Computational chemical analysis of Ru(II)-Pheox–catalyzed highly enantioselective intramolecular cyclopropanation reactions

Article abstract of DOI:10.1002/chir.23033

Computational chemical analysis of Ru(II)-Pheox–catalyzed highly enantioselective intramolecular cyclopropanation reactions was performed using density functional theory (DFT). In this study, cyclopropane ring–fused γ-lactones, which are 5.8?kcal/mol more stable than the corresponding minor enantiomer, are obtained as the major product. The results of the calculations suggest that the enantioselectivity of the Ru(II)-Pheox–catalyzed intramolecular cyclopropanation reaction is affected by the energy differences between the starting structures 5l and 5i. The reaction pathway was found to be a stepwise mechanism that proceeds through the formation of a metallacyclobutane intermediate. This is the first example of a computational chemical analysis of enantioselective control in an intramolecular carbene-transfer reaction using C₁-symmetric catalysts.

Full text of DOI:10.1002/chir.23033

Received: 25 June 2018
DOI: 10.1002/chir.23033
Revised: 16 October 2018
Accepted: 16 October 2018
R E G U L A R A R T I C L E
Computational chemical analysis of Ru(II)‐Pheox–catalyzed
highly enantioselective intramolecular cyclopropanation
reactions
1
2
3
1
Yoko Nakagawa
| Naofumi Nakayama | Hitoshi Goto
| Ikuhide Fujisawa |
1
1
1
Soda Chanthamath | Kazutaka Shibatomi | Seiji Iwasa
1
Department of Environmental and Life
Abstract
Sciences, Toyohashi University of
Technology, Toyohashi, Japan
Computational chemical analysis of Ru(II)‐Pheox–catalyzed highly
enantioselective intramolecular cyclopropanation reactions was performed
using density functional theory (DFT). In this study, cyclopropane ring–fused
γ‐lactones, which are 5.8 kcal/mol more stable than the corresponding minor
enantiomer, are obtained as the major product. The results of the calculations
suggest that the enantioselectivity of the Ru(II)‐Pheox–catalyzed intramolecu-
lar cyclopropanation reaction is affected by the energy differences between
the starting structures 5l and 5i. The reaction pathway was found to be a step-
wise mechanism that proceeds through the formation of a metallacyclobutane
intermediate. This is the first example of a computational chemical analysis of
enantioselective control in an intramolecular carbene‐transfer reaction using
C1‐symmetric catalysts.
2
CONFLEX Corporation, Tokyo, Japan
3
Department of Computer Science and
Engineering, Toyohashi University of
Technology, Toyohashi, Japan
Correspondence
Seiji Iwasa, Department of Environmental
and Life Sciences, Toyohashi University of
Technology, 1‐1 Tempaku‐cho, Toyohashi,
Aichi 441‐8580, Japan.
Email: iwasa@ens.tut.ac.jp
Funding information
JSPS KAKENHI, Grant/Award Numbers:
16J06159, 17H06373 and 26288087
KEYWORDS
asymmetric synthesis, chiral catalysts, chiral ligands, DFT calculation, mechanistic study, Ru(II)‐
Pheox
1
| INTRODUCTION
catalysts are able to highly control the enantioselectivity
of asymmetric cyclopropanation reactions.
Optically active cyclopropane compounds are a common
motif in natural products, medicinal drugs, and agricul-
In 2010, we developed a novel chiral complex, Ru(II)‐
phenyloxazoline (Ru(II)‐Pheox) complex, which has a
1,2
6-17
tural chemicals.
Intermolecular and intramolecular
unique C ‐symmetric structure
(Figure 1). Ru(II)‐
1
cyclopropanation reactions have been widely explored to
Pheox can be easily synthesized from a commercially
available benzoic acid derivative, an amino alcohol and a
Ru source, and effectively promotes enantioselective inter-
molecular and intramolecular cyclopropanations with
electron‐rich olefins, as well as electron‐deficient olefins,
under mild reaction conditions. Moreover, we have
reported Ru(II)‐Pheox–catalyzed highly enantioselective
give optically active cyclopropane derivatives via metal‐
3,4
carbene complexes. The most popular catalysts used
in asymmetric cyclopropanations are copper, rhodium,
ruthenium, and cobalt complexes. In this vast literature,
C2‐symmetric catalysts with large steric hindrance are
generally believed to be superior to simpler C ‐symmetric
1
5
3
18
catalysts with single chirality for asymmetric induction.
carbene insertion reactions into sp Si–H single bonds.
These results suggested that our catalytic system could
However, our recent study has shown that C ‐symmetric
1
Chirality. 2018;1–10.
wileyonlinelibrary.com/journal/chir
© 2018 Wiley Periodicals, Inc.
1

2
NAKAGAWA ET AL.
FIGURE 1 Ru(II)‐Pheox catalyst
FIGURE 2 X‐ray structure of the major cyclopropane‐fused γ‐
lactone (Table 1, entry 4) via intramolecular cyclopropanation
also be applied to carbene transfer reactions for σ‐bond
formation with high enantioselectivities and yields. How-
ever, to date, the asymmetric induction mechanism in
carbene transfer reactions catalyzed by the Ru(II)‐Pheox
complex has not been clarified.
computational studies of a variety of transition‐metal cat-
alysts have been performed, mainly using density func-
19-33
tional theory (DFT) methods.
In 2005, Cornejo and
An interesting example of the relationship between the
design of a catalyst and its enantioselectivity toward the
desired product is the intramolecular cyclopropanation
of allyl diazoacetate derivatives. We have reported that
Ru(II)‐Pheox is a highly efficient catalyst, affording chiral
cyclopropane–fused γ‐lactones in high yield (up to 98%
coworkers reported a computational mechanistic study
of asymmetric intermolecular cyclopropanation reactions
catalyzed by a pybox‐Ru catalyst (pybox = bis(oxazolinyl)
34
pyridine).
The theoretical results revealed that
cyclopropanation by the pybox‐Ru catalyst proceeded by
direct three‐membered ring formation. The origin of the
enantioselectivity was found to be due to both intermo-
lecular steric interactions between the styrene and the
ligand and intramolecular steric interactions between
the ester and the ligand. In 2011, Rosenberg and
coworkers reported a theoretical study on highly cis‐
diastereoselective intermolecular cyclopropanations
16
yield) and excellent enantioselectivity (up to 99% ee)
(Tables 1 and S1). The absolute stereochemistry of the
major product was determined to be (1R, 5R, 6R) by single
X‐ray crystal analysis (Figure 2 and the Supporting Infor-
mation). On the basis of these experimental results, we
envisioned that the enantioselectivity was induced by the
steric hindrance of the phenyl group of the catalyst at the
chiral carbon to form the favored transition state. How-
ever, Ru(II)‐Pheox has three vacant quadrants around
the ruthenium atom. Therefore, investigations into how
35
using a rhodium(I) catalyst. They explained that the
selectivity was probably due to unfavorable steric interac-
tions between the styrene phenyl group (the reactant)
and the methyl groups at the ortho‐positions of the ligand
N‐aryl ring. In addition, the results of this computational
and why this C ‐symmetric Ru(II)‐Pheox catalyst is able
1
to afford optically active cyclopropane compounds with
almost perfect enantioselectivity are of high interest.
To explain the observed product stereochemistry
and improve the efficiency of the catalytic systems,
study showed that the reaction proceeds via a
stepwise mechanism, involving the formation of Rh‐
metallacyclobutanes. In 2015, Xue and coworkers
reported a DFT study on the Rh (S‐PTTL) –catalyzed
2
4
3
6
intermolecular cyclopropanation of diazooxindole. They
suggested that the high trans‐diastereoselectivity might
be influenced by the π‐π interactions and steric interac-
tions between the substituents of the catalyst and the
phenyl group of styrene. An example of a computational
study that analyzed asymmetric induction in intramolec-
ular cyclopropanations was reported by Xu and
TABLE 1 Ru(II)‐Pheox–catalyzed intramolecular
cyclopropanation
37
coworkers. Using a Ru(salen) complex (salen = N,N′‐
a
b
Entry
R
Time
Yield , %
ee , %
1
2
bis(3‐R ‐5‐R ‐salicylidene)) model, they explained that
the observed variation in enantioselectivity resulted from
the equilibrium between cis‐ and trans‐ruthenium‐
carbene intermediates. However, systems with
1
2
3
4
H
1 min
1 min
1 min
1 min
24 h
66
98
89
90
75
99
96
98
99
99
6 5
C H
3 3
CON(CH )OCH
carbon–metal single bonds and C ‐symmetry, such as
1
COOCH
COOCH
3
Ru(II)‐Pheox, have never been analyzed theoretically.
The information obtained by computational chemical
analysis about the reaction mechanism, asymmetric
induction, and substituent effects will prompt the design
of more efficient catalysts.
c
5
3
a
Isolated yield.
b
Determined by chiral high‐performance liquid chromatography (HPLC).
CN was used as the solvent.
c
CH
3

NAKAGAWA ET AL.
3
Herein, we focus on two aspects of the Ru(II)‐Pheox–
2.3 | Preparation of allyl 2‐diazoacetate
catalyzed asymmetric intramolecular cyclopropanation
reactions: (1) the reaction mechanism and (2) the mecha-
nism of asymmetric induction. The calculations were
performed by means of a theoretical approach based on
DFT calculations using an allyl diazoacetate as a simpli-
fied model, initially used to optimize the metal‐carbene
complexes formed from Ru(II)‐Pheox and the allyl
diazoacetate.
A
mixture of allyl 2‐bromoacetate and N,N′‐
ditosylhydrazine (3.06 g, 9 mmol, 1.5 equiv) in tetrahy-
drofuran (THF) (18 mL) was cooled down to 0°C. Then
DBU (1.8 mL, 12 mmol, 2 equiv) was added dropwise,
and the mixture was stirred at 0°C for 15 minutes. After
quenching with aqueous NaHCO , the product was
extracted with diethyl ether three times, and the organic
phase was dried over Na SO and evaporated to give the
3
2
4
2
2
| MATERIALS AND METHODS
.1 | General
crude product. Purification was performed with column
chromatography on silica gel with Hex/EA = 9/1 to give
allyl 2‐diazoacetate as a yellow oil (440.4 mg, 3.2 mmol,
1
5
8% yield). H NMR (500 MHz, CDCl , δ): 5.93
3
3
3
3
All reactions were performed under an argon atmosphere
unless otherwise noted. Dichloromethane (CH Cl ) was
purchased from Kanto Chemical Co, Inc. All reactions
were monitored by thin‐layer chromatography (TLC),
using glass plates precoated with silica gel Merck KGaA
(ddt, J
= 16.05 Hz, J = 10.32 Hz, J = 5.73 Hz,
trans cis
3
2
2
1H, –CH CH¼CH ), 5.32 (brd, J
2
2
trans
= 17.20 Hz, 1H, –
3
CH¼CH ), 5.24 (brd, J = 10.32 Hz, 1H, –CH¼CH ),
2
cis
2
3
4.78 (brs, 1H, –CO–CH¼N ), 4.66 (dd, J = 5.73 Hz,
2
2
J
= 1.15 Hz, 2H, –CO–O–CH –) ppm (Figure S2).
geminal
2
13
6
0 F254, layer thickness 0.2 mm. The products were visu-
C NMR (100 MHz, CDCl ): δ 166.3 (–CO–O–), 132.0
3
alized by irradiation with UV light or by treatment with a
solution of p‐anisaldehyde. Flash column chromatogra-
phy was performed using silica gel (Merck, Art No.
(–CH –CH¼CH ), 118.1 (–CH –CH¼CH ), 65.2 (–O–
2
2
2
2
CH –CH¼CH ), 46.1(–CO–CH¼N ) ppm (Figure S3).
2
2
2
1
13
7
734). H NMR (500 MHz) and C NMR (125 MHz) spec-
tra were recorded on JEOL JNM‐ECX500 and JEOL JNM‐
ECS400 spectrometers. Chemical shifts are reported as δ
values (ppm) relative to CDCl (7.26 ppm). Enantiomeric
3
excesses were determined by high‐performance liquid
chromatography (HPLC) analyses with a JASCO GULLI-
VER using Daicel CHIRALPAK or CHIRALCEL
columns.
2.4 | Ru(II)‐Pheox–catalyzed
intramolecular cyclopropanation of allyl 2‐
diazoacetate
A solution of allyl 2‐diazoacetate (38.3 mg, 0.304 mmol, 1
2.2 | Preparation of allyl 2‐bromoacetate
equiv) in CH Cl (3 mL) was added to a mixture of
2
2
A suspension of allyl alcohol (0.68 mL, 10 mmol, 1 equiv)
and K CO (2.76 g, 10 mmol, 2 equiv) were dissolved in
Ru(II)‐Pheox catalyst in CH Cl (1 mL) under argon
2 2
atmosphere at room temperature (Table 1, entry 1). The
reaction mixture was stirred for 1 minute, and then the
solvent was removed under reduced pressure. Purifica-
tion was performed by column chromatography to give
(1S,5R)‐3‐oxabicyclo[3.1.0]hexan‐2‐one as a colorless oil
2
3
CH Cl2 (10 mL) and cooled down to 0°C. Then
2
bromoacetyl bromide (1.3 mL, 15 mmol, 1.5 equiv) was
added dropwise and stirred for 2 hours. After quenched
with H O (1 mL) and extracting with CH Cl three times,
2
2
2
2
5
the organic phase was dried over Na SO and evaporated
(19.6 mg, 0.2 mmol, 66% yield, 99% ee). [α]
D
= −60.54
, δ): 4.33
(dd, Jgeminal = 9.46 Hz, Jcis = 4.58 Hz, 1H, –O–CH –),
–), 2.18
–CH–CO–O–), 2.03 to 2.08 (m, 1H,
–O–), 1.22 to 1.28 (m, 1H, –CH–CH
CH–), 0.84 to 0.88 (m, 1H, –CH–CH –CH–) ppm (Figure
): δ 176.4 (–CO–O–),
–CH–CO–), 17.3 (–CH
–CH–) ppm (Figure S5).
The ee value was determined by chiral HPLC analysis.
Column (Chiral IC), UV detector 220 nm,
Hex/IPA = 7/3, flow late = 1.2 mL/min, tR = 15.5 minutes
2
4
1
to give the crude product. Purification was performed by
(c = 0.88 in CHCl ). H NMR (500 MHz, CDCl
3 3
2
3
column chromatography to give allyl 2‐bromoacetate as
2
1
2
a brown oil (1.67 g, 9.34 mmol, 93% yield). H NMR
4.21 (brd,
to 2.24 (m, 1H, –CH
–CH –CH–CH
J
geminal = 9.16 Hz, 1H, –O–CH
2
3
(
500 MHz, CDCl , δ): 5.93 (ddt, Jtrans = 16.05 Hz,
2
3
3
3
Jcis = 10.32 Hz, J = 5.73 Hz, 1H, –CH CH¼CH ), 5.37
2
2
2
–
2
2
3
2
(
dd, J
= 17.20 Hz, J
= 1.53 Hz, 1H, –CH¼C-
geminal
2
trans
geminal
3
2
13
H ), 5.29 (dd, J = 10.70 Hz, J
= 1.53, 1H, –
S4). C NMR (100 MHz, CDCl
69.4 (–CO–O–CH –), 17.5 (–CH
CH–CH –O–), 12.2 (–CH–CH
3
2
cis
3
CH¼CH ), 4.67 (brd, J = 5.73 Hz, 2H, –CH –O–CO–),
2
2
2
–
2
2
3
.86 (s, 2H, –CO–CH –Br) ppm (Figure S1).
2
2
2
=

4
NAKAGAWA ET AL.
(
(
minor product), tR = 16.2 minutes (major product)
Figures S6 and S7).
structure 1 that is an experimental model used as the
reactant in the Ru(II)‐Pheox–catalyzed cyclopropanation
experiments (Table 1, entry 1 and Figure 4).
Ru(II)‐Pheox–catalyzed intramolecular cyclopropanations
of allyl diazoacetate derivatives (Table 1, entries 2‐5):
These results were published in our previous report.
The most stable structure of 1 was found by scanning
the dihedral angles Ru–C–C¼O, O¼C–O–C, C–O–C–C,
and O–C–C¼C. All the structures were optimized by
6,14
3
9
DFT calculations using the M06‐2X hybrid functional.
We also calculated total energies, zero‐point correction,
and thermal‐free energies of S1 using both (i) M06 func-
tion with the same basis set (Table S10) and (ii) M06‐2X
function with 6‐311+G(2d,p) and LanL2DZ (Table S11)
for checking the reliability of this calculation. These
results are almost the same as that using M06‐2X with 6‐
3
| CALCULATION METHOD
31G(d) and LanL2DZ and suggested that there were little
We describe here the reaction mechanism from the
metal‐carbene complex to the cyclopropane product,
except for the analysis of the formation of the metal‐
carbene complex. It is generally accepted that metal‐
catalyzed cyclopropanation reactions proceed via the
formation of metal‐carbene species from transition metal
complexes and a carbene precursor. In 2002, Staub
reported the detailed mechanism of formation of a
metal‐carbene complex with an extruded nitrogen from
differences on computational accuracy. Therefore,
40-42
throughout this work, we used the LanL2DZ
set for Ru and 6‐31G(d) for C, H, N, and O. To
validate the cyclopropanation scheme of with
basis
4
3
1
simpler calculations, solvent effects were applied with a
polarizable continuum model (PCM) for the geometry
optimization of the minima corresponding to the reactant
and the product, transition state searches between them,
and the intrinsic reaction coordinate (IRC) calculations
for verifying their relationship. All the calculations were
38
a diazo compound and a transition metal complex.
4
4
Initially, we decided to investigate the structural sta-
bility of suitable metal‐carbene complexes, which would
then be used as models for further computational analy-
ses. To determine the coordination position of the α,β‐
unsaturated carbonyl diazoacetate in the Ru(II)‐Pheox
catalyst, we simplified the complex structure to S1 and
S2 and performed calculations considering four possibili-
ties (A, B, C, and C′ in Figure 3). Then, we chose a
performed using the Gaussian 09 program package.
4
| RESULTS AND DISCUSSION
The most stable coordination isomer of S1 and S2 was
metal‐carbene complex with coordination of the ester
group in (A) (Figures S11 and S12 and Tables S8 and S9,
and S12 and S13). The stability of the metal‐carbene com-
plex S1 and S2 may arise from intramolecular π‐electronic
effects with the phenyl group of the catalyst, such as
CH⋯π interactions between the phenyl group of the cat-
alyst and the α‐proton of the Ru(II)‐Pheox complex and
π‐π interactions between the phenyl group and the ester
carbonyl group, instead of the destabilization arising from
steric hindrance. We then proceeded to perform a compu-
tational analysis under the assumption that the ester
group is stably coordinated on the same side as the phenyl
group of the Ru(II)‐Pheox catalyst ((A) position).
FIGURE 3 Possible coordination isomers of diazoacetate binding
to the Ru(II)‐Pheox complex
4.1 | Conformation search to find the
most stable structure of 1
We found that the catalytic asymmetric intramolecular
cyclopropanation of the diazo ester was proceeding
rapidly in CH Cl in the presence of Ru‐Pheox catalyst
2
2
(
Table 1, entry 4). On the other hand, the
cyclopropanation required 24 hours in CH CN (Table 1,
3
FIGURE 4 Calculation model of the metal‐carbene complex
entry 5), and both reaction conditions gave almost the

NAKAGAWA ET AL.
5
same high enantioselectivities. Since the elimination of
TABLE 2 Relative energies and dihedral angles of the metal‐
carbene complexes 1a‐1m
CH CN from Ru(II)‐Pheox complex in the presence of
3
1
CH Cl was observed by H NMR (Figures S8 and S9),
2
2
Dihedral Angle, °
Relative
CH CN might be coordinated continuously to Ru in
3
Energy,
Conformer kcal/mol C¼O
Ru¼C– O¼C– (O¼)C– O–C–
O–C O–C–C
C¼C
CH CN as a solvent resulting in low catalytic activity
3
than in CH Cl . However, CH CN of Ru‐Pheox as a
2
2
3
1
a
16.82
16.53
19.59
20.57
10.56
10.23
11.40
10.91
5.82
−8.7
172.2 −65.6
129.0
ligand could be easily release into CH Cl to give vacant
2
2
site on Ru to react with diazo ester resulting high cata-
lytic activity. On the basis of these experimental results,
1b
−8.0
174.8 −67.4
−10.2
−133.3
−0.5
1c
1d
1e
1f
−12.7
−13.0
5.5
165.3
168.6
83.5
80.3
we expected that the effect of CH CN around Ru is no
3
influence or small influence for the reaction mechanism.
We began to investigate 1 by carrying out a dihedral
angle scan of the Ru–C–C¼O, O¼C–O–C, C–O–C–C,
and O–C–C¼C fragments without coordination of aceto-
nitrile (Figure 5 and Table 2).
The calculated structures 1a to 1h present a cis‐Ru–C–
C¼O dihedral angle, whereas 1i to 1m correspond to
trans‐Ru–C–C¼O structures. The trans‐Ru–C–C¼O
metal‐carbene complexes were apparently more stable
than the cis‐Ru–C–C¼O structures, as shown in Table 2.
The terminal C¼C bond located near the ruthenium
atom acts as a reactive site in the structure of the trans‐
Ru–C–C¼O metal‐carbene complexes. The stabilities are
likely influenced by the coordination of the C¼C bond
to the ruthenium atom.
176.2 −74.6
177.2 −70.4
127.8
−4.1
0.8
1g
5.3
−172.4
−174.4
68.0
67.2
69.4
97.2
88.6
67.2
72.9
−125.5
9.8
1h
4.4
1
1
1
1
i
−164.9
−154.8
−166.8
−177.4
−166.5
−87.6
−27.4
88.3
j
8.00
31.3
k
l
6.53
64.9
0.00
166.7
−4.2
1m
9.08
179.6 −79.1
14.4
1i was also on the same side (A) as the phenyl group of
the catalyst, forming a right‐handed spiral complex as a
minor structure. The structure of 1i displaying a right‐
handed spiral was found to be 5.8 kcal/mol higher in
energy than that of 1l with a left‐handed spiral, the major
structure (Figure 6). These results indicate that the
As a result, the metal carbene complex 1l with a left‐
handed spiral structure was confirmed to be the most sta-
ble metal‐carbene complex (Figure 6). To explain the
observed product stereochemistry, we focused on the sec-
ond stable structure 1i. The coordination of substituent
enantioselectivity
of
the
Ru(II)‐Pheox–catalyzed
FIGURE 5 Results from the search for the most stable metal‐carbene complexes as the starting structure for the intramolecular
cyclopropanation reaction

6
NAKAGAWA ET AL.
FIGURE 6 Plausible mechanism for
chiral induction in the Ru(II)‐Pheox–
catalyzed intramolecular cyclopropanation
of allyl 2‐diazoacetate
intramolecular cyclopropanation reaction is affected by
the energy difference between the starting structures 1l
and 1i.
4.2 | Reaction mechanism of the
intramolecular cyclopropanation of 1
To clarify the reaction mechanism of the Ru(II)‐Pheox–
catalyzed intramolecular cyclopropanation, the vibrations
of complex 1l in vacuum and in CH Cl and CH CN sol-
2
2
3
vents were analyzed. As a result, the vibrations for the
terminal olefin approaching the Ru atom were observed
−
1
−1
at 220.16 cm (in vacuum), 219.38 cm (in CH Cl ),
2
2
−
1
and 214.76 cm (in CH CN) (Figure 7). The new C–C
3
or Ru–C bond is probably formed via thermal vibrations
of the olefin in complex 1l. Next, the bond lengths Ru–
C2, Ru–C3, C1–C2, and C1–C3 from 1l were scanned to
confirm the formation of a new bond with the carbene
during the Ru(II)‐Pheox–catalyzed intramolecular
cyclopropanation process (Figures 8 and S13). The
FIGURE 8 Scan of the bond length C1–C2 and C1–C3 in metal‐
carbene complex 1l
calculated Ru–C2 and Ru–C3 bonds did not provide any
stable structures with new bonds. The results from the
bond length scans of C1–C2 and C1–C3 suggest a trans-
formation of the metal‐carbene complex 1l into another
stable structure that is not a cyclopropane‐fused γ‐lactone
via a transition‐state energy barrier. In particular, when
the C1 carbon approaches the C2 carbon, a stable
FIGURE 7 Vibrations of the most stable metal‐carbene complex
−
1
(
2 2
1l) in CH Cl (219.38 cm )

NAKAGAWA ET AL.
7
metal‐carbene complex
metallacyclobutane is generated.
3
with
a
Ru–C1–C3–C2
new intermediate structure is formed with a four‐
membered ring. The Ru–C1 bond length becomes lon-
ger, 1.978 Å. The energy of the intermediate is 3.5 kcal/
mol higher than that of the starting reactant (1l).
3. Intermediate (3)–TS2 (4): The C1–C3 bond length is
shortened, and the Ru–C3 bond length elongates con-
comitantly with the narrowing of the C1–Ru–C2
angle from 77.9° to 46.4°. Then, a second transition
state (TS2, 4) is formed. In the structure of 4, the
C1–C3 bond length is 1.513 Å, and the Ru–C1 and
Ru–C2 bond lengths are 2.296 and 2.205 Å, respec-
tively. This energy barrier is 4.1 kcal/mol above the
intermediate and 3.5 kcal/mol below 2.
4. TS2 (4)–product (5): A C1–C2 bond (1.525 Å) is
formed via reductive elimination. The Ru–C2 bond
length in the presence of a solvent increases because
of the stabilization of the product by solvent effects.
Cyclopropane‐fused γ‐lactone (5) products with an
absolute configuration consistent with that of the
experimental results are thus obtained.
On the basis of these results from the scans of bond
lengths, we concluded that no transition states could be
located for a concerted pathway from the metal‐carbene
complex 1l. These results suggest that the Ru(II)‐Pheox–
catalyzed intramolecular cyclopropanation reaction prob-
ably proceeds via a stepwise mechanism involving the
formation of a metallacyclobutane upon oxidative addi-
tion of ruthenium. Metallacyclobutane 3 undergoes a
reductive elimination, affording cyclopropane as the
product. On the basis of these results, the reaction
mechanism to yield cyclopropanes was investigated. The
details of the calculated reaction mechanism from 1l to
product 5 via intermediate 3 are shown in Figure 9 and
Table 3 (Tables S15 to S17).
Herein, we will focus on the calculation results in
CH Cl .
2
2
1
. Reactant (1l)–TS1 (2): The terminal C2¼C3 bond is
coordinated to the Ru–C1 bond; the first transition
state (TS1, 2) is formed upon C3 approaching C1. In
the structure of 2, the C1–C3 bond length is
Our previous report showed that dichloromethane
is
a
suitable solvent for Ru(II)‐Pheox–catalyzed
2
1
.015 Å, which is 1.0 Å shorter than in 1l, with an
1.1 kcal/mol energy barrier.
cyclopropanation reactions owing to its high polarity
and nonligating properties (Table 1, entry 4). On the
other hand, acetonitrile exhibited low activity owing to
its high ligating capability toward ruthenium centers
2
. TS1 (2)–intermediate (3): C1–C3 and Ru–C2 bonds are
formed at 2.474 and 1.956 Å, respectively, and a
FIGURE 9 Energy profile for the
cyclopropanation reaction using Ru(II)‐
Pheox

8
NAKAGAWA ET AL.
TABLE 3 Bond lengths of the metal‐carbene complex reactant‐product
Ru–C1, Å
Ru–C2, Å
Ru–C3, Å
C1–C2, Å
C1–C3, Å
Vacuum
Reactant
TS1
Intermediate
TS2
1.79
1.82
1.97
2.27
2.35
2.21
2.13
1.96
2.12
2.57
2.41
2.33
2.25
3.09
3.33
3.00
2.75
2.50
1.75
1.53
2.89
2.17
1.56
1.52
1.51
Product
CH
2
Cl
2
Reactant (1l)
TS1 (2)
Intermediate (3)
TS2 (4)
1.80
1.85
1.98
2.30
2.47
2.29
2.07
1.96
2.20
3.25
2.37
2.30
2.25
3.15
3.84
2.95
2.61
2.47
1.78
1.52
3.01
2.01
1.57
1.51
1.54
Product (5)
CH
3
CN
Reactant
TS1
Intermediate
TS2
1.80
1.84
1.98
2.30
2.60
2.30
2.09
1.96
2.23
3.34
2.37
2.30
2.25
3.17
3.97
2.96
2.61
2.46
1.78
1.52
3.02
2.02
1.55
1.51
1.53
Product
(the active sites) (Table 1, entry 5). As a result of the sol-
enantioselectivity of the simple C ‐symmetric catalyst
1
vent effects observed, by DFT calculations, the energy
barrier of the reaction in dichloromethane is lower than
in acetonitrile for the TS1 and intermediate 3. In vacuum,
a higher energy barrier for the product is calculated upon
formation of an unstable ruthenium complex with no
coordination of the substrate or solvent. The rate‐
determining step should be TS1 to construct a stable
intermediate. In addition, it was estimated that the inter-
mediate should react rapidly to form TS2 and the product
because of the low relative energy of the product
was affected by the energy gap between the metal‐
carbene complexes 1l and 1i. These results provide
important information about the origin of the observed
enantioselectivity, which will allow the design of new
and more efficient C ‐symmetric catalyst systems.
1
ACKNOWLEDGMENT
This research was funded by JSPS KAKENHI grant
numbers 26288087, 17H06373, and 16J06159 from the
Japan Society for the Promotion of Science.
(
−1.7 kcal/mol in CH Cl ). Thus, the computational
2 2
results demonstrate that using dichloromethane as the
solvent reduces the energy of the rate‐determining step
and increases the reactivity of the Ru(II)‐Pheox–catalyzed
intramolecular cyclopropanation. Moreover, these results
suggest that a suitable solvent setting is required for the
analysis of the Ru(II)‐Pheox–catalyzed reaction mecha-
nism to obtain plausible molecular behavior.
ORCID
Yoko Nakagawa http://orcid.org/0000-0001-8189-5552
Hitoshi Goto http://orcid.org/0000-0001-7739-0447
REFERENCES
1. Donaldson WA. Synthesis of cyclopropane containing natural
products. Tetrahedron. 2001;57(41):8589‐8627.
5
| CONCLUSION
2
. Chen DY‐K, Pouwer RH, Richard J‐A. Recent advances in the
total synthesis of cyclopropane‐containing natural products.
Chem Soc Rev. 2012;41(13):4631‐4642.
We have carried out the first mechanistic study of Ru(II)‐
Pheox–catalyzed highly enantioselective intramolecular
cyclopropanation reactions to elucidate the mechanisms
of the reaction and of chiral induction. The computa-
tional results indicate that the desired intramolecular
cyclopropanation by Ru(II)‐Pheox proceeds via a stable
metallacyclobutane and two transition states. Consider-
ing solvent effects, the relative energy of TS1 in CH Cl2
was 2.0 kcal/mol more stable than in CH CN. This result
supports the experimental data indicating that Ru(II)‐
Pheox–catalyzed intramolecular cyclopropanation is
faster in CH Cl2 than in CH CN. Moreover, the
3. Lebel H, Marcoux J‐F, Molinaro C, Charette AB. Stereoselective
cyclopropanation reactions. Chem Rev. 2003;103(4):977‐1050.
4. Pellissier
H.
Recent
developments
in
asymmetric
cyclopropanation. Tetrahedron. 2008;64(30‐31):7041‐7095.
5
6
. Desimoni G, Faita G, Jorgensen AK. C ‐symmetric chiral
2
2
bis(oxazoline) ligands in asymmetric catalysis. Chem Rev.
2006;106(9):3561‐3651.
3
. Abu‐Elfotoh A‐M, Phomkeona K, Shibatomi K, Iwasa S. Asym-
metric inter‐ and intramolecular cyclopropanation reactions
catalyzed by a reusable macroporous‐polymer‐supported chiral
2
3

NAKAGAWA ET AL.
9
ruthenium (II)/phenyloxazoline complex. Angew Chem Int Ed.
20. Ikeno T, Iwakura I, Yabushita S, Yamada T. Theoretical analysis
of the reaction pathway and the effect of the axial ligand for 3‐
oxobutylideneaminatocobalt (II)‐catalyzed cyclopropanation.
Org Lett. 2002;4(4):517‐520.
2
010;49(45):8439‐8443.
7
. Chanthamath S, Phomkeona K, Shibatomi K, Iwasa S.
Highly stereoselective Ru(II)‐Pheox catalyzed asymmetric
cyclopropanation of terminal olefins with succinimidyl
diazoacetate. Chem Commun. 2012;48(62):7750‐7752.
21. Straub FB, Gruber I, Rominger F, Hofmann P. Mechanism of
copper(I)‐catalyzed cyclopropanation: a DFT study calibrated
with copper(I) alkene complexes.
003;684(1‐2):124‐143.
J
Organomet Chem.
8
. Abu‐Elfotoh A‐M, Nguyen DT, Chanthamath S, Phomkeona K,
Shibatomi K, Iwasa S. Water‐soluble chiral ruthenium (II)
phenyloxazoline complex: reusable and highly enantioselective
catalyst for intramolecular cyclopropanation reactions. Adv
Synth Catal. 2012;354(18):3435‐3439.
2
22. Suenobu K, Itagaki M, Nakamura E. Reaction pathway and
stereoselectivity of asymmetric synthesis of chrysanthemate
1
with the Aratani C ‐symmetric salicylaldimine−copper catalyst.
J Am Chem Soc. 2004;126(23):7271‐7280.
9
. Chanthamath S, Thongjareun S, Shibatomi K, Iwasa S. Ru(II)‐
Pheox catalyzed N–H insertion reaction of diazoacetamides:
synthesis of N‐substituted α‐aminoamides. Tetrahedron Lett.
23. Meng Q, Li M, Tang D, Shen W, Zhang J. Density functional
studies on copper‐catalyzed asymmetric cyclopropanation of
2
012;52:4862‐4865.
diazoacetate with alkene.
004;711(1‐3):193‐199.
J
Mol Struct (THEOCHEM).
2
1
1
0. Chanthamath S, Nguyen DT, Shibatomi K, Iwasa S. Highly
enantioselective synthesis of cyclopropylamine derivatives via
Ru(II)‐Pheox–catalyzed direct asymmetric cyclopropanation of
vinylcarbamates. Org Lett. 2013;16:772‐775.
24. Howell SAJ. Carboxylate lability as
Rh (carboxylate)4‐catalysed cyclopropenation
cyclopropanation of alkynes and alkenes. Dalton Trans. 2007;
34):3798‐3803.
a
factor in the
2
and
(
1. Chanthamath S, Takaki S, Shibatomi K, Iwasa S. Highly
stereoselective cyclopropanation of α,β‐unsaturated carbonyl
compounds with methyl (diazoacetoxy) acetate catalyzed by a
chiral ruthenium (II) complex. Angew Chem Int Ed.
2
5. Wang F, Meng Q, Li M. Density functional computations of the
cyclopropanation of ethene catalyzed by iron (II) carbene com-
plexes Cp (CO)(L)Fe¼CHR, L¼CO, PMe
3
, R¼Me, OMe, ph,
CO Me. Int J Quant Chem. 2008;108(5):945‐953.
2
2
013;125(22):5930‐5933.
2
6. Zhang X, Geng Z‐Y, Wang Y‐C, Li W‐Q, Wang Z, Liu F‐X. A
theoretical study nickel‐catalyzed cyclopropanation reaction.
Nickel(0) versus nickel (II). J Mol Struct (THEOCHEM).
1
2. Chanthamath S, Ozaki S, Shibatomi K, Iwasa S. Highly
stereoselective synthesis of cyclopropylphosphonates catalyzed
by chiral Ru(II)‐Pheox complex. Org Lett. 2014;16(11):3012‐3015.
2
009;893(1‐3):56‐66.
7. Bonge TH, Hansen T. Computational study of cyclopropanation
reactions with halodiazoacetates. Org Chem.
010;75(7):2309‐2320.
1
3. Chanthamath S, Chua HW, Kimura S, Shibatomi K,
Iwasa S. Highly regio‐ and stereoselective synthesis of
alkylidenecyclopropanes via Ru(II)‐Pheox catalyzed asymmet-
ric inter‐ and intramolecular cyclopropanation of allenes. Org
Lett. 2014;16(12):3408‐3411.
2
2
J
2
8. Kefalidis EC, Kanakis AA, Gallos KJ, Tsipis AC. DFT study of
the mechanism of Cu(I)‐catalyzed and uncatalyzed intramolecu-
lar cyclopropanation of iodonium ylides. J Organomet Chem.
1
1
4. Nakagawa Y, Chanthamath S, Shibatomi K, Iwasa S. Ru(II)‐
Pheox–catalyzed asymmetric intramolecular cyclopropanation
of electron‐deficient olefins. Org Lett. 2015;17(11):2792‐2795.
2010;695(17):2030‐2038.
2
9. Dzik IW, Xu X, Zhang PX, Reek JHJ, Bruin B. ‘Carbene radicals’
in CoII(por)‐catalyzed olefin cyclopropanation. J Am Chem Soc.
5. Chanthamath S, Mandour ASH, Tong TMT, Shibatomi K, Iwasa
S. Highly stereoselective cyclopropanation of diazo Weinreb
amides catalyzed by chiral Ru(II)‐Amm‐Pheox complexes. Chem
Commun. 2016;52(50):7814‐7817.
2010;132(31):10891‐10902.
30. Wang H, Guptill MD, Varela‐Alvarez A, Musaev GD, Davies
LMH. Rhodium‐catalyzed enantioselective cyclopropanation of
electron‐deficient alkenes. Chem Sci. 2013;4(7):2844‐2850.
1
6. Chanthamath S, Iwasa S. Enantioselective cyclopropanation of a
wide variety of olefins catalyzed by Ru(II)‐Pheox complexes. Acc
Chem Res. 2016;49(10):2080‐2090.
31. Shi T, Luo Y, Wang X‐L, Lu S, Zhao Y‐L, Zhang J. Theoretical
studies on the mechanism, enantioselectivity, and axial ligand
effect of a Ru(salen)‐catalyzed asymmetric cyclopropanation
reaction. Organometallics. 2014;33(14):3673‐3682.
1
7. Mandour ASH, Chanthamath S, Shibatomi K, Iwasa S. Inter‐
and intramolecular cyclopropanations of diazo Weinreb amides
catalyzed by ruthenium(II)‐Amm‐Pheox. Adv Synth Catal.
2
017;359:1‐6.
32. Luo G, Luo Y, Maeda S, Qu J, Hou Z, Ohno K. Theoretical
mechanistic studies on methyltrioxorhenium‐catalyzed olefin
cyclopropanation: stepwise transfer of a terminal methylene
group. Organometallics. 2014;33(14):3840‐3846.
1
8. Nakagawa Y, Chanthamath S, Fujisawa I, Shibatomi K, Iwasa S.
Ru(II)‐Pheox–catalyzed Si–H insertion reaction: construction of
enantioenriched carbon and silicon centers. Chem Commun.
2
017;53(26):3753‐3756.
33. Wang W, Wang G‐C. Computational study on the mechanism
and enantioselectivity of Rh
4+3] cycloaddition between vinylcarbenoids and dienes. RSC
Adv. 2015;5(101):83459‐83470.
2 4
(S‐PTAD) catalyzed asymmetric
1
9. Ikeno T, Iwakura I, Yamada T. Highly enantioselective
cyclopropanation of styrenes and diazoacetates catalyzed by 3‐
oxobutylideneaminatocobalt(II) complexes, Part 2. Semiempiri-
cal analysis of diastereo and enantioselectivities. Bull Chem Soc
Jpa. 2001;74(11):2151‐2160.
[
34. Cornejo A, Fraile MJ, Garcia IJ, et al. Computational
mechanistic studies on enantioselective pybox‐ruthenium–

10
NAKAGAWA ET AL.
catalyzed cyclopropanation reactions. Organometallics. 2005;
4(14):3448‐3457.
41. Wadt RW, Hay JP. Ab initio effective core potentials for molec-
ular calculations. Potentials for main group elements Na to Bi.
J Chem Phys. 1985;82(1):284‐298.
2
3
3
3
5. Rosenberg LM, Krapp A, Tilset M. On the mechanism of
4
4
4
2. Hay JP, Wadt RW. Ab initio effective core potentials for molec-
ular calculations. Potentials for K to Au including the outermost
core orbitals. J Chem Phys. 1985;82(1):299‐310.
cyclopropanation reactions catalyzed by a rhodium(I) catalyst
bearing a chelating imine‐functionalized NHC ligand: a compu-
tational study. Organometallics. 2011;30(24):6562‐6571.
3. Hariharan CP, Pople AJ. The influence of polarization functions
on molecular‐orbital hydrogenation energies. Theor Chem Acc.
6. Xue Y‐S, Cai Y‐P, Chen Z‐X. Mechanism and stereoselectivity
of the Rh (II)‐catalyzed cyclopropanation of diazooxindole: a
density functional theory study. RSC Adv. 2015;5(71):
1973;28(3):213‐222.
4. Frisch JM, Trucks WG, Schlegel BH, et al. Gaussian 09, Revision
D.01, Gaussian, Inc., Wallingford CT, 2009.
5
7781‐57791.
7. Xu Z‐J, Fang R, Zhao C, et al. cis‐β‐Bis(carbonyl) ruthenium‐
salen complexes: X‐ray crystal structures and remarkable
catalytic properties toward asymmetric intramolecular alkene
cyclopropanation. J Am Chem Soc. 2009;131(12):4405‐4417.
SUPPORTING INFORMATION
Additional supporting information may be found online
in the Supporting Information section at the end of the
article.
3
8. Straub FB. Pd(0) mechanism of palladium‐catalyzed
cyclopropanation of alkenes by CH N : a DFT study. J Am
2 2
Chem Soc. 2002;124(47):14195‐14201.
3
9. Zhao Y, Truhlar GD. The M06 suite of density functionals for
main group thermochemistry, thermochemical kinetics,
noncovalent interactions, excited states, and transition
elements: two new functionals and systematic testing of four
M06‐class functionals and 12 other functionals. Theor Chem
Acc. 2008;120:215‐241.
How to cite this article: Nakagawa Y,
Nakayama N, Goto H, et al. Computational
chemical analysis of Ru(II)‐Pheox–catalyzed highly
enantioselective intramolecular cyclopropanation
reactions. Chirality. 2018;1–10. https://doi.org/
10.1002/chir.23033
4
0. Hay JP, Wadt RW. Ab initio effective core potentials for molec-
ular calculations. Potentials for the transition‐metal atoms Sc to
Hg. J Chem Phys. 1985;82(1):270‐283.