Ring-opening polymerization of lactide with group 3 metal complexes supported by dianionic alkoxy-amino-bisphenolate ligands: Combining high activity, productivity, and selectivity

Article abstract of DOI:10.1002/chem.200500856

A series of new alkoxy-amino-bis(phenols) (H₂L 1-6) has been synthesized by Mannich condensations of substituted phenols, formaldehyde, and amino ethers or diamines. The coordination properties of these dianionic ligands towards yttrium, lantha

Full text of DOI:10.1002/chem.200500856

FULL PAPER
DOI: 10.1002/chem.200500856
Ring-Opening Polymerization of Lactide with Group 3 Metal Complexes
Supported by Dianionic Alkoxy-Amino-Bisphenolate Ligands: Combining
High Activity, Productivity, and Selectivity
Abderramane Amgoune,[a] Christophe M. Thomas,[a] Thierry Roisnel,[b] and
Jean-FranÅois Carpentier*[a]
Abstract:
A
series of new alkoxy-
polydispersities of the isolated poly-
mers, together with the linear natures
of number average molecular weight
versus conversion plots and monomer-
to-catalyst ratios. Complex [Y(L6)-
{N(SiHMe2)2}(THF)] (17) polymerized
The in situ formation of [Y(L6)(OiPr)-
(THF)] (18) from 17 and 2-propanol
resulted in narrower molecular weight
distributions (PDI = 1.06). With com-
plex 18, highly heterotactic PLAs with
narrow molecular weight distributions
were obtained with high activities and
productivities at room temperature.
The natures of the ligand substituents
were shown to have a significant influ-
ence on the degree of control of the
polymerizations, and in particular on
the tacticity of the polymer.
amino-bis(phenols) (H2L 1–6) has been
synthesized by Mannich condensations
of substituted phenols, formaldehyde,
and amino ethers or diamines. The co-
ordination properties of these dianionic
ligands towards yttrium, lanthanum,
and neodymium have been studied.
The resulting Group 3 metal complexes
have been used as initiators for the
ring-opening polymerization of rac-lac-
tide to provide poly(lactic acid)s
(PLAs). The polymerizations are
living, as evidenced by the narrow
rac-lactide to heterotactic PLA (Pr
=
0.90 at 208C) and meso-lactide to syn-
diotactic PLA (Pr = 0.75 at 208C).
Keywords:
mers lactide
polymerization · tacticity
biodegradable poly-
·
·
lanthanides ·
Introduction
Among the most promising candidates in this class of mate-
rials are aliphatic polyesters and, in particular, poly(lactic
acid)s (PLAs).[2] These polymers, derived from 100% re-
newable resources such as corn or sugar beets, have many
properties similar and sometimes superior to those of tradi-
tional olefin-based polymers, with the added benefit of bio-
degradability.[3] Although several methods for synthesis of
PLAs exist, the most convenient and promising route is the
ring-opening polymerization (ROP) of lactide, in which the
driving force for polymerization is the relief of ring strain.[4]
Stereochemistry plays an important role in PLAs, determin-
ing the mechanical properties, biodegradability, and ulti-
mately the end use of the material.[1,3] There has been par-
ticular emphasis over the past decade on the synthesis of
discrete, well characterized complexes that function as
active polymerization initiators.[5] These recent advances in
catalyst design have resulted in a variety of PLA architec-
tures being obtained from the enantiomerically pure mono-
mers, the racemic mixture, or meso-lactide (Scheme 1).[6]
Systematic studies have demonstrated the influence both of
the initiating group and of the ligand architecture on poly-
merization behavior.[7]
Biodegradable polymers have recently attracted much inter-
est as replacements for conventional oil-based materials.[1]
[a] Dipl.-Chem. A. Amgoune, Dr. C. M. Thomas,
Prof. Dr. J.-F. Carpentier
Institut de Chimie de Rennes, UMR 6509
CNRS-UniversitØ de Rennes 1
OrganomØtalliques et Catalyse
35042 Rennes Cedex (France)
Fax : (+33)2–2323–6939
E-mail: jean-francois.carpentier@univ-rennes1.fr
[b] Dr. T. Roisnel
Chimie du Solide et Inorganique MolØculaire
Centre de DiffractomØtrie X
Institut de Chimie de Rennes
UniversitØ de Rennes 1
35042 Rennes (France)
Supporting information (NMR data for complexes 14 and 18 generat-
ed in situ; representative homonuclear decoupled H NMR and
MALDI-TOF spectra, and GPC trace of a poly(lactic acid) (entry 4,
Table 3) for this article is available on the WWW under http://
www.chemeurj.org/ or from the author.
1
Chem. Eur. J. 2006, 12, 169 – 179
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
169

Scheme 1. Stereoisomers of lactide.
Among the variety of initiators, zinc and magnesium com-
plexes bearing sterically demanding achiral b-diiminato li-
gands are notable as highly active for the polymerization of
rac-lactide to heterotactic poly(lactic acid) by a chain-end
control mechanism in which the last unit in the growing
polymer chain influences which enantiomeric form of the
monomer is incorporated next.[6e–i] Coates and co-workers
first reported the sensitivity of this polymerization to ligand
sterics.[6h] In particular, the authors described the importance
of steric bulk in the N-aryl substituents of b-diiminate li-
gands, with subtle ligand modifications resulting in dramatic
effects on activity and selectivity.[6h] We also recently report-
ed the synthesis of heterotactic PLA from rac-lactide with a
single-site Group 3 metal complex that controls stereochem-
istry by a chain-end control mechanism.[6n] To expand upon
this initial work, we were interested in investigating the
effect of complex architecture on lactide polymerization be-
havior (i.e., initiator selectivity and activity). Our approach
for determining the structure/reactivity dependence was
based on fine-tuning of the ancillary ligand and study of the
effect on the catalytic reactivity. For this purpose, we have
designed and synthesized new Group 3 metal complexes
bearing dianionic alkoxy-amino-bisphenolate ligands. De-
tails of the synthesis, structures, and polymerization per-
formances of these new complexes are reported here.
Scheme 3. Bisphenolate ligands screened.
product.[8] The nature of the substituent ortho to the phenol-
ic hydroxy group plays an important role in determining the
course of the reaction. All spectroscopic data for 1–6 are
given in the Experimental Section.
Group 3 precursors [MR3(THF)2] (M = Y, La ; R
=
CH2SiMe3, N(SiHMe2)2) and [Nd{N(SiMe3)2}3] were each
treated with one equivalent of each of the neutral bisphe-
nols H2L 1–6 to give the corresponding yttrium and lantha-
num complexes [(L)M(R)(THF)] 7–10, 12–14, 16, and 17,
and [(L)Nd{N(SiMe3)2}] 11 and 15 in good yields
(Scheme 4). A key event in this reaction is the protonolysis
of two of the bulky amido or carbyl ligands in the homolep-
tic precursor by the relatively acidic bisphenols.[9,10] The
products were easily isolated as white solids by evaporation
of volatiles and washing of the residues with pentane. The
compounds are quite air- and moisture-sensitive, readily
soluble in aromatic hydrocarbons (toluene, benzene), and
slightly soluble in aliphatic hydrocarbons (pentane, hex-
anes). These complexes were characterized by elemental
analysis and X-ray diffraction studies in some cases, and by
NMR spectroscopy for diamagnetic complexes 7–10, 12–14,
16, and 17.
Results and Discussion
Synthesis of bisphenolate Group 3 metal complexes: Diami-
no and alkoxy-amino-bisphenol ligands H2L 1–6 (of which 1
and 4–6 are new) were synthesized by double Mannich con-
densations of the corresponding substituted phenol, formal-
dehyde, and the appropriate alkoxy-amine or diamine
(Scheme 2).[8] They can be isolated in moderate to high
yields as white microcrystalline powders (Scheme 3). Some
of these N,N-substituted compounds were difficult to obtain
because of the side-formation of benzoxazines, which takes
place through cyclization of the intermediary N-substituted
The 1H and 13C NMR data for yttrium and lanthanum
complexes 8–10, 13, 14, 16, and 17 in [D6]benzene at room
temperature are in each case indicative of the formation of
a single, monomeric species with rigid Cs symmetry on the
NMR timescale, with the phenoxy groups in the trans con-
figuration. This is evidenced by the symmetry-related phe-
1
nolate rings, as well as the H AB spin systems for the two
benzylic methylene units and the single resonance for the si-
1
lylamido or carbyl group. The H and 13C resonances corre-
Scheme 2. Synthesis of bisphenolate ligands 1–6.
sponding to the coordinated THF molecule are clearly shift-
170
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 169 – 179

Ring-Opening Polymerization of Lactide with Group 3 Metal Complexes
FULL PAPER
with all four donors of the bis-
phenolate ligand bound to the
metal center. One THF ligand
and one bis(dimethylsilyl)amido
group occupy the remaining cis
positions. The oxygen phenolate
atoms of the ligand are ar-
ranged trans to each other, with
À
Y O distances of 2.148(2) Š
and 2.173(2) Š. Because of the
constraints imposed by the
ligand, however, the O(2)-Y(1)-
O(3) angle—of 151.88(9)8—is
significantly smaller than 1808.
The aryl rings of the phenolate
moieties fold back toward the
pendant 2-methoxyethyl arm
and are somewhat bent away
Scheme 4. Synthesis of Group 3 metal complexes 7–17.
ed from those observed for the free solvent (d(1H) = 3.57
Table 1. Selected bond lengths [Š] and angles [8] for complex 13.
and 1.40 ppm; d(13C) = 67.80 and 25.72 ppm in [D6]ben-
zene). Additionally, the 1H resonances for the side arm
OMe groups in complexes 8–10, 13, 14, 16, and 17 were
found to be shifted by approximately 0.2–0.6 ppm from
those in the corresponding free ligands, confirming a six-co-
ordinated geometry. This assignment of solution structures
was supported by single-crystal X-ray structure analyses of
8–10 and 13. The solid-state structures of 8–10 have been re-
ported in preliminary communications and will not be dis-
cussed further here.[6n,9a]
Single crystals of the new adamantyl-substituted complex
13 suitable for X-ray diffraction were obtained by slow
evaporation of a saturated toluene solution at room temper-
ature. The ORTEP plot of the complex with its atom num-
bering scheme is illustrated in Figure 1 and selected bond
lengths and angles are listed in Table 1. Overall, the struc-
ture of 13 is closely related to that of the parent tert-butyl-
substituted complex 8.[6n] Complex 13 adopts a distorted oc-
tahedral coordination geometry around the yttrium atom,
Y(1) O(2)
2.148(2)
2.173(2)
2.296(3)
2.318(2)
2.370(2)
2.597(3)
1.702(3)
1.700(3)
O(2)-Y(1)-O(3)
O(2)-Y(1)-N(1)
O(3)-Y(1)-N(1)
O(2)-Y(1)-O(1)
O(3)-Y(1)-O(1)
N(1)-Y(1)-O(1)
O(2)-Y(1)-O(4)
O(3)-Y(1)-O(4)
N(1)-Y(1)-O(4)
O(1)-Y(1)-O(4)
O(2)-Y(1)-N(2)
O(3)-Y(1)-N(2)
N(1)-Y(1)-N(2)
O(1)-Y(1)-N(2)
O(4)-Y(1)-N(2)
Si(1)-N(1)-Si(2)
151.88(9)
101.29(10)
106.80(9)
84.65(8)
82.23(8)
115.52(9)
90.88(9)
90.87(9)
88.96(9)
155.52(8)
76.02(9)
78.85(9)
155.81(10)
88.37(9)
67.22(9)
120.00(17)
À
À
Y(1) O(3)
À
Y(1) N(1)
À
Y(1) O(1)
À
Y(1) O(4)
À
Y(1) N(2)
À
Si(2) N(1)
À
Si(1) N(1)
from the THF ligand, leaving a relatively open cleft for the
bis(dimethylsilyl)amido group. The Y Oether bond length of
2.370(2) Š in 13 is particularly short in relation to those ob-
À
À
served for other Y Oether bonds, which are generally in the
2.45–2.50 Š range,[9] and also in the parent tert-butyl-substi-
À
À
tuted complex 8 (Y Oether = 2.414(2) Š). This short Y
Oether bond reflects a weak trans influence of the THF
ligand. The silylamido ligand is located trans to the nitrogen
atom of the bisphenolate ligand, the angle N(1)-Y(1)-N(2)
À
À
being 155.81(10) Š. The Y Namido (2.296(3) Š) and Y Namine
(2.597(3) Š) bond lengths are not far from those found in
other octahedral yttrium(iii) phenolate complexes.[9] Further-
À À
more, there is no evidence of any significant Si H Y agostic
interaction as frequently observed in similar compounds.[10]
Indeed, the Si(1)-N(1)-Si(2) bond angle of 120.0(2)8 in 13
falls into the range of values reported for classical Si-N-Si
angles (119.93–129.588).
The preparation of two complexes—7 and 12—was found
to be more complicated. Firstly, a mixture of species was ob-
tained upon treatment of 1 with one equivalent of [Y-
Figure 1. Molecular structure of complex 13; all hydrogen atoms and sol-
vent molecules are omitted for clarity.
Chem. Eur. J. 2006, 12, 169 – 179
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
171

J.-F. Carpentier et al.
{N(SiHMe2)2}3(THF)2], and complete purification could not
be achieved. Key 1H NMR resonances included four
ArCHH AB spin systems and at least two singlets for the
complete conversions of the monomer into PLA are a-
chieved within 1 h for monomer-to-initiator ratios of up to
500. All the polymers produced have narrow molecular
weight distributions and number-average molecular mass
(Mn) values close to the theoretical ones (calculated on the
assumption that each silylamide or carbyl group initiates the
polymerization). These data indicate that the polymerization
had proceeded in a living fashion: that is, without any signif-
icant side reactions. To illustrate this living character, the di-
block copolymerization of rac-lactide and caprolactone was
achieved, by a two-step sequential procedure, to provide a
well defined poly(lactide-block-caprolactone) (PLA-block-
PCL) copolymer (Scheme 6).[13] The first step involved
living ROP of rac-lactide ([LA]/[8] = 100, 208C, 1 h) to
À
Y N(Si(CH3)3)2 hydrogens, of which one series accounted
for about 50% of the total. The enhanced steric flexibility
within 1, originating from the presence of less bulky methyl
groups, and the corresponding more opened coordination
sphere at the metal center might be responsible for the for-
mation of a mixture of monomeric (7) and dimeric spe-
cies.[9d,11] Further complexes derived from ligand 1 were not
studied for that reason. Secondly, treatment of [Y-
{N(SiHMe2)2}3(THF)2] with bisphenol 3, which has a pend-
ant amine arm, proceeded in toluene with selective elimina-
tion of two equivalents of HN(SiHMe2)2, but the room tem-
1
À
À
perature H NMR spectrum showed broad resonances, indi-
yield a pre-polymer with an amino end group R2N PLA
1
À
cative of fluxional behavior. The H NMR spectrum at 608C
O [Y]. The reaction proceeded in a stereoselective way
displayed two sets of sharp resonances for the aryl and
SiHMe2 hydrogens, suggesting that another species is
formed in addition to 12.[12]
(vide infra) to give a macroinitiator with heterotactic PLA
sequence (Mn = 18300 gmolÀ1; Mw/Mn = 1.30, Pr = 0.80).
In the second step, polymerization of e-caprolactone ([CL]/
À À
[Y] = 100) with the new macroinitiator R2N-PLA O [Y]
Ring-opening polymerization of rac-lactide: Amido and
carbyl complexes 7–17 are active initiators for the controlled
ROP of rac-lactide under mild conditions (Scheme 5). Rep-
occurred under mild conditions (THF, 208C, 12 h) to give a
single diblock PCL-block-PLA (Mn,exp = 32900 gmolÀ1; Mw/
Mn = 1.25).
An important goal of this work was to study the asymmet-
ric incorporation of rac-lactide into the polymer backbone.
1
For this purpose, H NMR spectroscopy is applicable for de-
termination of the stereochemistry of the poly(lactic acid)s
by inspection of the methine region of homonuclear decou-
1
pled H NMR spectra of the polymers.[14] PLA derived from
rac-lactide can exhibit up to five tetrad sequences in relative
ratios determined by the ability of the initiator to control
racemic [r-diad] and meso [m-diad] connectivity of the mon-
omer units.
Scheme 5. Synthesis of heterotactic PLA.
resentative polymerization data are summarized in Table 2.
Homopolymerization of rac-lactide with the prepared
group 3 metal complexes proceeds rapidly at 208C, equally
in toluene and THF solvents in terms of activity. In all cases,
Microstructural analysis of the different PLAs formed
from rac-LA revealed that at room temperature the
Group 3 metal complexes 7–17 exert a significant influence
on the tacticity of the polymer formed, as the microstructure
is moderately to highly hetero-
tactic (Scheme 5). The peaks
Table 2. Promotion of ring-opening polymerization of rac-lactide by 7–17 in THF.[a]
were assigned to the appropri-
ate tetrads in accordance with
the correlations established in
the literature. The major tetrad
peaks correspond to those of
the mrm and rmr tetrads, indi-
cative of the heterotactic
-RRSSRRSS- sequence of the
polymer.[14]
One important feature of
these systems is that the poly-
merizations in THF were found
to be much more selective than
those in toluene.[6n] In a recent
paper, Rzepa and co-workers
have proposed as a result of
computational analysis that the
[b]
[c]
[c]
Entry
Initiator
[I]
[LA]/[I]
Reaction
time [min]
Mn,calcd
Mn,exp
Mw/Mn
Pr[d]
[gmolÀ1
]
[gmolÀ1
]
1
2
3
4
5
6
7
8
9
10
11
12
13
14
7
8
8
8
200
100
300
500
100
100
200
100
200
100
200
100
200
100
10
5
10
20
5
5
10
5
10
5
10
5
10
20
28800
14400
43200
72100
14400
14400
28800
14400
28800
14400
28800
14400
28800
14400
36000
15500
56200
87000
19000
20400
38500
20900
48000
17300
39400
22000
37800
26300
1.25
1.33
1.25
1.24
1.28
1.34
1.28
1.17
1.22
1.07
1.29
1.18
1.22
1.19
0.56
0.80
0.80
0.80
0.80
0.65
0.49
0.60
0.80
0.80
0.55
0.62
0.80
0.90
9
10
11
12
13
13
14
15
16
17
[a] All reactions performed with [rac-LA] = 0.44m at 208C in THF until completion as determined by integra-
tion of the methyl resonances of LA and PLA. [b] Mn of PLA calculated from Mn,calcd = 144.00”([LA]/[I])”
conversion LA (conv = 100%). [c] Mn and Mw/Mn of PLA determined by SEC-RI with use of polystyrene
standards. [d] Pr is the probability of racemic linkages between monomer units and is determined from the me- ROP of rac-lactide initiated by
1
thine region of the homonuclear decoupled H NMR spectrum.
172
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 169 – 179

Ring-Opening Polymerization of Lactide with Group 3 Metal Complexes
FULL PAPER
with the yttrium complex 8 with
variation of the monomer-to-
metal ratio (Table 2, entries 2–
4). As shown by the linear rela-
tionship between the monomer-
to-metal ratio and number-aver-
age molecular masses (Mn),
good control over polymeri-
zation is achieved with this
system, even at a catalyst load-
ing of 0.2%, which yielded Mn
values as high as 87000 gmolÀ1.
The same Pr value was obtained
in all cases. As expected, chang-
ing the amido initiator for an
Scheme 6. Synthesis of a polylactide-block-polycaprolactone diblock copolymer.
alkyl initiator does not affect
the selectivity for heterotactic
PLA (Table 2, entries 2 and 5).
a
single-site
b-diketiminate
magnesium
complex
proceeds
Complex 12 (X = NMe2) was found to be as reactive as
complex 8 (X = OMe) in the polymerization of lactide, but
significantly less stereoselective (Pr = 0.80 for 8, Table 2,
entry 2 versus Pr = 0.60 for 12, Table 2, entry 8). Because
of the uncertainty in the definition of 12 (vide supra), we
cannot conclude whether this decrease in stereoselectivity
reflects a direct influence of the pendant X functionality, or
competition between different initiating species. In contrast
with the Y complexes, the La complexes (Table 2, entries 6
and 11) and the Nd complexes (Table 2, entries 7 and 12)
gave poor selectivities. This trend probably reflects the influ-
ence of the ionic radius of the metal center. We assume
that, since lanthanum and neodymium are larger than yttri-
um,[18] the environment around the yttrium metal center is
more hindered, resulting in better selectivity. Complexes 8
and 16, which bear ligands 2 and 5 with tert-butyl and ada-
mantyl groups, respectively, in the ortho positions of the
phenolates, did not show any appreciable difference in poly-
merization (Table 2, entries 2–4 and 13). This is consistent
with the previously mentioned similarity of the solid-state
structures (i.e., steric crowding) of these two initiators. The
polymerization data also indicate that the para-substituents
on the ligand do not significantly affect the ability of the ini-
tiator to control monomer enchainment; changing the
ligand para-substituents from tert-butyl to methyl groups
(Table 2, entries 9 and 13), for instance, results in no change
in heterotacticity (Pr = 0.80). However, it appears that
steric and electronic modification of the ligand can give rise
to pronounced heterotacticity enhancement of the resulting
polymerization. Indeed, by using bulky and conformational-
ly flexible cumyl (a,a-dimethylbenzyl) groups as R1 and R2
substituents,[17a] we were able to improve heterotacticity sig-
nificantly (Table 2, entry 14); the PLA produced by 17 at
208C is substantially heterotactic, with a Pr of 0.90 (90% of
the linkages formed are between lactide units of opposite
stereochemistry).
[{HC(CMeN-2,6-iPr2C6H3)2}Mg(OMe)(THF)]
through two major transition states, with the highest-energy
transition state dictating the stereochemistry of monomer in-
sertion to give highly heterotactic PLAs.[15] The results sug-
gested that the solvent plays a key role, serving to balance
the system entropically. The authors claimed that this find-
ing should also be applicable to other coordinative initiator
systems.[15] This prediction is in agreement with our observa-
tions, in view of the high heteroselectivity displayed by the
bisphenolate yttrium initiators 7–9, 13, 16, and 17 in THF,
whereas propagating species in toluene generate nearly atac-
tic materials.
Role of the ligand on the stereoselectivity: To examine the
influence of the bisphenolate ligand substituents on the mi-
crostructures of the produced PLAs, the lactide polymeri-
zation abilities of complexes 7–17—with a series of substitu-
ents at the R1, R2, and X positions—were examined in
detail. The factors investigated include steric and possible
electronic effects originating from the substituents on the ar-
omatic rings, as well as those on the pendant amine arm.[16]
Bulky groups were chosen as the ortho-phenolate substitu-
ent R1 in order to favor mononuclear complex formation
and to limit chain aggregation during polymerization reac-
tions.[7b,16] The para-phenolate substituent R2 was varied to
provide ligands with differing electron-donating abilities and
soluble complexes.
Complex 7, which bears ortho-methyl groups at the phe-
nolate rings (Scheme 3), gave low selectivity for heterotactic
poly(lactic acid), probably due to the higher tendency to
form aggregated species (Table 2, entry 1).[9d,11] This lack of
tacticity is also consistent with previous observations that
bulky ligands are required for stereochemical control in the
polymerization of rac-lactide by a chain-end control mecha-
nism.[6e–i] In such a reaction, the presence of bulky ligands
increases the influence of the stereogenic center of the last
inserted monomer, which controls the stereochemistry of
propagation.[6g,17] Several polymerizations were performed
Having identified the most selective catalytic system, we
then examined complex 17 for the ROP of meso-lactide
(Scheme 7). After 0.5 h at 208C, the reaction had proceeded
Chem. Eur. J. 2006, 12, 169 – 179
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
173

J.-F. Carpentier et al.
vious observations, changing the initiating group from
amido (Table 2, entries 2 and 14) to isopropoxide (Table 3,
entries 1 and 4) results in no change in heterotacticity (Pr =
0.80 for 8 and 8/iPrOH; Pr = 0.90 for 17 and 17/iPrOH).
Notably, the yttrium alkoxide generated by treatment of 8
with 2-propanol in 1:1 ratio can polymerize 1000 equiv.
within 40 min at 208C (TOF = 1500 hÀ1; Table 3, entry 2).
In contrast, the system 17/iPrOH is less active than 8/
iPrOH; with use of this system, the conversion of
1000 equivalents of monomer is quantitative within 6 h
(Table 3, entry 5). We assume that this decrease in activity is
due to the presence of a more sterically demanding ligand,
which is also believed to be responsible for the better selec-
tivity observed during the polymerization (Pr = 0.90). This
result is consistent with the usual trade-off between activity
and selectivity. Moreover, the 17/iPrOH system gives signifi-
cantly higher Mn values than 8/iPrOH, with experimentally
determined Mn data in close agreement with the calculated
values. To the best of our knowledge, this is the first exam-
ple of a fast, room-temperature living lactide polymerization
that produces high molecular weight heterotactic PLAs (Mn
Scheme 7. Synthesis of syndiotactic PLA.
to 97% conversion ([LA] = 0.44m in THF, [LA]/[17] =
100). Analysis of the polymer microstructure by homonu-
1
clear decoupled H NMR revealed that 17 forms syndiotac-
tic enriched PLA (Pr = 0.75). This is in line with the find-
ings for the b-diiminate zinc complexes developed by
Coates and co-workers, which also proceed by a chain-end
control mechanism and give highly heterotactic PLA (Pr =
0.90 at 208C) from rac-lactide and moderately syndiotactic
PLA (Pr = 0.76 at 08C) from meso-lactide.[6h]
Polymerization of rac-lactide in the presence of 2-propanol:
Metal alkoxides have been reported to be effective initiators
for lactide ROP, giving polymers both in high yield and with
high molecular weight.[6a,c–e,g,h,j,l,7b,19–23] The isopropoxide ini-
tiating group mimics the propagating groups of the pre-
sumed active species, and complexes with these moieties
produce PLAs of predictable molecular weight and with
narrow molecular weight distributions.[6g] In contrast, amido
= 160000 gmolÀ1) with a narrow polydispersity (Mw/Mn
=
1.13). Increasing the amount of 2-propanol does not signifi-
cantly affect the polymerization rate. However, polymeri-
zation experiments with complexes 8 and 17 in the presence
of two or four equivalents of 2-propanol revealed that an
alkoxide/2-propanol exchange occurs during the chain prop-
agation process. Most interestingly, excess free alcohol acts
as an efficient chain transfer agent, eventually yielding poly-
mer chains with reduced molecular weights and narrow
polydispersities (Table 3, entries 6 and 7).[7b,24] End-group
analysis of PLAs prepared in the presence of one or more
equivalents of iPrOH by MALDI-TOF mass spectrometry
showed that the polymer chains were systematically end-
capped with an isopropyl ester and a hydroxy group; no
cyclic oligomer was detectable (see Supporting Informa-
tion).
À
À
À
and alkyl initiators such as N(SiMe3)2, N(SiHMe2)2, and
CH2SiMe3 are usually inferior initiating groups for the ROP
of rac-lactide at room temperature.[7b] To circumvent this
limitation, alkoxide initiators for lactide polymerization can
be generated in situ by alcoholysis of amido and carbyl com-
plexes with 2-propanol. Recently it was proposed by Okuda
et al. that such Group 3 metal isopropoxides are aggregated,
at least to dimers.[7b]
In an attempt to facilitate initiation and to improve fur-
ther on polymerization performances, we sought to evaluate
different alkoxide complexes generated in situ (Table 3). In
accordance with Okudaꢁs observations,[7b] addition of 2-
propanol to the polymerization reaction affected the activi-
ties of complexes 8 and 17 and resulted in narrower molecu-
lar weight distributions (PDI = 1.06–1.24). In line with pre-
To establish the nature of the active initiating species in
the polymerization of rac-lactide, the reactions of yttrium
complexes 8 and 17 with 2-propanol were examined by
NMR spectroscopy. These reactions and NMR studies were
carried out in two different sol-
vents: benzene, a noncoordinat-
Table 3. Ring-opening polymerization of rac-lactide in THF in the presence of 2-propanol.[a]
ing solvent, was used to investi-
gate the influence of the ligand
array on the nuclearity of the
species, whilst similar studies
[b]
[c]
[c]
Entry
Initiator
[I]
[LA]/[I]
[iPrOH]/[I]
Reaction
time [min]
Mn,calcd
[gmolÀ1
Mn,exp
Mw/Mn
Pr[d]
]
[gmolÀ1
]
1
2
3
4
5
6
7
8
8
8
17
17
17
17
100
1000
1000
100
1000
1000
1000
1
1
4
1
1
2
4
5
40
40
14400
144000
144000
14400
144000
144000
144000
17000
99000
37900
9900
160000
71000
35000
1.06
1.24
1.13
1.06
1.13
1.13
1.08
0.80
0.80
0.80
0.90
0.90
0.90
0.90
were
also
performed
in
[D8]THF, which is the solvent
used in polymerization.
20
360
360
360
In the reaction between 17
and 1 equiv. of 2-propanol in
[D8]THF, the room temperature
1H and 13C NMR spectra estab-
lished quantitative conversion
of the reagents and formation
of a single mononuclear prod-
[a] All reactions performed with [rac-LA] = 0.44m at 208C in THF until completion as determined by integra-
tion of the methyl resonances of LA and PLA. [b] Mn of PLA calculated from Mn,calcd = 144.00”([LA]/[I])”
conversion LA (conv = 100%), not taking into account chain transfer with excess iPrOH. [c] Mn and Mw/Mn
of polymer determined by SEC-RI with use of polystyrene standards. [d] Pr is the probability of racemic link-
ages between monomer units and is determined from the methine region of the homonuclear decoupled
1H NMR spectrum.
174
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 169 – 179

Ring-Opening Polymerization of Lactide with Group 3 Metal Complexes
FULL PAPER
uct:
[Y(L6)(OiPr)([D8]THF)]
([D8]-18) (Scheme 8).[25] The
1H NMR spectrum of 18 in
[D8]THF remains unchanged
after several days at room tem-
perature and shows Cs-symme-
try features down to À208C.
Key 1H NMR data for 18 in- Scheme 8. Synthesis of complex 18.
clude one set of resonances for
two equivalent phenolate rings
bound to the metal center in a trans configuration and a
single OiPr ligand. Additionally, the 13C{1H} NMR spectrum
of 18 in [D8]THF contains one signal for two equivalent
NCH2Ar groups, and two signals for the OCH(CH3)2 ligand.
Essentially the same H NMR characteristics were observed
for the reaction product of 17 with one equivalent of 2-prop-
ide initiators described in this study, and in particular of the
yttrium complex 18, is high polymerization activity and pro-
ductivity in conjunction with high selectivity for heterotactic
polymer formation. The steric bulk of the substituents on
the aromatic rings, and particularly the ortho-phenolate sub-
stituents, plays a crucial role in achieving high heteroselec-
tivity for the chain-end controlled polymerization of rac-lac-
tide. It can be anticipated that this class of Group 3 metal
complexes should continue to provide new and interesting
results. Future work will explore the mechanism and modifi-
cation of this system to develop new initiators that exhibit
higher stereochemical control during the polymerization re-
action.
1
anol in [D6]benzene, indicative of the formation of the same
1
complex 18.[26] The H resonances for the coordinated THF
molecule (d = 3.44 and 1.44 ppm) of 18 in [D6]benzene are
clearly shifted from those observed for the free solvent (d =
3.57 and 1.40 ppm in [D6]benzene).[26] The methine hydro-
gen of the isopropoxide group appears as a multiplet at d =
4.66 ppm, which is shifted downfield with respect to the
signal of free 2-propanol (d = 3.67 ppm). This signal is
broadened and further shifted to d = 4.22 and 4.05 ppm
upon addition of two and three equivalents of 2-propanol,
respectively, to 17, indicative of an exchange process be-
tween the isopropoxide ligand in 18 and free 2-propanol.[7b]
Experimental Section
General conditions: All manipulations requiring dry atmosphere were
performed under purified argon by use of standard Schlenk techniques
or in a glovebox. Solvents (toluene, pentane, THF) were freshly distilled
from Na/K alloy under nitrogen and were degassed thoroughly by freeze-
thaw-vacuum cycles prior to use. Deuterated solvents ([D6]benzene,
[D8]toluene, [D8]THF/99.5% D, Eurisotop) were freshly distilled from
sodium/potassium amalgam under argon and were degassed prior to use.
Starting materials for the synthesis of ligand precursors 1–7 were pur-
chased from Aldrich and were used without further purification. 2-Ada-
mantyl-4-tert-butylphenol[27] and ligands 2 and 3[28] were prepared by re-
ported methods. Ligands 1 and 4–7 were synthesized by modifications of
the methods reported in the literature.[8] Lanthanide precursors were pre-
pared by literature procedures.[29] rac-Lactide (Aldrich) was recrystallized
twice from dry toluene and was then sublimed under vacuum at 508C.
meso-Lactide (Purac) was recrystallized twice from dry propan-2-ol.
1
The H NMR spectrum of the reaction product of 8 with
one equivalent of 2-propanol in [D8]THF also contains only
one set of resonances, consistently with the presence of a
single symmetrical isopropoxide species, as observed in the
case of 17. In contrast, carrying out the reaction in [D6]ben-
1
zene resulted in complicated H NMR spectra, indicative of
the formation of mixtures of species that could not be iden-
tified.
These NMR data show that both isopropoxide complexes
generated in situ from the reactions between 2-propanol
and 8 or 17 are essentially single, monomeric species in
THF, which is the solvent used for ROP of lactide. It there-
fore seems that the difference in catalytic behavior of the
systems 8/iPrOH and 17/iPrOH (18) does not originate from
differences in the nuclearity of the active species. These re-
sults further confirm that subtle steric and possibly electron-
ic features induced by substituents in the phenolate ligands
significantly affect the activity and degree of stereocontrol
of the polymerization.
Instruments and measurements: NMR spectra were recorded on Bruker
AC 200, AC 300, and AC 500 spectrometers in Teflon valve NMR tubes.
1H and 13C chemical shifts are reported in ppm versus SiMe4 and were
determined by reference to the residual solvent peaks. Assignment of sig-
nals was carried out through multinuclear 1D (1H, 13C{1H}) and 2D
(COSY, HMQC, HMBC, NOESY) NMR experiments. Coupling con-
stants are given in Hertz. Size exclusion chromatography (SEC) of PLAs
was performed in THF at 208C with a Waters SIS HPLC pump, a
Waters 410 refractometer, a DAD-UV detector, and Waters styragel col-
umns (HR2, HR3, HR4, HR5E) or PL-GEL Mixte B and 100 A columns.
The number average molecular masses (Mn) and polydispersity indexes
(Mw/Mn) of the resultant polymers were calculated with reference to a
polystyrene calibration. Microstructures of PLAs were measured by ho-
modecoupling 1H NMR spectroscopy at 208C in CDCl3 on a Bruker
AC 500.
Conclusion
In summary, bisphenolate Group 3 metal complexes are effi-
cient initiators in the ROP of rac-lactide to form poly(lactic
acid) polymers with predominantly heterotactic placement
of repeat units. Despite recent significant advances in ROP
of lactide, the number of selective and productive initiators
remains modest. The most notable feature of the isopropox-
Compound 1: A solution of 2,4-dimethylphenol (1.95 g, 16.0 mmol), 2-
methoxyethylamine (0.52 mL, 6.00 mmol), and aqueous formaldehyde
(36%, 1.2 mL, 16 mmol) in methanol (4 mL) was stirred and heated at
reflux for 24 h. The mixture was cooled and filtered, and the residue was
dissolved in methanol and heated at reflux for 2 h. The mixture was
cooled and the white precipitate was filtered off. Recrystallization from
Chem. Eur. J. 2006, 12, 169 – 179
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
175

J.-F. Carpentier et al.
4
3
methanol gave white crystals of 1 (1.10 g, 52%). 1H NMR (200 MHz,
CDCl3, 258C): d = 8.35 (s, 2H; ArOH), 6.85 (s, 2H; ArH), 6.67 (s, 2H;
PhOH), 7.47 (d, J(H,H) = 1.8 Hz, 2H; ArH), 7.33 (t, J(H,H) = 7.7 Hz,
16H; phenyl cumyl), 7.06 (m, 4H; phenyl cumyl) 6.83 (d, 3J(H,H) =
3
ArH), 3.72 (s, 4H; ArCH2), 3.58 (t, J(H,H) = 5.1 Hz, 2H; NCH2CH2O),
1.3 Hz, 2H; ArH), 3.28 (s, 4H; ArCH2), 2.68 (brs, 5H; OCH3 +
3
3
3.47 (s, 3H; OCH3), 2.70 (t, J(H,H) = 4.9 Hz, 2H; NCH2CH2O), 2.20 (s,
NCH2CH2O), 2.15 (t, J(H,H) = 5.1 Hz, 2H; NCH2CH2O), 1.75 (s, 12H;
CH3 cumyl), 1.68 (s, 12H; CH3 cumyl) ppm; 13C{1H} NMR (75 MHz,
[D6]benzene, 258C): d = 153.52, 152.16, 151.70, 141.29, 136.51 (Cq Ar),
128.96, 128.40, 127.63, 126.80, 126.32, 123.57 (Ar), 71.57 (s, NCH2CH2O)
58.68 (OCH3), 57.60 (CH2Ar), 51.67 (s, NCH2CH2O), 43.26 (C(CH3)2),
42.95 (C(CH3)2), 31.82 (CH3), 30.17 (CH3) ppm; HRMS (4000 V, ESI):
m/z calcd for C53H62N1O3: 760.4729; found 760.4726 [M+H]+.
12H; CH3) ppm; 13C{1H} NMR (75 MHz, [D6]benzene, 258C):
d =
152.84, 131.37, 121.24 (Ar-Cq), 127.68, 127.36, 125.15 (Ar-CH), 70.89
(NCH2CH2O), 58.17 (OCH3), 57.04 (NCH2CH2O), 50.77 (CH2Ar), 20.24,
16.03 (CH3) ppm; HRMS (70 eV, EI): m/z calcd for C21H29N1O3:
343.2147; found 343.2139 [M]+.
Compound 4:
A solution of 2-adamantyl-4-methylphenol (0.38 g,
1.60 mmol), 2-methoxyethylamine (0.052 mL, 0.60 mmol), and aqueous
formaldehyde (36%, 0.12 mL, 1.60 mmol) in methanol (2 mL) was stirred
and heated at reflux for 48 h. The mixture was cooled and the superna-
tant liquid was decanted. The remaining oil was dissolved in methanol
and heated at reflux for 2 h. The mixture was cooled and a white precipi-
tate formed. The solid was filtered off to give 4 as white crystals (0.22 g,
Treatment of [Y{N(SiMe3)2}3] with 1—generation of “7”: A solution of 1
(53.4 mg, 0.105 mmol) in toluene (5 mL) was added to a stirred solution
of [Y{N(SiMe3)2}3] (60.0 mg, 0.105 mmol) in toluene (5 mL) at room tem-
perature. The mixture was stirred for 12 h at room temperature and for
2 h at 608C, and volatiles were then removed under vacuum. The residue
was washed with a minimal amount of cold pentane and dried under
vacuum, giving a white powder (43.9 mg, 67% based on 7). 1H NMR
(200 MHz, [D6]benzene, 258C) Key resonances: d = 5.60 (d, 2J(H,H) =
11.7 Hz, 0.5H; ArCH2), 5.45 (d, 2J(H,H) = 11.9 Hz, 0.5H; ArCH2), 5.22
(d, 2J(H,H) = 12.4 Hz, 0.8H; ArCH2), 5.09 (dd, 2J(H,H) = 4.0 Hz, 2J-
1
62%). H NMR (200 MHz, CDCl3, 258C): d = 8.37 (s, 2H; ArOH), 6.93
3
(brs, 2H; ArH), 6.69 (brs, 2H; ArH), 3.67 (s, 4H; ArCH2), 3.48 (brt, J-
3
(H,H) = 5.1 Hz, 2H; NCH2CH2O), 3.44 (s, 3H; OCH3), 2.74 (t, J(H,H)
= 5.1 Hz, 2H; NCH2CH2O), 2.22 (s, 6H; CH3), 2.15 (s, 12H; CH2-ada-
mantyl), 2.04 (s, 6H; CH-adamantyl), 1.76 (s, 12H; CH2-adamantyl) ppm;
13C{1H} NMR (75 MHz, [D6]benzene, 258C): d = 153.61, 137.39 (Ar-Cq),
128.69, 122.71 (Ar-CH), 71.07 (NCH2CH2O), 58.13 (OCH3), 57.44
(CH2Ar), 51.28 (NCH2CH2O), 40.73 (adamantyl), 37.44 (adamantyl),
37.13 (adamantyl), 29.60 (adamantyl), 20.83 (CH3) ppm; HRMS (4000 V,
ESI): m/z calcd for C39H54N1O3: 584.4103; found 584.4107 [M+H]+.
(H,H)
= = 12.4 Hz, 2H;
11.3 Hz, 0.8H; ArCH2), 4.71 (d, 2J(H,H)
ArCH2), 4.61 (d, 2J(H,H) = 12.4 Hz, 2H; ArCH2), 4.03 (d, 2J(H,H) =
11.9 Hz, 0.6H; ArCH2), 3.58 (d, 2J(H,H) = 12.4 Hz, 0.6H; ArCH2), 0.37
(brs, 18H; NSi(CH3)3), 0.31 (s, 16H; NSi(CH3)3) ppm. Resonances for
HNSi(CH3)3 were also observed [d
= 0.12 (s, 92H)], while those of
ArOH disappeared. The product was directly engaged in a polymeri-
zation experiment.
Compound 5: Pieces of sodium (0.50 g, 0.026 mol) were added to a solu-
tion of 4-tert-butylphenol (4.00 g, 0.026 mol) in a mixture of p-xylene
(20 mL) and N,N-dimethylformamide (10 mL). After complete dissolu-
tion of sodium, 1-chloroadamantane (4.55 g, 0.026 mol) was added, and
the mixture was heated at 858C for 24 h. The mixture was cooled, dis-
solved in ether (100 mL), and extracted with aqueous sodium hydroxide
(100 mLof a 10% solution). The aqueous layer was extracted with ether
(2”200 mL). Combined organic extracts were washed with water, dried
over anhydrous sodium sulfate, and concentrated in vacuum to give an
oil. The product was purified by column chromatography (silica gel SI 60
40–63 mm, pentane/ethyl acetate 80:20) to give 2-adamantyl-4-tert-butyl-
phenol as a white powder (2.85 g, 40%). 1H NMR (200 MHz, CDCl3,
258C): d = 7.25 (s, 1H; ArH), 7.09 (d, 3J(H,H) = 8.1 Hz, 1H; ArH),
6.59 (d, 3J(H,H) = 8.1 Hz, 1H; ArH), 4.59 (s, 1H; ArOH), 2.13 (brs,
9H; adamantyl), 1.78 (s, 6H; adamantyl), 1.29 (s, 9H; C(CH3)3).
Compound 8: A solution of 2 (0.153 g, 0.30 mmol) in pentane (5 mL) was
added at room temperature to a stirred solution of [Y{N(SiHMe2)2}3-
(THF)2] (0.189 g, 0.30 mmol) in pentane (5 mL). After the mixture had
been stirred for 2 h at room temperature, a white precipitate had formed.
To ensure complete reaction, the mixture was stirred for an additional
10 h. The solid was then filtered out, yielding 8 as a white powder
1
4
(0.163 g, 67%). H NMR (300 MHz, [D6]benzene, 258C): d = 7.60 (d, J-
(H,H) = 2.5 Hz, 2H; ArH), 7.10 (d, 4J(H,H) = 2.3 Hz, 2H; ArH), 5.14
2
(m, 2H; SiH), 3.87 (d, J(H,H) = 12.5 Hz, 2H; overlap with THF signal,
ArCH2), 3.84 (brm, 4H; a-CH2 THF), 2.97 (d, 2J(H,H) = 12.5 Hz, 2H;
ArCH2), 2.84 (s, 3H; OCH3), 2.71 (t, 3J(H,H)
= 5.2 Hz, 2H;
3
NCH2CH2O), 2.31 (t, J(H,H) = 5.2 Hz, 2H; NCH2CH2O), 1.79 (s, 18H;
C(CH3)3), 1.47 (s, 18H; C(CH3)3), 1.18 (m, 4H; b-CH2 THF), 0.49 (d, 4J-
(H,H) = 3.0 Hz, 12H; HSi(CH3)2) ppm; 1H NMR (300 MHz, [D8]THF,
258C): d = 7.19 (d, 4J(H,H) = 2.5 Hz, 2H; ArH), 6.91 (brs, 2H; ArH),
4.90 (m, 2H; SiH), 4.00 (d, 2J(H,H) = 12.5 Hz, 2H; ArCH2), 3.58 (m,
4H; a-CH2 THF, overlap with solvent resonances), 3.26 (m, 7H; ArCH2,
OCH3, NCH2CH2O), 2.70 (t, 3J(H,H) = 5.2 Hz, 2H; NCH2CH2O), 1.73
(m, 4H; b-CH2 THF, overlap with solvent resonances), 1.48 (s, 18H; C-
(CH3)3), 1.26 (s, 18H; C(CH3)3), 0.15 (d, 4J(H,H) = 3.0 Hz, 12H; HSi-
(CH3)2) ppm; 13C{1H} NMR (75 MHz, [D6]benzene, 258C): d = 161.38,
136.53, 125.40, 124.11 (Ar), 73.11 (NCH2CH2O), 70.96 (a-CH2 THF),
64.54 (CH2Ar), 60.43 (OCH3), 49.57 (NCH2CH2O), 35.46 (C(CH3)3),
34.04 (C(CH3)3), 32.06 (C(CH3)3), 30.30 (C(CH3)3), 24.95 (s; b-CH2
THF), 4.22 (HSi(CH3)2) ppm; elemental analysis calcd (%) for
C41H73N2O4Si2Y: C 61.32, H 9.17, N 3.49; found C 61.74, H 9.36, N 3.36.
A solution of 2-adamantyl-4-tert-butylphenol (0.50 g, 1.75 mmol), 2-me-
thoxyethylamine (0.057 mL, 0.65 mmol), and aqueous formaldehyde
(36%, 0.17 mL, 1.75 mmol) in methanol (2 mL) was stirred and heated at
reflux for 48 h. The mixture was cooled and the supernatant liquid was
decanted. The remaining oil was dissolved in methanol (10 mL) and
heated at reflux for 2 h. The mixture was cooled and a white precipitate
formed. The solid was filtered off to give 5 as a white powder (0.22 g,
52%). 1H NMR (200 MHz, [D6]benzene, 258C):
PhOH), 7.48 (d, 4J(H,H) 2.2 Hz, 2H; ArH), 6.96 (d, 4J(H,H)
2.2 Hz, 2H; ArH), 3.53 (s, 4H; ArCH2), 2.99 (brs, 5H; OCH3
d = 8.80 (s, 2H;
=
=
+
NCH2CH2O), 2.50 (s, 12H; CH2 adamantyl), 2.35 (t, 3J(H,H) = 2.2 Hz,
2H; NCH2CH2O), 2.18 (s, 6H; CH adamantyl), 1.92 (brm, 12H; CH2
adamantyl), 1.37 (s, 18H; C(CH3)3) ppm; 13C{1H} NMR (75 MHz,
[D6]benzene, 258C): d = 153.39, 140.94, 136.59, 122.07 (Cq Ar), 124.76,
123.36 (Ar), 70.79 (NCH2CH2O), 58.12 (OCH3), 57.63 (CH2Ar), 50.93
(NCH2CH2O), 40.70 (adamantyl), 37.37 (adamantyl), 37.33 (adamantyl),
34.07 (adamantyl), 31.65 (C(CH3)3), 29.51 (C(CH3)3) ppm; HRMS
(4000 V, ESI): m/z calcd for C45H66N1O3: 668.5042; found 668.5037
[M+H]+.
Compound 9: A solution of 2 (0.153 g, 0.30 mmol) in pentane (5 mL) was
added at 08C to a stirred solution of [Y(CH2SiMe3)3(THF)2] (0.148 g,
0.30 mmol) in pentane (5 mL). The mixture was stirred for 2 h at 08C
and volatiles were removed under vacuum. The residue was washed with
a minimal amount of cold pentane and dried under vacuum, giving 9 as a
1
colorless powder (0.16 g, 70%). H NMR (300 MHz, [D6]benzene, 258C):
d = 7.59 (d, 4J(H,H) = 2.5 Hz, 2H; ArH), 7.08 (d, 4J(H,H) = 2.5 Hz,
2H; ArH), 3.87 (brm, 4H; a-CH2 THF), 3.76 (d, 2J(H,H) = 12.5 Hz,
2H; ArCH2), 2.92 (d, 2J(H,H) = 12.5 Hz, 2H; ArCH2), 2.88 (s, 3H;
OCH3), 2.44 (t, 3J(H,H) = 5.3 Hz, 2H; NCH2CH2O), 2.21 (t, 3J(H,H) =
5.3 Hz, 2H; NCH2CH2O), 1.80 (s, 18H; C(CH3)3), 1.46 (s, 18H; C-
Compound 6:
A solution of 2,4-dicumylphenol (Aldrich, 5.00 g,
15.1 mmol), 2-methoxyethylamine (0.50 mL, 5.7 mmol), and aqueous
formaldehyde (36%, 1.12 mL, 15.1 mmol) in methanol (2 mL) was
heated at reflux for 72 h. The mixture was cooled and the supernatant
liquid was decanted. The remaining oil was dissolved in methanol and
heated at reflux for 24 h. The mixture was cooled and a white precipitate
formed. The solid was filtered off to give 6 as a white powder (1.10 g,
2
(CH3)3), 1.27 (brm, 4H; b-CH2 THF), 0.49 (s, 9H; Si(CH3)3), 0.40 (d, J-
(Y,H)
=
3.1 Hz, 2H; CH2Si(CH3)3) ppm; 13C{1H} NMR (75 MHz,
[D6]benzene, 258C): d = 161.38, 136.59, 136.40, 125.40, 124.21, 123.90
(Ar), 73.84 (NCH2CH2O), 70.66 (a-CH2 THF), 64.65 (CH2Ar), 61.09
30%). 1H NMR (200 MHz, [D6]benzene, 258C):
d = 8.11 (s, 2H;
176
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 169 – 179

Ring-Opening Polymerization of Lactide with Group 3 Metal Complexes
FULL PAPER
(OCH3), 49.09 (NCH2CH2O), 35.39 (C(CH3)3), 34.04 (C(CH3)3), 32.05 (C-
NMR-scale generation of 14: Solid 4 (10.5 mg, 0.016 mmol) was added at
room temperature to a solution of [La{N(SiHMe2)2}3(THF)2] (10.0 mg,
0.016 mmol) in [D6]benzene (ca. 0.5 mL) in an NMR tube. The tube was
vigorously shaken and left at room temperature. After 12 h, the reaction
(CH3)3), 30.11 (C(CH3)3), 24.92 (b-CH2 THF), 24.70 (d, 1J(C,Y)
=
46.4 Hz; YCH2), 4.64 (Si(CH3)3) ppm; elemental analysis calcd (%) for
C41H70NO4SiY: C 64.97, H 9.31, N 1.85; found: C 65.11, H 9.65, N 1.74.
1
was checked by H NMR spectroscopy, which indicated complete and se-
Compound 10: A solution of 2 (0.204 g, 0.40 mmol) in pentane (5 mL)
lective conversion of the ligand and starting lanthanum precursor into 14.
1H NMR (500 MHz, [D6]benzene, 258C): d = 7.20 (brs, 2H; ArH), 6.80
(brs, 2H; ArH), 5.34 (m, 2H; SiH(CH3)2), 3.69 (m, 4H; a-CH2 THF),
3.38 (m, 2H; ArCH2), 3.17 (s, 4H; NCH2CH2O), 2.82 (m, 2H; ArCH2),
2.43 (s, 12H; adamantyl), 2.40 (s, 6H; CH3), 2.19 (s, 9H; adamantyl +
was added at room temperature to
a stirred solution of [La-
{N(SiHMe2)2}3(THF)2] (0.272 g, 0.40 mmol) in pentane (5 mL). The mix-
ture was stirred for 24 h and volatiles were removed under vacuum. The
residue was washed with a minimal amount of cold pentane and dried
under vacuum, giving 10 as a colorless powder (0.18 g, 92%). 1H NMR
2
OCH3), 1.99 (d, J(H,H) = 12.8 Hz, 6H; adamantyl), 1.91 (d, 2J(H,H) =
(300 MHz, [D6]benzene, 258C): d
= = 2.3 Hz, 2H;
7.57 (d, 4J(H,H)
12.8 Hz, 6H; adamantyl), 1.36 (m, 4H; b-CH2 THF), 0.49 (d, 4J(H,H) =
2.9 Hz, 12H; SiH(CH3)2) ppm; 13C{1H} NMR (125 MHz, [D6]benzene,
258C): d = 162.30, 129.70, 128.00 (Ar), 71.29 (CH2Ar), 69.08 (a-CH2
THF), 61.48 (NCH2CH2O), 48.35 (OCH3), 40.83 (adamantyl), 37.64 (ada-
mantyl), 37.12 (adamantyl), 29.75 (adamantyl), 25.36 (b-CH2 THF), 20.98
(CH3), 3.44 (Si(CH3)2) ppm.
ArH), 7.11 (d, 4J(H,H) = 2.5 Hz, 2H; ArH), 5.25 (m, 2H; SiH), 3.76
(brm, 4H; a-CH2 THF), 3.58 (d, 2J(H,H) = 12.5 Hz, 2H; ArCH2), 3.32
(d, 2J(H,H) = 12.5 Hz, 2H; ArCH2), 3.07 (s, 3H; OCH3), 2.77 (t, 3J-
(H,H) = 5.2 Hz, 2H; NCH2CH2O), 2.27 (m, 2H; NCH2CH2O), 1.73 (s,
18H; C(CH3)3), 1.44 (s, 18H; C(CH3)3), 1.22 (m, 4H; b-CH2 THF), 0.48
(d, 4J(H,H)
=
3.0 Hz, 12H; Si(CH3)2) ppm; 13C{1H} NMR (75 MHz,
Generation of 15: A solution of 4 (9.3 mg, 0.016 mmol) in benzene
(1 mL) was added at room temperature to a solution of Nd[N(SiMe3)2]3
(10.0 mg, 0.016 mmol) in benzene (1.5 mL). The solution was stirred for
12 h at room temperature and for 2 h at 608C, and volatiles were then re-
moved under vacuum. The residue was washed with a minimal amount
of cold pentane and dried under vacuum, giving 15 as a light blue
powder. The product was directly engaged in a polymerization experi-
ment.
[D6]benzene, 258C): d = 161.8, 136.6, 135.6, 125.8, 124.3, 123.9 (Ar), 71.9
(NCH2CH2O), 69.9 (a-CH2 THF), 61.6 (CH2Ar), 60.9 (OCH3), 50.7
(NCH2CH2O), 35.3 (C(CH3)3), 34.0 (C(CH3)3), 32.0 (C(CH3)3), 30.1 (C-
(CH3)3), 25.0 (b-CH2 THF), 3.5 (Si(CH3)2) ppm; elemental analysis calcd
(%) for C41H73N2O4LaSi2: C 57.72, H 8.62, N 3.28; found: C 57.91, H
9.03, N 3.18.
Generation of 11: A solution of 2 (18.0 mg, 0.028 mmol) in benzene
(1 mL) was added at room temperature to a solution of [Nd{N(SiMe3)2}3]
(14.6 mg, 0.028 mmol) in benzene (1.5 mL). The mixture was stirred for
12 h at room temperature and 2 h at 608C, and volatiles were then re-
moved under vacuum. The residue was washed with a minimal amount
of cold pentane and dried under vacuum, giving 11 as a light blue
powder. The product was directly engaged in a polymerization experi-
ment.
Compound 16: A solution of 4 (0.105 g, 0.16 mmol) in toluene (5 mL)
was added at room temperature to a stirred solution of [Y{N(SiHMe2)2}3-
(THF)2] (0.100 g, 0.16 mmol) in pentane (10 mL). The mixture was stir-
red for 12 h at room temperature and volatiles were removed under
vacuum. The residue was washed with a minimal amount of cold pentane
and dried under vacuum, giving 16 as a colorless powder (0.108 g, 71%).
4
1H NMR (500 MHz, [D6]benzene, 258C): d = 7.54 (d, J(H,H) = 2.3 Hz,
NMR-scale reaction between [Y{N(SiHMe2)2}3(THF)2] and 3—genera-
tion of 12: Solid 3 (15.9 mg, 0.030 mmol) was added at room temperature
4
2H; ArH), 7.08 (d, J(H,H) = 2.2 Hz, 2H; ArH), 5.20 (m, 2H; SiHMe2),
3.82 (m, 4H; a-CH2 THF), 3.72 (brs, 2H; ArCH2), 3.11 (d, 2J(H,H) =
12.4 Hz, 2H; ArCH2), 2.82 (t, 3J(H,H) = 4.5 Hz, 4H; NCH2CH2O), 2.53
(s, 12H; adamantyl), 2.34 (s, 3H; OCH3), 2.28 (s, 6H; adamantyl), 2.08
(d, 2J(H,H) = 9.2 Hz, 6H; adamantyl), 1.92 (d, 2J(H,H) = 10.9 Hz, 6H;
adamantyl), 1.47 (s, 18H; C(CH3)3), 1.23 (m, 4H; THF), 0.49 (d, 4J(H,H)
= 2.9 Hz, 12H; SiH(CH3)2) ppm; 13C{1H} NMR (75 MHz, [D6]benzene,
258C): d = 162.39, 125.51, 124.21 (Ar), 72.41 (NCH2CH2O), 69.70 (a-
CH2 THF), 63.45 (CH2Ar), 48.35 (OCH3), 41.14 (adamantyl), 37.57 (ada-
mantyl), 37.48 (adamantyl), 32.42 (C(CH3)3), 29.67 (adamantyl), 24.36 (b-
CH2 THF), 4.12 (Si(CH3)2) ppm; elemental analysis calcd (%) for
C53H85N2O4Si2Y: C 66.36, H 8.93, N 2.92; found: C 66.24, H 8.73, N 2.57.
to
a solution of [Y{N(SiHMe2)2}3(THF)2] (17.1 mg, 0.030 mmol) in
[D8]toluene (ca. 0.5 mL) in a Teflon-valved NMR tube. The tube was vig-
orously shaken and left at room temperature. After 1 h, the reaction was
checked by 1H NMR spectroscopy, which indicated complete conversion
of the starting yttrium precursor and ligand, with release of two equiva-
lents of free amine HN(SiHMe2)2. Because of the fluxional behavior of
the products formed at room temperature, informative NMR data were
obtained at 608C. Key 1H NMR resonances (500 MHz, [D8]toluene,
608C): d = major species 7.48 (d, 4J(H,H) = 2.7 Hz, 4H; aryl), 6.88
(brs, 4H; aryl), 5.00 (m, 1H; SiH(CH3)2) ppm; minor species 7.49 (d, 4J-
(H,H) = 3.2 Hz, 4H; aryl), 6.81 (d, 4J(H,H) = 3.2 Hz, 4H; aryl), 4.90
(m, 1H; SiH(CH3)2) ppm.
Compound 17: A solution of 6 (0.100 g, 0.13 mmol) in toluene (5 mL)
was added at room temperature to a stirred solution of Y[N(SiHMe2)2]3-
(THF)2 (0.083 g, 0.13 mmol) in pentane (5 mL). The mixture was stirred
for 12 h at room temperature and volatiles were removed under vacuum.
The residue was washed with a minimal amount of cold pentane and
Compound 13: A solution of 4 (0.093 g, 0.159 mmol) in toluene (5 mL)
was added at room temperature to a stirred solution of [Y{N(SiHMe2)2}3-
(THF)2] (0.100 g, 0.159 mmol) in pentane (10 mL). The mixture was stir-
red for 12 h at room temperature and volatiles were removed under
vacuum. The residue was washed with a minimal amount of cold pentane
and dried under vacuum, giving 13 as a colorless powder (0.104 g, 75%).
Single crystals of 13 suitable for X-ray diffraction were obtained by slow
dried under vacuum, giving 17 as
a white powder (0.095 g, 68%).
1H NMR (500 MHz, [D6]benzene, 258C): d = 7.53 (brs, 2H; ArH), 7.41
(d, 3J(H,H)
=
7.1 Hz, 8H; cumyl), 7.22 (t, 3J(H,H)
=
7.7 Hz, 4H;
3
cumyl), 7.13 (m, 6H; cumyl), 6.95 (t, J(H,H) = 4.5 Hz, 2H; cumyl), 6.78
(brs, 2H; ArH), 4.80 (brs, 2H; SiH(CH3)2), 3.35 (brs, 2H; ArCH2), 2.93
(m, 4H; THF), 2.80 (brs, 3H; OCH3), 2.69 (brs, 2H; ArCH2), 2.48 (brs,
2H; NCH2CH2O), 2.16 (brs, 6H; CH3 cumyl), 2.04 (brs, 2H;
NCH2CH2O), 1.77 (brs, 12H; CH3 cumyl), 1.74 (brs, 6H; CH3 cumyl),
1.10 (m, 4H; THF), 0.48 (brs, 12H; SiH(CH3)2) ppm; 13C{1H} NMR
(75 MHz, [D6]benzene, 258C): d = 161.34 (Cq, Ar), 152.28 (Cq, cumyl),
137.64, 135.79 (cumyl-C), 128.11, 127.17 (Ar), 127.17, 126.21 (cumyl-C),
72.65 (NCH2CH2O), 69.70 (a-CH2 THF), 63.81 (CH2Ar), 59.87 (OCH3),
47.14 (NCH2CH2O), 42.49 (C(CH3)2), 31.46 (C(CH3)2), 27.58 (C(CH3)2),
24.81 (b-CH2 THF), 4.40 (Si(CH3)2) ppm; elemental analysis calcd (%)
for C61H81N2O4Si2Y: C 69.68, H 7.77, N 2.66; found: C 69.57, H 7.83, N
2.74.
evaporation of
a
saturated toluene solution at room temperature.
1H NMR (300 MHz, [D6]benzene, 258C): d = 7.23 (d, J(H,H) = 2.2 Hz,
2H; ArH), 6.18 (d, 4J(H,H) = 2.4 Hz, 2H; ArH), 5.21 (m, 2H; SiH-
(CH3)2), 3.93 (m, 4H; a-CH2 THF), 3.74 (d, 2J(H,H) = 11.9 Hz, 2H;
ArCH2), 2.98 (d, 2J(H,H) = 12.4 Hz, 2H; ArCH2), 2.84 (t, 3J(H,H) =
4.8 Hz, 4H; NCH2CH2O), 2.46 (s, 12H; adamantyl), 2.40 (s, 6H; CH3),
2.26 (s, 9H; adamantyl + OCH3), 2.03 (d, 2J(H,H) = 11.7 Hz, 6H; ada-
mantyl), 1.92 (d, 2J(H,H) = 11.3 Hz, 6H; adamantyl), 1.23 (m, 4H; b-
4
CH2 THF), 0.49 (d, 4J(H,H)
=
2.9 Hz, 12H; SiH(CH3)2) ppm;
13C{1H} NMR (75 MHz, [D6]benzene, 258C): d = 162.39, 129.78, 128.51
(Ar), 72.41 (NCH2CH2O), 69.60 (a-CH2 THF), 63.38 (CH2Ar), 58.43
(NCH2CH2O), 48.35 (OCH3), 40.64 (adamantyl), 37.40 (adamantyl),
37.06 (adamantyl), 29.09 (adamantyl), 24.36 (b-CH2 THF), 20.33 (CH3),
3.46 (Si(CH3)2) ppm; elemental analysis calcd (%) for C47H73N2O4Si2Y: C
64.50, H 8.41, N 3.20; found: C 64.72, H 8.27, N 3.18.
NMR-scale generation of 18 in [D8]THF: Dry propan-2-ol (0.67 mL,
0.0152 mmol, 1.0 equiv) was added by microsyringe at room temperature
to a solution of complex 17 (16.0 mg, 0.0152 mmol) in [D8]THF. The tube
Chem. Eur. J. 2006, 12, 169 – 179
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
177

J.-F. Carpentier et al.
was vigorously shaken and left at room temperature for 10 min. The reac-
tion was monitored by 1H NMR spectroscopy, which indicated quantita-
tive formation of 18, along with release of free amine HN(SiHMe2)2.
1H NMR (500 MHz, [D8]THF, 258C): d = 7.25 (m, 4H; cumyl), 7.20 (m,
8H; cumyl), 7.06 (m, 8H; cumyl), 6.92 (brs, 2H; ArH), 6.69 (brs, 2H;
could be found from the Fourier difference. Carbon-bound hydrogen
atoms were placed at calculated positions and forced to ride on the at-
tached carbon atom. The hydrogen atom contributions were calculated
but not refined. All non-hydrogen atoms were refined with anisotropic
displacement parameters. One of the methyl groups of one of the SiMe2
fragments is statistically disordered over two distinct positions. This disor-
der was introduced in the refinement procedure with the aid of the
PART instruction in SHELXL, which treated this methyl group as two
independent objects (carbon atoms labeled C502 and C512 and the corre-
sponding hydrogen atoms), with a refined occupancy of 62:38. The loca-
tions of the largest peaks in the final difference Fourier map calculation
and the magnitude of the residual electron densities were of no chemical
significance. The unit cell of 13 was found to contain one disordered mol-
ecule of toluene. Crystal data: C47H72N2O4Si2Y1·C7H8, Mr = 966.29, tri-
clinic, a = 12.453(5), b = 15.477(5), c = 16.117(5) Š, a = 63.756(5)8, b
= 81.559(5)8, g = 67.151(5)8, V = 2566.3(15) Š3, T = 150(2) K, space
ArH), 4.20 (brm, 1H; CH(CH3)2), 3.71 (d, 2J(H,H)
= 12.0 Hz, 2H;
ArCH2), 3.58 (m, 4H; THF), 3.40 (brs, 3H; OCH3), 3.01 (brs, 2H;
NCH2CH2O), 2.86 (d, 2J(H,H) = 12.0 Hz, 2H; ArCH2), 2.16 (brs, 2H;
NCH2CH2O), 2.04 (brs, 6H; CH3 cumyl), 1.77 (m, 4H; THF), 1.60 (brs,
12H; CH3 cumyl), 1.42 (s, 6H; CH3 cumyl), 1.15 (d, 3J(H,H) = 5.9 Hz,
6H; CH(CH3)2) ppm; 13C{1H} NMR (75 MHz, [D8]THF, 258C):
d =
162.14 (Cq Ar), 152.07 (Cq cumyl), 134.36 (cumyl-C), 127.38, 126.55
125.70, 125.08, 123.46 (Ar), 72.77 (NCH2CH2O), 69.70 (a-CH2 THF),
65.63 (CH(CH3)2) 63.72 (CH2Ar), 61.15 (OCH3), 49.75 (NCH2CH2O),
32.37 (C(CH3)2), 30.78 (C(CH3)2), 28.03 (CH(CH3)2), 26.27 (C(CH3)2),
24.81 (b-CH2 THF).
group P1 (no. 2), Z = 2, 1calcd = 1.250 gcmÀ3, crystal size = 0.40”0.25”
¯
NMR-scale generation of 18 in [D6]benzene: Dry propan-2-ol (0.67 mL,
0.0152 mmol, 1.0 equiv.) was added by microsyringe at room temperature
to a solution of 17 (16.0 mg, 0.0152 mmol) in [D6]benzene. The tube was
vigorously shaken and left at room temperature for 10 min. The reaction
was monitored by 1H NMR spectroscopy, which indicated quantitative
formation of 18, along with release of free amine HN(SiHMe2)2.
1H NMR (500 MHz, [D6]benzene, 258C): d = 7.63 (m, 4H; cumyl), 7.51
(m, 8H; cumyl), 7.44 (m, 8H; cumyl), 6.93 (brs, 2H; ArH), 6.79 (brs,
2H; ArH), 4.25 (brs, 1H; CH(CH3)2), 3.51 (d, 2J(H,H) = 11.97 Hz, 2H;
ArCH2), 3.27 (brs, 8H; THF), 2.94 (brs, 3H; OCH3), 2.41 (brs, 4H;
0.20 mm3, m(MoKa
) =
1.228 mmÀ1, 54735 reflections measured, 11817
unique and 8764 reflections with I>2s(I), which were used in all calcula-
tions. The final R1 was 0.0499 (observed data) and wR2 was 0.1500 (all
data).
CCDC-278313 contains the supplementary crystallographic data for this
paper. These data can be obtained free of charge from The Cambridge
Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
NCH2CH2O
+ ArCH2), 2.11 (s, 6H; CH3 cumyl), 1.99 (brs, 2H;
NCH2CH2O), 1.80 (brs, 12H; CH3 cumyl), 1.72 (s, 6H; CH3 cumyl), 1.32
(brs, 8H; THF), 1.07 (d, 3J(H,H) = 5.77 Hz, 6H; CH(CH3)2) ppm. A
minor series of resonances accounting for less than 5% of the total was
also observed.
Acknowledgement
The authors are grateful to the RØgion Bretagne (grant-in-aid COPO-
BIO), the Centre National de la Recherche Scientifique, and the French
Ministry for Higher Education and Research (MESR) for financial sup-
port. We thank S. Sinbandhit and N. Ajellal (University of Rennes 1) for
NMR experiments, and Dr. Guy Ricart (UniversitØ de Lille 1) for
MALDI-TOF MS analysis. A generous loan of meso-lactide from Purac
is gratefully acknowledged.
NMR-scale reaction between 8 and 2-propanol in [D8]THF: Dry propan-
2-ol (1.00 mL, 0.025 mmol, 1.0 equiv) was added by microsyringe at room
temperature to
a solution of complex 8 (20.0 mg, 0.025 mmol) in
[D8]THF. The tube was vigorously shaken and left at room temperature
for 10 min. The reaction was monitored by 1H NMR spectroscopy, which
indicated release of free amine HN(SiHMe2)2 and quantitative transfor-
mation of 8 into the corresponding isopropoxide complex. 1H NMR
(300 MHz, [D8]THF, 258C): d = 7.18 (d, 4J(H,H) = 2.2 Hz, 2H; ArH),
6.91 (d, 4J(H,H) = 2.2 Hz, 2H; ArH), 4.21 (m, 1H; CH(CH3)2), 4.16 (d,
2J(H,H) = 11.9 Hz, 2H; ArCH2), 3.59 (brs, 4H; a-CH2 THF, overlap
[1] R. E. Drumright, P. R. Gruber, D. E. Henton, Adv. Mater. 2000, 12,
1841–1846.
[2] a) W. Kuran, Prog. Polym. Sci. 1998, 23, 919–992; b) S. Mecking,
Angew. Chem. 2004, 116, 1096–1104; Angew. Chem. Int. Ed. 2004,
43, 1078–1085.
[3] a) E. Chiellini, R. Solaro, Adv. Mater. 1996, 8, 1375–1381; b) Y.
Ikada, H. Tsuji, Macromol. Rapid Commun. 2000, 21, 117–132;
c) M. Vert, S. N. Li, G. Spenlenhauer, P. J. Guerin, Pure Appl. Chem.
1995, A32, 787–796.
with solvent resonances), 3.40 (brs, 3H; OCH3), 3.16 (d, 2J(H,H)
=
12.2 Hz, 2H; ArCH2), 3.01 (t, 3J(H,H) = 5.3 Hz, 2H; NCH2CH2O), 2.56
(t, 3J(H,H) = 5.3 Hz, 2H; NCH2CH2O), 1.73 (brs, 4H; b-CH2 THF, over-
lap with solvent resonances), 1.49 (s, 18H; C(CH3)3), 1.26 (s, 18H; C-
3
(CH3)3), 1.16 (d, J(H,H) = 5.85 Hz, 6H; CH(CH3)2) ppm.
General polymerization procedure: A Schlenk flask was charged in a glo-
vebox with a solution of the initiator in THF, and a solution of the mono-
mer in the appropriate ratio in THF was added rapidly. The mixture was
immediately stirred with a magnetic stir bar at the desired temperature
for the desired reaction time. After a small sample of the crude material
had been removed with a pipette for characterization by 1H NMR, the
reaction was quenched with acidic methanol (0.5 mLof a 1.2 m HCl solu-
tion), and the polymer was precipitated with excess methanol. The poly-
mer was then dried under vacuum to constant weight.
[4] M. Okada, Prog. Polym. Sci. 2002, 27, 87–133.
[5] a) B. J. OꢁKeefe, M. A. Hillmyer, W. B. Tolman, J. Chem. Soc.
Dalton Trans. 2001, 2215–2224; b) O. Dechy-Cabaret, B. Martin-
Vaca, D. Bourissou, Chem. Rev. 2004, 104, 6147–6176.
[6] a) N. Spassky, M. Wisniewski, C. Pluta, A. Le Borgne, Macromol.
Chem. Phys. 1996, 197, 2627–2637; b) D. Jhurry, A. Bhaw-Luximon,
N. Spassky, Macromol. Symp. 2001, 175, 67–79; c) M. H. Chisholm,
N. W. Eilerts, Chem. Commun. 1996, 853–854; d) M. H. Chisholm,
N. W. Eilerts, J. C. Huffman, S. S. Iyer, M. Pacold, K. Phomphrai, J.
Am. Chem. Soc. 2000, 122, 11845–11854; e) M. H. Chisholm, J. C.
Huffman, K. Phomphrai, J. Chem. Soc. Dalton Trans. 2001, 222–
224; f) M. H. Chisholm, J. Gallucci, K. Phomphrai, Inorg. Chem.
2002, 41, 2785–2794; g) M. Cheng, A. B. Attygalle, E. B. Lobkovsky,
G. W. Coates, J. Am. Chem. Soc. 1999, 121, 11583–11584; h) B. M.
Chamberlain, M. Cheng, D. R. Moore, E. B. Lobkovsky, T. M. Ovitt,
G. W. Coates, J. Am. Chem. Soc. 2001, 123, 3229–3238; i) G. W.
Coates, M. Cheng, B. M. Chamberlain, WO 0134555, 2001; j) T. M.
Ovitt, G. W. Coates, J. Am. Chem. Soc. 2002, 124, 1316–1326; k) Z.
Zhong, P. J. Dijkstra, J. Feijen, J. Am. Chem. Soc. 2003, 125, 11291–
11298; l) N. Nomura, R. Ishii, M. Akakura, K. Aoi, J. Am. Chem.
Soc. 2002, 124, 5938–5939; m) K. Majerska, A. Duda, J. Am. Chem.
General kinetic procedure: The appropriate amount of monomer in THF
(1000 equiv., [LA] = 1.0m) was added to a solution of initiator in THF
(5 mL). The mixture was then stirred at 258C under N2. At appropriate
time intervals, 0.5 mLaliquots were removed and opened to air. The ali-
quots were then dried to constant weight in vacuo and analyzed by
1H NMR spectroscopy.
Crystal structure determination of 13: A suitable single crystal of 13 was
mounted on a glass fiber by the “oil-drop” method. Diffraction data were
collected at 150 K by use of an APEX 2 AXS-Bruker diffractometer with
graphite monochromatized Mo-Ka radiation (l = 0.71073 Š). The crystal
structure was solved by direct methods, remaining atoms being located
from difference Fourier synthesis, followed by full-matrix, least-squares
refinement based on F2 (program SIR-97).[30] Many hydrogen atoms
178
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 169 – 179

Ring-Opening Polymerization of Lactide with Group 3 Metal Complexes
FULL PAPER
Soc. 2004, 126, 1026–1027; n) C.-X. Cai, A. Amgoune, C. W. Leh-
mann, J.-F. Carpentier, Chem. Commun. 2004, 330–331.
1996, 29, 1375–1381; c) G. Schwach, J. Coudane, R. Engel, M. Vert,
J. Polym. Sci., A: Polym. Chem. 1997, 35, 3431–3440; d) A. Kowal-
ski, J. Libiszowski, A. Duda, S. Penczek, Macromolecules 2000, 33,
1964–1971; e) A. Finne, A.-C. Albertsson, Biomacromolecules 2002,
3, 684–690.
[7] a) M. H. Chisholm, E. E. Delbridge, New J. Chem. 2003, 27, 1177–
1183; b) H. Ma, J. Okuda Macromolecules 2005, 38, 2665–2673.
[8] a) W. J. Burke, R. P. Smith, C. Weatherbee, J. Am. Chem. Soc. 1952,
74, 602–605; b) W. J. Burke, M. J. Kolbezen, C. W. Stephens, J. Am.
Chem. Soc. 1952, 74, 3601–3605; c) W. J. Burke, E. L. M. Glennie,
C. Weatherbee, J. Org. Chem. 1964, 29, 909–912.
[9] a) C.-X. Cai, L. Toupet, C. W. Lehmann, J.-F. Carpentier, J. Organo-
met. Chem. 2003, 683, 131–136; b) H. Ma, T. P. Spaniol, J. Okuda,
Dalton Trans. 2003, 4770–4780; c) P. N. OꢁShaughnessy, P. D.
Knight, C. Morton, K. M. Gillespie, P. Scott, Chem. Commun. 2003,
1770–1771; d) O. Runte, T. Priermeier, R. Anwander, Chem.
Commun. 1996, 1385–1386; e) D. V. Gribkov, K. C. Hultzsch, F.
Hampel, Chem. Eur. J. 2003, 9, 4796–4810.
[10] a) W. A. Herrmann, J. Eppinger, M. Spiegler, O. Runte, R. Anwan-
der, Organometallics 1997, 16, 1813–1815; b) J. Eppinger, M. Spie-
gler, W. Hieringer, W. A. Herrmann, R. Anwander, J. Am. Chem.
Soc. 2000, 122, 3080–3096.
[11] P. B. Hitchcock, M. F. Lappert, A. Singh, J. Chem. Soc. Chem.
Commun., 1983, 1499–1501.
[21] For examples of Al-based alkoxide initiators, see: a) P. Dubois, C.
Jacobs, R. JØrôme, P. TeyssiØ, Macromolecules 1991, 24, 2266–2270;
b) P. A. Cameron, D. Jhurry, V. C. Gibson, A. J. P. White, D. J. Wil-
liams, S. Williams, Macromol. Rapid Commun. 1999, 20, 616–618;
c) T. M. Ovitt, G. W. Coates, J. Am. Chem. Soc. 1999, 121, 4072–
4073; d) C. P. Radano, G. L. Baker, M. R. Smith, J. Am. Chem. Soc.
2000, 122, 1552–1553.
[22] For examples of Fe-based alkoxide initiators, see: a) M. Stolt, A. So-
dergard, Macromolecules 1999, 32, 6412–6417; b) B. J. OꢁKeefe,
S. M. Monnier, M. A. Hillmyer, W. B. Tolman, J. Am. Chem. Soc.
2001, 123, 339–340; c) B. J. OꢁKeefe, L. E. Breyfogle, M. A. Hillmy-
er, W. B. Tolman, J. Am. Chem. Soc. 2002, 124, 4384–4393.
[23] For examples of Ln-based alkoxide initiators, see: a) W. M. Stevels,
M. J. K. AnkonØ, P. J. Dijkstra, J. Feijen, Macromolecules 1996, 29,
3332–3333; b) W. M. Stevels, M. J. K. AnkonØ, P. J. Dijkstra, J.
Feijen, Macromolecules 1996, 29, 6132–6138; c) B. M. Chamberlain,
Y. Sun, J. R. Hagadorn, E. W. Hemmesch, V. G. Young, M. Pink,
M. A. Hillmyer, W. B. Tolman, Macromolecules 1999, 32, 2400–
2402; d) B. M. Chamberlain, B. A. Jazdzewski, M. Pink, M. A. Hill-
myer, W. B. Tolman, Macromolecules 2000, 33, 3970–3977; e) G. R.
Giesbrecht, G. D. Whitener, J. Arnold, J. Chem. Soc. Dalton Trans.
2001, 923–927; f) M. Save, M. Schappacher, A. Soum, Macromol.
Chem. Phys. 2002, 203, 889–899; g) V. Simic, N. Spassky, L. G.
Hubert-Pfalzgraf, Macromolecules 1997, 30, 7338–7340.
[12] a) The low-temperature (213–253 K) 1H NMR spectra were quite
complicated, featuring several sets of resonances, and proved unin-
formative; b) Tetradentate coordination of 3 to Yb and Er in the
solid state has recently been reported; see Y. Yao, M. Ma, X. Xu, Y.
Zhang, Q. Shen, W.-T. Wong, Organometallics 2005, 24, 4014–4020.
[13] A. Amgoune, C. M. Thomas, E. Balnois, Y. Grohens, P. J. Lutz, J.-F.
Carpentier, Macromol. Rapid Commun. 2005, 26, 1145–1150.
[14] a) K. A. M. Thakur, R. T. Kean, E. S. Hall, J. J. Kolstad, T. A. Lindg-
ren, M. A. Doscotch, J. I. Siepmann, E. J. Munson, Macromolecules
1997, 30, 2422–2428; b) K. A. M. Thakur, R. T. Kean, E. S. Hall, J. J.
Kolstad, E. J. Munson, Macromolecules 1998, 31, 1487–1494;
c) K. A. M. Thakur, R. T. Kean, M. T. Zell, B. E. Padden, E. J.
Munson, Chem. Commun. 1998, 1913–1914; d) M. H. Chisholm,
S. S. Iyer, D. G. McCollum, M. Pagel, U. Werner-Zwanziger, Macro-
molecules 1999, 32, 963–973.
[15] E. L. Marshall, V. C. Gibson, H. S. Rzepa, J. Am. Chem. Soc. 2005,
127, 6048–6051.
[16] S. Groysman, I. Goldberg, M. Kol, E. Genizi, Z. Goldschmidt, Orga-
nometallics 2004, 23, 1880–1890.
[17] a) V. Busico, R. Cipullo, N. Friederichs, S. Ronca, G. Talarico, M.
Togrou, B. Wang, Macromolecules 2004, 37, 8201–8203; b) L. Resco-
ni, L. Abis, G. Franciscono, Macromolecules 1992, 25, 6814–6817;
c) M. Mitani, R. Furuyama, J.-I. Mohri, J. Saito, S. Ishii, H. Terao, T.
Nakano, H. Tanaka, T. Fujita J. Am. Chem. Soc. 2003, 125, 4293–
4305; d) B. L. Small, M. Brookhart, Macromolecules 1999, 32, 2120–
2130.
[24] E. Martin, P. Dubois, R. JØrôme, Macromolecules 2000, 33, 1530–
1535.
[25] The high sensitivity of complex 18 limited its analysis to NMR tech-
niques. Attempts to obtain diffraction-quality single crystals of any
of the cumyl-substituted bisphenolate complexes have so far been
unsuccessful.
[26] Only trace amounts (<5%) of a second species were observed in
1
this reaction; the unsymmetrical pattern of H NMR resonances sug-
gests it is a dimer or an aggregated species. Complex 18 loses its
THF ligand upon drying in vacuum to give an unsymmetrical com-
plex as well, as indicated by 1H NMR spectroscopy in [D6]benzene;
see reference [9b].
[27] S. H. Ong, J. Chem. Soc. Chem. Commun. 1970, 1180.
[28] a) E. Y. Tshuva, I. Goldberg, M. Kol, H. Weitman, Z. Goldschmidt,
Chem. Commun. 2000, 379–380; b) E. Y. Tshuva, I. Goldberg, M.
Kol, Z. Goldschmidt, Organometallics 2001, 20, 3017–3028; c) E. Y.
Tshuva, S. Groysman, I. Goldberg, M. Kol, Z. Goldschmidt, Orga-
nometallics 2002, 21, 662–670.
[18] R. Anwander, Top. Organomet. Chem. 1999, 2, 1–61.
[19] a) H. R. Kricheldorf, M. Berl, N. Scharnagl, Macromolecules 1988,
21, 286–293; b) S. J. McLain, T. M. Ford, N. E. Drysdale, N. Jones,
E. F. McCord, J. Shreeve, W. J. Evans, Polym. Prepr. 1994, 35, 534–
535; c) W. M. Stevels, M. J. K. AnkonØ, P. J. Dijkstra, J. Feijen, Mac-
romol. Chem. Phys. 1995, 196, 1155–1161; d) P. Dubois, R. JØrôme,
P. TeyssiØ, Polym. Prepr. 1994, 35, 536–537; e) M. Mçller, F. Neder-
berg, L. S. Lim, R. Kꢂnge, C. J. Hawker, J. L. Hedrick, Y. Gu, R.
Shah, N. L. Abbott, J. Polym. Sci., A: Polym. Chem. 2001, 39, 3529–
3538.
[29] a) M. F. Lappert, R. Pearce, J. Chem. Soc. Chem. Commun. 1973,
126; b) K. C. Hultzsch, P. Voth, K. Beckerle, T. P. Spaniol, J. Okuda,
Organometallics 2000, 19, 228–243; c) R. Anwander, O. Runte, J.
Eppinger, G. Gerstberger, E. Herdtweek, M. Spiegler, J. Chem. Soc.
Dalton Trans. 1998, 847–858.
[30] a) A. Altomare, M. C. Burla, M. Camalli, G. Cascarano, C. Giaco-
vazzo, A. Guagliardi, A. G. G. Moliterni, G. Polidori, R. Spagna, J.
Appl. Crystallogr. 1999, 32, 115–119; b) G. M. Sheldrick,
“SHELXL-97”, Universitꢃt Gçttingen, Gçttingen, Germany, 1999.
[20] For examples of Sn-based alkoxide initiators, see: a) A. J. Nijenhuis,
D. W. Grijpma, A. J. Pennings, Macromolecules 1992, 25, 6419–
6424; b) H. R. Kricheldorf, S.-R. Lee, S. Bush, Macromolecules
Received: July 20, 2005
Published online: October 14, 2005
Chem. Eur. J. 2006, 12, 169 – 179
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
179