Understanding the effect of allylic methyls in olefin cross-metathesis

Article abstract of DOI:10.1016/j.jorganchem.2005.09.030

A series of NMR spectroscopy experiments have been conducted with both the model compound, 3-methyl-1-pentene and the corresponding ADMET monomer 3,6,9-trimethylundeca-1,10-diene (11) to better understand the effect of allylic methyls during olefin metath

Full text of DOI:10.1016/j.jorganchem.2005.09.030

Journal of Organometallic Chemistry 691 (2006) 585–594
www.elsevier.com/locate/jorganchem
Understanding the effect of allylic methyls in olefin cross-metathesis
*
Florence C. Courchay, Travis W. Baughman, Kenneth B. Wagener
The George and Josephine Butler Polymer Research Laboratory, Department of Chemistry, University of Florida,
P.O. Box 117200, Gainesville, FL 32611, United States
Received 29 July 2005; received in revised form 21 September 2005; accepted 21 September 2005
Available online 5 December 2005
Abstract
A series of NMR spectroscopy experiments have been conducted with both the model compound, 3-methyl-1-pentene and the cor-
responding ADMET monomer 3,6,9-trimethylundeca-1,10-diene (11) to better understand the effect of allylic methyls during olefin
metathesis chemistry. Traditional ADMET catalysts such as SchrockÕs molybdenum (1), and GrubbsÕ ruthenium 1st and 2nd generation
(2 and 3) were examined under cross-metathesis and ADMET conditions. Regardless of catalyst selection, 50% or less metathesis con-
version was observed for all reactions, especially in the case of the more sterically encumbered diene. With SchrockÕs molybdenum cat-
alyst 1, the reaction leads to an accumulation of the non-productive metallacyclobutane, trapping the catalyst in an inactive form. With
GrubbsÕ ruthenium catalysts 2 and 3, the substrate coordinates to the metal center primarily to yield non-productive metathesis, which
results in a build-up of the methylidene complex leading to catalyst decomposition. These results are directly correlated to the orientation
of the substrateÕs bulk during the metallacyclobutane formation, the alkyl branch being adjacent to the metal center in the case of the
molybdenum catalyst 1, and opposite to it in the case of ruthenium catalyst 2 and 3.
Ó 2005 Elsevier B.V. All rights reserved.
Keywords: Olefin metathesis; Cross-metathesis; ADMET; Allylic methyls; Carbene catalysts
1. Introduction
hand, SchrockÕs highly active carbene complexes are oxo-
philic which renders them sensitive to air and moisture,
and thus inappropriate when used with certain functional-
ities such as aldehydes and alcohols [10]. On the other
hand, ruthenium complexes, which possess a much higher
functional group tolerance, are less active towards elec-
tron-poor or sterically demanding olefins [10,11]. Although
the 2nd generation catalyst 3 displays activities comparable
to early transition metal complexes [4,12], it also promotes
double bond isomerization at elevated temperatures, com-
petetively with metathesis, which can be problematic for
the synthesis of precise polymer microstructures [13].
Part of our research effort has recently focused on build-
ing a family of a-olefin/ethylene copolymers to better
understand the structure–properties relationship of widely
commercialized polyethylene materials [14]. Using acyclic
diene metathesis (ADMET), we have synthesized ethyl-
ene–propylene copolymers [EP(n + 1)] with exact ethylene
run length of n = 4, 6, 8, 14, 18 and 20 carbons, the methyl
branch content being determined during the monomer
Olefin metathesis has been applied to a variety of organic
synthetic challenges by allowing the simple formation of
carbon–carbon double bonds in a single step where other
routes would sometimes require several tedious steps [1].
The development of well-defined metathesis carbene cata-
lysts, such as SchrockÕs molybdenum catalyst 1 [2], and
the more recent GrubbsÕ ruthenium catalysts 2 [3] and 3
[4], has considerably widened the scope of olefin metathesis
in both organic and polymer chemistry (Fig. 1) [5]. This
success has inspired many catalyst modifications to intro-
duce heterogeneous catalysts [6], water-soluble catalysts
[7], recyclable catalysts [8], slower and faster initiators [9],
so as to accommodate any particular set of conditions.
Despite the broad range of applications of these cata-
lysts, challenges remain that must be addressed. On one
*
Corresponding author. Fax: +1 352 392 9741.
E-mail address: wagener@chem.ufl.edu (K.B. Wagener).
0022-328X/$ - see front matter Ó 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2005.09.030

586
F.C. Courchay et al. / Journal of Organometallic Chemistry 691 (2006) 585–594
methyl branches on every third carbon led to little if any
conversion (Fig. 3). Color changes during the reaction sug-
gested catalyst decomposition very early in the reaction;
olefin conversion was calculated at 4% for catalyst 1 and
8% for catalyst 2 at best. These results warranted a detailed
NMR study with catalysts 1, 2, and 3, probing for key
intermediates during the catalytic cycle. To facilitate inter-
pretation and broaden the scope of our study, the polymer-
ization of 11 was examined more carefully along with the
cross-metathesis reaction of 3-methyl-1-pentene, research
which had been previously reported by Grubbs with cata-
lyst 2 [16].
Fig. 1. Olefin metathesis catalysts.
design [15]. Although much progress has been made, we
have been unable to polymerize the EP3 diene monomer
(11 in Fig. 2), since allylic methyl groups seem to pose
problem.
3. Schrock molybdenum catalyst
Consequently, we decided to investigate the influence of
allylic methyls in condensation metathesis chemistry using
standard catalysts 1, 2, and 3, in both cross-metathesis
(CM) and ADMET conditions. Herein, we report an
NMR spectroscopy study highlighting mechanistic features
of both molybdenum and ruthenium metathesis catalysis in
this reaction, and the importance of steric arrangement in
the metallacyclobutane intermediate.
The experiments described below demonstrate that the
dominant reaction pathway when using catalyst 1 leads
to an accumulation of the non-productive metallacyclobu-
tane intermediate, rather than full metathesis conversion.
The experiments were set up using conditions that would
allow monitoring dynamic catalytic behavior by NMR
spectroscopy. For example, when examining the cross-
reaction of 3-methyl-1-pentene (the ‘‘control’’ reaction for
ADMET chemistry of monomer 11), a typical experiment
consisted of loading an NMR tube with a 0.2 M solution
of olefin in C6D6 and 30 mol% of complex 1 [17].
2. Results
ADMET monomer 11, the target monomer possessing
allylic groups, was synthesized starting with 3-methyl-4-
pentenoic acid which was reduced and tosylated to yield
diester 6 by addition to diethyl malonate (Fig. 2). Saponi-
fication and decarboxylation produced acid 8, which then
was reduced and tosylated. Substitution with LAH affor-
ded monomer 11 in 29% overall yield.
After 30 min at room temperature, the proton NMR
spectrum indicates the presence of the expected metathesis
product 3,6-dimethyl-oct-4-ene (A), and products of the
catalyst initiation (1,1-dimethyl-allyl)-benzene (B), and
(1,1,4-trimethyl-hex-2-enyl)-benzene (C) (Fig. 4). However,
product A only formed in 46% yield, which is insufficient
for ADMET chemistry, again illlustrating the difficulty
associated with the homodimerization of this class of ole-
fins. Products B and C, formed by reaction of one molecule
Efforts to polymerize 11, 3,6,9-trimethylundeca-1,
10-diene, to create unsaturated polyethylene containing
Fig. 2. Synthesis of monomer 11.

F.C. Courchay et al. / Journal of Organometallic Chemistry 691 (2006) 585–594
587
Fig. 3. ADMET reaction of 11.
Fig. 4. Cross-metathesis of 3-methyl-1-pentene catalyzed by 1.
of substrate with the original molybdenum catalyst (and
therefore present in catalytic quantities), normally are not
observed. They are visible here because of the use of high
catalyst concentrations.
Fig. 5 depicts the catalytic cycle involved in the forma-
tion of each products presented in Fig. 4. For clarity, note
that single arrows are drawn instead of equilibrium arrows,
but according to the metathesis mechanism it is implied
that each step actually exists as an equilibrium. Further,
it should be noted that steps 30, 7 and 8 do not represent
equilibrium conditions, but rather the non-productive
metathesis cycle.
As the reaction starts, the first feature we observe is the
major production of CH2@C(CH3)2Ph (B) compared to
the alternate product C, which indicates that formation
of the 2,4-metallacyclobutane is largely favored when 3-
methyl-1-pentene coordinates to the original SchrockÕs
alkylidene (Step 1 in Fig. 5). The two doublets at d 13.18
and 13.23 ppm (2JH,H = 5.6 Hz) confirm the presence of
the new alkylidene Mo@CHCH(CH3)Et, each signal corre-
sponding to one enantiomer of the racemic mixture
(Fig. 6(a)) [2a]. Two metallacyclobutanes can form from this
intermediate, the 2,3-addition leading to productive metath-
esis (step 3), and the 2,4-addition leading to non-productive
Fig. 5. Catalytic cycle for the metathesis of 3-methyl-1-pentene with a molybdenum carbene catalyst.

588
F.C. Courchay et al. / Journal of Organometallic Chemistry 691 (2006) 585–594
Fig. 6. 1H NMR alkylidene region for the reaction of 3-methyl-1-pentene with catalyst (a) 1; (b) 2; (c) 3.
metathesis (step 30). After 10 min, two broad singlets ap-
pear in equal ratios at ꢀ0.17 and ꢀ1.06 ppm, characteristic
of the two b protons of a trigonal–bipyramidal metallacyc-
lobutane ring, therefore implying the formation of the
2,4-metallacyclobutane, further confirmed by the two a pro-
tons singlets at 5.10 and 4.84 ppm [2a]. The 2,3-metallacycle
would have given rise to a single signal in the upper field
region. The fact that it is undetectable by NMR may be
simply a result of its instability, forcing the metallacycle
intermediate to decompose quickly into the methylidene
and the metathesis product (step 4), as proven by the inter-
nal olefin resonance at 5.32 ppm. Because the methylidene
complex is highly unstable, we cannot unambiguously as-
sign its signals. Moreover, the few methylidene complexes
that were ever observable by NMR were adducts stabilized
by a polar solvent such as dme or THF, which is not pos-
sible here [2c,18]. The most common decomposition routes
for high oxidation state (‘‘d0’’) alkylidene complexes in the
presence of olefins are rearrangement of metallacyclobu-
tanes complexes by b-hydride mechanism and bimolecular
coupling, methylidenes being the most susceptible to bimo-
lecular decomposition [18]. However, the broad singlet at d
12.36 ppm, assigned to the NH proton of an amido alkyli-
dyne complex, indicates yet another substantial decomposi-
tion pathway. This kind of proton transfer reaction from
carbon to nitrogen was observed by Shrock et al. [19] dur-
ing the preparation of Mo(NArCl)(CHCMe2Ph)[Biphen].
In any case, the metathesis reaction only reaches 46% con-
version after 30 minutes, while the 2,4-metallacyclobutane
already constitutes 50% of the catalyst mixture, which
highlights the significant extent of non-productive
metathesis.
The product distribution is similar when 3-methyl-1-
pentene is substituted with monomer 11, although AD-
MET conversions are slightly lower probably because of
the substrateÕs longer carbon chain.
4. GrubbsÕ ruthenium catalysts
The NMR experiments with 3-methyl-1-pentene (CM
conditions) and monomer 11 (ADMET conditions) were
repeated using ruthenium catalysts 2 and 3. Unlike what
is observed in the case of early transition metals, the ruthe-
nium metallacyclobutane is highly unstable, and it is still
debated whether it is an intermediate or a transition state
in the metathesis catalytic cycle [20]. As had been done be-
fore, an NMR tube was loaded with a 0.2 M solution of 3-
methyl-1-pentene in C6D6 and 30 mol% of complex 2.

F.C. Courchay et al. / Journal of Organometallic Chemistry 691 (2006) 585–594
589
After 30 min at 45 °C, the proton NMR spectrum indi-
cates the presence of the metathesis product 3,6-dimethyl-
oct-4-ene (A) in low yields and products of the catalyst
initiation: styrene (D), and (3-methyl-pent-1-enyl)-benzene
(E) that are formed by reaction of one molecule of sub-
strate with the ruthenium benzylidene (Fig. 7). Product A
is formed in 5% and 30% yields with catalyst 2 and 3,
respectively, again too low to be useful in ADMET step-
growth polymerization. As before, products D and E are
only visible because of the high catalyst concentrations.
Fig. 8 depicts the catalytic cycle involved in each prod-
uct formation. Downfield in the NMR spectrum, we ob-
serve the steady disappearance of ruthenium benzylidene
while the methylidene resonance at d 19.40 ppm appears al-
most at a comparable rate. At the end of the experiment,
three times more Ru@CH2 (15%) is present than the ex-
pected Ru@CHCH(CH3)Et (5% of the catalyst mixture),
evidenced by a doublet at d 19.28 ppm (Fig. 6(b)).
utane from the ruthenium benzylidene (steps 1 and 10 in
Fig. 8). Theoretically, the amount of styrene released
corresponds to the amount of catalyst activated for pro-
ductive metathesis, Ru@CHCH(CH3)Et. Since this alky-
lidene is not detected by NMR, it must react quickly
with a substrate molecule through the intermediate of a
2,3-metallacyclobutane to afford the metathesis product
A and the methylidene (steps 3 and 4). This analysis is fur-
ther supported by the fact that the amount of A equals the
amount of styrene generated. At this point, the methyli-
dene continues to increase in concentration, while the
metathesis reaction seems to cease, indicating that upon
coordination of a new olefin molecule, the 2,3-metallacyc-
lobutane is formed preferentially to regenerate the methyl-
idene through step 7. Step 10 and 20 could also generate
the methylidene, but this would involve a build-up of
product E, which we do not observe. Further, the satura-
tion of the reaction vessel with ethylene cannot be held en-
tirely responsible for the methylidene build-up either, since
little ethylene is present in solution compared to the excess
of starting material.
As observed with the molybdenum catalyst, styrene is
produced in much larger quantities than product E, which
indicates the preferred formation of the 2,4-metallacyclob-
Fig. 7. CM of 3-methyl-1-pentene catalyzed by 2 or 3.
Fig. 8. Catalytic cycle for the metathesis of 3-methyl-1-pentene with ruthenium carbene catalyst 2 or 3.

590
F.C. Courchay et al. / Journal of Organometallic Chemistry 691 (2006) 585–594
In any case, after 30 min, only 5% of starting material
tane is preferred during step 3. However, the amount of
metathesis product formed far exceeds the amount of sty-
rene detected, which implies that productive metathesis
does occur, unlike what was observed with catalyst 2. This
observation is consistent with the higher reactivity of
NHC-containing complexes towards olefinic substrates,
attributed to the stabilizing effect of the NHC ligand on
the metallacyclobutane [20b]. The ligand sphere may also
allow a better arrangement of the alkyl bulk during the for-
mation of the 2,4-metallacyclobutane.
As seen with 1st generation complex 2, the reaction pro-
file of catalyst 3 with the ADMET monomer 11 is identical
to the CM reaction, but affords much lower conversions.
Under the same conditions as CM chemistry, 70% of start-
ing material remains even though more catalyst has been
activated (46%). The amount of methylidene formed is also
higher than during the CM reaction, which suggests that
non-productive metathesis occurs to a larger extent
through the preferred formation of a 2,3-metallacyclobu-
tane. This favored conformation is likely a direct result
of the additional sterics brought about by the larger diene
monomer. In addition, four hydride complexes have
formed during the course of the reaction indicating the
start of the catalytic decomposition process, accelerated
by the accumulation of the unstable methylidene complex.
has reacted even though more than 20% of the benzylidene
precatalyst has been consumed, mostly forming the highly
unstable ruthenium methylidene. Apparently, decomposi-
tion of the catalyst system occurs at a faster rate than the
metathesis reaction.
Hence, complex 2 cannot be regarded as a suitable cat-
alyst for the homodimerization of 3-methyl-1-pentene, and
probably not for any other olefin containing allylic methyls
[16]. Indeed, when the reaction was repeated with the cor-
responding ADMET diene, we observed similar patterns;
in fact, conversions were even lower. Metathesis products
A and E were undetectable, and half less styrene was pro-
duced compared to the CM experiment, implying that less
metallacyclobutane had formed. After 30 min, the methyl-
idene constitutes 13% of the catalyst mixture and seems to
be the only species besides the original benzylidene. These
data indicate that with the ADMET monomer the rate of
formation of any metallacyclobutane is slower, more likely
because of the larger olefin.
Catalyst 3 was subjected to scrutiny under the same con-
ditions. At room temperature, approximately the same
amounts of styrene and E are detected at room tempera-
ture, while at 45 °C styrene seems to be in slight excess.
Phosphine dissociation is so slow at low temperatures that
an equilibrium between the 2,3- and 2,4-metallacyclobu-
tane is created. When the dissociation event becomes faster,
a greater amount of active catalyst is generated and the rate
difference between formation of the 2,3- or 2,4-metallacyc-
lobutane becomes evident because of the higher catalyst
concentration.
After 30 min the methylidene represents about 30% of
the catalyst mixture whereas Ru@CHCH(CH3)Et is only
present in 1%, appearing as a doublet at d 17.48 ppm
(Fig. 6(c)). A quadruplet at 19.0 ppm also indicates the
presence of Ru@CHCH3 formed from the metathesis of
isomerized starting material. Indeed, ruthenium catalysts,
particularly those containing N-heterocyclic carbene li-
gands (NHC), can isomerize olefins by migration of the
double bond along the backbone [13b,13c,13d]. The
appearance of a multiplet at d 5.19 ppm confirms the
presence of 3-methyl-2-pentene by structural isomeriza-
tion of the starting material. Traces of 2-methyl-1-butene
also become visible at d 4.67 ppm, formed by reaction of
3-methyl-2-pentene with the ruthenium methylidene. Nev-
ertheless, the expected metathesis product A, 3,6-di-
methyl-oct-4-ene, constitutes only 30% of the olefinic
mixture, while 50% of starting material still remains. This
level of conversion obviates any ADMET chemistry. The
reaction does proceed, albeit at slow rates. Noteworthy,
34% of the original precatalyst 3 has already been con-
sumed at this point, which is equivalent to a 10 mol% cat-
alyst loadings, already surpassing the typical catalytic
quantities.
5. Discussion
Although the conditions used in this study do not reflect
exact polymerization conditions (where the metathesis
equilibrium is driven by removal of ethylene) these experi-
ments reflect a general catalytic behavior, since they dem-
onstrate a steric conflict rather than a rate problem.
In the case of the SchrockÕs molybdenum catalyst, 3-
methyl-1-pentene (or its diene analog, monomer 11) seem
to consistently coordinate to the metal in order to minimize
steric interactions between the alkyl branches, hence form-
ing the 2,4-metallacyclobutane, which, during step 30
(Fig. 5), only leads to non-productive metathesis. More-
over, this somewhat stable intermediate traps the catalyst
in an inactive form. Even though productive metathesis oc-
curs to some extent, it is insufficient to promote ADMET
polymerization, a step-growth process that requires high
conversions (>99%) in order to reach high molecular weigh
polymer.
A different mechanism operates with ruthenium cata-
lysts to prevent the metathesis of allylic methyl-containing
substrates. The ruthenium methylidene accumulates when
catalyst 2 is used with either 3-methyl-1-pentene or the cor-
responding ADMET momomer, through the consecutive
steps 1–4 in Fig. 8. The original benzylidene first forms a
2,4-metallacyclobutane upon olefin coordination to create
a new alkylidene complex (steps 1 and 2 in Fig. 8), which
then forms the 2,3-metallacyclobutane by reacting with an-
other olefin (steps 3 and 4) [21]. From this point on, the
methylidene complex only reacts with incoming olefins to
Since Ru@CHCH(CH3)Et is not detected in significant
amounts, we can still assume that the 2,3-metallacyclobu-

F.C. Courchay et al. / Journal of Organometallic Chemistry 691 (2006) 585–594
591
promote non-productive metathesis and regenerate itself
7. Experimental
(step 7), even though formation of the methylidene is not
kinetically favored [16].
7.1. General
It has been suggested that the steric effects orient the
bulk away from the crowded ruthenium metal center, while
the electronic effects favor alkyl substituents adjacent to the
metal [16]. Consequently, when the olefin reacts with the
benzylidene complex, the benzene ring adjacent to ruthe-
nium may be sufficiently electron-withdrawing to place
the alkyl group next to the metal center, minimizing in
the same time the steric repulsions between the alkyl
branch and the phenyl group (step 1). On the other hand,
when the methylidene complex is involved, no electronic ef-
fect directs the coordination of the olefin, which positions
its steric bulk away from the metal center to form the
2,3-ruthenacyclobutane (step 7). The fact that the same
reaction with linear olefins affords the inverse conforma-
tion [16] reveals the directing effect of allylic methyls, more
likely by steric interactions with the ligand sphere of the
complex.
These conclusions also apply to catalyst 3, although the
steric directing effects seem to be diminished in the presence
of the NHC ligand. This could be due to a different
arrangement of the ligand sphere reducing steric interac-
tions, or to the propensity of the NHC ligand to stabilize
the ruthenacyclobutane. Despite better yields, we could
not achieve the conditions required by ADMET polycon-
densation. In addition, we observed a significant amount
of olefin isomerization interfering with the metathesis pro-
cess that render this 2nd generation of catalysts unsuitable
for the modeling of precise molecules.
1H NMR (300 MHz) and 13C NMR (75 Hz) spectra
were recorded on either a Mercury series or Varian
VXR-300 NMR superconducting spectrometer for small
molecule structure determination. Chemical shifts were ref-
1
erenced to residual C6H6 (7.15 for H and 128.39 for 13C)
1
or CHCl3 (7.27 for H and 77.23 for 13C), and the NMR
solvents were distilled, degassed by three freeze–pump–
thaw cycles, and stored in an argon-filled drybox prior to
use. Thin layer chromatography (TLC) was performed on
EMD silica gel coated (250 lm thickness) glass plates cut
to custom sizes. Developed TLC plates were stained with
iodine absorbed on silica to produce a visible signature.
Reaction conversions and relative purity of crude products
were monitored by TLC chromatography and NMR.
High-resolution mass spectral (HRMS) data were obtained
on a Finnegan 4500 gas chromatograph/mass spectrometer
using either the chemical ionization (CI) or electrospray
ionization (ESI) mode.
7.2. Materials and methods
All materials were purchased from Aldrich chemical and
used as received unless otherwise specified. Complex 1
[2a,2b,2c], and 3 [3a] were synthesized according to litera-
ture procedure. Complex 2 was a gift from Materia Inc.,
and was used as received. Catalysts were stored in an ar-
gon-filled drybox prior to use.
6. Conclusion
7.3. Bulk polymerization experiment
ADMET polymerization of dienes possessing allylic
methyl groups is not possible, principally due to interaction
of this methyl group with the metathesis catalyst. A series
of NMR spectroscopy experiments conducted with both
the diene monomer (11) and model compound 3-methyl-
1-pentene, the corresponding mono-olefin, demonstrated
that the reaction limitations depend on catalyst selection.
With SchrockÕs molybdenum catalyst 1, the reaction led
to an accumulation of metallacyclobutane, trapping the
catalyst into an inactive form. With GrubbsÕ ruthenium
catalysts 2 and 3, the substrate coordinates to the metal
center only to yield non-productive metathesis, which re-
sults in a build-up of the methylidene complex, more
prompt to decomposition. Although the NHC-containing
complex 3 affords better yields than its phosphine analog,
our experiments also illustrate the competitive nature of
double-bond isomerization during olefin metathesis. These
results are dictated by the steric arrangement of the sub-
strate within the catalyst ligand sphere during the metalla-
cyclobutane formation, the alkyl branch being adjacent to
the metal center for the molybdenum 1, and opposite to it
in the case of ruthenium catalysts 2 and 3.
Monomer 4 (1.5 g, 77 mmol) was vacuum transferred
from potassium mirror into a schlenk flask and taken into
an argon-filled glove box. The catalyst (monomer:catalyst
ratio is 1500:1 for 1, 450:1 for 2) was added to the mono-
mer in a 50 ml round-bottom flask equipped with a mag-
netic stirbar, and allowed to react approximately 15 min
before sealing the reactor with a schlenk adaptor and con-
necting to high vacuum line. Vacuum was applied intermit-
tently for the first 2 h, then polymerization was heated at
40 °C and put under full vacuum (10ꢀ3 Torr) for 4 days.
NMR samples were taken in d-chloroform directly from
the reactor with no purification. Conversions were calcu-
lated by the ratio of integral values of the internal olefin
(d 5.20 ppm, m, 2H) to the terminal olefin (d 4.95 ppm,
m, 2H).
7.4. NMR catalyst experiments
NMR spectra were recorded on a Varian Inova spec-
trometer equipped with a 5 mm indirect detection probe,
1
operating at 500 MHz for H and at 125 MHz for 13C.
The solvent was d6-benzene and the temperature was

592
F.C. Courchay et al. / Journal of Organometallic Chemistry 691 (2006) 585–594
25 °C for catalyst 1 and 45 °C for catalysts 2 and 3. Chem-
ical shifts were referenced to residual C6H6 (7.15 for 1H and
128.39 for 13C). In the drybox, 0.036 mmol of catalyst was
introduced in an NMR tube equipped with a Teflon valve.
Approximately 0.8 ml of C6D6 was added to the tube so as
not to dissolve the catalyst extensively, and 0.119 mmol of
substrate (3-methyl-1-pentene or monomer 11) was care-
fully layered on top of the mixture. The NMR tube was
sealed and shaken right before being introduced in the
spectrometer. Proton spectra were recorded every 5 min,
and quantitation was obtained by integration of the appro-
priate peaks against the solvent peak, which served as inter-
nal standard.
stirred cold for 1 h, then warmed to room temperature
for 18 h. The reaction was quenched by addition of water
(250 mL), extracted with diethyl ether (300 mL), and dried
with brine. Column chromatography (10% diethyl ether in
1
hexane) afforded 18.9 g of colorless oil. 59.5% yield. H
NMR (300 MHz, CDCl3): d (ppm) 0.95 (d, 6H), 1.14 (q,
4H), 1.24 (t, 6H), 1.85 (m, 4H), 2.08 (m, 2H), 4.17 (q,
4H), 4.96 (m, 4H), 5.64 (m, 2H). 13C NMR (300 MHz,
CDCl3): d (ppm) 14.33, 20.35, 29.96, 30.87, 38.10, 57.51,
61.16, 113.35, 144.11, 172.06. Elemental Anal. Calc. for
C19H32O4: C, 70.33; H, 9.94. Found: C, 69.93; H, 9.54%.
7.5.4. 2,2-Bis(3-methyl-4-pentenyl)malonic acid (7)
Potassium hydroxide (26.3 g, 470 mmol) was added to a
solution of 6 (18.9 g, 58 mmol) in ethanol (100 mL) and
water (20 mL) which was then brought to reflux for 3 h.
The reaction was cooled, quenched with water (150 mL)
and conc. HCl (until pH 3), and washed twice with diethyl
ether (200 mL). The ether phase was dried with brine and
concentrated to 15.0 g of a white solid with no further puri-
fication necessary. 95.7% yield. 1H NMR (300 MHz,
CDCl3): d (ppm) 1.01 (d, 6H), 1.25 (m, 4H), 1.83 (m,
4H), 2.11 (m, 2H), 4.98 (m, 4H), 5.65 (m, 2H), 11.18 (br,
2H). 13C NMR (300 MHz, CDCl3): d (ppm) 20.40, 31.18,
32.09, 38.14, 57.85, 113.73, 143.75, 177.96. CI/HRMS:
[M + H]+ Cacl. for C15H25O4: 269.1753. Found:
269.1749. Elemental Anal. Calc. for C15H25O4: C, 67.14;
H, 9.01. Found: C, 66.98; H, 8.89%.
7.5. Synthesis of EP3 monomer (4–11)
7.5.1. 3-Methyl-4-pentene-1-ol (4)
Under an argon atmosphere a solution of 3-methyl-4-
pentenoic acid (30.0 g, 263 mmol) in diethyl ether
(100 mL) was added dropwise to a suspension of LAH
(12.0 g, 342 mmol) in diethyl ether (350 mL) over 45 min
at 0 °C. When the addition was complete, the slurry was
stirred cold for 15 min, then the reaction was warmed to
room temperature for 2 h. The reaction was then cooled
to 0 °C and quenched by addition of water (200 mL) and
concentrated HCl (ꢁ30 mL) until the aqueous layer was
pH 3. The organic layer was collected, combined with a
second ether wash (200 mL), and dried with brine and
MgSO4. Filtration, followed by distillation of the filtrate
yielded a 21.5 g of colorless oil. 81.6% yield. b.p. =
7.5.5. 5-Methyl-2-(3-methyl-4-pentenyl)-6-heptenoic acid
(8)
1
149 °C, 760 mmHg. H NMR matched reported spectral
data [22]. 13C NMR (300 MHz, CDCl3): d (ppm) 20.61,
Decalinä (15 mL, 1:1 wt%) was added to 7 (14.5 g,
54 mmol) and heated to 185 °C in a 250 mL round bottom
flask equipped with an air cooled condenser under nitro-
gen. Production of CO2 was monitored with a mineral oil
bubbler, and the reaction was stirred vigorously until gas
evolution ceased after about 30 min. Upon cooling, Deca-
linä was removed via rotary evaporation affording crude
acid. Column chromatography (20% ethyl acetate in hex-
35.10, 39.48, 61.41, 113.32, 144.45.
7.5.2. 3-Ethenyl-1-butanol tosylate (5)
To a stirred solution of 4 (21.45 g, 214 mmol) in pyridine
(100 mL) at 0 °C was added tosyl chloride (53.06 g,
278 mmol). The suspension was stirred cold for 10 min,
then the viscous slurry was warmed to room temperature
for 2 h. Addition of water (200 mL) and diethyl ether
(200 mL) produced a biphasic mixture. The organic phase
was isolated and combined with a second diethyl ether
wash. Washing twice with 1 N HCl (200 mL), followed
by drying over MgSO4, and column chromatography
(15% diethyl ether in hexane) afforded 53.6 g of colorless
1
ane) afforded 11.4 g colorless oil. 94.0% yield. H NMR
(300 MHz, CDCl3): d (ppm) 1.01 (d, 6H), 1.33 (m, 4H),
1.50 (m, 2H), 1.61 (m, 2H), 2.12 (m, 2H), 2.31 (m, 1H),
4.96 (m, 4H), 5.67 (m, 2H). 13C NMR (300 MHz, CDCl3):
d (ppm) 20.34, 20.51, 29.96, 30.06, 34.25, 34.36, 37.96,
38.08, 45.70, 45.86, 46.05, 113.13, 113.17, 144.38, 144.45,
183.01. CI/HRMS: [M + H]+ Cacl. for C14H25O2:
225.1855. Found: 225.1845. Elemental Anal. Calc. for
C14H25O2: C, 74.95; H, 10.78. Found: C, 74.89; H, 10.68%.
1
oil. 98.3% yield. H NMR (300 MHz, CDCl3): d (ppm)
0.96 (d, 3H), 1.64 (m, 2H), 2.24 (m, 1H), 2.46 (s, 3H),
4.04 (m, 2H), 4.90 (m, 2H), 5.55 (m, 1H), 7.35 (d, 2H),
7.79 (d, 2H). 13C NMR (300 MHz, CDCl3): d (ppm)
20.27, 21.84, 34.34, 35.40, 69.01, 114.31, 128.10, 130.00,
133.40, 142.70, 144.86.
7.5.6. 5-Methyl-2-(3-methyl-4-pentenyl) hept-6-en-1-ol (9)
Same procedure as for 4 using 8 as starting material. 8.6 g
of colorless oil was isolated with no further purification
1
7.5.3. Diethyl 2,2-bis(3-methyl-4-pentenyl) malonate (6)
A solution of diethyl malonate (15.7 g, 98 mmol) in
THF (100 mL) was added to a stirred solution of 5
(53.6 g, 210 mmol) and sodium hydride (5.6 g, 233 mmol)
in THF (100 mL) over 30 min at 0 °C. The mixture was
needed. 84% yield. H NMR (300 MHz, CDCl3): d (ppm)
1.00 (d, 6H), 1.20–1.41 (br, 9H), 1.66 (br, 1H), 2.09 (m,
2H), 3.53 (d, 2H), 4.95 (m, 4H), 5.69 (m, 2H). 13C NMR
(300 MHz, CDCl3): d (ppm) 20.44, 20.47, 28.55, 28.58,
28.65, 33.97, 38.35, 40.94, 40.99, 65.74, 65.78, 112.78,

F.C. Courchay et al. / Journal of Organometallic Chemistry 691 (2006) 585–594
593
144.92, 144.94. CI/HRMS: [M + H]+ Cacl. for C14H27O:
211.2062. Found: 211.2062. Elemental Anal. Calc. for
C14H26O: C, 79.94; H, 12.46. Found: C, 79.71; H, 12.31%.
[5] (a) For recent reviews, see: T.M. Trnka, R.H. Grubbs, Acc. Chem.
Res. 34 (2001) 18–29;
(b) A. Furstner, Angew. Chem., Int. Edit. 39 (2000) 3012–3043;
¨
(c) R.H. Grubbs, S. Chang, Tetrahedron 54 (1998) 4413–4450;
(d) M. Schuster, S. Blechert, Angew. Chem., Int. Edit. 36 (1997)
2036–2056;
7.5.7. 5-Methyl-2-(3-methyl-4-pentenyl)-6-heptenyl tosylate
(10)
(e) S.K. Armstrong, J. Chem. Soc., Perk. Trans. 1 (1998) 371–388;
(f) U. Frenzel, O. Nuyken, J. Polym. Sci. Chem. 40 (2002) 2895–2916;
(g) M.R. Buchmeiser, Chem. Rev. 100 (2000) 1565–1604;
(h) D.L. Wright, Curr. Org. Chem. 3 (1999) 211–240;
(i) M.L. Randall, M.L. Snapper, J. Mol. Catal. A 133 (1998) 29–40;
(j) K.J. Ivin, J. Mol. Catal. A 133 (1998) 1–16;
(k) A.F. Noels, A. Demonceau, J. Phys. Org. Chem. 11 (1998) 602–
609.
Same procedure for 5 with 9 as starting material, 103%
crude yield after concentration. No further purification
performed. 1H NMR (300 MHz, CDCl3): d (ppm) 0.93
(d, 6H), 1.15 (m, 4H), 1.22 (m, 4H), 1.56 (br, 1H), 1.99
(m, 2H), 2.45 (s, 3H), 3.90 (d, 2H), 4.91 (m, 4H), 5.60
(m, 2H), 7.34 (d, 2H), 7.78 (d, 2H). 13C NMR (300 MHz,
CDCl3): d (ppm) 20.31, 20.39, 21.79, 28.29, 28.36, 28.40,
33.38, 33.41, 33.49, 37.97, 38.08, 38.10, 72.81, 112.94,
112.97, 128.11, 129.97, 133.25, 144.81, 144.43, 144.46. CI/
HRMS: [M + H]+ Cacl. for C21H33O3S: 365.2150. Found:
365.2143. Elemental Anal. Calc. for C21H33O3S: C, 69.19;
H, 8.85. Found: C, 68.93; H, 8.58%.
[6] (a) P. Nieczypor, W. Buchowicz, W.J.N. Meester, F.P.J.T. Rutjes,
J.C. Mol, Tetrahedron Lett. 42 (2001) 7103–7105;
(b) L. Jafarpour, S.P. Nolan, Org. Lett. 25 (2000) 4075–4078;
(c) M. Ahmed, A.G.M. Barrett, D.C. Braddock, S.M. Cramp, P.A.
Procopiou, Tetrahedron Lett. 40 (1999) 8657–8662;
(d) A.G.M. Barrett, S.M. Cramp, R.S. Roberts, Org. Lett. 7 (1999)
1083–1086;
(e) S.T. Nguyen, R.H. Grubbs, J. Organomet. Chem. 497 (1995) 195–
200;
7.5.8. 3,6,9-Trimethylundeca-1,10-diene (11)
Same procedure as for 9 with 10 as starting material.
Isolated 2.5 g of a colorless oil after column chromatogra-
(f) Q. Yao, A.R. Motta, Tetrahedron Lett. 45 (2004) 2447–2451;
(g) M.T. Zarka, O. Nuyken, R. Weberskirch, Macromol. Rapid
Commun. 25 (2004) 858–862;
(h) S.J. Connon, A.M. Dunne, S. Blechert, Angew. Chem., Int. Edit.
41 (2002) 3835–3838;
1
phy (hexane). 81.1% yield. H NMR (300 MHz, C6D6): d
(ppm) 0.87 (d, 3H), 0.98 (d, 6H), 1.12 (m, 2H), 1.30 (7H),
2.03 (m, 2H), 4.97 (m, 4H), 5.68 (2H). 13C NMR
(300 MHz, C6D6): d (ppm) 19.85, 19.92, 19.98, 20.40,
20.64, 33.34, 33.42, 33.54, 34.48, 34.89, 34.92, 34.99,
35.01, 38.54, 112.65, 112.75, 144.96, 145.07. CI/HRMS:
[M + H]+ Cacl. for C14H26O: 194.2035. Found: 194.2037.
Elemental Anal. Calc. for C14H26O: C, 86.52; H, 13.48.
Found: C, 86.32; H, 13.51%.
(i) K. Grela, M. Tryznowski, M. Bienik, Tetrahedron Lett. 43 (2002)
9055–9059;
(j) J.S. Kingsbury, S.B. Garber, J.M. Goftos, B.L. Gray, M.M.
Okamoto, M.A. Farrer, J.T. Fourkas, A.H. Hoveyda, Angew.
Chem., Int. Edit. 40 (2001) 4251–4256;
(k) J.O. Krause, O. Nuyken, K. Wurst, M.R. Buchmeiser, Chem.
Eur. J. 10 (2004) 777–784;
(l) S. Gessler, S. Randl, S. Blechert, Tetrahedron Lett. 41 (2000)
9977–9980.
[7] (a) J.P. Gallivan, J.P. Jordan, R.H. Grubbs, Tetrahedron Lett. 46
(2005) 2577–2580;
Acknowledgments
(b) M.T. Zarka, O. Nuyken, R. Weberskirch, Macromol. Rapid
Commun. 25 (2004) 858–862;
We thank the National Science Foundation (NSF), the
Army Research Office (ARO), and the Florida Space
Grant Consortium (NASA-FSGC) for generous funding
of this work. We also thank Materia, Inc. for providing
catalyst 2.
(c) D.M. Lynn, B. Mohr, R.H. Grubbs, L.M. Henling, M.W. Day, J.
Am. Chem. Soc. 122 (2000) 6601–6609;
(d) M. Saoud, A. Romerosa, M. Peruzzini, Organometallics 19
(2000) 4005–4007;
(e) T.A. Kirkland, D.M. Lynn, R.H. Grubbs, J. Org. Chem. 63
(1998) 9904–9909;
(f) B. Mohr, D.M. Lynn, R.H. Grubbs, Organometallics 15 (1996)
4317–4325.
References
[8] (a) For a review on Ru bearing bidentate carbenes: A.H. Hoveyda,
D.G. Gillingham, J.J. Van Veldhuizen, O. Kataoka, S.B. Garber,
J.S. Kingsbury, J.P.A. Harrity, Org. Biomol. Chem. 2 (2004) 8–23;
(b) S.B. Garber, J.S. Kingsbury, B.L. Gray, A.H. Hoveyda, J. Am.
Chem. Soc. 122 (2000) 8168–8179;
(c) J.S. Kingsbury, J.P.A. Harrity, P.J. Bonitatebus, A.H. Hoveyda,
J. Am. Chem. Soc. 121 (1999) 791–799.
[9] (a) J.A. Love, M.S. Sanford, M.W. Day, R.H. Grubbs, J. Am.
Chem. Soc. 125 (2003) 10103–10109;
[1] For olefin metathesis reviews, see: (a) R.H. Grubbs, Handbook of
Metathesis, Wiley–VCH: Weinheim, Germany, 2003;
(b) K.J. Ivin, J.C. Mol, Olefin Metathesis and Metathesis Polymer-
ization, Academic Press, London, 1997.
[2] (a) H.H. Fox, J.-K. Lee, L.Y. Park, R.R. Schrock, Organometallics
12 (1993) 759–768;
(b) R.R. Schrock, J.S. Murdzek, G.C. Bazan, M. Robbins, M.
Dimare, M. OÕRegan, J. Am. Chem. Soc. 112 (1990) 3875–3886;
(c) R.R. Schrock, Acc. Chem. Res. 23 (1990) 158–165.
[3] (a) M. Ulman, R.H. Grubbs, J. Org. Chem. 64 (1999) 7202–7207;
(b) M. Ulman, R.H. Grubbs, Organometallics 17 (1998) 2484–2489;
(c) E.L. Dias, S.T. Nguyen, R.H. Grubbs, J. Am. Chem. Soc. 119
(1997) 3887–3897;
(b) J.A. Love, J.P. Morgan, T.M. Trnka, R.H. Grubbs, Angew.
Chem., Int. Ed. 41 (2002) 4035–4037;
(c) M.S. Sanford, J.A. Love, R.H. Grubbs, J. Am. Chem. Soc. 123
(2001) 6543–6554.
[10] (a) For a survey on the functional group tolerance of various
catalysts, see [5e];
(b) A.K. Chatterjee, T.-L.C. Choi, D.P. Sanders, R.H. Grubbs, J.
Am. Chem. Soc. 125 (2003) 11360–11370.
[11] T.A. Kirkland, R.H. Grubbs, J. Org. Chem. 62 (1997) 7310–7318.
(d) P. Schwab, R.H. Grubbs, J. Ziller, J. Am. Chem. Soc. 118 (1996)
100–110.
[4] M. Scholl, S. Ding, C. Lee, R.H. Grubbs, Org. Lett. 1 (1999) 953–
956.

594
F.C. Courchay et al. / Journal of Organometallic Chemistry 691 (2006) 585–594
[12] (a) A.K. Chatterjee, R.H. Grubbs, Org. Lett. 1 (1999) 1751–1753;
(b) C.W. Bielawski, R.H. Grubbs, Angew. Chem., Int. Ed. 39 (2000)
2903–2906.
[13] For a short review on isomerization behavior of ruthenium carbene
see: (a) B. Schmidt, Eur. J. Org. Chem. 9 (2004) 1865–1880;
(b) S.E. Lehman, J.E. Schwendeman, P.M. OÕDonnell, K.B. Wagener,
Inorg. Chim. Acta 345 (2002) 190–198;
(b) K.B. Wagener, D.J. Valenti, S.F. Hahn, Macromolecules 30
(1997) 6688–6690.
[16] M. Ulman, R.H. Grubbs, Organometallics 17 (1998) 2484–2489.
[17] Typical catalyst loading for CM reactions are in the order of 0.1–
1 mol% for 1, 2–10 mol% for 2, and 1–5 mol% for 3.
[18] J. Robbins, G.C. Bazan, J.S. Murdzek, M.B. OÕRegan, R.R. Schrock,
Organometallics 10 (1991) 2902–2907.
(c) F.C. Courchay, J.C. Sworen, K.B. Wagener, Macromolecules 36
(2003) 8231–8239;
(d) V.I. Petkovska, T.E. Hopkins, D.H. Powell, K.B. Wagener,
Macromolecules 38 (2005) 5878–5885.
[19] R.R. Schrock, J.Y. Jamieson, J.P. Araujo, P.J. Bonitatebus Jr., A.
Sinha, L.P.H. Lopez, J. Organomet. Chem. 684 (2003) 56–67.
[20] (a) F. Bernardi, A. Bottoni, G.P. Miscione, Organometallics 22
(2003) 940–947;
[14] (a) For ADMET reviews, see: T.W. Baughman, K.B. Wagener, Adv.
Poly. Sci. 176 (2005) 1–42;
(b) L. Cavallo, J. Am. Chem. Soc. 124 (2002) 8965–8973;
(c) C. Aldhart, C. Hinderling, H. Baumann, P. Chen, J. Am. Chem.
Soc. 122 (2000) 8204–8214.
(b) S.E. Lehman, K.B. Wagener, in: R.H. Grubbs (Ed.), Handbook
of Metathesis, vol. 3, Wiley–VCH: Weinheim, Germany, 2003, pp.
283–353.
[21] Since the alkylidene is the most reactive intermediate, it is impossible to
assert the amount of non-productive metathesis occurring through step 30.
[22] R.C. Gadwood, R.M. Lett, J.E. Wissinger, J. Am. Chem. Soc. 108
(1986) 6343–6363.
[15] (a) J.A. Smith, K.R. Brzezinska, D.J. Valenti, K.B. Wagener,
Macromolecules 33 (2000) 3781–3794;