Lactide polymerization with chiral β-diketiminate zinc complexes

Article abstract of DOI:10.1021/om100154w

N,N--Di(S-phenylethyl)-2-amino-4-iminopent-2-ene, S,S-nacnac ^CH(Me)PhH, 1a, and N,N--dibenzyl-2-amino-4-iminopent-2-ene, nacnac^BnH, 1b, react with ZnEt₂ to form the corresponding nacnacZnEt complexes 2a and 2b. Neither complex is reactive with 2-propanol or methyl lactate to produce the corresponding alkoxide complexes. In reactions with 2b, ligand redistribution occurs and nacnac^Bn₂Zn was obtained. Reaction of 1a and 1b with Zn(N(SiMe₃)₂) ₂ yielded nacnacZnN(SiMe₃)₂, 6a and 6b. From further reactions with 2-propanol nacnacZnOiPr, 7a and 7b, were obtained. Both complexes were catalytically active for polymerization of rac-lactide with apparent first-order rate constants of k_app = 0.013-0.019 min ^-1 and 0.019-0.038 min^-1 for 7a and 7b, respectively. Obtained polymers were highly heterotactic, with P_r = 0.84-0.87 (7a) and 0.65-0.71 (7b). Analysis of remaining monomer after 75% conversion showed negligible ee and indicates that chiral 7a does not show enantioselectivity in rac-lactide polymerization. Complex 7a, but not 7b, catalyzes unselective transesterification of the polymer during and after polymerization. PLA microstructures in polymerizations with 7b were independent of temperature (23 or 0 °C) or monomer/catalyst ratio (100:1 to 400:1). However, slightly higher P_r values were obtained in the presence of 10-100 equiv of MeCN or pyridine. Complexes nacnac^Bn₂Zn, 2b, 6b, and 7a were characterized by an X-ray diffraction study.

Full text of DOI:10.1021/om100154w

Organometallics 2010, 29, 2139–2147 2139
DOI: 10.1021/om100154w
Lactide Polymerization with Chiral β-Diketiminate Zinc Complexes
ꢀ ꢀ
Frederic Drouin, Paul O. Oguadinma, Todd J. J. Whitehorne, Robert E. Prud’homme,
and Frank Schaper*
ꢀ ꢀ ꢀ ꢀ
Departement de Chimie, Universite de Montreal, 2900 Boulevard E.-Montpetit, Montreal, QC, H3T 1J4, Canada
Received February 25, 2010
N,N0-Di(S-phenylethyl)-2-amino-4-iminopent-2-ene, S,S-nacnacCH(Me)PhH, 1a, and N,N0-dibenzyl-2-
amino-4-iminopent-2-ene, nacnacBnH, 1b, react with ZnEt2 to form the corresponding nacnacZnEt com-
plexes 2a and 2b. Neither complex is reactive with 2-propanol or methyl lactate to produce the corresponding
alkoxide complexes. In reactions with 2b, ligand redistribution occurs and nacnacBn2Zn was obtained.
Reaction of 1a and 1b with Zn(N(SiMe3)2)2 yielded nacnacZnN(SiMe3)2, 6a and 6b. From further reactions
with 2-propanol nacnacZnOiPr, 7a and 7b, were obtained. Both complexes were catalytically active for poly-
merization of rac-lactide with apparent first-order rate constants of kapp=0.013-0.019 min-1 and 0.019-
0.038 min-1 for 7a and 7b, respectively. Obtained polymers were highly heterotactic, with Pr=0.84-0.87 (7a)
and 0.65-0.71 (7b). Analysis of remaining monomer after 75% conversion showed negligible ee and indi-
cates that chiral 7a does not show enantioselectivity in rac-lactide polymerization. Complex 7a, but not 7b,
catalyzes unselective transesterification of the polymer during and after polymerization. PLA microstruc-
tures in polymerizations with 7b were independent of temperature (23 or 0 °C) or monomer/catalyst ratio
(100:1 to 400:1). However, slightly higher Pr values were obtained in the presence of 10-100 equiv of MeCN
or pyridine. Complexes nacnacBn2Zn, 2b, 6b, and 7a were characterized by an X-ray diffraction study.
Introduction
more active indium system that polymerizes rac-lactide with
modest isoselectivity (Pm = 0.53-0.65).6 Zinc-based catalysts
are generally more active than those based on aluminum, but
obtaining isotactic enantiopure or stereoblock polymers from
rac-lactide with these catalysts remains a challenge.
In their seminal work on β-diketiminate zinc comp-
lexes, Coates and co-workers obtained highly stereoregular
heterotactic PLA7 from rac-lactide by a chain-end control
mechanism, which favors alternating insertion of lactide
enantiomers.8 In subsequent works, numerous groups varied
substitution patterns, denticity, and the metal center (Zn or
Mg) of β-diketiminate-based catalysts,9-15 but stayed close to
the “traditional” framework of a diketiminate ligand with
Polylactide (PLA) finds increasing use as a commodity poly-
mer due to its biodegradability and the possibility of deriving the
monomer from renewable resources. Numerous catalyst systems
for the polymerization of lactide to PLA have been developed.1
There is, however, still a remarkable lack of highly active catalyst
systems capable of polymerizing rac-lactide, the racemic mixture
of R,R- and S,S-lactide, to isotactic PLA, which is the tacticity of
highest commercial interest. Typically, Sn-based catalysts display
high activities, but no stereoselectivity.2 At most, a slight prefe-
rence for heterotacticity was observed with diketiminate tin com-
plexes.3 Chiral SALEN-based aluminum catalysts give rise to
isotactic (or block-isotactic) PLA, but only with low to moderate
activity.4,5 Mehrkhodavandi and co-workers recently reported a
(6) Douglas, A. F.; Patrick, B. O.; Mehrkhodavandi, P. Angew.
Chem., Int. Ed. 2008, 47, 2290.
*Corresponding author. E-mail: Frank.Schaper@umontreal.ca.
(1) Stanford, M. J.; Dove, A. P. Chem. Soc. Rev. 2010, 39, 486.
Wheaton, C. A.; Hayes, P. G.; Ireland, B. J. Dalton Trans. 2009, 4832.
Chisholm, M. H.; Zhou, Z. J. Mater. Chem. 2004, 14, 3081. Dechy-Cabaret,
O.; Martin-Vaca, B.; Bourissou, D. Chem. Rev. 2004, 104, 6147.
(2) Platel, R. H.; Hodgson, L. M.; Williams, C. K. Polym. Rev.
(Philadelphia, PA, U. S.) 2008, 48, 11.
(7) Since the monomer already contains two stereocenters with
identical stereochemistry, i.e., an m-dyad, it is not possbile to obtain
syndiotactic PLA from rac-lactide. Polymerization with preference for
alternating R,R/S,S enchainments affords heterotactic PLA with alter-
nating m- and r-dyads (mr)n.
(8) Cheng, M.; Attygalle, A. B.; Lobkovsky, E. B.; Coates, G. W. J.
Am. Chem. Soc. 1999, 121, 11583.
(9) Chamberlain, B. M.; Cheng, M.; Moore, D. R.; Ovitt, T. M.;
Lobkovsky, E. B.; Coates, G. W. J. Am. Chem. Soc. 2001, 123, 3229.
(10) Chisholm, M. H.; Huffman, J. C.; Phomphrai, K. J. Chem. Soc.,
Dalton Trans. 2001, 222. Chisholm, M. H.; Gallucci, J.; Phomphrai, K.
Inorg. Chem. 2002, 41, 2785.
(11) Chisholm, M. H.; Phomphrai, K. Inorg. Chim. Acta 2003, 350, 121.
(12) Dove, A. P.; Gibson, V. C.; Marshall, E. L.; White, A. J. P.;
Williams, D. J. Dalton Trans. 2004, 570.
(13) Yu, K.; Jones, C. W. J. Catal. 2004, 222, 558. Sanchez-Barba, L.
F.; Hughes, D. L.; Humphrey, S. M.; Bochmann, M. Organometallics 2006,
25, 1012. Ayala, C. N.; Chisholm, M. H.; Gallucci, J. C.; Krempner, C.
Dalton Trans. 2009, 9237.
(14) Chen, H.-Y.;Huang, B.-H.;Lin, C.-C. Macromolecules 2005, 38, 5400.
(15) Chisholm, M. H.; Gallucci, J. C.; Phomphrai, K. Inorg. Chem.
2005, 44, 8004.
(3) Dove, A. P.; Gibson, V. C.; Marshall, E. L.; White, A. J. P.;
Williams, D. J. Chem. Commun. 2001, 283.
(4) Spassky, N.; Wisniewski, M.; Pluta, C.; Le Borgne, A. Macromol.
Chem. Phys. 1996, 197, 2627. Radano, C. P.; Baker, G. L.; Smith, M. R. J. Am.
Chem. Soc. 2000, 122, 1552. Ovitt, T. M.; Coates, G. W. J. Polym. Sci., Part A:
Polym. Chem. 2000, 38, 4686. Ovitt, T. M.; Coates, G. W. J. Am. Chem. Soc.
2002, 124, 1316.
(5) Isotactic PLA was also obtained in some cases with achiral
aluminum catalysts, albeit normally with less stereocontrol. Depending
on the ligand, chain-end control for aluminum-based catalysts varies
between preferred isotactic and, more rarely, preferred syndiotactic
enchainment. Wisniewski, M.; Le Borgne, A.; Spassky, N. Macromol.
Chem. Phys. 1997, 198, 1227. Nomura, N.; Ishii, R.; Akakura, M.; Aoi, K. J.
Am. Chem. Soc. 2002, 124, 5938. Hormnirun, P.; Marshall, E. L.; Gibson, V.
C.; White, A. J. P.; Williams, D. J. J. Am. Chem. Soc. 2004, 126, 2688.
r
2010 American Chemical Society
Published on Web 04/15/2010
pubs.acs.org/Organometallics

2140 Organometallics, Vol. 29, No. 9, 2010
Drouin et al.
Scheme 1
aromatic N-substituents. In general, heterotactic PLAs with
different degrees of stereocontrol were obtained. In an attempt
to gain access to C1- or C2-symmetric catalysts and to over-
come chain-end-controlled monomer enchainment with a
catalytic-site control mechanism, Chisholm et al. investigated
diketiminato zinc and magnesium complexes with monosub-
stituted aryl N-substituents.11,15 While isomerization around
the N-aryl bond was slow compared to polymerization, the
Cs-symmetric meso rotamer proved to be the most active and
led to heterotactic PLA. Similar studies of Dove et al. using a
C1-symmetric ligand yielded nearly atactic PLA.12 Mehrkho-
davandi and co-workers incorporated the chiral cyclohexylen-
diamine framework, used successfully on indium as the central
metal, in tetrahedral zinc complexes, but obtained only low
activities and isotacticities.16
We recently became interested in derivatives of diketimi-
nate ligands with aliphatic N-substituents. In contrast to
N-aryl diketiminates, these ligands lend themselves more
easily to generate a C2-symmetric environment, and we were
able to prepare a chiral derivative from chiral amines and
the respective copper and zirconium complexes.17-19 Use of
diketiminate ligands with aliphatic substituents on nitrogen
offers access to complexes that either are inherently chiral
(Scheme 1, A) or form chiral rotamers (Scheme 1, B). We will
present here the synthesis of two examples for A and B and
their performance in the polymerization of rac-lactide.
Figure 1. X-ray crystal structure of 2b. Hydrogen atoms have been
omitted for clarity. The second independent molecule in the unit cell
(Zn2, N3, N4, C31-C51) of essentially identical geometry is not
shown. Thermal ellipsoids are drawn at the 50% probability level.
Scheme 2
Results and Discussion
nacnacZnEt Complexes.20 Following a protocol re-
ported for nacnacArZnEt,21 reaction of diketimines S,S-
nacnacCH(Me)PhH, 1a, and nacnacBnH, 1b, with ZnEt2 yielded
the corresponding zinc ethyl complexes 2a and 2b in 86% and
95% yield, respectively (Scheme 2). Both complexes display
similar NMR spectra: a C2/C2v-symmetric set of resonances
for the diketiminate ligand, a triplet at 1.1-1.3 ppm for
ZnCH2Me, and a quartet at 0.3-0.4 ppm for ZnCH2Me. As
in most asymmetric zinc ethyl complexes, the two diaster-
eotopic ZnCH2 hydrogen atoms in 2a are not differentiated
and appear as a single quartet. In contrast to other diketi-
mine zinc alkyl complexes, 2b forms a dimer in the solid
state (Figure 1, only one of two independent mole-
cules is shown). The compound is present as its meso rota-
mer, with the N-benzyl substituents in a syn orientation on
the same side of the ligand mean plane. This allows the
approach of a second molecule of meso-2b, which is related to
the first by a crystallographic inversion center and coordinates
with the central carbon atom in the ligand backbone (C3 or
C33) to Zn. Such a close approach of the two ligand mean
˚
planes (3.1 A) is not possible with N-aryl substituents for steric
reasons. Consequently, nacnacArZnR complexes display either
three-coordinated Zn centers12,22-24 or coordination of an
additional donor ligand (RCN, Py) to Zn.25 Zn-C distances
(Table 1) are at the higher end of the range observed
for nacnacArZnEt complexes (1.95-1.99 A).12,22,24,25 While
˚
(16) Labourdette, G.; Lee, D. J.; Patrick, B. O.; Ezhova, M. B.;
Mehrkhodavandi, P. Organometallics 2009, 28, 1309.
(17) Buch, F.; Harder, S. Z. Naturforsch., B: Chem. Sci. 2008, 63, 169.
(18) Oguadinma, P. O.; Schaper, F. Organometallics 2009, 28, 4089.
(19) El-Zoghbi, I.; Latreche, S.; Schaper, F. Organometallics 2010,
29, 1551.
(20) Diketiminate zinc complexes might exist as monomers or dimers
in the solid state or in solution. For the sake of simplicity, all complexes
are described in monomeric form.
(22) Cheng, M.; Moore, D. R.; Reczek, J. J.; Chamberlain, B. M.;
Lobkovsky, E. B.; Coates, G. W. J. Am. Chem. Soc. 2001, 123, 8738.
(23) Prust, J.; Stasch, A.; Zheng, W.; Roesky, H. W.; Alexopoulos,
E.; Uson, I.; Bohler, D.; Schuchardt, T. Organometallics 2001, 20, 3825.
(24) Cole, S. C.; Coles, M. P.; Hitchcock, P. B. Organometallics 2004,
23, 5159. Piesik, D. F. J.; Range, S.; Harder, S. Organometallics 2008, 27,
6178.
(25) Allen, S. D.; Moore, D. R.; Lobkovsky, E. B.; Coates, G. W. J.
Organomet. Chem. 2003, 683, 137. Lee, S. Y.; Na, S. J.; Kwon, H. Y.; Lee,
B. Y.; Kang, S. O. Organometallics 2004, 23, 5382.
(21) Cheng, M.; Lobkovsky, E. B.; Coates, G. W. J. Am. Chem. Soc.
1998, 120, 11018.

Article
Organometallics, Vol. 29, No. 9, 2010 2141
Table 1. Selected Bond Distances [A] and Bond Angles [deg] for Crystal Structures of 4, 6b, and 7a
˚
2b
4
6b
7a
Zn1-N1/Zn2-N3
Zn1-N2/Zn2-N4
Zn-Xa
1.990(1)/2.005(1)
1.988(1)/1.996(1)
1.984(2)/1.982(3)
95.06(6), 95.28(6)
119.8(1)-132.8(1)
1.983(1)
1.989(1)
1.942(2)
1.940(2)
1.890(2)
99.61(7)
1.984(2)/1.988(2)
2.007(2)/1.993(2)
1.971(2)-2.023(2)
99.21(8), 99.68(8)
108.91(8)-130.32(8)
82.52(6), 82.43(6)
N-Zn-Nb
97.93(7), 98.27(7)
114.80(5), 116.08(5)
N-Zn-Xc
O1-Zn-O2
129.99(7), 130.39(7)
a 2b: Zn1-C20, Zn2-C50a. 6b: Zn1-N3. 7a: Zn1/2-O1/2. b Diketiminate bite angle, 2b:N1-Zn1-N2, N3-Zn2-N4; 4:N1-Zn1-N1A, N2-Zn1-N2A;
6b: N1-Zn1-N2; 7a:N1-Zn1-N2, N3-Zn2-N4. c 2b: X = C20, C50a. 4: N1-Zn1-N2. 6b: N1-Zn1-N3, N2-Zn1-N3. 7a: X = O1, O2.
clearly present, coordination of the backbone carbon atom
˚
Scheme 3
C3/C33 to Zn is weak (Zn-C: 2.582(2) and 2.539(2) A) and
the Zn atom is only slightly distorted from its trigonal-planar
coordination with C3/33-Zn-CEt/N angles of 96-106°, the
sum of the N-Zn-N/CEt close to 360° (349° and 353°), and
˚
displacements of Zn out of the N/N/CEt planeoflessthan0.4 A.
Since nacnacZnEt complexes have been shown to be slow
initiators of lactide polymerization,9 we attempted to replace
the ethyl group by alkoxide. Reaction of 2a or 2b with either
2-propanol, methanol, or S-methyl lactate did not yield the
respective nacnacZnOR complexes in a variety of solvents
(toluene, C6D6, hexane, THF), at -80 °C or room temperature,
or with varying order of reagent addition. For 2a, heating or
prolonged reaction times yielded the free ligand 1a as the only
diketimine-containing species next to unreacted starting ma-
terial. Complex 2a proved also to be unreactive toward insertion
of benzaldehyde (60 °C, 2 h, C6D6). In an attempt to reverse the
order of ligand coordination, we reacted 3, prepared from
S-methyl lactate and ZnEt2, with 1a (Scheme 2). Only starting
material was observed after 24 h of reaction in toluene at
room temperature, accompanied by small amounts (5%) of
2a. Following the reaction at 65 °C for 24 h by NMR
spectroscopy showed slowly increasing amounts of 2a together
with some insoluble precipitate, but no indication for the
formation of the desired diketiminate zinc lactate complex.
Complex 2a is most likely formed by a ligand redistribution
of 3 to give the, probably insoluble, zinc bislactate and ZnEt2,
which in turn reacts with diketimine 1a.
Scheme 4
Scheme 5
1H NMR spectra of the reaction of 2b with 2-propanol
displayed several new signals, which could by independent
synthesis be assigned to the bis(nacnac) zinc complex 4 and to
ethyl zinc isopropanolate, 526 (Scheme 3). Again, no forma-
tion of nacnacBnZnOiPr was observed under a variety of
reaction conditions. Formation of a homoleptic bis(nacnac)
complex was observed only with nacnacBn, but not in reac-
tions with nacnacCH(Me)Ph, which can be attributed to the
sterically more demanding nature of the latter ligand. In-
deed, attempts to prepare (nacnacCH(Me)Ph)2Zn analogous to
4 from ZnEt2 and excess 1a yielded only the heteroleptic
complex 2a, next to unreacted diketimine.
Formation of homoleptic L2Zn complexes upon reaction of
LZnEt with alcohol has been reported previously for sterically
undemanding nacnacAr and tris(pyrazolyl)borate ligands.14,22,27
A possible mechanism is protonation of the diketiminate ligand
by 2-propanol to form 5 and trapping of nacnacBnH by 2b to
yield 4. Alternatively, 2b might undergo reversible ligand redis-
tribution to 4 and ZnEt2, the latter irreversibly reacting with
2-propanol to form 5. Since a solution of ZnEt2 in C6D6 reacted
completely with 4 in less than 30 min to the heteroleptic complex
2b (Scheme 4), the inverse reaction, i.e., the ligand redistribution
of 2b, seems feasible. Given the fact that protonation of the
diketiminate ligand in 2a by 2-propanol occurred only partially
and upon heating, we believe the latter mechanism to be
responsible for the observed reaction products (Scheme 3).28
1H NMR spectra of reactions of 1b with iPrOZnEt showed
mixtures indicative of ligand exchange reactions, which
contained 1b, 2b, 4, and an insoluble product, most likely
Zn(OiPr)2, in varying amounts, but no evidence for nac-
nacBnZnOiPr (Scheme 5). Attempted protonation of 4 with
iPrOH yielded only mixtures of 4 and small amounts of the
protonated ligand 1b.
(28) Formation of 4 was observed in some cases when solutions of 2b
are kept for extended periods of time. Varying amounts of 4 indicate that
2b is not inherentely prone to disproportionation and that in the absence
of alcohol protic impurities might be required to remove ZnEt2 from the
equilibrium.
(26) Herolds, R. J.; Aggarwal, L.; Neff, V. Can. J. Chem. 1963, 41,
1368.
(27) Looney, A.; Han, R.; Gorrell, I. B.; Cornebise, M.; Yoon, K.;
Parkin, G.; Rheingold, A. L. Organometallics 2002, 14, 274.

2142 Organometallics, Vol. 29, No. 9, 2010
Drouin et al.
Figure 3. X-ray crystal structure of 6b. Hydrogen atoms are
omitted for clarity. Thermal ellipsoids are drawn at the 50%
probability level.
Figure 2. X-ray crystal structure of 4. Hydrogen atoms have
been omitted for clarity. Thermal ellipsoids are drawn at the
50% probability level.
Scheme 6
Crystals suitable for an X-ray diffraction study were obtained
for 4 (Figure 2, Table 1). The structure of 4, which crystallizes on
a crystallographic C2-axis, is close to S4 symmetry with the
benzyl substituents oriented in a rac conformation on opposing
sides of the ligand mean plane (cf. the meso conformation
observed in 2a, Figure 1). The phenyl rings are positioned above
the benzylic CH2 group of the opposite ligand in a close CH-π
protonation of the respective zinc amides. Reaction of Zn-
{N(SiMe3)2}2 with ligands 1a and 1b afforded the amide
complexes 6a and 6b in good yields (Scheme 6). Complex 6a was
obtained as a highly viscous oil and could not be purified. The
impurities, mainly HN(SiMe3)2 and unreacted 1a, did not
interfere in the subsequent reaction with 2-propanol. The crystal
structure of 6b (Figure 3, Table 1) shows again a more sym-
˚
interaction (2.8 A). Short intermolecular π-π interactions
˚
(3.6 A) are observed between the C14-C19 rings of adjacent
molecules. In contrast to nacnacAr2Zn complexes,22,29 which
show the typical boat-like distortion of diketiminate metal
complexes as a consequence of steric congestion, the Zn atom
in 4 is located in the mean plane of the diketiminate ligands. This
is in line with observations for nacnacBnCu complexes: diketi-
minate ligands with primary alkyl substituents on nitrogen are
sterically less demanding than their counterparts with N-aryl
substituents.30 In agreement with this, Zn-N distances (1.983(1)
metric complex, when compared to its nacnacArZnN(SiMe3)2
Ar
˚
˚
analogues (Δ(Zn-N1/2): 6b, <0.01 A; nacnac , 0.02-0.04 A;
Δ(N-Zn-N3): 6b, <0.5°;nacnacAr, 2-8°).12,15,22 The Zn-N3
bond is found in the mean plane of the diketiminate ligand,
without the boat-like distortion observed in nacnacArZnN-
(SiMe3)2.12,15,22 The N(SiMe3)2 fragment is nearly perpendi-
cular (82°) to the mean ligand plane, indicating only small
interactions between the anti-orientated benzyl ligands and the
trimethylsilyl substituents.
˚
and 1.989(1) A) are shorter and the coordination geometry
around nitrogen is more symmetric (differences in N-Zn-N
angles, where NA and NB are of a different diketiminate ligands:
Δ(NA-Zn-NB) < 2°) than in nacnacAr2Zn complexes (Zn-N:
22,29
˚
1.99-2.04 A, Δ(NA-Zn-NB) = 12-26°).
Reaction of 6a or 6b with 2-propanol cleanly generated the
corresponding alkoxide complexes 7a and 7b, respectively
(Scheme 6). The presence of the chiral ligand in 7a is now
visible in the slight splitting of the isopropyl methyl groups into
two signals in the 1H and 13C NMR spectra. Complex 7b, on
the other hand, showed NMR spectra of apparent C2v sym-
metry: one singlet is observed for the NCH2Ph hydrogen atoms
2-Propanol has been reported not to react (or to react only
to a minor extent) with the Zn-Et group in nacnacdippZnEt
(dipp=2,6-diisopropylphenyl),22 (NNO)ZnEt (NNO=dia-
minophenol)16 or even in iPrOZnEt.26 Labourdette et al.
reported that protonation of (NNO)ZnEt was possible with
phenols,16 but given the slow initiation of lactide polymer-
ization by [Zn]OAr, we did not pursue this pathway further.
nacnacZn(amide) and nacnacZnOiPr Complexes. Follow-
ing the protocol established by Coates and co-workers,8
we attempted the synthesis of the targeted complexes by
1
in the H NMR and the methyl groups of the OiPr ligand
appear as one doublet in the 1H and one resonance in the 13
C
NMR spectra. Rotation around the N-Bn bond is thus fast on
the NMR time scale for 7b. Barriers of rotation around the
N-C bond seem to be strongly dependent on the steric demand
of the substituent. A mixture of rac and meso rotamers has been
observed in NMR spectra of nacnacArZnX complexes with
(29) Peng, Y.-L.; Hsueh, M.-L.; Lin, C.-C. Acta Crystallogr., Sect. E
2007, 63, m2388.
(30) Oguadinma, P. O.; Schaper, F. Organometallics 2009, 28, 6721.

Article
Organometallics, Vol. 29, No. 9, 2010 2143
mono-tBu-substituted aryl substituents,11 while fast rotation
was observed for their monomethoxy-substituted analogues.15
Despite the tendency of 2b to form the homoleptic complex 4 in
the presence of alcohol, solutions of 7b in C6D6 showed no
evidence of ligand redistribution over a period of 7 days at
room temperature.
The crystal structure of 7a displays a dimeric zinc complex
with bridging isopropanolate ligands (Figure 4). Formation of
a dimeric complex introduces significant steric strain, and the
secondary alkyl substituent seems to be too bulky to accommodate
the formation of the dimer in its lowest energy conformation,
i.e., with the phenyl groups in an anti conformation and the
CH(Me)Ph hydrogen atoms oriented toward the ligand back-
bone. Instead, rotations around the N2-C20 and the N4-C41
bonds orient the CH(Me)Ph hydrogen atoms toward each other
and lead to a syn conformation of the phenyl groups on the same
side of the ligand mean plane. Accordingly, the dimer distorts to
accommodate the CH(Me)Ph groups in their normal confor-
˚
mation with a N1-N3 distance of 6.0 A, while the rotated
CH(Me)Ph groups allow a closer approach of the two ligands
˚
with d(N2-N4) = 5.1 A. Zn-N and Zn-O bond lengths are
comparable to those observed in analogous complexes with
nacnacAr ligands, while N-Zn-N angles are slightly larger
Ar
˚
˚
(Table 1, nacnac : Zn-N: 2.00-2.07 A, Zn-O: 1.95-2.02 A,
N-Zn-N: 94.8-96.0°).8,9,31 As a consequence of its distorted
structure, 7a displays a high variation in Zn-O bond lengths
Figure 4. X-ray crystal structure of 7a. Most hydrogen atoms
are omitted for clarity. Thermal ellipsoids are drawn at the 50%
probability level.
Ar
˚
˚
(Δ(Zn-O) = 0.05 A, nacnac : Δ(Zn-O) = 0-0.06 A). While
there was no evidence for the rotation of a secondary alkyl
substituent around the N-C bond in planar nacnacRCuL (R=
iPr, CH(Me)Ph)18,32 or nacnaciPrPdX (X = nacnaciPr, allyl)33
complexes, it was previously observed in octahedral nac-
nacR2ZrCl2 (R = Cy, CH(Me)Ph)19 and square-pyramidal
nacnaciPrTiX3 complexes.34
Table 2. rac-Lactide Polymerizations (CH2Cl2, 23 °C) with 7a
[Zn] (mM)
[Zn]/[lactide]
T (°C)
kapp (min-1
)
Pra
#1
#2
#3
#4
#5
#6
#7
1.8
1.8
1.9
1.7
1.7
1.7
1.7
1:300
1:300
1:300
1:300
1:200
1:200
1:100
23
23
23
23
23
23
23
0.019(2)
0.015(1)
0.016(1)
0.016(1)
0.85
0.85 b
0.87 c
0.85 d
0.84
Lactide Polymerization. Both complexes 7a and 7b were
active for the polymerization of rac-lactide in CH2Cl2.35 Com-
plete conversion was achieved after 3 and 1.5 h, respectively
(>95%, 7/lactide = 1:300, [7]=1.5-2.0 mM, 23 °C). Analyses
of the polymer microstructure revealed that both complexes
exhibit a strong preference for alternating insertion of RR- and
SS-enantiomers, leading to predominantly heterotactic PLA
(Pr = 0.84-0.87 and 0.65-0.71 for 7a and 7b, respectively,
where Pr is the probability of forming an r-dyad by insertion).7
Similar preferences with Cs-symmetric diketiminate zinc com-
plexes have been attributed to chain-end control.8 Narrow poly-
dispersities of Mw/Mn=1.1 for selected samples are consistent
with a living polymerization mechanism, as expected for this
catalyst system. Polymer molecular weights are slightly higher
than expected (5-25%), probably due to catalyst decomposi-
tion. Determination of reaction kinetics with 7a and 7b showed
the expected first-order dependence on monomer concen-
tration. Pseudo-first-order rate constants of kapp = 0.019-
0.038 min-1 were found for 7b, which are comparable to
kapp = 0.054 min-1 observed for nacnacdippZnOiPr under
comparable conditions ([Zn]/lactide = 1:490, [Zn] = 2.1 mM).9
0.019(2)
0.013(1)
0.88 d
0.86 d
a Determined after complete conversion of monomer from decoupled
1H NMR by Pr = 2I1/(I1 + I2), with I1 = 5.20-5.25 ppm (rmr, mmr/
rmm), I2 = 5.13-5.20 ppm (mmr/rmm, mmm, mrm). b Mn = 53 000,
Mw/Mn=1.1. c Mn=58 200, Mw/Mn = 1.1. d After 75-85% conversion.
The sterically more congested 7a displays a reduced po-
lymerization rate constant of kapp = 0.013-0.019 min-1
(Table 2) and possesses a short induction period of ca.
10 min (Figure 5).
Samples were taken in regular intervals from polymerizations
with 7a, and the microstructure of the polymer was analyzed. As
expected, the obtained polymer microstructure is independent
of the 7a/lactide ratio (Table 2, 7a/lactide = 1:100 to 1:300). The
1
value of Pr, determined from decoupled H NMR spectra
(Table 2), decreased from around Pr ≈ 0.90 at 10% conversion
to Pr = 0.84-0.86 at 95% conversion (Figure 5). Partial kinetic
resolution by enantioselective polymerization cannot explain
the gradual decrease of Pr during polymerization. Nevertheless,
we analyzed the remaining rac-lactide monomer after 75-77%
conversion (7a/lactide = 1:300) by polarimetry and found
negligible enantiomeric excesses of less than 1%. There is thus
no preferential insertion of one enantiomer and the chiral
N-substituent in 7a does not impart any enantioselectivity in
the polymerization of rac-lactide.
(31) Marsh, R. E.; Clemente, D. A. Inorg. Chim. Acta 2007, 360, 4017.
(32) Oguadinma, P. O.; Schaper, F. Can. J. Chem. 2009, accepted.
€
(33) Tian, X.; Goddard, R.; Porschke, K. R. Organometallics 2006,
25, 5854.
(34) Nikiforov, G. B.; Roesky, H. W.; Magull, J.; Labahn, T.;
Vidovic, D.; Noltemeyer, M.; Schmidt, H.-G.; Hosmane, N. S. Poly-
hedron 2003, 22, 2669.
(35) Only complexes 7a and 7b were investigated in detail. Amide
complex 6b polymerized rac-lactide with lower activity (kapp smaller by a
factor of 6) but identical stereoselectivity compared to 7b. Since 6a was
not obtained in pure form, it was not tested for polymerization.
If a second batch of 200 equiv of monomer is added to
polymerizations of lactide with 7a (CDCl3, 7a/lactide = 1:200)
after either 2 or 4 h, the monomer is consumed with essentially

2144 Organometallics, Vol. 29, No. 9, 2010
Drouin et al.
Figure 7. Methine region of PLA 13C{1H} NMR spectra before
(top) and after (bottom) treatment with 5% 7a for 4 days.
Figure 5. Dependence of conversion and probability of alter-
nating R,R/S,S-enchainment (Pr) on polymerization time.
Table 3. Changes in Tetrad Distribution of PLA after Exposure to
7a for 4 days (determined from 13C NMR spectra)
PLA
1 mol % 7a
5 mol % 7a
Pra
0.78
13
17
39
30
0
0.72
16
18
32
29
3
0.55
25
17
20
13
17
8
% mmm
% mmr + rmmb
% mrm
% rmr
% mrr + rrmb
% rrr
% m
0
61
2
60
59
a Pr calculated from decoupled 1H spectra, presuming the absence of
rr-triads. b Both tetrads were assumed to have equal intensity.
Table 4. rac-Lactide Polymerizations (CH2Cl2) with 7b
[Zn] (mM) [Zn]/[lactide] T (°C) kapp (min-1
)
Pra
#1
#2
#3
#4
#5
#6
1.9
1.7
1.9
2.0
1.9
1.7
1:100
1:200
1:200
1:300
1:400
1:300
23
23
23
23
23
0
0.025(1)
0.038(1)
0.024(5)
0.035(3)
0.019(5)
0.65-0.68 b
0.66
Figure 6. Apparent decrease of Pr in PLA treated with 7a, deter-
1
mined from decoupled H NMR by Pr = 2(Irmr + Immr/rmm)/
(Itotal), presuming the absence of rr-triads.
0.67-0.68 b,c
0.68d
0.68-0.71 b
0.68
a Determined from decoupled 1H NMR by Pr = 2I1/(I1 + I2), with
I1 = 5.20-5.25 ppm (rmr, mmr/rmm), I2 = 5.13-5.20 ppm (mmr/rmm,
mmm, mrm). b Polymer samples precipitated after 12, 30, and 65 min.
c Mn = 35 900, Mw/Mn = 1.1. d Mn = 45 500, Mw/Mn = 1.1.
identical activity kapp and stereoselectivity Pr (see Supporting
Information). Decomposition of 7a during polymerization to
an active, but unselective species can thus also be excluded as a
reason for the apparent decrease of Pr.
An apparent decrease in Pr was also observed when samples
of precipitated and washed PLA were combined with 1, 2, or
5 mol % 7a (relative to mol monomer in PLA) and followed for
several days by NMR (CDCl3, room temperature). The change
in Pr correlated reasonably with the amount of 7a added
(Figure 6). Analyses of the carbonyl and methine regions in
the 13C NMR spectra showed a decrease in mrm- and rmr-
tetrads and an increase in mmm-tetrads (Table 3). At the same
time, resonances associated with rrr- and rrm/mrr-tetrads were
observed, which cannot be obtained from polymerizations of
rac-lactide (Figure 7, Table 3). The observed apparent decrease
in Pr during polymerizations with 7a can thus be attributed to
an unselective transesterification of PLA, catalyzed by 7a. The
Polymerizations with 7b did not show the time-dependent
change in Pr observed for 7a, but displayed in general lower Pr
values (Table 4). The reduced stereocontrol with 7b might be due
to an increased tendency to stereoerrors with a less bulky ligand,
due to a preference for isotactic monomer enchainment by the
chiral rotamers of 7b or due to the formation of Zn(OiPr)2 by
ligand redistribution of 7b into 4 and Zn(OiPr)2 (cf. Scheme 3).
Zinc diisopropanolate is an active, but unselective catalyst for
the polymerization of rac-lactide (Pr = 0.5).36 Since mixtures of
7b and 50 equiv of rac-lactide in CD2Cl2, followed by 1H NMR
spectroscopy, showed only traces of 4 appearing several hours
after polymerization was completed, we doubt that ligand
redistribution to 4and Zn(OiPr)2 occurs to any significant extent
during polymerizations with 7b.
1
appearance of rr-triads, which overlap in the decoupled H
NMR spectra, employed for the determination of Pr, with the
mmm and mrm resonances, is responsible for the apparent
increase in isotacticity.
ꢀ
ꢀ
ꢁ
(36) Claverie, J. Personal communication, Universite de Quebec a
ꢀ
Montreal, Canada, 2009.

Article
Organometallics, Vol. 29, No. 9, 2010 2145
Table 5. Influence of the Presence of Additional Lewis Base in
Lactide Polymerizations with 7b (CDCl3, [7b] = 1.9 mM,
7b/lactide = 1:300, ambient temperature)
responsible. We are currently investigating if we can stabilize
the lifetime of chiral rotamers and increase the steric differ-
entiation in complexes comparable to 7b.
L
L/Zn
kappa (min-1
)
Prb
Experimental Section
THF
0
10
100
1000
0
1
10
100
0
0.03
0.03
0.03
0.02
0.03
0.03
0.02
0.01
0.03
0.03
0.03
0.63
0.63
0.64
0.70
0.63-0.64
0.63
0.63-0.64
0.65-0.67
0.63-0.64
0.64-0.65
0.70-0.71
General Procedures. All reactions, except ligand synthesis, were
carried out under an inert atmosphere using Schlenk and glovebox
techniques under a nitrogen atmosphere. Zn(N(SiMe3)2)2,37 S,S-
nacnacCH(Me)PhH,18,38 and nacnacBnH30,38 were prepared according
to literature procedures. Solvents were dried by passage through
activated aluminum oxide (MBraun SPS) and deoxygenated by
repeated extraction with nitrogen. C6D6 was dried over sodium and
degassed by three freeze-pump-thaw cycles. CDCl3 and CD2Cl2
MeCN
pyridine
1
10
100
˚
were dried over 4 A molecular sieves. rac-Lactide (98%) was pur-
no reaction
chased from Sigma-Aldrich, kept at 5 °C, and used as received. All
other chemicals were purchased from common commercial suppli-
ers and used without further purification. 1H and 13C NMR spectra
were acquired on Bruker AMX 300, AV 400, or Avance 700 spec-
trometers. Chemical shifts were referenced to the residual signals of
the deuterated solvents (C6D6: 1H: δ 7.16 ppm, 13C: δ 128.38 ppm,
a kapp was estimated from conversions after 5-10, 60-90, and
120-140 min. b Determined after 5-10, 60-90, and 120-140 min from
decoupled 1H NMR by Pr = 2I(rmr + mmr/rmm)/I(total).
Polymerization at 0 °C yielded Pr = 0.68 (Table 4, #6),
which is within the margin of error indistinguishable from
values of Pr = 0.65-0.71 obtained at room temperature and
indicates that lower Pr values for 7b might not be simply
assigned to an increased tendency to stereoerrors. On the
other hand, parallel polymerizations with 7b/lactide ratios of
1:100, 1:200, and 1:400 under otherwise identical conditions
(Table 4, #1, #3, and #5) showed Pr values between 0.65 and
0.71, independent of conversion or the 7b/lactide ratio
and rule out a simple competition of chain propagation
and isomerization between rotamers as factors influencing
the polymer microstructure. The exact mechanism of lactide
polymerization with 7b thus remains unclear at the moment.
Polymerizations with 7b were also indifferent to the pre-
sence of THF, and comparable activities were obtained in the
presence of 0, 10, 100, or 1000 equiv of THF (Table 5). In the
presence of MeCN or pyridine, polymerization activity also
decreased only slightly with Lewis base concentration. The
value of Pr, however, increased upon addition of Lewis base.
A comparable increase in stereoselectivity by chain-end
control has been reported for Cs-symmetric magnesium
and zinc diketiminate systems when the reaction solvent
was switched from dichloromethane to THF.10,14 While the
reasons for this increase are unclear in our case (and might be
related simply to changes in solvent polarity), chain-end
control on the polymer microstructure could not be dimin-
ished by preventing the coordination of the pending polymer
chain with an additional Lewis base.
CDCl3: 1H: δ 7.26 ppm). Elemental analyses were performed by the
ꢀ
ꢀ
ꢀ
ꢀ
Laboratoire d’Analyse Elementaire (Universite de Montreal).
Molecular weight analyses were performed on a Waters 1525 gel
permeation chromatograph equipped with three Phenomenex col-
umns and a refractive index detector at 35 °C. Chloroform was used
as the eluant at a flow rate of 1.0 mL min-1, and polystyrene
standards (Sigma-Aldrich, 1.5 mg/mL, prepared and filtered
(0.2 μm) directly prior to injection) were used for calibration.
S,S-nacnacCH(Me)PhZnEt, 2a. S,S-nacnacCH(Me)PhH, 1a (250
mg, 0.82 mmol), and ZnEt2 (136 mg, 1.1 mmol) were dissolved in
toluene (8 mL) and stirred for 15 h at room temperature.
Evaporation of the solvent and drying under fine vacuum
yielded a yellow oil (86% yield), which contained less than 5%
impurities according to NMR and was used without further
purification.
1H NMR (C6D6, 300 MHz, 298 K): δ 7.26 (d, J = 8 Hz, 4H,
Ph), 7.14 (t, J = 8 Hz, 4H, Ph), 7.03 (t, J = 8 Hz, 2H, Ph), 4.70
(q, J = 7 Hz, 2H, CH(Me)Ph), 4.59 (s, 1H, CH(CdN)2), 1.81 (s,
6H, Me(CdN)), 1.52 (d, J = 7 Hz, 6H, CH(Me)Ph), 1.08 (t, J =
9 Hz, 3H, ZnCH2Me), 0.25 (q, J = 9 Hz, 2H, ZnCH2Me). 13
C
{1H} NMR (C6D6, 75 MHz, 298 K): δ 166.2 (CdN), 146.7 (ipso
Ph), 128.7 (Ph), 126.7, 126.6, 96.1 (CH(CdN)2), 57.9 (CH(Me)
Ph), 24.3 (Me(CdN)), 22.9 (CH(Me)Ph), 12.0 (ZnCH2Me), 5.57
(ZnCH2Me).
nacnacBnZnEt, 2b. Diketimine 1b (250 mg, 0.90 mmol) and
ZnEt2 (153 mg, 1.24 mmol) were dissolved in toluene (10 mL)
and stirred for 15 h. The solvent was evaporated and the
obtained yellow powder recrystallized from hexane at -30 °C
(319 mg, 95%).
1H NMR (C6D6, 300 MHz, 298 K): δ 7.10-7.14 (m, 8H, Ph),
7.00-7.05 (m, 2H, Ph), 4.67 (s, 1H, CH(CdN)2), 4.51 (s, 4H,
CH2Ph), 1.75 (s, 6H, Me(CdN)), 1.28 (t, J = 9 Hz, 3H,
ZnCH2Me), 0.36 (q, J = 9 Hz, 2H, ZnCH2). 13C{1H} NMR
(C6D6, 75 MHz, 298 K): δ 168.1 (CdN), 142.0 (ipso Ph), 128.8
(Ph), 127.1, 126.9, 96.8 (CH(CdN)2), 55.0 (CH2Ph), 21.9 (Me
(CdN)), 13.1 (ZnCH2Me), -1.93 (ZnCH2Me). Anal. Calcd for
C21H26ZnN2: C, 67.38; H, 7.05; N, 7.54. Found: C, 67.96; H,
6.95; N, 7.50.
Summary and Conclusions
The use of aliphatic substituents on nitrogen allowed the
synthesis of C2-symmetric zinc diketiminate complexes, which
were active initiators for the polymerization of rac-lactide.
Alternating R,R/S,S-enchainment by chain-end control, pre-
viously observed for C2v-symmetric diketiminato zinc com-
plexes,8 still prevails for these complexes, and heterotactic PLA
is obtained with 7a and 7b. The two chiral substituents in 7a do
not induce enantioselectivity in polymerization, which, judging
from the facile rotation around the N-C bond found in the
crystal structure of 7a, indicates that the ligand framework is
not sufficiently rigid and that its role is limited to provide
sufficient steric bulk for heterotactic chain-end control.
S-MeO2CC(H)(Me)OZnEt, 3. ZnEt2 (204 mg, 1.63 mmol)
and S-methyl lactate (169 mg, 1.63 mmol) were dissolved in
toluene (10 mL) and stirred at room temperature for 15 min.
Evaporation of the solvent and drying under vacuum yielded
colorless, wax-like 3 (258 mg, 80%).
ꢀ
ꢀ
(37) Rivillo, D.; Gulyas, H.; Benet-Buchholz, J.; Escudero-Adan, E.
C.; Freixa, Z.; van Leeuwen, P. W. N. M. Angew. Chem., Int. Ed. 2007,
46, 7247.
For 7b, reasons for the lack of isotactic enchainment
remain unclear: fast isomerization, higher activity of the
meso-isomer, or insufficient catalytic-site control might be
(38) El-Zoghbi, I.; Ased, A.; Oguadinma, P. O.; Tchirioua, E.;
Schaper, F. Can. J. Chem., submitted.

2146 Organometallics, Vol. 29, No. 9, 2010
Drouin et al.
Table 6. Details of X-ray Diffraction Studies
4
2a
6b
7a
formula
C21H26N2Zn
371.81; 1.314
C38H42N4Zn
620.13; 1.287
C25H39N3Si2Zn
503.14; 1.219
C48H64N4O2Zn2
859.77; 1.286
Mw (g/mol);
dcalcd. (g/cm3)
T (K); F(000)
cryst syst
space group
unit cell: a (A)
150; 784
triclinic
P1h
9.4077(6)
9.4416(6)
21.4629(14)
92.702(2)
93.216(2)
98.523(2)
1879.4(2); 4
150; 1312
monoclinic
C2/c
14.8210(3)
15.1994(3)
14.2091(3)
90
90.010(1)
90
3200.89(11); 4
150; 536
triclinic
P1
150; 1824
orthorhombic
P212121
12.9219(3)
18.4221(4)
18.6609(4)
90
˚
8.1588(2)
11.0074(2)
15.8174(3)
90.658(1)
98.133(1)
102.576(1)
1371.18(5); 2
2.8-63.5; 0.97
˚
b (A)
˚
c (A)
R (deg)
β (deg)
γ (deg)
90
90
3
˚
V (A ); Z
θ range (deg);
completeness
4442.20(17); 4
3.4-63.4; 1.00
2.1-72.6; 0.96
4.2-72.5; 0.99
collected reflns; Rsigma
indep reflns; Rint
24 585; 0.017
7173; 0.026
1.818; multiscan
0.037; 0.098; 1.07
20 811; 0.021
3143; 0.033
1.303; multiscan
0.031; 0.090; 1.09
18 435; 0.035
4362; 0.038
2.191; multiscan
0.032; 0.082; 0.97
59 422; 0.040
7216; 0.052
1.642; multiscan
0.029; 0.065; 0.97
μ (mm-1
)
R1(F); wR(F2);
GoF(F2) a
residual electron
density
0.46; -0.72
0.30; -0.28
0.26; -0.32
0.22; -0.38
a R1(F) based on observed reflections with I > 2σ(I), wR(F2) and GoF(F2) based on all data.
1H NMR (C6D6, 400 MHz, 298 K): δ 4.75 (q, J = 7 Hz, 1H,
OCHMe), 3.29 (s, 3H, CO2Me), 1.67 (d, J = 7 Hz, 3H,
-OCHMe), 1.49 (t, J = 8 Hz, 3H, ZnCH2Me), 0.48 (q, J = 8
Hz, 2H, ZnCH2Me). 13C{1H} NMR (C6D6, 101 MHz, 298 K): δ
178.1 (CdO), 70.4 (OCHMe), 52.0 (CO2Me), 22.5 (OCHMe),
13.0 (ZnCH2Me), 0.05 (ZnCH2Me). Anal. Calcd for C6H12-
O3Zn: C, 36.48; H, 6.12. Found: C, 36.28; H, 6.12.
nacnacBnZnN(SiMe3)2, 6b. Zn{N(SiMe3)2}2 (1.50 g, 3.88
mmol) and 1b (1.08 g, 3.88 mmol) were dissolved in hexane
(20 mL) and stirred for 12 h at room temperature. Evaporation
of the solvent afforded an off-white powder, which was recrys-
tallized from hexane (1.73 g, 89%).
1H NMR (CDCl3, 400 MHz, 298 K): δ 7.00-7.34 (m, 10H,
Ph), 4.80 (s, 4H, PhCH2), 4.72 (s, 1H, CH(CdN)2), 2.04 (s, 6H,
Me(CdN)), -0.20 (s, 18H, SiMe3). (C6D6, 400 MHz, 298 K): δ
7.13-7.16 (m, 8H, Ph), 7.11-7.12 (m, 2H, Ph), 4.68 (s, 4H,
PhCH2), 4.64 (s, 1H, CH(CdN)2), 1.71 (s, 6H, Me(CdN)), 0.13
(s, 18H, SiMe3). 13C{1H} NMR (C6D6, 75 MHz, 298 K): δ 170.2
(CdN), 141.0 (ipso Ph), 128.9 (Ph), 127.1, 126.9, 96.5 (CH
(CdN)2), 54.0 (PhCH2), 22.1 (Me(CdN)), 5.2 (SiMe3). Anal.
Calcd for C25H39N3ZnSi2: C, 59.68; H, 7.81; N, 8.35. Found: C,
60.33; H, 7.71; N, 8.22.
nacnacBn2Zn, 4. Diketimine 1b (250 mg, 0.900 mmol) and
ZnEt2 (56 mg, 0.45 mmol) were reacted in hexane (8 mL) for 15 h
at room temperature. The residue obtained after evaporation of
the solvent was washed with hexane and dried under vacuum to
yield a colorless powder of 4 (113 mg, 41%).
1H NMR (C6D6, 400 MHz, 298 K): δ 7.10-7.14 (d, J = 7 Hz,
4H, Ph), 7.04-7.10 (t, J = 7 Hz, 4H, Ph), 6.96-7.03 (t, J = 7
Hz, 2H, Ph), 4.43 (s, 1H, CH(CdN)2), 4.14 (s, 4H, CH2Ph), 1.71
(s, 6H, Me(CdN)). 13C{1H} NMR (C6D6, 75 MHz, 298 K): δ
167.9 (CdN), 142.1 (ipso Ph), 128.3 (Ph), 127.5, 126.3, 94.2 (CH
(CdN)2), 54.2 (CH2Ph), 22.0 (Me(CdN)). Anal. Calcd for
C38H42N4Zn: C, 73.60; H, 6.83; N, 9.04. Found: C, 73.37; H,
6.64; N, 8.96.
S,S-nacnacCH(Me)PhZnOiPr, 7a. Complex 6a (200 mg, 0.38
mmol) was reacted with 2-propanol (23 mg, 0.38 mmol) in
hexane (10 mL) for 20 h at room temperature. The solvent
was reduced until the onset of precipitation. The solution was
slightly warmed to redissolve any precipitate formed and kept
for 24 h at -20 °C to yield colorless crystals of 7a (90 mg, 55%).
1H NMR (C6D6, 400 MHz, 298 K): δ 7.40 (d, J = 8 Hz, 4H,
Ph), 7.19 (t, J = 8 Hz, 4H, Ph), 7.06 (t, J = 8 Hz, 2H, Ph), 5.05 (q,
J = 7 Hz, 2H, CH(Me)Ph), 4.48 (s, 1H, CH(CdN)2), 4.20 (sept,
J = 6 Hz, 1H, OCHMe2), 1.68 (d, J = 6 Hz, 6H, CH(Me)Ph), 1.67
(s, 6H, Me(CdN)), 1.275 (d, J = 7 Hz, 3H, OCHMe2), 1.27 (d, J =
7 Hz, 3H, OCHMe2). 13C{1H} NMR (C6D6, 75 MHz, 298 K): δ
168.8 (CdN), 147.0 (ipso Ph), 128.4 (Ph), 126.6 (Ph), 126.2 (Ph),
96.2 (CH(CdN)2), 66.0 (OCHMe2), 57.8 (CH(Me)Ph), 28.96
(OCHMe2), 28.91 (OCHMe2), 24.8 (Me), 24.2 (Me). Anal. Calcd
for C24H32N2ZnO: C, 67.00; H, 7.50; N, 6.50. Found: C, 66.64; H,
7.40; N, 6.42.
iPrOZnEt, 5. 26 ZnEt2 (50 mg, 0.40 mmol) and dry 2-propanol
(24 mg, 0.39 mmol) were reacted in toluene (1.5 mL) for 12 h.
The colorless powder obtained after evaporation of the solvent
is used without further purification in subsequent reactions
(46 mg, 85%).
1H NMR (C6D6, 300 MHz, 298 K): δ 3.99 (sept, J = 7 Hz,
1H, OCHMe2), 1.53 (t, J = 9 Hz, 3H, MeCH2Zn), 1.18 (d, J =
7 Hz, 6H, OCHMe2), 0.55 (q, J = 9 Hz, 2H, MeCH2Zn). 13C
{1H} NMR (C6D6, 75 MHz, 298 K): δ 68.7 (OCHMe2), 27.1
(2C, OCHMe2), 12.8 (MeCH2Zn), 1.61 (MeCH2Zn).
S,S-nacnacCH(Me)PhZnN(SiMe3)2, 6a. Zn{N(SiMe3)2}2 (993
mg, 2.57 mmol) and 1a (750 mg, 2.45 mmol) were dissolved in
hexane and stirred at room temperature for 24 h. Evaporation of
the solvent yielded an orange, highly air-sensitive oil (989 mg,
76%) in 85-90% purity, which was used in subsequent reac-
tions without further purification.
nacnacBnZnOiPr, 7b. Complex 6b (1.00 g, 1.99 mmol) was
dissolved in toluene (15 mL) and reacted with 2-propanol (152 μL,
1.99 mmol) for 12 h at room temperature, during which some white
precipitate appeared. The solvent was evaporated to a quarter of its
volume and the obtained precipitate isolated by filtration. Recrys-
tallization from hexane yielded colorless crystals (544 mg, 70%).
1H NMR (CDCl3, 300 MHz, 298 K): δ 7.05-7.24 (m, 10H, Ph),
4.45 (s, 4H, PhCH2), 4.39 (s, 1H, CH(CdN)2), 3.63 (sept, J = 7 Hz,
1H, OCHMe2), 1.81 (s, 6H, Me(CdN)), 0.77 (d, J = 7 Hz, 6H,
OCHMe2). (C6D6, 400 MHz, 298 K): δ 7.24 (d, J = 7 Hz, 4H, Ph),
7.18 (t, J = 7 Hz, 4H, Ph), 7.08 (t, J = 7 Hz, 2H, Ph), 4.57 (s, 4H,
PhCH2), 4.52 (s, 1H, CH(CdN)2), 3.95 (sept, J = 6 Hz, 1H,
OCHMe2), 1.71 (s, 6H, Me(CdN)), 1.08 (d, J = 6 Hz, 6H,
1H NMR (C6D6, 300 MHz, 298 K): δ 7.25 (d, J = 7 Hz, 4H,
Ph), 7.17 (t, J = 7 Hz, 4H, Ph), 7.04 (t, J = 7 Hz, 2H, Ph), 4.89
(q, J = 8 Hz, 2H, CH(Me)Ph), 4.55 (s, 1H, CH(CdN)2), 1.84 (d,
J = 8 Hz, 6H, CH(Me)Ph), 1.69 (s, 6H, Me(CdN)), 0.16 (s,
18H, SiMe3). 13C{1H} NMR (C6D6, 75 MHz, 298 K): δ 169.3
(CdN), 145.8 (ipso Ph), 128.8 (Ph), 126.5, 96.8 (CH(CdN)2),
58.4 (CH(Me)Ph), 25.0 (Me), 24.0 (Me), 5.7 (SiMe3). Anal.
Calcd for C27H43N3ZnSi2: C, 61.00; H, 8.10; N, 7.90. Found:
C, 63.33; H, 7.90; N, 6.92.

Article
Organometallics, Vol. 29, No. 9, 2010 2147
OCHMe2). 13C{1H} NMR (C6D6, 101 MHz, 298 K): δ 169.5
(CdN), 141.9 (ipso Ph), 128.5 (Ph), 127.0, 126.5, 94.3 (CH
(CdN)2), 66.5 (OCHMe2), 54.1 (PhCH2), 28.4 (OCHMe2)
21.9 (Me(CdN)). Anal. Calcd for C22H28N2ZnO: C, 65.75; H,
7.02; N, 6.97. Found: C, 65.85; H, 7.04; N, 6.88.
Typical Lactide Polymerization Conditions. A. Glovebox. In a
glovebox, a solution of the desired catalyst in CH2Cl2 (1-2 mL)
was added to a solution of rac-lactide in CH2Cl2 (5-10 mL). Typi-
cal conditions: [Zn] = 1.7-2.0 mM, Zn/lactide = 1:100 to 1:400,
total volume: 6-12 mL. After stirring at ambient temperature
(23 °C) for the desired time, the reaction or a reaction sample was
quenched with 2 equiv of AcOH in CH2Cl2, removed from the
glovebox, and freed from solvent immediately under vacuum. The
obtained colorless PLA was either washed with EtOH to remove
remaining monomer or dried without further treatment under
vacuum and analyzed as a polymer/monomer mixture (determina-
tion of conversion).
B. Schlenk Line. In a Schlenk flask, a solution of the desired
catalyst in CH2Cl2 (2 mL) was added via canula to a solution of
lactide in CH2Cl2 (200-1000 mg, 10 mL). The resulting solution
was stirred for the desired time, and reaction samples or final
product was isolated as described above. For polymerizations at
0 °C, both solutions were cooled to 0 °C prior to addition.
C. J.-Young Tube. In a J.-Young tube, a solution of 7b
in CDCl3 (26 mM, 75 μL, 1.9 μmol) was added by micropipet
to a solution containing rac-lactide (75-82 mg), Lewis base
(<40 μL), and CDCl3 (1 mL). After short mixing, the tube was
kept at ambient temperature and the reaction followed by NMR
in the required intervals.
Microstructure Analysis. Pr values were determined from the
integration of the methine region in homonuclear decoupled 1H
NMR spectra and calculated according to Pr =2I1/(I1 + I2),
with I1=5.20-5.25 ppm (rmr, mmr/rmm), I2=5.13 - 5.20 ppm
(mmr/rmm, mmm, mrm).9 Resonances in 1H and 13C spectra
were assigned according to the literature.39
X-ray Crystallography. Diffraction data were collected on a
Bruker SMART 6000 with Montel 200 monochromator,
equipped with a rotating anode source for Cu KR radiation.
Cell refinement and data reduction were done using APEX2.40
Absorption corrections were applied using SADABS.40 Struc-
tures were solved by direct methods using SHELXS97 and
refined on F2 by full-matrix least-squares using SHELXL97.41
All non-hydrogen atoms were refined anisotropically. Hydro-
gen atoms were refined isotropically on calculated positions
using a riding model. Further experimental details are listed in
Table 6 and given in the Supporting Information.
Acknowledgment. We thank Dr. Minh Tan Phan Viet
and Sylvie Bilodeau for assistance with polymer NMR
spectroscopy, Dr. Michel Simard for help with the struc-
ꢀ
ture of 2a, Pierre Menard-Tremblay for assistance with
the molecular weight determinations, and Dr. Carole
Fraschini for helpful discussions.
(39) Kasperczyk, J. E. Macromolecules 1995, 28, 3937. Kasperczyk,
J. E. Polymer 1999, 40, 5455. Zell, M. T.; Padden, B. E.; Paterick, A. J.;
Thakur, K. A. M.; Kean, R. T.; Hillmyer, M. A.; Munson, E. J. Macro-
molecules 2002, 35, 7700.
(40) APEX2, Release 2.1-0; Bruker AXS Inc.: Madison, WI, 2006.
(41) Sheldrick, G. M. Acta Crystallogr. 2008, A64, 112.
Supporting Information Available: Details of the crystal
structure determinations (CIF). This material is available free
of charge via the Internet at http://pubs.acs.org.