Aluminium complexes bearing functionalized trisamido ligands and their reactivity in the polymerization of ε-caprolactone and rac-lactide

Article abstract of DOI:10.1039/c0dt00005a

The addition of 1 and 2 equivalents of AlMe₃ to cis,cis-C ₆H₉(NHCH₂C₆H₄-o-R) ₃ (R = PPh₂ (3) and SPh (4)) gives complexes [cis,cis-C₆H₉(NCH₂C₆H ₄-o-R-κN)₂(NHCH₂C₆H ₄-o-R-κN)]AlMe (R = PPh₂ (7) and SPh (8)) and [cis,cis-C₆H₉(NCH₂C₆H ₄-o-R)₃-κ⁵μ²N]Al ₂Me₃ (R = PPh₂ (5) and SPh (6)), respectively. The bimetallic complexes are active in the polymerization of ε-caprolactone and rac-lactide whereas the monometallic complexes are not, although no cooperative behaviour is observed between the two aluminium atoms of 5 and 6. The polycaprolactone samples, which were characterized using ¹H NMR, MALDI-TOF, and SEC, show the presence of residual ligands 3 or 4 bound to the polymer and the in situ NMR studies confirm that the insertion occurs in an Al-N bond.

Full text of DOI:10.1039/c0dt00005a

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/dalton | Dalton Transactions
Aluminium complexes bearing functionalized trisamido ligands and their
reactivity in the polymerization of e-caprolactone and rac-lactide†
Marie-He´le`ne Thibault and Fre´de´ric-Georges Fontaine*
Received 19th February 2010, Accepted 13th April 2010
First published as an Advance Article on the web 19th May 2010
DOI: 10.1039/c0dt00005a
The addition of 1 and 2 equivalents of AlMe3 to cis,cis-C6H9(NHCH2C6H4-o-R)3 (R = PPh2 (3) and
SPh (4)) gives complexes [cis,cis-C6H9(NCH2C6H4-o-R-kN)2(NHCH2C6H4-o-R-kN)]AlMe (R = PPh2
5
2
(7) and SPh (8)) and [cis,cis-C6H9(NCH2C6H4-o-R)3-k m N]Al2Me3 (R = PPh2 (5) and SPh (6)),
respectively. The bimetallic complexes are active in the polymerization of e-caprolactone and rac-lactide
whereas the monometallic complexes are not, although no cooperative behaviour is observed between
the two aluminium atoms of 5 and 6. The polycaprolactone samples, which were characterized using 1H
NMR, MALDI-TOF, and SEC, show the presence of residual ligands 3 or 4 bound to the polymer and
the in situ NMR studies confirm that the insertion occurs in an Al–N bond.
Introduction
The worldwide consumption of plastics has risen steadily since the
revolutionary discovery of alkene polymerization by Ziegler and
Natta.1 While a significant percentage of the polymers currently
produced are recycled, there is still a large amount finding their way
into the ecosystem, having a negative impact on the environment.2
An ecological alternative to saturated polyolefins are biodegrad-
able polymers such as polylactides and polycaprolactones.3 In
addition to biodegradability, these materials are biocompatible
and are widely used for medical applications.3 However, in
order to obtain polymers with good mechanical properties, the
microstructure of the polymeric chains, including their molecular
weight and polydispersity, needs to be controlled; something that
can be done using catalysis. As such, interest over the past decade
for the catalytic ring-opening polymerization (ROP) of lactones
and lactides has spurred several review articles.4
Numerous metals have been known to catalyze the formation
of polylactones and polylactides with the most notable catalysts
being electrophilic metal ions, such as magnesium,5 calcium,5c,6
titanium,7 iron,8 zinc,5a-c,9 tin,10 and aluminium.11 Aluminium is
one of the most efficient catalysts for e-caprolactone and lactide
polymerization and is generally stabilized using N or O supported
ligands. Whereas single site catalysts are numerous, examples
where more than one metal site is required for efficient catalysis
are scarce.12-14 Examples of such collaborative work are found
in dizinc-monoalkoxide complexes that are believed to behave
similarly to metalloenzymes13 (Fig. 1A) and in a macrocyclic Schiff
base bisaluminium complex where a cooperative effect between
the two aluminium centers has been observed (R = H, Fig. 1B).14
In the latter example, one aluminium atom serves as the Lewis
Fig. 1 Two examples of bimetallic catalysts for ring-opening polymeriza-
tion of cyclic esters.
acid, and an alkyl functionality bound to the second metal atom
attacks the carbonyl group of the incoming ester during catalysis.
It was also proposed that adjacent aluminium centers linked in a
fashion similar to aluminoxane (R = AlMe2, Fig. 1B), hindered
the polymerization process, and that cooperation came instead
between aluminium atoms on both extremities of the macrocycle.
A specific arrangement of the metallic centers is thus crucial for
cooperative behaviour.
Our research group has been investigating the coordina-
tion chemistry of ambiphilic aluminium complexes having
both Lewis acidic and basic moieties.15 Derivatives of cis,cis-
triaminocyclohexane with pendant soft donor groups were chosen
as ligands for the synthesis of ambiphilic ligands. Hard elec-
trophilic metals such as Al3+ prefer to bind covalently to amido
moieties,16 generating a Lewis acidic coordination site, whereas
soft ligands are free to coordinate low oxidation state transition
metals, such as platinum(0).17 In our systematic study of the reac-
tivity of the functionalized triaminocyclohexane ligands, we wish
to report that the hexadentate ligands cis,cis-C6H9(NHCH2C6H4-
o-R)3 (R = PPh2, SPh) can bind one or two equivalents of AlMe3
depending on the stoichiometry of the reaction. The bimetallic
complexes formed are also active in ring-opening polymerization
of cyclic esters, such as e-caprolactone and lactides.
De´partement de Chimie, Universite´ Laval, 1045 Avenue de la Me´decine,
Que´bec, Canada G1V 0A6. E-mail: frederic.fontaine@chm.ulaval.ca
† Electronic supplementary information (ESI) available: Typical steric
1
exclusion chromatogram and 13C{ H}NMR for polylactides. CCDC
reference number 755385. For ESI and crystallographic data in CIF or
other electronic format see DOI: 10.1039/c0dt00005a
5688 | Dalton Trans., 2010, 39, 5688–5697
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
5 and at -0.15, -0.26, and -0.32 ppm for 6 did show that three
methyl groups remained bound to the aluminium atoms. Another
significant spectroscopic feature was the splitting of the single
resonance for the methylene protons on the functional arms into
three resonances integrating for two protons each. The resonances
for the protons on the cyclohexyl ring also indicated a loss of
symmetry; the signals that were observed as singlets in 3 and 4
were split into two series of resonances in a 1 : 2 ratio. Finally, two
Results and discussion
Ligand synthesis
The general procedure for the synthesis of the aminophos-
phine/thiol ligands is summarized in Scheme 1. Reaction of
three equivalents of the corresponding 2R-benzaldehyde with one
equivalent of cis,cis-1,3,5-triaminocyclohexane afforded the Schiff
=
bases cis,cis-C6H9(N CHC6H4-o-R))3 (1, 2) (R = PPh2, SPh) in
1
resonances were observed by 31P{ H}NMR spectroscopy at -14.1
approximately 70% yield. The reduction of 1 and 2 using excess
LiAlH4 gave the hexadentate ligands cis,cis-C6H9(NHCH2C6H4-
o-R)3 (3, 4) (R = PPh2, SPh) in excellent yields. It should be noted
that thiol analogues 2 and 4 show similar spectroscopic features
to what was observed for 1 and 3, which have been previously
reported.17
and -14.9 ppm for 5, integrating at a 1 : 2 ratio. These observations,
and literature precedent by the groups of Johnson18b and Chen,19
are in agreement with the structure depicted in Scheme 2, where
the functional arms of the cyclohexyl ring are in an axial position
and a plane of symmetry is passing through the Al2Me3 core.
Coordination to the aluminium atoms induces a flip in the
cyclohexane framework, with the functionalized arms now in axial
position, something that was previously observed for some cis,cis-
triamidocyclohexane complexes of early transition metals.16 While
the formation of 5 was clean by NMR spectroscopy, in the case
of 6, a small excess of trimethylaluminium remained coordinated
by the sulfide moieties and could not be removed under reduced
1
pressure, as observed by the presence of broad signals in the H
NMR spectrum. Adding triethylamine proved efficient to generate
6 cleanly by forming the Et3N·AlMe3 adduct (1H NMR d = 2.19,
0.68 and -0.40) which could be removed by subsequent washings
with pentane. For comparison, Chen and al. reported that upon
the addition of 2 equivalents of AlMe3 to the more flexible ligand
MeSi[SiMe2NH(4-MeC6H4)]3, only two of the amido moieties,
instead of three as observed for 5 and 6, were binding the metal
centers, generating a complex with a Al2N2 core.19
It was possible to obtain crystals of complex 6 by slow evapo-
ration of a pentane solution, thus confirming its connectivity. The
complex crystallizes in a P1 space group with two independent
Scheme 1 Synthesis of the aminophosphine/thiol ligands.
¯
molecules (Z = 4). The main difference between the molecules
is the conformation of one of the functionalized arms. Indeed,
the Al(1)–N(3)–C(36)–C(37) torsion angle is -135.2(1)◦ whereas
the equivalent torsion angle on the other molecule (Al(11)–
N(13)–C(136)–C(137)) is 147.9(2)◦. However, in both molecules
the thiophenyl moieties are in the same quadrant relative to the
metallic core. An ORTEP representation of one of the molecules
is shown in Fig. 2 and a simplified depiction of the conformers is
shown in Fig. 3.
As observed for the solid state structure of 3,17 the higher
symmetry observed in solution for the dialuminium complex is
not present in the solid state structure of 6, suggesting a fast
rotation of the functional arms in solution. In this complex, the
ligand is bound to an Al–Me fragment by three nitrogen atoms
while an AlMe2 fragment shares two nitrogen atoms with the other
metal center. Both aluminium centers are in distorted tetrahe-
dral environments, with the N–Al–N angles of the N2Al2 four-
membered ring being small (78.90 to 79.66◦). The nitrogen atoms
N(1), N(2), N(11), and N(12) are in tetrahedral environments. On
the other hand, N(3) and N(13), which are only bound to one Al
atom, are in a sp2 planar environment, as shown by the sum of
the angles of 359◦. The Al–N bond lengths are similar, ranging
Synthesis of the aluminium complexes
The protonolysis of an aluminium alkane precursor is a well-
known strategy for the synthesis of aluminium amido complexes.18
As expected, 3 and 4 were completely converted to amido
complexes 5 and 6 (Scheme 2) upon addition of more than 2
equivalents of trimethylaluminium (or 1 equivalent of hexamethyl-
dialuminium). Although the release of methane was observed as
a singlet at 0.15 ppm by 1H NMR spectroscopy in both reactions,
the presence of three singlets at -0.12, -0.22, and -0.52 ppm for
˚
˚
from 1.963(2) A to 2.000(2) A, with the exception of the Al(1)–
N(3) and Al(11)–N(13) distances that are significantly shorter
˚
(1.7806(17) and 1.7960(17) A, respectively). To our knowledge,
Scheme 2 Synthesis of aluminium amido complexes.
three other structurally characterized complexes bound to tripodal
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 5688–5697 | 5689
©

View Article Online
When one equivalent of AlMe3 was added to either 3 or 4 in
C6D6, new products were observed exhibiting only one Al–Me
resonance in the 1H NMR spectrum at -0.87 ppm and -0.82 ppm,
for 7 and 8, respectively. The 1H NMR spectra of the two
latter aluminium compounds were also much more complex, with
all hydrogen atoms on the organic framework integrating for
one each, indicating a total absence of symmetry in solution.
It can be proposed that the tripodal ligand is bound to the
methylaluminium fragment with the metal in a pseudo-tetrahedral
fashion. For this to happen, one of the nitrogen atom forms a
dative interaction with a secondary amine. Using HMQC and
COSY experiments, it was possible to locate the amine protons at
3.56 and 3.62 ppm, for 7 and 8 respectively, since they were the
only ones not correlated to a carbon atom. Due to this nitrogen
atom being chiral and sp3-hybridized, all of the diastereotopic
fragments become magnetically unequivalent, as proposed by
Chen et al.19 The isolation of these compounds in the solid state
proved impossible, since they did not precipitate from solution and
remained as oils with small amount of uncharacterized impurities.
Reaction of the bimetallic complexes with one equivalent of
their respective free ligand readily afforded the corresponding
monometallic complexes and the addition of one equivalent of
AlMe3 to 7 and 8 affords 5 and 6, respectively.
Fig. 2 ORTEP diagram of 6. Thermal ellipsoids are presented at 50%
probability. The hydrog◦en atoms were omitted for clarity. Selected bond
˚
lengths (A) and angles ( ): Molecule 1: Al(1)–N(1) 2.0000(16); Al(1)–N(2)
1.9727(16); Al(1)–N(3) 1.7806(17); Al(2)–N(1) 1.9783(16); Al(2)–N(2)
1.9890(17); Al(1)–C(7) 1.948(2); Al(2)–C(8) 1.971(2); Al(2)–C(9)
1.970(2); N(3)–Al(1)–C(7) 114.64(9); N(3)–Al(1)–N(2) 109.07(7);
N(3)–Al(1)–N(1) 107.00(7); N(2)–Al(1)–N(1) 79.45(6); N(1)–Al(2)–N(2)
79.59(6); C(36)–N(3)–C(3) 115.38(16); C(36)–N(3)–Al(1) 131.90(13);
C(3)–N(3)–Al(1) 111.59(12). Molecule 2: Al(11)–N(11) 1.9625(17);
Al(11)–N(12) 1.9824(17); Al(11)–N(13) 1.7960(17); Al(12)–N(11)
1.9976(18);
Al(12)–C(108) 1.971(2); Al(12)–C(109) 1.971(2); N(13)–Al(11)–C(107)
112.87(9); N(13)–Al(11)–N(12) 108.74(8); N(13)–Al(11)–N(11)
Al(12)–N(12)
1.9793(17);
Al(11)–C(107)
1.949(2);
108.09(8); N(12)–Al(11)–N(11) 79.66(7); N(11)–Al(12)–N(12) 78.90(7);
C(136)–N(13)–C(105) 113.10(16); C(136)–N(13)–Al(11) 124.94(13);
C(105)–N(13)–Al(11) 110.80(12).
Polymerization of e-caprolactone
The ring opening polymerization (ROP) of e-caprolactone (CL)
was performed using complexes 5, 6, and 8 as catalysts in a toluene
solution. Since the isolation and the purification of 7 proved
not feasible, no catalytic run was done with it. The yields were
calculated according to the mass of the polymer that precipitated
after quenching the solution with a CH2Cl2 and CH3OH mixture,
and the Mn and Mw values were calculated using steric exclusion
chromatography (SEC) for the high molecular weight domain
(ESI, Fig. S1,† before 20 min). The results are summarized in
Table 1.
Complex 6 was shown to afford better yields of polycaprolac-
tone at room temperature, or at 50 ◦C, than 5, as can be observed
in entries 1 to 6. It is believed that the larger steric hindrance of
PPh2 compared to SPh may play an important role in reducing the
polymer yield and the activity of the catalyst. The reaction with
both catalysts was slow at room temperature, since the polymer
isolated yields were very low after one◦ hour, but the reaction
proceeded more efficiently with 6 at 50 C. At higher monomer
to catalyst ratios, the TOFs ranged between 6.2 and 7.1 min-1
(entries 11 and 12). One limiting factor in the polymerization
was the jellification of the reaction mixture at high concentration
and temperature (entry 6), which occurred after 15 min. However,
running the experiment in a more dilute solution (0.8 M) instead
of a 2.4 M solution with 150 equivalents of CL, in order to prevent
jellification, gave lower yields of isolated polymer (entries 6 and
9). The use of benzyl alcohol (PhCH2OH) as activator for the
polymerization of cyclic esters has been reported.19,20 However,
as shown in entry 7, the presence of the alcohol greatly reduced
both the yield and the Mn of the polymer. The reaction between
one equivalent of benzyl alcohol and catalyst 6 in benzene-d6 did
Fig. 3 Depiction of the different orientations of the functionalized arm
in crystal of 6.
ligands have been reported having similar N3Al2R3 cores.18b,19 In
all examples, the terminal amido ligands have a short N–Al bond
length compared to the bridging amido. Complexes P[CH2N-3,5-
(CF3)2C6H3]3Al2Me3 and Me3Al.P(CH2NPh)3Al2Me3, reported by
Johnson,18b and complex MeSi[SiMe2N(4-MeC6H4)]3AlH(AlH2),
reported by Chen,19 have terminal amido N–Al bonds of 1.829(3),
˚
1.825(2), and 1.842(4) A, respectively, and bridging amido N–
˚
Al bond lengths between 1.958 and 1.997 A. The short bond
length and the sp2 hybridization could be induced by additional
p-donation of the nitrogen lone pair to the aluminium atom,
which would increase the N–Al bond strength while reducing
the nucleophilicity of the amido group. It should be noted that
no significant interaction between the sulfide and aluminium is
1
observed. The similarity in the 31P{ H} NMR chemical shifts
of the dialuminium complex (-14.1 and -14.9 ppm) to that of
free ligand 3 (-15.0 ppm), also suggests the absence of an Al–P
interaction in complex 5.
1
show by H NMR the presence of free ligand 4, thus indicating
the catalyst’s degradation in presence of PhCH2OH.
5690 | Dalton Trans., 2010, 39, 5688–5697
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
Table 1 Polymerization of e-caprolactone using 5, 6, and 8 as catalysts
Temp./◦C
Yield (%)
Mn
Mw
MRMN
PDId
Small : Highf
d
d
e
Entry
Catalyst
[Cat] : [CL]
Vol/ml
Time/h
1
2
3
4
5
6
7
8
5
5
5
6
6
6
6
6
6
6
6
6
1 : 150
1 : 150
1 : 150
1 : 150
1 : 150
1 : 150
1 : 150b
1 : 76c
1 : 150c
1 : 300c
1 : 400c
1 : 600c
1 : 76
1
1
1
1
1
1
1
3
3
3
3
3
1
1
1
1
1
6
1
1
6
1
6
1
1
1
1
1
1
1
1
6
RT
RT
50
RT
RT
50
RT
50
50
50
50
50
50
7
10900
18000
25400
NA
14600
19300
8700
12500
23800
52000
82000
69500
13700
46400
NA
14100
47100
47100
NA
1510
810
1.3
85 : 15
NA
65
78
17
99
98
64
35
68
88
94
71
79
78
6
2.62
1.85
NA
3.12
2.48
1.87
2.03
2.03
2.13
1.65
2.09
1.94
2.34
NA
2.91
4190
1100
7180
10800
7550
5030
7680
24500
26100
36100
5100
31800
NA
62 : 38
26 : 74
8 : 92
3 : 97
20 : 80
4 : 96
11 : 89
2 : 98
4 : 96
7 : 93
15 : 85
6 : 94
NA
45500
47900
16300
25500
48200
110800
135200
145200
26600
108500
NA
9
10
11
12
13
14
15
16
6
8a
8a
8a
1 : 76
1 : 76
1 : 76
50
RT
RT
54
11600
33700
NA
5 : 95
The experiments were carried out using a 2.4 mM solution of catalyst in toluene. The volume includes the volume of the e-caprolactone.a The catalyst
was prepared in situ and monitored by 1H NMR prior its transfer into the Schlenk vessel. b In presence of one equivalent of benzyl alcohol. c 0.8 mmol
solution of catalyst. d According to the higher molecular weight fractions of the chromatogram. e In CDCl3, using the HOCH2R end group of both
polymers terminated by the R-C(O)OMe and R-C(O)-X (X = 3 or 4). f Using SEC, where “small” and “high” represent the relative integrations of the
fractions after and before 20 min, respectively.
The effect of the caprolactone concentration on the yield
and Mn and Mw was assessed (entries 8 to 12). Experiments
were done using catalyst 6 with 76 to 600 equivalents of CL at
◦
50 C for one hour. Between 76 and 400 equivalents of CL, Mn
values increased linearly with CL concentration. The yields also
increased. At 600 equivalents, the trend no longer held and the Mn
value and yield were lower than expected. This could be due to
saturation of the system, since jellification of the reaction mixture
occurred too rapidly and slowed down the reaction.21
The activity of monoaluminic catalyst 8 was also compared
against its bimetallic counterpart (entries 14–16). 8 was not isolable
in the solid state and was thus prepared in situ in an NMR tube
prior to catalysis attempts. It was found that 8 afforded comparable
yields at 50 ◦C and higher Mn values compared to catalyst 6 (entries
13 and 14). However at room temperature, the reaction was slow
to start and low yields were obtained along with lower Mn values
(entries 15 and 16).
1
Fig. 4 Typical H NMR spectrum of polycaprolactone (entry 9) with a
blow up of the resonances of the terminal groups. T1 and T2 are for the
two triplets and S for the singlet.
1H NMR spectroscopy analysis of polycaprolactone
some broad and ill defined resonances at 7.3–7.1, 4.3–4.1, and
The end group analysis using NMR spectroscopy is a powerful
method to obtain the molecular weight of polycaprolactones in
solution.22 The 1H NMR spectra of the polycaprolactones isolated
in the catalytic experiments were typical of the reported values,
with d = 4.05 (t), 2.30 (t), 1.64 (m), and 1.37 (m). However,
the end group resonances were untypical for polycaprolactones
synthesized using aluminium catalysts that are quenched with
2.6–1.1 ppm; these chemical shifts are typical for ligands 3 and
1
4. The 31P{ H} NMR spectrum of the polymer obtained in entry
3 using catalyst 5 did exhibit three minor overlapping resonances
for ligand 3 at -15.6 and -15.9 ppm, along with major resonances
between 32.2 and 31.5 ppm. These resonances are typical for the
phosphine oxide analogue of 3. Therefore, quenching the reaction
with methanol allows to get rid of the aluminium from the catalyst,
but not of the ligand that remains bound by an amide linkage to
the polymeric chain. Indeed, the NC(O)R stretching frequency
was observed at 1651 cm-1 by FTIR spectroscopy. Such amide
linkage could be formed by a ROP initiated by an insertion within
an aluminium–amido bond (vide infra).
methanol. Indeed, two triplets, at 3.641 and 3.646 ppm (JH–H
=
1
5.1 Hz), and a singlet at 3.664 ppm, were observed by H NMR
spectroscopy (Fig. 4). In all samples, the upfield triplet and the
singlet were integrated in a 2 : 3 ratio, as would be expected for
a polycaprolactone having -CH2OH (triplet) and -C(O)OCH3
(singlet) as end groups.22a However, no other end group could
be assigned to the triplet corresponding to the -CH2OH (triplet)
The molecular weight of the polymers was calculated by
integrating both methylene triplets at 3.641 and 3.646 ppm for
a total of 2 protons. It can be observed that the Mn obtained by 1H
NMR spectroscopy is systematically lower than the Mn observed
1
at 3.646 ppm. After careful examination, the H NMR spectra
of polycaprolactones having a low molecular weight did show
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 5688–5697 | 5691
©

View Article Online
using SEC. However, it should be noted that the difference between
the Mn by NMR and SEC is much more acute with polymers
having a lower Mn. With these samples, the presence of 3 or 4, or
its oxidized analogue, as an end group will artificially increase the
molecular weight observed by SEC (3 has a molecular weight of
952 g mol-1), whereas the 1H NMR integrations will give a much
more reliable number of repetition units of the polycaprolactone.
m/z = 1114 with a repetition unit of 114.3, which corresponds to
the molecular weight of CL (molecular weight = 114.06, ESI, Fig.
S2†). The first value of 1114 corresponds to a unit of CL (m/z =
114) bound to ligand 3 where the phosphines are oxidized (m/z =
999) with an additional proton (H+), as previously assumed by
NMR spectroscopy.
In the mass spectrum of the fractions having a m/z < 1000,
some signals could be observed that could be attributed to
repeating units of CL (for example, m/z = 596.3, 710.4, and 824.3),
but no end group could be clearly assigned. However, the presence
of the oxidized analogue of 3 was clearly observed at m/z = 1000.4
as the main signal (expected value for 3 + H+ = 1000.4), suggesting
that not all of the ligands are incorporated in the polymeric
chains.
MALDI-TOF analysis
One general characteristic of this catalytic system is that poly-
dispersity indexes (PDI) are rather high at 50 ◦C. The SEC
traces clearly show the presence of two domains with different
molecular weights (ESI, Fig. S1†). A MALDI-TOF experiment
on the isolated product of entry 3 was carried out (Fig. 5) to have
a more reliable idea of the structure of the lower molecular weight
fraction. It was possible to observe a repetition pattern starting at
In situ study of the polymerization of e-caprolactone
NMR scale reactions were done to gain a better understanding of
the polymerization mechanism. Upon addition of 1 equivalent
of CL to a solution of 6 in C6D6, immediate changes in the
1H NMR spectrum were visible for the aluminium complex
suggesting the complete conversion to a new species (Fig. 6). The
methylene protons were split into 6 doublets ranging from 3.92
to 5.27 ppm, implying that the 6 no longer possessed a mirror
plane in solution, as stated above, and that the hydrogen atoms
were now diastereotopic. Two of the Al–Me signals were shifted
upfield at -0.58 and -0.68 ppm, respectively, and signals attributed
to bound CL emerged at 3.42, 2.07, 1.12, and 0.94 ppm. The fact
that the three Al–Me signals were still clearly visible in the upfield
region of the spectrum indicated that the ring opening insertion
of CL did not occur in the Al–Me bond. These observations are
in agreement with results reported by Milione et al. when using
a cationic heteroscorpionate complex.23 Furthermore, the signals
that shifted upfield (at -0.58 and -0.68 ppm) are reminiscent of the
aluminium complex Me2Al(m-OCH2CH2NMe2)Al(tBu)3 having a
Fig. 5 MALDI-TOF spectrum of entry 3.
Fig. 6 1H NMR spectra of catalyst 6 in the presence of 1 equivalent of CL in benzene-d6 after (a) 5 min at room temperature, (b) 18 h at 60 ◦C after the
addition of a second equiv. of CL, and (c) before the addition of CL.
5692 | Dalton Trans., 2010, 39, 5688–5697
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
Fig. 7 1H NMR spectra of catalyst 8 in the presence of 1 equiv. of CL in benzene-d6 after (a) 5 min at room temperature, (b) 18 h at 60 ◦C, and (c) before
the addition of CL.
R2N-AlMe2-OCH2 core, where the AlMe2 fragment exhibits a 1H
NMR chemical shift of -0.56 ppm.24 We therefore propose that
the insertion occurs into one of the two amido-aluminium bonds
of the N2AlMe2 four-member cycle, rather than in the N3AlMe
central core, to form the species depicted in Fig. 6. The presence
of the ligand bound to the isolated polymeric chain and the
low reactivity of 8, as seen in the next paragraph, support this
hypothesis.
Polymerization of rac-lactide
Complex 6 proved to be active for rac-lactide ROP. A first attempt
at room temperature with 75 equivalents of rac-lactide afforded
less than 5% yield after 5 days. A second attempt at 100 C lead
◦
to 96% yield after 4 days. SEC analyses gave a Mn of 32000 with a
PDI of 1.56.
The tacticity of the polymer was determined by 13C{ H} and
1
1
homodecoupled H NMR experiments. The resonances at 69.0
After addition of one equivalent of CL to 8 in C6D6 almost
and 69.2 ppm, integrating for 4 : 1 and representing the methyne
1
no reaction was detected by H NMR after one hour at room
1
carbon in the 13C{ H}, were attributed to the tetrads iii, iis, sii
temperature (Fig. 7). When heated at 60 ◦C for one hour, the
signals attributed to CL decreased and a new multiplet at 3.98 ppm
became more pronounced. After an additional 17 h at 60 ◦C, the
signals attributed to CL have decreased further but conversion
was still incomplete. However, the most important pieces of
information were that a large amount of 8 was still present in
solution, that the aluminium complex did not seem to be affected
by the presence of CL, and that no new signals were attributed to
an active catalyst. Whereas the exact nature of the active catalyst
with precatalyst 8 could not be confirmed, it can be speculated
that some species in low concentration was formed after heating
the solution. Since the ligand was also observed in the polymer
obtained by 8, as deduced from the series of triplet at 3.64 ppm
in the isolated polymer of entry 14, two possible pathways can
be imagined. First, it is possible that the CL inserts in one of the
N–Al bond at a very slow rate, thus allowing polymerization by
few activated species. Also, it might be possible for some residual
aluminium species to form an analogue of 6, which would be the
active species.
and isi (ESI, Fig. S3†). These are the stereosequences expected
in the polymerization of rac-lactide by an achiral catalyst.25
However, three weaker resonances present in the spectrum at
69.1, 69.3 and 69.4 ppm corresponding to the iss tetrads, which
cannot be normally present in polymers coming from rac-lactide,
were observed. This result implies that some transesterification
or racemization reaction occurred, which have been known to
induce inversion of the stereocenters.25 The methyne region of
1
the homodecoupled H NMR spectrum corroborates the results
1
obtained from the 13C{ H} spectrum. These preliminary results
show that the activity of catalyst 6 is slightly inferior to other
aluminium catalysts that are active in rac-lactide ROP.4a-b,26
Conclusion
The coordination chemistry of an aminothiol and a previously
reported aminophosphine ligand with aluminium was explored.
With both ligands, the reaction with two equivalents of AlMe3
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 5688–5697 | 5693
©

View Article Online
leads to the formation of bimetallic methylalane species. The
reaction of one equivalent afforded monometallic complexes with
one secondary amine bound to aluminium.
analysis calc. for C45H39N3S3: C, 75.27; H, 5.47; N, 5.85. Found:
C, 75.01; H, 5.56; N, 5.72%.
The reactivity of these complexes towards e-caprolactone ROP
was assessed and it was found that the bimetallic complexes 5 and
6 were active catalysts. Precatalyst 8 did exhibit some catalytic
cis,cis-C6H9(NHCH2C6H4SPh)3 (4)
To a solution of 2 in 40 ml of THF (600 mg, 0.84 mmol),
a suspension of LiAlH4 (96 mg, 2.5 mmol) in 10 ml of THF
was added. The suspension was stirred at room temperature for
48 h. The reaction was then filtered to remove excess LiAlH4 and
quenched with water. The mixture was extracted with 3 ¥ 20 ml
water. The organic fraction was dried with MgSO4 and filtered.
1
activity, but probing the reaction using H NMR spectroscopy
did show that the complex did not react significantly with e-
caprolactone and that the activity was probably a consequence
of a minor species not observed resulting from the degradation
of 8. No activator, such as benzyl alcohol, is needed for the
reaction to proceed, as the site of the CL insertion on the bimetallic
complexes was found to be at the N-AlMe2 moiety, as could be
demonstrated by the presence of the residual ligand bound to the
isolated polymer. In the presence of the sulfide containing catalyst
6, the polymers obtained consist mainly of high molecular weight
polymers, however, catalyst 5, with bulky PPh2 groups on the
functionalized arms, produced mainly oligomers. The synthesis
of analogues of 5 and 6 with other functional groups and the
formation of the bimetallic species are currently under way to
test their catalytic activity and probe and cooperative behaviour
between early and late metal systems.
The volatiles were removed under vacuum to yield 4 as yellow
3
oil (530 mg, 91% yield). dH (400 MHz; C6D6) 7.52 (d, JH–H
=
3
7.0, 3H, C6H4(SPh)), 7.37 (d, JH–H = 7.7, 3H, C6H4(SPh)), 7.20
3
(m, 6H, C6H4(SPh)), 7.09 (t, JH–H = 7.5, 3H, C6H4(SPh)), 6.92
(m, 12H, C6H4(SPh)), 3.95 (s, 6H, CHNHCH2C6H4), 2.23 (tt,
3
3JH–H = 11.2 and JH–H = 3.4, 3H, CHNHCH2C6H4), 2.00 (dt,
3
2JH–H = 11.6 and JH–H = 3.4, 3H, equ. CH2), 0.87 (br s, 3H,
2
3
CHNHCH2C6H4) and 0.79 (q, JH–H = 11.6 and JH–H = 11.6,
3H, ax. CH2); dC (100.57 MHz; C6D6) 143.6 (s, C6H4(SPh)), 137.6
(s, C6H4(SPh)), 134.1 (s, C6H4(SPh)), 133.8 (s, C6H4(SPh)), 130.0
(s, C6H4(SPh)), 129.8 (s, C6H4(SPh)), 129.5 (s, C6H4(SPh)), 127.9
(s, C6H4(SPh)), 126.5 (s, C6H4(SPh)), 53.9 (s, CHNHCH2C6H4),
49.5 (s, CHNHCH2C6H4) and 41.0 (s, CH2). One aromatic signal
is hidden beneath the solvent peak. Elemental analysis calc. for
C45H45N3S3: C, 74.75; H, 6.26; N, 5.80. Found: C, 74.36; H, 6.52;
N, 5.77%.
Experimental
cis,cis-1,3,5-Triaminocyclohexane·3HBr27 and 2-(phenylthio)-
benzaldehyde were prepared according to literature procedures.28
Syntheses for cis,cis-1,3,5-triaminocyclohexane and compounds 1
and 3 were previously reported.17 Trimethylaluminium was pur-
chased from Sigma-Aldrich and used as received. e-caprolactone
was heated at 80 ◦C with CaH2 and distilled at 0.07 mmHg
at 80 ◦C. Dry and deoxygenated solvents were used through-
out all syntheses. Toluene and pentane were distilled on
sodium/benzophenone and collected under nitrogen. The reac-
tions were carried out using usual Schlenk and glovebox method-
[cis,cis-C6H9(NCH2C6H4-o-PPh2)3-j5l2N]Al2Me3 (5)
In a glove box filled with nitrogen, 3 (300 mg, 0.32 mmol) and
trimethylaluminium (45 mg, 0.63 mmol) were mixed in 10 ml of
toluene. The Schlenk flask was sealed with a glass stopper and
the solution heated at 80 ◦C for 72 h after which the volatiles
were removed under vacuum. The crude product was dried under
vacuum at 40 ◦C for 2 h to assure the removal of all excess AlMe3.
In the glove box, the crude product was washed with several
portions of pentane to afford a white flaky powder (163 mg, 53%
yield). dH (400 MHz; C6D6) 8.09 (dd, 3JH–H = 7.1 and 3JH–P = 4.8,
1
ologies. 1H (400.0 MHz), 31P{H} (161.9 MHz) and 13C{ H}
(100.568 MHz) solution NMR spectra were recorded on a Varian
Inova NMR AS400 spectrometer or on a Bruker NMR AC-300
1
spectrometer (1H (300.0 MHz), 31P{H} (121.42 MHz) and 13C{ H}
3
3
1H), 7.63 (dd, JH–H = 7.7 and JH–P = 4.6, 2H), 7.38 (m, 15H),
3
3
(75.42 MHz). J values are given in Hz. The elemental analyses were
7.18–6.99 (m, 21H), 6.93 (dt, JH–H = 7.6 and JH–P = 1.1, 3H),
4.64 (dd, 4JH–P = 4.5 and 2JH–H = 14.2, 2H), 4.52 (d, 4JH–P = 2.8,
2H), 4.25 (d, 2JH–H = 14.2, 2H), 3.50 (s, 2H), 3.18 (d, 2JH–H = 15.2,
carried out by GCL & Chemisar Laboratories.
2
3
=
1H), 2.79 (s, 1H), 2.48 (d, JH–H = 13.8, 2H), 1.15 (td, JH–H
=
cis,cis-C6H9(N CHC6H4(SPh))3 (2)
2
2
3.6 and JH–H = 13.8, 2H), 1.04 (d, JH–H = 15.2, 1H), -0.12 (s,
Al–Me, 3H), -0.22 (s, Al–Me, 3H) and -0.52 (s, Al–Me, 3H); dP
(121.422 MHz; C6D6) -14.1 and -14.9; dC (100.57 MHz; C6D6)
149.3 (d, 1JC–P = 20.9), 144.0 (d, 1JC–P = 24.8), 138.0 (m), 137.3 (d,
3JC–P = 10.5), 137.2 (d, 3JC–P = 10.5), 136.0 (d, 3JC–P = 14.4), 134.3
(m), 133.1 (s), 129.7–129.1 (m), 126.6 (s), 54.7 (d, 5JC–P = 2.2), 51.5
cis,cis-1,3,5-Triaminocyclohexane (120 mg, 0.90 mmol) and 2-
(phenylthio)benzaldehyde (600 mg, 2.8 mmol) were dissolved in
˚
30 ml of anhydrous ethanol and molecular sieves (4 A) were added.
The solution was heated under reflux for 24 h and then cooled at
-35 ◦C. A white precipitate appeared which was filtered affording a
white crystalline powder (460 mg, 68% yield). dH (400 MHz; C6D6)
3
3
5
(d, JC–P = 24.0), 50.9 (s), 49.6 (d, JC–P = 22.8), 35.7 (d, JC–P
=
3
=
8.98 (s, 3H, N CH), 8.31 (br d, JH–H = 7.7, 3H, C6H4), 7.30 (br
5.2), 35.3 (s), -0.6 (s, Al–C), -5.2 (s, Al–C) and -11.5 (s, Al–C).
Elemental analysis calc. for C66H66Al2N3P3: C, 75.56; H, 6.30; N,
4.01. Found: C, 75.90; H, 6.58; N, 4.00%.
3
d, JH–H = 7.7, 3H, C6H4), 7.17 (m, 6H, o-SPh), 7.03–6.80 (m,
15H, m,p-SPh and C6H4), 3.15 (tt, 3JH–H = 11.4, 3JH–H = 3.5, 3H,
CHN CHC6H4), 2.22 (dd, JH–H = 11.8 and 3JH–H = 11.8, 3H, ax.
3
=
CH2) and 1.65 (td, 3JH–H = 11.8 and 3JH–H = 3.5, 3H, equ. CH2);
[cis,cis-C6H9(NCH2C6H4-o-SPh)3-j5l2N]Al2Me3 (6)
=
dC (100.57 MHz; C6D6) 157.3 (s, N CH), 138.0 (s, C6H4), 137.4
(s, C6H4), 135.2 (s, i-SPh), 134.3 (s, C6H4), 131.0 (s, p-SPh), 129.6
In a Schlenk flask with a Teflon stopper, 4 (420 mg, 0.58 mmol)
was dissolved in 10 ml toluene. 83 mg (1.2 mmol) of AlMe3 were
added and the solution heated at 100 ◦C for 24 h. The solution was
(s, m-SPh), 129.0 (s, C6H4), 128.5 (s, C6H4), 128.0 (s, o-SPh), 128.7
=
(s, C6H4), 66.7 (s, CH2) and 41.6 (s, CHN CHC6H4). Elemental
5694 | Dalton Trans., 2010, 39, 5688–5697
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
Table 2 Crystallographic information for 6
evaporated and dried under vacuum for 2 h. The resulting yellow
oil was then dissolved in 10 ml toluene and 64 mg (0.63 mmol)
of triethylamine was added. The reaction was stirred at ambient
temperature for 15 min after which the volatile materials were
removed. The crude product was washed with 5 portions of 2 ml of
pentane. Crystals were grown from slow evaporation of a pentane
solution (238 mg, 50% yield). dH (400 MHz; C6D6) 7.68 (dd,
6
Formula
Fw
C48H51Al2N3S3
820.06
0.09 ¥ 0.05 ¥ 0.02
Triclinic
Size/mm
Cryst syst
Space group
¯
P1
3JH–H = 7.6 and 3JH–H = 1.2, 1H), 7.55 (dd, 3JH–H = 7.7 and 3JH–H
=
˚
a/A
13.1455(11)
16.7733(14)
20.6807(17)
100.937(1)
99.373(1)
94.655(1)
4387.7(6)
4
0.71073
1.241
1736
200(2)
15420/44407
1.013
1.4, 2H), 7.37 (dd, 3JH–H = 7.8 and 3JH–H = 1.2, 1H), 7.31 (m, 4H),
˚
b/A
3
3
˚
7.19–7.14 (m, 4H), 7.05 (td, JH–H = 7.6 and JH–H = 1.4, 2H),
6.95 (m, 8H), 6.91–6.84 (m, 6H), 4.50 (d, 2JH–H = 13.9, 2H), 4.40
c/A
a/◦
b/◦
g /◦
V/A
Z
(s, 2H), 4.15 (d, 2JH–H = 13.9, 2H), 3.28 (s, 2H), 3.00 (dt, 2JH–H
=
15.2 and 3JH–H = 2.5, 1H), 2.72 (s, 1H), 2.43 (d, 2JH–H = 14.0, 2H),
1.19 (dt, 2JH–H = 14.0 and 3JH–H = 3.7, 2H), 0.94 (dt, 2JH–H = 15.2
and 3JH–H = 1.9 Hz, 1H), -0.15 (s, Al–Me, 3H), -0.26 (s, Al–Me,
3H) and -0.32 (s, Al–Me, 3H); dC (100.57 MHz; C6D6) 145.8 (s),
141.2 (s), 137.6 (s), 137.2 (s), 135.5 (s), 134.8 (s), 134.3 (s), 133.0 (s),
131.3 (s), 130.7 (s), 129.8 (s), 129.5 (s), 129.5 (s), 129.4 (s), 128.8 (s),
128.8 (s), 127.5 (s), 127.2 (s), 126.7 (s), 126.5 (s), 54.5 (s), 52.0 (s),
50.5 (s), 49.3 (s), 35.4 (s), 35.1 (s), -0.8 (s, AlMe), -5.0 (s, AlMe)
and -11.3 (s, AlMe). Elemental analysis calc. for C48H51Al2N3S3:
C, 70.23; H, 6.22; N, 5.12. Found: C, 69.80; H, 6.59;
N, 5.50%.
3
˚
˚
Wavelength/A
Dc/g cm-3
F000
T/K
No. of unique/total reflns
GOF
Rint
0.0301
0.0391
Final R indices [I > 2s(I)]
were removed under vacuum to afford an off-white powder (NMR
yield > 98%). Small impurities prevented from doing an elemental
3
analysis. dH (400 MHz; C6D6) 8.00 (d, JH–H = 7.3, 1H), 7.85 (d,
[cis,cis-C6H9(NCH2C6H4-o-PPh2-jN)2(NHCH2C6H4-o-PPh2-
jN)]AlMe (7)
3JH–H = 7.0, 1H), 7.39–7.45 (m, 25H), 4.62 (d, 2JH–H = 15.2, 1H),
2
2
4.50 (d, JH–H = 15.3, 1H), 4.32 (d, JH–H = 14.8, 1H), 4.26 (d,
2JH–H = 15.1, 1H), 4.10 (dd, JH–H = 14.0 and JH–H = 5.2, 1H),
3.81 (dd, 2JH–H = 14.1 and 3JH–H = 9.5, 1H), 3.62 (dd, 3JH–H = 8.0
and 3JH–H = 6.3, 1H, N–H), 3.01 (br, 1H), 2.90 (br, 1H), 2.80 (br,
2
3
In a glove box filled with nitrogen, 3 (140 mg, 0.15 mmol) and
AlMe3 (11 mg, 0.15 mmol) were dissolved in 2 ml of toluene. The
solution was heated at 100 ◦C for 48 h. The volatiles were removed
under vacuum to afford a white powder (NMR yield > 98%).
Small impurities prevented from doing an elemental analysis. dH
(400 MHz; C6D6) 8.18 (dd, 3JH–H = 7.2 and 3JH–P = 4.5, 1H), 8.12
3
3
1H), 2.49 (d, JH–H = 12.7, 1H), 2.28 (d, JH–H = 14.5, 1H), 1.48
3
3
3
(tt, JH–H = 12.2 and JH–H = 3.6 Hz, 1H), 1.45 (d, JH–H = 15.0,
1H), 1.24 (dt, 3JH–H = 13.8 and 3JH–H = 3.1, 1H), 0.97 (dt, 3JH–H
=
14.3 and 3JH–H = 3.0, 1H), -0.82 (s, 3H, AlMe); dC (100.57 MHz;
C6D6) 147.0 (s), 146.6 (s), 143.6 (s), 138.2 (s), 137.9 (s), 136.5 (s),
136.0 (s), 134.6 (s), 134.1 (s), 133.3 (s), 133.0 (s), 132.8 (s), 130.6
(s), 130.1 (s), 129.6 (s), 129.3 (s), 129.1 (s), 128.7 (s), 127.2 (s), 127.1
(s), 126.4 (s), 126.2 (s), 53.8 (s), 53.7 (s), 53.4 (s), 53.3 (s), 53.2 (s),
53.0 (s), 49.7 (s), 49.4 (s), 41.6 (s), 37.9 (s), 37.3 (s), 29.4 (s), -15.7
(s). Some aromatic signals are hidden beneath the solvent peak.
3
3
(dd, JH–H = 7.7 and JH–P = 4.4, 1H), 7.47–6.88 (m, 40H), 4.80
(dd, 2JH–H = 15.3 and 4JH–P = 3.3, 1H), 4.57 (d, 2JH–H = 15.3, 1H),
4.42 (dd, 2JH–H = 15.2 and4JH–P = 1.7, 1H), 4.31 (dd, 2JH–H = 15.3
and 4JH–P = 2.9, 1H), 4.27 (dd, 2JH–H = 11.3 and 3JH–H = 4.7, 1H),
3.92 (dd, 2JH–H = 14.1 and 3JH–H = 9.8, 1H), 3.56 (dd, 3JH–H = 14.9
and 3JH–H = 7.1, 1H, N–H), 3.05 (br, 1H), 2.92 (br, 1H), 2.86 (br,
3
3
1H), 2.45 (d, JH–H = 12.4, 1H), 2.33 (d, JH–H = 14.1, 1H), 1.48
(tt, 3JH–H = 12.2 and 3JH–H = 3.6, 1H), 1.45 (d, 3JH–H = 15.0, 1H),
1.24 (dt, 3JH–H = 13.8 and 3JH–H = 3.1, 1H), 0.97 (dt, 3JH–H = 14.3
and 3JH–H = 3.0, 1H) and -0.87 (s, 3H, AlMe); dP (121.422 MHz;
Crystallographic structural determination†
Crystallographic data are reported in Table 2. Single crystals were
coated with Paratone-N oil, mounted using a glass fibre and
frozen in the cold nitrogen stream of the goniometer. The data
were collected on a Bruker SMART APEX II diffractometer.
The data were reduced (SAINT)29 and corrected for absorp-
tion (SADABS).30 The structure was solved and refined using
SHELXS-97 and SHELXL-97.31 All non-H atoms were refined
anisotropically. The hydrogen atoms were placed at idealized
positions. Neutral atom scattering factors were taken from the
International Tables for X-Ray Crystallography.32 All calculations
and drawings were performed using the SHELXTL package.33
C6D6) -13.8 and -17.0; dC (100.57 MHz; C6D6) 150.0 (d, 1JC–P
=
1
1
20.9), 149.8 (d, JC–P = 20.9), 139.7 (d, JC–P = 25.0), 138.7 (d,
2JC–P = 12.5), 138.6 (d, 2JC–P = 11.8), 138.3 (d, 1JC–P = 11.7), 138.3
(d, 1JC–P = 12.0), 137.2 (d, 2JC–P = 13.9), 135.1 (s), 134.4 (s), 133.3
(s), 133.3 (s), 132.3 (d, 3JC–P = 5.8), 129.9 (s), 129.4 (s), 129.1 (s),
128.8 (s), 126.6 (s), 53.4 (s), 53.3 (s), 53.2 (d, 3JC–P = 25.0), 53.0 (s),
52.3 (d, 3JC–P = 23.0), 49.7 (d, 3JC–P = 16.0), 38.1 (s), 36.6 (s), 29.1
(s) and -15.8 (s, AlMe). Some aromatic signals are hidden beneath
the solvent peak.
[cis,cis-C6H9(NCH2C6H4-o-SPh-jN)2(NHCH2C6H4-o-SPh-
jN)]AlMe (8)
Polymerization procedures. The reaction mixtures were pre-
pared in a glove box. In a Schlenk tube with a magnetic stir bar,
a solution of the catalyst in toluene was added to the monomer
in toluene in order to obtain a 2.4 mM solution, unless stated
otherwise. The solutions were stirred for a specific period of time
In a glove box filled with nitrogen, compound 4 (140 mg,
0.15 mmol) and AlMe3 (11 mg, 0.15 mmol) were dissolved in 2 ml
of toluene. The solution was heated at 100 ◦C for 48 h. The volatiles
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 5688–5697 | 5695
©

View Article Online
at room temperature or at 50 ◦C. The reaction was quenched
by adding 1 mL of dichloromethane and pouring the solution
in cold methanol. The precipitate was filtered and dried under
vacuum. A fine suspension of the polymer in methanol could also
be centrifuged at 7000 rpm at 4 ◦C for 30 min.
Polymerization of rac-lactide was performed in a similar fashion
in 5 ml of toluene and heated at 100 ◦C for four days. The reaction
was quenched by pouring the solution into slightly acidic cold
methanol at pH 5. No significant reactivity was observed with
AlMe3 and CL. The precipitate was filtered and dried under
vacuum.
X. Chen, L. Jiang and X. Jing, Polymer, 2003, 44, 2331–2336; (c) S.-M.
Ho, C.-S. Hsiao, A. Datta, C.-H. Hung, L.-C. Chang, T.-Y. Lee and
J.-H. Huang, Inorg. Chem., 2009, 48, 8004–8011.
7 Some leading references: (a) R. R. Gowda, D. Chakraborty and V.
Ramkumar, Eur. J. Inorg. Chem., 2009, 2981–2993; (b) L. Postigo, J.
Sanchez-Nieves, P. Royo and M. E. G. Mosquera, Dalton Trans., 2009,
3756–3765; (c) F. Gornshtein, M. Kapon, M. Botoshansky and M. S.
Eisen, Organometallics, 2007, 26, 497–507.
8 Some leading references: (a) M.-Z. Chen, H.-M. Sun, W.-F. Li, Z.-G.
Wang, Q. Shen and Y. Zhang, J. Organomet. Chem., 2006, 691, 2489–
2494; (b) V. C. Gibson, E. L. Marshall, D. Navarro-Llobet, A. J. P.
White and D. J. Williams, J. Chem. Soc., Dalton Trans., 2002, 4321–
4322; (c) B. J. O’Keefe, L. E. Breyfogle, M. A. Hillmyer and W. B.
Tolman, J. Am. Chem. Soc., 2002, 124, 4384–4393.
9 Some leading references: (a) M. Cheng, A. B. Attygalle, E. B.
Lobkovsky and G. W. Coates, J. Am. Chem. Soc., 1999, 121, 11583–
11584; (b) C. Zhang and Z.-X. Wang, J. Organomet. Chem., 2008,
693, 3151–3158; (c) C. A. Wheaton, B. J. Ireland and P. G. Hayes,
Organometallics, 2009, 28, 1282–1285; (d) Z. Zheng, G. Zhao, R. Fablet,
M. Bouyahyi, C. M. Thomas, T. Roisnel, O. Casagrande Jr. and J.-F.
Carpentier, New J. Chem., 2008, 32, 2279–2291.
SEC experiments. The SEC measurements were performed in
chloroform on a Waters apparatus with 515 HPLC pumps and a
Waters Associates 441 detector and two Jordi columns. Mn and
Mw were calculated using polystyrene standards by integrating the
domain corresponding to the larger polymers.
10 Some leading references: (a) N. Nimitsiriwat, V. C. Gibson, E. L.
Marshall and M. R. J. Elsegood, Inorg. Chem., 2008, 47, 5417–5424;
(b) A. P. Dove, V. C. Gibson, E. L. Marshall, H. S. Rzepa, A. J. P.
White and D. J. Williams, J. Am. Chem. Soc., 2006, 128, 9834–9843;
(c) A. Kowalski, J. Libiszowski, K. Majerska, A. Duda and S. Penczek,
Polymer, 2007, 48, 3952–3960; (d) G. Jiang, I. A. Jones, C. D. Rudd
and G. S. Walker, J. Appl. Polym. Sci., 2009, 114, 658–662; (e) P. Dege´e,
P. Dubois, R. Jeroˆme, S. Jacobsen and H.-G. Fritz, Macromol. Symp.,
1999, 144, 289–302.
11 Some leading references: (a) M. H. Chisholm, J. C. Gallucci, K. T.
Quisenberry and Z. Zhou, Inorg. Chem., 2008, 47, 2613–2624; (b) H.
Du, A. H. Velders, P. J. Dijkstra, J. Sun, Z. Zhong, X. Chen and J. Feijen,
Chem.–Eur. J., 2009, 15, 9836–9845; (c) A. Arbaoui, C. Redshaw and
D. L. Hughes, Chem. Commun., 2008, 4717–4719; (d) M. Haddad, M.
Laghzaoui, R. Welter and S. Dagorne, Organometallics, 2009, 28, 4584–
4592; (e) S. Milione, F. Grisi, R. Centore and A. Tuzi, Organometallics,
2006, 25, 266–274.
MALDI-TOF experiments. The experiments were carried out
at the Mass Spectrometry service at the Durham University Chem-
istry department. The sample was prepared by first solubilizing a
small portion (~1 mg) in acetonitrile (1 mL). A ten fold dilution
of this was made in a solution MALDI matrix (saturated a-
cyano-4-hydroxy cinnamic acid prepared in 0.1% trifluoroacetic
acid : acetonitrile (2 : 1)). 1 mL of this was spotted onto a ground
steel target prior to running on the MALDI ToF instrument
(Autoflex MALDI ToF/ToF, Bruker, Coventry, UK).
Acknowledgements
F.-G. Fontaine is grateful to NSERC (Canada), CFI (Canada),
FQRNT (Que´bec), CCVC (Que´bec), and CERPIC (Univer-
site´ Laval) for financial support. M.-H. Thibault acknowledges
FQRNT for a scholarship. We acknowledge Prof. F. H. Schaper
and Prof. P. Hayes for helpful discussions. We acknowledge Prof.
S. A. Westcott, J. Boudreau, S. S. Barnes, and B. Macha for their
thoughtful input.
12 (a) S. Range, D. F.-J. Piesik and S. Harder, Eur. J. Inorg. Chem., 2008,
3442–3451; (b) D. F.-J. Piesik, R. Stadler, S. Range and S. Harder,
Eur. J. Inorg. Chem., 2009, 3569–3576; (c) Z. Zhang, X. Xu, S. Sun, Y.
Yao, Y. Zhang and Q. Shen, Chem. Commun., 2009, 7414–7416.
13 C. K. Williams, R. B. Brooks, M. A. Hillmayer and W. B. Tolman,
Chem. Commun., 2002, 3132–3133.
14 A. Arbaoui, C. Redshaw and D. L. Huges, Chem. Commun., 2008,
4717–4719.
15 (a) F.-G. Fontaine and D. Zargarian, J. Am. Chem. Soc., 2004, 126,
8786–8794; (b) M.-H. Thibault, J. Boudreau, S. Mathiotte, F. Drouin,
A. Michaud, O. Sigouin and F.-G. Fontaine, Organometallics, 2007,
26, 3807–3815; (c) F.-G. Fontaine, J. Boudreau and M.-H. Thibault,
Eur. J. Inorg. Chem., 2008, 5439–5454.
16 Some leading references: (a) J. E. Bollinger, J. T. Mague, W. A. Banks,
A. J. Kastin and D. M. Roundhill, Inorg. Chem., 1995, 34, 2143–2152;
(b) L. Turculet and T. D. Tilley, Organometallics, 2004, 23, 1542–1553;
(c) Y. Kajita, H. Arii, Y. Saito, S. Nagatomo, T. Kitagawa, Y. Fanahashi,
T. Ozawa and H. Masuda, Inorg. Chem., 2007, 46, 3322–3335.
17 M.-H. Thibault, B. E. G. Lucier, R. W. Schurko and F.-G. Fontaine,
Dalton Trans., 2009, 7701–7716.
18 (a) W. Yao, Y. Mu, A. Gao and L. Ye, Dalton Trans., 2008, 3199–
3206; (b) H. Han and S. A. Johnson, Organometallics, 2006, 25, 5594–
5602; (c) M. Garcia-Castro, A. Martin, M. Mena and C. Ye´lamos,
Organometallics, 2007, 26, 408–416.
19 H. Zhu and E. Y.-X. Chen, Organometallics, 2007, 26, 5395–
5405.
20 (a) T. Biela, A. Kowalski, J. Libiszowski, A. Duda and S. Penczek,
Macromol. Symp., 2006, 240, 47–55; (b) N. Wasa, S. Katao, J. Liu, M.
Fujiki, Y. Furukawa and K. Nomura, Organometallics, 2009, 28, 2179–
2187; (c) C. Zhang and Z. X. Wang, Appl. Organomet. Chem., 2009, 23,
9–18.
21 One of the referee suggested that the jellification of the solution was
certainly due to transesterification reactions. Although it was not
possible to confirm it, it is a likely reason for such change in the viscosity
of the solution.
References
1 (a) C. S. Bajgur and S. Sivaram, Curr. Science, 2000, 78, 1325–1335;
(b) H. Hagen, J. Boersma and G. van Koten, Chem. Soc. Rev., 2002, 31,
357–364 and references therein.
2 (a) C. J. Moore, S. L. Moore, M. K. Leecaster and S. B. Weisberg, Mar.
Pollut. Bull., 2001, 42, 1297–1300; (b) R. C. Thompson, Y. Olsen, R. P.
Mitchell, A. Davis, S. J. Rowland, A. W. G. John, D. McGonigle and
A. E. Russell, Science, 2004, 304, 838.
3 (a) Y. Ikada and H. Tsuji, Macromol. Rapid Commun., 2000, 21, 117–
132; (b) S. Mecking, Angew. Chem., Int. Ed., 2004, 43, 1078–1085.
4 (a) O. Dechy-Cabaret, B. Martin-Vaca and D. Bourissou, Chem. Rev.,
2004, 104, 6147–6176; (b) J. Wu, T.-L. Yu, C.-T. Chen and C.-C. Lin,
Coord. Chem. Rev., 2006, 250, 602–626; (c) C. K. Williams, Chem. Soc.
Rev., 2007, 36, 1573–1580; (d) C. A. Wheaton, P. G. Hayes and B. J.
Ireland, Dalton Trans., 2009, 4832–4846.
5 Some leading references: (a) B. M. Chamberlain, M. Cheng, D. R.
Moore, T. M. Ovitt, E. B. Lobkovsky and G. W. Coates, J. Am. Chem.
Soc., 2001, 123, 3229–3238; (b) M. H. Chisholm, J. C. Gallucci and
K. Phomphrai, Inorg. Chem., 2005, 44, 8004–8010; (c) V. Poirier, T.
Roisnel, J.-F. Carpentier and Y. Sarazin, Dalton Trans., 2009, 9820–
9827; (d) L. F. Sanchez-Barba, A. Garces, M. Fajardo, C. Alonso-
Moreno, J. Fernandez-Baeza, A. Otero, A. Antinolo, J. Tejeda, A. Lara-
Sanchez and M. I. Lopez-Solera, Organometallics, 2007, 26, 6403–6411;
(e) M.-Y. Shen, Y.-L. Peng, W.-C. Hung and C.-C. Lin, Dalton Trans.,
2009, 9906–9913.
22 (a) A. Duda, Z. Florjanczyk, A. Hofman, S. Slomkowski and S.
6 Some leading references: (a) M. H. Chisholm, J. C. Gallucci and K.
Phomphrai, Inorg. Chem., 2004, 43, 6717–6725; (b) L. Piao, M. Deng,
Penczek, Macromolecules, 1990, 23, 1640–1646; (b) G. Deshayes,
5696 | Dalton Trans., 2010, 39, 5688–5697
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
F. A. G. Mercier, P. Dege´e, I. Verbruggen, M. Biesemans, R. Willem
and P. Dubois, Chem.–Eur. J., 2003, 9, 4346–4352.
23 S. Milione, F. Grisi, R. Centore and A. Tuzi, Organometallics, 2006, 25,
266–274.
27 L. Cronin, S. P. Foxon, P. J. Lusby and P. H. Walton, JBIC, J. Biol.
Inorg. Chem., 2001, 6, 367–377.
28 S. Laue, L. Greiner, J. Wo¨tlinger and A. Liese, Adv. Synth. Catal., 2001,
343, 711–720.
24 C. N. McMahon, S. G. Bott and A. R. Barron, J. Chem. Soc., Dalton
29 SAINT Version 7.07a; Bruker AXS Inc.: Madison, WI, 2003. G. M.
Trans., 1998, 3301–3304.
Sheldrick.
25 K. A. M. Thakur, R. T. Kean, E. S. Hall, J. J. Kolstad, T. A. Lindgren,
A. M. Doscotch, J. I. Siepmann and E. J. Munson, Macromolecules,
1997, 30, 2422–2428.
26 (a) M. Bouyahyi, E. Grunova, N. Marquet, E. Kirillov, C. M. Thomas,
T. Roisnel and J.-F. Carpentier, Organometallics, 2008, 27, 5815–
5825; (b) Z. Tang and V. C. Gibson, Eur. Polym. J., 2007, 43, 150–
155.
30 SADABS Version 2004/1; Bruker AXS Inc.: Madison, WI, 2004.
31 G. M. Sheldrick, SHELXS-97 and SHELXL-97. Programs for the
refinement of crystal structuresUniversity of Gottingen: Germany, 1997.
32 International Tables for Crystallography, Vol C., A. J. C. Wilson, Ed.
Kluwer Academic Publishers: Dordecht, 1992, pp 219–222 and pp.
500–502.
33 SHELXTL. Version, 6.12; Bruker AXS: Madison, WI, 2001.
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 5688–5697 | 5697
©