Scandium versus yttrium{amino-alkoxy-bis(phenolate)} complexes for the stereoselective ring-opening polymerization of racemic lactide and β-butyrolactone

Article abstract of DOI:10.1039/c4dt01206b

Scandium and yttrium amide complexes Ln{ONXO^R1,R2}(N(SiHMe₂)₂)(THF)_n(Ln = Sc, n = 0 or Y, n = 1; X = NMe₂or OMe; R¹= Cumyl or p-Cl-Cumyl; R²= Me or Cumyl) were prepared by aminolysis of Ln[N(SiHMe₂)₂]₃(THF) with the corresponding tetradentate diamino- or alkoxy-amino-bis(phenol) pro-ligands {ONXO^R1,R2}H₂. In the solid state and in toluene solution, the scandium complexes are monomeric and 5-coordinated, while the analogous yttrium complexes all bear an extra THF-coordinated molecule and are 6-coordinated. Sc{ONXO^R1,R2}(N(SiHMe₂)₂) complexes are single-site initiators for the ring-opening polymerization (ROP) of racemic lactide but are less active than their yttrium analogues Y{ONXO^R1,R2}(N(SiHMe₂)₂)(THF); also, in contrast to the latter ones, they are inactive in the ROP of the more demanding racemic β-butyrolactone. On the other hand, the scandium amide complexes feature a significantly improved control over the ROP of lactide, yielding PLAs with much narrower molecular weight distributions (D_M< 1.1 for Sc vs. 1.5-2.0 for Y). The yttrium complex with the very bulky o,p-dicumyl-substituted ligand is more heteroselective than its scandium analogue (P_r= 0.88 vs. 0.83), while the opposite is observed with complexes based on p-methyl-substituted ligands (P_r= 0.50 in toluene or 0.72-0.75 in THF for Y vs. P_r= 0.75-0.83 for Sc in toluene). These reactivity and selectivity trends are rationalized by a much more sterically crowded coordination sphere in scandium than in yttrium complexes.

Full text of DOI:10.1039/c4dt01206b

Dalton
Transactions
View Article Online
View Journal
PAPER
Scandium versus yttrium{amino-alkoxy-bis-
(phenolate)} complexes for the stereoselective
ring-opening polymerization of racemic lactide
and β-butyrolactone†
Yulia Chapurina,a Joice Klitzke,a,b Osvaldo de L. Casagrande Jr.,b
Mouhamad Awada,a Vincent Dorcet,c Evgueni Kirillov*a and
Jean-François Carpentier*a
Cite this: DOI: 10.1039/c4dt01206b
Scandium and yttrium amide complexes Ln{ONXOR1,R2}(N(SiHMe2)2)(THF)n (Ln = Sc, n = 0 or Y, n = 1; X =
NMe2 or OMe; R1 = Cumyl or p-Cl-Cumyl; R2 = Me or Cumyl) were prepared by aminolysis of Ln[N(SiH-
Me2)2]3(THF) with the corresponding tetradentate diamino- or alkoxy-amino-bis(phenol) pro-ligands
{ONXOR1,R2}H2. In the solid state and in toluene solution, the scandium complexes are monomeric and
5-coordinated, while the analogous yttrium complexes all bear an extra THF-coordinated molecule and
are 6-coordinated. Sc{ONXOR1,R2}(N(SiHMe2)2) complexes are single-site initiators for the ring-opening
polymerization (ROP) of racemic lactide but are less active than their yttrium analogues Y{ONXOR1,R2}-
(N(SiHMe2)2)(THF); also, in contrast to the latter ones, they are inactive in the ROP of the more demanding
racemic β-butyrolactone. On the other hand, the scandium amide complexes feature a significantly
improved control over the ROP of lactide, yielding PLAs with much narrower molecular weight distri-
butions (ĐM < 1.1 for Sc vs. 1.5–2.0 for Y). The yttrium complex with the very bulky o,p-dicumyl-substi-
tuted ligand is more heteroselective than its scandium analogue (Pr = 0.88 vs. 0.83), while the opposite is
observed with complexes based on p-methyl-substituted ligands (Pr = 0.50 in toluene or 0.72–0.75 in
THF for Y vs. Pr = 0.75–0.83 for Sc in toluene). These reactivity and selectivity trends are rationalized by a
much more sterically crowded coordination sphere in scandium than in yttrium complexes.
Received 24th April 2014,
Accepted 13th May 2014
DOI: 10.1039/c4dt01206b
www.rsc.org/dalton
plexes supported by multidentate bis(phenolate) type ligands
are extremely efficient catalysts/initiators for the controlled
Introduction
The ring-opening polymerization (ROP) of cyclic esters such as ROP of such cyclic esters.2 In particular, yttrium complexes
the ubiquitous lactide but also the more demanding β-lactones with dianionic, tetradentate amino-bridged bis(phenolate)
is a highly topical research field due to the renewability of some ancillary ligands {ONXO}2− having an additional alkoxy or
of these monomers and the biocompatibility and valuable amino X cap (A, Fig. 1) have appeared as a unique class of cata-
properties of the corresponding polymers.1 Extensive research lysts due to their very high activity, living character, but also
over the past 15 years has demonstrated that rare earth com-
aInstitut des Sciences Chimiques de Rennes, Organometallics: Materials and
Catalysis Laboratories, UMR 6226 CNRS-Université de Rennes 1, F-35042 Rennes
Cedex, France. E-mail: evgueni.kirillov@univ-rennes1.fr,
jean-francois.carpentier@univ-rennes1.fr
bInstituto de Química, Laboratório de Catálise Molecular, Universidade Federal do
Rio Grande do Sul, Av. Bento Gonçalves, 9500, Porto Alegre, RS 90501-970, Brazil
cInstitut des Sciences Chimiques de Rennes, Centre de diffraction X, UMR 6226
CNRS-Université de Rennes 1, F-35042 Rennes Cedex, France
†Electronic supplementary information (ESI) available: X-ray crystallographic
data for pro-ligands L1, L2, and L3 and complexes Sc-1, Sc-3, Sc-4, and Y-2 as
CIF files; the molecular structure of pro-ligands L1, L2, and L3 and complexes
Sc-1, Sc-3, Sc-4, and Y-2; 1H and 13C{1H} NMR spectra of the prepared pro-
ligands and complexes. CCDC 1002394–1002400. For ESI and crystallographic
Fig. 1
ancillary ligands.
Rare earth complexes supported by tetradentate {ONXO}2−
data in CIF or other electronic format see DOI: 10.1039/c4dt01206b
This journal is © The Royal Society of Chemistry 2014
Dalton Trans.

View Article Online
Paper
Dalton Transactions
even more because of their capacity to fine control the stereo- catalytic performance of the scandium complexes with their
selectivity of the ROP when starting from racemic mixtures of yttrium analogues. The coordination chemistries of these two
chiral cyclic esters.3 Highly heterotactic polylactide (PLA) and families of complexes are also compared.
highly syndiotactic poly(β-hydroxyalkanoate)s4 were thus
respectively prepared from racemic lactide (rac-LA) and β-lac-
tones such as β-butyrolactone (rac-BL) and substituted
analogues.
Results and discussion
Pro-ligands
The stereoselectivity in those coordination–insertion poly-
To explore the steric and possible electronic influence of sub-
stituents on the phenolate rings and side arms on stereo-
merizations mediated by nonchiral catalysts originates from a
so-called chain-end control, in which the stereoselection of the
incoming monomer unit depends on the chirality of the last
inserted monomer unit in the growing polymer chain. It is
generally acknowledged that chain-end control is favored by a
sterically congested environment at the active metal center, so
that the chirality of the growing polymer chain can effectively
exert a selection between enantiomers of the monomer.1
Hence, it is not surprising to see that the most effective cata-
lysts of type A for achieving stereocontrolled ROP have very
bulky substituents at the ortho position of the phenolate rings,
namely tert-butyl or even better adamantyl, and especially
cumyl and trityl groups.3,4b,c,5 Similarly, small metal centers
are preferred because, once enwrapped by the {ONXO}2−
ligand, they leave little space for the growing polymer chain and
coordination of the incoming monomer. Yttrium and lutetium
have hence demonstrated clearly superior stereocontrol
abilities than the (much) larger rare earths (Sm, Nd, La, etc.).6
Surprisingly enough, there are very few examples of complexes
of the small7 scandium supported by the {ONXO}2− ligand,8
and none of them was used in ROP. More generally, examples
of scandium complexes in stereoselective ROP reactions are
quite limited.6a,f,i,j,o,9
selective ROP,
a set of four pro-ligands was targeted
(Scheme 1). Those pro-ligands were selected to afford direct
comparisons with reference yttrium systems that have proven
to offer some of the best catalytic performances thus far in
ROP of rac-LA and rac-BL,3b,4b,c that is, the methoxy-amino-bis
(phenolate) system bearing ortho,para-cumyl substituents
derived from {ONOOCum,Cum}H2 (L4). In addition, the new pro-
ligands {ONNOCum,Me}H2 (L1), {ONOOCum,Me}H2 (L2) and
{ONOOCumCl,Me}H2 (L3) bearing respectively an N,N-dimethyl-
amino cap in place of the methoxy cap, and a cumyl or a
p-chloro-substituted cumyl group, to explore potential elec-
tronic effects such as in the p-CF3-substituted cumyl com-
pounds,4c were prepared. Compounds L1, L2 and L3 were
synthesized by a regular double Mannich condensation of the
corresponding substituted phenol, formaldehyde, and
diamine or methoxyamine in methanol, and isolated in
36–90% yield after recrystallization as white microcrystalline
powders. They were authenticated by elemental analysis, 1H,
13C{1H} NMR spectroscopy (see the Experimental section), and
X-ray diffraction studies (see the ESI†).
Synthesis and solution structure of scandium- and yttrium-
amide complexes
Heteroleptic scandium-{ONXOR1,R2} amide complexes were
prepared straightforwardly by amine elimination reaction from
Sc[N(SiHMe2)2]3(THF) and one equivalent of the corresponding
Herein, we report on the preparation and structural charac-
terization of a series of Sc{ONXO} amide complexes. The
{ONXO}2− ligands for this study have been selected based on
the aforementioned trends, for a potentially high stereocontrol
in ROP of rac-LA and rac-BL and an optimal comparison of the
Scheme 1 Synthesis of scandium- and yttrium-{ONXOR1,R2} amide complexes.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
pro-ligands {ONXOR1,R2}H2 at 50 °C (Scheme 1). Complexes from the combined steric crowding between the bulky ortho-
Sc-1–Sc-4 were isolated in 64–73% yields as white solids, which and para-cumyl substituents.
are air- and moisture-sensitive and readily soluble in most
In contrast to the scandium complexes, a coordinated THF
usual organic solvents. The structure of these complexes was molecule was systematically observed by 1H and 13C NMR spec-
1
established on the basis of H, 13C{1H}, 29Si and 29Si{1H} NMR troscopy in yttrium complexes Y-1, Y-2 and Y-3, as evidenced
spectroscopy in solution, elemental analysis, and single-crystal by THF resonances clearly shifted from those of the free
X-ray diffraction studies for Sc-1, Sc-3, and Sc-4.
solvent (δ(1H) 1.40 and 3.57 ppm; δ(13C) 25.72 and 67.80 ppm
As for their previously reported analogue Y-4,3b,4b,c the new in benzene-d6). This observation, which is in line with the
yttrium complexes Y-1, Y-2 and Y-3 were also synthesized by solid-state structure of Y-2 (vide infra) and with previous obser-
direct aminolysis of Y[N(SiHMe2)2]3(THF) with the corres- vations made on related rare earth complexes of the type
ponding pro-ligands L1–L3. The reactions proceeded cleanly in Ln{ONXOR1,R2}(N(SiHMe2)2)(THF),3,4 indicates that Y-1–Y-3 are
benzene-d6, at room temperature, as evidenced by NMR moni- mononuclear and six-coordinated in benzene or toluene solu-
toring. However, as mentioned earlier,4b,c the preparation of tion. The chemical shifts for the SiHMe2 resonance
these compounds on a larger scale, in toluene at room temp- (δ 4.68–4.80 ppm in benzene-d6), which are shifted
erature, and their eventual isolation was found to be more upfield compared to the chemical shift in the corresponding
complicated. A mixture of the desired complex contaminated Y[N(SiHMe2)2]3(THF) precursor, argue against a strong β(Si–H)
with 5–10% of an unidentified species was often observed, and agostic interaction with the metal center in solution.
complete purification could not be achieved. For that reason,
Solid-state structure of scandium- and yttrium-amide
complexes
complexes Y-1, Y-2 and Y-3 were conveniently generated in situ
for further use in catalysis, but were not isolated.
The 1H and 13C{1H} NMR data for Sc-1 in benzene-d6 at
338 K, and for Sc-2 and Sc-3 in benzene-d6 at room tempera-
ture contain all single sets of sharp resonances that are con-
sistent with monomeric species in solution,6o with a rigid Cs
symmetry. Both methyl groups of the silylamide group SiHMe2
are equivalent on the NMR timescale and only one multiplet is
observed for the SiHMe2; also, in the 29Si{1H} and 29Si NMR
spectra of Sc-1 and Sc-3, the silylamide groups appear as a
sharp singlet and a doublet of multiplets (1JSi–H = ca.
Crystals suitable for X-ray diffraction studies were grown at
room temperature from a concentrated heptane solution for
Sc-1 and from toluene–hexane solutions for Sc-3, Sc-4 and Y-2.
The solid-state molecular structures of these complexes are
shown in Fig. 2–5. The main crystallographic details are
reported in Tables S3 and S4 of the ESI.† Selected bond dis-
tances and angles for these compounds are given in Table 1.
All three scandium complexes exhibited essentially the
same geometry, that is, a five-coordinated metal center in a
distorted square-based pyramidal (sbp) environment; one
2
162–172 Hz, JSi–H = ca. 6.5–7.1 Hz, see the ESI†), respectively,
evidencing the equivalence of the Si atoms. The 1H resonances
for the capping X group appear at δ 1.64 (for the NMe2 group
in Sc-1) and δ 2.79 and 2.90 ppm (for the OMe group in Sc-2
and Sc-3), shifted by approximately 0.3–0.6 ppm upfield with
respect to that in the pro-ligand; this indicates that this
X (N, O) group remains coordinated in solution onto the metal
center, which is thus 5-coordinated. This assignment of solu-
tion structures of scandium compounds was supported by
single-crystal X-ray structure analyses of Sc-1, Sc-3 and Sc-4
(vide infra).
O(phenolate) atom of the {ONXOR1,R2 2−
ligand occupies the
}
axial position and the three other heteroatoms (O, N, and X) of
the ligand and the N(amide) form the square basis (trigonal
index, τ = 0.17 (Sc-1); 0.00 (Sc-3), 0.02 (Sc-4)),10 the metal lying
ca. 0.5 Å above the square basis (0.51, 0.49 and 0.45 Å, respect-
ively). This five-coordinated sbp geometry11 contrasts with the
six-coordinated (distorted) octahedral geometry observed in
In contrast, the NMR spectra of Sc-4 in benzene-d6 are quite
broad at room temperature, and heating at 333 K was necess-
1
ary to sharpen significantly the resonances. Relevant H NMR
data at this temperature include two distinct multiplets for the
SiHMe2 hydrogens (δ 4.87 and 5.01 ppm), and one sharp
doublet and one broadened doublet for the SiHMe2 groups
(δ 0.29 and 0.33 ppm), each of equal intensity, respectively.
Also, two signals for the SiHMe2 groups are also observed in
the 13C{1H} NMR spectrum, while two distinct (one sharp and
one broadened), equal intensity resonances are observed in
the 29Si{1H} NMR spectrum (δ −20.4 and −20.3 ppm; see the
ESI†). These observations evidence that Sc-4 adopts a non-sym-
metric structure in solution at this temperature. Since Sc-4 is
monomeric in the solid state (vide infra) and since such fea-
tures are not observed in the NMR spectra of Sc-2 and Sc-3, we
Fig. 2 Solid-state molecular structure of Sc{ONNOCum,Me}(N(SiHMe2)2)
(Sc-1). All hydrogen atoms are omitted for clarity; thermal ellipsoids are
assume that this dissymmetry of Sc-4 in solution originates drawn at the 50% probability level.
This journal is © The Royal Society of Chemistry 2014
Dalton Trans.

View Article Online
Paper
Dalton Transactions
Fig. 3 Solid-state
molecular
structure
of
Sc{ONOOCumCl,Me}-
(N(SiHMe2)2) (Sc-3). All hydrogen atoms are omitted for clarity; thermal
ellipsoids are drawn at the 50% probability level.
Fig. 5 Solid-state molecular structure of Y{ONOOCum,Me}(N(SiHMe2)2)-
(THF) (Y-2). Thermal ellipsoids are drawn at the 50% probability level. In
the top view, all hydrogen atoms are omitted for clarity. The bottom view
shows the parallel orientation of the phenyl cumyl rings along the
N(amido)–Y–O(THF) plane, with the (THF)C–H⋯π(phenyl) close contacts.
Fig. 4 Solid-state
molecular
structure
of
Sc{ONOOCum,Cum}-
(N(SiHMe2)2) (Sc-4). All hydrogen atoms are omitted for clarity; thermal
ellipsoids are drawn at the 50% probability level.
similar. The Sc–O(methoxy) bond lengths in Sc-3 and Sc-4
(Sc(1)–O(3), 2.268(2) and 2.249(1) Å) are expectedly shorter
than the Sc–N(capping amine) bonds in Sc-1 (Sc(1)–N(3),
2.411(2) Å); the latter bond appears longer than that observed
(2.303(6) Å) for the corresponding Sc–N(capping 2-pyridyl) in
Mountford’s monomeric Sc{ONNOtBu,tBu}Cl(pyridine) and
dimeric phenoxy-bridged [Sc{ONNOMe,Me}Cl]2;8 obviously, the
steric hindrance induced by the bulky cumyl substituents pre-
vents such dimerization in Sc-1, Sc-3 and Sc-4. The Sc–O
(phenolate) bond lengths (1.947(2)–1.969(1) Å) are in the
normal range of values found in related Sc-phenolate com-
pounds.8,9 The bond angles involving the oxygen phenolate
atoms (O(1)–Sc(1)–O(2), 104.3(5)–107.34(5)°) and the trans-
located nitrogen atoms of the bridging amine and amide
ligands (N(1)–Sc(1)–N(2), 149.07(8)–154.76(6)°) are quite
the 6-coordinate Sc{ONNOtBu,tBu}Cl(pyridine),8 forming
a
similar 5-metallacycle. The Sc–N(bridging amine) bonds
(Sc(1)–N(1), 2.320(2)–2.341(1) Å) compare well with that
observed in Sc{ONNOtBu,tBu}Cl(pyridine) (2.387(2) Å)8 and are
expectedly ca. 0.2 Å longer than the Sc–N(amide) bonds (Sc(1)–
N(2), 2.089(1)–2.109(2) Å). All those M–N/O bond distances are
ca. 0.2–0.3 Å shorter than those observed in the corresponding
Dalton Trans.
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
Table 1 Selected bond distances (Å) and angles (°) for complexes Sc-1, Sc-3, Sc-4 and Y-1
Sc{ONNOCum,Me
(N(SiHMe2)2) (Sc-1)
}
Sc{ONOOCumCl
(N(SiHMe2)2) (Sc-3)
}
Sc{ONOOCum,Cum
(N(SiHMe2)2) (Sc-4)
}
Y{ONOOCum,Me
(N(SiHMe2)2)THF (Y-1)
}
M–N(1)
M–N(2)
M–N(3)
M–O(1)
M–O(2)
M–O(3)
M⋯Si(1)
M⋯Si(2)
O(1)–M–O(2)
O(1)–M–O(3)
O(2)–M–O(3)
N(1)–M–N(2)
N(1)–M–N(3)
N(2)–M–N(3)
Si(1)–N(2)–M
Si(2)–N(2)–M
Si(1)–N(2)–Si(2)
2.341(1)
2.109(1)
2.411(1)
1.968(1)
1.952(1)
—
3.079
3.387
107.34(5)
—
—
154.76(6)
92.14(5)
73.76(5)
107.16(7)
124.59(8)
126.85(9)
2.320(2)
2.095(2)
—
1.961(2)
1.947(2)
2.268(2)
2.937
2.331(1)
2.089(1)
—
1.969(1)
1.949(1)
2.249(1)
2.901
2.611(3)
2.275(3)
—
2.151(2)
2.144(2)
2.387(2)
3.222
3.433
3.490
3.626
106.06(7)
149.12(7)
92.54(7)
149.07(8)
—
104.03(5)
93.76(5)
150.18(5)
151.48(5)
—
151.76(10)
90.77(9)
90.74(9)
152.88(11)
—
—
—
—
100.80(10)
129.11(12)
128.62(13)
99.52(7)
133.40(8)
126.92(9)
107.95(18)
129.67(18)
122.3(2)
yet six-coordinated yttrium analogues (vide infra), reflecting 2.8–3.1 Å) and from Me groups of the N(SiHMe2)2 moiety
the same difference in ionic radius between scandium and (contacts: 3.0–3.3 Å) (Fig. 5). Such weak Z–H⋯π interactions
yttrium. The strong difference in the Si(1)–N(2)–Sc and Si(2)– are not seen in the scandium complexes if one excepts a single
N(2)–Sc bond angles (ca. 17–34°; see Table 1) points to monoa- (possibly incidental) contact between only one phenyl cumyl
gostic interactions of one silyl group of the amide ligand with group and the N(SiHMe2)2 hydrogen in Sc-1 and Sc-3 (Si–H⋯π,
the metal center.12 Close Sc⋯Si(1) (2.901–3.079 Å) contacts are 3.19 and 2.93 Å, respectively). Note also that the chlorine
accordingly observed, while the other Si(2) atom lies 0.3–0.4 Å atoms of the p-Cl-cumyl substituents in Sc-3 are not involved
further away from the metal center (Table 1).
in any peculiar intramolecular interaction.
In Y-2, the yttrium atom is six-coordinated by the
{ONOOCum,Me}, amide and THF ligands in a slightly distorted
octahedral geometry that is quite similar to that of Y44a and
related yttrium complexes supported by tert-butyl and p-CF3-
cumyl-substituted ligands.3b,4c One Si–N–Y bond angle is more
acute than the other one (107.95(18) vs. 129.67(18)°), indicat-
ing Si–H⋯Y monoagostic interaction; similar differences in
Si–N–Y bond angles of the Y–N(SiHMe2) moiety have been
noted in related 6-coordinate complexes, e.g. Y-4 (111.4(4) vs.
135.7(4)°).4a On the other hand, as observed previously for
Y44a and its p-CF3-cumyl-substituted analogue,3b,4c the phenyl
rings of the two cumyl substituents adopt each an orientation
flat parallel to the (THF)O–Y–N(amide) plane. This singular
orientation of both cumyl groups can be related to close
C–H⋯π interactions involving hydrogens from THF (contacts:
Ring-opening polymerization of racemic-lactide and
rac-β-butyrolactone
The prepared {ONXOR1,R2}-scandium- and yttrium-amide com-
plexes were investigated in the stereoselective ROP of rac-
lactide (rac-LA) and rac-β-butyrolactone (rac-BL) (Scheme 2). In
some cases, isopropanol was used as a co-initiator. Representa-
tive data from a series of polymerizations conducted in
toluene or THF solution are collected in Tables 2 and 3.
All four Sc{ONXOR}(N(SiHMe2)2) complexes promote the
controlled ROP of rac-LA at 60 °C in toluene (Table 2; entries 2,
4–15), with turnover frequencies (TOF) up to ca. 900 mol(LA)
−1
mol(Sc)
h−1. However, they are very poorly active in THF
(TOF = ca. 1 h−1 at 60 °C; entry 2); this observation can likely
be accounted for by the competing role of THF vs. rac-LA,
Scheme 2
This journal is © The Royal Society of Chemistry 2014
Dalton Trans.

View Article Online
Paper
Dalton Transactions
Table 2 ROP of racemic lactide promoted by {ONXOR1,R2}-scandium and -yttrium catalytic systemsa
[LA]0/[M]/
[iPrOH]0
Temp
(°C)
Timeb
(min)
Conv.c
(%)
Mn,theo
Mn,SEC
d
e
a
e
f
Entry
Complex
Solvent
(g mol−1) (×103)
(g mol−1) (×103)
ĐM
Pr
1
2
3
4
5
6
7
8
Sc-1
Sc-1
Sc-1
Sc-1
Sc-1
Sc-1
Sc-2
Sc-2
Sc-2
Sc-3
Sc-3
Sc-3
Sc-4
Sc-4
Sc-4
Y-1
Y-1
Y-1
Y-2
Y-2
Y-2
Y-3
Y-3
Y-3
100 : 1 : 0
100 : 1 : 0
100 : 1 : 0
100 : 1 : 0
100 : 1 : 1
500 : 1 : 1
100 : 1 : 0
100 : 1 : 1
500 : 1 : 1
100 : 1 : 0
500 : 1 : 0
100 : 1 : 1
100 : 1 : 0
500 : 1 : 0
100 : 1 : 1
100 : 1 : 0
100 : 1 : 0
500 : 1 : 0
100 : 1 : 0
100 : 1 : 0
500 : 1 : 0
100 : 1 : 0
100 : 1 : 0
500 : 1 : 0
100 : 1 : 0
100 : 1 : 1
THF
THF
20
60
20
60
60
60
60
60
60
60
60
60
60
60
60
20
20
20
20
20
20
20
20
20
20
20
72 × 60
420
72 × 60
60
40
180
30
20
120
20
50
15
20
30
10
60
20
40
60
20
0
5
25
94
99
59
99
99
74
88
50
93
99
85
99
99
100
100
68
100
100
58
99
99
100
100
—
nd
—
nd
—
nd
—
nd
Toluene
Toluene
Toluene
Toluene
Toluene
Toluene
Toluene
Toluene
Toluene
Toluene
Toluene
Toluene
Toluene
Toluene
THF
THF
Toluene
THF
THF
Toluene
THF
18.0
13.5
14.3
42.5
14.3
14.3
53.3
12.7
36.0
13.4
14.3
61.2
14.3
14.3
14.4
72.1
9.8
14.4
72.1
8.3
14.3
71.3
14.4
14.4
17.8
15.4
14.8
40.6
14.6
15.1
56.6
13.2
32.5
12.9
13.9
66.1
15.7
17.5
15.2
78.5
24.1
16.7
80.3
14.5
17.1
75.7
13.0
12.9
1.15
1.07
1.11
1.09
1.06
1.08
1.06
1.05
1.06
1.10
1.16
1.07
1.09
1.97
1.86
1.61
1.88
1.84
1.70
1.47
1.55
1.62
1.51
1.12
0.82
0.75
0.77
0.77
0.76
0.75
0.77
0.79
0.80
0.80
0.82
0.83
0.83
0.54
0.73
0.72
0.50
0.74
0.72
0.50
0.75
0.74
0.87
0.89
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
40
60
20
40
20
20
THF
THF
THF
Y-4
Y-4
1
a All reactions performed with [rac-LA]0 = 1.0 M. b Reaction times were not necessarily optimized. c Monomer conversion determined by H NMR
spectroscopy (CDCl3, 298 K). d Theoretical molecular weight calculated using Mn, theo = conv(LA) × [rac-LA]0/[M or iPrOH] × MLA e Experimental
.
molecular weight determined by SEC vs. polystyrene standards and corrected by a factor of 0.58; ĐM = Mw/Mn. f Pr is the probability of racemic
linkage between monomer units, as determined from the methine region of the homonuclear decoupled 1H NMR spectrum.
Table 3 ROP of racemic β-butyrolactone promoted by {ONXOR1,R2}-yttrium catalytic systemsa
d
e
[BL]0/[M]/
[iPrOH]0
Temp
(°C)
Timeb
(min)
Conv.c
(%)
Mn,theo
Mn,SEC
(g mol−1) (×103)
(g mol−1) (×103)
ĐM
Pr
Bg
a
e
f
Entry
Complex
27
28
29
30
31
32
33
34
35
36
37
Y-1
Y-1
Y-1
Y-1
Y-2
Y-2
Y-2
Y-3
Y-3
Y-3
Y-4
100 : 1 : 0
500 : 1 : 0
100 : 1 : 1
100 : 1 : 1
100 : 1 : 0
500 : 1 : 0
100 : 1 : 1
100 : 1 : 0
500 : 1 : 0
100 : 1 : 1
100 : 1 : 0
20
20
20
20
20
20
20
20
20
20
20
2
10
0.5
0.33
1
5
0.5
1
98
90
100
87
92
86
8.4
38.7
8.6
7.5
7.9
37.0
7.3
8.6
43.0
6.7
8.2
31.6
157.5
26.2
22.1
33.3
183.5
31.2
31.3
121.9
26.3
34.3
1.39
1.45
1.08
1.06
1.63
1.87
1.74
1.83
1.41
1.32
1.16
0.81
0.84
0.85
0.85
0.82
0.85
0.85
0.86
0.87
0.87
0.88
1.2
1.3
1.3
1.3
1.6
1.7
1.7
1.6
1.6
1.6
1.3
85
100
100
78
5
0.5
1
95
a All reactions performed with [rac-BL]0 = 2.4 M in toluene. b Reaction times were not necessarily optimized. c BL conversion to PHB determined by
1H NMR spectroscopy (CDCl3, 298 K) on the crude reaction mixture. d Theoretical molecular weight calculated using Mn, theo = conv(BL) × [rac-BL]0/
[Y or iPrOH] × MBL.
e Experimental (uncorrected) molecular weight determined by SEC in CHCl3; ĐM = Mw/Mn. f Pr is the probability of racemic linkage
between monomer units and is determined by 13C{1H} NMR spectroscopy. g Bernoulli model triad test B, where B = (ii)(ss)/[(is) + (si)]2.
preventing coordination of the later monomer onto the active systems was previously noted in several instances in the litera-
metal site. In fact, these scandium catalysts are systematically ture.6a,i,j,13 No dramatic change in activity was observed within
less active than their yttrium analogues, which operate at the series Sc-1–Sc-4, although the Me2N-capped complex (Sc-1)
room temperature, in toluene and even faster so in THF appeared somewhat less active than the OMe-capped systems
(entries 16–26; compare entries 16/17, 19/20, 22/23), with TOF (compare e.g. entry 6, TOF = 100 h−1 vs. entry 9, TOF =
−1
up to ca. 750 mol(LA) mol(Y)
h−1; scandium complexes 185 h−1); the reverse trend is observed with yttrium complexes
Sc-1–Sc-4 are very poorly active at 20 °C in toluene and essen- (compare entries 16 vs. 19 and 22).
tially inactive in THF at this temperature (see e.g. entries 1 and
On the other hand, the degree of control of the scandium
3). Slower reactivity of scandium vs. yttrium and lutetium catalyst systems over the molecular weights is remarkable,
Dalton Trans.
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
with experimental Mn values (determined by SEC) in excellent active for the ROP of rac-BL under the conditions investigated
agreement with the calculated ones and quite narrow dispersi- (toluene, 60 °C, 24 h; [BL]0–[Sc]–[iPrOH]0 = 100 : 1 : 0 or
ties (ĐM < 1.1 in nearly all cases; entries 3–15). This latter para- 100 : 1 : 1). It is indeed well-known that this monomer is much
meter control compares favorably with that of yttrium-amide more reluctant to ring-open and its polymerization requires
analogues, which provide much broader (yet still unimodal) highly active catalysts.1d,f,4 In fact, yttrium complexes Y-1–Y-4
molecular weight distributions (ĐM = 1.5–2.0).14 It is thus note- showed interesting performance in the ROP of rac-BBL in
worthy that the scandium-amide complexes themselves are toluene solution at 20 °C (Table 3, entries 27–37).15 As earlier
excellent initiators, since they offer the same control over the noted for yttrium complexes of this class,4 all the PHBs
molecular weights (i.e., 100% initiation efficiency and narrow formed under those conditions had unimodal, although
dispersities) as the corresponding isopropoxide complexes broadened molecular distributions (ĐM = 1.08–1.87); the SEC
generated in situ with 1 equiv. of isopropanol. This is in con- (uncorrected) molecular weights, in the range 22.1–183.5 × 103
trast to yttrium catalysts, for which only addition of 1 (or g mol−1, follow monotonously the theoretical values calculated
more) equiv. of isopropanol onto the amide precursors from the monomer-to-metal center or the added alcohol ratio.
allowed achieving truly narrow molecular weight distributions 13C NMR analysis revealed that all those PHBs had a signifi-
(e.g., compare entries 25 and 26 in Table 2, and refer to earlier cantly syndiotactic-enriched microstructure, with Pr in the
reports3b). In addition, the PLAs produced with these Sc com- range 0.81–0.88. The Bernoulli triad model test B was in all
plexes did not exhibit significant transesterification side-pro- cases close to 1, consistent with a chain-end stereocontrol
cesses, as a single population of macromolecules with 144 Da mechanism.1d,f,4 As noted for rac-LA polymerizations, only a
increments was observed by MALDI-ToF mass spectrometry marginal variation of the stereoselectivity was observed within
(see Fig. S38 in the ESI†); this also allowed us to identify the Y-1–Y-4 series.
unambiguously the expected –OCH(CH3)2 and –OH termini
when isopropanol was used as a co-initiator.
The ROP of rac-LA promoted by scandium complexes Sc-1–
Sc-4 in toluene allowed the formation of highly heterotactic-
Conclusions
enriched PLAs, with the probability of racemic linkages Due to its quite small ionic radius, scandium forms 5-coordi-
between monomer units (Pr) in the range 0.75–0.83 (Table 2, nate amide complexes with the tetradentate, dianionic
entries 3–15). As anticipated for a regular coordination–inser- {ONXOR1,R2 2− ligands incorporating bulky ortho-cumyl R1 sub-
}
tion mechanism, for a given ligand system, these stereo- stituents. This contrasts with previous reports on 6-coordinate
selectivity values are not affected by the nature of the initiating scandium chloro complexes incorporating related tetradentate
group (amide or in situ-generated isopropoxide), consistent diamino-bis(phenolate) ligands with ortho-tBu R1 substitu-
with positioning of this group at the terminus of the propagat- ents.8 This contrasts also with the analogous, but 6-coordinate
ing chain (this also stands for yttrium complexes, compare yttrium complexes Y{ONXOCum,R2}(N(SiHMe2)2)(THF), which
entries 25 and 26). Although noticeable, only a marginal vari- all incorporate a coordinated THF molecule. Sc{ONXOCum,R2
}
ation of the stereoselectivity was observed within the Sc-1–Sc-4 (N(SiHMe2)2) complexes are less active than their yttrium ana-
series: the o,p-dicumyl-substituted catalyst Sc-4 is just slightly logues for promoting the ROP of racemic lactide and are inac-
more stereoselective than the other ones. This contrasts with tive in the ROP of the more demanding racemic
the neat increase in heterotacticity between Y-4 (Pr = 0.87–0.89) β-butyrolactone. On the other hand, these scandium amide
and the other three yttrium catalysts (Pr = 0.72–0.75). We complexes feature a significantly improved control of the ROP
assume that this difference in stereocontrol between Sc-4 and of lactide, with much narrower molecular weight distributions
Y-4, as compared to their p-methyl-substituted Sc/Y-1–Sc/Y-3 and higher heteroselectivities. These reactivity and selectivity
analogues, reflects the similar highly sterically crowded trends can be rationalized by a much more sterically crowded
environment in the four scandium systems, while with the coordination sphere in scandium than in yttrium complexes.
larger yttrium center,7 the o,p-dicumyl-substituted ligand
enables us to make a further increase in steric hindrance.
Interestingly yet, the scandium systems Sc-1–Sc-3 are systemati-
cally more stereoselective than their yttrium analogues, not
Experimental section
General considerations
only in toluene where they are essentially atactic (Pr
=
0.50–054) but also in THF (Pr = 0.72–0.75). In contrast to All manipulations were performed under a purified argon
earlier observations on the p-CF3-substituted cumyl system,4c atmosphere using standard Schlenk techniques or in a glove-
the p-Cl-substituted cumyl ligand (L3), ever associated with box (<1 ppm O2, <5 ppm H2O). Solvents were distilled from Na/
scandium or yttrium (Sc/Y-3), does not show decreased stereo- benzophenone (THF and Et2O) or Na/K alloy (toluene, hexane
selectivity as compared to the regular (non-substituted) cumyl and pentane) under argon, degassed thoroughly and stored
systems (Sc/Y-2); this is likely accounted for by the significantly under argon prior to use. Benzene-d6 and toluene-d8 (>99.5%
lower electron-withdrawing ability of Cl as compared to CF3.
D, Eurisotop) used for NMR of complexes were stored over Na/
In line with the moderate activity noted above in the ROP of K alloy and vacuum-transferred just before use. Pro-ligand L4
rac-LA, none of the scandium complexes Sc-1–Sc-4 proved was prepared following the literature procedure.3b,4c Starting
This journal is © The Royal Society of Chemistry 2014
Dalton Trans.

View Article Online
Paper
Dalton Transactions
materials for the synthesis of pro-ligands L1, L2 and L3 were (CDCl3, 125 MHz, 298 K): δ 21.2 (ArCH3), 29.7 (C(CH3)2), 42.2
purchased and used without further purification. Rare earth (C(CH3)2), 44.5 (NCH2CH2NMe2), 49.1 (NCH2CH2NMe2), 56.2
precursors Ln[N(SiHMe2)2]3(THF) (Ln = Y, Sc) were synthesized (NCH2CH2NMe2), 56.6 (ArCH2N), 122.9, 124.8, 126.3, 126.6,
by literature procedures.16 Racemic lactide (rac-LA) was 127.6, 127.7, 129.4, 136.8 (aryl-CH), 151.8, 153.2 (aryl-C). Elem.
obtained from Aldrich and racemic β-butyrolactone (rac-BL) Anal. Calcd for C38H48N2O2: C, 80.81; H, 8.57; N, 4.96; found:
from TCI. Purification of rac-LA required a three-step pro- C, 80.96; H, 8.82; N, 4.93.
cedure involving first a recrystallization from a hot, concen-
{ONOOCum,Me}H2 (L2). This compound was prepared in an
trated iPrOH solution (80 °C), followed by two subsequent analogous manner to that described above for L1, starting
recrystallizations in hot toluene (100 °C). Racemic β-butyrolac- from 2-methoxyethylamine (1.35 mL, 1.16 g, 15.5 mmol),
tone was freshly distilled from CaH2 under argon prior to use. aqueous paraformaldehyde (2.32 mL of a 37 wt% solution in
After purification, rac-LA and rac-BL were stored at a tempera- water, 30.9 mmol) and 4-methyl-2-(2-phenylpropan-2-yl)
ture of −30 °C in the glovebox.
phenol18 (7.00 g, 30.9 mmol). Workup afforded L2 as a fine
white powder (3.10 g, 36%). 1H NMR (500 MHz, toluene-d8,
Instrumentation and measurements
298 K): δ 1.71 (s, 12H, C(CH3)CH3), 2.31 (s, 6H, ArCH3), 2.48 (t,
NMR spectra of complexes were recorded on Bruker AC-300, 3J = 5.0 Hz, 2H, OCH2CH2N), 2.99 (s, 3H, OCH3), 3.08 (t, J =
Avance DRX 400 and AM-500 spectrometers in Teflon-valved 5.0 Hz, 2H, OCH2CH2N), 3.59 (s, 2H, NCH2Ar), 7.29–6.74 (m,
NMR tubes at 25 °C unless otherwise indicated. 1H and 14H, Haro), 7.71 (s, 2H, OH). 13C{1H} NMR (CDCl3, 125 MHz,
3
13C NMR chemical shifts are reported in ppm vs. SiMe4 and 298 K):
δ 21.2 (ArCH3), 29.8 (C(CH3)2), 42.2 (C(CH3)2),
were determined by reference to the residual solvent peaks. 51.4 (NCH2CH2OMe), 56.8 (ArCH2N), 58.8 (OCH3), 71.2
Assignment of resonances for organometallic complexes was (NCH2CH2OMe), 123.5, 125.5, 126.2, 127.4, 127.7, 128.2, 129.5,
made from 2D 1H–13C HMQC and HMBC NMR experiments. 136.4 (aryl-CH), 151.0, 152.6 (aryl-C). Elem. Anal. Calcd for
29Si chemical shifts were determined by external reference to C37H45NO3: C, 80.54; H, 8.22; N, 2.54; found: C, 80.63; H, 8.29;
SiMe4.
Elemental analyses (C, H, N) were performed using a Flash
EA1112 CHNS Thermo Electron apparatus and results shown propan-2-yl)phenol: A Schlenk flask was charged with AlCl3
are the average of two independent measurements. (0.090 g, 0.655 mmol) and p-cresol (0.710 g, 6.55 mmol), and
N, 2.51.
{ONOOCumCl,Me}H2 (L3). (A) 4-Methyl-2-(2-(4-chlorophenyl)
Size exclusion chromatography (SEC) analyses of PLAs were then heated for 30 min at 160 °C until a melt of aluminum
performed in THF (1.0 mL min−1) at 20 °C using a Polymer phenolate is formed. The mixture was cooled down to 60 °C.
Laboratories PL-GPC 50 plus apparatus equipped with two 2-(4-Chlorophenyl)propene (0.200 g, 1.31 mmol) was added,
ResiPore 300 × 7.5 mm columns, and RI and Dual angle LS and the reaction was stirred for 18 h at 30 °C. The reaction
(PL-LS 45/90) detectors. The number-average molecular mixture was cooled down to room temperature and diluted
masses (Mn) and dispersity (ĐM = Mw/Mn) of the polymers were with water (ca. 10 mL) and CHCl3 (ca. 15 mL). The organic
calculated with reference to a universal calibration vs. poly- layer was separated and treated with 5% aqueous H2SO4
styrene standards. Reported experimental SEC molar mass (5 mL) and then 5% aqueous KOH (5 mL). The combined
values (Mn,SEC) for PLA samples were corrected by a factor of organic extracts were dried over anhydrous sodium sulfate and
0.58 as previously established.17 Mn,SEC values for PHB then concentrated under vacuum. The crude material was pur-
samples are uncorrected. Unless otherwise stated, the SEC ified by chromatography through a plug of silica gel using
traces of the polymers all exhibited a unimodal, and usually CH2Cl2–petroleum ether (1 : 1) as the eluent. 4-Methyl-2-(2-(4-
symmetrical, peak.
chlorophenyl)propan-2-yl)phenol was isolated as a pale yellow
1
MALDI-ToF mass spectra were obtained with a Bruker Dal- oil (1.36 g, 80%). H NMR (400 MHz, CDCl3, 298 K): δ 1.56 (s,
tonic MicroFlex LT, using a nitrogen laser source (337 nm, 6H, C(CH3)2), 2.25 (s, 3H, CH3), 4.33 (s, 1H, OH), 6.52 (d, J =
3 ns) in linear mode with a positive acceleration voltage of 8.0 Hz, 1H, Haro o-OH), 6.87 (d, J = 8.0 Hz, 1H, Haro m-OH),
20 kV and using α-cyano-4-hydroxycinnamic acid as the matrix. 7.28–7.04 (m, 5H, HArOCl and HAr meta OH). 13C{1H} NMR
{ONNOCum,Me}H2
(L1). N,N-Dimethylethylenediamine (100 MHz, CDCl3, 298 K): δ 21.0 (ArCH3), 29.6 (C(CH3)2), 41.4
(1.35 mL, 1.09 g, 12.35 mmol) and aqueous paraformaldehyde (C(CH3)2, 117.5, 127.2, 127.5, 128.6, 129.0, 129.8 (aryl-CH),
(1.85 mL of a 37 wt% solution in water, 24.7 mmol) were 147.7 (aryl-C), 151.4 (CArOH).
added to a solution of 4-methyl-2-(2-phenylpropan-2-yl)phenol
(B) Pro-ligand L3: This was prepared in an analogous
(5.60 g, 24.7 mmol) in methanol (25 mL). The reaction mixture manner to that described above for L1, starting from 2-methoxy-
was heated under stirring for 72 h at 80 °C. Upon cooling, a ethylamine (0.103 g, 1.35 mmol), aqueous paraformaldehyde
white precipitate formed, which was isolated by filtration, (0.235 g of a 37 wt% solution, 2.85 mmol) and 4-methyl-2-(2-
washed with ice-cold methanol (3 × 20 mL) and dried in vacuo (4-chlorophenyl)propan-2-yl)phenol (0.750 g, 2.85 mmol) in
1
to yield a fine white powder (2.90 g, 41%). H NMR (500 MHz, methanol (2 mL). The reaction was stirred for 24 h at 90 °C.
3
CDCl3, 298 K): δ 1.47 (s, 6H, ArCH3), 1.71 (t, J = 5.0 Hz, 2H, Workup afforded crude L3 as a white powder, which was
3
NCH2CH2N), 1.79 (s, 12H, C(CH3)CH3), 1.99 (t, J = 5.0 Hz, 2H, recrystallized from isopropanol at −30 °C to give colorless crys-
NCH2CH2N), 2.28 (s, 6H, N(CH3)2), 3.18 (s, 2H, NCH2Ar), tals of analytically pure L3 (0.80 g, 90%); some of these crystals
7.36–6.65 (m, 14H, Haro), 9.36 (s, 2H, OH). 13C{1H} NMR proved suitable for X-ray diffraction studies. M.p. = 125 °C.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
1H NMR (400 MHz, CDCl3, 298 K): δ 1.56 (s, 12H, C(CH3)2), (NCH2CH2OCH3), 124.1 (aryl-Cq), 124.8 (aryl-CH), 126.2 (aryl-
2.20 (s, 6H, ArCH3), 2.36 (t, J = 5.3 Hz, 2H, NCH2CH2OMe), CH), 127.6 (aryl-Cq), 127.7 (aryl-CH), 128.7 (aryl-CH), 128.9
2.87 (overlapped s and t, 5H, OCH3 and NCH2CH2OMe), 3.47 (aryl-CH), 136.2 (aryl-Cq), 151.4 (aryl-Cq), 159.5 (aryl-Cq). Anal.
(s, 4H, ArCH2N), 6.64 (br s, 2H, m-HArOH), 7.12–7.06 (m, 10H, Calcd for C41H57N2O3ScSi2: C, 67.73; H, 7.90; N, 3.85; found:
HAr), 7.79 (br s, 2H, OH). 13C{1H} NMR (100 MHz, CDCl3, C, 67.60; H, 7.96; N, 3.79.
298 K): δ 20.9 (ArCH3), 29.5 (C(CH3)2), 41.5 (C(CH3)2, 51.2
Sc{ONOOCumCl,Me}(N(SiHMe2)2) (Sc-3). This product was
(NCH2CH2OMe), 57.2 (ArCH2N), 58.4 (OCH3), 71.3 prepared as described above for Sc-1, starting from {ONOO-
(NCH2CH2OMe), 122.9, 127.1, 127.3, 127.5, 127.6, 129.3, 130.4, CumCl,Me}H2 (0.100 g, 0.161 mmol) and Sc[N(SiHMe2)2]3(THF)
138.8 (aryl-CH), 149.9 (aryl-C), 152.2 (CArOH). ESI-HRMS: m/z (0.083 g, 0.161 mmol) to give Sc-3 as a white powder (0.090 g,
1
3
calcd for C37H44NO3Cl2 (M + H)+: 620.2698; found: 620.2692. 70%). H NMR (400 MHz, C6D6, 298 K): δ 0.27 (d, JH–H = 3.0
Anal. Calcd for C37H43NO3Cl2: C, 71.60; H, 6.98; N, 2.26; Hz, 12H, SiH(CH3)2), 1.72 (s, 6H, C(CH3)2), 1.80 (s, 6H,
3
found: C, 71.52; H, 7.06; N, 2.23.
C(CH3)2), 1.97 (t, JH–H = 5.7 Hz, 2H, NCH2CH2OCH3), 2.27 (s,
Sc{ONNOCum,Me}(N(SiHMe2)2) (Sc-1). A Schlenk flask was 6H, ArCH3), 2.37 (br, 2H, ArCH2), 2.58 (t, JH–H = 5.7 Hz, 2H,
3
charged with {ONNOCum,Me}H2 (1.00 g, 1.77 mmol) and Sc[N NCH2CH2OCH3), 2.90 (s, 3H, NCH2CH2OCH3), 3.13–3.16 (m,
3
(SiHMe2)2]3(THF) (0.910 g, 1.77 mmol), and toluene (ca. 2H, ArCH2), 4.78 (hept, JH–H = 3.0 Hz, 2H, SiH(CH3)2), 6.55 (d,
10 mL) was vacuum transferred at room temperature. The reac- JH–H = 1.4 Hz, 2H, Haro), 6.99–7.02 (m, 1H, Haro), 7.10–7.12
tion mixture was stirred overnight at 50 °C. Volatiles were then (m, 1H, Haro), 7.16–7.18 (m, 2H, Haro). 13C{1H} NMR
removed under vacuum, and the residue was washed with (100 MHz, C6D6, 298 K): δ 2.5 (SiH(CH3)2), 20.6 (ArCH3),
hexane (2 × 3 mL) and dried under vacuum to give Sc-1 as a 29.4 (C(CH3)2), 30.3 (C(CH3)2), 41.9 (C(CH3)2), 55.5
white powder (0.96 g, 73%). Dissolution of this solid in (NCH2CH2OCH3), 61.4 (ArCH2), 61.8 (NCH2CH2OCH3), 71.7
heptane and recrystallization at room temperature for 15 days (NCH2CH2OCH3), 124.3 (aryl-Cq), 124.4 (aryl-Cq), 127.7 (aryl-
afforded Sc-1 as analytically pure colorless crystals, some of CH), 127.8 (aryl-CH), 128.2 (aryl-CH), 128.6 (aryl-CH), 128.8
which proved suitable for X-ray diffraction studies. 1H NMR (aryl-CH), 128.9 (aryl-CH), 130.5 (aryl-Cq), 135.7 (aryl-Cq),
3
(400 MHz, C6D6, 338 K): δ 0.34 (d, JH–H = 2.6 Hz, 12H, SiH 150.2 (aryl-Cq), 159.3 (aryl-Cq). 29Si{1H} NMR (79 MHz, C6D6,
(CH3)2), 1.58 (m, 2H, NCH2CH2N(CH3)2), 1.64 (s, 6H, 298 K): δ −19.79 (s). 29Si NMR (79 MHz, C6D6, 298 K): δ −19.79
1
2
NCH2CH2N(CH3)2), 1.87 (s, 6H, C(CH3)2), 1.91 (s, 6H, C(CH3)2), (d of multiplets, JSi–H = 162 Hz, JSi–H = 6.6 Hz). Anal. Calcd
2.06–2.08 (m, 2H, NCH2CH2N(CH3)2), 2.26 (s, 6H, ArCH3), 3.24 for C41H55Cl2N2O3ScSi2: C, 61.87; H, 6.97; N, 3.52; found: C,
(br, 4H, ArCH2), 5.03 (sept, 3JH–H = 2.6 Hz, 2H, SiH(CH3)2), 6.59 61.75; H, 7.02; N, 3.58.
(s, 2H, Haro), 6.95–6.99 (m, 2H, Haro), 7.11 (t, JH–H = 7.6 Hz,
Sc{ONOOCum,Cum}(N(SiHMe2)2) (Sc-4). This product was pre-
4H, Haro), 7.20 (s, 2H, Haro), 7.34–7.36 (m, 4H, Haro). 13C{1H} pared as described above for Sc-1 starting from {ONOOCum,
NMR (100 MHz, C6D6, 338 K): δ 3.26 (SiH(CH3)2), 20.5 (ArCH3), Cum}H2 (0.200 g, 0.263 mmol) and Sc[N(SiHMe2)2]3(THF)
26.3 (C(CH3)2), 29.4 (C(CH3)2), 31.3 (C(CH3)2), 35.3 (C(CH3)2), (0.135 g, 0.263 mmol) to give Sc-4 as a white powder (0.160 g,
1
3
42.4 (C(CH3)2), 46.9 (NCH2CH2N(CH3)2), 52.9 (NCH2CH2N 65%). H NMR (400 MHz, C6D6, 333 K): δ 0.29 (d, JH–H = 3.0
(CH3)2), 58.4 (NCH2CH2N(CH3)2), 61.6 (ArCH2), 124.2 (aryl-Cq), Hz, 6H, SiH(CH3)2), 0.33 (br d, 6H, SiH(CH3)2), 1.57 (s, 6H,
124.5 (aryl-Cq), 124.8 (aryl-CH), 126.2 (aryl-CH), 127.4 (aryl- C(CH3)2), 1.63–1.65 (m, 12H, C(CH3)2), 1.88 (d, JH–H = 3.4 Hz,
3
CH), 127.6 (aryl-CH), 128.8 (aryl-CH), 129.2 (aryl-CH), 135.9 2H, C(CH3)2), 2.02–2.06 (m, 3H, ArCH2 + NCHHCH2OCH3),
3
(aryl-Cq), 151.6 (aryl-Cq), 159.5 (aryl-Cq). 29Si{1H} NMR 2.57 (s, 1H, NCHHCH2OCH3), 2.69 (t, JH–H = 5.7 Hz, 2H,
(79 MHz, C6D6, 338 K): δ −20.85 (s). 29Si NMR (79 MHz, C6D6, NCH2CH2OCH3), 2.93 (s, 2H, NCH2CH2OCH3), 3.10–3.20 (s,
3
338 K): δ −20.85 (doublet of multiplets, 1JSi–H = 170 Hz, 2JSi–H
=
3H, ArCH2), 4.86 (hept, JH–H = 3.0 Hz, 1H, SiH(CH3)2), 5.01
7.1 Hz). Anal. Calcd for C42H60N3O2ScSi2: C, 68.16; H, 8.17; N, (br, 1H, SiH(CH3)2), 6.67 (d, JH–H = 2.4 Hz, 1H, Haro), 6.99–7.13
5.68; found: C, 68.01; H, 8.18; N, 5.74. (m, 11H, Haro), 7.17–7.20 (m, 4H, Haro), 7.26–7.28 (m, 5H,
Sc{ONOOCum,Me}(N(SiHMe2)2) (Sc-2). This product was pre- Haro), 7.35–7.37 (m, 3H, Haro). 13C{1H} NMR (100 MHz, C6D6,
pared as described above for Sc-1 starting from {ONOOCum,Me}- 333 K): δ 2.6 (SiH(CH3)2), 3.2 (SiH(CH3)2), 29.5 (C(CH3)2), 30.1
H2 (1.00 g, 1.81 mmol) and Sc[N(SiHMe2)2]3(THF) (0.930 g, (C(CH3)2), 30.2 (ArCH2), 30.7 (C(CH3)2), 30.8 (C(CH3)2), 30.9
1.81 mmol) to give Sc-2 as a white powder (0.84 g, 64%). (C(CH3)2), 31.0 (C(CH3)2), 42.1 (C(CH3)2), 42.2 (C(CH3)2), 42.3
3
1H NMR (500 MHz, C6D6, 298 K): δ 0.34 (d, JH–H = 2.7 Hz, (C(CH3)2), 42.6 (C(CH3)2), 42.8 (C(CH3)2), 43.5 (C(CH3)2), 55.3
12H, SiH(CH3)2), 1.86 (s, 6H, C(CH3)2), 1.91 (s, 6H, C(CH3)2), (NCH2CH2OCH3), 61.4 (ArCH2), 61.9 (NCH2CH2OCH3), 71.7
3
1.94 (d, JH–H = 5.7 Hz, 2H, NCH2CH2OCH3), 2.11 (br, 2H, (NCH2CH2OCH3), 123.8 (aryl-Cq), 124.8 (aryl-CH), 125.1 (aryl-
ArCH2), 2.28 (s, 6H, ArCH3), 2.38 (br, 2H, ArCH2), 2.56 (d, 3JH–H CH), 125.2 (aryl-CH), 125.3 (aryl-CH), 125.5 (aryl-CH), 125.9
= 5.5 Hz, 2H, NCH2CH2OCH3), 2.79 (s, 3H, NCH2CH2OCH3), (aryl-CH), 126.2 (aryl-CH), 126.5 (aryl-CH), 126.6 (aryl-CH),
3
4.90 (hept, JH–H = 2.7 Hz, 2H, HSi(CH3)2), 6.56 (s, 2H, Haro), 126.7 (aryl-CH), 126.8 (aryl-CH), 127.1 (aryl-CH), 127.6 (aryl-
7.02–7.06 (m, 2H, Haro), 7.16–7.20 (m, 4H, Haro), 7.23 (d, JH–H CH), 127.7 (aryl-CH), 127.8 (aryl-CH), 127.9 (aryl-CH), 128.1
= 1.7 Hz, 2H, Haro), 7.40 (d, JH–H = 7.5 Hz, 4H, Haro). 13C{1H} (aryl-CH), 128.8 (aryl-CH), 129.0 (aryl-CH), 135.2 (aryl-Cq),
NMR (125 MHz, C6D6, 298 K): δ 2.75 (SiH(CH3)2), 20.7 (ArCH3), 135.6 (aryl-Cq), 135.7 (aryl-Cq), 136.1 (aryl-Cq), 137.4 (aryl-Cq),
29.6 (C(CH3)2), 30.4 (C(CH3)2), 42.4 (C(CH3)2), 55.1 137.7 (aryl-Cq), 137.8 (aryl-Cq), 138.8 (aryl-Cq), 151.2 (aryl-Cq),
(NCH2CH2OCH3), 61.3 (ArCH2), 61.8 (NCH2CH2OCH3), 71.6 151.4 (aryl-Cq), 151.6 (aryl-Cq), 151.7 (aryl-Cq), 151.9 (aryl-Cq),
This journal is © The Royal Society of Chemistry 2014
Dalton Trans.

View Article Online
Paper
Dalton Transactions
3
153.1 (aryl-Cq), 158.9 (aryl-Cq), 159.5 (aryl-Cq), 159.7 (aryl-Cq), (s, 6H, C(CH3)2), 2.26 (t, JH–H = 5.4 Hz, 2H, ArCH2), 2.33 (s,
160.2 (aryl-Cq). 29Si{1H} NMR (79 MHz, C6D6, 333 K): δ −20.40 6H, ArCH3), 2.73–2.76 (m, 5H, NCH2CH2OCH3 + ArCH2
+
3
(br s), −20.30 (s). 29Si NMR (79 MHz, C6D6, 333 K): δ −20.31 (d NCH2CH2OCH3), 3.07 (br s, 4H, β-CH2 THF), 3.56 (d, 2H, JH–H
of multiplets, 1JSi–H = 171.6 Hz, 2JSi–H = 6.5 Hz). Anal. Calcd for = 12.3 Hz, NCH2CH2OCH3), 4.68–4.69 (m, 2H, SiH(CH3)2), 6.73
C57H73N2O3ScSi2: C, 73.19; H, 7.87; N, 3.00; found: C, 73.24; H, (s, 2H, Haro), 7.05 (s, 2H, Haro), 7.07 (s, 2H, Haro), 7.14 (s, 4H,
7.96; N, 3.09.
NMR-scale generation of Y{ONNOCum,Me}(N(SiHMe2)2) 333 K): δ 4.1 (SiH(CH3)2), 20.6 (ArCH3), 24.7 (α-CH2 THF), 27.6
(Y-1). A J. Young NMR-tube was charged with {ONNOCum,Me
(C(CH3)2), 32.7 (C(CH3)2), 41.9 (C(CH3)2), 48.3 (ArCH2), 59.7
Haro), 7.24 (s, 2H, Haro). 13C{1H} NMR (100 MHz, C6D6,
}
H2 (0.050 g, 0.088 mmol) and Y[N(SiHMe2)2]3(THF) (0.049 g, (ArCH3), 63.6 (NCH2CH2OCH3), 70.3 (β-CH2 THF), 72.7
0.088 mmol), and benzene-d6 (ca. 0.6 mL) was added at room (NCH2CH2OCH3), 122.3 (aryl-Cq), 124.3 (aryl-Cq), 127.4 (aryl-
temperature. NMR monitoring revealed rapid, quantitative Cq), 127.5 (aryl-CH), 127.8 (aryl-CH), 128.6 (aryl-CH), 130.0
conversion of the reagents to the targeted complex, with (aryl-CH), 135.9 (aryl-Cq), 151.5 (aryl-Cq), 161.2 (aryl-Cq). 29Si
release of two equiv. of HN(SiHMe2)2. 1H NMR (400 MHz, {1H} NMR (79 MHz, C6D6, 333 K): δ −23.27 (s). 29Si NMR
3
1
C6D6, 313 K): δ 0.42 (d, JH–H = 2.8 Hz, 12H, SiH(CH3)2), 1.29 (79 MHz, C6D6, 333 K): δ −23.3 (d of multiplets, JSi–H = 173
(m, 4H, α-CH2 THF), 1.64 (br s, 2H, ArCH2), 1.71 (s, 6H, Hz, 2JSi–H = 6.3 Hz).
NCH2CH2N(CH3)2), 1.85 (s, 6H, C(CH3)2), 2.06 (s, 6H, C(CH3)2),
3
2.31 (s, 8H, ArCH2 + ArCH3), 2.77 (d, JH–H = 12.5 Hz, 2H,
General procedure for polymerization of rac-lactide
3
NCH2CH2N(CH3)2), 3.31 (br s, 4H, β-CH2 THF), 3.64 (d, JH–H
=
In a typical experiment, in a glovebox, a Schlenk flask was
charged with a solution of complex Sc-1 (10.0 mg, 13.5 µmol)
in toluene (1.0 mL). To this solution, rac-lactide (0.195 g,
1.35 mmol, 100 equiv. vs. Sc) was added rapidly. The mixture
was immediately stirred with a magnetic stirrer bar at 60 °C for
1 h. An aliquot of the crude material was periodically sampled
by pipette for determining monomer conversion by 1H NMR
spectroscopy. Rac-LA conversions were calculated from 1H
NMR spectra of the crude reaction mixtures in CDCl3, from the
integration (Int.) ratio Int.polymer/[Int.polymer + Int.monomer],
using the methyl hydrogen resonances for PLA at δ 1.49 ppm
and for LA at δ 1.16 ppm. The reaction was quenched by
adding acidic methanol (ca. 1 mL of a 1.2 M HCl solution in
MeOH), and the polymer was precipitated with an excess
amount of methanol (ca. 100 mL). The polymer was then fil-
tered and dried under vacuum to a constant weight. The
microstructure of PLAs was determined by homodecoupling
1H NMR spectroscopy (methine region) at 20 °C in CDCl3 on a
Bruker AC-500 spectrometer.
12.5 Hz, 2H, NCH2CH2N(CH3)2), 4.80–4.82 (m, 2H, HSi(CH3)2),
6.78 (d, JH–H = 1.6 Hz, 2H, Haro), 6.95 (d, JH–H = 7.3 Hz, 2H,
Haro), 7.10–7.14 (m, 4H, Haro), 7.26 (d, JH–H = 1.7 Hz, 2H,
Haro), 7.39 (d, JH–H = 7.3 Hz, 4H, Haro). 13C NMR (100 MHz,
C6D6, 313 K): δ 4.4 (SiH(CH3)2), 20.6 (ArCH3), 24.8 (α-CH2
THF), 29.1 (C(CH3)2), 31.6 (C(CH3)2), 42.4 (C(CH3)2), 45.3
(NCH2CH2N(CH3)2), 48.7 (NCH2CH2N(CH3)2), 60.0 (ArCH2),
64.9 (NCH2CH2N(CH3)2), 66.9 (β-CH2 THF), 122.3 (aryl-Cq),
124.3 (aryl-CH), 124.9 (aryl-Cq), 126.4 (aryl-CH), 127.5 (aryl-
CH), 129.3 (aryl-CH), 129.8 (aryl-CH), 136.3 (aryl-Cq), 152.6
(aryl-Cq), 161.4 (aryl-Cq).
NMR-scale generation of Y{ONOOCum,Me}(N(SiHMe2)2)
(Y-2). This product was prepared in situ as described above for
Y-1, starting from {ONOOCum,Me}H2 (0.050 g, 0.091 mmol) and
Y[N(SiHMe2)2]3(THF) (0.051 g, 0.091 mmol). 1H NMR
3
(500 MHz, C6D6, 313 K): δ 0.45 (d, JH–H = 2.8 Hz, 12H, SiH
(CH3)2), 1.29 (m, 4H, α-CH2 THF), 1.72 (s, 6H, C(CH3)2), 2.15 (s,
6H, C(CH3)2), 2.29 (t, 3JH–H = 5.2 Hz, 2H, NCH2CH2OCH3), 2.35
(s, 6H, ArCH3), 2.73–2.80 (m, 7H, NCH2CH2OCH3 + ArCH2
+
NCH2CH2OCH3), 3.14 (br s, 4H, β-CH2 THF), 3.62 (br s, 2H,
ArCH2), 4.77 (m, 2H, SiH(CH3)2), 6.76 (s, 2H, Haro), 6.94 (t, JH–
General procedure for polymerization of rac-β-butyrolactone
= 7.3 Hz, 2H, Haro), 7.09 (t, JH–H = 7.5 Hz, 4H, Haro), 7.29 In a typical experiment, in the glovebox, a Schlenk flask was
H
(m, 2H, Haro), 7.37 (d, JH–H = 8.0 Hz, 4H, Haro). 13C{1H} NMR charged with a ligand {ONNOCum,Me}H2 (6.64 mg, 11.7 µmol)
(125 MHz, C6D6, 313 K): δ 4.2 (SiH(CH3)2), 20.7 (ArCH3), 24.8 and Y[N(SiHMe2)2]3(THF) (6.56 mg, 11.7 µmol). Toluene
(α-CH2 THF), 27.7 (C(CH3)2), 33.0 (C(CH3)2), 42.7 (C(CH3)2), (0.5 mL) was added at room temperature and the reaction
48.2 (NCH2CH2OCH3), 59.5 (ArCH2), 63.9 (NCH2CH2OCH3), mixture was stirred for 10 min at 20 °C. To this solution, rac-
69.9 (β-CH2 THF), 72.9 (NCH2CH2OCH3), 122.1 (aryl-Cq), 124.2 β-butyrolactone (0.101 g, 1.17 mmol, 100 equiv. vs. Y) was
(aryl-CH), 124.3 (aryl-Cq), 126.4 (aryl-CH), 127.4 (aryl-CH), added rapidly. The mixture was immediately stirred with a
128.9 (aryl-CH), 129.9 (aryl-CH), 136.5 (aryl-Cq), 152.9 (aryl- magnetic stirrer bar at 20 °C for 2 min. The reaction was pro-
Cq), 161.4 (aryl-Cq). 29Si{1H} NMR (79 MHz, C6D6, 313 K): cessed and worked-up similarly to that described above for
δ −23.3 (s). 29Si NMR (79 MHz, C6D6, 313 K): δ −23.3 (d of mul- lactide polymerization. Monomer (BL) conversions were calcu-
1
tiplets, 1JSi–H = 172 Hz).
lated from H NMR spectra of the crude reaction mixtures in
NMR-scale generation of Y{ONOOCumCl,Me}(N(SiHMe2)2) CDCl3, from the integration (Int.) ratio Int.polymer/[Int.polymer
+
(Y-3). This product was prepared in situ as described above for Int.monomer], using the methylene hydrogen resonances for
Y-1, starting from {ONOOCumCl,Me}H2 (0.100 g, 0.161 mmol) PHB at δ 1.49 ppm and for rac-BL at δ 1.16 ppm. The micro-
and Y[N(SiHMe2)2]3(THF) (0.090 g, 0.161 mmol). 1H NMR structure of PHBs was determined by 13C NMR spectroscopy
3
(400 MHz, C6D6, 333 K): δ 0.37 (d, JH–H = 2.8 Hz, 12H, SiH (methylene region) at 20 °C in CDCl3 on a Bruker AM-500
(CH3)2), 1.36 (m, 4H, α-CH2 THF), 1.60 (s, 6H, C(CH3)2), 2.03 spectrometer.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
Single crystal X-ray diffraction studies
J. R. Hagadorn, E. W. Hemmesch, V. G. Young Jr., M. Pink,
M. A. Hillmyer and W. B. Tolman, Macromolecules, 1999,
32, 2400; (f) B. M. Chamberlain, B. A. Jazdzewski, M. Pink,
M. A. Hillmyer and W. B. Tolman, Macromolecules, 2000,
33, 3970; (g) T. M. Ovitt and G. W. Coates, J. Am. Chem.
Soc., 2002, 124, 1316.
Diffraction data for pro-ligands L1, L2, L3 and complexes Sc-1,
Sc-3, Sc-4, and Y-2 were collected at 100(2) K using a Bruker
APEX CCD diffractometer with graphite-monochromatized
MoKα radiation (λ = 0.71073 Å). A combination of ω and ϕ
scans was carried out to obtain at least a unique data set. The
crystal structures were solved by direct methods, the remaining
atoms were located from difference Fourier synthesis followed
by full-matrix least-squares refinement based on F2 (programs
SIR97 and SHELXL-97)19 with the aid of the WINGX
program.20 In most cases, many hydrogen atoms could be
found from the Fourier difference analysis. Other hydrogen
atoms were placed at calculated 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 displacement parameters. The
locations of the largest peaks in the final difference Fourier
map calculation as well as the magnitude of the residual elec-
tron densities were of no chemical significance. Crystal data
and details of data collection and structure refinement for pro-
ligands L1, L2, and L3 and complexes Sc-1, Sc-3, Sc-4, and Y-2
are summarized in Tables S2–S4.†
3 (a) C.-X. Cai, A. Amgoune, C. W. Lehmann and
J.-F.
Carpentier,
Chem.
Commun.,
2004,
330;
(b) A. Amgoune, C. M. Thomas, T. Roisnel and
J.-F. Carpentier, Chem. – Eur. J., 2006, 12, 169; (c) X. Liu,
X. Shang, T. Tang, N. Hu, F. Pei, D. Cui, X. Chen and
X. Jing, Organometallics, 2007, 26, 2747; (d) H. E. Dyer,
S. Huijser, N. Susperregui, F. Bonnet, A. D. Schwarz,
R. Duchateau, L. Maron and P. Mountford, Organometallics,
2010, 29, 3602; (e) L. Clark, M. G. Cushion, H. E. Dyer,
A. D. Schwarz, R. Duchateau and P. Mountford, Chem.
Commun., 2010, 46, 273; (f) W. Zhao, D. Cui, X. Liu and
X. Chen, Macromolecules, 2010, 43, 6678; (g) K. Nie, X. Gu,
Y. Yao, Y. Zhang and Q. Shen, Dalton Trans., 2010, 39, 6832;
(h) K. Nie, L. Fang, Y. Yao, Y. Zhang, Q. Shen and Y. Wang,
Inorg. Chem., 2012, 51, 11133.
4 (a) A. Amgoune, C. M. Thomas, S. Ilinca, T. Roisnel and
J.-F. Carpentier, Angew. Chem., Int. Ed., 2006, 45, 2782;
(b) N. Ajellal, M. Bouyahyi, A. Amgoune, C. M. Thomas,
A. Bondon, I. Pillin, Y. Grohens and J.-F. Carpentier, Macro-
molecules, 2009, 42, 987; (c) M. Bouyahyi, N. Ajellal,
E. Kirillov, C. M. Thomas and J. F. Carpentier, Chem. – Eur.
J., 2011, 17, 1872; (d) C. G. Jaffredo, Y. Chapurina,
S. M. Guillaume and J.-F. Carpentier, Angew. Chem., Int.
Ed., 2014, 53, 2687.
Acknowledgements
This work was supported in part by the Ministère de l’En-
seignement Supérieur et de la Recherche, CNRS, CAPES-COFE-
CUB joined action Ph556-07_2007-2010 and CAPES-CNRS
joined action PICS05923. We thank S. Sinbandhit (University
of Rennes 1) for NMR experiments.
5 For the valuable use of cumyl substituents in ROP, see also:
(a) H.-L. Chen, S. Dutta, P.-Y. Huang and C.-C. Lin, Organo-
metallics, 2012, 31, 2016; (b) L. Wang and H. Ma, Macromol-
ecules, 2010, 43, 6535; see also ref. 9b. Note, however, that
this trend may not necessarily be true with some specific
substrates, such as β-malolactonates; see ref. 4d.
Notes and references
1 For leading reviews, see: (a) B. J. O’Keefe, M. A. Hillmyer
and W. B. Tolman, J. Chem. Soc., Dalton Trans., 2001, 2215;
(b) O. Dechy-Cabaret, B. Martin-Vaca and D. Bourissou,
Chem. Rev., 2004, 104, 6147; (c) C. K. Williams and
M. A. Hillmyer, Polym. Rev., 2008, 48, 1; (d) C. M. Thomas,
Chem. Soc. Rev., 2010, 39, 165; (e) M. J. Stanford and
A. P. Dove, Chem. Soc. Rev., 2010, 39, 486;
(f) J.-F. Carpentier, Macromol. Rapid Commun., 2010, 31,
1696; (g) N. Ajellal, J.-F. Carpentier, C. Guillaume,
S. M. Guillaume, M. Helou, V. Poirier, Y. Sarazin and
A. Trifonov, Dalton Trans., 2010, 39, 8363; (h) P. J. Dijkstra,
H. Du and J. Feijen, Polym. Chem., 2011, 2, 520;
(i) J.-C. Buffet and J. Okuda, Polym. Chem., 2011, 2, 2758.
2 For pioneering work in this field, see: (a) S. J. McClain,
T. M. Ford and N. E. Drysdale, Polym. Prepr. (Am. Chem.
Soc., Div. Polym. Chem.), 1992, 463; (b) A. Le Borgne,
C. Pluta and N. Spassky, Macromol. Rapid Commmun., 1994,
15, 955; (c) W. M. Stevels, M. J. K. Ankone, P. J. Dijkstra and
J. Feijen, Macromol. Chem. Phys., 1995, 196, 1153;
(d) V. Simic, N. Spassky and L. G. Hubert-Pfalzgraf, Macro-
molecules, 1997, 30, 7338; (e) B. M. Chamberlain, Y. Sun,
6 For selected references evidencing superiority of small rare
earth metal centers in achieving highly selective ROP reac-
tions, see: (a) H. Ma, T. P. Spaniol and J. Okuda, Inorg.
Chem., 2008, 47, 3328; (b) H. E. Dyer, S. Huijser,
A. D. Schwarz, C. Wang, R. Duchateau and P. Mountford,
Dalton Trans., 2008, 32; (c) L. M. Hodgson, R. H. Platel,
A. J. P. White and C. K. Williams, Macromolecules, 2008, 41,
8603; (d) N. Ajellal, D. M. Lyubov, M. A. Sinenkov,
G. K. Fukin, A. V. Cherkasov, C. M. Thomas,
J.-F. Carpentier and A. A. Trifonov, Chem. – Eur. J., 2008, 14,
5440; (e) R. H. Platel, A. J. P. White and C. K. Williams,
Chem. Commun., 2009, 4115; (f) E. Grunova, E. Kirillov,
T. Roisnel and J.-F. Carpentier, Dalton Trans., 2010, 39,
6739; (g) J. W. Kramer, D. S. Treiter, E. W. Dunn,
P. M. Castro, T. Roisnel, C. M. Thomas and G. W. Coates,
J. Am. Chem. Soc., 2009, 131, 16042; (h) R. H. Platel,
A. J. P. White and C. K. Williams, Inorg. Chem., 2011, 50,
7718; (i) Z. C. Zhang and D. M. Cui, Chem. – Eur. J., 2011,
17, 11520; ( j) G. Li, M. Lamberti, M. Mazzeo,
D. Pappalardo, G. Roviello and C. Pellecchia, Organometal-
This journal is © The Royal Society of Chemistry 2014
Dalton Trans.

View Article Online
Paper
Dalton Transactions
lics, 2012, 31, 1180; (k) C. Bakewell, T.-P.-A. Cao, N. J. Long,
group. The Y-isopropoxides are much better initiators.
There is no evidence for aggregation of the yttrium amide
precursors, nor of the active, propagating yttrium species;
this has been demonstrated by recent studies on model
O-lactate and β-alkoxy-butyrate yttrium complexes sup-
ported by bis(phenolate)-type ligands, which all proved to
be mononuclear in solution and in the solid state; see:
(a) J. S. Klitzke, T. Roisnel, E. Kirillov, O. de L. Casagrande
Jr. and J.-F. Carpentier, Organometallics, 2014, 33, 309;
(b) J. S. Klitzke, T. Roisnel, E. Kirillov, O. de L. Casagrande
Jr. and J.-F. Carpentier, Organometallics, 2014, DOI:
10.1021/om401214q, in press.
X. F. Le Goff, A. Auffrant and C. K. Williams, J. Am. Chem.
Soc., 2012, 134, 20577; (l) K. Nie, W. Gu, Y. Yao, Y. Zhang
and Q. Shen, Organometallics, 2013, 32, 2608;
(m) C. Bakewell, T.-P.-A. Cao, X. F. Le Goff, N. J. Long,
A. Auffrant and C. K. Williams, Organometallics, 2013, 32,
1475; (n) D. Pappalardo, M. Bruno, M. Lamberti and
C. Pellecchia, Macromol. Chem. Phys., 2013, 214, 1965;
(o) M. Mazzeo, R. Tramontano, M. Lamberti, A. Pilone,
S. Milione and C. Pellecchia, Dalton Trans., 2013, 42, 9338.
7 Ionic radius for 6-coordinate metals: Sc3+ = 0.745 Å, Y3+
=
0.90 Å, Lu3+ = 0.841 Å; R. D. Shannon, Acta Crystallogr.,
Sect. A: Cryst. Phys., Diffr., Theor. Gen. Cryst., 1976, 32, 751.
8 (a) M. E. G. Skinner, T. Toupence, B. R. Tyrrell, B. D. Ward,
C. R. Wilson, A. R. Cowley and P. Mountford, J. Organomet.
Chem., 2002, 647, 145; (b) C. L. Boyd, B. R. Tyrrell,
B. D. Ward and P. Mountford, Organometallics, 2005, 24, 309.
9 (a) H. Ma, T. P. Spaniol and J. Okuda, Angew. Chem., Int.
Ed., 2006, 45, 7818; (b) J.-C. Buffet, A. Kapelski and
J. Okuda, Macromolecules, 2010, 43, 10201; (c) J.-C. Buffet
and J. Okuda, Dalton Trans., 2011, 40, 7748; (d) A. Kapelski,
15 ROP of rac-BL was performed at 2.4 M, since a high
monomer concentration was previously noted to be a key
parameter for achieving high activities in ROP of such
β-lactones; see ref. 4a–c. A direct comparison with ROP of
rac-LA was not possible, since this less soluble monomer
cannot be loaded at such high concentrations. At 1.0 M
monomer concentration, ROP of rac-LA with yttrium cata-
lysts of the type Y-1–Y-4 proceeded significantly faster than
ROP of rac-BL.
J.-C. Buffet, T. P. Spaniol and J. Okuda, Chem. – Asian J., 16 (a) K. C. Hultzsch, P. Voth, K. Beckerle, T. P. Spaniol and
2012, 7, 1320.
J. Okuda, Organometallics, 2000, 19, 228; (b) R. Anwander,
O. Runte, G. Eppinder, G. Gerstberger, E. Herdtweek and
M. Spiegler, Dalton Trans., 1998, 847.
10 A. W. Addison, T. N. Rao, J. Reedijk, J. Van Rijn and
G. C. Verschoor, J. Chem. Soc., Dalton Trans., 1984, 1349–
1356.
17 A. Kowalski, A. Duda and S. Penczek, Macromolecules, 1998,
31, 2114.
11 For examples of 5-coordinate Sc-bis(phenolate) complexes
(yet structurally uncharacterized in the solid state), 18 A. I. Kochnev, I. I. Oleynik, I. V. Oleynik, S. S. Ivanchev and
see ref. 6n. G. A. Tolstikov, Russ. Chem. Bull., Int. Ed., 2007, 56, 1125.
12 J. Eppinger, M. Spiegler, W. Hieringer, W. A. Herrmann 19 (a) A. Altomare, M. C. Burla, M. Camalli, G. Cascarano,
and R. Anwander, J. Am. Chem. Soc., 2000, 122, 3080.
13 (a) H. Ma, T. P. Spaniol and J. Okuda, Dalton Trans., 2003,
4770; (b) M. Mazzeo, M. Lamberti, I. D’Auria, S. Milione,
J. C. Petrers and C. Pellecchia, J. Polym. Sci., Part A: Polym.
Chem., 2010, 48, 1374.
14 As earlier proposed for this class of catalysts and many
other yttrium amide complexes (ref. 3), the relatively high
PDI values observed are likely due to the moderate
C. Giacovazzo, A. Guagliardi, A. G. G. Moliterni, G. Polidori
and R. Spagna, J. Appl. Crystallogr., 1999, 32, 115;
(b) G. M. Sheldrick, SHELXS-97, Program for the Determi-
nation of Crystal Structures, University of Goettingen,
Germany, 1997; (c) G. M. Sheldrick, SHELXL-97, Program for
the Refinement of Crystal Structures, University of Goettin-
gen, Germany, 1997; (d) G. M. Sheldrick, Acta Crystallogr.,
2008, 64, 112.
initiation rate (as compared to propagation) of the amide 20 L. J. Farrugia, J. Appl. Crystallogr., 1999, 32, 837.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2014