Lanthanide chloride complexes of amine-bis(phenolate) ligands and their reactivity in the ring-opening polymerization of ε-caprolactone

Article abstract of DOI:10.1039/b802219d

Reaction of two equivalents of n-BuLi with sterically demanding amine-bis(phenol) compounds, H₂O₂NN′^R (Me₂NCH₂CH₂N{CH₂-3,5-R ₂-C₆H₂OH}₂; R = t-Bu or t-Pe (tert-pentyl)) yields isolable lithium complexes, Li₂(O ₂NN′^R), in good yields. Upon reaction with one equivalent of LnCl₃(THF)_x, the lithium salts afford rare earth amine-phenolate chloride complexes in good yields, Ln(O ₂NN′^R)Cl(THF); Ln = Y, Yb, Ho, Gd, Sm, Pr. Crystals of Y(O₂NN′^t-Bu)Cl(THF), 1, and Sm(O ₂NN′^t-Bu)Cl(DME), 2, suitable for single crystal X-ray crystallographic analysis were obtained. In contrast to previously reported [{Gd(O₂NN′^t-Pe)(THF)(μ-Cl)}₂] and related La and Sm complexes, these species are monomeric. 1 contains Y in a distorted octahedral environment bonded to two amine, two phenolate, one THF and one chloride donor. 2 contains Sm in a distorted capped trigonal prismatic environment bonded to two amine, two phenolate, two DME oxygens and one chloride donor. The Ln(O₂NN′^t-Pe)Cl(THF) complexes were active initators for the controlled ring-opening polymerization of ε-caprolactone with a tendency to form low molecular weight cyclic polyesters (M_n 3000-5000). The conversion rates, although slower than related amido and alkyl species, were different for monomeric and dimeric initiators. The size of the metal centre also affected the conversions and the molecular weights achieved.

Full text of DOI:10.1039/b802219d

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/dalton | Dalton Transactions
Lanthanide chloride complexes of amine-bis(phenolate) ligands and their
reactivity in the ring-opening polymerization of e-caprolactone†
Charlotte E. Willans,a Mikhail A. Sinenkov,b Georgy K. Fukin,b Kristina Sheridan,c Jason M. Lynam,a
Alexander A. Trifonov*b and Francesca M. Kerton*c
Received 8th February 2008, Accepted 16th April 2008
First published as an Advance Article on the web 20th May 2008
DOI: 10.1039/b802219d
Reaction of two equivalents of n-BuLi with sterically demanding amine-bis(phenol) compounds,
H2O2NNꢀR (Me2NCH2CH2N{CH2-3,5-R2-C6H2OH}2; R = t-Bu or t-Pe (tert-pentyl)) yields isolable
lithium complexes, Li2(O2NNꢀR), in good yields. Upon reaction with one equivalent of LnCl3(THF)x,
the lithium salts afford rare earth amine-phenolate chloride complexes in good yields,
Ln(O2NNꢀR)Cl(THF); Ln = Y, Yb, Ho, Gd, Sm, Pr. Crystals of Y(O2NNꢀt-Bu)Cl(THF), 1, and
Sm(O2NNꢀt-Bu)Cl(DME), 2, suitable for single crystal X-ray crystallographic analysis were obtained. In
contrast to previously reported [{Gd(O2NNꢀt-Pe)(THF)(l-Cl)}2] and related La and Sm complexes, these
species are monomeric. 1 contains Y in a distorted octahedral environment bonded to two amine, two
phenolate, one THF and one chloride donor. 2 contains Sm in a distorted capped trigonal prismatic
environment bonded to two amine, two phenolate, two DME oxygens and one chloride donor. The
Ln(O2NNꢀt-Pe)Cl(THF) complexes were active initators for the controlled ring-opening polymerization
of e-caprolactone with a tendency to form low molecular weight cyclic polyesters (Mn 3000–5000). The
conversion rates, although slower than related amido and alkyl species, were different for monomeric
and dimeric initiators. The size of the metal centre also affected the conversions and the molecular
weights achieved.
earth systems containing sterically-demanding ligands have been
developed and studied for the controlled ROP of e-CL and lac-
Introduction
tide, including bis(phosphanyl)amide,8 iminophenolato,9 pyridyl
amido,10 pyridylmethyl indenyl,11 bis(phenolate),12 guanidino,13
guanidinate,14 or amine(phenolate),15 and diamide-diamine
ligands.16 It has been shown through this extensive research that
the behaviour of lanthanide complexes for the living ROP of e-CL
can be tuned by altering the ancillary ligands attached to the metal
centre. It has also been shown that the size of the central metal
ion in these lanthanide based initiator systems also plays a crucial
role in their design.17,18
Important questions about the mechanism of polymerization
still need to be answered in order to clarify the relationships
between the structure and activity of well-defined initiators so
that improvements can be made. When a series of heteroleptic and
homoleptic 2,5-bis(N-aryliminomethyl) pyrrolyl complexes of Y
were prepared,19 the heteroleptic mono(pyrrolyl) and bis(pyrrolyl)
complexes were found to be initiators for ROP of e-CL, giving
polyesters with moderate molecular weight distributions. The
polymer made using the mono(pyrrolyl) complex, which has two
amido groups, was inferior in its polydispersity (Mw/Mn = 2.0)
to that made by the bis(pyrrolyl) complex (Mw/Mn = 1.3), with
one amido group, indicating that the number of Y–N(SiMe3)2
bonds significantly affects molecular weight distribution. The
bis(pyrrolyl) complex acted as a single site initiator. The homolep-
tic tris(pyrrolyl) complexes containing no amido groups were inac-
tive in the ROP of e-CL. Therefore, the balance between reactive
bonds and inert bonds within a complex needs to be properly
understood in the development of new initiators. After our initial
studies in this area,18 we were concerned about the reactivity of
The controlled ring-opening polymerization (ROP) of lactones
has been extensively studied in recent years, particularly as these
reactions yield biodegradable polymers. Various metal complexes
such as aluminium,1 titanium,2 tin,3 zinc4 and magnesium5 have
been used as initiators for ROP of e-caprolactone (e-CL). In
1992, McLain and Drysdale showed that the polymerization
of e-caprolactone initiated by rare earth alkoxides had some
living character, but the molecular weight distribution of poly(e-
caprolactone) got broader after the monomer was completely
consumed, which suggested transesterification to be occurring.6
These transesterifications are rather common during the polymer-
ization of e-CL by metal alkoxides, and the rate at which they
occur relates to both the nature of the metal ion and the groups
surrounding the ion. Subsequently, it was shown that sterically-
demanding groups attached to a rare earth ion can prevent poly(e-
caprolactone) chains from co-coordinating to the ion and, there-
fore, minimise transesterification reactions.7 Since then, many rare
aDepartment of Chemistry, University of York, Heslington, York, UK YO10
5DD
bG. A. Razuvaev Institute of Organometallic Chemistry of Russian Academy
of Sciences, Tropina str. 49, 603950, Nizhny Novgorod, Russia. E-mail: trif@
imoc.sinn.ru
cDepartment of Chemistry, Memorial University of Newfoundland, St.
John’s, NL A1B 3X7, Canada. E-mail: fkerton@mun.ca; Fax: +1-709-
7373702
† Electronic supplementary information (ESI) available: 1H NMR spectra
of selected polymer samples. CCDC reference numbers 676933 and 676934.
For crystallographic data in CIF or other electronic format see DOI:
10.1039/b802219d
3592 | Dalton Trans., 2008, 3592–3598
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
Table 1 Details of crystallographic data and refinements for 1 and 2
Ln–phenolate bonds towards lactone monomers, especially as it
had already been shown that shown Ln–O containing complexes,
such as alkoxide based initiators formed in situ,20 are also active
species. Indeed, Visseaux and co-workers have provided evidence
of Ln–O bonds being active in the polymerization process and
proposed that Ln–N bonds are converted to such species by
reaction with e-CL.21 In that work, it was proposed that –
N(SiMe3)2 extracts a proton in the aposition of the carbonyl group
of e-CL causing the alkoxide–Sm species to precipitate, with the
Sm–O bond being active in the polymerization process.
In this paper, we present the synthesis and e-CL ROP activity of
amine-bis(phenolate) lanthanide chloride complexes (see Fig. 1
for the ligands used in this work). These results indicate that
although Ln–phenolate and Ln–amine bonds are relatively inert—
compared to Ln–NR2 and Ln–OR bonds in amine-bis(phenolate)
lanthanide amido and alkoxide complexes—they are not com-
pletely unreactive and may contribute low molecular weight cyclic
polyester by-products in ROP reactions.
1
2·0.5(C7H8)
Chemical formula
Formula weight
T/K
C38H62ClN2O3Y
719.26
100(2)
Monoclinic
C2/c
48.516(1)
11.9131(3)
35.937(1)
130.934(1)
16
C41.50H68ClN2O4Sm
844.78
100(2)
Monoclinic
P2(1)/c
12.6169(4)
19.9748(6)
17.6003(5)
94.044(1)
4
Crystal system
Space group
˚
a/A
˚
b/A
˚
c/A
b/◦
Z
Dc/Mg m−3
l/mm−1
1.218
1.590
1.268
1.425
F(000)
6144
1.68/26.00
46613
15386
0.0436
0.0398
0.0820
0.981
0.531 and −0.364
1768
1.91/25.00
24314
7752
0.0224
0.0259
0.0636
1.009
1.367 and −0.523
h Range for collection/◦
No. of reflns colld
No. of indep. reflns
R(int)
R[I>2r(I)]
wR
GOOF on F2
Largest diff. peak and
hole/e A
−3
˚
of 1 and 2, Fig. 2 and 3, reveal monomeric species with the
rare earth metal centres in distorted octahedral and capped
trigonal prismatic environments, respectively. The Y–O bond
lengths in 1 (2.1360(10), 2.1467(10) and 2.1300(10), 2.1492(10)
Fig. 1 Protio ligands used in this work (t-Pe = tert-pentyl).
Results and discussion
˚
A) are close to the values previously reported for six-coordinate
bis(phenolate),23a,b aminophenolate,23c and salicylaldiminate23d,e
yttrium complexes. The O–Y–O bond angle in 1 (152.88(3) and
153.24(3)◦) is slightly larger than that in a related six-coordinate
alkylyttrium compound (150.94(7)◦).15d In general, the structure
of 1 is similar to other monomeric Y complexes containing these or
similar ligands,15c,d and the Sc complex, Sc(O2NPyt-Bu)(Cl)(py).23c
1 is isostructural with the previously reported Yb and Er
analogues.15f As expected the Ln–phenoxide bonds are marginally
longer in 1 due to the larger ionic radius of Y3+ compared with
Yb3+ and Er3+.
The monomeric structure of 2 is somewhat more surpris-
ing and is in contrast to the dimeric structure of the related
Gd complex, [{Gd(O2NNꢀt-Pe)(THF)(l-Cl)}2],18 the La complex,
[{La(O2NPyt-Bu)(l-Cl)}2] and Sm complex, [{Sm(O2NPyt-Bu)(l-
Cl)}2].15b Although, the use of a dibasic co-solvent, DME, in the
recrystallization process and the resultant formation of a seven-
coordinate Sm centre may have aided in its isolation. The Sm–O
Syntheses and structures of Ln–Cl complexes
We have previously reported the synthesis of lithium
diamine-bis(phenolate) complexes and used these to prepare
Y(O2NNꢀt-Bu)Cl(THF) by a salt metathesis route.22 This approach
can be extended to other rare earth metal centres, namely La, Pr,
Sm, Gd, Ho and Yb (Scheme 1), and Ln(O2NNꢀR)Cl(THF) species
can be isolated as pale coloured powders in moderate yields. Yb
and Er species have previously been prepared using sodium-ligand
complexes in the salt metathesis reaction.15f
Crystals of Y(O2NNꢀt-Bu)Cl(THF), 1, and Sm(O2NNꢀt-Bu)-
Cl(DME), 2, suitable for X-ray crystallographic studies were
obtained by slow concentration of toluene solutions at room
temperature. Complex 1 crystallizes in such a way that two
crystallographically distinct molecules are present in the asymmet-
ric unit cell, whereas complex 2 crystallizes as the toluene solvate
Sm(O2NNꢀt-Bu)Cl(DME)·0.5(C7H8). The geometric parameters of
the independent molecules of 1 are very similar. Crystal data
for 1 and 2 are shown in Table 1. The molecular structures
˚
bond distances in 2 (2.1819(13), 2.2113(14) A) fall into the interval
normally observed for seven-coordinate samarium phenoxides.23f ,g
Scheme 1 Synthesis of chlorobis(phenolate)diamine complexes of lanthanides.
The Royal Society of Chemistry 2008 Dalton Trans., 2008, 3592–3598 | 3593
This journal is
©

View Article Online
Fig. 2 Molecular structure of Y(O2NNꢀt-Bu)Cl(THF), 1, H atoms are
omitted for clarity. Two molecules are present in the asymmetric unit
cell, molecule A is shown. Thermal ellipsoids are drawn at the 20%
◦
˚
probability level. Selected bond lengths (A) and angles ( ) for both
molecules: Y(1A)–O(2A) 2.136(1), Y(1A)–O(3A) 2.147(1), Y(1A)–O(1A)
2.343(1), Y(1A)–N(1A) 2.515(1), Y(1A)–N(2A) 2.552(1), Y(1A)–Cl(1A)
2.5725(3), O(2A)–Y(1A)–O(3A) 152.88(3), N(1A)–Y(1A)–N(2A)
Fig. 3 Molecular structure of Sm(O2NNꢀt-Bu)Cl(DME), 2, H atoms are
omitted for clarity. Thermal ellipsoids are drawn at the 20% prob-
◦
˚
ability level. Selected bond lengths (A) and angles ( ): Sm(1)–O(2)
70.61(4);
Y(1B)–O(2B)
2.1300(10),
Y(1B)–O(3B)
2.1492(10),
2.182(1), Sm(1)–O(1) 2.211(1), Sm(1)–O(4) 2.546(1), Sm(1)–N(1) 2.624(2),
Sm(1)–O(3) 2.629(1), Sm(1)–Cl(1) 2.6792(5), Sm(1)–N(2) 2.685(2),
O(2)–Sm(1)–O(1) 150.34(5), N(1)–Sm(1)–N(2) 68.10(5).
Y(1B)–O(1B) 2.3343(12), Y(1B)–N(1B) 2.5216(10), Y(1B)–N(2B)
2.5437(4), Y(1B)–Cl(1B) 2.5769(3), O(2B)–Y(1B)–O(3B) 153.24(3),
N(1B)–Y(1B)–N(2B) 70.59(3).
represented a range of metal sizes across the lanthanide series,
As far as we are aware, this is the first monomeric samarium
chloride amine-bis(phenolate) complex reported.
and were used in combination with ligand (O2NNꢀt-Pe)2−. This t-
Pe substituted ligand was chosen in preference to (O2NNꢀt-Bu 2− for
)
these reactions as slightly narrower molecular weight distributions
were afforded by complexes bearing this ligand in our previous
study.18 The reaction time was increased from 1 h to 24 h (Table 2).
These data contrast with results obtained for the related Sm
species, [{Sm(O2NPyt-Bu)(l-Cl)}2], which did not yield any polymer
after 16 h reaction time.15b This shows that small changes to the
amine-bis(phenolate) ligand can result in quite significant changes
in reactivity. No attempts were made to investigate ROP of lactide
using the chloride complexes that we have reported here.
There are a number of mechanisms by which these polymeriza-
tions could be proceeding. One possibility is via insertion of the
monomer into the metal–phenoxide bond, as has been proposed
for Sm tris(phenolate) species.28 In previous studies of ROP using
amine-bis(phenolate) Ln species, the Ln–phenolate bonds were
considered inert and the monomer only inserted into the metal
Reactivity of Ln–Cl complexes towards e-caprolactone
At first, we thought that the lanthanide–chloride complexes,
Ln(O2NNꢀR)Cl(THF), would not initiate the polymerization of
e-CL, as using the same conditions as those employed with
lanthanide–amido initiators, polymerization did not occur.18 In
several previously reported studies,15,18,23 these and related com-
plexes appear to be fluxional in solution between dimeric and
monomeric forms, possibly resulting in there being no free coordi-
nation sites for the incoming ester to coordinate. However, during
attempts to isolate Ln–caprolactone adducts, it was discovered
that over a longer reaction time, ROP of e-CL could be brought
about using these complexes as initiators. Therefore, a series
of polymerization reactions were conducted in parallel, using
stirred vials in a nitrogen filled glove box. The chosen lanthanides
Table 2 Polymerization of e-caprolactone initiated by lanthanide–chloride complexes, Ln(O2NNꢀt-Pe)Cl(THF)
b
b
Metal
% Conversiona
Mw/Mn
Mn
Pr
Sm
Gd
Ho
Y
Yb
91
92
1.3
1.9
1.2
1.3
1.6
1.1
1.2
1.7
1.2
n/a
n/a
7390 (4138)
8130 (4550)
7720 (4323)
7540 (4222)
7890 (4418)
5200 (2912)
79 300 (44,408)
70 900 (39,704)
13 100 (7,336)
n/a
74
29 (35)c
89 (100)c
7
100
100
80
Sm(Nꢀꢀ)3 + [O2NNꢀt-Pe]H2
Gd(Nꢀꢀ)3 + [O2NNꢀt-Pe]H2
d
d
d
Y(Nꢀꢀ)3 + [O2NNꢀt-Pe]H2
[{Sm(O2NPyt-Bu)(l-Cl)}2]
0e
YCl3(THF)3
0d
n/a
a Calculated using 1H NMR spectroscopy, polymerization reaction time 24 h. b GPC analysis. Corrected Mn values with the coefficient 0.56 in parentheses.25
c Values in parentheses are for duplicate reactions performed with different batches of initiator. d Data from previous studies using in situ generated amido
species, polymerization time 1 h.18 e Data from previous studies using a pyridyl substituted amine-(bis)phenolate ligand.15b
3594 | Dalton Trans., 2008, 3592–3598
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
Scheme 2 Proposed mechanism of e-caprolactone ROP by Ln(O2NNꢀR)Cl(THF) species.
amide or alkyl bond in these related lanthanide–amide and –
alkyl lactone ROP initiators.15,18 However, in our study no ligand
related end groups that would suggest this insertion mechanism
Y(O2NNꢀt-Pe)Cl(THF) and Ho(O2NNꢀt-Pe)Cl(THF) to confirm
these observations. This result suggests that electronic factors
do also play a role in determining the polymerization activity
of the complex. The smallest of the metals, Yb, gave a polymer
with the narrowest polydispersity, perhaps owing to a reduced
possibility of side reactions such as transesterification. A clear
trend between the polydispersities and metal size is not evident
in these reactions. This may be attributed to the different extents
to which the complexes fluctuate between monomer and dimer in
solution, so the number and type of active sites are not constant.
To assess the affect of monomer–dimer equilibria on this process,
polymerization reactions initiated by {Gd(O2NNꢀt-Pe)(THF)(l-
Cl)}2 and Y(O2NNꢀt-Pe)Cl(THF) were monitored over time using
1H NMR spectroscopy, Fig. 4. Very different reaction profiles for
each initiator were revealed. When polymerization was initiated by
Y(O2NNꢀt-Pe)Cl(THF), the initial rate of conversion is far greater
compared to the Gd species. This suggests that the smaller metal,
Y, does not dimerize as readily as Gd, as it is sufficiently shielded
by the bulky ligand and perhaps, only exists in its monomeric form
in solution so initiates polymerization more rapidly. Interestingly,
1
were observed in the H NMR spectra of the polymer products.
Another possible mechanism is via Lewis acid catalysis, which is
the mechanism employed by widely used tin(II) species including
stannous octoate.29 In such processes, polymerization is thought to
occur through transesterification reactions between the activated
lactone (metal coordinated lactone, similar to I1, Scheme 2) and
hydroxyl groups. The hydroxyl groups can originate from an
added initiator to yield designer end-functionalized P(e-CL),29d–f
or through reaction contaminants or water. This is unlikely the
case for our complexes, as 1H NMR spectra of the polymers
do not show the expected and distinctive terminal –CH2OH
resonances. Therefore, we are likely forming cyclic polyesters as
have been reported with other initiators.15e,26,30 Unfortunately, at
the time of this study, we were unable to perform MALDI-TOF
MS analyses on the polymer samples to prove this unequivocally.
This cyclic ester formation may be occuring by one of two
coordination–insertion mechanisms, both proceed via an e-CL
adduct intermediate such as I1, Scheme 2. Initiation may occur
via migratory insertion into the Ln–phenolate bond or through
ring-opening attack by the proximal –NMe2 group. Propogation
would then proceed via a slightly different coordination–insertion
mechanism for each type of initiation, followed by similar ring-
closing/termination steps to yield the cyclic polymer. The latter
type of ROP of e-CL facilitated by a Ln metal centre and attack of
an amine or amide functional group within an ancillary ligand
has previously been proposed by the groups of Okuda,30 and
Mountford.15e In both these cases, the resultant polymer was cyclic.
It is worth noting that the activity of the Yb species is
significantly lower than the other complexes studied, presumably
because of the smaller ion size and a reduced ability to form the
necessary intermediate, I1. The smaller size of the Yb3+ ion may
also have led to the lower molecular weight for the polymers it
yielded. It is, however, somewhat surprising to find that Y, being
close to Ho in size, demonstrated such a high activity. These
polymerization reactions were repeated using freshly synthesised
Fig. 4 Polymerization of e-caprolactone by Y(O2NNꢀt-Pe)Cl(THF) (fit r2 =
0.988) and [{Gd(O2NNꢀt-Pe)(THF)(l-Cl)}2] (fit r2 = 0.995).
This journal is
The Royal Society of Chemistry 2008
Dalton Trans., 2008, 3592–3598 | 3595
©

View Article Online
when employing {Gd(O2NNꢀt-Pe)(THF)(l-Cl)}2 as the initiator,
the reaction appears to be zero order in e-CL giving a linear
conversion vs. time plot. This difference, compared to the results
from Y(O2NNꢀt-Pe)Cl(THF), is once again presumably due to
differences in the solution-state structures between these two
related species. It is also noted that for Y(O2NNꢀt-Pe)Cl(THF), the
plot of ln([e-CL]0/[e-CL]t) vs. time was not linear (fit r2 = 0.88),
unlike many other lactone ROP initiators.24,25 Given the variety of
coordination–insertion mechanisms that are possible using these
initiators, it may be that the Y species facilitates the ROP by a
different mechanism to its Gd analogue.
benzophenone (diethyl ether, hexane, pentane, THF, toluene)
under argon. Transfer of dry oxygen-free solutions was carried
out using cannula or vacuum transfers. Deuterated benzene,
toluene and pyridine were dried by stirring over potassium metal.
The deuterated solvents were then freeze–pump–thaw degassed
and trap-to-trap distilled. LaCl3·6H2O, PrCl3·6H2O, SmCl3·6H2O,
GdCl3·6H2O, HoCl3·6H2O, YCl3·6H2O, and YbCl3·6H2O were
purchased from Strem and converted to the LnCl3(THF)x species
using literature procedures.32 Diamine-bis(phenol) ligands were
prepared in methanol, ethanol or water following literature
procedures.33 e-CL was dried over CaH2 and trap-to-trap distilled
prior to use. NMR spectra were recorded on a Jeol EX 270
One possible polymerization mechanism for our systems is
shown in Scheme 2. e-CL adducts related to I1, LnCl3(e-CL)3,
have been isolated and structurally characterized by Evans and
co-workers.27 Samarium tris(phenolate) complexes have recently
been reported as initiators for e-CL ROP and their mechanism
studied in some detail.28 It was proposed that e-CL insertion
occurs at all three Sm–O bonds and although in Scheme 2
insertion is only shown into one Ln–O bond, it may well be
occuring at both reactive bond sites. After repeated coordination
and migratory–insertion steps, termination is likely occuring
via nucleophilic attack of the ‘phenol’ oxygen at the Ln metal
centre and subsequent ‘back-biting’ of the growing polymer chain
to yield cyclic polyesters of narrow polydispersity. The narrow
range of low molecular weights afforded by all our systems
(n = 25–40) suggests the size of the chelate rings in I2 and I3
are important in mediating and moderating the polymerization
process, especially when compared with other lanthanide amine-
bis(phenolate) species that generally afford higher molecular
weight polymers.15,18 The presence of the pendant NMe2 group and
the resulting 5-membered Ln–NCH2CH2N ring is likely crucial to
the activity of these species, given related pyridyl species do not
initiate polymerization.15b The generally slow but high conversions
to afford cyclic low molecular weight polyesters with a narrow
molecular weight distribution suggests a key role for the metal
centre and resulting chelate size which restricts the chain (ring)
growth in these systems. Although organic catalysts for ROP of
lactones have recently been discovered,26 the protio ligands in
this study do not initiate polymerization nor do Ln(Cl)3THFx
reagents. However, many metal complexes bearing these amine
bis-phenolate ligands (as discussed above) are known to initiate e-
CL and lactide polymerization. Some of these such as lithium,
barium, calcium and Ln2+ species only contain M–phenolate
bonds around the metal centre and therefore, may initiate lactone
ROP via a similar coordination–insertion mechanism to those
proposed here.24,31 Clearly, there is still a significant amount of
mechanistic study needed to fully understand the ROP activity
of these and related complexes, and that several mechanisms may
even be occurring simultaneously (despite the low Mw/Mn of the
polymers).
1
instrument or Bruker AMX-500 or Avance-500 instruments. H
NMR spectra were referenced to residual protons in the deuterated
solvent and 13C NMR spectra to the residual 13C atoms of
the solvent. Elemental analyses were performed by Elemental
Microanalysis Ltd., Devon or Guelph Chemical Laboratories,
Canada. GPC analyses were performed by Rapra Technology
Ltd., Shrewsbury. The GPC system was calibrated with narrow
distribution polystyrene calibrants obtained from Polymer Lab-
oratories Ltd. Analyses were made on chloroform solutiuons at
30 ◦C using a refractive index detector.
X-Ray crystallography
Low-temperature diffraction data were collected on a Bruker-
AXS Smart Apex I diffractometer with graphite-monochromated
˚
Mo-Ka radiation (k = 0.71073 A). All structures were solved
by direct methods and refined against F2 on all data by full-
matrix least squares with SHELXTL.34 Absorption correction was
applied using SADABS.35 All non-hydrogen atoms were refined
anisotropically. All hydrogen atoms in 1 and 2 were included in
idealized positions and their Uiso values were set to ride on the Ueq
values of the parent carbon atoms (Uiso(H) = 1.5Ueq for methyl
carbons and 1.2 Ueq for other carbons). In 1, for the second
independent molecule in the asymmetric unit cell, orientation
disorders on two positions of CH2 and two Me groups in the
–CH2NMe2 fragment were found. The disordered groups were
refined with population of positions 0.8 : 0.2. Crystallographic
data and structure refinement details are given in Table 1.
General procedure for the preparation of Li2(O2NNꢀR) complexes
This method has been optimized based on the original literature
procedure.22 To a THF solution (20 mL) of the appropriate
protonated ligand (5.6 mmol) at 0 ◦C was slowly added nBuLi
◦
(1.6M, 8 mL, 12.8 mmol). After stirring at 0 C for 10 min and
then at room temperature for 2 h the solvent was removed in vacuo
to give an off-white solid. The solid was washed twice with pentane
(2 × 20 mL) at 0 ◦C and yielded a colourless powder. Crystals of
the complex Li2(O2NNꢀt-Bu) suitable for X-ray diffraction analysis
can be grown by several methods.24,30b
Experimental
Data for Li2(O2NNꢀt-Pe). Yield 75%. 1H NMR (C5D5N,
4
500 MHz, 298 K): d 7.34 (d, JHH = 1.70 Hz, 2H, ArH), 7.07
General procedures
4
2
(d, JHH = 1.70 Hz, 2H, ArH), 4.28 (d, JHH = 11.70 Hz, 2H,
All manipulations were performed under an atmosphere of dry
oxygen-free nitrogen or argon by means of standard Schlenk line
or glove box techniques unless otherwise stated. Analytical grade
solvents were purchased from Fisher and distilled from sodium
ArCHH), 3.23 (d, 2JHH = 11.70 Hz, 2H, ArCHH), 2.81 (t, 3JHH
=
5.44 Hz, 2H, NCH2), 2.46 (m, 2H, CH2), 2.03 (t, 3JHH = 5.44 Hz,
2H, NCH2), 1.79 (m, 2H, CH2), 1.72 (m, 4H, CH2), 1.66 (d,
4JHH = 2.62 Hz, 12H, CH3), 1.53 (s, 6H, N(CH3)2), 1.39 (s, 6H,
3596 | Dalton Trans., 2008, 3592–3598
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
Data for Yb(O2NNꢀt-Pe)Cl(THF). Yield 70%. Anal. Found: C
58.93; H 8.30; N 3.11. C42H70ClN2O3Yb: requires C, 58.69; H,
8.21; N, 3.26%.
1
CH3), 1.35 (s, 6H, CH3), 0.82 (m, 12H, CH3). 13C{ H} NMR
(C5D5N, 75.50 MHz, 298 K): d 166.5 (q), 136.3 (q), 130.9 (q),
128.1 (CH), 126.6 (CH), 125.3 (q), 68.3 (CH2), 64.5 (CH2), 45.5
(CH2), 39.3 (CH3), 38.1 (q), 37.6 (q), 32.6 (CH3), 29.9 (CH3).
Anal. Found: C 77.02; H 10.44; N 4.59. C38H62Li2N2O2: requires
C, 76.99; H, 10.54; N, 4.73%.
Data for Y(O2NNꢀt-Bu)Cl(THF), 1. Yield 65%. 1H NMR (C7D8,
270 MHz, 333 K): d 7.52 (d, 2H, 4JHH = 2.3 Hz, ArH), 7.41 (d, 2H,
4JHH = 2.4 Hz, ArH), 4.11 (d, 2H, 2JHH = 11.2 Hz, ArCHH), 3.78
2
Data for Li2(O2NNꢀt-Bu). Yield 70%. 1H NMR (C5D5N,
500 MHz, 298 K): d 7.53 (br, 2H, ArH), 7.17 (br, 2H, ArH),
(br, 4H, a-CH2, THF), 2.88 (d, 2H, JHH = 11.2 Hz, ArCHH),
2.36 (br, 2H, CH2), 2.12 (br, 2H, CH2), 1.90 (br, 6H, N(CH3)2),
1.73 (s, 18H, C(CH3)3), 1.54 (br, 4H, b-CH2, THF), 1.44 (s, 18H,
2
2
4.20 (d, JHH = 10.2 Hz, 2H, ArCHH), 3.09 (d, JHH = 10.2 Hz,
2H, ArCHH), 2.98 (br, 2H, NCH2), 2.43 (br, 2H, NCH2), 1.64 (br,
1
C(CH3)3). 13C{ H} NMR (C7D8, 67.5 MHz, 333 K): d 161.5 (C–
18H, CH3), 1.49 (br, 6H, N(CH3)2), 1.45 (br, 18H, CH3). 13C{ H}
1
O), 143.1 (C), 136.2 (C), 126.6 (C), 122.6 (CH), one CH masked
by deutero-solvent, 71.4 (a-CH2, THF), 65.3 (CH2), 59.4 (CH2),
49.2 (CH2), 46.5 (N(CH3)2), 35.5 (C(CH3)3), 34.2 (C(CH3)3), 32.2
(C(CH3)3), 30.1 (C(CH3)3), 25.3 (b-CH2, THF). Anal. Found: C
63.53; H 8.27; N 4.13. C38H62ClN2O3Y: requires C, 63.45; H, 8.69;
Cl, 4.93; N, 3.89%. X-Ray diffraction data: vide supra.
NMR (C5D5N, 75.50 MHz, 298 K): d 165.7 (q), 136.8 (q), 132.5
(q), 127.6 (CH), 125.2 (CH), 122.5 (q), 63.7 (CH2), 59.2 (CH2),
50.9 (CH2), 44.6 (CH3), 35.2 (q), 33.7 (q), 32.1 (CH3), 29.7 (CH3).
Anal. Found: C 76.07, H 10.41, N 5.00. C34H54N2O2Li2 requires:
C 76.09, H 10.14, N 5.22%.
Data for Sm(O2NNꢀt-Bu)Cl(THF), THF analogue of 2. Yield
70%. Anal. Found: C 57.89; H 7.89; N 3.33. C38H62ClN2O3Sm:
requires C, 58.46; H, 8.00; N, 3.59%. X-Ray diffraction data: vide
supra.
General procedure for the preparation of Ln(O2NNꢀt-Pe)Cl(THF)
and Ln(O2NNꢀt-Bu)Cl(THF) (Ln = Pr, Sm, Gd, Ho, Y, Yb)
◦
To a THF solution (15 mL) of LnCl3(THF)x (1.9 mmol) at 0 C
was added a THF solution (15 mL) of the appropriate lithiated
ligand (1.9 mmol) which was also cooled to 0 ◦C. This was stirred
at room temperature for 16 h resulting in the formation of a
precipitate. The solvent was removed in vacuo and the product
extracted into toluene (2 × 50 mL). Crystals of 1 and 2 suitable
for X-ray diffraction analysis were grown by slow concentration
of toluene solutions at room temperature. In general, for Ln =
General polymerization procedure
Polymerizations were conducted in stirred vials in a glove box.
To toluene solutions (2 mL) of initiator (0.0118 mmol), e-CL
(131 lL, 1.18 mmol) was added. These were stirred at room
temperature for 24 h. The vials were removed from the glove
box and 0.5 mL from each was added to pentane (1 mL). The
solvent was removed in vacuo and the resulting samples were
subjected to GPC analysis. Conversions of monomer to polymer
were determined using 1H NMR spectroscopy. Due to the presence
in some cases of paramagnetic contaminants, the polymers were
further purified by repeated toluene dissolution and methanol
precipitation cycles.
1
Pr, Sm, Gd, Ho and Yb, the H NMR spectra of the complexes
were broad due to the paramagnetic and fluxional nature of the
compounds, as a result meaningful interpretation was not possible.
However, elemental analytical data (CHN) were obtained on all
powders and gave good agreement with theory.
Data for Pr(O2NNꢀt-Pe)Cl(THF). Yield 62%. Anal. Found: C
61.30; H 9.11; N 3.27. C42H70ClN2O3Pr: requires C, 60.97; H, 8.53;
N, 3.39%.
Conclusions
We have shown that lanthanide amine-bis-phenolate complexes,
Ln(O2NNꢀR)Cl(THF), can be formed in moderate yields via the
reaction of the corresponding lithium complex with LnCl3·THFx
in THF. Somewhat surprisingly, we see no evidence for ‘ate’
complex formation and solid state structures reveal monomeric
species. In attempts to isolate e-CL adducts of these species, it
was revealed that over a prolonged reaction time, polymerization
was initiated and that these complexes yield low molecular weight
polyesters. The polymer products show no end-groups in their 1H
NMR spectra indicating that cyclic polymers might have formed.
The similarity of Mn for all the polymers produced by these
catalysts may be a result of steric constraint enforced by the chelate
rings within the initiator’s structure. These rings enforce a specific
chain length when polymer ‘back-biting’ will be forced to occur
and yield the cyclic polymer. Therefore, the metal–phenoxide or
the metal–amine bonds in amine bis(phenolate) Ln complexes,
although not as reactive as metal–amido or metal–alkoxide bonds,
can facilitate the ROP of e-CL. This reactivity should be noted
when studying related initiator systems, as by-products may be
produced and molecular weight distributions increased via the
mechanisms at play here. We also note that dimeric Ln–phenolate
Data for Sm(O2NNꢀt-Pe)Cl(THF). Yield 73%. Anal. Found: C
60.97; H 8.55; N 2.93. C42H70ClN2O3Sm: requires C, 60.28; H,
8.43; N, 3.35%.
Data for Gd(O2NNꢀt-Pe)Cl(THF). Yield 69%. Anal. Found: C
60.11; H 8.52; N 3.12. C42H70ClGdN2O3: requires C, 59.79; H, 8.36;
N, 3.32. X-Ray diffraction data have been previously reported and
reveal a dimeric structure.18
Data for Ho(O2NNꢀt-Pe)Cl(THF). Yield 81%. Anal. Found: C
60.32; H 8.44; N 3.17. C42H70ClHoN2O3: requires C, 59.25; H,
8.29; N, 3.29%.
1
Data for Y(O2NNꢀt-Pe)Cl(THF). Yield 67%. H NMR (C7D8,
4
500 MHz, 298 K): d 7.61 (d, 2H, JHH = 1.9 Hz, ArH), 7.43
4
(4JHH = 1.9 Hz, ArH), 4.02 (d, JHH = 11.9 Hz, 2H, ArCHH),
2
3.42 (br, a-THF), 3.13 (d, JHH = 11.9 Hz, 2H, ArCHH), 2.88
3
3
(d, JHH = 12.2 Hz, 2H, NCH2), 2.50 (d, JHH = 12.2 Hz, 2H,
NCH2), 2.28 (s, 6H, N(CH3)2), 2.08 (m, 8H, CH2CH3), 1.49 (br,
b-THF), 1.22 (m, 36H, CH3). Anal. Found: C 66.12; H 9.57; N
3.55. C42H70ClN2O3Y: C, 65.06; H, 9.10; N, 3.61%.
This journal is
The Royal Society of Chemistry 2008
Dalton Trans., 2008, 3592–3598 | 3597
©

View Article Online
chloride complexes behave differently in this reaction compared
with monomeric species.
T. Tang, N. Hu, F. Pei, D. Cui, X. Chen and X. Jing, Organometallics,
2007, 26, 2747; (e) H. E. Dyer, S. Huijser, A. D. Schwarz, C. Wang,
R. Duchateau and P. Mountford, Dalton Trans., 2008, 32; (f) Y. Yao,
M. Ma, X. Xu, Y. Zhang, Q. Shen and W.–T. Wong, Organometallics,
2005, 24, 4014.
Acknowledgements
16 F. Bonnet, A. C. Hillier, A. Collins, S. R. Dubberley and P. Mountford,
Dalton Trans., 2005, 421.
17 S. Agarwal and M. Puchner, Eur. Polym. J., 2002, 38, 2365.
18 F. M. Kerton, A. C. Whitwood and C. E. Willans, Dalton Trans., 2004,
2237.
19 Y. Matsuo, K. Mashima and K. Tani, Organometallics, 2001, 20, 3510.
20 E. Martin, P. Dubois and R. Je´roˆme, Macromolecules, 2003, 36, 5934.
21 M. Visseaux, C. H. Brachais, C. Boisson and K. Tortosa, C. R. Acad.
Sci., Ser. IIc: Chim., 2000, 631.
22 F. M. Kerton, C. M. Kozak, K. Lu¨ttgen, C. E. Willans, R. J. Webster
and A. C. Whitwood, Inorg. Chim. Acta, 2006, 359, 2819.
23 (a) G. W. Rabe, M. Zhang-Presse, F. A. Rieder and G. P. A. Yap, Inorg.
Chem., 2003, 42, 3527; (b) H. Ma, T. P. Spaniol and J. Okuda, Dalton
Trans., 2003, 4770; (c) C. L. Boyd, T. Toupance, B. R. Tyrell, B. D. Ward,
C. R. Wilson, A. R. Cowley and P. Mountford, Organometallics, 2005,
24, 309; (d) D. J. H. Emslie, W. E. Piers, M. Parvez and R. McDonald,
Organometallics, 2002, 21, 4226; (e) P. N. O’Shaughnessy, P. D. Knight,
C. Morton, K. M. Gillespie and P. Scott, Chem. Commun., 2003, 1770;
(f) M. W. Essig, D. W. Keogh, B. L. Scott and J. G. Watkin, Polyhedron,
2001, 20, 373; (g) P. C. Leverd, D. Rinaldo and M. Nierlich, Eur. J. Inorg.
Chem., 2001, 2021.
F. M. K. thanks the Royal Society (University Research Fellow-
ship), the EPSRC (Studentship, C. E. W.), NSERC (Discovery
Grant and RTI) and Canada Foundation for Innovation (Leaders
Opportunity Grant). A. A. T. thanks the Russian Foundation for
Basic Research (Grants 08-03-00391, 06-03-32728-a and 07-03-
12164-ofi), Presidium of the Russian Academy of Science (RAS),
and RAS Chemistry and Material Science Division.
Notes and references
1 (a) A. Hamitou, T. Ouhadi, R. Jerome and P. H. Teyssie´, J. Polym. Sci.,
1977, 15, 865; (b) M. Endo, T. Aida and S. Inoue, Macromolecules, 1987,
20, 2982; (c) R. C. Yu, C. H. Hung, J. H. Huang, H. Y. Lee and J. T.
Chen, Inorg. Chem., 2002, 41, 6450; (d) D. Chakraborty and E. Y. X.
Chen, Organometallics, 2003, 22, 769; (e) C. T. Chen, C. A. Huang
and B. H. Huang, Macromolecules, 2004, 37, 7968; (f) P. Hormnirun,
E. L. Marshall, V. C. Gibson, J. P. White and D. J. Williams, J. Am.
Chem. Soc., 2004, 126, 2688; (g) N. Nomura, T. Aoyama, R. Ishii and
T. Kondo, Macromolecules, 2005, 38, 5363.
2 (a) J. Okuda and I. L. Rushkin, Macromolecules, 1993, 26, 5530; (b) Y.
Takashima, Y. Nakayama, K. Watanabe, T. Itono, N. Ueyama, A.
Nakamura, H. Yasuda and A. Harada, Macromolecules, 2002, 35, 7538.
3 (a) H. Yavuz, C. Babac¸, K. Tuzlakog˘lu and E. Pis¸kin, Polym. Degrad.
Stab., 2002, 75, 431; (b) A. Kowalski, J. Libiszowski, T. Biela, M.
Cypryk, A. Duda and S. Penczek, Macromolecules, 2005, 38, 8170.
4 (a) M. Vivas and J. Contreras, Eur. Polym. J., 2003, 39, 43; (b) H. Y.
Chen, B. H. Huang and C. C. Lin, Macromolecules, 2005, 38, 5400.
5 E. L. Marshall, V. C. Gibson and H. S. Rzepa, J. Am. Chem. Soc., 2005,
127, 6048.
24 M. G. Davidson, C. T. O’Hara, M. D. Jones, C. G. Keir, M. F. Mahon
and G. Kociok-Ko¨hn, Inorg. Chem., 2007, 46, 7686.
25 M. Save, M. Schappacher and A. Soum, Macromol. Chem. Phys., 2002,
203, 889.
26 D. A. Culkin, W. Jeong, S. Csihony, E. D. Gomez, N. P. Balsara, J. L.
Hedrick and R. M. Waymouth, Angew. Chem., Int. Ed., 2007, 46, 2627.
27 W. J. Evans, J. L. Shreeve, J. W. Ziller and R. J. Doedens, Inorg. Chem.,
1995, 34, 576.
28 F. Peng and Z. Shen, J. Appl. Polym. Sci., 2007, 106, 1828.
29 (a) A. J. Nijenhuis, D. W. Grijpma and A. J. Pennings, Macromolecules,
1992, 25, 6419; (b) G. Schwach, J. Coudane, R. Engel and M. Vert,
J. Polym. Sci., Part A: Polym. Chem., 1997, 35, 3431; (c) P. J. A. In,’t Veld,
E. M. Vekner, P. Van De Witte, J. Hamhuis, P. J. Dijkstra and J. Feijen,
J. Polym. Sci., Part A: Polym. Chem., 1997, 35, 219; (d) C.-M. Dong,
K.-Y Qiu, Z.-W. Gu and X.-D. Feng, Macromolecules, 2001, 34, 4691;
(e) R. Hoogenboom, B. C. Moore and U. S. Schubert, Chem. Commun.,
2006, 4010; (f) N. Xu, F.-Z Lu, F.-S Du and Z.-C. Li, Macromol. Chem.
Phys., 2007, 208, 730.
6 S. J. McLain and N. E. Drysdale, Chem. Abstr., 1992, 203, 160-poly;
S. J. McLain and N. E. Drysdale, Polym. Prepr. (Am. Chem. Soc., Div.
Polym. Chem.), 1992, 33, 174.
7 Y. Shen, Z. Shen, Y. Zhang and K. Yao, Macromolecules, 1996, 29,
8289.
8 P. W. Roesky, M. T. Gamer, M. Puchner and A. Greiner, Chem.–Eur. J.,
2002, 8, 5265.
30 K. C. Hultzch, T. P. Spaniol and J. Okuda, Organometallics, 1997, 16,
4845.
9 A. Lara-Sanchez, A. Rodriguez, D. L. Hughes, M. Schormann and M.
Bochmann, J. Organomet. Chem., 2002, 663, 63.
31 (a) C.-A. Huang and C.-T. Chen, Dalton Trans., 2007, 5561; (b) Y.
Sarazin, R. H. Howard, D. L. Hughes, S. M. Humphrey and M.
Bochmann, Dalton Trans., 2006, 340; (c) H. Zhou, H. Guo, Y. Yao,
L. Zhou, H. Sun, H. Sheng, Y. Zhang and Q. Shen, Inorg. Chem., 2007,
46, 958.
10 S. C. F. Kui, H. W. Li and H. K. Lee, Inorg. Chem., 2003, 42, 2824.
11 J. Cheng, D. Cui, W. Chen, N. Hu, T. Tang and B. Huang, J. Organomet.
Chem., 2004, 689, 2646.
12 (a) Y. Yao, M. Ma, X. Xu, Y. Zhang, Q. Shen and W. T. Wong,
Organometallics, 2005, 24, 4014; (b) Y. Yao, X. Xu, B. Liu, Y. Zhang,
Q. Shen and W. T. Wong, Inorg. Chem., 2005, 44, 513; (c) X. Xu, Z.
Zhang, Y. Yao, Y. Zhang and Q. Shen, Inorg. Chem., 2007, 46, 9379.
13 C. Jing-Lei, Y. Ying-Ming, L. Yun-Jie, Z. Li-Ying, Z. Yong and S. Qi,
J. Organomet. Chem., 2004, 689, 1019.
14 (a) Y. Luo, Y. Yao, Q. Shen, K. Yu and L. Weng, Eur. J. Inorg. Chem.,
2003, 318; (b) Y. Yao, Y. Luo, J. Chen, Z. Zhang, Y. Zhang and Q. Shen,
J. Organomet. Chem., 2003, 679, 229.
15 (a) A. Amgoune, C. M. Thomas and J.-F. Carpentier, Macromol. Rapid
Commun., 2007, 28, 693; (b) F. Bonnet, A. R. Cowley and P. Mountford,
Inorg. Chem., 2005, 44, 9046; (c) A. Amgoune, C. M. Thomas, T. Roisnel
and J.-F. Carpentier, Chem.–Eur. J., 2006, 12, 169; (d) X. Liu, X. Shang,
32 W. E. Piers, P. J. Shapiro, E. E. Bunel and J. E. Bercaw, Synlett, 1990,
74.
33 (a) E. Y. Tshuva, I. Goldberg and M. Kol, Organometallics, 2001, 20,
3017; (b) F. M. Kerton, S. Holloway, A. Power, R. G. Soper, K. Sheridan,
J. M. Lynam, A. C. Whitwood and C. E. Willans, Can. J. Chem., 2008,
86, 435.
34 G. M. Sheldrick, SHELXTL v. 6.12, Structure Determination Software
Suite, Bruker AXS, Madison, Wisconsin, USA, 2000.
35 G. M. Sheldrick, SADABS v.2.01, Bruker/Siemens Area Detector
Absorption Correction Program, Bruker AXS, Madison, Wisconsin,
USA, 1998.
3598 | Dalton Trans., 2008, 3592–3598
This journal is
The Royal Society of Chemistry 2008
©