Diruthenium tetracarbonyl sawhorse complexes bearing N-heterocylic carbene and phosphine ligands: Synthesis, structural characterization, and catalytic activity

Article abstract of DOI:10.1016/j.jorganchem.2015.11.006

A number of dinuclear ruthenium(I) sawhorse complexes substituted with an N-heterocyclic carbene or various phosphine ligands have been prepared and characterized by FT-IR, NMR, and elemental analysis. Treatment of [Ru₂(μ-O₂CCH₃)₂(CO)₄]_n with 2-electron donating ligands (L) yields the dimeric derivatives Ru₂(μ-O₂CCH₃)₂(CO)₄L₂, where L = IMes [1,3-bis(2,4,6-trimethylphenyl)imidazol-2-ylidene] (1), P(o-CH₃C₆H₄)₃ (2), P(C₆F₅)₃ (3), P(c-C₆H₁₁)₃ (4), and P(C₆H₅)₃ (5). The syntheses of 1-3 are reported herein; the syntheses of 4 and 5 have been previously reported. The crystal structures of 1-3 have been determined by single crystal X-ray diffraction. Factors influencing the structures of 1-5 are discussed and compared to DFT calculated geometries. An initial assay of the catalytic activities of 1-5, employing the isomerization of 1-hexene, has been performed.

Full text of DOI:10.1016/j.jorganchem.2015.11.006

Accepted Manuscript
Diruthenium tetracarbonyl sawhorse complexes bearing N-heterocylic carbene and
phosphine ligands: Synthesis, structural characterization, and catalytic activity
Thomas N. Rorhabaugh, Jr., Joshua C. Doverspike, Steven J. Geib, Evan D. Sawyer,
Mitchell R. Stibbard, Thomas J. Malosh
PII:
S0022-328X(15)30205-9
10.1016/j.jorganchem.2015.11.006
JOM 19293
DOI:
Reference:
To appear in: Journal of Organometallic Chemistry
Received Date: 7 September 2015
Revised Date: 4 November 2015
Accepted Date: 5 November 2015
Please cite this article as: T.N. Rorhabaugh Jr., J.C. Doverspike, S.J. Geib, E.D. Sawyer, M.R.
Stibbard, T.J. Malosh, Diruthenium tetracarbonyl sawhorse complexes bearing N-heterocylic carbene
and phosphine ligands: Synthesis, structural characterization, and catalytic activity, Journal of
Organometallic Chemistry (2015), doi: 10.1016/j.jorganchem.2015.11.006.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
Diruthenium tetracarbonyl sawhorse complexes bearing N-heterocylic
carbene and phosphine ligands: Synthesis, structural characterization,
and catalytic activity
Thomas N. Rorhabaugh Jr.a, Joshua C. Doverspikea, Steven J. Geibb, Evan D. Sawyera, Mitchell
R. Stibbarda, Thomas J. Malosha,*
a Department of Chemistry, University of Pittsburgh Johnstown, Johnstown, PA 15904, USA
b Department of Chemistry, University of Pittsburgh, Pittsburgh, PA 15260, USA
* Corresponding author. Email address: malosh@pitt.edu
A B S T R A C T
A number of dinuclear ruthenium(I) sawhorse complexes substituted with an N-heterocyclic
carbene or various phosphine ligands have been prepared and characterized by FT-IR, NMR, and
elemental analysis. Treatment of [Ru2(µ-O2CCH3)2(CO)4]n with 2-electron donating ligands (L)
yields the dimeric derivatives Ru2(µ-O2CCH3)2(CO)4L2, where L = IMes [1,3-bis(2,4,6-
trimethylphenyl)imidazol-2-ylidene] (1), P(o-CH3C6H4)3 (2), P(C6F5)3 (3), P(c-C6H11)3 (4), and
P(C6H5)3 (5). The syntheses of 1-3 are reported herein; the syntheses of 4 and 5 have been
previously reported. The crystal structures of 1-3 have been determined by single crystal X-ray
diffraction. Factors influencing the structures of 1-5 are discussed and compared to DFT
calculated geometries. An initial assay of the catalytic activities of 1-5, employing the
isomerization of 1-hexene, has been performed.
Keywords: Ruthenium; Dinuclear complexes; Catalysis; Carbene ligands; Phosphine ligands;
Carboxylato bridges
Introduction

ACCEPTED MANUSCRIPT
Interest in diruthenium(I) tetracarbonyl sawhorse-type compounds, Ru2(ꢀ-O2CCH3)2(CO)4L2,
particularly where L is a tertiary phosphine such as PBu3 or PPh3, is due to their identification as
catalysts or catalyst precursors for a variety of organic transformations. These transformations
include the conversion of dimethyl oxalate to methyl glycolate and ethylene glycol [1], the
conversion of acetic acid to ethyl acetate and methanol [2], the benzylation of phenol [3], alkene
isomerization [4], the hydrogenation of alkenes and ketones [5], the semi-hydrogenation of diaryl
alkynes [6], the hydrogenation of alkenes under scCO2 conditions [7], and the hydroformylation
of alkenes [8].
The sawhorse platform consists two pairs of eclipsed cis(carbonyl) legs, along with one pair
of cis(bridging carboxylate) ligands as the trestle top - Scheme 1. This platform is quite robust
and offers at least two opportunities for complex modification: alteration of the terminal or axial,
two-electron donating ligands, L; and alteration of the three-electron donating cis(bridging)
ligands. Indeed, while a variety of terminal ligands have been employed, a significant number of
bridging ligands/systems, beyond carboxylates, have also been explored [9]. Regarding these
bridging ligands, some recent work, involving carboxyphenyl porphyrin derivatives, has focused
on the areas of molecular recognition [10], and photosensitizing/chemotherapeutic agents [7,11].
Spurred by successes in the area of palladium-based, catalytic carbon-carbon coupling [12],
and ruthenium-based, catalytic ring closing olefin metathesis [13], the utilization of stable, N-
heterocyclic carbenes as ligands in metal based catalytic systems has rapidly proliferated over
the last twenty years. In addition to the above areas, it has been demonstrated that gold-NHC
complexes can function as catalyst precursors for a variety of organic transformations [14]. In
all three areas, it appears that N-heterocyclic carbenes have supplanted tertiary phosphines as the
supporting ligands of choice [15].

ACCEPTED MANUSCRIPT
In terms of the terminal ligand, L, we are interested in bringing both N-heterocyclic carbene
ligands and heretofore unexamined tertiary phosphine ligands to bear upon the small molecule
transformations mentioned above. Herein, we describe the synthesis and characterization of
Ru2(ꢀ-O2CCH3)2(CO)4L2, where L = IMes (1), P(o-tolyl)3 (2), and P(C6F5)3 (3), including the
solid-state structures of 1-3. The synthetic protocol employed for 1-3 was extended to (4), L =
PCy3. A procedure for the preparation of (5), L = PPh3, which has not been previously reported
in detail is contained within. Also included are the results of DFT determined, geometric
optimizations of 1-6, where axial L = CO for (6). The isomerization of 1-hexene under mild
conditions was utilized to assess the catalytic activities of 1-5. In a study of alkene isomerization
by select ruthenium compounds, η2-1-hexene intermediates have been proposed [16]. Optimized
models of η2-1-hexene substituted complexes 1a-5a, analogous to 1-5, were calculated and the
results are reported herein.
Results and discussion
Preparation and properties
In following the general procedure of Crooks et al. [17], we found that suspensions of the
oligomeric starting material, [Ru2(µ-O2CCH3)2(CO)4]n, were slow to react with solutions of
sterically bulky ligands such as PCy3. Adopting the approach of Hilts et al. [18], a circa 60 °C
acetonitrile suspension of the oligomer provides the labile bis(acetonitrile) adduct as an
intermediate. Both THF and the ultimate terminal ligand, L, were then added to the reaction
vessel to provide the complexes Ru2(ꢀ-O2CCH3)2(CO)4L2:
L
=
1,3-bis(2,4,6-
trimethylphenyl)imidazol-2-ylidene: IMes (1), L = P(o-tolyl)3 (2), L = P(C6F5)3 (3), and L =
PCy3 (4) - Scheme 1. The solid, yellow reaction residues were re-dissolved in either chloroform
or dichloromethane, and recrystallization by slow diffusions of either ethanol or hexane,

ACCEPTED MANUSCRIPT
provided compounds 1-4 in good yields. To our knowledge, complex 1 is the first reported
diruthenium(I) sawhorse-type complex with terminal N-heterocyclic carbene ligands. A dry
mixture of the oligomeric starting material and solid triphenylphosphine were heated beyond the
melting point of PPh3. After heating for 2 hours, the resulting yellow solution indicating the
formation of Ru2(ꢀ-O2CCH3)2(CO)4(PPh3)2, (5). A series of diethyl ether triturations, followed
by a diffusion of a saturated dichloromethane solution into hexane provided compound 5 in good
yield.
Elemental Analysis (C, H) established purity for compounds 1-5. In general, complexes 1-4
are soluble in chloroform and dichloromethane. Compound 1 is insoluble in pentane and
hexane; soluble in toluene and THF; slightly soluble in diethyl ether; and insoluble in ethanol.
Compound 2 is insoluble in pentane and hexane; soluble in toluene and THF; slightly soluble in
diethyl ether; and insoluble in ethanol. Compound 3 is insoluble in pentane and hexane; slightly
soluble in toluene and diethyl ether; soluble in THF; and insoluble in ethanol. Compound 4
reluctantly dissolves in chloroform and dichloromethane.
Solid-state structures
The molecular structures of 1-3 are presented in Figures 1-3. Relevant crystallographic and
structural data can be found in Tables 1 and 2. Each ruthenium atom is located at the center of
an irregular octahedron formed by coordination to five ligands, and with the sixth coordination
site occupied by a ruthenium-ruthenium bond. The 2-electron donating axial carbene, or
phosphine, ligands are located trans to the Ru-Ru bond, with each ruthenium-axial ligand bond
being roughly, but not exactly, collinear to the Ru-Ru bond. The equatorial plane around each
ruthenium center consists of a pair of ꢀ-acetate ligands along with a pair of carbon monoxide

ACCEPTED MANUSCRIPT
ligands, where both pairs of ligands are mutually cis. Thus, idealized structure presented in
Scheme 1 is generally achieved by 1-3 in the solid-state, Figures 1-3.
A limited number of diruthenium sawhorse compounds featuring bis(acetate) bridges have
been structurally characterized. When considering only those structures incorporating phosphine
ligands [5,6,19a-e], previously reported ruthenium-ruthenium bond lengths range from 2.718(5)
Å to 2.7614(10) Å. With Ru-Ru bond lengths of 2.6896(3) Å and 2.6918(2), respectively, 2 and
3 become the shortest examples to date, Table 2. When considering all bis(acetate) structures
[5,6,19a-k], previously reported Ru-Ru bond lengths range from 2.672(1) Å to 2.7614(10) Å.
With a bond length of 2.7805(3) Å, complex 1 possesses the longest Ru-Ru bond among
bis(acetate) bridged sawhorse compounds, Table 2.
The steric requirements of N-heterocyclic-carbene ligands are unlike those attributed to the
phosphine cone angle, Figure 1 [20]. In the solid-state structure of 1, a plane constructed by
using the four o-methyls of the bis(mesityl) substituents is approximately co-planar with the
equatorial planes of the ruthenium octahedra. Furthermore, the planar mesityl substituents are
substantially twisted with respect to the plane of the central imidazole ring, as interplane torsions
approach 90°. Following the method advanced by Müller and Mingos [21], the phosphine cone-
angles, as manifest in the solid-state, have been calculated for 2-5, and are presented in Table 2.
The calculated cone angles are comparable for 4 (164°) and 5 (154°), the PCy3 and PPh3 adducts,
respectively, as these data are within the ranges found by the initial study. Also, comparing 4
and 5 with the Tolman cone angles provides results similar to those exhibited by the average
crystallographic cone angles in the original study, where the Tolman cone angles are 170° and
145° for PCy3 and PPh3, respectively [22]. Müller and Mingos’ work did not extend to either
P(o-tolyl)3 or P(C6F5)3. For 2, the calculated crystallographic cone angle is 176°, while the

ACCEPTED MANUSCRIPT
putative cone angle for P(o-tolyl)3 is listed as 194° [22]. However, as visible in Figure 2, one
tolyl substituent has rotated such that its o-methyl faces away from the equatorial plane of the
ruthenium octahedron. Effectivelty then, the calculated cone angle for complex 2, of 176°, can
be considered that of a PPh(o-tolyl)2 adduct.
The ideal sawhorse configuration implies a (L-Ru-Ru-L)° torsion of 0°, along with eclipsed
carbon monoxide ”legs,” and strictly planar carboxylate bridges. As can be seen from the
various torsions presented in Table 2, the solid-state conformations generally deviate from the
ideal. For example, when considering only those bis(acetate) structures incorporating tertiary
phosphine ligands [5,6,19b-e], previously reported (P-Ru-Ru-P)° torsions range from 2.5(7)° to
57.0(11)° for the PiPr3 and the PPh2Py (dppy) adducts, respectively. There is no corrleation
between either of the phosphine cone angles and the torsions along the backbone of the
molecules. The carboxylate bridges themselves also have unique steric requirements as
suggested by the torsions of the hexacarbonyl adduct, 6. As evinced by the various torsions, the
core of complex 1 approaches an ideal sawhorse conformation. However, the N-hetreocyclic
carbene ligands are not eclipsed, but offset by circa 11°. Finally, the solid-state structures of 1-
3, and 5, are stabilized by variety of aromatic interactions. Additional details may be found in
the Supplemental Information.
Spectroscopic characterization
Compounds 1-5 are yellow, air stable, crystalline powders, which have been characterized by
infrared, proton NMR, fluorine-19 NMR (3 only), phosphorus-31 NMR (2-5 only), and
elemental analysis. The infrared spectra of 1-5 exhibit the expected pattern of three (very strong-
medium-very strong) bands attributed to carbonyl stretching, and zero to three additional weak
bands/shoulders in the 2200-1800 cm-1 region. These carbon monoxide stretching frequencies

ACCEPTED MANUSCRIPT
show an expected trend in the σ-donating abilities of the terminal ligands, L. The observed
trend: IMes > PCy3 > PPh3 ~ P(o-tolyl)3 > P(C6F5)3. Additionally, all infrared spectra feature at
least two bands assigned to the symmetric and asymmetric stretching of the carboxylate bridges,
found in the 1600-1400 cm-1 region. The phosphorus-31 NMR spectra of 2-5 each consist of a
single signal, and the proton NMR spectra of 1-5 contain but one singlet assigned to the methyl
1
substituents of the bridging acetate ligands. Furthermore, the H NMR spectra of 2 and 4 are
indicative of restricted rotation about the Ru-P bonds at ambient temperatures [5]. The fluorine-
19 NMR spectrum of 3 exhibits three signals corresponding to the ortho, meta, and para fluorine
substituents, respectively. The assignment of each signal to a specific ring position is based on
both integration, and previous reports [23].
The infrared and NMR data support a solution configuration for 1-5 in which the terminal
ligands and both of the bridging acetate ligands are symmetrically equivalent. Such
equivalencies are achieved by the C2v sawhorse configuration of Ru2(ꢀ-O2CCH3)2(CO)4L2 in
which the pairs of eclipsed, cis-carbonyl groups are the legs of the sawhorse. Thus, the solution
configuration of 1-5 is consistent with the solid-state configurations, as illustrated in Scheme 1,
and Figures 1-3. The IR bands, in the 2200-1800 cm-1 region exhibited by 1-5 are listed in Table
3, along with relevant phosphorus-31 NMR results.
Computational studies of 1-6
The exaggerated backbone torsion (P-Ru-Ru-P)° of complex 2 in the solid-state, prompted the
calculation of optimized molecular geometries for 1-6. The literature led to the Perdew, Burke,
and Ernzerhof functional, as well as the Becke Three Parameter functional. Selected results,
listed in Table 4, were provided by employing the Perdew, Burke, and Ernzerhof functional, with
the Stuttgart/Dresden ECP for ruthenium, and the 6-31G(d) basis for any additional elements,

ACCEPTED MANUSCRIPT
within Gaussian09 [24]. Through all six models, the computations provided zero imaginary
frequencies. With the inferred absence of intermolecular forces, the geometric parameters of the
gas phase molecules show considerable adjustment, particularly with regard to the various
torsions exhibited in the solid-state. Of interest, the backbone torsions both increase and
decrease across 1-6, with no discernable correlation to other geometric parameters. The main
torsion decreases for both 2 (82.9° to 54.9°) and 4 (20.2° to 18.7°), while this torsion increases
for 1, 3, 5 (21.1° to 43.5°), and 6; with the torsions of 2 and 5 exhibiting the extremes. Among
the tertiary phosphine adducts, 2-5, every calculated phosphine cone angle increases. With the
exception of 2, there is a general increase in the (O-Ru-Ru-O)° and (C-Ru-Ru-C)° torsions.
Regarding all three torsions, the various forces present in the solid-state appear to have the least
effect on complexes 4 and 6.
The solid-state structure of 2 exhibits two intramolecular close contacts between methyl
group hydrogen atoms on the P(o-tolyl)3 ligands and the oxygen atoms of the acetate bridges.
Specifically: (1) C-H···O = 3.13 Å with H···O = 2.26 Å, and θ C–H–O = 150°; (2) C-H···O =
3.35 Å with H···O = 2.45 Å, and θ C–H–O = 156°. For crystallographic refinement, a riding
model was employed for all hydrogen atoms. However, these two contacts persist in the
calculated structure of 2. Specifically (PBE0): (1) C-H···O = 3.17 Å with H···O = 2.16 Å, and θ
C–H–O = 154°; (2) C-H···O = 3.66 Å with H···O = 2.60 Å, and θ C–H–O = 163°. Similarly, the
solid-state structure of 4 exhibits two intramolecular close contacts between hydrogen atoms on
the PCy3 ligands and the oxygen atoms of the acetate bridges. Specifically: (1) C-H···O = 3.25
Å with H···O = 2.42 Å, and θ C–H–O = 142°; (2) C-H···O = 3.33 Å with H···O = 2.52 Å, and θ
C–H–O = 138°. For crystallographic refinement, a riding model was also employed for all
hydrogen atoms. Again however, the two contacts persist in the calculated structure of 4.

ACCEPTED MANUSCRIPT
Specifically (PBE0): (1) C-H···O = 3.23 Å with H···O = 2.32 Å, and θ C–H–O = 139°; (2) C-
H···O = 3.27 Å with H···O = 2.38 Å, and θ C–H–O = 137°. The C–H–O angles of 4 are more
acute than those of 2. Recently, an updated definition of the hydrogen bond was endorsed. The
catenated articles extend to observed characteristics and valid experimental criteria [25]. One
criterion is that hydrogen bonding leads to characteristic proton magnetic resonance signatures.
Specifically and typically, pronounced deshielding for H in X-H···Y-Z should be exhibited. In
1
the H NMR spectra of 2 and 4, pronounced desheilding of the relevant protons was not
observed.
Catalytic Assay of 1-5
Under a nitrogen atmosphere, 0.036 mmol of the appropriate sawhorse dimer (1-5) was dissolved
in 10.0 mL of toluene. To these yellow solutions, 0.011 mol, of 1-hexene was added by syringe.
The vessel was then placed in a 55 °C oil bath, with stirring, for 45 hours. Within moments, the
various solutions changed color. Green solutions were observed for every trial, and the color
persisted until the flasks were removed from the bath. The resulting mixtures were found to
contain 1-hexene, plus 2- and 3-hexenes. The following activities were observed: P(C6F5)3 >
PCy3 > PPh3 > P(o-tolyl)3 > IMes, Table 5. It is apparent that the P(C6F5)3 adduct (3) is an
efficient catalyst precursor under the stated conditions. There is no correlation between the
observed catalytic activity and either the calculated or putative cone angles, Tables 2 and 5.
In a pressure vessel under more robust conditions [P(N2) = 1 atm, T = 100 °C, t = 15 hrs],
Matteoli, et al. exposed 1-hexene to three (3) tertiary phosphine adducts of bis(acetate) bridged
ruthenium sawhorse complexes [5]. Isomerization was selected as this catalytic transformation
involves but a single substrate. A correlation was observed between isomerization activity and
the solid-state (P-Ru-Ru-P)° torsion: PtBu3 > PnBu3 > PiPr3; respective conversions = 42.4 % >

ACCEPTED MANUSCRIPT
34.2 % > 7.4 %; respective torsions = 21.9° > 8.1° > 2.5°. Their rationale being that greater
torsions along the backbone of the molecule provide progressively less hindered access to the
ruthenium centers. Regarding the isomerization activities of 1-5, the correlation, as observed by
Matteoli, et al., between percent conversion and backbone torsion does not hold, Tables 2 and 5.
Among the tertiary phosphine adducts only, 2-5, there are two inverse correlations between
catalytic activity and solid-state, geometric parameters: (1) isomerization activity increases with
decreasing (P-Ru-Ru-P)° torsions, and (2) isomerization activity increases with decreasing Ru-P
bond length, Tables 2 and 5. However, for 2-5, the two inverse correlations between catalytic
activity and geometric parameters do not hold for the DFT calculated molecular geometries,
Tables 4 and 5.
Alkene isomerization, specifically that of 1-hexene, by phosphine substituted ruthenium
carbonyl carboxylates was further examined by Salvini, et al. [16]. The diruthenium
tetracarbonyl sawhorse complex Ru2(ꢀ-O2CCH3)2(CO)4(PnBu3)2 was studied in this context.
Based on gas chromatography, infrared, and 31P NMR analyses, the authors postulate a
mechanism involving the substitution of 1-hexene for one terminal phosphine ligand to initially
form an η2-1-hexene intermediate.
Starting with the minimum energy structures
(PBE0/MWB28) for 1-5, optimized geometries for complexes 1a-5a were calculated, where one
terminal carbene ligand, or one terminal phosphine ligand, was substituted by η2-1-hexene.
Selected results are presented in Table 6. As illustrated in Figure 4, a characteristic of 1a-5a is
that the C1-C2 bond axis of the coordinated 1-hexene is roughly co-linear with an AcO-Ru-CO
axis. Of the various parameters listed in Tables 4 and 6, the main torsions of 1a-5a exhibit the
largest changes from the calculated main torsions for 1-5. Regarding the coordination of 1-
hexene to the ruthenium centers, the average distance, 1a-5a, from the ruthenium center to the

ACCEPTED MANUSCRIPT
centroid of the C1-C2 bond of the 1-hexene is 2.283 Å, Table 6. However, the Ru-centroid
distances of complexes 3a and 4a are 2.300 Å or greater, with an average of 2.303 Å.
Complexes 3 and 4 also exhibit the highest isomerization activities, Table 5. The Ru-centroid
distances of the remaining three putative intermediates (1a, 2a, and 5a) are all less than 2.300 Å,
with an average of 2.269 Å. Using the Student’s t-test, a comparison of these two mean
distances reveals that, with 95% confidence, 2.303 Å is statistically different from 2.269 Å. A
similar set of means can be constructed from the (L-Ru-Ru-centroid)° main torsions, Table 6.
Where again, the average torsion of 3a and 4a is statistically different from the average torsion
of the remaining three modeled intermediates. However, the respective main torsions of 3a
(5.9°) and 4a (23.8°) are also distinctly different from each other.
Experimental
General Information
All preparations were performed under dry gaseous nitrogen atmospheres using dual gas/vacuum
manifolds and standard schlenk techniques and glassware. Recrystallizations of compounds 1-5
were performed under aerobic conditions. The various phosphine ligands were used as received
from either Sigma-Aldrich or Strem. The compound 1,3-bis(2,4,6-trimethylphenyl)imidazol-2-
ylidene (IMes) was obtained from Strem and was recrystallized from hexane prior to use. The
various carbene and phosphine ligands were manipulated in glove bags under dry nitrogen
atmospheres. Organic solvents meeting ACS specifications, or better, were employed and were
degassed and saturated with dry nitrogen prior to use.
Synthesis
of
the
starting
material,
acetatodicarbonylruthenium
polymer,
[Ru(O2CCH3)(CO)2]n, was accomplished by following a previously published procedure [17].
The compound Ru3(CO)12 was obtained from Sigma-Aldrich and was used as received. The

ACCEPTED MANUSCRIPT
acetic acid was purified by the addition of acetic anhydride, followed by reflux over KMnO4 and
distillation under nitrogen.
Infrared spectra of starting materials and synthetic targets were recorded on a Perkin-Elmer
Model 1600 FTIR instrument. A Bruker Spectrospin Ultrashield 300 MHz FT NMR instrument
1
31
was employed to obtain H and P{1H} spectra of both starting materials and synthetic targets,
as well as 1H spectra of the isomerization reaction mixtures. The 31P chemical shifts are reported
versus 85% H3PO4. Single crystal X-ray structural analyses were performed at the Chemistry
Department X-ray Diffraction Facility at the University of Pittsburgh, Pittsburgh, PA, USA. At
the Indiana University of Pennsylvania, Indiana, PA, USA, an Agilent 6890N Gas
Chromatograph with a 5973 Mass Selective Detector was employed to obtain chromatographs of
the isomerization reaction mixtures. Fluorine-19 NMR was performed at Spectral Data Services,
Inc., located in Champaign, IL, USA. The 19F chemical shifts are reported versus C6F6.
Elemental Analyses were performed at Atlantic Microlab, Inc., located in Norcross, GA, USA.
Preparation of dimeric complexes 1-4: Ru2(µ-O2CCH3)2(CO)4L2
The dimeric, symmetric bis(substituted) complexes, 1-4, were prepared by the addition of a
carbene or phosphine ligand to the bis(acetonitrile) adduct, Ru2(ꢀ-O2CCH3)2(CO)4(CH3CN)2, in
the manner advanced by Hilts et al. [18].
Typical preparation: 3 mL of sparged acetonitrile were added by syringe to a 3-neck, 100-mL
RBF, with gas inlet, containing 200 mg (0.463 mmol) of [Ru2(ꢀ-O2CCH3)2(CO)4]n. With stirring
under N2, the flask was placed in a 60 °C oil bath for 2 hours. The resulting yellow solution
indicates full dissolution of the intractable, orange starting material, and the formation of Ru2(ꢀ-
O2CCH3)2(CO)4(CH3CN)2. After removal from the oil bath, a 20.0 mL aliquot of THF was
transferred to the flask. A solid addition funnel was then used to add a slight excess of ligand

ACCEPTED MANUSCRIPT
(circa 1.0 mmol). The mixture was returned to the oil bath, with stirring under N2, for an
additional 2 hours. The solvents were then removed under reduced pressure to yield a mixture of
Ru2(ꢀ-O2CCH3)2(CO)4L2 plus excess ligand, L.
Ru2(µ-O2CCH3)2(CO)4(C21H24N2)2: bis(IMes) adduct (1)
The reaction residue was dissolved in minimal dichloromethane. Under ambient conditions, a
layer of hexane was allowed to slowly diffuse into the dichloromethane solution of the complex.
The resulting pale yellow crystals were collected on a sintered glass crucible, washed with
chilled hexane, and dried under vacuum. Yield: (379 mg, 0.364 mmol, 79%). IR (CHCl3, cm–1)
1
ν(CO): 2005 (vs), 1951 (m), 1922 (vs), 1887 (w); ν(CO2): 1579 (m), 1438 (m). H NMR (300
MHz, CDCl3) δ: 6.87 (m, 12H), 2.31 (s, 12H), 2.10 (s, 24H), 1.19 (s, 6H, OAc). Anal. Calc. for
C50H54N4O8Ru2: C, 57.68; H, 5.23; N, 5.38. Found: C, 57.84; H, 5.34; N, 5.40%.
Ru2(µ-O2CCH3)2(CO)4{P(o-CH3C6H4)3}2: bis(tri-o-tolylphosphine) adduct (2)
The reaction residue was dissolved in minimal chloroform. Under ambient conditions, a layer of
hexane was allowed to slowly diffuse into the chloroform solution of the complex. The resulting
yellow crystals were collected on a sintered glass crucible, washed with chilled hexane, and dried
under vacuum. Yield: (415 mg, 0.399 mmol, 86%). IR (CHCl3, cm–1) ν(CO): 2025 (vs), 1982
1
(m), 1951 (vs), 1920 (w); ν(CO2): 1573 (m), 1440 (m). H NMR (300 MHz, CD2Cl2) δ: 7.28 (m,
24H), 2.17 (s, br 24H). 31P{1H} NMR (121 MHz, CDCl3) δ: 14.67 (s). Anal. Calc. for
C50H48O8P2Ru2: C, 57.69; H, 4.65. Found: C, 57.64; H, 4.75%.
Ru2(µ-O2CCH3)2(CO)4{P(C6F5)3}2: bis(tris(pentafluorophenyl)phosphine) adduct (3)
The reaction residue was dissolved in minimal chloroform. Under ambient conditions, a layer of
ethanol was allowed to slowly diffuse into the chloroform solution of the complex. The resulting
yellow crystals were collected on a sintered glass crucible, washed with chilled ethanol, and

ACCEPTED MANUSCRIPT
dried under vacuum. Yield, based on chloroform solvate: (482 mg, 0.298 mmol, 64%). IR
(CHCl3, cm–1) ν(CO): 2047 (vs), 2008 (m), 1979 (vs), 1953 (w); ν(CO2): 1572 (m), 1448 (m).
1H NMR (300 MHz, CDCl3) δ: 1.45 (s, 6H, OAc). 19F NMR (376 MHz, CDCl3) δ: -127.31 (s,
12F, o), -147.32 (s, 6F, p), -160.31 (t, 12F, m). 31P{1H} NMR (121 MHz, CDCl3) δ: -28.85 (m).
Anal. Calc. for C44H6F30O8P2Ru2·CHCl3: C, 33.45; H, 0.44. Found: C, 33.73; H, 0.32%.
Ru2(µ-O2CCH3)2(CO)4{P(c-C6H11)3}2: bis(tricyclohexylphosphine) adduct (4)
The reaction residue was dissolved in minimal chloroform. Under ambient conditions, a layer of
ethanol was allowed to slowly diffuse into the chloroform solution of the complex. The resulting
yellow crystals were collected on a sintered glass crucible, washed with chilled ethanol, and
dried under vacuum. Yield: (261 mg, 0.263 mmol, 57%). IR (CHCl3, cm–1) ν(CO): 2010 (vs),
1
1962 (m), 1933 (vs), 1900 (w); ν(CO2): 1579 (m), 1436 (m). H NMR (300 MHz, C6D6) δ: 2.21
(s, br 6H, OAc), 1.97-1.29 (all m, br 66H). 31P{1H} NMR (121 MHz, CDCl3) δ: 25.17 (s). Anal.
Calc. for C44H72O8P2Ru2: C, 53.21; H, 7.31. Found: C, 53.57; H, 7.46%.
Preparation of the dimeric complex 5: Ru2(µ-O2CCH3)2(CO)4(PPh3)2, via fusion
Ru2(µ-O2CCH3)2(CO)4{P(C6H5)3}2: bis(triphenylphosphine) adduct (5)
A 50-mL RBF, with a gas inlet, was charged with 200 mg (0.462 mmol) of [Ru2(µ-
O2CCH3)2(CO)4]n and 1.5 g of triphenylphosphine (circa 5 equivalents). With stirring under N2,
the flask was placed in a 90 °C oil bath for 2 hours. After cooling, the pale yellow solid was
triturated with diethyl ether - 3x. The resulting yellow residue was dissolved in minimal
dichloromethane. Under ambient conditions, a layer of hexane was allowed to slowly diffuse
into the dichloromethane solution of the complex. The resulting yellow-orange crystals were
collected on a sintered glass crucible, washed with chilled hexane, and dried under vacuum.
Yield: (390 mg, 0.408 mmol, 88%). IR (CHCl3, cm–1) ν(CO): 2023 (vs), 1978 (m), 1949 (vs),

ACCEPTED MANUSCRIPT
1
1919 (w); ν(CO2): 1569 (m), 1436 (m). H NMR (300 MHz, CDCl3) δ: 7.55 (m, 12H), 7.40 (m,
18H), 1.69 (s, 6H, OAc). 31P{1H} NMR (121 MHz, CDCl3) δ: 14.39 (s). Anal. Calc. for
C44H36O8P2Ru2: C, 55.23; H, 3.79. Found: C, 55.13; H, 3.87%.
Crystallographic analyses
Diffraction quality, off-white crystals of compound 1 were formed as hexane diffused into a
saturated dichloromethane solution under ambient conditions. Diffraction quality, yellow
crystals of compounds 2 and 4 were obtained as ethanol diffused into saturated chloroform
solutions under ambient conditions. Diffraction quality, yellow crystals of compound 3 were
formed as hexane diffused into a saturated chloroform solution under ambient conditions. The
studied crystals were mounted on glass capillaries and data was collected on a Bruker SMART
Apex II CCD system utilizing an IꢀS micro-focus source to provide Cu Kα radiation at λ =
1.54178 Å. Detailed crystallographic data for 1-3 are listed in Table 1. Other relevant, structural
information is presented in Table 2. A structure for Ru2(CO)4(ꢀ-O2CCH3)2(PCy3)2 has been
previously deposited in the Cambridge Structural Database [6]. Additional detail regarding the
crystallographic analysis of 4 can be found in the Supplemental Information.
Compounds 1 and 3. Systematic absences were consistent with the space groups P21/c and
C2/c, for 1 and 3, respectively. For both compounds, data were corrected for absorption effects
using multi-scan methods (SADABS), absorption coefficients: ꢀ = 5.522 mm-1 and 6.527 mm-1
for 1 and 3 respectively. The resulting 6943 data and 4377 data, for 1 and 3 respectively, were
employed in the least squares refinements. Compound 2. The triclinic unit cell was consistent
with either space group Pı, or Pī. The average values of the normalized structure factors favored
Pī. The data were corrected for absorption effects using the multi-scan method (SADABS),

ACCEPTED MANUSCRIPT
absorption coefficient: ꢀ = 7.185 mm-1. The resulting 15785 data were used in the least squares
refinements.
Compounds 1-3. All three structures were solved by employing direct methods within the
SHELXTL software package [26]. The correct positions for the ruthenium and phosphorous
atoms were deduced from a direct methods E-map. Subsequent least squares refinement and
difference Fourier calculations established the positions of the remaining non-hydrogen atoms.
Non-hydrogen atoms were refined with independent anisotropic displacement parameters.
Hydrogen atoms were fixed in idealized positions and their displacement parameters were tied to
those of the attached non-hydrogen atom. All structures were refined by full matrix, least
squares procedures, based on F2, of the positional, isotropic, and anisotropic thermal parameters
for all non-hydrogen atoms. A final analysis of variance between calculated and observed
structure factors exhibited no perceptible errors. PLATON was used to calculate a number of
geometric parameters [27].
Computational studies
Hybrid density functional theory calculations were performed to ascertain optimal molecular
geometries for complexes 1-6. As the starting point for the geometric optimizations, solid-state
structures were used to provide the molecular specifications for Gaussian09 [24]. The X-ray
structures of 1-3 are reported herein, and that of 4 in the Supplemental Information. Previously
reported structural data was used for 5 [19b]. The deposited solid-state structure of Ru2(CO)4(ꢀ-
O2CCH3)2(CO)2 (6) does not include any positional data for the six hydrogen atoms [19f]. A
model of 6 was constructed and optimized using Scigress computational software [28]. This
optimized model was employed as the molecular specification in lieu of complete CIF
information.

ACCEPTED MANUSCRIPT
Within Gaussian09, the Perdew, Burke, and Ernzerhof functional (PBE0) was employed.
The basis set was composed of the Stuttgart/Dresden ECP for ruthenium (MWB28), along with
the 6-31G(d) basis for any additional elements. Normal-mode analyses were then performed on
the optimized geometries. Selected results are presented in Table 4. Additionally, geometric
optimizations were obtained by utilizing the Becke Three Parameter functional with the LYP
correlation functional (B3LYP). The basis was composed of the Los Alamos ECP plus DZ for
ruthenium (LanL2DZ), along with the 6-31G(d) basis for any additional elements. Normal-mode
analyses were then performed on the optimized geometries. Selected results are presented in
Table 4S (Supplemental Information).
Optimized structures were also calculated for a series of complexes, 1a-5a, where one
terminal carbene or phosphine ligand was replaced by η2-1-hexene. As the starting point for the
geometric optimizations, the minimum energy structures for 1-5, above, were used to provide the
molecular specifications for Gaussian09. An idealized molecule of 1-hexene was constructed
and optimized using Scigress computational software [28]. The 1-hexene model was then
substituted for the appropriate terminal ligand. Within Gaussian09, the Perdew, Burke, and
Ernzerhof functional (PBE0) was employed.
The basis set was composed of the
Stuttgart/Dresden ECP for ruthenium (MWB28), along with the 6-31G(d) basis for any
additional elements. Normal-mode analyses were then performed on the optimized geometries.
Selected results are presented in Table 6.
Isomerization of 1-hexene in the presence of 1-5
Representative procedure: 10 mL of sparged toluene were added by syringe to a 50-mL RBF,
with gas inlet and condenser, containing 0.036 mmol of diruthenium sawhorse compound.

ACCEPTED MANUSCRIPT
Finally, 1.4 mL (0.011 mol) of 1-hexene were added by syringe. With stirring under N2, the
flask was placed in a 55 °C oil bath for 45 hours.
The reaction mixtures were examined by both proton NMR and gas chromatography
(GCMS). The average results, by adduct, are listed in Table 5.
Conclusions
Compounds 1-3 are readily synthesized by the addition of two equivalents of the desired axial
ligand to the labile intermediate Ru2(ꢀ-O2CCH3)2(CO)4(NCCH3)2 in THF. Compound 5 can be
synthesized via a fusion obtained from suspending oligomeric [Ru2(ꢀ-O2CCH3)2(CO)4]n in liquid
triphenylphosphine.
conformation. However, the various torsions achieved by 1-6 in the solid-state show numerous
departures from an ideal conformation. Spectroscopic characterization affirms a C2v
The solid-state structures of 1-3 exhibit the expected sawhorse
arrangement about the ruthenium centers. The Ru-Ru bond lengths of 1-3, and the P-Ru-Ru-P
torsion angles of 1 and 2 are structurally significant. The extent of the phosphine cone angles, as
achieved by 2-5 in the solid-state, have been calculated. Comparatively, the calculated cone
angles of 4 and 5 fall within the limits of the original study. The calculated cone angle of 5
exceeds the Tolman cone angle, whereas those of 2-4 fall short. Close X-H···Y-Z contacts are
found in the solid-state structures of 2 and 4, and these contacts persist in the respective DFT
optimized geometries. However, spectroscopic evidence of intramolecular hydrogen bonding is
lacking. Regarding the solid-state torsions displayed by 1-6, the DFT results show numerous,
irregular adjustments in the gas-phase conformations. Compounds 1-5 are catalytic precursors
toward the isomerization of 1-hexene under mild conditions. The compounds display a broad
range of capabilities. Among 1-5, the isomerization activity of the P(C6F5)3 adduct (3) is
markedly different. DFT calculated models of the proposed η2-1-hexene intermediates,

ACCEPTED MANUSCRIPT
constructed from the most active catalyst precursors, possess a longer average Ru-
hexene(centroid) distance than those built from less active adducts.
Acknowledgements
We gratefully acknowledge the Mentorship Fund for Faculty/Student Scholarship at the
University of Pittsburgh Johnstown for partial financial support of this project. T.J.M. thanks
Dr. Justin D. Fair at IUP, Dr. Robert Grimminger, and David R. Swanger III for their valuable
contributions.
Appendix A. Supplementary Data
CCDC 1418271, 1418272, 1418273, 1418274 contain the supplementary crystallographic data
for this paper. These data can be obtained free of charge from the Cambridge Crystallographic
Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
References
[1] (a) U. Matteoli, M. Bianchi, G. Menchi, P. Frediani, F. Piacenti, J. Mol. Catal. 29 (1985)
269;
(b) U. Matteoli, G. Menchi, M. Bianchi, F. Piacenti, J. Organomet. Chem. 299 (1986) 233;
(c) U. Matteoli, G. Menchi, M. Bianchi, F. Piacenti, J. Mol. Catal. 44 (1988) 347;
(d) U. Matteoli, G. Menchi, M. Bianchi, F. Piacenti, J. Mol. Catal. 64 (1991) 257.
[2] (a) M. Bianchi, G. Menchi, F. Francalanci, F. Piacenti, U. Matteoli, P. Frediana, C.
Botteghi, J. Organomet. Chem. 188 (1980) 109;
(b) U. Matteoli, G. Menchi, M. Bianchi, P. Frediani, F. Piacenti, Gazz. Chim. Ital. 115
(1985) 603;
(c) P. Frediani, M. Bianchi, A. Salvini, F. Piacenti, S. Ianelli, M. Nardelli, J. Chem. Soc.,
Dalton Trans. (1990) 1705;
(d) A. Salvini, P. Frediani, C. Gianelli, L. Rosi, J. Organomet. Chem. 690 (2005) 371.
[3] R. Jaouhari, Chem. Lett. 23 (1994) 1781.
[4] A. Salvini, P. Frediani, D. Rovai, M. Bianchi, F. Piacenti, J. Mol. Catal. 89 (1994) 77.
[5] U. Matteoli, G. Menchi, M. Bianchi, F. Piacenti, S. Ianelli, M. Nardelli, J. Organomet.
Chem. 498 (1995) 177.

ACCEPTED MANUSCRIPT
[6] J. Li, R. Hua, Chem. Eur. J. 17 (2011) 8462.
[7] J.P. Johnpeter, L. Plasseraud, F. Schmitt, L. Juillerat-Jeanneret, B. Therrien, J. Coord. Chem.
66 (2013) 1753.
[8] (a) J. Jenck, P. Kalck, E. Pinelli, M. Siani, A. Thorez, J. Chem. Soc., Chem. Commun.
(1988) 1428;
(b) P. Kalck, M. Siani, J. Jenck, B. Peyrille, Y. Peres, J. Mol. Catal. 67 (1991) 19.
[9] B. Therrien, G. Suss-Fink, Coord. Chem. Rev. 253 (2009) 2639.
[10] J.P. Johnpeter, J. Mohanraj, N. Amaroli, B. Therrien, Eur. J. Inorg. Chem. (2012) 3449.
[11] (a) J.P. Johnpeter, F. Schmitt, E. Denoyelle-Di-Muro, G. Wagnieres, L. Juillerat-Jeanneret,
B. Therrien, Inorg. Chim. Acta 393 (2012) 246.
(b) J.P. Johnpeter, B. Therrien, Inorg. Chim. Acta 394 (2013) 723.
[12] W.A. Herrmann, M. Elison, J. Fischer, C. Kocher, G.R.J. Artus, Angew. Chem. Int. Ed.
Engl. 34 (1995) 2371.
[13] M. Scholl, T.M. Trnka, J.P. Morgan, R.H. Grubbs, Tetrahedron Lett. 40 (1999) 2247.
[14] S.P. Nolan, Acc. Chem. Res. 44 (2011) 91.
[15] D.J. Nelson, Eur. J. Inorg. Chem. (2015) 2012.
[16] A. Salvini, P. Frediani, F. Piacenti, J. Mol. Catal. A 159 (2000) 185.
[17] G.R. Crooks, B.F.G. Johnson, J. Lewis, I.G. Williams, G. Gamlen, J. Chem. Soc. A (1969)
2761.
[18] R.W. Hilts, M. Cowie, Inorg. Chem. 29 (1990) 3349.
[19] (a) T.A. Bright, R.A. Jones, C.M. Nunn, J. Coord. Chem. 18 (1988) 361.
(b) K-B. Shiu, S-M. Peng, M-C. Cheng, J. Organomet. Chem. 452 (1993) 143.
(c) J.S. Field, R.J. Hines, C.J. Parry, J. Chem. Soc., Dalton Trans. (1997) 2843.
(d) T.J. Malosh, S.R. Wilson, J.R. Shapley, J. Organomet. Chem. 694 (2009) 3331.
(e) P. Stepnicka, I. Cisarova, Collect. Czech. Chem. Commun. 74 (2009) 799.
(f) M. Spohn, T. Vogt, J. Straehle, Z. Naturforsch., B: Chem. Sci. 41 (1986) 1373.

ACCEPTED MANUSCRIPT
(g) R.W. Hilts, S.J. Sherlock, M. Cowie, E. Singleton, M.M. de V. Steyn, Inorg. Chem. 29
(1990) 3161.
(h) K. Panneerselvam, T-H. Lu, C-H. Huang, S-F. Tung, K-B. Shiu, Acta Crystallogr., Sect.
C: Cryst. Struct. Commun. 53 (1997) 1782.
(i) C.M. Kepert, G.B. Deacon, L. Spiccia, G.D. Fallon, B.W. Skelton, A.H. White, J.
Chem. Soc., Dalton Trans. (2000) 2867.
(j) C.M. Kepert, G.B. Deacon, L. Spiccia, Inorg. Chim. Acta 355 (2003) 213.
(k) R. Dorta, G. Suss-Fink, H. Stoeckli-Evans, Acta Crystallogr., Sect. E: Struct. Rep.
Online 60 (2004) m1397.
[20] A.C. Miller, W.J. Sommer, B.S. Yong, J.L. Petersen, L. Cavallo, S.P. Nolan,
Organometallics 22 (2003) 4322.
[21] T.E. Müller, D.M.P. Mingos, Transition Met. Chem. 20 (1995) 533.
[22] C.A. Tolman, Chem. Rev. 77 (1977) 313.
[23] (a) M.G. Barlow, M. Green, R.N. Haszeldine, H.G. Higson, J. Chem. Soc. B (1966) 1026.
(b) M.G. Hogben, W.A.G. Graham, J. Am. Chem. Soc. 91 (1969) 283.
[24] M.J. Frisch, et al., Gaussian09, Revision D.01 WIN64 (2009) Gaussian, Inc.: Wallingford,
CT.
[25] (a) E. Arunan, G.R. Desiraju, R.A. Klein, J. Sadlej, S. Scheiner, I. Alkorta, D.C. Clary, R.H.
Crabtree, J.J. Dannenberg, P. Hobza, H.G. Kjaergaard, A.C. Legon, B. Mennucci, D.J.
Nesbitt, Pure Appl. Chem. 83 (2011) 1619.
(b) E. Arunan, G.R. Desiraju, R.A. Klein, J. Sadlej, S. Scheiner, I. Alkorta, D.C. Clary, R.H.
Crabtree, J.J. Dannenberg, P. Hobza, H.G. Kjaergaard, A.C. Legon, B. Mennucci, D.J.
Nesbitt, Pure Appl. Chem. 83 (2011) 1637.
[26] G.M. Sheldrick, SHELXTL, version 6.10 (2001) Bruker Analytical X-ray Systems:
Madison, WI.
[27] A.L. Spek, PLATON: A Multipurpose Crystallographic Tool, Ultrecht University, 2005.
[28] DGauss4.1, Oxford Molecular, Fujitsu Ltd., 2000.

ACCEPTED MANUSCRIPT
Table 1
Crystal data and refinement parameters for 1-3.________________
______Adduct: IMes (1)______P(o-tolyl)3 (2)___P(C6F5)3 (3)__
formula
C50H54N4O8Ru2 C50H48O8P2Ru2 C44H6F30O8P2Ru2
formula wt.
crystal system monoclinic
space group
a, Å
b, Å
c, Å
α, deg
β, deg
1041.11
1040.96
monoclinic
C2/c
10.5074(3)
14.8416(8)
29.3597(9)
90
99.364(1)
90
4517.5(2)
4
1496.57
triclinic
P-1
P21/c
13.8985(4)
14.9593(4)
23.3482(7)
90
95.947(1)
90
9.9245(4)
21.3845(8)
23.7181(9)
101.969(2)
92.086(2)
94.221(2)
4904.0(3)
4
γ, deg
Volume, Å3
Z
4828.2(2)
4
Temp, K
ρcalcd, g/cm3
ꢀ, mm-1
total data
unique data
parameters
Rint
230(2)
1.432
5.522
8808
6943
607
0.0547
0.0498
293(2)
1.531
6.527
4413
4377
284
0.1175
0.0376
0.1034
2.047, -1.459
230(2)
2.027
7.185
17353
15785
1565
0.0272
0.0283
R1 (all data)a
wR2 (all data)b 0.1038
0.0811
0.477, -0.573
max, min, e/Å3 1.591, -0.637
____________________________________________________________________________________
a R1 = ΣFo- Fc/ΣFo, b wR2 = [Σw(Fo2 – Fc2)2/Σw(Fo)2]1/2

ACCEPTED MANUSCRIPT
Table 2
Crystal Data: Selected bond lengths (Å), torsions (°), and angles (°) for 1-6a_________________
___Adduct:
Ru-Ru
Ru-L
IMes
2.7805(3)
2.184(3)
P(o-tolyl)3
2.6896(3)
2.5235(4)
1.848(2)
P(C6F5)3
2.6918(2)
2.4370(6)
1.840(1)
PCy3
2.7555(6)
2.4496(12) 2.4508(10) 1.988(9)
1.864(1)
PPh3
2.7360(9)
CO_____
2.6881(15)
P-C avg.
1.824(1)
L-Ru-Ru-L
0.9(5)
82.89(18)
18.29(6)
31.10(9)
114.18(6)
176
19.11(14)
13.61(7)
17.00(13)
115.64(6)
167
20.2(3)
5.07(11)
2.9(2)
114.05(11)
164
21.14(11)
1.47(8)
2.67(14)
115.1(3)
154
3.7(18)
9.8(3)
10.2(6)
O-Ru-Ru-Ob 1.05(9)
C-Ru-Ru-Cc
Ru-P-C avg.
Cone Angled
Cone Anglee
1.21(14)
194
184
170
145
____________________________________________________________________________________________________________
a) 1-4, this work; 5 [18b]; 6 [18f] b) avg. OAc bridges c) avg. CO ligands d) calculated avg., [21] e) Tolman, [22]

ACCEPTED MANUSCRIPT
Table 3
Selected spectral data for 1-5.__________________________________
IR
(CO)a cm-1 NMR 31P{1H}c
Ru2(ꢀ-OAc)2(CO)4L2
ν
_______________________________________________________________________________________
δ(ppm) δ(ppm) ꢁP(ppm)
Ligand
(vs)
(m)
(vs)
(w)
ligand complex complex
3 P(C6F5)3
2047 2008 1979 1953
-74.26 -28.85 45.41
2 P(o-tolyl)3 2025 1982 1951 1920
-29.64 14.67
-5.41 14.39
6.11 25.17
44.31
19.80
19.06
5 PPh3
4 PCy3
1 IMesb
2023 1978 1949 1919
2010 1962 1933 1900
2005 1951 1922 1887
_______________________________________________________________________________________
a) CHCl3, this work b) 1,3-bis(2,4,6-trimethylphenyl)imidazol-2-ylidene c) vs. H3PO4, this work

ACCEPTED MANUSCRIPT
Table 4
Optimized Models: Geometric parameters in (Å) and (°) for 1-6a
____Complex: 1 6____
2
3
4
5
Ru-Ru
Ru-L
P-C
2.771 2.722 2.684 2.752 2.700 2.729
2.166 2.509 2.423 2.448 2.437 1.961
1.851 1.842 1.871 1.836
L-Ru-Ru-L 13.3
O-Ru-Ru-Ob 13.1
C-Ru-Ru-Cc 17.5
Ru-P-C avg.
54.9
15.5
22.1
29.9
17.0
24.1
18.7
10.2
9.1
43.5
18.5
24.9
5.0
9.8
11.0
114.3 115.5 113.8 115.1
179
194
Cone Angled
169
184
167
170
158
145
Cone Anglee
_________________________________________________________________________________
a) PBE0/MWB28 b) avg. OAc, c) avg. CO, d) calc. avg., [21] e) Tolman, [22]

ACCEPTED MANUSCRIPT
Table 5
Isomerization of 1-hexene to 2- and 3-hexenes upon exposure to 1-5 in toluenea
Adduct: P(C6F5)3 (3) PCy3 (4) PPh3 (5) P(o-tolyl)3 (2) IMes (1)
Conversion %:
67.4
20.2
15.1
12.1
1.0____
a) conditions: P(N2) = 1 atm., T = 55 °C, t = 45 hrs., dimer = 0.036 mmol, substrate = 0.011 mol

ACCEPTED MANUSCRIPT
Table 6
Optimized Intermediates: Parameters in (Å) and (°) for 1a-5aa
____Complex:
Ru-Ru
1a
2a
3a
4a
5a_
2.701
2.251
2.432
1.835
2.733
2.287
2.147
2.717
2.270
2.503
1.851
2.681
2.300
2.427
1.843
2.729
2.306
2.449
1.871
Ru-centroidb
Ru-L
P-C
L-Ru-Ru-cent. 45.8
56.0
13.8
18.6
114.1
179
5.9
23.8
12.4
14.0
113.7
167
60.7
16.4
22.8
115.0
158
O-Ru-Ru-Oc
C-Ru-Ru-Cd
Ru-P-C avg.
Cone Anglee
Cone Anglef
14.7
20.9
17.7
23.7
115.5
169
194
184
170
145
____________________________________________________________________________
a) PBE0, b) C1-C2 centroid, c) avg. OAc, d) avg. CO, e) calc., [21] f) Tolman, [22]

ACCEPTED MANUSCRIPT