Proton Switch in the Secondary Coordination Sphere to Control Catalytic Events at the Metal Center: Biomimetic Oxo Transfer Chemistry of Nickel Amidate Complex

Article abstract of DOI:10.1002/chem.202005183

High-valent metal-oxo species are key intermediates for the oxygen atom transfer step in the catalytic cycles of many metalloenzymes. While the redox-active metal centers of such enzymes are typically supported by anionic amino acid side chains or porphyrin rings, peptide backbones might function as strong electron-donating ligands to stabilize high oxidation states. To test the feasibility of this idea in synthetic settings, we have prepared a nickel(II) complex of new amido multidentate ligand. The mononuclear nickel complex of this N5 ligand catalyzes epoxidation reactions of a wide range of olefins by using mCPBA as a terminal oxidant. Notably, a remarkably high catalytic efficiency and selectivity were observed for terminal olefin substrates. We found that protonation of the secondary coordination sphere serves as the entry point to the catalytic cycle, in which high-valent nickel species is subsequently formed to carry out oxo-transfer reactions. A conceptually parallel process might allow metalloenzymes to control the catalytic cycle in the primary coordination sphere by using proton switch in the secondary coordination sphere.

Full text of DOI:10.1002/chem.202005183

Full Paper
doi.org/10.1002/chem.202005183
Chemistry—A European Journal
&
Bioinorganic Chemistry
Proton Switch in the Secondary Coordination Sphere to Control
Catalytic Events at the Metal Center: Biomimetic Oxo Transfer
Chemistry of Nickel Amidate Complex
Soohyung Kim,[a] Ha Young Jeong,[b] Seonghan Kim,[c] Hongsik Kim,[a] Sojeong Lee,[b]
Jaeheung Cho,*[c, d] Cheal Kim,*[b] and Dongwhan Lee*[a]
Abstract: High-valent metal-oxo species are key intermedi-
ates for the oxygen atom transfer step in the catalytic cycles
of many metalloenzymes. While the redox-active metal cen-
ters of such enzymes are typically supported by anionic
amino acid side chains or porphyrin rings, peptide back-
bones might function as strong electron-donating ligands to
stabilize high oxidation states. To test the feasibility of this
idea in synthetic settings, we have prepared a nickel(II) com-
plex of new amido multidentate ligand. The mononuclear
nickel complex of this N5 ligand catalyzes epoxidation reac-
tions of a wide range of olefins by using mCPBA as a termi-
nal oxidant. Notably, a remarkably high catalytic efficiency
and selectivity were observed for terminal olefin substrates.
We found that protonation of the secondary coordination
sphere serves as the entry point to the catalytic cycle, in
which high-valent nickel species is subsequently formed to
carry out oxo-transfer reactions. A conceptually parallel pro-
cess might allow metalloenzymes to control the catalytic
cycle in the primary coordination sphere by using proton
switch in the secondary coordination sphere.
Introduction
Nitrile hydratases (NHases) hydrolyze nitriles into amides.
The active site of cobalt-based NHase (Figure 1a) is comprised
of two monoanionic amidate nitrogen atoms from the peptide
backbone and three sulfur atoms from amino acid side
chains.[7] For a finer control over the Lewis acidity and redox
properties of the metal center, hydrogen bonding networks co-
operatively exert secondary coordination sphere effects.[8] As
another example, nickel superoxide dismutase (NiSOD) catalyz-
es the disproportionation of superoxide to produce dioxygen
and hydrogen peroxide.[9] The reaction mechanism requires cy-
cling between nickel(II) and nickel(III) oxidation states, al-
Nature has evolved efficient strategies to transfer oxygen
atoms to hydrocarbons by using iron,[1] copper,[2] or manga-
nese[3] dependent metalloenzymes. In such biological catalysis,
high-valent metal-oxo species are often involved as key inter-
mediates.[4] To help access the high oxidation states of the
metal, the active sites of metalloenzymes exploit anionic li-
gands derived from amino acid side chains or porphyrin deriv-
atives.[5] A few rare examples of metalloenzymes, however, uti-
lize peptide backbone itself to supply anionic ligand donors.[6]
[a] S. Kim, H. Kim, Prof. Dr. D. Lee
Department of Chemistry, Seoul National University
1 Gwanak-ro, Gwanak-gu, Seoul 08826 (Korea)
E-mail: dongwhan@snu.ac.kr
[b] H. Y. Jeong, S. Lee, Prof. Dr. C. Kim
Department of Fine Chemistry
Seoul National University of Science and Technology
232 Gongneung-ro, Nowon-gu, Seoul 01811 (Korea)
E-mail: chealkim@seoultech.ac.kr
[c] S. Kim, Prof. Dr. J. Cho
Department of Emerging Materials Science, DGIST
Daegu 42988 (Korea)
[d] Prof. Dr. J. Cho
Department of Chemistry
Ulsan National Institute of Science and Technology (UNIST)
Ulsan 44919 (Korea)
E-mail: jaeheung@unist.ac.kr
Supporting information and the ORCID identification numbers for the
authors of this article can be found under:
https://doi.org/10.1002/chem.202005183.
Figure 1. Chemical structures of the active sites of (a) cobalt-based NHase
and (b) NiSOD. Shown next is the X-ray structure of each enzyme (PDB
code: 1IRE for cobalt-based NHase; 1T6U for NiSOD).
Chem. Eur. J. 2021, 27, 4700 –4708
4700
ꢀ 2021 Wiley-VCH GmbH

Full Paper
doi.org/10.1002/chem.202005183
Chemistry—A European Journal
though the redox chemistry of nickel in non-porphyrinoid set-
ting is challenging under biological conditions.[10] In the active
site of NiSOD, the metal center is coordinated to the ligands
derived from cysteine thiolates, N-terminal amine, and depro-
tonated amide backbone (Figure 1b). The peptidyl N-amidate
ligand here helps stabilize the oxidized nickel(III) species and
prevent the deleterious oxidation of cysteine sulfur atoms.[11]
As synthetic systems inspired by metalloenzymes, nickel-
based catalysts can carry out hydroxylation of alkanes and ep-
oxidation of alkenes using H2O2, mCPBA, or NaOCl as terminal
oxidant.[12] To accommodate high-valent nickel-oxo species as
active oxidants in the catalytic cycle, various nickel complexes
have been developed that are supported by electron-rich li-
gands having anionic amidate or neutral guanidyl motif.[12f–k]
Only a few nickel-based catalytic systems, however, are cur-
rently available for efficient epoxidation of terminal olefins
with high selectivity under mild conditions.[13]
chemistry.[15] The first-generation tridentate ligand 2 was pre-
pared by installing a flexible pyridyl pendant to the rigid and
p-conjugated anilido-pyridyl [N,N]-bidentate motif 3. To stabi-
lize the high-valent metal-oxo species, we have increased the
ligand denticity by appending the tridentate bis(2-pyridylme-
thyl)amine tether to the bidentate 3 by using an amide linker
(Figure 2a).
Upon metallation, the amide NÀH group of H-dpap is depro-
tonated to afford monoanionic ligand dpap. As shown in Fig-
ure 2b, the amide carbonyl group of the dpap ligand is not di-
rectly coordinated to the metal, but protonation or metal-bind-
ing to this secondary coordination sphere can reversibly modu-
late the relative contribution of the amido vs. iminol/iminolate
Lewis structure descriptions in resonance, thereby enabling
remote control of the ligand donor ability.[16] The ligand
H-dpap was conveniently prepared by an amide coupling
reaction between 3[15] and 4[17] in excellent yield (>95%)
(Scheme 1).
In this study, we report the chemistry of a new pentadentate
ligand H-dpap (dpap = 2-(di(pyridin-2-ylmethyl)amino)-N-(2-(5-
methylpyridin-2-yl)phenyl)acetamidate) and its mononuclear
nickel complex [NiII(dpap)](ClO4) (1), which functions as an effi-
cient catalyst for the epoxidation of various terminal olefins.
Due to their electron deficient nature, terminal olefins are a
particularly challenging class of substrates to epoxidize. Since
terminal epoxides are useful building blocks for the synthesis
of value-added organic molecules, developing an efficient cata-
lyst is a topic of both fundamental and practical importance.[14]
Scheme 1. Synthesis of H-dpap ligand.
Results and Discussion
Preparation and characterization of nickel(II) complex
Ligand design and synthesis
Mononuclear nickel complex 1 was synthesized from Ni(-
ClO4)2·6H2O and H-dpap in the presence of Et3N in MeCN. The
resulting light purple precipitate was isolated by filtration, and
washed with MeCN and Et2O. The crude product was purified
by crystallization from MeOH/Et2O, which also produced
single-crystals suitable for X-ray crystallography (Figure 3a).
As anticipated, the nickel center in the cation of 1 is coordi-
nated to all of the five nitrogen atoms of the monoanionic
ligand. Among the five NiÀN bonds, the axial NiÀN1 distance
(2.016(2) Š) is the shortest, which reflects the strong electron
donor ability of the anionic amidate-N. Intriguingly, the open
axial site of the complex is taken up by the carbonyl oxygen
atom of adjacent 1 unit to complete a pseudo-octahedral ge-
ometry (Figure 3b). As shown in Figure 2b, metal binding en-
hances the contribution of the iminolate over the amido char-
acter, thereby attenuating the electron-donating ability of the
nitrogen atom of the amide group. Such a cascade transmis-
sion of electronic effect is best accommodated in a one-dimen-
sional (1D) polymeric structure in the solid state.[18] When the
1D coordination polymer dissolves in polar solvent MeCN,
however, the weak NiÀOcarbonyl bond would dissociate so that a
pentacoordinate nickel complex could be generated with a
vacant axial site for substrate binding and subsequent reaction
chemistry. We proceeded to test this idea with oxo group
transfer reactions to olefinic substrates.
The dpap ligand system is based on the N3 ligand 2 (Fig-
ure 2a) which we recently reported for biomimetic iron
Figure 2. (a) Functionalization of anilido-pyridine to increase the denticity
and electron-donating ability of the ligand. (b) Modulation of secondary co-
ordination sphere by protonation or binding of Lewis acidic metal to the
amide carbonyl group.
Chem. Eur. J. 2021, 27, 4700 –4708
www.chemeurj.org
4701
ꢀ 2021 Wiley-VCH GmbH

Full Paper
doi.org/10.1002/chem.202005183
Chemistry—A European Journal
Table 1. Epoxidation of olefins catalyzed by 1.[a]
Entry Substrate
Product
Conversion Yield
(%)[b]
(%)[b]
1
2
3
4
cyclopentene epoxide
cycloheptene epoxide
ꢀ100
ꢀ100
63.2Æ2.8
ꢀ100
cyclooctene
cyclohexene
epoxide
epoxide
98.0Æ0.1
59.2Æ0.5
97.9Æ0
82.3Æ3.5
trace
2-cyclohexenol
2-cyclohexenone
epoxide
4.3Æ0
5
6
7
1-hexene
63.1Æ1.4
71.5Æ1.1
97.0Æ0.2
50.5Æ5.2
46.5Æ0.9
30.0Æ0.2
62.6Æ0.7
85.5Æ1.2
32.1Æ3.0
61.0Æ6.9
72.1Æ1.2
62.7Æ1.8
3.8Æ0.2
12.1Æ0.7
9.3Æ0.1
63.4Æ4.8
7.3Æ0.7
trace
1-octene
epoxide
cis-2-hexene
cis-oxide
trans-oxide
trans-2-hexene trans-oxide
8
9
ꢀ100
ꢀ100
cis-2-octene
cis-oxide
trans-oxide
trans-2-octene trans-oxide
10
11
99.2Æ0.4
styrene
epoxide
ꢀ100
benzaldehyde
phenylacetaldehyde
cis-stilbene oxide
trans-stilbene oxide
benzaldehyde
12
13
cis-stilbene
ꢀ100
2-phenylacetophenone
trans-stilbene trans-stilbene oxide
benzaldehyde
Figure 3. (a) ORTEP diagram of the cation of 1 with thermal ellipsoids at the
50% probability level. Hydrogen atoms are omitted for clarity. (b) Capped-
stick representation of 1D polymeric packing of 1 and a close-up view. The
ꢀ100
53.0Æ1.2
7.5Æ0.1
trace
2-phenylacetophenone
À
ClO4 counter anions are omitted for clarity.
[a] Reaction conditions: substrates (0.035 mmol), 1 (0.0017 mmol), and
mCPBA (0.10 mmol) in MeCN (1 mL), T = 258C. [b] Based on the sub-
strate.
Olefin epoxidation reactions
Using 1 as an oxo-transfer catalyst and mCPBA as a terminal
oxidant, olefin epoxidation reactions were carried out for a
wide range of olefin substrates listed in Table 1. At r.t., the re-
actions were complete within <10 min in MeCN. In the ab-
sence of 1, however, direct epoxidation of substrates by
mCPBA alone was negligible (Table S2).
Styrene was oxidized to styrene oxide (62.7%, entry 11)
along with small amounts of benzaldehyde (3.8%) and phenyl-
acetaldehyde (12.1%). For cis-stilbene (entry 12), trans-stilbene
oxide (63.4%) was formed as the major product with a small
amount of cis-stilbene oxide (9.3%); trans-stilbene (entry 13)
was transformed to trans-stilbene oxide (53.0%) as the major
product along with benzaldehyde byproduct (7.5%). The for-
mation of aldehyde and ketone products from aromatic olefins
suggests that certain amount of peroxyl radical species partici-
pates as active oxidants in the catalytic reactions.[12a,20]
As reported in Table 1, cyclic olefins (entries 1–3) were con-
verted to the corresponding epoxides in moderate to excellent
yields (59.2–100%). Cyclohexene was oxidized to cyclohexene
oxide (82.3%, entry 4) as the major product, along with small
amounts of byproducts such as 2-cyclohexenone (4.3%) and 2-
cyclohexenol (trace). Terminal olefins such as 1-hexene and 1-
octene (entries 5 and 6) are challenging substrates to oxidize.
We found that 1 epoxidized them in moderate yields (47–
51%). Previously reported non-heme nickel complexes and
porphyrin nickel complexes typically show low catalytic activity
towards terminal olefins.[12c,j]
Epoxidation of terminal olefins
The epoxidation of terminal olefins is an important class of
functional group transformation. The resulting 1,2-epoxide
products are versatile building blocks for the synthesis of
useful organic molecules.[14] Since our nickel-based catalytic
system showed promising reactivity towards terminal olefins
(entries 5 and 6, Table 1), we proceeded to examine the epoxi-
dation reactions of an expanded set of terminal olefins. Under
standard conditions, various terminal olefins reacted in excel-
lent conversion (93.6–100%) to afford the epoxidation prod-
ucts up to 90% yield (Table 2). The selectivity showed a gradu-
al decrease with increasing chain length. No nickel-based cata-
lysts are currently known that effect catalytic epoxidation of
terminal olefins with such high efficiency and selectivity.[12c,j]
To study the stereochemistry of the epoxidation reaction,
cis-2-hexene and cis-2-octene (entries 7 and 9) were used. They
were mainly converted to trans-2-hexene oxide (62.6%) and
trans-2-octene oxide (61.0%), along with lower amounts of cis-
2-hexene oxide (30.0%) and cis-2-octene oxide (32.1%). When
trans-2-hexene and trans-2-octene (entries 8 and 10) were
used, trans-2-hexene oxide (85.5%) and trans-2-octene oxide
(72.1%) were obtained. These results implicate that the puta-
tive Ni-oxo-olefin radical intermediate has sufficiently long life-
time, to allow for the rotation of the initially formed Ni-oxo-cis-
olefin radical to the thermodynamically more stable trans-
olefin radical (Scheme S1).[19]
Chem. Eur. J. 2021, 27, 4700 –4708
www.chemeurj.org
4702
ꢀ 2021 Wiley-VCH GmbH

Full Paper
doi.org/10.1002/chem.202005183
Chemistry—A European Journal
m/z = 240.6 and m/z = 261.1 appeared, which correspond to
the (calcd m/z 240.6) and
[NiII(dpap)(H)]2+
Table 2. Epoxidation of terminal olefins catalyzed by 1.[a]
=
Entry Substrate
Product
Conversion Yield
[NiII(dpap)(MeCN)(H)]2+ (calcd m/z = 261.1) species, respective-
ly. However, its conjugate base [NiII(dpap)(MeCN)]+ (calcd m/z
= 521.2) was not detected under the same condition (Fig-
ure 4b). To confirm that this shift in solution population is
indeed triggered by protonation, 1 equiv of HClO4 was added
to the solution sample of 1. The same peak at m/z = 261.1 for
[NiII(dpap)(MeCN)(H)]2+ was detected, but no peak was ob-
served at m/z = 521.2 for [NiII(dpap)(MeCN)]+ (Figure S2 and
Figure S3).
(%)[b]
(%)[b]
1
1-hexene
1,2-epoxyhexane
1,2-epoxyoctane
1,2-epoxynonane
1,2-epoxydecane
1,2-epoxyundecane
1,2-epoxydodecane
1,2-epoxytridecane
1,2-epoxytetradecane
1,2-epoxypentadecane
1,2-epoxyhexadecane
1,2-epoxyoctadecane
ꢀ100
89.5Æ0.7
2
1-octene
93.9Æ0.5 72.8Æ1.2
96.4Æ0.5 69.4Æ2.1
96.9Æ0.4 52.8Æ2.0
95.9Æ0.6 44.0Æ1.2
99.9Æ0.1 44.3Æ1.9
98.0Æ0.2 35.4Æ3.8
99.1Æ0.1 20.3Æ3.8
96.5Æ0.1 31.6Æ0.8
3
1-nonene
4
1-decene
5
6
7
8
1-undecene
1-dodecene
1-tridecene
1-tetradecene
1-pentadecene
1-hexadecene
1-octadecene
9
These results suggest that protonation lowers the electron
donating ability of the ligand (Figure 2b), so that MeCN sol-
vent molecule is recruited as the sixth ligand to balance the
overall electronic demand of the metal center. It is reasonable
to speculate that the position of protonation is the amide car-
bonyl oxygen atom when analogy is drawn to the X-ray struc-
ture shown in Figure 3b.[21] Here, Lewis acidic nickel(II) center
of the adjacent repeat unit of the coordination polymer in the
solid-state takes the role of proton in solution.
10
11
12
ꢀ100
ꢀ100
37.8Æ3.8
18.4Æ0.9
vinylcyclohexane vinylcyclohexane oxide
93.6Æ2.1 60.0Æ1.7
[a] Reaction conditions: substrates (0.020 mmol), 1 (0.0017 mmol), and
mCPBA (0.20 mmol) in MeCN (1 mL), T = 258C. [b] Based on the sub-
strate.
Reversible protonation of nickel complex
In support of this interpretation, the DFT computed (UBP86/
Def2-TZVP level of theory) Mulliken charge of the carbonyl
oxygen atom is À0.33, which is the most negative among all
the heteroatoms within 1 (Table S3). The molecular electrostat-
ic potential (MEP) map of 1 (Figure S4) also shows that the car-
bonyl oxygen atom is the most electron-rich site to be target-
ed by the proton. Indeed, the geometry optimized DFT models
of protonated [NiII(dpap)(H)]2+ (Figure S5) predict that the
oxygen-protonated isomer should be the energetically most
preferred form when the proton is allowed to choose any of
the Brønsted basic N- or O-donor atoms of [NiII(dpap)]+.
In our mass spectrometric analysis, a peak at m/z = 652.2
was also observed (Figure 4b), which corresponds to the
[NiII(dpap)(mCPBA)]+ (calcd m/z = 652.1) species. This finding
implies that mCPBA used as a terminal oxidant can also coordi-
nate to the vacant site of nickel center. A peak at m/z = 496.2
was also observed, which could be assigned to the
[NiIV(dpap)(O)]+ (calcd m/z = 496.1) species.
To elucidate the mechanism of oxygen atom transfer reaction
described above, we first performed a cold spray ionization
mass spectrometry (CSI-MS) analysis of an MeCN solution
sample of 1. Only one prominent peak was observed at m/z =
480.2, which corresponds to the mass of discrete [NiII(dpap)]+
species (calcd m/z = 480.1) (Figure 4a). The isotopic patterns
for the observed ions further validated this assignment (Fig-
ure S1). Upon addition of mCPBA to the solution, new peaks at
A coherent mechanistic picture emerges from the CSI-MS
analysis of 1 in solution. As postulated in Scheme 2, protona-
tion event in the ligand sphere controls the electronic demand
of the metal sphere. When mCPBA is added to the solution for
the epoxidation reaction, the released proton should first bind
Figure 4. Positive-ion CSI-MS spectra (T = 233 K) of (a) 1 in MeCN (concen-
tration = 0.1 mm), and (b) after treatment with mCPBA (2 equiv). Inset
shows the observed isotope distribution pattern.
Scheme 2. Proposed mechanism of oxidant binding to the metal center trig-
gered by protonation of the secondary coordination sphere.
Chem. Eur. J. 2021, 27, 4700 –4708
www.chemeurj.org
4703
ꢀ 2021 Wiley-VCH GmbH

Full Paper
doi.org/10.1002/chem.202005183
Chemistry—A European Journal
to the amide carbonyl group. The attenuated electron-donat-
ing ability of the amide group of the ligand facilitates coordi-
nation of the peroxycarboxylate unit of mCPBA to the vacant
site of [NiII(dpap)(H)]2+. Both intermediates [NiII(dpap)(H)]2+
and [NiII(dpap)(mCPBA)]+ were observed by CSI-MS analysis
(Figure 4b). The entry point into the catalytic cycle is thus
gated by mCPBA functioning first as a Brønsted acid, and sub-
sequently as metal-bound oxidant (Scheme 2).
new feature at n = 1643 cmÀ1 observed by solution IR spec-
troscopy (Figure S8a). Similar spectral changes were observed
when 1 was treated with hard and Lewis acidic Sc3+ ion (deliv-
ered as Sc(OTf)3 salt) to function as a proton surrogate (Fig-
ure S7 and Figure S8b). Back-titration of protonated 1 with
Et3N, however, restored the original spectrum of 1 (Figure 5b
and Figure S6). HPLC-MS analysis on the reaction mixture (Fig-
ure S9) further established that the reversible changes in both
UV/Vis (Figure 5a) and 1H NMR spectra (Figure 5b and Fig-
ure S6) are not associated with protonation-induced demetalla-
tion of 1.
To follow this intriguing solution chemistry, we carried out
UV/Vis spectroscopic studies while adding HClO4 to the metal
complex and back-titrating with Et3N. As a proton donor,
perchloric acid was chosen to avoid complications from anion
coordinating to the metal center. As shown in Figure 5a, the
UV/Vis spectra of 1 in MeCN displayed an increase in the ab-
sorption at lmax = 250 nm with concomitant decrease at lmax
= 350 nm upon addition of up to 1 equiv HClO4. Subsequent
addition of Et3N (1 equiv) restored the original spectrum,
thus establishing the reversible nature of the protonation-
Low-temperature UV/Vis and EPR studies on the reaction in-
termediates
With the protonation-gated entry point to the catalytic cycle
established, we investigated the nature of the reactive species
developing from the initial reaction between the nickel(II) com-
plex and mCPBA (Figure 6). The UV/Vis spectrum of 1 in MeCN
showed weak bands at lmax = 540 nm (e = 30 cmÀ1 mÀ1) and
lmax = 800 nm (e = 35 cmÀ1 mÀ1). Upon adding 1 equiv of
mCPBA to the solution at À108C, thermally unstable yellow in-
termediate was formed, which exhibited intense absorption at
lmax = 430 nm (e = 1150 cmÀ1 mÀ1) and lmax = 890 nm (e =
440 cmÀ1 mÀ1) and subsequently decayed with t1/2 = ca. 200 s
(Figure 6a and Figure S10). The X-band EPR spectrum of 1 is
induced
interconversion
between
[NiII(dpap)]+
and
[NiII(dpap)(MeCN)(H)]2+ (Scheme 2). The presence of isosbestic
points is consistent with the equilibrium between two spectro-
scopically observable species.
Protonation-induced spectral changes of 1 were also fol-
1
lowed by H NMR spectroscopy (Figure 5b and Figure S6) and
solution IR spectroscopy (Figure S8a). Upon titration with up to
1 equiv of HClO4, the paramagnetically shifted proton resonan-
ces at d = 18–54 ppm showed significant shifts and broaden-
ing, in addition to the appearance of new signals. This process
is accompanied by the diminution of strong carbonyl stretch-
ing at n = 1606 cmÀ1, which correlates with the build-up of a
featureless (Figure S11), but an axial signal with g = 2.17 and
?
g
= 2.25 developed upon treatment with 1 equiv of mCPBA
k
Figure 6. (a) Time-dependent changes in the UV/Vis spectra to follow the
thermal decay of the intermediate (red line) generated from the reaction be-
tween 1 and mCPBA (1 equiv) in MeCN at À108C. Inset shows changes in
A890 (absorbance at l = 890 nm) as a function of time. (b) Plot of pseudo-
first-order rate constants (kobs) vs. the concentration of cyclohexene to deter-
Figure 5. (a) UV/Vis absorption and (b) 1H NMR spectral changes of 1 in
MeCN (concentration = 0.1 mm and 4.0 mm respectively) upon treatment
with HClO4 and back-titration with Et3N at T = 298 K.
mine the second-order rate constant k2 = 5.2(Æ0.2)”10À2 mÀ1 sÀ1
.
Chem. Eur. J. 2021, 27, 4700 –4708
www.chemeurj.org
4704
ꢀ 2021 Wiley-VCH GmbH

Full Paper
doi.org/10.1002/chem.202005183
Chemistry—A European Journal
(Figure S12), which is characteristic of nickel(III) species with
low-spin d7 (S = 1/2) electronic configuration.[12f,j,13a,22]
Unlike the chemistry of high-valent Fe or Mn, studies on
nickel-oxo species do not benefit much from the scarce
number of well-characterized examples of synthetic ori-
gin.[4g,12b] To probe the electronic structure and optical transi-
tions of the yellow intermediate generated from the reaction
between 1 and mCPBA, we carried out exploratory DFT and
TD-DFT computational studies. As summarized in Table S4, the
putative NiIII-oxo species (Figure S13) prefers the S = 1/2 spin-
state as the ground-state electronic configuration, which is
consistent with the experimentally observed EPR spectrum.
TD-DFT calculations on this species predicted multiple elec-
tronic transitions at around l = 440 nm, along with broad and
weak band at l = 800–1000 nm (Figure S14). While such spec-
tral features are consistent with the low-temperature UV/Vis
spectrum shown in Figure 6a, DFT models of high-spin NiIII-
oxo (S = 3/2) or NiIV-oxo (S = 1) also predicted similar elec-
tronic transitions (Figure S14).
Scheme 3. Reaction pathways of PPAA as mechanistic probe.
dation reaction of substrate by nickel-acylperoxo intermediate
affords PAA (pathway B). A homolytic OÀO bond cleavage of
Ni-OOC(=O)R produces a mixture of benzaldehyde (I), benzyl
alcohol (II), and toluene (III) through a rapid b-scission of the
acyloxy radical intermediate (pathway C).
We first conducted control experiments on the reaction of 1
with PPAA in the absence of the substrate (Table 3, entry 1). A
mixture of the heterolytic cleavage product, phenylacetic acid
(60.1%), and the homolytic cleavage products, benzaldehyde
(21.3%) and benzyl alcohol (2.9%), were produced. This result
indicates that nickel-acylperoxo intermediate underwent both
heterolytic cleavage (70%) and homolytic cleavage (30%) to
produce putative NiIV-oxo and NiIII-oxo species, respectively
(Scheme 3).
To investigate the oxidative reactivity of this yellow inter-
mediate, the reaction between 1 and mCPBA was carried out
in the presence of an increasing concentration of cyclohexene
substrate. The pseudo-first-order rate constants increased pro-
portionally with increasing the concentration of cyclohexene.
The second-order rate constant for this bimolecular decay pro-
cess was determined to be 5.2”10À2 mÀ1 À1 (Figure 6b).[23]
s
Collectively, our findings suggest that binding of mCPBA is
followed by oxidation of nickel(II) to generate reactive inter-
mediate that can be trapped by olefin substrate, although a
definitive assignment of the oxidation state and electronic con-
figuration of this intermediate remains currently elusive.
To probe the OÀO bond cleavage pattern in the presence of
olefin substrate, we employed cyclohexene as a reactive sub-
strate.[24c,d] Even when the concentration of cyclohexene was
increased from 0 to 160 mm, the ratios of the heterolysis to ho-
molysis reaction products remained essentially identical
(Table 3, entries 1–5). This observation suggests that the Ni-
OOC(=O)R intermediate was diverted into heterolytic cleavage
(70%) and homolytic cleavage (30%) pathways regardless of
the amount of cyclohexene substrate present in the reaction
mixture.
Mechanistic studies of oxidant activation by OÀO bond
cleavage
What reaction pathways are available for the initial adduct in
the catalytic cycle? As a mechanistic probe to address this
question, we employed peroxyphenylacetic acid (PPAA) as a
terminal oxidant in the epoxidation of olefin by 1. The chemi-
cal structure and relative distribution of the degradation prod-
ucts of PPAA help to understand the mechanism of OÀO bond
cleavage of the M-OOC(=O)R intermediate.[24] When heterolyt-
ic OÀO bond cleavage of Ni-OOC(=O)R occurs (Scheme 3,
pathway A), phenylacetic acid (PAA) is formed. The direct oxi-
To examine the effect of the type of the substrate, we
changed the substrate from cyclohexene to 1-octene, which is
Table 3. Analysis of the products derived from PPAA and 1 in the absence or the presence of cyclohexene.[a]
Entry
Cyclohexene
(mm)
Heterolysis[b]
PAA
Homolysis[b]
Heterolysis/homolysis
PAA/(I+II+III)
Oxidation products[c]
-one -ol
I
II
III
oxide
1
2
3
4
5
0
20
40
80
160
60.1Æ4.8
65.0Æ0.4
63.8Æ1.6
60.2Æ1.0
56.4Æ2.4
21.3Æ0.5
29.4Æ1.9
31.4Æ1.1
29.0Æ2.6
26.3Æ0.9
2.9Æ2.0
0.9Æ0.1
0.8Æ0.0
0.8Æ0.0
0.8Æ0.0
–
–
–
–
–
2.48
2.14
1.99
2.02
2.09
–
–
–
41.6Æ0.1[d]
26.5Æ0.3[b]
31.3Æ0.7[b]
37.0Æ0.5[b]
4.8Æ0.2[d]
2.7Æ0.0[b]
3.1Æ0.5[b]
3.6Æ0.3[b]
3.0Æ0.1[d]
2.0Æ0.0[b]
2.6Æ0.4[b]
3.4Æ0.3[b]
[a] Reaction conditions: substrate (0–0.16 mmol), 1 (0.0017 mmol), and PPAA (0.04 mmol) in MeCN (1 mL), T = 258C. [b] Based on PPAA; I–III indicate ben-
zaldehyde, benzyl alcohol, and toluene, respectively. [c] Oxide, -one and -ol denote cyclohexene oxide, 2-cyclohexen-1-one, and 2-cyclohexen-1-ol, respec-
tively. [d] Based on the substrate.
Chem. Eur. J. 2021, 27, 4700 –4708
www.chemeurj.org
4705
ꢀ 2021 Wiley-VCH GmbH

Full Paper
doi.org/10.1002/chem.202005183
Chemistry—A European Journal
more difficult to oxidize. A nearly identical ratio (70:30) of het-
erolysis to homolysis products was also observed in this case
as well (Table S5). These results suggest that 1 effects OÀO
bond cleavage of the metal-bound acylperoxo species through
a combination of heterolysis (70%) and homolysis (30%) path-
ways, irrespective of the concentration or the type of olefinic
substrate present. At the same time, the Ni-OOC(=O)R inter-
mediate species derived from PPAA is not a competent oxidant
for olefin epoxidation, or has a short lifetime.
Electronic nature of the reactive intermediate
To gain insights into the electronic nature of the reactive spe-
cies involved in the olefin epoxidation reaction, we carried out
competitive epoxidation reactions of various para-substituted
styrenes (4-X-C6H4CHCH2, X = OMe; Me; H; Cl; CN). The Ham-
mett analysis afforded a 1 value of À0.75 (Figure S15), which
indicates the electrophilic nature of the reactive species similar
to reported nickel complexes.[12c,g,j,25]
Scheme 4. Mechanism of proton-switched olefin epoxidation via branched
reaction pathways.
Isotope exchange studies
Based on the mechanistic studies with PPAA, the Ni-OOC(=
O)R species partitions into the heterolysis (70%) and homolysis
(30%) pathways to afford putative NiIV-oxo and NiIII-oxo spe-
cies, respectively, that can exchange with H218O. Presumably,
this process is facilitated by the strong s-donating amido-N
ligand in the trans-position (Scheme 2 and Scheme 4). Under-
standably, this branching in the reaction coordinate is not af-
fected by the presence of the olefin substrate, as previously
been observed.[12j] In contrast, certain nickel-acylperoxo species
of porphyrin ligand is competent to effect direct epoxidation
of olefins.[12c]
Isotope exchange experiments help understand the mecha-
nism of catalytic oxygenation reactions. If the high-valent
metal-oxo species is sufficiently long lived, its oxygen atom
could exchange with exogenously added H218O, so that the sol-
vent-derived oxygen atom ends up in the 18O-labelled oxygen-
ation products.[26] Epoxidation reaction of cyclohexene was
thus conducted with 1 and mCPBA in a mixture of distilled
MeCN and excess amount (55–220 equiv relative to mCPBA) of
H218O (Table S6). The percentage of 18O incorporated cyclohex-
ene oxide increased from 7.4% to 10.3% with the increasing
amount of H218O. This result suggests that a nickel species
having solvent-exchangeable oxo ligand is involved as a reac-
tive intermediate in the catalytic cycle.
Conclusions
A new monoamidate nickel(II) complex 1 was developed that
can efficiently epoxidize various olefins including terminal ole-
fins using mCPBA as a terminal oxidant. To stabilize high-
valent nickel-oxo species involved in oxygen-atom transfer, a
robust pentadentate chelate H-dpap was built, which functions
as a monoanionic ligand by deprotonation of the amide NÀH
group upon metallation.
Mechanism of olefin epoxidation
From the combination of X-ray crystallographic, spectroscopic,
kinetic, and mechanistic probe studies, a coherent mechanistic
picture emerges, as summarized in Scheme 4. Protonation of
the amide carbonyl group of 1 attenuates the electron-donat-
ing ability of the ligand, so that mCPBA binds as a ligand to
the vacant axial site of [NiII(dpap)]+ to furnish nickel-acylper-
oxo intermediate.
We found that protonation of the amide carbonyl group in-
creases the electronic demand of the metal center, so that
mCPBA is recruited as a ligand to initiate the catalytic cycle. A
combination of spectroscopic, isotope exchange, and product
analysis studies using PPAA as a mechanistic probe suggests
that OÀO bond cleavage of nickel-acylperoxo species gener-
ates high-valent nickel-oxo species for subsequent oxygen
atom transfer to olefin substrates. For terminal olefins, remark-
ably high catalytic efficiency and selectivity were observed,
which is highly unusual for a nickel-based system.
The self-regulating nature of this acid-base chemistry is
highlighted by a significant decrease in the substrate conver-
sion as well as the epoxidation product yield in the presence
of acid (HClO4, Table S7) or base (NH4OH, Table S8) additives.
Exogenously added proton would shift the acid-base equilibri-
um of mCPBA to suppress its deprotonation, thereby prevent-
ing the formation of nickel-acylperoxo intermediate. On the
other hand, Brønsted base OHÀ functions as a proton scaveng-
er to help convert mCPBA to its conjugate base. Without pro-
tonation of the ligand carbonyl group, however, m-chloroper-
benzoate cannot coordinate to the metal center, and would be
diverted away from the catalytic cycle for olefin epoxidation.
All chemical reactions can be classified into either substitu-
tion or oxidation-reduction.[27] As the smallest and ubiquitous
electrophile, the proton nicely couples the non-redox event in
the ligand sphere to the redox event at the metal center of 1.
It would be interesting to see if conceptually parallel processes
Chem. Eur. J. 2021, 27, 4700 –4708
www.chemeurj.org
4706
ꢀ 2021 Wiley-VCH GmbH

Full Paper
doi.org/10.1002/chem.202005183
Chemistry—A European Journal
Jitsukawa, T. Ozawa, H. Masuda, Eur. J. Inorg. Chem. 2006, 3753–3761;
c) M. Prejanò, T. Marino, C. Rizzuto, J. C. Madrid, N. Russo, M. Toscano,
Inorg. Chem. 2017, 56, 13390–13400.
also occur in the metalloenzymes involved in group-transfer
chemistry to shuttle the system between the off-cycle and on-
cycle using a proton switch installed at the peptidyl backbone.
[9] a) D. P. Barondeau, C. J. Kassmann, C. K. Bruns, J. A. Tainer, E. D. Getzoff,
Biochemistry 2004, 43, 8038–8047; b) J. Wuerges, J.-W. Lee, Y.-I. Yim, H.-
S. Yim, S.-O. Kang, K. D. Carugo, Proc. Natl. Acad. Sci. USA 2004, 101,
8569–8574.
[10] M. J. Maroney, S. Ciurli, Chem. Rev. 2014, 114, 4206–4228.
[11] a) J. Shearer, Acc. Chem. Res. 2014, 47, 2332–2341; b) J. O. CampeciÇo,
L. W. Dudycz, D. Tumelty, V. Berg, D. E. Cabelli, M. J. Maroney, J. Am.
Chem. Soc. 2015, 137, 9044–9052.
Experimental Section
For the full experimental details see the Supporting Information.
Deposition number 2015122 contains the supplementary crystallo-
graphic data for this paper. These data are provided free of charge
by the joint Cambridge Crystallographic Data Centre and Fachin-
[12] a) M. Sankaralingam, M. Balamurugan, M. Palaniandavar, P. Vadivelu,
C. H. Suresh, Chem. Eur. J. 2014, 20, 11346–11361; b) T. Corona, A. Com-
pany, Chem. Eur. J. 2016, 22, 13422–13429; c) H. M. Ahn, J. M. Bae, M. J.
Kim, K. H. Bok, H. Y. Jeong, S. J. Lee, C. Kim, Chem. Eur. J. 2017, 23,
11969–11976; d) S. K. Padamati, D. Angelone, A. Draksharapu, G. Primi,
D. J. Martin, M. Tromp, M. Swart, W. R. Browne, J. Am. Chem. Soc. 2017,
139, 8718–8724; e) M. Sankaralingam, M. Balamurugan, M. Palanianda-
var, Coord. Chem. Rev. 2020, 403, 213085; f) F. F. Pfaff, F. Heims, S.
Kundu, S. Mebs, K. Ray, Chem. Commun. 2012, 48, 3730–3732; g) T.
Corona, F. F. Pfaff, F. AcuÇa-ParØs, A. Draksharapu, C. J. Whiteoak, V.
Martin-Diaconescu, J. Lloret-Fillol, W. R. Browne, K. Ray, A. Company,
Chem. Eur. J. 2015, 21, 15029–15038; h) T. Corona, A. Draksharapu, S. K.
Padamati, I. Gamba, V. Martin-Diaconescu, F. AcuÇa-ParØs, W. R. Browne,
A. Company, J. Am. Chem. Soc. 2016, 138, 12987–12996; i) P. Pirovano,
E. R. Farquhar, M. Swart, A. R. McDonald, J. Am. Chem. Soc. 2016, 138,
14362–14370; j) K. H. Bok, M. M. Lee, G. R. You, H. M. Ahn, K. Y. Ryu, S.-J.
Kim, Y. Kim, C. Kim, Chem. Eur. J. 2017, 23, 3117–3125; k) J. Singh, G.
Hundal, R. Gupta, Eur. J. Inorg. Chem. 2008, 2052–2063.
formationszentrum
www.ccdc.cam.ac.uk/structures.
Karlsruhe
Access
Structures
service
Acknowledgements
This work was supported by the National Research Foundation
(NRF) of Korea (2020R1A2C2006381) funded by the Ministry of
Science and ICT (MSIT). The work at SNUST and UNIST was sup-
ported
by
NRF
(2018R1A2B6001686)
and
CGRC
(2016M3D3A01913243), respectively.
Conflict of interest
[13] a) A. R. Jeong, J. W. Shin, J. H. Jeong, K. H. Bok, C. Kim, D. Jeong, J. Cho,
S. Hayami, K. S. Min, Chem. Eur. J. 2017, 23, 3023–3033; b) D. Chatter-
jee, S. Mukherjee, A. Mitra, J. Mol. Catal. A 2000, 154, 5–8; c) M. T. Risp-
ens, O. J. Gelling, A. H. M. de Vries, A. Meetsma, F. van Bolhuis, B. L. Fer-
inga, Tetrahedron 1996, 52, 3521–3546.
[14] a) J. P. Collman, Z. Wang, A. Straumanis, M. Quelquejeu, J. Am. Chem.
Soc. 1999, 121, 460–461; b) H. Tian, X. She, J. Xu, Y. Shi, Org. Lett. 2001,
3, 1929–1931; c) M. C. White, A. G. Doyle, E. N. Jacobsen, J. Am. Chem.
Soc. 2001, 123, 7194–7195; d) A. Murphy, T. D. P. Stack, J. Mol. Catal. A:
Chem. 2006, 251, 78–88; e) I. Garcia-Bosch, X. Ribas, M. Costas, Adv.
Synth. Catal. 2009, 351, 348–352.
The authors declare no conflict of interest.
Keywords: bioinorganic chemistry · enzyme models
epoxidation · ligand design · nickel
·
[1] a) M. Costas, M. P. Mehn, M. P. Jensen, L. Que, Jr., Chem. Rev. 2004, 104,
939–986; b) A. Decker, E. I. Solomon, Curr. Opin. Chem. Biol. 2005, 9,
152–163; c) P. R. Ortiz de Montellano, Chem. Rev. 2010, 110, 932–948;
d) A. J. Jasniewski, L. Que, Jr., Chem. Rev. 2018, 118, 2554–2592; e) H.
Park, D. Lee, Chem. Eur. J. 2020, 26, 5916–5926.
[2] a) E. A. Lewis, W. B. Tolman, Chem. Rev. 2004, 104, 1047–1076; b) M. Bor-
deaux, A. Galarneau, J. Drone, Angew. Chem. Int. Ed. 2012, 51, 10712–
10723; Angew. Chem. 2012, 124, 10870–10881; c) S. Sirajuddin, A. C.
Rosenzweig, Biochemistry 2015, 54, 2283–2294.
[3] a) V. L. Pecoraro, M. J. Baldwin, A. Gelasco, Chem. Rev. 1994, 94, 807–
826; b) R. Manchanda, G. W. Brudvig, R. H. Crabtree, Coord. Chem. Rev.
1995, 144, 1–38; c) S. Sahu, D. P. Goldberg, J. Am. Chem. Soc. 2016, 138,
11410–11428.
[15] H. Park, H. M. Ahn, H. Y. Jeong, C. Kim, D. Lee, Chem. Eur. J. 2018, 24,
8632–8638.
[16] a) S. Hong, Y.-M. Lee, M. Sankaralingam, A. K. Vardhaman, Y. J. Park, K.-B.
Cho, T. Ogura, R. Sarangi, S. Fukuzumi, W. Nam, J. Am. Chem. Soc. 2016,
138, 8523–8532; b) D. B. Rice, E. N. Grotemeyer, A. M. Donovan, T. A.
Jackson, Inorg. Chem. 2020, 59, 2689–2700; c) S.-S. Xue, X.-X. Li, Y.-M.
Lee, M. S. Seo, Y. Kim, S. Yanagisawa, M. Kubo, Y.-K. Jeon, W.-S. Kim, R.
Sarangi, S. H. Kim, S. Fukuzumi, W. Nam, J. Am. Chem. Soc. 2020, 142,
15305–15319.
[4] a) M. H. Lim, J.-U. Rohde, A. Stubna, M. R. Bukowski, M. Costas, R. Y. N.
Ho, E. Münck, W. Nam, L. Que, Jr., Proc. Natl. Acad. Sci. USA 2003, 100,
3665–3670; b) A. S. Borovik, Chem. Soc. Rev. 2011, 40, 1870–1874;
c) H. M. Neu, R. A. Baglia, D. P. Goldberg, Acc. Chem. Res. 2015, 48,
2754–2764; d) X. Engelmann, I. Monte-PØrez, K. Ray, Angew. Chem. Int.
Ed. 2016, 55, 7632–7649; Angew. Chem. 2016, 128, 7760–7778;
e) B. E. R. Snyder, M. L. Bols, R. A. Schoonheydt, B. F. Sels, E. I. Solomon,
Chem. Rev. 2018, 118, 2718–2768; f) M. Guo, T. Corona, K. Ray, W. Nam,
ACS Cent. Sci. 2019, 5, 13–28; g) V. A. Larson, B. Battistella, K. Ray, N.
Lehnert, W. Nam, Nat. Rev. Chem. 2020, 4, 404–419.
[5] a) M. Sono, M. P. Roach, E. D. Coulter, J. H. Dawson, Chem. Rev. 1996, 96,
2841–2888; b) J. N. Rodríguez-López, D. J. Lowe, J. Hernµndez-Ruiz,
A. N. P. Hiner, F. García-Cµnovas, R. N. F. Thorneley, J. Am. Chem. Soc.
2001, 123, 11838–11847; c) J. S. de Ropp, S. Sham, A. Asokan, S. L. New-
myer, P. R. Ortiz de Montellano, G. N. La Mar, J. Am. Chem. Soc. 2002,
124, 11029–11037.
[17] C. Incarvito, A. L. Rheingold, A. L. Gavrilova, C. J. Qin, B. Bosnich, Inorg.
Chem. 2001, 40, 4101–4108.
[18] a) D. M. L. Goodgame, D. J. Williams, R. E. P. Winpenny, Polyhedron 1989,
8, 1913–1921; b) J.-P. Costes, F. Dahan, B. Donnadieu, M.-J. Rodriguez
Douton, M.-I. Fernandez Garcia, A. Bousseksou, J.-P. Tuchagues, Inorg.
Chem. 2004, 43, 2736–2744.
[19] a) C. Bousquet, D. G. Gilheany, Tetrahedron Lett. 1995, 36, 7739–7742;
b) R. V. Ottenbacher, D. G. Samsonenko, E. P. Talsi, K. P. Bryliakov, ACS
Catal. 2016, 6, 979–988.
[20] a) W. Adam, C. Mock-Knoblauch, C. R. Saha-Mçller, M. Herderich, J. Am.
Chem. Soc. 2000, 122, 9685–9691; b) C. Linde, M. F. Anderlund, B. Šker-
mark, Tetrahedron Lett. 2005, 46, 5597–5600.
[21] a) S. H. van Rijt, A. J. Hebden, T. Amaresekera, R. J. Deeth, G. J. Clarkson,
S. Parsons, P. C. McGowan, P. J. Sadler, J. Med. Chem. 2009, 52, 7753–
7764; b) R. M. Lord, S. J. Allison, K. Rafferty, L. Ghandhi, C. M. Pask, P. C.
McGowan, Dalton Trans. 2016, 45, 13196–13203.
[22] R. I. Haines, A. McAuley, Coord. Chem. Rev. 1981, 39, 77–119.
[23] While the involvement of multiple reaction pathways (Schemes 3 and
4) makes the interpretation of the rate constant k2 less than straightfor-
ward, the linear relationship between kobs and [olefin] unambiguously
establishes the bimolecular nature of the reaction mechanism.
[6] C. Panda, A. Sarkar, S. Sen Gupta, Coord. Chem. Rev. 2020, 417, 213314.
[7] a) M. Kobayashi, S. Shimizu, Curr. Opin. Chem. Biol. 2000, 4, 95–102;
b) A. Miyanaga, S. Fushinobu, K. Ito, T. Wakagi, Biochem. Biophys. Res.
Commun. 2001, 288, 1169–1174.
[8] a) A. Miyanaga, S. Fushinobu, K. Ito, H. Shoun, T. Wakagi, Eur. J. Biochem.
2004, 271, 429–438; b) T. Yano, H. Arii, S. Yamaguchi, Y. Funahashi, K.
Chem. Eur. J. 2021, 27, 4700 –4708
www.chemeurj.org
4707
ꢀ 2021 Wiley-VCH GmbH

Full Paper
doi.org/10.1002/chem.202005183
Chemistry—A European Journal
[24] a) T. G. Traylor, S. Tsuchiya, Y. S. Byun, C. Kim, J. Am. Chem. Soc. 1993,
115, 2775–2781; b) W. Nam, H. J. Choi, H. J. Han, S. H. Cho, H. J. Lee, S.-
Y. Han, Chem. Commun. 1999, 387–388; c) N. Suzuki, T. Higuchi, T.
Nagano, J. Am. Chem. Soc. 2002, 124, 9622–9628; d) M. Y. Hyun, Y. D.
Jo, J. H. Lee, H. G. Lee, H. M. Park, I. H. Hwang, K. B. Kim, S. J. Lee, C.
Kim, Chem. Eur. J. 2013, 19, 1810–1818.
[26] a) K. A. Lee, W. Nam, J. Am. Chem. Soc. 1997, 119, 1916–1922; b) K.
Nehru, S. J. Kim, I. Y. Kim, M. S. Seo, Y. Kim, S.-J. Kim, J. Kim, W. Nam,
Chem. Commun. 2007, 4623–4625.
[27] H. Taube, Science 1984, 226, 1028–1036.
Manuscript received: December 2, 2020
Accepted manuscript online: January 11, 2021
Version of record online: February 5, 2021
[25] J. Nakazawa, S. Terada, M. Yamada, S. Hikichi, J. Am. Chem. Soc. 2013,
135, 6010–6013.
Chem. Eur. J. 2021, 27, 4700 –4708
www.chemeurj.org
4708
ꢀ 2021 Wiley-VCH GmbH