Mechanistic Studies on the Rh(III)-Mediated Amido Transfer Process Leading to Robust C-H Amination with a New Type of Amidating Reagent

Article abstract of DOI:10.1021/jacs.5b01324

Mechanistic investigations on the CpRh(III)-catalyzed direct C-H amination reaction led us to reveal the new utility of 1,4,2-dioxazol-5-one and its derivatives as highly efficient amino sources. Stepwise analysis on the C-N bond-forming process showed th

Full text of DOI:10.1021/jacs.5b01324

Article
pubs.acs.org/JACS
Mechanistic Studies on the Rh(III)-Mediated Amido Transfer Process
Leading to Robust C−H Amination with a New Type of Amidating
Reagent
Yoonsu Park,†,‡ Kyung Tae Park,†,‡ Jeung Gon Kim,*,‡,† and Sukbok Chang*,‡,†
†Department of Chemistry, Korea Advanced Institute of Science and Technology (KAIST), Daejeon, 305-701, Republic of Korea
‡Center for Catalytic Hydrocarbon Functionalizations, Institute for Basic Science (IBS), Daejeon, 305-701, Republic of Korea
S
* Supporting Information
ABSTRACT: Mechanistic investigations on the Cp*Rh(III)-
catalyzed direct C−H amination reaction led us to reveal the
new utility of 1,4,2-dioxazol-5-one and its derivatives as highly
efficient amino sources. Stepwise analysis on the C−N bond-
forming process showed that competitive binding of rhodium
metal center to amidating reagent or substrate is closely related
to the reaction efficiency. In this line, 1,4,2-dioxazol-5-ones
were observed to have a strong affinity to the cationic Rh(III)
giving rise to dramatically improved amidation efficiency when
compared to azides. Kinetics and computational studies
suggested that the high amidating reactivity of 1,4,2-dioxazol-
5-one can also be attributed to the low activation energy of an
imido-insertion process in addition to the high coordination
ability. While the characterization of a cationic Cp*Rh(III) complex bearing an amidating reagent was achieved, its facile
conversion to an amido-inserted rhodacycle allowed for a clear picture on the C−H amidation process. The newly developed
amidating reagent of 1,4,2-dioxazol-5-ones was applicable to a broad range of substrates with high functional group tolerance,
releasing carbon dioxide as a single byproduct. Additional attractive features of this amino source, such as they are more
convenient to prepare, store, and use when compared to the corresponding azides, take a step closer toward an ideal C−H
amination protocol.
en route for an ideal catalytic C−H amidation system: coupling
of unactivated hydrocarbons with aminating reagents which are
readily available and convenient to use under mild conditions
while producing minimal amounts of waste.3 We recently
answered to those demands by utilizing organic azides as the
unique amino source which also work as internal oxidants, thus
not necessitating any external oxidants enabling an environ-
mentally benign catalytic process.6,7 Our original development
and detailed mechanistic studies on the Cp*Rh(III)-catalyzed
direct C−H amination using azides6 have been a stepping stone
for the expansion to more efficient and selective catalytic
systems showing broader synthetic applications.7,8
In our continuing mechanistic investigations on the C−H
amination reactions, we were able to draw a more
comprehensive picture on the C−N bond-forming process.
Kinetics and coordination studies including the isolation of a
key intermediate revealed that a competitive binding of a metal
center to substrate or nitrogen source is directly related to the
reaction efficiency. This breakdown study allowed us to
introduce a robust catalytic system for the efficient C−H
INTRODUCTION
■
Construction of a carbon−nitrogen bond is one of the most
fundamental operations in nature and organic synthesis. The
resulting amino compounds are widely present in natural
products, pharmaceutical drugs, and functional materials. Thus,
the development of efficient C−N bond-forming methods has
been extensively explored.1 While conventional synthetic
approaches employ functionalized compounds such as alkenyl
or aryl (pseudo)halides to react with nitrogen sources,2 direct
catalytic amination on carbon−hydrogen bonds has drawn
significant attention based on a recent notable advance achieved
in the C−H activation procedures.3 With a better under-
standing of C−H activation process dealing with high activation
barriers and selectivity issues, the direct use of abundant
carbon−hydrogen connections would eliminate prefunctional-
ization steps.4 In this regard, transition-metal-mediated C−H
functionalization offers great promise to chemo-, regio-, and
stereoselective synthetic procedures operating under mild
conditions.
In case of C−H amination reactions, a seminal work by
Breslow et al. using iminoiodinane with Fe- and Rh-based
catalysts blazed a trail for the significant catalytic C−H
amination procedures.5 Currently, major research efforts are
Received: February 10, 2015
© XXXX American Chemical Society
A
DOI: 10.1021/jacs.5b01324
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Scheme 1. Development of Highly Reactive C−H Amidation Driven by Mechanistic Studies
Scheme 2. Mechanistic Proposal on the Rh(III)-Catalyzed C−H Amination Using an Organic Azides
amidation reactions using 1,4,2-dioxazol-5-ones as the con-
venient amino source (Scheme 1). This new catalytic system
allowed low catalyst loading, broad synthetic scope, and
external oxidant-free mild reaction conditions releasing CO2
as a single byproduct, altogether directing an ideal C−H
amination protocol, which is described herein as follows.
Rh(III)-catalyzed direct amidation of arene, alkene, and alkane
C−H bonds using organic azides as the nitrogen source.6,7
Extensive experimental and computational studies allowed us to
propose a preliminary mechanistic depiction (Scheme 2).
During our subsequent investigations, one interesting observa-
tion especially caught our attention: while each of stoichio-
metric C−N bond formation of a rhodacycle (1a) with benzoyl
azide (5a) (Scheme 3a) and protodemetalation of the
consequent amido rhodium complex (4a) by using 2-
phenylpyridine substrate (6) are facile at room temperature
RESULTS AND DISCUSSIONS
■
Identification of a Rate-limiting Factor in the C−H
Amidation using Azides. We previously reported the
B
DOI: 10.1021/jacs.5b01324
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Scheme 3. Analyses of Individual Stage
Figure 1. NMR study of C−N bond formation with a resting species
2a (see the Supporting Information for details).
(Scheme 3b), its catalytic conversion using a rhodacycle species
(1a) did not take place (Scheme 3c).
metal complex is present in the resting state 2. This result led
us to reason that one factor directly related to the reactivity is a
coordination competition between substrate and azide to a
metal center, and the facile generation of an azide-bound
rhodacycle 3 becomes crucial for the efficient C−H amination
(Scheme 4).
Finding a New Amino Source with Higher Affinity:
1,4,2-Dioxazol-5-one. We started to search for a different
type of amino source with strong coordination ability, thus
eventually leading to an efficient catalytic C−H amination
reaction. In this research program, one class of dioxazole
compounds especially caught our attention. In 1960s, Sauer and
Meyer showed that N-acyl nitrenes can be generated from
1,4,2-dioxazol-5-one, 1,4,2-dioxazol-5-thione, and 1,3,2,4-dioxa-
thiazole-2-oxide.10 Similar to acyl azides, those compounds
undergo thermal or photo-initiated decomposition leading to
N-acyl nitrenes (Scheme 5).10,11
We decided to initiate a detailed analysis on whole C−N
bond-forming processes to find a clue to this nonreactivity
under the catalytic conditions. We hypothesized that the C−H
amination pathway can be divided into four individual stages
(Scheme 2): (i) generation of a cationic rhodacycle 1 from the
resting state species 2; (ii) coordination of an azide to a metal
center to form an intermediate 3; (iii) insertion of an imido
moiety into the rhodium−carbon bond to form the C−N bond,
thus generating a metal−amido complex 4; and (iv)
protodemetalation with concomitant C−H bond activation of
a second substrate to release product 7 with the regeneration of
an active species 1. The successful conversion from a
rhodacycle 1a to a product 7a (Scheme 3a,b) led us to
postulate that the above presupposed three stages (ii−iv) are
facile under the applied conditions. Consequently, this
reasoning warranted a further study especially with regard to
the stage (i).
Ellman and co-workers observed the inhibitory effects of
substrates in the Cp*Rh(III)-catalyzed addition of 2-phenyl-
pyridines to imines.9 We also figured out the analogous
behavior of substrates in the Rh-catalyzed C−H amination
reactions.6 As was revealed by our previous mechanistic
studies,6 a substrate will occupy the vacant site of a cationic
rhodacycle 1 to generate a resting species 2, which accounts for
an inverse first-order kinetics on 2-phenylpyridine (Scheme 2).
Consistent with this reasoning, when 2a was allowed to react
with benzoyl azide 5a, the amidation proceeded almost
negligibly even after prolonged time (Figure 1). Again, this
reactivity of 2a is totally opposite when compared to a cationic
rhodacycle 1a having one vacant site that reacts promptly with
benzoyl azide 6a at room temperature (Scheme 3a).
Recently, Bolm and co-workers elegantly demonstrated the
synthetic utility of 1,4,2-dioxazol-5-ones for the synthesis of N-
acyl sulfimides and sulfoximides via a light-induced catalytic
method, in which a ruthenium N-acyl nitrene was proposed as a
key intermediate.12 Dube
́
also employed 1,4,2-dioxazol-5-ones
as a safe alternative to acyl azides for the synthesis of
isocyanates.13 On the basis of our previous development of the
Rh(III)-catalyzed C−H amination using organic azides,6 we
envisioned that dioxazoles could also be a good candidate as
another type of efficient nitrene sources. Considering the
observation by Ellman9 and Shi14 that imines displayed
comparable binding ability with 2-phenylpyridine, we antici-
pated that dioxazoles would have stronger affinity than acyl
azides, again contributing to the increase of amination
efficiency.
To validate our working hypothesis, three different types of
dioxazoles were examined along with benzoyl azide (Table 1).
Initial tests were conducted with a cationic Rh(III) catalytic
system generated in situ from [Cp*RhCl2]2 and AgNTf2.
Benzoyl azide (5a) and 3-phenyl-5,5-dimethyl-1,4,2-dioxazole
(8a) gave no detectable product at room temperature (entries
1−2). The latter, however, started to display reasonable
The observed poor conversion of 2a to a metal−amido
complex 4a implies that an organic azide is not capable of
binding effectively to a metal center in the presence of 2-
phenylpyridine. Under this circumstance, the effective concen-
tration of a rhodium complex entering into an operating
catalytic cycle can be regarded to be very low, while most of the
C
DOI: 10.1021/jacs.5b01324
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Scheme 4. Coordination Equilibrium of Cp*Rh(III) with 2-Phenylpyridine and Azide
−
−
replacement of a counteranion of NTf2 by SbF6 or the use
of a pregenerated cationic rhodium catalyst did not deteriorate
the catalytic efficiency (entries 8−9). Significantly, the product
yield was still well maintained at 40 °C even with 0.5 mol % of
catalyst loading (entries 10−12). In a stark contrast, other
catalytic systems such as [Ru(p-cymene)Cl2]2 and [Cp*IrCl2]2,
previously shown to be effective on the direct C−H amidation
in our studies,7 were not effective for the present amidation
using 3-phenyl-1,4,2-dioxazol-5-one (8c) as an amidating
reagent (entries 13−16).
Scheme 5. N-Acyl Nitrenes from Dioxazoles
Origin of High Reactivity. To understand the origin of
high reactivity of 1,4,2-dioxazol-5-ones, we conducted a series
of experimental and computational studies. At first, the
stoichiometric reactions of cationic Cp*Rh(III) complexes
(1a and 2a) with 3-phenyl-1,4,2-dioxazol-5-one (8c) were
examined (Scheme 6). Both of 1a and its resting species 2a
underwent instant conversion to the amido product 4a. We also
monitored the reaction of 1a with 8c using a time-resolved IR
spectrometer (Figure 2). A new peak appeared at 2342 cm−1
upon the addition of dioxazolone into a solution of 1a, and it
was assigned to be an asymmetric stretching CO bond of
reactivity upon heating (14% at 80 °C and 75% at 100 °C)
(entries 3−4). We were pleased to observe that while 3-phenyl-
1,4,2-dioxazol-5-thione (8b) exhibited satisfactory reactivity at
ambient temperatures (entries 5−6), an amidation with its
carbonyl analogue 3-phenyl-1,4,2-dioxazol-5-one (8c) was
quantitative even at room temperature (entry 7).
As a result, this amidating reagent 8c was chosen for further
optimization studies. It should be mentioned that 1,4,2-
dioxazol-5-ones can easily be prepared from the corresponding
hydroxamic acids, and they can be stable to store and
convenient to use without a special precaution.10−13
A
a
Table 1. Optimization of Direct C−H Amidation with Dioxazoles
b
entry
N source
catalyst (mol %)
[Cp*RhCl2]2 (5)
[Cp*RhCl2]2 (5)
[Cp*RhCl2]2 (5)
[Cp*RhCl2]2 (5)
[Cp*RhCl2]2 (5)
[Cp*RhCl2]2 (5)
[Cp*RhCl2]2 (5)
[Cp*RhCl2]2 (5)
[Cp*Rh(MeCN)3][SbF6]2 (10)
[Cp*RhCl2]2 (1)
[Cp*RhCl2]2 (1)
[Cp*RhCl2]2 (0.5)
[Ru(p-cymene)Cl2]2 (5)
[Ru(p-cymene)Cl2]2 (5)
[Cp*IrCl2]2 (5)
additive (mol %)
temp (°C)
yield (%)
1
2
5a
8a
8a
8a
8b
8b
8c
8c
8c
8c
8c
8c
8c
8c
8c
8c
AgNTf2 (20)
AgNTf2 (20)
AgNTf2 (20)
AgNTf2 (20)
AgNTf2 (20)
AgNTf2 (20)
AgNTf2 (20)
AgSbF6 (20)
−
AgNTf2 (4)
AgNTf2 (4)
AgNTf2 (2)
AgNTf2 (20)
AgNTf2 (20)
AgNTf2 (20)
AgNTf2 (20)
rt
rt
<1
<1
3
80
100
rt
14
4
74
5
75
6
40
rt
80
7
99 (97)
97
8
rt
9
rt
98
10
11
12
13
14
15
16
rt
80
40
40
rt
99
92
<1
50
rt
<1
4
[Cp*IrCl2]2 (5)
50
10
a
b
6 (0.20 mmol), nitrogen source 8 (0.22 mmol), catalyst, additive, and 1,2-dichloroethane (0.5 mL) at the indicated temperature. NMR yields
(internal standard: 1,1,2,2-tetrachloroethane, isolated yield in parentheses).
D
DOI: 10.1021/jacs.5b01324
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
a direct competition reaction (Scheme 7). A cationic rhodium
complex 1a (1.0 equiv) was allowed to react with an equal
mixture of dioxazolone 8c (2.0 equiv) and deuterium-labeled
d5-benzoyl azide 5a-d5 (2.0 equiv) with the anticipation of
tracing the origin of obtained product(s). After protonolysis,
1H NMR analysis of the isolated product clearly showed no
noticeable deuterium incorporation (7a:7a-d5, >19:1). This
result strongly suggests that the reactivity of 1,4,2-dioxazol-5-
one excelled acyl azide under identical conditions.
Scheme 6. Stoichiometric Reactions of 1,4,2-Dioxazol-5-one
with Cp*Rh(III) Complexes
For a better understanding of the reactivity difference among
amino sources, we conducted density functional theory (DFT)
calculations to obtain thermodynamic parameters including
Gibbs free energy for the equilibrium between a rhodacycle
complex 2 (resting species bound to a substrate) and 3
(replaced by an amino precursor) employing various nitrogen
sources (Table 2).17,18 It was seen that the replacement of 2-
Table 2. Calculated Coordination Equilibrium between 2-
Phenylpyridine and Nitrogen Sources
Figure 2. Time-resolved IR monitoring of a reaction of 1a with 8c.
released CO2 in accordance with the precedent reports.15,16
The observed fast conversion of a resting species 2a to an
amido-inserted rhodacycle 4a was in a stark contrast to the
reaction of 2a with benzoyl azide (5a) as shown in Figure 1,
where much slower conversion was monitored. This result
clearly showed that 1,4,2-dioxazol-5-one readily displaces
bound 2-phenylpyridine from the resting species 2a with
much higher efficiency than benzoyl azide.
Interestingly, as benzoyl azide can react with 1a to form 4a
via N2 release (Scheme 3a), the same product can also be
obtained from a reaction with 1,4,2-dioxazol-5-one upon release
of CO2. To measure the reactivity difference between
dioxazolone and azide in the reaction with 1a, we performed
phenylpyridine with acetyl azide was thermodynamically least
favorable among examined amino sources (entry 1). It was
calculated to require 10.1 kcal/mol for this process, which will
consequently yield low concentration of 3 where Y−NX is
acetyl azide (Kcalcd = 3.7 × 10−8). Change of amino precursors
from azide to dioxazolone or dioxathiazole resulted in
significantly higher coordination propensity by the magnitude
of 103 to 105 scales (entries 2 and 3, respectively), which is in a
good agreement with our working hypothesis. Interestingly, in
case of dioxazole (entry 4), the equilibrium lies favorably
toward the replaced complex 3 (ΔG° = −1.2 kcal/mol). Since
the subsequent migratory imido insertion of an isolated
Scheme 7. Competitive C−N Bond Formation between 1,4,2-Dioxazol-5-one and Acyl Azide
E
DOI: 10.1021/jacs.5b01324
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 3. Potential energy surfaces for the formation of Rh-nitrenoid from a resting species 2 with azide or 1,4,2-dioxazol-5-one (see Supporting
Information for details).
complex 3 (Y−NX: 3-phenyl-5,5-dimethyl-1,4,2-dioxazole)
did not occur at room temperature, we were able to obtain
experimentally the equilibrium constant with 3-phenyl-5,5-
dimethyl-1,4,2-dioxazole (8c) (Kexp = 6.3), which is within the
range of the calculated number (Kcald° = 8.1, entry 4).19
Extended calculations on the postulated subsequent metal-
nitrenoid formation were also performed using acyl azide and
1,4,2-dioxzol-5-one as the amino sources (Figure 3). It revealed
that ΔG⧧ in the formation of a rhodacyclic nitrene species (9)
by a reaction of 2 with 1,4,2-dioxzol-5-one (CO2 release) was
lower by 12.7 kcal/mol than that with acyl azide (N2 release):
20.3 and 33.0 kcal/mol, respectively. We reasoned this
difference may be reflected in the reaction efficiency observed
experimentally (Scheme 3 and Scheme 6). As a result, the high
reactivity of 1,4,2-dioxazol-5-one as an amino source can be
attributed to two synergistic effects: facilitation of the nitrenoid
formation as well as more favorable replacement of a substrate
in a resting species (shown in Table 2).
expected, an X-ray crystallographic analysis clearly showed the
coordination of a nitrogen atom to the rhodium center (Figure
4). The bond length of rhodium-dioxazole (Rh1−N2, 2.122 Å)
Figure 4. ORTEP of 3c with selected bond lengths and angles.
Thermal ellipsoid depicted at the 50% probability level. Hydrogen
atoms and SbF6 anion are omitted for clarity.
Isolation and Characterization of a Rhodacycle Bound
to an Amino Source (3). Whereas mechanistic details on the
C−H bond activation process have been well studied,
experimental investigations on the C−N bond-forming step
are still less explored mainly due to the highly labile nature of
plausible intermediates.20 In this line, while we recently cleared
the mist in the direct C−H amidation reactions utilizing
organic azides as the amino source,6b there are still some issues
remained unanswered, representatively the characterization of
key intermediates. In fact, while several catalytic systems have
been developed for the C−H amination, only a few examples
are known reporting the isolation or identification of nitrogen-
containing critical intermediates in the amido transfer
process.21,22 In our case of using azide precursors, the
immediate conversion of substrates to amido products under
the developed amination conditions hampered an intermediate
study.
is similar to that of rhodium-pyridyl nitrogen (Rh1−N1, 2.102
Å). Again, this result proved our working hypothesis that
dioxazole derivatives would have a strong binding affinity. More
pleasingly, this isolated species was observed to undergo an
imido insertion smoothly at 100 °C to furnish the
corresponding amido rhodacycle 4a in 54% after 2 h. In this
transformation, not surprisingly, acetone was observed to
discharge by a NMR analysis (Scheme 8).
Synthetic Scope. With the optimized amidation condition
in hand (Table 1 and also see the Supporting Information for
screening additional reaction parameters), the scope of
substrates was finally scrutinized (Scheme 9). The reaction
occurred smoothly under mild conditions at 40 °C within 12 h
by employing the dimeric rhodium catalyst in 1 mol % together
with 4 mol % of AgNTf2 in most cases. With 2-phenylpyridine
derivatives, regardless of electronic and/or steric variations,
excellent amidation efficiency was obtained (7a−g). Alteration
on either phenyl or pyridyl ring did not degrade the catalytic
activity. However, a substrate bearing an ortho-ester substituent
in phenyl side was ineffective under the present conditions
(7d). Functional group tolerance was found to be excellent, and
Hinted by the favorable equilibrium shift toward the
dioxazole-bound rhodacycle (Table 2, entry 4), we tried to
obtain the anticipated intermediate. We were delighted to
isolate indeed the desired intermediate 3c, an adduct of a
rhodacycle 1a and 3-phenyl-5,5-dimethyl-1,4,2-dioxazole 8a. As
F
DOI: 10.1021/jacs.5b01324
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Scheme 8. Stepwise C−N Bond Formation
a
Scheme 9. Substrate Scope of C−H Amidations
a
b
Substrate (0.20 mmol) and 7 (0.22 mmol) in 1,2-dichloroethane (0.5 mL), otherwise indicated: isolated yields. C−C amidation product (10%)
c
was formed. [Cp*RhCl2]2 (5.0 mol %) and AgNTf2 (20 mol %) were applied.
a wide range of potentially chelating groups such as ester (7bf
and 7c), ketone (7ec), and aldehyde (7be) did not affect the
reaction progress. The amidation of benzo[h]quinoline was
almost quantitative (7g) and that on a heterocycle was also
facile (7h and 7k). A variation on the amidating reagent, 1,4,2-
dioxazol-5-one, was subsequently examined. Whereas substitu-
ents at the phenyl moiety with electronically neutral or deficient
property did not affect the excellent amidation efficiency (7ia−
ic), a reaction with 1,4,2-dioxazol-5-one having 4-methox-
yphenyl substituent was sluggish (7id, 49%) with a
contaminant C−C amidated side product (10%).7g The
introduction of alkyl N-acyl amide group is of special interest
considering the fact that alkyl acyl azides, an alternative
amidating reagent for giving rise to the same products by our
previously developed procedures,7c are less convenient to
employ in that they are prone to rearrange into the
corresponding isocyanates.7g In addition, alkyl acyl azides of
low molecular weight often cause a safety issue.13,23 In this
regard, we envisioned that 1,4,2-dioxazol-5-one derivatives
could be viewed as a more convenient and safer alternative to
G
DOI: 10.1021/jacs.5b01324
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Notes
the acyl azides. We were pleased to observe that a number of
1,4,2-dioxazol-5-ones having alkyl substituents at the 3-position
were highly facile to afford alkyl amidated products in excellent
yields (7j−l). An additional array of modified directing groups
was also investigated to find out that amide (7m−n), ketoxime
(7o), and N-oxide (7p) worked well albeit under slightly more
demanding conditions. However, ketone, a weakly coordinating
directing group, was not effective (7q).
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank Dr. Sung-Woo Park, Dr. Jaesung Kwak, and Mr.
Kwangmin Shin (Institute for Basic Science, Korea) for
valuable discussions and Ms. Seohee Oh (KAIST) for XRD
analysis. Y.P. is a recipient of Kwanjeong Educational
Foundation scholarship. This research was supported by the
Institute for Basic Science (IBS-R010-D1) in Korea.
The present arene C−H amidation with 1,4,2-dioxazol-5-
ones was highly efficient under mild conditions to allow for a
facile performance even in a large scale (eq 1). For instance, the
REFERENCES
■
(1) (a) Amino Group Chemistry, From Synthesis to the Life Science;
Ricci, A., Ed.; Wiley-VCH; Weinheim, 2008. (b) Candeias, N. R.;
Branco, L. C.; Gois, P. M. P.; Afonso, C. A. M.; Trindade, A. F. Chem.
Rev. 2009, 109, 2703. (c) Hili, R.; Yudin, A. K. Nat. Chem. Biol. 2006,
2, 284.
(2) (a) Ullmann, F. Ber. Dtsch. Chem. Ges. 1903, 36, 2382.
(b) Monnier, F.; Taillefer, M. Angew. Chem., Int. Ed. 2009, 48, 6954.
(c) Evano, G.; Blanchard, N.; Toumi, M. Chem. Rev. 2008, 108, 3054.
(d) Paul, F.; Patt, J.; Hartwig, J. F. J. Am. Chem. Soc. 1994, 116, 5969.
(e) Hartwig, J. F. Acc. Chem. Res. 2008, 41, 1534. (f) Guram, A. S.;
Buchwald, S. L. J. Am. Chem. Soc. 1994, 116, 7901. (g) Surry, D. S.;
Buchwald, S. L. Angew. Chem., Int. Ed. 2008, 47, 6338.
(3) Selected reviews on C−H aminations: (a) Collet, F.; Dodd, R.
H.; Dauban, P. Chem. Commun. 2009, 5061. (b) Collet, F.; Lescot, C.;
Dauban, P. Chem. Soc. Rev. 2011, 40, 1926. (c) Jeffrey, J. L.; Sarpong,
R. Chem. Sci. 2013, 4, 4092. (d) Ramirez, T. A.; Zhao, B.; Shi, Y. Chem.
Soc. Rev. 2012, 41, 931. (e) Armstrong, A.; Collins, J. C. Angew. Chem.,
Int. Ed. 2010, 49, 2282. (f) Louillat, M.-L.; Patureau, F. W. Chem. Soc.
Rev. 2014, 43, 901. (g) Mahy, J.-P.; Ciesielski, J.; Dauban, P. Angew.
Chem., Int. Ed. 2014, 53, 6862.
(4) Selected reviews on C−H activations: (a) Labinger, J. A.; Bercaw,
J. E. Nature 2002, 417, 507. (b) Fagnou, K.; Lautens, M. Chem. Rev.
2003, 103, 169. (c) Bergman, R. G. Nature 2007, 446, 391.
(d) Seregin, I. V.; Gevorgyan, V. Chem. Soc. Rev. 2007, 36, 1173.
(e) Li, C.-J. Acc. Chem. Res. 2009, 42, 335. (f) Colby, D. A.; Bergman,
R. G.; Ellman, J. A. Chem. Rev. 2010, 110, 624. (g) Sun, C.-L.; Li, B.-J.;
Shi, Z.-J. Chem. Commun. 2010, 46, 677. (h) Daugulis, O. Top. Curr.
Chem. 2010, 292, 57. (i) Yeung, C. S.; Dong, V. M. Chem. Rev. 2011,
111, 1215. (j) McMurray, L.; O’Hara, F.; Gaunt, M. J. Chem. Soc. Rev.
2011, 40, 1885. (k) Cho, S. H.; Kim, J. Y.; Kwak, J.; Chang, S. Chem.
Soc. Rev. 2011, 40, 5068. (l) White, M. C. Science 2012, 335, 807.
(m) Arockiam, P. B.; Bruneau, C.; Dixneuf, P. H. Chem. Rev. 2012,
112, 5879. (n) Song, G.; Wang, F.; Li, X. Chem. Soc. Rev. 2012, 41,
3651. (o) Engle, K. M.; Mei, T.-S.; Wasa, M.; Yu, J.-Q. Acc. Chem. Res.
2012, 45, 788. (p) Hartwig, J. F. Acc. Chem. Res. 2012, 45, 864.
(q) Neufeldt, S. R.; Sanford, M. S. Acc. Chem. Res. 2012, 45, 936.
(r) Yamaguchi, J.; Yamaguchi, A. D.; Itami, K. Angew. Chem., Int. Ed.
2012, 51, 8960. (s) Kuhl, N.; Hopkinson, M. N.; Wencel-Delord, J.;
Glorius, F. Angew. Chem., Int. Ed. 2012, 51, 10236. (t) Rouquet, G.;
Chatani, N. Angew. Chem., Int. Ed. 2013, 52, 11726. (u) Pan, S.;
installation of acetamido and benzamido into 2-phenylpyridine
could be conducted in a gram scale with excellent product
yields (7a and 7j, respectively) with low catalyst loading (0.5
mol %).
CONCLUSIONS
■
We have described herein the development of a robust direct
C−H amidation using a new type of amidating reagent.
Stepwise analysis on the Cp*Rh(III)-catalyzed C−H amidation
showed that competitive binding of rhodium metal center to
amidating reagent or substrate is closely related to the reaction
efficiency. In this line, 1,4,2-dioxazol-5-ones with a strong
affinity could be employed as a new amidating reagent, thus
resulting in excellent amidation efficiency. Experimental and
computational works proved that the combination of the strong
coordination ability of 1,4,2-dioxazol-5-ones and low activation
energy of an imido insertion process is attributed to the origin
of high reactivity. In addition, isolation and characterization of a
cationic Cp*Rh(III) complex bearing an amino source were
realized. The understanding of coordination propensity of
metal center and nitrogen sources and the observation of its
turnover to an amido inserted rhodacycle allowed a clear
picture on the C−H amination process. The newly developed
Cp*Rh(III)-catalyzed reaction with 1,4,2-dioxazol-5-ones was
found to be highly efficient over a broad range of substrates
with high functional group tolerance releasing carbon dioxide as
a single byproduct. Additional attractive features of this amino
source such that they are more convenient to prepare, store,
and use when compared to the corresponding azides will make
a step closer toward an ideal C−H amination protocol.
Shibata, T. ACS Catal. 2013, 3, 704. (v) Kuhl, N.; Schroder, N.;
̈
Glorius, F. Adv. Synth. Catal. 2014, 356, 1443. (w) Davies, H. M. L.;
Morton, D. Chem. Soc. Rev. 2011, 40, 1857. (x) Newhouse, T.; Baran,
P. S. Angew. Chem., Int. Ed. 2011, 50, 3362.
ASSOCIATED CONTENT
* Supporting Information
(5) (a) Breslow, R.; Gellman, S. H. J. Chem. Soc., Chem. Comm. 1982,
1400. (b) Breslow, R.; Gellman, S. H. J. Am. Chem. Soc. 1983, 105,
6728. (c) Svastits, E. W.; Dawson, J. H.; Breslow, R.; Gellman, S. H. J.
Am. Chem. Soc. 1985, 107, 6427.
(6) (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) Park, S. H.; Kwak, J.; Shin,
K.; Ryu, J.; Park, Y.; Chang, S. J. Am. Chem. Soc. 2014, 136, 2492.
(7) (a) Ryu, J.; Shin, K.; Park, S. H.; Kim, J. Y.; Chang, S. Angew.
Chem., Int. Ed. 2012, 51, 9904. (b) Kim, J.; Kim, J.; Chang, S. Chem.−
Eur. J. 2013, 19, 7328. (c) Ryu, J.; Kwak, J.; Shin, K.; Lee, D.; Chang,
S. J. Am. Chem. Soc. 2013, 135, 12861. (d) Shin, K.; Baek, Y.; Chang, S.
Angew. Chem., Int. Ed. 2013, 52, 8031. (e) Kim, J.; Chang, S. Angew.
■
S
Detailed experimental procedure and characterization of new
compounds, Cartesian coordinates of computed structures, and
X-ray analysis. This material is available free of charge via the
Internet at http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Authors
*jeunggonkim@kaist.ac.kr
*sbchang@kaist.ac.kr
H
DOI: 10.1021/jacs.5b01324
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Chem., Int. Ed. 2014, 53, 2203. (f) Hwang, H.; Kim, J.; Jeong, J.;
Chang, S. J. Am. Chem. Soc. 2014, 136, 10770. (g) Shin, K.; Ryu, J.;
Chang, S. Org. Lett. 2014, 16, 2010. (h) Lee, D.; Kim, Y.; Chang, S. J.
Org. Chem. 2013, 78, 11102. (i) Kang, T.; Kim, Y.; Lee, D.; Wang, Z.;
Chang, S. J. Am. Chem. Soc. 2014, 136, 4141. (j) Kang, T.; Kim, H.;
Kim, J. G.; Chang, S. Chem. Commun. 2014, 50, 12073. (k) Shin, K.;
Chang, S. J. Org. Chem. 2014, 79, 12197. (l) Kim, H.; Park, J.; Kim, J.
G.; Chang, S. Org. Lett. 2014, 16, 5466. (m) Gwon, D.; Lee, D.; Kim,
J.; Park, S.; Chang, S. Chem.−Eur. J. 2014, 20, 12421. (n) Figg, T. M.;
Park, S.; Park, J.; Chang, S.; Musaev, D. G. Organometallics 2014, 33,
4076.
(8) (a) Hou, W.; Yang, Y.; Ai, W.; Wu, Y.; Wang, X.; Zhou, B.; Li, Y.
Eur. J. Org. Chem. 2015, 395. (b) Shin, Y.; Han, S.; De, U.; Park, J.;
Sharma, S.; Mishra, N. K.; Lee, E.-K.; Lee, Y.; Kim, H. S.; Kim, I. S. J.
Org. Chem. 2014, 79, 9262. (c) Pan, C.; Jin, N.; Zhang, H.; Han, J.;
Zhu, C. J. Org. Chem. 2014, 79, 9427. (d) Peng, J.; Xie, Z.; Chen, M.;
Wang, J.; Zhu, Q. Org. Lett. 2014, 16, 4702. (e) Zhou, X.; Luo, P.;
Long, L.; Ouyang, M.; Sang, X.; Ding, Q. Tetrahedron 2014, 70, 6742.
(f) Ryu, T.; Min, J.; Choi, W.; Jeon, W. H.; Lee, P. H. Org. Lett. 2014,
16, 2810. (g) Wang, N.; Li, R.; Li, L.; Xu, S.; Song, H.; Wang, B. J. Org.
Chem. 2014, 79, 5379. (h) Sun, B.; Yoshino, T.; Matsunaga, S.; Kanai,
M. Adv. Synth. Catal. 2014, 356, 1491. (i) Jia, X.; Han, J. J. Org. Chem.
2014, 79, 4180. (j) Pan, C.; Abdukader, A.; Han, J.; Cheng, Y.; Zhu, C.
Chem.−Eur. J. 2014, 20, 3606. (k) Yadav, M. R.; Rit, R. K.; Sahoo, A.
K. Org. Lett. 2013, 15, 1638. (l) Bhanuchandra, M.; Yadav, M. R.; Rit,
R. K.; Kuram, M. R.; Sahoo, A. K. Chem. Commun. 2013, 49, 5225.
(m) Zheng, Q.-Z.; Liang, Y.-F.; Qin, C.; Jiao, N. Chem. Commun. 2013,
49, 5654. (n) Yu, D.-G.; Suri, M.; Glorius, F. J. Am. Chem. Soc. 2013,
135, 8802.
(20) (a) Zhang, L.-L.; Li, L.-H.; Wang, Y.-Q.; Yang, Y.-F.; Liu, X.-Y.;
Liang, Y.-M. Organometallics 2014, 33, 1905. (b) Dick, A. R.; Remy, M.
S.; Kampf, J. W.; Sanford, M. S. Organometallics 2007, 26, 1365.
(c) Ke, Z.; Cundari, T. R. Organometallics 2010, 29, 821. (d) Sau, Y.-
K.; Yi, X.-Y.; Chan, K.-W.; Lai, C.-S.; Williams, I. D.; Leung, W.-H. J.
Organomet. Chem. 2010, 695, 1399.
(21) Kim, H.; Shin, K.; Chang, S. J. Am. Chem. Soc. 2014, 136, 5904.
(22) Turlington, C. R.; White, P. S.; Brookhart, M.; Templeton, J. L.
J. Am. Chem. Soc. 2014, 136, 3981.
(23) Azide of low molecular weight (more than 25% nitrogen
content) should be handled as a solution below 10%. Bretherick’s
Handbook of Reactive Chemical Hazards; Urben, P. G., Ed.;
Butterworth-Heinemann Ltd.: Oxford, 1999.
(9) (a) Tsai, A. S.; Tauchert, M. E.; Bergman, R. G.; Ellman, J. A. J.
Am. Chem. Soc. 2011, 133, 1248. (b) Tauchert, M. E.; Incarvito, C. D.;
Rheingold, A. L.; Bergman, R. G.; Ellman, J. A. J. Am. Chem. Soc. 2012,
134, 1482.
(10) (a) Sauer, J.; Mayer, K. K. Tetrahedron Lett. 1968, 9, 319.
(b) Sauer, J.; Mayer, K. K. Tetrahedron Lett. 1968, 9, 325. (c) Eibler,
E.; Sauer, J. Tetraheron Lett. 1974, 15, 2565. (d) Eibler, E.; Kasbauer,
̈
J.; Pohl, H.; Sauer, J. Tetrahedron Lett. 1987, 28, 1097.
(11) (a) Middleton, W. J. J. Org. Chem. 1983, 48, 3845. (b) Zhong,
C. L.; Tang, B. Y.; Yin, P.; Chen, Y.; He, L. J. Org. Chem. 2012, 77,
4271.
(12) (a) Bizet, V.; Buglioni, L.; Bolm, C. Angew. Chem., Int. Ed. 2014,
53, 5639. (b) Boglioni, L.; Bizet, V.; Bolm, C. Adv. Synth. Catal. 2014,
356, 2209.
(13) Dube, P.; Nathel, N. F. F.; Vetelino, M.; Couturier, M.;
́
Aboussafy, C. L.; Pichette, S.; Jorgensen, M. L.; Hardink, M. Org. Lett.
2009, 11, 5622.
(14) (a) 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. (b) Li, Y.;
Zhang, X.-S.; Li, H.; Wang, W.-H.; Chen, K.; Li, B.-J.; Shi, Z.-J. Chem.
Sci. 2012, 3, 1634.
(15) Stevens, R. W., Jr.; Chuang, S. S. C. J. Phys. Chem. B 2004, 108,
696.
(16) IR measurement of CO2, dissolved in 1,2-DCE, showed a peak
at 2342 cm−1; see Supporting Information for details.
(17) Selected examples of computed equilibrium constants analysis:
(a) Zhao, Y.-L.; Bartberger, M. D.; Goto, K.; Shimada, K.; Kawashima,
T.; Houk, K. N. J. Am. Chem. Soc. 2005, 127, 7964. (b) Schmidt, Y.;
Lam, J. K.; Pham, H. V.; Houk, K. N.; Vanderwal, C. D. J. Am. Chem.
Soc. 2013, 135, 7339. (c) Gom
Perez-Prior, M. T.; Calle, E.; Casado, J. J. Phys. Chem. A 2009, 113,
́
́ ́ ́
ez-Bombarelli, R.; Gonzalez-Perez, M.;
11423. (d) Guthrie, J. P.; Povar, I. J. Phys. Org. Chem. 2013, 26, 1077.
(18) Computational equilibrium constants for the given reaction can
be calculated by substituting the differences of standard Gibbs energy
differences for equilibrium constant−Free energy relationship, ΔG° =
−RTln K°.
(19) An experimental equilibrium constant was measured by mixing
2a and 3-phenyl-5,5-dimethyl-1,4,2-dioxazole with 1H NMR spec-
trometer; see the Supporting Information for details.
I
DOI: 10.1021/jacs.5b01324
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX