Synthesis and evaluation of new RuCl2(p-cymene)(ER2R′) and (η1:η6-phosphinoarene)RuCl2 complexes as ring-opening metathesis polymerization catalysts

Article abstract of DOI:10.1016/S0022-328X(00)00287-4

New RuCl₂(p-cymene)(ER₂R′) complexes (E = P, As, Sb; R, R′ = H, alkyl, arylalkyl) have been synthesized and used as catalyst precursors for the ring-opening metathesis polymerization (ROMP) of cyclooctene, cyclopentene, and norbornen

Full text of DOI:10.1016/S0022-328X(00)00287-4

www.elsevier.nl/locate/jorganchem
Journal of Organometallic Chemistry 606 (2000) 55–64
Synthesis and evaluation of new RuCl2(p-cymene)(ER2R%) and
(h1:h6-phosphinoarene)RuCl2 complexes as ring-opening metathesis
polymerization catalysts
Dominique Jan, Lionel Delaude, Franc¸ois Simal, Albert Demonceau,
Alfred F. Noels *
Center for Education and Research on Macromolecules (CERM), Institut de Chimie (B6a), Uni6ersite´ de Lie`ge, Sart-Tilman,
B-4000 Lie`ge, Belgium
Received 9 December 1999; received in revised form 1 May 2000
Abstract
New RuCl2(p-cymene)(ER2R%) complexes (E=P, As, Sb; R, R%=H, alkyl, arylalkyl) have been synthesized and used as catalyst
precursors for the ring-opening metathesis polymerization (ROMP) of cyclooctene, cyclopentene, and norbornene. When ER2R%
was a phosphinoarene, the p-cymene ligand could be displaced upon heating and tethered (h1:h6-phosphinoarene)RuCl2
complexes were obtained. Simple thermogravimetric analysis (TGA) of the complexes provided clear-cut indication on their
potential catalytic activity in ROMP. © 2000 Elsevier Science S.A. All rights reserved.
Keywords: Ruthenium; Carbene; Cyclooctene; Ring-opening metathesis polymerization; Cyclopropanation; Homogeneous catalysis
parameters that govern the chemistry of different
metal–carbene complexes, no ‘unified theory’ of metal–
carbene reactivity being available nowadays.
1. Introduction
Chemists can exert a profound influence on the reac-
tivity of organometallic complexes through molecular
engineering, i.e. modification of the ligand environ-
ment. Metal–carbene complexes are no exception to
this rule and transition-metal mediated reactions of
carbene fragments with substrates containing one or
more unsaturations are among the most important
catalytic strategies for constructing new hydrocarbon
frameworks [1–4]. Whereas reactions of metal–carbene
moieties with CꢀC double bonds result in cycloaddition
products with some catalysts (in other words cyclo-
propanation takes place), other carbene complexes in-
duce olefin metathesis. In some cases, cycloaddition and
olefin metathesis occur as competing processes [5,6].
Thus, the reaction of metalꢁcarbene bonds with olefins
can yield different products depending on the nature of
the metal, its oxidation state, and the ancillary ligands
present. We have only a limited understanding of the
The mechanism most commonly accepted for the
metal-catalyzed olefin metathesis reaction involves the
interconversion of metal–carbene–alkene complexes
with metallacyclobutanes as key intermediates. The for-
mation of metallacycles in olefin metathesis clearly
points out the necessity of olefin coordination to the
metal center. No metathesis occurs in the absence of
olefin coordination. Many observations bear out this
hypothesis. In particular, it was shown that some classi-
cal ruthenium- and rhodium-based cyclopropanation
catalysts could also act as olefin metathesis catalysts
simply by promoting coordinative unsaturation at the
metal center [7–9].
Our laboratory has recently reported on the excep-
tional efficacy of RuCl2(arene)(PR3) complexes as cata-
lyst precursors for the ring-opening metathesis
polymerization (ROMP) of low-strain cycloolefins after
reaction with a stoichiometric amount of a diazo com-
pound. This initiator is required to generate well-
defined ruthenium–carbene species in situ [10,11]. It
was shown unambiguously that in solution, the active
* Corresponding author. Tel.: +32-4-3663463; fax: +32-4-
3663497.
E-mail address: af.noels@ulg.ac.be (A.F. Noels).
0022-328X/00/$ - see front matter © 2000 Elsevier Science S.A. All rights reserved.
PII: S0022-328X(00)00287-4

56
D. Jan et al. / Journal of Organometallic Chemistry 606 (2000) 55–64
ligands in RuCl2(arene)(ER2R%) complexes. Hereafter we
also show that simple thermogravimetric analyses of
type 2 precatalysts give a clear-cut indication of their
potential catalytic activity in ROMP.
ruthenium(II)–carbene species retain only one phos-
phane ligand and are no longer bound to the arene
ligand. Moreover, the phosphane ligand has to be quite
bulky and basic to afford high catalytic activities [10].
Practically, only bulky trialkylphosphanes and/or basic
Arduengo-type carbenes [12] can impart sufficient activ-
ity and stability to the active species.
2. Results and discussion
Attempts to improve the catalytic performances of the
archetypical Grubbs’ catalyst (1) have focused mainly on
varying the carbene, the phosphane, or the anionic
ligands. This search for better initiators has led in the last
two years to the discovery inter alia of active cationic
18-electron allenylidene species [13], of cationic Grubbs-
type catalysts with a rigid cis-stereochemistry of the
phosphane ligands [14], and also of cationic carbynehy-
dridoruthenium complexes [15]. Along with these
metal–alkylidene or metal–alkylidyne complexes,
RuCl2(p-cymene)(PCy3) (2a) remains an attractive cata-
lyst precursor because of its ready availability and its air
stability, even when ligated to the basic tricyclo-
hexylphosphane. Complex 2a can be prepared in situ by
addition of the phosphane ligand to the ruthenium(II)
dimer [RuCl2(p-cymene)]2 (3). It promotes the ROMP of
strained olefins [10] and the ring closing metathesis
(RCM) of dienes in a photo-assisted manner simply by
heating a solution of the diene substrate under neon light
[16]. Upon activation with trimethylsilyldiazomethane
(TMSD), it allows the synthesis of poly(norbornene-g-o-
caprolactone) copolymers with an excellent control of
molecular weight distributions [17,18].
Using the ROMP of cyclooctene as a test reaction, we
have investigated the catalytic activity of type 2 com-
plexes where the PCy3 ligand has been replaced by: (i)
the homologous AsCy3 and SbCy3 arsine and stibine; (ii)
various simple PR3 and PR2R% phosphanes; or (iii) new
chelating phosphinoarene ligands (Scheme 1).
2.1. Synthesis and catalytic acti6ity of
RuCl2(p-cymene)(ECy3) complexes (2a–c)
The tricyclohexylarsine and stibine ligands were syn-
thesized by reacting AsCl3 and SbCl3, respectively, with
cyclohexylmagnesium bromide [19,20]. Complexes 2b
and 2c were obtained by addition of a stoichiometric
amount of [RuCl2(p-cymene)]2 (3) to the arsine and
stibine ligands (see Section 4). Table 1 summarizes the
results obtained for the polymerization of cyclooctene
both with the preformed ruthenium(II) complexes and
with the same complexes prepared in situ by addition of
four equivalents of ligand to the ruthenium(II) dimer 3.
This corresponds to a phosphane-to-ruthenium ratio of
2, a value found optimum for the PCy3-based catalytic
system [10].
A comparison with the corresponding PCy3 complexes
indicates the superiority of the phosphine-based com-
plexes over the arsine- and stibine-ones. It also appears
that molecular weight distribution and |cis (which repre-
sents the relative amount of cis double bonds in the
polyoctenamers) vary not only with the different ECy3
ligands, but also with the way of preparing the catalyst
(excess of phosphine). Monomer conversion decreases
in the order PCy3\AsCy3\SbCy3, as do polymer
We now report on various attempts at improving
catalyst efficiency in the ROMP of strained and low-
strain cycloolefins by fine-tuning the arene and the base
Scheme 1.

D. Jan et al. / Journal of Organometallic Chemistry 606 (2000) 55–64
57
Table 1
ROMP of cyclooctene with RuCl2(p-cymene)(ECy3) catalysts 2a–c preformed or generated in situ a
b
Catalyst
T (°C)
Conversion (%)
Mn (kg mol−1
)
Mw/Mn
|
cis
2a
2b
2c
20
20
20
60
60
60
72
39
21
99
41
11
68.2
31.1
5.5
42.3
51.3
3.9
1.67
1.76
2.78
2.00
1.55
1.82
0.59
0.78
0.70
0.26
0.72
0.70
3 + 4PCy3
3 + 4AsCy3
3 + 4SbCy3
a Reaction conditions: 6×10−5 mol of 2a–c with 1×10−4 mol of TMSD or 3×10−5 mol of 3 and 12×10−5 mol of ECy3 with 2×10−4 mol
of TMSD, 1 g of cyclooctene, 5 ml of PhCl, 4 h.
b Fraction of cis double bonds in the polyoctenamer.
molecular weights. Only oligomers are obtained with the
stibine complex. These observations can be rationalized
by invoking a combination of steric and electronic effects,
the arsine and phosphine ligands being both less basic and
less sterically demanding than the corresponding stibine
[21].
2.3. Synthesis of phosphinoarene complexes 6a–c and
7a–c
Having established that the addition of a diazocom-
pound to complexes 2 leads to arene disengagement, we
considered tethering the phosphane and the arene into
2.2. Synthesis and catalytic acti6ity of
RuCl2(p-cymene)(PR2R%) complexes 2d–g
In an exploratory work from our laboratory, numerous
phosphane ligands PR3 were screened for use in conjunc-
tion with 3 as catalysts for the ROMP of cyclooctene [10].
To refine this study, we have tested new PR2R% ligands
whose basicities (8.5BpKaB10.5) and steric bulk
(defined by their cone angle q, see Table 6) matched those
of tricylohexylphosphane, our lead contender so far.
Thus, complexes 2d–g were synthesized (cf. Scheme 1)
andtheircatalyticactivitiesinvestigated. Resultsobtained
for the polymerization of cylooctene are presented in Fig.
1 and Table 2. For comparison’s sake, control experi-
ments were also carried out with complexes 1 and 2a.
Variations in catalyst activities are magnified when the
polymerizations are carried out at room temperature. The
different results obtained with catalyst precursors 2a, 2d,
2e, and 2f at 20°C highlight the fact that very small
variations of the phosphane steric bulk induce large
variations of catalyst efficiencies. Particularly striking is
the difference brought about by replacing tricyclo-
hexylphosphane by tricyclopentylphosphane (2a versus
2f), two ligands of apparently very similar cone angles
and basicities (but spatial conformation may vary). It also
appears that 2d is superior to 2a for the ROMP of
cyclooctene. Its overall relative efficacy is very much alike
to that of 1: same kinetics of polymerization, same
molecular weight distributions, but higher content of cis
double bonds and higher Mn for 2d relative to 1. Yet, the
two catalytic systems show quite different behaviors for
the polymerization of cyclopentene, a cycloolefin that has
seldom been polymerized with ruthenium-based catalysts
[15]. In that case, the superiority of 2a and 2d over 1 is
blatant (Table 3).
Fig. 1. Time course of the cyclooctene polymerization at 20°C using
various RuCl2(p-cymene)(PR2R%) complexes or 1 as catalysts (reac-
tion conditions as in Table 2).
Table 2
ROMP of cyclooctene at 20°C with catalysts 1, 2a, and 2d–g a
b
Catalyst
Conversion (%)
Mn (kg mol−1
)
Mw/Mn
|
cis
1
99
72
99
53
13
0
49.1
68.2
80.3
51.6
26.2
0
1.80
1.67
1.72
1.62
1.77
–
0.26
0.59
0.45
0.64
0.66
–
2a
2d
2e
2f
2g
a Reaction conditions: 6×10−5 mol of ruthenium catalyst, 2×10−4
mol of TMSD, 1 g of cyclooctene, 5 ml of PhCl, 5 h, 20°C.
b Fraction of cis double bonds in the polyoctenamer.

58
D. Jan et al. / Journal of Organometallic Chemistry 606 (2000) 55–64
Table 3
ROMP of cyclopentene at 20°C with catalysts 1, 2a, and 2d a
b
c
c
c
Catalyst
Conversion (%)
Mn (kg mol−1
)
Mw/Mn
|cis
rcis
rtrans
rcis×rtrans
1
2a
2d
8
63
64
3.2
57.8
45.3
1.86
1.66
1.72
–
0.18
0.17
–
0.26
0.09
–
4.94
5.20
–
1.26
0.50
a Reaction conditions: 6×10−5 mol of ruthenium catalyst, 2×10−4 mol of TMSD, 0.7 g of cyclopentene, 3 ml of PhCl, 2 h, 20°C.
b Fraction of cis double bonds in the polypentenamer.
c For a definition of rcis, rtrans, and rcis×rtrans see Ref. [1], pp. 242–243.
new phosphinoarene ligands that could act either as
monodentate h1 or as chelating h1:h6 ligands. Prece-
dents for the synthesis and the chemistry of such three-
legged piano-stool complexes are found in the work of
Ward et al. [22,23].
phorus atom. The efficiency of 6c remains, however,
much lower than that of 2a or 2d (cf. Table 2), al-
though chelation of the pending arm to yield 7c is very
slow under the polymerization conditions, as evidenced
by NMR spectroscopy. Substitution on the remote
arene ring is therefore expected not to have any signifi-
cant influence on the metal center. Indeed, complexes
6a and 6b display identical behaviors and polymerize
cyclooctene at the same very low rate.
Cyclooctene does not polymerize with catalysts of the
series 7. The reaction occurs with norbornene, a
strained cycloolefin more prone to ring-opening, but
conversion remains low (see Table 5). Again, the cata-
lyst bearing ligand 5c displays a higher efficiency than
Ligand synthesis is straightforward and outlined in
Scheme 2, starting from the appropriate halogenated
arene molecules and Cy2P− Li+. The yield decreases
substantially when the secondary bromide 4c is reacted
in place of the primary halogenated derivatives 4a and
4b, because of the increased competition between elimi-
nation and nucleophilic substitution. Addition of a
stoichiometric amount of ruthenium dimer 3 to ligands
5 affords the corresponding complexes 6 in good yield
(see Section 4 and Scheme 1). Heating 6 in chloroben-
zene for several hours results in the quantitative forma-
tion of 7 where the phosphinoarene molecule acts now
as a chelating ligand (Scheme 3). Alternatively, the
one-pot reaction of ligands 5 with dimer 3 at high
temperature also affords complexes 7 in high yields.
Structures of 7b and 7c were ascertained by X-ray
crystallography [24].
Complexes 7, where the phosphinoarene molecules
act as chelating ligands could be seen as ‘dormant
species’, arene disengagement (h1:h6 to h1 ligation)
providing room for the active species to form upon
reaction with the diazocompound and, subsequently,
the arene possibly acting as a two- or four-electron
ligand during the polymerization process (h1 to h1:h2
or h1:h4 ligation). It is also expected, if arene ligation is
controlling the chemistry, that carbene transfer reac-
tions might be favored, resulting in cyclopropane
formation.
Scheme 2.
Scheme 3.
2.4. Catalytic acti6ity of complexes 6a–c and 7a–c
in ROMP and in olefin cyclopropanation
Table 4
ROMP of cyclooctene at 20°C with catalysts 6a–c a
Complexes 6a–c and 7a–c yield poor ROMP cata-
lysts after activation with TMSD or EtDA (ethyl dia-
zoacetate). Some typical results obtained for the
polymerization of cyclooctene in the presence of TMSD
are summarized in Table 4. Among the ‘open arm’
series, compound 6c (R=Me, R%=H) comes out as the
best catalyst precursor, probably because of a slightly
higher basicity and higher steric hindrance at the phos-
b
Catalyst
Conversion (%)
Mn (kg mol−1
)
Mw/Mn
|
cis
6a
6b
6c
5
5
57
0.8
0.8
71.7
–
–
1.64
–
–
0.61
a Reaction conditions: 6×10−5 mol of ruthenium catalyst, 2×10−4
mol of TMSD, 1 g of cyclooctene, 5 ml of PhCl, 5 h, 20°C.
b Fraction of cis double bonds in the polyoctenamer.

D. Jan et al. / Journal of Organometallic Chemistry 606 (2000) 55–64
59
Table 5
ROMP of norbornene at 60°C with catalysts 7a–c a
b
c
d
d
d
Catalyst
Isolated yield (%)
Mn (kg mol−1
)
Mw/Mn
|
cis
rcis
rtrans
rcis×rtrans
7a
7b
7c
19
5
40
12.8
35.8
32.6
6.32
15.47
10.25
0.80
0.74
0.76
5.55
4.00
4.94
1.11
1.20
1.41
6.17
4.80
6.98
a Reaction conditions: 3×10−5 mol of ruthenium catalyst, 1×10−4 mol of TMSD, 1.0 g of norbornene, 30 ml of PhCl, 2 h, 60°C.
b Multimodal distributions.
c Fraction of cis double bonds in the polynorbornene.
d For a definition of rcis, rtrans, and rcis×rtrans see Ref. [1], pp. 242–243.
those based on 5a or 5b. The resulting polynorbornenes
are somewhat blocky (rcis×rtransꢀ1) and have a broad
multimodal molecular weight distribution, indicating
that different active species are operative and/or that
initiation of the polymerization is slow. Indeed, it was
checked that TMSD decomposition was very slow un-
der our reaction conditions. Furthermore, when stoi-
chiometric amounts of TMSD and cyclooctene were
reacted with 7a as catalyst, two new isomeric products
were formed in a 78/22 ratio (80% yield, m/z=196).
Although these compounds were not fully character-
ized, the lack of CꢀC double bond absorption in their
IR spectra and the lack of vinyl proton peaks in their
1H-NMR spectra suggest that they are cyclopropanes
resulting from carbene transfer to the double bond of
cyclooctene.
A more thorough study revealed that with EtDA as
the carbene source, cyclopropanation reactions take
over metathesis. Activated olefins such as styrene
derivatives are cyclopropanated in up to 82% yield
based on EtDA. The scope and limitations of com-
plexes 6 and 7 as cyclopropanation catalysts have been
reported elsewhere [25]. Experimental observations sup-
port the idea that arene de-coordination is crucial for
observing ROMP, the more labile the arene, the more
efficient the catalyst. Arene disengagement requires a
close spatial fit between the phosphane and arene lig-
ands. The role of the diazocompound in promoting
arene removal remains, however, largely speculative so
far.
catalytic efficiency. Experimental data supporting this
assumption are provided in Table 6, which links the
temperature at which the p-cymene ligand is liberated
from complexes 2a–g and the activity of the resulting
active species in the ROMP of cyclooctene at room
temperature. Coupled TGA–MS and TGA–IR analy-
ses unambiguously confirmed that it is indeed the arene
ligand that is disengaged from the metal complexes 2
upon heating.
4. Experimental
4.1. General considerations
All syntheses were carried out under a dry argon
atmosphere using standard Schlenk and glove-box tech-
niques. NMR spectra were recorded on a Brucker AM
1
400 spectrometer. H and 13C chemical shifts are listed
in parts per million downfield from TMS and are
referenced by the solvent peaks (7.25 and 77.0 in CDCl3
1
for H and 13C spectra, respectively). 31P data are listed
in parts per million downfield from 85% H3PO4 and are
externally referenced. Infrared spectra were recorded on
a Perkin–Elmer 1720X series FT-IR spectrometer with
a selected resolution of 2 cm−1. Gel permeation chro-
matographic (GPC) analyses of the polymers were per-
formed in THF on a Hewlett–Packard HP 1090
Table 6
Influence of the phosphane cone angle and of the arene lability on the
catalytic activity of complexes 2
q (°) a
TD (°C) b
Cyclooctene conversion (%) c
3. Predictive value of TGA measurements
Catalyst
2a
2b
2c
2d
2e
2f
170
166
161
174
160
n.a.
145
162
212
219
139
172
165
222
72
39
21
99
53
13
0
The existence of a relationship between the p-cymene
release from a ruthenium–phosphane complex and the
catalyst activity was proposed by Hafner et al. to
rationalize differential scanning calorimetry (DSC)
measurements carried out on RuCl2(p-cymene)(PR3)
complexes used as photoinitiators in ROMP [26]. It was
confirmed and substantiated by thermogravimetric
analysis (TGA) of catalyst precursors in our group. An
easy liberation of the arene ligand (corresponding to a
low TD value in DSC or TGA) is indicative of a good
2g
a Cone angle of the phosphane ligand.
b Temperature at which the p-cymene ligand is liberated as deter-
mined by TGA.
c Reaction at 20°C (same conditions as in Table 2).

60
D. Jan et al. / Journal of Organometallic Chemistry 606 (2000) 55–64
instrument equipped with a HP 1037A refractive index
detector and a battery of four PL gel columns fitted in
series (particle size: 5 mm; pore sizes: 100 000, 10 000,
9.7 Hz), 30.33 (d, C(CH3)3, 2JCꢁP=13.2 Hz), 27.69,
27.61 (2 s, C3 C6H11), 26.41 (s, C4, C6H11)%. 31P-NMR:
28.58.
,
1000, and 100 A). The molecular weights (not cor-
rected) are reported versus monodisperse polystyrene
standards used to calibrate the instrument. The GPC
values are internally consistent but are not necessary
directly comparable to values obtained in different sol-
vents. The polymer microstructures were determined by
comparison of their 1H- and 13C-NMR spectra with
those reported in the literature. Results are accurate
within 2% when the integrations of the vinyl and allyl
protons and of all carbon atoms are averaged. For
analogous complexes, elemental analyses were per-
formed only on select representative materials.
4.3.2. 1-Chloro-3-(3,5-dimethyl)phenylpropane (4b)
To a solution of 1,3,5-trimethylbenzene (6.30 g, 52.4
mmol) in THF (20 ml) cooled at −78°C were added 22
ml of a n-butyllithium solution (2.5 M in hexanes, 55
mmol). The solution was stirred overnight at r.t. and
was then added dropwise at 0°C to 49 g of 1,2-
dichloroethane (505 mmol). The mixture was stirred for
2 h at r.t. and the fine white precipitate was removed by
filtration through Celite. Evaporation of the volatile
fraction and distillation under reduced pressure af-
forded the title product as a colorless liquid. Yield 3.10
g (33%); b.p. 70°C (0.07 mm Hg). GC–MS: m/z (%)
1
4.2. Materials
184 (13), 182 (35) [M+], 119 (100), 105 (38). H-NMR
(CDCl3, l ppm): 6.91 (s, 1H, CHpara), 6.88 (s, 2H,
3
Solvents and monomers were distilled from appropri-
ate drying agents and deoxygenated prior to use. 1-
CHortho), 3.58 (t, 2H, CH2Cl, JHꢁH=6.8 Hz), 2.76 (t,
3
2H, CH2Ar, JHꢁH=7.2 Hz), 2.36 (s, 6H, CH3Ar), 2.12
3
Bromo-3-phenylpropane,
1,2-dichloroethane,
and
(pseudo-quint, 2H, CH2CH2CH2, JHꢁH=7 Hz). 13C-
1,3,5-trimethylbenzene were dried over calcium chloride
and distilled before use. Cyclohexyl bromide, AsCl3,
SbCl3, PBr3, RuCl3·xH2O, trimethylsilyldiazomethane
(2.0 M in hexanes), 4-phenylbutan-2-one, n-butyl-
NMR: 140.57 (Cipso Ar), 137.87 (CCH3), 127.68 (Cpara
Ar), 126.32 (Cortho Ar), 44.29 (CH2Cl), 34.08
(CH2CH2Cl), 32.61 (CH2Ar), 21.14 (CH3). IR (cm−1):
3014 (m), 2952 (s), 2919 (s), 2860 (m), 1607 (s), 1443
(m), 837 (m), 703 (m).
lithium (2.5
M in hexanes), PCy3, Cy2PH, and
[RuCl2(p-cymene)]2 were purchased from commercial
suppliers and used without further purification. Tricy-
clopentylphosphine (PCp3, 50% wt solution in toluene)
was a generous loan from CYTEC Canada Inc. AsCy3
[19], SbCy3 [20], RuCl2(p-cymene)(PCy3) (2a), and
RuCl2(p-cymene)(PiPr3) (2e) [10] were synthesized ac-
cording to published procedures. 4-Phenylbutan-2-ol
was obtained by reduction of 4-phenylbutan-2-one with
sodium borohydride.
4.3.3. 2-Bromo-4-phenylbutane (4c)
Phosphorus tribromide (3.25 g, 12 mmol) was slowly
added at 0°C to 4.95 g of neat 4-phenylbutan-2-ol (33
mmol). The resulting yellow solution was stirred 2 h at
0°C and overnight at r.t. The reaction mixture was
carefully hydrolyzed with 15 ml of water and extracted
twice with 30 ml of diethyl ether. The ethereal phase
was washed with a saturated Na2CO3 solution, dried
over Na2SO4, and the solvent was evaporated. Distilla-
tion under reduced pressure afforded the pure product
as a colorless liquid. Yield 4.57 g (65%); b.p. 64°C (0.08
mm Hg). GC–MS: m/z (%) 214 (26), 212 (27) [M+],
4.3. Synthesis of phosphine and phosphinoarene ligands
4.3.1. tert-Butyldicyclohexylphosphine
1
To a solution of chlorodicyclohexylphosphine (5.75
g, 24.7 mmol) in THF (20 ml) cooled at −78°C were
added dropwise 18 ml of tert-butyllithium (1.5 M in
pentane, 27.2 mmol). The yellow suspension was al-
lowed to warm to room temperature (r.t.) and was
stirred overnight at this temperature. The reaction mix-
ture was evaporated to dryness and the phosphine was
extracted twice with 20 ml of pentane. The pentane
solution was filtered through Celite, concentrated to 20
ml, and cooled to −78°C. After 2 h, the white crystals
obtained were filtered under inert atmosphere and
washed with small fractions of cold pentane. Cy2PtBu is
highly air-sensitive and melts at r.t. Yield 5.60 g (89%).
1H-NMR (CDCl3, l ppm): 1.90–1.05 (m, 22H, C6H11),
1.10 (d, 9H, C(CH3)3, 3JHꢁP=10.8 Hz). 13C-NMR:
117 (53), 91 (100). H-NMR (CDCl3, l ppm): 7.35–
1
7.22 (m, 5H, Ph), 4.11 (m, H, CHBr), 2.83 (m, 2H,
CH2Ph), 2.15 (m, 2H, CH2CH), 1.75 (d, 3H, CH3,
3JHꢁH=10.7 Hz). 13C-NMR: 140.87, 128.47, 128.44,
126.05 (C6H5), 50.84 (CHBr), 42.64 (CH2CH), 33.92
(CH2Ph), 26.47 (CH3). IR (cm−1): 3063 (m), 3027 (s),
2980 (m), 2922(s), 2861 (m), 1603 (m), 1495 (s), 1455
(s), 1209 (m), 700 (s).
4.3.4. Dicyclohexyl(3-phenylpropyl)phosphine (5a)
To a solution of dicyclohexylphosphine (1.79 g, 9
mmol) in THF (25 ml) cooled at −78°C were added
3.6 ml of a n-butyllithium solution (2.5 M in hexanes, 9
mmol). The resulting deep yellow suspension was
stirred for 2 h at r.t. 1-Bromo-3-phenylpropane (2.19 g,
11 mmol) was added at −78°C and the mixture was
stirred 18 h at r.t. After evaporation of the solvent, the
1
33.11 (d, CMe3, JCꢁP=19.4 Hz), 33.61 (d, C1 C6H11,
2
1JCꢁP=16.2 Hz), 30.96, 27.82 (2 d, C2 C6H11, JCꢁP
=

D. Jan et al. / Journal of Organometallic Chemistry 606 (2000) 55–64
61
phosphine was extracted with pentane and the resulting
solution was filtered through Celite. Evaporation of the
solvent and recrystallization from cold (−78°C) diethyl
ether afforded the product as white crystals which gave
an air-sensitive, oily liquid at ambient temperature.
30H, C6H11), 2.10 (s, 3H, CH3 p-cym), 1.26 (d, 6H,
CHCH3, 3JHꢁH=7.2 Hz). 13C-NMR: 106.75 (CꢁMe
p-cym), 93.65 (CꢁCHMe2 p-cym), 85.85, 81.21 (CH
p-cym), 36.78 (C1 C6H11), 30.56 (CHMe2 and C2
C6H11), 27.93 (C3 C6H11), 26.54 (C4 C6H11), 22.35
(CHCH3), 18.04 (ArCH3).
1
Yield 2.24 g (78%). H-NMR (CDCl3, l ppm): 7.25–
3
7.13 (m, 5H, Ph), 2.65 (t, 2H, CH2Ph, JHꢁH=7.6 Hz),
1.76–1.09 (m, 26H, C6H11 and CH2CH2P). 13C-NMR:
144.22, 128.44, 128.21, 125.67 (C6H5), 37.58 (d, CH2Ph,
4.4.2. RuCl2(p-cymene)(SbCy3) (2c)
To a solution of [RuCl2(p-cymene)]2 (0.26 g, 0.42
mmol) in dichloromethane (15 ml) were added 0.91 g
(2.45 mmol) of SbCy3 and the mixture was stirred 14 h
at r.t. The solvent was then evaporated to dryness and
the orange–red residue was washed with pentane (2×
15 ml) and with diethyl ether (8 ml). Yield 0.38 g (67%);
1
3JCꢁP=12.7 Hz), 33.28 (d, C1 C6H11, JCꢁP=11.4 Hz),
2
30.41, 28.99 (both d, C2 C6H11, JCꢁP=8.1 Hz), 30.15
1
(d, CH2P, JCꢁP=19.5 Hz), 27.43, 27.26 (both s, C3
C6H11), 27.32, 26.53 (2 s, C4, C6H11), 20.86 (d,
2
CH2CH2CH2, JCꢁP=16.5 Hz). 31P-NMR: −4.09.
1
m.p. 179°C (dec.). H-NMR (CD2Cl2, l ppm): 5.54 (m,
4.3.5. Dicyclohexyl(3-(3,5-dimethyl)phenylpropyl)-
phosphine (5b)
4H, CHarom p-cym), 2.82 (sept, 1H, CHMe2, 3JHꢁH=7.2
Hz), 2.47 (pseudo t, 3H, CH C6H11), 2.15–1.15 (m,
30H, C6H11), 2.08 (s, 3H, CH3 p-cym), 1.26 (d, 6H,
CHCH3, 3JHꢁH=7.3 Hz). 13C-NMR: 105.86 (C-Me
p-cym), 93.40 (C-CHMe2 p-cym), 84.74, 81.29 (CH
p-cym), 32.21 (C1 C6H11), 31.41 (C2 C6H11), 31.03
(CHMe2), 29.07 (C3 C6H11), 27.02 (C4 C6H11), 22.40
(CHCH3), 18.57 (ArCH3).
The procedure given above for 5a was followed using
2.65 g (13.3 mmol) of dicyclohexylphosphine, 5.4 ml
(13.5 mmol) of n-butyllithium solution and 2.77 g (13.3
mmol) of 1-chloro-3-(3,5-dimethyl)phenylpropane (4b).
1
Yield 3.40 g (74%). H-NMR (CDCl3, l ppm): 6.81 (s,
3
3H, CH arom), 2.62 (t, 2H, CH2Ar, JHꢁH=7.6 Hz),
2.28 (s, 6H, CH3), 1.75–1.20 (m, 26H, C6H11 and
CH2CH2P). 13C-NMR: 142.14, 137.66, 127.31, 126.31
3
(C6H3), 37.37 (d, CH2Ar, JCꢁP=11.4 Hz), 33.30 (d, C1
4.4.3. RuCl2(p-cymene)(Cy2PtBu) (2d)
C6H11, 1JCꢁP=11.4 Hz), 30.35, 29.02 (both d, C2 C6H11,
2JCꢁP=7.2 Hz), 30.11 (d, CH2P, 1JCꢁP=17.9 Hz),
27.45, 27.27 (both s, C3 C6H11), 27.34, 26.54 (both s, C4,
C6H11), 21.24 (ArCH3), 20.86 (d, CH2CH2CH2,
2JCꢁP=16.3 Hz). 31P-NMR: −4.02.
To 0.47 g (1.84 mmol) of tert-butyldicyclohexylphos-
phine in dichloromethane (20 ml) were added 0.47 g of
[RuCl2(p-cymene)]2. The red–brown solution was
stirred 1 h at r.t. and was then evaporated to dryness.
The crude product was washed with pentane (2×20
ml) and with diethyl ether (2×8 ml). The complex was
obtained as a red microcrystalline powder. Yield 0.70 g
4.3.6. Dicyclohexyl(2-(4-phenyl)butyl)phosphine (5c)
The procedure given above for 5a was followed using
2.88 g (14.5 mmol) of dicyclohexylphosphine, 6.1 ml
(15.2 mmol) of n-butyllithium solution and 3.09 g (14.5
mmol) of 2-bromo-4-phenylbutane (4c). Yield 2.73 g
1
(81%); m.p. 108°C (dec.). H-NMR (CDCl3, l ppm):
1
5.57 (s, 4H, CHarom p-cym), 2.87 (sept, H, CHMe2,
3JHꢁH=7.2 Hz), 2.43 (m, 2H, CH C6H11), 2.25–1.11
(m, 26H, C6H11 and CH2CH2CH2P), 2.11 (s, 3H, CH3
p-cym), 1.41 (d, 9H, C(CH3)3, 3JHꢁP=12.4 Hz), 1.30 (d,
6H, CH(CH3)2, 3JHꢁH=7.2 Hz). 13C-NMR: 108.43
(CꢁMe p-cym), 107.46, 94.24, 89.20, 83.66, 81.29, 80.53
1
(57%). H-NMR (CDCl3, l ppm): 7.30–7.10 (m, 5H,
Ph), 2.70 (m, PCHCH3 and CH2Ph, 3H), 2.14–1.13 (m,
C6H11 and PCH(CH3)CH2, 27H). 13C-NMR: 141.09,
128.49, 128.32, 126.14 (C6H5), 37.40–23.28 (m, C6H11
and PCHCH2CH2), 17.26 (d, PCHCH3, 2JCꢁP=8.2
Hz). 31P-NMR: 11.64.
1
(C p-cym), 38.81 (d, C1 C6H11, JCꢁP=24.7 Hz), 38.27
(d, CMe3, 1JCꢁP=11.5 Hz), 30.49 (CHMe2), 32.14,
2
30.03 (both s, C2 C6H11), 29.78 (d, C(CH3)3, JCꢁP=3.3
3
Hz), 28.55, 26.43 (both d, C3 C6H11, JCꢁP=8.6 Hz),
4.4. Synthesis of RuCl2(p-cymene)(ER2R%) complexes
26.63 (s, C4 C6H11), 22.24 (CHMe2), 22.63 (Ar(CH3)2),
17.71 (d, CH2CH2P, 2JCꢁP=22.8 Hz), 18.02 (ArCH3
4.4.1. RuCl2(p-cymene)(AsCy3) (2b)
p-cym).
31P-NMR:
35.68.
Anal.
Calc.
for
To a solution of [RuCl2(p-cymene)]2 (0.10 g, 0.16
mmol) in dichloromethane (8 ml) were added 0.32 g
(0.98 mmol) of AsCy3 and the mixture was stirred 14 h
at r.t. The solvent was then evaporated to dryness and
the orange–red residue was washed with pentane (2×
15 ml) and with diethyl ether (8 ml). Yield 0.17 g (82%);
C26H45Cl2PRu: C, 55.71; H, 8.09. Found: C, 55.17; H,
8.99%.
4.4.4. RuCl2(p-cymene)(P(C5H9)3) (2f)
A solution of tricyclopentylphosphine (50% wt in
toluene, 0.90 g, 1.88 mmol) was added to 0.51 g of
[RuCl2(p-cymene)]2 (0.83 mmol) in 10 ml of
dichloromethane. The reaction mixture was stirred 1 h
at r.t. and evaporated to dryness. The crude product
1
m.p. 165°C (dec.). H-NMR (CD2Cl2, l ppm): 5.53 (m,
4H, CHarom p-cym), 2.88 (sept, 1H, CHMe2, 3JHꢁH=7.2
Hz), 2.51 (pseudo t, 3H, CH C6H11), 2.15–1.15 (m,

62
D. Jan et al. / Journal of Organometallic Chemistry 606 (2000) 55–64
2
was washed several times with pentane and diethyl
ether. The complex was obtained as an orange micro-
crystalline powder. Yield 0.73 g (82%); m.p. 149°C
C6H11, JCꢁP=3.2 Hz), 28.58, 26.43 (both s, C3 C6H11),
1
27.37 (d, CH2P, JCꢁP=22.8 Hz), 27.09, 27.01 (both s,
C4 C6H11), 22.29 (CHMe2), 18.54 (d, CH2CH2P,
2JCꢁP=24.4 Hz), 18.02 (ArCH3 p-cym). 31P-NMR:
24.96.
1
(dec.). H-NMR (CDCl3, l ppm): 5.55 (s, 4H, CHarom
1
3
p-cym), 2.78 (sept, H, CHMe2, JHꢁH=6.8 Hz), 2.66
(m, 3H, CHP), 2.05 (s, 3H, ArCH3), 1.98–1.50 (m,
24H, C5H9), 1.25 (d, 6H, CHCH3, 3JHꢁH=6.8 Hz).
13C-NMR: 106.23 (C-Me p-cym), 93.63 (CꢁCHMe2
p-cym), 89.79, 84.01 (both d, CH p-cym, J=3.3 Hz),
4.4.7. RuCl2(p-cymene)(Cy2P(CH2)3C6H3Me2) (6b)
This complex was prepared in the same way as 6a by
using 0.42 g (0.68 mmol) of [RuCl2(p-cymene)]2 and
0.55 g (1.60 mmol) of phosphine 5b. Yield 0.69 g (78%);
1
37.22 (d, C1 C5H9, JCꢁP=22.8 Hz), 30.46 (CHMe2),
2
1
29.78 (C2 C5H9), 25.63 (d, C3 C5H9, JCꢁP=8.1 Hz),
m.p. 161°C (dec.). H-NMR (CDCl3, l ppm): 6.78 (s,
22.47 (CHMe2), 17.78 (ArCH3). 31P-NMR: 51.78.
1H, CHpara Ar), 6.74 (s, 2H, CHortho Ar), 5.51 (s, 4H,
CHarom p-cym), 2.82 (sept, 1H, CHMe2, 3JHꢁH=6.8
Hz), 2.48 (m, 2H, CH C6H11), 2.23–1.18 (m, 26H,
4.4.5. RuCl2(p-cymene)(Cy2PH) (2g)
A solution of dicyclohexylphosphine (0.31 g, 1.44
mmol) in 5 ml of dichloromethane was added via a
cannula to 0.40 g of [RuCl2(p-cymene)]2 (0.65 mmol) in
12 ml of dichloromethane. The resulting red–brown
solution was stirred 1 h at r.t. The volatiles were
evaporated under vacuum and the crude product was
washed several times with pentane and diethyl ether to
afford a deep orange powder. Yield 0.61 g (93%); m.p.
C6H11 and CH2CH2CH2P), 2.26 (s, 6H, ArCH3), 2.06
3
(s, 3H, CH3 p-cym), 1.24 (d, 6H, CH(CH3)2, JHꢁH
=
6.8 Hz). 13C-NMR: 141.79, 137.69, 127.41, 126.26
(C6H5), 108.43 (CꢁMe p-cym), 93.76 (CꢁCHMe2 p-
cym), 88.4 (d, CH p-cym, J=3.2 Hz), 82.96 (d, CH
3
p-cym, J=4.9 Hz), 37.62 (d, CH2Ar, JCꢁP=11.1 Hz),
1
37.26 (d, C1 C6H11, JCꢁP=21.1 Hz), 30.64 (CHMe2),
2
29.18, 27.37 (both d, C2 C6H11, JCꢁP=3.2 Hz), 28.55,
1
1
222°C (dec.). H-NMR (CDCl3, l ppm): 5.49 (m, 4H,
26.43 (both s, C3 C6H11), 27.35 (d, CH2P, JCꢁP=22.8
CHarom p-cym), 4.34 (dt, 1H, PH, 1JHꢁP=366.9 Hz,
3JHꢁH=3.3 Hz), 2.82 (sept, 1H, CHMe2, 3JHꢁH=6.8
Hz), 2.30 (m, 2H, CH C6H11), 2.15–1.15 (m, 20H,
C6H11), 2.09 (s, 3H, CH3 p-cym), 1.21 (d, 6H, CHCH3,
3JHꢁH=6.8 Hz). 13C-NMR: 107.15 (C-Me p-cym),
95.86 (CꢁCHMe2 p-cym), 87.45, 87.42, 84.12, 84.09
Hz), 27.08, 27.00 (both s, C4 C6H11), 22.24 (CHMe2),
21.19 (Ar(CH3)2), 18.65 (d, CH2CH2P, 2JCꢁP=22.8
Hz), 18.02 (ArCH3 p-cym); 31P-NMR: 24.88. Anal.
Calc. for C33H51Cl2PRu: C, 60.91; H, 7.90. Found: C,
61.40; H, 9.13%.
1
(CH p-cym), 34.28 (C1 C6H11, JCꢁP=22.8 Hz), 32.50,
4.4.8. RuCl2(p-cymene)(Cy2PCH(Me)(CH2)2Ph) (6c)
This complex was prepared in the same way as 6a by
using 0.37 g (0.60 mmol) of [RuCl2(p-cymene)]2 and
0.50 g (1.50 mmol) of phosphine 5c. Yield 0.50 g (66%);
30.50 (both d, C2 C6H11, 2JCꢁP=3.2 Hz), 30.48
(CHMe2), 27.19, 26.98 (both s, C3 C6H11), 27.08, 25.70
(both s, C4 C6H11), 22.05 (CH(CH3)2), 18.05 (ArCH3).
31P-NMR: 40.91. Anal. Calc. for C22H37Cl2PRu: C,
51.55; H, 8.84. Found: C, 52.04; H, 8.55%.
1
m.p. 96°C (dec.). H-NMR (CDCl3, l ppm): 7.21–7.17
(m, 5H, CH Ph), 5.53–5.49 (m, 4H, CHarom p-cym),
1
3
2.80 (sept, H, CHMe2, JHꢁH=6.8 Hz), 2.75–1.24 (m,
28H, C6H11 and CH2CH(CH3)P), 2.03 (s, 3H, CH3
4.4.6. RuCl2(p-cymene)(Cy2P(CH2)3Ph) (6a)
3
A solution of phosphine 5a (0.50 g, 1.57 mmol) in
dichloromethane (5 ml) was added to 0.42 g of
[RuCl2(p-cymene)]2 (0.68 mmol) in 10 ml of
dichloromethane. The reaction mixture was stirred for 1
h and then evaporated to dryness. The crude product
was washed several times with pentane and diethyl
ether. The complex was obtained as an orange micro-
crystalline powder. Yield 0.72 g (84%); m.p. 152°C
p-cym), 1.24 (d, 6H, CH(CH3)2, JHꢁH=6.8 Hz). 13C-
NMR: 141.82, 128.64, 128.33, 125.87 (C6H5), 107.04
(CꢁMe p-cym), 94.57 (CꢁCHMe2 p-cym), 88.94, 87.91
(both d, CH p-cym, J=3.2 Hz), 84.37, 83.62 (both d,
CH p-cym, J=4.8 Hz), 36.35, 36.18 (both d, C1 C6H11,
1JCꢁP=13.1 Hz), 35.19 (s, CH2Ph), 34.55 (d, PCHMe,
1JCꢁP=10.1 Hz), 30.57 (CHMe2), 29.72, 27.67 (both d,
C2 C6H11, 3JCꢁP=5.0 Hz), 29.57, 26.47 (both s, C4
C6H11), 29.23, 29.04 (both s, C3 C6H11), 22.74
(PCHCH3), 22.26 (CH(CH3)2), 17.89 (ArCH3 p-cym),
16.40 (CH2CH2CH). 31P-NMR: 30.30.
1
(dec.). H-NMR (CDCl3, l ppm): 7.24–7.21 (m, 2H,
CHortho Ph), 7.15–7.11 (m, 3H, CHmeta+para Ph), 5.51
(s, 4H, CHarom p-cym), 2.80 (sept, 1H, CHMe2, 3JHꢁH
=
6.8 Hz), 2.55 (m, 2H, C1 C6H11), 2.21–1.16 (m, 26H,
C6H11 and CH2CH2CH2P), 2.05 (s, 3H, CH3 p-cym),
1.24 (d, 6H, CH(CH3)2, 3JHꢁH=6.8 Hz). 13C-NMR:
141.91, 128.47, 128.28, 125.79 (C6H5), 108.48 (CꢁMe
p-cym), 93.82 (CꢁCHMe2 p-cym), 88.45 (d, CH p-cym,
J=3.2 Hz), 82.99 (d, CH p-cym, J=4.9 Hz), 37.75 (d,
4.5. Synthesis of tethered phosphinoarene–ruthenium
complexes
4.5.1. RuCl2(p1:p6-Cy2P(CH2)3Ph) (7a)
A solution of complex 6a (0.36 g, 0.58 mmol) in 20
ml of chlorobenzene was heated overnight at 120°C.
The volatiles were removed under vacuum and the
3
1
CH2Ph, JCꢁP=11.1 Hz), 37.38 (d, C1 C6H11, JCꢁP
=
21.1 Hz), 30.67 (CHMe2), 29.21, 27.39 (both d, C2

D. Jan et al. / Journal of Organometallic Chemistry 606 (2000) 55–64
63
orange residue was washed several times with pentane
4.6. Typical procedure for the ROMP of cyclooctene
and diethyl ether. Yield 0.23 g (82%); m.p. 252°C (dec.).
1H-NMR (CDCl3, l ppm): 6.25 (t, 1H, CHpara Ph,
3JHꢁH=6.0 Hz), 5.66 (t, 2H, CHmeta Ph, 3JHꢁH=6.0
To a ruthenium complex (6×10−5 mol) placed in a
flask under argon were added 3 ml of chlorobenzene
and 1.0 g of cyclooctene (9.1 mmol). The mixture was
stirred for 5 min and 2 ml of trimethylsilyldia-
zomethane (TMSD, 0.1 M in chlorobenzene, 0.2 mmol)
were added via a syringe. The conversion was followed
by gas chromatography using the cyclooctane impurity
of cyclooctene as an internal standard. The solution
was kept at 20°C for 5 h, then diluted in CHCl3 before
precipitation in a large volume of methanol acidified
with HF (600 ml). The resulting polymer was dried
overnight under vacuum and analyzed by GPC and
NMR spectroscopy.
3
Hz), 5.08 (d, 2H, CHortho Ph, JHꢁH=6.0 Hz), 2.52 (m,
3
2H, CH C6H11), 2.39 (t, 2H, CH2Ph, JHꢁH=4.8 Hz),
2.36–1.15 (m, 24H, C6H11 and CH2CH2P). 13C-NMR:
97.17, 97.06, 95.96, 93.27, 93.24, 80.09 (arene), 33.05 (d,
C1 C6H11, 1JCꢁP=24.3 Hz), 29.89, 25.00 (both s,
CH2Ph), 29.04 (s, C2 C6H11), 27.68 (d, C3 C6H11,
1
3JCꢁP=3.3 Hz), 27.46, 26.77 (both d, CH2P, JCꢁP
=
10.5 Hz), 26.08 (s, C4 C6H11), 15.43, 15.19 (both s,
CH2CH2CH2P). 31P-NMR: 29.57.
4.5.2. RuCl2(p1:p6-Cy2P(CH2)3C6H3Me2) (7b)
4.7. Typical procedure for the ROMP of norbornene
Heating 6b (0.40 g, 0.62 mmol) in chlorobenzene as
described above for 7a afforded the title complex as an
orange solid. Yield 0.30 g (95%); m.p. 272°C (dec.).
To a ruthenium complex (3×10−5 mol) placed in a
flask under argon were added 25 ml of chlorobenzene
and 1.0 g of norbornene (10.6 mmol) dissolved in 4 ml
of chlorobenzene. The flask was heated to 60°C over 5
min and 1 ml of trimethylsilyldiazomethane (TMSD,
0.1 M in chlorobenzene, 0.10 mmol) was added via a
syringe. The solution was kept at 60°C for 2 h, cooled
to r.t. and diluted in CHCl3 before precipitation in a
large volume of methanol acidified with HF (600 ml).
The resulting polymer was dried overnight under vac-
uum and analyzed by GPC and NMR spectroscopy.
1
1H-NMR (CDCl3, l ppm): 5.63 (s, H, CHpara Ar), 4.60
(s, 2H, CHortho Ar), 2.43 (m, 2H, CH C6H11), 2.29 (t,
2H, CH2Ph, 3JHꢁH=5.6 Hz), 2.09 (s, 3H, CH3Ar),
2.00–1.10 (m, 24H, C6H11 and CH2CH2P). 13C-NMR:
107.23, 107.20, 96.39, 95.99, 95.86, 77.47 (arene), 33.69
1
(d, C1 C6H11, JCꢁP=24.3 Hz), 30.31, 24.41 (both s,
CH2Ar), 29.18 (s, C2 C6H11), 27.72 (d, C3 C6H11,
3JCꢁP=3.3 Hz), 27.56, 27.01 (both d, CH2P, JCꢁP
=
1
10.5 Hz), 26.27 (s, C4 C6H11), 18.23 (ArCH3), 16.32,
16.09 (both s, CH2CH2CH2P). 31P-NMR: 28.61. Anal.
Calc. for C22H35Cl2PRu: C, 53.49; H, 7.22. Found: C,
53.56; H, 8.13%.
Acknowledgements
D.J. thanks the ‘Services Fe´de´raux des Affaires Sci-
entifiques, Techniques et Culturelles’ for financial sup-
port in the frame of the PAI 4/11-Supramolecular
Chemistry and Supramolecular Catalysis. Support of
the ‘Fonds National de la Recherche Scientifique’
(FNRS) for the purchase of major instrumentation is
gratefully acknowledged. TGA-MS and TGA-FTIR
analyses were kindly performed by M. Alexandre (Uni-
versity of Mons-Hainaut).
4.5.3. RuCl2(p1:p6-Cy2PCH(Me)(CH2)2Ph) (7c)
Heating 6c (0.77 g, 1.22 mmol) in chlorobenzene as
described above for 7a afforded the title complex as an
orange solid. Yield 0.38 g (63%); m.p. 242°C (dec.).
1H-NMR (CDCl3, l ppm): 6.21 (m, 2H, CHortho Ph),
1
1
5.28 (m, H, CHmeta Ph), 5.23 (m, H, CHmeta Ph), 4.83
(m, 1H, CHpara Ph), 2.69–0.90 (m, 30H, C6H11
and CH2CH2CH(CH3)P). 13C-NMR: 102.83, 102.75
(Cipso Ph), 97.95, 97.84, 91.64, 85.62, 81.88, 76.01
(CHortho+meta+paraPh), 36.02–13.34 (m, not completely
assigned because of the multiplicity of the signals).
31P-NMR: 36.44.
References
[1] K.J. Ivin, J.C. Mol, Olefin Metathesis and Metathesis Polymer-
ization, Academic Press, London, 1997.
[2] M. Schuster, S. Blechert, Angew. Chem. Int. Ed. Engl. 36 (1997)
2036.
[3] R.H. Grubbs, E. Khosravi, Material Science and Technology,
Wiley-VCH, Weinheim, 1999, pp. 64–104.
[4] R.H. Grubbs, S. Chang, Tetrahedron 54 (1998) 4413.
[5] A. Demonceau, A.F. Noels, E. Saive, A.F. Hubert, J. Mol.
Catal. A 76 (1992) 123.
[6] A.F. Noels, A. Demonceau, E. Carlier, A.J. Hubert, R.L. Mar-
quez-Silva, R.A. Sanchez-Delgado, J. Chem. Soc. Chem. Com-
mun. (1988) 783.
4.5.4. Alternati6e preparation of 7a–c
The reaction of [RuCl2(p-cymene)]2 (0.7 mmol) with
the chelating phosphines 5a–c (1.6 mmol) in chloroben-
zene (20 ml) at 120°C for 16 h led to the corresponding
tethered phosphinoarene–ruthenium complexes after
washing with pentane and diethyl ether. Complexes
7a–c were obtained in 91, 85 and 80% yield,
respectively.

64
D. Jan et al. / Journal of Organometallic Chemistry 606 (2000) 55–64
[7] A.F. Noels, A. Demonceau, D. Jan, Russ. Chem. Bull. 48 (1999)
[17] P. Lecomte, D. Meccereyes, P. Dubois, A. Demonceau, A.F.
Noels, R. Je´roˆme, Polym. Bull. 40 (1998) 631.
[18] D. Meccereyes, D. Dahan, P. Lecomte, P. Dubois, A. Demonceau,
A.F. Noels, R. Je´roˆme, J. Polym. Sci. Part A: Polym. Chem. 37
(1999) 2447.
[19] R. Appel, D. Rehbahn, Chem. Ber. 102 (1969) 3955.
[20] K. Issleib, B. Hamann, Z. Anorg. Allg. Chem. 332 (1964)
179.
[21] N.S. Imyanitov, Russ. J. Coord. Chem. (Transl. Koord. Khim.)
11 (1985) 663.
[22] B. Therrien, T.R. Ward, M. Pilkington, C. Hoffmann, F. Gilar-
doni, J. Weber, Organometallics 17 (1998) 330.
[23] B. Therrien, T.R. Ward, Angew. Chem. Int. Ed. Engl. 38 (1999)
405.
1206.
[8] A.F. Noels, A. Demonceau, J. Phys. Org. Chem. 11 (1998) 602.
[9] F. Simal, A. Demonceau, A.F. Noels, D.R.T. Knowles, S.
O’Leary, P.M. Maitlis, O. Gusev, J. Organomet. Chem. 558 (1998)
163.
[10] A. Demonceau, A.W. Stumpf, E. Saive, A.F. Noels, Macro-
molecules 30 (1997) 3127.
[11] L. Delaude, A. Demonceau, A.F. Noels, Macromolecules 32
(1999) 2091.
[12] T. Weskamp, W.C. Schattenmann, M. Spiegler, W.A. Herrmann,
Angew. Chem. Int. Ed. Engl. 37 (1998) 2490.
[13] A. Fu¨rstner, M. Picquet, C. Bruneau, P.H. Dixneuf, J. Chem. Soc.
Chem. Commun. (1998) 1315.
[14] S.M. Hansen, M.A.O. Volland, F. Rominger, F. Eisentra¨ger, P.
Hofmann, Angew. Chem. Int. Ed. Engl. 38 (1999) 1273.
[15] W. Stu¨er, J. Wolf, H. Werner, P. Schwab, M. Schulz, Angew.
Chem. Int. Ed. Engl. 37 (1998) 3421.
[24] D. Jan, A.F. Noels, submitted for publication.
[25] F. Simal, D. Jan, A. Demonceau, A.F. Noels, Tetrahedron Lett.
40 (1999) 1653.
[26] A. Hafner, A. Mu¨hlebach, P.A. van der Schaaf, Angew. Chem.
Int. Ed. Engl. 36 (1997) 2121.
[16] A. Fu¨rstner, L. Ackermann, J. Chem. Soc. Chem. Commun.
(1999) 95.
.