Si-CN bond cleavage of silyl cyanides by an iron catalyst. A new route of silyl cyanide formation

Article abstract of DOI:10.1246/bcsj.20130206

The reaction of silyl cyanide (R₃SiCN) with hydrosilane (R'₃SiH) in the presence of a catalytic amount of [(η⁵-C₅H₅)Fe(CO)₂Me] formed R'₃SiCN and R₃SiH. This reaction involves Si?CN bond cleavage and provides a new method for the preparation of silyl cyanides. DFT calculation showed that coordinatively unsaturated [(η⁵-C₅H₅)Fe(CO)(SiMe₃)] reacts exothermically with Me₃SiCN to give the CN π-coordinated complex, [(η⁵-C₅H₅)Fe(CO)(η²-Me₃SiCN)(SiMe₃)], followed by exothermic silyl migration from the iron to the nitrogen atom of the η²-coordinated Me₃SiCN with the activation energy of 8.8 kcal mol^-1 to give [(η⁵-C₅H₅)Fe(CO)(Me₃SiC=NSiMe₃-κC,κN)]. The complex is one of the intermediates in the catalytic cycle. The related complex [(η⁵-C₅Me₅)Fe(CO)(κ-C,N-t-BuMe₂SiC=NSiPh₃)] was isolated in the reaction of [(η⁵-C₅Me₅)Fe(CO)(py)(SiPh₃)] with t-BuMe₂SiCN and characterized by ¹H, ¹³C, and ²⁹Si NMR spectroscopy. The activation energy of the Si?CN bond cleavage in [(η⁵-C₅H₅)Fe(CO)(κC,N-Me₃SiC=NSiMe₃)] was evaluated by DFT calculation to be 33.7 kcal mol^-1 which is comparable to that in the reaction of acetonitrile (32.9 kcal mol^-1).

Full text of DOI:10.1246/bcsj.20130206

© 2013 The Chemical Society of Japan
Bull. Chem. Soc. Jpn. Vol. 87, No. 1, 59-68 (2014)
59
BCSJ Award Article
Si-CN Bond Cleavage of Silyl Cyanides by an Iron Catalyst.
A New Route of Silyl Cyanide Formation
Andrea Renzetti,1 Nobuaki Koga,2 and Hiroshi Nakazawa*1
1Department of Chemistry, Graduate School of Science, Osaka City University, Sumiyoshi-ku, Osaka 558-8585
2Graduate School of Information Science, Nagoya University, Nagoya, Aichi 464-8601
Received July 24, 2013; E-mail: nakazawa@sci.osaka-cu.ac.jp
The reaction of silyl cyanide (R3SiCN) with hydrosilane (R¤3SiH) in the presence of a catalytic amount of [(η5-
C5H5)Fe(CO)2Me] formed R¤3SiCN and R3SiH. This reaction involves Si-CN bond cleavage and provides a new method
for the preparation of silyl cyanides. DFT calculation showed that coordinatively unsaturated [(η5-C5H5)Fe(CO)(SiMe3)]
reacts exothermically with Me3SiCN to give the CN π-coordinated complex, [(η5-C5H5)Fe(CO)(η2-Me3SiCN)(SiMe3)],
followed by exothermic silyl migration from the iron to the nitrogen atom of the η2-coordinated Me3SiCN with the
activation energy of 8.8 kcal mol¹1 to give [(η5-C5H5)Fe(CO)(Me3SiC=NSiMe3-κC,κN)]. The complex is one of the inter-
mediates in the catalytic cycle. The related complex [(η5-C5Me5)Fe(CO)(κ-C,N-t-BuMe2SiC=NSiPh3)] was isolated
in the reaction of [(η5-C5Me5)Fe(CO)(py)(SiPh3)] with t-BuMe2SiCN and characterized by 1H, 13C, and 29Si NMR
spectroscopy. The activation energy of the Si-CN bond cleavage in [(η5-C5H5)Fe(CO)(κ-C,N-Me3SiC=NSiMe3)] was
¹1
evaluated by DFT calculation to be 33.7 kcal mol which is comparable to that in the reaction of acetonitrile (32.9
kcal mol¹1).
Most chemical reactions consist of bond cleavage and for-
mation. Finding suitable reaction conditions that allow break-
ing of a particular bond in a reagent and forming a new bond
selectively is a challenging task. Bond cleavage is an impor-
tant topic in organic chemistry as much as bond formation.1
Selective cleavage of weak bonds is relatively easy, whereas
that of strong bonds is difficult and usually requires the use
of a transition-metal catalyst to reduce the activation energy
of bond breaking. In this context, we have reported a C-CN
bond cleavage reaction of organonitriles catalyzed by an iron
complex.2 The reaction is noteworthy as the C-CN bond is
one of the strongest C-C single bonds: the bond energy of
C-C bond in acetonitrile is 133 kcal mol¹1, whereas that of
C-C bond in alkanes is ca. 83 kcal mol¹1. The key steps of
the catalytic cycle are shown in Scheme 1. The cyano group
coordinates in η1-fashion to a 16e iron complex bearing a silyl
group. Then, the organonitrile changes its coordination mode
from η1 to η2. The silyl group migrates from iron to nitrile
nitrogen to form an N-silylimino complex. Finally, C-R bond
cleavage affords the iron complex bearing silyl isocyanide and
R ligands. N-CN bond cleavage of cyanamides (R2N-CN)3
and O-CN bond cleavage of cyanates (RO-CN)3b,4 have been
reported using a molybdenum catalyst, and a catalytic cycle
similar to that for C-CN bond cleavage has been proposed.
In all cases above, the silyl group plays a key role because
its migration triggers the cleavage of strong C-CN, N-CN,
and O-CN bonds. We reasoned that this type of reaction,
called Silyl-Migration-Induced reaction (SiMI reaction), has
general validity and can be applied to the cleavage of E-CN
bonds where E is other than alkyl, aryl, amino, and alkoxy
groups. Based on this hypothesis, we investigated the cleav-
age of strong Si-CN bonds in silyl cyanides (BDESi-CN = 85
SiR'3
SiR'3
N
SiR'3
[Fe]
SiR'3
[Fe]
SiR'3
[Fe]
N
C
N
C
C
N
C
R
[Fe]
[Fe]
R
RCN
R
R
[Fe] = [(η5-C5H5)Fe(CO)]
Scheme 1. Key step in the catalytic cycle of R-CN bond cleavage by an iron complex.
Published on the web October 19, 2013; doi:10.1246/bcsj.20130206

60
Bull. Chem. Soc. Jpn. Vol. 87, No. 1 (2014) BCSJ AWARD ARTICLE
kcal mol¹1).5 We examined the reaction shown in Scheme 2,
where silyl cyanide and hydrosilane bear different silyl groups,
causing the formation of a new Si-CN bond if Si-CN bond
cleavage occurs.
by ferrocene, which showed no activity under both conditions
probably due to its high stability (Entries 3 and 4). In the light
of this observation, complexes in Entries 15-22 were only used
under photoirradiation. [CpFe(CO)2Me] proved to be a better
catalyst than the analogous complexes of molybdenum and
tungsten (compare Entries 9 and 10 with 11-14). Replacement
of methyl group in [CpFe(CO)2Me] with other groups, as well
as the introduction of substituents into the cyclopentadienyl
ring, caused a marked decrease in catalytic activity (compare
Entries 15-18 and 19-22 with Entry 9, respectively). Overall,
[CpFe(CO)2Me] under photoirradiation was the best catalytic
system among those tested. The same trend was observed for
C-CN bond cleavage of organonitriles.2
Results and Discussion
A mixture of t-BuMe2SiCN (10 equiv), Et3SiH (100 equiv),
and [CpFe(CO)2Me] (1 equiv) (Cp: η5-C5H5) in deuterated
benzene was photoirradiated under nitrogen at room temper-
1
ature for 12 h. The H, 13C, and 29Si NMR spectra of the reac-
tion mixture showed the formation of Et3SiCN and t-BuMe2-
SiH in 44% yield based on the amount of t-BuMe2SiCN used
(Scheme 3). The remaining part of the reaction mixture was
unreacted starting materials, indicating that no side reaction
occurred. After the sample was exposed to air, (t-BuMe2Si)2O,
(Et3Si)2O, and Et3SiOH coming from the hydrolysis of t-
BuMe2SiCN and Et3SiCN were observed by GC-MS.6 Et3SiCN
could be isolated by distillation under reduced pressure,
exploiting the large difference in boiling point between silane
(108 °C for Et3SiH, 83 °C for t-BuMe2SiH) and Et3SiCN
(183 °C), and the fact that t-BuMe2SiCN is a solid. These
results show that the starting silyl cyanide and hydrosilane
underwent a recombination of their Si-CN and Si-H bonds,
affording a new silyl cyanide and a new silane. This reaction
is the first example of metathesis between Si-CN and Si-H
bonds. It is also one of a few examples of Si-CN bond-
breaking reactions promoted by organometallic complexes.7
Moreover, the reaction should be noted because it involves Si-
CN bond cleavage without Si-O bond formation, which is the
driving force in the majority of reactions of silyl cyanides.
In the reaction shown in Scheme 3, silyl cyanide reacts with
hydrosilane to give a different silyl cyanide and hydrosilane.
As the reaction involves Si-CN bond cleavage and Si-CN
bond formation, we reasoned that the reverse reaction may
occur, in other words, the reaction would reach an equilib-
rium point. In order to check whether the reverse reaction
takes place, Et3SiCN (10 equiv), t-BuMe2SiH (20 equiv), and
[CpFe(CO)2Me] (1 equiv) were treated under the same reaction
conditions. The formation of t-BuMe2SiCN (28% NMR yield
based on the amount of Et3SiCN) was observed, confirming
that the metathesis reaction is reversible.
Next, we varied the amount of catalyst and reactants
(Table 2). A progressive decrease in the catalyst amount from
0.1 to 0.01 equiv with respect to silyl cyanide caused a dramatic
lowering in reaction yield from 62 to 3% (Entries 1-4).
However, the amount of silane could be reduced from 100 to
Table 1. Reaction of Et3SiH and t-BuMe2SiCN in the
Presence of Various Transition-Metal Catalysts
Et3SiH
+
t-BuMe2SiCN
(100 equiv)
(10 equiv)
Metal catalyst (1 equiv)
C6D6, conditions, 12 h
Et3SiCN
t-BuMe2SiH
+
Entry Metal catalyst
Conditions Yield/%a)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
none
none
[Cp2Fe]
[Cp2Fe]
[Fe(CO)5]
[Fe(CO)5]
Fp2
Fp2
[CpFe(CO)2Me]
[CpFe(CO)2Me]
[CpMo(CO)3Me]
[CpMo(CO)3Me]
[CpW(CO)3Me]
[CpW(CO)3Me]
[CpFe(CO)2CH2Ph]
[CpFe(CO)2SiMe3]
[CpFe(CO)2Cl]
[CpFe(CO)2I]
[(C5H4Me)Fe(CO)2Me]
h¯, 25 °C
¦, 80 °C
h¯, 25 °C
¦, 80 °C
h¯, 25 °C
¦, 80 °C
h¯, 25 °C
¦, 80 °C
h¯, 25 °C
¦, 80 °C
h¯, 25 °C
¦, 80 °C
h¯, 25 °C
¦, 80 °C
h¯, 25 °C
h¯, 25 °C
h¯, 25 °C
h¯, 25 °C
h¯, 25 °C
0b)
0b)
0b)
0b) (0)c)
12
6 (15)c)
36
5 (4)c)
44
0b) (0)c),d)
22
4 (0)c),d)
24
0b) (0)c)
25
25
17
5
3
25
9
The catalytic activity was examined for several related iron
complexes and [CpM(CO)3Me] (M = Mo, W). Results are
shown in Table 1. No reaction took place in the absence of
metal complex (Entries 1 and 2), indicating that a catalyst is
necessary for the reaction. Complexes in Entries 3-14 were
tested under both photoirradiation at room temperature and
heating at 80 °C. In all cases, photoirradiation turned out to be
more effective than heating. The only exception was provided
[(C5H4SiMe3)Fe(CO)2Me] h¯, 25 °C
[(C5H4I)Fe(CO)2Me]
[(C5Me5)Fe(CO)2Me]
h¯, 25 °C
h¯, 25 °C
8
cat.
1
a) Yield calculated by H NMR using 50 equiv of toluene as
an internal standard. b) No reaction. c) Yields after 120 h are in
parentheses. d) Metal complex decomposed.
R'3SiCN + R3SiH
R3SiCN + R'3SiH
Scheme 2. Reaction investigated in this work.
[CpFe(CO)2Me] (1 equiv)
C6D6, hν, 25 °C, 12 h
t-BuMe2SiCN + Et3SiH
(10 equiv) (100 equiv)
Et3SiCN + t-BuMe2SiH
44% NMR yield
Scheme 3. Metathesis reaction of t-BuMe2SiCN and Et3SiH.

A. Renzetti et al.
Bull. Chem. Soc. Jpn. Vol. 87, No. 1 (2014)
61
90
80
70
60
50
40
30
20
10
0
Table 2. Photoreaction of t-BuMe2SiCN with Et3SiH in the
Presence of Various Amounts of Reactants and Catalyst
t-BuMe2SiCN + Et3SiH
[CpFe(CO)2Me]
t-BuMe2SiH
Et3SiCN
+
C6D6, hν, 25 °C
[M]:[CN]:[SiH]a)
Entry
Time/h
Yield/%b)
1
2
3c)
4c)
5
6
7
8
9
1:10:100
1:20:200
1:50:500
1:100:1000
1:10:10
1:10:20
1:10:50
1:10:500
1:20:10
78
78
78
78
24
24
24
24
24
62
18
9
0
10
20
30
40
50
60
70
80
Time/h
3
37
60
48
1
Figure 1. 1H NMR spectral monitoring of the reaction
between Et3SiH (20 equiv) and t-BuMe2SiCN (10 equiv)
in C6D6 at 25 °C in the presence of [CpFe(CO)2Me]
(1 equiv).
2
a) Molar ratio of [CpFe(CO)2Me] (M), t-BuMe2SiCN (CN),
and Et3SiH (SiH). b) Yield calculated by 1H NMR using
50 equiv of toluene as an internal standard. c) Reaction was
performed in neat.
yield in benzene was twice that of neat, indicating that a solvent
is necessary to assist the reaction (compare Entries 2 and 1,
respectively). Benzene, toluene, 1,4-dioxane, and diethyl ether
proved to be the best solvents among those tested, affording the
products in 60-69% yield (Entries 2, 3, 5, and 6). Their effect
can be attributed to the stabilization of reaction intermedi-
ates and/or transition states through coordination of π-bonds
(in aromatic solvents) and oxygen lone pairs (in oxygenated
solvents). Chlorinated solvents afforded a complex mixture of
compounds, presumably by the reaction of solvent with catalyst
and/or with reactants/products (Entries 8-10). Coordinative
solvents such as THF, pyridine, and nitrobenzene also turned
out not to be good solvents, probably because they coordinate to
catalyst reducing or killing its activity (Entries 7, 11, and 12).
In order to optimize reaction time, the reaction was moni-
tored by 1H NMR spectroscopy under appropriate reaction
conditions [Et3SiH (20 equiv), t-BuMe2SiCN (10 equiv), and
[CpFe(CO)2Me] (1 equiv) in C6D6 at 25 °C; Figure 1]. An
induction period of a few minutes was observed which may be
used to create a catalytically active species (see below). After
60 h, 80% of t-BuMe2SiCN was converted into t-BuMe2SiH
and the reaction reached equilibrium.
Finally, we varied the reactants to explore the reaction scope
and limitations. Various tertiary silanes were used (Table 4).
Because the metathesis was reversible, each reaction gave a
mixture of four compounds (two silanes and two silyl cya-
nides). Results show that the reaction is quite sensitive to steric
hindrance, because it becomes slower by increasing the length
of the alkyl chain of silane or using a branched silane (compare
Entries 2-5 with Entry 1). The same trend was observed for
aromatic compounds upon increasing the number of phenyl
rings on silicon (Entries 6-8), although Ph3SiH performed
slightly better than MePh2SiH (Entries 8 and 7, respectively).
In all cases the reaction was clean, with 29Si NMR spectrum
of the reaction mixture only displaying four signals: two for
the starting hydrosilane and silyl cyanide, two for the pro-
ducts hydrosilane and silyl cyanide. Because the reaction was
clean and was monitored until equilibrium was attained, the
relatively low yield of new hydrosilane and silyl cyanide is due
to the equilibrium shifting toward the starring position and not
due to the formation of side products. This metathesis reaction
Table 3. Reaction of Et3SiH and t-BuMe2SiCN in Various
Solvents
Et3SiH
+
t-BuMe2SiCN
(20 equiv)
(10 equiv)
[CpFe(CO)2Me] (1 equiv)
Et3SiCN + t-BuMe2SiH
solvent, hν, 25 °C, 24 h
a)
Entry
Solvent
¾
Yield/%b)
1
2
3
4
5
6
7
8
9
neat
C6D6
®
30
60
62
41
69
2.27
2.38
2.02
2.21
4.20
7.58
2.24
4.89
8.93
PhCH3-d8
cyclohexane-d12
1,4-dioxane-d8
Et2O-d10
THF-d8
CCl4
CDCl3
CD2Cl2
pyridine-d5
PhNO2-d5
67
24
c.m.c)
c.m.c)
c.m.c)
0d)
10
11
12
12.4
34.8
0c)
a) ¾: dielectric constant. Values refer to the corresponding
nondeuterated solvents and were taken from Ref. 8. b) Yield
calculated by 1H NMR using 50 equiv of toluene as an internal
standard. c) c.m.: complex mixture of compounds. d) A ligand-
exchange reaction between pyridine and CO was observed by
13C NMR spectroscopy.
20 equiv without a significant effect on the conversion (com-
pare Entries 1 and 6). Almost no reaction took place using a
large excess (500 equiv) of silane (Entry 8). The reason will
be discussed later. The use of an excess of silyl cyanide also
provided a low yield (Entry 9).
Solvent effect was also investigated (Table 3). The choice
was mainly limited to apolar solvents, as most polar solvents are
chemically incompatible with reactants or catalyst.9 Reaction

62
Bull. Chem. Soc. Jpn. Vol. 87, No. 1 (2014) BCSJ AWARD ARTICLE
allowed formation of some silyl cyanides that have not been
synthesized so far (Entries 9-12). Unfortunately, the new
compounds could not be isolated due to their high boiling
point that made purification by distillation difficult. Therefore,
elemental analysis data or mass data could not be obtained for
these compounds. However, expected products were observed
in the reaction mixture by 1H, 13C, and 29Si NMR spectroscopy.
Silyl cyanides are a very useful class of compounds in
organic synthesis. They are mainly used as sources of cyanide
group in nucleophilic substitution and addition reactions.10
For example, they react with aldehydes and ketones to form
silylated cyanohydrins,11 which can be readily converted into
β-amino alcohols and α-hydroxy acids, two important precur-
sors of drugs.12 They are also used in silylation reactions for the
protection of functional groups, especially alcohols, phenols,
and carboxylic acids.10 Several methods for the synthesis of
silyl cyanides have been developed so far.10 The most common
one is the reaction of silyl halides with a metal cyanide
(lithium, sodium, potassium, or silver cyanide). This method
affords yields up to 90%, but usually requires an excess of
metal cyanide,13,14 expensive catalysts (solid polymer15 or
KI16) or high-boiling point solvents (N-methylpyrrolidone13,14
or DMF17) that are difficult to recycle. Therefore, an alternative
method for the synthesis of silyl cyanides is needed. The
reaction mentioned in this paper opens new perspectives in
silyl cyanide formation, although it needs further development
before finding broad synthetic applications.
A plausible catalytic mechanism for the metathesis reaction
between silyl cyanides and silanes is shown in Scheme 4. The
cycle involves the following steps: (1) dissociation of CO from
Fe in [CpFe(CO)2Me], (2) oxidative addition of silane, (3)
reductive elimination of methane to generate the 16e active
species, (4) η1-coordination of silyl cyanide to iron, (5) change
in the coordination mode from η1 to η2, (6) migration of the silyl
group from iron to nitrogen (through a concerted mechanism or
a two-step process involving the formation of a pentacoordi-
nated silicon species), (7) cleavage of Si-CN bond with
migration of the silyl group to the iron center, (8) dissociation
of silyl isocyanide, that isomerizes to the thermodynamically
more stable silyl cyanide,2a,3c,18 (9) oxidative addition of a
Table 4. Catalytic Reaction of Various Silanes with
t-BuMe2SiCN
t-BuMe2SiCN + Silane
(10 equiv)
(20 equiv)
[CpFe(CO)2Me] (1 equiv)
t-BuMe2SiH
Yield/%a)
Silyl cyanide +
1,4-dioxane-d8, hν, 25 °C
Entry
Silane
Time/h
1
2
3
4
5
6
7
8
9
Et3SiH
n-Pr3SiH
n-Bu3SiH
n-Hex3SiH
i-Pr3SiH
Me2PhSiH
MePh2SiH
Ph3SiH
(PhCH2)Me2SiH
(allyl)Me2SiH
Me(Ph)(vinyl)SiH
(EtO)2MeSiH
24
81
81
81
81
69
64
41
25
13
57
28
36
25
5
64
103
103
148
124
124
130
10
11
12
37
58
a) Yield calculated by 1H NMR using 50 equiv of toluene as an
internal standard.
+ R3SiH
(2)
– CO
(1)
Fe
Fe
Fe
C
CH3
C
CH3
C
CH3
O
O
O
R3Si
H
C
O
(3)
– CH4
R'3SiH
R'3SiCN
(4)
Fe
C
SiR3
(10)
O
Fe
Fe
R3Si
C
C
O
SiR3
(5)
C
SiR'3
R3SiH
O
N
H
(9)
R'3Si
Fe
Fe
C
O
SiR'3
SiR3
(6)
C
O
C
N
(7)
Fe
Fe
R'3Si
R3Si NC
R3Si CN
C
O
C
C
O
C
N
N
R'3Si
SiR3
R'3Si
SiR3
Scheme 4. Proposed catalytic cycle.

A. Renzetti et al.
Bull. Chem. Soc. Jpn. Vol. 87, No. 1 (2014)
63
Fe Si O N C H
2.322
2.325
2.088
2.358
1.856
2.103
1.161
A
B
TSBC
C
2.322
1.941
2.063
2.638
1.789
1.186
1.254
1.262
TSCD
D
TSDE
E
Figure 2. Optimized structures in the reaction of the Si-CN bond cleavage by an iron complex. All bond lengths are in ¡.
TSDE
second molecule of silane (R3SiH), and (10) reductive elimi-
nation of the new silane (R¤3SiH) with regeneration of the active
A +
Me3SiCN
8.3
[19.3]
species. This reaction mechanism is similar to that proposed for
C-CN bond cleavage of organonitriles by an iron complex,2 and
0.0
[0.0]
N-CN bond cleavage of cyanamides,3 and O-CN bond cleavage
of cyanates4 by a molybdenum complex. The cleavage of the
TSCD
TSBC
strong Si-CN bond is presumably triggered by a silyl migration
from iron to nitrogen (step (6) in Scheme 4).
-12.7
[0.7]
-14.5
[-2.0]
C
Density functional theory (DFT) calculations were per-
formed to verify the feasibility of the cycles in Scheme 4,
especially focusing on the Si-CN bond cleavage. SiMe3 was
used as a silyl group of R3SiH and R¤3SiCN in the calcula-
tions, analogously to our previous theoretical studies.2a,3b,3c,4
Accordingly, the active species in step (4) is coordinatively
unsaturated [CpFe(CO)SiMe3] (complex A in Figure 2). The
optimized structures of the intermediates and transition states
(TSs) of the reactions and the associated energy profile are
shown in Figures 2 and 3, respectively.
-21.5
[-6.9]
D
B
-25.4
[-12.5]
-30.0
[-17.3]
Fe
C
O
SiMe3
N
C
Fe
C
Me3Si
C
O
N
SiMe3
Fe
N
E
C
O
SiMe3
Me3Si
-48.6
[-35.1]
C
SiMe3
Fe
In the first step of the reaction pathway, silyl cyanide
coordinates to A in the η1-coordination mode via the lone
pair of electrons on the N atom (step (4)) with a binding energy
C
O
C
N
SiMe3
SiMe3
¹1
¹1
of 30.0 kcal mol (Figure 3). Then, silyl cyanide complex B
Figure 3. Energy profile in kcal mol for the Si-CN bond
cleavage by iron silyl complex. All energies are ZPE-
corrected and relative to [CpFe(CO)(SiMe3)] (A) and
Me3SiCN. Energies in parenthesis are free energies and
relative to A and Me3SiCN.
isomerizes to silyl cyanide complex C in the η2-coordination
mode through TSBC (step (5)). This reaction is endothermic
¹1
by 8.5 kcal mol and requires an activation energy of 15.5
kcal mol¹1. We have shown that the energy of this reaction
step is dependent on the electronic effect of the RnE group
in cyanide RnE-CN (R3C-CN, R2N-CN, and RO-CN).3b
An electron-withdrawing group enhances back-donation from
metal to cyano group, stabilizing the side-on complex and
making the reaction less endothermic. In the present case, the
electron-donating silyl group makes back-donation in C diffi-
cult and thus this rearrangement is more endothermic compared

64
Bull. Chem. Soc. Jpn. Vol. 87, No. 1 (2014) BCSJ AWARD ARTICLE
SiMe3
SiMe3
SiMe3
N
N
C
C
N
C
[Fe]
[Fe]
[Fe]
ERn
ERn
ERn
D'(ERn)
I'(ERn)
E'(ERn)
Scheme 5. C-E bond cleavage in an iron coordination sphere.
with the rearrangement of other nitriles. For instance, the
rearrangement of acetonitrile is 7.1 kcal mol exothermic at
the present level of calculations.
The formation of the isocyanide complex E is exothermic
by 48.6 kcal mol¹1, relative to A + Me3SiCN, and E is much
more stable than other intermediates such as B and D. This
is different from the reactions of acetonitrile, cyanamide,
¹1
In the next step (step (6)), the silyl group migrates from iron
to the nitrile N atom, converting C to the N-silylated η2-
silylimino intermediate D with a dative N-Fe bond. The activa-
tion energy of this step was calculated to be 8.8 kcal mol¹1. The
relatively small activation energy for this silyl migration can be
ascribed to the hypervalent character of the migrating Si atom
and the electrostatic attractive interaction between a significant
positive charge on the Si and a negative charge on nitrile
N.2a This reaction is less exothermic (3.9 kcal mol¹1) than the
reactions of the other nitriles; for instance the silyl migration in
¹1
and cyanate (Scheme 5).3b E¤(Me) is only 3.4 kcal mol more
stable than three-membered intermediate D¤(Me). In the reac-
tions of cyanamides, E¤(NR2) is even less stable than the
three-membered intermediate D¤(NR2). This difference can be
explained using the difference in the bond dissociation energies.
¹1
Me3Si-CN bond is only 2.4 kcal mol stronger than H3C-CN
bond, whereas Fe-Si bond in E is 18.7 kcal mol¹1 stronger than
the Fe-C bond in E¤(Me). The above results show that the
reaction pathway including the SiMI reaction is feasible for
Si-CN bond cleavage.
¹1
the case of acetonitrile is 9.7 kcal mol exothermic. This is
presumably because of the steric hindrance between the bulky
silyl groups in D.
After E is formed, the isomerization of silyl isocyanide to
silyl cyanide which is detected experimentally takes place.
We have previously shown that this isomerization takes place
out of the coordination sphere after the dissociation of silyl
isocyanide from the Fe atom.2a
Dissociation of silyl isocyanide from E forms a 16e species
A, which in principle can react in two ways: (i) a reaction with
Me3SiCN to give B and (ii) a reaction with Me3SiH to give
[CpFe(CO)(H)(SiMe3)2] (F). We performed the calculation
In step (7), the Si-CN bond is cleaved. Theoretical calcu-
lations for the reaction of acetonitrile, cyanamide, and cyanate
showed that in this step there is an intermediate I¤ bearing the
interaction between the Fe atom and the ERn group as shown
in Scheme 5. Thus, for instance, the reaction of acetonitrile
requires two reaction steps, the N dissociation from the Fe atom
and the C-CN bond cleavage. However, in the reaction of silyl
cyanide such an intermediate does not exist. Instead, D trans-
and found that the reaction E + Me3SiCN ¼ B + Me3SiNC is
¹1
forms directly to the silyl isocyanide complex E through TSDE
.
19.5 kcal mol
endothermic, whereas E + HSiMe3 ¼ F +
¹1
The structural feature in TSDE is similar to that in the TS
structures leading to I¤ in the reactions of acetonitrile, cyana-
mide, and cyanate, suggesting that TSDE is a transition state
for the N dissociation. However, an investigation of the reac-
tion coordinate vector shows that the Si-CN bond lengthening
is also an important structural feature of TSDE. Actually, the
Si-CN bond is stretched from 1.885 ¡ at D to 1.921 ¡ at
TSDE. Analysis of the electronic structure of I¤ in the case of
acetonitrile showed that agostic interactions between both the
C-H and C-CN sigma bonds and the Fe vacant d orbital activate
the C-CN sigma bond.3b Because the Si-CN bond orbital is
larger and higher in energy than the C-CN bond orbital, its
interaction with iron is stronger and more effectively activates
the Si-CN bond.19 This merges the two steps, N dissociation
and Si-CN bond cleavage, into a single step. Activation energy
of this step was calculated to be 33.7 kcal mol¹1 which is com-
Me3SiNC is 21.7 kcal mol
endothermic (Figure 4). The
results show that [CpFe(CO)SiR3] reacts more feasibly with
R¤3SiCN than with R3SiH, as is shown in Scheme 4.
Experimental evidence is in agreement with theoretical
calculations. Methane was detected in the reaction mixture by
1H NMR spectroscopy, confirming the reductive elimination
of step (3) and hence the formation of the active species
[CpFe(CO)SiR3].
In order to isolate any of the intermediates in the catalytic
cycle, we replaced [CpFe(CO)2Me] with pyridine complexes 1
and 3 (Scheme 6). Such complexes are also precursors of
the active species, as they can easily generate a 16e complex
by dissociation of pyridine.3a Heating a solution of 1 and t-
BuMe2SiCN in equimolar amounts in benzene at 80 °C for 3 h
yielded 2 quantitatively according to NMR measurements, but
the isolation as a solid failed. In contrast, reaction of 3 with t-
BuMe2SiCN yielded 4, which was isolated as a brown powder
in 86% yield. Although 4 was not obtained in crystalline form
suitable for X-ray analysis, its structure in solution could be
determined by comparison of its NMR data with an analo-
gous intermediate whose crystalline structure is known. The
29Si NMR spectrum of 4 displays two signals (see Experimental
¹1
parable to that in the reaction of acetonitrile (32.9 kcal mol
at the present level of calculations). The N dissociation is a
dominant structure change before reaching the TSDE, and the
energy required by the N dissociation does not depend on the
nature of the ERn group so much. Accordingly, the activation
energy for this step is similar to that of the other nitriles.

A. Renzetti et al.
Bull. Chem. Soc. Jpn. Vol. 87, No. 1 (2014)
65
Section). The signal at 40.7 ppm is close in value to that of the
N-silylated η2-amidino iron complex of dimethylcyanamide
(¤ = 35.4 ppm),3a and can therefore be attributed to the silicon
atom in the SiPh3 group. The signal at ¹4.2 ppm is close in
value to that of t-BuMe2SiCN (¤ = ¹2.8 ppm), so it can be
assigned to the t-BuMe2Si group. The product coming from the
reaction of t-BuMe2SiCN with pyridine complex 1 (or 3) is a
resonance hybrid of two contributing forms: the imino complex
2 (or 4), bearing a C=N double bond and a N¼Fe dative bond,
and the Fisher-type carbene 2¤ (or 4¤), with an Fe=C double
bond and a N-Fe covalent bond (Scheme 7). In the absence of
data on the crystalline structure of 2 and 4, one has to rely on
NMR to evaluate the contribution of the resonance forms to
the hybrid structure. The 13C NMR spectra of carbene com-
plexes are characterized by marked deshielding of the carbene
carbon atom.20 The chemical shift of the carbene carbon atom
in two cyclopentadienyl iron carbene complexes related to
2 and 4 {[η5-C5H5(CO)2Fe=CHC6H5]PF6 and [η5-C5H5(CO)-
{P(C6H5)3}Fe=CHC6H5]PF6} has been reported to be about
340 ppm.20d The chemical shift measured for the CN carbon
atom in complexes 2 and 4 is dramatically lower (¤ = 144.2
and 143.4 ppm, respectively). Although the chemical shift of
the carbene carbon atom depends on the nature of substituents
on carbon, such a large difference in chemical shift of the
carbon atom on iron between complexes 2 and 4 and carbene
complexes above suggests that carbene forms 2¤ and 4¤ do not
significantly contribute to the structure of hybrid. Photoreaction
of 2 with 2 equiv of Et3SiH over 4 days generated Et3SiCN and
t-BuMe2SiH in 58% yield. This result confirms that the N-
silylated η2-silylimino iron complex is an intermediate in the
catalytic cycle.
21.7
[21.5]
19.5
[18.5]
F
Fe
Fe
SiMe3
C
O
C
O
SiMe3
H
Me3Si
N
+
C
B
Me3SiCN
+
Me3SiNC
SiMe3
0.0
[0.0]
+
Me3SiNC
+
Me3SiH
E
Fe
C
O
C
N
SiMe3
+
SiMe3
Me3SiCN
+
Me3SiH
¹1
The low conversion measured in the presence of a large
excess of silane (Table 2, Entry 8) can be rationalized in the
light of a side reaction involving the active species [CpFe-
(CO)SiR3]. This complex reacts with R¤3SiCN and also with
Figure 4. Energy profile in kcal mol for exchange of
Me3SiNC ligand by Me3SiCN and Me3SiH. All energies
are ZPE-corrected and relative to E + Me3SiCN + Me3-
SiH. Energies in parenthesis are free energies.
R5
R5
N
Δ, 80 °C
toluene, 2h
Fe
Fe
C
+
+ t-BuMe2SiCN
N
C
SiR'3
C
O
O
SiR'3
(1.0 equiv)
N
t-BuMe2Si
N-silylated η2-silylimino iron complex
(1.0 equiv)
1
3
R = H, R' = Et
R = Me, R' = Ph
2
4
100% (NMR yield)
86% (Isolated yield)
Scheme 6. Synthesis of N-silylated η2-silylimino iron complex.
R5
R5
Fe
C
Fe
C
C
N
C
N
O
O
SiR'3
SiR'3
t-BuMe2Si
t-BuMe2Si
2
4
2'
4'
R = H, R' = Et
R = Me, R' = Ph
Scheme 7. Possible resonance structures of 2 and 4.

66
Bull. Chem. Soc. Jpn. Vol. 87, No. 1 (2014) BCSJ AWARD ARTICLE
R3SiH. In the former case, the reaction proceeds according to
the sequence shown in Scheme 4. In the latter case, Si-H
bond in R3SiH undergoes oxidative addition to the iron center
generating [CpFe(CO)(H)(SiR3)2]. This complex has been
previously isolated from the reaction of [CpFe(CO)2Me] with
R3SiH, and it has been characterized by NMR spectroscopy
and X-ray crystallography.21 In the reaction of 10 equiv of t-
BuMe2SiCN, 20 equiv of Et3SiH, and 1 equiv of [CpFe(CO)2-
Me], disilyl complex [CpFe(CO)(H)(SiEt3)2] was detected
according to literature methods. The other chemicals were
commercially available and were used without further purifi-
cation. Photoirradiation was performed with a 400 W medium-
pressure mercury arc lamp (Pyrex filtered) at room temperature
under nitrogen atmosphere. NMR spectra (1H, 13C, and 29Si)
were recorded on a JNM-AL400 spectrometer. The residual
1
peak of solvent was used as reference for H and 13C NMR
spectra of pure compounds. The peak of toluene was used as
1
reference for H and 13C NMR spectra of reaction mixtures.
1
by H NMR spectroscopy after 0.5 h of photoirradiation, and
Peak positions in 29Si NMR spectra were referenced to external
tetramethylsilane (¤ = 0 ppm). GC-MS spectra were acquired
on a GCMS-QP2010Plus spectrometer using a temperature
gradient (25-300 °C).
disappeared after a prolonged reaction time (10 h). The molar
ratio of R¤3SiCN and R3SiH is responsible for which reaction
occurs mainly. When silane and silyl cyanide are nearly in
equimolar ratio, the reaction mainly follows the catalytic cycle
in Scheme 4, affording the metathesis products as shown by
DFT calculation (Figure 3). When silane is in a large excess
with respect to silyl cyanide, the side reaction becomes
prevalent and reaction yield is low.
General Procedure for the Metathesis Reactions.
A
solution of [CpFe(CO)2Me] (4.1 mg, 21.4 ¯mol, 1.0 equiv),
silane (428 ¯mol, 20 equiv), silylcyanide (214 ¯mol, 10 equiv)
and toluene (22.6 ¯L, 214 ¯mol, 10 equiv) in 1,4-dioxane-d8 or
benzene-d6 (500 ¯L)32 under nitrogen was photoirradiated at
1
Another possible mechanism for the metathesis reaction
involves the oxidative addition of the Si-CN bond in silyl cya-
nide. The oxidative addition of E-CN bond in acetonitrile and
cyanamide has been investigated2a,3b and found to be disfavored
for two reasons: 1) the high activation energy required for
this step (52.72a and 31.33b kcal mol¹1, respectively), and 2) the
instability of product [CpFeCO(SiR3)(E)(CN)], which would
bear iron in the +4 oxidation state. Based on these findings, the
oxidative addition pathway for acetonitrile and cyanamide has
been ruled out. Because the E-CN bond cleavage reactions
investigated in our group all proceed through silyl migration,2-4
we proposed the silyl migration pathway for Si-CN bond
cleavage reaction rather than Si-CN oxidative addition.
In conclusion, this paper describes an iron-catalyzed meta-
thesis reaction of silyl cyanides and hydrosilanes involving
the recombination of their Si-CN and Si-H bonds. The study of
mechanism by a combination of DFT calculations and experi-
mental techniques revealed that the reaction is triggered by
the migration of a silyl group from the iron center to nitrogen
of silyl cyanide, and follows a mechanism similar to that of
other strong bond cleavage reactions. This work is important
for two reasons. First, it is the first example of metathesis
reaction of Si-H and Si-CN bonds. Therefore, it sheds new
light on the activation of strong Si-CN bonds by transition-
metal complexes. Second, it is potentially a new method for
the preparation of silyl cyanides. The reaction offers several
advantages with respect to traditional methodologies, such as
the use of cheap catalyst, low-boiling solvents, and stoichio-
metric amounts of cyanide source. Moreover, it allows synthe-
sis of silyl cyanides never obtained so far, thus opening new
perspectives on the use of this class of compounds.
25 °C. The reaction was monitored by H, 13C, and 29Si NMR
spectroscopy until equilibrium was attained.
Diethoxymethylsilyl Cyanide. The title compound was
obtained from diethoxymethylsilane and tert-butyldimethylsi-
lyl cyanide according to the general procedure. 1H NMR (¤, 1,4-
dioxane-d8): 0.07 (s, 3H, CH3), 1.14 (t, 6H, J = 6.8 Hz, CH3-
CH2O), 3.66 (q, 4H, J = 6.8 Hz, CH3CH2O). 13C{1H} NMR (¤,
C6D6): ¹5.3 (s, CH3), 18.8 (s, CH3CH2O), 58.2 (s, CH3CH2O),
125.4 (s, CN). 29Si NMR (¤, C6D6): ¹27.6 (s).
Benzyldimethylsilyl Cyanide. The title compound was
obtained from benzyldimethylsilane and tert-butyldimethylsilyl
cyanide according to the general procedure. H NMR (¤, 1,4-
dioxane-d8): 0.25 (s, 6H, CH3), 2.33 (s, 2H, CH2), 7.01-7.27
(m, 5H, Ph). 13C{1H} NMR (¤, C6D6): ¹4.3 (s, CH3), 23.6 (s,
CH2), 125.1 (s, CN), 125.6 (Ph), 128.6 (Ph), 129.0 (Ph), 136.4
(Ph). 29Si NMR (¤, C6D6): ¹12.9 (s).
Methylphenylvinylsilyl Cyanide. The title compound was
obtained from methylphenylvinylsilane and tert-butyldimethyl-
1
1
silyl cyanide according to the general procedure. H NMR (¤,
1,4-dioxane-d8): 0.33 (s, 1H, CH3), 5.82 (dd, 1H, Jtrans
23.9 Hz, Jvic = 3.4 Hz, CH=CH2), 6.06 (dd, 1H, Jcis = 15.6 Hz,
vic = 3.4 Hz, CH=CH2), 6.28 (ddd, 1H, Jtrans = 23.9 Hz, Jcis
=
J
=
2
15.6 Hz, JSi-H = 3.9 Hz, CH=CH2), 7.41 (m, 3H, Ph), 7.63
(m, 2H, Ph). 13C{1H} NMR (¤, C6D6): ¹4.5 (s, CH3), 124.3
(s, CN), 128.8 (s, Ph), 130.1 (s, Ph), 131.2 (s, CH=CH2), 134.5
(s, Ph), 134.8 (s, Ph), 138.3 (s, CH=CH2). 29Si NMR (¤, C6D6):
¹21.5 (s).
Triethylsilyl Cyanide. A solution of [CpFe(CO)2Me] (41
mg, 214 mmol, 1.0 equiv), Et3SiH (4.28 mmol, 20 equiv), and
t-BuMe2SiCN (2.14 mmol, 10 equiv) in benzene (5.0 mL) was
photoirradiated at room temperature for 24 h. After removal of
volatile materials at 4000 Pa, the residue was distilled at 80 °C
Experimental
and 100 Pa to give Et3SiCN as a colorless liquid. H, 13C, and
1
General Remarks. All reactions were carried out under an
atmosphere of dry nitrogen by using Schlenk tube techniques.
Toluene, hexane, and pentane were distilled from sodium
and benzophenone prior to use and stored under nitrogen.
[(C5R5)Fe(CO)2Me] (C5R5 = C5H5,22 C5H4Me,23 C5Me5,24
C5H4(SiMe3),25 and C5H4I26), [(C5H5)Fe(CO)2X] (X = Cl,27
I,28 CH2Ph,29 SiMe3,30 and SiEt331), [CpMo(CO)3Me],22
[CpW(CO)3Me],22 and [CpFe(CO)(py)SiEt3],2c were prepared
29Si NMR spectra of this compound have not been described in
the literature,33 so they are reported below. 1H NMR (¤, C6D6):
0.31 (q, 6H, J = 8.0 Hz, CH2CH3), 0.79 (t, 9H, J = 8.0 Hz,
CH2CH3). 13C{1H} NMR (¤, C6D6): 3.0 (s, CH2CH3), 7.1 (s,
CH2CH3), 124.8 (s, CN). 29Si NMR (¤, C6D6): ¹1.3 (s).
Preparation of [Cp*Fe(CO)(py)SiPh3] (py: pyridine) (3).
A solution of [Cp*Fe(CO)2SiPh3] (812 mg, 1.60 mmol, 1.0
equiv) in pyridine (9.0 mL, 112 mmol, 70 equiv) was photo-

A. Renzetti et al.
Bull. Chem. Soc. Jpn. Vol. 87, No. 1 (2014)
67
irradiated at 5 °C for 48 h. Removal of volatile materials under
reduced pressure left a dark-red solid, which was washed ten
times with 2 mL of hexane at ¹70 °C. Recrystallization with
hexane at ¹20 °C afforded pure crystals of 3 (495 mg, 0.89
Gibbs free energies were also calculated at a temperature of
298.15 K and a pressure of 1.00 atm. Unless otherwise explic-
itly stated, the ZPE-corrected potential energies are used in the
discussion.
1
mmol, 55% yield). H NMR (¤, C6D6): 1.34 (s, 15H, C5Me5),
5.91 (t, 2H, J = 6.8 Hz, m-py), 6.40 (t, 1H, J = 8.0 Hz, p-py),
7.20 (m, 9H, Ph), 7.71 (m, 6H, Ph), 8.26 (m, 2H, o-py).
13C{1H} NMR (¤, C6D6): 9.6 (s, C5Me5), 90.9 (s, C5Me5),
123.4 (s, m-py), 128.2 (s, Ph), 128.9 (s, Ph), 134.0 (s, p-py),
136.9 (s, Ph), 145.5 (s, Ph), 157.6 (s, o-py), 224.2 (s, CO).
29Si NMR (¤, C6D6): 38.3 (s).
This work was supported by a Challenging Exploratory
Research (No. 23655056) grant from the Ministry of Educa-
tion, Culture, Sports, Science and Technology, and by a Grant-
in-Aid for Scientific Research from the Japan Society for the
Promotion of Science.
Synthesis of N-Silylated η2-Silylimino Complexes 2
and 4. Complex 1 (580 mg, 1.69 mmol, 1.0 equiv) was treated
with t-BuMe2SiCN (239 mg, 1.69 mmol, 1.0 equiv) in toluene
(5 mL) at 80 °C for 2 h. Removal of volatile materials under
reduced pressure led to formation of the corresponding N-
silylated η2-amidino complex 2 as a light brown oil in 100%
1H NMR yield. In a similar way, complex 3 (753 mg, 1.35
mmol, 1.0 equiv) was treated with t-BuMe2SiCN (191 mg, 1.35
mmol, 1.0 equiv) in toluene (5 mL) at 80 °C for 2 h. Removal
of solvent under reduced pressure, followed by washing of
residue with pentane at ¹70 °C, provided 4 as a brown-orange
powder (1.16 mmol, 719 mg, 86%). Complex 4 was so air- and
moisture-sensitive that correct elemental analysis data could
not be obtained, although satisfactory spectroscopic data were
obtained. 2; 1H NMR (¤, C6D6): 0.44 (q, 6H, J = 7.8 Hz,
SiCH2CH3), 0.57 and 0.58 (s, 6H, t-BuMe2Si), 0.89 (t, 9H,
J = 7.8 Hz, SiCH2CH3), 1.21 (s, 9H, t-BuMe2Si), 4.42 (s, 5H,
Cp). 13C{1H} NMR (¤, C6D6): ¹6.6 (s, t-BuMe2Si), ¹6.4
(s, t-BuMe2Si), 9.9 (s, SiCH2CH3), 11.2 (s, SiCH2CH3), 16.6
(s, Me3CSi), 25.4 (s, Me3CSi), 81.9 (s, Cp), 144.2 (s, CN), 222.3
(s, CO). 29Si NMR (¤, C6D6): ¹4.7 (s, t-BuMe2Si), 53.4 (s,
References
1
For some recent reviews on bond cleavage reactions, see:
C-H bonds: a) S. H. Cho, J. Y. Kim, J. Kwak, S. Chang, Chem.
Soc. Rev. 2011, 40, 5068. C-H bonds: b) J. Wencel-Delord, T.
Dröge, F. Liu, F. Glorius, Chem. Soc. Rev. 2011, 40, 4740. C-C
bonds: c) A. Korotvicka, D. Necas, M. Kotora, Curr. Org. Chem.
2012, 16, 1170. C-CN and CÔC bonds: d) M. Tobisu, N. Chatani,
Chem. Soc. Rev. 2008, 37, 300. C-H and C-C bonds: e) N.
Chatani, Organomet. News 2009, 140. C-F bonds: f) E. Clot, O.
Eisenstein, N. Jasim, S. A. Macgregor, J. E. McGrady, R. N.
Perutz, Acc. Chem. Res. 2011, 44, 333. C-F bonds: g) T. Braun, F.
Wehmeier, Eur. J. Inorg. Chem. 2011, 613. C-F bonds: h) H. Amii,
K. Uneyama, Chem. Rev. 2009, 109, 2119. C-F bonds: i) M.
Klahn, U. Rosenthal, Organometallics 2012, 31, 1235. C-F bonds:
j) C. Bianchini, A. Meli, F. Vizza, Eur. J. Inorg. Chem. 2001, 43.
Si-Si bonds: k) H. K. Sharma, K. H. Pannell, Chem. Rev. 1995, 95,
1351. P-H and P-P bonds: l) S. Greenberg, D. W. Stephan, Chem.
Soc. Rev. 2008, 37, 1482.
2
a) H. Nakazawa, T. Kawasaki, K. Miyoshi, C. H. Suresh,
N. Koga, Organometallics 2004, 23, 117. b) H. Nakazawa, K.
Kamata, M. Itazaki, Chem. Commun. 2005, 4004. c) H. Nakazawa,
M. Itazaki, K. Kamata, K. Ueda, Chem.®Asian J. 2007, 2, 882.
1
SiCH2CH3). 4; H NMR (¤, C6D6): ¹0.16 (s, 3H, t-BuMe2Si),
3
a) K. Fukumoto, T. Oya, M. Itazaki, H. Nakazawa, J. Am.
¹0.11 (s, 3H, t-BuMe2Si), 0.81 (s, 9H, t-Bu), 1.49 (s, 15H,
C5Me5), 7.22 (m, 9H, Ph), 7.96 (m, 6H, Ph). 13C{1H} NMR (¤,
C6D6): ¹6.1 (s, t-BuMe2Si), 9.5 (s, C5Me5), 16.4 (s, Me3CSi),
25.5 (s, Me3CSi), 92.1 (s, C5Me5), 127.2 (s, Ph), 136.6 (s, Ph),
137.0 (s, Ph), 143.4 (s, CN), 145.6 (s, Ph), 223.2 (s, CO).
29Si NMR (¤, C6D6): ¹4.2 (s, t-BuMe2Si), 40.7 (s, SiPh3).
Photoreaction of 2 with Et3SiH. A solution of complex 2
(22.3 mg, 46.0 ¯mol, 1.0 equiv) in 500 ¯L of benzene-d6 was
treated with Et3SiH (14.7 ¯L, 92.0 ¯mol, 2.0 equiv) under
nitrogen and photoirradiated at 25 °C. After 96 h of photo-
irradiation, the solution changed color from brown to yellow,
Chem. Soc. 2009, 131, 38. b) A. A. Dahy, N. Koga, H. Nakazawa,
Organometallics 2012, 31, 3995. c) A. A. Dahy, N. Koga, H.
Nakazawa, Organometallics 2013, 32, 2725.
4
K. Fukumoto, A. A. Dahy, T. Oya, K. Hayasaka, M. Itazaki,
N. Koga, H. Nakazawa, Organometallics 2012, 31, 787.
J. Y. Corey, in The Chemistry of Organic Silicon Com-
5
pounds Part 1, ed. by S. Patai, Z. Rappoport, John Wiley & Sons,
New York, 2004, pp. 5-6.
6
Reaction in the NMR tube (in anhydrous solvent under
nitrogen) was clean, affording only hydrosilane and silyl cyanide.
No silanols or disiloxanes were observed by NMR spectroscopy.
When the NMR tube was opened and the sample diluted under air
for injection into the GC-MS instrument, both silyl cyanides
hydrolyzed generating silanols and disiloxanes. These products
were detected by GC-MS. Silanols and disiloxanes are not
products of the metathesis reaction because they only formed after
reaction upon exposure to air, not during the reaction under
nitrogen.
1
and H NMR spectroscopy revealed the formation of Et3SiCN
and t-BuMe2SiH in 58% yield based on the amount of complex
2 used.
Computational Methods.
Molecular geometries were
optimized at the density functional theory level using the
M06 functional with 6-311+G(d,p) basis functions using
the Gaussian09 program.34-36 Normal coordinate analysis was
performed for all stationary points to characterize the equi-
librium structures and intermediates (the Hessian matrix has
positive eigenvalues for equilibrium structures and one nega-
tive value for each TS). Calculations of intrinsic reaction coor-
dinates near the TS followed by geometry optimization were
performed for all reactions to confirm the connectivity of the
TSs.37 Unscaled vibrational frequencies were used to calculate
zero-point-energy (ZPE) correction to the total energies. The
7
a) R. P. Alexander, G. R. Stephenson, J. Organomet. Chem.
1986, 299, C1. b) R. P. Alexander, T. D. James, G. R. Stephenson,
J. Chem. Soc., Dalton Trans. 1987, 2013. c) A. J. L. Pombeiro,
D. L. Hughes, C. J. Pickett, R. L. Richards, J. Chem. Soc.,
Chem. Commun. 1986, 246. d) S. S. P. R. Almeida, M. F. C.
Guedes da Silva, J. J. R. Fraústoda Silva, A. J. L. Pombeiro,
J. Chem. Soc., Dalton Trans. 1999, 467. e) W. J. Evans, E.
Montalvo, T. M. Champagne, J. W. Ziller, A. G. DiPasquale, A. L.
Rheingold, Organometallics 2009, 28, 2897. f) W. J. Evans, T. J.

68
Bull. Chem. Soc. Jpn. Vol. 87, No. 1 (2014) BCSJ AWARD ARTICLE
Mueller, J. W. Ziller, Chem.®Eur. J. 2010, 16, 964. g) L. Zhang,
C. W. Fung, K. S. Chan, Organometallics 2006, 25, 5381. h) N.
Chatani, T. Takeyasu, T. Hanafusa, Tetrahedron Lett. 1988, 29,
3979. i) T. Pohlmann, A. de Meijere, Org. Lett. 2000, 2, 3877.
Chem. 1974, 69, 131.
25 S. R. Berryhill, B. Sharenow, J. Organomet. Chem. 1981,
221, 143.
26 C. Lo Sterzo, M. M. Miller, J. K. Stille, Organometallics
1989, 8, 2331.
27 B. F. Hallam, O. S. Mills, P. L. Pauson, J. Inorg. Nucl.
Chem. 1955, 1, 313.
8
Y. Marcus, The Properties of Solvents, John Wiley & Sons,
New York, 1998, pp. 95-102.
Water, alcohols, and ketones react with silyl cyanides
9
(Ref. 10); CH3CN (Ref. 2) and DMF (Ref. 21a) may also undergo
bond breaking reaction in the presence of [CpFe(CO)2Me].
10 a) J. K. Rasmussen, S. M. Heilmann, L. R. Krepski, in
Advances in Silicon Chemistry, ed. by G. L. Larson, JAI Press,
Greenwich, CT, 1991, Vol. 1, pp. 65-187. b) W. P. Weber, Silicon
Reagents for Organic Synthesis in Reactivity and Structure
Concepts in Organic Chemistry, Springer-Verlag, Berlin, 1983,
Vol. 14. doi:10.1007/978-3-642-68661-0. c) M. North, in Science
of Synthesis: Houben Weyl Methods of Molecular Transforma-
tions, ed. by I. Fleming, Thieme, Stuttgart, 2002, Vol. 4, pp. 499-
512.
11 M. North, D. L. Usanov, C. Young, Chem. Rev. 2008, 108,
5146.
12 M. Breuer, K. Ditrich, T. Habicher, B. Hauer, M. Keßeler,
R. Stürmer, T. Zelinski, Angew. Chem., Int. Ed. 2004, 43, 788.
13 S. Hünig, G. Wehner, Synthesis 1979, 522.
14 J. K. Rasmussen, S. M. Heilmann, Synthesis 1979, 523.
15 K. Sukata, Bull. Chem. Soc. Jpn. 1987, 60, 2257.
16 M. T. Reetz, I. Chatziiosifidis, US Patent 4429145, 1984.
17 J. M. Aizpurua, C. Palomo, Nouv. J. Chim. 1984, 8, 51.
18 a) M. R. Booth, S. G. Frankiss, Chem. Commun. (London)
1968, 1347. b) T. A. Bither, W. H. Knoth, R. V. Lindsey, Jr., W. H.
Sharkey, J. Am. Chem. Soc. 1958, 80, 4151.
19 N. Koga, K. Morokuma, J. Am. Chem. Soc. 1988, 110, 108.
20 a) C. P. Casey, Transition Metal Organometallics in
Organic Synthesis, ed. by H. Alper, Academic Press, New York,
1976, Vol. 1, pp. 190-193. b) M. H. Chisholm, S. Godleski, Prog.
Inorg. Chem. 1976, 20, 299. c) C. G. Kreiter, V. Formáček, Angew.
Chem., Int. Ed. Engl. 1972, 11, 141. d) C. P. Casey, S. W.
Polichnowski, J. Am. Chem. Soc. 1977, 99, 6097.
21 a) M. Itazaki, K. Ueda, H. Nakazawa, Angew. Chem. Int.
Ed. 2009, 48, 3313; M. Itazaki, K. Ueda, H. Nakazawa, Angew.
Chem. Int. Ed. 2009, 48, 6938. b) M. Itazaki, M. Kamitani, K.
Ueda, H. Nakazawa, Organometallics 2009, 28, 3601.
22 T. S. Piper, G. Wilkinson, J. Inorg. Nucl. Chem. 1956, 3,
104.
28 B. F. Hallam, P. L. Pauson, J. Chem. Soc. 1956, 3030.
29 J. P. Bibler, A. Wojcicki, J. Am. Chem. Soc. 1966, 88, 4862.
30 R. B. King, K. H. Pannel, Inorg. Chem. 1968, 7, 1510.
31 E. Scharrer, S. Chang, M. Brookhart, Organometallics
1995, 14, 5686.
32 1,4-Dioxane-d8 was used for 1H NMR spectra as it afforded
the highest yield among the tested solvents (Table 3). Benzene-d6
was used for 13C and 29Si NMR spectra due to its lower cost.
33 The 1H NMR spectrum of Et3SiCN has been generically
reported as a multiplet at 0.53-1.30 ppm in CDCl3. See: U.
Hertenstein, S. Hünig, M. Öller, Chem. Ber. 1980, 113, 3783.
34 Y. Zhao, D. G. Truhlar, Theor. Chem. Acc. 2008, 120, 215.
35 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B.
Mennucci, G. A. Petersson, 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, 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, Ö. Farkas, J. B. Foresman, J. V. Ortiz,
J. Cioslowski, D. J. Fox, Gaussian 09, (Revision C.01), Gaussian,
Inc., Wallingford CT, 2009.
36 a) R. Krishnan, J. S. Binkley, R. Seeger, J. A. Pople,
J. Chem. Phys. 1980, 72, 650. b) A. J. H. Wachters, J. Chem.
Phys. 1970, 52, 1033. c) P. J. Hay, J. Chem. Phys. 1977, 66, 4377.
d) A. D. McLean, G. S. Chandler, J. Chem. Phys. 1980, 72, 5639.
37 a) K. Fukui, Acc. Chem. Res. 1981, 14, 363. b) H. P.
Hratchian, H. B. Schlegel, Theory and Applications of Computa-
tional Chemistry: The First 40 Years, ed. by C. E. Dykstra, G.
Frenking, K. S. Kim, G. E. Scuseria, Elsevier, Amsterdam, 2005,
pp. 195-249. doi:10.1016/B978-044451719-7/50053-6.
23 [(C5H4Me)Fe(CO)2Me] was prepared according to the
method in Ref. 22 by using methylcyclopentadiene.
24 R. B. King, W. M. Douglas, A. Efraty, J. Organomet.