Why is the Ir(III)-Mediated Amido Transfer Much Faster Than the Rh(III)-Mediated Reaction? A Combined Experimental and Computational Study

Article abstract of DOI:10.1021/jacs.6b08211

The mechanism of the Ir(III)- and Rh(III)-mediated C-N coupling reaction, which is the key step for catalytic C-H amidation, was investigated in an integrated experimental and computational study. Novel amidating agents containing a 1,4,2-dioxazole moiety allowed for designing a stoichiometric version of the catalytic C-N coupling reaction and giving access to reaction intermediates that reveal details about each step of the reaction. Both DFT and kinetic studies strongly point to a mechanism where the M(III)-complex engages the amidating agent via oxidative coupling to form a M(V)-imido intermediate, which then undergoes migratory insertion to afford the final C-N coupled product. For the first time, the stoichiometric versions of the Ir- and Rh-mediated amidation reaction were compared systematically to each other. Iridium reacts much faster than rhodium (～1100 times at 6.7 °C) with the oxidative coupling being so fast that the activation of the initial Ir(III)-complex becomes rate-limiting. In the case of Rh, the Rh-imido formation step is rate-limiting. These qualitative differences stem from a unique bonding feature of the dioxazole moiety and the relativistic contraction of the Ir(V), which affords much more favorable energetics for the reaction. For the first time, a full molecular orbital analysis is presented to rationalize and explain the electronic features that govern this behavior.

Full text of DOI:10.1021/jacs.6b08211

Subscriber access provided by CORNELL UNIVERSITY LIBRARY
Article
Why is the Ir(III)-Mediated Amido Transfer Much Faster Than the Rh(III)-
Mediated Reaction? - A Combined Experimental and Computational Study
Yoonsu Park, Joon Heo, Mu-Hyun Baik, and Sukbok Chang
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.6b08211 • Publication Date (Web): 03 Oct 2016
Downloaded from http://pubs.acs.org on October 4, 2016
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of the American Chemical Society is published by the American Chemical
Society. 1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 11
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
Why is the Ir(III)-Mediated Amido Transfer Much Faster Than the
Rh(III)-Mediated Reaction? - A Combined Experimental and Com-
putational Study
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Yoonsu Park,†,‡ Joon Heo,† MuꢀHyun Baik,*,‡,† and Sukbok Chang*,‡,†
†Department of Chemistry, Korea Advanced Institute of Science and Technology (KAIST), Daejeon, 34141, Republic of Korea
‡Center for Catalytic Hydrocarbon Functionalizations, Institute for Basic Science (IBS), Daejeon, 34141, Republic of Korea
Supporting Information Placeholder
ABSTRACT: The mechanism of the Ir(III) and Rh(III)ꢀmediated C–N coupling reaction, which is the key step of catalytic C−H
amidation, was investigated in an integrated experimental and computational study. Novel amidating agents containing a 1,4,2ꢀ
dioxazole moiety allowed for designing a stoichiometric version of the catalytic C–N coupling reaction and giving access to reacꢀ
tion intermediates that reveal details about each step of the reaction. Both DFT and kinetic studies strongly point to a mechanism
where the M(III) complex engages the amidating agent via oxidative coupling to form a M(V)−imido intermediate, which then unꢀ
dergoes migratory insertion to afford the final C–N coupled product. For the first time, the stoichiometric versions of the Ir and Rhꢀ
mediated amidation reaction were compared systematically to each other. Iridium reacts much faster than rhodium (~ 1100 times at
6.7 oC) with the oxidative coupling being so fast that the activation of the initial Ir(III)ꢀcomplex becomes rateꢀlimiting. In the case
of Rh, the Rh−imido formation step is rateꢀlimiting. These qualitative difference stems from a unique bonding feature of the dioxaꢀ
zole moiety and the relativistic contraction of the Ir(V), which affords much more favorable energetics for the reaction. For the first
time, a full molecular orbital analysis is presented to rationalize and explain the electronic features that govern this behavior.
possible to qualitatively predict how an Ir−nitrenoid will beꢀ
have differently from a Rh−nitrenoid.
Introduction
Catalytic amination of C–H bonds is an efficient and diꢀ
rect method for accessing molecular fragments that contain C–
N bonds, which play a pivotal role in functional materials,
natural products and pharmaceutical agents.1 In general, two
distinct mechanisms are widely accepted.2 The first, often
referred to as outer-sphere C−H amination, is illustrated on the
leftꢀhand side in Scheme 1 and assumes that a metal−nitrenoid
complex A is formed by oxidative coupling from an appropriꢀ
ate precursor, such as iminoiodinane or organic azides.3 Interꢀ
or intraꢀmolecular insertion of the nitrene functionality into the
C–H bond may afford the aminated product.4 Although this
relatively simple mechanism is plausible and is accepted wideꢀ
ly, recent work provided strong evidence for a more compliꢀ
cated involvement of a highꢀvalent metal−nitrenoid complex
in the catalytic process. For example, Che and Gallo reported
the characterization of one such intermediate, namely
bis(imido)ruthenium(VI) porphyrine.5 In another study, a putaꢀ
tive Rh−nitrenoid intermediate was detected using mass specꢀ
trometric techniques.6 Furthermore, an iron(III)−imidyl comꢀ
plex capable of catalyzing C–H bond cleavage was recently
reported.7 These examples illustrate that metal−nitrenoids relꢀ
evant to catalytic aminations are readily accessible with a vaꢀ
riety of transition metals and ligand systems. Yet, there mechꢀ
anistic roles remain poorly understood and it is currently imꢀ
Scheme 1. General mechanism of catalytic C−H amination.
The catalytic systems based on the outer-sphere pathway
are known to work more effectively with benzylic or allylic
C(sp3)–H bonds, whereas unactivated primary or sp2ꢀ
hybridized C–H bonds proved difficult to functionalize.5c,8
A
promising strategy to force these strong C–H bonds to react in
a selective manner is to utilize a chelating group to direct the
C–H bond cleavage, affording a metallacycle, as illustrated on
the rightꢀhand side of Scheme 1.9 The resulting metallacyclic
intermediate may react with a nitrogen source to form the C–N
1
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 11
1
2
3
4
5
bond. Over the last decade, a few catalysts based on late tranꢀ
Scheme 2. Proposed M−nitrenoids for inner-sphere C−H amiꢀ
nation reactions.
sition metals were discovered and proposed to function folꢀ
lowing this mechanism.10 Because both components of the
reaction are tightly bound to the metal, this mechanism is reꢀ
ferred to as inner-sphere C–H amination.
6
Whereas the outer-sphere amination mechanism is reaꢀ
sonably well established, the inner-sphere pathway to C–N
coupling remains under debate. Specifically, relevant key inꢀ
termediates, such as complex B in Scheme 1, could not be
detected thus far presumably owing to the highly reactive naꢀ
ture of these putative intermediates.11 As a result, most of the
mechanistic proposals are based on theoretical studies. Previꢀ
ously, Che and Sanford proposed a Pd−nitrenoid complex as a
key intermediate for the ligandꢀdirected C−H amination, as
shown in Scheme 2a.12 Ke and Cundari examined such an
intermediate by DFT calculations and concluded that a singlet
Pd(IV)−imido complex is a plausible intermediate.12e We proꢀ
posed previously that Cp*Rh(V)− and Cp*Ir(V)−nitrenoid
complexes are key intermediates during catalytic C−H aminaꢀ
tion using organic azides, as described in Scheme 2b.13,14 Sevꢀ
eral theoretical studies by us15 and others16 suggested that the
formations of highꢀvalent Cp*M(V)−imido complexes are
feasible. Regrettably, direct experimental evidence for the
putative metal−nitrenoid complex, such as characterization in
a kinetic experiment, spectroscopic detection or isolation reꢀ
mained elusive to date. An isolobal Ru(II) complex was reꢀ
ported to promote olefin aziridination under catalytic amiꢀ
dation conditions,17 suggesting that a nitrene is formed during
the reaction, but direct examination of such metal−nitrenoid
intermediate was again not possible.
Recently, we found that the catalytic C−H amidation can
be improved significantly by introducing 1,4,2ꢀdioxazole deꢀ
rivatives as a novel class of aminating agents.18 Compared to
organic azides, the 1,4,2ꢀdioxazoles allow for much milder
reaction conditions and they are easier to handle owing to inꢀ
creased thermal stability. During the course of our mechanistic
investigation we realized that the origin of the increased effiꢀ
ciency lies in much faster C−N formation. We found that the
dioxazoles give amido transfer reactions that are both thermoꢀ
dynamically and kinetically more favorable. Motivated by
these observations, we envisioned that the kinetics with 1,4,2ꢀ
dioxazolꢀ5ꢀone may provide valuable information on the
mechanism of the C−N bond forming step.
Described herein is a fully integrated experimental and
computational study aimed at quantitatively assessing whether
the stepwise formation of metal−nitrenoid complex, as deꢀ
scribed in Scheme 2c, is plausible. Surprising differences in
the amidation kinetics are proved to be present between with
Rhꢀ and Irꢀbased organometallic complexes. This work constiꢀ
tutes the first inꢀdepth study that explains both quantitatively
and qualitatively how and why iridium is so much more comꢀ
petent in promoting C–N bond coupling. This new level of
insight was possible because we were able to make a stoichiꢀ
ometric amidation model of the catalytic reaction exploiting
the unique features of the new amidating agent, which was not
found previously with traditional nitrogen sources, such as
organic azides. Moreover, frontier orbital analyses provided
fundamental general insights for the design of efficient groupꢀ
transfer agents when the oxidation of metal center is involved
in the reaction.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Results and Discussions
Whereas organic azides were successfully used as aminatꢀ
ing agents within the inner-sphere C–H amination strategy,
these reactions typically require variable reaction temperatures
making them less desirable in synthesis. The thermally instaꢀ
ble nature of the azides requires special caution during prepaꢀ
ration and handling of these reagents. We recently found that
1,4,2ꢀdioxazole derivatives19 can serve as an efficient Nꢀ
sources in the Cp*Rh(III)ꢀcatalyzed C–H amidation reaction,
as highlighted in eq 1.18a Importantly, the aforementioned
problems associated with the organic azides can be avoided by
using dioxazoles as the amino precursors. Because dioxazoles
react under much milder conditions than organic azides, they
offer a more efficient and safer route to nitrogenꢀcontaining
molecules. The potential relevance of this new methodology in
large scale production regarding the safety, sustainability and
scalability was demonstrated by successfully carrying out a
decagramꢀscale reaction.18b Additional studies by Li,20 Jiao,21
Ackermann,22 Glorius23 and others24 proved that this novel
amidating agents are also effective for isolobal d6 catalytic
systems containing Ir(III), Co(III) and Ru(II) likely following
2
ACS Paragon Plus Environment

Page 3 of 11
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
useful.12b,13a,17,25 We sought to design a model system26 that
Scheme 3. Mechanism of Cp*Rh(III)ꢀcatalyzed C−H amiꢀ
dation.
may capture the key component of the catalytic C–H amiꢀ
dation. Cyclometalated Cp*Rh(III) and Ir(III) complexes with
2ꢀphenylpyridine substrate (1-Rh and 1-Ir in Scheme 5) were
chosen as model systems, since both systems are found to
enable the C–N bond formation in a reaction with 1,4,2ꢀ
dioxazole derivatives in our preliminary study. A benzonitrile
carrying a 3,5ꢀbis(trifluoromethyl) substituents (ArFCN) was
used as an innocent ligand to block an empty coordination site
for mimicking the catalytic process in this stoichiometric reacꢀ
tion.18a
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 4. Putative reaction pathways for C−N formation.
similar mechanistic reaction pathways.
Elementary steps of the catalytic C–H amidation under
this system are illustrated in Scheme 3. The metalation of subꢀ
strate, 2ꢀphenylpyridine (ppy) as a model, via C–H activation
gives the rhodacycle C at the formally Rh(III)ꢀd6 center in the
pseudoꢀoctahedral ligand environment provided by the Cp*
and ppy ligands. The coordinating ppy can be exchanged by
the aminating reagent dioxazole to give an intermediate comꢀ
plex D. As will be explained in greater detail below, both
complexes C and D coexist in equilibrium within the reaction
mixture. Intermediate D may undergo C–N coupling accomꢀ
panied by extrusion of the leaving group. The resulting
Rh−amido complex E may subsequently react with another
substrate via a concertedꢀmetalation deprotonation (CMD)
mechanism to regenerate complex C affording the desired
product. A few acyl nitrene precursors were examined as alꢀ
ternative amidating agents and 1,4,2ꢀdioxazolꢀ5ꢀones were
identified as being particularly reactive, leading to room temꢀ
perature reactions.
Despite having identified these elementary steps of the
catalytic cycle, the key C–N bond forming process remains
poorly understood. The proposed mechanism of the amido
transfer is depicted in Scheme 4. Coordination of the dioxazoꢀ
lone affords the Nꢀbound, pseudoꢀoctahedral 18ꢀelectron comꢀ
plex II, from which the C–N bond can be formed in two funꢀ
damentally different ways: Path a invokes a concerted SN2ꢀ
type attack of the metal–carbon bond and synchronous cleavꢀ
age of CO2. In this case, the formal oxidation state of the metal
remains unchanged throughout the reaction. On the other hand,
in Path b the metal−nitrogen multiple bond is formed in a
stepwise fashion, to give a M(V)–imido species III. Subseꢀ
quent migratory insertion of the imido group into metꢀ
al−carbon bond produces the C–N coupled product, the amido
complex IV. The Rh(V)–imido intermediate was suggested by
previous computational studies using tosyl azide as the amiꢀ
dating agent15b but these findings have not been corroborated
by experiments. Attempts at isolating the M−nitrenoid interꢀ
mediate have thus far not been successful and we were also
unable to obtain spectroscopic evidence that would either supꢀ
port or disprove the computationally identified mechanistic
proposal. In part, this lack of experimental support is due to
the highly reactive nature of the key intermediates that continꢀ
ue to evade detection. In lieu of direct evidence within the
catalytic reaction, the stoichiometric conversion of metallacyꢀ
clic intermediates toward analogous nitrene precursors may be
Comparison of the Mechanisms. We sought to comꢀ
pare the mechanisms promoted by the rhodium and iridium
complexes by carefully constructing the energy landscapes of
the stoichiometric C–N coupling reactions using density funcꢀ
tional theory, as shown in Figure 1.27 Not surprisingly, both
metals display similar reactivity patterns in general, but signifꢀ
icant and surprising differences are also found. Before the
metal can mediate the amidation reaction, the preꢀcoordinated
benzonitrile ligand ArFCN must be exchanged with the amiꢀ
dating agent, as illustrated in Scheme 4. In principle, such
ligand exchange may occur following an associative or a disꢀ
sociative mechanism. It is difficult to predict a priori which
displacement pathway will be followed because late transition
metal complexes bearing Cp* ligands are known to engage
both in associative and dissociative ligand exchange mechaꢀ
nisms.28
Our computer models revealed that the dissociative pathꢀ
way is much preferred and we were unable to locate any reaꢀ
sonable intermediate with either of the metals carrying both
the nitrile and the dioxazole ligands that would be expected in
an associative process. The calculations showed that the nitrile
ligand can dissociate in a unimolecular fashion with a solution
phase free energy barrier of 12.8 and 15.5 kcal/mol, respecꢀ
tively, for Rh and Ir. As will be discussed below, this seemingꢀ
ly simple step becomes mechanistically important for the iridꢀ
iumꢀmediated reaction and, therefore, we chose to interrogate
it in greater detail.
To investigate this initial step without interference by the
subsequent steps of the C–N coupling reaction, 5,5ꢀdimethylꢀ
3ꢀphenylꢀ1,4,2ꢀdioxazole (2a) was chosen as the amidating
agent as summarized in Scheme 5. 2a is particularly well suitꢀ
ed for this purpose, as it is capable of acting as an amidating
agent, but requires an elevated reaction temperature for the
amidation. It allows for observing how it binds to the metal
without the reaction
3
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 11
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. DFTꢀcalculated reaction energy profiles for (a) rhodium and (b) iridium.27 Transition states marked * were not located.
Scheme 5. Stoichiometric ligand exchange with 2a.
as a consequence, forming more covalent, stronger and thereꢀ
fore shorter Rh–C bond.
Another approach to describe the distinct interactions of
Ir and Rh with ligands is using the hard/softꢀacid/base (HSAB)
concept: due to the relativistic contraction of Ir, Ir(III) comꢀ
plexes are expected to be much harder Lewis acids than the
corresponding Rh(III) counterparts. Thus, Ir will interact more
strongly with the hard nitrogenꢀbased Lewis bases, whereas
the softer carbonꢀbased phenylꢀanion ligand will prefer the
softer Rhꢀcenter. These trends are reliably reproduced in our
computational models, as enumerated in Table 1, confirming
that these bond lengths are electronically invoked and are not
based on environmental effects like crystal packing.
progressing towards C–N bond formation, if we keep the reacꢀ
tion temperature low.18a The higher activation temperature is
needed because the chemical driving force is lower when 2a is
used: instead of CO2, acetone is liberated during the amido
formation. When metallacycles 1-Rh and 1-Ir were exposed
to 2a, the nitrile ligand ArFCN was liberated and the metꢀ
al−dioxazole adducts 3-Rh or 3-Ir were formed.
To determine the ligand displacement rate in the forꢀ
mation of dioxazoleꢀbound metallacycles, the appearance of
the 1H NMR signatures of 3-Ir or 3-Rh were monitored below
o
ꢀ35 C in the presence of 10 equivalent of 2a. Reaction temꢀ
perature was calibrated internally with methanolꢀd4 prior to the
measurement, as described in detail in the Supporting Inforꢀ
mation. The pseudoꢀfirst order rate constants were obtained by
taking the average of three independent experiments and they
showed that the appearance of 3-Rh is much faster than that of
3-Ir, as summarized in Table S1. At ꢀ38.7 oC, the rate constant
for the appearance of 3-Ir was 4.27 ± 0.58 x 10ꢀ3 secꢀ1, whereꢀ
as the reaction with 3-Rh complex was too fast to be measured
reliably at this temperature. Therefore, the rate constant for the
formation of 3-Rh (4.22 ± 0.65 x 10ꢀ3 secꢀ1) was determined at
ꢀ50.0 oC. This finding is in excellent agreement with the DFTꢀ
calculated barriers for this process. As highlighted in Figure 1,
the transition state for nitrile dissociation for i-Rh-TS was
located at 12.8 kcal/mol, whereas 15.5 kcal/mol was found for
i-Ir-TS. As often encountered in theoretical studies, we cannot
reliably estimate the reaction rate from simple quantum chemꢀ
ical transition state calculations, as the preꢀexponential colliꢀ
sion factor in the Arrhenius equation cannot be obtained in this
manner. But, the transition state energy difference of nearly 3
kcal/mol should translate into faster reaction for the Rhꢀ
complex, as is seen. Consistent with this mechanistic scenario
is that increasing the concentration of 2a up to 20 equivalents
showed no appreciable change in the rate.
A single crystal of 3-Ir was obtained by slow diffusion of
petroleum ether into a saturated dichloromethane solution at ꢀ
20 C and the composition of 3-Ir was unambiguously deterꢀ
o
mined by Xꢀray diffraction studies. The fact that we are able to
isolate these key intermediates is a testament of an additional
benefit of the dioxazole system. Previously, it was difficult to
isolate the analogous intermediates when other aminating
agents, such as organic azides, were utilized.15b,29 An ORTEP
drawing of the structure of 3-Ir is shown in Figure 2. The
structure of the analogous rhodium complex 3-Rh was deterꢀ
mined in our previous work18a and displays very similar feaꢀ
tures as seen in 3-Ir. The most salient bond lengths found in
these two crystal structures are compared in Table 1 and S7. It
was predicted that the strong relativistic contraction of the
third row metal iridium gives rise to a smaller ionic radius
when compared to the second row metal rhodium, which in
turn affords shorter dative bonds in the iridium complex.30 The
Ir–N(ppy) and Ir–N(2a) bonds are 2.090 and 2.095 Å, respecꢀ
tively, which are 0.012 and 0.027 Å shorter than what is seen
in the Rhꢀcomplex. Interestingly, the Ir–C(ppy) bond is 0.048
Å longer than the Rh–C(ppy) bond, which may be a reflection
on the Rhꢀorbitals being closer in energy to the Cꢀorbitals and,
4
ACS Paragon Plus Environment

Page 5 of 11
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
Rhꢀcomplex.16c Once the M−nitrenoids iv-Rh/Ir are formed,
the C–N bond coupling proceeds with little difficulty to afford
the final product. These mechanistic differences are somewhat
surprising and demand additional supporting evidence and
ultimately an intuitive explanation. It should be noted that
recent computational studies by Xia identified that similar
Cp*Rh(V)−nitrenoid complexes are key intermediates catalyzꢀ
ing C−H functionalizations when oxidizing directing groups
are employed.16a,b Houk and Wu also found that the pathway
involving the formation of Cp*Rh(V)−nitrenoid is more kinetꢀ
ically favored than the nonꢀnitrenoid pathway.16d
As indicated in Figure 1, we have explicitly considered an
alternative reaction pathway. After association of dioxazolone
to form intermediate iii-Rh or iii-Ir, the C–N bond may be
formed following a SN2ꢀtype mechanism, as outlined in
Scheme 4 as Path a. The transition state for the formation of
C–N bond and synchronous cleavage of CO2 was found to be
much higher at 30.8 (iii-Rh-TS2) or 34.0 kcal/mol (iii-Ir-TS2)
than the asynchronous pathway described above. Thus, this
alternative mechanism can be excluded from further considerꢀ
ations.
9
10
11
12
13
14
15
Figure 2. The XRD structure of 3-Ir. 50% probability.
Table 1. Selected bond lengths of 3-Ir and 3-Rh18a in Å.
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
M−N(ppy)
M−N(2a)
M−C(ppy)
Exp.
DFT
Exp.
DFT
Exp. DFT
2.090
2.102
2.127
2.095
2.141 2.080 2.057
2.185 2.032 2.039
3-Ir
3-Rh
ꢁ
Scheme 6. Stoichiometric amidation with 2b.
2.144
2.122
ꢀ0.012 ꢀ0.017 ꢀ0.027 ꢀ0.044 0.048 0.018
O
-
BArF
-
4
BArF
4
O
N
M
NCArF
M
N
O
NCArF
All these observations suggest strongly that the ligand exꢀ
change follows a dissociative mechanism. The observation
that Ir is much more reluctant to eliminate the nitrile than Rh
is easy to understand considering the relativistic contraction
and the resulting increase in Lewis acidity that in turn leads to
stronger dative bonding interactions, as mentioned above. Just
as was the case in 3-Ir, the dative bond between Ir and the
nitrile ligand in i-Ir was stronger and shorter at 2.045 Å than
in the rhodium analogue i-Rh, where 2.120 Å is found in our
calculations.
+
N
ArF
C(O)ArF
N
CD2Cl2
-CO2
1-Rh
M= Irh: 1-Ir
M= Rh:
2b (10 equiv.)
4-Rh or 4-Ir
ArF = 3,5-(CF3)2C6H3
Amidation with 1,4,2-dioxazol-5-one. Having established
a firm understanding of the ligand exchange reaction in the
precursor complex, the overall amidation reactions were invesꢀ
tigated with the most efficient amidating agent, 1,4,2ꢀ
dioxazolꢀ5ꢀone (2b). Our DFTꢀcalculations summarized in
Figure 1 indicate a significant difference in the abilities of the
two metals in promoting an oxidative M=N coupling with
concurrent extrusion of carbon dioxide, where the formal oxiꢀ
dation states of the Rh/Ir metal centers are changed from +III
to +V. Whereas the Irꢀcomplex iii-Ir is able to accomplish this
step with ease traversing the transition state iii-Ir-TS1 that lies
only 9.2 kcal/mol higher in energy, the same transformation is
much more difficult when rhodium is used. The energy differꢀ
ence between iii-Rh and iii-Rh-TS1 is 22.8 kcal/mol, which is
by far the highest barrier of all reaction steps and suggests that
the M−nitrenoid formation accompanied by carbon dioxide
extrusion may be rate determining when Rh is used. Ir is much
more reactive and the most difficult step becomes the initial
loss of the nitrile ligand with a computed barrier of 15.5
kcal/mol. Interestingly, in previous DFT studies using the aryl
azides as the aminating agents, such superior reactivity of Ir
compared with Rh was not found for the analogous oxidation
step.15a Another study with tosyl azide also expected that Irꢀ
complex would have marginally lower activation barrier than
Figure 3. Amidation profile at ꢀ47.7 oC; black triangle: [1-Ir],
blue circle: [4-Ir], red square: [4-Rh].; data were fit (R2 =
0.999) to firstꢀorder exponential decay.
In search of corroborating evidence for the proposal that
the rateꢀlimiting steps are different depending on the metal
used, we carried out a series of experiments: 1,4,2ꢀdioxazolꢀ5ꢀ
one (2b) was added to a chilled solution of 1-Rh or 1-Ir,
which converted both complexes quantitatively to the correꢀ
sponding metal−amido complexes 4-Rh and 4-Ir with the
extrusion of stoichiometric amounts of carbon dioxide, as ilꢀ
lustrated in Scheme 6. The formations of metal−amido comꢀ
plexes were confirmed using NMR techniques by comparing
with reported spectra26e,31 and the concentrations of the reacꢀ
tants and products were monitored, as shown in Figure 3.
5
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 11
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 4. Eyring plots for (a) rhodium and (b) iridium.
Figure 5. Reaction rate constants for (a) rhodium and (b) iridium as a function of [2b].
Table 2. Summary of activation parameters.
to be 20.0 and 15.8 kcal/mol for 1-Rh and 1-Ir, respectively,
as shown in Table 2. These experimental numbers are in good
k (sꢀ1)
ꢁH‡
ꢁS‡
(e.u.)
ꢁG298
‡
agreement with the computed values of 22.8 and 15.5 kcal/mol,
mentioned above. The decay constants provide a clear comꢀ
parison between the Rhꢀ and Irꢀbased amidation reactions: as
shown in Table 2 and S5, the iridium complex at ꢀ47.7 oC and
6.7 oC was ~400 and ~1100 times faster in producing the C–N
coupled product than rhodium, respectively. The good agreeꢀ
ment of the barriers between these experimental findings and
our DFTꢀcalculated results were encouraging, but we sought
to find additional support for the conclusion that the nitrile
ligand extrusion is rate limiting in the Irꢀmediated reaction,
while the formation of a metal−imido species by releasing CO2
is most difficult when Rh is used for the same reaction. It is
noteworthy that the SN2ꢀtype mechanism via iii-Rh-TS2 or iii-
Ir-TS2 is not in accord with the experimentally determined
activation parameters and does not reproduce the superior
reactivity of iridium in the amidation.
Inspecting the components of the activation free energies
in greater detail, we note that the sign of the activation entroꢀ
pies (ꢁS‡) are different. In the case of the iridium complex, we
found a ꢁS‡ of +5.2 e.u., but in the case of rhodium the entroꢀ
py of activation is ꢀ14.6 e.u., indicating that the nature of rateꢀ
limiting transition state is fundamentally different, consistent
with what we concluded based on our DFTꢀcalculations. The
magnitude of the experimentally determined entropies of actiꢀ
vation suggest that only very little free particle character is
present at the respective transition states. The cleavage of the
6.7 oC
(kcal molꢀ1)
(kcal molꢀ1)
20.0 ±0.7
15.8 ±1.0
Rh
Ir
2.08 ×10ꢀ3
2.2a
15.7 ±0.5
17.4 ±0.6
ꢀ14.6 ±1.7
5.2 ±2.7
a The value was extrapolated from Eyring plot.
In the presence of 10 equivalents of 2b, the reactions showed
pseudo firstꢀorder kinetics behavior over a time period correꢀ
sponding to at least 3 times the estimated halfꢀlifetime of the
reactant species. Interestingly, the reaction rates were dramatiꢀ
cally different for the two metal systems. At ꢀ47.7 oC, the firstꢀ
order decay constant of 1-Ir to 4-Ir was determined to be 8.63
x 10ꢀ4 secꢀ1 (black triangle and blue circle in Figure 3), whereꢀ
as no formation was observable for 4-Rh within a reaction
time window of 3 hours (red square in Figure 3). Only when
the temperature was raised significantly were we able to detect
o
the amidation product and at 6.7 C the rate constant was
measured to be 2.08 x 10ꢀ3 secꢀ1.
To extract the enthalpy and entropy components of the
free energy of activation, the rate of the reaction was obtained
in variable temperature experiments and the results were used
to construct the Eyring plots, shown in Figure 4. Linear reꢀ
gressions gave excellent correlation in both cases and we were
able to extract the activation free energies at 298.15 K (ꢁG298
‡
)
6
ACS Paragon Plus Environment

Page 7 of 11
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Ir–NCArF bond is a simple bond dissociation event with very
oxidative M–N coupling with the M(III) center. Detailed disꢀ
cussion on the electronic reorganization is placed on the Supꢀ
porting Information.
little electronic rearrangement and thus, it is plausible that the
entropy of activation, +5.2 e.u., is slightly positive due to
some of the translational entropy gain materializing at the
transition state, while the Ir–NCArF bond is weakened. The
carbon dioxide release step is accompanied by a significantly
more pronounced electronic reorganization. Concomitant to
the carbon dioxide being released, the Rhꢀcenter is oxidized
and the Rh–nitrenoid double bond is formed formally, which
leads to a much stronger Lewis acidic Rh(V) center. This inꢀ
crease in formal oxidation state will naturally lead to stronger
interactions of the metal center with all ligands present in the
molecule, which in turn will decrease the number of accessible,
lowꢀenergy vibronic quantum states, as they shift to higher
energies. Such a reduction of accessible vibrational miꢀ
crostates is expected to lead to a decrease in the entropy. Thus,
the experimentally observed entropy of activation of ꢀ14.6 e.u.
for the Rhꢀsystem is fully consistent with the transition state
being that of Rh−nitrenoid formation.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 6. Conceptual MOꢀdiagram showing the electronic
structure change for the M−nitrenoid formation.
Examining the reaction rates at various temperatures careꢀ
fully, it is interesting that the amidation rate of 1-Ir with 2b
was essentially identical with the nitrile ligand exchange rate
with 2a. As listed in Table S1 and S2, the rate constant disꢀ
Inspecting the reaction energy profiles shown in Figure 1
in greater detail, the relative energy differences between the Irꢀ
and Rhꢀcomplexes at corresponding stages of the reaction are
interesting: compared to the initial reactant complex i-M, the
dioxazoloneꢀbound intermediate iii-Rh is more stable at ꢀ4.6
kcal/mol than its iridium analogue iii-Ir, which registered a
relative solution free energy of only ꢀ0.9 kcal/mol. The CO2
eliminating transition state iii-Ir-TS1 was found at a relative
energy of 8.3 kcal/mol that is nearly 10 kcal/mol lower than
iii-Rh-TS1, which found at 18.2 kcal/mol. Interestingly, the
Ir−imido complex iv-Ir was found at ꢀ24.6 kcal/mol and also
about 10 kcal/mol lower in energy than iv-Rh, which has a
relative energy of only ꢀ14.4 kcal/mol.
These almost identical energy differences of the transition
states and M−imido intermediates are of course not coinciꢀ
dental and indicate that the transition states are “late” with
respect to their electronic structure distortions. In other words,
at the transition state, the electronic reorganization that gives
rise to the energetic difference between the two intermediates
has already developed fully. Indeed, the energy decomposition
analysis of the complexes iii-M, iii-M-TS1 and iv-M clearly
supports the idea that the free energy differences are the reꢀ
sults of the electronic interactions between [Cp*M(ppy)]n+
fragment and N source moiety, as described in Supporting
crepancy for the formation of 3-Ir (4.27
±
0.58 x 10ꢀ3 secꢀ1)
and 4-Ir (4.79
±
0.15 x 10ꢀ3 secꢀ1) at ꢀ38.7 oC is within experiꢀ
mental error range. The almost same rates between two differꢀ
ent but related reactions substantiate the DFTꢀbased proposal
that the nitrile dissociation is the rateꢀlimiting step for Irꢀ
mediated amidation.
To further support the mechanistic assignment that dioxaꢀ
zolone is intimately involved in the rateꢀlimiting step in the
Rhꢀsystem, but it is not at all important for the rateꢀlimiting
step in the Irꢀsystem, we repeated the amidation experiments
with varying amounts of the dioxazolone substrate in the reacꢀ
tion mixture.27 If the transition state for the Rh−nitrenoid forꢀ
mation is involved in the rateꢀlimiting step, the pseudoꢀfirst
order rate constant should be proportional to the concentration
of dioxazolone.15b If the dissociation of nitrile ligand is the
rateꢀdetermining step, as we propose for the Irꢀsystem, the
reaction rate should not be affected by the amount of dioxazoꢀ
lone in solution. As shown in Figure 5a, a clear firstꢀorder
dependence was observed for the rhodium complex as a funcꢀ
tion of [2b] in the range of 10 to 20 equivalents. For the iridiꢀ
um complex, the same experiment shows that the rate of the
reaction is independent of the concentration of 2b within the
same concentration range, as illustrated in Figure 5b.
Molecular Orbital Analysis. From a fundamental perꢀ
spective, the difference in mechanism discussed above stems
from the fact that Ir is much more reactive towards oxidative
Ir−nitrenoid formation, leading to facile C–N bond coupling.
The barrier is so low that the otherwise relatively innocent
elimination of the placeholder ligand, ArFCN, at the initiation
step has become rate determining. To better understand the
electronic foundation for this behavior, we examined the fronꢀ
tier molecular orbitals (FMOs) of the molecular species inꢀ
volved in the oxidative M–N coupling, namely, iii-M, iii-M-
TS1 and iv-M. Qualitatively, the changes in the FMOs during
the M–imido formation are identical for both metals, as sumꢀ
marized conceptually in Figure 6. As the carbon dioxide fragꢀ
ment is expelled, a vacant pz orbital is generated on the Nꢀ
fragment of the dioxazolone, which is used to carry out the
Information.
Significantly
shortened
distances
of
M−N(dioxazolone) bond in the transition states strongly sugꢀ
gest that changes in orbital interactions between M(dyz) and
N(pz) have already matured: Ir−N bond length of iii-Ir, iii-Ir-
TS1 and iv-Ir are 2.136 Å, 1.867 Å and 1.857 Å, respectively,
as shown in Figure 7 and Supporting Information. It is for this
reason that the relative thermodynamic stabilities of M−imidos
are transposed onto the transition states. It is important to note
that dioxazolone is unique in facilitating such “late” transition
state because the N−O and C−O bonds are being broken in a
concerted fashion, as there is no intermediate. But the process
is asynchronous, as the N−O bond is cleaved early and is alꢀ
most broken at the transition state, while the C−O bond reꢀ
mains intact, as shown in Figure 7. The C−O bond breaking
7
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 11
1
2
3
4
5
6
7
8
9
takes place after the transition state is traversed. Bond lengths
of N−O and C−O bonds in the transition state are 2.244 Å and
1.612 Å, respectively, and elongated N−O distance facilitates
the orbital overlap with N(pz) and M(dyz) to be maximized. It
is interesting to compare this mode of action with that of N2
cleavage of organic azides, as described in previous
studies.15a,16c Since N2 dissociation is a oneꢀbond cleaving
event, it is difficult to fully develop an empty N(pz) orbital at
the transition state and the interaction between the metal and
the azide moiety is necessarily weaker. Therefore, the nature
of N2ꢀcleaving transition state is largely unperturbed by the
stability of the M−imido species and it can be further generalꢀ
ized that the N2ꢀcleaving transition state should be higher in
energy than the CO2ꢀcleaving analogs. Whereas the structural
differences between the organic azides and the dioxazolones
are obvious, its consequence on the electronics of the C–N
coupling is impossible to predict without the precise analysis
of the DFT calculations presented here.
in spatial extension and lower in energy, thus making the Ir(V)
center a harder Lewis acid than the Rh(V) analogue. This elecꢀ
tronic structure is reflected in lower energy fragment orbitals,
as indicated in Figure 8 for the Mꢀfragments. Note that this
energy ordering is classically not expected, since the Irꢀ
orbitals carry a higher main quantum number than the Rhꢀ
orbitals, and are the results of the more pronounced relativistic
contraction of Ir. Consequently, the Ir–N σꢀbonding orbital
was found at ꢀ17.577 eV, which is nearly 0.3 eV lower than
the corresponding orbital that promotes the Rh–N σꢀbonding
interaction at ꢀ17.270 eV. Similarly, the M–N πꢀbond is
formed between the pzꢀlone pair orbital of the imido ligand
and the metalꢀd(yz) orbitals of the metal centers, as discussed
above and marked in blue in Figure 8. In the Ir–imido fragꢀ
ment, this frontier orbital was found at ꢀ12.262 eV, whereas
the same MO was located at ꢀ12.217 eV in the iv-Rh. These
energy differences in the M–N bonding manifold are a direct
illustration of the HSAB principle and provide an intuitively
understandable rationale for the energetics that we found in
our computational studies. At the transition state, the same
interactions exist and give rise ultimately to the observed effiꢀ
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ciency in M=N bond formation. The calculated Ir–N π and π*
orbitals are shown in Figure 9.
CONCLUSIONS
Combining computational reaction modeling with experꢀ
imental techniques of mechanistic inquiry, we were able to
establish a solid mechanistic understanding of how a Rh(III)
and a Ir(III) centers are able to promote oxidative M–N couꢀ
pling to afford a M(V)−imido complex, which can subsequentꢀ
ly undergo C–N bond forming reactions. These studies were
carried out as stoichiometric models of the catalytic amination
reactions and reveal new insights that are highly relevant for
Figure 7. DFTꢀoptimized structure of iii-Ir-TS1 and iv-Ir.
This qualitative FMO analysis highlights that the key to
understanding the kinetics of imido forming step lies in delinꢀ
eating the relative energies of the intermediates iv-Rh vs iv-Ir.
Figure 8 shows an MOꢀdiagram that compares the frontier
orbitals of the two complexes, iv-Rh and iv-Ir. As mentioned
above, the fragment orbitals of the iridium center are smaller
the C–H functionalization.
+
+
3+
3+
N
N
2-
N
N
O
R
O
O
IrV
RhV
C
RhV
C
N
IrV
C
N
N
R
Me
C
3+
[Cp*Ir(ppy)]3+
NAc2-
iv-Ir
iv-Rh
[Cp*Rh(ppy)]
E
LUMO
-6.156 eV
LUMO
-6.593 eV
dx2-z2
dx
2
2
-z
2
dy
dy2
dyz
dxy
dyz
dxy
dxz
dxz
-12.217 eV
-17.270 eV
-12.262 eV
-17.577 eV
HOMO-11
HOMO-46
z
y
x
Figure 8. Quantitative MOꢀdiagram comparing the Ir–imido (iv-Ir) to the Rh–imido (iv-Rh) bond.
8
ACS Paragon Plus Environment

Page 9 of 11
Journal of the American Chemical Society
ACKNOWLEDGMENT
This article is dedicated to Professor Naoto Chatani on the occaꢀ
sion of his 60th birthday. We thank Ms. Seohee Oh and Mr. Jin
Kim (KAIST) for helpful discussions.
1
2
3
4
5
6
7
8
9
REFERENCES
(1) (a) Ricci, A. Amino Group Chemistry: From Synthesis to the
Life Sciences; WileyꢀVCH: Weinheim, 2008. (b) Candeias, N. R.;
Branco, L. C.; P.Gois, P. M.; Afonso, C. A. M.; Trindade, A. F.
Chem. Rev. 2009, 109, 2703.
Figure 9. KohnꢀSham orbital plot (isodensity value: 0.05 a.u.)
(2) (a) Crabtree, R. H. J. Chem. Soc., Dalton Trans. 2001, 2437.
(b) Dick, A. R.; Sanford, M. S. Tetrahedron 2006, 62, 2439. (c)
Collet, F.; Dodd, R. H.; Dauban, P. Chem. Comm. 2009, 5061.
(3) Selected examples on the synthesis of metal−imido complex:
(a) Harlan, E. W.; Holm, R. H. J. Am. Chem. Soc. 1990, 112, 186.
(b) Glueck, D. S.; Wu, J.; Hollander, F. J.; Bergman, R. G. J. Am.
Chem. Soc. 1991, 113, 2041. (c) Perez, P. J.; White, P. S.;
Brookhart, M.; Templeton, J. L. Inorg. Chem. 1994, 33, 6050. (d)
Au, S.ꢀM.; Fung, W.ꢀH.; Huang, J.ꢀS.; Cheung, K.ꢀK.; Che, C.ꢀM.
Inorg. Chem. 1998, 37, 6564. (e) Dai, X.; Kapoor, P.; Warren, T.
H. J. Am. Chem. Soc. 2004, 126, 4798. (f) Kogut, E.; Wiencko, H.
L.; Zhang, L.; Cordeau, D. E.; Warren, T. H. J. Am. Chem. Soc.
2005, 127, 11248. (g) Ni, C.; Fettinger, J. C.; Long, G. J.; Brynda,
M.; Power, P. P. Chem. Comm. 2008, 6045. (h) Moret, M.ꢀE.;
Peters, J. C. Angew. Chem. Int. Ed. 2011, 50, 2063. (i) Wiese, S.;
McAfee, J. L.; Pahls, D. R.; McMullin, C. L.; Cundari, T. R.;
Warren, T. H. J. Am. Chem. Soc. 2012, 134, 10114. (j) Spasyuk,
D. M.; Carpenter, S. H.; Kefalidis, C. E.; Piers, W. E.; Neidig, M.
L.; Maron, L. Chem. Sci. 2016, 7, 5939.
(4) Selected reviews on the outersphere C−H amination: (a) Lu,
H.; Zhang, X. P. Chem. Soc. Rev. 2011, 40, 1899. (b) Strokes, B.
J.; Driver, T. G. Eur. J. Org. Chem. 2011, 4071. (c) Ramirez, T.
A.; Zhao, B.; Shi, Y. Chem. Soc. Rev. 2012, 41, 931. (d) Jeffrey,
J. L.; Sarpong, R. Chem. Sci. 2013, 4, 4092. (e) Uchida, T.;
Katsuki, T. Chem. Rec. 2014, 14, 117. (f) Du Bois, J. Org.
Process Res. Dev. 2011, 15, 758. (g) Schafer, A. G.; Blakey, S. B.
Chem. Soc. Rev. 2015, 44, 5969. (h) Jana, N.; Driver, T. G. Org.
Biomol. Chem. 2015, 13, 9720.
(5) (a) Au, S.ꢀM.; Huang, J.ꢀS.; Yu, W.ꢀY.; Fung, W.ꢀH.; Che, C.ꢀ
M. J. Am. Chem. Soc. 1999, 121, 9120. (b) Liang, J.ꢀL.; Huang,
J.ꢀS.; Yu, X.ꢀQ.; Zhu, N.; Che, C.ꢀM. Chem.–Eur. J. 2002, 8,
1563. (c) Leung, S. K.ꢀY.; Tsui, W.ꢀM.; Huang, J.ꢀS.; Che, C.ꢀM.;
Liang, J.ꢀL.; Zhu, N. J. Am. Chem. Soc. 2005, 127, 16629. (d)
Fantauzzi, S.; Gallo, E.; Caselli, A.; Ragaini, F.; Casati, N.;
Macchi, P.; Cenini, S. Chem. Comm. 2009, 3952.
(6) (a) Perry, R. H.; Cahill, T. J., III; Roizen, J. L.; Du Bois, J.;
Zare, R. N. Proc. Natl. Acad. Sci. U. S. A. 2012, 109, 18295. (b)
Harvey, M. E.; Musaev, D. G.; Du Bois, J. J. Am. Chem. Soc.
2011, 133, 17207. (c) Kornecki, K. P.; Berry, J. F. Chem.–Eur. J.
2011, 17, 5827. (d) Kornecki, K. P.; Berry, J. F. Chem. Comm.
2012, 48, 12097. (e) Warzecha, E.; Berto, T. C.; Berry, J. F.
Inorg. Chem. 2015, 54, 8817. (f) VarelaꢀAlvarez, A.; Yang, T.;
Jennings, H.; Kornecki, K. P.; Macmillan, S. N.; Lancaster, K.
M.; Mack, J. B. C.; Du Bois, J.; Berry, J. F.; Musaev, D. G. J. Am.
Chem. Soc. 2016, 138, 2327.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
of π and π* orbitals of iv-Ir.
We quantified the rates of the initial ligand dissociation and
the amidation, which allowed for studying each of the catalytꢀ
ic steps in an isolated manner. The observed higher reactivity
of the Irꢀcomplex in the amidation was successfully rationalꢀ
ized by the novel bonding feature of a dioxazole based amidatꢀ
ing reagent and the transposition of thermodynamic stability
of Ir(V) intermediate to the kinetic barrier. Another possible
mechanism, the synchronous CO2 extrusion and C−N forꢀ
mation without requiring a change in oxidation state of the
metal, is not found to be consistent with the experimentally
determined activation parameters and does not explain the
dramatically higher reactivity of iridium compared to rhodium
in the amidation reaction. This work constitutes the first fully
integrated experimental and theoretical study supporting the
involvement of the highꢀvalent M(V)−imido intermediates in
an inner-sphere type of amination mechanism. In addition to
evaluating reaction energies, the unique advantage of quantum
chemical methods32 lies in the frontier molecular orbital analyꢀ
sis that provides a deep and intuitively understandable conꢀ
cepts of bonding that are valid and useful. In this case, the key
concept that is responsible for the low barriers of the Irꢀbased
reactions is that Ir(III) and Ir(V) are harder and stronger Lewis
acids than the Rh(III) and Rh(V) analogues. The dioxazolone
gives a late transition state for the oxidative coupling step that
can fully take advantage of the strong M=N bonding. They are
able to interact much more favorably with the Nꢀbased ligands
that are strong and hard Lewis bases, which is ultimately reꢀ
sponsible for the superior performance of the Ir catalysts for
these type of reactions. We anticipate that the principles we
outlined herein will be valid for many reactions of this type
and this work may serve as a showcase example for the rationꢀ
al design of novel and efficient agents for groupꢀtransfer reacꢀ
tions.
ASSOCIATED CONTENT
Supporting Information. Detailed experimental procedures,
characterization of new compounds, kinetic profiles, Xꢀray analyꢀ
sis of 3-Ir and Cartesian coordinates of DFTꢀoptimized structures.
This material is available free of charge via the Internet at
http://pubs.acs.org.
(7) (a) King, E. R.; Hennessy, E. T.; Betley, T. A. J. Am. Chem.
Soc. 2011, 133, 4917. (b) Iovan, D. A.; Betley, T. A. J. Am.
Chem. Soc. 2016, 138, 1983.
(8) Bryant, J. R.; Mayer, J. M. J. Am. Chem. Soc. 2003, 125,
10351.
(9) Selected literatures on C−H functionalization: (a) Murai, S.;
Kakiuchi, F.; Sekine, S.; Tanaka, Y.; Kamatani, A.; Sonoda, M.;
Chatani, N. Nature 1993, 366, 529. (b) Davies, D. L.; Donald, S.
M. A.; AlꢀDuaij, O.; Macgregor, S. A.; Pölleth, M. J. Am. Chem.
Soc. 2006, 128, 4210. (c) Davies, D. L.; Donald, S. M. A.; Alꢀ
Duaij, O.; Fawcett, J.; Little, C.; Macgregor, S. A.
Organometallics 2006, 25, 5976. (d) Li, L.; Brennessel, W. W.;
AUTHOR INFORMATION
Corresponding Author
MuꢀHyun Baik (mbaik2805@kaist.ac.kr)
Sukbok Chang (sbchang@kaist.ac.kr)
Funding Sources
This research was supported by the Institute for Basic Science
(IBSꢀR010ꢀD1) in Korea.
9
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 11
Jones, W. D. J. Am. Chem. Soc. 2008, 130, 12414. (e) Seregin, I.
Huang, F.; Chen, D.ꢀZ. ACS Catal. 2016, 6, 2452. (d) Yang, Y.ꢀ
1
2
3
4
5
6
7
8
V.; Gevorgyan, V. Chem. Soc. Rev. 2007, 36, 1173. (f) Colby, D.
A.; Bergman, R. G.; Ellman, J. A. Chem. Rev. 2010, 110, 624. (g)
Cho, S. H.; Kim, J. Y.; Kwak, J.; Chang, S. Chem. Soc. Rev.
2011, 40, 5068. (h) Li, H.; Lia, B.ꢀJ.; Shi, Z.ꢀJ. Catal. Sci. Tech.
2011, 1, 191. (i) Davies, H. M. L.; Morton, D. Chem. Soc. Rev.
2011, 40, 1857. (j) WencelꢀDelord, J.; Droge, T.; Liu, F.; Glorius,
F. Chem. Soc. Rev. 2011, 40, 4740. (k) Ackermann, L. Chem.
Rev. 2011, 111, 1315. (l) Rouquet, G.; Chatani, N. Angew. Chem.
Int. Ed. 2013, 52, 11726. (m) Daugulis, O.; Roane, J.; Tran, L. D.
Acc. Chem. Res. 2015, 48, 1053. (n) Gensch, T.; Hopkinson, M.
N.; Glorius, F.; WencelꢀDelord, J. Chem. Soc. Rev. 2016, 45,
2900. (o) He, G.; Wang, B.; Nack, W. A.; Chen, G. Acc. Chem.
Res. 2016, 49, 635. (p) Davies, H. M. L.; Morton, D. J. Org.
Chem. 2016, 81, 343.
(10) Selected reviews on the innersphere C−H amination: (a)
Matsuda, N.; Hirano, K.; Satoh, T.; Miura, M. Synthesis 2012, 44,
1792. (b) Mei, T.ꢀS.; Kou, L.; Ma, S.; Engle, K. M.; Yu, J.ꢀQ.
Synthesis 2012, 44, 1778. (c) Louillat, M.ꢀL.; Patureau, F. W.
Chem. Soc. Rev. 2014, 43, 901. (d) Jiao, J.; Murakami, K.; Itami,
K. ACS Catal. 2016, 6, 610. (e) Kim, H.; Chang, S. ACS Catal.
2016, 6, 2341. (f) Subramanian, P.; Rudolf, G. C.; Kaliappan, K.
P. Chem.–Asian J. 2016, 11, 168.
(11) (a) Wendlandt, A. E.; Suess, A. M.; Stahl, S. S. Angew.
Chem. Int. Ed. 2011, 50, 11062. (b) Huffman, L. M.; Stahl, S. S.
J. Am. Chem. Soc. 2008, 130, 9196. (c) Suess, A. M.; Ertem, M.
Z.; Cramer, C. J.; Stahl, S. S. J. Am. Chem. Soc. 2013, 135, 9797.
(d) Pendleton, I. M.; PérezꢀTemprano, M. H.; Sanford, M. S.;
Zimmerman, P. M. J. Am. Chem. Soc. 2016, 138, 6049.
(12) (a) Thu, H.ꢀY.; Yu, W.ꢀY.; Che, C.ꢀM. J. Am. Chem. Soc.
2006, 128, 9048. (b) Dick, A. R.; Remy, M. S.; Kampf, J. W.;
Sanford, M. S. Organometallics 2007, 26, 1365. (c) Ng, K.ꢀH.;
Chan, A. S. C.; Yu, W.ꢀY. J. Am. Chem. Soc. 2010, 132, 12862.
(d) Xiao, B.; Gong, T.ꢀJ.; Xu, J.; Liu, Z.ꢀJ.; Liu, L. J. Am. Chem.
Soc. 2011, 133, 1466. (e) Ke, Z.; Cundari, T. R. Organometallics
2010, 29, 821.
(13) (a) Kim, J. Y.; Park, S. H.; Ryu, J.; Hwan, S. C.; Kim, S. H.;
Chang, S. J. Am. Chem. Soc. 2012, 134, 9110. (b) Ryu, J.; Kwak,
J.; Shin, K.; Lee, D.; Chang, S. J. Am. Chem. Soc. 2013, 135,
12861. (c) Kang, T.; Kim, Y.; Lee, D.; Wang, Z.; Chang, S. J.
Am. Chem. Soc. 2014, 136, 4141. (d) Park, S. H.; Park, Y.;
Chang, S. Org. Synth. 2014, 91, 52. (e) Shin, K.; Kim, H.; Chang,
S. Acc. Chem. Res. 2015, 48, 1040. (f) Kim, Y.; Park, J.; Chang,
S. Org. Lett. 2016, 18, 1892.
(14) (a) Yu, D.ꢀG.; Suri, M.; Glorius, F. J. Am. Chem. Soc. 2013,
135, 8802. (b) Zheng, Q.ꢀZ.; Liang, Y.ꢀF.; Qin, C.; Jiao, N. Chem.
Comm. 2013, 49, 5654. (c) Pan, C.; Abdukader, A.; Han, J.;
Cheng, Y.; Zhu, C. Chem.–Eur. J. 2014, 20, 3606. (d) Sun, B.;
Yoshino, T.; Matsunaga, S.; Kanai, M. Adv. Synth. Catal. 2014,
356, 1491. (e) Chen, H.; Huestis, M. P. ChemCatChem 2015, 7,
743. (f) Hou, W.; Yang, Y.; Ai, W.; Wu, Y.; Wang, X.; Zhou, B.;
Li, Y. Eur. J. Org. Chem. 2015, 2015, 395. (g) Pi, C.; Cui, X.;
Wu, Y. J. Org. Chem. 2015, 80, 7333. (h) Ali, M. A.; Yao, X.; Li,
G.; Lu, H. Org. Lett. 2016, 18, 1386. (i) Hermann, G. N.; Becker,
P.; Bolm, C. Angew. Chem. Int. Ed. 2016, 55, 3781. (j) Zhang, T.;
Hu, X.; Wang, Z.; Yang, T.; Sun, H.; Li, G.; Lu, H. Chem.–Eur.
J. 2016, 22, 2920. (k) Zhu, B.; Cui, X.; Pi, C.; Chen, D.; Wu, Y.
Adv. Synth. Catal. 2016, 358, 326.
F.; Houk, K. N.; Wu, Y.ꢀD. J. Am. Chem. Soc. 2016, 138, 6861.
(17) Zhang, L.ꢀL.; Li, L.ꢀH.; Wang, Y.ꢀQ.; Yang, Y.ꢀF.; Liu, X.ꢀ
Y.; Liang, Y.ꢀM. Organometallics 2014, 33, 1905.
(18) (a) Park, Y.; Park, K. T.; Kim, J. G.; Chang, S. J. Am. Chem.
Soc. 2015, 137, 4534. (b) Park, Y.; Jee, S.; Kim, J. G.; Chang, S.
Org. Process Res. Dev. 2015, 19, 1024. (c) Park, J.; Chang, S.
Angew. Chem. Int. Ed. 2015, 54, 14103.
(19) (a) Sauer, J.; Mayer, K. K. Tetrahedron Lett. 1968, 9, 319.
(b) Zhong, C. L.; Tang, B. Y.; Yin, P.; Chen, Y.; He, L. J. Org.
Chem. 2012, 77, 4271. (c) Bizet, V.; Buglioni, L.; Bolm, C.
Angew. Chem. Int. Ed. 2014, 53, 5639. (d) Bizet, V.; Bolm, C.
Eur. J. Org. Chem. 2015, 2015, 2854.
(20) (a) Wang, H.; Tang, G.; Li, X. Angew. Chem. Int. Ed. 2015,
54, 13049. (b) Wang, F.; Wang, H.; Wang, Q.; Yu, S.; Li, X. Org.
Lett. 2016, 18, 1306.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(21) Liang, Y.; Liang, Y.ꢀF.; Tang, C.; Yuan, Y.; Jiao, N. Chem.–
Eur. J. 2015, 21, 16395.
(22) (a) Mei, R.; Loup, J.; Ackermann, L. ACS Catal. 2016, 6,
793. (b) Wang, H.; Lorion, M. M.; Ackermann, L. Angew. Chem.
Int. Ed. 2016, 55, 10386.
(23) Wang, X.; Lerchen, A.; Glorius, F. Org. Lett. 2016, 18, 2090.
(24) (a) Barsu, N.; Rahman, M. A.; Sen, M.; Sundararaju, B.
Chem.–Eur. J. 2016, 22, 9135. (b) Wang, J.; Zha, S.; Chen, K.;
Zhang, F.; Song, C.; Zhu, J. Org. Lett. 2016, 18, 2062.
(25) (a) Sau, Y.ꢀK.; Yi, X.ꢀY.; Chan, K.ꢀW.; Lai, C.ꢀS.; Williams,
I. D.; Leung, W.ꢀH. J. Organomet. Chem. 2010, 695, 1399. (b)
Turlington, C. R.; White, P. S.; Brookhart, M.; Templeton, J. L.
Organometallics 2015, 34, 4810.
(26) (a) Sanford, M. S.; Love, J. A.; Grubbs, R. H. J. Am. Chem.
Soc. 2001, 123, 6543. (b) Powers, D. C.; Benitez, D.; Tkatchouk,
E.; Goddard, W. A.; Ritter, T. J. Am. Chem. Soc. 2010, 132,
14092. (c) Powers, D. C.; Xiao, D. Y.; Geibel, M. A. L.; Ritter, T.
J. Am. Chem. Soc. 2010, 132, 14530. (d) Ball, N. D.; Gary, J. B.;
Ye, Y.; Sanford, M. S. J. Am. Chem. Soc. 2011, 133, 7577. (e)
Turlington, C. R.; White, P. S.; Brookhart, M.; Templeton, J. L. J.
Am. Chem. Soc. 2014, 136, 3981. (f) Lotz, M. D.; Remy, M. S.;
Lao, D. B.; Ariafard, A.; Yates, B. F.; Canty, A. J.; Mayer, J. M.;
Sanford, M. S. J. Am. Chem. Soc. 2014, 136, 8237.
(27) See Supporting Information for details.
(28) (a) Hinderling, C.; Plattner, D. A.; Chen, P. Angew. Chem.
Int. Ed. 1997, 36, 243. (b) Rerek, M. E.; Basolo, F. J. Am. Chem.
Soc. 1984, 106, 5908. (c) Fan, H.ꢀJ.; Hall, M. B. Organometallics
2001, 20, 5724. (d) O'Connor, J. M.; Casey, C. P. Chem. Rev.
1987, 87, 307. (e) Janowicz, A. H.; Bryndza, H. E.; Bergman, R.
G. J. Am. Chem. Soc. 1981, 103, 1516. (f) Glueck, D. S.;
Bergman, R. G. Organometallics 1991, 10, 1479.
(29) (a) Tauchert, M. E.; Incarvito, C. D.; Rheingold, A. L.;
Bergman, R. G.; Ellman, J. A. J. Am. Chem. Soc. 2012, 134,
1482. (b) Tsai, A. S.; Tauchert, M. E.; Bergman, R. G.; Ellman, J.
A. J. Am. Chem. Soc. 2011, 133, 1248. (c) Yu, S.; Tang, G.; Li,
Y.; Zhou, X.; Lan, Y.; Li, X. Angew. Chem. Int. Ed. 2016, 55,
8696.
(30) (a) Siegbahn, P. E. M. J. Am. Chem. Soc. 1996, 118, 1487.
(b) Gorin, D. J.; Toste, F. D. Nature 2007, 446, 395. (c) Pyykko,
P. Chem. Rev. 1988, 88, 563. (d) LeyvaꢀPérez, A.; Corma, A.
Angew. Chem. Int. Ed. 2012, 51, 614.
(31) Turlington, C. R.; White, P. S.; Brookhart, M.; Templeton, J.
L. J. Organomet. Chem. 2015, 792, 81.
(32) Selected reviews on computational organometallic catalysis:
(a) Balcells, D.; Clot, E.; Eisenstein, O. Chem. Rev. 2010, 110,
749. (b) Lodewyk, M. W.; Siebert, M. R.; Tantillo, D. J. Chem.
Rev. 2012, 112, 1839. (c) Sperger, T.; Sanhueza, I. A.; Kalvet, I.;
Schoenebeck, F. Chem. Rev. 2015, 115, 9532. (d) Park, Y.; Ahn,
S.; Kang, D.; Baik, M.ꢀH. Acc. Chem. Res. 2016, 49, 1263. (e)
Sperger, T.; Sanhueza, I. A.; Schoenebeck, F. Acc. Chem. Res.
2016, 49, 1311. (f) Guan, C. S.; Cundari, T. R. Comput. Theor.
Chem. 2016, 1091, 64.
(15) (a) Figg, T. M.; Park, S.; Park, J.; Chang, S.; Musaev, D. G.
Organometallics 2014, 33, 4076. (b) Park, S. H.; Kwak, J.; Shin,
K.; Ryu, J.; Park, Y.; Chang, S. J. Am. Chem. Soc. 2014, 136,
2492. (c) Ajitha, M. J.; Huang, K.ꢀW.; Kwak, J.; Kim, H. J.;
Chang, S.; Jung, Y. Dalton Trans. 2016, 45, 7980.
(16) (a) Xu, L.; Zhu, Q.; Huang, G.; Cheng, B.; Xia, Y. J. Org.
Chem. 2012, 77, 3017. (b) Zhou, T.; Guo, W.; Xia, Y. Chem.–
Eur. J. 2015, 21, 9209. (c) Liu, J.ꢀB.; Sheng, X.ꢀH.; Sun, C.ꢀZ.;
10
ACS Paragon Plus Environment

Page 11 of 11
Journal of the American Chemical Society
1
2
3
4
5
Table of Contents
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
11
ACS Paragon Plus Environment