Stereoselective controlled polymerization of DL-lactide with [Ti(trisphenolate)O-i-Pr]2 initiators

Article abstract of DOI:10.1021/ma051811w

The potential of stereoselective controlled polymerization of DL-Lactide with [Ti(trisphenolate)O-i-Pr]₂ initiators was analyzed. The anhydrous solvent used in polymerization were passed through a plug of alumina immediately prior to use. The e

Full text of DOI:10.1021/ma051811w

10336
Macromolecules 2005, 38, 10336-10340
Stereoselective Controlled Polymerization of
DL-Lactide with [Ti(trisphenolate)O-i-Pr]2
Initiators
Sarah K. Russell,† Carolyn L. Gamble,‡
Karen J. Gibbins,† Kirsten C. S. Juhl,‡
William S. Mitchell, III,† Aistis J. Tumas,† and
Gretchen E. Hofmeister*,†
Department of Chemistry, Carleton College,
Northfield, Minnesota 55057, and Department of Chemistry,
Gustavus Adolphus College, St. Peter, Minnesota 56083
Received August 16, 2005
Revised Manuscript Received September 28, 2005
Biodegradable polymers, particularly poly(lactic acid)
(PLA), have become important industrial commodities1,2
as well as a focus of research efforts aimed at developing
more selective polymerization catalysts.3 One approach
that has been successful in controlling the chain length
and stereochemistry of PLA is the metal alkoxide-
initiated ROP of the cyclic condensed dimer of lactic acid
(LA). Controlled polymerization reactions have been
discovered for a number of well-defined metal alkoxide
complexes,4 including those of aluminum,5,6 calcium,7
magnesium,8-10 lanthanides,11 titanium,12,13 and
zinc,9,10,14,15 so that isotactic PLA can be prepared
starting from L-LA (Figure 1a). Enantiomerically pure
or racemic chiral aluminum(III) alkoxide initiators have
enabled the preparation of isotactic stereoblock PLA
from rac-LA16,17 (Figure 1b) and syndiotactic PLA from
meso-LA,17 presumably via enantiomorphic site control
(ESC). The ROP of rac-LA using achiral metal alkoxide
complexes generally favors heterotactic PLA (Figure 1c)
via chain-end control (CEC),10,18-20 the exception being
achiral aluminum salen21 or salan20 complexes, which
yield isotactic stereoblock PLA via CEC. These recent
advances and the commercial utility of PLA indicate
that further research in this area is valuable, both for
improving synthetic methods and for better understand-
ing the mechanisms and constraints of ROP catalytic
systems.
Our group has been characterizing titanium(IV) and
aluminum(III) trisphenoxide compounds, with the goal
of preparing well-defined single site Lewis acid catalysts
that can be used in asymmetric synthesis or stereose-
lective polymerization reactions. We have reported the
solution- and solid-state structures of dimeric titanium-
(IV) alkoxide complexes of trisphenols 1a,b (2a,b, 3a,
eq 1 of Scheme 1),22 which possess structural features
that encouraged us to explore their reactivity in ROP
reactions. The four-coordinate titanium centers in these
compounds retain one terminal alkoxide ligand, which
is ideal for ROP via the coordination-insertion mech-
anism established for known metal alkoxide initiators
(eq 2 of Scheme 1).5,23 In addition, the trisphenolate
adopts a chiral conformation when coordinated to
titanium, resulting in chiral (although racemic) tita-
nium centers. We decided to evaluate the reactivity of
these complexes by investigating the ROP of LA. We
Figure 1. Stereoregular poly(lactic acid) microstructures from
(a) L-lactide and (b, c) rac-lactide (D-lactide + L-lactide).
Scheme 1
envisioned that if the chirality at titanium were pre-
served under the ROP reaction conditions, rac-LA could
polymerize with isotactic stereoselectivity via ESC.
Titanium alkoxides are attractive targets to investi-
gate in ROP reactions because the low toxicity of
titanium minimizes concerns regarding its presence in
commercial PLA products. At the time that we began
this work, there were no published accounts of titanium
initiators for the controlled ROP of LA, although Aida,
Endo, and co-workers had described similar titanium
bis(phenolate)s in the controlled polymerization of ꢀ-ca-
prolactone and cyclic carbonates.24 Subsequently, Verk-
ade and co-workers12 and Harada and co-workers13
described the controlled ROP of LA using different
titanium tris- and bis(phenolate)s, respectively. In this
report, we describe the ROP of LA by 2a and 2b, which
result in well-behaved catalytic systems, comparable to
these recently described titanium initiators.12,13 In
addition, 2a and 2b show a greater degree of stereo-
control in the ROP of rac-LA than has been observed
for previous titanium initiators.25
Results and Discussion
Our initial studies examined the potential of complex
2a as a ROP initiator for LA under different reaction
conditions and indicated that 2a produces a well-
controlled catalyst in solution, yielding PLA that retains
the stereochemical integrity of the starting monomer.
The reaction of L-LA with 2a at 80 °C results in close
to monodisperse, stereochemically pure isotactic poly-
† Carleton College.
‡ Gustavus Adolphus College.
* To whom correspondence should be addressed: e-mail
ghofmeis@carleton.edu.
10.1021/ma051811w CCC: $30.25 © 2005 American Chemical Society
Published on Web 10/29/2005

Macromolecules, Vol. 38, No. 24, 2005
Table 1. Polymerization of Lactide (LA) Using 2a, 2b, or 3a as Initiatora
Notes 10337
d
e
entry initiator monomer
solvent
concn (M) temp (°C) time (h) convb (%) Mn × 10-3 c Mn × 10-3 d Mw/Mn
Ps
1
2
3
4
5
6
7
8
9
2a
2a
2a
2a
2a
2a
2a
2b
3a
L-LA
toluenef
CH2Cl2
CH2Cl2
CH2Cl2
toluenef
toluenef
toluenef
CH2Cl2
CH2Cl2
0.63
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.26
80
60
312
72
12
24
72
46
24
6
75
95
91
92
96
96
97
89
87
10.9
13.8
13.2
13.3
13.9
13.9
14.0
12.9
12.6
24.3
20.1
19.7
19.9
24.1
18.5
18.4
20.6
22.6
1.19
1.06
1.06
1.07
1.10
1.10
1.21
1.05
1.69
rac-LA
rac-LA
rac-LA
rac-LA
rac-LA
rac-LA
rac-LA
rac-LA
0.82
0.79
0.81
0.80
80
100
60
80
100
80
80
0.80
72
a [LA]/[Ti-O-i-Pr] ) 100. b Determined by integration of the methine region in the 1H NMR spectrum of the crude product. c Calculated
using percent conversion, assuming that each isopropoxide is an initiator and that each polymer chain is terminated by a proton: (%
conv) × (144.13) + (60.1) g/mol. d Size exclusion chromatography relative to polystyrene standards. e Ps is the probability of syndiotactic
addition of monomers and is determined from the methine region of the homonuclear decoupled 1H NMR spectrum. f 7:1 toluene:CH2Cl2.
Figure 3. Semilogarithmic plot of rac-lactide (LA) conversion
with time for initiators 2a (diamonds) and 2b (circles); [LA]0
) 1.00 M, [LA]/[Ti-O-i-Pr] ) 100, CH2Cl2, 80 °C; [LA]eq
measured as 0.07 M. A linear least-squares fit of each data
set resulted in the lines and equations shown.
Figure 2. Plot of poly(rac-lactide) Mn (squares, size exclusion
chromatography relative to polystyrene standards) and poly-
dispersity (circles) as a function of conversion, with rac-lactide
(LA) and 2a (CH2Cl2, 80 °C, [LA]/[Ti-O-i-Pr] ) 100).
throughout the polymerization (Figure 2), are indicative
of a well-controlled polymerization. At longer reaction
times (e.g., 72 h), SEC reveals a bimodal molecular
weight distribution accompanied by an increase in the
PDI (i.e., final data point in Figure 2), which we
attribute to intermolecular chain transfer (via trans-
esterification) reaction rates becoming significant when
the polymerization rate is diminished at high conver-
sion.30 A semilogarithmic plot of conversion vs time for
initiator 2a shows a first-order dependence of monomer
conversion on [LA], which is consistent with a coordina-
tion-insertion mechanism (Figure 3, kapp ) 0.52 h-1).31
The plot also reveals that there is an induction period
before 2a initiates the ROP reaction (vide infra).
Evidence that the isopropoxide ligand is transferred
to the end of the growing polymer chain during initia-
tion was obtained by preparing oligomers of LA and
determining their mass distribution by electrospray
mass spectrometry (ESI-MS).32 The masses of the
observed ions correspond to sodium-cationized oligomers
of LA, which have been initiated by the addition of O-i-
Pr and terminated by protonation (m/z ) [H(OC(H)MeC-
(dO))2nO-i-Pr + Na]+, where n ) 3-13).33 MS-MS data
obtained by collisional dissociation of m/z ) 803 (n ) 5)
reveal the loss of isopropene (m/z ) 761), in addition to
losses of (C(H)MeC(dO)O)n (n ) 1-3, m/z ) 689, 617,
545).32
mer26 (Table 1, entry 1). Under similar conditions, rac-
LA also yields nearly monodisperse polymer and shows
a high probability for syndiotactic addition (Ps ) 0.79,
Table 1, entry 3),27,28 rather than the anticipated
isotactic selectivity (vide infra). Consistent with this
observed stereoselectivity, rac-LA polymerizes in far less
time than L-LA (cf. Table 1, entries 1 and 6).
After exploring several different reaction conditions
for this polymerization, we concluded that changes in
temperature or solvent affect the ROP rate but have
very little effect on the stereoregularity or polydispersity
of the resultant polymer (cf. Table 1, entries 2-4 and
5-7). At 40 °C, the rate of polymerization is very slow
(i.e., trace polymerization within 24 h), while temper-
atures of 60-100 °C result in reasonable rates of
polymerization. We routinely observe excellent poly-
dispersities and very good stereoselectivities under the
conditions described in Table 1. It appears that the
molecular weight control is better in dichloromethane
than in toluene, as is evidenced by the consistent Mn
values and smaller PDIs in entries 2-4, as compared
with entries 5-7 (Table 1). Therefore, we elected to
perform further kinetic and mechanistic studies at 80
°C in dichloromethane.
The ROP of rac-LA using 2a as an initiator was
1
monitored by H NMR spectroscopy and size exclusion
chromatography (SEC) periodically over the course of
the reaction, and plots of molecular weight (Mn) and PDI
vs percent conversion were constructed (Figure 2).29 The
linear correlation between the polymer molecular weight
and monomer conversion, coupled with very low PDIs
To determine whether both of the Ti-O-i-Pr ligands
in 2a are initiating, we integrated the isopropoxide
methyl resonances relative to the PLA methine reso-
nances in the 1H NMR spectrum of these oligomers.
These measurements reveal a 12:1 ratio of LA:O-i-Pr

10338 Notes
Macromolecules, Vol. 38, No. 24, 2005
in the NMR sample, which agrees reasonably well with
the 10:1 LA:Ti-O-i-Pr reaction stoichiometry and the
results of the ESI-MS experiment. Thus, it appears that
the majority of the titanium isopropoxide ligands are
initiating this reaction by transfer to the carboxyl group
of the first inserted LA.
carbon-phenol carbon bonds and should be easier to
achieve with a smaller para substituent on phenol. We
have evidence that isomerization of this type occurs in
these systems over a period of hours at room tempera-
ture.22 A second possibility is that the dimeric complexes
2a and 2b are either dissociating to monomers or
associating to form tetramers or other higher-order
species. Associative processes should be faster for the
less-hindered complex 2b, and both associative and
dissociative processes would require some reorganiza-
tion of the ligand, which again should be faster for the
less hindered trisphenolate. Although the reason for
these kinetic differences is still unclear, it is remarkable
that steric differences in ligands, which apparently have
little effect on the metal coordination environment,
significantly impact the kinetics of polymerization.
Despite these differences in induction, complexes 2a
and 2b show the same degree of selectivity for syndio-
tactic addition, rather than the isotactic addition that
would be more consistent with ESC. Syndiotactic pref-
erence has been commonly observed in the ROP of rac-
LA with sterically hindered achiral initiators, arising
from a CEC mechanism.3,18,19 The degree of syndiotactic
selectivity has also been shown to correlate with the
increased size of the ligands at the metal center.10,19,20
The heterotactic bias that we observe in our systems
suggests that our titanium trisphenolate complexes,
even though the titanium centers are formally chiral,
are influencing the ROP reaction as if they were simply
sterically hindered achiral catalysts. Furthermore, our
kinetic studies suggest that the trisphenolate ligand
may be conformationally isomerizing under the ROP
reaction conditions; this makes it difficult to envision
stereocontrol via ESC. Therefore, we attribute our
observed stereoselectivity to CEC.
The mass spectra also reveal that there is little
intermolecular chain transfer occurring in the early
stages of polymerization because the observed masses
correspond to whole number multiples of LA. However,
long reaction times (e.g., 72 h) produce lower intensity
ions for masses that represent multiples of 0.5 × LA
(m/z ) 587, 731, 875, 1019, etc.),32 a phenomenon that
has been observed in the ROP of LA using aluminum
isopropoxide initiators and attributed to intermolecular
chain transfer.34 Consistent with this observation, the
stereoselectivity is unchanged through the early stages
of ROP (Ps ) 0.79 up to 12 h), but at long reaction times
the polymer stereoregularity begins to erode (e.g., Ps )
0.75 at 72 h). Altogether, the evidence indicates that
chain transfer reactions become important at late stages
in this ROP reaction.
In addition to 2a, we have evaluated complexes 2b
and 3a as initiators for the ROP of rac-LA. We predicted
that complex 3a would be relatively slow to initiate
because the transfer of the O-t-Bu substituent to the
first inserted LA monomer would be hindered by the
sterically larger alkoxide. In fact, complex 3a yields
polymer with much broader PDIs (Table 1, entry 9) and
a distribution skewed toward lower molecular weight,
indicating slow initiation. The overall rates of conversion
are also much slower using 3a. Together, these data
confirm that initiation is more difficult with large
alkoxides, which has also been observed in other metal
alkoxide ROP reactions.3
In conclusion, complexes 2a and 2b are initiators for
the controlled polymerization of rac-LA, providing mono-
disperse PLA with a higher degree of syndiotactic
addition than has been observed for other titanium
initiators. The reaction shows a first-order dependence
on [LA], consistent with a coordination-insertion mech-
anism, and the titanium isopropoxide ligand is trans-
ferred to the end of the growing polymer chain during
initiation. Both complexes 2a and 2b have the same
apparent rate constant for polymerization and show the
same degree of stereoselectivity; however, reactions with
2a have a significant (ca. 0.5 h) induction period,
whereas those with 2b do not. This difference in
induction is attributed to the steric size of the para
substituent on the central phenol ring, suggesting that
a conformational isomerization of the trisphenolate
ligand is necessary to produce the active catalyst. The
stereoselectivity that we observe is high, although the
bias toward syndiotactic addition suggests that the
chirality at titanium, if it is retained under the reaction
conditions, does not provide good ESC. Instead, it
appears that the titanium trisphenolate complexes are
enhancing the stereoselectivity analogous to sterically
hindered achiral initiators, via a CEC mechanism.
We anticipated that initiators 2a and 2b would give
comparable polymerization rates because the steric
difference between these complexes is at a site distal to
the presumed active site of catalysis. In fact, the
apparent rate constant for initiator 2b (kapp ) 0.53 h-1
,
Figure 3) is identical (within the error of our measure-
ments) to that of 2a, although 2b does not appear to
require an induction period before initiating ROP
(Figure 3). The ROP using initiator 2b is also well
controlled, as is evidenced by a linear correlation
between molecular weight and conversion32 and narrow
PDIs at high conversion (Table 1, entry 8). The struc-
1
tures of 2a and 2b are almost identical by H NMR
spectroscopy,22,32 so the lack of an induction period for
2b appears to be related solely to the steric size of the
para substituent on the central ring. However, we
believe that both initiators must undergo the same kind
of isomerization or reorganization process before initiat-
ing ROP because the active catalysts arising from 2a
and 2b appear to be virtually identical; not only are the
apparent rate constants for the two essentially the
same, but also they show identical stereoselectivities in
the ROP of rac-LA (Table 1, entries 3 and 8). Therefore,
the induction process must simply be more facile for the
smaller para-methyl derivative because the period of
induction was not observed for initiator 2b in our kinetic
studies.
Experimental Section
General. All reagents and solvents were purchased from
commercial suppliers and used as received, unless noted
otherwise. Anhydrous solvents were purchased in Sure/Seal
bottles from Aldrich Chemical Co., transferred into an M.
Braun glovebox without exposure to air, and used as received,
except as specifically noted. Ethyl acetate was distilled from
calcium hydride under a nitrogen atmosphere and transferred
In our view, two possibilities emerge to explain these
interesting results. One is that the trisphenolate ligand
undergoes a conformational change during induction,
such as the apparent inversion of a bridging methylene
carbon, which necessitates rotations around methylene

Macromolecules, Vol. 38, No. 24, 2005
Notes 10339
into the glovebox without exposure to air. Lactide (L and DL)
was sublimed at reduced pressure (0.1 mmHg) and recrystal-
lized from ethyl acetate in the glovebox. Titanium complexes
2a and 3a were prepared as described in the literature.22
Complex 2b was prepared in an analogous fashion.32
M. Frick and Mr. Nathaniel A. Lynd for assistance with
SEC measurements in the Hillmyer laboratory at the
University of Minnesota. We acknowledge NSF-CCLI
and Carleton College’s Howard Hughes Medical Insti-
tute grants for support of the ESI-MS.
NMR spectra were recorded on a Varian Unity Plus spec-
trometer operating at 400 MHz (1H) in either CDCl3 or C6D6.
SEC was performed either using a Waters Breeze HPLC
system equipped with a Waters 2414 refractive index detector
or a Hewlett-Packard 1100 series chromatograph equipped
with a Hewlett-Packard 1047A refractive index detector. Both
systems were equipped with three Jordi Gel DVB columns of
104, 103, and 500 Å pore sizes, and THF was used as the mobile
phase (either 35 °C (Waters) or 40 °C (H-P) at 1.0 mL/min).
Column calibration was performed with polystyrene standards
(Waters or Polymer Laboratories).
Supporting Information Available: Figures S1-S4 and
additional experimental details. This material is available free
of charge via the Internet at http://pubs.acs.org.
References and Notes
(1) Ritter, S. K. Chem. Eng. News 2002, 80, 26-30.
(2) NatureWorks LLC is a joint venture by Cargill-Dow to
develop new consumer products from PLA. See: http://
www.natureworksllc.com/corporate/nw_pack_home.asp and
http://www.ingeofibers.com/ingeo/home.asp.
Polymerization Reactions. The following is a typical
polymerization procedure. The anhydrous solvents used in
polymerization were passed through a plug of alumina im-
mediately prior to use. In the glovebox, a solution of DL-LA
(0.721 g, 5.00 mmol) and 2a (0.500 mL of a [Ti-O-i-Pr] ) 0.100
M solution in CH2Cl2) was prepared in CH2Cl2 at room
temperature in a 5 mL volumetric flask. This solution was
divided between two pressure reaction tubes (Ace no. 8648,
15 mL), which were then introduced into an oil bath, the
temperature of which was regulated to 80 ( 1 °C. After 12 h,
a pressure tube was removed from the heat and cooled in an
ice bath for 15 min to stop polymerization, and the reaction
solution was concentrated by rotary evaporation at room
temperature to an orange residue. The extent of conversion
(91%) was determined by obtaining a 1H NMR spectrum of
the residue and integrating the methine resonances of the
polymer (δ 5.10 ppm, C6D6) relative to the methine resonance
of the monomer (δ 3.65 ppm, C6D6). A sample for SEC analysis
was prepared by dissolving a portion of the residue in THF (3
mg/mL) and adding 2 drops of methanol (Mn ) 19.708 kDa,
Mw/Mn ) 1.06). Stereochemical analysis was performed by
analyzing the methine region of a homonuclear decoupled 1H
NMR spectrum of the polymer (CDCl3, Ps ) 0.79).28,35 The
polymer was precipitated in methanol at 0 °C (83% yield). 1H
NMR (C6D6) δ: 5.11 (m, 1 H), 1.40 (m, 2 H), 1.29 (m, 1 H).
Kinetic Studies. The following is a typical procedure for
determining the conversion, Mn, and PDI during the course of
the polymerization reaction. The general conditions for setting
up polymerization reactions were identical to those described
above. A CH2Cl2 solution of DL-lactide (1.00 M) and initiator
2a ([Ti-O-i-Pr] ) 0.0100 M) was prepared in a 10 mL
volumetric flask and divided approximately equally among 10
pressure reaction tubes. The reaction tubes were removed from
the glovebox and arrayed in a temperature-controlled (80 ( 1
°C) oil bath. At specified time intervals, tubes were removed
from the oil bath and cooled to 0 °C for 10 min, after which
the solutions were concentrated to orange residues. The
residues were analyzed by 1H NMR spectroscopy to determine
conversion, as described above. Samples for SEC were pre-
pared as described above. Plots of Mn and PDI vs conversion
and semilogarithmic plots of conversion vs time for initiators
2a, 2b, and 3a were obtained. Since this ROP reaction is
reversible, the logarithmic expression, (ln([LA]0 - [LA]eq) -
ln([LA] - [LA]eq)), was used for the kinetic plots. The equilib-
rium concentration, [LA]eq, was determined to be 0.07 M at
80 °C and [LA]0 ) 1.0 M.36
(3) Dechy-Cabaret, O.; Martin-Vaca, B.; Bourissou, D. Chem.
Rev. 2004, 104, 6147-6176.
(4) O’Keefe, B. J.; Hillmyer, M. A.; Tolman, W. B. J. Chem. Soc.,
Dalton Trans. 2001, 2215-2224 and references therein.
(5) Dubois, Ph.; Jacobs, C.; Jerome, R.; Teyssie, Ph. Macro-
molecules 1991, 24, 2266-2270.
(6) (a) Aida, T.; Inoue, S. Acc. Chem. Res. 1996, 29, 39-48. (b)
Kowalski, A.; Duda, A.; Penczek, S. Macromolecules 1998,
31, 2114-2122. (c) Liu, Y. C.; Ko, B. T.; Lin, C. C.
Macromolecules 2001, 34, 6196-6201.
(7) (a) Zhong, Z.; Dijkstra, P. J.; Birg, C.; Westerhausen, M.;
Feijen, J. Macromolecules 2001, 34, 3863-3868. (b) Zhong,
Z.; Schneiderbauer, S.; Dijkstra, P. J.; Westerhausen, M.;
Feijen, J. Polym. Bull. (Berlin) 2003, 51, 175-182.
(8) (a) Chisholm, M. H.; Eilerts, N. W. Chem. Commun. 1996,
853-854. (b) Shueh, M. L.; Wang, Y. S.; Huang, B. H.; Kuo,
C. Y.; Lin, C. C. Macromolecules 2004, 37, 5155-5162. (c)
Eifler, J.; Kobylka, M.; Jerzykiewicz, L. B.; Sobota, P. Dalton
Trans. 2005, 2047-2050.
(9) Chisholm, M. H.; Eilerts, N. W.; Huffman, J. C.; Iyer, S. S.;
Pacold, M.; Phomphrai, K. J. Am. Chem. Soc. 2000, 122,
11845-11854.
(10) Chamberlain, B. M.; Cheng, M.; Moore, D. R.; Ovitt, T. M.;
Lobkovsky, E. B.; Coates, G. W. J. Am. Chem. Soc. 2001,
123, 3229-3238.
(11) (a) Stevels, W. M.; Ankone´, M. J. K.; Dijkstra, P. J.; Feijen,
J. Macromolecules 1996, 29, 6132-6138. (b) Ma, H.; Spaniol,
T. P.; Okuda, J. Dalton Trans. 2003, 4770-4780. (c) Ma,
H.; Okuda, J. Macromolecules 2005, 38, 2665-2673.
(12) (a) Kim, Y.; Verkade, J. G. Organometallics 2002, 21, 2395-
2399. (b) Kim, Y.; Jnaneshwara, G. K.; Verkade, J. G. Inorg.
Chem. 2003, 42, 1437-1447.
(13) Takashima, Y.; Nakayama, Y.; Watanabe, K.; Itono, T.;
Ueyama, N.; Nakamura, A.; Yasuda, H. Harada, A.; Okuda,
J. Macromolecules 2002, 35, 7538-7544.
(14) Williams, C. K.; Breyfogle, L. E.; Choi, S. K.; Nam, W.;
Young, V. G., Jr.; Hillmyer, M. A.; Tolman, W. B. J. Am.
Chem. Soc. 2003, 125, 11350-11359.
(15) (a) Cheng, M.; Ovitt, T. M.; Hustad, P. D.; Coates, G. W.
Polym. Prepr. 1999, 40 (1), 542-543. (b) Chen, H. Y.; Huang,
B. H.; Lin, C. C. Macromolecules 2005, 38, 5400-5405.
(16) (a) Spassky, N.; Wisniewski, M.; Pluta, C.; Le Borgne, A.
Macromol. Chem. Phys. 1996, 197, 2627-2637. (b) Radano,
C. P.; Baker, G. L.; Smith, M. R., III J. Am. Chem. Soc. 2000,
122, 1552-1553. (c) Ovitt, T. M.; Coates, G. W. J. Polym.
Sci., Part A: Polym. Chem. 2000, 38, 4686-4692. (d) Zhong,
Z.; Dijkstra, P. J.; Feijen, J. Angew. Chem., Int. Ed. 2002,
41, 4510-4513. (e) Zhong, Z.; Dijkstra, P. J.; Feijen, J. J.
Am. Chem. Soc. 2003, 125, 11291-11298.
(17) Ovitt, T. M.; Coates, G. W. J. Am. Chem. Soc. 2001, 124,
1316-1326.
Acknowledgment. The authors gratefully acknowl-
edge the Donors of the American Chemical Society
Petroleum Research Fund, Research Corporation, the
Howard Hughes Medical Institute (for a grant to Car-
leton College), Carleton College, and Gustavus Adolphus
College for support of this research. We thank Profes-
sors Marc Hillmyer and Kris McNeill in the Chemistry
Department at the University of Minnesota and Profes-
sor Bradley Chamberlain in the Chemistry Department
at Luther College for helpful discussions and Ms. Esther
(18) For selected examples, see: (a) Cheng, M.; Attygalle, A. B.;
Lobkovsky, E. B.; Coates, G. W. J. Am. Chem. Soc. 1999,
121, 11583-11584. (b) Dove, A. P.; Gibson, V. C.; Marshall,
E. L.; White, A. J. P.; Williams, D. J. Chem. Commun. 2001,
283-284. (c) Aubrecht, K. B.; Hillmyer, M. A.; Tolman, W.
B. Macromolecules 2002, 35, 644-650. (d) Chisholm, M. H.;
Gallucci, J.; Phomphrai, K. Inorg. Chem. 2002, 41, 2785-
2794. (e) Jensen, T. R.; Breyfogle, L. E.; Hillmyer, M. A.;
Tolman, W. B. Chem. Commun. 2004, 2504-2505. (f)
Chivers, T.; Fedorchuk, C.; Parvez, M. Organometallics
2005, 24, 580-586. (g) Ma, H.; Melillo, G.; Oliva, L.; Spaniol,
T. P.; Englert, U.; Okuda, J. Dalton Trans. 2005, 721-727.

10340 Notes
Macromolecules, Vol. 38, No. 24, 2005
(19) Chisholm, M. H.; Gallucci, J. C.; Phomphrai, K. Inorg. Chem.
2004, 43, 6717-6725.
in the homonuclear decoupled methine region of the 1H
NMR spectrum of PLA.28 The intensities were analyzed as
probabilities of syndiotactic (Ps) or isotactic (Pi) enchain-
ment, using Bernoullian statistics. See: Bovey, F. A.; Mirau,
P. A. NMR of Polymers; Academic Press: San Diego, 1996.
(28) (a) Thakur, K. A. M.; Kean, R. T.; Hall, E. S.; Kolstad, J. J.;
Lindgren, T. A. Doscotch, M. A.; Siepmann, J. I.; Munson,
E. J. Macromolecules 1997, 30, 2422-2428. (b) Zell, M. T.;
Padden, B. E.; Paterick, A. J.; Thakur, K. A. M.; Kean, R.
T.; Hillmyer, M. A.; Munson, E. J. Macromolecules 2002,
35, 7700-7707.
(29) Molecular weight (Mn) and PDI (Mw/Mn) values were
obtained by size exclusion chromatography (SEC) and were
calibrated relative to polystyrene standards; therefore, they
do not reflect the absolute (or theoretical) molecular weights.
(30) Baran, J.; Duda, A.; Kowalski, A.; Szymanski, R.; Penczek,
S. Macromol. Rapid Commun. 1997, 18, 325-333.
(31) The value for [LA]eq was measured at equilibrium by NMR
spectroscopy and determined to be 0.07 M when [LA]0 was
1.00 M.
(20) Hormnirun, P.; Marshall, E. L.; Gibson, V. C.; White, A. J.
P.; Williams, D. J. J. Am. Chem. Soc. 2004, 126, 2688-2689.
(21) (a) Wisniewski, M.; LeBorgne, A.; Spassky, N. Macromol.
Chem. Phys. 1997, 198, 1227-1238. (b) Nomura, N.; Ishii,
R.; Akakura, M.; Aoi, K. J. Am. Chem. Soc. 2002, 124, 5938-
5939. (c) Ishii, R.; Nomura, N.; Kondo, T. Polym. J. 2004,
36, 261-264. (d) Tang, Z.; Chen, X.; Pang, X.; Yang, Y.;
Zhang, X.; Jing, X. Biomacromolecules 2004, 5, 965-970.
(e) Tang, Z.; Chen, X.; Yang, Y.; Pang, X.; Sun, J.; Zhang,
X.; Jing, X. J. Polym. Sci., Part A: Polym. Chem. 2004, 42,
5974-5982.
(22) Appiah, W. O.; DeGreeff, A. D.; Razidlo, G. L.; Spessard, S.
J.; Pink, M.; Young, V. G., Jr.; Hofmeister, G. E. Inorg.
Chem. 2002, 41, 3656-3667.
(23) Kricheldorf, H. R.; Berl, M.; Scharnagl, N. Macromolecules
1988, 21, 286-293.
(24) (a) Takeuchi, D.; Aida, R.; Endo, R. Macromol. Rapid
Commun. 1999, 20, 646-649. (b) Takeuchi, D.; Nakamura,
R.; Aida, R. Macromolecules 2000, 33, 725-729. (c)
Takeuchi, D.; Aida, R.; Endo, T. Macromol. Chem. Phys.
2000, 201, 2267-2275.
(32) See Supporting Information.
(33) The extra proton results from terminating the polymeriza-
tion by the addition of methanol.
(25) A visual inspection of the published homonuclear decoupled
1H NMR spectra of the methine region of PLA produced
from rac-LA by the Verkade initiators indicates that they
are less stereoselective than 2a and 2b.12 Harada and co-
workers did not publish results concerning the ROP of rac-
LA.13
(34) Montaudo, G.; Montaudo, M. S.; Puglisi, C.; Samperi, F.;
Spassky, N.; LeBorgne, A.; Wisniewski, M. Macromolecules
1996, 29, 6461-6465.
(35) According to Bernoullian statistics, [sis] ) Ps2/2; [sii] ) PsPi/
2, where Ps ) probability of syndiotactic addition and Pi )
probability of isotactic addition; see ref 27.
(26) Confirmed by analysis of the homonuclear decoupled 1H
NMR spectrum of the methine region of PLA.28
(36) This value for [LA]eq is similar to that (0.06 M) determined
by Duda and Penczek in dioxane: Duda, A.; Penczek, S.
Macromolecules 1990, 23, 1636-1639.
(27) Ps represents the probability of syndiotactic enchainment
(i.e., D-LA followed by L-LA, etc.), also referred to as Pr,
the probability of racemic enchainment. Ps was measured
by integration of the sis, sii, iis, iii, and isi tetrad resonances
MA051811W