Catalytic Hydrogenation of Benzonitrile by Triruthenium Clusters: Consecutive Transformations of Benzonitrile on the Face of a Ru3 Plane

Article abstract of DOI:10.1021/acs.organomet.8b00165

Reactions of the triruthenium clusters {Cp?Ru(μ-H)}₃(μ₃-H)₂ (1) and (Cp?Ru)₃(μ₃-H)(μ-H)₂(μ-CO) (12) (Cp? = η⁵-C₅Me₅) with benzonitrile were investigated in relation to the selective hydrogenation of benzonitrile to benzylamine. Benzonitrile undergoes consecutive transformations into μ₃-η²:η²(⊥)-nitrile, μ₃-η²:η η²(⊥)-imidoyl, μ₃-η²(∥)-imidoyl, μ₃-η²-alkylideneamido, μ₃-imido, and μ-amido ligands on the Ru₃ plane accompanied by the uptake of dihydrogen. The reactions are analogues of nitrile hydrogenation on a metal surface. The complexes are structural models of chemisorbed species and catalyze the hydrogenation of benzonitrile. Complex 1 catalyzes benzonitrile hydrogenation without additives but exhibits only moderate selectivity toward benzylamine. Although the μ₃-benzylimido complex {Cp?Ru(μ-H)}₃(μ₃-NCH₂Ph) (4) was obtained by reaction of 1 with benzonitrile, it was readily transformed into to the μ₃-imidoyl complexes (Cp?Ru)₃(μ-H)₂{μ₃-η²:η²(⊥)-PhCNH} (3) and (Cp?Ru)₃(H)₄{μ₃-η²(∥∥)-PhCNH} (4), which are key intermediates in secondary imine formation. Two benzonitrile molecules were incorporated on the Ru₃ plane under the reaction conditions, which decreases the selectivity of primary amine formation. In contrast, μ-carbonyl complex 12 suppresses the incorporation of additional benzonitrile ligands and the formation of μ₃-imidoyl species due to the presence of CO. These features of 12 bring about significant improvement in the selectivity of benzonitrile hydrogenation and produce benzylamine in 92% yield.

Full text of DOI:10.1021/acs.organomet.8b00165

Article
Cite This: Organometallics XXXX, XXX, XXX−XXX
Catalytic Hydrogenation of Benzonitrile by Triruthenium Clusters:
Consecutive Transformations of Benzonitrile on the Face of a Ru3
Plane
,
†,‡
†
†
†
†
Toshiro Takao,*
Sachie Horikoshi, Takashi Kawashima, Sachio Asano, Yuta Takahashi,
†
†
Akira Sawano, and Hiroharu Suzuki
†Department of Chemical Science and Engineering, School of Materials and Chemical Technology, Tokyo Institute of Technology,
2
-12-1 O-okayama, Meguro-ku, Tokyo 152-8552, Japan
‡
JST, ACT-C, 4-1-8 Honcho, Kawaguchi, Saitama 332-0012, Japan
S Supporting Information
*
ABSTRACT: Reactions of the triruthenium clusters {Cp*Ru-
(
(
μ-H)} (μ -H) (1) and (Cp*Ru) (μ -H)(μ-H) (μ-CO) (12)
3 3 2 3 3 2
5
Cp* = η -C Me ) with benzonitrile were investigated in rela-
5
5
tion to the selective hydrogenation of benzonitrile to benzylamine.
2
2
Benzonitrile undergoes consecutive transformations into μ -η :η (⊥)-
3
2
2
2
2
nitrile, μ -η :η (⊥)-imidoyl, μ -η (∥)-imidoyl, μ -η -alkylideneamido,
3
3
3
μ -imido, and μ-amido ligands on the Ru plane accompanied by
3
3
the uptake of dihydrogen. The reactions are analogues of nitrile
hydrogenation on a metal surface. The complexes are structural models
of chemisorbed species and catalyze the hydrogenation of benzonitrile.
Complex 1 catalyzes benzonitrile hydrogenation without additives but
exhibits only moderate selectivity toward benzylamine. Although
the μ -benzylimido complex {Cp*Ru(μ-H)} (μ -NCH Ph) (4)
3
3
3
2
was obtained by reaction of 1 with benzonitrile, it was readily transformed into to the μ -imidoyl complexes (Cp*Ru) (μ-H) {μ -
3
3
2
3
2
2
2
η :η (⊥)-PhCNH} (3) and (Cp*Ru) (H) {μ -η (∥∥)-PhCNH} (4), which are key intermediates in secondary imine formation.
Two benzonitrile molecules were incorporated on the Ru plane under the reaction conditions, which decreases the selectivity of
3
4
3
3
primary amine formation. In contrast, μ-carbonyl complex 12 suppresses the incorporation of additional benzonitrile ligands and
the formation of μ -imidoyl species due to the presence of CO. These features of 12 bring about significant improvement in the
3
selectivity of benzonitrile hydrogenation and produce benzylamine in 92% yield.
INTRODUCTION
selective hydrogenation of nitriles into the corresponding
■
6c,7
primary amines, although most catalysts require the addition
Hydrogenation of nitriles using heterogeneous catalysts, such as
Raney-Ni and Co, is an important basis for the synthesis of
t
of significant amounts of base, typically BuOK, to suppress
1
byproduct formation.
primary amines. Nitrile hydrogenation is considered to pro-
Catalysts exhibiting high primary amine selectivity withou8t
ceed via formation of an aldimine-like intermediate on a metal
surface. Because the nucleophilicity of the product, primary
amine, is higher than that of the nitrile, primary amines react
readily with the “surface aldimine”, leading to the formation of
alkylideneamines, which are transformed into secondary amines
additives have been reported recently in both homogeneous
9
and heterogeneous systems. These examples are of interest,
because of their wide functional group tolerance and reduction
of waste. Sabo-Etienne and co-workers reported the selective
hydrogenation of benzonitrile via an ortho-metalated imine
intermediate and observed that the N-benzylidenebenzylamine
(
Scheme 1). Reaction of the aldimine with a secondary amine
then affords a tertiary amine via formation of an enamine.
Because these side reactions decrease the selectivity of primary
amine formation, catalysis with high selectivity toward primary
amines is desirable. Addition of large quantities of ammonia is
commonly employed to prevent the undesirable condensation
8a
formed initially is transformed into benzylamine. This sug-
gests that liberated ammonia plays a crucial role in the selective
formation of benzylamine. A similar reaction pathway has been
8e
reported by Prechtl and co-workers.
2
In contrast to nitrile reduction at a metal center, metal−
ligand cooperation is effective in suppressing aldimine con-
densation. An outer-sphere mechanism of hydrogenation by
reaction. Protonation and acylation of primary amines, which
3
reduce their nucleophilicity, are also sometimes used. How-
ever, these procedures require deprotection steps, which make
the process more complicated and expensive.
4
Homogeneous catalysts can selectively produce primary
amines. For example, cooperative metal−ligand systems promote
Received: March 22, 2018
5,6
©
XXXX American Chemical Society
A
DOI: 10.1021/acs.organomet.8b00165
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
2
Scheme 1. Plausible Mechanistic Model of Nitrile
Hydrogenation on a Metal Surface
monometallic complexes with an η -RCN ligand have been
19
synthesized. Trinuclear complexes with a π-coordinated nitrile
ligand also have been prepared. These complexes are suitable
models for the parallel adsorption of nitrile on a metal sur-
face. Kaesz et al. synthesized the first triply bridged π-nitrile
complex, Fe {μ -η (N):η (C,N):η (C,N)-RCN}(CO) , by oxi-
20
1
2
2
3
3
9
2
2
dation of a μ -η (∥)-imidoyl complex, Fe (μ-H){μ -η (∥)-RCNH}-
3
3
3
(
CO) , and elucidated its interconversion via hydrogenation
9
20a
and dehydrogenation (Scheme 2). They also demonstrated
Scheme 2. Sequential Transformation of Benzonitrile on a
2
0a
Triiron Cluster As Elucidated by Kaesz and Co-workers
i
[
{HN(CH P Pr ) }Fe(H)(BH )(CO)] has been proposed on
2
2 2
4
8c,d,10
the basis of experimental and theoretical studies.
Two distinctive pathways, the “M−N route” and the “M−C
1a
2
route”, have been proposed in heterogeneous catalysis. Krupka
et al. suggested that the dramatic difference in selectivity among
various catalysts is derived from the preferred adsorption mode
the sequential transformation of the μ -η (∥)-imidoyl complex
3
2
into μ -η -alkylideneamido and μ -imido complexes, Fe (μ-H){μ -
3
3
3
3
2
η (C,N)-NCRH}(CO) and Fe (μ-H) (μ -NCH R)(CO) ,
9
3
2
3
2
9
11
of the surface-bound reaction intermediate. The surface nitrene
in the “M−N route” is selective toward primary amines, whereas
the surface aminiocarbene or the π-aldimine intermediate in the
respectively. Although elimination of primary amines was not
observed, these transformations have clarified the mechanism of
nitrile hydrogenation on a metal surface.
“
M−C route” allows condensation with amines and leads to the
Keller et al. reported that the same triiron complex can be
prepared by reaction of Fe (CO) with nitrile under an atmo-
formation of higher amines. Tada and co-workers reported that
Ru nanoparticles supported by K-doped Al O catalyze the
selective hydrogenation of nitriles into the corresponding primary
amines without additives and proposed that the selective for-
mation of a nitrene-like intermediate is responsible for the high
selectivity on the basis of density functional theory (DFT) cal-
3
12
20c
sphere of hydrogen. In contrast, the corresponding triruthe-
nium and triosmium complexes were not obtained from the same
procedure. Reaction of Ru (CO) with nitriles under H afforded
2
3
3
12
2
21
a μ-alkylideneamido complex, Ru (μ-H)(μ-NCRH)(CO) ,
3
10
whereas a mixture of μ-alkylideneamido and μ-amido complexes,
Os (μ-H)(μ-NCRH)(CO) and Os (μ-H)(μ-NHCH R)-
(CO) , was obtained from Os (CO) . Cabeza and co-workers
10 3 12
9
f
culation.
3
10
3
2
22
Because the identification of surface intermediates is crucial
in the mechanism evaluation, nitrile adsorption modes have
been investigated extensively by near-edge X-ray fine structure
2
synthesized the first μ -η -alkylideneamido ruthenium complex
3
by thermolysis of the μ-alkylideneamido complex formed by
reaction of Ru (CO) with LiNCPh followed by protona-
12
(
NEXAFS) spectroscopy,1 high-resolution electron energy loss
3
12
2
3
23
spectroscopy (HREELS), reflection−absorption infrared spec-
troscopy (RAIRS),
infrared (ATR-IR) spectroscopy. These studies demonstrated
that nitriles sometimes adsorb through π interaction with the
CN bond parallel or slightly inclined to the metal surface
rather than by common κ(N) coordination to a metal center.
A preference for parallel adsorption of HCN on Ni(111) and
Co(111) has been indicated theoretically by Oliva and
co-workers. In some cases, a surface intermediate with a par-
tially hydrogenated carbon−nitrogen bond is detected. These
species, which are proposed to be adsorbed imidoyl (RCNH)
units, are regarded as key intermediates in the “M−C route”.
However, their role in the catalytic cycle has yet to be verified
experimentally.
Although nitriles typically are coordinated to a metal center
through their nitrogen atom (κ(N) coordination), an electron-
rich metal center can stabilize π coordination of a nitrile. Several
tion. They showed that the μ-alkylideneamido ligand is
eliminated as an amine upon hydrogenation and proposed that
μ-alkylideneamido species are key intermediates in the hydrogena-
13d,14
and in situ attenuated total reflection
15
24
tion of nitriles.
We previously reported the synthesis of a triruthenium com-
plex having a triply bridging nitrile ligand, (Cp*Ru) (μ -H)-
3
3
16
2
2
5
25
(μ-H) {μ -η :η (⊥)-PhCN} (2) (Cp* = η -C Me ). Unlike
Kaesz’s triiron complex, the nitrile ligand in 2 is coordinated to
2
3
5
5
1
7
the Ru plane with its nitrogen atom directed away from the
3
18
Ru core. Søtofte and co-workers reported the same coordi-
3
26
nation mode of a nitrile ligand on a triniobium plane. This
orientation resembles that of the alkyne ligand in {Cp*Ru-
2
2
27
(μ-H)} {μ -η :η (⊥)-RCCR′}, and suggests that the nitrile
3
3
ligand acts as a four-electron donor in a manner similar to the
alkyne. It also suggests that the lone pair electrons at the nitro-
gen atom in 2 do not participate in coordination. Thus,
intramolecular hydrido migration occurs upon thermolysis to
B
DOI: 10.1021/acs.organomet.8b00165
Organometallics XXXX, XXX, XXX−XXX

Organometallics
yield a μ -imidoyl complex, (Cp*Ru) (μ-H) {μ -η :η (⊥)-
Article
2
2
3
3
2
3
Scheme 3. Switchback Motion of the (⊥)-Nitrile Ligand in 2
2
8
2
2
PhCNH} (3). The formation of μ -η :η (⊥)-nitrile and
on the Ru Plane
3
3
2
2
μ -η :η (⊥)-imidoyl complexes contrasts with the common reac-
3
tivity of hydrido clusters with nitriles, which leads to formation of
a bridging alkylideneamido ligand via insertion into an M−H
2
9
bond. Fryzuk et al. synthesized μ-alkylideneamido complexes
by reaction of {(dippe)Rh(μ-H)} (dippe = 1,2-diisopropyl-
2
phosphinoethane) with nitriles and the transformation into
29j
μ-amido complexes by treatment with pressurized hydrogen.
Because 2 and 3 adopt a 46-electron configuration, it is anti-
cipated that they will react readily with dihydrogen. Formation
2
2
ligand moves around the Ru plane by altering positions of
of the μ -η (∥)-imidoyl complex (Cp*Ru) (H) {μ -η (∥)-
3
3
3
4
3
the acetylenic carbons, the isomer containing the phenyl group
PhCNMe} was indicated upon reaction of the perpendicularly
31
2
2
on the inner carbon is much more stable than the other. The
population of the minor isomer, which has the phenyl group
coordinated μ -η :η (⊥)-imidoyl complex (Cp*Ru)(μ-H) -
3
2
2
2
2
{
μ -η :η (⊥)-PhCNMe} with H , although the μ -η (∥)-
imidoyl complex was not isolated due to facile dehydrogenation
3
2
3
oriented outside the Ru core, is negligible, because the energy
3
30
difference between the two isomers is large. This allows the
kinetic parameters of the motion to be evaluated by dynamic
NMR study. Dynamic NMR studies for 2 were not successful,
however, because 2 equilibrates with the κ(N) isomer 5, whose
that regenerated the perpendicular form. Herein, we report
on the consecutive transformations of a nitrile ligand on a Ru3
plane into a primary amine and the catalytic properties of
the triruthenium complexes {Cp*Ru(μ-H)} (μ -H) (1) and
3
3
2
μ -nitrile ligand is coordinated with its nitrogen atom directed
(
Cp*Ru) (μ -H)(μ-H) (μ-CO) (12) toward benzonitrile
3
3
3
2
20a
inside the Ru core as in Kaesz’s μ -nitrile complex. Because
hydrogenation. The reactions are suitable models for the “M−C”
and “M−N” routes, respectively, of nitrile hydrogenation on a
metal surface.
3
3
of the increased population of 5, NMR spectra could not
be simulated using only the rate of the switchback motion.
Interaction of the inner nitrogen atom with the unique Ru atom
efficiently stabilizes 5, unlike the case of the phenylacetylene
complex. 1
RESULTS AND DISCUSSION
■
Reaction of 1 with Benzonitrile. Reaction of the
triruthenium pentahydrido complex {Cp*Ru(μ-H)} (μ -H)
In the H NMR spectrum recorded at −50 °C, one set of
3
3
2
hydrido signals of 2 at δ −4.01 and −24.50 ppm and a second
(
(
1) with benzonitrile at 25 °C affords a mixture of the
2
set of hydrido signals of 5 at δ −15.08 (d, J
= 4.9 Hz) and
H−H
⊥)-nitrile complex 2 and μ -benzylimido complex 4 (eq 1).
3
2
−
25.61 (t, J
= 4.9 Hz) ppm with an intensity of 2:1 were
H−H
observed. The structure of 5 was not fully determined, because
its Cp* and phenyl signals were not identified owing to the low
population of the isomer (ca. 9%). However, spin saturation
transfer (SST) between the hydrido signals of 2 and 5 was
observed, which indicates that 5 equilibrates with 2. When the
signal of 2 at δ −4.01 was irradiated at −10 °C, the signal at
δ −25.61 disappeared and the intensity of the signal at δ −15.08
decreased by 44% (Figure S6, Supporting Information). Although
no structural data are available for 5, DFT calculations provide a
The time course of the reaction conducted with 10 equiv of
benzonitrile in a sealed NMR tube shown in Figure S1 in the
Supporting Information clearly illustrates the consecutive
formation of 2 and 4. In the initial stage of the reaction, the
plausible structure containing a μ -κ(N)-nitrile ligand (Figure S41,
3
Supporting Information) and show that 5 is less stable than 2
−1
by ca. 4 kJ mol , which agrees with the 5:2 ratio of 6:94
1
estimated from the H NMR spectrum recorded at −10 °C.
liberation of H from 1 is accompanied by the formation of 2,
2
Maintaining a THF solution of 2 at room temperature leads
which is detected at δ 4.46 ppm. This suggests that μ -benzylimido
2
3
to formation of the μ -η -benzylideneamido complex (Cp*Ru) -
3
3
complex 4 is formed from 2 by uptake of dihydrogen. In fact,
2
(
μ-H) (μ -η -NCPhH) (6), which is another isomer of 2
2
3
(
⊥)-nitrile complex 2 is obtained exclusively by reaction at
(
eq 2). The population of 6 gradually increases and stabilizes
reduced pressure. Treatment of 2 with 0.1 MPa of H at 25 °C
immediately affords 4 in quantitative yield.
2
1
The H NMR spectrum of 2 recorded at 20 °C displays two
broad Cp* signals, which sharpen with decreasing temperature
and resonate at δ 1.99 and 1.53 ppm with an intensity ratio of
2
:1 at −50 °C. The spectral change suggests that the μ -nitrile
3
ligand in 2 moves among the three Ru−Ru edges in a
switchback motion (Scheme 3). This motion averages the mag-
netic environments of the three Cp* groups. Accordingly, a
time-averaged spectrum displaying a single resonance should be
observed at elevated temperature. However, coalescence of the
after 3 h with an 2:6 ratio of 73:27 (Figure S7, Supporting
1
Information). The H NMR spectrum recorded at −50 °C dis-
Cp* signals is not observed due to isomerization to a μ -
plays sharp Cp* signals at δ 1.90, 1.68, and 1.57 ppm. Doublet
3
2
2
2
2
2
η :η (⊥)-imidoyl complex, (Cp*Ru) (μ-H) {μ -η :η (⊥)-
PhCNH} (3), above 40 °C (vide infra).
In the switchback motion of the μ -phenylacetylene ligand in
signals at δ −8.23 and −9.89 ppm (d, J
= 2.8 Hz), which
3
2
3
H−H
broaden with increasing temperature, are observed in the
hydrido region. The signal for the benzylideneamido proton,
which occurs as a sharp singlet at δ 8.06 ppm, lies within the
3
2
2
{
Cp*Ru(μ-H)} {μ -η :η (⊥)-PhCCH}, where the alkyne
3
3
C
DOI: 10.1021/acs.organomet.8b00165
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
range reported for μ-alkylideneamido ligands in di- and
2
1a,b,29l,m,32
triruthenium complexes (δ 6.5−9.1 ppm).
These
spectral features suggest that the μ -benzylideneamido complex
3
also is fluxional. However, this behavior will not be discussed
further, because of the low population of 6. Fluxional behavior
2
2
of the μ -η -benzylideneamido ligand in (Cp*Ru) (μ-H) (μ−η -
3
3
2
NCPhH)(μ-CO) (13) will be discussed later.
Because a single crystal of 6 suitable for X-ray diffraction
(
XRD) could not be obtained from the mixture, distinction
between μ and μ coordination modes of the benzylideneamido
3
1
ligand in 6 cannot be established on the basis of H NMR data
13
alone. However, μ coordination is strongly indicated by C NMR.
3
A doublet signal appears for the benzylideneamido carbon at
1
δ 98.4 ppm (d, J
= 170 Hz). Its chemical shift differs
C−H
considerably from that of μ-alkylideneamido ligands (δ 155−
2
0e,29j−m,o,33
179 ppm).
Cabeza and co-workers noted a similar
upfield shift of the alkylideneamido carbon resonance by ca.
2
3
3
1−37 ppm upon the μ coordination. The upfield shift is
Figure 1. Molecular structure and labeling scheme of 7 with thermal
ellipsoids at the 30% probability level. Selected bond lengths (Å) and
angles (deg): Ru(1)−Ru(2) 2.91365(18), Ru(1)−Ru(3) 2.87680(18),
Ru(2)−Ru(3) 2.73689(17), Ru(1)−N(1) 2.0497(13), Ru(1)−C(1)
2.2290(17), Ru(2)−N(1) 1.9806(13), Ru(2)−C(8) 2.1392(17),
Ru(3)−N(1) 1.9987(13), Ru(3)−C(8) 2.0186(17), N(1)−C(1)
3
2
likely due to the additional η coordination to the Ru center in 6.
The μ mode seems to be preferable according to the effective
atomic number (EAN) rule. The number of valence electrons in
3
2
6
is 46, although η -coordination of the CN bond is involved.
Both μ and μ coordination modes are known for
3
1
.351(2), N(2)−C(8) 1.227(2), N(2)−C(9) 1.478(2); Ru(2)−
alkylideneamido ligands on a trinuclear plane. The μ -alkylidenea-
mido structures have been reported for Kaesz’s triiron complex and
Although it was not confirmed in 6, μ3
coordination of a benzylideneamido ligand is unambiguously
identified in 7, which is obtained by reaction of 6 with BuNC
eq 3). Reaction of the mixture of 2 and 6 with BuNC at 25 °C
3
Ru(1)−Ru(3) 56.409(4), Ru(1)−Ru(2)−Ru(3) 61.116(4), Ru(1)−
Ru(3)−Ru(2) 62.475(4), Ru(1)−N(1)−Ru(2) 92.58(5), Ru(1)−
N(1)−Ru(3) 90.56(5), Ru(1)−N(1)−C(1) 78.99(9), Ru(2)−
N(1)−Ru(3) 86.91(5), Ru(2)−N(1)−C(1) 146.89(11), Ru(3)−
N(1)−C(1) 124.73(10), C(8)−N(2)−C(9) 134.68(16), N(1)−
C(1)−C(2) 124.34(14), Ru(2)−C(8)−Ru(3) 82.28(6), Ru(2)−
C(8)−N(2) 125.22(13), Ru(3)−C(8)−N(2) 152.47(14).
20e,34
related compounds.
t
t
(
The μ-isocyanido ligand in 7 bridges the Ru(2) and Ru(3)
atoms and adopts a bent structure (∠C(8)−N(2)−C(9)
1
34.68(16)°). Accordingly, the Ru(2)−Ru(3) bond
(
2.73689(17) Å) is noticeably shorter than the Ru(1)−Ru(2)
and Ru(1)−Ru(3) bonds (2.91365(18) and 2.87680(18) Å,
respectively).
t
Because 6 reacts with BuNC much more quickly than does
2
, it can be regarded as the more reactive isomer. Thus, it is
reasonable to assume that the hydrogenation of 2 occurs via its
isomerization to 6. When complex 6 takes up dihydrogen,
subsequent insertion of the benzylideneamido moiety into a
Ru−H bond leads to the formation of the μ -benzylimido com-
3
plex 4. As shown in Scheme 2, Kaesz and co-workers showed
2
affords the μ-isocyanido complex 7, which is isolated in 87%
that the triiron μ -benzylideneamido complex Fe (μ-H)(μ -η -
3
3
3
yield. Complex 6 reacts much more quickly than the μ -nitrile
NCPhH)(CO)9 was transformed into a μ -benzylimido
3
3
20a
complex 2. Thus, the formation of 7 is accompanied by the
immediate disappearance of 6 in the initial stage of the reaction,
with 2 being gradually consumed via its slow isomerization of
complex upon treatment with H (1.4 MPa) for 48 h.
2
Vahrenkamp and co-workers showed that the μ-benzylidenea-
mido complex Ru (μ-H)(μ-NCPhH)(CO) is hydrogen-
ated by 0.1 MPa H to yield the μ -benzylimido complex
2 3
3
10
2
to 6.
The XRD results in Figure 1 clearly establish μ coordination
Ru (μ-H) (μ -NCH Ph)(CO) with elimination of CO. How-
3 2 3 2 9
21a
3
of the benzylideneamido ligand in 7, where the Ru(1)−N(1)
2.0497(13) Å) and Ru(1)−C(1) (2.2290(17) Å) distances
ever, heating at 135 °C is required.
In contrast to the
(
preceding examples, 6 reacts smoothly with H to yield 4 at
2
2
describe the η coordination of the CN moiety to the Ru(1)
atom. The N(1)−C(1) distance (1.351(2) Å) is noticeably
greater than the N(2)−C(8) distance (1.227(2) Å) of the
μ-isocyanido ligand and comparable to the N−C distance in
ambient temperature, as shown in Figure S1. The unsaturated
nature of 6 may be responsible for its high reactivity toward H2.
Cp* and hydrido signals with an intensity ratio of 15:1 are
1
observed at δ 1.70 and −14.14 ppm, respectively, in the H NMR
2
34
Fe (μ -η -NCRH)(CO) (1.322−1.347 Å). The Ru(1)−
spectrum of 4. The methylene protons of the μ -benzylimido
3
3
9
3
C(1) distance also is comparable to those of the triiron
complex (2.186−2.223 Å). The C signal derived from the
μ -η -benzylideneamido carbon in 7 is observed at δ 106.9
group resonate as a singlet at δ 4.81 ppm. Formation of a
μ3-benzylimido ligand was confirmed by XRD, as shown in
Figure 2. Although three independent molecules with similar
structural parameters occur in the unit cell, only one molecule
is depicted. The three Ru−Ru distances are nearly equal
1
3
2
3
1
(
d, J
= 171 Hz), which is slightly downfield of the resonance
C−H
in 6 (δ 98.4).
D
DOI: 10.1021/acs.organomet.8b00165
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Figure 2. Molecular structure and labeling scheme of 4 with thermal
ellipsoids at the 30% probability level. Selected bond lengths (Å) and
angles (deg): Ru(1)−Ru(2) 2.7091(3), Ru(1)−Ru(3) 2.7003(3),
Ru(2)−Ru(3) 2.7160(3), Ru(1)−N(1) 2.0000(19), Ru(2)−N(1)
at 80 °C produces a mixture of μ -imidoyl complexes 3 and 8,
3
2
where μ -η (∥)-imidoyl complex 8 can be prepared by
3
hydrogenation of 3 at 25 °C (eq 5). As previously reported
2.000(2), Ru(3)−N(1) 2.016(2), N(1)−C(1) 1.457(3); Ru(2)−
Ru(1)−Ru(3) 60.276(7), Ru(1)−Ru(2)−Ru(3) 59.702(7), Ru(1)−
Ru(3)−Ru(2) 60.022(7), Ru(1)−N(1)−Ru(2) 85.25(8), Ru(1)−
N(1)−Ru(3) 84.49(7), Ru(1)−N(1)−C(1) 120.49(14), Ru(2)−
N(1)−Ru(3) 85.09(7), Ru(2)−N(1)−C(1) 133.42(16), Ru(3)−
N(1)−C(1) 131.55(17), N(1)−C(1)−C(2) 117.8(2).
(
Ru(1)−Ru(2) 2.7091(3) Å, Ru(1)−Ru(3) 2.7003(3) Å, and
2
Ru(2)−Ru(3) 2.7160(3) Å), and the μ -imido ligand lies above
for the analogous μ -η (∥)-imidoyl complex (Cp*Ru) -
3
3
3
2
30
the center of the Ru triangle. Thus, the three Ru−N distances
(μ-H) {μ -η (∥)-PhCNMe}, 8 cannot be isolated due to
facile elimination of H upon evacuation. Hydrogenation of 8
3
4
3
are almost equal (Ru(1)−N(1) 2.0000(19) Å, Ru(2)−N(1)
2
2
.000(2) Å, Ru(3)−N(1) 2.016(2) Å) and are comparable to
with 0.5 MPa of H at 80 °C results in the formation of 1.
2
2
those of the isostructural μ -phenylimido analogue (2.013(4)
Transformation of the μ -η (∥)-imidoyl ligand into a
3
3
35
and 2.0243(3) Å).
μ-aminocarbene has been proposed in the “M−C route” of nitrile
1a
Extrusion of Benzylamine from the Ru Site. We pre-
hydrogenation on a metal surface (Scheme 1). Migration of a
3
2
viously reported that treatment of the μ -imido complex
hydride to the nitrogen atom of the μ -η (∥)-imidoyl ligand
affords μ -aminocarbene intermediate A, as shown in Scheme 4.
Due to its unsaturated nature, A can accept dihydrogen to yield
μ-aminocarbene intermediate B. B is then transformed into
terminal isomer E, which is susceptible to nucleophilic attack by
benzylamine, yielding N-benzylidenebenzylamine. Adams et al.
have demonstrated that Os (μ-OMe)(μ-H)(CO) reacts with
3
3
{
Cp*Ru(μ-H)} (μ -NH) with 0.1 MPa of H generates the
3
3
2
3
pentahydrido complex 1 and eliminates NH via formation
3
3
6
of the μ-amido intermediate (Cp*Ru) (μ-H) (μ-NH ).
As shown in Scheme 2, Kaesz and co-workers also synthesized
a μ -imido complex upon hydrogenation of the triiron μ -nitrile
complex. However, the triiron μ -benzylimido complex did not
3
4
2
3
3
3
3
10
20a
react further with dihydrogen. In contrast to this observation,
CH (NMe ) to yield a mixture of terminal- and μ -amino-
2
2
2
3
the triruthenium complex Ru (μ-H) (μ -NCPh H)(CO) syn-
carbene complexes, Os (μ-OMe)(μ-H){C(NMe )H}(CO)
and Os (μ-OMe)(μ-H){μ -C(NMe )H}(CO) , by elimination
3 3 2 9
3
2
3
2
9
3
2
9
thesized by Cabeza and co-workers reacts with H to liberate
2
38
H NCPh H, but the triruthenium skeleton is not retained, as
of dimethylamine. They also showed that the μ -aminocarbene
2
2
3
24
Ru (μ-H) (CO) forms in the reaction. Unlike these coordi-
complex Os (μ-OMe)(μ-H){μ -C(NMe )Et}(CO) , which is
4
4
12
3
3
2
9
natively saturated carbonyl clusters, the μ -imido complex
obtained by reaction of Os (μ-H) (CO) (MeCN) with MeC
3
3
2
9
{
4
Cp*Ru(μ-H)} (μ -NH) adopts a coordinatively unsaturated
CNMe , is converted by thermolysis to a terminal complex,
3
3
2
39
6-electron configuration that presumably facilitates the uptake
Os (μ-OMe)(μ-H){C(NMe )H}(CO) . The need for
3
2
9
of dihydrogen and ammonia elimination.
pressurized H in the hydrogenation of 8 is rationalized as
2
Isostructural 4 is expected to react readily with hydrogen to
produce benzylamine. However, 4 does not react with 0.1 MPa
follows. Intermediate B is stabilized under an H atmosphere at
2
high pressure. Although A is unstable, the elevated H pressure
2
of H at 80 °C. Complex 1 is obtained eventually in quantitative
drives the equilibrium between 8 and A to A. Subsequent hydro-
genation leads to elimination of benzylamine and condensation to
aldimine intermediate F.
2
37
yield, but 0.5 MPa of H is required (eq 4). Benzylamine is
2
formed in this reaction, but considerable N-benzylidenebenzyl-
amine also is produced (Figure S20, Supporting Information).
Unlike {Cp*Ru(μ-H)} (μ -NH), μ -benzylimido complex 4 is
The reaction of 1 with benzylamine also was examined. The
reaction proceeds at 100 °C and produces a mixture of μ -imidoyl
3
3
3
3
not the true intermediate in benzylamine elimination, because
the secondary imine is formed and 4 is not smoothly hydro-
genated under H2.
complexes of 3 and 8. Formation of μ -benzylimido complex 4
3
is not observed in the early stages. Although it cannot be
excluded that the mixture of 3 and 8 is produced from a small
Although 6 is hydrogenated to 4 at ambient temperature, the
amount of initially formed 4, the fact suggests that benzylamine
2
reverse process, β-hydrogen elimination from the μ -benzylimido
is eliminated from the μ -η (∥)-imidoyl complex 8 rather than
3
3
ligand, is operative at an elevated temperature. Thermolysis of 4
the μ -benzylimido complex 4 on the basis of the principle of
3
E
DOI: 10.1021/acs.organomet.8b00165
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Scheme 4. Plausible Mechanism for Hydrogenation of the
μ -η (∥)-Imidoyl Complex 8
3
intensity of the methylene proton decreases by ca. 50%, and the
shape of the signal changes from a sharp singlet to a broad
triplet-like shape alongside incorporation of deuterium at the
CH site. The scrambling likely occurs within the amine adduct
D in Scheme 4. These results imply that the benzylic C−H
2
bond is readily broken at the Ru site and leads to the pref-
3
erential formation of the μ -imidoyl species. The presence of a
3
phenyl group, which stabilizes π coordination of the CN
2
bond, also is responsible for formation of the μ -η (∥) imidoyl
3
complex 8. In other words, the electron-rich Ru site com-
3
prising three Cp* groups has a propensity to form imidoyl species.
Catalytic Hydrogenation of Benzonitrile. Although the
“
M−C route” appears to be the preferred reaction mechanism,
hydrogenation of 4 cleanly regenerates 1. Aminocarbene inter-
mediate B has 46 valence electrons, and its hydrogenation is
accelerated at an elevated H pressure, which should increase
2
selectivity for benzylamine. Thus, we attempted the catalytic
hydrogenation of benzonitrile using 1 mol % of 1 under 5 MPa
of H . When the reaction is carried out at 100 °C, almost all
2
benzonitrile is consumed within 66 h (Table 1, entry 3). Although
selectivity for benzylamine is moderate at the beginning (entry
1
, 87%), it decreases to 48% in the later stages of reaction.
A similar decrease in primary amine selectivity in the late stages
of the reaction is often observed in nitrile hydrogenation using
heterogeneous catalysts. It is widely accepted that accumulation
of the primary amine promotes condensation, yielding the
9d
secondary imine.
Reaction of 1 under 1 MPa of H results in low benzonitrile
2
conversion and low selectivity (Table 1, entry 4). Although
hydrogenation proceeds sluggishly below 80 °C (entry 5), the
rate increases significantly above 100 °C. When the reaction is
carried out at 110 °C, benzonitrile is 93% consumed after 20 h
with 82% selectivity for benzylamine (Table 1, entry 6). How-
ever, the selectivity decreases to 48% at 130 °C (Table 1, entry 7).
Moderate selectivity for benzylamine therefore is achieved
microscopic reversibility. In addition, the reaction of deuterated
using H at high pressure. Because conversion of B to C is
2
pentahydride 1-d with benzylamine at 50 °C results in the
scrambling of deuterium into the CH and NH sites of
benzylamine (Figure S22, Supporting Information). The signal
promoted by elevated H pressure, the opportunity for con-
densation via formation of E decreases. The pathway involving
5
2
hydrogenation of the μ -benzylimido complex 4 also is possible
3
Table 1. Catalytic Hydrogenation of Benzonitrile by Ruthenium Clusters under 5 MPa of H2
yield (selectivity)/% ,
ab
entry
cat.
time/h
temp/°C
conversn/%
BA
BB
DB
1
2
1
1
1
1
1
1
1
1
1.5
20
20
66
20
20
20
20
20
20
100
100
100
100
100
80
6.8
56
98
98
18
3.9
93
98
24
98
34
5.9 (87)
0.9 (13)
0
40 (71)
16 (28)
0.3 (0.5)
c
3
27 (28)
47 (48)
12 (69)
19 (19)
40 (41)
52 (53)
d
4
11 (11)
5
6
7
8
9
5.6 (31)
0.7 (18)
15 (16)
0
0
3.2 (82)
110
130
110
120
110
76 (82)
47 (48)
trace
1.2 (1.3)
15 (16)
4.1 (17)
3.3 (3.2)
27 (80)
36 (37)
20 (83)
e
10
1
0
1
12
15
92 (93)
3.7 (3.8)
6.8 (20)
1
20
0
a
b
c
d
e
Determined by GLC analysis. Based on benzonitrile. Methanol was used as solvent. Reaction carried out under 1 MPa of H A mixture of 10
2
and unidentified complexes was prepared by reacting 3 with benzonitrile. The amount of 3 used in preparing the mixture is taken as the amount of
catalyst.
F
DOI: 10.1021/acs.organomet.8b00165
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
under this condition. However, formation of N-benzylidene-
benzylamine is not completely suppressed even under 5 MPa
of H2.
Remarkable solvent effects have been often reported, in
particular using a metal−ligand cooperation system, where a
protic solvent such as 2-propanol is effective for selective pri-
8c−f
mary amine formation.
In our system, however, protic sol-
vent (methanol) was not effective for the selective benzylamine
formation (Table 1, entry 3). In methanol, conversion of
benzonitrile is accelerated but it also causes acceleration of con-
densation leading to the formation of secondary amine. Con-
sequently, benzylamine selectivity decreases in comparison to
the reaction performed in THF.
2
2
Reactivity of μ -η :η (⊥)-Imidoyl Complex 3. We showed
3
previously that the μ -nitrile complex 2 begins to isomerize to
3
2
2
28
the μ -η :η (⊥)-imidoyl complex 3 above 60 °C (eq 6).
3
Figure 3. Molecular structure and labeling scheme of 9 with thermal
ellipsoids at the 30% probability level. Selected bond lengths (Å) and
angles (deg): Ru(1)−Ru(2) 2.9237(2), Ru(1)−Ru(3) 2.7741(2),
Ru(2)−Ru(3) 2.9650(2), Ru(1)−N(1) 2.0486(15), Ru(2)−C(1)
2
.1082(17), Ru(2)−C(8) 1.9000(19), Ru(3)−N(1) 2.0934(15),
Ru(3)−C(1) 2.0701(17), N(1)−C(1) 1.386(2), N(2)−C(8)
.177(2), N(2)−C(9) 1.462(2); Ru(2)−Ru(1)−Ru(3) 62.650(5),
1
Ru(1)−Ru(2)−Ru(3 56.204(5), Ru(1)−Ru(3)−Ru(2) 61.145(5),
Ru(1)−N(1)−Ru(3) 69.65(10), Ru(1)−N(1)−C(1) 116.12(12),
Ru(2)−C(1)−Ru(3) 90.40(7), Ru(2)−C(1)−N(1) 106.52(11),
N(1)−C(1)−C(2) 119.52(16), Ru(2)−C(8)−N(2) 174.82(17),
C(8)−N(2)−C(9) 150.3(2).
synthesis. Site exchange buries the hydrido signals in the
1
baseline of the H NMR spectrum at 25 °C, but two doublets
appear at δ −19.87 and −10.15 ppm at −80 °C. Two singlets at
δ 157.9 and 179.1 ppm are observed in the low-field region of
the 13C NMR spectrum of 9. They arise from the terminal
isocyanido and the imidoyl carbons but cannot be further
assigned.
Complex 3 is expected to react smoothly with two-electron
donors, because it has a 46-electron configuration. Imidoyl spe-
cies are regarded as key intermediates in the “M−C route”.
Thus, we examined the reactivity of 3 toward two-electron
donors to explore the basis of this functionality.
Complex 3 is expected to react with primary amines in a
t
As shown in eq 5, 3 reacts smoothly with dihydrogen at
manner similar to its reaction with BuNC. If the reaction
2
2
5 °C to form the μ -η (∥)-imidoyl complex 8. Complex 3 also
results in the formation of secondary imines, the complex may
be a promoter of side reactions. However, 3 does not react with
benzylamine (100 equiv) at 110 °C. After 20 h, the majority of
3
t
reacts rapidly with BuNC at 25 °C to yield the isocyanido
adduct 9 (eq 7). The orientation of the μ -imidoyl ligand
3
3
remains unchanged (8% conversion), and 8 is the identifiable
product. Although a small amount of N-benzylidenebenzyl-
amine formation (ca. 4% based on 3) is observed, the results
suggest that the reaction of 3 with benzylamine does not con-
tribute to secondary imine formation during catalysis. Thermolysis
of an equilibrated mixture of 2 and 6 in the presence of 100 equiv
of benzylamine also was examined. The result does not differ
from the thermolysis of 2 in the absence of amine, and only 3 is
formed in 94% yield. These facts indicate that complexes 2, 3,
and 6 are not involved in side reactions.
changes from perpendicular to parallel relative to its adjacent
t
Ru−Ru vector upon incorporation of BuNC. In contrast to the
Although 3 does not react with primary amines, it does with
benzonitrile. Reaction of 3 with benzonitrile at 70 °C yields a
t
μ-isocyanido complex 7, the BuNC group in 9 coordinates to
Ru in a terminal manner, as unambiguously described by XRD
mixture of several complexes, including (Cp*Ru) (μ-H)-
3
2
(
Figure 3).
(μ -NCH Ph){μ -η (∥)-PhCN} (10) (eq 8). Complex 10
was isolated in 26% yield by column chromatography on
3
2
3
The μ -imidoyl group in 9 is σ-bonded to the Ru(1) and
3
Ru(2) atoms via its N(1) and C(1) atoms, respectively, and
π-bonded to the Ru(3) atom. The N(1)−C(1) distance
(
1.386(2) Å) is comparable to the reported range of N−C
2
distances in triruthenium μ -η (∥)-imidoyl complexes (1.319−
1
complex 3 (1.391(6) Å). The BuNC group is coordinated to
the Ru(2) atom, where the imidoyl C(1) atom is attached.
Hydrides H(1) and H(2) were successfully located at the
Ru(1)−Ru(2) and Ru(1)−Ru(3) edges during the Fourier
3
4
0
2
2
.381 Å) and to the N−C distance of μ -η :η (⊥)-imidoyl
3
28
t
G
DOI: 10.1021/acs.organomet.8b00165
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
alumina. Because benzonitrile is abundant under the catalytic
conditions, incorporation of second nitrile into the triruthe-
nium core of 3 is possible.
1
In the H NMR spectrum of 10 at 20 °C, two Cp* units
appear as a broad signal near 1.5 ppm, while the third Cp* yields
a sharp signal at δ 1.45 ppm. The broad signal resolves into two
signals resonating at δ 1.30 and 1.58 ppm at −80 °C. The
methylene protons of the μ -benzylimido ligand, which are equiv-
3
alent at 20 °C, separate into doublets at δ 6.21 and 6.42 ppm
The 13C NMR chemical shift of the imidoyl carbon of 11
2
at −80 °C with a large geminal coupling constant ( J
=
H−H
(
δ 179.2) is similar to that of 10. Three Cp* signals, two of which
1
7.5 Hz). These spectral changes were successfully simulated
⧧
are broad, occur at δ 1.34 (br), 1.51 (s), and 1.84 (br) ppm in
and produced estimated activation parameters of ΔH = 11.2 ±
1
.2 kcal mol−1 and ΔS = −7.77 ± 0.8 cal mol K . Details
⧧
−1 −1
the H NMR spectrum at 25 °C. This suggests a windshield
0
2
1
wiper motion of the μ -η (∥)-imidoyl ligand in 11, but the rate
of the variable-temperature H NMR study of 10 are shown
in Figures S26 and S27 in the Supporting Information. The
spectral changes originate from a single process that involves a
windshield wiper motion of the nitrile ligand in concert with
hydride migration, as shown in Scheme 5. Although we could
3
is much less than in 10.
The molecular structure of 11 determined by XRD is shown
in Figure 4. Each face of the Ru plane is capped by a
3
2
Scheme 5. Windshield Wiper Motion of the μ -η (∥)-Nitrile
3
Ligand in 10
not determine whether the nitrile ligand pivots on carbon or
the nitrogen atom, a pivoting motion on carbon by related
2
μ -η (∥)-imidoyl ligands has been proposed to be lower in
Figure 4. Molecular structure and labeling scheme of the cationic part
of 11 with thermal ellipsoids at the 30% probability level. The anionic
part was omitted for clarity. Selected bond lengths (Å) and angles
3
4
0a,c,41
energy.
Although the position of the hydride was not determined, its
presence was confirmed unambiguously by observation of a
singlet resonating at δ −27.80 ppm. We tentatively locate the
hydride at the Ru−Ru edge where the bridging nitrile ligand is
(
deg): Ru(1)−Ru(2) 2.8305(4), Ru(1)−Ru(3) 2.7383(4), Ru(2)−
Ru(3) 2.7529(5), Ru(1)−N(1) 2.012(3), Ru(1)−N(2) 2.089(5),
Ru(1)−C(8) 2.131(5), Ru(2)−N(1) 1.995(4), Ru(2)−C(8)
2
.068(4), Ru(3)−N(1) 2.034(4), Ru(3)−N(2) 2.029(5), N(1)−
bound, analogous to the position of the hydride in the cationic
C(1) 1.457(6), N(2)−C(8) 1.374(7); Ru(2)−Ru(1)−Ru(3)
5
2
μ -η (∥)-imidoyl complex 11, obtained by protonation of 10.
3
9.227(11), Ru(1)−Ru(2)−Ru(3) 58.718(11), Ru(1)−Ru(3)−
It is well-known among isoelectronic trinuclear alkyne com-
plexes that perpendicular coordination is typical for coor-
dinatively unsaturated 46-electron complexes, whereas parallel
coordination of an alkyne ligand is typical for coordinatively
Ru(2) 62.055(11), Ru(1)−N(1)−Ru(2) 89.90(14), Ru(1)−N(1)−
Ru(3) 85.20(13), Ru(2)−N(1)−Ru(3) 86.21(15), Ru(1)−N(2)−
Ru(3) 83.33(19), Ru(1)−C(8)−Ru(2) 84.76(16), N(2)−C(8)−C(9)
119.4(4).
42
saturated 48-electron complexes. According to the EAN rule,
the number of valence electrons in 10 is 48 when the nitrile
ligand acts as a four-electron donor. Therefore, the nitrile ligand
in 10 is proposed to adopt the parallel mode shown in eq 8.
The nitrile signal lies at δ 182.2 ppm in the C NMR spectrum,
which is at much lower field than the carbon signal of the
2
μ -benzylimido or μ -η (∥)-imidoyl group. The Ru−Ru
3
3
distances are 2.7383(4), 2.7529(5), and 2.8305(4) Å. The
Ru(1)−Ru(2) bond length is the greatest, likely because of
1
3
2
steric repulsion between the phenyl group of the μ -η (∥)-
3
imidoyl moiety and the Cp* groups at Ru(1) and Ru(2). The
25
(
⊥)-nitrile ligand in 2 (δ 97.8).
phenyl group of the μ -imido group is directed toward the
3
2
Although μ -η (∥) coordination of nitrile is unprecedented in
midpoint of the Ru(1)−Ru(2) bond.
3
2
2
trinuclear complexes, a similar μ -η (∥) coordination involving
Although the hydrogen atom of the μ -η (∥)-imidoyl moiety
4
3
additional κ(N) coordination to a metal center is known for a
was not located, the hydride is positioned at the Ru(2)−Ru(3)
edge. The μ-hydride likely moves to the Ru(1)−Ru(3) edge on
the NMR time scale in concert with the windshield wiper
motion of the imidoyl ligand to produce a time-averaged Cs
spectrum. The imidoyl group is π-bonded to the Ru(1) atom
and σ-bonded to Ru(2) and Ru(3) through the carbon and
nitrogen atoms, respectively. Due to structural disorder, precise
evaluation of the bond distances and angles involving the
2
0c,43
2
few clusters.
Despite the inability to establish μ -η (∥)-
3
coordination of PhCN by XRD, this bonding motif is suggested
2
by the structure of the cationic μ -η (∥)-imidoyl complex 11.
3
Treatment of 10 with HBF ·Et O results in quantitative for-
4
2
mation of 11 (eq 9). Protonation likely occurs at the nitrogen
2
atom of the μ -η (∥)-nitrile ligand in a manner similar to the
protonation of 2 that yields a cationic (⊥)-imidoyl complex.
3
25
H
DOI: 10.1021/acs.organomet.8b00165
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
imidoyl group is hampered, but the N(2)−C(8) distance of
electron configuration of 12 is the same as that of 1 (44e), 12 is
expected to react with nitriles in a similar manner. Furthermore,
the CO ligand on 12 is not readily eliminated from the Ru3
core during the reaction. Because CO decreases electron den-
1
.374(7) Å is comparable to that found in 9.
2
2
As mentioned above, μ -η :η (⊥)-imidoyl complex 3 reacts
3
with dihydrogen to regenerate the starting hydrido complex 1
2
via formation of the μ -η (∥)-imidoyl complex 8. In contrast,
sity at the Ru core, its presence destabilizes the π coordination
3
3
regeneration of 1 is not observed upon treating the mixture
obtained by reacting 3 with benzonitrile. Only a complicated
mixture of unidentified complexes results. Thus, formation of
of a CN bond and suppresses the uptake of the second nitrile
molecule. In fact, when the reaction is conducted using 1 mol %
of 12 under 5 MPa of H at 120 °C, benzylamine is obtained
2
1
0 may be related to the deactivation process.
When hydrogenation of benzonitrile is carried out in the
in 92% yield (Table 1, entry 10). Although small amounts of
N-benzylidenebenzylamine (3.7%) and dibenzylamine (3.3%)
are produced, primary amine selectivity is improved remarkably
in comparison to the reaction catalyzed by 1.
presence of the mixture obtained by reacting 3 with benzonitrile,
N-benzylidenebenzylamine is produced in 20% yield with
83% selectivity (Table 1, entry 9). Although a small amount of
The stoichiometric reaction of 12 with benzonitrile also
dibenzylamine (4%) is formed in the reaction, only a trace
amount of benzylamine is detected. The mechanism of secondary
imine formation is not clear at present, but the results suggest that
species containing two (or more) benzonitrile units promote the
formation of secondary imines. Therefore, suppressing the incor-
was examined. Complex 12 reacts readily with benzonitrile at
2
5 °C and exclusively affords a μ -benzylideneamido complex,
3
2
(
Cp*Ru) (μ-H) (μ -η -NCPhH)(μ-CO) (13) (eq 10).
3
2
3
poration of additional nitriles onto the Ru core seems to be
3
crucial for the selective formation of primary amines.
The series of transformations of benzonitrile at the Ru3
center derived from 1 is summarized in Scheme 6. Benzonitrile
Scheme 6. Transformations of Benzonitrile at the Ru3
Center Derived from 1
In contrast to the reaction of 1, benzonitrile inserts into a
2
Ru−H bond to form a μ -η -benzylideneamido ligand without
3
dihydrogen elimination. Suppression of H elimination from
2
the Ru core upon replacement of two hydrides with a CO
3
ligand also is observed in the reaction of 12 with butadiene and
44
ethylene.
Complex 13 was isolated in 56% yield by recrystallization. Its
molecular structure determined by XRD is shown in Figure 5.
There are two independent molecules with similar structural
features in the unit cell, but the structure of only one is
is hydrogenated to the μ -imido complex 4 at the Ru site by
3
3
formation of the π-coordinated nitrile complex 2. This process
corresponds to the “M−N route”, but the Ru unit derived
3
from 1 exhibits a tendency to form μ -imidoyl complexes 3 and
3
2
8
at elevated temperature. The μ -η (∥)-imidoyl complex 8
3
apparently promotes secondary imine formation via formation
of an aminocarbene intermediate, which is characteristic of the
“
M−C route” on a metal surface. The coordinatively unsat-
Figure 5. Molecular structure and labeling scheme of 13 with thermal
ellipsoids at the 30% probability level. Selected bond lengths (Å) and
angles (deg): Ru(1)−Ru(2) 2.8444(6), Ru(1)−Ru(3) 2.8747(6),
Ru(2)−Ru(3) 2.7123(6), Ru(1)−N(1) 2.062(4), Ru(1)−C(1)
urated (⊥)-imidoyl complex 3 binds another nitrile molecule
to yield 10. Under an atmosphere of H , 10 decomposes
to unidentified complexes, which catalyze the formation of
N-benzylidenebenzylamine. These results clearly show, as noted
by Krupka et al., that suppression of imidoyl species is very
important for the selective hydrogenation of benzonitrile.
Reactivity of μ-CO Complex 12. We recently reported the
synthesis of a triruthenium hydrido complex containing a μ-CO
ligand, (Cp*Ru) (μ -H)(μ-H) (μ-CO) (12). We have shown
that 12 reacts with phenylacetylene in a manner similar to that
2
2
.263(6), Ru(2)−N(1) 2.003(4), Ru(2)−C(8) 1.978(6), Ru(3)−
1
1
N(1) 2.011(4), Ru(3)−C(8) 2.083(5), N(1)−C(1) 1.329(7), C(8)−
O(1) 1.192(6); Ru(2)−Ru(1)−Ru(3) 56.618(14), Ru(1)−Ru(2)−
Ru(3) 62.253(16), Ru(1)−Ru(3)−Ru(2) 61.129(15), Ru(1)−N(1)−
Ru(2) 88.81(17), Ru(1)−N(1)−Ru(3) 89.77(17), Ru(1)−N(1)−
C(1) 80.6(3), Ru(2)−N(1)−Ru(3) 85.02(16), Ru(2)−N(1)−C(1)
126.9(4), Ru(3)−N(1)−C(1) 146.0(4), N(1)−C(1)−C(2) 126.0(5),
Ru(2)−C(8)−Ru(3) 83.8(2), Ru(2)−C(8)−O(1) 142.0(4), Ru(3)−
C(8)−O(1) 134.2(4).
44
3
3
2
2
2
for 1, yielding a μ -η :η (⊥)-alkyne complex. Because the
3
I
DOI: 10.1021/acs.organomet.8b00165
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
depicted. The benzylideneamido ligand lies on the Ru plane in
hydrides and the alkylideneamido proton at 50 °C suggests that
3
2
a μ -η fashion with the N(1)C(1) moiety π-bonded to the
a μ -imido intermediate formed by migration of a hydride to
3
3
Ru(1) atom and σ-bonded to the Ru(2) and Ru(3) atoms
through the N(1) atom. The N(1)C(1) distance (1.329(7) Å)
is comparable to that of the isostructural μ-isocyanido complex 7
the β-carbon atom is not involved in this fluxional process.
Thus, dynamic behavior via a vertically oriented μ -alkylidenea-
3
mido intermediate likely occurs in 13.
(1.351(2) Å).
Although the features of the isostructural μ -isocyanido
3
The CO ligand bridges the Ru(2)−Ru(3) edge on the face of
complex 7 are similar to those of 13, complex 7 is not fluxional
on the NMR time scale. The π-acidity of CO dominates the
σ-donating property of isocyanide and reduces back-donation
the Ru plane opposite to the μ -benzylideneamido ligand. The
3
3
presence of the μ-CO ligand also was confirmed by IR
−1
spectroscopy, which showed strong absorption at 1735 cm .
The bridging hydrides were successfully located on the Ru(1)−
Ru(2) and Ru(1)−Ru(3) edges.
from the Ru
-benzylideneamido ligand in 13 fluxional. The difference in
mobility of the μ -benzylideneamido ligand between 13 and 7 is
reflected in the difference in their reactivity toward H
3
core to the π*(CN) orbital, which makes the
μ
3
3
Complex 13 shows dynamic behavior arising from rotation of
.
2
the μ -benzylideneamido ligand (Scheme 7a). Two Cp* signals
Despite its saturated electron configuration, 13 reacts slowly
3
with 0.6 MPa of H at 25 °C to yield the μ-benzylamido
2
2
complex 14 in 70% yield (eq 11). Complex 14 was purified by
Scheme 7. Possible Fluxional Behaviors of the μ -η (∥)-
3
Benzylideneamido Ligand in 13
column chromatography and isolated in 45% yield. Prior to
uptake of H , a vacant site must be created by migration of a
2
hydride to the alkylideneamido carbon or decoordination of the
μ3-benzylideneamido ligand to form a μ-benzylideneamido
intermediate. The isostructural μ-isocyanido complex 7 does
not react with dihydrogen. Only 4% of 7 is converted upon
treatment with 0.6 MPa of H at 25 °C for 50 h to afford small
2
amounts of unidentified complexes. This result implies that the
mobility of the μ -benzylideneamido ligand on the Ru plane is
crucial for uptake of dihydrogen.
3
3
are observed at δ 1.57 (s) and 1.82 (br) ppm with an intensity
ratio of 1:2 by H NMR at 25 °C. The broad signal at δ 1.82 ppm
1
A broad signal from the amido proton is observed at δ 3.74 ppm
1
separates into sharp signals at δ 1.79 and 1.83 ppm at −30 °C.
Sharp hydrido signals are observed at δ −21.22 and −19.47 ppm
at −30 °C and become broad at higher temperatures. In con-
trast, the alkylideneamido proton signal (δ 6.56 ppm) is sharp
from −30 to 50 °C. These spectral changes were successfully
in the H NMR spectrum of 14, whereas an adsorption
assignable to ν(NH) is not observed in the IR spectrum.
1
A triplet resonating at δ 93.1 ppm ( J
= 134 Hz) in the
C−H
1
3
C NMR spectrum also indicates that the CN moiety of 13
is hydrogenated. Formation of a μ-benzylamido ligand with the
⧧
simulated to provide estimated activation parameters of ΔH =
benzyl group oriented inside the Ru core was confirmed by a
3
−
1
⧧
−1
−1
48
1
6.5 ± 0.3 kcal mol and ΔS = 4.0 ± 0.9 cal mol K .
preliminary XRD study (Figure S42, Supporting Information).
1
Details of the variable-temperature H NMR study of 13 are
Formation of the μ-benzylamido ligand is accompanied by a
shown in Figures S33 and S34 of the Supporting Information.
change from μ to μ coordination of the CO group, which is
confirmed by a strong adsorption at 1665 cm in the IR
3
2
−1
Although the dynamics of the μ -η -alkylideneamido ligand
3
have never been documented, the related μ -alkenylidene
motion in several trinuclear complexes has been discussed.
spectrum.
3
Hydrogenation of 14 to 12 occurs at 100 °C (eq 12).
Complex 14 is 64% converted after 24 h, and 12 is regenerated
in 47% yield. These reactions are accompanied by formation of
13 in 11% yield as a consequence of dehydrogenation despite
We documented the pivoting motion of a μ -vinylidene ligand
3
2
2
in (Cp*Ru) (μ-H)(μ−η -CCH ){μ−η (∥)-HCCH} to pro-
3
2
4
5
duce a time-averaged plane of C symmetry. Pivoting of the
s
μ3-alkylideneamido ligand may explain the observed spectral
changes (Scheme 7b). However, this is unlikely for 13, because
the motion requires simultaneous migration of the μ-CO and
μ-hydride ligands between two Ru−Ru edges.
the presence of 0.6 MPa of H . Thermolysis of 14 at 80 °C in
2
the absence of H results in the regeneration of 13 in 97% yield.
2
This fact indicates that coordinatively saturated 13 equilibrates
with unsaturated 14 via the release and uptake of H and that
2
Dynamic movement of a μ -vinylidene ligand via a vertical
the equilibrium is shifted largely toward 13. Owing to this
3
vinylidene intermediate has been described by Deeming and
equilibrium, high-pressure H seems to be required for the
hydrogenation of benzonitrile using 12.
A plausible mechanism for the hydrogenation of benzonitrile
by 12 is proposed on the basis of the reactions in Scheme 8.
2
co-workers in explanation of the interconversion of two isomers
2
46
of Os (μ-H) {μ -η -CC(OEt)H}(CO) . A similar mecha-
3
2
3
9
nism has been proposed for the enantiomerization of [Co {μ -
3
3
2
+
47
η -CC(CHMe )H}(CO) ] . The motion produces equiv-
Although unsaturated μ -benzylimido intermediate A is not
2
9
3
alent environments for the two Cp* groups in 13 without
migration of the μ-CO ligand. The lack of SST between the
observed, hydrogenation catalyzed by 12 is considered to
proceed via the “M−N route”. Complex 13, unlike 6, is not
J
DOI: 10.1021/acs.organomet.8b00165
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
2
2
2
(
1) and have identified μ -η :η (⊥)-nitrile complex 2, μ -η -
3
3
2
benzylideneamine complex 6, μ -η (∥)-imidoyl complex 8, and
μ -imido complex 4 as key intermediates. The transformations
3
3
of 2 to 6 and 4 correspond to the “M−N route” proposed for
nitrile hydrogenation on a metal surface. However, benzylamine
is not directly eliminated from 4. The true intermediate in the
2
extrusion of benzylamine is the μ -η (∥)-imidoyl complex 8.
3
This step, which follows the model of the “M−C route”, affords
considerable N-benzylidenebenzylamine in the catalytic hydro-
genation of benzonitrile. An aminocarbene intermediate formed
from 8 seems to be responsible for secondary imine formation.
2
2
The reactivity of coordinatively unsaturated μ -η :η (⊥)-
3
imidoyl complex 3 also was investigated. Complex 3 reacts
readily with H and BuNC but not with benzylamine, although
t
2
transformed into a μ-imidoyl species. It is presumed that
π-coordination of the CN bond is destabilized by the reduced
a mechanism involving nucleophilic attack of a primary amine
at the imidoyl carbon has been proposed for imine formation.
2
electron density at the Ru site arising from CO ligation.
Complex 3 binds benzonitrile to yield the μ
3
-imido−μ
3
-η (∥)-
3
Complex 13 also is obtained by reaction of 12 with benzylamine,
most likely by the formation of 14. The sum of these results
supports the mechanism shown in Scheme 8.
nitrile complex 10. The mixture obtained by reacting 3 with
benzonitrile catalyzes secondary imine formation. The mech-
anism of secondary imine formation by the mixture cannot be
established at present, but it is suggested that species formed by
the incorporation of two or more benzonitrile molecules, such
as 10, may be responsible for secondary imine formation.
Scheme 8. Plausible Mechanism for the Hydrogenation of
Benzonitrile Catalyzed by 12
In contrast to 1, the μ-carbonyl−hydrido complex (Cp*Ru) (μ-
3
H) (μ -H)(μ-CO) (12) reacts with one molecule of benzonitrile
2
3
to catalyze highly selective hydrogenation of benzonitrile with-
out additives. The reaction of 12 with benzonitrile affords
μ3-benzylideneamido complex 13, which upon subsequent hydro-
genation yields benzylamine via formation of μ-benzylamido
complex 14. The most important feature of 12 is that a
μ -imidoyl species was not formed on the Ru core, probably
3
3
due to reduced electron density within the Ru unit. Thus,
3
hydrogenation of benzonitrile proceeds preferentially via the
“
M−N route”. Although 13 is coordinatively saturated, it does
react with H to yield μ-amido complex 14. This reactivity is
2
attributed to the flexible nature of the μ -benzylideneamido
3
ligand in 13. In contrast, the isostructural μ-isocyanido complex
7
does not react with H . This is likely due to the π-acidity of
2
the CO ligand on Ru . The results show that subtle differences
3
acting at the Ru core can produce marked differences in
3
reactivity.
Degradation of the triruthenium skeleton occurs at elevated
temperatures and forms {Cp*Ru(μ-H)} (μ-CO) (15). Com-
2
plex 15 catalyzes secondary imine formation. Suppressing
degradation of the trinuclear skeleton is crucial for increasing
the selectivity of primary amine formation. We will continue to
optimize reaction conditions to minimize degradation and seek
a robust trinuclear skeleton that can catalyze more efficient
nitrile hydrogenation.
A small amount of {Cp*Ru(μ-H)} (μ-CO) (15) (6%) formed
2
1
during the hydrogenation of 14 is confirmed by H NMR
49
signals (δ 1.79 (Cp*) and −12.88 ppm (RuH)) in C D .
6
6
Complex 15 is produced by thermolysis of 13 at 110 °C under
an Ar atmosphere. Complex 13 decomposes slightly upon heating
at 110 °C in THF-d (ca. 6% conversion after 20 h), affording
EXPERIMENTAL SECTION
8
■
1
5 in ca. 2% yield. Although the rate is not significant, degrada-
General Procedures. All compounds were manipulated using
standard Schlenk and high-vacuum-line techniques under an
atmosphere of argon. Dehydrated toluene, tetrahydrofuran (THF),
pentane, hexane, and methanol used in this study were purchased from
Kanto Chemicals and stored under an atmosphere of argon. Diethyl
ether was dried over sodium−benzophenone ketyl and distilled under
tion of the trinuclear skeleton occurs at elevated temperature.
We also examined the catalytic activity of 15 toward hydrogena-
tion of benzonitrile (Table 1, entry 11). Although the reactions
are fairly slow, formation of N-benzylidenebenzylamine (6.8%)
and dibenzylamine (27%) is observed. The result indicates that
secondary imine is formed from complex 15 that is generated
by the decomposition of 13.
an atmosphere of argon. C D and THF-d were dried over sodium−
6
6
8
benzophenone ketyl and stored under an atmosphere of argon.
Acetone-d was dried over MS-4A and stored under an atmosphere of
6
argon. Benzonitrile was dried over P O5 and stored under an
atmosphere of argon. Other reagents were used as received.
2
CONCLUSION
In summary, we have accomplished the consecutive hydrogenation
of benzonitrile to benzylamine using {Cp*Ru(μ-H)} (μ -H)
■
Triruthenium complexes {Cp*Ru(μ-H)} (μ -H) (1), {Cp*Ru(μ-
3
3
2
5
0
44
D)} (μ -D) (1-d5), and (Cp*Ru) (μ -H)(μ-H) (μ-CO) (12)
3 3 2 3 3 2
3
3
2
K
DOI: 10.1021/acs.organomet.8b00165
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
were prepared according to published methods. H and 13C NMR
1
performed using the Gaussian 09 software suite. The molecular
54
55
spectra were recorded on Varian INOVA-400 and Varian 400MR
spectrometers. H NMR spectra were referenced to tetramethylsilane
structures were drawn using the GaussView version 5.0 program.
1
Frequency calculations at the same level of theory as geometry opti-
mizations were performed on the optimized structure to ensure that
minima exhibit only positive frequency. Information on the atom
coordinates for the optimized structure is given in the Supporting
Information.
Reaction of {Cp*Ru(μ-H)} (μ -H) (1) with Benzonitrile in a
Sealed NMR Tube. An NMR tube equipped with a Teflon valve was
charged with 1 (8.9 mg, 12.5 μmol), C D (0.5 mL), and cyclooctane
as an internal standard. 13C NMR spectra were referenced to the
natural-abundance carbon signal of the solvent employed. IR spectra
were recorded on a Nicolet Avatar 360 FT-IR instrument with a
Ge-ATR cell and a JASCO FT/IR-4200 spectrophotometer by the diffuse
reflection method using KBr. Elemental analyses were performed on a
PerkinElmer 2400II CHN analyzer. GLC analyses were performed on
Shimadzu GC-17A and GC-2010 instruments using a capillary column
3
3
2
6 6
(
J&W DB-1; 30 m × 0.53 mm × 1.50 μm) with helium gas as a carrier.
X-ray Diffraction Studies. Single crystals of 4, 7, 9, 11, and 13 for
an XRD analysis were obtained directly from the preparations
described below and mounted on nylon Cryoloops with Paratone-N
Hampton Research corp.). Diffraction experiments were performed
(1.0 μL) as an internal standard. Then, 10 equiv of benzonitrile (12.8 μL,
1
0.124 mmol) was added to the solution at 25 °C. The H NMR spectra
of the solution were periodically recorded, and the populations of 1, 2,
and 4 were measured.
2
2
(
Preparation of (Cp*Ru) (μ -H)(μ-H) {μ -η :η (⊥)-PhCN} (2).
3 3 2 3
on a Rigaku R-AXIS RAPID imaging plate diffractometer with
graphite-monochromated Mo Kα radiation (λ = 0.71069 Å). In all
samples, cell refinement and data reduction were performed using the
PROCESS-AUTO program. Intensity data were corrected for Lorentz−
polarization effects and for numerical or empirical absorption. The
structure of 4 was solved by the Patterson method, and those of 7, 9,
A 50 mL Schlenk tube was charged with THF (15 mL) and 1
(299 mg, 0.42 mmol). Benzonitrile (0.85 mL, 8.3 mmol) was added
with vigorous stirring. Then, the pressure in the flask was reduced
using the vacuum line to remove the dihydrogen produced in the reac-
tion. The reaction mixture was stirred at 25 °C for 6 h. Dihydrogen
was periodically removed from the reaction flask every 30 min during
the reaction. The solution turned from reddish brown to brownish
yellow. The solvent and the remaining benzonitrile were removed
under reduced pressure. The residual solid was purified by the use of
column chromatography on alumina (Merck, Art. No. 1097) with
THF. The third reddish brown band including 2 was collected, and the
solvent was removed under reduced pressure. Complex 2 was obtained
as a brownish yellow solid (204 mg, 60% yield). Complex 2 equi-
librated with 5 in solution, and the 2:5 ratio was estimated to be 94:6
51
1
1, and 13 were solved by the direct method using SHELXT-2014/5
52
and further refined with the SHELXL-2016/6 program package. All
non-hydrogen atoms were found by a difference Fourier synthesis and
were refined anisotropically except for the disordered atoms described
below. The refinement was carried out by least-squares methods based
on F2 with all measured reflections. For 11, the disorder at the
nitrogen atom of the μ -imidoyl ligand, N(2), was refined in the ratio
3
of 75:25 and the hydrogen atom attached to the N(2) atom could not
be located. For 13, disorder at the Cp* ligand bonded to Ru(4) was
refined in the ratio of 67:33 and refined isotropically. Hydrogen atoms
attached to the Cp* ligand were not located. The metal-bound
hydrogen atoms in 4, 7, 9, 11, and 13 were located in a difference
Fourier map and refined isotopically. Crystal data and results of the
analyses are given in Table S5 in the Supporting Information. The CIF
data of 4, 7, 9, 11, and 13 are deposited with the Cambridge
Crystallographic Data Center with the deposition numbers 1827738
at −10 °C; the population of 5 decreased to zero above 10 °C. Data
1
for 2 are as follows. H NMR (400 MHz, acetone-d
, −50 °C):
6
δ −24.50 (s, 1H, Ru−H), −4.01 (s, 2H, RuH), 1.53 (s, 15H, Cp*),
3
3
1.99 (s, 30H, Cp*), 5.87 (d, JH−H = 7.6 Hz, 2H, o-Ph), 6.73 (t, JH−H
=
3
13
7.6 Hz, 1H, p-Ph), 6.93 ppm (t, JH−H = 7.6 Hz, 2H, m-Ph). C NMR
1
(100 MHz, THF-d
, −40 °C): δ 11.9 (q, JC−H = 126 Hz, C
(q, JC−H = 126 Hz, C Me ), 84.4 (s, C Me ), 91.2 (s, C
(s, CN), 123.2 (d, JC−H = 160 Hz, Ph), 126.3 (d, JC−H = 158 Hz, Ph),
Me
), 12.7
8
5
5
1
Me
), 97.8
5
5
5
5
5
5
1
1
1
2
(4), 1827742 (7), 1827741 (9), 1827739 (11), and 1827740 (13),
126.8 (d, JC−H = 158 Hz, Ph), 143.3 ppm (t, JC−H = 8 Hz, ipso-Ph).
−1
respectively.
IR (ATR, cm ): ν 2976, 2902, 1600,1592, 1478, 1448, 1373, 1284,
1248, 1156, 1070, 1025, 859. Anal. Calcd for C37 : C, 54.53;
H, 6.55; N, 1.72. Found: C, 54.18; H, 6.93; N, 1.60. 5: H NMR
(400 MHz, acetone-d , −50 °C): δ −25.61 (t, JH−H = 4.9 Hz, 1H,
RuH), −15.08 ppm (d, JH−H = 4.9 Hz, 2H, RuH). Other signals
derived from 5 were not identified due to its low population.
Variable-Temperature NMR Spectra and NMR Simulations.
Variable-temperature NMR studies were performed in NMR tubes
equipped with a J. Young valve using a Varian INOVA-400 Fourier
H53NRu
3
1
2
6
1
2
transform spectrometer. The NMR simulation for the H NMR
spectrum of 8 recorded at 25 °C was performed using gNMR v5.0.6.0
2
2
(
2006, Ivory Soft). Final simulated line shapes were obtained via an
Preparation of (Cp*Ru) (μ-H) {μ -η :η (⊥)-PhCNH} (3). A 20 mL
3 2 3
iterative parameter search upon the coupling constants, and the detail
of the fitting procedure is shown in Figure S16 in the Supporting
Information. The NMR simulations for the temperature-dependent
Cp* and methylene signals of 10 and Cp* and hydrido signals of 13
were also performed via an iterative parameter search upon the
exchange constants k, which was the rate for the pivoting motion of
glass tube equipped with a Teflon valve was charged with THF
(10 mL) and 2 (57.2 mg, 70.2 μmol). The reaction mixture was stirred
at 100 °C for 2 h. The solution turned from brownish yellow to brown.
The solvent was removed under reduced pressure. Complex 3 was
1
obtained as a brown solid (52.3 mg, 91% yield). H NMR (400 MHz,
acetone-d
(s, 30H, Cp*), 6.12 (d, J
8.0 Hz, 1H, p-Ph), 6.99 (dd, J
(br s, 1H, NH). C NMR (100 MHz, benzene-d , 25 °C): δ 12.2 (br q,
w1/2 = 18 Hz, J
126 Hz, C Me ), 80.0 (br s, w = 9 Hz, C Me ), 90.6 (br s, w
, −20 °C): δ −8.07 (s, 2H, RuH), 1.66 (s, 15H, Cp*), 1.76
6
2
2
3
3
the μ -η (∥)-nitrile ligand in 10 and the rotation of the μ -η -
= 8.4 Hz, 2H, o-Ph), 6.86 (t, J
=
H−H
3
3
H−H
3
benzylideneamido ligand in 13. The details of the fitting procedure are
also shown in Tables S2 and S3 and Figures S27 and S34 in the
Supporting Information. The rate constants k that accurately modeled
the experimental spectra at each temperature are also given in the
Supporting Information (Figures S25 and S26 for 10 and Figures S32
= 8.4, 8.0 Hz, 2H, m-Ph), 12.61 ppm
H−H
13
6
1
1
= 126 Hz, C Me ), 12.6 (br q, w = 13 Hz, J
=
=
=
C−H
5
5
1/2
C−H
5
5
1/2
5
5
1/2
1
1
10 Hz, C Me ), 124.4 (d, J
= 161 Hz, p-Ph), 125.9 (d, J
5
5
C−H
C−H
⧧
⧧
1
and S33 for 13). The activation parameters ΔH and ΔS were
determined from the plot of ln(k/T) versus 1/T. Estimated standard
deviations (σ) in the slope and y intercept of the Eyring plot deter-
158 Hz, m-Ph), 126.7 (d, J
161.4 ppm (s, CN). IR (ATR, cm ): ν 2976, 2956, 2901, 2854, 1618,
1593, 1477, 1461, 1427, 1373, 1284, 1248, 1070, 1025, 763, 702 cm .
= 157 Hz, o-Ph), 151.9 (s, ipso-Ph),
C−H
−1
−1
⧧
⧧
mined the error in ΔH and ΔS , respectively. The standard deviation
Anal. Calcd for C H NRu : C, 54.53; H, 6.55; N, 1.72. Found: C,
37
53
3
⧧
⧧
2
⧧ 2
in ΔG was determined from the formula σ(ΔG ) = σ(ΔH ) +
54.14; H, 6.54; N, 1.77.
Preparation of {Cp*Ru(μ-H)} (μ -NCH Ph) (4). A 50 mL
⧧
2
⧧
⧧
[
Tσ(ΔS )] − 2Tσ(ΔH )σ(ΔS ).
3
3
2
Computational Details. DFT calculations for 5, which is the
Schlenk tube was charged with THF (5 mL) and 2 (44.1 mg,
54.1 μmol). After the solution was degassed by one freeze−pump−
thaw cycle, hydrogen (0.1 MPa) was added at room temperature and
the mixture stirred for 24 h. The solution turned from reddish brown
to dark green. The solvent was removed under reduced pressure. The
residual solid was purified by column chromatography on alumina
(Merck, Art. No. 1097) with a mixed solvent of hexane and THF
possible isomer of 2, were carried out at the ωB97XD level in
conjunction with the Stuttgart/Dresden ECP and associated with
53
triple-ζ SDD basis sets for Ru. For C, H, and O, 6-31(d) basis sets
were employed. No simplified model compounds were used for the
calculations. Initial geometries for the optimization were based on the
crystallographically determined structure for 2. All calculations were
L
DOI: 10.1021/acs.organomet.8b00165
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
(5/1 in volume). The first green band including 4 was collected.
JH−H values among the hydrido ligands were estimated on the basis
A 38.6 mg amount of 4 was obtained as a dark green solid on removal
of the solvent under reduced pressure (87% yield). A single crystal
of spectral simulations (Figure S16 in the Supporting Information).
13
1
C NMR (100 MHz, THF-d , 25 °C): δ 11.7 (q, J
= 126 Hz,
= 126 Hz,
8
C−H
1
1
used for the diffraction studies was prepared from the pentane solution
C Me ), 12.1 (q, J
= 126 Hz, C Me ), 12.6 (q, J
C−H 5 5 C−H
5
5
1
stored at 2 °C. H NMR (400 MHz, THF-d , 25 °C): δ −14.14 (s, 3H,
C Me ), 86.8 (s, C Me ), 91.0 (s, C Me ), 93.4 (s, C Me ), 124.3
8
5
5
5
5
5
5
5
5
1
1
RuH), 1.70 (s, 45H, Cp*), 4.81 (s, 2H, NCH ), 7.08 (t, J
= 7.6 Hz,
(d, J
(d, J
= 159 Hz, Ph), 126.9 (d, J
= 158 Hz, Ph), 153.6 (s, ipso-Ph). 169.1 ppm (s, PhCNH).
= 156 Hz, Ph), 127.4
2
H−H
C−H
C−H
1
1
=
H, p-Ph), 7.16 (dd, J
= 7.6, 7.2 Hz, 2H, m-Ph), 7.34 ppm (d, J
H−H
H−H
C−H
7.2 Hz, 2H, o-Ph). 13C{ H} NMR (100 MHz, THF-d , 25 °C):
1
8
Hydrogenation of {Cp*Ru(μ-H)} (μ -NCH Ph) (4). Complex 4
3 3 2
δ 12.5 (C Me ), 81.7 (NCH Ph), 91.4 (C Me ), 126.2 (Ph), 127.5
(20.0 mg. 24.5 μmol) and THF-d (2 mL) were charged in a glass
5
5
2
5
5
8
(
Ph), 128.1 (Ph), 144.3 ppm (ipso-Ph). Anal. Calcd for C H NRu :
autoclave together with hexamethylbenzene (4.0 mg, 24.7 μmol) as an
internal standard. After the reactor was degassed with a liquid nitrogen
37
55
3
C, 54.39; H, 6.79; N, 1.71. Found: C, 54.27; H, 6.47; N, 1.87.
2
Formation of (Cp*Ru) (μ-H) (μ -η -NCPhH) (6). An NMR
bath, the reactor was filled with 0.5 MPa of H . The reactor was then
3
2
3
2
tube equipped with a J. Young valve was charged with 2 (4.3 mg,
.3 μmol), THF-d (0.45 mL), and cyclooctane (1 μL) as an internal
heated at 100 °C for 21 h. After H gas was purged from the reactor,
2
1
5
the solution was transferred to an NMR tube. The H NMR spectrum
8
standard. The solution was allowed to stand at 25 °C and monitored
of the solution showed that all of 4 was converted into 1. (The
formation of a small amount of the bis(μ -oxo) complex (Cp*Ru)
1
periodically by H NMR spectroscopy. The distribution of 6 gradually
3
3-
increased and became steady after 3 h, where the ratio between 2 and
(μ -O) (μ-H) was also observed in the spectrum. The bis(μ -oxo)
3
2
3
1
6
was estimated at 73:27. H NMR (400 MHz, acetone-d , −50 °C):
complex was formed by the reaction of 1 with air during the transport
of the solution into the NMR tube.) Two singlet signals were observed
at δ 3.77 and 3.71 ppm with a ratio of 7:3, which were assigned to the
methylene signals of benzylamine and N-benzylidenebenzylamine, respec-
tively. Formation of benzylamine and N-benzylidenebenzylamine was also
confirmed by GC analysis of the solution.
6
2
2
δ −9.89 (d, J
= 2.8 Hz, 1H, RuH), −8.23 (d, J
= 2.8 Hz, 1H,
H−H
H−H
RuH), 1.57 (s, 15H, Cp*), 1.68 (s, 15H, Cp*), 1.90 (s, 15H, Cp*),
3
3
7
7
.07 (t, J
= 7.8 Hz, 1H, p-Ph), 7.20 (t, J
= 7.8 Hz, 2H, m-Ph),
H−H
H−H
3
.44 (d, J
= 7.8 Hz, 2H, o-Ph), 8.06 ppm (s, 1H, NCHPh).
H−H
1
3
1
C NMR (100 MHz, acetone-d , 25 °C): δ 11.8 (q, J
= 127 Hz,
= 126 Hz,
6
C−H
C−H
1
1
C Me ), 12.6 (q, J
= 126 Hz, C Me ), 13.2 (q, J
Reaction of 1 with Benzylamine at 100 °C. An NMR tube
equipped with a Teflon valve was charged with 1 (4.2 mg, 5.9 μmol),
5
5
C−H
5
5
C Me ), 81.7 (s, C Me ), 85.8 (s, C Me ), 86.5 (s, C Me ), 98.4
5
5
5
5
5
5
5
5
1
1
(
1
d, J
27.2 (d, J
= 170 Hz, N = CPhH), 125.6 (d, J
= 159 Hz, p-Ph),
= 160 Hz, Ph),
C−H
THF-d (0.4 mL), and hexamethylbenzene (1.8 mg) as an internal
C−H
C−H
8
1
1
= 158 Hz, Ph), 128.2 (d, J
standard. Then, benzylamine (3.0 μL, 28 μmol) was added to the
solution. The tube was heated at 100 °C using an oil bath. The
C−H
1
46.3 ppm (s, ipso-Ph).
Preparation of (Cp*Ru) (μ-H) (μ -η -NCPhH)(μ-CN Bu) (7).
2
t
1
3
2
3
H NMR spectrum showed that 90% of 1 was converted after 24 h
A 50 mL Schlenk tube was charged with toluene (5 mL) and an
and afford a mixture of 3 and 8 in a ratio of 5:4. Formation of
μ3-benzylimido complex 4 was not detected.
equilibrated mixture of 2 and 6 (58.0 mg, 0.071 mmol). An excess
t
amount of BuNC (65 μL, 0.58 mmol) was then added to the reaction
Reaction of 1-d with Benzylamine at 50 °C. An NMR tube
5
mixture at 25 °C, and the solution was stirred for 8 h. The solution
equipped with a Teflon valve was charged with 1-d5 (8.6 mg,
turned from brownish yellow to red. The solvent and remaining
12.0 μmol), THF-d (0.4 mL), and hexamethylbenzene (1.2 mg) as an
8
t
BuNC were removed under reduced pressure. The residual solid was
internal standard. Then, benzylamine (1.2 μL, 11.0 μmol) was added
to the solution. The tube was heated at 50 °C using an oil bath, and
the 1H NMR spectra were periodically recorded. The H NMR
spectrum recorded after 8 h showed that the signal intensity of the
methylene proton of benzylamine decreased by ca. 50%.
purified by the use of column chromatography on alumina (Merck,
Art. No. 1097) with a mixed solvent of toluene and THF (5/2 in
volume). The first reddish brown band including 7 was collected, and
the solvent was removed under reduced pressure. Complex 7 was
obtained as a reddish brown solid (55.8 mg, 62.1 μmol, 87% yield).
A brown single crystal used for the diffraction study was obtained from
1
2
t
Preparation of (Cp*Ru) (μ-H) {μ -η (∥)-PhCNH}(CN Bu) (9).
3
2
3
A 50 mL Schlenk tube was charged with THF (4 mL) and 3 (28.6 mg,
35.1 μmol). A 1.5 equiv amount of BuNC (6 μL, 53 μmol) was then
1
t
a cold pentane solution of 7 stored at 0 °C. H NMR (400 MHz,
THF-d , 25 °C): δ −22.01 (s, 1H, RuH), −19.74(s, 1H, RuH), 1.40
added to the reaction mixture at 25 °C, and the solution was stirred for
8
t
(
(
s, 9H, μ-CN Bu), 1.57 (s, 15H, Cp*), 1.77 (s, 15H, Cp*), 1.80
5 min. The solution immediately turned from brown to green.
t
s, 15H, Cp*), 6.62 (s, 1H, μ -N = CPhH), 7.09 (t, J
= 7.4 Hz, 1H,
Removal of the solvent and remaining BuNC under reduced pressure
3
H−H
p-Ph), 7.17 (dd, J
7
(
(
= 7.4, 7.0 Hz, 2H, p-Ph), 7.39 ppm (t, JH−H
=
gave 9 as a green solid (31.2 mg, 34.7 μmol, 99% yield). A green single
crystal used for the diffraction study was obtained by slow evaporation
H−H
1
3
.0 Hz, 2H, o-Ph). C NMR (100 MHz, THF-d , −30 °C): δ 11.5
8
1
1
1
q, J
= 127 Hz, C Me ), 11.7 (q, J
= 127 Hz, C Me ), 11.9
= 126 Hz, CNCMe ), 56.0
of the solvent from a pentane solution of 9 stored at 25 °C. H NMR
C−H
5
5
C−H
5
5
1
1
t
q, J
= 127 Hz, C Me ), 31.9 (q, J
(400 MHz, C D , 25 °C): δ 0.96 (s, 9H, CN Bu), 1.60 (s, 15H, Cp*),
C−H
5
5
C−H
3
6
6
(
s, CNCMe ), 87.5 (s, C Me ), 91.3 (s, C Me ), 92.9 (s, C Me ),
1.93 (s, 15H, Cp*), 2.09 (s, 15H, Cp*), 6.34 (br s, 1H, PhCNH), 7.12
3
5
5
5
5
5
5
1
2
1
3
3
1
1
7
2
1
7
06.9 (d, J
= 171 Hz, μ -η -N = CPhH), 126.8 (dt, J
=
=
(t, J
= 7.2 Hz, 1H, p-Ph), 7.31 (dd, J
= 7.2, 7.2 Hz, 2H,
C−H
3
C−H
H−H
H−H
3
1
3
3
60 Hz, J
= 7.5 Hz, p-Ph), 127.8 (dd, J
= 160 Hz, J
m-Ph), 7.46 ppm (d, J
= 7.2 Hz, 2H, o-Ph). Hydrido signals were
C−H
C−H
C−H
H−H
1
.7 Hz, o-Ph), 128.2 (br d, J
= 158 Hz, m-Ph), 144.0 (s, ipso-Ph),
not observed due to severe broadening at ambient temperature.
C−H
t
−1
1
2
06.4 ppm (s, μ-CN Bu). IR (KBr, cm ): ν 3393, 2970, 2898, 2716,
712 (ν(CN)), 1599, 1454, 1367, 1199, 1069, 1025, 876, 822,
H NMR (400 MHz, THF-d , −80 °C): δ −19.87 (d, J
= 3.7 Hz,
8
H−H
2
1H, RuH), −10.15 (d, J
CN Bu), 1.47 (s, 15H, Cp*), 1.83 (s, 15H, Cp*), 1.92 (s, 15H, Cp*),
= 3.7 Hz, 1H, RuH), 1.05 (s, 9H,
H−H
−1
t
56 cm . Anal. Calcd for C H N Ru : C, 56.16; H, 6.96; N, 3.12.
42
62
2
3
13
Found: C, 56.51; H, 7.25; N, 3.27.
Preparation of (Cp*Ru) (μ-H) {μ -η (∥)-PhCNH} (8). An NMR
6.80 (br s, 1H, PhCNH), 7.0−7.7 ppm (m, 5H, Ph). C NMR
1
2
(100 MHz, C D , 25 °C): δ 11.4 (q, J
= 126 Hz, C Me ), 12.65
5 5
3
4
3
6
6
C−H
1
1
tube equipped with J. Young valve was charged with 3 (12.1 mg,
(q, J
(q, J
= 126 Hz, C Me ), 12.69 (q, J
= 126 Hz, C Me ), 31.79
C−H
5
5
C−H 5 5
1
1
4.8 μmol), THF-d (0.4 mL), and 2,2,4,4-tetramethypentane (1 μL) as an
= 127 Hz, CNCMe ), 55.5 (s, CNCMe ), 82.5 (s, C Me ),
8
C−H
3 3 5 5
1
internal standard. After the solution was frozen by a liquid nitrogen
86.1 (s, C Me ), 91.9 (s, C Me ), 123.3 (d, J
(d, J
= 157 Hz, Ph), 127.1
C−H
= 157 Hz, Ph), 155.1 (s, ipso-Ph), 157.9 (s, CN Bu or
5
5
5
5
1
t
bath, the tube was degassed. The solution was then warmed to 25 °C,
C−H
2
t
2
and 0.1 MPa of hydrogen was introduced. The solution immediately
μ -η (∥)-PhCNH), 179.1 ppm (s, CN Bu or μ -η (∥)-PhCNH). One
3
3
1
13
turned from brown to red. The H NMR spectrum recorded after
C signal derived from the phenyl group was not observed owing to
1
−1
1
0 min showed the quantitative formation of 8. H NMR (400 MHz,
obstruction by the solvent signals. IR (KBr, cm ): ν 3442, 2972, 2900,
1999 (ν(CN)), 1598, 1453, 1369, 1205, 1038, 1025. Anal. Calcd for
C H N Ru : C, 56.16; H, 6.96; N, 3.12. Found: C, 55.87; H, 6.99; N,
2
THF-d , 25 °C): δ −18.48 (ddd, J
−
5
= 5.1, 3.9, 1.3 Hz 1H, RuH),
8
H−H
2
2
16.76 (ddd, J
.3, 5.1, 1.4 Hz, 1H, RuH), −13.67 (ddd, J
= 8.3, 5.3, 3.9 Hz, 1H, RuH), −15.03 (ddd, J
=
H−H
H−H
42 62
2
3
2
= 8.3, 1.4, 1.3 Hz, 1H,
2.94.
H−H
2
RuH), 1.65 (s, 15H, Cp*), 1.75 (s, 15H, Cp*), 1.90 (s, 15H, Cp*),
.15 (br s, w1/2 = 8.7 Hz, 1H, PhCNH), 6.91−7.15 ppm (m, 5H, Ph).
Preparation of (Cp*Ru) (μ-H)(μ -NCH Ph){μ -η (∥)-PhCN}
3
3
2
3
6
(10). A 50 mL glass tube equipped with a Teflon valve was charged
M
DOI: 10.1021/acs.organomet.8b00165
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
1
1
with THF (7.5 mL) and 3 (92.8 mg, 0.114 mmol). After a large excess
amount of benzonitrile (0.24 mL, 2.30 mmol) was added to the
solution, the reaction mixture was heated at 70 °C for 45 h with
vigorous stirring. The solvent and remaining benzonitrile were then
removed under reduced pressure. The residual solid including 10 was
dissolved in pentane (3 mL) and purified by column chromatography
on alumina (Merck, Art. No. 1097). The eluent was gradually changed
from pentane to toluene, THF, and methanol. The fourth green band
including 10 that was eluted with THF and methanol was collected.
A 26.7 mg amount of 10 was obtained as a dark green solid on removal
of the solvent under reduced pressure (26% yield). Due to the
difficulty in the complete removal of methanol from the residual solid,
109.9 (d, J
= 171 Hz, N = CPhH), 127.0 (d, J
= 160 Hz, Ph),
C−H
C−H
1
1
127.6 (d, J
= 159 Hz, Ph), 127.9 (d, J
= 160 Hz, Ph), 143.7
C−H
C−H
−
1
(s, ipso-Ph). 234.3 ppm (s, μ-CO). IR (KBr, cm ): ν 2976, 2952,
2900, 1735 (ν(CO)), 1596, 1483, 1454, 1374, 1070, 1070, 1027, 816,
760. Anal. Calcd for C H NORu : C, 54.14; H, 6.34; N, 1.66. Found:
C, 54.11; H, 6.36; N, 1.83.
Thermolysis of 13. An NMR tube equipped with a J. Young valve
was charged with 13 (1.1 mg, 1.3 μmol), THF-d (0.4 mL), and
2,2,4,4-tetramethylpentane (1 μL) as an internal standard. The tube
was then heated at 110 °C for 20 h. The H NMR spectrum of the
38
53
3
8
1
reaction solution showed that ca. 7% of 13 was consumed. Formation
of 15 (ca. 2%) was confirmed by the hydrido signal of 15 resonating at
δ −13.55 ppm.
1
elemental analysis of 10 did not fit with the calculated values. H NMR
(
400 MHz, THF-d , −80 °C): δ −27.80 (s, 1H, RuH), 1.30 (s, 15H,
Preparation of (Cp*Ru) (μ-H) {μ-N(CH Ph)H)}(μ-CO) (14).
8
3
2
2
2
Cp*), 1.51 (s, 15H, Cp*), 1.58 (s, 15H, Cp*), 6.21 (d, J
1
=
A 50 mL glass autoclave was charged with THF (10 mL) and 13
(68.3 mg, 80.8 μmol). After the solution was degassed by one freeze−
pump−thaw cycle, hydrogen (0.6 MPa) was introduced at 25 °C and
the mixture stirred for 1 week. The solution turned from red to
brownish yellow. The solvent was removed under reduced pressure.
The residual solid was purified by column chromatography on alumina
(Merck, Art. No. 1097) with a mixed solvent of hexane and toluene
(3/2 in volume). The first green band including 14 was collected.
Removal of the solvent under reduced pressure gave a 30.9 mg amount
H−H
2
7.7 Hz, 1H, NCH Ph), 6.42 (d, J
= 17.7 Hz, 1H, NCH Ph), 6.86
2
H−H 2
(
dd, 3JH−H = 8.4, 8.0 Hz, 2H, m-Ph), 7.06−7.50 ppm (m, 10H,
13
NCH Ph and NCPh). C NMR (100 MHz, C D , 25 °C): δ 10.9
2
6
6
1
1
(
(
(
(
q, J
= 126 Hz, C Me ), 11.3 (q, J
= 126 Hz, C Me ), 79.1
= 134 Hz, NCH Ph), 93.1 (s, C Me ), 96.1 (s, C Me ), 125.3
C−H
5
5
C−H 5 5
1
t, J
C−H
2 5 5 5 5
Ph), 126.2 (Ph), 126.8 (Ph), 130.1 (br, Ph), 144.0 (s, ipso-Ph), 148.1
s, ipso-Ph), 182.2 ppm (s, PhCN). Several 13C signals derived from
two phenyl groups in 10 were not identified probably due to the
1
obstruction by the solvent signal and broadening arising from
of 14 as a green solid (0.37 mmol, 45% yield). H NMR (400 MHz,
C D , 25 °C): δ −10.59 (s, 2H, RuH), 1.74 (s, 30H, Cp*), 1.77
(s, 15H, Cp*), 2.67 (d, J
2
restricted rotation of the μ -η (∥)-nitrile ligand.
3
6
6
3
Preparation of a Tetrafluoroborate Salt of [(Cp*Ru) (μ-H)(μ -
= 7.3 Hz, 2H, N(CH Ph)H), 3.71 (br t,
H−H 2
3
3
2
+
NCH Ph){μ -η (∥)-PhCNH}] (11). A 50 mL Schlenk tube was
3
13
2
3
JH−H = 7.3 Hz, 1H, NH), 7.07−7.23 ppm (m, 5H, Ph). C NMR
charged with diethyl ether (5.0 mL) and 10 (31.5 mg, 34.2 μmol).
After the flask was cooled with a dry ice/methanol bath, a slight excess
amount of tetrafluoroboric acid diethyl ether complex (5.5 μL,
1
(
(
(
100 MHz, THF-d , 25 °C): δ 10.5 (q, J
= 126 Hz, C Me ), 12.7
C−H 5 5
8
1
q, J
= 126 Hz, C Me ), 79.6 (s, C Me ), 92.9 (s, C Me ), 93.1
C−H
5 5 5 5 5 5
1
1
t, J
= 134 Hz, N(CH Ph)H), 127.5 (d, J
= 160 Hz, Ph),
C−H
= 159 Hz, Ph), 145.6
C−H
2
40 μmol) was added to the solution. The solution was then gradually
1
1
1
(
2
29.1 (d, J
= 156 Hz, Ph), 129.3 (d, J
C−H
C−H
warmed to 25 °C and stirred for 0.5 h at ambient temperature. A dark
brown precipitate immediately formed, and the supernatant was
removed. The residual solid was rinsed three times with 1 mL of
diethyl ether. Drying under vacuum gave 11 as a tetrafluoroborate salt
−1
s, ipso-Ph). 271.4 ppm (s, μ-CO). IR (KBr, cm ): ν 2976, 2904,
842, 1665 (ν(CO)), 1492, 1451, 1370, 1072, 1070, 1025, 752, 694.
Anal. Calcd for C H NORu : C, 54.01; H, 6.56; N, 1.66. Found: C,
4.09; H, 6.76; N, 2.03.
Reaction of 12 with Benzylamine. An NMR tube equipped with
a J. Young valve was charged with 12 (5.7 mg, 7.7 μmol), THF-d8
0.4 mL), and 2,2,4,4-tetramethylpentane (1 μL) as an internal stan-
dard. After benzylamine (8 μL, 73 μmol) was added to the solution,
the tube was heated at 120 °C for 70 h. The H NMR spectrum of the
reaction solution showed that 13 was formed in 24% yield. The
formation of a trace amount of 15 was also observed.
Hydrogenation of (Cp*Ru) (μ-H) {μ-N(CH Ph)H)}(μ-CO) (14).
Complex 14 (15.2 mg. 18.0 μmol) and THF (2 mL) were charged in a
glass autoclave. After the reactor was degassed with a liquid nitrogen
bath, the reactor was filled with 0.6 MPa of H . The reactor was heated
at 100 °C for 24 h. The solution was then transferred to a 50 mL
Schlenk tube, and the solution was removed under reduced pressure.
Conversion of 14 was estimated to be 64% on the basis of the
H NMR spectrum of the residual solid, and the spectrum also showed
that complexes 12, 13, and 15 were formed in 47, 11, and 6% yields,
38
55
3
5
(
33.4 mg, 97% yield). A single crystal suitable for the diffraction study
was obtained from the acetone solution of a tetraphenylborate salt of
1, which was prepared by exchanging the counteranion with
1
(
tetraphenylborate, stored at 25 °C by vapor diffusion of diethyl
ether. Data for 11-BF are as follows. H NMR (400 MHz, acetone-d ,
1
1
4
6
2
5 °C): δ −25.98 (s, 1H, RuH), 1.34 (br, 15H, Cp*), 1.51 (s, 15H,
2
Cp*), 1.84 (br, 15H, Cp*), 6.28 (d, J
= 18.1 Hz, 1H, NCH Ph),
2
H−H
2
6.70 (d, J
= 18.1 Hz, 1H, NCH Ph), 7.24−7.52 (m, 10H, NCH Ph
H−H
2
2
3
2
2
and PhCNH), 9.08 ppm (br s w = 14 Hz, 1H, PhCNH).
1
/2
1
3
1
C NMR (100 MHz, acetone-d , 25 °C): δ 10.8 (br q, J
=
=
6
C−H
1
1
1
1
9
26 Hz, C Me ), 10.9 (q, J
= 126 Hz, C Me ), 11.1 (br q, J
= 136 Hz, NCH Ph), 97.6 (s, C Me ),
5
5
C−H
5 5 C−H
1
2
26 Hz, C Me ), 81.3 (t, J
5
5
C−H
2 5 5
8.8 (s, C Me ), 101.1 (s, C Me ),127.8 (d, J = 161 Hz, Ph), 128.0
5
5
5
5
C−H
1
(
d, J
= 158 Hz, Ph), 128.66 (d, J
= 161 Hz, Ph), 128.70
C−H
C−H
C−H
1
(d, J
= 161 Hz, Ph), 143.2 (s, ipso-Ph), 146.2 (s, ipso-Ph), 179.2 ppm
1
(
s, PhC=NH). Elemental analysis was performed on 11-BPh . Anal.
4
Calcd for C H BN Ru : C, 65.95; H, 6.43; N, 2.26. Found: C, 65.95; H,
68
79
2
3
respectively.
6
.70; N, 2.26.
Catalytic Hydrogenation of Benzonitrile. A series of catalytic
hydrogenations of benzonitrile were performed in a 100 mL stainless-
steel autoclave equipped with a glass cup and magnetic stirrer bar. The
reaction solution (catalyst (ca. 6−10 mg), benzonitrile (100 equiv),
and THF (3 mL)) was prepared in a 20 mL Schlenk tube, and the
solution was transferred into the autoclave under an Ar atmosphere.
The reactor was filled with H gas at an appropriate pressure (typically
MPa). The reactor was heated using an oil bath with vigorous
stirring. After the appropriate reaction time (typically 20 h), the
solution was analyzed by GLC. The results of catalytic reactions are
given in Table 1.
2
Preparation of (Cp*Ru) (μ-H) (μ -η -NCPhH)(μ-CO) (13).
3
2
3
A 50 mL Schlenk tube was charged with THF (20 mL) and 12
263.2 mg, 0.356 mmol). An excess amount of benzonitrile (0.36 mL,
.50 mmol) was then added to the solution at 25 °C, and the solution
(
3
was stirred for 0.5 h. The solution turned from dark brown to red. The
solvent and remaining benzonitrile were removed under reduced
pressure. Recrystallization from a hexane solution of the residual solid
stored at −30 °C gave a 167 mg amount of analytically pure 13 as a
red single crystal (0.198 mmol, 56% yield). A single crystal used for the
2
5
diffraction study was prepared from the THF solution of 13 stored at
1
−
30 °C. H NMR (400 MHz, THF-d , −40 °C): δ −21.21 (s, 1H,
8
RuH), −19.46 (s, 1H, RuH), 1.56 (s, 15H, Cp*), 1.80 (s, 15H, Cp*),
3
1
.83 (s, 15H, Cp*), 6.56 (s, 1H, N = CPhH), 7.10 (t, J
= 7.6 Hz,
ASSOCIATED CONTENT
H−H
■
3
1
H, p-Ph), 7.18 (dd, J
= 7.6, 7.2 Hz, 2H, m-Ph), 7.34 ppm
H−H
3
13
*
S
Supporting Information
The Supporting Information is available free of charge on the ACS
(
d, J
= 7.2 Hz, 2H, o-Ph). C NMR (100 MHz, THF-d , 25 °C):
H−H 8
1 1
δ 11.1 (q, J
= 127 Hz, C Me ), 11.8 (br q, J = 127 Hz,
C−H
C−H
5
5
C Me ), 88.1 (s, C Me ), 93.4 (br s, C Me ), 94.0 (br s, C Me ),
Publications website at DOI: 10.1021/acs.organomet.8b00165.
5
5
5
5
5
5
5
5
N
DOI: 10.1021/acs.organomet.8b00165
Organometallics XXXX, XXX, XXX−XXX

Organometallics
NMR spectra of 2−11, 13, and 14, VT-NMR spectra of
Article
(8) (a) Reguillo, R.; Grellier, M.; Vautravers, N.; Vendier, L.; Sabo-
Etienne, S. J. Am. Chem. Soc. 2010, 132, 7854−7855. (b) Gunanathan,
2, 10, and 13, crystal data of 4, 7, 9, 11, and 13,
̈
C.; Holscher, M.; Leitner, W. Eur. J. Inorg. Chem. 2011, 2011, 3381−
preliminary results of an XRD study for 14, and results of
DFT calculations for 2 and 5 (PDF)
Atom coordinates of the optimized structures of 2 and 5
3
386. (c) Bornschein, C.; Werkmeister, S.; Wendt, B.; Jiao, H.;
Alberico, E.; Baumann, W.; Junge, H.; Junge, K.; Beller, M. Nat.
Commun. 2014, 5, 4111. (d) Neumann, J.; Bornschein, C.; Jiao, H.;
Junge, K.; Beller, M. Eur. J. Org. Chem. 2015, 2015, 5944−5948.
e) Choi, J. H.; Prechtl, M. H. G. ChemCatChem 2015, 7, 1023−1028.
f) Adam, R.; Bheeter Charles, B.; Jackstell, R.; Beller, M.
(
XYZ)
(
(
Accession Codes
CCDC 1827738−1827742 contain the supplementary crystal-
lographic data for this paper. These data can be obtained free of
charge via www.ccdc.cam.ac.uk/data_request/cif, or by email-
ing data_request@ccdc.cam.ac.uk, or by contacting The
Cambridge Crystallographic Data Centre, 12 Union Road,
Cambridge CB2 1EZ, UK; fax: +44 1223 336033.
ChemCatChem 2016, 8, 1329−1334. (g) Adam, R.; Alberico, E.;
Baumann, W.; Drexler, H. J.; Jackstell, R.; Junge, H.; Beller, M. Chem. -
Eur. J. 2016, 22, 4991−5002.
(
9) (a) Li, H.; Wu, Y.; Wan, Y.; Zhang, J.; Dai, W.; Qiao, M. Catal.
Today 2004, 93−95, 493−503. (b) Li, H.; Wu, Y.; Zhang, J.; Dai, W.;
Qiao, M. Appl. Catal., A 2004, 275, 199−206. (c) Chojecki, A.;
Veprek-Heijman, M.; Mu
Lercher, J. A. J. Catal. 2007, 245, 237−248. (d) Schar
̈
ller, T. E.; Schar
̈
ringer, P.; Veprek, S.;
ringer, P.; Muller,
AUTHOR INFORMATION
Corresponding Author
E-mail for T.T.: takao.t.aa@m.titech.ac.jp.
̈
̈
■
T. E.; Lercher, J. A. J. Catal. 2008, 253, 167−179. (e) Segobia, D. J.;
Trasarti, A. F.; Apesteguía, C. R. Appl. Catal., A 2012, 445−446, 69−
*
7
5. (f) Muratsugu, S.; Kityakarn, S.; Wang, F.; Ishiguro, N.; Kamachi,
ORCID
T.; Yoshizawa, K.; Sekizawa, O.; Uruga, T.; Tada, M. Phys. Chem.
Chem. Phys. 2015, 17, 24791−24802.
Toshiro Takao: 0000-0002-5393-112X
Hiroharu Suzuki: 0000-0002-9718-1375
Notes
(10) Jiao, H.; Junge, K.; Alberico, E.; Beller, M. J. Comput. Chem.
2
(
016, 37, 168−176.
11) Krupka, J.; Pasek, J. Curr. Org. Chem. 2012, 16, 988−1004.
The authors declare no competing financial interest.
(12) (a) Kordesch, M. E.; Lindner, T.; Somers, J.; Stenzel, W.;
Conrad, H.; Bradshaw, A. M.; Williams, G. P. Spectrochim. Acta Part A
1987, 43, 1561−1566. (b) Cassuto, A.; Tourillon, G. Vacuum 1995,
ACKNOWLEDGMENTS
■
4
(
1
1
6, 583−585.
13) (a) Friend, C. M.; Muetterties, E. L.; Gland, J. J. Phys. Chem.
981, 85, 3256−3262. (b) Sexton, B. A.; Avery, N. R. Surf. Sci. 1983,
29, 21−36. (c) Avery, N. R.; Matheson, T. W.; Sexton, B. A. Appl.
This work was supported by JST ACT-C Grant No.
JPMJCR12YA. The numerical calculations were carried out
on the TSUBAME2.5 supercomputer at the Tokyo Institute of
Technology, Tokyo, Japan.
Surf. Sci. 1985, 22−23, 384−391. (d) Kordesch, M. E.; Stenzel, W.;
Conrad, H. Surf. Sci. 1986, 175, L687−L692. (e) Serafin, J. G.; Friend,
C. M. J. Phys. Chem. 1988, 92, 6694−6700. (f) Hubbard, A. T.; Cao, E.
Y.; Stern, D. A. Electrochim. Acta 1994, 39, 1007−1014.
REFERENCES
■
(
4
1) (a) de Bellefon, C.; Fouilloux, P. Catal. Rev.: Sci. Eng. 1994, 36,
59−506. (b) Gomez, S.; Peters Joop, A.; Maschmeyer, T. Adv. Synth.
(14) Villegas, I.; Weaver, M. J. J. Am. Chem. Soc. 1996, 118, 458−466.
(15) (a) Ortiz-Hernandez, I.; Owens, D. J.; Strunk, M. R.; Williams,
Catal. 2002, 344, 1037−1057.
2) Kukula, P.; Studer, M.; Blaser, H. U. Adv. Synth. Catal. 2004, 346,
487−1493.
3) Xie, X.; Liotta, C. L.; Eckert, C. A. Ind. Eng. Chem. Res. 2004, 43,
907−7911.
4) Chatterjee, M.; Sato, M.; Kawanami, H.; Yokoyama, T.; Suzuki,
T.; Ishizaka, T. Adv. Synth. Catal. 2010, 352, 2394−2398.
5) (a) Werkmeister, S.; Junge, K.; Beller, M. Org. Process Res. Dev.
C. T. Langmuir 2006, 22, 2629−2639. (b) Ortiz-Hernandez, I.;
(
1
(
7
Williams, C. T. Langmuir 2007, 23, 3172−3178.
(16) Oliva, C.; van den Berg, C.; Niemantsverdriet, J. W.; Curulla-
Ferre,
17) Oliva, C.; van den Berg, C.; Niemantsverdriet, J. W.; Curulla-
Ferre, D. J. Catal. 2007, 248, 38−45.
́
D. J. Catal. 2007, 245, 436−445.
(
(
́
(
18) (a) Friend, C. M.; Serafin, J. G. J. Chem. Phys. 1988, 88, 4037−
(
4
2
(
045. (b) Stevens, P. A.; Madix, R. J.; Friend, C. M. Surf. Sci. 1988,
05, 187−206.
2
2
(
014, 18, 289−302. (b) Bagal, D. B.; Bhanage, B. M. Adv. Synth. Catal.
015, 357, 883−900.
19) (a) Storhoff, B. N.; Lewis, H. C., Jr Coord. Chem. Rev. 1977, 23,
6) (a) Yoshida, T.; Okano, T.; Otsuka, S. J. Chem. Soc., Chem.
1
1
(
−29. (b) Michelin, R. A.; Mozzon, M.; Bertani, R. Coord. Chem. Rev.
996, 147, 299−338.
20) (a) Andrews, M. A.; Kaesz, H. D. J. Am. Chem. Soc. 1979, 101,
Commun. 1979, 870−871. (b) Grey, R. A.; Pez, G. P.; Wallo, A. J. Am.
Chem. Soc. 1981, 103, 7536−7542. (c) Enthaler, S.; Addis, D.; Junge,
K.; Erre, G.; Beller, M. Chem. - Eur. J. 2008, 14, 9491−9494.
(
d) Enthaler, S.; Junge, K.; Addis, D.; Erre, G.; Beller, M.
7255−7259. (b) Andrews, M. A.; Knobler, C. B.; Kaesz, H. D. J. Am.
Chem. Soc. 1979, 101, 7260−7264. (c) Keller, E.; Wolters, D. Chem.
Ber. 1984, 117, 1572−1582. (d) Suter, P.; Vahrenkamp, H. Chem. Ber.
1995, 128, 793−797. (e) Donnecke, D.; Halbauer, K.; Imhof, W. J.
̈
Organomet. Chem. 2004, 689, 2707−2719.
ChemSusChem 2008, 1, 1006−1010. (e) Addis, D.; Enthaler, S.;
Junge, K.; Wendt, B.; Beller, M. Tetrahedron Lett. 2009, 50, 3654−
3656. (f) Miao, X.; Bidange, J.; Dixneuf Pierre, H.; Fischmeister, C.;
Bruneau, C.; Dubois, J. L.; Couturier, J. L. ChemCatChem 2012, 4,
911−1916. (g) Werkmeister, S.; Bornschein, C.; Junge, K.; Beller, M.
Chem. - Eur. J. 2013, 19, 4437−4440.
7) (a) Takemoto, S.; Kawamura, H.; Yamada, Y.; Okada, T.; Ono,
A.; Yoshikawa, E.; Mizobe, Y.; Hidai, M. Organometallics 2002, 21,
897−3904. (b) Li, T.; Bergner, I.; Haque, F. N.; Zimmer-De Iuliis,
1
(21) (a) Bernhardt, W.; Vahrenkamp, H. Angew. Chem., Int. Ed. Engl.
1984, 23, 381. (b) Lausarot, P. M.; Turini, M.; Vaglio, G. A.; Valle, M.;
Tiripicchio, A.; Camellini, M. T.; Gariboldi, P. J. Organomet. Chem.
1984, 273, 239−246.
(
3
(22) Lausarot, P. M.; Vaglio, G. A.; Valle, M.; Tiripicchio, A.;
Tiripicchio Camellini, M.; Gariboldi, P. J. Organomet. Chem. 1985, 291,
221−229.
M.; Song, D.; Morris, R. H. Organometallics 2007, 26, 5940−5949.
(
(
c) Lu, Z.; Williams, T. J. Chem. Commun. 2014, 50, 5391−5393.
d) Mukherjee, A.; Srimani, D.; Chakraborty, S.; Ben-David, Y.;
(23) Andreu, P. L.; Cabeza, J. A.; del Río, I.; Riera, V.; Bois, C.
Organometallics 1996, 15, 3004−3010.
Milstein, D. J. Am. Chem. Soc. 2015, 137, 8888−8891. (e) Chakraborty,
S.; Leitus, G.; Milstein, D. Chem. Commun. 2016, 52, 1812−1815.
(24) Bois, C.; Cabeza, J. A.; Franco, R. J.; Riera, V. c.; Saborit, E. J.
Organomet. Chem. 1998, 564, 201−207.
(
f) Elangovan, S.; Topf, C.; Fischer, S.; Jiao, H.; Spannenberg, A.;
Baumann, W.; Ludwig, R.; Junge, K.; Beller, M. J. Am. Chem. Soc. 2016,
(25) Takao, T.; Kawashima, T.; Matsubara, K.; Suzuki, H.
Organometallics 2005, 24, 3371−3374.
138, 8809−8814.
O
DOI: 10.1021/acs.organomet.8b00165
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
(
26) Joensen, H. A. N.; Hansson, G. K.; Kozlova, S. G.; Gushchin, A.
L.; Søtofte, I.; Ooi, B.-L. Inorg. Chem. 2010, 49, 1720−1727.
27) Takao, T.; Takaya, Y.; Murotani, E.; Tenjimbayashi, R.-i.;
Suzuki, H. Organometallics 2004, 23, 6094−6096.
28) Kawashima, T.; Takao, T.; Suzuki, H. Angew. Chem., Int. Ed.
006, 45, 485−488.
29) (a) Mays, M. J.; Prest, D. W.; Raithby, P. R. J. Chem. Soc., Chem.
M.; Wolf, E.; Espitia, D. Organometallics 1997, 16, 2665−2673.
(e) Gervasio, G.; Marabello, D.; Sappa, E.; Secco, A. J. Cluster Sci.
(
2007, 18, 67−85. (f) Cabeza, J. A.; del Río, I.; Per
́
̃
ez-Carreno, E.;
Pruneda, V. Chem. - Eur. J. 2010, 16, 5425−5436.
(
2
(
(41) (a) Howell, J. A. S.; Mathur, P. J. Chem. Soc., Chem. Commun.
1981, 263−265. (b) Howell, J. A. S.; Mathur, P. J. Chem. Soc., Dalton
Trans. 1982, 43−50. (c) Rosenberg, E.; Kabir, S. E.; Hardcastle, K. I.;
Day, M.; Wolf, E. Organometallics 1990, 9, 2214−2217. (d) Beringhelli,
T.; D’Alfonso, G.; Minoja, A.; Ciani, G.; Moret, M.; Sironi, A.
Organometallics 1991, 10, 3131−3138. (e) Day, M.; Espitia, D.;
Hardcastle, K. I.; Kabir, S. E.; Rosenberg, E.; Gobetto, R.; Milone, L.;
Osella, D. Organometallics 1991, 10, 3550−3559. (f) Rosenberg, E.;
Freeman, W.; Carlos, Z.; Hardcastle, K.; Yoo, Y. J.; Milone, L.;
Gobetto, R. J. Cluster Sci. 1992, 3, 439−457. (g) Casey, C. P.;
Widenhoefer, R. A.; Hallenbeck, S. L.; Gavney, J. A. J. Chem. Soc.,
Chem. Commun. 1993, 1692−1694. (h) Day, M.; Espitia, D.;
Hardcastle, K. I.; Kabir, S. E.; McPhillips, T.; Rosenberg, E.;
Gobetto, R.; Milone, L.; Osella, D. Organometallics 1993, 12, 2309−
2324. (i) Casey, C. P.; Widenhoefer, R. A.; Hallenbeck, S. L.; Hayashi,
R. K.; Gavney, J. A. Organometallics 1994, 13, 4720−4731. (j) Cabeza,
Commun. 1980, 171−173. (b) Adams, R. D.; Katahira, D. A.; Yang, L.-
W. J. Organomet. Chem. 1981, 219, 85−101. (c) Adams, R. D.;
Katahira, D. A.; Yang, L.-W. J. Organomet. Chem. 1981, 219, 241−249.
(
d) Dawoodi, Z.; Mays, M. J.; Raithby, P. R. J. Organomet. Chem. 1981,
2
19, 103−113. (e) Prest, D. W.; Mays, M. J.; Raithby, P. R. J. Chem.
Soc., Dalton Trans. 1982, 2021−2028. (f) Churchill, M. R.;
Wasserman, H. J.; Belmonte, P. A.; Schrock, R. R. Organometallics
1
982, 1, 559−561. (g) Evans, W. J.; Meadows, J. H.; Hunter, W. E.;
Atwood, J. L. J. Am. Chem. Soc. 1984, 106, 1291−1300. (h) Banford, J.;
Mays, M. J.; Raithby, P. R. J. Chem. Soc., Dalton Trans. 1985, 1355−
1360. (i) Edwards, K. J.; Field, J. S.; Haines, R. J.; Homann, B.;
Sundermeyer, J.; Woollam, S. F. J. Organomet. Chem. 1990, 386, C1−
C6. (j) Fryzuk, M. D.; Piers, W. E.; Rettig, S. J. Can. J. Chem. 1992, 70,
2
381−2389. (k) Garcia Alonso, F. J.; Garcia Sanz, M.; Riera, V.; Anillo
́
̃
J. A.; del Río, I.; Goite María, C.; Perez-Carreno, E.; Pruneda, V. Chem.
Abril, A.; Tiripicchio, A.; Ugozzoli, F. Organometallics 1992, 11, 801−
8
- Eur. J. 2009, 15, 7339−7349. (k) Takao, T.; Kawashima, T.; Kanda,
H.; Okamura, R.; Suzuki, H. Organometallics 2012, 31, 4817−4831.
(42) Deabate, S.; Giordano, R.; Sappa, E. J. Cluster Sci. 1997, 8, 407−
459.
08. (l) He, Z.; Neibecker, D.; Lugan, N.; Mathieu, R. Organometallics
1
992, 11, 817−821. (m) Tada, K.-i.; Oishi, M.; Suzuki, H.; Tanaka, M.
Organometallics 1996, 15, 2422−2424. (n) Edwards, K. J.; Field, J. S.;
Haines, R. J.; Homann, B. D.; Stewart, M. W.; Sundermeyer, J.;
Woollam, S. F. J. Chem. Soc., Dalton Trans. 1996, 4171−4181.
(43) (a) Couture, C.; Farrar, D. H. J. Chem. Soc., Dalton Trans. 1987,
2245−2252. (b) Bantel, H.; Suter, P.; Vahrenkamp, H. Organometallics
1995, 14, 4424−4426.
(
o) Schollhammer, P.; Cabon, N.; Petillon, F. Y.; Talarmin, J.; Muir, K.
W. Chem. Commun. 2000, 2137−2138. (p) Cabon, N.; Petillon, F. Y.;
Schollhammer, P.; Talarmin, J.; Muir, K. W. Dalton Trans. 2004,
(44) Takahashi, Y.; Nakajima, Y.; Suzuki, H.; Takao, T. Organo-
metallics 2017, 36, 3539−3552.
2
(
708−2719.
30) Kanda, H.; Kawashima, T.; Takao, T.; Suzuki, H. Organo-
metallics 2012, 31, 1917−1926.
31) Takao, T.; Kakuta, S.; Tenjimbayashi, R.-i.; Takemori, T.;
Murotani, E.; Suzuki, H. Organometallics 2004, 23, 6090−6093.
32) Michelin Lausarot, P.; Operti, L.; Vaglio, G. A.; Valle, M.;
(45) Takao, T.; Takemori, T.; Moriya, M.; Suzuki, H. Organometallics
2002, 21, 5190−5203.
(46) Boyar, E.; Deeming, A. J.; Felix, M. S. B.; Kabir, S. E.; Adatia, T.;
Bhusate, R.; McPartlin, M.; Powell, H. R. J. Chem. Soc., Dalton Trans.
1989, 5−12.
(
(
(47) Edidin, R. T.; Norton, J. R.; Mislow, K. Organometallics 1982, 1,
561−562.
(48) The XRD study of 14 was carried out, and the formation of μ-
Tiripicchio, A.; Tiripicchio Camellini, M.; Gariboldi, P. Inorg. Chim.
Acta 1986, 122, 103−109.
(
33) (a) Zippel, T.; Arndt, P.; Ohff, A.; Spannenberg, A.; Kempe, R.;
benzylamido and μ -CO ligands were confirmed. However, owing to
3
Rosenthal, U. Organometallics 1998, 17, 4429−4437. (b) Lin, C.-J.;
the presence of disordered structures regarding the positions of the μ-
amido ligand, we could not refine the structure. The preliminary
results of the XRD study are shown in Figure S42 in the Supporting
Information.
(49) (a) Forrow, N. J.; Knox, S. A. R. J. Chem. Soc., Chem. Commun.
1984, 679−681. (b) Kang, B. S.; Koelle, U.; Thewalt, U. Organo-
metallics 1991, 10, 2569−2573.
(50) Suzuki, H.; Kakigano, T.; Tada, K.-i.; Igarashi, M.; Matsubara,
K.; Inagaki, A.; Oshima, M.; Takao, T. Bull. Chem. Soc. Jpn. 2005, 78,
67−87.
(51) PROCESS-AUTO: Automatic Data Acquisition and Processing
Package for Imaging Plate Diffractometer; Rigaku Corporation: Tokyo,
Japan, 1998.
(52) (a) Sheldrick, G. M. Acta Crystallogr., Sect. A: Found. Crystallogr.
2008, 64, 112−122. (b) Sheldrick, G. M. Acta Crystallogr., Sect. C:
Struct. Chem. 2015, 71, 3−8.
(53) Chai, J. D.; Head-Gordon, M. Phys. Chem. Chem. Phys. 2008, 10,
6615−6620.
(54) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci,
B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H.
P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.;
Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima,
T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A., Jr.;
Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin,
K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.;
Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega,
N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.;
Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.;
Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.;
Hwang, W.-S.; Chiang, M. Y. J. Organomet. Chem. 2001, 640, 85−92.
(
c) Wu, C.-Y.; Chen, Y.; Jing, S.-Y.; Lee, C.-S.; Dinda, J.; Hwang, W.-S.
Polyhedron 2006, 25, 3053−3065.
34) (a) Andrews, M. A.; Kaesz, H. D. J. Am. Chem. Soc. 1979, 101,
238−7244. (b) Andrews, M. A.; Van Buskirk, G.; Knobler, C. B.;
Kaesz, H. D. J. Am. Chem. Soc. 1979, 101, 7245−7254. (c) Donnecke,
(
7
̈
D.; Wunderle, J.; Imhof, W. J. Organomet. Chem. 2004, 689, 585−594.
(
(
35) Nakajima, Y.; Suzuki, H. Organometallics 2003, 22, 959−969.
36) Nakajima, Y.; Kameo, H.; Suzuki, H. Angew. Chem., Int. Ed.
2
(
006, 45, 950−952.
37) Nakajima, Y.; Suzuki, H. Organometallics 2005, 24, 1860−1866.
In this paper, we reported the wrong description for the hydrogenation
of 4. We confirmed quantitative formation of 1 by the H NMR
1
analysis of the residual solid, which was obtained by the removal of the
solvent. However, we overlooked the generation of N-benzylidene-
benzylamine owing to its small content. We reinvestigated the reaction
and analyzed the reaction solution by GC, which showed the
formation of a considerable amount of N-benzylidenebenzylamine in
addition to benzylamine. We correct the description of the
hydrogenation of 4 as shown ineq 4.
(
38) Adams, R. D.; Babin, J. E. Organometallics 1987, 6, 1364−1365.
(
39) Adams, R. D.; Pompeo, M. P.; Tanner, J. T. Organometallics
1
(
991, 10, 1068−1078.
40) (a) Aime, S.; Gobetto, R.; Padovan, F.; Botta, M.; Rosenberg, E.;
Gellert, R. W. Organometallics 1987, 6, 2074−2078. (b) Suss-Fink, G.;
̈
Jenke, T.; Heitz, H.; Angela Pellinghelli, M.; Tiripicchio, A. J.
Organomet. Chem. 1989, 379, 311−323. (c) Imhof, W. J. Chem. Soc.,
Dalton Trans. 1996, 1429−1436. (d) Rosenberg, E.; Milone, L.;
Gobetto, R.; Osella, D.; Hardcastle, K.; Hajela, S.; Moizeau, K.; Day,
P
DOI: 10.1021/acs.organomet.8b00165
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.;
̈
Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, O.;
Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian 09,
Revision D.01; Gaussian, Inc., Wallingford, CT, 2013.
(
55) Dennington, R.; Keith, T.; Millam, J. GaussView, version 5.0;
Semichem Inc., Shawnee Mission, KS, 2009.
Q
DOI: 10.1021/acs.organomet.8b00165
Organometallics XXXX, XXX, XXX−XXX