significantly increased in the presence of additives.
Among various additives screened,15 a Brønsted acid of high
acidity was most effective leading to the desired C−N
amidation product (4a) in high yield and selectivity (entry
5).14,16 The C−N amidation proceeded even at 25 °C (entry
6), highlighting the effect of the acid additive on the reactivity.
With the optimized conditions in hand, we investigated the
scope of the Ru-catalyzed selective C−N amidation reaction
(Scheme 3). In general, product yields of 4 were higher than
those of C−C amidated products (3) that employed the Rh
catalytic system (Scheme 2). Electronic influence on the
reaction efficiency was negligible (4a−4d). Substrates possess-
ing various sensitive functional groups such as bromo, chloro,
ketone, ester, or aldehyde all smoothly underwent the C−N
amidation to provide the desired products in synthetically
acceptable yields (4e−4i). The scope of benzoyl azides under
the present Ru-catalyzed conditions was also broad to include
either electron-donating or -withdrawing substituents (4j−4m).
More notably, benzoyl azides substituted with ester or halide
groups were also amidated without difficulty (4n−4p). It needs
a
1 (0.2 mmol) and 2 (0.36 mmol) in 1,2-DCE (0.5 mL). Isolation
b
c
yields are given. [RhCp*(MeCN)3][SbF6]2 (8 mol %). A side
product (C−N amidated 4j) was also detected (3j/4j, 8:1).
electron-neutral or -donating substituents were smoothly
amidated at the ortho-position relative to the 2-pyridyl group
with moderate to high yields (3a−3d). Substrates bearing a
bromo or chloro group were selectively reacted to afford the
desired products (3e−3f), thus offering potential for further
functionalizations. The reaction conditions were compatible
2023
dx.doi.org/10.1021/ol500602b | Org. Lett. 2014, 16, 2022−2025

Organic Letters
Letter
a
Scheme 3. Scope of the Ru-Catalyzed C−N Amidation
Scheme 5. Stoichiometric Reactions with Metallacycles
conditions (Scheme 5c).15 This last result suggests that the
observed high selectivity in the case of the ruthenium catalyst
system is not intrinsic, but it is originated rather from the
relative concentrations of acyl azides and isocyanates present
under the reaction conditions (vide infra).
The postulated nonintrinsic chemoselectivity of the
ruthenium catalytic system was further verified by a series of
control experiments (Scheme 6). When equimolar amounts of
a
1 (0.2 mmol) and 2 (0.36 mmol) in 1,2-DCE (0.5 mL). Isolation
b
yields are given. NaOAc (30 mol %) was used instead of (o-
NO2)C6H4CO2H. A side product (C−C amidated 3k) was also
detected (3k/4k, 1:6).
c
Scheme 6. Control Studies Using Ru-Catalytic System
to be mentioned that the C−N amiadtion occurred with high
selectivity to form C−C amidated products in <5% in all cases
examined except 4k, where 4k/3k ≈ 6:1.
Considering the notable chemoselectivity between Rh- and
Ru-systems, a working mode leading to this dichotomy was
preliminarily investigated. As anticipated, when separately
prepared metallacycles of rhodium (5) and ruthenium (7)17
were employed as catalysts in the reaction of 2-phenylpyridine
(1a) with benzoyl azide (2a), each species was found to
catalyze the C−C and C−N amidations, respectively (Scheme
4).15
benzoyl azide (2a) and phenyl isocyanate (2a′) were allowed to
react with 2-phenylpyridine under the Ru-catalyzed conditions,
2a′ showed higher reactivity than 2a at an early stage of
conversion (Scheme 6a).15 In addition, while the C−N
amidation gave a mixture of product 3b and 4b at 70 °C, the
same reaction at 50 °C gave 4b exclusively (Scheme 6b).15 The
higher selectivity seen at 50 °C is probably due to the slower
decomposition rate of azide 2b at that temperature (Scheme
6c).15 These results suggest that the observed high chemo-
selectivity for the C−N amidation under the ruthenium
catalytic conditions is mainly due to the fast amidation with
the concomitant slow rearrangement of acyl azides.
On the basis of above studies, a plausible rationale for the
mechanistic dichotomy between C−C and C−N amidation is
depicted in Scheme 7. In the Rh-catalyzed system, the selective
C−C bond formation may be attributed to the difference in
binding affinity of a metal center between acyl azides and in situ
generated isocyanates. It is assumed that while acyl azides
display low binding affinity to an acetonitrile-bound cationic Rh
species (5), those azides rather undergo a thermal Curtius
rearrangement to isocyanates that readily replaces an
acetonitrile ligand, eventually leading to a C−C amidated
product. On the other hand, the Ru-catalyzed amidation is
proposed to be governed by kinetic parameters. At temper-
atures 50 °C or below, since the formation of isocyanates is
Scheme 4. Catalytic Reactivity of Metallacycle Catalysts
To further shed light on orthogonal reactivity depending on
catalytic conditions used, stoichiometric reactions of rhoda-
cycles were subsequently examined (Scheme 5). It was found
that an acetonitrile-bound cationic rhodacycle (5) reacted with
isocyanate (2a′) to lead to C−C amidation (3a), but reaction
with benzoyl azide (2a) did not occur (Scheme 5a).15 In
contrast, a rhodacycle (6)17 containing a 2-phenylpyridine
ligand displayed an interesting dual reactivity: both C−C and
C−N amidations took place by changing reactants (Scheme
5b).15 This notable difference in reactivity between complexes
5 and 6 toward acyl azides and isocyanates clearly indicates the
critical role of the acetonitrile ligand on the selectivity. On the
other hand, a neutral ruthenacycle (7) exhibited dual activity
for both C−C and C−N amidations under otherwise identical
2024
dx.doi.org/10.1021/ol500602b | Org. Lett. 2014, 16, 2022−2025

Organic Letters
Letter
Ellman, J. A. Angew. Chem., Int. Ed. 2013, 52, 629. (e) Yoshino, T.;
Ikemoto, H.; Matsunaga, S.; Kanai, M. Angew. Chem., Int. Ed. 2013, 52,
2207.
Scheme 7. Catalytic Dichotomy between C−C (Rh) and C−
N (Ru) Amidations
(4) For selected examples of C−H addition to isocyanates, see:
(a) Kuninobu, Y.; Tokunaga, Y.; Kawata, A.; Takai, K. J. Am. Chem.
Soc. 2006, 128, 202. (b) Kuninobu, Y.; Tokunaga, Y.; Takai, K. Chem.
Lett. 2007, 36, 872. (c) Hesp, K. D.; Bergman, R. G.; Ellman, J. A. J.
Am. Chem. Soc. 2011, 133, 11430. (d) Muralirajan, K.; Parthasarathy,
K.; Cheng, C.-H. Org. Lett. 2012, 14, 4262. (e) Zhou, B.; Hou, W.;
Yang, Y.; Li, Y. Chem.Eur. J. 2013, 19, 4701.
(5) Curtius, T. Ber. Dtsch. Chem. Ges. 1890, 23, 3023.
(6) (a) Brase, S.; Gil, C.; Knepper, K.; Zimmermann, V. Angew.
̈
Chem., Int. Ed. 2005, 44, 5188. (b) Lebel, H.; Leogane, O.; Huard, K.;
Lectard, S. Pure Appl. Chem. 2006, 78, 363.
(7) (a) Wang, Y.-F.; Chiba, S. J. Am. Chem. Soc. 2009, 131, 12570.
(b) Nguyen, Q.; Sun, K.; Driver, T. G. J. Am. Chem. Soc. 2012, 134,
7262. (c) Nishioka, Y.; Uchida, T.; Katsuki, T. Angew. Chem., Int. Ed.
2013, 52, 1739. (d) Hennessy, E. T.; Betley, T. A. Science 2013, 340,
591. (e) Yu, D.-G.; Suri, M.; Glorius, F. J. Am. Chem. Soc. 2013, 135,
8802. (f) Lian, Y.; Hummel, J. R.; Bergman, R. G.; Ellman, J. A. J. Am.
Chem. Soc. 2013, 135, 12548.
(8) (a) Kim, J. Y.; Park, S. H.; Ryu, J.; Cho, S. H.; Kim, S. H.; Chang,
S. J. Am. Chem. Soc. 2012, 134, 9110. (b) Ryu, J.; Shin, K.; Park, S. H.;
Kim, J. Y.; Chang, S. Angew. Chem., Int. Ed. 2012, 51, 9904. (c) Shin,
K.; Baek, Y.; Chang, S. Angew. Chem., Int. Ed. 2013, 52, 8031. (d) Park,
S. H.; Park, Y.; Chang, S. Org. Synth. 2014, 91, 52. (e) Kim, H. J.;
Ajitha, M. J.; Lee, Y.; Ryu, J.; Kim, J.; Lee, Y.; Jung, Y.; Chang, S. J. Am.
Chem. Soc. 2014, 136, 1132. (f) Park, S. H.; Kwak, J.; Shin, K.; Ryu, J.;
Park, Y.; Chang, S. J. Am. Chem. Soc. 2014, 136, 2492.
negligible residing in low concentration, a direct reaction of acyl
azides with a ruthenacyclic intermediate (7) will be
predominant (k1 ≫ k2).
In summary, the dual reactivity of acyl azides was controlled
in the C−H functionalization by altering catalyst systems.
Rhodium catalysis enabled the selective C−C amidation with
isocyanates in situ generated from acyl azides. In contrast, a
ruthenium catalyst system facilitated the selective C−N
amidation with acyl azides.18 Mechanistic studies revealed
that the dichotomy between two catalytic systems might
originate from chemoselective and kinetic control. This work
opens the way to utilizing acyl azides in C−H functionalization
by controlling their dual reactivity as a nitrogen or carbon
source.
(9) Kim, J.; Kim, J.; Chang, S. Chem.Eur. J. 2013, 19, 7328.
(10) (a) Ryu, J.; Kwak, J.; Shin, K.; Lee, D.; Chang, S. J. Am. Chem.
Soc. 2013, 135, 12861. (b) Lee, D.; Kim, Y.; Chang, S. J. Org. Chem.
2013, 78, 11102. (c) Kim, J.; Chang, S. Angew. Chem., Int. Ed. 2014,
53, 2203.
ASSOCIATED CONTENT
* Supporting Information
■
S
Experimental procedure and characterization of new com-
pounds (1H and 13C NMR spectra). This material is available
free of charge via the Internet at http://pubs.acs.org.
(11) (a) Umeda, N.; Tsurugi, H.; Satoh, T.; Miura, M. Angew. Chem.,
Int. Ed. 2008, 47, 4019. (b) Guimond, N.; Gouliaras, C.; Fagnou, K. J.
Am. Chem. Soc. 2010, 132, 6908. (c) Rakshit, S.; Grohmann, C.;
Besset, T.; Glorius, F. J. Am. Chem. Soc. 2011, 133, 2350. (d) Li, H.; Li,
Y.; Zhang, X.-S.; Chen, K.; Wang, X.; Shi, Z.-J. J. Am. Chem. Soc. 2011,
133, 15244. (e) Chan, W.-W.; Lo, S.-F.; Zhou, Z.; Yu, W.-Y. J. Am.
AUTHOR INFORMATION
Corresponding Author
■
Chem. Soc. 2012, 134, 13565. (f) Hyster, T. K.; Knorr, L.; Ward, T. R.;
̈
*E-mail: sbchang@kaist.ac.kr.
Notes
Rovis, T. Science 2012, 338, 500.
(12) White, C.; Thompson, S. J.; Maitlis, P. M. J. Chem. Soc., Dalton
Trans. 1977, 1654.
(13) An iridium catalyst system (Cp*IrIII, ref 10a) showed low
reactivity when 2-phenylpyridine was employed as a substrate (see the
Supporting Information.).
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
This research was supported by the Institute of Basic Science
(IBS).
■
(14) (a) Ackermann, L. Chem. Rev. 2011, 111, 1315. (b) Arockiam, P.
B.; Bruneau, C.; Dixneuf, P. H. Chem. Rev. 2012, 112, 5879.
(c) Ackermann, L. Acc. Chem. Res. 2014, 47, 281.
(15) See the Supporting Information for details.
REFERENCES
■
(16) For selected examples of the additive effects of carboxylates in
Ru-catalyzed C−H functionalizations, see: (a) Ackermann, L.; Vicente,
R.; Althammer, A. Org. Lett. 2008, 10, 2299. (b) Ackermann, L.;
Vicente, R.; Potukuchi, H. K.; Pirovano, V. Org. Lett. 2010, 12, 5032.
(c) Flegeau, E. F.; Bruneau, C.; Dixneuf, P. H.; Jutand, A. J. Am. Chem.
Soc. 2011, 133, 10161.
(17) Rhodacylic intermediates 5, 6 and ruthenacyclic complex 7 are
well characterized in the previous reports (see the Supporting
Information).
(1) (a) Ackermann, L.; Vicente, R.; Kapdi, A. R. Angew. Chem., Int.
Ed. 2009, 48, 9792. (b) Colby, D. A.; Bergman, R. G.; Ellman, J. A.
Chem. Rev. 2010, 110, 624. (c) Lyons, T. W.; Sanford, M. S. Chem.
Rev. 2010, 110, 1147. (d) Yeung, C. S.; Dong, V. M. Chem. Rev. 2011,
111, 1215. (e) Engle, K. M.; Mei, T.-S.; Wasa, M.; Yu, J.-Q. Acc. Chem.
Res. 2012, 45, 788. (f) Rouquet, G.; Chatani, N. Angew. Chem., Int. Ed.
2013, 52, 11726.
(2) (a) Jun, C.-H.; Moon, C. W.; Lee, D.-Y. Chem.Eur. J. 2002, 8,
2422. (b) Chen, X.; Engle, K. M.; Wang, D.-H.; Yu, J.-Q. Angew.
Chem., Int. Ed. 2009, 48, 5094. (c) Satoh, T.; Miura, M. Chem.Eur. J.
2010, 16, 11212. (d) Song, G.; Wang, F.; Li, X. Chem. Soc. Rev. 2012,
41, 3651.
(18) When cyclohexanecarbonyl azide was applied to either Rh- or
Ru-catalyzed amidation conditions, the reactions were sluggish only
leading to low conversions (<30%).
(3) For selected examples of C−H additions to C−N multiple bonds,
see: (a) Zhou, C.; Larock, R. C. J. Am. Chem. Soc. 2004, 126, 2302.
(b) Tsai, A. S.; Tauchert, M. E.; Bergman, R. G.; Ellman, J. A. J. Am.
Chem. Soc. 2011, 133, 1248. (c) Li, Y.; Li, B.-J.; Wang, W.-H.; Huang,
W.-P.; Zhang, X.-S.; Chen, K.; Shi, Z.-J. Angew. Chem., Int. Ed. 2011,
50, 2115. (d) Lian, Y.; Huber, T.; Hesp, K. D.; Bergman, R. G.;
2025
dx.doi.org/10.1021/ol500602b | Org. Lett. 2014, 16, 2022−2025