Homobimetallic ruthenium-arene complexes bearing vinylidene ligands: Synthesis, characterization, and catalytic application in olefin metathesis

Article abstract of DOI:10.1021/om1006177

Five new arylvinylidene complexes with substituents ranging from electron-donating to strongly withdrawing (p-OMe, p-Me, p-Cl, p-CF₃, and m-(CF₃)₂) were isolated in high yields by reacting [(p-cymene)Ru(μ-Cl)₃RuCl(η² C₂H ₄)(PCy₃)] (3) with the corresponding phenylacetylene derivatives. The known phenylvinylidene complex [(p-cymene)Ru(μ-Cl) ₃RuCl(=C=CHPh)(PCy₃)] (5) was also obtained from [RuCl₂(p-cymene)]₂, tricyclohexylphosphine, and phenylacetylene under microwave irradiation. The influence of the remote aryl substituents on structural features was investigated by IR, NMR, and XRD spectroscopies. A very good linear relationship was observed between the chemical shift of the vinylidene α-carbon atom and the Hammett σ-constants of the aryl group substituents. The catalytic activity of the six homobimetallic complexes was probed in various types of olefin metathesis reactions. Unsubstituted phenylvinylidene compound 5 served as a lead structure for these experiments. Its reaction with norbornene afforded high molecular weight polymers with a broad polydispersity index and mostly trans double bonds. Aluminum chloride was a suitable cocatalyst for the ring-opening metathesis polymerization of cyclooctene and led to the formation of high molecular weight polyoctenamer with a rather narrow polydispersity index (M_w/M _n = 1.25) and an almost equimolar proportion of cis and trans double bonds. No major changes were observed in the polymer yields and microstructures when complexes bearing donor groups on their aryl rings were employed as catalyst precursors. On the other hand, compounds bearing strongly electron-withdrawing substituents were significantly less active. Model vinylidene compound 5 and its ruthenium-ethylene parent (3) both required the addition of phenylacetylene to achieve the ring-closing metathesis of diethyl 2,2-diallylmalonate. Thus, the role of this terminal alkyne cocatalyst goes beyond the facile replacement of the η²-alkene ligand with a vinylidene fragment.

Full text of DOI:10.1021/om1006177

Organometallics 2010, 29, 6675–6686 6675
DOI: 10.1021/om1006177
Homobimetallic Ruthenium-Arene Complexes Bearing Vinylidene
Ligands: Synthesis, Characterization, and Catalytic Application
in Olefin Metathesis
Yannick Borguet,† Xavier Sauvage,† Guillermo Zaragoza,‡ Albert Demonceau,† and
Lionel Delaude*,†
†
ꢀ
Laboratory of Macromolecular Chemistry and Organic Catalysis, Institut de Chimie (B6a), Universite de
Liege, Sart-Tilman par 4000 Liege, Belgium, and Unidade de Difraccion de Raios X, Edificio CACTUS,
‡
ꢁ
ꢁ
ꢀ
Universidade de Santiago de Compostela, Campus Sur, 15782 Santiago de Compostela, Spain
Received June 27, 2010
Five new arylvinylidene complexes with substituents ranging from electron-donating to strongly
withdrawing (p-OMe, p-Me, p-Cl, p-CF3, and m-(CF3)2) were isolated in high yields by reacting [(p-
cymene)Ru(μ-Cl)3RuCl(η2-C2H4)(PCy3)] (3) with the corresponding phenylacetylene derivatives.
The known phenylvinylidene complex [(p-cymene)Ru(μ-Cl)3RuCl(dCdCHPh)(PCy3)] (5) was also
obtained from [RuCl2(p-cymene)]2, tricyclohexylphosphine, and phenylacetylene under microwave
irradiation. The influence of the remote aryl substituents on structural features was investigated by
IR, NMR, and XRD spectroscopies. A very good linear relationship was observed between the
chemical shift of the vinylidene R-carbon atom and the Hammett σ-constants of the aryl group
substituents. The catalytic activity of the six homobimetallic complexes was probed in various types
of olefin metathesis reactions. Unsubstituted phenylvinylidene compound 5 served as a lead structure
for these experiments. Its reaction with norbornene afforded high molecular weight polymers with a
broad polydispersity index and mostly trans double bonds. Aluminum chloride was a suitable
cocatalyst for the ring-opening metathesis polymerization of cyclooctene and led to the formation of
high molecular weight polyoctenamer with a rather narrow polydispersity index (Mw/Mn =1.25)
and an almost equimolar proportion of cis and trans double bonds. No major changes were observed
in the polymer yields and microstructures when complexes bearing donor groups on their aryl rings
were employed as catalyst precursors. On the other hand, compounds bearing strongly electron-
withdrawing substituents were significantly less active. Model vinylidene compound 5 and its
ruthenium-ethylene parent (3) both required the addition of phenylacetylene to achieve the ring-
closing metathesis of diethyl 2,2-diallylmalonate. Thus, the role of this terminal alkyne cocatalyst
goes beyond the facile replacement of the η2-alkene ligand with a vinylidene fragment.
Introduction
from several groups have led to improved second- and third-
generation ruthenium-alkylidene complexes, which display
extremely high catalytic activities and excellent tolerance
toward polar functional groups.3
From a practical point of view, the introduction of an
alkylidene fragment onto a group 8 transition metal usually
requires the use of either cyclopropenes4 or diazoalkanes.5
Because these two types of reagents are rather difficult to
prepare and to handle, especially on a large scale, the straight-
forward access to ruthenium-benzylidene complexes remains
Thanks to the development of the well-defined ruthenium-
benzylidene catalyst [RuCl2(dCHPh)(PCy3)2] (PCy3 is tri-
cyclohexylphosphine) initiated by Grubbs in the late 1990s,1
olefin metathesis has become a key methodology in organic
synthesis and in polymer chemistry.2 Countless subsequent
studies have been aimed at improving the catalytic efficiency
of this archetypal compound. The most significant advances
were achieved by replacing one of its phosphine ligands with
anN-heterocycliccarbene (NHC), and tirelessresearchefforts
(3) (a) Beligny, S.; Blechert, S. In N-Heterocyclic Carbenes in Synthesis;
Nolan, S. P., Ed.; Wiley-VCH: Weinheim, 2006; pp 1-25. (b) Despagnet-
Ayoub, E.; Ritter, T. In N-Heterocyclic Carbenes in Transition Metal
Catalysis; Glorius, F., Ed.; Topics in Organometallic Chemistry, Vol. 21;
Springer: Berlin, 2007; pp 193-218. (c) Colacino, E.; Martinez, J.; Lamaty,
F. Coord. Chem. Rev. 2007, 251, 726–764. (d) Samojzowicz, C.; Bieniek,
M.; Grela, K. Chem. Rev. 2009, 109, 3708–3742. (e) Vougioukalakis, G. C.;
Grubbs, R. H. Chem. Rev. 2010, 110, 1746–1787.
(4) Nguyen, S. T.; Johnson, L. K.; Grubbs, R. H. J. Am. Chem. Soc.
1992, 114, 3974–3975.
(5) Schwab, P.; France, M. B.; Ziller, J. W.; Grubbs, R. H. Angew.
Chem., Int. Ed. 1995, 34, 2039–2041.
*To whom correspondence should be addressed. E-mail: l.delaude@
ulg.ac.be.
(1) (a) Grubbs, R. H.; Miller, S. J.; Fu, G. C. Acc. Chem. Res. 1995,
28, 446–452. (b) Grubbs, R. H. Tetrahedron 2004, 60, 7117–7140.
(c) Grubbs, R. H. Adv. Synth. Catal. 2007, 349, 34–40.
(2) For monographs, see: (a) Ivin, K. J.; Mol, J. C. Olefin Metathesis
and Metathesis Polymerization; Academic Press: San Diego, 1997.
(b) Handbook of Metathesis; Grubbs, R. H., Ed.; Wiley-VCH: Weinheim,
2003. (c) Green Metathesis Chemistry; Dragutan, V.; Demonceau, A.;
Dragutan, I.; Finkelshtein, E. S., Eds.; NATO Science for Peace and Security
Series-A: Chemistry and Biology, Springer: Dordrecht, 2010.
r
2010 American Chemical Society
Published on Web 11/18/2010
pubs.acs.org/Organometallics

6676 Organometallics, Vol. 29, No. 24, 2010
Borguet et al.
Scheme 1. Synthesis of Monometallic Ruthenium-Vinylidene
Complexes
Scheme 2. Synthesis of Homobimetallic Ruthenium-Ethylene
Complex 3
a major issue, and a great deal of attention has been paid to
other catalyst precursors bearing a metal-carbon multiple
bond that are more convenient to synthesize. Among them,
ruthenium-vinylidene,6-8 -allenylidene,7-9 and -indenyl-
idene8,10 species have generated much interest, because they
are readily obtained from commercially available alkyne
derivatives using safe and efficient experimental procedures.
In particular, monometallic ruthenium-vinylidene com-
plexes are easily isolated from the reaction of [RuCl2(p-cymene)]2
(1) with an excess of phosphine and terminal alkyne (Scheme 1).
This simple and efficient methodology was devised by Ozawa
and co-workers in 1998.11 The Japanese team successfully
applied it to prepare a range of complexes with the generic
formula [RuCl2(dCdCHR)L2] (L = PPri3 or PCy3) (2),
which were probed as catalysts for the ring-opening metath-
esis polymerization (ROMP) of norbornene derivatives and
the ring-closing metathesis (RCM) of R,ω-dienes.12 Subse-
quent reports from other groups further demonstrated the
validity of this approach,13 although the first-generation vinyl-
idene complexes turned out to be less efficient metathesis initia-
tors than their benzylidene counterparts. Recourse to [RuCl2-
(dCdCHR)(PCy3)(NHC)] complexes with mixed phosphine/
NHC ligands helped reduce this gap. Such second-generation
ruthenium-vinylidene catalysts were first reported by Louie
and Grubbs in 200114 and further investigated by Opstal and
Verpoort in 2003.15 It should be pointed out that the catalytic
activity of the nonsubstituted vinylidene complex [RuCl2(dCd
CH2)(PCy3)2] in the ROMP of norbornene and cyclooctene
was already evidenced by Grubbs et al. in 1996.16 At that
time, installation of the vinylidene fragment was achieved via
stoichiometric metathesis between [RuCl2(dCHPh)(PCy3)2]
and 1,2-propadiene (allene).
In 2005, Severin and co-workers investigated the reaction
of [RuCl2(p-cymene)]2 (1) with 1 equiv of PCy3 under an
ethylene atmosphere. Under these conditions, the ruthenium
dimer afforded a new type of molecular scaffold (3), in which
a RuCl(η2-C2H4)(PCy3) fragment was connected via three μ-
chloro bridges to a ruthenium-(p-cymene) moiety (Scheme 2).17
In view of the enhancements brought by the replacement of
phosphines with NHCs in monometallic ruthenium-arene
catalyst precursorsfor olefin metathesis18,19 and atom transfer
radical reactions,19,20 we have adopted the same strategy to
synthesize two homobimetallic complexes of type 3 bearing
NHC instead of phosphine ligands, which were found suitable
for promoting olefin metathesis.21 Results from this study
indicated that the ethylene ligand was highly labile and that
adding a small amount of phenylacetylene to the reaction
media had a beneficial influence on the metathetical activity.
We attributed this catalytic enhancement to the in situ forma-
tion of a vinylidene complex.
While work was in progress in our laboratory to validate
this hypothesis, Severin et al. independently reported the
synthesis of two homobimetallic ruthenium-vinylidene
complexes by displacing the ethylene ligand in 3 with tert-
butylacetylene or phenylacetylene (Scheme 3).22 Complexes
4 and 5 were fully characterized by NMR and single-crystal
X-ray diffraction (XRD) analyses, but their catalytic activity
was not assessed. Hence, we decided to study their behavior
in various ruthenium-promoted organic transformations.
We also chose to prepare additional complexes bearing electron-
donating or -withdrawing substituents on the phenylvinyl-
idene ligand and to investigate their catalytic activity. In this
contribution, we disclose the synthesis and characterization
of five new homobimetallic ruthenium-vinylidene complexes
bearing a tricyclohexylphosphine ligand (6-10), and we probe
the influence of their remote aryl substituents on structural
(6) (a) Dragutan, V.; Dragutan, I. Platinum Met. Rev. 2004, 48, 148–
153. (b) Katayama, H.; Ozawa, F. Coord. Chem. Rev. 2004, 248, 1703–1715.
(c) Trost, B. M.; McClory, A. Chem. Asian J. 2008, 3, 164–194.
(7) (a) Bruneau, C.; Dixneuf, P. H. Angew. Chem., Int. Ed. 2006, 45,
2176–2203. (b) Metal Vinylidenes and Allenylidenes in Catalysis; Bruneau,
C.; Dixneuf, P. H., Eds.; Wiley-VCH: Weinheim, 2008.
(8) Lozano-Vila, A.; Monsaert, S.; Bajek, A.; Verpoort, F. Chem.
Rev. 2010, 110, 4865–4909.
(9) Dragutan, I.; Dragutan, V. Platinum Met. Rev. 2006, 50, 81–97.
(10) (a) Dragutan, V.; Dragutan, I.; Verpoort, F. Platinum Met. Rev.
2005, 49, 33–40. (b) Boeda, F.; Clavier, H.; Nolan, S. P. Chem. Commun.
2008, 2726–2740.
(11) Katayama, H.; Ozawa, F. Organometallics 1998, 17, 5190–5196.
(12) (a) Katayama, H.; Ozawa, F. Chem. Lett. 1998, 67–68.
(b) Katayama, H.; Urushima, H.; Ozawa, F. Chem. Lett. 1999, 369–370.
(c) Katayama, H.; Urushima, H.; Ozawa, F. J. Organomet. Chem. 2000, 606,
16–25. (d) Katayama, H.; Yonezawa, F.; Nagao, M.; Ozawa, F. Macro-
molecules 2002, 35, 1133–1136.
(17) Quebatte, L.; Solari, E.; Scopelliti, R.; Severin, K. Organo-
metallics 2005, 24, 1404–1406.
(18) (a) Delaude, L.; Demonceau, A.; Noels, A. F. Chem. Commun.
2001, 986–987. (b) Delaude, L.; Szypa, M.; Demonceau, A.; Noels, A. F.
Adv. Synth. Catal. 2002, 344, 749–756.
(19) Delaude, L.; Demonceau, A.; Noels, A. F. Curr. Org. Chem.
2006, 10, 203–215.
(20) (a) Delaude, L.; Delfosse, S.; Richel, A.; Demonceau, A.; Noels,
A. F. Chem. Commun. 2003, 1526–1527. (b) Richel, A.; Delfosse, S.;
Cremasco, C.; Delaude, L.; Demonceau, A.; Noels, A. F. Tetrahedron Lett.
2003, 44, 6011–6015.
(13) (a) del Rıo, I.; van Koten, G. Tetrahedron Lett. 1999, 40, 1401–
´
1404. (b) Saoud, M.; Romerosa, A.; Peruzzini, M. Organometallics 2000,
19, 4005–4007. (c) Maya, V. G.; Contreras, A. P.; Canseco, M.-A.; Tlenko-
patchev, M. A. React. Funct. Polym. 2001, 49, 145–150. (d) Contreras, A. P.;
Cerda, A. M.; Tlenkopatchev, M. A. Macromol. Chem. Phys. 2002, 203,
1811–1818.
(14) Louie, J.; Grubbs, R. H. Angew. Chem., Int. Ed. 2001, 40, 247–
249.
(15) Opstal, T.; Verpoort, F. J. Mol. Catal. A: Chem. 2003, 200,
49–61.
(16) Schwab, P.; Grubbs, R. H.; Ziller, J. W. J. Am. Chem. Soc. 1996,
118, 100–110.
(21) Sauvage, X.; Borguet, Y.; Noels, A. F.; Delaude, L.; Demonceau,
A. Adv. Synth. Catal. 2007, 349, 255–265.
(22) Solari, E.; Antonijevic, S.; Gauthier, S.; Scopelliti, R.; Severin,
K. Eur. J. Inorg. Chem. 2007, 367–371.

Article
Organometallics, Vol. 29, No. 24, 2010 6677
Scheme 3. Synthesis of Homobimetallic Ruthenium-Vinylidene
Complexes 4-10
Scheme 4. Microwave-Assisted Synthesis of Homobimetallic
Ruthenium-Vinylidene Complex 5
Table 1. Selected Wavenumbers (cm-1), Chemical Shifts (ppm),
and Coupling Constants (Hz) Recorded on IR Spectroscopy (KBr)
or NMR Spectroscopy (CD2Cl2, 25 °C) for Complexes 5-10
features determined by NMR and XRD spectroscopies. We
also present the results of our investigations on the ROMP of
norbornene and cyclooctene using these catalyst precursors.
A few CM and RCM reactions were also carried out on
model substrates to help us better evaluate the potentials of
[(p-cymene)Ru(μ-Cl)3RuCl(dCdCHR)(PCy3)] species for
olefin metathesis and to gain insight into their mode of action.
ν
CdC
δ Hβ
(d)
4JPH
Hβ
δ CR
(d)
2JPC
CR
δCβ
(s)
δ PCy3
(s)
h
complex
5
6
7
8
9
10
1621
1631
1630
1619
1600
1598
4.87
4.85
4.85
4.86
4.91
5.01
3.2
3.5
3.3
3.0
3.3
3.3
355.0
357.6
356.4
353.6
351.2
348.5
19.4 114.2
19.4 113.7
19.4 114.0
19.4 113.5
19.4 113.6
18.6 113.2
54.2
55.0
55.0
54.9
54.9
55.4
Results and Discussion
Synthesis and Characterization of Complexes 6-10. The
reaction of homobimetallic ruthenium-ethylene complex 3
with a 2-fold excess of various phenylacetylene derivatives
was carried out in dichloromethane at room temperature
(Scheme 3). The course of the transformation was monitored
by 31P NMR analysis of samples spiked with CD2Cl2. Within
2 h, the resonance of the starting material located at 44.3 ppm
vanished, while a single new line at ca. 55 ppm indicated the
clean formation of the desired products. A simple workup
involving solvent removal and washing of the alkyne in
excess with n-pentane afforded new arylvinylidene com-
plexes 6-10, with substituents ranging from electron-donating
(like the para-methoxy group in 6) to strongly electron-
withdrawing (such as the bis-meta-trifluoromethyl groups in
10). Our procedure closely matched the one reported by Severin
et al. for the preparation of compounds 4 and 5, starting from
tert-butylacetylene and phenylacetylene, respectively.22 High
yields (>80%) were obtained in all cases. The mechanism
most likely involves the displacement of the labile ethylene
ligand, followed by an alkyne-to-vinylidene tautomerization.23
A similar reaction with a face-bridged Ru dimer was also
reported by Fogg et al., who found that the addition of tert-
butylacetylene to [(dcypb)ClRu(μ-Cl)3Ru(dcypb)(N2)] (dcypb
is 1,4-bis(dicyclohexylphosphino)butane) afforded the vinylidene
complex [(dcypb)ClRu(μ-Cl)3Ru(dCdCH-t-Bu)(dcypb)] with
evolution of dinitrogen gas.24
31P NMR analysis revealed the formation of [(p-cymene)Ru-
(μ-Cl)3RuCl(dCdCHPh)(PCy3)] (5) together with [RuCl2-
(p-cymene)(PCy3)] in a ca. 1:3 molar ratio. Performing the
synthesis in toluene at 70 °C and extending the reaction time
increasedthe selectivitytoward the desired bimetallic product,
but the monometallic intermediate was still present. We reasoned
that recourse to microwave irradiation would prove a more
convenient way to speed up the displacement of the arene
π-ligand.25 Indeed, a rapid optimization of the microwave
heating conditions allowed us to isolate complex 5 in almost
quantitative yield starting from the widely available ruthe-
nium dimer 1 (Scheme 4). Thus, heating a dichloromethane
solution of the reagents for 15 min at 105 °C in a pressure vial
led to the pure homobimetallic vinylidene compound in 93%
yield, with no trace of monometallic [RuCl2(p-cymene)(PCy3)]
byproduct detectable by 31P NMR spectroscopy.
Complexes 5-10 were isolated as stable microcrystalline
solids that could be stored in open-air vials for more than a
year. They readily dissolved in dichloromethane and chloro-
form, but were only sparingly soluble in THF, benzene, or
toluene, and quickly decomposed in acetonitrile. They were
fully characterized by various analytical techniques, viz., 1H,
13C, and 31P NMR, IR, XRD, and elemental analyses. Their
most salient spectroscopic features are summarized in Table
1. In addition to strong aliphatic C-H stretching vibration
bands at 2928 and 2849 cm-1, a sharp absorption due to the
vinylidene CdC bond stretching was visible on IR spectros-
copy at wavenumbers ranging from 1598 to 1631 cm-1. 1H
NMR analysis confirmed that the vinylidene and the arene
ligands were coordinated to ruthenium in a 1:1 ratio. This
stoichiometry was most easily established by comparing the
integrals of the signals arising from the vinylidene proton,
In order to streamline the preparation of homobimetallic
ruthenium-vinylidene complexes, we investigated the possi-
bility of starting directly from dimer 1 instead of intermediate 3,
bearing a sacrificial ethylene ligand. Unsubstituted phenyl-
vinylidene derivative 5 was chosen as a representative target
for these experiments. Preliminary tests were carried out by
heating a mixture of [RuCl2(p-cymene)]2, 1 equiv of tricyclo-
hexylphosphine, and a small excess of phenylacetylene (1.2
equiv) in dichloromethane at 40 °C under argon. After 16 h,
(25) (a) Borguet, Y.; Richel, A.; Delfosse, S.; Leclerc, A.; Delaude, L.;
Demonceau, A. Tetrahedron Lett. 2007, 48, 6334–6338. (b) Delfosse, S.;
Borguet, Y.; Delaude, L.; Demonceau, A. Macromol. Rapid Commun. 2007,
28, 492–503. (c) Nicks, F.; Libert, L.; Delaude, L.; Demonceau, A. Aust. J.
Chem. 2009, 62, 227–231. (d) Albrecht, C.; Gauthier, S.; Wolf, J.; Scopelliti,
R.; Severin, K. Eur. J. Inorg. Chem. 2009, 1003–1010.
(23) Wakatsuki, Y. J. Organomet. Chem. 2004, 689, 4092–4109.
(24) Amoroso, D.; Yap, G. P. A.; Fogg, D. E. Organometallics 2002,
21, 3335–3343.

6678 Organometallics, Vol. 29, No. 24, 2010
Borguet et al.
which resonated as a doublet at about 4.9 ppm due to its
small, albeit visible 4JPH coupling with the phosphorus atom,
together with the methine proton of the p-cymene isopropyl
group that appeared as a septet at 2.95 ppm.
In 13C{1H} NMR spectroscopy, the most interesting
signals were observed in the downfield region of the spectra.
A noticeable feature was the highly deshielded doublet due to
the RudCdC R-carbon atom located at about 350 ppm
(Table 1). The fine structure of this signal was ascribed once
again to the occurrence of a strong coupling with the phos-
2
phorus nucleus, in this case a JPC coupling through the
metal center. The corresponding vinylidene Cβ remote end
resonated as a singlet at ca. 114 ppm. It is noteworthy that the
six aromatic carbons of the η6-arene ligand gave rise to six
distinct absorptions. Further evidence for the asymmetric
nature of complexes 6-10 came from the observation of two
separate doublets for the methyl groups of the p-cymene
Figure 1. Plot of Δδ CR vs Hammett σ-constants for complexes
5-10.
1
isopropyl units in H NMR spectroscopy, although these
signals often collapsed into a pseudotriplet, due to their
proximity. Last but not least, the 31P{1H} NMR spectra
displayed only one sharp absorption at ca. 55 ppm (Table 1).
The chemical shift of this signal remained roughly unaffected
(less than 1 ppm variation) when the substituents on the
vinylidene ligand were modified.
as a reference. Gratifyingly, a very good linear relationship
between the two sets of data was observed (Figure 1). This
means that the electron density at CR is directly related to the
polar and resonance effects of the aryl ring substituents.
None of the other 1H, 13C, or 31P nuclei within the P-Rud
CRdCHβ-C6H4X sequence afforded a satisfactory correla-
Hammett Correlation. Although the Hammett equation (eq 1)
was originally formulated to quantify the effect of structure on
reactivity,26 it was also applied to correlate physical measure-
ments, such as NMR chemical shifts (eq 2),27 IR absorp-
tions,28 or redox potentials,28,29 with the electron-donating or -
withdrawing properties of substituents attached to a phenyl
ring. Attempts to correlate the chemical shifts of side-chain
protons with Hammett σ-constants have met with success in
some instances, but in the majority of cases, the F values are
rather small and obtaining data of sufficient precision for mean-
ingful analyses remains difficult.30 Correlations based on 13C
NMR spectroscopy usually give much better results, and the
terminal carbon atom of a vinyl group attached to a meta- or
para-substituted aromatic ring is particularly sensitive to sub-
stituent effects.31
tion, and the IR ν CdC wavenumber was also of no use for
h
that matter. Yet, the fact that the electronic properties of the
vinylidene carbon atom coordinated to the metal center
could be fine-tuned simply by changing the substituent on
a distant meta or para position was deemed a good omen for
potential catalytic applications.
Crystal Structures. The solid-state structures of homobi-
metallic Ru-vinylidene complexes 6-10 were determined
by X-ray diffraction analysis (XRD). Single crystals of all
five compounds were obtained by slow diffusion of cyclo-
hexane into a dichloromethane solution. In several cases,
these solvents cocrystallized with the organometallic products
in variable proportions (see Supporting Information for
details). Before discussing the results of this crystallographic
study, it should be pointed out that compounds 6-10 are
chiral due to the presence of two stereogenic centers located
on the ruthenium atoms and that these asymmetric units are
not independent, because of geometric constraints imposed
by the three μ-chloro bridges. Hence, our syntheses always
afforded racemic mixtures of two enantiomers. Moreover,
1H and 13C NMR analyses had shown that the products were
configurationally stable in solution and that the p-cymene
ligand did not rotate freely on the NMR time scale (vide
supra). Further evidence of this asymmetry was obtained by
XRD. Thus, the unit cells of complexes 7, 8, and 9 contained
two nonequivalent molecules, which differed primarily in the
relative orientation of their p-cymene and vinylidene ligands
(Figure 2). Of course, the mirror images of these two
molecules were also present. This is in sharp contrast with
the crystals obtained previously for complexes 4 and 5, which
contained only molecules having the isopropyl group of their
p-cymene ligand oriented in the same direction as the vinyl-
idene ligand (designated as configuration A in Figure 2).22
Complex 6 provided an additional example of this spatial
arrangement, while the reverse situation occurred with com-
plex 10. Indeed, the solid-state structure of this compound
revealed only the presence of molecules with a roughly
antiparallel orientation between the isopropyl group of their
p-cymene ligand and the vinylidene fragment (designated as
k
ð1Þ
ð2Þ
log
¼ σF
k0
δ - δ0 ¼ σF
This prompted us to examine the possible correlation between
the chemical shift of the Ru-bonded vinylidene carbon atom
in complexes 6-10 (labeled CR in Table 1) and the Hammett
σ-constants for p-OMe, p-Me, p-Cl, p-CF3, and m-(CF3)2
groups,32 using unsubstituted phenylvinylidene compound 5
ꢀ
(26) (a) Hammett, L. P. J. Am. Chem. Soc. 1937, 59, 96–103. (b) Jaffe,
H. H. Chem. Rev. 1953, 53, 191–261.
(27) Ewing, D. F. In Correlation Analysis in Chemistry: Recent
Advances; Chapman, N. B.; Shorter, J., Eds.; Plenum: New York, 1978;
p 357.
(28) Paul, F.; Ellis, B. G.; Bruce, M. I.; Toupet, L.; Roisnel, T.;
Costuas, K.; Halet, J.-M.; Lapinte, C. Organometallics 2006, 25,
649–665.
(29) (a) Halbach, D. P.; Hamaker, C. G. J. Organomet. Chem. 2006,
691, 3349–3361. (b) Hamaker, C. G.; Halbach, D. P. Polyhedron 2009, 28,
2228–2232.
(30) Happer, D. A. R. Aust. J. Chem. 1976, 29, 2607–2614.
(31) See, for example: (a) Nagy, A. G. J. Organomet. Chem. 1985,
291, 335–340. (b) Aun, C. E.; Clarkson, T. J.; Happer, D. A. R. J. Chem.
ꢂ
ꢀ
ꢂ ꢀ
ꢀ
Soc., Perkin Trans. 2 1990, 645–649. (c) Jovanovic, B. Z.; Misic-Vukovic,
ꢀ
ꢀ
M.; Marinkovic, A. D.; Csanadi, J. J. Mol. Struct. 1999, 482-483, 371–374.
(32) Hansch, C.; Leo, A.; Taft, R. W. Chem. Rev. 1991, 91, 165–195.

Article
Organometallics, Vol. 29, No. 24, 2010 6679
Figure 2. Atom-numbering system and molecular structures of homobimetallic ruthenium-vinylidene complex 7 showing two
nonequivalent configurations of the p-cymene ligand relative to the vinylidene fragment.
Figure 3. ORTEP diagrams of complexes 6-9 showing their molecular structure in configuration A. Thermal ellipsoids were drawn at
the 50% probability level. For the sake of clarity, hydrogen atoms and the solvent molecules that cocrystallized with complexes 7, 8, and
9 were omitted. See Tables 2 and 3 for selected bond lengths and angles.
Figure 4. ORTEP diagrams of complexes 7-10 showing their molecular structure in configuration B. Thermal ellipsoids were drawn
at the 50% probability level. For the sake of clarity, hydrogen atoms and the cocrystallized solvent molecules were omitted. See Tables 4
and 5 for selected bond lengths and angles.
configuration B in Figure 2). ORTEP plots of individual
molecules displaying configurations A and B are represented
in Figures 3 and 4, respectively, while relevant bond lengths
and angles are listed in Tables 2-5. For the sake of compar-
ison, data pertaining to complex 5 are also included.
Except for the orientation of the p-cymene ligand, the
molecular structures of complexes 6-10 were very similar to
those already reported for compounds 4 and 5.22 They consisted
of a (p-cymene)Ru unit connected to the RuCl(dCdCHAr)-
(PCy3) fragment via three μ-chloro bridges. A typical piano
stool geometry was observed for the (arene)RuCl3 moiety,
with three Ru-Cl bonds of similar lengths (2.39-2.46 A).
The other ruthenium atom lay in a highly distorted octa-
hedral environment with Ru-Cl bond lengths varying between
2.34 and 2.63 A. In all the complexes under scrutiny, the
shortest distances corresponded to the terminal halogen (Cl1
or Cl5). The bridging chloro substituent trans to it (Cl2 or
Cl6) was slightly more distant from the metal center. A
further elongation was recorded for Cl3 (or Cl7) facing the
tricyclohexylphosphine ligand. The vinylidene unit displayed
the strongest trans effect within the coordination sphere of
the octahedral ruthenium center and was responsible for the

6680 Organometallics, Vol. 29, No. 24, 2010
Borguet et al.
˚
Table 2. Selected Bond Lengths (A) Derived from the Crystal
Scheme 5. ROMP of Norbornene
Structures of Complexes 5-9 (Configuration A)a
5b
6
7
8
9
Ru1-Cl1 2.362(2)
Ru1-Cl2 2.403(2)
Ru1-Cl3 2.541(2)
Ru1-Cl4 2.630(2)
2.360(3)
2.389(3)
2.531(3)
2.617(3)
2.303(3)
2.3614(11) 2.3609(8) 2.358(3)
2.4132(16) 2.4137(8) 2.408(4)
2.5059(12) 2.5103(8) 2.499(5)
2.6012(12) 2.6081(8) 2.592(4)
2.3107(12) 2.3150(8) 2.313(4)
Table 6. ROMP of Norbornene Catalyzed by Complex 5 at
Various Temperaturesa
Ru1-P1
2.325(2)
temperature
(°C)
conversion
(%)b
yield
(%)
10-3Mn
Mw/Mn
σcis
c
c
d
Ru1-C1 1.778(8)
1.786(13) 1.785(5)
1.329(11) 1.329(17) 1.314(6)
1.780(3)
1.311(5)
1.799(15)
1.243(19)
C1-C2
Ru2-Cl2 2.428(2)
Ru2-Cl3 2.437(2)
Ru2-Cl4 2.427(2)
2.440(3)
2.429(4)
2.422(3)
2.4379(12) 2.4349(8) 2.429(4)
2.3923(13) 2.3977(8) 2.386(6)
2.4568(13) 2.4565(9) 2.449(5)
30
40
60
95
97
>99
76
81
80
77
39
22
2.2
2.9
4.3
0.15
0.14
0.16
a See Figure 2 for the atom-numbering system. b Data from ref 22.
a Experimental conditions: Ru cat. (0.015 mmol), norbornene (3.75
mmol), PhCl (12.5 mL), 2 h under Ar. b Determined by GC using
norbornane as internal standard. c Determined by SEC in THF with
polystyrene calibration. d Fraction of cis double bonds in the polymer,
determined by 13C NMR spectroscopy.
Table 3. Selected Bond Angles (deg) Derived from the Crystal
Structures of Complexes 5-9 (Configuration A)a
5b
6
7
8
9
Cl6 < Ru3-Cl7 < Ru3-Cl8). Moreover, the trans effect of the
vinylidene unit in homobimetallic complexes 4-10 may be
ranked slightly inferior to that of an indenylidene group33
and considerably stronger than the one exerted by the ethylene
ligand in complex 3.22
Ru1-Cl2-Ru2 87.61(7) 86.58(10) 85.58(4) 85.75(3) 85.53(12)
Ru1-Cl3-Ru2 84.39(7) 83.73(11) 84.53(4) 84.43(3) 84.46(13)
Ru1-Cl4-Ru2 82.69(7) 82.07(10) 81.26(4) 81.23(3) 81.27(15)
P1-Ru1-Cl3 176.17(8) 174.18(12) 177.45(4) 178.04(3) 177.48(14)
Cl1-Ru1-Cl2 164.32(8) 166.58(11) 164.74(4) 163.97(3) 165.19(13)
C1-Ru1-Cl4 167.0(3) 165.1(4) 167.72(14) 167.35(10) 167.2(5)
Cl4-Ru1-Cl2 77.17(7) 77.76(11) 78.10(4) 78.85(3) 78.06(13)
Cl1-Ru1-Cl4 88.85(7) 89.69(11) 89.52(4) 88.93(3) 89.69(12)
Cl2-Ru1-C1 96.0(3) 95.1(4)
Cl1-Ru1-C1 96.3(3) 96.0(4)
a See Figure 2 for the atom-numbering system. b Data from ref 22.
The RudC and CdC bond lengths within the polyunsa-
turated fragments of complexes 5-10 were similar to those
recorded for complex 422 and other monometallic, hexacoor-
dinated ruthenium-vinylidene complexes, such as [RuCl-
(dCdCHPh)(κ2P,O-Pri2PCH2CH2OMe)2](OTf)34 or [TpRuCl-
(dCdCHPh)(PPh3)] (Tp is tris(pyrazolyl)borate).35 Further-
more, they remained seemingly independent from the exact
nature of the aryl group attached to them. At ca. 1.78 A, the
vinylidene RudC bonds in complexes 4-10 are, however,
slightly longer than in unsaturated, pentacoordinated com-
plexes of type 2 having a square-pyramidal structure with the
vinylidene ligand in the apical position, like [RuCl2(dCd
CHPh)(PPri3)2] (1.750(4) A) and [RuCl2(dCdCHPh)(PCy3)2]
(1.761(2) A) (cf. Scheme 1).11 Conversely, they are significantly
shorter than the RudC distances measured in pincer com-
plexes [RuCl2(dCdCHPh)(κ3P,N,P-dcpmp)] (1.816(6) A)
and [RuCl2{dCdC(SiMe3)Ph}(κ3P,N,P-dcpmp)] (1.845(4) A)
(dcpmp is 2,6-bis{(dicyclohexylphosphino)methyl}pyridine),
in which the vinylidene fragment is facing a nitrogen donor.36
Thus, the RudCR distance between a ruthenium center and
the carbon atom of the vinylidene ligand attached to it is
mainly determined by the presence or the absence of a coaxial
ligand and its donor strength, while the substituents on Cβ
are considerably less influential.
93.76(14) 94.08(10) 94.2(5)
96.90(14) 97.27(10) 96.2(5)
˚
Table 4. Selected Bond Lengths (A) Derived from the Crystal
Structures of Complexes 7-10 (Configuration B)a
7
8
9
10
Ru3-Cl5
Ru3-Cl6
Ru3-Cl7
Ru3-Cl8
Ru3-P2
Ru3-C37
C37-C38
Ru4-Cl6
Ru4-Cl7
Ru4-Cl8
2.3627(11)
2.4015(11)
2.5171(10)
2.6004(11)
2.3139(12)
1.779(4)
2.3585(8)
2.4020(7)
2.5167(8)
2.5991(8)
2.3173(8)
1.776(3)
2.357(4)
2.398(4)
2.523(4)
2.601(4)
2.319(4)
1.774(14)
1.296(18)
2.456(4)
2.408(4)
2.434(4)
2.3416(13)
2.4043(13)
2.5127(14)
2.6176(14)
2.3270(14)
1.777(6)
1.313(6)
1.325(4)
1.313(8)
2.4574(11)
2.4204(11)
2.4104(12)
2.4578(7)
2.4209(8)
2.4117(8)
2.4484(13)
2.4197(14)
2.4166(14)
a See Figure 2 for the atom-numbering system.
Table 5. Selected Bond Angles (deg) Derived from the Crystal
Structures of Complexes 7-10 (Configuration B)a
7
8
9
10
ROMP of Norbornene and Cyclooctene. In order to assess
the catalytic activity of homobimetallic ruthenium-vinylidene
complexes in olefin metathesis, we first investigated the ROMP
of norbornene in chlorobenzene using a monomer-to-catalyst
ratio of 250 (Scheme 5). Due to its high strain, ring-opening
metathesis of this bicyclic monomer is particularly easy to
carry out and occurs under almost any circumstances, provided
that enough time is allowed for the reaction.37 It was therefore
not surprising to observe the full conversion of norbornene
into polynorbornene using unsubstituted phenylvinylidene
Ru3-Cl6-Ru4
Ru3-Cl7-Ru4
Ru3-Cl8-Ru4
P2-Ru3-Cl7
Cl5-Ru3-Cl6
Cl8-Ru3-C37
Cl6-Ru3-Cl8
Cl5-Ru3-Cl8
Cl6-Ru3-C37
Cl5-Ru3-C37
85.65(3)
83.95(3)
82.38(3)
177.55(4)
165.73(4)
85.66(2)
83.99(2)
82.42(2)
177.61(3)
165.76(3)
85.90(13)
84.22(14)
82.04(13)
177.82(14) 177.03(5)
165.16(13) 164.64(5)
85.97(4)
84.24(4)
82.08(4)
165.66(15) 165.75(10) 165.6(4)
164.86(17)
77.11(4)
88.04(5)
97.70(17)
95.69(17)
77.71(3)
88.96(4)
96.21(14)
95.58(14)
77.70(2)
89.04(3)
96.21(10)
95.50(10)
77.73(13)
88.34(12)
96.8(5)
95.5(5)
a See Figure 2 for the atom-numbering system.
significantly longer distances measured for Ru1-Cl4 (or
Ru3-Cl8) in comparison with other Ru-Cl bonds. Thus,
the bond lengths always increased in the order Ru1-Cl1<
Ru1-Cl2 < Ru1-Cl3 < Ru1-Cl4 (or Ru3-Cl5 < Ru3-
(34) Martın, M.; Gevert, O.; Werner, H. J. Chem. Soc., Dalton Trans.
´
1996, 2275–2283.
(35) Slugovc, C.; Mereiter, K.; Zobetz, E.; Schmid, R.; Kirchner, K.
Organometallics 1996, 15, 5275–5277.
(36) Katayama, H.; Wada, C.; Taniguchi, K.; Ozawa, F. Organo-
metallics 2002, 21, 3285–3291.
(37) Lebedev, B.; Smirnova, N.; Kiparisova, Y.; Makovetsky, K.
(33) Sauvage, X.; Borguet, Y.; Zaragoza, G.; Demonceau, A.;
Delaude, L. Adv. Synth. Catal. 2009, 351, 441–455.
Makromol. Chem. 1992, 193, 1399–1411.

Article
Organometallics, Vol. 29, No. 24, 2010 6681
Scheme 6. ROMP of Cyclooctene
Table 8. Influence of the Reaction Time on the ROMP of
Cyclooctene Catalyzed by Complex 5 at 140 °Ca
time (h) conversion (%)b yield (%) 10-3Mn
Mw/Mn
σcis
c
c
d
0.5
2
6
8.5
31
33
34
34
19
26
25
26
100
115
122
119
1.5
1.8
1.6
2.0
0.58
0.46
0.41
0.54
Table 7. Influence of the Temperature on the ROMP of
Cyclooctene Catalyzed by Complex 5a
a Experimental conditions: Ru cat. (0.0075 mmol), cyclooctene (3.75
mmol), PhCl (2.5 mL), 140 °C under Ar. b-d See footnotes in Table 7 for
details.
temperature
(°C)
conversion
(%)b
yield
(%)
10-3Mn
Mw/Mn
σcis
c
c
d
100
120
140
160
140e
2
20
33
24
50
<1
15
26
19
40
85
115
71
1.9
1.8
1.7
1.7
0.47
0.46
0.47
0.52
Table 9. Influence of the Lewis Acid Cocatalyst on the ROMP of
Cyclooctene Catalyzed by Complex 5 at 60 °Ca
202
Lewis acid conversion (%)b yield (%) 10-3Mn Mw/Mn σcis
c
c
d
a Experimental conditions: Ru cat. (0.0075 mmol), cyclooctene (3.75
mmol), PhCl (2.5 mL), 2 h under Ar. b Determined by GC using cyclo-
octane as internal standard. c Determined by SEC in THF with poly-
styrene calibration. d Fraction of cis double bonds in the polymer, deter-
mined by 13C NMR spectroscopy. e Reaction in a sealed tube heated in
an oil bath instead of a monomodal microwave reactor.
AlCl3
AgOTf
BF3 OEt2
FeCl3
Sc(OTf)3
TiCl4
>99
>99
48
83
93
50
90
93
84
82
37
64
80
37
80
81
453
427
544
439
523
322
125
270
1.26
1.90
1.24
1.29
1.31
1.38
1.63
1.52
0.48
0.46
0.51
0.39
0.43
0.45
0.44
0.42
3
WCl6
ZnCl2
compound 5 as a model initiator at 60 °C (Table 6). Even
when the temperature was decreased to 30 or 40 °C, high
molecular weight polymers with a broad polydispersity index
and mostly trans double bonds were isolated in high yields.
Similar results were already obtained when homobimetallic
ruthenium-ethylene complex 3 served as catalyst precur-
sor.21 Thus, the presence of a vinylidene fragment is of little
added value for this reaction. Activation of the monomer via
ill-defined mechanisms is sufficient to generate metathetically
active species from the bulk of the ruthenium precursor.38
The consequence is a poor control over the initiation and the
formation of polymers with a broad molecular weight dis-
tribution. It should be pointed out, however, that narrow
polydispersities are not essential in many polymer applications
and can even hamper processing in some cases.39
Compared to norbornene, cyclooctene is significantly less
strained and more difficult to ring-open.40 Hence, formation
of polyoctenamer occurs only at a reasonable rate with highly
active catalytic systems. In order to determine whether or not
homobimetallic Ru-vinylidene complexes fall into this cate-
gory, polymerization tests were performed in chlorobenzene
using a monomer-to-catalyst ratio of 500 (Scheme 6). A mono-
modal microwave reactor served as a convenient heating
device for these experiments. Under these conditions, only
traces of polymer were isolated when complex 5 served as
catalyst precursor for 2 h at 60 or 100 °C (Table 7). When the
reaction was carried out at higher temperatures (120 and 140 °C),
conversion increased and high molecular weight polymers
with an almost equimolar proportion of cis and trans double
bonds were isolated in modest yields. A further experiment
performed at 160 °C eventually led to a drop in activity, most
likely due to a thermal decomposition of the catalyst. Switching
from microwave irradiation to conventional heating in an oil
bath at 140 °C slightly increased the monomer conversion
a Experimental conditions: Ru cat. (0.0075 mmol), Lewis acid (0.0075
mmol), cyclooctene (3.75 mmol), PhCl (2.5 mL), 2 h at 60 °C under Ar.
b-d See footnotes in Table 7 for details.
and the polymer yield, but did not affect the microstructure
of the macromolecular chains.
Next, we examined the influence of the reaction time on
the ROMP of cyclooctene carried out in a microwave reactor
at 140 °C using complex 5 as initiator (Table 8). These
experiments showed that conversion stopped increasing al-
ready after ca. 30 min and that extending the reaction time
beyond this period did not significantly improve the outcome
of the polymerization. Thus, the need for a long induction
time before the active species are generated from the catalyst
precursor may be ruled out under these conditions.
Because the two ruthenium centers in complex 5 are
coordinatively saturated, dissociation of a ligand is required
prior to any interaction between the catalyst and the mono-
mer. For the ROMP of cyclooctene, results gathered in
Tables 7 and 8 clearly indicated that a thermal activation
was not sufficient to promote this mandatory step. Alterna-
tively, we reasoned that adding a Lewis acid to the reaction
media might help generate active species. Indeed, a 1998 report
by Ozawa and co-workers had shown that the catalytic
activity of the 18-electron monometallic complex [TpRuCl-
(dCdCHPh)(PPh3)] (Tp is tris(pyrazolyl)borate) in the ROMP
of norbornene was considerably enhanced by the addition of
boron trifluoride diethyl etherate.41 This additive was assumed
to facilitate the decoordination of a phosphine ligand. Accord-
ingly, we have probed the influence of various Lewis acids on
the ROMP of cyclooctene catalyzed by complex 5 (Table 9).
A mixture of the two catalyst components in 1:1 molar propor-
tion was suspended in chlorobenzene under argon and stirred
for a few minutes in an oil bath at 60 °C before the monomer
was added. Under these conditions, BF3 OEt2 displayed a
ꢀ
(38) (a) Bencze, L.; Kraut-Vass, A.; Prokai, L. Chem. Commun.
3
moderate, albeit encouraging, activating effect. Addition of
1 equiv of this cocatalyst to the ruthenium-phenylvinylidene
precursor afforded a high molecular weight polyoctenamer
in 37% yield after 2 h, whereas the use of complex 5 alone did
not lead to any conversion of the cycloolefin within the same
1985, 911–912. (b) M€uhlebach, A.; Bernhard, P.; B€uhler, N.; Karlen, T.; Ludi,
A. J. Mol. Catal. 1994, 90, 143–156. (c) Al Samak, B.; Amir-Ebrahimi, V.;
Corry, D. G.; Hamilton, J. G.; Rigby, S.; Rooney, J. J.; Thompson, J. M.
J. Mol. Catal. A: Chem. 2000, 160, 13–21. (d) Alvarez, P.; Gimeno, J.;
Lastra, P. Organometallics 2002, 21, 5678–5680.
€
(39) Bohm, L. L.; Enderle, H. F.; Fleissner, M. In Catalyst Design for
Tailor-Made Polyolefins; Soga, K.; Terano, M., Eds.; Stud. Surf. Sci.
Catal., Vol. 89; Elsevier Science: Amsterdam, 1994; pp 351-363.
(40) Lebedev, B.; Smirnova, N. Macromol. Chem. Phys. 1994, 195,
35–63.
(41) Katayama, H.; Yoshida, T.; Ozawa, F. J. Organomet. Chem.
1998, 562, 203–206.

6682 Organometallics, Vol. 29, No. 24, 2010
Borguet et al.
period of time. Another liquid catalyst modifier, TiCl4, behaved
in pretty much the same way as the boron trifluoride adduct.
Recourse to other strong Lewis acids based on transition
metals, such as FeCl3, Sc(OTf)3, WCl6, or ZnCl2 further
boosted the catalytic activity and led to almost quantitative
conversions of cyclooctene. Although it is a weak Lewis acid,
the halide-abstracting AgOTf salt was another serious con-
tender. Yet, the best results were obtained with aluminum
chloride, which allowed a total consumption of the monomer
within 2 h and the formation of high molecular weight
polyoctenamer with a rather narrow polydispersity index
(Mw/Mn=1.25) and an almost equimolar proportion of cis
and trans double bonds (σcis = 0.48). Similar yields and
macromolecular parameters were obtained when the molar
proportion of AlCl3 relative to complex 5 was doubled and
when a 1 or 2 h delay was introduced prior to the addition of
the monomer. These experiments confirmed that the Lewis
acid acted as a fast and efficient activator. Conversely,
Table 10. ROMP of Cyclooctene Catalyzed by Complexes 5-10
in the Presence of AlCl3 at 60 °Ca
complex conversion (%)b yield (%) 10-3Mn
Mw/Mn
σcis
c
c
d
5
6
7
8
9
10
>99
>99
95
>99
33
86
82
86
84
24
0
453
392
341
355
149
1.25
1.29
1.37
1.79
1.96
0.48
0.50
0.43
0.37
0.35
<5
a Experimental conditions: Ru cat. (0.0075 mmol), AlCl3 (0.0075 mmol),
cyclooctene (3.75 mmol), PhCl (2.5 mL), 2 h at 60 °C under Ar. b-d See
footnotes in Table 7 for details.
Scheme 7. Self-Metathesis of Styrene
complex 4 and AlCl3 was tested under the same experimental
conditions, no sign of reaction was detected after 2 h at 60 °C.
This was an unexpected result, because related monometallic
complexes of the type [RuCl2(dCdCHBut)(PR3)2] (PR3 =
PPri3 or PCy3) were found to catalyze various ROMP and
RCM reactions.12 Therefore, an aryl group on the β-position
of the vinylidene unit seems essential to the catalytic activity
of homobimetallic species 4-10, but except for strongly
electron-withdrawing groups, the exact nature of the sub-
stituents on the aromatic ring has little influence on the poly-
merization outcome.
Other Metathesis Reactions. In order to further evaluate
the scope of homobimetallic ruthenium-vinylidene catalysts
for olefin metathesis, we have carried out the homodimer-
ization of styrene using complexes 5 and 6 (Scheme 7). In the
absence of any cocatalyst, no reaction took place after 2 h at
85 °C under low catalyst loading conditions (0.2 mol %).
Addition of AlCl3 (2 equiv) led to the formation of trans-
stilbene, as evidenced by GC analysis of the reaction mixture,
but conversion remained very limited (<5%). Previous work
from our group had already shown that first-generation
ruthenium catalysts were not suitable for styrene homocoupling,
whether they weremono-43 or bimetallic.21 Second-generation
NHC-based species were required to achieve satisfactory re-
action rates, in good agreement with the general ranking of
olefin reactivity proposed by Grubbs and co-workers for
predicting the selectivity of cross-metatheses.44 Therefore,
we did not further investigate this transformation.
Instead, we focused on the RCM of diethyl 2,2-diallylmal-
onate (DEDAM). Preliminary tests were carried out in toluene
at 85 °C using 2 mol % of complex 5. Under these conditions,
conversion of the R,ω-diene stagnated below 10% even when
the reaction time was extended to more than 5 h (Figure 5).
These results were only slightly superior to those obtained
with the ruthenium-ethylene complex 3. When phenylacetylene
(6 mol %) was added to the reaction media, consumption of
the substrate occurred much faster, and diethyl cyclopent-3-ene-
1,1-dicarboxylate was the sole product formed (Scheme 8). It
is noteworthy that complexes 3 and 5 displayed the same
reaction profile in the presence of the terminal alkyne co-
catalyst. Thus, the role of this additive goes beyond the replace-
ment of the ethylene ligand with a vinylidene fragment, as we
neither SnCl2, CuCl2, CuCl2 xH2O, nor CuCl afforded
3
any polymer. This last result has important mechanistic
implications (vide infra), because copper(I) salts are well-known
phosphine scavengers, which have been frequently employed
to displace PCy3 from metathetically active ruthenium-
alkylidene or -indenylidene complexes.42 Therefore, the
absence of ROMP when complex 5 was treated with CuCl
suggests that the formation of active species does not involve
a dissociation of the tricyclohexylphosphine ligand.
Having identified aluminum chloride as the most suitable
cocatalyst to activate ruthenium-phenylvinylidene complex
5, we launched a final series of ROMP experiments to probe
how the introduction of electron-donating or -withdrawing
groups on the vinylidene moiety affected the metathetical
process. Thus, complexes 6-10 were tested as catalyst pre-
cursors for the polymerization of cyclooctene in the presence
of AlCl3 (1 equiv) at 60 °C (Table 10). After 2 h of reaction,
high yields of polyoctenamer were isolated with complexes 6
and 7 bearing para-methoxy and para-methyl donor groups.
No major differences were observed in the polymer micro-
structures compared to those obtained with unsubstituted
compound 5. In all three cases, the molecular weights
remained very high, which indicates that only small amounts
of active species were generated, regardless of the actual
substituent present on the arylvinylidene unit. Indeed, the
initiation efficiencies f = Mn,theor/Mn,exp (calculated with
Mn,theor = ([monomer]/[Ru cat.])o ꢀ MWmonomer ꢀ conversion)
never exceeded 15%. Introduction of a para-chloro electron-
withdrawing group in catalyst precursor 8 did not have any
adverse influence on the polymer yield, but slightly increased
the polydispersity index and altered the cis/trans ratio of the
double bonds. These changes became more obvious with
complexes 9 and 10, bearing para-trifluoromethyl and bis-
meta-trifluoromethyl substituents, which gradually lost any
activity. When a 1:1 mixture of tert-butylvinylidene-ruthenium
(42) (a) Dias, E. L.; Nguyen, S. T.; Grubbs, R. H. J. Am. Chem. Soc.
1997, 119, 3887–3897. (b) Wakamatsu, H.; Blechert, S. Angew. Chem., Int.
Ed. 2002, 41, 794–796. (c) Connon, S. J.; Rivard, M.; Zaja, M.; Blechert, S.
Adv. Synth. Catal. 2003, 345, 572–575. (d) Buschmann, N.; Wakamatsu, H.;
Blechert, S. Synlett 2004, 667–670. (e) Vehlow, K.; Maechling, S.; Blechert,
S. Organometallics 2006, 25, 25–28. (f) Grela, K.; Harutyunyan, S.;
Michrowska, A. Angew. Chem., Int. Ed. 2002, 41, 4038–4040. (g) Michrowska,
A.; Harutyunyan, S.; Sashuk, V.; Dolgonos, G.; Grela, K. J. Am. Chem. Soc.
2004, 126, 9318–9325. (h) Bieniek, M.; Michrowska, A.; Guzajski, Ł.; Grela,
K. Organometallics 2007, 26, 1096–1099. (i) Rix, D.; Caijo, F.; Laurent, I.;
Boeda, F.; Clavier, H.; Nolan, S. P.; Mauduit, M. J. Org. Chem. 2008, 73,
4225–4228.
ꢀ
(43) Ferre-Filmon, K.; Delaude, L.; Demonceau, A.; Noels, A. F.
Eur. J. Org. Chem. 2005, 3319–3325.
(44) Chatterjee, A. K.; Choi, T.-L.; Sanders, D. P.; Grubbs, R. H.
J. Am. Chem. Soc. 2003, 125, 11360–11370.

Article
Organometallics, Vol. 29, No. 24, 2010 6683
initially assumed. Its possible implication to the catalytic
process will be further discussed in the next section.
Mechanistic Indications. Except for the facile ROMP of
norbornene, all the other catalytic tests that we performed
required the use of a co-promoter to enhance the metath-
etical activity of complexes 5-10. Recourse to aluminum
chloride afforded a convenient way to initiate the ROMP
of cyclooctene. The Lewis acid most likely favored the
dissociation of a ligand from the saturated homobimetallic
scaffold, thereby leading to a more reactive species able to
coordinate the monomer. Experiments carried out with
copper salts indicated that tricyclohexylphosphine was not
involved in this activation process, whereas the good result
obtained with silver triflate strongly suggested that halide
abstraction was probably at stake. Therefore, we tentatively
propose that aluminum chloride interacts with a bridging
chloro ligand to afford cationic complex 12 and a highly
reactive 14-electron species of type 13, as illustrated for
compound 5 in Scheme 9. The intimate details of the
mechanism remain unclear. However, ESI-MS analyses of
solutions prepared by titrating complex 5 with AlCl3 (1.3
equiv) showed the presence of a major signal centered at
m/z = 577, whose molecular weight and isotopic pattern
matched those computed for the well-known cationic dimer
12.45 Additionally, 31P NMR spectroscopy revealed that a
main product containing a phosphine ligand resonated at
59.0 ppm, but the intermediacy of compound 13 remained
otherwise elusive. Yet, it is noteworthy that a 14-electron
species of the type [RuCl2(dCdCHBut)(NHC)] was also
postulated by Louie and Grubbs to be the key intermediate
in catalytic systems generated in situ by adding 2 equiv of 1,3-
dimesitylimidazolium chloride, sodium tert-butoxide, and
tert-butylacetylene to the [RuCl2(p-cymene)]2 dimer (1).14
For the RCM of DEDAM, we reasoned that AlCl3 would
not be a suitable activator, because of its affinity for the
diester functionality. In this case, addition of a small amount
of phenylacetylene to homobimetallic ruthenium-phenylvinyl-
idene precursor 5 provided an efficient workaround to
boost its activity. A similar enhancement was also observed
starting from ruthenium-ethylene complex 3. Thus, we believe
that the role of the terminal alkyne is 2-fold. First, it displaces
rapidly and quantitatively the labile ethylene ligand of com-
plex 3 to afford the homobimetallic vinylidene precursor 5
(cf. Scheme 3). A direct interaction between this compound
and the olefinic substrate is able to generate a metathetically
active species of type A (Scheme 10). However, this process is
rather inefficient and slow. In the presence of excess phenyl-
acetylene, complex 5 has the possibility to react with an alkyne
Figure 5. Time course of the RCM of DEDAM catalyzed by
complexes 3 (2,4) and 5 (b,O) at 85 °C in the presence (2,b) or
in the absence (4,O) of phenylacetylene.
Scheme 8. RCM of DEDAM
Scheme 9. Possible Activation of Complex 5 by AlCl3
Scheme 10. Possible Mechanism for the Activation of Homobimetallic Ruthenium-Ethylene or -Vinylidene Complexes by
Phenylacetylene

6684 Organometallics, Vol. 29, No. 24, 2010
Borguet et al.
instead of an alkene partner. This second intervention should
ease the required opening of a coordination site, because the
alkyne is a better π-donor and a sterically less encumbered
ligand than the alkene. Moreover, a subsequent enyne meta-
thesis event would then lead to the formation of an alkylidene
species of type B, which is a faster initiator than vinylidene
intermediate A. As a matter of fact, previous work from our
group had already shown that phosphino complex 3 and
related NHC-based catalyst precursors efficiently promoted
the enyne metathesis between phenylacetylene and ethylene.21
double bonds were obtained almost quantitatively within
2 h. When the ROMP of cyclooctene was performed at tem-
peratures ranging from 60 to 160 °C, only modest yields of
polyoctenamer were isolated after 2 h. Thus, thermal activa-
tion was not sufficient to trigger the catalytic activity of
complex 5 using a low-strain monomer. Alternatively, we
found that aluminum chloride was a suitable cocatalyst to
allow the formation of high molecular weight polyoctenamer
with a rather narrow polydispersity index (Mw/Mn =1.25)
and an almost equimolar proportion of cis and trans double
bonds within 2 h at 60 °C. No major changes were observed
in the polymer yields and microstructures when complexes 6
and 7 bearing para-methoxy and para-methyl donor groups
on their aryl rings were employed as catalyst precursors. On
the other hand, compound 9 or 10 (and to a lesser extent 8),
bearing electron-withdrawing substituents, were significantly
less active.
Model vinylidene derivative 5 and its ruthenium-ethylene
parent (3) were equally inactive in the RCM of DEDAM,
and they both required the addition of phenylacetylene to
achieve the ring-closure of this R,ω-diene. Thus, the role of
the terminal alkyne cocatalyst goes beyond the facile re-
placement of the η2-alkene ligand with a vinylidene frag-
ment. As a matter of fact, compounds 5-10 are highly stable,
coordinatively saturated species that cannot enter a catalytic
cycle without prior dissociation of a ligand. Recourse to
AlCl3 is believed to activate a μ-chloro bridge rather than a
phosphine ligand and could result in the formation of the
well-known cationic complex 12 and a metathetically active
14-electron species of the type [RuCl2(dCdCHPh)(PCy3)]
(13). The influence of phenylacetylene is less obvious to
rationalize, although it can be reasonably assumed that an
alkyne has a better chance than an alkene to find its way to
the metal center, where it would allow an enyne metathesis
to take place, thereby affording a highly active mono- or
bimetallic ruthenium-alkylidene species.
Conclusion
The labile ruthenium-ethylene complex 3 is a convenient
starting material to prepare more elaborate homobimetallic
ruthenium-arene architectures containing polyunsaturated
carbon-rich fragments. Its reaction with various phenyl-
acetylenederivatives in dichloromethane at roomtemperature
afforded five new arylvinylidene complexes with substituents
ranging from electron-donating (like the para-methoxy
group in 6) to strongly electron-withdrawing (such as the bis-
meta-trifluoromethyl groups in 10) in high yields. The synth-
esis of unsubstituted phenylvinylidene complex 5 was also
successfully carried out in a single step starting from the
widely available [RuCl2(p-cymene)]2 dimer (1), tricyclohexyl-
phosphine, and a small excess of phenylacetylene under
microwave irradiation. This procedure is particularly attrac-
tive in terms of atom economy and efficiency. All the new
compounds were fully characterized by various analytical
techniques, and the influence of their remote aryl substitu-
ents on structural features was carefully investigated by IR,
NMR, and XRD spectroscopies. A very good linear rela-
tionship was observed between the chemical shift of the
vinylidene R-carbon atom and the Hammett σ-constants
for p-OMe, p-Me, p-H, p-Cl, p-CF3, and m-(CF3)2 groups.
X-ray diffraction analysis showed that the unit cells of
complexes 7, 8, and 9 contained two nonequivalent mole-
cules, which differed primarily in the relative orientation of
their p-cymene and vinylidene ligands. Comparison of the
various bond lengths and angles showed that the introduc-
tion of electron-donating or -withdrawing groups on the aryl
ring of the vinylidene ligand did not have any significant
influence on the magnitude of its trans effect. Moreover, the
distance between a ruthenium center and the carbon atom of
the vinylidene ligand attached to it was mainly determined by
the presence or the absence of a coaxial ligand and its donor
strength, while the substituents on Cβ were considerably less
influential.
Experimental Section
General Comments. Unless otherwise stated, all the syntheses
were carried out under a dry argon atmosphere using standard
Schlenk techniques. Solvents were distilled from appropriate
drying agents and deoxygenated prior to use. The [RuCl2(p-
cymene)]2 dimer (1) and tricyclohexylphosphine were purchased
from Strem. Homobimetallic ruthenium-ethylene and tert-butyl-
vinylidene complexes (3 and 5) were prepared according to the
literature.17,22 All the other chemicals were obtained from
Aldrich. A CEM Discover instrument was used for microwave-
assisted syntheses. 1H and 13C{1H} NMR spectra were recorded
at 298 K with a Bruker DRX 400 spectrometer operating at
400.13 and 100.62 MHz, respectively. Chemical shifts are listed
in parts per million downfield from TMS and are referenced
from the solvent peaks or TMS. 31P{1H} NMR spectra were
recorded at 298 K with a Bruker Avance 250 spectrometer
operating at 101.25 MHz using an inlay capillary tube with 85%
H3PO4 as external reference. Infrared spectra were recorded
with a Perkin-Elmer Spectrum One FT-IR spectrometer. Elemental
analyses were carried out in the Laboratory of Pharmaceutical
The catalytic activity of complexes 5-10 was probed in
various types of olefin metathesis reactions. Unsubstituted
phenylvinylidene compound 5 served as a lead structure for
these experiments. Its reaction with norbornene, a typical
strained cycloalkene, was investigated at temperatures ran-
ging from 30 to 60 °C. In all cases, high molecular weight
polymers with a broad polydispersity index and mostly trans
(45) (a) Bennett, M. A.; Smith, A. K. J. Chem. Soc., Dalton Trans.
1974, 233–241. (b) Arthur, T.; Stephenson, T. A. J. Organomet. Chem.
1981, 208, 369–387. (c) Tocher, D. A.; Walkinshaw, M. D. Acta Crystallogr.
1982, B38, 3083–3085. (d) Brost, R. D.; Bruce, G. C.; Grundy, S. L.; Stobart,
ꢁ
Chemistry at the University of Liege. Gas chromatography was
carried out with a Varian 3900 instrument equipped with a flame
ionization detector and a WCOT fused silica column (stationary
phase: CP-Sil 5CB, column length: 15 m, inside diameter: 0.25 mm,
outside diameter: 0.39 mm, film thickness: 0.25 μm). Size-
exclusion chromatography (SEC) was performed in THF at 45 °C
with a SFD S5200 autosampler liquid chromatograph equipped
with a SFD 2000 refractive index detector and a battery of 4 PL
ꢂ
S. R. Inorg. Chem. 1993, 32, 5195–5200. (e) Cubrilo, J.; Hartenbach, I.;
Lissner, F.; Schleid, T.; Niemeyer, M.; Winter, R. F. J. Organomet. Chem.
2007, 692, 1496–1504. (f) Sumiyoshi, T.; Gunnoe, T. B.; Petersen, J. L.;
Boyle, P. D. Inorg. Chim. Acta 2008, 361, 3254–3262. (g) Mutseneck, E. V.;
€
Reus, C.; Schodel, F.; Bolte, M.; Lerner, H.-W.; Wagner, M. Organometallics
2010, 29, 966–975.

Article
Organometallics, Vol. 29, No. 24, 2010 6685
3
gel columns fitted in series (particle size: 5 μm; pore sizes: 105,
104, 103, and 102 A; flow rate: 1 mL/min). The molecular weights
(not corrected) are reported versus monodisperse polystyrene
standards used to calibrate the instrument. Mass spectra were
recorded with a Waters Q-TOF Ultima spectrometer (ESIþ
mode, source temperature: 100 °C, capillary voltage: 3000 V, RF
lens intensity: 100, source pressure: 3.3 mbar).
General Procedure for the Synthesis of Complexes 6-10. A
phenylacetylene derivative (ca. 380 mmol, 2 equiv) was added to
a solution of [(p-cymene)Ru(μ-Cl)3RuCl(η2-C2H4)(PCy3)] (3)
(150 mg, 190 μmol) in dichloromethane (7.5 mL), and the
mixture was stirred for 2 h at room temperature under a static
argon atmosphere. After evaporation of the solvent under
reduced pressure, the residue was washed with n-pentane (3ꢀ
5 mL) and dried overnight under high vacuum.
3JPC = 3 Hz, CH2, PCy3), 27.8 (d, JPC = 3 Hz, CH2, PCy3),
28.9 (d, 2JPC = 24 Hz, CH2, PCy3), 31.3 (CH(CH3)2), 35.3 (d,
1JPC = 24 Hz, CH, PCy3), 78.3, 78.7, 79.5, 80.2 (CHar, p-cym),
97.2, 101.3 (Car, p-cym), 113.5 (RudCdC), 126.8, 128.2, 129.3,
130.0 (C6H4), 353.6 (d, 2JPC = 19.4 Hz, RudC) ppm. 31P NMR
(101 MHz, CD2Cl2): δ 54.92 (s) ppm. IR (KBr): ν 3071(w),
h
2924(s), 2849(m), 1619(s), 1585(m), 1489(m), 1090(m), 833(m),
486(m) cm-1. Anal. Calcd for C36H52Cl5PRu2 (895.18): C, 48.30;
H, 5.86. Found: C, 48.05; H, 5.86.
[(p-cymene)Ru(μ-Cl)3RuCl(dCdCH-C6H4-4-CF3)(PCy3)] (9):
red-brown solid (160 mg, 90% yield) obtained from 4-ethynyl-R,
1
R,R-trifluorotoluene (40 μL, 381 μmol). H NMR (400 MHz,
CD2Cl2): δ 1.12-1.35 (m, 9 H, PCy3), 1.33 (d, 3J = 6.8 Hz, 6 H,
CH(CH3)2), 1.52-2.19 (m, 24 H, PCy3), 2.25 (s, 3 H, CH3,
=
p-cym), 2.93 (sept, 3J = 6.8 Hz, 1 H, CH(CH3)2), 4.91 (d, 4JPH
3
[(p-cymene)Ru(μ-Cl)3RuCl(dCdCH-C6H4-4-OMe)(PCy3)]
(6): brown-beige solid (164 mg, 97% yield) obtained from
4-ethynylanisole (50 μL, 386 μmol). 1H NMR (400 MHz,
3.2 Hz, 1 H, RudCdCH), 5.42 (d, J = 5.5 Hz, 2 H, CHar,
p-cym), 5.58 (d, 3J = 5.5 Hz, 1 H, CHar, p-cym), 5.65 (d, 3J = 5.5
Hz, 1 H, CHar, p-cym), 7.18 (d, 3J = 8.0 Hz, 2 H, C6H4), 7.35 (d,
3J = 8.0 Hz, 2 H, C6H4) ppm. 13C NMR (100 MHz, CD2Cl2): δ
18.6 (CH3, p-cym), 21.9, 22.0 (CH(CH3)2), 26.3 (CH2, PCy3),
27.7 (d, 3JPC = 4 Hz, CH2, PCy3), 27.8 (d, 3JPC = 4 Hz, CH2,
3
CD2Cl2): δ 1.12-1.25 (m, 9 H, PCy3), 1.35 (t, J=6.8 Hz, 6
H, CH(CH3)2), 1.51-1.74 (m, 15 H, PCy3), 1.87 (m, 6 H, PCy3),
2.12 (q, 3J = 11.0 Hz, 3H, PCy3), 2.28 (s, 3 H, CH3, p-cym), 2.95
(sept, 3J = 6.8 Hz, 1 H, CH(CH3)2), 3.73 (s, 3 H, OCH3), 4.84 (d,
2
PCy3), 28.9 (d, JPC = 20 Hz, CH2, PCy3), 31.3 (CH(CH3)2),
35.2 (d, 1JPC = 23 Hz, CH, PCy3), 78.3, 78.8, 79.6, 80.2 (CHar,
p-cym), 97.2, 101.3 (Car, p-cym), 113.6 (RudCdC), 124.7 (q,
3
4JPH = 3.5 Hz, 1 H, RudCdCH), 5.40 (d, J= 5.2 Hz, 2 H,
CHar, p-cym), 5.57 (d, 3J = 5.2 Hz, 1 H, CHar, p-cym), 5.65 (d,
3J = 5.2 Hz, 1 H, CHar, p-cym), 6.72 (d, 3J = 8.8 Hz, 2 H, CHar),
1JCF = 270 Hz, CF3), 125.0 (q, JCF = 4 Hz, C6H4), 125.2
3
3
7.00 (d, J = 8.8 Hz, 2 H, CHar) ppm. 13C NMR (100 MHz,
(C6H4), 125.7 (q, 2JCF = 33 Hz, C-CF3), 136.7 (C6H4), 351.1 (d,
2JPC = 19 Hz, RudC) ppm. 31P NMR (101 MHz, CD2Cl2): δ
CD2Cl2): δ 18.5 (CH3, p-cym), 21.9, 22.0 (CH(CH3)2), 26.4
(CH2, PCy3), 27.7 (d, JPC = 4 Hz, CH2, PCy3), 27.8 (d,
3
54.84 (s) ppm. IR (KBr): ν 3066(w), 2928(s), 2851(m), 1625(m),
h
3JPC = 4 Hz, CH2, PCy3), 28.9 (d, 2JPC = 26 Hz, CH2, PCy3),
31.2 (CH(CH3)2), 35.2 (d, JPC = 24 Hz, CH, PCy3), 55.2
1600(s), 1446(m), 1323(s), 1160(m), 1116(m), 1065(m), 845(m)
cm-1. Anal. Calcd for C37H52Cl4F3PRu2 (928.73): C, 47.85; H,
5.64. Found: C, 47.13; H, 5.58.
[(p-cymene)Ru(μ-Cl)3RuCl{dCdCH-C6H3-3,5-(CF3)2}(PCy3)]
(10): orange solid (158 mg, 83% yield) obtained from 4-ethynyl-
1
(OCH3), 78.2, 78.7, 79.5, 80.2 (CHar, p-cym), 97.2, 101.2 (Car,
p-cym), 113.7 (RudCdC), 113.9, 122.1, 126.8 (C6H4), 157.3 (C-
OCH3), 357.5 (d, 2JPC = 19.4 Hz, RudC) ppm. 31P NMR (101
1
MHz, CD2Cl2): δ 55.02 (s) ppm. IR (KBr): ν 3065(w), 2928(s),
3,5-bis(trifluoromethyl)benzene (67 μL, 381 μmol). H NMR
h
(400 MHz, CD2Cl2): δ 1.05-1.30 (m, 12 H, PCy3), 1.34 (d, 3J =
6.9 Hz, 6 H, CH(CH3)2), 1.50-1.75 (m, 14 H, PCy3), 1.80-1.95
(m, 4 H, PCy3), 2.14 (q, 3J = 11.9 Hz, 3 H, PCy3), 2.26 (s, 3 H,
CH3, p-cym), 2.95 (sept, 3J = 6.9 Hz, 1 H, CH(CH3)2), 5.01 (d,
4JPH = 3.3 Hz, 1 H, RudCdCH), 5.41 (m, 2 H, CHar, p-cym),
5.58 (d, 3J = 5.8 Hz, 1 H, CHar, p-cym), 5.67 (d, 3J = 5.8 Hz,
1 H, CHar, p-cym), 7.38 (s, 1 H, C6H3), 7.58 (s, 2 H, C6H3) ppm.
13C NMR (100 MHz, CD2Cl2): δ 18.4 (CH3, p-cym), 21.9, 22.0
2849(m), 1631(s), 1573(w), 1507(s), 1444(m), 1293(w), 1244(s),
1176(m), 1033(m), 828(m) cm-1. Anal. Calcd for C37H55Cl4O-
PRu2 (890.76): C, 49.89; H, 6.22. Found: C, 49.77; H, 6.22.
[(p-cymene)Ru(μ-Cl)3RuCl(dCdCH-C6H4-4-Me)(PCy3)] (7):
brown-beige solid (153 mg, 92% yield) obtained from 4-ethynyl-
1
toluene (50 μL, 394 μmol). H NMR (400 MHz, CD2Cl2): δ
3
1.13-1.24 (m, 12 H, PCy3), 1.35 (t, J = 6.7 Hz, 6 H, CH-
(CH3)2), 1.50-1.74 (m, 14 H, PCy3), 1.87 (m, 4 H, PCy3), 2.14 (q,
3J = 11.7 Hz, 3 H, PCy3), 2.28 (s, 3 H, CH3, p-cym), 2.30 (s, 3 H,
CH3, p-tolyl), 2.95 (sept, 3J = 6.8 Hz, 1 H, CH(CH3)2), 4.85 (d,
3
(CH(CH3)2), 26.3 (CH2, PCy3), 27.7 (d, JPC = 3 Hz, CH2,
PCy3), 27.8 (d, 3JPC = 3 Hz, CH2, PCy3), 28.9 (d, 2JPC = 35.5
Hz, CH2, PCy3), 31.3 (CH(CH3)2), 35.3 (d, 1JPC = 24.3 Hz, CH,
PCy3), 78.4, 78.7, 79.4, 80.2 (CHar, p-cym), 97.3, 101.4 (Car, p-
cym), 113.2 (RudCdC), 126.8, 128.2, 129.3, 130.0 (C6H3), 348.5
(d, 2JPC = 18.6 Hz, RudC) ppm. 31P NMR (101 MHz, CD2Cl2):
3
4JPH = 3.3 Hz, 1 H, RudCdCH), 5.41 (d, J = 5.5 Hz, 2 H,
CHar, p-cym), 5.58 (d, 3J = 5.5 Hz, 1 H, CHar, p-cym), 5.65 (d,
3J = 5.5 Hz, 1 H, CHar, p-cym), 6.97 (s, 4 H, C6H4) ppm. 13
C
NMR (100 MHz, CD2Cl2): δ 18.5 (CH3, p-cym), 20.8 (CH3, p-
=
tolyl), 21.9, 22.0 (CH(CH3)23), 26.4 (CH2, PCy3), 27.7 (d, 3JPC
δ 55.43 (s) ppm. IR (KBr): ν 3067(w), 2930(m), 2852(m), 1622(m),
h
1598(s), 1447(m), 1376(s), 1277(s), 1176(s), 1131(s), 681(m) cm-1
.
4 Hz, CH2, PCy3), 27.8 (d, JPC = 4 Hz, CH2, PCy3), 29.0 (d,
2JPC = 27 Hz, CH2, PCy3), 31.3 (CH(CH3)2), 35.2 (d, 1JPC = 24
Hz, CH, PCy3), 78.2, 78.7, 79.6, 80.2 (CHar, p-cym), 97.2, 101.2
(Car, p-cym), 114.0 (RudCdC), 125.9, 127.4, 128.9, 134.3 (C6H4),
Anal. Calcd for C38H51Cl4F6PRu2 (996.73): C, 45.79; H, 5.16.
Found: C, 45.48; H, 5.18.
Microwave-Assisted Synthesis of [(p-cymene)Ru(μ-Cl)3RuCl-
(dCdCH-C6H5)(PCy3)] (5). A 10 mL pressure vial equipped
with a magnetic stirring bar and capped with a septum was
charged with [RuCl2(p-cymene)]2 (1) (120 mg, 196 μmol), tri-
cyclohexylphosphine (55 mg, 196 μmol, 1 equiv), phenylacetylene
(33 μL, 300 μmol, 1.5 equiv), and dichloromethane (3 mL). The
reaction mixture was heated for 15 min at 105 °C under stirring
in a microwave reactor with a 150 W maximum power. After
cooling to room temperature, the solvent was removed on a
rotary evaporator. The residue was washed with n-pentane (3 ꢀ
3 mL) and dried under high vacuum. Complex 5 was isolated as
2
356.4 (d, JPC = 19.4 Hz, RudC) ppm. 31P NMR (101 MHz,
CD2Cl2): δ 54.95 (s) ppm. IR (KBr): ν 3068(w), 3046(w), 2922(s),
h
2848(s), 1630(s), 1606(m), 1509(m), 1445(m) 817(m), 490(m)
cm-1. Anal. Calcd for C37H55Cl4PRu2 (874.76): C, 50.80; H,
6.34. Found: C, 50.64; H, 6.29.
[(p-cymene)Ru(μ-Cl)3RuCl(dCdCH-C6H4-4-Cl)(PCy3)] (8):
red-brown solid (139 mg, 82% yield) obtained from 4-ethynyl-
chlorobenzene (52 mg, 381 μmol). 1H NMR (400 MHz, CD2Cl2):
δ 1.13-1.24 (m, 12 H, PCy3), 1.35 (t, 3J = 6.3 Hz, 6 H,
CH(CH3)2), 1.48-1.74 (m, 14 H, PCy3), 1.87 (m, 4 H, PCy3),
2.14 (q, 3J = 11.8 Hz, 3 H, PCy3), 2.28 (s, 3 H, CH3, p-cym), 2.95
(sept, 3J = 6.8 Hz, 1 H, CH(CH3)2), 4.86 (d, 4JPH = 3.0 Hz, 1 H,
RudCdCH), 5.41 (d, 3J = 5.5 Hz, 2 H, CHar, p-cym), 5.58 (d,
3J = 5.5 Hz, 1 H, CHar, p-cym), 5.65 (d, 3J = 5.5 Hz, 1 H, CHar,
p-cym), 7.03 (d, 3J = 8.2 Hz, 2 H, C6H4), 7.10 (d, 3J = 8.2 Hz,
2 H, C6H4) ppm. 13C NMR (100 MHz, CD2Cl2): δ 18.6 (CH3,
p-cym), 21.9, 22.0 (CH(CH3)2), 26.4 (CH2, PCy3), 27.7 (d,
1
an orange-brown solid (156 mg, 93%). H NMR (400 MHz,
CD2Cl2): δ 1.08-1.29 (m, 9 H, PCy3), 1.35 (t, 3J = 6.6 Hz, 6 H,
CH(CH3)2), 1.45-1.79 (m, 15 H, PCy3), 1.89 (m, 6 H, PCy3),
2.14 (q, 3J = 11.6 Hz, 3 H, PCy3), 2.28 (s, 3 H, CH3, p-cym), 2.95
(sept, 3J = 6.8 Hz, 1 H, CH(CH3)2), 4.87 (d, 4JPH = 3.2 Hz, 1 H,
RudCdCH), 5.42 (s, 2 H, CHar, p-cym), 5.59 (d, 3J = 5.6 Hz,
1 H, CHar, p-cym), 5.66 (d, 3J = 5.6 Hz, 1 H, CHar, p-cym), 6.92

6686 Organometallics, Vol. 29, No. 24, 2010
Borguet et al.
(t, 3J = 6.8 Hz, 1 H, p-CH, Ph), 7.07 (d, 3J = 7.6 Hz, 2 H, o-CH,
Ph), 7.15 (t, 3J = 7.2 Hz, 2 H, m-CH, Ph) ppm. 13C NMR (100
MHz, CD2Cl2): δ 18.6 (CH3, p-cym), 21.8, 22.0 (CH(CH3)2),
stirring in a microwave reactor with a 300 W maximum power.
The conversion was monitored by gas chromatography using
the cyclooctane impurity of cyclooctene as internal standard.
The resulting gel was diluted with chloroform (10 mL) and
slowly poured into methanol (250 mL) under vigorous stirring.
The precipitated polyoctenamer was filtered with suction, dried
overnight under dynamic vacuum, and characterized by SEC
and NMR.
Alternatively, a 25 mL round-bottom flask equipped with a
magnetic stirring bar and capped with a three-way stopcock was
charged with a homobimetallic ruthenium complex (7.5 μmol)
and a Lewis acid (7.5 μmol). Air was expelled by applying three
vacuum/argon cycles before dry chlorobenzene (2.5 mL) and
cyclooctene (0.5 mL, 3.75 mmol) were added with dried syringes
under argon. The reaction mixture was stirred for 2 h in an oil
bath at 60 °C. Analyses and workup were carried out as
described above.
Self-Metathesis of Styrene. A homobimetallic ruthenium
complex (4 μmol) and anhydrous aluminum chloride (1.1 mg,
8 μmol) were placed in a 15 mL Schlenk tube containing a
magnetic stirring bar and capped with a septum. Air was
expelled by three vacuum/argon cycles before 2 mL of a styrene
solution (1 M in toluene, 2 mmol) was added with a dried syringe
under argon. The reaction mixture was heated for 2 h in an oil
bath at 85 °C under inert atmosphere. The conversion was
monitored by gas chromatography using n-dodecane as internal
standard.
RCM of Diethyl 2,2-Diallylmalonate. A homobimetallic
ruthenium complex (4 μmol) was placed in a 15 mL Schlenk
tube containing a magnetic stirring bar and capped with a
septum. Air was expelled by three vacuum/argon cycles before
2 mL of a diethyl 2,2-diallylmalonate solution (0.1 M in toluene,
0.2 mmol) possibly containing 6 mol % of phenylacetylene was
added with a dried syringe under argon. The reaction mixture
was stirred in an oil bath at 85 °C. Conversion was monitored by
GC using n-dodecane as internal standard.
3
26.3 (CH2, PCy3), 27.6 (d, JPC = 4 Hz, CH2, PCy3), 27.7 (d,
3JPC = 4 Hz, CH2, PCy3), 28.9 (d, 2JPC = 26 Hz, CH2, PCy3),
31.2 (CH(CH3)2), 35.2 (d, 1JPC = 24 Hz, CH, PCy3), 78.3, 78.8,
79.6, 80.2 (CHar, p-cym), 97.1, 101.1 (Car, p-cym), 114.0 (Rud
=
2
CdC), 124.5, 125.5, 128.1, 131.1 (C6H5), 355.0 (d, JPC
19.4 Hz, RudC) ppm. 31P NMR (101 MHz, CD2Cl2): δ 54.21
(s) ppm. IR (KBr): ν 3073(w), 3053(w), 2929(s), 2849(m),
h
1621(s), 1593(s), 1491(m), 1445(m), 1376(s), 1004(w), 756(m)
cm-1. Anal. Calcd for C36H53Cl4PRu2 (860.73): C, 50.23; H,
6.21. Found: C, 50.23; H, 6.30.
X-ray Crystallographic Studies. Crystals suitable for X-ray
diffraction analysis were obtained by slow diffusion of cyclo-
hexane into a CH2Cl2 solution of complexes 6-10. Crystal data
were collected on a Bruker APPEX II diffractometer using
graphite-monochromated Mo KR radiation (λ = 0.71073 A)
from a fine-focus sealed tube source at 100 K (d = 0.95 A for
complexes 6 and 9). Computing data and reduction was made
with the APPEX II software.46 The structure was solved using
DIRDIF47 and finally refined by full-matrix, least-squares
based on F2 by SHELXL.48 An empirical absorption correction
was applied using SADABS.49 All non-hydrogen atoms were
anisotropically refined, and the hydrogen atom positions were
calculated and refined using a riding model. A highly disordered
distribution of solvent around the inversion center in complex 6
was eliminated with SQUEEZE.50 More details on the crystals,
data collection parameters, and structure refinements are given
in the Supporting Information.
Typical Procedure for the ROMP of Norbornene. A 50 mL
round-bottom flask equipped with a magnetic stirring bar and
capped with a three-way stopcock was charged with a homo-
bimetallic ruthenium complex (15 μmol). Air was expelled by
applying three vacuum/argon cycles before dry chlorobenzene
(10 mL) and a 1.5 M solution of norbornene in chlorobenzene
(2.5 mL, 3.75 mmol) were added with dried syringes under
argon. The reaction mixture was stirred for 2 h in an oil bath
at 60 °C. The conversion was monitored by gas chromatography
using norbornane as an internal standard. The resulting gel was
diluted with chloroform (2 ꢀ 10 mL) and slowly poured into
methanol (500 mL) under vigorous stirring. The precipitated
polynorbornene was filtered with suction, dried overnight under
dynamic vacuum, and characterized by SEC and NMR.
Typical Procedure for the ROMP of Cyclooctene. A 10 mL
pressure vial equipped with a magnetic stirring bar and capped
with a septum was charged with a homobimetallic ruthenium
complex (7.5 μmol). Air was expelled by applying three vacuum/
argon cycles before dry chlorobenzene (2.5 mL) and cyclooctene
(0.5 mL, 3.75 mmol) were added with dried syringes under
argon. The reaction mixture was heated for 2 h at 140 °C under
Reaction of Complex 5 with AlCl3. An NMR tube capped with
septum was charged with [(p-cymene)Ru(μ-Cl)3RuCl-
a
(dCdCH-C6H5)(PCy3)] (5) (28.4 mg, 33 μmol) and CD2Cl2
(0.6 mL) under Ar. A stock solution prepared by dissolving
AlCl3 (36.3 mg, 272 μmol) in CD3CN (0.5 mL) was added in 20
μL portions with a microsyringe, and the reaction was mon-
itored by 1H and 31P NMR spectroscopy at 25 °C. The mixture
obtained upon addition of 80 μL of stock solution (1.3 equiv of
AlCl3) was also analyzed by ESI-MS in toluene.
Acknowledgment. The financial support of the “Fonds
de la Recherche Scientifique-FNRS”, Brussels, is grate-
fully acknowledged for the purchase of major instrumen-
tation. We thank Mr. Dario Bicchielli for his help with the
synthesis and characterization of complex 6, Ms. Itziar
Otazo Aseguinolaza, Ms. Radostina Kalinova, and Mr.
Alexandre Herman for their help with the ROMP cata-
lytic tests, and Prof. Christo Jossifov (Institute of Poly-
mers, Bulgarian Academy of Sciences) for stimulating
discussions.
(46) APPEX II; Bruker AXS Inc.; Madison, WI, USA, 2004.
(47) Beurskens, P. T.; Beurskens, G.; de Gelder, R.; Smits, J. M. M.;
Garcia-Granda, S.; Gould, R. O. DIRDIF-2007; Radboud University:
Nijmegen (The Netherlands), 2007.
(48) Sheldrick, G. M. SHELX-97 (SHELXS 97 and SHELXL 97),
€
€
Programs for Crystal Structure Analyses; University of Gottingen: Gottingen
(Germany), 1998.
(49) Sheldrick, G. M. SADABS, Programs for Scaling and Correction
Supporting Information Available: Crystal data, detailed
comparison of bond lengths and angles, and an X-ray crystal-
lographic file in CIF format for complexes 6-10. This material
is available free of charge via the Internet at http://pubs.acs.org.
€
€
of Area Detection Data; University of Gottingen: Gottingen (Germany),
1996.
(50) van der Sluis, P.; Spek, A. L. Acta Crystallogr. Sect. A: Found.
Crystallogr. 1990, 46, 194–201.