Zinc N-heterocyclic carbene complexes and their polymerization of d,l-lactide

Article abstract of DOI:10.1016/j.jorganchem.2005.07.070

A series of zinc complexes of monodentate N-heterocyclic carbenes (NHCs) and a new sterically bulky bidentate pyridyl-NHC ligand have been synthesized and characterized by spectroscopic and X-ray crystallographic methods. Dinuclear alkoxide complexes of monodentate NHC complexes with 2,4,6-trimethylphenyl substituents appear to form monomeric species in solution and show good control and activity for lactide polymerization, including mild stereoelectivity as indicated by formation of heterotactic-enriched polylactide in d,l-lactide polymerizations. Kinetics studies revealed an overall second order rate law, first order in [LA] and [catalyst]. Efforts to obtain Zn-alkoxide complexes of a more sterically hindered NHC with 2,6-diisopropylphenyl groups were unsuccessful due to Zn-NHC bond scission. Ligand dissociation was also observed in attempts to prepare Zn-alkoxide complexes of the bidentate pyridyl-NHC system, despite its chelating nature.

Full text of DOI:10.1016/j.jorganchem.2005.07.070

Journal of Organometallic Chemistry 690 (2005) 5881–5891
www.elsevier.com/locate/jorganchem
Zinc N-heterocyclic carbene complexes and their polymerization
of D,L-lactide
a
b
a,
a,
*
Tryg R. Jensen , Chris P. Schaller , Marc A. Hillmyer *, William B. Tolman
a
Department of Chemistry, University of Minnesota, 207 Pleasant Street SE, Minneapolis, MN 55455, USA
Department of Chemistry, College of St. Benedict/St. Johnꢀs University, 37S College Avenue, St. Joseph, MN 56374, USA
b
Received 21 June 2005; received in revised form 20 July 2005; accepted 20 July 2005
Available online 16 September 2005
Abstract
A series of zinc complexes of monodentate N-heterocyclic carbenes (NHCs) and a new sterically bulky bidentate pyridyl-NHC
ligand have been synthesized and characterized by spectroscopic and X-ray crystallographic methods. Dinuclear alkoxide complexes
of monodentate NHC complexes with 2,4,6-trimethylphenyl substituents appear to form monomeric species in solution and show
good control and activity for lactide polymerization, including mild stereoelectivity as indicated by formation of heterotactic-
enriched polylactide in D,L-lactide polymerizations. Kinetics studies revealed an overall second order rate law, first order in [LA]
and [catalyst]. Efforts to obtain Zn–alkoxide complexes of a more sterically hindered NHC with 2,6-diisopropylphenyl groups were
unsuccessful due to Zn–NHC bond scission. Ligand dissociation was also observed in attempts to prepare Zn–alkoxide complexes of
the bidentate pyridyl-NHC system, despite its chelating nature.
ꢀ 2005 Elsevier B.V. All rights reserved.
Keywords: Zinc; N-heterocyclic carbenes; Polymerization catalysis; Lactide, Polylactide
1. Introduction
(LA). Because LA can be obtained through fermenta-
tion of corn and soybeans, production of PLA does
not depend on the consumption of petrochemical
feedstocks but is instead based on renewable resources
[4]. PLA is currently produced commercially for food
packaging; a number of biomedical applications have
also been developed. Ideally, an LA polymerization
catalyst should not only induce rapid polymerization
of LA, but should also demonstrate control of PLA
molecular weight and stereochemistry. A number of
recent efforts have focused on the development of
single site metal alkoxides for this purpose [5]. Alumi-
num alkoxides with supporting salen-type ligands have
shown good control of tacticity in the production of
PLA from racemic D,L-LA and from meso-LA but
require high temperatures and long reaction times for
adequate conversion [6]. On the other hand, zinc
complexes are more active but demonstrate less
control over tacticity [7,8]. Consequently, additional
N-heterocylic carbenes (NHCꢀs) are now ubiquitous
in their use as ligands for transition metals [1]. Although
they are strong sigma donors analogous to phosphine li-
gands, NHCꢀs are generally less labile. Pd- and Ru–N-
heterocylic carbene complexes have been used for C–C
coupling reactions and reactions involving olefin
metathesis; systems employing NHC ligands have
shown superior performance in many cases, and chiral
NHCꢀs have demonstrated promise in asymmetric catal-
ysis [2,3].
Polylactide (PLA) is a biodegradable polymer pro-
duced by the ring-opening polymerization of lactide
*
Corresponding authors. Tel.: +1 612 625 7834/4061; fax: +1 612
624 7029.
E-mail addresses: hillmyer@chem.umn.edu (M.A. Hillmyer),
tolman@chem.umn.edu (W.B. Tolman).
0022-328X/$ - see front matter ꢀ 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2005.07.070

5882
T.R. Jensen et al. / Journal of Organometallic Chemistry 690 (2005) 5881–5891
investigation is warranted to determine factors condu-
cive to rapid and well-controlled polymerization of LA.
As part of an ongoing effort to address these issues
[9], we are exploring the use of Zn complexes of NHCꢀs
as LA polymerization catalysts. The synthesis of NHC
complexes of diethyl zinc and bis(pentamethylcyclopen-
tadienyl) zinc have been reported previously [10]. A
recent report detailed several zinc complexes of 1,3-
bis(2,4,6-trimethylphenyl)imidazol-2-ylidene and their
homopolymerizations of e-caprolactone and cyclohex-
ene oxide, and copolymerizations of cyclohexene oxide
and carbon dioxide [11]. Work on the coupling of epox-
ides and carbon dioxide with ill-defined imidazolium ha-
lide/zinc halide complexes has also been reported [12].
In our continuing studies of N-heterocyclic carbene
zinc complexes and LA polymerization [13] we have pur-
sued a number of different avenues designed to address
issues of catalyst stability and selectivity. We report
herein the development of synthetic routes to a series
of such complexes and studies of their LA polymeriza-
tion behavior. Because the relative lability of zinc com-
plexes raised the possibility that the NHC ligand could
dissociate from these initiators, we also employed a
new bidentate pyridyl-NHC ligand.
isolable carbene 2a, which was then coordinated to
diethylzinc. Complex 4a was fully characterized, includ-
ing by X-ray crystallography; the structure is repro-
duced for comparative purposes (vide infra) in
1
Fig. 1(a). H NMR spectroscopy in THF-d8 revealed
two mesityl CH3 resonances and one resonance due to
the methylene group of the alkoxide ligand, consistent
with either a monomeric structure or a highly fluxional
dimer in solution.
We have now turned to another, somewhat simplified
synthetic approach that has led to several new com-
plexes 5a,b and 6a (Scheme 1). Rather than deprotonat-
ing the starting imidazolium salt and isolating the
resulting carbene before metal complexation, diethylzinc
was reacted directly with the appropriate imidazolium
salts. This protocol has been demonstrated previously
with the non-bulky 1,3-bis(methyl)imidazolium bromide
and diethyl zinc [14]. In this way, the monomeric
(NHC)Zn(Cl)Et complex 5a was prepared from 1,3-
mesitylimidazolium chloride and diethylzinc in high
yield (91%). Spectroscopically pure 5a was then treated
with one equivalent of benzyl alcohol to afford dinuclear
6a (77%).
A single crystal X-ray structure determination of 6a
(Fig. 1(b)) showed a number of similarities with 4a. Both
complexes are dimeric in the solid state and in both cases
the two zinc atoms are bridged by alkoxide oxygen
atoms. Both cores share C2 symmetry about the bridg-
ing O–O interatomic axis, and in both cases this symme-
try is broken by the capping NHC ligands, which are
almost perpendicular to each other in the case of 6a.
2. Results and discussion
2.1. Complexes of monodentate NHC ligands
We recently reported the synthesis of a new NHC
complex of zinc, 4a, which was highly active in the poly-
merization of lactide. The alkoxide complex 4a was pre-
pared through alcoholysis of the previously reported
[10a] NHC zinc diethyl complex 3a (Scheme 1). Com-
pound 3a, in turn, was made by deprotonation of the
1,3-dimesitylimidazolium salt 1a to provide the stable,
˚
The zinc–carbon bond distances, 2.040 A in 6a and
˚
˚
2.054 A in 4a, are typical; distances of 2.022–2.096 A
have been reported elsewhere [10,11]. However, the
˚
zinc–zinc distance is noticeably shorter in 6a (2.935 A)
˚
than in 4a (3.021 A), indicative of a more tightly coupled
dimeric structure for the former complex.
Scheme 1.

T.R. Jensen et al. / Journal of Organometallic Chemistry 690 (2005) 5881–5891
5883
Fig. 1. Representations of the X-ray crystal structures of (a) 4a (from [13]) and (b) 6a, showing non-hydrogen atoms as 50% thermal ellipsoids.
˚
Selected bond distances (A) and angles (deg): 4a: Zn1–O1, 2.0135(14); Zn1–O2, 1.9198(15); Zn1–C1, 2.054(2); Zn1–Zn2, 3.0211(6); O1–Zn1–O2,
112.31(7); O2–Zn1–O3, 81.98(6); Zn1–O2–Zn2, 98.02(6); O2–Z1–C1, 124.80(8); O1–Zn1–C1, 107.75(7). 6a: Zn1–Cl1 = 2.2619(10); Zn1–
O1 = 1.983(2); Zn1–C1 = 2.040(3); Zn1–Zn2 = 2.9351(6); O1–Zn1–O2 = 84.65(10); Cl1–Zn1–O1 = 108.60(7); Zn1–O1–Zn2 = 96.12(10); O1–Z1–
C1 = 115.80(11); Cl1–Zn1–C1, 108.97(10).
In addition, the solution structure of 6a as deter-
mined by H NMR spectroscopy in THF-d8 is more
(Fig. 2). Molecular weights relative to polystyrene were
generally consistent with one growing chain per alkoxide
ligand (four growing chains per dimeric zinc complex 4a
and two growing chains per dimeric complex 6a). The
reaction resulted in a narrow molecular weight distribu-
tion, with a polydispersity index (PDI) resulting from
initiation by 4a of approximately 1.2 at lactide conver-
sions up to 96%. The polydispersity index was slightly
1
complicated than that of 4a and exhibits a concentration
dependence. At high concentration ([6a] = 20 mM), the
principal species (80%) displays two doublets in the
PhCH2O region of the spectrum, indicating inequivalent
benzylic protons, but only two mesityl CH3 resonances
are observed [15]. At lower concentrations, a minor
species becomes increasingly prevalent (50% at [6a] =
2 mM). This species displays equivalent benzylic protons
but also two inequivalent resonances for the Ar–H and
mesityl ortho-CH3 positions. The ratio of these two spe-
cies can be altered reversibly either by diluting with
THF-d8 or by adding more 6a. We speculate that the
species predominating at high concentration is the dimer
with bridging benzyloxide ligands observed in the solid
state, with the species growing in at lower concentration
being a monomer, presumably (NHC)Zn(Cl)OBn. Inter-
estingly, however, coalescence of the signals for these
two species was not observed in THF solutions at tem-
peratures as high as 55 ꢁC, suggesting a relatively high
energy barrier for their interconversion.
An investigation of the polymerization of D,L-LA ini-
tiated by complexes 4a and 6a in CH2Cl2 at 25 ꢁC re-
vealed good control over polymer molecular weight,
which grew linearly with increasing LA conversion
Fig. 2. Molecular weight (Mn, filled markers) and polydispersity index
(PDI, open markers) of PLA produced using 4a (diamonds) and 6a
(circles) as initiators. Linear fits of molecular weight vs. LA conversion
are shown as solid (4a) or dashed (6a) lines.

5884
T.R. Jensen et al. / Journal of Organometallic Chemistry 690 (2005) 5881–5891
higher when 6a was used as an initiator and this value
increased substantially later in the reaction.
Table 1 illustrates the results of further analysis of
PLA produced using 4a and 6a as initiators. Complex
6a was much more soluble in THF and CH2Cl2 than
4a, allowing more straightforward comparisons of poly-
merizations for 6a performed with different mono-
mer:initiator ratios at a constant starting concentration
of LA. At all catalyst loadings, the molecular weight of
polymer produced was in good agreement with weights
calculated based on one growing chain per initiating alk-
oxide ligand. Analysis of a sample at 10% conversion
showed very good agreement between the theoretical
molecular weight (1980 g/mol) and the experimental
molecular weight obtained both by size exclusion chro-
matography (2.0 kg/mol) and by mass spectrometry
(base peak at 2149 g/mol, corresponding to HO(CHM-
eCO)14OCH2Ph Æ Na+, as well as peaks corresponding
to other oligomers). The latter result confirmed the pres-
ence of a benzyloxide initiating group in the polymer
chain and also revealed a small fraction of chains con-
taining odd numbers of lactic acid units, which indicates
some degree of transesterification processes. Polydisper-
sity indices were moderately narrow, and were especially
low at high catalyst loading (PDI = 1.18 when [LA]0/
[Zn]0 = 50). However, the polydispersity index increased
greatly at longer reaction times.
Fig. 3. Consumption of LA over time during polymerization by 6a at
25 ꢁC in THF, monitoring LA at 1775 cmꢀ1 by FT-IR spectroscopy
([LA]0 =1 M, [Zn]0 = 15 mM). The line is a non-linear fit of the data
(except the first 5 points) to the first order expression At = (A0 ꢀ
A
)eꢀkt + A , which yielded kobs = 0.0119 sꢀ1 (R2 = 0.993).
1
1
is observed in the initial part of the curve in a plot of
[LA] vs. time, indicating autocatalysis or an induction
period (Fig. 3). Given the complicated structure of 6a
indicated by its NMR spectra (vide supra), it is possible
that a zinc dimer/monomer equilibrium is responsible
for this behavior.
In further kinetic studies, kobs values for the polymer-
ization of LA by 6a were obtained at various initial cat-
alyst concentrations. A linear fit of the data (Fig. 4) is
consistent with a rate law that is first order in [Zn]0
and second order overall, such that rate = kp[Zn][LA]
and kp = 0.9 Mꢀ1 sꢀ1, with a non-zero x intercept sug-
gestive of a small amount of a catalyst deactivator
(ꢁ3.7 mM) [9a,9d]. Indeed, polymerizations performed
using concentrations of 6a lower than those shown in
Fig. 4 failed to go to full conversion ([Zn]0 = 2 mM,
[LA0]/[Zn]0 = 500). In these cases, initial consumption
of LA was observed but the reaction ceased abruptly
Further mechanistic insight was obtained by kinetic
analysis of the polymerizations initiated by 4a and 6a.
We previously reported that the reaction with 4a was
very fast, with kobs = 1.9 · 10ꢀ3 sꢀ1 at [LA]0 = 1 M
and [4a]0 = 2.7 mM ([Zn]0 = 5.3 mM). The reaction
with 6a occurs at essentially the same rate at a very
similar concentration (kobs = 2.0 · 10ꢀ3 sꢀ1 at [6a]0 =
2.5 mM). These rates approach that of the fastest
reported Zn catalyst [9a], which is about five times more
rapid at the same monomer and catalyst loading (kobs
=
1.2 · 10ꢀ2 sꢀ1 with [LA]0 = 1 M and [Zn]0 = 5.3 mM).
During polymerization with either 4a or 6a, consump-
tion of LA may be fit to an expression for first-order de-
cay, but with 6a a small but reproducible inflection point
Table 1
Data for the polymerization of D,L-LA by compounds 4a and 6aa
Catalyst
[LA]0/[Zn]0
Polymerization time (min)
LA conv. (%)b
Calculated Mn (kg/mol)
Mn (kg/mol)c
PDI (Mw/Mn)c
Prd
4a
6a
6a
6a
6a
6a
a
260
200
130
130
130
50
20
16
75
4
0.5
2
96
97
99
99
10
95
18.6
28.0
18.6
18.6
2.0
17
30
19
20
2.0; 2.2e
7.4
1.25
1.38
1.68
1.52
1.39
1.18
0.6
0.6
0.6
0.6
0.6
0.6
7.0
Conditions: CH2Cl2 at 25 ꢁC; [LA]0 = 1 M.
Determined by 1H NMR spectroscopy.
SEC (relative to polystyrene in THF).
Probability of racemic linkage between monomer units as determined from the methine region of the homonuclear decoupled 1H NMR spectrum.
Base peak in MALDI-TOF mass spectrum.
b
c
d
e

T.R. Jensen et al. / Journal of Organometallic Chemistry 690 (2005) 5881–5891
5885
interactions that favor one orientation of LA over
another.
Consequently, we applied the synthetic strategies
outlined above to prepare more sterically hindered zinc
complexes of 1,3-bis(2,6-diisopropylphenyl)imidazol-2-
ylidene (2b, Scheme 1). We previously reported the
preparation of the zinc diethyl complex 3b, prepared
via a route analogous to 3a [13]. The X-ray crystal
structure of 3b, unavailable at that time, is shown in
Fig. 5(a). Alternatively, complex 5b was prepared
through direct addition of diethylzinc to the imidazo-
lium chloride 1b. This reaction proceeded in good yield
to give a spectroscopically pure complex that was also
characterized by X-ray crystallography (Fig. 5(b)).
Both structures feature a monomeric species with a
Zn(II) ion in a trigonal planar coordination environ-
˚
Fig. 4. Plot of pseudo-first order rate constants kobs as a function of
[Zn]0 polymerizations of D,L-LA by 6a at 25 ꢁC in THF ([LA]0 = 1 M).
From the slope of the linear fit shown kp = 0.9 Mꢀ1 (R2 = 0.93).
ment. The Zn–C1 bond in 3b, at 2.100 A, is at the
higher end of the range observed in other zinc carbene
˚
complexes (2.022–2.096 A) [10,11]. This bond is pre-
sumably lengthened due to the steric influence of the
neighboring isopropyl groups. The analogous bond in
5b is somewhat shorter, however, at 2.032 A. It may
be that the two ethyl groups in 3b exert a more dra-
matic steric effect than the single ethyl ligand and the
chloride in 5b, allowing the bulky NHC ligand to bind
more tightly to 5b see Fig. 6.
Unfortunately, treatment of either 3b or 5b with ben-
zyl alcohol failed to result in alcoholysis of the zinc ethyl
groups. The reaction of benzyl alcohol with 3b instead
yielded the liberated NHC 2b. It may be that the steric
bulk of the carbene ligand results in a zinc–carbon bond
that is too weak to survive reaction with the strongly
coordinating alcohol; addition of alcohol consequently
results in ligand dissociation. On the other hand, the
reaction of benzyl alcohol with 5b resulted in a complex
mixture of products, and it is unclear what role carbene
dissociation may have played in that case.
after approximately 1 min. These data are consistent
with catalyst deactivation at low catalyst loadings, sim-
ilar to what had been seen previously for other systems
[9a,9d], but in this case with the rate of deactivation
being similar to the rate of LA polymerization.
˚
1H NMR analysis [16] showed that both 4a and 6a
exhibit modest stereocontrol in the polymerization of
D,L-LA, with a slight preference for racemic linkages
between monomer units in the PLA chains (Pr = 0.6,
Table 1). This stereobias presumably arises from
chain-end control, since 4a and 6a contain no chiral cen-
ters. Thus, the orientation of an approaching LA unit is
influenced by the stereochemistry of the last unit of LA
previously incorporated. We speculated that this subtle
stereochemical influence might be enhanced by increas-
ing the steric bulk at the site of polymerization. In-
creased crowding at the site would exacerbate any
Fig. 5. Representations of the X-ray crystal structure of (a) 3b and (b) 5b, showing non-hydrogen atoms as 50% thermal ellipsoids. The hydrogen
˚
atoms are omitted for clarity. Selected bond distances (A) and angles (ꢁ) for 3b: Zn1–C1 = 2.1000(18); Zn1–C2 = 1.998(2); Zn1–C3 = 2.027(3); C1–
Zn1–C2 = 117.62(9); C1–Zn1–C3 = 111.58(10); C2–Zn1–C3 = 129.83(12). Disorder in one of the ethyl groups that was modeled in the crystal
structure solution is omitted here for clarity. 5b: Zn1–C1 = 2.0319(16); Zn1–C2 = 1.979(2); Zn1–Cl1 = 2.2551(8); C1–Zn1–C2 = 132.99(9); C1–Zn1–
Cl1 = 109.57(4); C2–Zn1–Cl1 = 117.42(8).

5886
T.R. Jensen et al. / Journal of Organometallic Chemistry 690 (2005) 5881–5891
Isolating the imidazol-2-ylidene from structures simi-
lar to 7 has generally proven difficult because of multiple
acidic protons in the molecule (e.g., the benzylic protons
on the picolyl group) as well as radical rearrangements
at the benzylic position [18]. Indeed, we were unable
to isolate 3-mesityl-1-picolylimidazol-2-ylidene by addi-
tion of base to 7. On the other hand, the reaction of 7
with diethylzinc resulted in the formation of either 8
or the bis-ligand adduct 9, depending on the stoichiom-
etry of the reaction and the temperature. Because of the
volatility of diethylzinc and the heat required to induce
ethane elimination, 9 was more commonly produced by
the reaction of 8 with imidazolium salt 7. Both the imi-
dazolium salt 7 [15] and complex 9 were characterized
by X-ray crystallography, and the structure of 9 is illus-
trated in Fig. 6 [19]. The Zn(II) ion is bound by two
NHC-pyridyl ligands (albeit with long Zn–N(pyridyl)
distances indicative of a weak bonding interaction)
and one iodide bound to the zinc; the second iodide is
present as a counter anion. The zinc–carbon bond
Fig. 6. Representation of the X-ray crystal structure of 9, showing
non-hydrogen atoms as 50% thermal ellipsoids. The hydrogen atoms
˚
and the anion are omitted for clarity. Selected bond distances (A) and
angles (ꢁ): Zn1–C7 = 2.049(6); Zn1–C25 = 2.055(6); Zn1–I1 =
2.5850(9); C7–Zn1–C25 = 113.7(2); C7–Zn1–N4 = 94.8(2); C25–Zn1–
N4 = 83.7(2); C7–Zn1–I1 = 123.78(16); C25–Zn1–I1 = 122.35(16);
˚
N4–Zn1–I1 = 95.24(13). Selected interatomic distances (A): Zn1–
N1 = 2.557(6); Zn1–N4 = 2.408(5).
˚
2.2. Complexes of a bidentate pyridyl-NHC ligand
lengths are both typical (2.049(6) and 2.055(6) A) of re-
ported values [10,13].
In an effort to prevent the aforementioned ligand dis-
sociation problem and further probe ligand structural
effects on cyclic ester polymerization behavior, we
turned our attention to multidentate NHC derivatives.
A number of recent reports have underscored the utility
of NHC ligands that incorporate chelating phenolic,
alkoxide or pyridyl moieties [17]. In initial efforts toward
this end, we synthesized a new, bidentate pyridyl-NHC
ligand precursor 7 (Scheme 2). Thus, in a procedure sim-
ilar to that reported previously for the preparation of 3-
methyl-1-picolylimidazolium iodide [22], picolyl chloride
was treated with sodium iodide and mesityl imidazole
[23]. The resulting 3-mesityl-1-picolylimidazolium io-
dide, 7, was formed in 84% yield and could be recrystal-
lized in THF.
Unfortunately, reaction of the picolylcarbene zinc
ethyl complex 8 with benzyl alcohol did not yield an alk-
oxide complex. At the same time, straightforward disso-
ciation of the carbene ligand, as observed with
complexes 3b and 5b of the non-chelating NHC ligands,
did not occur here. Instead, protonation of the carbene
ligand resulted in formation of the picolylimidazole salt
7. The exact mechanism of this transformation is un-
clear because formation of 7 was accompanied by the
appearance of intractable side products.
Despite this drawback, we were interested in seeing
whether the complexes of bidentate carbene ligands ini-
tiated polymerization in the absence of alkoxide ligands.
Although polymerization of lactide did not occur in
solution, melt polymerizations of D,L-LA at 140 ꢁC
Scheme 2.

T.R. Jensen et al. / Journal of Organometallic Chemistry 690 (2005) 5881–5891
5887
Table 2
Data for the polymerization of D,L-LA by compounds 7–9a
Catalyst
[LA]0/[Zn]0
Polymerization time (min)
LA conv. (%)b
Calc. Mn (kg/mol)
Mn (kg/mol)c
PDI (Mw/Mn)c
Prd
7
7
8
8
9
9
a
50
200
100
200
50
5
5
5
5
5
5
0
0
96
86
91
73
13.8
24.8
6.6
12
20
6.1
12
2.45
2.04
2.51
1.78
0.6
0.6
0.6
0.6
200
21.0
Conditions: neat, 140 ꢁC.
b
c
Determined by 1H NMR spectroscopy.
SEC (relative to polystyrene in THF).
d
Probability of racemic linkage between monomer units as determined from the methine region of the homonuclear decoupled 1H NMR spectrum.
using 8 and 9 produced PLA (Table 2). The melt poly-
merization produces heterotatic enriched PLA (Pr =
0.6) with good conversion of LA at a rapid rate. The
polydispersity indices are somewhat broad (Mw/
Mn = 1.75–2.45), but the molecular weight distributions
are monomodal. While alcohol impurities in the LA
would probably cause 8 to degrade to the imidazolium
salt, 7, the latter does not initiate polymerization. It is
still possible that some other type of decomposition of
either 8 or 9 occurs to yield an active initiator, but at
high catalyst loadings ([LA]0/[Zn]0 6 100) the similarity
between calculated Mn and experimental Mn suggests
that a single site catalyst is active.
The demonstrated lability of carbenes on zinc that we
have described in this manuscript suggests it is reason-
able to assume that some 1,3-bis(2,4,6-trimethylphenyl)
imidazol-2-ylidene dissociation from complexes such as
6a or 8 could occur. Given that 1,3-bis(2,4,6-trimethyl-
phenyl)imidazol-2-ylidene in the presence of alcohol is
known to be a rapid, stereoelective, and controlled poly-
merization catalyst of D,L-LA [13,20], there could be
some concern that the polymerization we have observed
is not due to zinc complexes, but to dissociated imidazol-
2-ylidenes such as 2a. However, as was shown previously
with 4a, the PLA formed from these reactions is hetero-
tactic enriched (Pr = 0.6) while PLA made with 1,3-
bis(2,4,6-trimethylphenyl)imidazol-2-ylidene is isotactic
enriched [13]. This observation suggests that, while we
cannot discount participation to some degree of 1,3-
bis(2,4,6-trimethylphenyl)imidazol-2-ylidene, the main
catalytic species in polymerizations catalyzed by 6a, for
example, must be distinctly different than the free car-
bene, 1,3-bis(2,4,6-trimethylphenyl)imidazol-2-ylidene.
Unlike some of the other late transition metals, zinc
complexes of NHCꢀs are prone to dissociation, particu-
larly when the NHC is sterically bulky. For example, we
found that addition of alcohol to 3b and 5b comprising
the hindered 1,3-bis(2,6-diisopropylphenyl)imidazol-2-
ylidene ligand failed to yield zinc alkoxides via alcohol-
ysis of the zinc ethyl group. In at least one case, this
complication resulted from dissociation of the carbene
ligand. In order to counter problems with NHC ligand
dissociation, a new imidazolium salt containing a pyr-
idyl arm, 7, was explored. Ligand 7 reacts directly with
diethyl zinc to produce either the bidentate carbene
complex 8 (at room temperature) or 9 (at elevated tem-
peratures). While the use of chelation in 8 failed to pre-
vent loss of the carbene ligand during alcoholysis, both 8
and 9 can polymerize D,L-LA in the melt with high activ-
ity to make heterotactic enriched PLA.
Continuing studies with NHC complexes of zinc
should allow us to produce new catalysts capable of ra-
pid, controlled, highly stereoelective LA polymerization,
and provide further insight into the mechanism of me-
tal-mediated LA polymerization. In particular, we will
continue to examine other chelating NHC ligands on
zinc in an effort to obtain complexes with greater stabil-
ity towards decomposition during polymerization of
lactide.
4. Experimental procedures
4.1. General procedures
All reactions were performed under a nitrogen atmo-
sphere, using either standard Schlenk techniques or adini-
trogen-filled MBraun glovebox. All solvents and reagents
were obtained from commercial sources and used as
received unless otherwise stated. Pentane, dichlorometh-
ane, tetrahydrofuran and toluene were passed through
purification columns (Glass Contour, Laguna, CA). Deu-
terated solvents (all 99 atom%D) were purchased from
Cambridge Isotope Laboratories, dried over calcium
hydride, distilled under vacuum, and stored under nitro-
gen. Benzyl alcohol was distilled from calcium hydride.
3. Concluding remarks
The new NHC zinc alkoxide complex 6a has been
synthesized and characterized. It polymerizes D,L-LA
with good control of molecular weight, high activity,
and moderate stereoelectivity to make heterotactic-en-
riched PLA. The polymerization of D,L-LA by 6a fol-
lows kinetics that are first order in [LA] and [6a].

5888
T.R. Jensen et al. / Journal of Organometallic Chemistry 690 (2005) 5881–5891
1
Diethyl zinc (1 M in hexanes) was purchased from
Aldrich and used as received. D,L-Lactide was purified
by recrystallization from toluene, followed by repeated
(2–3·) vacuum sublimation. 1,3-bis(2,4,6-trimethylphenyl)
limidazol-2-ylidene (2a), 1,3-bis(2,4,6-trimethylphenyl)
limidazolium chloride (1a), 1,3-bis(2,6-diisopropylphe-
nyl)imidazolium chloride (1b), and 1,3-bis(2,6-diisopropyl-
phenyl)imidazol-2-ylidene were prepared as previously
described [21]. The complexes 3a [10a], 4a [13], and 4b
[13] were prepared by established procedures.
was broken prior to characterization by SEC and H
NMR spectroscopy.
4.1.3. Kinetics measurements
The polymerization of D,L-LA was monitored by IR
spectroscopy using a Mettler Toledo ReactIR 4000 spec-
trometer. A reaction flask was charged with LA and sol-
vent in the glovebox (the total volume varied, but the
average total volume was approximately 3.5 mL). The
ReactIR probe was then attached to the flask via a
ground glass joint, and the apparatus was removed from
the glovebox and attached to the IR spectrometer so that
the reaction flask was immersed in a temperature con-
trolled bath at 24 ꢁC with rapid stirring. The solution
temperature was allowed to equilibrate over 5 min, and
catalyst (as a solution prepared in the glovebox) was in-
jected through a septum via syringe. Peaks due to LA at
1776 and 1250 cmꢀ1 and peaks due to PLA at 1750 and
1184 cmꢀ1 were monitored; analysis of LA decay and
PLA growth gave very similar results. The observed rate
constants (kobs) were extracted by fitting an exponential
curve to the plot of absorbance versus time (for the peak
at 1776 cmꢀ1) using At = (A0 ꢀ A )eꢀkt + A , allowing
NMR spectra were collected with a Varian VI-300,
Varian VXR-300, or Varian VXR-500 spectrometer.
1
Chemical shifts for H and 13C spectra were referenced
using internal solvent resonances and are reported rela-
tive to tetramethylsilane. Elemental analyses were deter-
mined by Robertson Microlit Laboratories Inc.,
Madison, NJ. IR spectra for characterization were
recorded on a Thermo Nicolet Avatar 370 FT-IR spec-
trometer, by ATR with a Ge crystal. Molecular weights
(Mn and Mw) and polydispersity indices (PDI = Mw/
Mn) were determined by size exclusion chromatography.
Size exclusion chromatography (SEC) was performed
using a Hewlett–Packard 1100 series liquid chromato-
graph with THF as the mobile phase. The SEC instru-
ment was equipped with three Jordi divinylbenzene
1
1
A0, A , and k to vary freely. The data were analyzed
over at least three half-lives. All linear and non-linear
curve fits were performed with KaleidaGraph.
1
3
4
˚
columns (pore sizes 500, 10 , and 10 A; column temper-
ature 40 ꢁC), and output was detected with an HP1047A
differential refractive index detector, an eluent flow rate
of 1 mL/min, and a 50 lL injection loop. The instru-
ment was calibrated using narrow polydispersity
polystyrene standards (Polymer Laboratories) with
molecular weights ranging from 5000 to 1 · 106 g/mol.
4.2. Ethyl-zinc-chloride Æ 1,3-bis(2,4,6-trimethylphenyl)
imidazol-2-ylidene adduct (5a)
In a 19 mL vial in the glove box, 1,3-bis(2,4,6-trimeth-
ylphenyl)imidazolium chloride (1a, 1.00 g, 2.93 mmol)
was dissolved in THF (50 mL). Diethylzinc (1.0 M in
hexanes) was added by syringe (3.1 mL, 3.1 mmol). The
solution was stirred at room temperature overnight. Vol-
atiles were removed in vacuo to yield a white powder
(1.2 g, 2.7 mmol, 91% yield). 1H NMR (300 MHz,
THF-d8) d ꢀ0.061 (q, J = 8.1 Hz, 2H), 0.69 (t, J =
8.1 Hz, 3H), 2.15 (s, 6H), 2.33 (s, 3H), 7.00 (s, 4H),
7.34 (s, 2H). Anal. Calc. for C23H29ClN2Zn: C, 63.60;
H, 6.73; N, 6.45. Found: C, 60.06; H, 6.39; N, 6.27%.
4.1.1. General solution polymerization procedure
Glassware used for polymerizations was oven or
flame-dried. In the glovebox a vial was charged with
D,L-LA (375 mg, 2.59 mmol) and solvent (2.6 mL) with
rapid stirring. Once the LA had dissolved, a solution
of the catalyst was injected to start the reaction. After
an appropriate time, the reaction mixture was poured
into pentane to precipitate the polymer. Excess pentane
was removed be decantation and the polymer mixture
was dried at 80 ꢁC prior to characterization by SEC
4.3. Dimer of benzyloxy-zinc-chloride Æ 1,3-bis(2,4,6-
trimethylphenyl)imidazol-2-ylidene adduct (6a)
1
and H NMR spectroscopy.
4.1.2. General melt polymerization procedure
To a solution of 5a (750 mg, 1.72 mmol) in cold THF
(3 mL, ꢀ35 ꢁC) was slowly added a solution of benzyl
alcohol (186 mg, 1.72 mmol, in 2 mL THF, ꢀ35 ꢁC).
The resulting pale, yellow solution was stirred for 3 h,
the solvent volume was concentrated in vacuo to
2 mL, and the product was precipitated with cold pen-
tane (15 mL) and collected by filtration to obtain an
off-white product (680 mg, 1.33 mmol, 77%). Single
crystals for examination by X-ray diffraction were
grown by cooling a THF solution to ꢀ40 ꢁC. 1H
NMR (300 MHz, THF-d8): d 7.0–7.25 (m, 5H), 7.09
In the glovebox, a vial was loaded with a Teflon stir
bar, D,L-LA, and the appropriate amount of catalyst,
and the solid components were mixed thoroughly. The
vial was capped tightly, and removed from the glovebox.
The vial was submerged in an oil bath set at 140 ꢁC.
After the appropriate reaction time, the vial was quickly
removed from the oil bath and dropped in a vacuum de-
war filled with liquid nitrogen. When the nitrogen
stopped boiling, the vial was removed. The polymer
was scraped out with a screw driver or the glass vial

T.R. Jensen et al. / Journal of Organometallic Chemistry 690 (2005) 5881–5891
5889
(s, 2H), 6.49 (s, 4H), 3.49 (d, J = 16 Hz, 1H), 3.30 (d,
J = 16 Hz, 1H), 2.10, (s, 6H), 1.85 (s, 12H) ppm.
13C{1H} NMR (300 MHz, THF-d8): 177.1, 149.1,
139.1, 136.1, 129.9, 129.8, 127.4, 126.4, 124.7, 124.5,
35.1, 23.3, 21.4 ppm. IR: 3030(w), 2930(m), 1490(m),
1063(s),739(s) cmꢀ1. Anal. Calc. for C56H62Cl2N4O2Zn2:
C, 65.63; H, 6.10; N, 5.47. Found: C, 63.06; H, 6.19; N,
5.24%.
54.0, 21.2, 17.8. Anal. Calc. for C18H20IN3: C, 53.34;
H, 4.97; N, 10.37. Found: C, 53.11; H, 4.76; N, 10.24%.
4.6. Ethyl-zinc iodide Æ 1-mesityl-3-picolyl imidazol-2-
ylidene adduct (8)
In a 120 mL vial in the glove box, diethylzinc (1.0 M
in hexanes) was added by syringe (5.9 mL, 5.9 mmol, 1.2
equiv.) to THF (100 mL) at ꢀ40 ꢁC and stirred for 30 s.
Compound 7 (2.00 g, 4.94 mmol) was added to the solu-
tion. The solution was stirred at room temperature over-
night and everything dissolved. The solution was
transferred to a 500 mL Erlenmeyer and pentane
(300 mL) was added to precipitate a solid. The solid
was filtered, washed consecutively with pentane
(50 mL), toluene (20 mL), pentane (20 mL) and dried
in vacuo to yield a white powder (2.11 g, 4.2 mmol,
86% yield). An analytically pure sample was made by
4.4. Ethyl-zinc chloride Æ 1,3-bis(2,6-diisopropylphenyl)
imidazol-2-ylidene adduct (5b)
In a 19 mL vial in the glove box 1,3-bis(2,6-diiso-
propylphenyl)imidazolium chloride (1b, 0.321 g,
0.733 mmol) was suspended in THF (10 mL) and
cooled to ꢀ40 ꢁC. Diethylzinc (1.0 M in hexanes)
was added by syringe (1.1 mL, 1.1 mmol). The solu-
tion was allowed to warm to room temperature over
20 min. The reaction was stirred for another 60 min
at room temperature. The solvent volume was reduced
to ꢁ2 mL, the product was precipitated by adding the
mixture to Et2O (15 mL), and then collected by filtra-
tion to obtain an off-white powder (0.34 g, 0.66 mmol,
90% yield). Single crystals were grown by slowing
cooling a toluene solution. 1H NMR (300 MHz,
THF-d8) d ꢀ0.52 (q, J = 8.1 Hz, 2H), 0.68 (t, J =
8.1 Hz, 3H), 1.18 (d, J = 6.9 Hz, 12H), 1.28 (d,
J = 6.9 Hz, 12H), 2.6 (sept, J = 6.9 Hz, 4H), 7.25 (s,
2H), 7.32 (s, 2H), 7.35 (s, 4H). Anal. Calc. for
C29H41ClN2Zn: C, 67.18; H, 7.97; N, 5.40. Found:
C, 66.10; H, 7.93; N, 5.39%.
1
reprecipitation in THF. H NMR (300 MHz, THF-d8)
d ꢀ0.153 (q, J = 8.1 Hz, 2H), 0.822 (t, J = 8.1 Hz,
3H), 1.98 (s, 6H), 2.31 (s, 3H), 5.93 (s, 2H), 6.97 (s,
2H), 7.12 (d, J = 1.8 Hz, 1H), 7.45 (m, 1H), 7.75–7.85
(m, 2H), 7.92 (m, 1H), 8.62 (m, 1H). 13C{1H} NMR
(75 MHz, C6D6, 22 ꢁC) d 182.0, 155.6, 150.6, 140.1,
140.0, 136.1, 130.5, 129.9, 125.8, 125.5, 125.0, 123.7,
121.6, 52.5, 21.3, 18.3, 14.0, 0.4. Anal. Calc. for
C20H24IN3Zn: C, 48.17; H, 4.85; N, 8.43; I, 25.45; Zn,
13.11. Found: C, 48.01; H, 4.69; N, 8.29%.
4.7. (C18H19N3)2ZnI+ Iꢀ (9)
A 100 mL Schlenk flask in the glove box was charged
with 8 (0.221 g, 0.44 mmol), 7 (0.173 g, 0.43 mmol), and
toluene (50 mL). The flask was heated with stirring to
110 ꢁC for 3 h; the solution becomes homogeneous.
The reaction was allowed to cool to room temperature.
The solid was filtered and wash consecutively with tolu-
ene (50 mL) and pentane (30 mL) before being dried in
vacuo to yield a white powder (0.33 g, 0.36 mmol, 83%
yield). Single crystals of 9 were grown by slow evapora-
4.5. 1-Mesityl-3-picolylimidazolium iodide (7)
1-Mesityl-3-picolylimidazolium iodide was synthe-
sized with minor modification to a similar procedure
[22]. To a solution of picolyl chloride, prepared by treat-
ing picolyl chloride hydrochloride (4.63 g, 28.2 mmol)
with sodium hydrogen carbonate (4.19 g, 49.9 mmol)
in 200 mL of acetone were added 1-mesityl imidazole
[23] (5.01 g, 25.1 mmol) and NaI (4.63 g, 30.9 mmol).
After the mixture was stirred for 36 h, volatiles were re-
moved in vacuo. The solid was dissolved in dichloro-
methane (100 mL), and the solution filtered through
Celite. Addition of diethyl ether (400 mL) caused a solid
to precipitate. The solvent was removed by filtration, the
solid taken up in dichloromethane (100 mL), and diethyl
ether (400 mL) added to precipitate the product. After
filtration, drying in vacuo yielded a yellow/orange solid
(9.63 g, 84%). Single crystals were grown by slow evap-
oration from a THF solution. 1H NMR (500 MHz,
CDCl3, 21 ꢁC): d 9.82 (s, 1H), 8.50 (s, 1H), 8.13 (s,
1H), 7.80 (m, 1H), 7.74 (m, 1H), 7.31 (m, 2H), 7.00 (s,
2H), 6.09 (s, 2H), 2.34 (s, 3H), 2.09 (s, 6H). 13C{1H}
NMR (75 MHz, CDCl3): d 152.2, 149.8, 141.4, 137.7,
137.6, 134.3, 130.6, 129.9, 124.3, 124.0, 123.8, 122.8,
1
tion of a pentane solution into a THF solution of 9. H
NMR (300 MHz, acetonitrile-d8) d 2.05 (s, 2H), 2.35 (s,
6H), 5.60 (s, 4H), 7.12 (s, 4H), 7.37 (m, 2H), 7.45 (d,
J = 2.1 Hz, 2H), 7.52 (m, 2H), 7.70 (d, J = 2.1 Hz,
2H), 7.86 (m, 2H), 8.54 (m, 2H). Anal. Calc. for
C36H38IN6Zn: C, 49.48; H, 4.38; N, 9.62; I, 29.06; Zn,
7.48. Found: C, 49.83; H, 4.53; N, 9.39%.
Acknowledgements
Financial support from the NSF (CHE-023666) is
gratefully acknowledged. C.P.S. wishes to acknowledge
support from the NSF-RSEC program at the University
of Minnesota. We thank Lyndal M.R. Hill, Benjamin E.
Kucera, Victor G. Young, Jr., and the X-ray Crystallo-

5890
T.R. Jensen et al. / Journal of Organometallic Chemistry 690 (2005) 5881–5891
(i) M. Scholl, T.M. Trnka, J.P. Morgan, R.H. Grubbs,
Tetrahedron Lett. 40 (1999) 2247;
(j) C.W. Bielawski, R.H. Grubbs, Angew. Chem. 112 (2000)
3025;
graphic Laboratory at the University of Minnesota for
the crystallographic determinations, and Dana Reed
for assistance with mass spectrometry.
(k) T.M. Trnka, R.H. Grubbs, Acc. Chem. Res. 34 (2001) 18;
(l) R.H. Grubbs, Handbook of Metathesis, Wiley–VCH, Wein-
heim, 2003;
Appendix A. Supplementary data
(m) H.M. Lee, D.C. Smith Jr., Z. He, E.D. Stevens, C.S. Yi,
S.P. Nolan, Organometallics 20 (2001) 794.
[4] (a) R.W. Drumwright, P.R. Gruber, D.E. Henton, Adv. Mater.
12 (2000) 1841;
(b) M. Okada, Prog. Polym. Sci. 27 (2002) 87.
[5] (a) O. Dechy-Cabaret, B. Martin-Vaca, D. Bourissou, Chem.
Rev. 104 (2004) 6147;
(b) B.J. OꢀKeefe, M.A. Hillmyer, W.B. Tolman, J. Chem. Soc.
Dalton Trans. (2001) 2215.
[6] (a) N. Spassky, M. Wisniewski, C. Pluta, A. Le Borgne,
Macromol. Chem. Phys. 197 (1996) 2627;
NMR spectra for 6a and a representation of the X-
ray crystal structure of 7 are provided. Crystallographic
data (CIFs) for the structural analyses have been depos-
ited with the Cambridge Cystallographic Data Centre
(CCDC No. 275448 for compound 3b, No. 275331 for
compound 5b, No. 275452 for compound 7, Nos.
275450 and 275449 for compound 9 (forms 1 and 2,
respectively) [19], and No. 275451 for compound 6a).
Copies of this information may be obtained free of
charge from The Director, CCDC, 12 Union Road,
Cambridge CB2 1EZ, UK (Fax: +44 1223 336033; de-
posit@ccdc.cam.ac.uk or www.ccdc.ac.uk). Supplemen-
tary data associated with this article can be found, in the
online version, at doi:10.1016/j.jorganchem.2005.07.070.
(b) A. Bhaw-Luximon, D. Jhurry, N. Spassky, Polym. Bull. 44
(2000) 31;
(c) D. Jhurry, A. Bhaw-Luximon, N. Spassky, Macromol. Symp.
175 (2001) 67;
(d) C.P. Radano, G.L. Baker, M.R. Smith III, J. Am. Chem. Soc.
122 (2000) 1552;
(e) T.M. Ovitt, G.W. Coates, J. Am. Chem. Soc. 124 (2002) 1316;
(f) Z. Zhong, P.J. Dijkstra, J. Feijen, J. Am. Chem. Soc. 125
(2003) 11291;
(g) N. Nomura, R. Ishii, M. Akakura, K. Aoi, J. Am. Chem. Soc.
124 (2002) 5938;
References
(h) K. Majerska, A. Duda, J. Am. Chem. Soc. 126 (2004) 1026;
(i) J. Belleney, M. Wisniewski, A. Le Borgne, Eur. Polym. J. 40
(2004) 523;
[1] (a) D. Bourissou, O. Guerret, F.P. Gabbaie, G. Bertrand, Chem.
Rev. 100 (2000) 39;
(b) W.A. Herrmann, Angew. Chem. Int. Ed. 41 (2002) 1290.
[2] (a) M.T. Powell, D.-R. Hou, M.C. Perry, X. Cui, K. Burgess, J.
Am. Chem. Soc. 123 (2001) 8878;
(j) P. Hormnirun, E.L. Marshall, V.C. Gibson, A.J.P. White,
D.J. Williams, J. Am. Chem. Soc. 126 (2004) 2688.
[7] B.M. Chamberlain, M. Cheng, D.R. Moore, T.M. Ovitt,
E.B. Lobkovsky, G.W. Coates, J. Am. Chem. Soc. 123 (2001)
3229.
(b) Y. Ma, C. Song, C. Ma, Z. Sun, Q. Chai, M.B. Andrus,
Angew. Chem., Int. Ed. 42 (2003) 5871;
(c) L.C. Bonnet, R.E. Douthwaite, B.M. Kariuki, Organometal-
lics 22 (2003) 4187;
[8] M.H. Chisholm, C.-C. Lin, J.C. Gallucci, B.-T. Ko, Dalton
Trans. (2003) 406.
[9] For recent relevant examples, see (a) C.K. Williams, L.E.
Breyfogle, S.K. Choi, W. Nam, V.G. Young Jr., M.A. Hillmyer,
W.B. Tolman, J. Am. Chem. Soc. 125 (2003) 11350;
(b) L.M. Alcazar-Roman, B.J. OꢀKeefe, M.A. Hillmyer, W.B.
Tolman, Dalton Trans. (2003) 3082;
(d) A. Alexakis, C.L. Winn, F. Guillen, J. Pytkowicz, S. Roland,
P. Mangeney, Adv. Synth. Catal. 345 (2003) 345;
(e) D.R. Jensen, M.S. Sigman, Org. Lett. 5 (2003) 63;
(f) L.H. Gade, V. Cesar, S. Bellemin-Laponnaz, Angew. Chem.,
Int. Ed. 43 (2004) 1014;
(g) E. Bappert, G. Helmchen, Synlett (2004) 1789;
(h) P.L. Arnold, M. Rodden, K.M. Davis, A.C. Scarisbrick, A.J.
Blake, C. Wilson, Chem. Commun. (2004) 1612;
(i) S. Tominaga, Y. Oi, T. Kato, D.K. An, S. Okamoto,
Tetrahedron Lett. 45 (2004) 5585.
(c) C.K. Williams, N.R. Brooks, M.A. Hillmyer, W.B. Tolman,
Chem. Commun. (2002) 2132;
(d) B.J. OꢀKeefe, L.E. Breyfogle, M.A. Hillmyer, W.B. Tolman,
J. Am. Chem. Soc. 124 (2002) 4384.
[10] (a) A.J. Arduengo III, H.V. Rasika Dias, F. Davidson, R.L.
Harlow, J. Organomet. Chem. 462 (1993) 13;
(b) A.J. Arduengo III, F. Davidson, R. Krafczyk, W.J. Marshall,
M. Tamm, Organometallics 17 (1998) 3375.
[11] D. Wang, K. Wurst, M.R. Buchmeiser, J. Organomet. Chem. 689
(2004) 2123–2130.
[3] (a) D.S. McGuinness, K.J. Cavell, B.W. Skelton, A.H. White,
Organometallics 18 (1999) 1596;
(b) D.S. McGuinness, K.J. Cavell, Organometallics 19 (2000)
741;
[12] H.S. Kim, J.J. Kim, H. Kim, H.G. Jang, J. Catal. 220 (2003) 44–
46.
[13] T.R. Jensen, L.E. Breyfogle, M.A. Hillmyer, W.B. Tolman, J.
Chem. Soc. Chem. Commun. 21 (2004) 2504.
[14] M.C. Law, K.-Y. Wong, T.H. Chan, Green Chem. 6 (2004) 241.
[15] See supporting information.
[16] (a) M.T. Bell, B.E. Padden, A.J. Paterick, K.A.M. Thakur, R.T.
Kean, M.A. Hillmyer, E.J. Munson, Macromolecules 35 (2002)
7700;
(c) A.M. Magill, D.S. McGuinness, K.J. Cavell, G.J.P. Brit-
ovsek, V.C. Gibson, A.J.P. White, D.J. Williams, A.H. White,
B.W. Skelton, J. Organomet. Chem. 617–618 (2001) 546;
(d) T. Weskamp, W.C. Schattenmann, M. Spiegler, W.A.
Herrmann, Angew. Chem. 110 (1998) 2631;
(e) T. Weskamp, F.J. Kohl, W. Hieringer, D. Gleich, W.A.
Herrmann, Angew. Chem. 111 (1999) 2573;
(f) L. Ackermann, A. Furstner, T. Weskamp, F.J. Kohl, W.A.
¨
Herrmann, Tetrahedron Lett. 40 (1999) 4787;
(g) T. Weskamp, F.J. Kohl, W.A. Herrmann, J. Organomet.
Chem. 582 (1999) 362;
(b) H.R. Kricheldorf, C. Boettcher, K.U. Toennes, Polymer 33
(1992) 2817.
[17] (a) E. Peris, R.H. Crabtree, Coord Chem. Rev. 248 (2004) 2239–
2246;
(h) M. Scholl, S. Ding, C.W. Lee, R.H. Grubbs, Org. Lett. 1
(1999) 953;

T.R. Jensen et al. / Journal of Organometallic Chemistry 690 (2005) 5881–5891
5891
(b) A.W. Waltman, R.H. Grubbs, Organometallics 23 (2004)
3105–3107;
show the information for form 1 here. For further details, see the
supporting information and the CIFs.
(c) A.O. Larsen, W. Leu, C.N. Oberhuber, J.E. Campbell, A.H.
Hoveyda, J. Am. Chem. Soc. 126 (2004) 11130–11131;
(d) P.L. Arnold, M. Rodden, C. Wilson, Chem. Commun. 13
(2005) 1743–1745.
[20] (a) G.W. Nyce, T. Glauser, E.F. Connor, A. Moeck, R.M.
Waymouth, J.L. Hedrick, J. Am. Chem. Soc. 125 (2003) 3046;
(b) E.F. Connor, G.W. Nyce, M. Myers, A. Moeck, J.L. Hedrick,
J. Am. Chem. Soc. 124 (2002) 914.
[18] B. Cetinkaya, E. Cetinkaya, J.A. Chamizo, P.B. Hitchcock, H.A.
Jasim, H. Kucukbay, M.F. Lappert, J. Chem. Soc., Perkin Trans.
1 (1998) 2047–2054.
[21] A.J. Arduengo III, R. Krafczyk, R. Schmutzler, H.A. Craig, J.R.
Goerlich, W.J. Marshall, M. Unverzagt, Tetrahedron 55 (1999)
14523.
[19] Complex 9 crystallized in two forms, and X-ray crystal structure
determinations were performed for both. In form 2, significant
disorder of the main molecule was observed which required fixing
of key bond distances to obtain a reasonable solution; thus, we
[22] D.S. McGuinness, K.J. Cavell, Organometallics 19 (2000)
741.
[23] B.E. Ketz, A.P. Cole, R.M. Waymouth, Organometallics 23
(2004) 2835–2837.