α-Substitution effect of fluorine atoms in the cobalt-catalyzed hydrosilylation of fluorine-containing aromatic ketones

Article abstract of DOI:10.1016/j.jfluchem.2017.12.003

Hydrosilylation of fluorine-containing aromatic ketones with triethylsilane in the presence of a catalytic amount of a low-cost and an environmentally benign cobalt catalyst proceeded very smoothly to give the corresponding adducts. Interestingly, this hy

Full text of DOI:10.1016/j.jfluchem.2017.12.003

JournalofFluorineChemistry206(2018)8–18
Contents lists available at ScienceDirect
Journal of Fluorine Chemistry
journal homepage: www.elsevier.com/locate/fluor
α-Substitution effect of fluorine atoms in the cobalt-catalyzed
hydrosilylation of fluorine-containing aromatic ketones
Tatsuya Kumon, Siti Asiah Binti Mohd Sari, Shigeyuki Yamada, Tsutomu Konno⁎
Faculty of Molecular Chemistry and Engineering, Kyoto Institute of Technology, Matsugasaki, Sakyo-ku, Kyoto 606-8585, Japan
A R T I C L E I N F O
A B S T R A C T
Keyword:
Fluorine
Cobalt catalyst
Hydrosilylation
Fluorinated ketone
Silane
Hydrosilylation of fluorine-containing aromatic ketones with triethylsilane in the presence of a catalytic amount
of a low-cost and an environmentally benign cobalt catalyst proceeded very smoothly to give the corresponding
adducts. Interestingly, this hydrosilylation reaction was not applicable to non-fluorinated ketones. As revealed
by density functional theory analysis, a fluorine substituent at the α-position plays a crucial role in facilitating
the hydrosilylation reaction.
1. Introduction
metal catalysts. We have reported that the Co2(CO)8-catalyzed hydro-
silylation of fluoroalkylated internal alkynes [12a] or β-fluoroalkyl-α,β-
Transition metal-catalyzed hydrosilylation reaction of carbon–-
carbon multiple bonds or a carbonyl group is well recognized as one of
the most elegant and reagent economical approaches to prepare orga-
nosilicon compounds, which are useful intermediates in the fields of
medicinal and materials chemistry [1]. Owing to the broad utility of the
hydrosilylation reaction, extensive studies on the development of effi-
cient hydrosilylation protocols using transition metal catalysts have
been conducted so far, and many useful hydrosilylative transformations
using catalysts based on late transition metals such as Pt, Pd, Rh, Ru,
and Ir have been reported [2–5]. However, owing to the environmental
impact and resource depletion associated with the use of precious-metal
catalysts, the development of alternative hydrosilylation protocols
using environmentally friendly and inexpensive transition metal cata-
lysts is necessary. Consequently, significant research effort has been
devoted to the development of hydrosilylation reactions using catalysts
based on first-row transition metals such as Fe [6], Co [7], and Cu [8],
which are naturally abundant, economical, and less toxic. The broad
utility of such catalysts for hydrosilylation reactions has been demon-
strated over the past decade [6–8].
unsaturated ketones [12b] proceed successfully to afford the corre-
sponding fluorine-containing vinyl silanes or 1,4-hydrosilylated enol
silane derivatives, respectively, demonstrating the potent utility of low-
cost and environmentally benign cobalt catalysts for the effective
synthesis of organofluorine compounds (Fig. 1A).
During our extensive investigation, fluorine-containing ketones
were effectively transformed into the corresponding silyl-protected
fluorinated alcohols by cobalt-catalyzed hydrosilylation (Fig. 1B) [13].
In addition, it was observed that the catalyst system is not applicable to
non-fluorinated ketones, i.e., fluorine atoms in the substrate are crucial
for the cobalt-catalyzed hydrosilylation process. In this article, we
discuss the cobalt-catalyzed hydrosilylation of fluorine-containing ke-
tones in detail. Furthermore, a reaction mechanism that rationalizes the
unique α-substitution effect exerted by a fluorine atom is proposed
based on the results of density functional theory (DFT) calculations.
2. Results and discussion
2.1. Screening of reaction conditions
Our research group has hitherto devoted extensive effort to the
development of fluorinated functional materials [9], fluorinated syn-
thetic intermediates [10], and transition metal-catalyzed synthetic
transformations [11] in order to establish useful synthetic protocols for
a wide variety of organofluorine compounds. Notably, in the course of a
recent research project on transition metal-catalyzed transformations,
our interest was directed toward the development of novel synthetic
protocols for organofluorine compounds using first-row transition
An initial screening of the reaction conditions for cobalt-catalyzed
hydrosilylation reactions was undertaken using the α,α,α-tri-
fluoroacetophenone derivative 1aA. The results are summarized in
Table 1.
Upon refluxing 1aA with Et3SiH (1.2 equiv.) in dichloroethane in
the presence of a catalytic amount of Co2(CO)8 (5.0 mol%) for 3 h, the
desired hydrosilylation occurred smoothly to afford the corresponding
⁎
Corresponding author.
E-mail address: konno@kit.ac.jp (T. Konno).
https://doi.org/10.1016/j.jfluchem.2017.12.003
Received 7 November 2017; Received in revised form 29 November 2017; Accepted 4 December 2017
Availableonline06December2017
0022-1139/©2017ElsevierB.V.Allrightsreserved.

T. Kumon et al.
JournalofFluorineChemistry206(2018)8–18
Fig. 1. Cobalt-catalyzed transformations of fluorine-containing molecules.
Table 1
Initial screening of hydrosilylation conditions.
Entry
Solvent
Si-H
X/mol%
Yielda/% of 2aA
Recoverya/% of 1aA
1
2
3
4
5
6
7
8
ClCH2CH2Cl
Hexane
Et3SiH
Et3SiH
Et3SiH
Et3SiH
Et3SiH
Et3SiH
Et3SiH
PhMe2SiH
(EtO)3SiH
Me(EtO)2SiH
i-Pr3SiH
5.0
5.0
5.0
3.0
1.0
0.5
0.0
1.0
1.0
1.0
1.0
75
18
00
00
00
00
18
90
73
Quant.
Quant.
Quant.
Quant. (84)
80
Trace
23
76
Toluene
Toluene
Toluene
Toluene
Toluene
Toluene
Toluene
Toluene
Toluene
9
10
11
10
18
Quant.
69
00
a
Determined by 19F NMR. Value in parentheses is the isolated yield.
adduct 2aA in 75% yield (Table 1, entry 1). As shown in Table 1, entries
2 and 3, replacing the solvent with hexane or toluene provided sig-
nificant improvement in the yield, leading to quantitative formation of
2aA. With a lower catalyst loading (1.0–3.0 mol%), the hydrosilylated
adduct 2aA is obtained in a quantitative yield as well (entries 4 and 5).
However, further decrease in catalyst loading (0.5 mol%) resulted in
the sluggish formation of the hydrosilylation product (entry 6). Fur-
thermore, no 2aA was obtained in the absence of the Co2(CO)8 catalyst
(entry 7). Thus, the presence of the cobalt catalyst was crucial for the
hydrosilylation of fluorinated ketone 1aA. Other reductants instead of
Et3SiH, including PhMe2SiH, (EtO)3SiH, and Me(EtO)2SiH, also parti-
cipated in the cobalt-catalyzed hydrosilylation reaction, although the
yields of 2aA were slightly lower (entries 8–10). In sharp contrast, the
use of i-Pr3SiH as a reductant was not suitable for the present cobalt-
catalyzed hydrosilylation protocol, resulting in quantitative recovery of
1aA. These results indicate that hydrosilanes possessing bulky sub-
stituents tend to suppress the formation of the desired hydrosilylation
adduct. Informed by these experiments, we selected the reaction con-
ditions shown in Table 1, entry 5 as optimal for the present cobalt-
catalyzed hydrosilylation.
2.2. Substrate scope of the cobalt-catalyzed hydrosilylation reaction
Using the optimal reaction conditions, we performed cobalt-cata-
lyzed hydrosilylation reactions using various fluorinated ketones (1a).
The results obtained are presented in Table 2.
Hydrosilylation of α,α,α-trifluoroacetophenone (1aB), which has no
substituents on the aromatic ring, with 1.2 equiv. of Et3SiH in the
9

T. Kumon et al.
JournalofFluorineChemistry206(2018)8–18
Table 2
2aG both being obtained in only 19% yield, respectively (entries 8 and
9). In these cases, the unreacted starting substrates 1aF and 1aG were
obtained in 39% and 81% recovery yields, respectively, as mixtures of
the hydrate forms. The distinct difference in the recovery yields of 1aF
and 1aG may be caused by their thermal properties, including boiling
point and volatility. 1aF had a lower boiling point (66–67 °C/34 Torr
[14a]), which was close to the reflux temperature of toluene (110.6 °C/
760 Torr), than that of 1aG (183 °C/760 Torr [14b]). However, upon
treating 1aG with increased amounts of Et3SiH (1.5 equiv.) and
Co2(CO)8 (5.0 mol%) in toluene at reflux for 3 h, the hydrosilylation
proceeded very smoothly to afford 2aG in 94% NMR (89% isolated)
yield. Although the hydrosilylation of the sterically bulky 1-naphthyl
trifluoromethyl ketone (1aH) under the optimal conditions was some-
what sluggish (Table 2, entry 11), a dramatic improvement in the yield
of 2aH (95% isolated yield) became observed when the reaction was
carried out using 1.5 equiv. of Et3SiH and 3.0 mol% Co2(CO)8 (entry
12). As shown in entries 13 and 14, the use of the aliphatic fluorinated
ketone 1aI was not applicable to the present cobalt-catalyzed hydro-
silylation protocol and results in a large recovery of 1aI (> 93%) as a
mixture with the hydrate forms, despite extensive attempts to identify
suitable reaction conditions.
Cobalt-catalyzed hydrosilylation of different fluorinated ketones.
Entry
R
Product
Yielda/% of 2a
Recoverya,b/% of 1a
01
4-MeC6H4 (A)
Ph (B)
Ph (B)
2aA
2aB
2aB
2aC
2aD
2aE
2aE
2aF
2aG
2aG
2aH
2aH
2aI
Quant. (84)
71
Quant. (83)
Quant. (95)
Quant. (84)
24
91 (80)
19
19
94 (89)
47
96 (95)
Trace
00
18
00
00
Trace
66
09
39
81
Trace
46
00
95
93
02
03c
04
4-MeOC6H4 (C)
3-MeOC6H4 (D)
2-MeOC6H4 (E)
2-MeOC6H4 (E)
4-FC6H4 (F)
4-ClC6H4 (G)
4-ClC6H4 (G)
1-Naphthyl (H)
1-Naphthyl (H)
n-C10H21 (I)
n-C10H21 (I)
05
06
07c
08
09
10d
11
12e
13
14e
2aI
Trace
Subsequently, we examined the cobalt-catalyzed hydrosilylation
reaction using different aromatic ketones with a variety of fluoroalk-
anoyl moieties (1b–e). The results obtained are summarized in Table 3.
Using the optimal reaction conditions presented in entry 5 in
Table 1, reaction of the fluorinated aromatic ketone 1bA carrying a n-
C3F7 moiety with 1.2 equiv. of Et3SiH in the presence of 1.0 mol%
Co2(CO)8 in toluene at reflux for 3 h did not provide satisfactory results,
leading instead to a large recovery of 1bA (entry 2). Upon further in-
vestigation, the reaction efficiency using 1bA was significantly im-
proved, with the corresponding hydrosilylated adduct 2bA being ob-
tained in 93% isolated yield when the reaction was conducted with 1.5
equiv. of Et3SiH in the presence of 6.0 mol% Co2(CO)8 for 24 h (entry
3). As can be seen in entries 4–5, the use of 2,3,4,5,6-penta-
fluorobenzophenone (1cB), where Rf = C6F5 and R = Ph, as a sub-
strate was not applicable to the present cobalt-catalyzed hydrosilylation
protocol, despite several attempts to identify suitable reaction condi-
tions. This drastic suppression of the hydrosilylation process is most
likely caused by a combination of the steric bulkiness and low solubility
of 1cB in toluene. When the α,α-difluoroacetophenone derivative 1dA
was used in the present protocol, the reaction proceeded very well to
give rise to the corresponding adduct 2dA in quantitative (97% iso-
lated) yield (entry 6). In contrast, the use of the α-fluoroacetophenone
derivative 1eA as a substrate drastically retarded the reaction and a
large amount (83%) of unreacted 1eA and its hydrate were recovered
a
Determined by 19F NMR. Values in parentheses are the isolated yields.
Indicated combined yield of 1a and the corresponding hydrate.
Carried out for 18 h.
With 5.0 mol% of Co2(CO)8 and 1.5 equiv. of Et3SiH.
With 3.0 mol% of Co2(CO)8 and 1.5 equiv. of Et3SiH.
b
c
d
e
presence of 1.0 mol% Co2(CO)8 in toluene at reflux temperature for 3 h
proceeded smoothly to afford the corresponding hydrosilylated adduct
2aB in 71% yield along with a small amount of unreacted 1aB (entry 2).
However, extending the reaction time from 3 to 18 h allowed complete
consumption of 1aB, resulting in the formation of 2aB in 83% isolated
yield (Table 2, entry 3). Reaction of 1aC or 1aD with a strong electron-
donating group, i.e. a methoxy group, on the benzene ring at the para or
meta position, respectively, proceeded very well to afford the corre-
sponding silyl ether 2aC or 2aD in 95% or 84% yield, respectively
(entries 4 and 5). In sharp contrast, the reaction of fluorinated ketone
1aE with a methoxy group at the ortho position under the same con-
ditions became sluggish, leading to the formation of 2aE in only 24%
NMR yield and 66% recovery of 1aE as a mixture of the hydrate forms
(entry 6). After several attempts, it was found that extending the re-
action time to 18 h led to the formation of 2aE in high isolated yield
(entry 7). Conversely, fluorinated ketones 1aF and 1aG having an
electron-withdrawing fluoro or chloro substituent, respectively, on the
benzene ring at the para position were observed to be less reactive for
the present hydrosilylation reaction, with the corresponding 2aF and
Table 3
Cobalt-catalyzed hydrosilylation using different aromatic ketones.
Entry
Rf
R
Product
Yielda/% of 2
Recoverya,b/% of 1
01
02
03b
04
05b
06
07
08
09
CF3 (a)
4-MeC6H4 (A)
4-MeC6H4 (A)
4-MeC6H4 (A)
Ph (B)
2aA
2bA
2bA
2cB
2cB
2dA
2eA
2fA
Quant. (84)
6
Quant. (93)
0
00
94
00
99
73
0
83
(94)
(89)
n-C3F7 (b)
n-C3F7 (b)
C6F5 (c)
C6F5 (c)
CF2H (d)
CFH2 (e)
CH3 (f)
Ph (B)
20
4-MeC6H4 (A)
4-MeC6H4 (A)
4-MeC6H4 (A)
Ph (B)
Quant. (97)
Trace
0
0
CO2Et (g)
2gB
a
Determined by 19F NMR. Values in parentheses are the isolated yields.
Carried out for 24 h using 6.0 mol% of Co2(CO)8 and 1.5 equiv. of Et3SiH.
b
10

T. Kumon et al.
JournalofFluorineChemistry206(2018)8–18
Scheme 1. Possible reaction mechanism for cobalt-catalyzed hydro-
silylation.
(entry 7). These results imply that the number of fluorine atoms at the
α-position is an important factor for the cobalt-catalyzed hydrosilyla-
tion reaction.
at room temperature for 0.5 h followed by removal of the excess Et3SiH,
Et3Si–Co(CO)4 was successfully obtained as a brown oil. The structure
was confirmed by typical spectroscopic measurements such as NMR and
IR. IR analysis confirmed the disappearance of the Si–H stretching vi-
bration (2100 cm−1) of Et3SiH and the signal associated with the
bridging C]O bond (1860 cm−1) of Co2(CO)8. With the pre-generated
Et3Si–Co(CO)4 in hand, we performed a cobalt-catalyzed hydrosilyla-
tion reaction using 1aB in toluene at reflux temperature for 3 h and
successfully obtained the corresponding product 2aB in 91% isolated
yield. Thus, it was safely concluded that the actual pre-catalyst was
Et3Si–Co(CO)4.
To gain further information on the reaction mechanism, a crossover
experiment for the hydrosilylation reaction was carried out using an
equimolar amount of α,α,α-trifluoroacetophenone (1aB) and the non-
fluorinated acetophenone 1fB under the optimal conditions shown in
Table 1, entry 5 (Scheme 3).
In order to clarify the effect of the fluorine substituent at the α-
position in the present hydrosilylation protocol, we also examined co-
balt-catalyzed hydrosilylation reaction using ketones and aldehydes
without any fluorine substituents such as 1fA and 1gB, and the results
are also listed in Table 3. As shown in Table 3, entry 8, the reaction
using the non-fluorinated acetophenone derivative 1fA did not proceed
at all, resulting in almost quantitative recovery (94%) of unreacted 1fA.
When ethyl phenylglyoxylate (1gB), where Rf = CO2Et and R = Ph
and in which the CO2Et group had a similar electron-withdrawing
nature to that of the CF3 moiety in 1aA [15], was adopted as the sub-
strate, no reaction occurred. These results indicate that the cobalt-cat-
alyzed hydrosilylation is far more influenced by the steric bulkiness
around the carbonyl moiety than its electronic nature.
The treatment of a mixture of 1aB and 1fB (1.0 equiv. each) with
1.2 equiv. of Et3SiH in the presence of 1.0 mol% Co2(CO)8 in toluene at
reflux temperature for 3 h unexpectedly afforded neither 2aB nor 2fB,
resulting instead in the quantitative recovery of both 1aB and 1fB.
2.3. Possible reaction mechanism
A possible reaction mechanism for the cobalt-catalyzed hydro-
silylation of fluorinated ketones is shown in Scheme 1 [16,17].
The reaction of Co2(CO)8 with two equivalents of Et3SiH generates
the silylcobalt precatalyst Et3Si–Co(CO)4 accompanied by hydrogen gas
evolution. This precatalyst immediately eliminates CO to generate the
active silylcobalt complex Et3Si–Co(CO)3 (Complex A) [16]. The active
Complex A has a vacant d orbital on the cobalt atom and undergoes
coordination by the carbonyl oxygen of 1 to form Int-1, which is sub-
jected to silylmetalation to form Int-2 via a transition state (TS-1).
Subsequent oxidative addition of the Co(I) species in Int-2 to the Si–H
bond of unreacted Et3SiH furnishes Co(III)-intermediate (Int-3) via TS-
2, which immediately undergoes reductive elimination via TS-3 to af-
ford the desired hydrosilylated adduct 2 along with regeneration of the
active Et3Si–Co(CO)3 complex. The detailed reaction mechanism using
a quantum chemical calculation will be discussed in Section 2.4.
In order to figure out the actual composition of the cobalt catalyst,
we demonstrated the hydrosilylation reaction of α,α,α-tri-
fluoroacetophenone (1aB) using pre-generated Et3Si–Co(CO)4 prepared
in advance as a catalyst (Scheme 2).
2.4. Theoretical evaluation of the reaction mechanism
2.4.1. Effect of fluorine-substitution at the α-position
In order to gain more information on the reaction mechanism, we
carried out a theoretical study using a Gaussian 09W with a density
functional theory (DFT) at the B3LYP/6-31+G(d,p) level of theory as a
basis set [18]. To explore the pathway of the cobalt-catalyzed hydro-
silylation reaction using fluorinated ketone A as well as the non-
fluorinated ketone B by DFT calculation, Me3Si–Co(CO)3 was adopted
as a model catalyst structure and Me3SiH as a reductant. The DFT
calculation elucidated the presence of three transition states, i.e., those
for the silylmetalation step (TS-1), the oxidative addition step (TS-2),
and the reductive elimination step (TS-3), and the geometry of the
transition states was characterized by vibrational analysis, which con-
firmed that the obtained geometries possess a single imaginary fre-
quency. Reaction pathways were traced by the intrinsic reaction co-
ordinate (IRC) method [19] to obtain the energy-minimum geometries,
and IRC analysis of TS-1 provided two energy-minimum geometries,
assigned as Int-1 for the reactant and Int-2 for the product. Similarly,
The silylcobalt complex was prepared according to a previous lit-
erature method [16]. Upon treating Co2(CO)8 with 3.0 equiv. of Et3SiH
11

T. Kumon et al.
JournalofFluorineChemistry206(2018)8–18
Scheme 2. Cobalt-catalyzed hydrosilylation of 1aB using pre-generated
Et3Si–Co(CO)4 as a catalyst.
IRC analysis of TS-2 also provided two energy-minimum geometries
(Int-3 for the reactant and Int-4 for the product). On the basis of the
calculation results, energy profiles of the reaction pathways are illu-
strated in Fig. 2.
the sterically bulky phenyl group in Int-1A.
(3) The dihedral angle ϕ (C2eC1eOeCo) in Int-1A was 18.3°, which
deviated significantly from the plane structure of an ideal carbonyl
moiety (ϕ = 0°), whereas a virtually planar structure (ϕ = 6.6°)
was observed in the case of Int-1B.
The energy profiles revealed that the energy-minimum geometry for
fluorinated ketone (Int-1A) was 4.5 kcal mol−1 higher than that for the
non-fluorinated ketone (Int-1B). Thus, the stabilities of Int-1 were in
the order Int-1A < Int-1B, whereas the stabilities for TS-1 presented
the opposite trend, i.e., TS-1A > TS-1B. To understand the crossing
stability of Int-1 and TS-1 between fluorinated ketone A and non-
fluorinated ketone B, the geometries for Int-1, TS-1, and Int-2 for A and
B were evaluated in detail. The geometries calculated are shown in
Fig. 3 and the observed structural data are summarized in Table 4.
Upon careful observation of the energy-minimum geometries for
Int-1A and Int-1B shown in Table 4, large differences were observed
for the CoeO bond lengths, the coordination bond angles ∠C1eOeCo,
and the dihedral angle (ϕ) between C2eC1eOeCo, as follows:
Taking the above factors into account, it can be concluded that
larger deviations from the ideal values for bond length, angle, and di-
hedral angle in the case of Int-1A cause an increase in its total energy
and thus a decrease in its relative stability.
Fig. 3 also provided us several understandings on the transition state,
TS-1, for the silylmetallation process. Thus, analyzing a bond-length of
CoeSi for Int-1 and TS-1, dramatically elongated interatomic distance
(18 pm each) between Co⋯Si in the TS-1 was observed in both fluori-
nated A and non-fluorinated B; the significant elongation of Co⋯Si bond
demonstrates to be completed its bond-breaking during silylmetallation
process across the TS-1. On the other hand, comparing interatomic dis-
tance between Co⋯C1 for both TS-1 and Int-2, a little (only 4–9 pm)
change of the interatomic distance was observed in spite of the presence
or absence of fluorine-substituents. The similar Co–C1 bond length be-
tween TS-1 and Int-2 indicates that bond formation between Co and C1
atoms is almost completely achieved during the silylmetallation process.
Judging from these theoretical insights, the TS-1 for silylmetallation
process could be safely posited as a late-transition state, which generally
possesses product-like character and stability; accordingly, it is reason-
ably concluded that total energy of TS-1A for fluorinated ketone is lower
than that of TS-1 B for non-fluorinated one due to the stabilization effect
reflected in the stability of Int-2. The energetic difference between a
transition state and a substance, TS-1 and Int-1 in this case, indicates the
activation energy (Ea), which can be defined as the minimum energy
required for a chemical reaction to occur. The Ea values at each transition
state are summarized in Table 5.
(1) The interatomic distance of CoeO in Int-1A was 216 pm, which
was 4 pm longer than the coordination bond length in Int-1B. The
longer CoeO bond length in Int-1A may be brought about by the
electron-withdrawing character of a CF3 group, which strongly
weakens the Lewis basicity of the carbonyl oxygen.
(2) The coordination bond angle, ∠C1eOeCo, was 153.8° for Int-1A, in
which the Co atom was existed at the opposite side to the CF3
moiety. In contrast, the angle for Int-1B was 136.2°, and the Co
atom was presented at the same side as the CH3 group (Fig. 3a and
d). The steric bulkiness of the substituents were in the order:
Ph > CF3 > CH3 [20], indicating that both Int-1A and Int-1B
should undergo coordination from the same side as the CF3 or CH3
moiety. The opposite observation for Int-1A indicates that not only
steric bulkiness but also the electron-abundance of the CF3 moiety
strongly affect the coordination process to form Int-1A. In addition,
the bond angle for Int-1A deviated more from the ideal angle of the
sp2 hybridized orbital for oxygen (120°) than that for Int-1B, which
may be due to steric repulsion between the silylcobalt complex and
Comparing the activation energies for fluorinated A and non-fluori-
nated B to across TS-2 and TS-3, there was no difference in the Ea values
(Ea-2A, Ea-2B for TS-2 and Ea-3A, Ea-3B for TS-3), whereas a significant
difference was observed in Ea-1A and Ea-1B to cross TS-1A or TS-1B,
Scheme 3. Crossover experiment for the hydrosilylation reaction using a
1:1 ratio of 1aB and 1fB as substrates.
12

T. Kumon et al.
JournalofFluorineChemistry206(2018)8–18
Fig. 2. Energy profiles for the hydrosilylation of fluorinated and non-fluorinated ketones calculated by IRC analysis.
respectively, with Ea-1A being much lower than Ea-1B. This means that
the fluorinated ketone can cross TS-1A to form the more stable Int-2A
more easily than non-fluorinated substrate can cross TS-1B.
cross TS-1B and thus a drastic suppression of further silylmetalation.
Taking the results of the crossover experiment shown in Scheme 3 into
consideration, the hypothesis is concluded to be reasonable.
Considering all the theoretical arguments above, it can be assumed
that the silylmetalation process is the key reaction step in the cobalt-
catalyzed hydrosilylation reaction for the fluorinated ketone, which is
attributable to the low value of Ea and the formation of the thermally
stable Int-2. In sharp contrast, the non-fluorinated ketone forms a co-
ordination complex (Int-1B) that is more stable than Int-1A, which
causes a significant increase in the activation energy Ea required to
2.4.2. Effect of the number of fluorine atoms at the α-position
As mentioned above, our investigation of the present cobalt-cata-
lyzed hydrosilylation revealed an interesting phenomenon in that the
CF3- and CF2H-substituted ketones 1aA and 1dA are very smoothly
converted to the corresponding product, but no adduct was obtained in
the case of the CFH2-substituted ketone 1eA. To make this intriguing
Fig. 3. (a, d) Energy-minimum geometries (Int-1A for the
fluorinated ketone and Int-1B for the non-fluorinated ketone)
calculated by IRC analysis, (b, e) geometries for TS-1A and
TS-1B, and (c, f) energy-minimum geometries for Int-2A and
Int-2B calculated by IRC analysis.
13

T. Kumon et al.
JournalofFluorineChemistry206(2018)8–18
Table 4
Structural data for Int-1, TS-1, and Int-2 for A and B calculated by DFT calculations.
For fluorinated ketone A
For non-fluorinated ketone B
Int-1A
TS-1A
Int-2A
Int-1B
TS-1B
Int-2B
∠C1eOeCo [°]
ϕ (C2eC1eOeCo) [°]
Co⋯O [pm]
153.8
18.3
216
–
–
–
–
–
–
136.2
6.6
212
–
–
–
–
–
–
Co⋯Si [pm]
242
–
115.3
119.1
125.5
260
202
111.1
113.1
120.9
–
241
–
120.9
120.2
118.9
259
207
113.6
114.3
120.3
–
Co⋯C1 [pm]
198
112.9
113.2
114.0
198
109.3
111.6
118.6
∠C2eC1eO [°]
∠C2eC1eC3 [°]
∠C3eC1eO [°]
Table 5
largest values (∠C1eOeCo = 151.1° and ϕ = 9.95°), which was due to
the effects of both steric hindrance and electron-withdrawal induced by
multiple fluorine atoms at the α-position. The large bond angle and
dihedral angle in Int-1C1 presented the largest deviations from an ideal
sp2-hybridized structural conformation of the five intermediates, which
led to higher total energies for Int-1C (with the CF2H-substituent) than
those for Int-1D (with the CFH2-substituent).
Activation energies for each transition state for the fluorinated and non-fluorinated ke-
tones.
Activation energy [kcal mol−1
]
Fluorinated ketone (A)
Non-fluorinated ketone (B)
Ea-1A (13.38)
Ea-1B (21.90)
Ea-2A (10.21)
Ea-2B (10.36)
Ea-3A (10.73)
Ea-3B (9.28)
The fact that the calculated structures at TS-1 are conformationally
similar to that of the intermediate Int-2 for both the CF2H- and CFH2-
derivatives clearly demonstrates that the silylmetalation process for
these analogues also proceeds via a late transition state [21]. Accord-
ingly, it is clear that the energy level of TS-1C for the CF2H-substituted
derivative is lower than that of TS-1D for the CFH2-substituted ana-
logue. The activation energies Ea-1C and Ea-1D required to cross TS-1C
and TS-1D, respectively, were 13.63 for Ea-1C and 16.45 kcal mol−1 for
Ea-1D. The lower value of Ea-1C for the case of the CF2H-containing
substrate significantly facilitates the silylmetalation process, allowing it
to afford the corresponding hydrosilylation product smoothly. Thus, it
can be concluded that fluorine atoms substituted at the α-position of
acetophenone play a pivotal role in decreasing the activation energy
required to cross TS-1 and, thereby, dramatically facilitates the cobalt-
catalyzed hydrosilylation reaction.
phenomenon clear, we carried out further reaction pathway analysis for
the cobalt-catalyzed hydrosilylation of CF2H-substituted (C) and CFH2-
substituted (D) ketones by DFT calculation with the same theory level
and basis set. The energy profiles calculated are shown in Fig. 4, in
which only the energy profile for the initial silylmetalation process is
shown as it is the key reaction step in the hydrosilylation reaction.
As shown in Fig. 4, DFT calculations revealed the presence of two
coordination intermediates (Int-1C) for the CF2H-containing substrate
(C) and three coordination intermediates (Int-1D) for the CFH2-con-
taining substrate (D) in the silylmetalation process. Among the five
intermediates, as shown in Fig. 5 and Table 6, the two intermediates
Int-1C1 and Int-1C2 possessed larger coordination bond angles of
∠C1eO1eCo and dihedral angles (ϕ) between C2eC1eOeCo than those
of the other three intermediates Int-1D, and Int-1C1 exhibited the
Fig. 4. Energy profile for the hydrosilylation of CF2H- (C) and CFH2-substituted (D) substrates calculated by IRC analysis.
14

T. Kumon et al.
JournalofFluorineChemistry206(2018)8–18
Fig. 5. (a, b) Energy-minimum geometries for Int-1C (with a
CF2H moiety) calculated by IRC analysis, (c) geometries for
TS-1C, (d–f) energy-minimum geometries for Int-1D (with a
CFH2 moiety) calculated by IRC analysis, and (g) geometries
for TS-1D.
3. Conclusion
4. Experimental
In this study, we have demonstrated that replacement of hydrogen
atoms at the α-position of acetophenone derivatives by fluorine atoms
made it possible to proceed cobalt-catalyzed silylmetalation reactions,
and we have elucidated the effects of fluorine substitution by theore-
tical calculations. Cobalt-catalyzed hydrosilylation of α,α,α-trifluoro-
or α,α-difluoroacetophenone derivatives with Et3SiH occurred
smoothly to provide the corresponding hydrosilylated products in high
yields, whereas no reaction occurred when non-fluorinated or α-fluor-
oacetophenone derivatives were used as substrates. In addition, various
α,α,α-trifluoroacetophenone derivatives having an electron-donating
or -withdrawing substituent on the aromatic ring also participated well
in cobalt-catalyzed hydrosilylation, leading to fluorinated silyl ether
derivatives in good to excellent yields. From theoretical evaluation
using DFT calculations, the dramatic difference between the fluorinated
and non-fluorinated substrates was found to significantly influence the
activation energy required to across the transition state in the si-
lylmetalation process; increasing fluorine substitution at the α-position
of acetophenone derivatives causes a dramatic decrease in the activa-
tion energy. The findings obtained in this study provide valuable in-
formation on the fluorine-substitution effect in organofluorine mole-
cules and will facilitate the design and development of new organic
materials as well as pharmaceuticals consisting of organofluorine
compounds.
4.1. General experimental procedures
Infrared spectra (IR) were determined in a liquid film on a NaCl
plate with a JASCO FT/IR-4100 type A spectrometer and all spectra are
reported in wavenumber (cm−1). 1H and 13C NMR spectra were mea-
sured with
a
Bruker AVANCE III 400 NMR spectrometer (1H:
400 MHz and 13C: 100 MHz) in chloroform-d (CDCl3) solution and the
chemical shifts are reported in parts per million (ppm) using the re-
sidual proton in the NMR solvent. 19F NMR (376 MHz) spectra were
measured with a Bruker AVANCE III 400 NMR spectrometer in CDCl3
solution with CFCl3 (δF = 0 ppm) as an internal standard. The Bruker
AVANCE III 400 NMR spectrometer was used for determining the yield
of the products with trifluoromethylbenzene (CF3C6H5) or hexa-
fluorobenzene (C6F6). High-resolution mass spectra (HRMS) were taken
on a JEOL JMS-700MS spectrometer with fast atom bombardment
(FAB) methods.
4.1.1. Materials
All reactions were carried out using dried glassware with a magnetic
stirrer bar and routinely monitored by 19F NMR spectroscopy or thin-
layer chromatography (TLC). All chemicals were of reagent grade and,
if necessary, were purified in the usual manner prior to use. Column
chromatography was carried out on silica gel (Wako gel® 60N,
38–100 μm) and TLC analysis was performed on silica gel TLC plates
(Merck, Silica gel 60F254).
Table 6
Structural data for Int-1 and TS-1 for 1dA and 1eA calculated by DFT calculation.
For difluorinated 1dA
For monofluorinated 1eA
Int-1C1
Int-1C2
TS-1C
Int-1D1
Int-1D2
Int-1D3
TS-1D
∠C1eOeCo [°]
ϕ (C2eC1eOeCo) [°]
Co⋯O [pm]
151.1
9.9
217
139.5
9.6
214
–
–
–
147.8
8.7
218
136.6
1.1
213
135.1
9.4
212
–
–
–
Co⋯C1 [pm]
–
–
204
110.1
115.5
121.3
–
–
–
206
113.4
115.2
120.7
∠C2eC1eO [°]
∠C2eC1eC3 [°]
∠C3eC1eO [°]
119.8
118.2
121.9
116.9
121.3
121.7
121.3
117.9
120.8
116.7
123.4
119.9
116.8
123.1
120.0
15

T. Kumon et al.
JournalofFluorineChemistry206(2018)8–18
4.2. Typical procedure for the cobalt-catalyzed hydrosilylation of
fluorinated ketones
4.2.6. Triethyl[2,2,2-trifluoro-1-(4-chlorophenyl)ethoxy]silane (2aG)
Yield: 89% (colorless oil); IR (neat): ν 2959, 2915, 2880, 1494,
1280, 1242, 1172, 1135, 1092, 860 cm−1; HRMS (FAB): calcd for [M
In a 30 mL two-necked round bottomed-flask, equipped with a
magnetic stirring bar, reflux condense and an inlet tube for argon, were
+H]+ C14H21ClF3OSi: 325.1002, Found: 325.0992; 1H NMR:
δ
0.55–0.65 (m, 6H), 0.91 (t, J = 7.9 Hz, 9H), 4.91 (q, J = 6.4 Hz, 1H),
7.36 (d, J = 8.7 Hz, 2H), 7.40 (d, J = 8.7 Hz, 2H); 13C NMR: δ 4.5, 6.4,
72.8 (q, J = 32.8 Hz), 124.0 (q, J = 282.4 Hz), 128.6, 128.9, 134.2,
135.1; 19F NMR: δ −79.09 (d, J = 6.4 Hz, 3F).
placed
2,2,2-trifluoro-4′-methylacetophenone
(1aA,
0.092 g,
0.49 mmol) and triethylsilane (0.10 mL, 0.63 mmol) and Co2(CO)8
(1.7 mg, 5.0 μmol) in toluene (2.5 mL), and the whole was stirred at
reflux temperature. After 3 h, the reaction mixture was subjected to a
flash column chromatography using a silica gel as a stationary phase
and AcOEt as a mobile phase. After removal of solvent from the eluent
using a rotary evaporator under reduced pressure, the residue obtained
was purified by silica gel column chromatography (hexane/
AcOEt = 50/1) to give the corresponding triethyl[2,2,2-trifluoro-1-(4-
methylphenyl)ethoxy]silane 2aA in 84% yield (0.13 g, 0.41 mmol) as a
colorless oil.
4.2.7. Triethyl[2,2,2-trifluoro-1-(1-naphthyl)ethoxy]silane (2aH)
Yield: 95% (colorless oil); IR (neat): ν 2958, 2914, 2879, 1272,
1235, 1173, 1131, 1007, 852, cm−1; HRMS (FAB): calcd for [M]+
C18H23F3OSi: 340.1470, Found: 340.1465; 1H NMR: δ 0.55–0.69 (m,
6H), 0.90 (t, J = 7.9 Hz, 9H), 5.82 (q, J = 5.9 Hz, 1H), 7.55–7.60 (m,
3H), 7.85–7.93 (m, 3H), 8.13 (d, J = 8.0 Hz, 1H); 13C NMR: δ 4.6, 6.6,
69.50–70.80 (m, Ar-CH(OH)CF2), 123.1, 124.9 (q, J = 283.3 Hz),
125.3, 125.8, 126.6, 127.0, 129.1, 129.9, 131.2, 131.7, 133.8; 19F
NMR: δ −78.36 (brs, 3F).
4.2.1. Triethyl[2,2,2-trifluoro-1-(4-methylphenyl)ethoxy]silane (2aA)
Yield: 84% (colorless oil); IR (neat): ν 2958, 2915, 2880, 1273,
4.2.8. Triethyl[2,2,3,3,4,4,4-heptafluoro-1-(4-methylphenyl)butoxy]silane
(2bA)
1205, 1170, 1132, 1108, 1007, 842 cm−1; HRMS (FAB): calcd for [M+
]
C
15H23F3OSi: 304.1470, Found: 304.1466; 1H NMR: δ 0.53–0.70 (m,
Yield: Quant. (colorless oil); IR (neat): ν 2959, 2916, 2881, 1460,
1222, 1195, 1133, 1007, 970 cm−1; HRMS (FAB): calcd for [M]+
C17H23F7OSi: 404.1406, Found: 404.1414; 1H NMR: δ 0.48–0.57 (m,
6H), 0.87 (t, J = 7.9 Hz, 9H), 2.37 (s, 3H), 5.09 (dd, J = 17.2, 5.2 Hz,
1H), 7.18 (d, J = 7.9 Hz, 2H), 7.32 (d, J = 7.9 Hz, 2H); 13C NMR: δ 4.7,
6.5, 21.4, 73.2 (dd, J = 30.5, 22.1 Hz), 109.9 (tq, J = 268.7, 37.5 Hz),
114.7 (dddd, J = 263.8, 251.5, 31.8, 25.6 Hz), 119.6 (tt, J = 287.8,
33.4 Hz), 128.5, 129.0, 132.3, 139.3; 19F NMR: δ −126.96 (dm,
J = 289.9 Hz, 1F), −126.75 (dm, J = 281.5 Hz, 1F), −124.03 (dm,
J = 289.8 Hz, 1F), −118.04 (dm, J = 281.5 Hz, 1F), −81.40 (dd,
J = 12.4, 9.1 Hz, 3F).
6H), 0.94 (t, J = 7.9 Hz, 9H), 2.39 (s, 3H), 4.93 (q, J = 6.5 Hz, 1H),
7.20 (d, J = 7.9 Hz, 2H), 7.37 (d, J = 7.9 Hz, 2H); 13C NMR: δ 4.7, 6.6,
21.4, 73.4 (q, J = 32.0 Hz), 124.5 (q, J = 282.3 Hz), 127.6, 129.1,
132.9, 139.1; 19F NMR: δ −79.60 (d, J = 6.5 Hz, 3F).
4.2.2. Triethyl(2,2,2-trifluoro-1-phenylethoxy)silane (2aB)
Yield: 83% (colorless oil); IR (neat): ν 2959, 2915, 2880, 1457,
1272, 1206, 1170, 1133, 1006, 854 cm−1; HRMS (FAB): calcd for [M+
]
C
14H21F3OSi: 290.1314, Found: 290.1306; 1H NMR: δ 0.52–0.67 (m,
6H), 0.90 (t, J = 7.9 Hz, 9H), 4.93 (q, J = 6.6 Hz, 1H), 7.35–7.40 (m,
3H), 7.43–7.49 (m, 2H); 13C NMR: δ 4.6, 6.5, 73.6 (q, J = 32.0 Hz),
124.4 (q, J = 282.4 Hz), 127.8, 128.4, 129.2, 135.9; 19F NMR: δ
−78.97 (d, J = 6.6 Hz, 3F).
4.2.9. Triethyl[2,2-difluoro-1-(4-methylphenyl)ethoxy]silane (2dA)
Yield: 97% (colorless oil); IR (neat): ν 2957, 2914, 2879, 1459,
1414, 1241, 1118, 1102, 1007, 842 cm−1; HRMS (FAB): calcd for [M
+Na]+ C15H24F2NaOSi: 309.1462, Found:309.1452; 1H NMR:
δ
4.2.3. Triethyl[2,2,2-trifluoro-1-(4-methoxyphenyl)ethoxy]silane (2aC)
Yield: 95% (colorless oil); IR (neat): ν 2958, 2914, 2880, 1614,
1516, 1275, 1207, 1170, 1131, 1036, 843 cm−1; HRMS (FAB): calcd for
[M+] C15H23F3O2Si: 320.1419, Found: 320.1411; 1H NMR: δ 0.52–0.66
(m, 6H), 0.91 (t, J = 7.9 Hz, 9H), 3.82 (s, 3H), 4.88 (q, J = 6.6 Hz,
1H), 6.90 (d, J = 8.6 Hz, 2H), 7.38 (d, J = 8.6 Hz, 2H); 13C NMR: δ 4.6,
6.6, 55.3, 73.2 (q, J = 32.1 Hz), 113.8, 124.5 (q, J = 282.3 Hz), 127.9,
129.0, 160.3; 19F NMR: δ −79.25 (d, J = 6.6 Hz, 3F).
0.52–0.66 (m, 6H), 0.91 (t, J = 7.9 Hz, 9H), 2.36 (s, 3H), 4.67–4.74 (m,
1H), 5.61 (ddd, J = 56.9, 55.8, 5.0 Hz, 1H), 7.18 (d, J = 7.9 Hz, 2H),
7.29 (d, J = 7.9 Hz); 13C NMR: δ 4.8, 6.7, 21.3, 74.4 (dd, J = 26.7,
23.5, Hz), 116.3 (dd, J = 248.1, 244.4, Hz), 127.2, 129.2, 134.7 (d,
J = 5.2 Hz), 138.4; 19F NMR: δ −127.66 (ddd, J = 277.9, 56.9,
10.9 Hz, 1F), −125.59 (ddd, J = 277.9, 55.8, 8.3 Hz, 1F).
4.3. Computational details
4.2.4. Triethyl[2,2,2-trifluoro-1-(3-methoxyphenyl)ethoxy]silane (2aD)
Yield: 94% (colorless oil); IR (neat): ν 2958, 2915, 2880, 1604,
1493, 1262, 1242, 1172, 1133, 855, cm−1; HRMS (FAB): calcd for
[M+] C15H23F3O2Si: 320.1419, Found: 320.1423; 1H NMR: δ 0.55–0.68
(m, 6H), 0.91 (t, J = 7.9 Hz, 9H), 3.82 (s, 3H), 4.90 (q, J = 6.4 Hz,
1H), 6.88–6.96 (m, 1H), 7.00–7.12 (m, 2H), 7.28 (t, J = 8.1 Hz, 1H);
13C NMR: δ 4.6, 6.6, 55.3, 73.4 (q, J = 32.0 Hz), 113.2, 114.8, 120.1,
124.4 (q, J = 282.5 Hz), 129.4, 137.4, 159.7; 19F NMR: δ −78.84 (d,
J = 6.4 Hz, 3F).
All calculations were carried out using the Gaussian 09 program
(Rev. C.01) and density functional theory (DFT) method. We used the
standard basis set, 6-31+G(d,p) levels of theory for the geometry op-
timizations. All stationary points on the potential energy surfaces were
ascertained to be the minima or transition states.
References
[1] (a) For representative books and reviews on transition-metal-catalyzed hydro-
silylation, see: Y. Nakajima, S. Shimada, Hydrosilylation reaction of olefins: recent
advances and perspectives, RSC Adv. 5 (2015) 20603–20616;
(b) B. Marciniec (Ed.), Hydrosilylation, A Comprehensive Review on Recent
Advances, Springer, Netherlands, 2009;
4.2.5. Triethyl[2,2,2-trifluoro-1-(2-methoxyphenyl)ethoxy]silane (2aE)
Yield: 80% (colorless oil); IR (neat): ν 2958, 2915, 2880, 1494,
1272, 1248, 1173, 1134, 1030, 849 cm−1; HRMS (FAB): calcd for [M]+
C15H23F3O2Si: 320.1419, Found: 320.1409; 1H NMR: δ 0.50–0.65 (m,
6H), 0.88 (t, J = 7.9 Hz, 9H), 3.84 (s, 3H), 5.56 (q, J = 6.5 Hz, 1H),
6.88 (dd, J = 8.4, 0.7 Hz, 1H), 7.00 (td, J = 7.5, 0.7 Hz, 1H), 7.32
(ddd, J = 8.4, 7.5, 1.7 Hz, 1H), 7.60 (br s, J = 7.5 Hz, 1H); 13C NMR: δ
4.6, 6.5, 55.8, 66.0 (q, J = 32.8 Hz), 110.6, 120.8, 124.6, 124.8 (q,
J = 282.5 Hz), 129.0, 130.1, 156.9; 19F NMR: δ −79.08 (d, J = 6.5 Hz,
3F).
(c) S. Díez-González, S.P. Nolan, Transition metal-catalyzed hydrosilylation of
carbonyl compounds and imines. A review, Org. Prep. Proc. Int. (2007), pp.
523–559;
(d) A.K. Roy, A Review of Recent Progress in Catalyzed Homogeneous
Hydrosilation (Hydrosilylation), Adv. Organomet. Chem. (2008), pp. 1–59.
[2] (a) A.J. Holwell, For platinum- or palladium-catalyzed hydrosilylation, see: Platin.
Met. Rev. 52 (2008) 243–246;
(b) M. Igarashi, T. Fuchikami, Transition-metal complex-catalyzed reduction of
amides with hydrosilanes: a facile transformation of amides to amines, Tetrahedron
Lett. 42 (2001) 1945–1947.
16

T. Kumon et al.
JournalofFluorineChemistry206(2018)8–18
[3] (a) K. Riener, M.P. Högerl, P. Gigler, F.E. Kühn, Rhodium-catalyzed hydrosilylation
of ketones: catalyst development and mechanistic insights, For rhodium-catalyzed
hydrosilylation, see: ACS Catal. 2 (2012) 613–621;
cycloaddition with 2,2, 3-trifluoroacrylate, Eur. J. Org. Chem. (2016) 4561–4568;
(d) S. Yamada, K. Hondo, T. Konno, T. Ishihara, Concise preparation of fluorine-
containing carbocycles involving Diels-Alder reaction using fluorinated alkene or
diene derivatives, RSC Adv. 6 (2016) 28458–28469;
(e) T. Nihei, Y. Nishi, N. Ikeda, S. Yokotani, T. Ishihara, S. Arimitsu, T. Konno,
Highly stereoselective synthesis of fluoroalkene dipeptides via the novel Cr(II)-
mediated carbon-fluorine bond cleavage/new carbon–carbon bond formation,
Synthesis 48 (2016) 865–881 and references cited therein.
(b) M. Sawamura, R. Kuwano, Y. Ito, trans-Chelating chiral diphosphane ligands
bearing flexible P-Alkyl substituents (AlkylTRAPs) and their application to the
rhodium-catalyzed asymmetric hydrosilylation of simple ketones, Angew. Chem.
Int. Ed. 33 (1994) 111–113;
(c) H. Nishiyama, H. Sakaguchi, T. Nakamura, M. Horihata, M. Kondo, K. Itoh,
Chiral and C2-symmetrical bis(oxazolinylpyridine)rhodium(III) complexes: effec-
tive catalysts for asymmetric hydrosilylation of ketones, Organometallics 8 (1989)
846–848.
[11] (a) T. Konno, Trifluoromethylated internal alkynes: versatile building blocks for
the preparation of various fluorine-containing molecules, Synlett (2014)
1350–1370;
[4] (a) G. Zhu, M. Terry, X. Zhang, Asymmetric hydrosilylation of ketones catalyzed by
ruthenium complexes with chiral tridentate ligands, For ruthenium-catalyzed hy-
drosilylation, see: J. Organomet. Chem. 547 (1997) 97–101;
(b) T. Nihei, Y. Kubo, T. Ishihara, T. Konno, Highly regioselective SN2′ reaction of
β-fluoroallylic phosphates with organocopper reagents, and its application to the
synthesis of fluorine-containing carbocycles, J. Fluor. Chem. 167 (2014) 110–121;
(c) T. Konno, M. Kishi, T. Ishihara, S. Yamada, Stereoselective synthesis of CF3-
substituted trans- and cis-enediynes via carbocupration or hydrostannation reac-
tions of CF3-containing diyne, Tetrahedron 70 (2014) 2455–2463;
(d) T. Konno, R. Kinugawa, T. Ishihara, S. Yamada, A remarkable regiocontrol in
the palladium-catalyzed silylstannylation of fluoroalkylated alkynes – highly regio-
and stereoselective synthesis of multi-substituted fluorine-containing alkenes, Org.
Biomol. Chem. 12 (2014) 1611–1617;
(e) A. Moriagki, M. Kawamura, S. Arimitsu, T. Ishihara, T. Konno, Application to
tertiary amines synthesized by catalytic dehydrogenation of enamies as nucleo-
philic C2 synthons for 1,4-conjugate addition with fluoroalkylated olefins, Asian J.
Org. Chem. 2 (2013) 239–243;
(f) A. Morigaki, K. Tsukade, S. Arimitsu, T. Konno, T. Kubota, Unusual annulation
reaction of electron-deficient alkenes with enamines: an easy access to stereo-
controlled 4-fluoroalkylated 3,4-dihydro-2H-pyrans, Tetrahedron 69 (2013)
1521–1525;
(g) A. Morigaki, T. Tanaka, T. Miyabe, T. Ishihara, T. Konno, Rhodium(I)-catalyzed
1,4-conjugate addition toward β-fluoroalkylated electron-deficient alkenes: a new
entry to a construction of a tertiary carbon center possessing a fluoroalkyl group,
Org. Biomol. Chem. 11 (2013) 586–595;
(h) T. Konno, T. Kida, A. Tani, T. Ishihara, A novel synthesis of fluorine-containing
cyclopentenones via Pauson-Khand reaction, J. Fuor. Chem. 144 (2012) 147–156;
(i) T. Konno, K. Moriyasu, R. Kinugawa, T. Ishihara, Rhodium-catalyzed [2+2+2]
cycloaddition of various fluorine-containing alkynes – novel synthesis of multi-
substituted fluoroalkylated aromatic compounds, Org. Biomol. Chem. 8 (2010)
1718–1724 and references cited therein.
(b) Y. Nishibayashi, I. Takei, S. Uemura, M. Hidai, Ruthenium-catalyzed asym-
metric hydrosilylation of ketones and imine, Organometallics 17 (1996)
3420–3422.
[5] (a) R. Malacea, R. Poli, E. Manoury, Asymmetric hydrosilylation, transfer hydro-
genation and hydrogenation of ketones catalyzed by iridium complexes, For ir-
idium-catalyzed hydrosilylation, see: Coord. Chem. Rev. 254 (2010) 729–752;
(b) A.R. Chianese, R.H. Crabtree, Axially chiral bidentate n-heterocyclic carbene
ligands derived from BINAM: rhodium and iridium complexes in asymmetric ketone
hydrosilylation, Organometallics 24 (2005) 4432–4436.
[6] (a) M.D. Greenhalgh, A.S. Jones, S.P. Thomas, Iron-catalyzed hydrofunctionalisa-
tion of alkenes and alkynes, For a selected review of Fe-catalyzed hydrosilylation,
see, ChemCatChem 7 (2015) 190–222;
(b) M. Zhang, A. Zhang, Iron-catalyzed hydrosilylation reactions, Appl.
Organomet. Chem. 24 (2010) 751–757;
(c) R.H. Morris, Asymmetric hydrogenation, transfer hydrogenation and hydro-
silylation of ketones catalyzed by iron complexes, Chem. Soc. Rev. 38 (2009)
2282–2291.
[7] (a) J. Sun, L. Deng, Cobalt complex-catalyzed hydrosilylation of alkenes and al-
kynes, For a selected review of co-catalyzed hydrosilylation, see: ACS Catal. 6
(2016) 290–300;
(b) X. Du, Z. Huang, Advances in base-metal-catalyzed alkene hydrosilylation, ACS
Catal. 7 (2017) 1227–1243;
(c) Q. Niu, H. Sun, X. Li, H.-F. Klein, U. Flörke, Synthesis and catalytic application
in hydrosilylation of the complex mer-hydrido(2-mercaptobenzoyl)tris(trimethyl-
phosphine)cobalt(III), Organometallics 32 (2013) 5235–5238.
[8] (a) S.-B. Qi, M. Li, S. Li, J.-N. Zhou, J.-W. Wu, F. Yu, X.-C. Zhang, A.S.C. Chan,
J. Wu, Copper-dipyridylphosphine-catalyzed hydrosilylation: enantioselective
synthesis of aryl- and heteroaryl cycloalkyl alcohols, For selected examples of
copper-catalyzed hydrosilylation, see: Org. Biomol. Chem. 11 (2013) 929–937;
(b) Y.-Z. Sui, X.-C. Zhang, J.-W. Wu, S. Li, J.-N. Zhou, M. Li, W. Fang, A.S.C. Chan,
J. Wu, CuII-Catalyzed asymmetric hydrosilylation of diaryl- and aryl heteroaryl
ketones: application in the enantioselective synthesis of orphenadrine and neobe-
nodine, Chem. Eur. J. 18 (2012) 7486–7492;
[12] (a) T. Konno, K. Taku, S. Yamada, K. Moriyasu, T. Ishihara, Org. Biomol. Chm. 7
(2009) 1167–1170;
(b) N. Ikeda, T. Konno, Highly regio- and stereoselective hydrosilylation of β-
fluoroalkylated α,β-unsaturated ketones, J. Fluor. Chem. 173 (2015) 69–76.
[13] (a) Recent reports on hydrosilylation of α,α,α-trifluoroacetophenone using tris
(pentafluorophenyl)borane as a catalyst, see: N.D. Rio, M. Lopez-Reyes,
A. Baceiredo, N. Saffon-Merceron, D. Lutters, T. Müller, T. Kato, Angew. Chem. Int.
Ed. 56 (2017) 1365–1370;
(c) K. Junge, B. Wendt, D. Addis, S. Zhou, S. Das, M. Beller, Copper-catalyzed
enantioselective hydrosilylation of ketones by using monodentate binaphthopho-
sphepine ligands, Chem. Eur. J. 16 (2010) 68–73;
(b) I. Mallov, D.W. Stephan, Dalton Trans. 45 (2016) 5568–5574.
[14] (a) F.E. Herkes, D.J. Burton, Fluoro olefins. I. Synthesis of β-substituted perfluoro
olefins, J. Org. Chem. 32 (1967) 1311–1318;
(d) X.-C. Zhang, Y. Wu, F.-F. Yu, F. Wu, J. Wu, A.S.C. Chan, Application of copper
(II)-dipyridylphosphine catalyst in the asymmetric hydrosilylation of simple ke-
tones in air, Chem. Eur. J. 15 (2009) 5888–5891;
(e) S. Diez-Gonzalez, S.P. Nolan, Copper, silver, and gold complexes in hydro-
silylation reactions, Acc. Chem. Res. 41 (2008) 349–358;
(f) B.H. Lipshutz, B.A. Frieman, A.E. Tomaso Jr., Copper-in-charcoal (Cu/C): het-
erogeneous, copper-catalyzed asymmetric hydrosilylations, Angew. Chem. Int. Ed.
45 (2006) 1259–1264.
(b) A. Kaluszyner, Cyclic analogs of DDT-like compounds, J. Org. Chem. 25 (1960)
473–474.
[15] (a) The electron-withdrawing nature of a substituent can be quantitatively eval-
uated using the Hammett substituent constant (σ), and the σp values for CO2Et and
CF3 are reported as +0.46 and +0.54, demonstrating the similar electron-with-
drawing characteristics of these substituent, see: C. Hansch, A. Leo, R.W. Taft, A
survey of Hammett substituent constants and resonance and field parameters,
Chem. Rev. 91 (1991) 165–195;
[9] (a) S. Yamada, M. Morita, T. Konno, Multi-color photoluminescence induced by
electron-density distribution of fluorinated bistolane derivatives, J. Fluor. Chem.
202 (2017) 54–64;
(b) L.P. Hammett, The effect of structure upon the reactions of organic compounds.
Benzene derivatives, J. Am. Chem. Soc. 59 (1937) 96–103.
[16] A.J. Chalk, J.F. Harrod, Homogeneous catalysis. IV. Some reactions of silicon hy-
drides in the presence of cobalt carbonyls, J. Am. Chem. Soc. 89 (1967) 1640–1647.
[17] (a) For related reports on the mechanisms of transition-metal-catalyzed hydro-
silylation, see: For Mn catalyst: T.K. Mukhopadhyay, C.L. Rock, M. Hong,
D.C. Ashley, T.L. Groy, M.-H. Baik, R.J. Trovitch, Mechanistic investigation of bis
(imino)pyridine manganese catalyzed carbonyl and carboxylate hydrosilylation, J.
Am. Chem. Soc. 139 (2017) 4901–4915;
(b) S. Yamada, S. Yamaguchi, O. Tsutsumi, Electron-density distribution tuning for
enhanced thermal stability of luminescent gold complexes, J. Mater. Chem. C 5
(2017) 7977–7984;
(c) S. Yamada, K. Miyano, T. Konno, T. Agou, T. Kubota, T. Hosokai, Fluorine-
containing bistolanes as light-emitting liquid crystalline molecules, Org. Biomol.
Chem. 15 (2017) 5949–5958;
(d) S. Yamada, S. Hashishita, H. Konishi, Y. Nishi, T. Kubota, T. Asai, T. Ishihara,
T. Konno, Unprecedented 3,6-disubstituted 1,1,2,2-tetrafluorocyclohexane deriva-
tives, J. Fluor. Chem. 200 (2017) 47–58;
(e) S. Yamada, K. Kinoshita, S. Iwama, T. Yamazaki, T. Kubota, T. Yajima,
K. Yamamoto, S. Tahara, Synthesis of perfluoroalkylated pentacenes and evaluation
of their fundamental physical properties, Org. Biomol. Chem. 15 (2017)
2522–2535;
(b) For Ru catalyst: M.C. Lipke, T.D. Tilley, Hypercoordinate ketone adducts of
electrophilic η3-H2SiRR′ Ligands on ruthenium as key intermediates for efficient
and robust catalytic hydrosilation, J. Am. Chem. Soc 136 (2014) 16387–16398;
(c) D.V. Gutsulyak, S.F. Vyboishchikov, G.I. Nikonov, Cationic silane σ-complexes
of ruthenium with relevance to catalysis, J. Am. Chem. Soc. 132 (2010) 5950–5951;
(d) For Rh catalyst: I. Ojima, M. Nihonyanagi, T. Kogure, M. Kumagai, S. Horiuchi,
K. Nakatsugawa, Y. Nagai, Reduction of carbonyl compounds via hydrosilylation: I.
Hydrosilylation of carbonyl compounds catalyzed by tris(triphenylphosphine)
chlororhodium, J. Organomet. Chem. 94 (1975) 449–461;
(e) J.A. Osborn, F.H. Jardine, J.F. Young, G. Wilkinson, The preparation and
properties of tris(triphenylphosphine)halogenorhodium(I) and some reactions
thereof including catalytic homogenous hydrogenation of olefins and acetylenes
and their derivatives, J. Chem. Soc. A (1966) 1711–1732.
(f) S. Yamada, S. Hashishita, T. Asai, T. Ishihara, T. Konno, Design, synthesis, and
evaluation of new fluorinated liquid crystals bearing CF2CF2 fragment with nega-
tive electronic anisotropy, Org. Biomol. Chem. 15 (2017) 1495–1509.
[10] (a) Y. Sakaguchi, S. Yamada, T. Konno, T. Agou, T. Kubota, Stereochemically de-
fined various multi-substituted alkenes bearing a tetrafluoroethylene (–CF2CF2–)
fragment, J. Org. Chem. 82 (2017) 1618–1631;
(b) K. Tamamoto, S. Yamada, M. Higashi, T. Konno, T. Ishihara, Radical reactivity
of α,β,β-trifluorocrylic ester: facile approach to γ-silyloxy-α,β,β-trifluorobutyric
acid derivatives involving regioselective nucleophilic radical addition reaction, J.
Fluorine Chem. 190 (2016) 23–30;
(c) S. Yamada, M. Higashi, T. Konno, T. Ishihara, Straightforward and highly ste-
reoselective synthesis of 3,3,4-trifluoropyrrolidines involving 1, 3-dipolar
[18] M.J. Frish, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheeseman,
G. Scalmani, V. Barone, B. Mennucci, G.A. Peterson, H. Nakatsuji, M. Caricato, X. Li,
H.P. Hratchian, A.F. Izmaylov, J. Bloino, G. Zheng, J.L. Sonnenberg, M. Hada,
M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda,
O. Kitao, H. Nakai, T. Vreven, J.A. Montgomery Jr., J.E. Peralta, F. Ogliaro,
M. Bearpark, J.J. Heyd, E. Brothers, K.N. Kudin, V.N. Staroverov, T. Keith,
17

T. Kumon et al.
JournalofFluorineChemistry206(2018)8–18
(c) K. Fukui, The path of chemical reactions – the IRC approach, Acc. Chem. Res.
14 (1981) 363–368.
R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J.C. Burant, S.S. Iyengar,
J. Tomasi, M. Cossi, N. Rega, J.M. Millam, M. Klene, J.E. Knox, J.B. Cross,
V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R.E. Stratmann, O. Yazyev,
A.J. Austin, R. Cammi, C. Pomelli, J.W. Ochterski, R.L. Martin, K. Morokuma,
V.G. Zakrzewski, G.A. Voth, P. Salvador, J.J. Dannenberg, S. Dapprich,
A.D. Daniels, O. Farkas, J.B. Foresman, J.V. Ortiz, J. Cioslowski, D.J. Fox, Gaussian
09W, Revision C.01, Gaussian, Inc., Wallingford CT, 2013.
[20] J.A. MacPhee, A. Panaye, J.-E. Dubois, Steric effects – I: A critical examination of
the taft steric parameter – Es. Definition of a revised, broader and homogeneous
scale. Extention to highly congested alkyl groups, Tetrahedron 34 (1978)
3553–3562. The steric bulkiness of the substituents around the carbonyl moiety are
in the order Ph (Es′ = –2.31) > CF3 (Es′ = –0.78) > CH3 (Es′ = 0) A larger ne-
gative Es′ value indicates a larger steric effect, see:.
[19] (a) H.P. Hratchian, H.B. Schlegel, Accurate reaction paths using a Hessian based
predictor-corrector integrator, J. Chem. Phys. 120 (2004) 9918–9924;
(b) C. Gonzalez, H.B. Schlegel, An improved algorithm for reaction path following,
J. Chem. Phys. 90 (1989) 2154–2161;
[21] The calculation results of Int-2C and Int-2D were described in the supporting in-
formation.
18