Olefin-metathesis reactions using vinylideneruthenium(II) complexes as catalyst precursors

Article abstract of DOI:10.1016/S0022-328X(00)00094-2

Vinylideneruthenium(II) complexes of the type RuCl₂(=C=CHR)L₂ (R = Bu^t, ferrocenyl; L = PPrⁱ₃, PCy₃), which are easily accessible from [RuCl₂(p-cymene)]₂ and terminal

Full text of DOI:10.1016/S0022-328X(00)00094-2

www.elsevier.nl/locate/jorganchem
Journal of Organometallic Chemistry 606 (2000) 16–25
Olefin-metathesis reactions using vinylideneruthenium(II)
complexes as catalyst precursors
Hiroyuki Katayama, Hideto Urushima, Fumiyuki Ozawa *
Department of Applied Chemistry, Faculty of Engineering, Osaka City Uni6ersity, Sumiyoshi-ku, Osaka 558-8585, Japan
Received 6 December 1999
Abstract
Vinylideneruthenium(II) complexes of the type RuCl (ꢀCꢀCHR)L2 (R=Bu , ferrocenyl; L=PPr , PCy ), which are easily
t
i
3
2
3
accessible from [RuCl (p-cymene)] and terminal alkynes, have been found to serve as good catalyst precursors for ring-opening
2
2
metathesis polymerization (ROMP) of cyclic alkenes and ring-closing metathesis (RCM) of a,v-dienes, a,v-enynes, and dienynes.
These complexes possess high stability toward air and heat, compared with the metathesis-active alkylidenerutheniums. Molecular
3
weight control of poly(norbornene) in extremely wide range (M /10 =500–4) has been achieved by using heteroatom-substituted
n
vinylic compounds such as ethyl vinyl ether and phenyl vinyl sulfide as chain-transfer agents for ROMP. © 2000 Elsevier Science
S.A. All rights reserved.
Keywords: Vinylidene complexes; Ruthenium; Ring-opening metathesis polymerization (ROMP); Ring-closing metathesis (RCM); Molecular
weight control
a variety of ruthenium complexes, including allenyli-
1. Introduction
dene [5] and alkylidyne complexes [6], have been exam-
ined as precursors of metathesis catalysts [5–8]. This is
mainly due to a drawback relating to the preparation of
alkylidene complexes, which generally requires either
cyclopropenes or diazoalkanes; these reagents are
rather difficult to make and handle, especially on a
large scale [9]. In this paper we wish to report that
Olefin–metathesis reactions catalyzed by transition
metal complexes have attracted a great deal of attention
owing to their versatility in organic and polymer syn-
theses [1]. The rapid progress in recent years has been
triggered off by the discovery of the Grubbs catalysts in
mid 1990s [2,3]. It has been demonstrated that the
vinylideneruthenium complexes with the formula
single-component
alkylideneruthenium
complexes
i
3
RuCl (ꢀCꢀCHR)L (L=PPh , PPr , PCy ) can be used
RuCl (ꢀCHR)(PCy ) (R=Ph, CHꢀCPh ) exhibit ex-
2
2
3
3
2
3 2
2
as the other entries of catalyst precursors for olefin-
metathesis reactions [10,11].
tremely high catalytic activity and excellent tolerance
toward polar functional groups. The latter feature,
which is the principal advantage of the ruthenium
catalysts over the conventional early transition metal-
based ones, has brought about a remarkable extension
of the scope of applications of olefin–metathesis reac-
tions in synthetic organic chemistry [1]. Therefore, fol-
low-up studies focusing on improvement in the catalytic
activity of alkylidene complexes have been intensively
carried out by several research groups [4]. Furthermore,
2. Results and discussion
2.1. Synthesis of 6inylideneruthenium complexes
Vinylidene complexes employed in this study are
listed in Chart 1. All of the complexes could be easily
prepared in high yields using common ruthenium com-
plexes and alkynes. Thus, 1a bearing PPh ligands was
3
synthesized from RuCl (PPh ) and t-butylacetylene,
2
3 3
*
Corresponding author.
according to the Wakatsuki’s method [12]. On the other
0022-328X/00/$ - see front matter © 2000 Elsevier Science S.A. All rights reserved.
PII: S0022-328X(00)00094-2

H. Katayama et al. / Journal of Organometallic Chemistry 606 (2000) 16–25
17
i
3
hand, the PPr and PCy complexes (2a–2d, 3a, 3b) were
2.2. ROMP of cyclic alkenes
3
prepared by the treatment of [RuCl (p-cymene)] [13]
2
2
with phosphines (two equivalents/Ru) and alkynes (one
equivalent/Ru) in toluene, as we recently reported [14].
All the reactions proceeded selectively, without any
side-reaction products, detectable by H- and P{ H}-
NMR spectroscopy.
Table 1 summarizes the results of ring-opening
metathesis polymerization (ROMP) of cyclic alkenes.
i
3
Similarly to alkylidene complexes [2a], the PPr and
1
31
1
PCy complexes 2a and 3a exhibited much higher reac-
3
tivity than the PPh complex 1a (entries 1, 2, 6). The
3
ROMP of norbornene (4) with 2a or 3a was completed
within a few minutes at room temperature to give an
almost quantitative yield of polymer with high molecu-
lar weight (entries 2, 6) [16,17]. The catalytic activities
observed were comparable to those so far reported for
the highly active ruthenium catalysts (e.g. RuCl2-
(
CHPh)(PCy3)2 [2a], RuCl (ꢀCHPh)(imidazol-2-yli-
2
dene)2 [4d,f], RuCl (p-cymene)(PCy )/(trimethylsilyl-
diazomethane) [8a], [Ru(h ,h -C H )(NCMe) ](BF ) /
(
2
3
3
3
10
16
3
4 2
ethyl diazoacetate) [8b], RuCl (ꢀCꢀCꢀCPh )(PCy )
2
2
3 2
[5], and [RuHCl(ꢁCCH )(OEt )(PCy ) ]BF [6]; room
3
2
3 2
4
temperature, a few minutes, nearly quantitative yields
for all compounds), and much higher than the follow-
6
2
ing ruthenium catalysts ([(h -C Me )RuH(h -O,P-
6
6
Ph PCH CH OCH )]BF (room temperature, 16 h, 81%
2
2
2
3
4
yield) [7d], RuCl (p-cymene)(PCy ) (60°C, 2 h, 100%
2
3
yield) [8a], TpRuCl(ꢀCꢀCHPh)(PPh ) (80°C, 72 h, 99%
3
yield) [18], and RuCl (PPh ) (50°C, 18 h, 69% yield)
2
3 3
Scheme 1.
[8d]).
In addition to the easy accessibility, the vinylidene
complexes have the advantage of high stability. Thus,
while ruthenium alkylidene complexes are known to
readily undergo bimolecular decomposition in solution
Three kinds of norbornene derivatives (5–7 in
Scheme 1), cyclooctene (8), and dicyclopentadiene (9)
could be also polymerized in high yields (entries 7–11).
The reactions of 5, 6, 8 and 9 required heated condi-
tions, whereas that of 7 proceeded at room tempera-
ture.
[
15], the present complexes were fairly stable in solution
as well as in the solid. For example, no decomposition of
a or 3a was observed by NMR spectroscopy in toluene
13
1
2
C{ H}-NMR spectrum of the poly(norbornene)
at 80°C for 24 h. The solid samples could be safely stored
thus produced exhibited two vinylic carbon signals at l
in the air for a few months at ambient temperature.
133.9 and 133.0 in a 1:9 ratio, which are assignable to
Table 1
a
ROMP of cyclic olefins using vinylideneruthenium complexes as catalyst precursors
Catalyst b
Monomer (equiv.) b,c
Temperature (°C)
Time
Yield (%)
M d/104
M /Mn
w
d
Entry
n
1
2
3
4
5
6
7
8
9
1a
2a
2a
2a
2a
3a
2a
2a
2a
2a
2a
4 (100)
4 (100)
4 (50)
4 (20)
4 (10)
4 (100)
5 (100)
6 (100)
7 (100)
8 (100)
9 (100)
40
24 h
83
98
99
98
96
\99
87
95
97
10.6
59.9
47.6
36.9
10.9
48.3
18.0
5.5
2.31
1.44
1.60
1.28
2.06
2.03
2.81
2.59
2.14
r.t.
r.t.
r.t.
r.t.
r.t.
60
60
r.t.
60
10 min
10 min
10 min
10 min
10 min
24 h
48 h
24 h
12 h
12 h
66.5
10
1
82
\99
37.9
2.13
e
e
1
60
a
Initial concentration: [catalyst] =10 mM (entries 1, 7–11), 2 mM (entries 2–6). Solvent: CH Cl (entries 1–6, 9), ClCH CH Cl (entries 7, 8).
0
2
2
2
2
Entries 10 and 11 were run without solvent.
b
See Chart 1 for the numbering.
Molar ratio relative to catalyst.
Molecular weight was determined by GPC based on polystyrene standards [16].
GPC analysis was infeasible due to insolubility of the polymer.
c
d
e

18
H. Katayama et al. / Journal of Organometallic Chemistry 606 (2000) 16–25
the cis- and trans-CꢀC bonds in the polymer main
bornene using 2a. The results are listed in Table 2. It is
seen that the addition of heteroatom-substituted vinylic
compounds reduces the molecular weight of polymer to
a great extent. Phenyl vinyl sulfide was particularly
effective, and the polymerization was completed within
2 h (entries 2, 3). The molecular weight could be varied
in an extremely wide range (ca. 476 000–6100) by
changing the molar ratio of ethyl vinyl ether to norbor-
nene (entries 1, 4–11). Vinyl acetate and N-vinylpyrro-
lidinone also reduced the molecular weight, though
elongated reaction time was needed for the completion
chain, respectively [19]. The high trans content (ca.
90%), which is commonly observed for the polymer
synthesized by ruthenium complex-catalyzed ROMP
reactions [2a], was also supported by IR spectroscopy.
Thus, the trans-CHꢀCH out-of-plane bending absorp-
−
1
tion at 966 cm
was observed much more strongly
than the cis-CHꢀCH in-plane bending absorption at
−
1
1445 cm
[20].
The vinylidene precursors exhibited the catalytic ac-
tivity without isolation. A representative example is as
(
entries 12, 13).
follows. A solution of 2a (30 mM) was prepared by
i
3
We have originally employed the vinylic compounds
heating a mixture of [RuCl (p-cymene)] , PPr (two
2
2
on a working hypothesis that these reagents may serve
as a terminator, because vinylic compounds bearing a
heteroatom-substituent are commonly added to the
ROMP systems after the polymerization, in order to
eliminate the metathesis activity of ruthenium alkyli-
dene species by converting them into Fischer-type car-
bene complexes [21]. It has been considered that the
initiation step in the present ROMP reactions is much
slower than the propagation step owing to the much
lower reactivity of the vinylidene precursors than the
alkylidene species toward olefin–metathesis. Accord-
ingly, most of the vinylidene precursor is left in the
catalytic system after the polymerization, and the
molecular weight of polymer is therefore much higher
than that expected from the monomer to catalyst pre-
cursor ratio. On the other hand, if the vinylic com-
pounds added to the catalytic systems efficiently trap
the propagating alkylidene intermediates, so as to make
the resulting ruthenium species inactive toward the
ROMP, the remaining part of vinylidene precursors
may start the polymerization in succession. As a result,
most part of the vinylidene complex should participate
in the catalytic reaction and the molecular weight of
polymer will be reduced.
However, different from our original prospect, the
following results have strongly suggested that the
vinylic compounds serve as chain-transfer agents, not
the terminator. Thus, when the vinylic compounds act
as terminator, one may expect that the molecular
weight of poly(norbornene) is mainly controlled by the
ratio of monomer to vinylidene precursor, but little
dependent upon the amount of vinylic compound, espe-
cially in the presence of an excess amount of vinylic
compound relative to the ruthenium precursor. How-
ever, the relation indeed observed was a clear depen-
dence of the molecular weight upon the amount of
vinylic compound (see entries 4–8, 10, 11 in Table 2);
equivalents/Ru), and t-butylacetylene (one equivalent/
Ru) in toluene at 80°C for 16 h. Addition of the
resulting solution into a CH Cl solution of norbornene
2
2
(
100 equivalents/Ru) at room temperature caused rapid
polymerization, giving poly(norbornene) with Mn of
4
5
2.2×10 (M /M =1.82) in 95% yield. The result was
w n
comparable to that conducted with isolated 2a (entry 2
in Table 1), while the starting complex [RuCl (p-
2
cymene)] and its monophosphine derivative RuCl (p-
2
2
cymene)(PCy ) are known to be much less reactive,
3
requiring heated conditions and longer reaction times
[8a].
2.3. ROMP of norbornene in the presence of 6inylic
compounds
As judging from the Mn values in Table 1, the
polymers synthesized with the vinylidene precursors
have much higher molecular weight than that expected
from the monomer to catalyst precursor ratio. The Mn
values were still significantly large even in the presence
of higher ratio of the catalyst precursor (entries 3–5).
This fact strongly indicates low efficiency of the vinyli-
dene complexes as the initiators, i.e. the slow initiation
as compared with the propagation. Based on the data
in entries 2 and 6, the efficiency of 2a and 3a in the
polymerization of norbornene can be estimated to be
1.6 and 2.0%, respectively. Indeed, most parts of the
vinylidene precursors remained unreacted in the reac-
tion solutions after the polymerization, as confirmed by
3
1
1
P{ H}-NMR spectroscopy. In addition, 2a was recov-
ered in 95% yield from the reaction solution of entry 2,
by pouring the solution into MeOH, filtering the result-
ing precipitate of polymer out, and then concentrating
the filtrate to dryness. The recovered complex exhibited
the catalyst activity almost identical with pure 2a, giv-
4
ing poly(norbornene) with M of 68.6×10 (M /M =
plot of the M values against the ratio of monomer to
n
w
n
n
1.81) in 94% yield under the same reaction conditions
ethyl vinyl ether exhibited a linear correlation (Fig. 1).
It has been also observed that a large amount of
vinylidene complex 2a (ca. 90%) is still remained to be
unreacted in the polymerization system of entry 8,
where 5 times excess of ethyl vinyl ether against 2a was
initially present.
as entry 2.
In order to complement this drawback in the ROMP
using vinylidene initiators and to achieve the variable
control of the molecular weight, a variety of vinylic
compounds were added to the ROMP system of nor-

H. Katayama et al. / Journal of Organometallic Chemistry 606 (2000) 16–25
19
Table 2
spectrum was fully consistent with the polymer bearing
ꢂCHꢀCH2 and ꢂCHꢀCHSPh (E/Z=3/2) groups at
each terminus. The M value estimated from the peak
n
integration was 1800, the value was in fair agreement
with the GPC data (M =3600) [16]. Similar structural
n
features having the ꢂCHꢀCH and ꢂCHꢀCHY (Y=
2
OEt, OAc, and N-pyrrolidinonyl) groups at each termi-
nus were observed for all the poly(norbornene)s
prepared in the presence of vinylic compounds (see
Section 3).
The formation of poly(norbornene)s capped with the
ꢂCHꢀCH and ꢂCHꢀCHY groups at each terminus is
2
rationalized by the catalytic cycle in Scheme 2, which
involves a Fischer-type ruthenium carbene complex 10,
generated from vinylidene precursors, as a key interme-
diate. The polymer is formed by successive olefin–
metathesis of 10 with norbornenes, followed by a
chain-transfer process between the resulting 11 with
vinylic compounds.
The catalytic cycle was proposed on the basis of the
following experimental results, which were obtained by
using phenyl vinyl sulfide and 3a. (1) The Fischer-type
carbene complex RuCl (ꢀCHSPh)(PCy ) (10a) (ca. 5%)
2
3 2
31
1
was detected in the polymerization solution by P{ H}-
NMR spectroscopy, together with unreacted 3a; no
other species was observed. (2) The isolated 10a pro-
moted the polymerization of norbornene much faster
than 3a; the reaction with 1 mol% of 10a was com-
pleted within a few seconds, giving poly(norbornene)
Fig. 1. Plot of the Mn values of poly(norbornene) versus the ratio of
norbornene to ethyl vinyl ether (entries 4–8, 10, and 11 in Table 2).
A more suggestive information for the role of vinylic
compounds was gained by the structural analysis of the
resultant polymers. Fig. 2 shows the vinylic proton
1
region of the H-NMR spectrum of poly(norbornene)
isolated from the reaction system of entry 3 in Table 2.
Besides the large signals at l 5.15–5.42 arising from the
d
i
vinylic protons of polymer main chain (H ꢂH ), seven
sets of signals assignable to terminal vinyl group
Fig. 2. 1H-NMR spectrum of poly(norbornene) isolated from the
reaction system of entry 3 in Table 2. The spectrum was recorded in
CDCl3 at 300.11 MHz.
a
c
(
H ꢂH ), and (E)- and (Z)-CHꢀCHSPh groups (trans-
j k
and cis-H and H ), respectively, were observed. The

20
H. Katayama et al. / Journal of Organometallic Chemistry 606 (2000) 16–25
(
4) Complex 10a did not react with terminal alkenes
such as styrene and 1-octene.
The regioselective olefin–metathesis in result (3)
probably reflects the higher thermodynamic stability of
the Fischer-type carbene complex 10a than
RuCl (ꢀCH )(PCy ) which is expected for the metathe-
2
2
3 2
sis reaction with the reverse regioselectivity [22]. If the
reactions of 11 with heteroatom-substituted vinylic
compounds proceed with the same regioselectivity, and
the resulting 10 possesses much higher reactivity toward
norbornene than the vinylidene precursors as observed
in result (2), the propagation intermediate 11 may
always have the ꢀCHY group at the end of polymer
chain. Accordingly, the subsequent olefin–metathesis
leads to the selective formation of the polymer with
ꢂCHꢀCH and ꢂCHꢀCHY groups at each terminus.
2
Furthermore, as suggested from result (4), the terminal
vinyl group in the product polymer will remain unre-
acted in the polymerization system.
Poly(norbornene) prepared in the presence of ethyl
vinyl ether was subjected to further transformations
(
Scheme 3). The ethoxyethenyl group at the one termi-
nus readily hydrolyzed to a formylmethyl group under
an acidic condition. On the other hand, the hydrogena-
Scheme 2. Proposed catalytic cycle for the selective formation of
end-functionalized poly(norbornene)s in the ROMP in the presence
of vinylic compounds (CH ꢀCHY).
tion proceeded at room temperature under H pressure
2
2
(
70 atm) to give the polymer bearing an EtO pendant.
In this reaction, 2a was used as a hydrogenation cata-
lyst. Thus, the ROMP of norbornene and the subse-
quent hydrogenation could be conducted with the same
catalyst precursor.
2.4. RCM of dienes, enynes, and dienynes
Similarly to the other ruthenium-based ROMP cata-
lysts [2a,4d,5b,5c,7b], the vinylideneruthenium com-
plexes exhibited the catalytic activity toward
ring-closing metathesis reactions (RCM). Table 3 sum-
marizes the results using 3b as the catalyst precursor. A
variety of dienes (12, 13, 14) and dienynes (16, 17) were
cleanly converted into the corresponding cyclic alkenes
in high yields without any detectable side reaction
products. On the other hand, the reaction of enyne 15
involved the formation of some unidentified com-
pounds. The RCM of diethyl diallylmalonate (12) could
be also conducted with 2a and 3a in place of 3b, while
the reaction was somewhat slower than that with 3b:
the yields of 18 after 24 h at 60°C were 83 (2a) and 90%
(
3a), respectively.
Scheme 3.
2.5. Consideration for the initiation process
There are several possibilities of the initiation process
4
with Mn of 54.1×10 (M /M =1.81) (compare to
for the present catalytic reactions (processes (a)–(c) in
Scheme 4). Process (a) involves a direct cycloaddition
between the RuꢀCꢀCHR moiety and the alkene sub-
strate to give an exo-methylene ruthenacyclobutane
intermediate, which is a potential precursor for a cata-
w
n
entry 6 in Table 1). (3) RuCl (ꢀCHPh)(PCy ) [2a] as a
simple model of the alkylidene intermediate 11 rapidly
reacted with PhSCHꢀCH in solution at room tempera-
ture (r.t.) to give 10a and styrene in quantitative yields.
2
3 2
2

H. Katayama et al. / Journal of Organometallic Chemistry 606 (2000) 16–25
21
lytically active alkylidene species. Such a cycloaddition
process has been suggested for several vinylideneruthe-
nium complexes [23]. Process (b) includes tautomeriza-
tion of vinylidene to alkyne ligand, followed by
cycloaddition of the resulting alkyne complex with an-
other alkyne. The ruthenacyclopentadiene complex thus
formed has been shown to possess a bis-carbenoid
character [24]. On the other hand, in process (c), the
alkyne complex formed by tautomerization undergoes
dissociation of the alkyne ligand to give a highly coor-
dinatively unsaturated ruthenium(II) species, which re-
acts with an alkene substrate to form an alkylidene
species, as depicted for the case of norbornene [25].
In processes (a) and (b), the vinylidene part of the
precursor complex must be included in the polymer
produced by ROMP, whereas no incorporation of the
vinylidene ligand will be expected for process (c).
Therefore, we next carefully examined the NMR spec-
tra of poly(norbornene).
As we already mentioned, the efficiency of the vinyli-
Scheme 4. Possible initiation processes. [Ru]=RuCl L (n=1 or 2).
2 n
Table 3
a
RCM reactions using 3b as a catalyst precursor
Entry Substrate
Product
Time (h) Yield (%) b
Fig. 3. 1H-NMR spectrum of poly(norbornene) prepared by using
RuCl (ꢀCꢀCHFc)(PPh ) as a catalyst precursor. The spectrum was
2
3 2
recorded in CDCl3 at 300.11 MHz.
i
3
dene precursors bearing PPr and PCy ligands was
3
significantly low, and the polymers formed by the
ROMP had high molecular weight. Therefore, it was
infeasible to inspect the possibility of incorporation of
vinylidene ligands into the products. On the other
hand, since the reactions with PPh -coordinated com-
3
plexes afforded poly(norbornene) with relatively low
molecular weight (see entry 1 in Table 1), the presence
of the substituent derived from the vinylidene ligand
could be detected by NMR spectroscopy.
Fig. 3 shows the H-NMR spectrum of poly(norbor-
nene), which was prepared in 92% yield using the
a
All reactions were run in CDCl3 at 60°C. Initial concentration:
[
3b] =5.0 mM, [substrate] =0.25 M.
0
0
1
b
1
Determined by H-NMR analysis using mesitylene as an internal
standard.

22
H. Katayama et al. / Journal of Organometallic Chemistry 606 (2000) 16–25
vinylidene complex RuCl (ꢀCꢀCHFc)(PPh ) (1 mol%)
P O (Merck, SICAPENT). IR spectra were recorded
2 5
2
3 2
in situ generated from RuCl (PPh ) and ferrocenyl-
on a JASCO FT/IR-410 instrument. NMR spectra were
2
3 3
1
acetylene (FcCꢁCH) (ten equivalents/Ru). In addition
to the peaks of poly(norbornene) chain at l 1.0–2.9
recorded on a Varian Mercury 300 spectrometer ( H-
13
NMR, 300.11 MHz; C-NMR, 75.47 MHz) or a Jeol
1
13
(
cyclopentane framework) and at around l 5.3 ((E)-
a-400 spectrometer ( H-NMR, 399.65 MHz; C-NMR,
and (Z)-CHꢀCH groups), small signals arising from the
ferrocenyl group were observed. Two triplets at l 4.51
and 4.34 (each 2H) are assignable to the Cp ring bound
100.40 MHz). The chemical shifts are reported in l
1
13
(ppm), referred to H (of residual protons) and
C
signals of the deuterated solvents as internal standards.
The number- and weight-average molecular weight (Mn
and M ) and polydispersity indices (M /M ) of poly-
a
to the polymer chain (Cp ), and the singlet at l 4.28
b
(
5H) to the other Cp ring (Cp ), respectively. The
w
w
n
molecular weight estimated from the peak integration
of these signals relative to the vinylic proton signals of
mers were determined by gel permeation chromatogra-
phy (THF, 38°C) using polystyrene standards and a
Tosoh 8000 GPC system equipped with TSK gel
4
the polymer chain was 4.7×10 , which was in good
4
agreement with the GPC data (M =6.9×10 ) [26].
columns
(G7000HHR
,
G4000HHR
,
G3000HHR,
n
Further implication of the participation of vinylidene
complexes in the initiation process has been provided
by the following results. Thus, the reactivity of vinyli-
dene precursors was significantly affected by the sub-
stituents at the b-carbon of vinylidene ligands. Table 4
G2000HHR; the molecular-weight range=2 890 000–
946).
Dichloromethane and 1,2-dichloroethane were dried
over CaH . Toluene and THF were dried over sodium
2
benzophenone ketyl. These solvents were distilled and
stored over activated molecular sieves (MS4A) under
summarizes the results of RCM of diethyl diallyl-
i
3
malonate (12) using four kinds of PPr -coordinated
an argon atmosphere. CDCl was purified by passage
3
vinylidene complexes (2a–2d) (entries 1–4), which were
examined under a controlled condition (at 60°C, for 24
h). It is seen that the more donating substituent tends
to give the higher reactivity. A similar trend on the
effect of b-substituents has previously been observed
through a basic alumina column and stored over acti-
vated MS4A under a nitrogen atmosphere. Norbornene
(4) was distilled from sodium prior to use. The vinyli-
deneruthenium complexes RuCl {=C=C(H)R}L
2
2
t
i
(R=Bu , Fc; L=PPh , PPr , PCy ) [12,14], RuCl -
3
3
3
2
i
i
3 2 2
for the relative stability of RhCl(ꢀCꢀCHR)(PPr ) com-
(PPr ) (NCMe) [14], endo,endo-5,6-bis(methoxycar-
3
2
plexes, where the more donating substituent gives rise
to the higher stability of vinylidene complex [27]. It was
also found that the ruthenium(II) complex without a
vinylidene ligand is inactive toward the RCM (entry 5).
bonyl)-2-norbornene (5) [28], endo,endo-5,6-bis(me-
thoxymethyl)-2-norbornene (6) [29], exo,exo-5,6-bis-
(methoxycarbonyl)-7-oxa-2-norbornene (7) [25], N,N-
diallyl-p-toluenesulfonamide
(13)
[30],
diethyl
allyl(3-butenyl)malonate (14) [31], 2-butyne-1,4-diol di-
allyl ether (16) [32], and 5-(1-propynyl)-5-[(triethylsi-
lyl)oxy]1,8-nonadiene (17) [32] were synthesized
according to the literature. All other compounds were
obtained from commercial sources and used without
purification.
3
. Experimental
3.1. General procedures
All manipulations were performed under a nitrogen
or argon atmosphere using common Schlenk tech-
niques. The inert gases were dried by passage through
3.2. ROMP of cyclic alkenes
A typical procedure (entry 2 in Table 1) is as follows.
A solution of norbornene (4) (191 mg, 2.03 mmol) in
Table 4
CH Cl2 (2.3 ml) was added to
a
solution of
2
RCM of diethyl diallylmalonate (12) using 2a–2d as catalyst precur-
t
i
3 2
sors a
RuCl {ꢀCꢀC(H)Bu }(PPr ) (2a) (11.6 mg, 0.0202
2
mmol) in CH Cl (7.6 ml) at r.t. The reddish brown
2
2
Catalyst precursor b
Yield of 18 (%)
c
solution immediately set to gel. The resulting viscous
mixture was poured into a vigorously stirred MeOH
Entry
t
1
2
3
4
5
2a (Bu )
83
75
55
10
0
(
ca. 40 ml) containing 0.1% of 2,6-di-tert-butyl-4-
2b (ferrocenyl)
2c (Ph)
methylphenol (BHT) to give white precipitate. The
crude product was collected by filtration and purified
by silica gel column chromatography using CH Cl
2d (p-MeO C H )
2
6
4
i
3 2
RuCl (PPr ) (NCMe)
2
2
2
2
containing 0.1% BHT as eluent. The eluate was concen-
trated and poured into a vigorously stirred MeOH (ca.
100 ml, containing 0.1% BHT) to give a white solid of
poly(norbornene) (poly(4)), which was collected by
a
All reactions were run in CDCl3 at 60°C for 24 h. Initial
concentration: [2] =5.0 mM, [12] =0.25 M.
0
0
b
c
See Scheme 1 for the numbering.
1
Determined by H-NMR analysis using mesitylene as an internal
standard.

H. Katayama et al. / Journal of Organometallic Chemistry 606 (2000) 16–25
23
filtration and dried under vacuum (187 mg, 98%). The
reactions listed in Table 1 were similarly carried out,
except for entries 10 and 11, where the polymerizations
were conducted in neat cyclooctene (8) or dicyclopenta-
diene (9) (2.0 ml) in the presence of 2a (10.0 mg, 0.0174
mmol). The polymers thus obtained were characterized
by IR and NMR spectroscopy and GPC analysis, ex-
cept for poly(9), whose characterizations were infeasible
due to its insolubility. The spectroscopic and analytical
was added to a solution of 2a (18.4 mg, 0.0320 mmol)
and phenyl vinyl sulfide (21.8 mg, 0.160 mmol) in
CH Cl (14 ml) at r.t. The reaction mixture was stirred
2
2
at r.t. for 2 h. GLC analysis revealed the full consump-
tion of 4. The resulting reddish brown solution was
poured into a vigorously stirred MeOH (ca. 50 ml)
containing 0.1% BHT to give white suspension. A part
of the solvent was removed by rotary evaporator, and
the white precipitate was collected by filtration, washed
with MeOH, and dried under vacuum (127 mg, 84%).
The reactions listed in Table 2 were similarly carried
out. The polymers thus obtained were characterized by
IR and NMR spectroscopy and GPC analysis. The
spectroscopic data are as follows.
data are as follows.
1
poly(4): H-NMR (CDCl , 21°C) l 5.34 (dd, J=5.0,
3
2
.4 Hz, ꢀCH of trans-polymer), 5.20 (dd, J=5.4, 2.0
Hz, ꢀCH of cis-polymer), 2.45–2.32, 1.91–1.70, 1.44–
13
1
1
2
4
2
.24, 1.12–0.97 (each m, CH2). C{ H}-NMR (CDCl ,
3
1°C) l 133.9, 133.0 (each s, ꢀCH), 43.4, 43.1, 42.1,
1.4, 38.4, 32.3, 32.2 (each s, CH and CH ). IR (KBr)
2
−
1
944, 2864, 1445, 1034, 966, 746 cm . Anal. Calc. for
C H : C, 89.30; H, 10.70. Found: C, 88.89; H, 10.63%.
7
10
1
poly(5): H-NMR (CDCl , 18°C) l 5.52 (br, ꢀCH),
3
3
.63 (m, CO Me), 3.09, 2.85 (each br, CH), 1.94 (br,
2
1
1
3
1
Y=OEt: H-NMR (CDCl , 19°C) l 5.86 (d, J=9.0
CH2). C{ H}-NMR (CDCl , 21°C) l 172.5 (br,
3
3
k
c
Hz, H ), 5.80 (ddd, J=17.5, 10.0, 7.5 Hz, H ), 5.34 (dd,
CO Me), 132.4–130.2 (m, ꢀCH), 51.2, 44.9, 44.5, 44.2,
2
d
i
J=5.0, 2.4 Hz, H ꢂH of trans-polymer), 5.20 (dd,
3
2
9.4, 39.0, 38.5, 37.9 (each br, CH and CH ). IR (KBr)
952, 1737, 1438, 1388, 1200, 807 cm . Anal. Calc. for
2
d
i
−
1
J=5.6, 1.7 Hz, H ꢂH of cis-polymer), 4.95 (dd, J=
b a
1
(
7.5, 2.1 Hz, H ), 4.86 (dd, J=10.0, 2.1 Hz, H ), 4.31
C H O : C, 62.85; H, 6.71. Found: C, 62.29; H,
1
1
14
4
j
dd, J=9.0, 7.0 Hz, H ), 3.77 (q, J=7.5 Hz, CH CH ),
6
.77%.
2 3
1
1.24 (t, J=7.5 Hz, CH CH ), 2.82–2.73, 2.50–2.35,
poly(6): H-NMR (CDCl , 18°C) l 5.39 (br, ꢀCH),
.45–3.28 (m, CH OCH ), 2.98, 2.68, 2.28, 1.94, 1.42
2 3
3
1
.92–1.69, 1.43–1.30, 1.11–0.98 (each m, CH and
3
2
3
1
3
1
(
each m, CH and CH2). C{ H}-NMR (CDCl , 21°C)
CH2).
3
1
Y=OAc: H-NMR (CDCl , 19°C) l 6.96 (d, J=
l
132.2–131.6 (m, ꢀCH), 71.3, 71.1 (each s,
2
3
k
k
1
7.8 Hz, trans-H ), 6.94 (d, J=6.4 Hz, cis-H ), 5.80
CH OCH ), 58.6 (s, CH OCH ), 45.7, 44.1, 39.5, 38.9,
3
1
3
2
3
c
(
ddd, J=17.2, 9.6, 7.2 Hz, H ), 5.34 (dd, J=4.4, 2.0
8.1 (each s, CH and CH ). IR (KBr) 2976, 2870, 2806,
2
450, 1390, 1197, 1107, 957, 795 cm . Anal. Calc. for
d
i
−
1
Hz, H –H of trans-polymer), 5.20 (dd, J=6.4, 1.2 Hz,
d
i
b
H –H of cis-polymer), 4.96 (d, J=17.2 Hz, H ), 4.86
C H O : C, 72.49; H, 9.95. Found: C, 72.15; H,
1
1
18
2
a
j
(
2
d, J=9.6 Hz, H ), 4.82 (dd, J=9.6, 6.4 Hz, cis-H ),
.85–2.70, 2.55–2.35, 1.89–1.70, 1.43–1.27, 1.12–1.01
(each m, CH and CH ), 2.14 (s, OCOCH of Z isomer),
2.11 (s, OCOCH of E isomer). IR (KBr) 1760 (wCꢀO
9.95%.
1
poly(7): H-NMR (CDCl , 21°C) l 5.88 (br), 5.60
3
1
3
1
(
br), 5.06 (br), 4.68 (br), 3.68 (s), 3.08 (br). C{ H}-
2
3
NMR (CDCl , 21°C) l 170.9, 170.7 (each s, CO Me),
3
)
3
2
−
1
1
5
1
32.3, 132.1, 131.2, 130.8 (each s, ꢀCH), 80.4, 77.2,
3.2, 52.8, 52.5, 52.3, 52.2. IR (KBr) 3002, 2954, 1729,
436, 1209, 1010, 872, 736, 686 cm . Anal. Calc. for
cm
Y=SPh: H-NMR (CDCl
7.21–7.16 (each m, Ph), 6.11 (dd, J=14.3, 0.8 Hz,
.
1
, 19°C) l 7.34–7.29,
3
−
1
k
k
C H O : C, 56.60; H, 5.70. Found: C, 56.15; H,
trans-H ), 6.10 (dd, J=10.2, 1.2 Hz, cis-H ), 5.98 (dd,
J=15.2, 8.0 Hz, trans-H ), 5.80 (ddd, J=16.8, 10.0,
7.2 Hz, H ), 5.79 (dd, J=9.2, 9.0 Hz, cis-H ), 5.34 (dd,
1
0
12
5
j
5
.76%.
poly(8): H-NMR (CDCl , 19°C): l 5.40 (br, ꢀCH),
1
c
j
3
1
3
1
d
i
2
.00 (br, ꢀCHCH ꢂ), 1.30 (br, CH ). C{ H}-NMR
J=4.4, 2.0 Hz, H ꢂH of trans-polymer), 5.20 (dd,
d i
2
2
(
2
CDCl , 21°C) l 130.3, 129.9 (each s, ꢀCH), 32.6, 29.8,
J=6.4, 2.0 Hz, H ꢂH of cis-polymer), 4.98 (ddd,
3
b
9.6, 29.2, 29.1, 27.2 (each s, CH ). IR (KBr) 3004,
923, 2852, 1463, 1052, 966, 808, 724 cm . Anal. Calc.
J=14.8, 2.0, 1.2 Hz, H ), 4.86 (ddd, J=10.4, 2.0, 1.2
2
−
1
a
2
Hz, H ), 2.80–2.76, 2.45–2.41, 1.95–1.55, 1.46–1.23,
for C H : C, 87.19; H, 12.81. Found: C, 86,71; H,
1.06–1.01 (each m, CH and CH ).
2
8
14
1
12.99%.
Y=N-pyrrolidinonyl: H-NMR (CDCl , 19°C) l
3
k
6
.87 (d, J=14.4 Hz, H ), 5.80 (ddd, J=17.2, 10.0, 7.6
c
d
i
3.3. ROMP of norbornene in the presence of 6inylic
Hz, H ), 5.34 (dd, J=4.4, 2.0 Hz, H ꢂH of trans-poly-
d i
compounds
mer), 5.20 (dd, J=5.6, 1.6 Hz, H ꢂH of cis-polymer),
b
4
.97 (d, J=17.2 Hz, H ), 4.90 (dd, J=14.4, 6.0 Hz,
j
a
A typical procedure (entry 3 in Table 2) is as follows.
H ), 4.87 (d, J=10.0 Hz, H ), 3.49 (t, J=7.2 Hz, CH
2
A solution of 4 (151 mg, 1.60 mmol) in CH Cl (2.3 ml)
of pyrrolidinone ring), 2.08 (m, CH of pyrrolidinone
2
2
2

24
H. Katayama et al. / Journal of Organometallic Chemistry 606 (2000) 16–25
ring), 2.82–2.73, 2.49–2.35, 1.91–1.70, 1.41–1.25,
3.6. Hydrogenation of poly(norbornene)
1
cm
.09–0.98 (each m, CH and CH ). IR (KBr) 1710 (wCꢀO)
2
−
1
.
A stainless-steel autoclave was charged with complex
2
a (40.3 mg, 0.0701 mmol) and poly(norbornene) (132
4
mg, M /10 =2.0). Toluene (15 ml) was added and the
n
3
.4. Reaction of RuCl (ꢀCHPh)(PCy ) with
system was flushed with hydrogen. Then the reaction
mixture was pressurized with 70 atm of hydrogen and
stirred at r.t. for 16 h. The resulting mixture was
transferred into a round-bottom flask and concentrated
to dryness by a rotary evaporator. The residue was
washed with MeOH (30 ml) and dried under vacuum to
give a white solid (108 mg). The hydrogenated
2
3 2
PhSCHꢀCH2
The complex RuCl (ꢀCHPh)(PCy ) [2a] (11.2 mg,
2
3 2
1
3.6 mmol) was loaded in a 5 mm NMR sample tube
equipped with a rubber septum cap, and the system was
replaced with nitrogen gas. CDCl (0.7 ml) and phenyl
vinyl sulfide (18.5 mg, 13.6 mmol) was added at r.t., and
the mixture was gently shaken. The H- and P{ H}-
NMR spectra measured after 5 min revealed quantita-
tive formation of RuCl (ꢀCHSPh)(PCy ) (10a) and
3
1
poly(norbornene) was characterized by H- and
1
31
1
13
1
1
C{ H}-NMR spectroscopy [33]. H-NMR (CDCl ,
3
13
1
60°C) l 2.05–1.50, 1.30–0.55 (each m). C{ H}-NMR
(CDCl , 60°C) l 40.8, 40.5, 35.8, 31.8 (each s).
2
3 2
3
styrene; no trace of the starting materials were detected.
Complex 10a was independently prepared and iden-
tified. To a solution of RuCl (ꢀCHPh)(PCy ) (454 mg,
3.7. RCM of dienes, enynes, and dienynes
2
3 2
0
.552 mmol) in CH Cl (10 ml) was added phenyl vinyl
A typical procedure (entry 1 in Table 3) is as follows.
2
2
sulfide (83 mg, 0.61 mmol) at r.t. with stirring. After 10
min, volatile materials were thoroughly removed by
pumping. The resulting solid was washed with MeOH
To a solution of RuCl {ꢀCꢀC(H)Fc}(PCy ) (3b) (5.1
2
3 2
mg, 5.4 mmol) and mesitylene (internal standard; 10 ml)
in CDCl (1.1 ml) was added diethyl diallylmalonate
3
(
3 ml×3) at r.t. to give a purple microcrystalline solid
(12) (57.4 mg, 0.270 mmlol) at r.t. A part of the
resulting solution (0.6 ml) was transferred into a 5 mm
NMR sample tube via a cannula and frozen in liquid
nitrogen. The sample was evacuated by vacuum pump
and fire-sealed. The reaction mixture was slowly
warmed to r.t., and then heated to 60°C. The reaction
of 10a (420 mg, 89%), which was spectroscopically
pure. Analytically pure compound containing one
equivalent of THF in the crystal was obtained by
1
recrystallization from THF (65% yield). H-NMR
(
CDCl , 23°C): l 17.65 (s, 1H, RuꢀCH), 7.46–7.34 (m,
3
1
5
1
1
2
1
H, Ph), 3.78–3.71 (m, 4H, THF), 2.74–2.56, 2.02–
.90, 1.86–1.48, 1.38–1.12 (each m, 66H, Cy), 1.89–
progress was followed by H-NMR spectroscopy, and
the yield of the RCM product was determined based on
the relative peak integration of methyl protons of
mesitylene (l 2.27) to allylic protons of 12 (l 2.63) and
18 (l 3.01). All the reactions listed in Table 3 and Table
1
3
1
.84 (m, 4H, THF). C{ H}-NMR (CDCl , 23°C): l
3
2
80.5 (t, JPC=9 Hz, RuꢀCH), 141.0 (s, ipso-C of Ph),
29.2, 129.1, 128.3 (each s, Ph), 67.9 (s, THF), 32.5
virtual triplet, Japp=9 Hz, C of Cy), 29.7 (s, C of
1
3,5
1
(
4 were similarly carried out. H-NMR data of the
2,6
Cy), 27.7 (virtual triplet, J =5 Hz, C of Cy), 26.4
RCM products (18–23) were identical with those re-
ported [30–32].
app
4
31
1
(
s, C of Cy), 25.6 (s, THF). P{ H}-NMR (CDCl ,
3
2
3°C): l 30.9 (s). Anal. Calc. for C H Cl P -
43 72 2 2
RuS·C H O: C, 60.89; H, 8.70. Found: C, 60.71; H,
4
8
8.70%.
4. Conclusions
We have demonstrated that the vinylidenerutheni-
i
3
3.5. Acid hydrolysis of poly(norbornene) capped with a
um(II) complexes bearing PPr and PCy ligands (2a,
3
(
Z)-ethoxyethenyl group
2b, 3a, 3b) serve as good catalyst precursors for ROMP
of cyclic alkenes and RCM of dienes, enynes, and
dienynes. These complexes are easily prepared in almost
quantitative yields, only by heating a toluene solution
of [RuCl (p-cymene)] , phosphines, and alkynes. The
resulting complexes are fairly stable in solution as well
as in the solid. In addition, the vinylidene complexes
exhibit the catalytic activity without isolation. The easy
accessibility and the high stability are predominant over
the other ruthenium catalysts and catalyst precursors,
recently developed for the olefin–metathesis reactions.
We have also found that the molecular weight of
poly(norbornene) can be controlled in an extremely
Poly(norbornene) (50 mg) isolated from the reaction
system of entry 8 in Table 2 was dissolved in CDCl3
(
0.7 ml). To this solution, 5% aqueous HCl (10 ml) was
2
2
added at r.t. The mixture was transferred into a 5 mm
NMR sample tube and gently shaken. After 30 min, the
1
H-NMR spectrum showed the disappearance of the
signals due to the (Z)-ethoxyethenyl group, and a new
triplet at l 9.75 (J=3.0 Hz) assignable to aldehyde
proton of the formylmethyl group appeared. The other
signals were almost identical with the starting
poly(norbornene).

H. Katayama et al. / Journal of Organometallic Chemistry 606 (2000) 16–25
25
Makromol. Chem. 176 (1975) 3121. (d) K. Hiraki, A. Kuroiwa,
H. Hirai, J. Polym. Sci. A-1 9 (1971) 2323. (e) F.W. Michelotti,
W.P. Keaveney, J. Polym. Sci. A 3 (1965) 895.
wide range by using heteroatom-substituted vinylic
compounds as chain-transfer agents. The polymers pro-
duced by this process have well-defined structures.
Thus, the one terminus is selectively capped with the
substituents originated from the chain-transfer agents,
whereas the other terminus remains unsubstituted.
These structural features allow the further transforma-
tion of poly(norbornene) into highly functionalized
polymers.
[
9] For alternative synthetic routes to alkylidene complexes, see
Refs. [3g] and [3h].
[10] Portions of this work have been communicated: (a) H.
Katayama, F. Ozawa, Chem. Lett. (1998) 67. (b) H. Katayama,
H. Urushima, F. Ozawa, Chem. Lett. (1999) 369.
[
11] The ROMP activity of RuCl (ꢀCꢀCH )(PCy ) , which is synthe-
2 2 3 2
sized by the reaction of RuCl (ꢀCHPh)(PCy ) with 1,2-propadi-
2
3 2
ene, has been briefly mentioned by Grubbs et al.: Ref. [2a].
12] Y. Wakatsuki, H. Yamazaki, N. Kumegawa, T. Satoh, J.Y.
Satoh, J. Am. Chem. Soc. 113 (1991) 9604.
[
[
13] M.A. Bennett, A.K. Smith, J. Chem. Soc. Dalton Trans. (1974)
References
2
33.
[
[
[
14] H. Katayama, F. Ozawa, Organometallics 17 (1998) 5190.
15] M. Ulman, R.H. Grubbs, J. Org. Chem. 64 (1999) 7202.
16] Actual Mn is known to be approximately half of the GPC value
based on polystyrene standards: (a) L.R. Gilliom, R.H. Grubbs,
J. Am. Chem. Soc. 108 (1986) 733. (b) T.J. Katz, S.J. Lee, N.
Acton, Tetrahedron Lett. (1976) 4247. (c) T.J. Katz, N. Acton,
Tetrahedron Lett. (1976) 4251.
[
1] (a) F.Z. D o¨ rwald, Metal Carbenes in Organic Synthesis, Wiley-
VCH, Weinheim, 1999. (b) A. F u¨ rstner, Alkene Metathesis in
Organic Synthesis, Springer, Berlin, 1998. (c) S. Armstrong, J.
Chem. Soc. Perkin Trans. 1 (1998) 371. (d) M. Schuster, S.
Blechert, Angew. Chem. Int. Ed. Engl. 36 (1997) 2036.
[
2] (a) P. Schwab, R.H. Grubbs, J.W. Ziller, J. Am. Chem. Soc. 118
(1996) 100. (b) S.-B.T. Nguyen, R.H. Grubbs, J.W. Ziller, J. Am.
[
[
[
17] Complexes 2b–2d and 3b gave almost the same results (M /
n
Chem. Soc. 115 (1993) 9858.
4
1
0 =47.6–61.3, M /M =1.55–1.99, \98%).
w
n
[
3] For further studies of Grubbs group, see: (a) M. Scholl, S. Ding,
C.W. Lee, R.H. Grubbs, Org. Lett. 1 (1999) 953. (b) D.M. Lynn,
B. Mohr, R.H. Grubbs, J. Am. Chem. Soc. 120 (1998) 1627. (c)
E.L. Dias, R.H. Grubbs, Organometallics 17 (1998) 2758. (d)
M.S. Sanford, L.M. Henling, R.H. Grubbs, Organometallics 17
18] H. Katayama, T. Yoshida, F. Ozawa, J. Organomet. Chem. 562
1998) 203.
19] K.J. Ivin, D.T. Laverty, J.J. Rooney, Makromol. Chem. 178
1977) 1545.
(
(
[
[
20] M.A. Golub, J. Polym. Sci. Polym. Lett. Ed. 12 (1974) 295.
21] (a) M.A. Hillmyer, W.R. Laredo, R.H. Grubbs, Macromolecules
(1998) 5384. (e) T.A. Kirkland, D.M. Lynn, R.H. Grubbs, J.
Org. Chem. 63 (1998) 9904. (f) E.L. Dias, S.B.-T. Nguyen, R.H.
Grubbs, J. Am. Chem. Soc. 119 (1997) 3887. (g) T.E. Wilhelm,
T.R. Belderrain, S.N. Brown, R.H. Grubbs, Organometallics 16
2
2
8 (1995) 6311. (b) S. Kanaoka, R.H. Grubbs, Macromolecules
8 (1995) 4707.
[
[
22] The higher stability of Fischer-type ruthenium carbenes than
alkylidenerutheniums has been documented [2a].
(1997) 3867. (h) T.R. Belderrain, R.H. Grubbs, Organometallics
1
6 (1997) 4001. (i) M. Weck, P. Schwab, R.H. Grubbs, Macro-
23] (a) C. Slugovc, K. Mereiter, R. Schmid, K. Kirchner, J. Am.
Chem. Soc. 120 (1998) 6175. (b) C. Slugovc, K. Mereiter, R.
Schmid, K. Kirchner, Organometallics 18 (1999) 1011. (c) K.
Onitsuka, H. Katayama, K. Sonogashira, F. Ozawa, J. Chem.
Soc. Chem. Commun. (1995) 2267.
24] (a) K. Mauthner, K.M. Soldouzi, K. Mereiter, R. Schmid, K.
Kirchner, Organometallics 18 (1999) 4681. (b) J. Le Paih, S.
D e´ rien, P.H. Dixneuf, Chem. Commun. (1999) 1437. (c) M.O.
Albers, D.J.A. de Waal, D.C. Liles, D.J. Robinson, E. Singleton,
M.B. Wiege, J. Chem. Soc. Chem. Commun. (1986) 1680.
25] A. M u¨ hlebach, P. Bernhard, N. B u¨ hler, T. Karlen, A. Ludi, J.
Mol. Catal. 90 (1994) 143.
molecules 29 (1996) 1789.
[
4] (a) T. Weskamp, F.J. Kohl, W. Hieringer, D. Gleich, W.A.
Herrmann, Angew. Chem. Int. Ed. Engl. 38 (1999) 2416. (b)
S.M. Hansen, F. Rominger, M. Metz, P. Hofmann, Chem. Eur.
J. 5 (1999) 557. (c) J.S. Kingsbury, J.P.A. Harrity, P.J. Bonitate-
bus, Jr., A.H. Hoveyda, J. Am. Chem. Soc. 121 (1999) 791. (d)
J. Huang, E.D. Stevens, S.P. Nolan, J.L. Petersen, J. Am. Chem.
Soc. 121 (1999) 2674. (e) C. Hinderling, C. Adlhart, P. Chen,
Angew. Chem. Int. Ed. Engl. 37 (1998) 2685. (f) T. Weskamp,
W.C. Schattenmann, M. Spiegler, W.A. Herrmann, Angew.
Chem. Int. Ed. Engl. 37 (1998) 2490.
5] (a) A. F u¨ rstner, A.F. Hill, M. Liebl, J.D.E.T. Wilton-Ely, Chem.
Commun. (1999) 601. (b) M. Picquet, C. Bruneau, P.H. Dixneuf,
Chem. Commun. (1998) 2249. (c) A. F u¨ rstner, M. Picquet, C.
Bruneau, P.H. Dixneuf, Chem. Commun. (1998) 1315.
6] W. St u¨ er, J. Wolf, H. Werner, P. Schwab, M. Schulz, Angew.
Chem. Int. Ed. Engl. 37 (1998) 3421.
7] For other single-component ruthenium catalysts, see: (a) I. del
R ´ı o, G. van Koten, Tetrahedron Lett. 40 (1999) 1401. (b) A.
F u¨ rstner, L. Ackermann, Chem. Commun. (1999) 95. (c) M.
Bartz, J. K u¨ ther, R. Seshadri, W. Tremel, Angew. Chem. Int.
Ed. Engl. 37 (1998) 2466. (d) E. Lindner, S. Pautz, R. Fawzi, M.
Steimann, Organometallics 17 (1998) 3006. (e) B. Moreno, S.
Sabo-Etienne, F. Dahan, B. Chaudret, J. Organomet. Chem. 498
[
[
[
[
26] A similar attempt to identify the fate of t-butylvinylidene moiety
t
in 1a was unsuccessful owing to overlap of the Bu signal with
1
the signals of cyclopentane frameworks in the H-NMR spec-
trum.
[
[
[
27] H. Katayama, K. Onitsuka, F. Ozawa, Organometallics 15
(
1996) 4642.
[
[
28] A.C. Cope, E. Herrick, Org. Synth. Collect. 4 (1963) 304.
29] D.M. Lynn, S. Kanaoka, R.H. Grubbs, J. Am. Chem. Soc. 118
(
1996) 784.
[
30] (a) T. Kataoka, M. Yoshimatsu, Y. Noda, T. Sato, H. Shimizu,
M. Hori, J. Chem. Soc. Perkin Trans. 1 (1993) 121. (b) B.L.
Pagenkopf, T. Livinghouse, J. Am. Chem. Soc. 118 (1996) 2285.
(
3
(
1995) 139. (f) W.J. Feast, D.B. Harrison, Polym. Bull. 25 (1991)
43. (g) B.M. Novak, R.H. Grubbs, J. Am. Chem. Soc. 110
1988) 7542.
8] For further studies of ruthenium-based metathesis catalysts, see:
a) A. Demonceau, A.W. Stumpf, E. Saive, A.F. Noels, Macro-
[31] (a) T.A. Kirkland, D.M. Lynn, R.H. Grubbs, J. Org. Chem. 63
(1998) 9904. (b) R. Grigg, J.F. Malone, T.R.B. Mitchell, A.
Ramasubbu, R.M. Scott, J. Chem. Soc. Perkin Trans. 1 (1984)
1745.
[32] S.H. Kim, W.J. Zuercher, N.B. Bowden, R.H. Grubbs, J. Org.
Chem. 61 (1996) 1073.
[
(
molecules 30 (1997) 3127. (b) W.A. Herrmann, W.C. Schatten-
mann, O. Nuyken, S.C. Glander, Angew. Chem. Int. Ed. Engl.
3
[33] B. Al-Samak, V. Amir-Ebrahimi, A.G. Carvill, J.G. Hamilton,
J.J. Rooney, Polym. Int. 41 (1996) 85.
5 (1996) 1087. (c) L. Porri, P. Diversi, A. Lucherini, R. Rossi,