Iron-catalyzed cross coupling of aryl chlorides with alkyl Grignard reagents: Synthetic scope and FeII/FeIV mechanism supported by x-ray absorption spectroscopy and density functional theory calculations

Article abstract of DOI:10.1246/bcsj.20180333

A combination of iron(III) fluoride and 1,3-bis(2,6-diiso-propylphenyl)imidazolin-2-ylidene (SIPr) catalyzes the high-yielding cross coupling of an electron-rich aryl chloride with an alkyl Grignard reagent, which cannot be attained using other iron catalysts. A variety of alkoxy-or amino-substituted aryl chlorides can be cross-coupled with various alkyl Grignard reagents regardless of the presence or absence of β-hydrogens in the alkyl group. A radical probe experiment using 1-(but-3-enyl)-2-chlorobenzene does not afford the corresponding cyclization product, therefore excluding the intermediacy of radical species. Solution-phase X-ray absorption spectroscopy (XAS) analysis, with the help of density functional theory (DFT) calculations, indicates the formation of a high-spin (S = 2) heteroleptic difluorido organoferrate(II), [MgX][Fe^IIF₂(SIPr)-(Me/alkyl)], in the reaction mixture. DFT calculations also support a feasible reaction pathway, including the formation of a difluorido organoferrate(II) intermediate which undergoes a novel Lewis acid-assisted oxidative addition to form a neutral organoiron(IV) intermediate, which leads to an Fe^II/Fe^IV cata-lytic cycle, where the fluorido ligand and the magnesium ion play key roles.

Full text of DOI:10.1246/bcsj.20180333

Advance Publication Cover Page
Iron-Catalyzed Cross Coupling of Aryl Chlorides with Alkyl Grignard Reagents:
II
IV
Synthetic scope and Fe /Fe mechanism supported by X-ray absorption spectroscopy
and density functional theory calculations
Ryosuke Agata, Hikaru Takaya, Hiroshi Matsuda, Naoki Nakatani, Katsuhiko Takeuchi, Takahiro Iwamoto,
Takuji Hatakeyama, and Masaharu Nakamura*
Advance Publication on the web December 13, 2018
doi:10.1246/bcsj.20180333
©
2018 The Chemical Society of Japan
Advance Publication is a service for online publication of manuscripts prior to releasing fully edited, printed versions. Entire
manuscripts and a portion of the graphical abstract can be released on the web as soon as the submission is accepted. Note that the
Chemical Society of Japan bears no responsibility for issues resulting from the use of information taken from unedited, Advance
Publication manuscripts.

Iron-Catalyzed Cross Coupling of Aryl Chlorides with Alkyl Grignard Reagents:
II
IV
Synthetic scope and Fe /Fe
mechanism supported by X-ray absorption
spectroscopy and density functional theory calculations
1
,2
1,2
1
3
1,2,‡
Ryosuke Agata,
Hikaru Takaya,
Hiroshi Matsuda, Naoki Nakatani, Katsuhiko Takeuchi,
Takahiro
1
,2,4
1,2,†
1,2
Iwamoto,
Takuji Hatakeyama,
and Masaharu Nakamura*
1
International Research Center for Elements Science (IRCELS), Institute for Chemical Research (ICR), Kyoto University, Uji, Kyoto
6
11-0011, Japan
2
3
4
Department of Energy and Hydrocarbon Chemistry, Graduate School of Engineering, Kyoto University, Nishikyo-ku, Kyoto 615-8510,
Japan
Department of Chemistry, Graduate School of Science, Tokyo Metropolitan University, 1-1 Minami-Osawa, Hachioji, Tokyo 192-0397,
Japan
CREST, Japan Science and Technology Agency (JST), 4-1-8 Honcho Kawaguchi, Saitama 332-0012, Japan
E-mail: masaharu@scl.kyoto-u.ac.jp
Masaharu Nakamura
He received his B. Sc. in 1991 from Tokyo University of Science under supervision of Prof. Teruaki Mukaiyama and D.
Sc. in 1996 from Tokyo Institute of Technology under the supervision of Prof. Eiichi Nakamura. He worked at the
University of Tokyo as an assistant and associate professor during 1996–2005 with Professor Eiichi Nakamura. Since
2
006, he has been a professor of Institute of Chemical Research at Kyoto University. His research interest pursues the
best synthesis for better society and humanity.
1
Abstract
pharmaceuticals, and bio-active natural products. Despite
significant advancement of these classical cross-coupling
reactions, their drawback of requiring precious and toxic
metal as catalysts has been widely recognized and is of
A
combination
of
iron(III)
fluoride
and
1
,3-bis(2,6-diisopropylphenyl)imidazolin-2-ylidene (SIPr)
catalyzes the high-yielding cross coupling of an
electron-rich aryl chloride with an alkyl Grignard reagent,
which cannot be attained using other iron catalysts. A
variety of alkoxy- or amino-substituted aryl chlorides can be
cross-coupled with various alkyl Grignard reagents
regardless of the presence or absence of β-hydrogens in the
2
concern in the synthetic community. As a result, iron
catalysts have attracted considerable attention in
cross-coupling chemistry because of their environmental and
economic advantages, and, more importantly, their unique
3
reactivity and selectivity, which are often not available in
alkyl
group.
Radical
probe
experiment
using
conventional palladium and nickel catalysts.
1
-(but-3-enyl)-2-chlorobenzene does not afford the
Significant progress has been made in the iron-catalyzed
cross-coupling reactions of organic electrophiles and
corresponding cyclization product, therefore excluding the
intermediacy of radical species. Solution-phase X-ray
absorption spectroscopy (XAS) analysis, with the help of
density functional theory (DFT) calculations, indicates the
formation of a high-spin (S = 2) heteroleptic difluorido
organoferrate(II), [MgX][Fe F
reaction mixture. DFT calculations also support a feasible
reaction pathway, including the formation of a difluorido
organoferrate(II) intermediate which undergoes a novel
Lewis acid-assisted cyclic oxidative addition to form a
neutral organoiron(IV) intermediate, which leads to an
4
–6
organometallic reagents.
In 2002, Fürstner and
co-workers reported the first iron-catalyzed cross-coupling
reaction of aryl chlorides with alkyl Grignard reagents by
using catalytic iron(III) salts in the presence of excess
II
2
(SIPr)(Me/alkyl)], in the
5
amounts of 1-methylpyrrolidin-2-one (NMP) as an essential
7a
additive (Figure 1a). Since then, a variety of coupling
reactions between alkyl Grignard reagents possessing
β-hydrogens and electron-deficient aryl/hetero aryl chlorides
7
b–e
(
alkyl–aryl coupling) have been developed by his group
8
II
IV
and others . Nonetheless, these iron catalysts are incapable
of promoting a cross-coupling reaction with electron-rich
Fe /Fe catalytic cycle, where the fluorido ligand and the
magnesium ion play key roles.
(
1
deactivated)
-chloro-4-methoxybenzene.
aryl
chlorides
Furthermore,
such
as
Key words: iron catalyst
N-heterocyclic carbene
| cross coupling |
Grignard
reagents lacking a β-hydrogen cannot participate in the
reaction due to their inability to generate a catalytically
active low-valent organoiron species.
methyl and aryl Grignard reagents could not be used for
cross coupling with even reactive aryl chlorides such as
7
d,7e,9,10
Therefore,
1
. Introduction
Transition-metal-catalyzed cross couplings of organic
halides and pseudohalides with organometallic reagents are
powerful tools for carbon–carbon bond formation in the
synthesis of various functional organic molecules.
Palladium- and nickel-catalyzed cross-coupling reactions
have been widely applied in the synthesis of a vast array of
useful compounds such as organic electronic materials,
7d,7e,11–13
methyl 4-chlorobenzoate.
After Bedford’s seminal discovery of the favorable
effect of N-heterocyclic carbene (NHC) on the
iron-catalyzed cross-coupling reaction of alkyl halides,
number of iron/NHC-catalyzed cross-coupling reactions of
aryl electrophiles such as biaryl
14
a
15,16
17,18
and alkyl–aryl
have

1
9
been developed (Figure 1b). However, the narrow substrate
scope of the reaction remains a problem. There are also
synthetic challenges, such as obtaining a high-yielding cross
coupling of electron-rich (deactivated) aryl electrophiles and
incorporating Grignard reagents without β-hydrogens.
2. Results and Discussion
2-1. Synthetic Scope of the iron fluoride/SIPr-Catalyzed
Cross Coupling between Aryl Chlorides and Alkyl
Grignard Reagents
Catalytic efficiency of various iron catalysts. We
evaluated the catalytic efficiency of various combinations of
iron salts, ligands, and additives in the reaction of
1-chloro-4-methoxybenzene 1a with methylmagnesium
bromide. As shown in entry 1 of Table 1, the use of iron(III)
fluoride and the NHC ligand SIPr gave the best result for
this methylation, affording the coupling product
1-methoxy-4-methylbenzene 2aa in 92% yield. The
(
a) Iron/NMP combination reported by Fürstner
R1
R1
cat. iron
NMP
O
Cl + XMg
X = Cl, Br )
N
CH3
NMP
(
limited to reactive aryl chlorides and 1º and 2º alkyl Grignard reagents
3 2
corresponding hydrate, FeF ·3H O, can also be used for this
(
b) Iron/NHC-catalyzed cross-coupling reactions of aryl electrophiles
reaction, resulting in a slightly higher yield (>99%) of the
coupling product, likely because of its better solubility in
THF. Unsaturated or less hindered NHC precursors, IPr·HCl
R1
R1
cat. iron
cat. NHC
R3
Ar
N
N
Ar
L
+
XMg–R3
L = Cl, OR2) (R3 = Ar or Alkyl)
NHC
(
Table 1. Optimization of the cross-coupling reaction between
1
-chloro-4-methoxybenzene (1a) and methylmagnesium
(
c) Our work communicated in 201520
a
bromide.
Cl
N
N
iron salt (5 mol %)
additive 1 (15 mol %)
additive 2 (30 mol %)
R1
R1
cat. FeF3 cat. SIPr·HCl
MeO
Cl
+
MeMgBr
MeO
Me
Cl
+
AlkylMgX
Alkyl
THF, 80 ºC, 24 h
(
1.5 equiv)
1
a
2aa
Yieldb (%) RSMb,c (%)
(
X = Cl, Br )
ArCl: including deactivated aryl chlorides
Entry
Iron salt
Additive 1 Additive 2
AlkylMgX: with or without β-hydrogens
1
2
3
FeF (FeF ·3H O) SIPr·HCl
3
3
2
none
none
92 (>99)
0
29
57
97
94
>99
76
94
1
Figure 1. Iron-catalyzed cross-coupling reactions of aryl
FeF3
IPr·HCl
58
33
0
electrophiles with Grignard reagents.
FeF3
SIMes·HCl none
4d FeF3
5e FeF3
NMP
none
none
none
none
none
none
none
none
KF
Recently, we developed cross-coupling reactions of
deactivated aryl chlorides with a variety of alkyl Grignard
reagents with or without β-hydrogens by using
TMEDA
1
6
FeF3
Pt-Bu3
0
a
2
0
combination of FeF
3
and SIPr·HCl (Figure 1c). Based on
7f FeCl (SciOPP)
none
1
2
this reaction, highly selective alkylation, including
methylation of electron-rich aryl chlorides, can be achieved
with the minimal formation of undesirable by-products. This
iron fluoride/SIPr-catalyzed cross-coupling reaction showed
contrasting reactivity and selectivity compared with the
8
9
0
FeF3
none
0
FeF2
SIPr·HCl
SIPr·HCl
SIPr·HCl
SIPr·HCl
SIPr·HCl
SIPr·HCl
88
86
21
72
49
42
1
Fe(OTf)2
FeCl3
2
11
57
16
31
38
7
,17,18
above iron-catalyzed alkyl-aryl cross-coupling reaction.
2g
FeCl3
1
1
As described in our previous report of the iron
fluoride/SIPr-catalyzed selective biaryl cross-coupling
3
FeCl3
NaOt-Bu
none
1
5
15b
14
FeCl2·4H2O
reaction, a high-valent heteroleptic organoferrate(II),
II
[
2
MgX][Fe F (SIPr)(Me/alkyl)], is likely generated as an
a
b
Reactions were carried out on a 1 mmol scale. Determined by
active intermediate. This hypothesis motivated us to further
investigate the reaction mechanism of this iron
fluoride/SIPr-catalyzed alkyl-aryl cross-coupling reaction.
Herein, we report a full account of the synthetic scope
and mechanistic study of the iron fluoride/SIPr-catalyzed
alkyl-aryl coupling reaction. We performed solution-phase
XAS analysis combined with DFT calculations for this
reaction, confirming the formation of a high-spin (S = 2)
heteroleptic difluorido organoferrate(II) intermediate as a
catalytically active intermediate. The results of DFT
calculations indicate that oxidative addition of aryl chloride
to an organoferrate(II) intermediate, assisted by a Lewis
acidic magnesium ion, efficiently proceeds to provide a
neutral organoiron(IV) intermediate. These mechanistic
investigations suggest that the fluorido ligand and the
magnesium ion of the difluorido organoferrate(II)
intermediate are important factors for attaining high
reactivity and selectivity.
quantitative GC analysis using undecane as an internal standard.
c
d
Recovery of starting material. The THF:NMP ratio is 10:1.
e
f
TMEDA (1.5 equiv) was used. Iron/bisphosphine complex
g
FeCl
2
(SciOPP) was used. FeCl
3
was treated with KF in
MeOH/THF for 1 h and then the solvent was removed in vacuo.
Cl
Cl
N
N
N
N
Me N
NMe2
2
SIPr·HCl
IPr·HCl
TMEDA
t-Bu
t-Bu
Cl
t-Bu
t-Bu
t-Bu
t-Bu
P
P
N
N
Fe
Cl
Cl
t-Bu
t-Bu
SIMes·HCl
FeCl2(SciOPP)

and SIMes·HCl, show lower activity than SIPr·HCl (entries
Table 2. Substrate scope of iron fluoride/SIPr-catalyzed
2
1
5,7
a
2
and
3).
NMP
and
(TMEDA),
alkyl-aryl coupling.
2
2
N,N,N′,N′-tetramethylethylenediamine
Entry
1
Ar–Cl
AlkylMgX (equiv)
Product
Me
Yield (%)
b
representative additives in iron-catalyzed cross-coupling
reactions, showed detrimental or no effects (entries 4 and 5).
MeO
Cl
MeMgBr
1.5)
MeO
MeO
MeO
The bulky monodentate phosphine Pt-Bu
the coupling reaction (entry 6), while it showed
comparable effect to NHC in a nickel-catalyzed biaryl
3
did not promote
92 (11)
(
1
a
2aa
a
C H17MgBr
8
8 17
C H
2
3
2
3
4
1a
1a
96 (93)
>99 (94)
60 (59)c
>99 (79)
cross-coupling reaction. A bulky bidentate phosphine,
(1.5)
2
4
2ab
2ac
SciOPP, was also ineffective and gave no coupling product
entry 7). FeF did not catalyze the reaction without an
(
3
CyMgCl
(1.5)
additive (entry 8). Thus, SIPr is the optimum ligand for this
reaction. The iron fluoride/SIPr catalyst was further
evaluated by comparison with various other iron salts. Iron
SiMe
3
Me SiCH MgCl MeO
3
2
1a
(2.0)
1
6b
2ad
fluorides and Fe(OTf)
2
were more effective than iron
OMe
OMe
chloride (entries 1, 9, and 10 versus entry 11) and other iron
salts (see SI, Table S1). The fluoride ion was indispensable
as shown in entries 11 and 12; the addition of KF to the
MeMgBr
1.5)
5d
Cl
Me
(
1
b
2b
FeCl
yield of 2aa than that without the metal fluoride. Other
iron/NHC combinations, such as Duong’s FeCl /SIPr with
O/SIPr, also exhibit
catalytic activity to give 2aa albeit in moderate yields
3
/SIPr combination resulted in a substantially higher
MeO
MeO
1
5b
C H MgBr
(1.4)
d,e
3
7
6
Cl
Cl
C H
(92)
3
7
3
1
6a
17
NaOt-Bu and Perry’s FeCl
2
·4H
2
1c
2c
MeO
MeO
MeO
MeO
(
entries 13 and 14).
7
30
MeMgBr
(2.0)
Me
Scope of aryl halides and alkyl Grignard reagents.
8f
72 (21)
With the optimal conditions in hand, we investigated the
substrate scope using various combinations of an aryl
chloride with an alkyl Grignard reagent (Table 2). The iron
fluoride/SIPr combination showed a high catalytic activity
1
d
2d
2e
Me
Me
2
N
N
Cl
C
8
H
17MgCl Me
1.5)
2
N
8 17
C H
9
94 (76)
93 (44)
(
1e
in the reaction of
a
deactivated aryl chloride,
2
Me
2
N
1
-chloro-4-methoxybenzene 1a, with alkyl Grignard
MeMgBr
1.5)
1
0
Cl
Me
(
reagents with or without β-hydrogens (entries 1–4). The
primary and secondary alkyl Grignard reagents,
octylmagnesium bromide and cyclohexylmagnesium
chloride, could be used as coupling partners to afford the
1
f
2f
F
3
C
Cl
MeMgBr
F
3
C
C
Me
11
81 (16)
76 (8)
(1.5)
1
g
2ga
2
5
corresponding products 2ab and 2ac in excellent yields.
The reaction with less reactive Grignard reagent,
trimethylsilyl)methylmagnesium
proceeded to give the desired product 2ad in 60% yield
SiMe
3
a
Me
3
SiCH
MgCl F
2
3
1
2
1g
2
6a,b
(2.0)
(
chloride,
also
2
gb
Ph
Cl
MeMgBr
1.5)
Ph
Me
(
entry 4). The trimethylsilylmethyl group in this product is
13
>99 (85)
>99 (97)
(92)
(
2
6c
amenable to further synthetic elaborations.
1h
1h
2ha
Gram-scale syntheses demonstrate the synthetic utility of
the present protocol; the reaction of the electronically and
sterically demanding substrate 1b (20 mmol) with
methylmagnesium bromide (30 mmol) afforded the desired
product 2b in >99% yield: 1.92 g (79%) of the analytically
pure product 2b was obtained (entry 5). In addition, the
reaction of 1c (70 mmol) with propylmagnesium bromide
C H17MgBr Ph
8 17
C H
8
1
4
(1.5)
2
hb
hc
CyMgCl
Ph
1
5
1h
1h
(1.5)
2
SiMe3
Me SiCH MgCl Ph
3
2
16
>99 (96)
(2.0)
(
98 mmol) gave 9.71 g of 2c in 92% yield, even with a
2hd
reduced catalytic load of 2 mol % (entry 6). The reaction of
the more electron-rich substrate 1d was very sluggish and
provided the desired product in poor yield (entry 7);
however, by increasing the catalytic loading to 15 mol %,
the yield was significantly improved to 72% (entry 8).
Cl
Me
MeMgBr
(1.5)
17
53 (40)
(57)
N
N
1i
2i
TsO
Cl
C
8
H
17MgBr TsO
1.3)
8 17
C H
18g
(
The electron-rich aryl chlorides 1e and 1f, possessing a
dimethylamino group, also participated in the reaction to
give the coupling products 2e and 2f in 94% and 93% yields,
respectively (entries 9 and 10). High-yielding methylations
1j
2j
a
3
Reactions were carried out using FeF (5 mol %) and SIPr·HCl (15
mol %) in THF; see the SI for details of the reaction conditions and
b
and
electron-deficient aryl chloride 1g, biaryl chloride 1h, and
-chloroquinoline 1i to give the corresponding coupling
various
alkylations
proceeded
with
the
scales for each entry. Yield were determined by quantitative GC or
1
H NMR analysis using an internal standard. Isolated yields are
c
2
given in parentheses. The starting material was recovered in 20%
1
8
d
products (entries 11–17). The coupling reaction of 1j with
octylmagnesium bromide proceeded preferably at the
chloride site to give 2j in 57% yield along with the
formation of the double octylation product in 15% yield
yield. Reactions were performed on 20 mmol and 70 mmol scales,
e
respectively. FeF
3
·3H
2
O (2 mol %) and SIPr·HCl (6 mol %) were
f
g
3 3
used. FeF (15 mol %) and SIPr·HCl (45 mol %) were used. FeF
(10 mol %) and SIPr·HCl (30 mol %) were used. For the mass
2
7
(
entry 18).
balance, see text and ref. 27.

Effect of the halogens of aryl halides on product
selectivity. To study reactivity differences of the leaving
halogen groups, we carried out reactions of
combination to facilitate the isomerization of the isopropyl
group. In contrast, in the FeF /SIPr combination,
3
a
1
5b
four-coordinate organoferrate(II) species is formed owing
to the strong binding ability of the fluorido ligand to retard
the isomerization of the isopropyl group (see section 2-2).
The branched versus linear selectivity was thus completely
1
-halo-4-methoxybenzenes 1a–X with methyl and octyl
Grignard reagents. As shown in Table 3, the aryl chloride
a–Cl afforded the corresponding coupling products 2aa
1
2
8
32,33
and 2ab in quantitative yields (entries 1 and 4). In contrast,
the reactions of the corresponding aryl bromide 1a–Br and
iodide 1a–I gave the products 2aa and 2ab in drastically
lower yields owing to significant production of the reductive
different depending on the halogen ligand used.
Table 4. Comparison between the FeF /SIPr and FeCl /SIPr
3
3
a
combinations using various alkyl Grignard reagents.
2
9
dehalogenation by-product 3 (entries 2, 3, 5, and 6).
iron salt (5 mol %)
SIPr·HCl (15 mol %)
AlkylMgX
MeO
Alkyl
a
2
Table 3. Comparison of leaving group ability.
MeO
Cl
+
THF
FeF3 (5 mol %)
SIPr·HCl (15 mol %)
AlkylMgBr (1.5 equiv)
MeO
Alkyl
H
MeO
H
1a
2
3
MeO
X
+
THF
AlkylMgX
(equiv)
Yieldb (%)
MeO
Entry Iron salt
Product 2
2ae : 2af
1
a–X
RSMc
2
3
3
1
2
FeF3
FeCl
92 <1
27
0
MeMgBr
(1.5)
Yieldb (%)
RSMc
2aa
–
Entry
1a–X
AlkylMgBr
MeMgBr
Product 2
3
8
57
2
3
3
4
FeF3
>99 <1
68 12
0
7
CyMgCl
1
2
3
1a–Cl
1a–Br
1a–I
>99 <1
0
25
17
2ac
–
(1.5)
FeCl3
2aa
2ab
12
3
57
61
5d,e
FeF3
MeO
38 : 62
97 <1
0
i-PrMgCl
2.0)
2ae
(
4
5
6
1a–Cl
1a–Br
1a–I
>99
4
<1
65
68
0
24
0
6d,e FeCl3
MeO
<1 : >99 37 35 10
C8H17MgBr
2af
a
3
Reactions were carried out on a 1 mmol scale; see the SI for details
b
1
a
of the reaction conditions. Determined by quantitative GC or H
Reactions were carried out on a 1 mmol scale; see SI for details of
c
b
1
NMR analysis using an internal standard. Recovery of starting
the reaction conditions. Determined by quantitative GC or H NMR
d
c
3
material. FeF (10 mol %) and SIPr·HCl (30 mol %) were used.
analysis using an internal standard. Recovery of starting material.
e
1
Branched versus linear selectivity was determined by H NMR
analysis using an internal standard.
Effect of the fluorido ligand on product selectivity.
We conducted the coupling reactions of aryl chloride 1a
with various alkyl Grignard reagents in the presence of the
Radical probe experiment. To gain insight into the
reaction mechanism, we conducted
a
radical probe
experiment using aryl chloride 1k (Figure 2). The
corresponding aromatic methylation proceeded
FeF
product selectivity of the fluoride system (Table 4). As
shown in entries 1 and 2, the FeF /SIPr combination showed
higher reactivity and selectivity than the FeCl /SIPr
combination; the FeF /SIPr combination gave 2aa in 92%
yield, whereas mixture of 2aa and reductive
dehalogenation product 3 was obtained in low yield using
the FeCl /SIPr combination. The same tendency was
3 3
/SIPr and FeCl /SIPr combinations to access the unique
3
4
3
quantitatively without formation of cyclization products 2kb
or 2kc. This result indicates that the reaction proceeds via a
two-electron mechanism
3
3
3
5
rather than the radical
which is widely accepted for the
a
3
6
mechanism
3
6h,i
iron-catalyzed cross-coupling reactions of alkyl halides.
3
observed in the reaction with cyclohexylmagnesium chloride,
resulting in the quantitative formation of coupling product
FeF3 (5 mol %)
SIPr·HCl (15 mol %)
2
ac with the FeF
and 3 with the FeCl
The reaction of 1a with isopropylmagnesium chloride
confirmed the similar cross-coupling/reductive
3
/SIPr combination but a mixture of 2ac
+
MeMgBr
/SIPr combination (entries 3 and 4).
Cl
(1.5 equiv)
THF, 60 ºC, 36 h
3
1
k
dehalogenation selectivity of the fluoride system. It also
provided further mechanistic insight into the reaction; the
+
+
Me
ka
99% (NMR yield)
Me
Me
2kc
2
2kb
not observed
FeF
3
/SIPr combination provided the branched 2ae and linear
38:62 ratio, while the FeCl /SIPr
>
not observed
2
af isomers in
a
3
Figure 2. Radical probe experiment.
combination furnished linear 2af as the sole isomer owing to
the complete isomerization of the isopropyl group (entries 5
and 6). The isomerization of the isopropyl group to a propyl
group can be ascribed to the β-hydrogen elimination of the
isopropyliron intermediate and subsequent regio-inversion
2
-2. Mechanistic Study of iron fluoride/SIPr-Catalyzed
Cross-Coupling Reaction by XAS analysis and DFT
calculations
3
0
re-insertion. Danopolous reported that three-coordinate
dialkyliron(II)/NHC complexes were formed by using alkyl
Grignard reagents without β-hydrogens, such as
Based on the results both described above and
15b
previously reported,
mechanism in Figure 3. The catalyst precursor FeF
undergoes reduction by RMgX to give the divalent
difluoridoiron(II) intermediate Fe F
of SIPr. Intermediate 4 further reacts with RMgX to form
we depict a plausible reaction
3
3
1
(
trimethylsilyl)methyl and benzyl Grignard reagents. We
thus consider that coordinatively unsaturated
organoiron(II) species is generated in the FeCl /SIPr
a
II
2
(SIPr) 4 in the presence
3

difluorido organoferrate intermediate 5. Oxidative addition
of the aryl chloride Ar–Cl proceeds via intermediate 6 to
give neutral organoiron(IV) intermediate 7. Reductive
elimination from 7 proceeds almost concertedly with a very
low activation barrier, as discussed later, to give
π-coordinated intermediate 8. Dissociation of the product,
Ar–R, gives intermediate 9, which reacts with RMgX to
regenerate the catalytically active organoferrate(II)
intermediate 5, completing the catalytic cycle. This catalytic
(a)
FeF3·3H2O + SIPr + Me3SiCH2MgCl
FeF2 pellet
2
1
FeF3 pellet
II
IV
cycle consists of an Fe /Fe redox shuttle which is
completely different from the one proposed for the
0
7
100
7110
7120
7130
7140
7150
7160
3
6
iron-catalyzed cross-coupling reactions of alkyl halides.
Energy (eV)
We assessed this mechanistic proposal by using
solution-phase synchrotron XAS analysis and DFT
calculations as discussed in the following sections.
(b)
6
4
FeF3·3H2O + SIPr + Me3SiCH2MgCl
FEFF fitting
R factor = 3.47
-
1
Δk = 2.0-10 Å
ΔR = 1.10-3.14 Å-
1
FeF3
N
F
MgX
F
C1
FeII
Fe
L
Cl
2
0
Mg
n RMgX
SIPr
Ar–Cl
C2
Si
Ar
R
F
Cl
6
F
MgX
F
F
MgClX
F
F
FeII
RMgX
0
1
2
3
4
5
L
L
FeII
L
FeIV
Radial distance (Å)
F
R
Ar
R
Figure 5. (a) Solution-phase Fe K-edge XANES spectrum of
the reaction mixture of FeF ·3H O, SIPr (7.0 equiv), and
Me SiCH MgCl (12 equiv) in THF (fluorescence mode, blue
line). BN diluted pellets of FeF (transmission mode, dotted
black line) and FeF (transmission mode, black line) as a
references (b) FEFF fitting analysis on the EXAFS spectrum
of the reaction mixture of FeF ·3H O, SIPr (7.0 equiv), and
Me SiCH MgCl (12 equiv) in THF (fluorescence mode, blue
line). FEFF simulated EXAFS spectrum of
MgCl][Fe F (SIPr)(CH SiMe )] 10 (S = 2) (red line).
4
5
7
MgClX
3
2
3
2
RMgX
L = SIPr
X = Cl, Br
F
F
2
FeII
MgClX
R
FeII
MgClX
L
L
3
F
F
9
R
Ar
8
Ar
3
2
3
2
Figure 3. Proposed reaction mechanism.
II
[
2
2
3
Stoichiometric reaction of the reactive organoferrate(II)
intermediate and solution-phase Fe K-edge X-ray
absorption spectroscopy of the reaction mixture. To
confirm the formation of organoferrate(II) intermediate 5
depicted in Figure 3, and to examine its reactivity toward an
aryl chloride, we conducted a stoichiometric reaction and its
solution-phase XAS analysis. As shown in Figure 4, a
We predicted the formation of organoferrate(II)
1
5b
intermediate 10 based on our previous report. However,
the paramagnetic nature of the reaction mixture prevented us
from obtaining structural information about the intermediate
by NMR, and we thus conducted synchrotron XAS analysis
of the reaction mixture prior to the addition of the aryl
3
8
reaction mixture was prepared by mixing FeF
3 2
·3H O (0.06
chloride. The solution-phase structure of the reactive
intermediate was determined by the combination of XAS
mmol) and large excess amounts of SIPr (7.0 equiv) and
3
7
3 2
Me SiCH MgCl (12 equiv) in THF (80 °C, 1h).
analysis
PCMTHF/UB3LYP-GD3BJ/6-311G(d,p) level of theory to be
the high-spin (S 2) difluorido organoferrate(II)
intermediate [MgCl][Fe F
and
DFT
calculations
at
a
Subsequent treatment of the reaction mixture with aryl
chloride 1h (0.06 mmol) was added to the reaction mixture
and the cross-coupling reaction proceeded smoothly to give
the corresponding product 2hd in 74% yield.
=
II
2
2 3
(SIPr)(CH SiMe )] (10). As
shown in Figure 5a, XANES analysis clearly indicated the
reduction of iron(III) fluoride to an iron(II) intermediate by
the lower energy edge shift caused by increasing radial
distribution of electron density around the iron center. The
intense pre-edge peak at 7110 eV indicates the tetrahedral
SIPr (7.0 equiv)
Me3SiCH2MgCl (12 equiv)
N
N
3
9
FeF3·3H2O
FeII
geometry of the organoiron(II) intermediate. Furthermore,
as shown in Figure 5b, the EXAFS spectrum of the reaction
mixture shows excellent agreement with the FEFF simulated
EXAFS spectrum of the DFT-optimized structure 10 (high
F
F
(0.06 mmol)
THF, 80 ºC, 1 h
MgCl
Si
1
0
Ph
Cl
solution-phase XAS
4
0,41
spin, S = 2).
We thus conclude that non-labile fluorido
1
h (0.06 mmol )
SiMe3
ligand have a critical role in preventing the formation of
Ph
42
7d,7e
unfavorable dialkyl, low-valent,
species,
or peralkylated iron
THF, 80 ºC, 38 h
43,44
2
hd 74% (GC yield)
even in the presence of excess amounts of
Grignard reagent.
DFT calculations of the reaction pathways. The
proposed reaction mechanism shown in Figure is
supported by DFT calculations. The oxidative
addition/reductive elimination process that includes an
Figure 4. Stoichiometric reaction of reactive organoferrate(II)
intermediate 10 with 4-chlorobiphenyl 1h and solution-phase
XAS analysis.
3

Ph–Cl
Me
F
MgCl
Cl
TSox_t
+26.2
Cl
FeII
F
Mg
Ph–Me
L
F
Cl
F
Mg
Cl
Cl
F
Cl
FeII
Cl
L
FeII
Me
Cl
L
F
MgCl
Mg
L
FeIV
F
F
Mg
F
F
TSox_q
+23.5
6
′q
F
Me
L
FeII
F
5
′t
FeIV
Cl
Me
L
9
′t
‡
6
′q
ΔG = 23.5
Me
8′t
5
′t + Ph–Cl
1.0
+
2.4
TSred_t
+
7
′q
Cl
Ph–Me
TSred_t
–8.2
Mg
6
′t
F
F
5
′q + Ph–Cl
MgCl
7′q
Cl
Mg
+
1.6
F
0
F
L
FeII
Cl
L
FeII
–
12.0
F
F
Cl
Fe II
Cl
TSred_q
–9.6
Me
L
F
Me
F
MgCl
9
′q
L
FeII
F
Me
7′t
–17.3
MgCl
Cl
FeII
F
8′q
L
9
′t + Ph–Me
TSox_t
Cl
Me
′q
8
′t
Cl
–33.2
5
F
Mg
Mg
–36.5
6
′t
MgCl
F
Ph–Cl
F
FeIV
Cl
L
F
F
F
FeIV
Cl
9′q + Ph–Me
48.2
L
8′q
FeII
Cl
L
Me
–
Me
–49.3
Me
7
′t
TSox_q
TSred_q
Figure 6. Energy diagram of the reaction pathway calculated by PCMTHF/UTPSS-GD3BJ/6-311G(d). Relative Gibbs free energies
(
ΔG) are presented in kcal/mol (quintet state, S = 2: red; triplet state, S = 1: blue).
organoferrate(II) intermediate such as 10 was confirmed by
elimination were successfully located as TSred_t and TSred_q
for the triplet and quintet states, respectively. The quintet
4
5
triplet and quintet potential energy surfaces. Figure 6
depicts the detailed energy profiles of the reaction pathways
from organoferrate(II) intermediate 5′ to the product Ph–Me
and the starting neutral iron(II) species 9′ for both quintet (S
TSred_q is 1.4 kcal/mol more stable than the triplet TSred_t
and the quintet product 8′ is 12.8 kcal/mol more stable than
the triplet product 8′ , suggesting the possibility of spin
crossover from the triplet to the quintet state. The
,
q
t
5
1,52
=
2) and triplet (S = 1) spin states. We employed the
chemical model of the iron fluoride/SIPr-catalyzed
methylation of phenyl chloride Ph–Cl by using an
activation barriers of the reductive elimination were
calculated to be 2.4 kcal/mol and 9.1 kcal/mol for the quintet
unrestricted
PCMTHF/UTPSS-GD3BJ/6-311G(d)
proposed catalytic cycle. The ground state of the starting
organoferrate(II) species was calculated as 5′ , a quintet
state (S =2) species. The coordination of chlorobenzene to
the magnesium ion by the chlorine atom resulted in the
open-shell
DFT
calculation
to corroborate the
of
and triplet states, respectively, reflecting the instability of
4
6
15b
the iron(IV) intermediates.
Dissociation of the Ph–Me
product may easily occur to generate a complex between
iron(II) fluoride and MgCl (9′). The DFT calculation
q
2
IV
II
demonstrates that the Fe /Fe mechanism has advantages
for both the oxidative addition and reductive elimination
steps; the Lewis acidic magnesium ion of the
organoferrate(II) intermediate contributes to the facile cyclic
oxidative addition of aryl chloride, and the resulting
iron(IV) intermediate is prone to undergo rapid reductive
elimination.
formation of precomplex 6′
Thus, the equilibrium between the starting 5′
precomplex 6′ is shifted to the substrate side. The same
tendency is observed for the triplet spin state, yet the energy
difference is rather small: 5′ is 0.6 kcal/mol more stable
than 6′ . These four species have a free energy range of 2.4
q
, which is an uphill process.
q
and the
q
t
t
kcal/mol and thus may exist in solution as an equilibrium
mixture.
3. Conclusion
The iron fluoride/SIPr combination enables the versatile
alkylation of deactivated aryl chlorides. The advantage of
this combination over previous catalysts has been clearly
demonstrated by the wide scope of alkyl Grignard reagents
with or without β-hydrogens that can be used. Not only
methyl Grignard reagents, but also primary and secondary
alkyl Grignard reagents undergo cross-coupling reactions
with various electron-rich aryl chlorides to selectively afford
the corresponding methylated/alkylated aryl compounds.
The reactivity of organoferrate(II) intermediate 10, the
existence of which was confirmed by solution-phase XAS
analysis using the DFT-based structure, was assessed by its
stoichiometric reaction with aryl chloride to provide the
The transition structures of the oxidative addition of
chlorobenzene to the organoferrate(II) species were located
with an activation barrier of 23.5 kcal/mol for the quintet
state TSox_q and 25.2 kcal/mol for the triplet state TSox_t. We
thus consider that the oxidative addition step proceeds via
5′ and TSox_q on the quintet energy surface. The geometry
q
of the transition structure is notable; the Cl atom of Ph–Cl
coordinates to the Lewis acidic magnesium ion, and the
characteristic
five-centered
interaction
of
Cl···Mg···F···Fe···C in TSox results in the formation of the
neutral organoiron(IV) intermediate 7′. The cooperation of
the two metal centers, Mg and Fe, of difluorido
organoferrate(II) reduces the activation energy to accelerate
II
IV
corresponding coupling product. A high-valent Fe /Fe
4
7
the cleavage of the Ph–Cl bond. In addition, strong
catalytic cycle via Lewis acid-assisted cyclic oxidative
addition was supported by DFT calculations. We expect that
the present catalyst combination will provide facile access to
useful aromatic compounds, such as organic electronic
materials and pharmaceuticals, in practical settings.
Moreover, further development of new catalysts based on
1
9
48,49
σ-donation of SIPr and π-donation of the fluorido
ligands may contribute to stabilization of the organoiron(IV)
intermediate .
5
0
As for the spin state of the organoiron(IV) intermediate,
the triplet species 7′
t
is 5.3 kcal/mol more stable than the
II
IV
quintet species 7′ . Transition structures for the reductive
q
the Fe /Fe
mechanism will advance cross-coupling

reactions that are not currently achievable not only by iron,
but also by conventional precious metal catalysts.
the glovebox. The cell was thoroughly dried under elevated
temperature and evacuated prior to the measurements. For
solid-state XAS, BN pellets (ca. 0.5mm thickness) were
prepared by mixing of the powders of iron salts and a
commercial BN powder for dilution, and then
uniaxially-compressing by tableting machine placed in the
glovebox. All the processes for sample preparation were
performed in an argon-filled glovebox.
4
. Experimental
General procedure for cross coupling: synthesis of
1
-methoxy-4-methylbenzene (2aa). A THF solution of
methyl magnesium bromide (1.59 mL, 0.94M, 1.5 mmol)
was added to a mixture of FeF (5.7 mg, 0.050 mmol) and
,3-bis(2,6-diisopropylphenyl)imidazolinium chloride (64.4
3
1
Acknowledgement
mg, 0.15 mmol) in 2.0 mL of THF at 0 ºC. The substrate
aryl chloride, 1-chloro-4-methoxybenzene (143 mg, 1.0
mmol), and undecane were then added at room temperature.
The mixture was stirred at 80 ºC for 24 h. After cooling to
room temperature, an aliquot of the reaction mixture was
filtered through a Florisil pad. The product yield was
determined by GC analysis (92% yield) using undecane as
an internal standard. To avoid loss of products, isolation was
performed in the large-scale experiment using 5.6 mmol of
the starting aryl chloride. After the reaction was performed
at 80 ºC for 84 h, saturated aqueous sodium potassium
tartrate and aqueous HCl (1 N) were added. The aqueous
layer was extracted with ether four times. The organic layers
This work is supported in part by the Core Research for
Evolutional Science and Technology (CREST 1102545)
Program, Advanced Low Carbon Technology Research and
Development Program (ALCA JPMJAL1504) from the
Japan Science and Technology Agency (JST), and JSPS
Core-to-Core
Program
"Elements
Function
for
Transformative Catalysis and Materials". R.A. thanks
‘Research Fellowships of Japan Society for the Promotion of
Science for Young Scientists’ (JSPS 11488). We are grateful
to Integrated Research Consortium on Chemical Sciences
(IRCCS). The authors thank Dr. Yasuhiro Ohki for
providing iron/NHC complexes. FT-ICR-MS and 800-MHz
NMR analyses were supported by the JURC at ICR, Kyoto
University. The authors thank RIKEN for the provision of
beamtime in SPring-8 (BL14B2: 2015A0121, 2015B0121,
2016A0121, and BL27SU: 2015A0122, 2015B0122,
2016A0122). The authors thank Drs. Tetsuo Honma,
Masafumi Takagaki, and Yusuke Tamenori for their kind
and generous support on the XAS analysis and also thank
Prof. Seiji Mori for advice on the DFT calculations. We are
grateful to Tosoh Finechem Corporation and Nissan
Chemical Industries Corporation for their financial support.
2 4
were combined, washed with brine, dried over Na SO , and
filtered through a Florisil pad. After removal of the solvent,
the crude product was purified by silica gel column
chromatography using pentane as the eluent (R
f
= 0.06) and
subsequent GPC to obtain the title compound (78.7 mg,
1
1
(
2
1%) as a colorless oil. H NMR (392 MHz, CDCl
s, 3H, -CH ), 3.78 (s, 3H, -OCH ), 6.80 (d, J = 8.6 Hz, 2H,
-), 7.09 (d, J = 7.8 Hz, 2H, 3,5-C
) δ 20.59, 55.43, 113.84 (2C), 129.98,
3
) δ 2.29
3
3
1
3
,6-C
6
H
4
6 4
H -); C NMR
(
99 MHz, CDCl
3
+
1
1
8 10
30.03 (2C), 157.61; HRMS (EI) m/z [M] calcd for C H O
22.0732, found 122.0730; Anal. calcd for C
H
10O C, 78.65;
Supporting Information
8
H, 8.25, found C, 78.81; H, 8.17.
Stoichiometric reaction
organoferrate(II) intermediate. The sample reaction
mixture was prepared in the argon-filled glovebox. A THF
solution of trimethylsilylmethylmagnesium chloride (0.67
mL, 0.97 M, 0.65 mmol) was added to a mixture of
Details of the experimental procedure, computational
study, XAS analysis, and spectral date of the products. This
material is available electronically on J-STAGE.
of
the
reactive
References
‡
Current address: National Institute of Advanced
Industrial Science and Technology (AIST), 1-1-1
Higashi, Tsukuba, Ibaraki 305-8561, Japan
FeF
3
·3H
2
O
(9.1
mg,
0.055
mmol)
and
1
0
,3-bis(2,6-diisopropylphenyl)imidazolin-2-ylidene (149 mg,
.38 mmol) in 0.42 mL of THF at room temperature. The
†
Current address: Department of Chemistry, School of
Science and Technology, Kwansei Gakuin University,
2-1 Gakuen, Sanda, Hyogo 669-1337, Japan
mixture was stirred at 80 ºC for 1 h. The substrate aryl
chloride, 4-chlorobiphenyl (10.3 mg, 0.060 mmol), and
undecane were then added at room temperature. The mixture
was stirred at 80 ºC for 38 h. After cooling to room
temperature, an aliquot of the reaction mixture was filtered
through a Florisil pad. The product yield (74%) was
determined by GC analysis using undecane as an internal
standard.
X-ray absorption spectroscopy (XAS) analysis.
XANES and EXAFS measurements were performed at the
BL14B2 beamline of the synchrotron SPring-8 in Japan
under standard beamline conditions. The Fe K-edge (7.11
keV) XAS data were corrected by fluorescence or
1.
a) K. Tamao, In Comprehensive Organic Synthesis;
Vol. 3; B. M. Trost, I. Fleming, Eds; Pergamon Press:
Oxford, 1991, p. 435. For review palladium-catalyzed
cross-coupling reactions, see: b) S. R. Chemler, D.
Trauner, S. J. Danishefsky, Angew. Chem. Int. Ed.
2001, 40, 4544. c) K. C. Nicolaou, P. G. Bulger, D.
Sarlah, Angew. Chem. Int. Ed. 2005, 44, 4442. For
review of palladium, nickel, or iron-catalyzed
cross-coupling reactions, see: d) R. Jana, T. P. Pathak,
M. S. Sigman, Chem. Rev. 2011, 111, 1417.
2.
3.
a) E. Nakamura, K. Sato, Nature Mater. 2011, 10, 158.
b) J. E. Zweig, D. E. Kim, T. R. Newhouse, Chem.
Rev. 2017, 117, 11680. c) W. M. Czaplik, M. Mayer. J.
Cvengroš, A. J. von Wangelin, ChemSusChem 2009, 2,
396.
a) E. Nakamura, T. Hatakeyama, S. Ito, K. Ishizuka, L.
Ilies, M. Nakamura, Organic Reactions, 2014, 83, pp.
1–210. b) J. Legros, B. Figadère, Nat. Prod. Rep. 2015,
32, 1541. c) A Piontek, E. Bisz, M. Szostak, Angew.
Chem. Int. Ed. 2018, 57, 11116.
transmission modes using N
chambers with optimized gas ratio and pressure with Si
111) double-crystal monochromator. To correct energy
2
/Ar mixed gas-filled ionization
(
deviations of XANES spectra at different beamtimes, an
energy calibration was performed. For Fe K-edge, an iron
metal foil was used as reference sample to set the white line
determined by first-derivative of XANES spectra to
7
128.456 eV. For solution-phase XAS, sample solutions
were transferred into a specially designed PEEK-made
solution cell equipped with 150 nm-thickness Si fillter in
3
N
4

4
5
.
.
a) M. Tamura, J. K. Kochi, J. Am. Chem. Soc. 1971,
3, 1487. b) S. M. Neumann, J. K. Kochi, J. Org.
Chem. 1975, 40, 599.
Suzuki-type biaryl cross-coupling reaction has also
been reported, see: c) M. O’Brien, M. Manzotti, R. D.
Abrams, D. Elorriaga, H. A. Sparkes, S. A. Davis, R.
B. Bedford, Nature Catalysis 2018, 1, 429.
9
The iron/NMP catalyst system was originally
developed by Cahiez for the coupling of alkenyl
halides, see: a) G. Cahiez, S. Marquais, Pure Appl.
Chem. 1996, 68, 53. b) G. Cahiez, H. Avedissian,
Synthesis 1998, 1199. The reaction of acyl halides
with alkyl zinc reagents was also developed using an
iron/NMP catalyst system, see: c) C. K. Reddy, P.
Knochel, Angew. Chem. Int. Ed. 1996, 35, 1700.
B. D. Sherry, A. Fürstner, Acc. Chem. Res. 2008, 41,
16. a) Y.-Y. Chua, H. A. Duong, Chem. Commun. 2014,
50, 8424. b) Y.-Y. Chua, H. A. Duong, Chem.
Commun. 2016, 52, 1466. See also ref. 21a.
17. The
FeCl
2
·4H
2
O/SIPr
combination
catalyzes
cross-coupling reactions of deactivated aryl chlorides
with primary and secondary alkyl Grignard reagents,
but the use of excess amounts of Grignard reagents
and high temperature conditions are required to obtain
the coupling products, see: M. C. Perry, A. N. Gillet, T.
C. Law, Tetrahedron Lett. 2012, 53, 4436.
6
7
.
.
1
500.
a) A. Fürstner, A. Leitner, Angew. Chem. Int. Ed. 2002,
1, 609. b) A. Fürstner, A. Leitner, M. Méndez, H.
4
2 3 2
18. The FeCl /SIMes·HCl and FeF ·3H O/IPr·HCl
Krause, J. Am. Chem. Soc. 2002, 124, 13856. c) A.
Fürstner, A. Leitner, G. Seidel, Org. Synth. 2005, 81,
combinations are effective for the alkylation of aryl
sulfamates, sulfonates, and carbamates, see: a) A. L.
Silberstein, S. D. Ramgren, N. K. Garg, Org. Lett.
2012, 14, 3796. b) T. Agrawal, S. P. Cook, Org. Lett.
2013, 15, 96. One example of cross coupling between
biphenyl sulfamate and methyl Grignard reagent in the
3
3. d) A. Fürstner, H. Krause, C. W. Lehmann, Angew.
Chem. Int. Ed. 2006, 45, 440. e) A. Fürstner, R. Martin,
H. Krause, G. Seidel, R. Goddard, C. W. Lehmann, J.
Am. Chem. Soc. 2008, 130, 8773.
8
.
.
For selected papers, see: a) L. K. Ottesen, F. Ek, R.
Olsson, Org. Lett. 2006, 8, 1771. b) C. Risatti, K. J.
Natalie, Z. Shi, D. A. Conlon, Org. Proccess Res. Dev.
3 2
presence of FeF ·3H O/IPr has been reported, see: c)
T. Agrawal, S. P. Cook, Org. Lett. 2014, 16, 5080.
19. Recent reviews: a) K. Riener, S. Haslinger, A. Raba,
M. P. Högerl, M. Cokoja, W. A. Herrmann, F. E. Kühn,
Chem. Rev. 2014, 114, 5215. b) D. Bézier, J.-B.
Sortais, C. Darcel, Adv. Synth. Catal. 2013, 355, 19. c)
M. J. Ingleson, R. A. Layfield, Chem. Commun. 2012,
48, 3579.
20. R. Agata, T. Iwamoto, N. Nakagawa, K. Isozaki, T.
Hatakeyama, H. Takaya, M. Nakamura, Synthesis
2015, 47, 1733.
21. A systematic evaluation of NHC ligands in an
iron-catalyzed biaryl coupling shows the sterically
demanding NHC ligand led to the high yields of the
desired coupling products, see: a) W. Wu, Q. Teng,
Y.-Y. Chua, H. V. Huynh, H. A. Duong,
Organometallics 2017, 36, 2293. Backbone and
N-substituents on alkylated iron/NHC complex
contribute to the reactivity of the cross-coupling
reaction, see: b) S. B. Muñoz III, V. E. Fleischauer, W.
W. Brennessel, M. L. Neidig, Organometallics 2018,
37, 3093.
2
013, 17, 257.
9
1
Fürstner proposed a low-valent organoiron species as
the catalytically active intermediate in their coupling
reaction.
0. a) L. E. Aleandri, B. Bogdanović, P. Bons, C. Dürr, A.
Gaidies, T. Hartwig, S. C. Huckett, M. Lagarden, U.
Wilczok, R. A. Brand, Chem. Mater. 1995, 7, 1153. b)
G. Siedlaczek, M. Schwickardi, U. Kolb, B.
Bogdanović, D. G. Blackmond, Catal. Lett. 1998, 55,
6
7. c) B. Bogdanović, M. Schwickardi, Angew. Chem.
Int. Ed. 2000, 39, 4610.
1
1. The introduction of a methyl group is limited to highly
reactive substrates such as electron-deficient
heteroaromatic compounds and enol triflates, see: a) M.
Hocek, H. Dvoráková, J. Org. Chem. 2003, 68, 5773.
b) S. Malhotra, P. S. Seng, S. G. Koenig, A. J. Deese,
K. A. Ford, Org. Lett. 2013, 15, 3698. c) B. Scheiper,
M. Bonnekessel, H. Krause, A. Fürstner, J. Org. Chem.
2
004, 69, 3943.
1
2. For reviews of methyl effects in medicinal chemistry,
see: a) E. J. Barreiro, A. E. Kümmerle, C. A. M. Fraga,
Chem. Rev. 2011, 111, 5215. b) H. Schonherr, T.
Cernak, Angew. Chem. Int. Ed. 2013, 52, 12256, and
the references cited therein.
22. For the effectiveness of the TMEDA ligand, see: P. J.
Rushworth, D. G. Hulcoop, D. J. Fox, J. Org. Chem.
2013, 78, 9517.
23. M. Nakamura, T. Hatakeyama, H. Eguchi, H. Yano,
WO 2009/008447 A1, January 15, 2009.
1
1
3. Iron-bisphosphine-catalyzed cross-coupling reactions
of arylboron compounds with iodomethane have been
reported, see: S. Nakajima, H. Takaya, M. Nakamura,
Chem. Lett. 2017, 46, 711.
4. For the favorable effect of NHC ligands on
iron-catalyzed cross-coupling reactions, see: a) R. B.
Bedford, M. Betham, D. W. Bruce, A. A. Danopoulos,
R. M. Frost, M. Hird, J. Org. Chem. 2006, 71, 1104. b)
S. K. Ghorai, M. Jin, T. Hatakeyama, M. Nakamura,
Org. Lett. 2012, 14, 1066.
24. T. Hatakeyama, T. Hashimoto, Y. Kondo, Y. Fujiwara,
H. Seike, H. Takaya, Y. Tamada, T. Ono, M.
Nakamura, J. Am. Chem. Soc. 2010, 132, 10674.
25. In the reaction employing octylmagnesium bromide,
trace amounts of the by-products 1-octene and
hexadecane were observed by GC analysis.
26. For the trimethylsilylmethyl group acting as
a
nontransferable or unreactive dummy ligand in an
iron-catalyzed Negishi-type coupling, see: a) M.
Nakamura, S. Ito, K. Matsuo, E. Nakamura, Synlett
2005, 1794. b) R. B. Bedford, M. Huwe, M. C.
Wilkinson, Chem. Commun. 2009, 600. The
trimethylsilylmethyl group can be applied to
subsequent transformations, see: c) M. Leiendecker, C.
Hsiao, L. G. Nurtalya, N. Alandini, M. Rueping,
Angew. Chem. Int. Ed. 2014, 53, 12912.
1
5. a) T. Hatakeyama, M. Nakamura, J. Am. Chem. Soc.
2
007, 129, 9844. For the effects of the fluorido and
NHC ligands on iron catalysis, see: b) T. Hatakeyama,
S. Hashimoto, K. Ishizuka, M. Nakamura, J. Am.
Chem. Soc. 2009, 131, 11949. The application of an
iron fluoride/NHC combination for the coupling of
aryl sulfonates and sulfamates were reported by Cook
et al., see: ref. 18b and 18c. Iron-catalyzed

2
2
7. Double octylation product and starting material were
observed in ca. 15% and 12% yields, respectively,
which were determined by non-calibrated GC analysis.
8. Aryl chlorides are useful chemicals in industry and
widely available. However, aryl chlorides are less
reactive for cross-coupling reactions compared with
the corresponding bromides and iodides. For
palladium-catalyzed coupling reactions of aryl
chlorides, see: a) A. F. Littke, G. C. Fu, Angew. Chem.
Int. Ed. 2002, 41, 4176. The palladium-catalyzed
cross-coupling reaction of aryl chlorides with alkyl
Grignard reagents is limited, see: b) T. Katayama, M.
Umeno, Chem. Lett. 1991, 2073.
Chem. Soc. 2004, 126, 3686. i) T. Hatakeyama, Y.
Fujiwara, Y. Okada, T. Itoh, T. Hashimoto, S.
Kawamura, K. Ogata, H. Takaya, M. Nakamura, Chem.
Lett. 2011, 40, 1030. An Fe /Fe catalytic cycle for
the coupling with alkyl halides, see: j) T. Iwasaki, R.
Shimizu, R. Imanishi, H. Kuniyasu, N. Kambe, Chem.
Lett. 2018, 47, 763.
II
IV
37. Due to the low solubility of anhydrous FeF
FeF ·3H O was used with large excess amounts of
SIPr (7.0 equiv) and Me SiCH MgCl (12 equiv).
3
in THF,
3
2
3
2
38. The effectiveness of XAS-based studies for iron
catalysis was thoroughly demonstrated in the
mechanistic study of iron-SciOPP complex-catalyzed
cross-coupling reactions to determine the oxidation
state and precise geometry of the organoiron species,
see: ref. 36c.
2
9. 1-Octene was observed at 0.09, 4.27, and 3.40
equivalents based on the iron salt in Table 3, entries 4–
6
.
3
3
0. M. O’Neil, J. Cornella, Synthesis 2018, 50, 3974.
1. a) A. A. Danopoulos, P. Braunstein, M. Wesolek, K. Y.
Monakhov, P. Rabu, V. Robert, Organometallics 2012,
39. The XANES spectrum of the resulting organoiron(II)
intermediate 10 (high spin, S = 2) is shown in the
supporting information, Figure S1. A DFT-level core
3
1, 4102. b) K. L. Fillman, J. A. Przyojski, M. H.
Al-Afyouni, Z. J. Tonzetich, M. L. Neidig, Chem. Sci.
015, 6, 1178.
2. A similar trend was observed in the iron/NHC
catalyzed cross-coupling reactions, see: ref. 17 and
8b. Iron-catalyzed isopropylation of hetero aryl
excitation
[MgCl][Fe F
calculation based
on
II
2
(SIPr)(CH SiMe )] (S = 2) provided a
2
3
2
reasonable result for the 1s-3d electron transition for
the corresponding pre-edge peak.
3
3
-
1
40. The fitting calculation was performed by the FEFF6
program (revision 9.6.4) embedded with Artemis. For
the FEFF6 program, see: a) J. J. Rehr, J. J. Kas, F. D.
Vila, M. P. Prange, K. Jorissen, Phys. Chemi. Chem.
Phys. 2010, 12, 5503. b) J. J. Rehr, R. C. Albers, Rev.
Mod. Phys. 2000, 72, 621. For Artemis software, see:
c) M. Newville, J. Synchrotron Rad. 2001, 8, 322. d)
B. Ravel, M. Newville, J. Synchrotron Rad. 2005, 12,
537.
41. The fitting simulation with other possible iron
intermediates did not provide any reasonable matches
(see SI Figures S2–12). Five types of iron
intermediates were provided for the FEFF fitting of the
obtained EXAFS spectra. The multiple-scattering path
calculation was performed using atomic coordinates
obtained from DFT-calculated models of high (S = 2)
and low (S = 1) spin organoferrate(II) intermediates of
chlorides primarily provides the branched products,
see: J. N. Sanderson, A. P. Dominey, J. M. Percy, Adv.
Synth. Catal. 2017, 359, 1007.
3. The color of the reaction mixture was completely
different among the iron salts discussed in Table S1.
Iron fluoride affords a brown suspension, while iron
chloride, bromide, and iodide gave a black solution
after the addition of alkyl Grignard reagents. These
results indicate that a different catalytically active
intermediate is generated when using fluorido as
opposed to other halogen ligands, see: ref. 42.
3
3
4. E. Yamamoto, K. Izumi, Y. Horita, H. Ito, J. Am.
Chem. Soc. 2012, 134, 19997.
5. For a recent mechanistic study on iron-catalyzed
cross-couplings of aryl electrophiles, see: a) J.
Kleimark, P. Larsson, P.-F. Emamy, A. Hedström,
P.-O. Norrby, Adv. Synth. Catal. 2012, 354, 448. b) R.
Schoch, W. Desens, T. Werner, M. Bauer, Chem. Eur.
J. 2013, 19, 15816. See also ref. 15b.
6. For a recent mechanistic study on iron-catalyzed
cross-couplings of alkyl halides, see: a) D. Noda, Y.
Sunada, T. Hatakeyama, M. Nakamura, H. Nagashima,
J. Am. Chem. Soc. 2009, 131, 6078. b) S. L. Daifuku,
M. H. Al-Afyouni, B. E. R. Snyder, J. L. Kneebone, M.
L. Neidig, J. Am. Chem. Soc. 2014, 136, 9132. c) H.
Takaya, S. Nakajima, N. Nakagawa, K. Isozaki, T.
Iwamoto, R. Imayoshi, N. J. Gower, L. Adak, T.
Hatakeyama, T. Honma, M. Takagaki, Y. Sunada, H.
Nagashima, D. Hashizume, O. Takahashi, M.
Nakamura, Bull. Chem. Soc. Jpn. 2015, 88, 410. d) M.
Jin, L. Adak, M. Nakamura, J. Am. Chem. Soc. 2015,
[MgCl][FeF
iron
2
(SIPr)(CH
intermediates
SiMe ), and FeSIPr(CH
2 3
SiMe )] and high (S = 2) spin
of
FeF (SIPr),
3 2
SiMe ) at the
2
FeF(SIPr)(CH
PCMTHF/UB3LYP/6-311G (d,p) level. Among various
candidate molecular structures examined, the model of
2
3
2
3
the
high-spin
organoferrate(II)
intermediate
[MgCl][FeF
2
(SIPr)(CH
2
SiMe )] (S = 2) shows the best
3
agreement over the range of the second coordination
sphere with reasonable bond lengths.
2 3 2
42. The DFT-optimized structure of FeSIPr(CH SiMe )
which is in good accordance with the X-ray
crystallographic structure reported by Danopoulos.
This X-ray crystallographic structure resulted in the
appropriate R factor on the THF solution of FeCl
SIPr, and Me SiCH MgCl fitted by the same method,
3
,
3
2
showing the applicability of the technique, see: SI
Figures S18 and S19. The dialkyliron(II)/NHC
1
37, 7128. e) L. Adak, S. Kawamura, G. Toma, T.
Takenaka, K. Isozaki, H. Takaya, A. Orita, H. C. Li, T.
K. M. Shing, M. Nakamura, J. Am. Chem. Soc. 2017,
complex, FeSIPr(CH
FeCl (thf)1.5 and Me SiCH
2
2
SiMe
3 2
) , was synthesized from
2
3
MgCl. See also ref. 31.
1
39, 10693. f) A. K. Sharma, W. M. C. Sameera, M.
43. For a review of the anionic ate complex, see: a) T.
Kauffmann, Angew. Chem. Int. Ed. 1996, 35, 386. b) J.
Terao, N. Kambe, Acc. Chem. Res. 2008, 41, 1545.
44. For the alkyl ferrate complex, see: a) B. K. Bower, H.
G. Tennent, J. Am. Chem. Soc. 1972, 94, 2512. b) M.
H. Al-Afyouni, K. L. Fillman, W. W. Brennessel, M.
L. Neidig, J. Am. Chem. Soc. 2014, 136, 15457. c) S.
Jin, L. Adak, C. Okuzono, T. Iwamoto, M. Kato, M.
Nakamura, K. Morokuma, J. Am. Chem. Soc. 2017,
1
Chem. Soc. 2017, 139, 16126, and the references cited
therein. See also selected papers from our group: h) M.
Nakamura, K. Matsuo, S. Ito, E. Nakamura, J. Am.
39, 16117. g) W. Lee, J. Zhou, O. Gutierrez, J. Am.

B. Muñoz III, S. L. Daifuku, W. W. Brennessel, M. L.
Neidig, J. Am. Chem. Soc. 2016, 138, 7492. d) T.
Parchomyk, S. Demeshko, F. Meyer, K. Koszinowski,
J. Am. Chem. Soc. 2018, 140, 9709. For the aryl
ferrate complex, see: e) R. B. Bedford, P. B. Brenner,
E. Carter, P. M. Cogswell, M. F. Haddow, J. N.
Harvey, D. M. Murphy, J. Nunn, C. H. Woodall,
Angew. Chem. Int. Ed. 2014, 53, 1804. f) F. E.
Zhurkin, M. D. Wodrich, X. Hu, Organometallics
2
017, 36, 499. g) M. Clémancey, T. Cantat, G.
Blondin, J.-M. Latour, P. Dorlet, G. Lefèvre, Inorg.
Chem. 2017, 56, 3834.
4
4
5. A singlet (S = 0) potential energy surface is quite high
compared with triplet (S = 1) and quintet (S = 2)
potential energy surfaces, which is confirmed by a
preliminary
DFT
study
at
the
PCMTHF/UB3LYP-GD3BJ/6-31G(d) level. Thus, we
excluded a singlet energy surface in the energy profile.
6. a) T. Weymuth, E. P. A. Couzijn, P. Chen, M. Reiher,
J. Chem. Theory Comput. 2014, 10, 3092. b) Y. Zhao,
D. G. Truhlar, J. Chem. Physics. 2006, 124, 224105. c)
L. Goerigk, S. Grimme, Phys. Chem. Chem. Phys.
2
011, 13, 6670. d) M. Mammen, E. I. Shakhnovich, J.
M. Deutch, G. M. Whitesides, J. Org. Chem. 1998, 63,
821. The detailed energy profile is shown in Figure
3
S21.
4
7. For nickel-mediated oxidative addition assisted by a
magnesium ion as a Lewis acid, see: a) N. Yoshikai, H.
Matsuda, E. Nakamura, J. Am. Chem. Soc. 2009, 131,
9
7
590. b) J.-R. Wang, K. Manabe, Org. Lett. 2009, 11,
41. c) C. Lohre, T. Dröge, C. Wang, F. Glorius,
Chem. Eur. J. 2011, 17, 6052. d) H. Ogawa, H.
Minami, T. Ozaki, S. Komagawa, C. Wang, M.
Uchiyama, Chem. Eur. J. 2015, 21, 13904.
4
4
8. For effects of the fluorido ligand in transition-metal
catalysis, see: a) K. Fagnou, M. Lautens, Angew.
Chem. Int. Ed. 2002, 41, 26.
9. a) M. Tilset, J.-R. Hamon, P. Hamon, Chem. Commun.
1
998, 765. b) M. Tilset, I. Fjeldahl, J.-R. Hamon, P.
Hamon, L. Toupet, J.-Y. Saillard, K. Costuas, A.
Haynes, J. Am. Chem. Soc. 2001, 123, 9984.
5
0. a) V. S. Asirvatham, Z. Yao, K. J. Klabunde, J. Am.
Chem. Soc. 1994, 116, 5493. b) A. Casitas, J. A. Rees,
R. Goddard, E. Bill, S. DeBeer, A. Fürstner, Angew.
Chem. Int. Ed. 2017, 56, 10108.
5
5
1. For a review of spin-forbidden reactions, see: R. Poli,
J. N. Harvey, Chem. Soc. Rev. 2003, 32, 1.
2. S. M. Bellows, T. R. Cundari, P. L. Holland,
Organometallics 2013, 32, 4741.