Trifluoromethanesulfonate (triflate) as a moderately coordinating anion: Studies from chemistry of the cationic coordinatively unsaturated mono- and diruthenium amidinates

Article abstract of DOI:10.1016/j.jorganchem.2006.08.069

Triflate complexes of mono- and diruthenium amidinates, (η⁶-C₆R₆)Ru(κ¹-OTf){η²-R′N{double bond, long}C(R′′)NR′} (1: R = Me; 2: R = H) and (η⁵-C₅Me₅)Ru(μ-η²-ⁱPrN{double bond, long}C(Me)NⁱPr)Ru(κ¹-OTf)(η⁵-C₅R₅) (3: R = Me; 4: R = H), are synthesized, and coordination behavior of the triflate anion to the coordinatively unsaturated ruthenium species is investigated by crystallography and variable temperature (VT) NMR spectroscopy (¹⁹F, ¹H). The monoruthenium amidinate complexes have three-legged piano-stool structures in single crystals, which include a κ¹-OTf ligand with the Ru-O bond of 2.15-2.20 A?. In contrast, reversible dissociation of OTf is observed in variable temperature ¹H NMR spectroscopy in liquid states; the activation energy for the dissociation and recombination of the OTf ligand is varied with the substituents on the arene and amidinate ligand in the corresponding ruthenium cation and the solvent used. A typical example of moderately coordinating ability of the OTf ligand is seen in ¹⁹F NMR spectra of (η⁶-C₆Me₆)Ru(κ¹-OTf){η²-ⁱPrN{double bond, long}C(Me)NⁱPr} (1a) and (η⁶-C₆H₆)Ru(κ¹-OTf){η²-ⁱPrN{double bond, long}C(Me)NⁱPr} (2a) in CD₂Cl₂ at the temperature range from -90 to 20 °C, in which the OTf anion is dissociated in 1a, whereas 2a has a relatively robust Ru-OTf bond. Combination of crystallography and VT NMR contributes to understanding the difference in coordination behavior of the OTf ligand between two diruthenium amidinates, (η⁵-C₅Me₅)Ru(μ-η²-ⁱPrN{double bond, long}C(Me)NⁱPr)Ru(κ¹-OTf)(η⁵-C₅Me₅) (3) and (η⁵-C₅Me₅)Ru(μ-η²-ⁱPrN{double bond, long}C(Me)NⁱPr)Ru(κ¹-OTf)(η⁵-C₅H₅) (4); the results suggest that the electron-donating and sterically demanding η⁵-C₅Me₅ helps for dissociation of the triflate ligand. Moderate coordinating ability of the triflate anion sometimes provides characteristic reactions of mono- and diruthenium amidinates which differ from the corresponding neutral halogeno-compounds or cationic coordinatively unsaturated homologues bearing fluorinated tetraarylborates; a typical example is given by inhibition of coordination of ethylene to the [(η⁶-C₆H₆)Ru{η²-^tBuN{double bond, long}C(Ph)N^tBu}]⁺ species by the OTf ligand.

Full text of DOI:10.1016/j.jorganchem.2006.08.069

Journal of Organometallic Chemistry 692 (2007) 382–394
www.elsevier.com/locate/jorganchem
Trifluoromethanesulfonate (triflate) as a moderately coordinating
anion: Studies from chemistry of the cationic coordinatively
unsaturated mono- and diruthenium amidinates
a
a
a
c
Taizo Hayashida , Hideo Kondo , Jun-ichi Terasawa , Karl Kirchner ,
Yusuke Sunada a,b, Hideo Nagashima
a,b,
*
a
Graduate School of Engineering Sciences, Kyushu University, Kasuga, Fukuoka 816-8580, Japan
Institute for Materials Chemistry and Engineering, Kyushu University, Kasuga, Fukuoka 816-8580, Japan
Institute of Applied Synthetic Chemistry, Vienna University of Technology, Getreidemarkt 9/163/AC, A-1060 Vienna, Austria
b
c
Received 28 February 2006; accepted 18 April 2006
Available online 30 August 2006
Abstract
Triflate complexes of mono- and diruthenium amidinates, (g6-C6R6)Ru(j1-OTf){g2-R0N@C(R0 0)NR0} (1: R = Me; 2: R = H) and
(g5-C5Me5)Ru(l-g2-iPrN@C(Me)NiPr)Ru(j1-OTf)(g5-C5R5) (3: R = Me; 4: R = H), are synthesized, and coordination behavior of
the triflate anion to the coordinatively unsaturated ruthenium species is investigated by crystallography and variable temperature (VT)
NMR spectroscopy (19F, 1H). The monoruthenium amidinate complexes have three-legged piano-stool structures in single crystals, which
1
˚
include a j -OTf ligand with the Ru–O bond of 2.15–2.20 A. In contrast, reversible dissociation of OTf is observed in variable temperature
1H NMR spectroscopy in liquid states; the activation energy for the dissociation and recombination of the OTf ligand is varied with the
substituents on the arene and amidinate ligand in the corresponding ruthenium cation and the solvent used. A typical example of mod-
erately coordinating ability of the OTf ligand is seen in 19F NMR spectra of (g6-C6Me6)Ru(j1-OTf){g2-iPrN@C(Me)NiPr} (1a) and
(g6-C6H6)Ru(j1-OTf){g2-iPrN@C(Me)NiPr} (2a) in CD2Cl2 at the temperature range from ꢀ90 to 20 ꢀC, in which the OTf anion is dissociated
in 1a, whereas 2a has a relatively robust Ru–OTf bond. Combination of crystallography and VT NMR contributes to understanding the
difference in coordination behavior of the OTf ligand between two diruthenium amidinates, (g5-C5Me5)Ru(l-g2-iPrN@C(Me)NiPr)-
Ru(j1-OTf)(g5-C5Me5) (3) and (g5-C5Me5)Ru(l-g2-iPrN@C(Me)NiPr)Ru(j1-OTf)(g5-C5H5) (4); the results suggest that the electron-
donating and sterically demanding g5-C5Me5 helps for dissociation of the triflate ligand. Moderate coordinating ability of the triflate
anion sometimes provides characteristic reactions of mono- and diruthenium amidinates which differ from the corresponding neutral halo-
geno-compounds or cationic coordinatively unsaturated homologues bearing fluorinated tetraarylborates; a typical example is given by
inhibition of coordination of ethylene to the [(g6-C6H6)Ru{g2-tBuN@C(Ph)NtBu}]+ species by the OTf ligand.
ꢁ 2006 Elsevier B.V. All rights reserved.
Keywords: Ruthenium amidinate; Triflate; Coordinatively unsaturated
1. Introduction
ClOꢀ4 , PF6ꢀ, and BFꢀ4 and non-classical ‘‘larger and more
weakly coordinating’’ anions such as fluorinated tetraa-
rylborates, anionic methylaluminoxane, and carborane
anions [1]. This area has lately expanded to homogeneous
catalysis; in particular, their deep knowledge is necessary
for developing new metallocene catalysts for industrially
important olefin polymerization processes [2]. Trifluorome-
Importance of weakly coordinating anions in organome-
tallic chemistry has been widely recognized in a number of
review articles, which covers fundamentals in coordination
behavior of classical ‘‘weakly coordinating’’ anions such as
thanesulfonate (triflate; CF3SOꢀ; TfOꢀ) has long been
*
Corresponding author. Tel./fax: +81 925837819.
3
believed as a classical ‘‘non-coordinating’’ anion, and in
E-mail address: nagasima@cm.kyushu-u.ac.jp (H. Nagashima).
0022-328X/$ - see front matter ꢁ 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2006.08.069

T. Hayashida et al. / Journal of Organometallic Chemistry 692 (2007) 382–394
383
fact, many transition metal triflates act as a cationic coord-
inatively unsaturated species, which allow facile coordina-
tion of other auxiliary ligands such as water, pyridine,
and phosphines to result in throwing away the triflate
anion out of the coordination sphere: M–OTf + L !
M(L)+ (OTf)ꢀ. However, it has been recognized that tri-
flate has relatively strong coordinating ability to the metal
center similar to ClO4ꢀ compared with PF6ꢀ, BFꢀ4 , and fluo-
rinated tetraarylborates; many complexes having metal–
OTf bonds have been isolated and characterized [1]. The
progress in understanding the weakly coordinating anions
has given unique insights in differences in the coordination
behavior among both classical and non-classical weakly
coordinating anions; for example, chemistry of transition
metal Lewis acids has provided catalysis for asymmetric
Diels–Alder reactions, of which rate and selectivity are
dependent on the weakly coordinating counter anions used
[3]. In this context, systematic studies on the organometal-
lic chemistry of weakly coordinating anions bound to a ser-
ies of cationic transition metal species with similar
structures should be important for understanding the effect
of weakly coordinating anions; in particular, triflate having
a relatively strong coordination ability compared with
other weakly coordinating anions is of interest to be
investigated.
R
R
R
R
R
R
R
R
'R R
N
Ru
R'
Ru
Ru
N
R
OTf
R
N
OTf
iPr
N
iPr
R"
1a: R=Me, R'=iPr, R"=Me
2a: R=H, R'=iPr, R"=Me
2b: R=H, R'=iPr, R"=Ph
2c: R=H, R'=tBu, R"=Ph
Me
3: R = Me
4: R = H
Fig. 1. Triflate complexes of ruthenium amidinates.
atively unsaturated form in less-polar solvents: M–
OTf « M+ + ꢀOTf. The equilibrium would shift to the
right side, when M+ is stable. If not very stable, the equilib-
rium should be favored for the left side, so that the unsta-
ble M+ is protected by bonding with ꢀOTf. Our cationic
coordinatively unsaturated ruthenium amidinate com-
plexes are good candidates to discuss about the triflate
chemistry from this side [5]. Coordinatively unsaturated
yet isolable cationic ruthenium complexes, [(g6-C6R6)-
Ru{g-R0N@C(R0 0)NR0}]+Xꢀ (R = Me, H, X = PF6,
BF4, TFPB), were synthesized from the corresponding neu-
tral halogeno complexes [5b]. In contrast to the three-leg-
ged piano-stool structures of the halogeno-precursors [6],
center of the C6Me6 ligand, the ruthenium atom, and the
two nitrogen atoms are in a plane in the molecular struc-
Chemistry of j1-OTf complexes of ruthenium has
received considerable attention of organometallic chemists
since mid 90s; molecular structures of several complexes
have been published, and discussion on a possible existence
of j1-OTf species in solution was performed using NMR
techniques [4]. A typical example is [Ru(OTf)(60-diph-
enylphosphino-1-naphthyl)(PPh2OH)]OTf [4h], of which
crystallography showed existence of two triflates; one is
bonded to the ruthenium center with the Ru–O bond dis-
þ
ꢀ
ture of ½ðg6-C6Me6ÞRuf PrN@C(Me)NiPrgꢁ PF6 , and
possible interaction of the central carbon of the amidinate
ligand to the ruthenium center is indicative of p-stabiliza-
tion to compensate the coordinative unsaturation. Coord-
inatively unsaturated diruthenium amidinates, [(g5-C5Me5)-
Ru(l-g2-iPrN@C(Me)NiPr)Ru(g5-C5R5)]+Xꢀ (R = H, Me;
X = TFPB, PF6, BF4, etc.), are also synthesized from the
corresponding halogeno-precursors, and crystallographic
studies showed that the molecular structures of [(g5-
C5Me5)Ru(l-g2-iPrN@C(Me)NiPr)Ru(g5-C5Me5)]+Xꢀ
differ from the corresponding halogeno-precursors [5c,5d].
The aim of this paper is preparation and characterization
of triflate complexes of these coordinatively unsaturated
ruthenium amidinates, 1–4 (Fig. 1). As indicated from the
moderate coordination ability of triflate, the OTf is bonded
to the ruthenium center in crystal structures. In contrast,
i
˚
tance of 2.155(4) A, whereas the other is out of the coordi-
nation sphere. Two distinct 19F signals due to the triflates
are seen at room temperature; one is assigned to the j1-
OTf (d ꢀ78.2), whereas the other is due to the ꢀOTf (d
ꢀ79.5) in CD2Cl2. In many other j1-OTf ruthenium com-
plexes, the Ru–O bond distance in the crystal structures
is ca. 2.15–2.25 A, the 19F resonance in solution appears
˚
in lower fields than that of the free ꢀOTf [4]. Although
these data indicate that the triflate anion is tightly bound
to the cationic ruthenium center, treatment of the j1-OTf
ruthenium complexes with several donor ligands result in
their coordination to the cationic ruthenium center accom-
panied by movement of the triflate out of the coordination
sphere. These are good examples suggesting the triflate is a
moderately weak coordinating anion to the cationic ruthe-
nium species; the Ru–O interaction is strong enough to
detect by crystallography and 19F NMR spectroscopy but
weak enough to react with other ligands.
solution dynamics in H and 19F NMR spectroscopy indi-
1
cates how facile is the dissociation of the OTf, that depends
on the nature of the ruthenium cation. The moderate coor-
dination ability also affects the reactivity of the triflate
complexes, which sometimes differs from both the neutral
halogeno-precursor and the cationic coordinatively unsatu-
rated complexes.
2. Results and discussion
We are interested in coordination chemistry of triflate
from the side of cationic ruthenium species. When we con-
sider the dissociation (or solvent-assisted dissociation) of
the M–OTf bond, an important factor should be stability
of cationic species M+, which may exist as the coordin-
2.1. Synthesis and the molecular structures of triflate
complexes of mono- and diruthenium amidinates
Triflate complexes of monoruthenium amidinates, (g6-
C6R6)Ru(j1-OTf){g2-R0N@C(R0 0)NR0} (1a, 2a, 2b, and

384
T. Hayashida et al. / Journal of Organometallic Chemistry 692 (2007) 382–394
2c; see Fig. 1) were synthesized from the corresponding
halogeno-precursors, (g6-C6R6)Ru(X){g2-R0N@C(R0 0)NR0}
(5, 6), by treatment with AgOTf in CH2Cl2. The diruthe-
nium amidinates, [(g5-C5Me5)Ru(l-g2-iPrN@C(Me)NiPr)-
Ru(g-OTf)(g5-C5Me5)] (3) and [(g5-C5Me5)Ru(l-g2-iPrN@
C(Me)NiPr)Ru(g-OTf)(g5-C5H5)] (4; R = H), were pre-
pared from the corresponding halogeno-precursors by
treatment with AgOTf and TlOTf, respectively. As
reported previously, the TFPB homologues of these triflate
complexes are coordinatively unsaturated, showing intense
blue to violet color [5b]. In contrast, the halogeno-homo-
logues of these complexes are orange or red and neutral.
From X-ray structure determination of several mono-
and diruthenium amidinates, these coordinatively
saturated and unsaturated complexes can typically be illus-
trated as shown in Fig. 2, where the angle h is defined by
the centroid of the ring (the arene or cyclopentadienyl
ring)—the Ru atom with the plane consisting of the two
nitrogen atoms of the amidinate ligand and the ruthenium
atom. It is clearly seen from the value of h that the coord-
inatively saturated monoruthenium amidinates A have typ-
ical three-legged piano-stool structures (h = ca. 145ꢀ) [6,7]
(structural details of 9-OTf and 10-OTf are discussed
later), whereas the coordinatively unsaturated analogue B
is a two-legged piano-stool (h = 174ꢀ) [5]. Structural fea-
tures of diruthenium amidinates are similar to those of
monoruthenium amidinates; one of the ruthenium atoms
is close to A with three-legged piano-stool structures in
the coordinatively saturated C (h = 156–160ꢀ), whilst those
of the coordinatively unsaturated analogues D are close to
B with two-legged piano-stool structures (h = 172–174ꢀ).
The molecular structures of 1a, 3, and 4 were determined
by crystallography. A single crystal suitable for X ray struc-
ture determination of 2a was unfortunately not obtained;
however, its molecular structure is believed to be similar
to that of the analogous compound 2c. The ORTEP draw-
ings (Fig. 3) revealed that the ligand arrangement around
the ruthenium is similar to that seen in the coordinatively
saturated halogeno-analogues; this is clearly demonstrated
by the angle h [148.9ꢀ (1a), 138ꢀ (2c), 160.3ꢀ (3), and
154.9ꢀ (4)]. The Ru–O bond distance is another indication
of coordinatively unsaturated nature, though the crystallog-
raphy contains some influence of crystal packing. However,
comparison in the bond distances and angles among the
molecules with similar structures seems to have some sense
for discussion. The three-legged piano-stool structures of 1a
and 2c, suggest the coordinative saturation by the tightly
bound OTf ligand; however, the longer Ru–O bond of 1a
˚
[2.199(3) A] may suggest the weaker Ru–O interaction than
˚
that seen in 2c [Ru–O = 2.155(5) A]. The electron-donating
property of the C6Me6 ligand in 1a possibly contribute to
stabilizing the coordinatively unsaturated ruthenium cat-
ion, giving the weaker Ru–O interaction. The diruthenium
amidinate 4 has a slightly longer Ru–O bond distance
˚
[Ru–O = 2.213(5) A] than those seen in 1a and 2c, suggest-
ing the coordinatively saturated nature in a single crystal. In
sharp contrast, 3 has a significantly longer Ru–O distance
˚
(2.360(2) A), and relatively large Ru–O–S angle (150.63ꢀ).
These suggest that 3 may have some coordinatively unsatu-
rated nature even in a single crystal. This is inconsistent
with the angle h not close to those of [(g5-C5Me5)
Ru(l-g2-iPrN@C(Me)NiPr)Ru(g5-C5Me5)]+Yꢀ [Y = PF6,
B(C6F5)4], but in accord with an intense purple color of 3
in solid states. Steric circumstances in 3, in which the bulky
C5Me5 group and isopropyl moieties interfere the coordina-
tion of OTf, is one explanation of this unusual bond dis-
tance and angle, whereas electron-donation by the C5Me5
groups is another reason. In a similar crystal structure
reported in trans-[Ru(OTf)(CN)(dppe)2] having a long
˚
Ru–O distance of 2.410(5) A and large Ru–O–S angle
(160.5(3)ꢀ), strong trans-influence of the CN ligand and ste-
ric circumstances are discussed as the possible explanation
[4l] (Tables 1 and 2).
In summary, the data of the angle h suggests that the tri-
flate ligand has relatively strong coordinating ability to give
the crystal structures close to the neutral halogeno-homo-
logues, whereas electronic and steric circumstances provide
the variation of the Ru–OTf interaction, which is estimated
from the Ru–O bond distance of each complex.
R
OTf
PF6
θ
N
C
θ
N
C
Ru
Ru
Ru
N
θ
X
L
C
θ = 145o
For several complexes
where X=halogen (left)
and 9-OTf, 10-OTf (right)
θ = 173.4o
For (η6-C6Me6)Ru+(η-amidinate)
amidinate = iPrNC(Me)=NiPr
2.2. Variable temperature NMR studies of triflate complexes
of mono- and diruthenium amidinates
R
θ
θ
R
Y
R
R
R
Ru
Ru
In contrast to the crystal structures suggesting the
coordinatively saturated nature in solid states, it is known
that solution dynamics measured by variable temperature
NMR spectroscopy can predict the coordinating ability
of OTf in liquid states. There are several factors which
affect the solution dynamics, steric and electronic proper-
ties of the auxiliary ligands, e.g. C6H6 vs. C6Me6 (1 vs.
2), C5Me5 vs. C5H5 (3 vs. 4), and substituents of the amidi-
nate ligands (2), and the polarity and coordinating ability
Cl
Ru
Ru
N
N
C
C
θ = 174.4o (Y=PF6)
θ = 159.9o (R=Me)
θ = 155.8o (R=H)
θ = 171.8o [Y=B(C6F5)4]
For diruthenium amidinates
amidinate = iPrNC(Me)=NiPr
For diruthenium amidinates
amidinate = iPrNC(Me)=NiPr
Fig. 2. Dihedral angles h.

T. Hayashida et al. / Journal of Organometallic Chemistry 692 (2007) 382–394
385
Fig. 3. The ORTEP drawings of 1a, 2c, 3, and 4.
Table 2
˚
Representative bond distances (A) and angles (ꢀ) of diruthenium amidi-
nates 3 and 4
3
4
˚
Bond distances (A)
Table 1
Ru(1)–Ru(2)
Ru(1)–O
2.9135(2)
2.360(2)
1.7926(2)
1.8320(2)
2.133(2)
2.132(2)
2.092(2)
2.122(3)
2.125(3)
1.381(4)
1.379(4)
2.7589(5)
2.213(3)
1.7899(4)
1.7998(4)
2.116(3)
2.119(3)
2.091(3)
2.098(3)
2.131(4)
1.372(5)
1.378(5)
˚
Representative bond distances (A) and angles (ꢀ) of monoruthenium
amidinates, 1a, 2c, 9-OTf, and 10-OTf
Ru(1)–R1
Ru(2)–R2
1a
2c
9-OTf
10-OTf
˚
Bond distances (A)
Ru(1)–N(1)
Ru(1)–N(2)
Ru(2)–N(1)
Ru(2)–N(2)
Ru(2)–C(1)
N(1)–C(1)
N(2)–C(1)
Ru–O(1)
Ru–N(3)
Ru–C(2)
Ru–N(1)
Ru–N(2)
Ru–C(1)
2.199(3)
2.155(5)
2.151(4)
1.956(4)
2.112(4)
2.112(4)
2.546(5)
2.083(3)
2.083(3)
2.55(3)
2.092(3)
2.122(4)
2.561(3)
2.093(3)
2.087(4)
2.57(3)
C(2)–N(3)
Ru–C6 ring (centroid)
N(1)–C(1)
N(2)–C(1)
1.151(6)
1.704(1)
1.318(5)
1.328(5)
Bond angles (ꢀ)
h1
h2
1.678(3)
1.329(8)
1.324(7)
1.662(1)
1.318(5)
1.318(5)
1.7475(3)
1.328(5)
1.327(5)
161.81(4)
123.95(4)
118.80(7)
79.22(9)
84.95(8)
61.48(9)
104.3(2)
154.71(6)
115.06(7)
125.07(9)
81.5(1)
81.3(1)
61.4(1)
h3
N(1)–Ru(1)–O(1)
N(2)–Ru(1)–O(1)
N(1)–Ru(1)–N(2)
N(1)–C(1)–N(2)
Bond angles (ꢀ)
N(1)–Ru–N(2)
N(1)–Ru–O(1)
N(2)–Ru–O(1)
N(1)–Ru–N(3)
N(2)–Ru–N(3)
N(1)–Ru–C(2)
N(2)–Ru–C(2)
N(1)–C(1)–N(2)
62.1(2)
86.5(2)
87.0(2)
61.9(2)
81.2(2)
81.2(2)
61.9(1)
62.0(1)
103.7(3)
87.4(2)
84.3(1)
R2
R1
θ2
θ1
Ru2 Ru1 θ3
R
1 = center of Cp or Cp*
R2 = center of Cp*
85.5(2)
86.0(2)
108.9(4)
N1
C1
N2
O(1)
110.4(5)
108.6(5)
109.5(4)

386
T. Hayashida et al. / Journal of Organometallic Chemistry 692 (2007) 382–394
Ru
Ru
N
Ru
iPr
Ru
N
Ru
Ru
iPr
iPr
N
iPr
N
N
N
iPr
iPr
Me
Me
Me
R
R
R
R
R
R
R
R
R
R
R
R
R
R
R
R
R
'R
Ru
'R R
Ru
R'
Ru
R'
N
N
N
N
N
N
R"
R'
R"
R'
R"
E
R
R
R
R
R
R
R
R
R
R
R
R
R
R
R
R
R
R
R
R
R
OTf
R
R
R
R
- OTf
R
'R
+ OTf
- OTf
R
R
N
N
Ru
Ru
Ru
R'
R'
'R R
'R R
Ru
R'
Ru
R'
N
N
TfO
N
N
N
OTf
+ OTf
N
N
N
R'
R"
R"
R"
R"
R'
R'
R"
OTf
- OTf
Ru Ru
Ru Ru
+ OTf
- OTf
Ru
Ru
N
Ru
Ru
N
Ru Ru
OTf
iPr
iPr
iPr
N
N
N
N
iPr
Me
iPr
TfO
iPr
N
N
N
N
+ OTf
iPr
iPr
iPr
iPr
Me
Me
Me
Me
Fig. 4. Simplified schemes for flipping of the amidinate ligand in coordinatively unsaturated complexes vs. triflate complexes. A possibility that the
cationic species may be stabilized by solvation or coordination of the solvents is not taken into consideration in these schemes for clarity.
of the solvent used. An advantage of the mono- and diru-
thenium amidinates 1–4 is that coordinatively unsaturated
cationic ruthenium species formed by dissociation of ꢀOTf
are stable enough to isolate as the TFPB anion. As dis-
cussed in our previous papers [5a,5b,5c,5d], the amidinate
ligands of all of these coordinatively unsaturated TFPB
complexes potentially show dynamic behavior as shown
in Fig. 4 (blue scheme),1 in which the flipping of the amidi-
nate ligand leads to conversion of a pseudo-three-legged
piano-stool form to the other through the transition state
with a two-legged piano-stool structure. In fact, variable
vent the flipping even below ꢀ78 ꢀC. The crystal structures
suggesting the coordination of the OTf ligand to 1, 2, and 3
indicate that the coordination may stabilize the pseudo-
three-legged form. Thus, the flipping of the amidinate
ligand takes place from the j1-OTf three-legged piano-
stool form to its enantiomer through a cationic two-legged
piano-stool transition state; this provides substantial
increase of activation energy for as shown in Fig. 4 (red
1
scheme). In other words, solution dynamics of H NMR
of the triflate homologues of 1, 2, and 3 may provide good
evidence that the OTf ligand interferes the flipping of the
amidinate ligand by coordination.
1
temperature H NMR spectroscopy of the TFPB homo-
logues of 1, 2, and 3 showed rapid flipping of the amidinate
ligand, in which the activation energy is low enough to pre-
1H NMR studies on 1a, 2a and 2b clearly showed that
the OTf ligand prevents the flipping at low temperatures.
A typical example for the dynamism is seen in the diastero-
topic Me signals of isopropyl group on the amidinate
1
For interpretation of the references in color in Fig. 4, the reader is
1
ligand in 2a in H NMR spectrum (Scheme 1). In CD2Cl2,
referred to the web version of this article.

T. Hayashida et al. / Journal of Organometallic Chemistry 692 (2007) 382–394
387
R
R
R
OTf
Ru
H
H
Ru
H
Ru
TfO
OTf
Me1
Me2
Me1
Me2
C
Me1
Me2
C
C
OTf
Ru
Ru
Ru
Ru
Me2
Ru
Ru
H
H
H
OTf
TfO
Me1
Me2
Me1
Me1
Me2
C
C
C
Scheme 1. Dynamisms of the isopropyl group.
two methyl signals appeared independently at ꢀ90 ꢀC, coa-
lesced at ꢀ30 ꢀC, and became a single doublet at 25 ꢀC
(Fig. 5). Similar spectral change was seen at 20, 40, and
80 ꢀC, respectively, in toluene-d8. These are due to revers-
ible coordination of OTf providing the flipping of the ami-
dinate ligand as shown in Scheme 1; the calculated DGs are
dependent on the solvent used, 11.4 (CD2Cl2) and 15.7 (tol-
uene-d8) kcal/mol (Table 3). Similar experiments on 1a and
2b revealed that DG is dependent on the substituents on the
arene ring [DG(g6-C6H6) > DG(g6-C6Me6)], those of the
amidinate group [DG(g-iPrNC(Ph)NiPr) > DG(g-iPrNC-
(Me)NiPr)]. As discussed in the first paper showing evidence
of the coordinating ability of triflate to ruthenium, the cat-
ionic ruthenium species formed by dissociation of OTf is
possibly stabilized by the solvent. In extreme cases using
polar and coordinating solvents, dissociation of OTf are
assisted by coordination of the solvent, whereas dissocia-
tion of OTf occurs spontaneously and the solvent plays a
small role for the stabilization of the cationic species in
the other cases using less-polar and weakly coordinating
solvents. The difference in DG between that in CD2Cl2
and in toluene-d8 may reflect polarity and coordinating
ability of CD2Cl2 to the ruthenium cation. The result of
DG(g6-C6H6) > DG(g6-C6Me6) can be explained by stabil-
ization of the cationic coordinatively unsaturated ruthe-
nium species by more electron-donating g6-C6Me6 ligand.
The stabilized ruthenium cation does not need further sta-
bilization by coordination of OTf. The result of
DG(g-iPrNC(Ph)NiPr) > DG(g-iPrNC(Me)NiPr) may be
attributed to bulkiness of the phenyl group in 2b compared
with the methyl group in 2a, which protect the coordin-
atively unsaturated ruthenium center from the coordina-
tion of OTf. Resonance stabilization by the phenyl group
is difficult to consider, because the crystal structure of 2c
suggests that the phenyl group of 2b is perpendicular to
the amidinate plane.
Table 3
The coalescent temperature on the 1H NMR and the calculated activation
energy of the reversible coordination of the triflate to [(g6-C6R6)Ru(g2-
amidinate)]+
Complex
In CD2Cl2
In toluene-d8
Coalescent
temperature
(ꢀC)
DGà
(kcal/mol)
Coalescent
temperature
(ꢀC)
DGà
(kcal/mol)
1a
2a
2b
<ꢀ100
ꢀ30
Not determined
11.4
Not determined
0
40
100
13.0
15.7
18.6
>40
Fig. 5. Variable temperature 1H NMR signals for diastereotopic Me protons of the N–iPr group of 2a in CD2Cl2 (left) and in toluene-d8 (right).

388
T. Hayashida et al. / Journal of Organometallic Chemistry 692 (2007) 382–394
The effect of the arene ring between 1a and 2a was further
tronic nature and steric circumstances of the corresponding
coordinatively unsaturated cationic species. Among the tri-
flates we examined, the complexes, (g6-C6H6)Ru(j1-
OTf){g2-R0N@C(R0 0)NR0} 2a, 2b, and 2c, have the most
strongly bound OTf ligand to the ruthenium center; this
is evidenced by 19F resonance assignable to the coordinated
triflate at the temperature range from 20 to ꢀ70 ꢀC. VT-1H
NMR also supports this. The relatively strong coordina-
tion of triflate to the metal center in 2a–2c may provide dif-
ferent reactivity of 2a–2c with auxiliary ligands from that
of the corresponding coordinatively unsaturated TFPB
homologues. For the examination of this, we carried out
the reactions of several ligands with 2c.
investigated by variable temperature 19F NMR. In several
earlier reports are discussed chemical shifts of 19F resonances
due to the coordinated and uncoordinated triflate ligand: the
latter appears around ꢀ79 ppm in CD2Cl2, and the former is
observed at downfield by 1–2 ppm. A single 19F peak was
seen at ꢀ78.95 ppm for 1a at room temperature; no spectral
change was observed even at ꢀ90 ꢀC. In contrast, 2a pro-
vided a single 19F signal at room temperature at
ꢀ77.25 ppm; a small broad peak appeared at ꢀ78.1 ppm
below ꢀ50 ꢀC, which became sharp as the temperature down
to ꢀ70 ꢀC. The peak ratio of the former (ꢀ77.25) to the latter
(ꢀ78.1) is 9:1. We consider the new peak at ꢀ78.1 ppm to be
uncoordinated ꢀOTf, which is in equilibrium with the coor-
dinated OTf seeing at ꢀ77.25 ppm. Thus, the triflate is unco-
ordinated in a CD2Cl2 solution of 1a, whereas that is mainly
coordinated and only partly dissociated in that of 2a. This is
consistent with DG(g6-C6H6) > DG(g6-C6Me6).
As described in our previous paper, [(g6-C6H6)Ru-
{g2-tBuN@C(Ph)NtBu}]+(TFPB)ꢀ (7) showing intense
blue color reacted with PPh3, pyridine, CNtBu, CO, and
CH2@CH2 to give the corresponding adducts almost
instantly [5b]. The reactivity of 2c was smaller than 7. As
shown in Scheme 2, PPh3, pyridine, and CNtBu reacted
smoothly with 2c to give the corresponding adducts,
whereas treatment of CO with 2c results in decomposition
of the product (vide infra). The products were characterized
by spectroscopic methods, and X-ray structure determina-
tion of two of the compounds, 9-OTf and 10-OTf, showed
their three-legged piano-stool structures (Fig. 6). Of impor-
tance is the reaction of 2c with ethylene having the weakest
donor property among we examined, which afforded no
adduct. In other words, ꢀOTf is a stronger ligand than eth-
ylene to [(g6-C6H6)Ru{g2-tBuN@C(Ph)NtBu}]+; ethylene
cannot replace the triflate on the ruthenium center due to
the weak coordination ability. This clearly demonstrates
moderately weak coordinating ability of triflate, which dif-
fers from other weakly coordinated anions. There are num-
ber of reports suggesting differences in reactivity between
transition metal halide complexes and their triflate homo-
logues, which are classified into ‘‘neutral’’ and ‘‘cationic’’,
respectively; the former is less reactive than the latter. In
fact, we have preliminary results showing that the reactions
of neutral halogeno-homologues of 2 with pyridine, CO,
and CH2@CH2 gave no adduct, though the adduct forma-
tion was detectable by NMR in the reactions with PPh3
Similar NMR experiments were done with diruthenium
amidinates 3 and 4. Although the l-amidinate ligand in 4
does not show any dynamic behavior, that in 3 and its
TFPB homologue is rapidly flipping in solution. Variable
temperature 19F NMR of 3 in CD2Cl2 showed only a signal
due to the uncoordinated OTf, whereas that of 4 provide a
major and minor peak at ꢀ78.2 and ꢀ79.3 ppm, respec-
tively, below ꢀ20 ꢀC, suggesting that the triflate is mainly
coordinated and partly dissociated. The results can be
explained by better donor property and steric bulkiness
of the C5Me5 ligand in 3, which give favorable influence
for the stabilization of cationic coordinatively unsaturated
diruthenium amidinate species.
2.3. Representative reactions showing the moderate
coordinating ability of ꢀOTf
As described above, the crystal structures and solution
dynamics of the triflate complexes of mono- and diruthe-
nium amidinates indicate that the triflate anion is coordi-
nated in a j1-mode in solid states, whereas it reversibly
dissociates in liquid states. Of importance is how easy is
the dissociation, which apparently depends on the elec-
tBu
Ru
tBu
L
tBu Ru
TFPB
Ru
or
N
OTf or
TFPB
N
OTf
tBu
L
CH2Cl2
N
N
N
N
Ph
tBu
Ph
tBu
Ph
TFPB homologueb
L
the OTf complex 2c)
t
(
he
84%
98%
94%
82%
PPh3
pyridine
tBuNC
CO
89%
98%
90%
60%
decomposed
No reaction
ethylene
a. Isolated yields are listed. b. see ref 5b.
Scheme 2. Reaction of 2c and its TFPB homologue with auxiliary ligands.

T. Hayashida et al. / Journal of Organometallic Chemistry 692 (2007) 382–394
389
Fig. 6. The molecular structures of 9-OTf (left) and 10-OTf (right).
and CNtBu [6c]. In contrast to this, the reactivity difference
between the triflate complexes and their TFPB homologues,
which are both classified into the ‘‘cationic complexes’’ has
not been fully investigated in organometallilc chemistry yet.
The above results clearly showing the reactivity order of
[Ru]+[TFPB]ꢀ > [Ru]d+[OTf]dꢀ > Ru–X, are one of the
rare example demonstrating the moderately weak coordi-
nating ability of the triflate anion.
trast, the complexes which are not well stabilized by the
auxiliary ligands require the tight coordination of the tri-
flate for the stabilization of the complexes; these are visible
in the complexes 2 and 4. In the extreme case, the triflate
tightly bound to the ruthenium inhibits its replacement
by a weak ligand, ethylene; this clearly predicts that use
of the triflate should be careful when the homogeneous
catalysis include cationic ruthenium intermediates and acti-
vation of weakly coordinating substrates such as olefins.
These aspects have first been provided by nice contribution
of organometallic chemistry of coordinatively unsaturated
ruthenium amidinates, and would contribute to chemistry
of the weakly coordinating anion from fresh insights.
3. Conclusion
Cationic transition metal species plays an important role
in some of the homogeneous catalysis, and it has become
popular that coordinating ability of the corresponding
counter anion sometimes affects the catalytic activity and
selectivity [2,3]. Such counter anion effect is well investi-
gated in highly oxophilic cationic early transition metal
complexes [2], in particular, in terms of mechanisms for
olefin polymerization by metallocene catalysts; however,
less oxophilic late transition metal cations have relatively
been ill investigated [3]. Among studies on the triflate com-
plexes of ruthenium, which mainly dealt with crystal struc-
tures and spectroscopy, the present report is unique to let
the readers understand how the triflate is bound to the
metal center from both crystal structures, solution dynam-
ics, and reactions with auxiliary ligands. The particular
uniqueness is attributed to the cationic mono- and diruthe-
nium amidinates 1–4, which can be isolated as coordin-
atively unsaturated forms by using TFPB as the counter
anion. In this context, comparison in the structures,
dynamics, and reactions between the triflate complexes 1–
4 with their TFPB homologues, as well as that between
the triflate complexes with coordinatively saturated halo-
geno-homologues, provides many clues for understanding
the coordinating ability of the triflate anion, which appar-
ently differs from both halides and TFPB. The results
described in this paper clearly demonstrated the nature of
the triflate anion as a moderately coordinating anion,
which facilely dissociates from the metal center when the
counter cationic species is stabilized by electronic or steric
reasons; these are seen in the complexes 1 and 3. In con-
4. Experimental
4.1. General methods
Manipulation of air- and moisture sensitive organome-
tallic compounds was carried out under a dry argon atmo-
sphere using standard Schlenk techniques associated with a
high-vacuum line and a nitrogen-filled glove-box. All sol-
vents were distilled over appropriate drying reagents prior
to use (toluene, pentane, Et2O; Ph2CO/Na:MeCN,
CH2Cl2; CaH2:acetone; MS4A). 1H, 13C, 19F, and 31P
NMR spectra were recorded on a JEOL Lambda 400 or
600 MHz spectrometer. The chemical shifts were recorded
in ppm relative to the solvent signal or standard references
(19F; external C6F6, 31P; external 85% H3PO4). IR spectra
were measured on a JASCO FT/IR-550 spectrometer.
Melting points were measured on a Yanaco micromelting
point apparatus. ESI mass spectra were recorded on a
JEOL JMS-T100CS spectrometer. Elemental analyses were
preformed by the Elemental Analysis center, Faculty of
Science, Kyushu University.
4.2. Preparation of (g6-C6R6)Ru(j1-OTf)
{g2-R0N@C(R0 0)NR0} (1a, 2a, 2b, and 2c)
The triflates of monoruthenium amidinates, 1a, 2a, 2b,
and 2c, were prepared by treatment of AgOTf with the

390
T. Hayashida et al. / Journal of Organometallic Chemistry 692 (2007) 382–394
corresponding halogeno-precursors, 5a-Br, 6a-Br, 6b-Cl,
and 6c-Cl, respectively. In a typical example, the complex
6c-Cl (69 mg, 0.16 mmol) and AgOTf (41 mg, 0.16 mmol)
were dissolved in CH2Cl2 at ꢀ78 ꢀC. The mixture was
slowly warmed to room temperature, and stirred for 1 h.
Insoluble AgCl was filtered off, and the filtrate was concen-
trated. The crude product was recrystallized from a mix-
ture of CH2Cl2 and pentane at ꢀ35 ꢀC to give 2c as
air-sensitive red-orange crystals (58 mg, 0.10 mmol, 67%).
cal example, synthesis of Li{iPrN@C(Ph)NiPr} was per-
formed by a similar process to that reported for
Li{tBuN@C(Ph)NtBu}. In a Schlenk tube, a cyclohex-
ane–Et2O solution of PhLi (0.865 M, 6.8 mL, 5.9 mmol)
i
was added to a Et2O (50 mL) solution of PrN@C@NiPr
at 0 ꢀC. The mixture was warmed to room temperature
and stirred for 2 h. Removal of solvents in vacuo, the
formed air- and moisture sensitive white solids (1.35 g,
>99%) were stored in a glove box, and used for the next
1
1a: yield 64%, mp 148 ꢀC (dec). ESI-TOF: MꢀꢀOTf:
step without purification. H NMR (400 MHz, THF-d8) d
12
405.18. Exact mass (ESI-TOF): Calcd. for
C
1H35-
0.88 (br, 12H, CHMe2), 2.96 (br, 2H, CHMe2), 7.04 (br,
2H, Ph), 7.17 (br, 1H, Ph), 7.27 (m, 2H, Ph). 13C NMR
(100 MHz, CD2Cl2) d 25.32 (br, CHMe2), 67.40 (CHMe2),
126.57 (br, Ph), 127.94, 128.32, 129.03 (Ph). The 13C res-
onance due to the central carbon of the amidinate ligand
was not observed. In THF (ca. 20 mL) were dissolved
[(g6-C6H6)RuCl2]2 (265 mg, 1.06 mmol) and Li{iPr-
N@C(Ph)NiPr} (224 mg, 1.07 mmol), and the solution
was stirred for 5 h at room temperature. The mixture
was concentrated, and the residue was suspended in tolu-
ene. After insoluble materials were filtered off, the vola-
tiles were removed in vacuo to form 6b-Cl as orange
solids (354 mg, 79%). The lithium salt, Li{iPrN@C(Me)-
NiPr}, contains substantial amounts of LiBr unless other-
wise salt-free MeLi was used. The reaction of
Li{iPrN@C(Me)NiPr} containing LiBr with [(g6-C6R6)-
RuCl2]2 was accompanied by halogen exchange of the
formed (g6-C6R6)Ru(Cl){g2-iPrN@C(R0 0)NiPr} to the
corresponding bromide, affording 5a-Br or 6a-Br as a sin-
gle product.
20
14N2102Ru1: 405.1844. Found: 405.1845. 1H NMR
(400 MHz, CD2Cl2) d 1.24 (d, J = 6.6 Hz, 12H, CHMe2),
1.60 (s, 3H, NC(Me)N), 2.17 (s, 18H, g6-C6Me6), 3.33 (sep,
J = 6.6 Hz, 2H, CHMe2). 13C NMR (100 MHz, CD2Cl2) d
14.83 (NC(Me)N), 16.64 (g6-C6Me6), 24.71 (CHMe2),
48.21 (CHMe2), 88.39 (g6-C6Me6), 176.83 (NCN). The 13
C
resonance due to the TfO ligand was not visible. 19F NMR
(376 MHz, CD2Cl2) d ꢀ78.95 (s, CF3).
2a: yield 70%, mp 143 ꢀC (dec). Anal. Calc. for
C15H23N2O3F3SRu: C, 38.37; H, 4.94; N. 5.97. Found:
C, 38.43; H, 5.21; N, 6.03%. ESI-TOF: MꢀꢀOTf: 321.09.
Exact mass (ESI-TOF): Calcd. for
C
1H2314N2102Ru1:
14
12
1
321.0905. Found: 321.0886. H NMR (400 MHz, CD2Cl2)
d 1.22 (d, J = 6.3 Hz, 12H, CHMe2), 1.64 (s, 3H,
NC(Me)N), 3.32 (sep, J = 6.3 Hz, 2H, CHMe2), 5.70 (s,
6H, g6-C6H6). 13C NMR (100 MHz, CD2Cl2) d 12.24
(NC(Me)N), 24.42 (CHMe2), 48.00 (CHMe2), 79.71 (g6-
C6H6), 160.26 (NCN). The 13C resonance due to the TfO
ligand was not visible. 19F NMR (376 MHz, CD2Cl2) d
ꢀ77.3 (s, CF3).
(g6-C6Me6)Ru(Br){g2-iPrN@C(Me)NiPr} (5a-Br): yield
82%, mp 185 ꢀC (dec). Anal. Calc. for C20H35N2BrRu: C,
49.58; H, 7.28; N. 5.78. Found: C, 49.53; H, 7.31; N,
2b: yield 82%, mp 145 ꢀC (dec). Anal. Calc. for
C20H25N2O3F3SRu: C, 45.19; H, 4.74; N. 5.27. Found:
C, 45.13; H, 4.79; N, 5.43%. 1H NMR (600 MHz, CD2Cl2)
d 1.12 (br, 12H, CHMe2), 2.97 (sep, J = 6.2 Hz, 2H,
CHMe2), 5.81 (s, 6H, g6-C6H6), 6.99 (br, 2H, Ph), 7.35
(br, 3H, Ph). 13C NMR (150 MHz, CD2Cl2) d 24.41 (br,
CHMe2), 48.33 (CHMe2), 79.74 (g6-C6H6), 118.72 (q,
JCF = 319.2 Hz, CF3), 126.70 (br, Ph), 128.02, 128.64,
131.78 (Ph), 177.72 (NCN).
1
5.60%. H NMR (400 MHz, C6D6) d 1.21 (d, J = 6.3 Hz,
6H, CHMe2), 1.25 (d, J = 6.3 Hz, 6H, CHMe2), 1.62 (s,
3H, NC(Me)N), 2.16 (s, 18H, g6-C6Me6), 3.36 (sep,
J = 6.3 Hz, 2H, CHMe2). 13C NMR (100 MHz, C6D6) d
16.48 (NC(Me)N), 17.14 (g6-C6Me6), 24.83 (CHMe2),
26.63 (CHMe2), 48.74 (CHMe2), 88.74 (g6-C6Me6),
174.23 (NCN).
2c: mp 200 ꢀC (dec). Anal. Calc. for C22H29N2O3F3SRu:
C, 47.22; H, 5.22; N. 5.01. Found: C, 47.21; H, 5.22; N,
5.02%. 1H NMR (400 MHz, CD2Cl2) d 1.09 (s, 18H,
C(CH3)3), 5.82 (s, 6H, g6-C6H6), 7.08 (m, 2H, ortho-Ph),
7.23 (m, 2H, meta-Ph), 7.32 (m, 1H, para-Ph). 13C NMR
(100 MHz, CD2Cl2) d 32.71 (CMe3), 55.43 (CMe3), 80.82
(g6-C6H6), 126.67, 128.35, 129.47, 143.61 (Ph), 163.93
(NCN). The 13C resonance due to the TfO ligand was
not visible.
(g6-C6H6)Ru(Br){g2-iPrN@C(Me)NiPr} (6a-Br): yield
79%, mp 140 ꢀC (dec). Anal. Calc. for C14H23N2BrRu: C,
42.00; H, 5.79; N. 7.00. Found: C, 41.64; H, 5.79; N,
1
6.85%. H NMR (400 MHz, C6D6) d 1.13 (d, J = 6.1 Hz,
6H, CHMe2), 1.33 (s, 3H, NC(Me)N), 1.39 (d,
J = 6.1 Hz, 6H, CHMe2), 3.31 (sep, J = 6.1 Hz, 2H,
CHMe2), 4.90 (s, 6H, g6-C6H6). 13C NMR (100 MHz,
C6D6)
d 13.22 (NC(Me)N), 25.87 (CHMe2), 26.19
(CHMe2), 48.27 (CHMe2), 81.03 (g6-C6H6), 172.94 (NCN).
(g6-C6H6)Ru(Cl){g2-iPrN@C(Ph)NiPr} (6b-Cl): mp
210 ꢀC (dec). Anal. Calc. for C20H25N2ClRu: C, 54.60;
4.3. Preparation of (g6-C6R6)Ru(X){g2-R0N@C(R0 0)NR0}
(5a-Br, 6a-Br, 6b-Cl, 6c-Cl)
1
H, 6.03; N. 6.70. Found: C, 54.26; H, 6.07; N, 6.68%. H
NMR (400 MHz, C6D6) d 1.12 (d, J = 6.4 Hz, 6H,
CHMe2), 1.16 (d, J = 6.4 Hz, 6H, CHMe2), 3.05 (sep,
J = 6.4 Hz, 2H, CHMe2), 5.61 (s, 6H, g6-C6H6), 7.09 (br,
1H, Ph), 7.17 (br, 1H, Ph), 7.32 (br, 3H, Ph). 13C NMR
(100 MHz, C6D6) d 25.23 (CHMe2), 26.83 (CHMe2),
The halogeno complexes, 5a-Br, 6a-Br, 6b-Cl, and 6c-
Cl, were prepared by treatment of Li{R0N@C(R0 0)NR0},
which was synthesized from R0N@C@NR0 with R0 0Li,
with [(g6-C6Me6)RuCl2]2 or [(g6-C6H6)RuCl2]2. In a typi-

T. Hayashida et al. / Journal of Organometallic Chemistry 692 (2007) 382–394
391
48.56 (CHMe2), 81.14 (g6-C6H6), 127.60 (br), 127.93,
128.51, 128.80 (br), 132.95 (Ph), 175.93 (NCN).
[(g5-C5Me5)Ru(l-g2-iPrN@C(Me)NiPr)Ru(g-Cl)(g5-C5H5)]
(100 mg, 0.173 mmol) with TlOTf (100 mg, 0.282 mmol) in
CH2Cl2 (5 mL) afforded 4 as red crystals in 90% yield
(108 mg).
(g6-C6H6)Ru(Cl){g2-tBuN@C(Ph)NtBu} (6c-Cl): yield
57%, mp 220 ꢀC (dec). Anal. Calc. for C21H29N2ClRu: C,
56.55; H, 6.55; N. 6.28. Found: C, 56.21; H, 6.61; N,
3: mp 142 ꢀC (dec). Anal. Calc. for C29H47F3N2O3Ru2S:
C, 45.66; H, 6.27; N, 3.67. Found: C, 45.22; H, 6.12; N,
1
t
6.00%. H NMR (400 MHz, C6D6) d 1.26 (s, 18H, Bu),
5.02 (s, 6H, g6-C6H6), 6.94–7.08 (m, 3H, Ph), 7.11 (m,
1H, Ph), 7.49 (m, 1H, Ph). 13C NMR (100 MHz, C6D6) d
34.11 (CMe3), 55.82 (CMe3), 82.01 (g6-C6H6), 126.96,
126.97, 128.42, 129.67, 132.09, 143.61 (Ph), 174.51 (NCN).
3.60%. ESI-TOF: MꢀꢀOTf: 615.18. Exact mass (ESI-
12
TOF): Calcd. for
C
1H4714N2102Ru2: 615.1826. Found
28
615.1838.1H NMR (600 MHz, CD2Cl2) d 1.58 (s, 3H,
NC(Me)N), 1.59 (s, 30H, C5Me5), 1.61 (d, J = 6.6 Hz,
12H, CHMe2), 3.26 (sep, J = 6.6 Hz, 2H, CHMe2). 13C
4.4. Preparation of [(g5-C5Me5)Ru(l-g2-iPrN@C(Me)N-
iPr)Ru(j-OTf)(g5-C5R5)] (3; R = Me; 4; R = H)
NMR (150 MHz, CD2Cl2)
d
11.9 (C5Me5), 15.1
(NC(Me)N), 24.3 (CHMe2), 54.6 (CHMe2), 82.2 (C5Me5),
124.4 (NCN). 19F NMR (565 MHz, CD2Cl2) d ꢀ79.0. IR
(KBr, cmꢀ1) 1296, 1230.
Preparation of [(g5-C5Me5)Ru(l-g2-iPrN@C(Me)NiPr)-
Ru(g-X)(g5-C5R5)] was reported previously. In a Schlenk
tube were placed [(g5-C5Me5)Ru(l-g2-iPrN@C(Me)NiPr)-
Ru(g-Br)(g5-C5Me5)] (200 mg, 0.288 mmol) and AgOTf
(150 mg, 0.583 mmol), and CH2Cl2 (5 mL) was added.
The suspension was stirred for 1 h at room temperature.
After removal of the silver salt by filtration, the filtrate
was concentrated in vacuo. The residue was recrystallized
from a mixture of CH2Cl2 and pentane to give 3 as purple
crystals in 92% yield (203 mg). Similar treatment of
4: mp 142 ꢀC (dec). Anal. Calc. for C24H37F3N2O3Ru2S:
C, 41.61; H, 5.38; N, 4.04. Found: C, 41.36; H, 5.31; N,
3.99%. ESI-TOF: MꢀꢀOTf: 545.10. Exact mass (ESI-
12
TOF): Calcd. for
C
1H3714N2102Ru2: 545.1044. Found
23
545.1036. 1H NMR (600 MHz, CD2Cl2) d 1.06 (d,
J = 6.6 Hz, 6H, CHMe2), 1.26 (d, J = 6.6 Hz, 6 H,
CHMe2), 1.65 (s, 15 H, C5Me5), 1.91 (s, 3 H, NC(Me)N),
2.86 (sep. J = 6.6 Hz, 2H, CHMe2), 4.44 (s, 5 H, C5H5).
13C NMR (150 MHz, CD2Cl2) d 10.3 (C5Me5), 15.7
Table 4
Crystallographic tables
1
2c
3
4
Empirical formula
Crystal size (mm)
Crystal color, habit
Formula weight
Temperature (K)
Radiation
C
21H35F3O3RuSN2
C22H29F3O3RuSN2
0.35 · 0.20 · 0.20
Red, prismatic
559.61
C29H47N2O3F3Ru2S
0.30 · 0.30 · 0.10
Red prism
C24H37F3N2O3Ru2S
0.20 · 0.03 · 0.02
Red, platelet
692.76
0.20 · 0.20 · 0.20
Orange, block
553.64
762.90
123(2)
293(2)
Mo Ka (0.71069 A)
123(2)
Mo Ka (0.71069 A)
123(2)
Mo Ka (0.71069 A)
˚
˚
˚
˚
Mo Ka (0.71069 A)
Orthorhombic
Pbca (#61)
Crystal system
Space group
Orthorhombic
Pnma (#62)
Monoclinic
P21/c (#14)
Orthorhombic
Pna21 (#33)
Unit cell parameters
˚
a (A)
15.950(4)
15.311(4)
19.471(5)
90
90
90
4755.2(19)
8
1.547
18.830(4)
10.799(2)
12.005(3)
90
90
90
2441.1(9)
4
1.523
10.6176(2)
15.9728(3)
19.0453(4)
90
101.9811(8)
90
3159.6(1)
4
1.604
14.176(2)
18.841(2)
10.2680(13)
90
90
90
2742.5(6)
4
1.678
12.26
1400.00
3.1–27.5
6254 (I > 0.00r(I))
354
˚
b (A)
˚
c (A)
a(ꢀ)
b (ꢀ)
c (ꢀ)
3
˚
V (A )
Z
qcalc (g cmꢀ3
)
l(Mo Ka) (cmꢀ1
F(000)
)
7.95
7.76
10.72
2288.00
3.1–27.5
5441 (I > 0.00r(I))
315
1144.00
2.54–27.49
1876(I > 2.00r(I))
166
1560.00
2.3–27.5
7216 (I > 0.00r(I))
408
h Range (ꢀ)
No. observations
No. variables
Reflection/parameter ratio
R (all reflections)
R1(I > 2.00r(I))a
wR2 (all reflections)b
GOF
17.27
0.109
0.054
0.127
11.30
0.097
0.048
0.146
17.69
0.043
0.033
0.096
17.67
0.034
0.029
0.063
1.000
1.019
1.000
1.008
Flack parameter
Max shift/error in final cycle
Maximum peak in final diff. map (e /A )
–
–
–
0.00(3) (Friedel pairs = 2494)
0.000
1.55
ꢀ0.95
0.000
1.94
ꢀ1.02
0.000
0.57
ꢀ0.697
0.000
0.76
ꢀ0.70
ꢀ
ꢀ
3
˚
˚
3
Minimum peak in final diff. map (e /A )
P
P
a
R1 = jFoj ꢀ jFcj/ jFoj.
P
P
2
2
1=2
b
wR2 ¼ ½ ðwðF 2o ꢀ F c2Þ Þ= ðwðF 2oÞ Þꢁ
.

392
T. Hayashida et al. / Journal of Organometallic Chemistry 692 (2007) 382–394
(NC(Me)N), 21.5 and 25.8 (CHMe2), 52.6 (CHMe2),
66.3 (C5H5), 82.2 (C5Me5), 129.8 (NCN). 19F NMR
(565 MHz, CD2Cl2) d ꢀ78.0. IR (KBr, cmꢀ1) 1285, 1260.
4.5. Reactions of (g6-C6H6)Ru(X){g2-tBuN@C(Ph)NtBu}
(6c-Br: X = Br, 2c: X = OTf) with auxiliary ligands
4.6. X-ray data collection and reduction
Single crystals of 1a, 2c, 3, 4, 9-OTf, 10-OTf were grown
from CH2Cl2/pentane. X-ray crystallography were per-
formed on a Rigaku Saturn CCD area detector in the case
of 1a and 4, and on a Rigaku RAXIS RAPID imaging
plate diffraction meter in the case of 2c, 3, 9-OTf and 10-
OTf with graphite monochromated Mo Ka radiation
(k = 0.71070A). The data were collected at 123(2) K for
1a, 3 and 4, 223(2) K for 9-OTf and 10-OTf, and
293(2) K for 2c. The data were corrected for Lorentz and
polarization effects. The structures were solved by direct
methods (SIR92) [8] in the case of 2c and 9-OTf, by direct
method (SIR97) [9] in the case of 1a, by Patterson method
(DIRDIF94 PATTY) [10] for 10-OTf, and by Patterson method
(DIRDIF 99 PATTY) [11] in the case of 3 and 4, and expanded
using Fourier techniques [12]. The non-hydrogen atoms
were refined anisotropically. Hydrogen atoms were refined
using the riding model. The final cycle of full-matrix least-
squares refinement on F2 was based on 5441 observed
reflections and 315 variable parameters for 1a, 1876
observed reflections and 166 variable parameters for 2c,
7216 observed reflections and 408 variable parameters for
3, 6254 observed reflections and 354 variable parameters
Reactions of [(g6-C6H6)Ru{g2-tBuN@C(Ph)NtBu}]+-
(TFPB)ꢀ (7) with PPh3, pyridine, tBuNC, CO, and
CH2@CH2 were reported elsewhere. The bromide 6c-Br
t
reacted with BuNC to give the corresponding adduct 10-
Br, whereas formation of the PPh3 adduct 8-Br was detected
by NMR. The triflate 2c reacted with PPh3, pyridine and
tBuNC to afford the corresponding adduct, 8-OTf, 9-OTf,
and 10-OTf, respectively. All of the reactions were per-
formed in a similar fashion, and a typical example is as fol-
lows: In CH2Cl2 (5 mL) were dissolved 2c (34 mg,
0.061 mmol) and PPh3 (17 mg, 0.065 mmol) at ꢀ78 ꢀC.
The mixture was slowly warmed to room temperature and
stirred for 1 h. Removal of the solvent in vacuo, and the
formed crude product was recrystallized from CH2Cl2/pen-
tane at ꢀ35 ꢀC to give 8-OTf (47 mg) in 94% yield.
8-OTf: mp. 180 ꢀC (dec). Anal. Calc. for C40H44-
N2O3F3P1SRu: C, 58.45; H, 5.40; N, 3.41. Found: C,
1
58.03; H, 5.36; H, 3.29%. H NMR (400 MHz, CDCl3) d
t
0.83 (s, 18H, Bu), 6.04 (s, 6H, g6-C6H6), 7.29–7.32 (m,
Table 5
Crystallographic tables
2H, Ph), 7.37–7.47 (m, 9H, Ph), 7.51–7.60 (m, 9H, Ph).
13C NMR (100 MHz, CDCl3) d 30.32 (CMe3), 55.20
(CMe3), 89.83 (d, JC-P = 2.9 Hz, g6-C6H6), 126.99,
127.61, 127.89 (d, JC-P = 1.2 Hz), 128.93, 128.98 (d,
JC-P = 9.9 Hz), 129.49, 131.35 (br), 131.64, 133.52 (d,
JC-P = 9.9 Hz), and 138.13 (d, JC-P = 2.5 Hz) (Ph), 171.48
(d, JC-P = 3.7 Hz, NCN). 31P NMR (162 MHz, CDCl3) d
28.03 (PPh3).
9-OTf
10-OTf
Empirical formula
Crystal size (mm)
Crystal color, habit
Formula weight
Temperature (K)
Radiation
Crystal system
Space group
Unit cell parameters
C27H34F3N3O3RuS C27H38O3N3F3SRu
0.35 · 0.13 · 0.08
Yellow, plate
638.71
0.30 · 0.20 · 0.10
Yellow, prism
642.73
223(2)
223(2)
˚
˚
Mo Ka (0.71069 A) Mo Ka (0.71069 A)
Triclinic
Monoclinic
P21/n (#14)
ꢀ
P1 (#2)
9-OTf: yield 82%. mp. 180 ꢀC (dec). Anal. Calc. for
C27H34N3O3F3SRu: C, 50.77; H, 5.37; N, 6.58. Found:
˚
a (A)
10.719(2)
16.148(3)
8.892(2)
103.95(1)
94.38(2)
71.84(1)
1419.3(5)
2
1.494
6.79
656.00
2.73–27.50
3978 (I > 3.00r(I))
343
11.60
0.045
9.4434(2)
29.0548(7)
11.3224(3)
90
98.9770(4)
90
3068.5(1)
4
1.391
1
C, 50.60; H, 5.39; H, 6.53%. H NMR (400 MHz, CDCl3)
˚
b (A)
t
d 0.97 (s, 18H, Bu), 6.00 (s, 6H, g6-C6H6), 6.72 (m, 1H,
˚
c (A)
a (ꢀ)
b (ꢀ)
c (ꢀ)
Ph), 7.16–7.25 (m, 2H, Ph), 7.31–7.39 (m, 2H, Ph), 7.62
(ddd, J = 7.6, 6.6, 1.5 Hz, 2H, C5H5N), 7.95 (tt, J = 7.6,
1.5 Hz, 1H, C5H5N), 9.09 (dd, J = 6.6, 1.5 Hz, 2H). 13C
NMR (150 MHz, CDCl3) d 30.34 (CMe3), 55.28 (CMe3),
84.71 (g6-C6H6), 126.18 (C5H5N), 127.25, 127.82, 128.98,
129.36, 129.50, and 138.16 (Ph), 139.13, 154.88 (C5H5N),
171.50 (s, NCN).
3
˚
V (A )
Z
qcalc (g cmꢀ3
)
l (Mo Ka) (cmꢀ1
F(000)
)
6.28
1328.00
2.59–27.48
4063 (I > 3.00r(I))
343
11.85
0.050
0.155
0.87
0.000
0.41
h Range (ꢀ)
No. observations
No. variables
Reflection/parameter ratio
R1(I > 3.00r(I))a
wR2 (all reflections)b
GOF
10-OTf: yield 89%. mp. 140 ꢀC (dec). Anal. Calc. for
C27H38N3O3F3SRu: C, 50.45; H, 5.96; N, 6.54. Found:
1
C, 49.97; H, 5.83; H, 6.47%. H NMR (400 MHz, CDCl3)
d 0.94 (s, 18H, CMe3 of the amidinate), 1.68 (s, CMe3 of
the isocyanide), 6.12 (s, 6H, g6-C6H6), 7.19–7.30 (m, 2H,
Ph), 7.33 (m, 2H, Ph), 7.39 (m, 1H, Ph). 13C NMR
(150 MHz, CDCl3) d 30.49 (CMe3 of the isocyanide),
33.26 (CMe3 of the amidinate), 55.38 (CMe3 of the amidi-
nate), 84.71 (g6-C6H6), 127.19, 128.012, 129.09, 129.36,
130.030, and 138.34 (Ph), 172.61 (NCN). The 13C reso-
nance due to the TfO, CMe3 of the isocyanide, CN of the
isocyanide were not visible. IR (KBr, cmꢀ1) 2135.
0.109
1.007
Max shift/error in final cycle 0.001
Maximum peak in final
1.26
ꢀ
3
˚
diff. map (e /A )
Minimum peaꢀk in final
ꢀ0.85
ꢀ0.47
3
˚
diff. map (e /A )
P
P
a
R1 = jF0j ꢀ jFcj/ jF0j.
P
P
2
2
1=2
b
wR2 ¼ ½ ðwðF 2o ꢀ F c2Þ Þ= ðwðF 2oÞ Þꢁ
.

T. Hayashida et al. / Journal of Organometallic Chemistry 692 (2007) 382–394
393
´
(i) R.A.T.M. Abbenhuis, I. del Rıo, M.M. Bergshoef, J. Boersma,
N. Veldman, A.L. Spek, G. van Koten, Inorg. Chem. 37 (1998)
1749;
for 4, 3978 observed reflections and 343 variable parame-
ters for 9-OTf, 4063 observed reflections and 343 variable
parameters for 10-OTf. Neutral atom scattering factors
were taken from Cromer and Waber [13]. All calculations
were performed using the CrystalStructure [14,15] crystal-
lographic software package for 1a and 4, and all calcula-
tions for 2c, 3, 9-OTf and 10-OTf were performed using
the teXsan [16] crystallographic software package of
Molecular Structure Corporation except for refinement,
which was performed using SHELXL-97 [17]. Details of final
refinement are summarized in Tables 4 and 5, and the num-
bering scheme employed is shown in Figs. 3 and 6, which
was drawn with ORTEP at 50% probability ellipsoid.
Detailed data as well as the bond distances and angles
are shown in Supporting information.
(j) T. Funaioli, C. Cavazza, F. Marchetti, G. Fachinetti, Inorg.
Chem. 38 (1999) 3361;
(k) I. del Rıo, R.A. Gossage, M.S. Hannu, M. Lutz, A.L. Spek,
´
G. van Koten, Organometallics 18 (1999) 1097;
(l) T.P. Fong, C.E. Forde, A.L. Lough, R.H. Morris, P. Rigo, E.
Rocchini, T. Stephan, J. Chem. Soc., Dalton Trans. (1999) 4475;
(m) C.J. den Reijer, M. Wo¨rle, P.S. Pregosin, Organometallics 19
(2000) 309;
(n) D. Huang, J.C. Bollinger, W.E. Streib, K. Folting, V. Young
Jr., O. Eisenstein, K.G. Caulton, Organometallics 19 (2000) 2281;
(o) C.P. Casey, T.E. Vos, S.W. Singer, I.G. Guzei, Organometallics
21 (2002) 5038;
(p) M.D. Hagereaves, M.F. Mahon, M.K. Whittlesey, Inorg.
Chem. 41 (2002) 3137;
(q) M.A.O. Volland, S.M. Hansen, F. Rominger, P. Hofmann,
Organometallics 23 (2004) 800;
(r) J.M. Bikley, A.A. La Pensee, S.J. Higgins, C.A. Stuart, Dalton
Trans. (2003) 4663;
Acknowledgements
(s) J.O. Krause, O. Nuyken, K. Wurst, M.R. Buchmeiser, Chem.
Eur. J. 10 (2004) 777;
This work was supported by a Grant-in-Aid for Scien-
tific Research on Priority Areas (No. 14078101, ‘‘Reaction
Control of Dynamic Complexes’’) from Ministry of Educa-
tion, Culture, Sports, Science and Technology, Japan.
´
(t) F.A. Cotton, C.A. Murillo, J.H. Reinbenspies, D. Villagran, X.
Wang, C.C. Wilkinson, Inorg. Chem. 43 (2004) 8373.
[5] (a) Y. Yamaguchi, H. Nagashima, Organometallics 19 (2000) 725;
(b) T. Hayashida, Y. Yamaguchi, K. Kirchner, H. Nagashima,
Chem. Lett. (2001) 954;
Appendix A. Supplementary data
(c) H. Kondo, H. Nagashima, J. Am. Chem. Soc. 123 (2001) 500;
(d) H. Kondo, K. Matsubara, H. Nagashima, J. Am. Chem. Soc. 124
(2002) 534;
(e) J. Terasawa, H. Kondo, T. Matsumoto, K. Kirchner, Y.
Motoyama, H. Nagashima, Organometallics 24 (2005) 2713;
(f) For an account: H. Nagashima, H. Kondo, T. Hayashida, Y.
Yamaguchi, M. Gondo, S. Masuda, K. Miyazaki, K. Matsubara, K.
Kirchner, Coord. Chem. Rev. 245 (2003) 177;
(g) Application to catalysis: Y. Motoyama, M. Gondo, S. Masuda,
Y. Iwashita, H. Nagashima, Chem. Lett. 33 (2004) 442;
(h) Application to catalysis: Y. Motoyama, S. Hanada, S. Niibayashi,
K. Shimamoto, N. Takaoka, H. Nagashima, Tetrahedron 61 (2005)
10216.
The CIF files for the compound 1a, 2c, 3, 4, 9-OTf and
10-OTf, 1H and 13C spectra of 1a, 2a, 2c, 3, 4, 6-Cl, 9-OTf
1
and 10-OTf, variable temperature H and 19F NMR data
of 1a, 2a, 3 and 4, and ESI mass spectra of 1a, 2a, 3 and
4. Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/j.jorganchem.
2006.08.069.
References
[6] (a) T. Hayashida, H. Nagashima, Organometallics 21 (2002) 3884;
(b) T. Hayashida, H. Nagashima, The Reports of Institute of
Advanced Material Study, Kyushu University, vol. 15, 2000, p. 31;
(c) Crystal structures and reactions of several other (g6-arene)Ru(g2-
amidinate)X complexes were investigated; T. Hayashida, H. Naga-
shima, unpublished results.
[1] For reviews: (a) G.A. Lawrance, Chem. Rev. 86 (1986) 17;
(b) W. Beck, K. Sunkel, Chem. Rev. 88 (1988) 1405;
¨
(c) S.H. Strauss, Chem. Rev. 93 (1993) 927.
[2] For a review: E.Y.-X. Chen, T. Marks, Chem. Rev. 100 (2000)
1391.
[3] (a) D.A. Evans, J.A. Murry, P. von Matt, R.D. Norcross, S. Miller,
Angew. Chem., Int. Ed. Engl. 34 (1995) 798;
[7] For a related guanizine complex: (a) P.J. Bailey, L.A. Mitchel, S.
Parsons, J. Chem. Soc., Dalton Trans. (1996) 2839;
(b) E.P. Kundig, C.M. Saudan, G. Bernardinelli, Angew. Chem., Int.
Ed. Engl. 38 (1999) 1219.
(b) P.J. Bailey, K.J. Grant, L.A. Mitchel, S. Pace, A. Parkin, S.
Parsons, J. Chem. Soc., Dalton Trans. (2000) 1987.
[4] (a) M.J.A. Kraakman, B. de Klerk-Engels, P.P.M. de Lange, K.
Vrieze, W.J. Smeets, A.L. Spek, Organometallics 11 (1992) 3774;
(b) P.W. Blosser, J.C. Gallucci, A. Wojcicki, Inorg. Chem. 31 (1992)
2376;
[8] SIR92: A. Altomare, M. Cascarano, C. Giacovazzo, A. Guagliardi, J.
Appl. Crystallogr. 26 (1993) 343.
[9] SIR 97 A. Altomare, M. Burla, M. Camalli, M. Cascarano, C.
Giacovazzo, A. Guagliardi, A. Moliterni, G. Polidori, R. Spagna, J.
Appl. Crystallogr. 32 (1999) 115.
[10] DIRDIF94: P.T. Beurskens,G. Admiraal, G. Beurskens, W.P. Bosman, S.
Garcia-Granda, R.O. Gould, J.M.M. Smits, C. Smykalla, The DIRDIF-
94 program system; Technical Report of the Crystallography Labora-
tory, University of Nijmegen, Nijmegen, The Netherlands, 1992.
[11] DIRDIF99: P.T. Beurskens, G. Admiraal, G. Beurskens, W.P. Bosman,
R. de Gelder, R. Israel, J.M.M. Smits, The DIRDIF-99 program system,
Technical Report of the Crystallography Laboratory, University of
Nijmegen, Nijmegen, The Netherlands, 1999.
(c) B. de Klerk-Engels, J.H. Groen, M.J.A. Kraakman, J.M. Ernsting,
K. Vrieze, K. Goubitz, J. Fraanje, Organometallics 13 (1994) 3279;
(d) R.M. Barns, J.L. Hubbard, J. Am. Chem. Soc. 116 (1994)
9514;
(e) J.-P. Sutter, S.L. James, P. Steenwinkel, T. Karlen, D.M.
Grove, N. Veldman, W.J.J. Smeets, A.L. Spek, G. van Koten,
Organometallics 15 (1996) 941;
(f) A. Svetlanova-Larsen, C.R. Zoch, J.L. Hubbard, Organomet-
allics 15 (1966) 3076;
(g) C. Gemel, D. Kalt, K. Mareiter, V.N. Sapunov, R. Schmid,
K. Kirchner, Organometallics 16 (1997) 427;
(h) A.C. Ontko, J.F. Houlis, R.C. Schnabel, D.M. Roddick, T.P.
Fong, A.J. Lough, R.H. Morris, Organometallics 17 (1998) 5467;
[12] PATTY: P.T. Beurskens, G. Admiraal, G. Beurskens, W.P. Bosman, S.
Garcia-Granda, R.O. Gould, J.M.M. Smits, C. Smykalla, The DIRDIF
program system, Technical Report of the Crystallography Laboratory,
University of Nijmegen, Nijmegen, The Netherlands, 1992.

394
T. Hayashida et al. / Journal of Organometallic Chemistry 692 (2007) 382–394
[13] D.T. Cromer, J.T. WaberInternational Tables for X-ray Crystallog-
raphy, vol. 4, Kynoch Press, Birmingham, UK, 1974.
[14] CrystalStructure 3.6.0: Crystal Structure Analysis Package, Package,
Rigaku and Rigaku/MSC, 9009 New Trails Dr. The Woodlands, TX
7738,1 USA, 2000–2004.
[15] CRYSTALS Issue 10: D.J. Watkin, C.K. Prout, J.R. Carruthers, P.W.
Betteridge, Chemical Crystallography Laboratory, Oxford, UK, 1996.
[16] teXsan for Windows: Crystal Structure Analysis Package, Molecular
Structure Corporation, 1997.
[17] SHELX97: G.M. Sheldrick, 1997.