Exploring electronic versus steric effects in stereoselective ring-opening polymerization of lactide and β-butyrolactone with amino-alkoxy- bis(phenolate)-yttrium complexes

Article abstract of DOI:10.1002/chem.201002779

A series of methoxy-amino-bis(phenol)s (ONOOR1,R2)H₂ possessing on the phenol rings R¹ ortho substituents with variable steric and electronic properties (R¹=CMe₂Ph, 1; CMe₂tBu, 3; CMe₂(4-CF

Full text of DOI:10.1002/chem.201002779

DOI: 10.1002/chem.201002779
Exploring Electronic versus Steric Effects in Stereoselective Ring-Opening
Polymerization of Lactide and b-Butyrolactone with Amino-alkoxy-
bis(phenolate)–Yttrium Complexes
Miloud Bouyahyi,[a] Noureddine Ajellal,[a] Evgueni Kirillov,[a]
Christophe M. Thomas,[a, b] and Jean-FranÅois Carpentier*[a]
Abstract: A series of methoxy-amino-
anism. Most of these polymerizations
proceeded in controlled fashion,
giving polymers with narrow polydis-
persities and experimental molecular
weights in good agreement with calcu-
lated values. The nature of the R1
(Pr =0.93–0.94)ꢀCMe2tBu (Pr =0.94–
0.95)ꢀCPh3 (Pr =0.95–0.96). On the
other hand, the syndiotactic stereocon-
trol in the polymerization of rac-BBL
follows a quite different trend: R1 =Cl
(Pr =0.42–0.45)!CMe2tBu (Pr =0.62–
1
2
bis
A
a
on the phenol rings R1 ortho substitu-
ents with variable steric and electronic
properties (R1 =CMe2Ph, 1; CMe2tBu,
3; CMe2(4-CF3C6H4), 5; CPh3, 9; Cl,
10) has been synthesized and further
À
ortho substituents has
a
profound
0.70)<CMe3 (Pr =0.80)ꢀCMe
2ACHTUNGTRENNUNG(4CF3
reacted with [Y{N
G
N
impact on the rates and, more spectac-
ularly, on the stereocontrol of the poly-
merizations. The heterotactic stereo-
control in the ROP of rac-LA appears
to be governed essentially by steric
considerations; the larger the substitu-
ent, the higher the heterotacticity: R1 =
Cl (Pr =0.56)!CMe3 (Pr =0.80)!
Ph) (Pr =0.82–0.84)<CMe2Ph (Pr =
0.89)<CPh3 (Pr =0.94), which suggests
the involvement of electronic interac-
tions. DFT computations on model in-
to give cleanly the corresponding yttri-
1
2
um compounds [Y(ONOOR ,R ){N-
(SiHMe2)2}(thf)n] (Y-x); the solid-state
A
ACHTUNGTRENNUNG
À
structures of Y-3 and Y-10 have been
determined. These amido complexes
have been used as initiators for the
ring-opening polymerization (ROP) of
rac-lactide (LA) and rac-b-butyrolac-
tone (BBL) to provide heterotactically
enriched poly(lactic acid)s (PLAs) and
syndiotactically enriched poly(3-hy-
droxybutyrate)s (PHBs), respectively,
by means of a chain-end control mech-
termediates confirmed a stabilizing C
H···p interaction between a methylene
À
C H of the ring-opened BBL unit and
the p system of one of the ortho-aryl
1
CMe2Ph (Pr =0.90)<CMe
N
substituents of the ONOOR ligand; by
contrast, for model intermediates in
À
the ROP of LA, no such C H···p inter-
Keywords: electronic effects · lac-
tide · ring-opening polymerization ·
stereoselectivity · yttrium
action involving the methyl group of
lactate was observed.
Introduction
Among the numerous metal-based catalysts that have been
disclosed for the ring-opening polymerization (ROP) of lac-
tones and other cyclic (di)esters,[1] organolanthanide and re-
lated group 3 metal derivatives hold a special position.[2]
Indeed, neutral compounds of the type (LnX)Ln–Nu, in
which LnX is a mono- or dianionic ancillary ligand and Nu
an active nucleophilic group (typically, an alkoxide or an
amide), often feature high reactivity for ROP processes.
This is believed to result from the conjugated high Lewis
acidity of the metal center and the strong ionicity of the lan-
thanide (Ln)–heteroatom (Nu) bonds. Under proper condi-
[a] Dr. M. Bouyahyi, Dr. N. Ajellal, Dr. E. Kirillov, Prof. C. M. Thomas,
Prof. J.-F. Carpentier
Organomꢀtalliques et Catalyse
UMR 6226 CNRS-Universitꢀ de Rennes 1
35042, Rennes Cedex (France)
Fax : (+33)2-23-23-69-39
E-mail: jean-francois.carpentier@univ-rennes1.fr
[b] Prof. C. M. Thomas
Present address: Chimie Paris Tech
UMR CNRS 7223, 75005 Paris (France)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem201002779. It contains crystal
structures of proligands 3 and 5, crystallographic data for 3, 5, Y-3,
tions, many of these catalysts/initiators also feature
a
1
1
and Y-10 as CIF files, representative H, H-1H NOESY, and
13C NMR spectra for some complexes and polymers, and computa-
tional details including Cartesian coordinates of computed structures.
“living” behavior, rapidly initiating with almost 100% effi-
ciency the growth of a polyester chain through cleavage of
1872
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 1872 – 1883

FULL PAPER
the oxygen–carbonyl bond in the monomer. Moreover, with
adequate ancillaries, that is, usually very bulky multidentate
ligands, installed on the active lanthanide center, high levels
of stereocontrol can be achieved in the ROP of chiral mono-
mers. A combination of these properties—activity, “living-
ness”, and stereoselectivity—makes some organolanthanides
particularly efficient initiators/catalysts for preparing unique
polyesters with controlled architectures.
of 0.91 at 208C, in both cases; Pr is the probability of race-
mic linkages between monomer units and is determined by
NMR spectroscopy (see the Experimental Section)) than
those based on tert-butyl or adamantyl R1 groups (Pr =0.80–
0.82 at 208C, for both PLA and PHB); on the other hand,
yttrium complexes supported by methyl-substituted ligands
proved essentially nonstereoselective (Pr ꢀ0.53 at 208C).
Yet, although a cumyl group is clearly bulkier than tert-
butyl or adamantyl groups, close examination of the solid-
Among such compounds,[3] we reported several years ago
yttrium derivatives supported by tetradentate dianionic ami-
noalkoxybis(phenolate) ligands (hereafter abbreviated as
(ONOOR)2À).[4] Amido or alkoxide complexes of this class
proved to be particularly active and productive for the con-
trolled ROP of racemic lactide (rac-LA)[5,6] and racemic b-
butyrolactone (rac-BBL),[7] which afforded highly heterotac-
tic polylactides (PLA) and highly syndiotactic poly(3-hy-
droxybutyrate)s (PHB), respectively (Scheme 1). Detailed
state structures of the yttrium–amido precursors [Y-
1
A
E
N
crease in the steric crowding around the metal center; in
fact, the closest C atoms of the R1 substituents are located
more than 4.0 ꢂ away from the yttrium center or N atom of
the amido group (vide infra). The exact influence of the sub-
stituents in this series of compounds, and more generally in
other compounds having tunable ligand platforms, on chain-
end stereocontrol remains not necessarily well understood
and rationalized. To address this issue, we have prepared
1
new (ONOOR )H2 proligands bearing specific R1 ortho sub-
stituents on the phenolate rings and investigated the perfor-
mance of the corresponding yttrium–amido complexes in
the stereoselective ROP of rac-LA and rac-BBL. Unexpect-
ed, subtle effects were observed, with contrasting behaviors
for rac-LA and rac-BBL. The results suggest that electronics
are also involved in the stereocontrol of the ROP of rac-
BBL.
Results and Discussion
Scheme 1. Stereoselective ROP of racemic lactide and racemic b-butyro-
lactone with (ONOOR)Ln–Nu complexes (Nu=N
ACHTNUGTRENUNG(SiHMe2)2 or OiPr).
Design and synthesis of proligands: To explore the steric
and possible electronic influence of substituents installed on
the phenolate rings on stereoselective ROP, a set of proli-
gands was targeted. Those proligands were selected to
afford direct, valuable comparisons with the reference
system that has proven to offer, within this series, the best
catalytic performance thus far in ROP of rac-LA and rac-
BBL, that is, the amino-methoxy-bis(phenolate) system
microstructural NMR spectroscopy investigations[5b,7b] re-
vealed that those reactions proceed by means of a “chain-
end control mechanism” (CEM), in which the stereogenic
center of the last inserted monomer in the growing polymer
chain biases the catalyst to enchain a monomer of opposite
stereochemistry. Chain-end control is well known to be as-
sisted by the catalyst/initiator that effects the polymerization
reaction. Coates and co-workers first reported the sensitivity
of the ROP of lactides to ligand sterics with (N-aryl-substi-
tuted b-diiminate)–zinc complexes, with subtle ligand modi-
fications resulting in dramatic effects on activity and stereo-
selectivity.[8]
Along the same lines, we have shown that a most interest-
ing feature of the (ONOOR)Ln–Nu catalyst systems is the
possibility to finely tune the stereoselectivity of the polymer-
ization through the nature of the R substituents installed on
the phenolate ligand platforms. Initial studies in the ROP of
rac-LA indicated that the ortho substituents (Scheme 1, R1;
i.e., the R substituent in the abbreviation (ONOOR)) appa-
rently play the most important role,[5b] which turned out to
be true also in the ROP of rac-BL.[7a,b] Actually, yttrium
complexes bearing phenolate ligands ortho substituted by
cumyl (CMe2Ph) groups offered PLAs (in THF) and PHBs
(in toluene) with higher hetero/syndiotacticities (Pr values
bearing ortho-cumyl substituents (ONOOCMe Ph)H2 (1). Pro-
2
2ACHTUNGTRENNUNG
ligands (ONOOCMe tBu)H2 (3) and (ONOOCMe (CF Ph))H2 (5),
which have CMe2tBu and CMe2(4-CF3C6H4) ortho substitu-
ents, respectively, were prepared by means of a Mannich
condensation of the corresponding phenols (2 and 4) with 2-
methoxyethylamine (Scheme 2); yet, those reactions were
hampered by the formation of cyclic benzoxazines,[9] espe-
cially when a low concentration of reagents was used, as ex-
pected for intramolecular reactions. Optimization of the
2
3
Scheme 2.
Chem. Eur. J. 2011, 17, 1872 – 1883
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1873

J.-F. Carpentier et al.
procedures led to isolated yields of 60 and 52% for 3 and 5,
respectively.
Attempts to prepare the pro
gand (ONOOCPh )H2 (9)
3
li
that bears a very bulky trityl ortho substituent by means of
the Mannich condensation or reductive amination starting
from the ortho-trityl-substituted phenol or salicaldehyde (6),
respectively, were unsuccessful. In those cases, only the cor-
responding benzoxazine was isolated. Effective preparation
of proACHTUNGTRENNUNGliACHTUNGTRENNUNGgand 9 was achieved by stepwise amination/reduction
sequences, starting from salicaldehyde 6 (Scheme 3).
Scheme 5. Synthesis of [amino-alkoxy-bis(phenolate)]yttrium–amido
complexes.
Scheme 3.
smoothly at room temperature, equally in toluene or pen-
tane, to afford selectively yttrium complexes [Y-
1
To complete this set of proligands, the ortho,para-di-
G
G
ACHTUNGTRENN(UNG thf)] (Y-1, Y-3, and Y-5) as air-sen-
chloro-substituted aminobisACHTUNGTRNEUNG
(phenol) (ONOOCl)H2 (10) was
prepared in 63% yield, in a one-pot experiment by means
of the reductive amination of the commercially available
2,4-dichlorosalicaldehyde with 2-methoxy-ethylamine and
sodium cyanoborohydride (Scheme 4). Again, this route
proved more efficient than the Mannich condensation from
2,4-dichlorophenol, which led systematically to significant
amounts of the corresponding benzoxazine.
A
A
G
ACHTUNGTRENNUNG
mental analysis, that is, a THF-free complex. These com-
pounds are all readily soluble in THF and aromatic hydro-
carbons (benzene, toluene), but sparingly soluble in aliphat-
ic hydrocarbons (pentane, hexanes), a feature that helped in
their recrystallization and eventually the growth of single
crystals suitable for X-ray diffraction in the case of Y-3 and
Y-10.
The molecular structure of [Y(ONOOCMe tBu){N-
2
ACHUTNRGEN(NUG SiHMe2)2}ACHTUGNTREN(NUNG thf)] (Y-3) is depicted in Figure 1. This com-
pound adopts in the solid state a mononuclear structure
with a metal center hexacoordinated in an octahedral envi-
ronment that is quite similar to that of related yttrium com-
plexes supported by tert-butyl and cumyl-substituted li-
Scheme 4.
gands.[5] The yttrium–O
ACTHNUTRGNE(UNG methoxy) bond length in Y-3
À
(Y(1) O(3), 2.455(2) ꢂ) is just slightly longer than that in
related Y(ONOOCMe Ph
)
and
Y
G
)
compounds
2
Proligands 3, 5, 9, and 10 were recovered as air-stable, col-
1
orless, crystalline compounds that were characterized by H
(2.384(5) and 2.414(2) ꢂ, respectively). Comparable distan-
ces are similarly observed for the yttrium–O(phenolate)
and 13C NMR spectroscopy, mass spectrometry, as well as
single-crystal X-ray diffraction studies for (ONOOCMe tBu)H2
[(Y(1) O(1), 2.164(17); Y(1) O(2), 2.145(17) ꢂ) vs.
2
À
À
(3) and (ONOOCPh )H2 (9) (see the Experimental Section
2.145(4), 2.152(5) ꢂ in Y(ONOOCMe Ph
)
and 2.157(2),
3
2
(ONOOtBu)], yttrium–N
G
À
and Supporting Information).
2.147(2) ꢂ in Y
2.558(2) ꢂ vs. 2.610(5) ꢂ in Y(ONOOCMe Ph) and 2.533(2) ꢂ
2
(ONOOtBu)), and yttrium–N
G
À
Synthesis, solid-state, and solution structure of yttrium com-
plexes: Heteroleptic [amino-alkoxy-bis(phenolate)]yttrium–
amido complexes were prepared by direct aminolysis of
in
Y
2.279(2) ꢂ vs. 2.297(6) ꢂ in Y(ONOOCMe Ph) and 2.276(3) ꢂ
2
in YACHTNUTGRNEUNG
(ONOOtBu)). Also, the bond angles involving the trans-
[Y{N
G
G
located oxygen atoms of the phenolates (O(1)-Y(1)-O(2),
sponding proligands (Scheme 5). The reactions proceeded
152.2(7)8 in Y-3 vs. 152.3(17) and 153.9(8)8 in
1874
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 1872 – 1883

Polymerization of Lactide and b-Butyrolactone
FULL PAPER
still remain rather far away from the metal center or the
amido group.
Compound Y10 adopts a dimeric, nonsymmetric structure
in the solid state (Figure 2). Both yttrium centers are hexa-
Figure 1. Molecular structure of [Y(ONOOCMe tBu){N
C
G
2
3). All hydrogen atoms and solvent molecules are omitted for clarity;
thermal ellipsoids are drawn at the 50% probability level. Selected bond
À
À
lengths [ꢂ] and angles [8]: Y(1) N(1) 2.558(2), Y(1) N(2) 2.279(2),
À
À
À
À
Y(1) O(1) 2.164(17), Y(1) O(2) 2.145(17), Y(1) O(3) 2.455(2), Y(1)
O(4) 2.371(19); N(2)-Y(1)-N(1) 175.96(8), O(1)-Y(1)-N(1) 78.56(7),
O(2)-Y(1)-N(1) 75.25(7), O(4)-Y(1)-N(1) 86.81(7), O(3)-Y(1)-N(1)
68.66(7), O(1)-Y(1)-N(2) 101.98(8), O(2)-Y(1)-N(2) 104.80(8), O(4)-
Y(1)-O(3) 153.52(6), O(4)-Y(1)-N(2) 97.23(8), O(3)-Y(1)-N(2) 107.37(8),
O(1)-Y(1)-O(2) 152.24(7), O(1)-Y(1)-O(4) 82.02(7), O(1)-Y(1)-O(3)
83.40(6), O(2)-Y(1)-O(4) 87.48(7), O(2)-Y(1)-O(3) 95.39(7).
Figure 2. Molecular structure of [(YACHTUNGRTENNUG HCATUNGTRENN(NGU SiHMe2)2})2] (Y-10).
(ONOOCl){N
All hydrogen atoms and solvent molecules are omitted for clarity; ther-
mal ellipsoids are drawn at the 50% probability level.
Y(ONOOCMe Ph) and Y
A
2
trans-located oxygen atoms of the methoxy and thf ligands
(O(3)-Y(1)-O(4), 153.52(6)8 in Y-3 vs. 157.89(16) and
157.09(8)8, respectively) are quite similar. The only signifi-
cant difference is observed for the bond angle involving the
coordinated in a distorted-octahedral environment. Each
metal center retains a k1-amido and a k4-amino-alkoxy-
bis(phenolate) ligand and an additional oxygen(phenolate)
atom that forms a m bridge between the two metal centers
(Y···Y, 3.833 ꢂ). Apparently, the dimeric nature of this com-
pound originates from the insufficient bulkiness of the
ortho-chloro substituents. Main bond lengths and angles are
trans-located NACHTUNGTRENNUNG(amine) and NHCATUNGTRENN(UGN amide), which is significantly
larger in Y-3 (N(2)-Y(1)-N(1), 175.9(8)8 vs. 160.8(2)8 in
Y(ONOOCMe Ph) and 165.80(9)8 in Y
C
2
Another way to assess the possible difference in steric
hindrance introduced by the ortho substituents on the phe-
nolate rings is to compare the
spatial distance between the
Table 1. Spatial distances [ꢂ] in [Y(ONOOCMe R){N
A
G
2
latter substituents (e.g., the
closest methyl in the CMe2R
group or the closest C atom in
the R group) and either the
metal center or the nitrogen
atom of the amido group
(Table 1). In fact, comparison
CMe2Ph (Y-1), and CMe3; see below for numbering schemes).
À
À
of the C(1) Y(1) and C(1’)
Y(1)
distances
in
Y(ONOOCMe tBu
)
(Y-3 ; 4.176,
2
4.272 ꢂ) suggests
a
metal
center more hindered than that
[Y(ONOOCMe tBu)]
[Y(ONOOCMe Ph)]
[YACTHNGUTERNNUG
(ONOOtBu)]
2
2
Y(ONOOCMe Ph
)
(4.542,
2
in
4.452 ꢂ) and
À
À
À
C(1) Y(1)
4.176
4.272
4.993
4.724
3.965
4.367
5.342
4.815
C(1) Y(1)
4.542
C(1) Y(1)
4.431
4.847
Y
U
)
À
À
À
C(1’) Y(1)
C(1’) Y(1)
4.452
5.344
6.292
4.449
4.329
5.081
5.724
C(2) Y(1)
(4.431, 4.847 ꢂ). The same
trend is observed when compar-
À
À
C(2) Y(1)
C(2) Y(1)
À
À
C(2’) Y(1)
C(2’) Y(1)
À
À
ing the C(1) N(2) and C(1’)
N(2) distances, as well as the
À
À
À
C(1) N(2)
C(1) N(2)
C(1) N(2)
4.194
4.841
À
À
À
C(1’) N(2)
C(1’) N(2)
C(2) N(2)
À
À
C(2) N(2)
C(2) N(2)
À
À
C(2) N(2) and C(2’) N(2) dis-
tances. Yet, those substituents
À
À
C(2’) N(2)
C(2’) N(2)
Chem. Eur. J. 2011, 17, 1872 – 1883
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1875

J.-F. Carpentier et al.
Table 2. Main bond lengths [ꢂ] and angles [8] in [(YACHTNUGTRENNUG CAHTUNGTREN(NGUN SiHMe2)2})2] (Y-10).
(ONOOCl){N
tion is in line with previous
NOESY experiments conducted
À
Y(2) N(2’)
2.283(3)
2.344(3)
2.381(3)
2.389(3)
2.522(3)
2.973(12)
2.111(3)
2.144(3)
2.260(3)
2.380(3)
2.425(3)
2.511(3)
2.247(3)
O(5)-Y(2)-N(2’)
O(5)-Y(2)-O(1)
N(2’)-Y(2)-O(1)
O(5)-Y(2)-O(4)
N(2’)-Y(2)-O(4)
O(1)-Y(2)-O(4)
O(5)-Y(2)-O(6)
N(2’)-Y(2)-O(6)
O(1)-Y(2)-O(6)
O(4)-Y(2)-O(6)
O(5)-Y(2)-N(2)
N(2’)-Y(2)-N(2)
O(1)-Y(2)-N(2)
O(4)-Y(2)-N(2)
O(6)-Y(2)-N(2)
O(5)-Y(2)-Cl(1)
89.13(11)
120.79(10)
121.75(11)
82.49(10)
169.05(11)
68.91(9)
O(2)-Y(1)-O(4)
O(2)-Y(1)-N(1’)
O(4)-Y(1)-N(1’)
O(2)-Y(1)-O(1)
O(4)-Y(1)-O(1)
N(1’)-Y(1)-O(1)
O(2)-Y(1)-O(3)
O(4)-Y(1)-O(3)
N(1’)-Y(1)-O(3)
O(1)-Y(1)-O(3)
O(2)-Y(1)-N(1)
O(4)-Y(1)-N(1)
N(1’)-Y(1)-N(1)
O(1)-Y(1)-N(1)
O(3)-Y(1)-N(1)
90.01(10)
90.16(12)
114.78(11)
95.44(10)
70.53(9)
172.35(11)
146.02(10)
115.99(10)
97.38(11)
75.08(9)
78.17(11)
145.32(10)
97.92(11)
78.21(10)
68.00(10)
À
Y(2) O(1)
on Y-1 and its isopropoxide
À
Y(2) O(4)
2
congener
(OiPr)
evidenced
[Y(ONOOCMe Ph)-
À
Y(2) O(6)
N
ACHTUNGTRENNUNG
À
Y(2) N(2)
À
Y(2) Cl(1)
À
Y(2) O(5)
146.65(10)
91.22(11)
86.82(9)
OCHMe2 interactions, respec-
tively.
À
Y(1) O(2)
À
Y(1) N(1’)
The 1H NMR spectra of [Y-
À
Y(1) O(1)
91.93(9)
ACHUTNRGENNUG AHCTUNGTREN(NUGN SiHMe2)2}] (Y-10)
(ONOOCl){N
À
Y(1) O(3)
78.24(11)
94.46(12)
136.94(10)
76.99(10)
68.48(10)
74.91(8)
À
in [D6]benzene at different tem-
peratures showed many reso-
nances and proved uninforma-
tive due to their complexity
caused by severe broadening
and overlapping. We assume
Y(1) N(1)
À
Y(1) O(4)
reported in Table 2. The metallacycle Y(1)O(1)Y(2)O(4) is
that this may reflect the coexistence of mononuclear and di-
nuclear (and possibly higher aggregated) species in solu-
tion;[11] however, in the context of the present study directed
toward stereoselectivity of ROP reactions, this aspect was
not investigated further.
À
perfectly planar, with slightly different
Y
G
À
À
lengths (Y(1) O(1), O(4) Y(2), 2.380(3) ꢂ; Y(1) O(4),
À
2.247(3) ꢂ; Y(2) O(1), 2.344(3) ꢂ). The N’ atoms of the bis-
(silyl)amide groups belong to this plane and are located in a
position cis to the
NACTHNGUTER(NNUG amine) atoms (N(1)-Y(1)-N(1’),
97.9(11)8; N(2)-Y(2)-N(2’), 94.5(12)8). Some Cl ortho sub-
Stereoselective ROP of racemic lactide and racemic b-buty-
À
stituents are close to the yttrium centers (Cl(1) Y(2),
2.973(12) ꢂ; Y(1)-Cl(2), 3.216(12) ꢂ), which suggests addi-
tional contacts and a possible tendency towards a 7-coordi-
ACHTUNGERTNrNUNG olactone: The prepared [amino-alkoxy-bis(phenolate)]yttri-
um–amido complexes were investigated in the stereoselec-
tive ROP of rac-LA and rac-BBL and their performances
were compared to those of the reference system (Y-1) that
is based on the cumyl-substituted ligand. Previous investiga-
À
À
nation. Overall, the Y O and Y N bond lengths involving
the k1-amido and the k4-aminoalkoxybis(phenolate) ligand
are comparable to those found in related complexes of the
tions revealed that ROP of rac-LA with [Y
(SiHMe2)2}(thf)] complexes proceed with higher stereoselec-
(ONOOR){N-
ACHTUNGTRENNUNG
type [Y(ONOOCMe R){N
C
(thf)] (e.g., Y-3 ; vide
A
ACHTUNGTRENNUNG
2
supra) and the bimetallic complex of divalent ytterbium
[{Yb
(ONNOtBu)}2].[10]
The solution structure of [Y(ONOOCMe R){N
(thf)] compounds Y-3, Y-5, and Y-9 was investigated by H
tivities in THF than in toluene,[5b] whereas the reverse was
observed in the ROP of rac-BBL.[7a,b] Most ROP experi-
ments of rac-LA and rac-BBL were therefore conducted in
THF and toluene, respectively (Scheme 1). Those polymeri-
zations were at least triplicated using different batches of
catalyst precursor or monomer to ensure the reproducibility
and meaningfulness of the results; in particular, Pr values
were found to fall within a range of Æ1%. Representative
results are summarized in Tables 3 and 4.
ACHTUNGTRENNUNG
2
AHCTUNGRTEN(GNUN SiHMe2)2}-
1
ACHTUNGTRENNUNG
and 13C NMR spectroscopy. The spectra in [D6]benzene or
in [D8]THF (two solvents closely related to those used for
ROP catalysis, that is, toluene and THF) at room tempera-
ture are consistent with the solid-state structures retained in
solution. Diagnostic features include one set of signals for
two equivalent phenolate rings, with a single characteristic
AB pattern in the 1H NMR spectra for the diastereotopic
CHH benzylic hydrogen atoms (see the Supporting Informa-
tion for a representative 1H NMR spectrum of Y-3). At
room temperature, both methyl groups of the silylamido
group SiHMe2 are equivalent on the NMR spectroscopic
timescale and distinct resonances for a coordinated THF
ligand are observed. Also, the resonance for the methoxy
group in the complexes is shifted by approximately d=0.2–
All compounds proved to be active toward the ROP of
rac-LA, and gave PLAs with relatively narrow molecular
weight distributions and number-average molecular weight
(Mn) values generally in good agreement with those calcu-
lated from the monomer-to-yttrium ratio. Although detailed
kinetic investigations were not performed, because they are
outside the scope of this study that is directed toward ste-
reoselectivity, the data reported in Table 3 evidence a de-
crease in activity with the trityl-substituted system (Y-9), as
0.6 ppm upfield with respect to that in the pro
G
N
(ONOOR){N
compared to other [YACHTUNGRTENUNGN ACHTNUTRGEG(NNUN SiHMe2)2}ACHTNUGTRNE(NUGN thf)] com-
indicates that this group remains coordinated in solution
plexes. This might be related to the large bulkiness of this
substituent. The chloro-substituted complex Y-10 also
turned out to be relatively less active than other compounds,
possibly because of the aggregated/dimeric nature of this
compound.
1
onto the metal center, which is thus 6-coordinated. H–1H
NOESY NMR spectroscopic experiments conducted on Y-3
confirmed that both the methyl and tert-butyl groups of the
ortho-CMe2tBu substituents are—at some time—in close vi-
cinity with the N
mation), which is consistent with the short distances ob-
served in the solid-state structure (see above); this observa-
(SiHMe2)2 group (see the Supporting Infor-
In terms of stereoselectivity, all systems but the chloro-
substituted compound led to highly heterotactically enriched
PLAs, provided the reaction is performed in THF and not in
1876
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 1872 – 1883

Polymerization of Lactide and b-Butyrolactone
FULL PAPER
(ONOOR){N (thf)n] complexes.[a]
Table 3. Ring-opening polymerization of rac-lactide with [YACHTUNGTRENNGUN ACHTUNTRGENNG(UN SiHMe2)2}ACHTGUNTRENNUGN
be ruled out. The trityl system
Y-9 also combines an obvious
large bulkiness with a high p-
[d]
[e]
[e]
Entry Complex R1
[LA]/[Y] Solvent Time Conv. Mn,calcd
Mn,exptl
(ꢃ103)
Mw/Mn
Pr[f]
[h][b]
[%][c]
G
ACHTUNGTRENNUNG
1[5b]
2[5b]
3
4
5
6
7
8
9
CMe3
100
100
100
400
400
THF
THF
THF
THF
toluene
THF
THF
THF
THF
toluene
THF
5
0.34
1
1
1
1
1
2
2
100
100
100
95
80
92
79
100
74
52
60
14.4
14.4
14.4
51.8
46.1
13.2
22.7
14.4
21.3
14.9
8.6
15.5
26.3
23.8
53.6
33.0
22.3
32.5
24.9
31.1
17.4
15.3
1.33
1.19
1.22
1.64
1.94
1.33
1.15
1.25
1.32
1.35
1.66
0.80 electron density, of which indi-
CMe2Ph (Y-1)
CMe2tBu (Y-3)
CMe2tBu (Y-3)
CMe2tBu (Y-3)
0.90
0.94
0.95
0.50
0.93
0.94
0.96
0.95
0.61
0.56
vidual effects cannot be sepa-
rated. Nevertheless, the slight
increase in heterotacticity be-
tween the CMe2tBu and CPh3
systems again suggests that
chain-end control in ROP of
rac-LA mainly depends on the
degree of crowding around the
metal center in these [Y-
CMe
CMe
N
CPh3 (Y-9)
CPh3 (Y-9)
CPh3 (Y-9)
Cl (Y-10)
100
200
200
100
10
11
2
2
[a] All reactions performed with [rac-LA]=0.44m at 208C. [b] Reaction times were not necessarily optimized.
AHCTUNGTRENNUNG
(ONOOR)] complexes. It is
[c] PLA conversion determined by 1H NMR spectroscopy. [d] Mn [gmolÀ1] of PLA calculated from Mn,calcd
=
144.00ꢃ([LA]/[Y])ꢃconv.(LA). [e] Experimental (corrected; see Experimental Section) Mn [gmolÀ1] and Mw/
Mn values were determined by GPC in THF versus polystyrene standards. [f] Pr is the probability of racemic
linkages between monomer units and is determined from the methine region of the homonuclear decoupled
1H NMR spectrum.
noteworthy that detailed micro-
structural analyses of the heter-
otactic PLAs produced with the
new systems confirmed again a
pure chain-end control mecha-
nism (see the Supporting Infor-
mation).
(ONOOR){N (thf)n] complexes.[a]
Table 4. Ring-opening polymerization of rac-BBL with [YACHTUNGTRENNNUG AHCTUNRTGEG(NNUN SiHMe2)2}ACHTGUNTRENNUGN
[d]
[e]
[e]
Entry
Complex R1
[BBL]/[Y]
Time
[min][b]
Yield
[%][c]
Mn,calcd
(ꢃ103)
Mn
ACHTUNGTRENNUNG
Mw/Mn
Pr[f]
exptl
G
N
Polymerizations of rac-BBL
with the new [YACTHNUTRGNEUNG
(ONOOR)]
1[7]
2[7]
3
4
5
6
7
8
CMe3
200
400
100
200
400
1000
100
400
60
1
20
45
60
120
20
60
>98
92
95
96
98
81
97
85
17.2
31.6
8.2
16.5
33.7
69.7
8.3
1.03
1.06
1.08
1.10
1.10
1.10
1.16
1.20
0.80
0.89
0.70
0.68
0.64
0.62
0.84
0.82
CMe2Ph (Y-1)
CMe2tBu (Y-3)
CMe2tBu (Y-3)
CMe2tBu (Y-3)
CMe2tBu (Y-3)
26.1
7.5
15.1
27.6
48.9
7.6
complexes Y-3, Y-5, Y-9, and Y-
10 also proceeded in a con-
trolled fashion, which led to
PHBs with quite narrow molec-
ular weight distributions (Mw/
Mn <1.20; except for Y-9,
CMe
CMe
2A
2A(CF3Ph)
29.2
23.5
ACHTUNGTRENNUNG(Y-5)
1.21<Mw/Mn <1.88)
and
9
CPh3 (Y-9)
CPh3 (Y-9)
CPh3 (Y-9)
Cl (Y-10)
100
500
1500
100
1
3
10
180
320
95
94
88
35
28
8.2
40.4
113.5
3.0
4.8
7.9[g]
1.21[g]
1.42[g]
1.88[g]
1.18
0.94
0.94
0.94
0.45
0.42
47.0[g]
number-average
molecular
10
11
12
13
118.2[g]
2.5
3.2
weights determined by GPC in
good agreement with the Mn
values calculated from the mo-
nomer conversions (Table 4).
Larger loading of rac-BBL re-
sulted in proportionally higher
molecular weight PHBs (com-
pare entries 14–17, 18 and 19,
and 20–22). As for polymeri-
zations of rac-LA, the system
based on the chloro-substituted
Cl (Y-10)
200
1.20
[a] All reactions performed with [rac-BBL]=2.4m at 208C in toluene. [b] Reaction times were not necessarily
optimized. [c] Isolated yields of PHB. [d] Mn [gmolÀ1] of PHB calculated from Mn,calcd =86.00ꢃ([BBL]/[Y])
conv.ACHTUNGTRENNUNG
(BBL). [e] Experimental (uncorrected) Mn [gmolÀ1] and Mw/Mn values determined by GPC in THF
versus polystyrene standards. [f] Pr is the probability of racemic linkages between b-butyrolactone units and is
determined by 13C{1H} NMR spectroscopy (see the Experimental Section). [g] Mn and Mw/Mn of polymer de-
termined by SEC-RI in CH2Cl2 at RT using polystyrene standards.
toluene (see entries 5 and 10 in Table 3). Apparently, the
heterotacticity of these polymerizations follows grossly the
bulkiness of the ortho substituents installed on the pheno-
late rings: Cl (Pr =0.56)!CMe3 (Pr =0.80)!CMe2Ph (Pr =
compound Y-10 proved to be quite sluggish toward rac-BBL
(turnover frequencies (TOFs)<12 hÀ1 at 208C; entries 12
and 13 in Table 4), whereas the other three compounds af-
forded remarkably active systems. In contrast to rac-LA, the
trityl-substituted system Y-9 was the most active, with a
TOF of up to 9400 hÀ1 (entry 10 in Table 4), that is, of the
same order of magnitude but less active than its cumyl ana-
logue (Y-1, TOF=22080 hÀ1; entry 2 in Table 4).[7] This de-
creasing activity trend may be logically accounted for by the
increasing bulkiness of the ortho substituents, going from
CMe2Ph to CPh3. Yet, the activity of the CMe2tBu-substitut-
ed system Y-3 (TOFꢁ400 hÀ1 at 208C; entries 5 and 6 in
Table 4) may appear somewhat inconsistent with the above
trend, as it was expected (on steric grounds) to be in be-
tween that of Y-1 and Y-9.
0.90)<CMe
N
0.94–0.95)ꢀCPh3 (Pr =0.95–0.96). The high heterotacticity
achieved using the CMe2tBu-substituted complex Y-3 is evi-
dence that the stereocontrol of the polymerization is essen-
tially, if not only, governed by steric factors. The slight but
significant increase in heterotacticity noticed going from
compound Y-1 to Y-5, that is, upon installing a para-trifluor-
omethyl group on the phenyl ring of the cumyl substituent,
may also arise from a small increase in bulkiness; yet, addi-
tional effects such as variation of the electron density on the
cumyl phenyl ring (vide infra) or F···H interactions[12] cannot
Chem. Eur. J. 2011, 17, 1872 – 1883
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1877

J.-F. Carpentier et al.
Quite unexpected results were also observed in the ste-
reoselectivity of these rac-BBL polymerizations. The syndio-
tacticity degree of the resulting PHBs was assessed by close
examination of the methyl, methylene, and carbonyl regions
of the 125 MHz 13C{1H} NMR spectra (see the Supporting
Information),[7b] and led to the following trend: Cl (Pr =
0.42–0.45)!CMe2tBu (Pr =0.62–0.70)<CMe3 (Pr =0.80)ꢀ
À
mization of several steric interactions between the methyl
groups of LA and the bulky aryl substituents, possibly rein-
À
forced by an attractive electronic C H···p interaction be-
À
tween a methyl C H of the Mg–lactate unit and the p
system of one of the N-aryl substituents of the ligand.
The apparent beneficial effect of phenyl groups in ortho
substituents installed on the phenolate rings in our series of
CMe
(4CF3 Ph) (Pr =0.82–0.84)<CMe2Ph (Pr =0.89)<
[YACTHNGUTERNNUG
(ONOOR)] complexes for the ROP of rac-BBL led us to
CPh3 (Pr =0.94) (Table 4). Thus, as compared to the cumyl-
substituted reference system Y-1, the trityl-substituted
system Y-9 led to a superior Pr value of 0.94 (entries 20–22),
which is to our knowledge the highest syndioselectivity thus
far achieved in the ROP of rac-BBL. Yet, unexpectedly
enough, the CMe2tBu-substituted system Y-3 gave PHBs
with modest Pr values in the range 0.62–0.70 (lower syndio-
tacticities being achieved at higher BBL loadings; en-
tries 14–17). These observations markedly contrast with the
high heterotacticities observed in the ROP of rac-LA, in
which Y-3 performed equally to Y-9 (vide supra). Also,
going from compound Y-1 to Y-5, that is upon installing a
para-trifluoromethyl group on the phenyl ring of the cumyl
substituent, resulted in a slight but noticeable drop in syn-
diotacticity of the PHBs (from Pr =0.89 to 0.82–0.84); this is
to be compared to the case of rac-LA, for which a slight in-
crease in stereoselectivity was noticed (from Pr =0.90 with
Y-1 to Pr =0.93–0.94 with Y-5).
perform DFT computations, to assess the pertinence of simi-
À
lar weak C H···p electronic interactions. The cumyl-substi-
tuted framework ONOOCMe Ph (1) was used to run these
2
computations. We examined, for both LA and BBL poly-
merizations, four possible intermediate structures in which
the propagating chain has been modeled with a ring-opened
L-LA and (S)-BBL unit, respectively. In line with the
models by Rzepa et al.[13] and general belief about mechanis-
tic scenarios of these ROP reactions,[1] the latter alkoxy-
ester unit has been considered to be chelated on the metal
center; due to the nature of the [Y
two coordination modes were investigated, with either a
trans-alkoxy/amino–trans-ester/OMe or trans-alkoxy/
(ONOOR)] framework,
ACHTUNGTRENNUNG
a
OMe–trans-ester/amino arrangement, resulting in an overall
(distorted) octahedral geometry (Schemes 6 and 7).
All structures of model intermediates were constructed
starting from the solid-state structure of Y-1,[7a] in which the
cumyl groups at the para position of the phenolate rings
were replaced by methyl groups, to quicken computations.
Of the four optimized structures of type I and II model
intermediates in the ROP of rac-LA, the most stable ap-
peared to be the one with the cumyl groups positioned to-
wards the yttrium–lactate core (II-trans-alkoxy/amino;
Scheme 6). The closest distance between any of the hydro-
gen atoms of the lactate unit and any atom of the two cumyl
groups in this structure is 3.163 ꢂ; this rather long distance
The above results evidence a markedly distinct behavior
within this series of [YACHTNUTGRNEUNG
(ONOOR)] systems in terms of activi-
ty and, even more, of stereoselectivity for the ROP of rac-
LA and rac-BBL. Clearly, caution must be taken in the
comparison of these results since different solvents have
been used and they relate to intrinsically different mono-
mers, possessing either a single (BBL) or two (LA) stereo-
genic centers. On the other hand, it is noteworthy that these
two reactions share close similarities since they both operate
under pure chain-end stereocontrol mechanism. In this
regard, the most striking result of this study is the apparent
strict dependence of stereoselectivity on steric factors for
the ROP of rac-LA, whereas the ROP of rac-BBL does not
obey this trend. Rather, our results suggest that electronic
factors may also be involved in the stereocontrol of the
latter reaction. This hypothesis is supported by the following
observations: 1) the highest Pr values (0.89–0.94) for the
ROP of rac-BBL are obtained only with ortho substituents
containing a phenyl ring: CMe2Ph and CPh3; alkyl ortho
substituents, even quite bulky (CMe3, CMe2tBu), gave much
lower Pr values (0.62–0.80); 2) introduction of an electron-
withdrawing CF3 group on the phenyl ring of the cumyl
ortho substituents has a noticeable (negative) influence on
the stereoselectivity.
À
argues against any significant stabilizing C H···p interac-
tion.[14] Interestingly, in II-trans-alkoxy/OMe, the shorter dis-
DFT computational analysis: In 2005, Rzepa et al. reported
DFT computations on model intermediates and transition
states in the heterospecific ROP of rac-LA mediated by
magnesium–(b-diketiminate) complexes.[13] Their detailed
examination of the rate-limiting transition-state geometries
revealed that heterotactic PLA arises by means of the mini-
Scheme 6. Schematic representation of model intermediates in the ROP
of LA mediated by [YACTHNUTRGNEUNG
(ONOOR)] compounds. It shows the different ar-
rangements of the chelated alkoxy-ester units investigated and relative
computed energies.
1878
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 1872 – 1883

Polymerization of Lactide and b-Butyrolactone
FULL PAPER
trans-alkoxy/OMe, being +4.9 kcalmolÀ1 higher in energy.
The other two intermediates III-trans-alkoxy/amino and III-
trans-alkoxy/OMe with the cumyl groups orientated away
from the butyrate moiety were significantly destabilized (by
+9.8 and +12.1 kcalmolÀ1, respectively).
Conclusion
We have experimentally evidenced that subtle modifications
in the ligand framework of [amino-alkoxy-bis(phenolate)]–
yttrium complexes can have a major impact on the ROP of
rac-LA and rac-BBL. Modification of polymerization rates
and, more spectacularly, of stereocontrol have been noticed
within the series of newly prepared compounds. For the
polymerization of rac-LA, the heterotactic stereocontrol op-
erates by means of a chain-end control mechanism and ap-
pears to be governed essentially, if not only, by steric consid-
erations. Higher heterotacticties are achieved with com-
plexes that possess bulkier ortho substituents installed on
the phenolate rings, independently of their electronic
nature. On the other hand, for the polymerization of rac-
BBL, the syndiotactic stereocontrol operates also by means
of a chain-end control mechanism, but electronic interac-
tions appear to be involved, as higher syndiotacticities are
obtained only with ligands bearing a phenyl group in the
ortho substituents.
Scheme 7. Schematic representation of model intermediates in the ROP
of BBL mediated by [YACHTUNGTRENNUNG
(ONOOR)] compounds. It shows the different ar-
rangements of the chelated alkoxy-ester units investigated and relative
computed energies.
tance of 2.984 ꢂ between one of the hydrogen atoms of the
lactate methyl group and a meta-carbon atom of the phenyl
ring of the cumyl substituent positions these atoms in such a
À
way that enables them to engage in such a C H···p interac-
tion; yet, this intermediate appeared to be 12.7 kcalmolÀ1
less stable than the former one, which means that if such a
stabilizing interaction takes place, it is quite weak.[13] Also,
the two other computed intermediates, II-trans-alkoxy/
amino and I-trans-alkoxy/OMe, which bear cumyl groups
oriented out of the yttrium–lactate core, were found to be
higher in energy by +8.9 and +21.3 kcalmolÀ1, respectively.
Strikingly different results were obtained upon optimizing
the similar four model intermediates in the ROP of rac-
DFT computations performed on model intermediates
À
confirmed these hypotheses. A C H···p interaction between
À
a methylene C H of the ring-opened BBL unit and the p
system of one of the ortho-aryl substituents of the ligand has
been found to stabilize some intermediates; by contrast, no
À
such C H···p interaction involving the methyl group of lac-
À
BBL (Scheme 7). In fact, close C H···p interactions between
tate has been observed. Overall, our computational data are
in line with the hypothesis of a ligand-assisted stereocontrol
mechanism pointed out by Rzepa et al. with magnesium–(b-
diketiminate) complexes.[13] Yet, in our series of [Y-
one methylene hydrogen of the butyrate moiety (Ha) and
both the ortho- and meta-carbon atoms of a phenyl ring of
the cumyl substituent (H···C distances: 2.898 and 3.083 ꢂ,
respectively) appeared in the most stable structure, that is,
AHCTUNGTRENNUNG
(ONOOR)] complexes, this stereocontrol mechanism ap-
IV-trans-alkoxy/amino
(Figure 3).
Similarly,
short
pears to be operative for the ROP of rac-BBL but is unlike-
ly in the ROP of rac-LA.
(butyrate)H···C(cumyl) distances of 3.033 and 2.969 ꢂ, re-
ACHTUNGTRENNUNG
spectively, were found in the optimized structure of IV-
These first experimental and computational results consti-
tute a rare example of electronic implications in the stereo-
selective ROP of cyclic esters.[15] Clearly, more experimental
and theoretical work is required to get deeper insight into
these subtle effects and eventually take profit of these for
the design of even more efficient catalyst systems. Efforts di-
rected at the isolation of model compounds similar to those
computed, which remain very scarce in the case of LA[16]
and are virtually nonexistent in the case of BBL,[17] most
likely due to their elusive character, are ongoing in our labo-
ratories.
Experimental Section
Figure 3. DFT-optimized structure of the IV-trans-alkoxy/amino inter-
mediate. Hydrogen atoms, except those of the butyrate fragment, are
omitted for clarity.
General conditions: All manipulations requiring a dry atmosphere were
performed under a purified argon atmosphere using standard Schlenk
Chem. Eur. J. 2011, 17, 1872 – 1883
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1879

J.-F. Carpentier et al.
techniques or in a glovebox (<1 ppm O2, <5 ppm H2O). Solvents (tolu-
ene, THF, hexane) were freshly distilled from a Na/K alloy under nitro-
gen and degassed thoroughly by freeze–thaw–vacuum cycles prior to use.
Deuterated solvents (except CDCl3) were freshly distilled from a Na/K
amalgam under argon and degassed prior to use. Starting materials for
the synthesis of proligands 1–4 were purchased and used without further
purification. 2-Hydroxy-5-methyl-3-tritylbenzaldehyde (6) was prepared
by using a method reported in the literature.[18] The yttrium precursor
NCH2CH2OMe), 3.13 (s, 3H; OCH3), 3.53 (s, 4H; ArCH2), 6.72 (s,
2HAr), 7.31 (s, 2HAr), 8.47 ppm (s, 2HAr; OH); 13C{1H} NMR (75 MHz,
CDCl3, 298 K): d=20.78 (C
37.83 (C(CH3)2), 44.38
A
ACHTUNGTRENNUNG
A
ACHTUNGTRENNUNG
(NCH2CH2OMe), 58.18 (OCH3), 71.04 (NCH2CH2OMe), 123.04, 126.34,
129.23, 132.33, 134.14 (Ar), 154.14 ppm (CArOH); ESI-HRMS: m/z: calcd
for C33H54NO3 [M+H]+: 512.410; found: 512.411; calcd for C33H53NO3Na
[M+Na]+: 534.3923; found: 534.3923.
[Y{NACHTUNGTRENNUNG(SiHMe2)2}3ACHTUNGTRENNUNG
(thf)2] was prepared by using a literature procedure.[19]
4-Methyl-2-{2-[4-(trifluoromethyl)phenyl]propan-2-yl}phenol
(4):
A
Racemic lactide (Aldrich) was recrystallized twice from dry toluene and
then sublimed under vacuum at 508C. Racemic b-butyrolactone (Al-
drich) was freshly distilled from CaH2 under nitrogen and degassed thor-
oughly by freeze–thaw–vacuum cycles prior to use.
Schlenk flask was charged with AlCl3 (20 mg, 0.15 mmol) and p-cresol
(1.00 g, 9.25 mmol), and then heated for 30 min at 1608C. The mixture
was cooled down to 608C, 1-(prop-1-en-2-yl)-4-(trifluoromethyl)benzene
(0.43 g, 2.31 mmol) was added, and the reaction was allowed to proceed
with stirring for 24 h at 1108C. The reaction mixture was cooled down to
room temperature and diluted with water (ca. 20 mL) and CHCl3 (ca.
20 mL). The organic layer was separated and treated with 5% aqueous
H2SO4 (15 mL) and 5% aqueous KOH (15 mL). The combined organic
extracts were dried over anhydrous sodium and then concentrated under
vacuum. The crude material was purified by flash chromatography
through a plug of silica gel, using CH2Cl2 as the eluent. Pure 4 was isolat-
ed as a pale yellow oil (0.37 g, 55%). 1H NMR (500 MHz, CDCl3,
Instruments and measurements: NMR spectra were recorded on Bruker
AC 200, AC 300, and AC 500 spectrometers in Teflon-valved NMR spec-
troscopy tubes. 1H and 13C chemical shifts are reported in ppm versus
SiMe4 and were determined by reference to the residual solvent resonan-
ces. Assignment of signals was carried out by means of multinuclear 1D
(1H, 13C{1H}) and 2D (1H–1H NOESY, 1H–13C HMBC, and HMQC)
NMR spectroscopy experiments.
Size-exclusion chromatography (SEC) of PLAs and PHBs was performed
in THF (1 mLminÀ1) at 208C using a Polymer Laboratories PL50 appara-
tus equipped with PLgel 5 mm MIXED-C columns (300ꢃ7.5 mm), and
combined RI and dual-angle LS (PL-LS 45/908) detectors. The number-
average molecular weight (Mn) values and polydispersity index (Mw/Mn)
of the polymers were calculated with reference to a universal calibration
versus polystyrene standards. The Mn values of PLAs were corrected by
a factor of 0.58 to account for the difference in hydrodynamic volumes
between polystyrene and polylactide.[20] The microstructure of PLAs was
298 K): d=1.72 (s, 6H; CACTHNUTRGNEUNG(CH3)2), 2.38 (s, 3H; ArCH3), 4.18 (s, 1H; OH),
6.60 (d, J=7.8 Hz, 1H; HAr ortho OH), 7.01 (d, J=7.8 Hz, 1H; HAr meta
OH), 7.33 (s, 1H; HAr ortho cumyl), 7.43 (d, J=8.1 Hz, 2H; HAr meta
CF3), 7.58 ppm (d, J=8.1 Hz, 2H; HAr ortho CF3); 13C{1H} NMR
(75 MHz, CDCl3, 298 K): d=20.94 (ArCH3), 29.47 (C
AHCTUNGTRENNUNG
ACHTUGNTRENN(UNG CH3)2), 41.89 (C-
(CH3)2), 117.31, 125.54, 126.24, 126.93 (q, 1JC,F =284.5 Hz; CF3), 127.36,
128.61, 129.92, 134.38, 151.13 (Ar), 153.82 ppm (CArOH); HRMS (70 eV,
EI): m/z calcd for C17H17OF3 [M]+: 294.123; found: 294.122.
1
2ACHTUNGTRENNUNG
gand (ONOOCMe (CF Ph))H2 (5): 2-Methoxyethylamine (63.0 mg,
ProACTHUNGTRENNUlG iHCATUNGTRENNNUG
3
determined by homodecoupling H NMR spectroscopy at 208C in CDCl3
with a Bruker AC-500 spectrometer operating at 500 MHz. The micro-
structure of PHBs (reprecipitated with methanol from a solution in di-
chloromethane) was determined by analyzing the carbonyl region of the
13C{1H} NMR spectra recorded using approximately 0.2 molLÀ1 solutions
in CDCl3 in 5 mm tubes at 408C with a Bruker AC-500 spectrometer op-
erating at 125 MHz (see text for details). NMR spectroscopy data acquis-
ition was carried out with the Bruker pulse program Zgpg and the fol-
lowing parameters: pulse time (P1)=9.40 ms, relaxation time (D1)=1.5 s,
real transform size (SI)=64 k, number of scans=30000.
0.85 mmol) and aqueous paraformaldehyde (69.0 mg of a 37% solution,
0.85 mmol) were added to a solution of 4 (0.50 g, 0.33 mmol) in methanol
(5 mL). The reaction was allowed to proceed with stirring for 24 h at
908C. The reaction mixture was cooled down to room temperature; con-
centrated to approximately 3 mL under vacuum, and left to proceed
under stirring at room temperature for 24 h. The precipitated solids were
filtered off and dried under reduced pressure to give a white powder. Re-
crystallization of the latter crude product from isopropanol at À308C
gave colorless crystals of analytically pure 5 (0.30 g, 52%). M.p. 119–
1208C; 1H NMR (500 MHz, CDCl3, 298 K): d=1.66 (s, 12H; C
ACHTUNGTRENNUNG(CH3)2),
4-Methyl-2-(2,3,3-trimethylbutan-2-yl)phenol (2): A Schlenk flask was
charged with AlCl3 (20.0 mg, 0.15 mmol) and p-cresol (1.00 g,
9.25 mmol), and the mixture was heated for 30 min at 1608C. The reac-
tion mixture was cooled to room temperature and 2,3,3-trimethyl-1-
butene (0.90 g, 9.16 mmol) was added. The reaction was allowed to pro-
ceed under stirring for 24 h at 1108C. The reaction mixture was cooled
down to room temperature and was treated with 5% aqueous H2SO4
(20 mL) and 5% aqueous KOH (20 mL). The organic layer was separat-
ed, dried with anhydrous Na2SO4, and volatiles were evaporated under
vacuum. The pure product was obtained after fractional distillation under
vacuum at 1008C, to give an oil that rapidly solidified to a white powder
(0.56 g, 30%). 1H NMR (500 MHz, CDCl3, 298 K): d=0.94 (s, 9H; C-
2.30 (s, 6H; ArCH3), 2.44 (t, J=5.1 Hz, 2H; NCH2CH2OMe), 2.80 (s,
3H; OCH3), 2.91 (t, J=5.1 Hz, 2H; NCH2CH2OMe), 3.54 (s, 4H;
ArCH2), 6.73 (s, 2HAr), 7.17 (s, 2HAr), 7.31 (d, J=8.0 Hz, 2H; HAr), 7.47
(d, J=8.0 Hz, 2H; HAr), 8.97 ppm (s, 2H; OH); 13C{1H} NMR (75 MHz,
CDCl3, 298 K): d=20.48 (ArCH3), 28.96 (C
51.26 (ArCH2), 57.25 (NCH2CH2OMe), 57.95 (OCH3), 71.42
(NCH2CH2OMe), 124.27, 126.96 (q, 1J
(C,F)=298.7 Hz, CF3), 127.14,
ACHTUNTGRNENGU(CH3)2), 41.86 (CACHTUNGTRENNUNG(CH3)2),
AHCTUNGTRENNUNG
127.52, 129.39, 135.45, 151.95 (Ar), 156.04 ppm (CArOH); 19F{1H} NMR
(188 MHz, CDCl3, 298 K): d=À62.68 ppm (s, 6F); ESI-HRMS: m/z
calcd for C39H44NO3F6 [M+H]+: 688.3225; found: 688.3224; calcd for
C39H43NO3F6Na [M+Na]+: 710.304; found: 710.303.
A
E
(E)-2-[(2-Methoxyethylimino)methyl]-4-methyl-6-tritylphenol (7): 2-Me-
thoxyethylamine (0.40 g, 5.24 mmol) was added to a solution of 2-hy-
droxy-5-methyl-3-tritylbenzaldehyde (6) (2.00 g, 5.24 mmol) in MeOH/
CHCl3 (20 mL, 50:50 v/v). The reaction mixture was heated at reflux
under stirring for 6 h. After cooling down to room temperature, the pre-
cipitated solids were removed by filtered and dried under reduced pres-
sure to give 7 as a yellow powder (1.92 g, 83%). 1H NMR (500 MHz,
CDCl3, 298 K): d=2.24 (s, 3H; ArCH3), 3.31 (s, 3H; OCH3), 3.57–3.64
(m, 2H; NCH2CH2OMe), 3.78 (m, 2H; NCH2CH2OMe), 7.04–7.24 (m,
17H; HAr), 8.30 (s, 1H; N=CH), 13.33 ppm (s, 1H; OH); 13C{1H} NMR
(75 MHz, CDCl3, 298 K): d=20.85 (ArCH3), 58.98 (OCH3), 63.16
(CH2N), 71.69 (CH2OMe), 118.71, 125.51, 126.13, 127.15, 130.93, 131.09,
134.38, 145.63 (Ar), 157.93 (CArOH), 166.43 ppm (C=N).
6.59 (d, J=7.9 Hz, 1HAr), 6.89 (d, J=9.4 Hz, 1HAr), 7.09 ppm (s, 1HAr);
13C{1H} NMR (75 MHz, CDCl3, 298 K): d=20.83 (ArCH3), 26.01 (C-
ACHTUNGTRENNUNG(CH3)2), 26.49 (CACHTUNGTRENNUNG(CH3)3), 37.70 (CAHCNUTEGRTGUN(NN CH3)3), 43.98 (CACHTUNGTRENN(UNG CH3)2), 117.26,
127.55, 128.52, 132.32, 132.71 (Ar), 152.61 ppm (CArOH); HRMS (70 eV,
EI): m/z calcd for C17H17NO3Cl4 [M]+: 422.996; found: 422.999.
Pro
3.61 mmol) and aqueous paraformaldehyde (0.69 g of a 37% solution,
7.27 mmol) were added to a solution of 2 (1.52 g, 7.28 mmol) in methanol
(5 mL). The reaction was allowed to proceed under stirring for 72 h at
808C. The reaction mixture was cooled down to room temperature; a
white powder precipitated, which was filtered off from the liquid and
dried under reduced pressure to give a white solid. Recrystallization of
the latter solid from isopropanol at À308C gave white crystals of analyti-
cally pure 3 (2.26 g, 60%). M.p. 149–1508C; 1H NMR (CDCl3, 500 MHz,
2-[(2-Methoxyethylamino)methyl]-4-methyl-6-tritylphenol (8): NaBH3CN
(0.22 g, 3.43 mmol) was added to a solution of 7 (1.00 g, 2.29 mmol) in
ethanol (10 mL) at room temperature. The reaction mixture was stirred
for 72 h at room temperature; the solution was acidified by adding con-
298 K): d=1.23 (s, 18H; C
ArCH3), 2.37 (t, J=5.1 Hz, 2H; NCH2CH2OMe), 3.06 (t, J=5.1 Hz, 2H;
1880
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 1872 – 1883

Polymerization of Lactide and b-Butyrolactone
FULL PAPER
centrated hydrochloric acid (ca. 0.1 mL of a 12m solution) and volatiles
were removed under vacuum. The residue was dissolved in saturated
aqueous Na2CO3 (ca. 20 mL), the aqueous layer was extracted with
CHCl3 (3ꢃ20 mL), the combined organic extracts were dried over anhy-
drous Na2SO4, and then concentrated under vacuum. Recrystallization of
the residue from ethanol gave colorless crystals of 8 (0.58 g, 58%).
1H NMR (500 MHz, CDCl3, 298 K): d=2.20 (s, 3H; ArCH3), 2.53 (t, J=
3.1 Hz, 2H; NCH2CH2OMe), 3.23 (t, J=3.1 Hz, 3H; NCH2CH2OMe),
3.29 (s, 3H; OCH3), 3.93 (s, 2H; ArCH2), 6.81 (s, 1HAr), 6.91 (s, 1HAr),
7.20–7.27 ppm (m, 15H; HAr); 13C{1H} NMR (75 MHz, CDCl3, 298 K):
d=20.94 (ArCH3), 46.94 (ArCH2), 51.99 (NCH2CH2OMe), 58.84 (CPh3),
63.24 (OCH3), 70.89 (NCH2CH2OMe), 120.02, 122.66, 125.40, 126.63,
127.08, 128.59, 130.66, 131.38. 134.20, 146.15 (Ar), 154.42 ppm (CArOH).
vacuum to give Y-3 as a white powder. Recrystallization from hexane at
À308C gave colorless crystals of Y-3 (0.240 g, 75%); crystals suitable for
X-ray diffraction were obtained from this batch. 1H NMR (500 MHz,
C6D6, 298 K): d=0.60 (d, 3J
18H; C
(s, 6H; ArC
(s, 6H; 2ArCH3), 2.83 (t, J
3H; OCH3), 3.06 (d, 2J(H,H)=12.0 Hz, 2H; ArCH2), 3.82 (d, 2J
ACHTUNGTRENNUNG
A
ACHTUNGTRENNUNG
A
ACHTUNGTRENNUNG
A
ACHTUNGTRENNUNG
3
AHCTUNGTRENNUNG
U
12.0 Hz, 2H; ArCH2), 3.91 (m, 4H; a-CH2 THF), 5.17 (m, 2H; SiH),
1
6.87 (s, 2HAr), 7.44 ppm (s, 2HAr); H NMR (500 MHz, [D8]THF, 298 K):
d=0.17 (d, 3J
N
N
ACHTUNGTRENNUNG(CH3)3),
A
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
ProACHTUNGTRENNUNGliACHTUNGTRENNUNG
gand (ONOOCPh )H2 (9): Salicaldehyde 6 (0.19 g, 0.43 mmol) was
3
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
added to a solution of aminophenol 8 (0.17 g, 0.43 mmol) in ethanol/
chloroform (10 mL, 50:50 v/v) with two drops of acetic acid. The reaction
mixture was stirred for 24 h at room temperature. Then, the precipitated
solids were filtered off and dried under reduced pressure to give a white
powder (0.26 g, 76%). This material was dissolved in ethanol/chloroform
(50:50 v/v, 15 mL) and NaBH4 (18.5 mg, 0.48 mmol, 1.5 equiv) was
added. The reaction mixture was stirred for 72 h at room temperature;
the solution was acidified by adding concentrated hydrochloric acid (ca.
0.1 mL of a 12m solution) and volatiles were removed under vacuum.
The solid residue was dissolved in saturated aqueous Na2CO3 (25 mL),
the aqueous layer was extracted with CHCl3 (3ꢃ20 mL), the combined
organic extracts were dried over anhydrous Na2SO4, and then concentrat-
ed under vacuum to give a white powder. Recrystallization of this prod-
uct from isopropanol/dichloromethane (=50:50 v/v) gave colorless crys-
tals of 9 (0.16 g, 60%). M.p. 181–1888C; 1H NMR (500 MHz, CDCl3,
298 K): d=2.16 (s, 6H; ArCH3), 2.31 (t, J=2.2 Hz, 2H; NCH2CH2OMe),
2.97 (t, J=2.1 Hz, 2H; NCH2CH2OMe), 3.02 (s, 3H; OCH3), 3.63 (s, 4H;
ArCH2), 6.75 (s, 2HAr), 6.86 (s, 2HAr), 7.16–7.21 (m, 30H; H trityl),
7.80 ppm (m, 2H; OH); 13C{1H} NMR (75 MHz, CDCl3, 298 K): d=
21.03 (ArCH3), 49.99 (NCH2CH2OMe), 53.50 (ArCH2), 58.69 (OCH3),
63.10 (CPh3), 70.84 (NCH2CH2OMe), 120.02, 123.23, 125.77, 127.28,
127.40, 130.08, 130.79, 131.11, 133.90 (Ar), 145.55 (Cipso trityl),
152.61 ppm (CArOH); HRMS (70 eV, EI): m/z calcd for C57H53NO3 [M]+:
799.402; found: 799.404; ESI-HRMS: m/z calcd for C57H54NO3 [M+H]+:
800.403; found: 800.404; calcd for C57H53NO3Na [M+Na]+: 822.393;
found: 822.394.
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
G
R
G
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
CH2 THF), 74.67 (NCH2CH2OMe), 122.09, 126.43, 130.58, 133.26, 133.84,
163.16 ppm; elemental analysis calcd (%) for C41H73N2O4Si2Y (803.11): C
61.32, H 9.16, N 3.49; found: C 61.5, H 9.4, N 3.4.
CHTUNGTRENNUNG
{ONOOCMe (CF Ph)}{N
ACHTUGNTERNNUGN AHCUNGTRENNUNG(T SiHMe2)2}ACUNGTNERTHUNG(thf)] (Y-5): A solution of
2ACTHNGUTRENNUG
3
3
Complex [Y
proli
gand {ONOOCMe (CF Ph)}H2 (5) (100 mg, 0.15 mmol) in hexane (3 mL)
was added dropwise at room temperature over 10 min to a stirred solu-
tion of [Y{N(SiHMe2)2}3A(thf)2] (92.0 mg, 0.15 mmol) in toluene (4 mL).
A
A
E
CHTUNGTRENNUNG
The reaction mixture was stirred for 24 h at room temperature and vola-
tiles were removed under vacuum. The residue was washed with hexane
(2ꢃ3 mL) and dried under vacuum to give complex Y-5 as a white
powder (61.0 mg, 42%). 1H NMR (500 MHz, C6D6, 298 K): d=0.40 (d,
3J
6H; C
(s, 6H; ArCH3), 2.71 (s, 3H; OCH3), 2.73 (m, 2H; NCH2CH2OMe), 2.77
(d, 2J
(H,H)=12.0 Hz, 2H; ArCH2), 2.86 (m, 4H; a-CH2 THF), 3.55 (d,
2J
(H,H)=12.0 Hz, 2H; ArCH2), 4.62 (m, 2H; SiH), 6.80 (s, 2HAr), 7.06
(s, 2HAr), 7.29 (d, JACTHNUGRTNE(NUNG H,H)=8.0 Hz, 4HAr), 7.33 ppm (d, JACTHUNGRTENNU(G H,H)=8.0 Hz,
N
ACHTUNGTRENNUNG
G
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
3
3
4HAr); 19F{1H} NMR (188 MHz, C6D6, 298 K): d=À62.04 ppm (s, 6F);
13C{1H} NMR (75 MHz, [D8]THF, 298 K): d=4.66 (SiH
ACHTUNGTRENNUNG
(ArCH3), 25.90 (b-CH2 THF), 26.36 (C
ACHTUGTNRNENUG(CH3)2), 30.62 (CACTHUNGTRENNNUG
ProACHTUNGTRENNUNGliACHTUNGTRENNUNG
gand (ONOOCl)H2 (10): 2-Methoxyethylamine (0.52 mL, 6.00 mmol)
(C
AHCTUNGTRENNUNG
(a-CH2 THF), 68.19 (NCH2CH2OMe), 123.06, 124.95, 125.29, 125.74,
127.47, 128.15, 128.47, 128.79, 131.22 (CAr), 161.15 ppm (CArO); quartet
resonances for CF3 groups were not observed due to their low intensity;
elemental analysis calcd (%) for C47H63F6N2O4Si2Y (979.08): C 57.66, H
6.49, N 2.86; found: C 57.9, H 6.8, N 3.0.
and two drops of acetic acid were added to a solution of 3,5-dichloro-2-
hydroxybenzaldehyde (1.00 g, 5.26 mmol) in methanol (20 mL). The reac-
tion mixture was stirred for 10 min at room temperature and then
sodium cyanoborohydride (0.33 g, 5.26 mmol) was added to the solution
under stirring. The reaction was allowed to proceed for 72 h at room tem-
perature. Then, the solution was acidified by adding concentrated hydro-
chloric acid (ca. 0.5 mL of an aqueous 12.0m solution) and volatiles were
removed under vacuum. The residue was dissolved in saturated aqueous
Na2CO3 (ca. 20 mL), the aqueous layer was extracted with CHCl3 (3ꢃ
20 mL), the combined organic extracts were dried over anhydrous
sodium sulfate, and then concentrated under vacuum to give crude 10 as
a yellow powder. Recrystallization from isopropanol gave white crystals
of analytically pure 10 (1.40 g, 63%). M.p. 118–1198C; 1H NMR
(500 MHz, CDCl3, 298 K): d=2.76 (t, J=5.3 Hz, 2H; NCH2CH2OMe),
3.43 (s, 3H; OCH3), 3.62 (t, J=5.3 Hz, 2H; NCH2CH2OMe), 3.82 (s, 4H;
ArCH2), 7.00 (s, 2HAr), 7.29 ppm (s, 2HAr); 13C{1H} NMR (75 MHz,
CDCl3, 298 K): d=51.73 (ArCH2), 55.91 (NCH2CH2OMe), 59.05
(OCH3), 70.36 (NCH2CH2OMe), 121.67, 124.23, 124.47, 128.53, 128.92
(Ar), 150.88 ppm (CArO); HRMS (70 eV, EI): m/z calcd for C17H17NO3Cl4
[M]+: 422.996; found: 422.999; m/z calcd for C15H12NO2Cl4
[MÀCH2OMe]+: 377.962; found: 377.960.
Complex [Y
(ONOOCPh ){N
E
E
li-
3
gand (ONOOCPh )H2 (9) (100 mg, 0.13 mmol) in toluene (4 mL) was
added dropwise at room temperature over 10 min to a stirred solution of
[Y{N(SiHMe2)2}3A(thf)2] (82.0 mg, 0.15 mmol) in toluene (4 mL). The mix-
3
A
CHTUNGTRENNUNG
ture was stirred for 24 h at room temperature and volatiles were removed
under vacuum. The residue was washed with hexane (2ꢃ3 mL) and dried
under vacuum to give Y-9 as a white powder (82.0 mg, 57%). 1H NMR
(500 MHz, [D8]THF, 298 K): d=0.11 (d, 3J
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
58.82 (OCH3), 62.49 (ArCH2), 64.10 (CPh3), 64.40 (CPh3), 68.0 (a-CH2
THF), 79.49 (NCH2CH2OMe), 115–140 (CAr), 146.87 (Cipso trityl),
153.82 ppm (CArO); elemental analysis calcd (%) for C65H73N2O4Si2Y
(1091.36): C 71.53, H 6.74, N 2.57; found: C 71.8, H 6.9, N 2.5.
Complex [Y(ONOOCMe tBu){N
(SiHMe2)2}
(thf)] (Y-3): A solution of pro
li-
2
gand (ONOOCMe tBu)H2 (3) (200 mg, 0.39 mmol) in hexane/toluene (50:50
2
v/v, 2 mL) was added at room temperature to a stirred solution of [Y{N-
(SiHMe2)2}3A(thf)2] (248 mg, 0.39 mmol) in hexane (2 mL). The reaction
Complex [(Y
lar to that for the synthesis of Y-3, a solution of [Y{N
(92.0 mg, 0.15 mmol) in hexane (5 mL) and a solution of pro
(ONOOCl)H2 (10) (300 mg, 0.70 mmol) in toluene (5 mL) were reacted at
(ONOOCl){N
ACHUTGTNRENNUG CAHTUNGTRENNUNG
A
U
A
CHTUNGTRENNUNG
mixture was stirred for 12 h at 458C and volatiles were removed under
vacuum. The residue was washed with hexane (2ꢃ3 mL) and dried under
A
A
Chem. Eur. J. 2011, 17, 1872 – 1883
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1881

J.-F. Carpentier et al.
room temperature to yield after evap-
oration of volatiles and washing with
Table 5. Crystal data and structure refinement for complexes Y-3 and Y-10.
[Y(ONOOCMe tBu){N
C
N
(ONOOCl){N
ACHTUTGNRENNGU AHCUTNGTREN(NUGN SiHMe2)2})2]
2
hexane (2ꢃ5 mL) crude Y-10 as
a
3)
ACHTUNGTRENNUNG
white powder (355 mg, 74%). Dissolu-
tion of this solid in hexane/THF
(50:50 v/v, 6 mL) and recrystallization
at À308C for 20 days afforded Y-10
(90 mg, 19%) as analytically pure col-
orless crystals, some of which proved
suitable for X-ray diffraction studies.
Elemental analysis calcd (%) for
C42H58Cl8N4O6Si4Y2 (1288.71): C 39.14,
H 4.54, N 4.35; found: C 39.4, H 4.7,
formula
C41H73N2O4Si2Y1
803.1
orthorhombic
P212121
10.306(14)
17.511(2)
24.191(3)
90
2(C42H58Cl8N4O6Si4Y2)·5
3038.08
ACHTUNGTRENNUNG(toluene)
Mr [gmolÀ1
]
crystal system
space group
a [ꢂ]
b [ꢂ]
c [ꢂ]
triclinic
¯
P1
14.838(6)
22.684(10)
22.818 (11)
70.524(2)
87.449(2)
79.397(2)
7115.9(6)
2
a [8]
b [8]
90
90
N
4.2. Room temperature 1H NMR
g [8]
V [ꢂ3]
Z
4365.6(9)
4
1.222
1.430
1728
spectra of this compound in C6D6 and
[D8]THF showed many broad overlap-
ping resonances and proved uninfor-
mative.
1calcd [gcmÀ3
]
1.418
2.039
3124
absorption coefficient [mmÀ1
(000)
crystal size [mm3]
]
F
A
Crystal structure determination of 5, 9,
Y-3, and Y-10: Diffraction data were
collected at 100(2) K using a Bruker
APEX CCD diffractometer with
graphite-monochromatized MoKa radi-
ation (l=0.71073 ꢂ). A combination
of w and f scans was carried out to
obtain at least a unique data set. The
crystal structures were solved by direct
methods; remaining atoms were locat-
ed from difference Fourier synthesis
followed by full-matrix least-squares
refinement based on F2 (programs
SIR97 and SHELXL-97)[21] with the
aid of the WINGX program.[22] Many
hydrogen atoms could be found from
the Fourier difference analysis. Other
hydrogen atoms were placed at calcu-
lated positions and forced to ride on
the attached atom. The hydrogen atom contributions were calculated but
not refined. All non-hydrogen atoms were refined with anisotropic dis-
placement parameters. The locations of the largest peaks in the final dif-
ference Fourier map calculation as well as the magnitude of the residual
electron densities were of no chemical significance. Crystal data and de-
tails of data collection and structure refinement for the different com-
pounds are given in Table 5. The principal crystallographic data (exclud-
ing structure factors) are available as Supporting Information, as cif files.
0.20ꢃ0.12ꢃ0.10
3.05 to 27.48
À13ꢀhꢀ13
À22ꢀkꢀ22
À31ꢀlꢀ31
0.20ꢃ0.15ꢃ0.08
3.4 to 27.5
À18ꢀhꢀ19
À29ꢀkꢀ29
À29ꢀlꢀ29
100689/32241 (R
0.986
q range for data collection [8]
index ranges
reflections collected/unique
completeness to qmax
absorption correction
max. and min. transmission
data/restraints/params
goodness-of-fit on F2
41615/9971 (R
0.997
N
ACHTUNGTREN(NUNG int)=0.0719)
multiscan
0.867 and 0.746
9971/0/457
multiscan
0.850 and 0.650
32241/0/1348
1.018
R1=0.0505, wR2=0.1151
R1=0.1139, wR2=0.139
1.578 and À2.487
0.974
final R indices (I>2s(I))
R indices (all data)
R1=0.0393, wR2=0.0704
R1=0.0627, wR2=0.0764
0.312 and À0.375
largest diff. peak and hole
[eꢂÀ3
]
Acknowledgements
This work was supported by the Ministꢄre de l’Enseignement Supꢀrieur
et de la Recherche (Ph.D. fellowship to M.B.), the Region Bretagne
(PRIR POLYBIO; postdoctoral fellowship to N.A.), the CNRS, and the
Institut Universitaire de France (IUF, J.F.C.).
Polymerization of rac-lactide: In a typical experiment (Table 3, entry 7),
in a glovebox, a Schlenk flask was charged with a solution of complex Y-
5 (10.0 mg, 10.2 mmol) in THF (1.0 mL). To this solution, rac-lactide
(0.290 g, 2.04 mmol, 200 equiv versus Y) in THF (3.6 mL) was added rap-
idly. The mixture was immediately stirred with a magnetic stirrer bar at
208C for 1 h. After an aliquot of the crude material was sampled by pip-
ette for determining monomer conversion by 1H NMR spectroscopy, the
reaction was quenched with acidic methanol (ca. 1.0 mL of a 1.2m HCl
solution in MeOH), and the polymer was precipitated with an excess
amount of methanol (ca. 100 mL). The polymer was then filtered and
dried under vacuum to a constant weight.
[1] a) M. Okada, Prog. Polym. Sci. 2002, 27, 87; b) O. Dechy-Cabaret,
B. Martin-Vaca, D. Bourissou, Chem. Rev. 2004, 104, 6147; c) J. Wu,
T.-L. Yu, C.-T. Chen, C.-C. Lin, Coord. Chem. Rev. 2006, 250, 602;
d) C. K. Williams, Chem. Soc. Rev. 2007, 36, 1573; e) C. A. Wheaton,
P. G. Hayes, B. J. Ireland, Dalton Trans. 2009, 4832; f) M. J. Stanford,
A. P. Dove, Chem. Soc. Rev. 2010, 39, 486; g) O. Coulembier, P.
Dubois in Handbook of Ring-Opening Polymerization (Eds.: P.
Dubois, O. Coulembier, J.-M. Raquez), Wiley-VCH, Weinheim,
2009, pp. 227–254; h) C. M. Thomas, Chem. Soc. Rev. 2010, 39, 165;
i) J.-F. Carpentier, Macromol. Rapid Commun. 2010, 31, 1696; j) N.
Ajellal, J.-F. Carpentier, C. Guillaume, S. M. Guillaume, M. Helou,
V. Poirier, Y. Sarazin, A. Trifonov, Dalton Trans. 2010, 39, 8363.
[2] For pioneering work in this field, see: a) S. McLain, T. Ford, N. Dry-
sdale, Polym. Prepr. (Am. Chem. Soc. Polym. Chem.) 1992, 33, 463;
b) A. Le Borgne, C. Pluta, N. Spassky, Macromol. Rapid Commun.
1994, 15, 955; c) T. Ovitt, G. W. Coates, J. Am. Chem. Soc. 2002,
124, 1316.
[3] a) N. Ajellal, D. M. Lyubov, M. A. Sinenkov, G. K. Fukin, A. V.
Cherkasov, C. M. Thomas, J.-F. Carpentier, A. A. Trifonov, Chem.
Eur. J. 2008, 14, 5440; b) J. W. Kramer, D. S. Treiter, E. W. Dunn,
P. M. Castro, T. Roisnel, C. M. Thomas, G. W. Coates, J. Am. Chem.
Soc. 2009, 131, 16042; c) J.-F. Carpentier, Angew. Chem. 2010, 122,
2720; Angew. Chem. Int. Ed. 2010, 49, 2662; d) N. Ajellal, G. Dur-
ieux, L. Delevoye, G. Tricot, C. Dujardin, C. M. Thomas, R. M.
Polymerization of rac-b-butyrolactone: In a typical experiment (Table 4,
entry 8), in a glovebox, a Schlenk flask was charged with a solution of
complex Y-5 (10.0 mg, 10.2 mmol) in toluene (0.40 mL). To this solution,
rac-b-butyrolactone (0.35 g, 4.08 mmol, 400 equiv vs. Y) in toluene
(1.30 mL) was added rapidly. The mixture was immediately stirred with a
magnetic stirrer bar at 208C. The reaction was processed and worked-up
similarly to that described above for lactide polymerization.
1882
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 1872 – 1883

Polymerization of Lactide and b-Butyrolactone
FULL PAPER
Gauvin, Chem. Commun. 2010, 46, 1032; e) E. Grunova, E. Kirillov,
T. Roisnel, J.-F. Carpentier, Dalton Trans. 2010, 39, 6739.
[4] C.-X. Cai, L. Toupet, J.-F. Carpentier, J. Organomet. Chem. 2003,
683, 131.
[5] a) C.-X. Cai, A. Amgoune, C. W. Lehmann, J.-F. Carpentier, Chem.
Commun. 2004, 330; J.-F. Carpentier, Chem. Commun. 2004, 330;
b) A. Amgoune, C. M. Thomas, T. Roisnel, J.-F. Carpentier, Chem.
Eur. J. 2006, 12, 169; c) A. Amgoune, C. M. Thomas, J.-F. Carpen-
tier, Pure Appl. Chem. 2007, 79, 2013; d) A. Amgoune, C. M.
Thomas, J.-F. Carpentier, Macromol. Rapid Commun. 2007, 28, 693.
[6] See also: X. Liu, X. Shang, T. Tang, N. Hu, F. Pei, D. Cui, X. Chen,
X. Jing, Organometallics 2007, 26, 2747.
[7] a) A. Amgoune, C. M. Thomas, S. Ilinca, T. Roisnel, J.-F. Carpentier,
Angew. Chem. 2006, 118, 2848; Angew. Chem. Int. Ed. 2006, 45,
2782; b) N. Ajellal, M. Bouyahyi, A. Amgoune, C. M. Thomas, A.
Bondon, I. Pillin, Y. Grohens, J.-F. Carpentier, Macromolecules
2009, 42, 987; c) for the syndiospecific ROP of b-functionalized-b-
lactones, see: N. Ajellal, C. M. Thomas, J.-F. Carpentier, J. Polym.
Sci. Polym. Chem. Ed. 2009, 47, 3177.
[8] B. M. Chamberlain, M. Cheng, D. R. Moore, T. M. Ovitt, E. B. Lob-
kovsky, G. W. Coates, J. Am. Chem. Soc. 2001, 123, 3229.
[9] a) W. J. Burke, R. P Smith, C. Weatherbee, J. Am. Chem. Soc. 1952,
74, 602; b) W. J. Burke, M. J. Kolbezen, C. W. Stephens, J. Am.
Chem. Soc. 1952, 74, 3601; c) W. J. Burke, E. L. M. Glennie , C .
Weatherbee, J. Org. Chem. 1964, 29, 909.
[15] Gibson et al. have reported that, in the ROP of rac-lactide promot-
ed by aluminum initiators supported by tetradentate aminophenox-
ide (SALAN) ligands, a dramatic switch in tacticity, from isotactical-
ly enriched to highly heterotactic, occurred upon small changes to
the ligand ortho substituents (i.e., H versus Cl, respectively). Yet,
exact reasons for this switch remained unclear. See: a) P. Hormnir-
un, E. L. Marshall, V. C. Gibson, A. J. P. White, D. J. Williams, J.
Am. Chem. Soc. 2004, 126, 2688; b) P. Hormnirun, E. L. Marshall,
V. C. Gibson, R. I. Pugh, A. J. P. White, Proc. Natl. Acad. Sci. USA
2006, 103, 15343.
[16] a) Scandium: H. Ma, T. P. Spaniol, J. Okuda, Angew. Chem. 2006,
118, 7982; Angew. Chem. Int. Ed. 2006, 45, 7818; b) yttrium: E. Gru-
nova, E. Kirillov, T. Roisnel, J.-F. Carpentier, Organometallics 2008,
27, 5691; c) For other examples of discrete alkyl lactate metal com-
plexes used as initiators for ROP of lactide, see reference [8].
[17] A. Steudel, J. Stehr, E. Siebel, R. D. Fischer, J. Organomet. Chem.
1998, 570, 89.
[18] a) R. X. Wang, X. Z. You, Q. J. Meng, E. A. Mintz, X. R. Bu, Synth.
Commun. 1994, 24, 1757; b) G. Casiraghi, G. Casnati, G. Puglia, G.
Sartori, G. Terenghi, J. Chem. Soc. Perkin Trans. 1 1980, 1862; c) W.
Zhang, E. N. Jacobsen, J. Org. Chem. 1991, 56, 2296.
[19] a) K. C. Hultzsch, P. Voth, K. Beckerle, T. P. Spaniol, J. Okuda, Or-
ganometallics 2000, 19, 228; b) R. Anwander, O. Runte, J. Eppinger,
G. Gerstberger, E. Herdtweek, M. Spiegler, J. Chem. Soc. Dalton
Trans. 1998, 847.
[20] a) I. Barakat, P. Dubois, R. Jerome, P. Teyssiꢀ, J. Polym. Sci. Polym.
Chem. Ed. 1993, 31, 505; b) H. Ma, J. Okuda, Macromolecules 2005,
38, 2665.
[10] D. T. Dugah, A. B. W. Skeltonb, E. E. Delbridge, Dalton Trans.
2009, 1436.
[11] a) O. Runte, T. Priermeier, R. Anwander, Chem. Commun. 1996,
1385; b) P. B. Hitchcock, M. F. Lappert, A. Singh, J. Chem. Soc.
Chem. Commun. 1983, 1499.
[12] a) S. C. F. Kui, N. Zhu, M. C. W. Chan, Angew. Chem. 2003, 115,
1666; Angew. Chem. Int. Ed. 2003, 42, 1628; b) M. C. W. Chan,
S. C. F. Kui, J. M. Cole, G. J. McIntyre, S. Matsui, N. Zhu, K. H.
Tam, Chem. Eur. J. 2006, 12, 2607; c) M. C. W. Chan, Chem. Asian J.
2008, 3, 18.
[21] 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; b) SHELXS-97, Program for the
Determination of Crystal Structures, G. M. Sheldrick, University of
Gçttingen (Germany), 1997; c) SHELXL-97, Program for the Re-
finement of Crystal Structures, G. M. Sheldrick, University of Gçt-
tingen (Germany), 1997; d) G. M. Sheldrick, Acta Crystallogr. Sect.
A 2008, 64, 112.
[13] E. L. Marshall, V. C. Gibson, H. S. Rzepa, J. Am. Chem. Soc. 2005,
127, 6048.
[22] L. J. Farrugia, J. Appl. Crystallogr. 1999, 32, 837.
[14] Y. Umezawa, S. Tsuboyama, H. Takahashi, J. Uzawa, M. Nishio, Tet-
rahedron 1999, 55, 10047.
Received: September 28, 2010
Published online: January 7, 2011
Chem. Eur. J. 2011, 17, 1872 – 1883
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1883