Synthesis and characterization of Salalen lanthanide complexes and their application in the polymerization of rac -lactide

Article abstract of DOI:10.1021/om4001023

A series of neutral lanthanide complexes supported by ONNO Salalen-type ligands were synthesized, and their catalytic activity for the polymerization of rac-lactide (rac-LA) was explored. The amine elimination reactions of Ln[N(SiMe₃)₂]₃(μ-Cl)Li(THF)₃ with the ONNO Salalen-type ligand L¹H₂ (L¹ = (2-O-C₆H₂-Bu^t₂-3,5)CH-NCH ₂CH₂N(Me)CH₂(2-O-C₆H ₂-Bu^t₂-3,5)) in a 1:1 molar ratio in tetrahydrofuran (THF) gave the neutral lanthanide amides L ¹Ln[N(SiMe₃)₂](THF) (Ln = Y (1), Sm (2), Nd (3)). Reaction of the lanthanide amides with benzyl alcohol produces the dimeric lanthanide alkoxo complex (L¹LnOCH₂Ph)₂ (Ln = Y (4), Sm (5)) in high isolated yield. Y[N(SiMe₃)₂] ₃(μ-Cl)Li(THF)₃ reacted with the Salalen-type ligand L²H₂ (L² = (2-O-C₆H ₂-Bu^t₂-3,5)CH-NCH₂CH ₂N(Me)CH₂{2-O-C₆H₂-(CPhMe ₂)₂-3,5}) in a 1:l molar ratio in THF also gave the desired yttrium amide, but this complex could not be separated because of its very good solubility even in n-pentane. The proton exchange reactions of L ¹H₂ and L²H₂ with (C ₅H₅)₃Ln(THF) in a 1:1 molar ratio in THF and then with 1 equiv of benzyl alcohol gave the desired lanthanide alkoxides [L¹Ln(OCH₂Ph)]₂ (Ln = Y (4), Sm (5), Yb (6)) and [L²Y(OCH₂Ph)]₂ (7), respectively. Complexes 1-7 were well characterized by elemental analyses, IR spectra, X-ray single-crystal structure determination, and NMR spectroscopy in the case of complexes 1, 4, and 7. Complexes 1-3 are isostructural and have a solvated monomeric structure. The coordination geometry around the lanthanide atom can be best described as a distorted trigonal bipyramid. Complexes 4-7 are dimeric species in the solid state. They all contain a Ln₂O₂ core bridging through the oxygen atoms of the two OCH₂Ph groups. Each of the lanthanide atoms is also six-coordinated to form a distorted octahedron. It was found that all the complexes are efficient initiators for the ring-opening polymerization of rac-LA, giving PLA with good heterotacticity (P_r up to 0.85). The observed order of increase in activity is in agreement with the order of the ionic radii, whereas the stereoselectivity is in reverse order. The steric bulkiness of the substituents on the phenol ring has no obvious impact on the rate and stereocontrolability of the polymerizations. The Ln-O species resulted in more controllable polymerization than the corresponding Ln-N species, and complex 4 can initiate rac-LA polymerization in a controlled manner.

Full text of DOI:10.1021/om4001023

Article
pubs.acs.org/Organometallics
Synthesis and Characterization of Salalen Lanthanide Complexes and
Their Application in the Polymerization of rac-Lactide
Kun Nie,† Weikai Gu,† Yingming Yao,*,†,‡ Yong Zhang,† and Qi Shen†
†Key Laboratory of Organic Synthesis of Jiangsu Province, College of Chemistry, Chemical Engineering and Materials Science, Dushu
Lake Campus, Soochow University, Suzhou 215123, People’s Republic of China
‡The Institute of Low Carbon Economy, Suzhou 215123, People’s Republic of China
S
* Supporting Information
ABSTRACT: A series of neutral lanthanide complexes
supported by ONNO Salalen-type ligands were synthesized,
and their catalytic activity for the polymerization of rac-lactide
(rac-LA) was explored. The amine elimination reactions of
Ln[N(SiMe3)2]3(μ-Cl)Li(THF)3 with the ONNO Salalen-
type ligand L1H2 (L1 = (2-O-C6H2-But -3,5)CH
2
NCH2CH2N(Me)CH2(2-O-C6H2-But2-3,5)) in a 1:1 molar
ratio in tetrahydrofuran (THF) gave the neutral lanthanide
amides L1Ln[N(SiMe3)2](THF) (Ln = Y (1), Sm (2), Nd
(3)). Reaction of the lanthanide amides with benzyl alcohol
produces the dimeric lanthanide alkoxo complex
(L1LnOCH2Ph)2 (Ln = Y (4), Sm (5)) in high isolated
yield. Y[N(SiMe3)2]3(μ-Cl)Li(THF)3 reacted with the Salalen-
type ligand L2H2 (L2 = (2-O-C6H2-But -3,5)CH
2
NCH2CH2N(Me)CH2{2-O-C6H2-(CPhMe2)2-3,5}) in a 1:l
molar ratio in THF also gave the desired yttrium amide, but
this complex could not be separated because of its very good
solubility even in n-pentane. The proton exchange reactions of
L1H2 and L2H2 with (C5H5)3Ln(THF) in a 1:1 molar ratio in THF and then with 1 equiv of benzyl alcohol gave the desired
lanthanide alkoxides [L1Ln(OCH2Ph)]2 (Ln = Y (4), Sm (5), Yb (6)) and [L2Y(OCH2Ph)]2 (7), respectively. Complexes 1−7
were well characterized by elemental analyses, IR spectra, X-ray single-crystal structure determination, and NMR spectroscopy in
the case of complexes 1, 4, and 7. Complexes 1−3 are isostructural and have a solvated monomeric structure. The coordination
geometry around the lanthanide atom can be best described as a distorted trigonal bipyramid. Complexes 4−7 are dimeric
species in the solid state. They all contain a Ln2O2 core bridging through the oxygen atoms of the two OCH2Ph groups. Each of
the lanthanide atoms is also six-coordinated to form a distorted octahedron. It was found that all the complexes are efficient
initiators for the ring-opening polymerization of rac-LA, giving PLA with good heterotacticity (Pr up to 0.85). The observed
order of increase in activity is in agreement with the order of the ionic radii, whereas the stereoselectivity is in reverse order. The
steric bulkiness of the substituents on the phenol ring has no obvious impact on the rate and stereocontrolability of the
polymerizations. The Ln−O species resulted in more controllable polymerization than the corresponding Ln−N species, and
complex 4 can initiate rac-LA polymerization in a controlled manner.
Throughout the past decade, significant effort has been
invested in the development of discrete, well-characterized
metal complexes to initiate the stereocontrolled ROP of
lactides.2 In most cases, the catalytic efficiency and the degree
of chain stereocontrol of the resulting polymers is strongly
influenced by the nature of the ancillary ligands and the metal
center. Subtle modifications in the ligand framework may have
a major effect on the ROP of rac-lactide (rac-LA). It has been
found that the lanthanide derivatives stabilized by bridged
bis(phenolate) ligands are efficient catalysts for the ROP of
INTRODUCTION
■
Poly(lactides) (PLA) are among the most important synthetic
biodegradable polymers investigated for biomedical and
pharmaceutical applications, such as controlled drug delivery,
resorbable sutures, medical implants, and scaffolds for tissue
engineering.1 The most efficient synthesis, and the commercial
route to PLA, involves the ring-opening polymerization (ROP)
of lactide.1d,g Because the chain stereochemistry of the
monomeric units in the polymer chains plays a decisive role
in the mechanical, physical, and degradation properties of PLA
materials, the design and synthesis of catalysts to prepare
different stereospecific PLA architectures is a major topic.
Received: February 6, 2013
© XXXX American Chemical Society
A
dx.doi.org/10.1021/om4001023 | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
NMR spectra for the other complexes were obtained. Molecular
weights and molecular weight distributions (PDI) were determined
against a polystyrene standard by gel permeation chromatography
(GPC) on a PL 50 apparatus, and THF was used as an eluent at a flow
rate of 1.0 mL/min at 40 °C. The microstructures of PLAs were
measured by homodecoupling 1H NMR spectroscopy at 20 °C in
CDCl3 on a Unity Varian AC-400 spectrometer. MALDI-TOF mass
spectra were recorded using a Bruker Reflex II mass spectrometer
(Bremen, Germany).
lactides; especially, the lanthanide complexes supported by
amine-bridged bis(phenolate) groups can initiate the highly
stereoselective ROP of rac-LA to give PLAs with heterotacticity
up to 0.99.3 To further understand the effect of the backbone of
the bis(phenolate) ligands on the stereoselectivity of rac-LA
polymerization, the design and synthesis of lanthanide initiators
with different bis(phenolate) ancillary ligands are still mean-
ingful.
Synthesis of Ligand Precursor L2H2. A solution of 3,5-di-tert-
butyl-2-hydroxybenzaldehyde (4.68 g, 20 mmol) was added to N-
methylethylenediamine (1.8 mL, 20 mmol) in methanol (30 mL). The
mixture was stirred to give a clear yellow solution at room
temperature, and then a solution of 2,4-bis(α,α-dimethylbenzyl)-
phenol (8.25 g, 25 mmol) and paraformaldehyde (0.90 g, 30 mmol) in
methanol was added. The mixture was refluxed for 24 h and then was
cooled. The precipitate was filtered and washed with ice-cold methanol
to give the yellow product (9.01 g, 71%). Anal. Calcd for C43H56N2O2:
C, 81.60; H, 8.92; N, 4.43. Found: C, 81.78; H, 8.74; N, 4.66. MS: m/z
calcd for C43H57N2O2 633.4420, found 633.4407. 1H NMR (400 MHz,
DCCl3, 25 °C): δ 8.12 (s, 1H, NCHAr), 7.37 (s, 1H, ArH), 7.20−
7.06 (m, 12H, ArH), 6.72 (s, 1H, ArH), 3.61 (s, 2H, ArCH2N), 3.45
(s, 2H, NCH2), 2.63 (s, 2H, NCH2), 2.25 (s, 3H, NCH3), 1.68 (s, 6H,
C(CH3)2Ph), 1.64 (s, 6H, C(CH3)2Ph), 1.43 (s, 9H, C(CH3)3), 1.32
(s, 9H, C(CH3)3). 13C{1H} NMR (100 MHz, DCCl3, 25 °C): δ
167.31 (NCHAr), 158.09, 153.90, 151.92, 151.60, 140.10, 136.76,
136.18, 135.29, 128.05, 127.88, 126.99, 126.68, 125.75, 125.03, 124.47,
121.42, 117.89, 109.85 (Ar−C), 62.20 (ArCH2N), 57.62 (NCH2),
56.41 (NCH2), 42.62 (NCH3), 42.15 (C(CH3)2Ph), 35.15 (C-
(CH3)3), 34.27 (C(CH3)3), 31.66 (C(CH3)3), 31.24 (C(CH3)2Ph),
29.57 (C(CH3)3). IR (KBr pellet, cm−1): 2956 s, 1617 m, 1465 s, 1374
m, 1245 m, 1214 m, 1023 w, 872 w, 767 m, 693 s.
Salen and Salan are two types of popular ligands in
transition-metal chemistry, and the corresponding metal
complexes are now widely used as catalysts in a variety of
metal-catalyzed organic transformations and polymerizations.3,4
Salalen ligands, which are half-Salan/half-Salen hybrid ligands,
have received little attention. Just like Salen and Salan, Salalen
ligands can be easily synthesized via simple condensation of
primary amine with 1 equiv of a substituted salicylaldehyde to
form the “half-Salen” part, followed by a Mannich condensation
of the secondary amine with 1 equiv of formaldehyde and 1
equiv of a substituted phenol to complete the “half-Salan” part.
Accordingly, it is relatively easy to evaluate the steric and
electronic effects of Salalen ligands, which is advantageous for
fine tuning of the structures of the corresponding Salalen metal
complexes and for systematically exploiting their catalytic
behaviors. Recently, Salalen ligands have been used to stabilize
metal-based catalysts, which can catalyze a number of
transformations, such as asymmetric oxidations,5 hydrophos-
phonylation of aldehydes and aldimines,6 hydrosilylation of
organic carbonyl compounds,7 copolymerization of epoxide and
CO2,8 and polymerization of olefins9 and lactide.10 However, all
of these catalysts focused on Al(III), Zn(II), Cr(III), and group
4 elements. To our knowledge, there has been no report
concerning the application of Salalen ligands to rare-earth
organometallic chemistry.
In this paper, the Salalen ligand was introduced to
organolanthanide chemistry for the first time, and a series of
lanthanide bis(trimethylsilyl)amido and alkoxo complexes
supported by unsymmetrical Salalen ligands (amine/imine
bis(phenolate)) were prepared and well-characterized, using
Ln[N(SiMe3)2]3(μ-Cl)Li(THF)3 and (C5H5)3Ln(THF) as the
precursors. It was found that these lanthanide Salalen
complexes are efficient initiators for the ROP of rac-LA to
give PLAs with good heterotacticity, and the lanthanide
alkoxides can initiate rac-LA polymerization in a controlled
manner. Herein we report these results.
Synthesis of L1Y[N(SiMe3)2](THF) (1). To a THF solution of
Y[N(SiMe3)2](μ-Cl)Li(THF)3 (10 mL, 2.89 mmol) was added a THF
solution of L1H2 (10 mL, 1.47 g, 2.89 mmol). The mixture was stirred
for 2 h at room temperature, and then THF was evaporated
completely under reduced pressure. Hexane (10 mL) was added to
extract the residue twice, and the precipitate formed was removed by
centrifugation. The solvent was concentrated to about 8 mL, and
colorless crystals were obtained at room temperature in several days
(1.91 g, 80%). Anal. Calcd for C43H76N3O3Si2Y: C, 62.36; H, 9.25; N,
1
5.07; Y, 10.74. Found: C, 62.09; H, 9.43; N, 5.25; Y, 10.98. H NMR
(400 MHz, C6D6, 25 °C): δ 7.71 (d, J = 2.5 Hz, 1H, ArH), 7.50 (d, J =
2.5 Hz, 1H, ArH), 7.26 (s, 1H, NCHAr), 6.96 (d, J = 2.4 Hz, 1H,
ArH), 6.89 (d, J = 2.4 Hz, 1H, ArH), 4.37 (d, J = 12.8 Hz, 1H,
ArCH2N), 3.58 (br, 4H, α-CH2 THF), 3.21 (m, 1H, NCH2), 2.75 (d, J
= 12.8 Hz, 1H, ArCH2N), 2.56 (m, 1H, NCH2), 2.24 (s, 3H, NCH3),
2.10 (m, 1H, NCH2), 1.77 (s, 9H, C(CH3)3), 1.51 (s, 9H, C(CH3)3),
1.41 (s, 9H, C(CH3)3), 1.40 (br, 4H, β-CH2 THF), 1.31 (s, 9H,
C(CH3)3), 1.27 (m, 1H, NCH2), 0.38 (s, 18H, SiMe3). 13C{1H} NMR
(100 MHz, C6D6, 25 °C): δ 171.2 (NCHAr), 163.9, 160.2, 139.7,
137.6, 137.3, 130.7, 129.8, 125.7, 124.9, 122.8, 121.3 (Ar−C), 68.09
(α-CH2, THF), 65.1 (ArCH2N), 57.2 (NCH2), 54.0 (NCH2), 43.9
(NCH3), 35.8 (C(CH3)3), 35.4 (C(CH3)3), 34.3 (C(CH3)3), 34.1
(C(CH3)3), 32.3 (C(CH3)3), 31.7 (C(CH3)3), 30.2 (C(CH3)3), 25.8
(β-CH2, THF), 5.01 (SiMe3). IR (KBr pellet, cm−1): 2960 s, 2900 s,
2869 s, 1620 s, 1540 m, 1470 s, 1440 s, 1410 m, 1360 m, 1310 m, 1250
m, 1200 w, 1160 m, 933 m, 881 m, 835 s, 742 w, 652 w.
EXPERIMENTAL SECTION
■
General Procedures. All of the manipulations were performed
under a purified argon atmosphere using standard Schlenk techniques.
The solvents were degassed and distilled from sodium benzophenone
ketyl under argon prior to use. HN(SiMe3)2 and n-BuLi are
commercially available. HN(SiMe3)2 was dried over CaH2 for 4 days
and distilled before use. Ln[N(SiMe3)2]3(μ-Cl)Li(THF)3 (Ln = Y,
Sm, Nd),11a−c (C5H5)3Ln(THF),11d and the ligand L1H2 (L1 = (2-O-
C6H2-But2-3,5)CHNCH2CH2N(Me)CH2(2-O-C6H2-But2-3,5))12
were prepared according to the procedures reported in the literature.
Benzyl alcohol was dried over 4 Å molecular sieves for 1 week and
then distilled before use. rac-LA was recrystallized twice from dry
toluene and then was sublimed under vacuum at 50 °C. Lanthanide
analyses were performed by EDTA titration with a xylenol orange
indicator and a hexamine buffer.13 Carbon, hydrogen, and nitrogen
analyses were performed by direct combustion with a Carlo-Erba EA-
1110 instrument. The IR spectra were recorded with a Nicolet-550
FTIR spectrometer as KBr pellets. The 1H and 13C NMR spectra were
recorded in a C6D6 solution for complexes 1, 4, and 7 with a Unity
Varian spectrometer. Because of their paramagnetism, no resolvable
Synthesis of L1Sm[N(SiMe3)2](THF) (2). The synthesis of
complex 2 was carried out in the same way as that described for
complex 1, but Sm[N(SiMe3)2](μ-Cl)Li(THF)3 (10 mL, 2.82 mmol)
was used instead of Y[N(SiMe3)2](μ-Cl)Li(THF)3. Yellow crystals
were obtained in a hexane (7 mL)/THF (1 mL) solution (1.86 g,
74%). Anal. Calcd for C43H76N3O3Si2Sm: C, 58.05; H, 8.61; N, 4.72;
Sm, 16.90. Found: C, 58.36; H, 8.80; N, 4.07; Sm, 16.79. IR (KBr
pellet, cm−1): 2960 s, 2900 s, 2869 s, 1620 s, 1540 m, 1470 s, 1440 s,
1410 m, 1360 m, 1310 m, 1250 m, 1200 w, 1160 m, 928 m, 877 m,
839 s, 742 w, 617 w.
Synthesis of L1Nd[N(SiMe3)2](THF) (3). The synthesis of
complex 3 was carried out in the same way as that described for
B
dx.doi.org/10.1021/om4001023 | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
complex 1, but Nd[N(SiMe3)2](μ-Cl)Li(THF)3 (12 mL, 2.95 mmol)
was used instead of Y[N(SiMe3)2](μ-Cl)Li(THF)3. Green crystals
were obtained in a hexane (9 mL)/THF (1 mL) solution (2.01 g,
77%). Anal. Calcd for C43H76N3O3Si2Nd: C, 58.46; H, 8.67; N, 4.76;
Nd, 16.33. Found: C, 58.66; H, 8.41; N, 4.93; Nd, 16.45. IR (KBr
pellet, cm−1): 2960 s, 2910 s, 2869 s, 1620 s, 1530 m, 1470 s, 1440 s,
1410 m, 1360 m, 1310 m, 1250 m, 1200 w, 1160 m, 933 m, 876 m,
835 s, 741 w, 611 w.
g, 83%). Anal. Calcd for C100H122N4O6Y2: C, 72.62; H, 7.44; N, 3.39;
Y, 10.75. Found: C, 72.82; H, 7.77; N, 3.21; Y, 11.08. IR (KBr pellet,
cm−1): 2952 s, 2900 s, 2866 s, 1620 s, 1530 m, 1470 s, 1440 s, 1410 s,
1360 m, 1310 m, 1230 m, 1208 m, 1160 m, 1015 m, 879 w, 839 w, 744
w, 696 w, 643 w. 1H NMR (400 MHz, C6D6, 25 °C): δ 7.80 (d, J = 2.3
Hz, 2H, ArH), 7.35 (m, 8H, ArH), 7.30 (d, J = 2.3 Hz, 2H, ArH), 7.22
(s, 2H, NCHAr), 7.14 (s, 2H, ArH), 7.02 (m, 10H, ArH), 6.93 (d, J
= 2.3 Hz, 2H, ArH), 6.85 (m, 10H, ArH), 6.75 (d, J = 2.3 Hz, 2H,
ArH), 4.76 (d, J = 12.4 Hz, 2H, OCH2Ph), 4.54 (d, J = 12.4 Hz, 2H,
OCH2Ph), 3.68 (d, J = 13.2 Hz, 2H, ArCH2N), 2.93 (m, 2H, NCH2),
2.58 (d, J = 13.2 Hz, 2H, ArCH2N), 2.52 (s, 6H, NCH3), 2.09 (m, 2H,
NCH2), 1.87 (s, 18H, C(CH3)3), 1.78 (m, 2H, NCH2), 1.67 (m, 12H,
C(CH3)2Ph), 1.58 (s, 6H, C(CH3)2Ph), 1.43 (s, 18H, C(CH3)3), 1.41
(m, 6H, C(CH3)2Ph), 1.29 (m, 2H, NCH2). 13C{1H} NMR (100
MHz, C6D6, 25 °C): δ 170.2 (NCHAr), 164.0, 160.6, 152.4, 143.7,
138.6, 136.8, 136.2, 135.8, 130.7, 129.8, 128.6, 128.6, 127.5, 127.3,
126.9, 125.6, 124.7, 124.0, 122.5 (Ar−C), 68.7 (OCH2Ph), 64.9
(ArCH2N), 58.2 (NCH2), 54.5 (NCH2), 45.1 (NCH3), 42.9
(C(CH3)2Ph), 42.6(C(CH3)2Ph), 36.1 (C(CH3)3), 34.2 (C(CH3)3),
31.9 (C(CH3)3), 31.6 (C(CH3)2Ph), 31.0 ((C(CH3)2Ph)).
Typical Procedure for the Polymerization Reaction. The
procedures for the polymerization of rac-LA initiated by the complexes
1−7 were similar, and a typical polymerization procedure is given
below. A 20 mL Schlenk flask, equipped with a magnetic stirring bar,
was charged with the desired amount of monomer and solvent. After
the monomer was dissolved, a solution of the initiator was added to
this solution by syringe. The mixture was immediately stirred
vigorously for the desired time, during which time an increase in the
viscosity was observed. The reaction mixture was quenched by the
addition of ethanol and then poured into ethanol to precipitate the
polymer, which was dried under vacuum and weighed.
Oligomer Preparation. The oligomerization of rac-LA was carried
out with complex 4 as the initiator in THF and toluene, respectively, at
25 °C under the condition of a [monomer]/[initiator] molar ratio of
of 20. The reaction mixture was stirred for 0.5 h and then quenched by
adding n-hexane and 1 drop of water. The precipitated oligomers were
collected, dried under vacuum, and used for 1H NMR, 13C NMR, and
MALDI-TOF measurement.
X-ray Crystallographic Structure Determination. Suitable
single crystals of complexes 1−7 were sealed in a thin-walled glass
capillary for determination of the single-crystal structures. Intensity
data were collected with a Rigaku Mercury CCD area detector in ω
scan mode using Mo Kα radiation (λ = 0.71070 Å). The diffracted
intensities were corrected for Lorentz/polarization effects and
empirical absorption corrections. Details of the intensity data
collection and crystal data are given in Table S1 (Supporting
Information).
Synthesis of [L1Y(OCH2Ph)]2 (4). Method A. To a hexane
solution of complex 1 (1.87 g, 2.26 mmol) was added a hexane
solution of benzyl alcohol (0.23 mL, 2.26 mmol), and a white powder
formed immediately. The reaction mixture was stirred for 1 h at room
temperature, and then the precipitate that formed was collected by
centrifugation. Benzene (10 mL) and THF (1 mL) were added to
extract the residue. Colorless crystals were obtained at 5 °C in several
days (1.35 g, 85%). Anal. Calcd for C80H114N4O6Y2: C, 68.36; H, 8.18;
1
N, 3.99; Y, 12.65. Found: C, 68.79; H, 8.77; N, 3.68; Y, 12.84. H
NMR (400 MHz, C6D6, 25 °C): δ 7.82 (d, J = 2.4 Hz, 2H, ArH), 7.41
(m, 6H, ArH), 7.32 (s, 2H, NCHAr), 7.00 (m, 8H, ArH), 6.88 (d, J
= 2.4 Hz, 2H, ArH), 5.30 (d, J = 12.5 Hz, 2H, OCH2Ph), 5.14 (d, J =
12.5 Hz, 2H, OCH2Ph), 3.94 (d, J = 12.8 Hz, 2H, ArCH2N), 2.99 (m,
2H, NCH2), 2.80 (d, J = 12.8 Hz, 2H, ArCH2N), 2.68 (s, 6H, NCH3),
2.54 (m, 2H, NCH2), 2.12 (m, 2H, NCH2), 1.94 (s, 18H, C(CH3)3),
1.44 (s, 18H, C(CH3)3), 1.43 (s, 18H, C(CH3)3), 1.37 (s, 18H,
C(CH3)3), 1.34 (m, 2H, NCH2). 13C{1H} NMR (100 MHz, C6D6, 25
°C): δ 170.2 (NCHAr), 164.1, 160.9, 138.7, 136.8, 130.5, 129.8,
127.1, 125.3, 124.2, 123.5, 122.4 (Ar−C), 69.2 (OCH2Ph), 65.2
(ArCH2N), 58.2 (NCH2), 54.5 (NCH2), 45.3 (NCH3), 36.1
(C(CH3)3), 35.4 (C(CH3)3), 34.2 (C(CH3)3), 32.3 (C(CH3)3),
31.8 (C(CH3)3), 30.7 (C(CH3)3), 30.0 (C(CH3)3). IR (KBr pellet,
cm−1): 2950 s, 2900 s, 2860 s, 1620 s, 1530 m, 1470 s, 1440 s, 1410 s,
1360 m, 1310 m, 1230 m, 1200 m, 1160 m, 1010 m, 879 w, 839 w, 744
w, 696 w, 650 w.
Method B. To a THF solution of (C5H5)3Y(THF) (1.00 g, 2.81
mmol) was added a THF solution of L1H2 (1.44 g, 2.81 mmol). The
reaction mixture was stirred for 1 h at room temperature, and then
benzyl alcohol (0.29 mL, 2.81 mmol) was added by a syringe. The
mixture was stirred for 24 h, and then THF was evaporated completely
under vacuum. Toluene (10 mL) and THF (1 mL) were added to
extract the residue. Colorless crystals were obtained at 5 °C in several
days (1.58 g, 80%). 1H NMR (400 MHz, C6D6, 25 °C): δ 7.81 (s, 2H,
ArH), 7.45 (m, 6H, ArH), 7.32 (s, 2H, NCHAr), 6.94 (m, 8H,
ArH), 6.88 (s, 2H, ArH), 5.30 (d, J = 12.6 Hz, 2H, OCH2Ph), 5.13 (d,
J = 12.3 Hz, 2H, OCH2Ph), 3.94 (d, J = 12.9 Hz, 2H, ArCH2N), 2.99
(m, 2H, NCH2), 2.80 (d, J = 12.8 Hz, 2H, ArCH2N), 2.68 (s, 6H,
NCH3), 2.59 (m, 2H, NCH2), 2.07 (m, 2H, NCH2), 1.94 (s, 18H,
C(CH3)3), 1.45(m, 36H, C(CH3)3), 1.37 (s, 18H, C(CH3)3), 1.31 (m,
2H, NCH2).
The structures were solved by direct methods and refined by full-
matrix least-squares procedures based on |F|2. The hydrogen atoms in
these complexes were generated geometrically, assigned appropriate
isotropic thermal parameters, and allowed to ride on their parent
carbon atoms. All of the hydrogen atoms were held stationary and
included in the structure factor calculations in the final stage of full-
matrix least-squares refinement. The structures were solved and
refined using SHELXL-97 programs.
Synthesis of [L1Sm(OCH2Ph)]2 (5). The synthesis of complex 5
was carried out in the same way as that described for complex 4
(method A), but complex 2 (2.14 g, 2.40 mmol) was used instead of
complex 1. Yellow crystals were obtained in a hexane (10 mL)/THF
(3 mL) solution (1.40 g, 76%). Anal. Calcd for C80H114N4O6Sm2: C,
62.86; H, 7.52; N, 3.67; Sm, 19.67. Found: C, 62.55; H, 7.69; N, 3.81;
Sm, 19.54. IR (KBr pellet, cm−1): 2950 s, 2901 s, 2860 s, 1620 s, 1530
m, 1470 s, 1440 s, 1410 s, 1360 m, 1310 m, 1230 m, 1200 m, 1160 m,
1010 m, 879 w, 841 w, 744 w, 695 w, 650 w.
RESULTS AND DISCUSSION
Synthesis of [L1Yb(OCH2Ph)]2 (6). The synthesis of complex 6
was carried out in the same way as that described for complex 4
(method B), but (C5H5)3Yb(THF) (1.23 g, 2.80 mmol) was used
instead of (C5H5)3Y(THF). Yellow crystals were obtained in a hexane
(12 mL)/THF (2 mL) solution (1.70 g, 77%). Anal. Calcd for
C80H114N4O6Yb2: C, 61.05; H, 7.30; N, 3.56; Yb, 21.99. Found: C,
61.32; H, 7.68; N, 3.24; Yb, 22.25. IR (KBr pellet, cm−1): 2950 s, 2900
s, 2860 s, 1620 s, 1530 m, 1470 s, 1440 s, 1410 s, 1360 m, 1310 m,
1230 m, 1200 m, 1160 m, 1010 m, 879 w, 841 w, 744 w, 695 w, 650 w.
Synthesis of [L2Y(OCH2Ph)]2 (7). The synthesis of complex 7 was
carried out in the same way as that described for complex 4 (method
B), but L2H2 (1.74 g, 2.75 mmol) was used instead of L1H2. Colorless
crystals were obtained in a hexane (9 mL)/THF (5 mL) solution (1.89
■
Synthesis and Characterization of Salalen Lanthanide
Complexes. An amine elimination reaction is a straightforward
method for the synthesis of lanthanide amides, and the
standard precursors are Ln[N(SiMe3)2]3 and Ln[N-
(SiMe3)2]3(μ-Cl)Li(THF)3. However, for the Salen ligands,
the popular precursor used is the sterically less demanding but
valuable bis(dimethylsilyl)amido lanthanide complex Ln[N-
(SiHMe2)2]3(THF)2. The cheaper Ln[N(SiMe3)2]3 reacted
with the ligand precursor SalenH2 to form generally an
oligomeric, THF-insoluble product.14 When the Salalen ligand
precursors were prepared, we wanted to know whether
C
dx.doi.org/10.1021/om4001023 | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Scheme 1
Scheme 2
Ln[N(SiMe3)2]3(μ-Cl)Li(THF)3 could be used as the
precursor to synthesize the Salalen lanthanide amido
complexes. Thus, an NMR-scale reaction was conducted first.
When Y[N(SiMe3)2]3(μ-Cl)Li(THF)3 was added to a C6D6
solution of 1 equiv of L1H2, the yellow solution turned slightly
An NMR monitoring reaction revealed that Y[N(SiMe3)2]3(μ-
Cl)Li(THF)3 reacted with the Salalen ligand precursor L2H2
(L2 = (2-O-C6H2-But2-3,5)CHNCH2CH2NMeCH2{2-O-
C6H2-(CPhMe2)2-3,5}), possessing larger substituents on the
phenol ring, to give also the yttrium amide L2Y[N(SiMe3)2]-
(THF). However, this compound could not be isolated because
of its very good solubility even in n-pentane.
A further NMR-tube reaction revealed that these Salalen
lanthanide complexes can be transferred to the corresponding
Salalen lanthanide alkoxides. When equimolar PhCH2OH was
added to a C6D6 solution of complex 1, the peak at 0.38 ppm
disappeared and two doublet peaks at 5.30 and 5.14 ppm
appeared, which can be attributed to the resonances of the
PhCH2O group of the target yttrium alkoxide. Thus, reactions
of the Salalen lanthanide amides with benzyl alcohol produced
the dimeric Salalen lanthanide alkoxo complexes [L1Ln-
(OCH2Ph)]2 (Ln = Y (4), Sm (5)) in high isolated yields, as
shown in Scheme 1.
1
muddy, indicating that LiCl might be formed. The H NMR
spectrum showed that the OH signals of the ligand precursor at
13.92 and 10.92 ppm disappeared, and single sharp resonances
at 0.10 ppm for HN(SiMe3)2 and at 0.38 ppm for LY-
N(SiMe3)2 arose, which demonstrated that the Salalen yttrium
amide was formed. On a preparative scale, Y[N(SiMe3)2]3(μ-
Cl)Li(THF)3 reacted with L1H2 in THF in a 1:1 molar ratio for
2 h at room temperature; after workup, the desired neutral
yttrium amido complex L1Y[N(SiMe3)2](THF) (1) was
isolated from a concentrated hexane solution as colorless
microcrystals in high yield (80%). This method can also be
used to synthesize the Salalen lanthanide amides with larger
ionic radii, and L1Ln[N(SiMe3)2](THF) (Ln = Sm (2), Nd
(3)) were also obtained, as shown in Scheme 1. These results
proved that the cheap Ln[N(SiMe3)2]3(μ-Cl)Li(THF)3 is a
useful starting material for preparing Salalen lanthanide amides.
We recently found that the easily available Cp3Ln(THF) is a
useful precursor for preparing bridged bis(phenolate) lantha-
nide alkoxo complexes.3c,d Thus, we tried to synthesize the
D
dx.doi.org/10.1021/om4001023 | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
lanthanide alkoxo complex stabilized by L2 using Cp3Y(THF)
as the precursor. A 1H NMR monitoring reaction revealed that
the reaction did not occur immediately when Cp3Y(THF) was
added to a C6D6 solution of 1 equiv of L2H2, because of the
insolubility of Cp3Y(THF) in C6D6. However, when the
mixture was heated slightly, the precipitate of Cp3Y(THF)
disappeared, and the solution changed from yellow to brilliant
yellow, indicating that the reaction took place. In the 1H NMR
spectrum, the signals of OH of the ligand precursor disappeared
and a single sharp resonance at 6.33 ppm for the cyclo-
pentadienyl group appeared. In addition, the splitting peaks for
the eliminated cyclopentadiene were also observed at 6.49,
6.30, and 2.69 ppm. These results indicated the formation of a
Salalen yttrium cyclopentadienyl complex. After equimolar
PhCH2OH was added, the brilliant yellow solution turned
slightly muddy immediately, the single sharp resonance at 6.33
ppm almost disappeared, and two doublet peaks at 4.76 and
4.54 ppm of the PhCH2O group appeared, which indicated the
transformation of the Salalen yttrium cyclopentadienyl complex
to a Salalen yttrium alkoxo complex. On a preparative scale, the
expected Salalen yttrium alkoxo complex [L2Y(OCH2Ph)]2 (7)
was obtained in high isolated yield. A further study revealed
that complexes 4−6 can also be synthesized via the proton
exchange reactions of (C5H5)3Ln(THF) with L1H2 in a 1:l
molar ratio in THF for 1 h and then with 1 equiv of benzyl
alcohol, as shown in Scheme 2.
Table 2. Selected Bond Lengths (Å) and Bond Angles (deg)
for Complexes 4−7
4
5
6
7
Bond Lengths
2.196(4)
Ln1−O1
Ln1−O2
Ln1−O3
Ln1−O3A
Ln1−N1
Ln1−N2
2.139(6)
2.108(6)
2.261(6)
2.265(5)
2.427(7)
2.551(7)
2.133(5)
2.106(5)
2.237(6)
2.222(5)
2.393(7)
2.488(7)
2.150(4)
2.129(4)
2.273(3)
2.281(3)
2.438(5)
2.522(4)
2.183(4)
2.342(4)
2.318(4)
2.504(5)
2.611(5)
Bond Angles
141.20(2)
88.00(16)
160.03(16)
72.31(16)
171.13(14)
99.31(15)
99.02(15)
O1−Ln1−N2
N1−Ln1−O2
O2−Ln1−O3
O3−Ln1−O3A
O3A−Ln1−N1
N1−Ln1−O3
O2−Ln1−O3A
144.2(2)
91.7(2)
97.3(2)
72.5(2)
96.3(2)
167.0(2)
162.2(2)
147.8(2)
92.1(2)
96.3(2)
73.4(2)
96.4(2)
167.8(2)
164.1(2)
144.69(14)
92.25(15)
92.31(13)
73.07(13)
99.43(14)
163.42(15)
162.84(14)
X-ray diffraction analyses displayed that complexes 1−3 are
isostructural and have solvated monomeric structures. Thus,
only the ORTEP diagram of complex 1 is shown in Figure 1.
The compositions of complexes 1−7 were confirmed by
elemental analysis and NMR spectroscopy in the cases of
complexes 1, 4, and 7. The definitive molecular structures were
determined by single-crystal structure analysis. Complexes 1−7
are extremely sensitive to air and moisture. The crystals
decompose in a few minutes when they are exposed to air, but
neither the crystals nor the solution showed any sign of
decomposition after several months when they were stored
under argon. Complexes 1−3 are freely soluble in hexane, while
compounds 4−7 are freely soluble in THF and toluene and
slightly soluble in hexane.
Crystal Structures. Crystals suitable for an X-ray structure
determination of complexes 1−3 were obtained from a hexane
solution at room temperature, whereas the crystals of complex
4 were obtained from a benzene solution and complexes 5−7
were obtained from a hexane/THF solution. The selected bond
lengths and angles for these complexes are provided in Tables 1
and 2.
Figure 1. ORTEP diagram of complex 1 showing the atom-numbering
scheme. Thermal ellipsoids are drawn at the 10% probability level, and
hydrogen atoms are omitted for clarity. Complexes 2 and 3 are
isomorphous with complex 1.
Table 1. Selected Bond Lengths (Å) and Bond Angles (deg)
for Complexes 1−3
Just as for Salan and Salen lanthanide complexes,4b,14 the two
nitrogen atoms of the imine and amine bridges were found to
bind to the metal center in the solid state. The lanthanide ion is
six-coordinated by two oxygen atoms and two nitrogen atoms
from the dianionic Salalen ligand, one nitrogen atom from the
amido group, and one oxygen atom from one THF molecule.
The coordination geometry around the metal center can be
best described as a slightly distorted octahedron, in which
O(1), O(2), N(2), and N(3) occupy equatorial positions and
O(3) and N(1) occupy axial positions. In complexes 1−3, the
average Ln−O(Ar) bond lengths are 2.142(5), 2.193(4), and
2.218(5) Å, respectively. The increasing trend Y < Sm < Nd
reflects the usual lanthanide contraction from Nd3+ to Y3+.
Similar consequences are also observed from the Ln−N bond
1
2
3
Bond Lengths
Ln1−O1
Ln1−O2
Ln1−O3
Ln1−N1
Ln1−N2
Ln1−N3
2.131(5)
2.152(5)
2.410(6)
2.419(8)
2.617(6)
2.330(6)
2.186(4)
2.199(4)
2.497(4)
2.498(5)
2.684(4)
2.374(5)
2.234(5)
2.201(4)
2.554(5)
2.552(5)
2.773(6)
2.397(5)
Bond Angles
O3−Ln1−N1
O1−Ln1−O2
O2−Ln1−N2
N2−Ln1−N3
N3−Ln1−O1
161.53(16)
96.0(2)
159.31(16)
95.81(15)
70.93(14)
93.75(15)
107.14(15)
155.82(18)
100.03(18)
70.05(16)
92.80(18)
114.05(18)
72.5(2)
93.5(2)
105.3(2)
E
dx.doi.org/10.1021/om4001023 | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
lengths in these complexes. As expected, the Ln−N(imine)
bond length is considerably shorter than the Ln−N(amine)
bond length.
In complex 1, the Y−O(Ar) bond length of 2.142(5) Å is in
accord with the values in bridged bis(phenolate) yttrium
complexes, such as (Salen)Y[N(SiHMe2)2](THF) (2.16 Å;
Salen
= N,N′-bis(3,5-di-tert-butylsalicylidene)-
ethylenediamine]),14a [NNOO]Y(OC6H4-4-CH3)(THF)
(2.152(4) Å; NNOO = Me2NCH2CH2N{CH2-(2-O-C6H2-
But2-3,5)}2),3c [NNOO]Y(OCH2CF3)(THF) (2.158(3) Å),3d
and [ONOO]Y[N(SiHMe2)2](THF) (2.151(2) Å; ONOO =
MeOCH2CH2N{CH2-(2-O-C6H2-tBu2-3,5)}2).15
Complexes 4−6 are isostructural and have unsolvated
dimeric structures. The ORTEP diagram of complex 4 is
shown in Figure 2. Complexes 4−6 contain a Ln2O2 core
Figure 3. ORTEP diagram of complex 7 showing the atom-numbering
scheme. Thermal ellipsoids are drawn at the 10% probability level, and
hydrogen atoms are omitted for clarity.
behaviors of complexes 1−7 for the ROP of rac-β-
butyrolactone and rac-LA were examined. To our surprise,
these Salalen lanthanide complexes are inert for rac-β-
butyrolactone polymerization. It has reported that the
lanthanide amido and alkoxo complexes stabilized by Salan
and amine bridged bis(phenolate) ligands showed good activity
and stereoselectivity for rac-β-butyrolactone.3,4b These results
indicated that the backbone of the ancillary ligands has a crucial
influence on the catalytic property of the corresponding
lanthanide derivatives for rac-β-butyrolactone polymerization.
It was found that these Salalen lanthanide complexes are active
for rac-LA polymerization, and the representative polymer-
ization data are summarized in Table 3.
It can be seen that all of the lanthanide amido complexes can
initiate the ROP of rac-LA to give the PLAs with high
molecular weights. Polymerization media play a key role not
only in polymerization activity but also in stereoselectivity. For
example, using complex 1 as the initiator, the yield is 99% and
Pr is 0.84 (Pr = probability of racemic enchainment) in THF,
when the molar ratio of monomer to initiator is 400 (Table 3,
entry 2), whereas the yield is only 85% and Pr is 0.60 in toluene
even when the reaction time is extended to 5 h under the same
polymerization conditions (Table 3, entry 5).
The ionic radii of the lanthanide metals have an obvious
effect on the catalytic activity and stereoselectivity for rac-LA
polymerization. Using neodymium complex 3 as the initiator,
the yield is 90% when the molar ratio of monomer to initiator is
2000 in 1 h (Table 3, entry 9), whereas the yield is 68% using
yttrium complex 1 as the initiator (Table 3, entry 4). The
decreasing activity order Nd > Sm > Y is in agreement with the
order of their ionic radii, which is also consistent with the active
trend observed in the other lanthanide initiator systems for rac-
LA polymerization.3d,16 This may be attributed to the larger
ionic radii resulting in a greater opening of the metal
Figure 2. ORTEP diagram of complex 4 showing the atom-numbering
scheme. Thermal ellipsoids are drawn at the 10% probability level, and
hydrogen atoms are omitted for clarity. Complexes 5 and 6 are
isomorphous with complex 4.
bridging through the oxygen atoms of the two OCH2Ph groups.
Each of the lanthanide atoms is six-coordinated by two oxygen
atoms and two nitrogen atoms from the dianionic Salalen
ligand and two oxygen atoms from the two OCH2Ph groups.
The average Y−O(Ar) bond length of 2.124(6) Å in complex 4
is comparable with the corresponding value in complex 1. The
average Y−O(CH2Ph) bond length (2.263(5) Å) is in accord
with the values in [(Salan)Y(OPri)]2 (2.276(2) Å; Salan =
MeN(CH2)2MeN{CH2-(2-O-C6H2-But2-3,5)}2).4b Complex 7
also has an unsolvated dimeric structure, and the ORTEP
diagram is shown in Figure 3. The coordination environment
around yttrium atom in complex 7 is identical with those in
complexes 4−6. The Y−O(Ar) and Y−O(CH2Ph) bond
lengths are 2.140(4) and 2.277(3) Å, respectively, which are
consistent with the corresponding bond lengths in complex 4.
Ring-Opening Polymerization of rac-LA by Com-
plexes 1−7. To understand the relationship of the structures
of these Salalen lanthanide complexes on the polymerization
activity, controllability, and stereoselectivity, the catalytic
F
dx.doi.org/10.1021/om4001023 | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
complex 4 under the same polymerization conditions (Table 3,
entry 12). A similar phenomenon was also observed in the
dithiaalkanediyl-bridged lanthanide amide systems.18 This
difference might also be attributed to the fact that the Salalen
lanthaide alkoxides have dimeric structures, and the cleavage of
the dimeric structure by the monomer is slow.
The property of the initiating groups has an obvious effect
not only on the catalytic activity but also on the controllability
for rac-LA polymerization. The Salalen lanthanide alkoxides
showed better controllability for rac-LA polymerization than
the lanthanide amides, to give polymers with narrow molecular
weight distributions (1.17−1.33). To further elucidate the
controlled character of the polymerization, the relationship
between the number-average molecular weight (Mn) and the
molar ratio of monomer to initiator ([M]0/[I]0) was measured
as shown in Figure 4 (Table 3, entries 10, 11, and 13−15).
Table 3. Polymerization of rac-LA Initiated by Complexes
1−6
a
c
d
[M]0/
[I]0
yield
Mc
Mn
b
d
e
cntry cat.
t
(%)
(×104) (×104) PDI
Pr
1
2
3
4
5
6
7
8
9
1
1
1
1
1
2
2
3
3
4
4
4
4
4
4
5
6
7
7
400
400
30 min
1 h
1 h
1 h
5 h
1 h
1 h
1 h
1 h
2 h
2 h
1 h
2 h
2 h
2 h
2 h
3 h
2 h
2 h
90
99
90
68
85
92
82
95
90
90
94
77
94
90
87
95
85
96
91
5.18
5.70
4.66
4.47
1.44
0.85
0.84
0.85
0.85
0.60
0.72
0.72
0.69
0.69
0.85
0.85
0.84
0.85
0.85
0.83
0.73
0.76
0.85
0.83
1.51
1.47
1.47
1.58
1.57
1.66
1.40
1.51
1.30
1.33
1.23
1.17
1.27
1.22
1.25
1.18
1.29
1.14
1000
2000
400
12.96
19.58
4.90
12.02
9.23
f
5.73
1000
2000
1000
2000
100
13.25
23.62
13.68
25.92
1.30
11.70
10.44
11.33
13.06
1.51
10
11
12
13
14
15
16
17
18
19
200
2.71
3.40
400
4.44
3.92
400
5.41
4.67
600
7.78
7.22
800
10.02
10.94
4.90
10.57
11.24
5.45
800
400
400
5.53
5.79
600
7.86
8.16
a
General polymerization conditions: THF as the solvent, [rac-LA] = 1
b
mol/L, at 25 °C. Yield: (weight of polymer obtained)/(weight of
c
monomer used). Mc = 144.13 × [M]0/[I]0 × (polymer yield) (%).
d
Measured by GPC calibrated with standard polystyrene samples.
e
f
1
Measured by homodecoupling H NMR spectroscopy at 25 °C. In
toluene.
coordination sphere in the vicinity of the σ ligand, which makes
the insertion of rac-LA into Ln−N bonds easier. These Salalen
lanthanide amido complexes polymerize rac-LA to give PLAs
with moderate to good heterotacticity. The increasing order in
heterotacticity Nd (Pr = 0.69) < Sm (Pr = 0.72) < Y (Pr =
0.84−0.85) follows the decreasing order of ionic radii of the
central metal, which indicated that a crowded coordination
environment around the metal center is essential for higher
stereoselectivity for rac-LA polymerization. The influence of the
ionic radii of the metal center in this system is in agreement
with the observations reported in the amine-bridged bis-
(phenolate)−lanthanide systems.3a,d,17 It has been reported
that the aluminum methyl complex of the same Salalen ligand
Figure 4. Polymerization of rac-LA initiated by complex 4 in THF at
25 °C: relationship between the number-average molecular weight
(Mn) and the molar ratio of monomer to initiator.
When the molar ratio of monomer to initiator increased, the
molecular weights of the resultant polymers increased linearly,
whereas the molecular weight distributions were kept almost
unchanged, indicating that the polymerization process is
controlled.
showed isotactic selectivity for rac-LA polymerization (Pm
=
The initiation groups had no obvious effect on the
stereoselectivity. These Salalen lanthanide akoxo complexes
showed similar stereoselectivities for rac-LA polymerization in
comparison with the corresponding lanthanide amides. These
results are consistent with those observed in the literature.3a,18
To elucidate the influence of the bulkiness of the substituent on
the phenol ring on the polymerization stereoselectivity, the
polymerization behavior of [L2Y(OCH2Ph)]2 (7) for the ROP
of rac-LA was further examined. It was found that the steric
bulkiness of the substituent group on the phenol ring has no
obvious effect on the stereoselectivity in our system. The Pr
value of the PLA initiated by complex 7 is almost the same as
that of the PLA initiated by complex 4. A similar phenomenon
was also found by Okuda’s group in the dithiaalkanediyl-
bridged lanthanide amide systems.18 However, Carpentier and
co-workers found that bulky and conformationally flexible ortho
substituents enhanced the heteroselectivity in the amine
bridged bis(phenolate) lanthanide systems.3a,19
0.61),10c whereas the Ti(IV) counterpart complex showed
atactic selectivity (Pr = 0.50).10b However, the heterotactic
selectivity can reach Pr = 0.85 in our cases. These results
indicated that the stereoselectivity of a metal complex for the
ROP of rac-LA is affected by the Lewis acidity of the central
metal.
Generally, the polymerization initiated by lanthanide amides
was not well-controlled, because the silylamide group is less
nucleophilic than alkoxide, which caused a relatively slow
initiation.3a,18 To explore the difference in the initiation of the
polymerization between the amide and the alkoxide groups in
our system, the catalytic behavior of the lanthanide alkoxides
4−6 toward the polymerization of rac-LA was also examined.
As anticipated, the polymerization was more slowly initiated by
complexes 4−6 in comparison with complexes 1−3. For
example, using the amido complex 1 as the initiator, the
polymerization was complete in 1 h when the molar ratio of
monomer to initiator was 400 (Table 3, entry 1). However, the
yield was 77% when the polymerization was initiated by
The initiation mechanism was elucidated by end group
analysis of the oligomer of rac-LA, which was prepared by the
G
dx.doi.org/10.1021/om4001023 | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
1
Figure 5. H NMR spectrum of rac-LA oligomer initiated by complex 4 in CDCl3 ([rac-LA]0/4 = 20:1, 25 °C).
Figure 6. 13C NMR spectrum of rac-LA oligomer initiated by complex 4 in CDCl3 ([rac-LA]0/4 = 20:1, 25 °C).
reaction of complex 4 with rac-LA in a 1:20 molar ratio. End-
group analysis by H NMR spectroscopy showed clearly the
revealed that the Salalen group was not involved in the
polymerization process. The existence of a benzyloxy group was
also confirmed by 13C NMR spectroscopy, according to the
resonance at about 128.6 ppm (Figure 6). A MALDI-TOF
mass spectrum further revealed that only oligomers with
PhCH2O− end groups were formed (Figure 7), and it also
indicated that there is a small degree of transesterification side
1
existence of a benzyloxy group and a HOCH(CH3)CO− group
according to the resonances at about 7.35 ppm for the former
and at 1.26, 2.75, and 4.34 ppm for the latter, as shown in
Figure 5. Meanwhile, no proton resonance of the Salalen ligand
was observed in the 1H NMR spectrum of the oligomer, which
H
dx.doi.org/10.1021/om4001023 | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Figure 7. MALDI-TOF mass spectrum of rac-LA oligomer initiated by complex 4 ([rac-LA]0/4 = 20:1, in THF, 25 °C; doped with NaI).
reactions during the lactide polymerization initiated by the
Salalen yttrium alkoxide, giving rise to series of chains separated
by the mass of a lactide unit (144) rather than that of a lactic
acid (72). Thus, the polymerization proceeds via a common
“coordination−insertion” mechanism.
REFERENCES
■
(1) (a) Ragauskas, A. J.; Williams, C. K.; Davison, B. H.; Britovsek,
G.; Cairney, J.; Eckert, C. A.; Frederick, W. J.; Hallett, J. P.; Leak, D. J.;
Liotta, C. L.; Mielenz, J. R.; Murphy, R.; Templer, R.; Tschaplinski, T.
Science 2006, 311, 484. (b) Williams, C. K.; Hillmyer, M. A. Polym. Rev.
2008, 48, 1. (c) Dove, A. P. Chem. Commun. 2008, 48, 6446.
(d) Platel, R. H.; Hodgson, L. M.; Williams, C. K. Polym. Rev. 2008,
48, 11. (e) Inkinen, S.; Hakkarainen, M.; Albertsson, A. C.; Sodergard,
A. Biomacromolecules 2011, 12, 523. (f) Mecking, S. Angew. Chem., Int.
Ed. 2004, 43, 1078. (g) Dechy-Cabaret, O.; Martin-Vaca, B.;
Bourissou, D. Chem. Rev. 2004, 104, 6147. (h) Tschan, M. J.-L.;
Brule, E.; Haquette, P.; Thomas, C. M. Polym. Chem. 2012, 3, 836.
(2) (a) Thomas, C. M. Chem. Soc. Rev. 2010, 39, 165. (b) Stanford,
M. J.; Dove, A. P. Chem. Soc. Rev. 2010, 39, 486. (c) Bakewell, C.; Cao,
T.-P.-A.; Long, N.; Le Goff, X. F.; Auffrant, A.; Williams, C. K. J. Am.
Chem. Soc. 2012, 134, 20577.
(3) (a) Amgoune, A.; Thomas, C. M.; Roisnel, T.; Carpentier, J.-F.
Chem. Eur. J. 2006, 12, 169. (b) Liu, X.; Shang, X.; Tang, T.; Hu, N.;
Pei, F.; Cui, D.; Chen, X.; Jing, X. Organometallics 2007, 26, 2747.
(c) Nie, K.; Gu, X.; Yao, Y.; Zhang, Y.; Shen, Q. Dalton Trans. 2010,
39, 6832. (d) Nie, K.; Fang, L.; Yao, Y.; Zhang, Y.; Shen, Q.; Wang, Y.
Inorg. Chem. 2012, 51, 11133. (e) Amgoune, A.; Thomas, C. M.; Ilinca,
S.; Roisnel, T.; Carpentier, J. F. Angew. Chem., Int. Ed. 2006, 45, 2782.
(f) Alaaeddine, A.; Thomas, C. M.; Roisnel, T.; Carpentier, J. F.
Organometallics 2009, 28, 1469.
(4) (a) Matsumoto, K.; Saito, B.; Katsuki, T. Chem. Commun. 2007,
3619. (b) Kramer, J. W.; Treitler, D. S.; Dunn, E. W.; Castro, P. M.;
Roisnel, T.; Thomas, C. M.; Coates, G. W. J. Am. Chem. Soc. 2009,
131, 16042.
(5) (a) Fujisaki, J.; Matsumoto, K.; Matsumoto, K.; Katsuki, T. J. Am.
Chem. Soc. 2010, 133, 56. (b) Matsumoto, K.; Yamaguchi, T.; Fujisaki,
J.; Saito, B.; Katsuki, T. Chem. Asian J. 2008, 3, 351. (c) Matsumoto,
K.; Kubo, T.; Katsuki, T. Chem. Eur. J. 2009, 15, 6573. (d) Xiong, D.;
Hu, X.; Wang, S.; Miao, C.-X.; Xia, C.; Sun, W. Eur. J. Org. Chem.
2011, 2011, 4289. (e) Berkessel, A.; Brandenburg, M.; Leitterstorf, E.;
CONCLUSION
■
In summary, a Salalen-type ligand was introduced in organo-
lanthanide chemistry for the first time, and a series of neutral
lanthanide amido and alkoxo complexes were synthesized via
amine elimination reactions and proton exchange reactions
using the readily available Ln[N(SiMe3)2]3(μ-Cl)Li(THF)3 and
Cp3Ln(THF) as the precursors, respectively. Their structural
features have been provided by an X-ray diffraction study. It
was found that these lanthanide complexes are efficient
initiators for the ROP of rac-LA, displaying good activity and
stereoselectivity, to give PLAs with moderate to good
heterotacticity. The nature of the initiating groups has a
significant effect on the activity and the controllability but has
no obvious effect on the stereoselectivity. The observed
increasing order of activity is in agreement with the order of
the ionic radii, but the heterotacticity follows the opposite
sequence. The steric bulkiness of the substituent group on the
phenol ring has no obvious effect on the stereocontrollability of
rac-LA polymerization. However, these Salalen lanthanide
complexes are inert for rac-β-butyrolactone, indicating that
the backbone of the ancillary ligands has a crucial influence on
the catalytic property of the corresponding lanthanide
derivatives for rac-β-butyrolactone polymerization.
ASSOCIATED CONTENT
* Supporting Information
■
S
CIF files and a table giving X-ray crystallographic data for 1−7
and figures giving additional characterization data. This material
is available free of charge via the Internet at http://pubs.acs.org.
Frey, J.; Lex, J.; Schafer, M. Adv. Synth. Catal. 2007, 349, 2385.
(f) Sawada, Y.; Matsumoto, K.; Katsuki, T. Angew. Chem., Int. Ed.
2007, 46, 4559.
̈
(6) (a) Saito, B.; Katsuki, T. Angew. Chem., Int. Ed. 2005, 44, 4600.
(b) Saito, B.; Egami, H.; Katsuki, T. J. Am. Chem. Soc. 2007, 129, 1978.
(c) Li, W.; Qin, S.; Su, Z.; Yang, H.; Hu, C. Organometallics 2011, 30,
2095.
(7) Ziegler, J. E.; Du, G.; Fanwick, P. E.; Abu-Omar, M. M. Inorg.
Chem. 2009, 48, 11290.
AUTHOR INFORMATION
Corresponding Author
*Y.Y.: e-mail, yaoym@suda.edu.cn; tel, (86)512-65882806; fax,
(86)512-65880305.
■
(8) Nakano, K.; Nakamura, M.; Nozaki, K. Macromolecules 2009, 42,
6972.
(9) Press, K.; Cohen, A.; Goldberg, I.; Venditto, V.; Mazzeo, M.; Kol,
Notes
The authors declare no competing financial interest.
M. Angew. Chem., Int. Ed. 2011, 50, 3529.
ACKNOWLEDGMENTS
■
(10) (a) Whitelaw, E. L.; Davidson, M. G.; Jones, M. D. Chem.
Commun. 2011, 47, 10004. (b) Whitelaw, E. L.; Jones, M. D.; Mahon,
M. F. Inorg. Chem. 2010, 49, 7176. (c) Whitelaw, E. L.; Loraine, G.;
Mahon, M. F.; Jones, M. D. Dalton Trans. 2011, 40, 11469.
Financial support from the National Natural Science
Foundation of China (Grants 21174095, and 21132002), the
PAPD, and the Qing Lan Project is gratefully acknowledged.
I
dx.doi.org/10.1021/om4001023 | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
(11) (a) Andersen, R. A.; Templeton, D. H.; Zalkin, A. Inorg. Chem.
1978, 17, 2317. (b) Edelmann, F. T.; Steiner, A.; Stalke, D.; Gilje, J.
W.; Jagner, S.; Hakansson, M. Polyhedron 1994, 13, 539. (c) Zhou, S.-
L.; Wang, S.-W.; Yang, G.-S.; Liu, X.-Y.; Sheng, E.-H.; Zhang, K.-H.;
Cheng, L.; Huang, Z.-X. Polyhedron 2003, 22, 1019. (d) Birmingham,
J. M.; Wilkinson, G. J. Am. Chem. Soc. 1956, 78, 42.
(12) Yeori, A.; Gendler, S.; Groysman, S.; Goldberg, I.; Kol, M. Inorg.
Chem. Commun. 2004, 7, 280.
(13) Atwood, J. L.; Hunter, W. E.; Wayda, A. L.; Evans, W. J. Inorg.
Chem. 1981, 20, 4115.
(14) (a) Runte, O.; Priermeier, T.; Anwander, R. Chem. Commun.
1996, 1385. (b) Evans, W. J.; Fujimoto, C. H.; Ziller, J. W. Polyhedron
2002, 21, 1683.
(15) Cai, C.-X.; Toupet, L.; Lehmann, C. W.; Carpentier, J.-F. J.
Organomet. Chem. 2003, 683, 131.
(16) (a) Zhang, J.; Qiu, J.; Yao, Y.; Zhang, Y.; Shen, Q.
Organometallics 2012, 31, 3138. (b) Li, W.; Zhang, Z.; Yao, Y.;
Zhang, Y.; Shen, Q. Organometallics 2012, 31, 3499.
(17) Cai, C.-X.; Amgoune, A.; Lehmann, C. W.; Carpentier, J.-F.
Chem. Commun. 2004, 330.
(18) Ma, H.; Spaniol, T. P.; Okuda, J. Angew. Chem., Int. Ed. 2006, 45,
7818.
(19) (a) Amgoune, A.; Thomas, C. M.; Carpentier, J. F. Macromol.
Rapid Commun. 2007, 28, 693. (b) Bouyahyi, M.; Ajellal, N.; Kirillov,
E.; Thomas, C. M.; Carpentier, J. F. Chem. Eur. J. 2011, 17, 187.
J
dx.doi.org/10.1021/om4001023 | Organometallics XXXX, XXX, XXX−XXX