Aggregated initiators: defining their role in the ROP of rac-lactide

Article abstract of DOI:10.1039/c8dt01229f

Reported examples of aggregated initiators for the ring-opening polymerisation (ROP) of lactide often lack detailed investigations as to the nature of the active species, making it difficult to reconcile ligand design with performance. Here, we offer addi

Full text of DOI:10.1039/c8dt01229f

View Article Online
View Journal
Dalton
Transactions
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: R. D. McIntosh, J.
E. Cols, V. Hill and S. Williams, Dalton Trans., 2018, DOI: 10.1039/C8DT01229F.
This is an Accepted Manuscript, which has been through the
Volume 45 Number
1 7 January 2016 Pages 1–398
Royal Society of Chemistry peer review process and has been
accepted for publication.
Dalton
Transactions
Accepted Manuscripts are published online shortly after
An international journal of inorganic chemistry
www.rsc.org/dalton
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
author guidelines.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the ethical guidelines, outlined
in our author and reviewer resource centre, still apply. In no
event shall the Royal Society of Chemistry be held responsible
for any errors or omissions in this Accepted Manuscript or any
consequences arising from the use of any information it contains.
ISSN 1477-9226
PAPER
Joseph T. Hupp, Omar K. Farha et al.
Efficient extraction of sulfate from water using
framework
a Zr-metal–organic
rsc.li/dalton

Please do not adjust margins
Dalton Transactions
Page 1 of 11
View Article Online
DOI: 10.1039/C8DT01229F
Journal Name
ARTICLE
Aggregated initiators: Defining their role in the ROP of rac-lactide
Jean-Marie E. P. Cols,a Victoria G. Hill,a Stella K. Williamsa and Ruaraidh D. McIntosh*a
Received 00th January 20xx,
Accepted 00th January 20xx
Reported examples of aggregated initiators for the ring-opening polymerisation (ROP) of lactide often lack detailed
investigations as to the nature of the active species, making it difficult to reconcile ligand design with performance. Here,
we offer additional stability to the polynuclear titanium complexes, TiL(OiPr) (L = 9-14), through a bridging carboxylate
anchored to the supporting amine bis(phenolate) ligands. An in-depth study of solution-state behaviour determined the
process of assembly was driven by interactions between the carboxylate and a vacant site on a neighbouring titanium
centre. Furthermore, we establish that mononuclear units form dynamic mixtures of polynuclear aggregates, with a clear
relationship between nuclearity of the aggregates and the steric bulk on the ligand. Smaller aggregates displayed increased
activity towards the ROP of rac-lactide. Furthermore, addition of a chiral centre, on the ligand framework, was
DOI: 10.1039/x0xx00000x
www.rsc.org/
investigated as
a
route to influence the selectivity of the polymerisation via an easily-accessible initiators.
larger aggregates can retain the activity of the monometallic
complexes. In the work of Tolman et al. (M = Fe),14 the activity
of the pentanuclear activator was found to be similar to a
Introduction
Single-site, Lewis acidic metal centres are often applied to the
ring-opening polymerisation (ROP) of lactide as these are well
defined initiators that provide high levels of control over the
polymerisation process. While these initiators show high
activity towards the ROP of lactide,1 some can lack stability
outside of the glovebox. Polynuclear or aggregated initiators
generally have improved stability2 but structure-activity
relationships are more challenging to establish for these
species due to complications associated with defining their
nuclearity in solution.3-10
dinuclear activator but in contrast, a different iron(III) system,
found the dinuclear species to be more active than the
mononuclear species.15 These studies highlight that the
relationship between aggregate size and activity is not
straightforward.
The aforementioned examples are polynuclear complexes13-15
bridged by alkoxide ligands. These ligands are labile by
definition, as they also act as initiating groups in the ROP of
lactide and therefore have limited control over cluster
stability. Here, a multidentate amine bis(phenolate) ligand is
bound to the metal during the polymerisation process.
Utilising an amino acid,16 we can incorporate a pendant
carboxylate arm into the ligand, which is capable of providing a
stabilising, bridging interaction between two metal centres
and promote stability of the aggregates. The polynuclear
complexes presented herein represent out first efforts in the
synthesis of polynuclear titanium initiators for the ROP of rac-
lactide that are assembled and stabilised through ligand
design.
Aluminium isopropoxide is a well-studied example of an
aggregate that applied to the ROP of lactide.11,12 It was
reported that trinuclear and tetranuclear aluminium clusters,
bridged by the isopropoxide ligand, were formed under the
polymerisation conditions. The trinuclear species was found to
be the more reactive and more abundant of the two species.
Furthermore, varying the conditions of the reaction to
promote formation of the trinuclear species resulted in a
decrease in the PDI of the polymer. This increase in control
was attributed to the exchange between the trinuclear and
tetranuclear species which disrupts propagation.
Further examples of metal alkoxide aggregates include a group
of compounds with general formula M5(μ5-O)(OR)13 (M = Fe, Y,
La, Sm, Yb; R = alkyl). In the case of M = Y,13 similar levels of
activity in the ROP of lactide was reported for the
pentanuclear and mononuclear activators, suggesting that
Results and Discussion
Ligand synthesis and characterisation
Starting from the amino acid glycine, two synthetic routes
were used to access compounds
reductive aminations were required to obtain compound
while a one-pot Mannich condensation reaction produced
compounds
(H)3.17-19 Altering the peripheral substituents on
1-4
(H)3: two successive
1(H)3
2-4
Electronic supplementary information (ESI) available. CCDC 1562058 and
1562059.
the phenol rings allowed for analysis of structural trends in
assembling a polynuclear titanium complex as well as assessing
the tunability of the polynuclear titanium initiators for the ROP
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 1
Please do not adjust margins

Please do not adjust margins
Dalton Transactions
Page 2 of 11
View Article Online
DOI: 10.1039/C8DT01229F
ARTICLE
Journal Name
of lactide. Varying the starting amino acid, from glycine to centre onto the pendant arm of the ligand. This allows us to
alanine or phenylalanine, allows the incorporation of a chiral assess the effect
Scheme 1. Summary of synthetic routes to ligand precursors 1-8(H)3 and proposed mononuclear forms of titanium isopropoxide complexes
suitable donor moiety.
9-14, where L′ denotes a
it has on aggregation as well as its ability to influence the to the magnetic anisotropy of the phenol rings and the
stereoselectivity of the polymerisation. A reported procedure presence of the chiral X group. This was verified by EXSY NMR
for the synthesis of enantiomerically pure S-
followed20 but several attempts to obtain the optically pure exchange. In addition, the methylene protons for
compound were unsuccessful as polarimetry indicated the (H)3 are no longer observed as singlets, appearing as
6(H)3 was initially experiments that confirmed these signals were under
5-6
compound had racemised under the relatively harsh separate doublets for each diastereotopic protons.
conditions of the reaction.21-23
Large, colourless, block crystals of
6(H)3 were grown from a
We sought to synthesise compounds 5-8(H)3 in higher yields saturated methanol solution. These were analysed by SCXRD
using the reaction conditions previously employed in the using Cu radiation to determine the absolute configuration of
synthesis of compounds
(H)3.17-19 Unreacted phenol was the molecule by anomalous dispersion. The molecule was
2-4
removed by acidifying the reaction mixture in methanol with 1 determined to be in the S-configuration with high certainty
M aqueous hydrochloric acid and collecting the resulting shown by the reported flack parameter of 0.00(2).
solids. Further purification by column chromatography gave Unfortunately this single crystal sample was found to not be
the pro-ligands in high purity. Despite attempts to optimise the representative of the bulk sample.
methodology, the isolated yields for
unsatisfactory (< 5 % yield). We interpreted the cause as the and
bulkier benzyl group in the X position of (H)3 and (H)3, which the S-amino acids had racemised during the Mannich
7(H)3 and
8(H)3 were Polarimetry was used to examine the enantiopurity of
5(H)3
6(H)3. In both cases it was found that, in the bulk samples,
7
8
would sterically hinder the attack of the phenol on the condensation reaction to give rac-5(H)3 and rac-6(H)3.
imine/iminium ion during the electrophilic aromatic
substitution step in the mechanism of the Mannich Dynamic polynuclear titanium complexes.
condensation, especially during the addition of the second
A ligand substitution reaction was employed to synthesise
titanium complexes 14 using titanium isopropoxide as the
source of titanium. Compounds (H)3 were stirred in dry THF
phenol group. Indeed the monophenolate analogue of 8(H)3
9
-
was isolated from the reaction (see Supporting Information).
Heating to reflux in higher boiling solvents was attempted to
increase yields but this was ultimately unsuccessful in
1-6
under a dry N2 atmosphere. Following the addition of titanium
isopropoxide, an instant change from cloudy white suspension
to a clear yellow/orange solution occurred. The exception was
producing sufficient quantities of 7(H)3 or 8(H)3 for further
reaction.
compound
1(H)3 which, when treated with titanium
1
Characterisation of
1
-6
(H)3, by H and 13C NMR spectroscopy,
isopropoxide, swiftly formed a yellow precipitate.
showed the expected products had been obtained. However,
in the cases of (H)3 and (H)3, the spectra were more complex
Following the examples of other amine bis(phenolate)
complexes from the literature,24-26 the complexes in Scheme 1
were proposed to form in the first instance. Considering the
preference of amine bis(phenolate) titanium complexes for a
5
6
than had been expected. Bond rotations about the methylene
bridges result in magnetically inequivalent environments due
2 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Please do not adjust margins
Dalton Transactions
Page 3 of 11
View Article Online
DOI: 10.1039/C8DT01229F
Journal Name
ARTICLE
six-coordinate geometry, this would leave one coordination O‒Ti (2.058(2) and 2.063(2) Å) and C‒O (1.261(3) and 1.255(3)
site on the titanium centre occupied by a labile donor, L′ (e.g. Å) bond distances are in agreement with related
structures.19,28,29 The similarity in bond distances between
solvent).
The aforementioned yellow precipitate from the reaction of native and bridging interactions supports the view that the
compound
compound
1
(H)3 and titanium isopropoxide was identified as carboxylates are delocalised in their bridging interaction with
by mass spectrometry. Solution-based the titanium centres in the structure.
was Crystals of 123(HOiPr) were dissolved in CDCl3 and a H NMR
9
1
characterisation was not possible since compound
9
largely insoluble in common laboratory solvents. However, spectrum obtained. Comparing this spectrum with that of the
CPMAS NMR was used to obtain data that supported original dynamic mixture of 12 revealed that the trinuclear
characterisation by mass spectrometry and elemental analysis. form obtained in the solid state crystal structure forms part of
The NMR spectra of 9-14 in CDCl3 were more complex than the mixture. After leaving the dissolved crystals of 123(HOiPr)
anticipated. By integrating the broad regions in these spectra, in CDCl3 for one day at RT under a dry N2 atmosphere, further
the expected integrals for the aromatic, methylene, R′/R′′ and 1H NMR spectra were taken to reveal the trinuclear aggregate
isopropoxide groups are present, indicating that the had reverted to the dynamic mixture previously observed. This
1
compounds were formed. EXSY experiments revealed an was strong evidence that the mixtures observed by H NMR
exchange process was the effect behind the larger number of spectroscopy of 10
-14 are of a pure product undergoing
peaks in these complex spectra. We attributed the nature of dynamic exchange in solution, where each spectrum displays
10-14 in solution to the dynamic coordination of a variety of an equilibrium of species of differing nuclearity in solution. In
labile donors present in solution. As a representative example, both the 1H NMR spectra of 123(HOiPr) and 12, a sharp doublet
1
the H NMR spectrum of 12 was repeated in dry D6-DMSO, a is noticeable at 10.41 ppm. A key feature in the crystal
strong donor solvent that could inhibit exchange and/or structure is the intramolecular hydrogen bond between the
aggregation. Indeed, we observed significantly fewer signals in propan-2-ol bound to Ti1 and a carboxylate native to Ti3. The
a far less complex spectrum in comparison to the spectrum of 1H COSY NMR experiment showed this proton is coupling with
12 in CDCl3. However, multiple species under exchange were a proton at 4.36 ppm, which corresponds to the expected
still present. This confirmed that a strong donor could slow the chemical shift of an propan-2-ol CH. This coupling would
exchange process and supported our proposition that the explain why this appears as a doublet with a typical vicinal
vacant site was integral to the exchange process. Further to coupling constant of 7.60 Hz. An EXSY NMR experiment shows
this, 4-dimethylaminopyridine (DMAP) was added to an an exchange signal that corresponds to the propan-2-ol OH at
1
equimolar quantity of the complexes in CDCl3. The H NMR 1.60 ppm. From this evidence, we propose this doublet to arise
spectra of the samples showed the presence of a single species from the intramolecular hydrogen bonded propan-2-ol OH.
in solution with DMAP under exchange. These detailed The presence of the doublet at 10.41 ppm in both the spectra
experiments confirm the proposed structures of
9-14 exist in of the crystals of 12 and the dynamic mixture of 12 confirms
solution and that the nature of the species is solvent the solid-state trinuclear structure exists in solution as part of
1
dependent. A feature in all of the 1H NMR spectra of
DMAP in CDCl3 is the significant line broadening of the NMR experiments both concur with the proposed dynamic
methylene bridge protons. We interpret this as an indication nature of 10 14, whereby through-bond and through-space
9-14 with the dynamic mixture. Further analysis by H COSY and ROESY
-
that the ligand framework is flexible despite coordination to Ti. coupling was present for the varied species in solution. By
Through-space contacts between protons on the pendant focusing on specific regions of these spectra, we could
carboxylate, phenolate moieties in the ligand framework, and tentatively estimate the nuclearity of the aggregates in the
1
DMAP are clearly defined in the well-resolved spectra of 12 dynamic mixture. For example, we noted that the H NMR
and rac-14. This evidence placed a DMAP binding at the L′ spectrum of 11 in CDCl3 was better resolved with a set of
position.
signals with similar intensity in the methylene region. A COSY
Crystals of 12 were grown from a saturated, dry THF solution experiment showed there were nine pairs of diastereotopic
and analysed by SCXRD. The structure obtained showed a protons, which correspond to a trinuclear aggregate.
trinuclear form of 12 had been crystallised, whereby the three
Ti atoms are linked by carboxylates and one terminal Ti is
coordinated to propan-2-ol (Figure 1). This terminal propan-2-
ol can be distinguished from the isopropoxides in the structure
with its longer O‒Ti bond length (2.091(2) Å versus 1.771(2) Å)
as well as the freely-refined proton that was located in the
difference Fourier map (attached to O16). The D∙∙∙A and H∙∙∙A
distances of 2.556(3) Å and 1.80(4) Å respectively are within
expected hydrogen bonding values.27 The longer O‒Ti bond
length of 2.091(2) Å was found to be in agreement with other
neutral propan-2-ol titanium complexes (Figure S2).
Furthermore, no counter-ions are present in the asymmetric
unit. Analysis of the average native and bridging carboxylate
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 3
Please do not adjust margins

Please do not adjust margins
Dalton Transactions
Page 4 of 11
View Article Online
DOI: 10.1039/C8DT01229F
ARTICLE
Journal Name
intramolecular hydrogen bond (dashed red lines) between the isopropanol
ligand and the carboxylate native to Ti1. Ti = purple; C = grey; N = blue; O = red,
Figure
[Ti4(OiPr)]2[Ti4(OiPr)(HOiPr)]. Ligand framework displayed as tubes. Hydrogen
atoms omitted for clarity with exception of the proton that displays an
1.
Crystal
structure
of
123(HOiPr)
with
formula
H
=
white.
Figure 2. Plot of Gaussian fitting curves for datasets of estimated nuclearities of 10-14 in CDCl3 versus the count of data points corresponding to said nuclearity. The
estimated nuclearities were obtained by 1H DOSY NMR experiments via estimated MWs. See Supporting Information for further details.
1
Using H COSY and ROESY NMR experiments, we tentatively when adding the X group in rac-13 and rac-14. The dynamic
found the possibility of trinuclear/tetranuclear aggregates nature of the aggregates in solution and the overlapping of
under exchange for 10, dinuclear/trinuclear aggregates for 12 signals in the NMR spectra can result in an average diffusion
and mononuclear/dinuclear aggregates for 13 and 14. A more coefficient being estimated from a signal, resulting in
detailed analysis is provided in the Supporting Information.
broadening of the distributions presented here. The estimated
Having observed aggregation in the crystal structure of 12, we nuclearities of the compounds correlate with the observations
sought to gain insight into the extent of aggregation in samples made previously by EXSY, COSY and ROESY NMR experiments.
1
14 in solution. A H DOSY NMR experiment would allow Using the aggregated structure 123(HOiPr), we can look at how
of 10
-
the determination of the diffusion coefficients of the species in bulk in the R′/R′′ positions affect aggregation. Close steric
solution that can be used to estimate the MWs of the interactions between C24D/C47C show how bulk in the R′′
aggregates using external calibration curves (ECCs). These position impedes aggregation. Importantly, close proximity
ECCs, published by Stalke and coworkers,30,31 take into account between C12/C29B and C22/C45C demonstrates how bulk in
variables when measuring the diffusion coefficient such as the the R′ also hinders aggregation, explaining the trends observed
solvent, analyte shape and concentration, temperature and for the aggregation of compounds 10-12.
the spectrometer used. The authors noted that the ECCs The proposed structures (Scheme 1) are based on the evidence
worked well for a particular molar density range. Molecules from the crystal structure of 12 as aggregates of the
with heavy atoms, in relation to organic atoms, exceed the mononuclear form of 12, which is composed of a single
molar density range stated for the ECCs, which results in an titanium centre bound by ligand
4 and an isopropoxide. It
underestimation of the MW. For the analytes studied here, the should be noted that the dynamic nature of this system limits
heavier titanium atom leads us to expect an underestimation our ability to probe the structures present in the mixture.
of MW that will be consistent throughout 10
-14. For further Further conceivable entities in solution are aggregates with
details, refer to the Supporting Information.
various donors at the vacant site (bridging carboxylate,
Estimated MW distributions were obtained for compounds 10
-
propan-2-ol, isopropoxide). The ambidentate nature of the
to switch from LX3 to L2X2
14 using the most appropriate ECCs. A Gaussian distribution carboxylate arm allows ligands
1-6
was fitted for each dataset to estimate the mean size of the type donation, allowing for both L- and X-type donors on the
distributions. Figure 2 overlays the MW distributions of vacant site.
compounds 10
at the R′/R′′ positions of 10
of the aggregates in solution. The same effect was associated highlights the challenges faced in understanding
-
14. As expected, an increase in the steric bulk In comparison to 10
-14, complex 9 was not able to be studied
-
12 results in a decrease in the size to the same level of detail due to its heterogeneity. This
a
4 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Please do not adjust margins
Dalton Transactions
Page 5 of 11
View Article Online
DOI: 10.1039/C8DT01229F
Journal Name
ARTICLE
heterogeneous system in comparison to a homogeneous mononuclear form to be the most active form. Our
analogue.
observations agree with these systems whereby the additional
unit(s) during aggregation cause an increase in steric bulk
around the active site.
Application in the ROP of rac-lactide
Further studies of 10-12 under melt polymerisation conditions
The low relative cost and toxicity of Ti(IV) make it a highly
desirable initiator for the production of a biodegradable
polymer with applications where leaching of a toxic metal is of
great concern.32-34
were undertaken with 300:1 monomer:titanium loading in a
sealed flask for 24 h. Under these conditions, all three
compounds 10-12 achieved high conversion after 24 h. To
reveal differences in activity between these compounds, the
polymerisation time was reduced. Conversion fell below 15%
when the polymerisation time was 2 h and efforts to recover
the polymer were unsuccessful, an indication that short chains
were synthesised. A polymerisation time of 6 h was successful
The performance of complexes 9-14 was studied in the ROP of
rac-lactide to investigate how the ligand design and nuclearity
of the complexes affect the polymerisation process. We found
ambiguity relating the nuclearity of polynuclear species to
their performance in the ROP of lactide. The in-depth
understanding of the solution-state behaviour of polynuclear
will be key in allowing us to draw conclusions between
performance and structure. Furthermore, our aim was to
determine the extent of stereocontrol that could be imparted
by the chiral X group in rac-13 and rac-14 despite the high
temperature (130 °C) and dynamic nature of the complexes.
The data collected from these experiments is summarised in
Table 1. The monomer-to-titanium loadings were calculated
assuming the solid samples were of the pure mononuclear
in revealing the difference in activity between compounds 10
-
12. The PDI values are consistently lower for 12 in comparison
to 10 and 11. This can be explained by the tert-butyl R′′ group
that imparts greater control over the polymer formed, which is
in accord with what is found for a variety of other ROP of
lactide systems.24,36,37 Increasing the extent of alkyl groups
through the R′/R′′ groups on the titanium initiators 10-12
correlates with an increase in conversion. In contrast to what
was observed under solution polymerisation conditions,
compound 11 is more active than compound 12 under melt
conditions. The difference between 11 and 12 is the additional
tert-butyl group in the R′′ positions, which primary
contribution will be greater steric bulk around the active site
on the titanium centre. Under melt polymerisation conditions,
rac-lactide acts as a strong donor solvent that will reduce the
extent of aggregation much like when using strong donors
during NMR characterisation of these complexes. We propose
that the tert-butyl group in the R′′ position for initiator 12
would sterically hinder the interaction of the monomer and
polylactide chains with the titanium centre, resulting in a
forms of 10
-
14.
Compound
9
showed near-zero conversions under various
conditions and thus no polymer
polymerisation
characterisation data could be gathered. This can be explained
by the heterogeneous nature of the compound hampering
access to the active sites on the titanium centre. Using the
more soluble analogues of
achieving an active compound for the ROP of rac-lactide.
By comparing the conversion achieved by 10 12 under
different [M]0/[Ti]0 loadings under solution polymerisation
conditions, an increase in activity described by 10 11 12
9 with alkyl R′/R′′ groups was key to
-
<
<
reduced activity in comparison to compound 11
. To
was apparent. The significant difference in performance was
unexpected for a simple change from methyl to tert-butyl to
di-tert-butyl R′/R′′ groups, prompting us to further investigate
the underlying cause.
summarise, aggregation effects are proposed to be less
pronounced under melt conditions and are outweighed by the
extent to which the monomer and growing chains can access
the active site. Further testing of the catalytic system under
melt conditions was undertaken by increasing the
monomer:titanium ratio to 600:1 for compounds 11 and 12. A
slight decrease in the conversion was observed in comparison
to a monomer:titanium ratio of 300:1. In addition, the Mn
values are similar, which is unexpected as a lower loading of
active sites typically leads to decreased polymer chain lengths.
In this case, the values obtained for conversion, Mn and
Mn(NMR) are similar despite reducing the number of active sites
by half. This is indicative of polymeryl exchange, which we
attribute to the high temperature (130 °C) of the reaction and
competing aggregation of the initiator.
Having observed aggregation to be
complexes, we sought to investigate the tendency of 10
aggregate in solution. The 1H DOSY NMR data revealed the
complexes formed larger aggregates in the order 10 11 12
a
feature of these
-
12 to
>
>
,
which correlated with the increase of steric bulk from 10 to 11
to 12. Since aggregational effects would compete with the
monomer and polymer chains for time on the active site on
the titanium centre, it would follow that a greater tendency for
aggregation would lead to a decrease in activity. For this
reason, we propose aggregational effects to correlate with the
activity of these compounds, whereby 10 exhibits low activity
in solution due to its propensity to aggregate while 12 has a
lower tendency to aggregate and displayed better
performance than 10 or 11. In these systems comprised of
Ligand influence over stereoselectivity
dynamic polynuclear aggregates, we found reducing nuclearity Stereoselective ROP of rac-lactide is sought to control the
resulted in increased activity towards the ROP of rac-lactide. tacticity of the resulting polylactide product. Increasing the
Other systems10-14 typically found nuclearity to have little or no isotacticity of polylactide has been shown to influence the
effect on performance. In contrast, some systems with metal properties of the polymer product and altering these is
centres supported by multidentate ligands15,35 showed the
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 5
Please do not adjust margins

Please do not adjust margins
Dalton Transactions
Page 6 of 11
View Article Online
DOI: 10.1039/C8DT01229F
ARTICLE
Journal Name
essential for the successful application of polylactide in the examine the effect of adding a chiral centre to the ligand. For
plastics industry and widen its potential uses.38,39
the initiator S-14 we see that there are two potential sites for
Compounds 11 and 12 were the most active initiators for the polymerisation on the Ti centre (Figure 3). In the crystal
ROP of rac-lactide under both solution and melt structure, site is bound by the isopropoxide initiating group
A
polymerisation conditions. For this reason, using the analogous while site A* is bound by the
compounds rac-13 and rac-14 provide an opportunity to
Table 1. Polymerisation data for 10-
14 in the ROP of rac-lactide under solution and melt polymerisation conditions. a Conversion from 1H NMR spectra evaluated by
c
integration of the methine regions of polylactide versus lactide. b Polymer number average molar mass determined from GPC traces. Calculated polymer molecular
weight by 1H NMR end-group analysis. d Polydispersity indices (Mw / Mn) obtained from GPC traces.
bridging carboxylate. The initiating site
A is proposed to be too rac-lactide under melt polymerisation conditions, with good
far away from the chiral centre on the pendant arm to be agreement between the conversion, Mn and PDI values.
Solution polymerisation conditions ([M]0 = 1 mol∙L-1, dry toluene, oil bath temperature 130 °C)
[b]
[c]
Entry
1
Activator
10
[M]0/[Ti]0
20
Time
24 h
24 h
24 h
24 h
24 h
24 h
24 h
24 h
24 h
24 h
24 h
24 h
Conv.[a]
Mn
Mn(NMR)
PDI[d]
-
4
-
-
2
10
100
20
0
-
3
11
92
13
19
90
62
86
26
13
93
59
4120
-
7260
4790
1.24
-
4
11
50
5
11
100
50
4070
7360
9420
6080
4400
-
1.09
1.24
1.18
1.54
1.24
-
6
12
11300
13500
4780
4330
-
7
12
100
20
8
rac-13
rac-13
rac-13
rac-14
rac-14
9
50
10
11
12
100
20
4050
5260
5640
5880
1.27
1.22
50
Melt polymerisation conditions (oil bath temperature 130 °C)
[b]
[c]
Entry
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
Activator
10
[M]0/[Ti]0
300
300
300
300
300
300
600
300
300
300
600
300
300
300
600
300
300
300
Time
2 h
Conv.[a]
5
Mn
Mn(NMR)
-
PDI[d]
-
-
10
6 h
21
83
18
74
96
93
14
47
89
80
45
82
92
87
8
5040
10000
-
4640
9810
-
1.32
1.47
-
10
24 h
2 h
11
11
6 h
13500
15000
14800
-
13000
10700
11800
-
1.64
1.54
2.03
-
11
24 h
24 h
2 h
11
12
12
6 h
6780
14100
12200
6300
9700
10500
13700
-
8320
12000
11500
5470
7550
8910
10900
-
1.26
1.40
1.39
1.46
1.58
1.65
1.55
-
12
24 h
24 h
2 h
12
rac-13
rac-13
rac-13
rac-13
rac-14
rac-14
rac-14
6 h
24 h
24 h
2 h
6 h
60
90
7830
11000
5200
12000
1.22
1.42
24 h
influenced by it. In contrast, the A* sit is in close proximity and The microstructure of the polylactide produced by 10
therefore able to impart a degree of stereoselective control inspected using 1H{1H} NMR spectroscopy. The Pm values
over the polymerisation. indicate a slightly heterotactic polymer was obtained using
Under solution polymerisation conditions, there is a disparity initiators 10
12 40-43 Calculating the Mn from the 1H NMR of the
-14 was
-
.
in the activity between the achiral/chiral analogues 11/rac-13 polylactide chain agree for the most part with the Mn values
and 12/rac-14. Lower conversion from monomer to polymer obtained from GPC, suggesting linear polylactide is produced.
was observed. Although the X group was seen to reduce Isotactic enrichment was inspected by comparing the
aggregation in rac-13 and rac-14, we envisage the X group effectiveness of site control mechanism (SCM)41 or chain end
causes greater steric hindrance for both monomer and control mechanism (CEM)42 statistics for tetrad intensity
polymer chain approach to the titanium centre, in turn prediction followed by comparison of the Pm value between
reducing the activity of the compounds. In contrast, 11/rac-13 the achiral and chiral initiators (Table S2). The SCM statistical
and 12/rac-14 show very similar performance in the ROP of model failed to predict the tetrad intensities for polylactide
samples produced by rac-13, in which case CEM statistics were
6 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Please do not adjust margins
Dalton Transactions
Page 7 of 11
View Article Online
DOI: 10.1039/C8DT01229F
Journal Name
more appropriate. For polylactide samples produced by rac-14
ARTICLE
,
the ROP of rac-lactide under solution and melt polymerisation
SCM statistics were more successful in the prediction of tetrad conditions. Increasing the steric bulk of the peripheral
intensities than CEM statistics. Thus, for rac-14 we see an phenolate groups resulted in an increase in activity. The
increase in intensity of the mmm tetrad (and Pm) associated largest, with di-tert-butyl phenolate donors, was observed to
with an increase in influence of a SCM over enchainment significantly increase the activity of the polynuclear complex,
during the polymerisation. The lack of stereoselectivity which we attribute to restricted aggregation of the titanium
imparted by rac-13 in comparison to rac-14 can be explained species. Unfortunately, this group also hindered
by referring to Figure 3. The achiral site
A is not sterically polymerisation under melt conditions due to the restriction it
disfavoured in rac-13 due to the lack of tert-butyl groups in the causes about the titanium centre – a contrasting result to our
R′′ positions. With no differentiation between the achiral and previous findings whereby enhanced solubility had the
chiral sites, the influence on stereoselectivity imparted by rac- greatest influence over activity under melt polymerisation
13 was minimal.
consumption against time gives a linear fit for both rac- and L- ligand framework to form analogous titanium species with
A
semi logarithmic plot of lactide conditions.19 Finally, a chiral backbone was introduced to the
lactide, from which the
both achiral and chiral sites. When steric bulk hindered access
to the achiral site, a SCM imparted a greater degree of
isotactic enchainment in the polylactide chain. These
polynuclear titanium aggregates showcase how careful design
on the molecular scale can have a great effect on the
performance of initiators and macromolecular properties of
the polymer they produce. Using this new understanding, our
focus will be to increase the activity of new polynuclear
initiators and the isotactic enchainment they can impart to the
polymer under industrially-viable conditions.
Experimental
The starting materials were purchased and used as received
from Sigma-Aldrich and Acros except for rac- and L-lactide,
which were sublimed and stored under a dry N2 atmosphere
prior to use. Dry solvents were purified in an MBRAUN SPS-800
and stored over 4 Å molecular sieves under a dry N2
atmosphere. NMR spectroscopy data was acquired with a
Bruker AVIII 300 MHz instrument or Bruker AVIII 400 MHz at
298 K unless otherwise stated. 1H DOSY NMR experiments
were recorded on a Bruker AVIII 400 MHz with samples at 50
Figure 3. Models showing the achiral (A) and chiral (A*) sites for: A) S-13 B) S-14;
constructed using a mononuclear unit from the asymmetric unit of 123(HOiPr). A
portion of the model is displayed as wireframe for clarity. Yellow: tert-butyl
groups in the R′′ positions. Green: methyl group in the X position. Orange: achiral
(
A
) and chiral (
A*) sites. Purple: titanium. Grey: carbon. Blue: nitrogen. Red:
oxygen.
observed rate constant kapp was calculated to be 2.89 × 10-2
min-1 and 2.44 × 10-2 min-1 respectively. The almost equal rates
indicate there is no preference for rac- or L-lactide
consumption by rac-14. However, we remain cautious over our
assignment of mechanism as the challenges in distinguishing
between SCM and CEM are well documented.43-46
nM
concentration
and
tetramethylsilane
standard.
Electrospray ionisation (ESI) and electron impact ionisation (EI)
were recorded using a Bruker micrOTOF II. Nanoelectrospray
ionisation (NSI) and matrix-assisted laser desorption (MALDI)
mass spectra were obtained by the EPSRC National Mass
Spectrometry Facility (NMSF), Swansea, UK. Elemental
microanalysis was carried out on an Exeter CE-440 Elemental
Analyse. Single crystal X-ray diffraction data was acquired
using a Bruker Apex-II operating at 173 K with Mo-Kα radiation
or Rigaku SuperNova operating at 120 K with Cu-Kα radiation.
CCDC 1562058 and 1562059 contain the supplementary
crystallographic data for this paper. These data can be
obtained free of charge from The Cambridge Crystallographic
Data Centre. Molecular weights (Mn) and molecular mass
distributions (Mw / Mn) of polymers were determined by GPC
at 35°C using THF as eluent with a flow rate of 1.0 mL min-1.
Conclusions
To study the effect of aggregation of the initiator in the ROP of
rac-lactide we undertook an in-depth study to define a trend
between nuclearity and activity, We successfully synthesised
and characterised a series of polynuclear titanium complexes
stabilised by a multidentate amine bis(phenolate) ligand
featuring a bridging carboxylate donor. These compounds
presented themselves as dynamic mixtures of different
nuclearities. An in-depth solution-state NMR study was key to
develop our understanding of the aggregates in solution and
relate their behaviour to their performance in the ROP of rac-
lactide. Subsequently, 1H DOSY NMR studies probed the size
distributions of aggregates in solution, establishing a trend:
increasing steric bulk around the ligand periphery reduced the
nuclearity of the aggregates. These aggregates were applied to
The experimental values were obtained relative to
a
calibration curve using polystyrene standards, which were
corrected with Mark-Houwink parameters and by a factor of
0.58.47
Synthesis of ligand precursors
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 7
Please do not adjust margins

Please do not adjust margins
Dalton Transactions
Page 8 of 11
View Article Online
DOI: 10.1039/C8DT01229F
ARTICLE
Journal Name
48,49
4(H)3
:
Synthesis of titanium complexes
The dynamic nature of these polynuclear titanium aggregates
leads to complex NMR spectra. For 10 14, NMR data will be
Sodium hydroxide (0.80 g, 20 mmol) was dissolved in methanol
(30 mL) with 2,4-di-tert-butylphenol (8.253 g, 40 mmol) and
glycine (1.50 g, 20 mmol). Paraformaldehyde (1.20 g, 40 mmol)
was added and the reaction mixture was heated to reflux for
48 h under N2. A white precipitate formed and was filtered,
washed with 3×30 mL ice-cold methanol and dried. Yield:
6.834 g (66.8 %). 1H NMR (300 MHz, CDCl3): δ 7.22 (d, 2H, ArH),
6.89 (d, 2H, ArH), 3.69 (s, 4H, CH2), 3.29 (s, 2H, CH2), 1.40 (s,
18H, CH3), 1.29 (s, 18H, CH3); 13C NMR (75.5 MHz, CDCl3): δ
171.4, 152.7, 141.2, 136.3, 125.2, 124.2, 120.0, 60.5, 57.3,
34.9, 34.1, 31.6, 29.6; HRMS (ESI+): m/z calcd for C32H50NO4:
512.3734 [M+H]+; found 512.3709; elemental analysis calcd
(%) for C32H49NO4: C, 75.11; H, 9.65; N, 2.74; found: C 74.98; H
9.52; N 2.69.
-
presented for the complex in CDCl3 followed by CDCl3 with
DMAP inserted as a strong donor to reduce aggregation.
Additional data for 12 in CD6SO was also obtained and
displayed two sets of signals with a 1:0.4 ratio.
General method
Under N2, Ti(OiPr)4 (0.154 mL, 0.500 mmol) was added to a
suspension of the ligand precursor (0.500 mmol) in dry THF (5
mL). The yellow/orange reaction mixture was stirred for 2 h
after which solvents were removed under vacuum.
Complex
9
The yellow solid was suspended in dry toluene, filtered, dried
1
and weighed. Yield: 0.3764 g (96.2 %). H NMR (400 MHz, 293
K, inferred by HMQC): δ 9.67-5.49 (ArH), 6.70-5.92 (CH), 5.25-
4.01 (CH2), 3.66-2.21 (CH3); 13C CP-MAS NMR (100.6 MHz, 293
K): δ 182.2, 181.7, 181.5, 165.6, 161.9, 160.3, 133.5, 131.3,
130.4, 129.6, 129.0, 127.8, 127.2, 125.6, 125.3, 124.6, 124.0,
122.9, 121.5, 120.0, 118.3, 115.9, 115.0, 114.5, 82.6, 81.0,
64.7, 63.7, 63.2, 62.9, 60.9, 59.9, 27.0, 26.1, 24.8; MS (EI): m/z
calcd for C19H21NO5Ti: 391.2 [M]∙+; found 391.1; elemental
analysis calcd (%) for C19H21NO5Ti: C, 58.33; H, 5.41; N, 3.58;
found: C 59.76; H 5.48; N 3.49.
rac-5(H)3:
Sodium hydroxide (0.80 g, 20 mmol) was dissolved in methanol
(30 mL) with 4-tert-butylphenol (4.326 g, 40 mmol) and S-
alanine (1.782 g, 20 mmol). Paraformaldehyde (1.20 g, 40
mmol) was added and the reaction mixture was heated to
reflux for 72 h under N2. The solvents were removed to give
off-white solids that were dissolved in methanol (30 mL). The
yellow solution was acidified with aqueous 1 M hydrochloric
acid and the white precipitate collected and dried. The product
was isolated by column chromatography (gradient 90:10 to
50:50, dichloromethane : ethyl acetate with 0.1 % acetic acid)
Complex 10
The yellow solid was suspended in dry acetonitrile, filtered,
1
dried and weighed. Yield: 0.3668 (87.5 %). H NMR (400 MHz,
CDCl3): δ 7.14-5.99 (m, 6H, ArH), 5.27-2.50 (m, 7H, CH / CH2),
1
2.41-0.74 (m, 15H, CH3); H NMR (400 MHz, CDCl3 / DMAP): δ
1
as a white powder. Yield: 4.036 g (48.8 %). H NMR (400 MHz,
CDCl3): δ 7.17 (d, 2H, ArH), 7.03 (s, 2H, ArH), 6.87 (d, 2H, ArH),
4.53 (br, 2H, CH2), 4.18 (br, 2H, CH2). 3.86 (br, 1H, CH), 1.58
(br, 3H, CH3), 1.18 (s, 18H, CH3); 13C NMR (100.6 MHz, CDCl3): δ
176.2, 154.3, 142.2, 128.0, 127.7, 116.7, 116.1, 60.6, 59.9,
53.8, 31.5, 14.2; HRMS (ESI+): m/z calcd for C25H36NO4:
414.2642 [M+H]+; found 414.2657; elemental analysis calcd
(%) for C25H35NO4: C, 72.61; H, 8.53; N, 3.39; found: C 72.32; H
8.22; N 3.31.
8.23 (br, 2H, ArH DMAP), 7.03-6.24 (br, 8H, ArH / ArH DMAP),
4.66 (br, 1 H, CH), 3.96 (m, 0.5H, CH), 3.68 (br, 2H, CH2), 3.24
(br, 2H, CH2), 2.96 (s, 6H, CH3 DMAP), 2,18 (br, 6H, CH3), 1.17
(br, 10H, CH3); 13C NMR (100.6 MHz, CDCl3 / DMAP): δ 181.8,
174.6 (br), 164.0, 158.7, 155.1, 149.2 (br), 130.3 (br), 130.1,
130.0, 129.8, 129.7, 129.0 (br), 128.8, 125.7, 125.0, 123.9,
116.7, 115.9, 115.8, 106.6, 81.1 (br), 64.6, 64.4, 62.5, 60.0,
39.1, 25.4, 24.5, 20.6, 20.5; MS (MALDI): m/z calcd for
C24H34NO6Ti: 480.2 [M+H+HOiPr]+; found 480.1. HRMS (ESI–):
m/z calcd for C21H28NO5Ti: 422.1447 [M+(H)3]–; found
422.1146; elemental analysis calcd (%) for C21H25NO5Ti: C,
60.15; H, 6.01; N, 3.34; found: C 61.99; H 6.74; N 3.28.
Complex 11
rac-6(H)3:
Sodium hydroxide (0.80 g, 20 mmol) was dissolved in methanol
(30 mL) with 2,4-di-tert-butylphenol (8.253 g, 40 mmol) and
glycine (1.50 g, 20 mmol). Paraformaldehyde (1.20 g, 40 mmol)
was added and the reaction mixture was heated to reflux for
96 h under N2. The solvents were removed to give off-white
solids that were dissolved in methanol (30 mL). The yellow
solution was acidified with aqueous 1 M hydrochloric acid and
the white precipitate collected and dried. The product was
isolated by column chromatography (gradient 95:5 to 70:30,
dichloromethane : ethyl acetate with 0.1 % acetic acid) as a
white powder. Yield: 6.224 g (59.2 %). 1H NMR (300 MHz,
CDCl3): δ 7.16 (d, 2H, ArH), 6.81 (d, 2H, ArH), 4.11 (d, 2H, CH2),
3.73 (q, 1H, CH), 3.36 (d, 2H, CH2), 1.38 (d, 3H, CH3), 1.30 (s,
18H, CH3), 1.19 (s, 18H, CH3); 13C NMR (75.5 MHz, CDCl3): δ
174.5, 152.8, 141.2, 136.3, 125.2, 124.1, 119.9, 60.3, 55.9,
53.0, 34.9, 34.1, 31.6, 29.6, 14.3; HRMS (ESI+): m/z calcd for
C33H52NO4: 526.3890 [M+H]+; found 526.3883 ; elemental
analysis calcd (%) for C33H51NO4: C, 75.39; H, 9.78; N, 2.66;
found: C 75.30; H 9.54; N 2.67.
The orange solid was suspended in dry acetonitrile, filtered,
1
dried and weighed. Yield: 0.4286 (85.1 %). H NMR (400 MHz,
CDCl3): δ 7.26-6.21 (m, 6H, ArH), 5.42-2.61 (m, 7H, CH / CH2),
1
2.11-0.79 (m, 24H, CH3); H NMR (400 MHz, CDCl3 / DMAP): δ
8.22 (br, 2H, ArH DMAP), 7.17 (d, 2H, ArH), 6.96 (s, 2H, ArH),
6.71 (br, 2H, ArH), 6.41 (d, 2H, ArH DMAP), 4.67 (br, 1H, CH),
3.96 (m, 0.5H, CH), 3.77 (br, 2H, CH2), 3.24 (br, 2H, CH2), 2.96
(s, 6H, CH3 DMAP), 1.17 (br, 28H, CH3); 13C NMR (100.6 MHz,
CDCl3 / DMAP): δ 181.7, 174.7 (br), 163.5, 158.5, 154.8 (br),
149.3 (br), 142.5 (br), 126.5 (br), 126.0, 124.7, 123.5 (br), 116.4
(br), 106.6, 80.8 (br), 64.4, 63.4 (br), 39.2, 34.1, 31.6, 25.4,
24.6; HRMS (ESI–): m/z calcd for C27H40NO5Ti: 506.2386
[M+(H)3]–; found 506.2104; elemental analysis calcd (%) for
8 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Please do not adjust margins
Dalton Transactions
Page 9 of 11
View Article Online
DOI: 10.1039/C8DT01229F
Journal Name
ARTICLE
C27H37NO5Ti: C, 64.41; H, 7.41; N, 2.78; found: C 65.37; H 7.35; ArH), 4.49 (sept, 1H, CH), 4.03 (m, 3.5H, CH2), 3.82 (q, 1H, CH),
N 2.62.
3.40 (d, 0.75H, CH2), 3.32 (d, 0.75H, CH2), 1.56 (d, 3H, CH3),
Complex 12
1.49 (d, 9H, CH3), 1.44 (d, 9H, CH3), 1.29 (d, 9H, CH3), 1.26 (d,
1
The orange solid was dissolved in dry THF. Clear, orange 15H, CH3); H NMR (400 MHz, CDCl3 / DMAP): δ 8.26 (br, 2H,
crystals were obtained at –20 °C which were filtered, dried and ArH DMAP), 7.21 (d, 1H, ArH), 7.10 (br, 1H, ArH), 6.87 (m, 2H,
weighed. Yield: 0.5236 g (85.0 %). 1H NMR (400 MHz, CDCl3): δ ArH), 6.38 (d, 2H, ArH DMAP), 4.46 (br, 1H, CH), 3.94 (sept,
10.41 (d, 0.1H, C=O∙∙∙HOiPr), 7.38-6.55 (m, 4H, ArH), 5.45-2.34 0.5H, CH), 3.56 (m, 2H, CH2 / CH), 3.36 (d, 1H, CH2), 3.00 (d, 2H,
1
(m, 7H, CH / CH2), 1.83-0.67 (m, 42H, CH3); H NMR (400 MHz, CH2), 2.95 (s, 6H, CH3 DMAP), 1.48 (s, 9H, CH3), 1.37 (m, 12H,
CDCl3 / DMAP): δ 8.46 (d, 0.75H, ArH DMAP), 8.20 (br, 3.25H, CH3), 1.22 (s, 9H, CH3), 1.16 (s, 9H, CH3), 1.12 (d, 3H, CH3), 1.03
ArH DMAP), 7.19 (d, 2H, ArH), 6.83 (d, 2H, ArH), 6.41 (m, 4H, (d, 3H, CH3), 0.95 (d, 3H, CH3); 13C NMR (100.6 MHz, CDCl3 /
DMAP), 4.46 (sept, 0.75H, CH), 3.95 (sept, 1H, CH), 3.78 (d, 1H, DMAP): δ 176.7, 163.1 (br), 158.8, 155.0, 149.5 (br), 141.4,
CH2), 3.35 (s, 1H, CH2), 3.22 (br, 1H, CH2), 3.15 (d, 1H, CH2), 140.8, 136.3, 135.1, 124.7, 124.5 (br), 124.3 (br), 124.1, 123.6,
2.96 (m, 12H, DMAP), 1.45 (m, 18H, CH3), 1.21 (m, 18H, CH3), 123.3, 106.6, 79.6 (br), 65.1 (br), 64.3, 61.4, 53.0 (br), 39.1,
1
1.13 (d, 6H, CH3), 1.08 (d, 2.5H, CH3), 1.00 (d, 2.5H, CH3); H 35.2, 34.8, 34.3, 34.2, 31.7, 31.7, 29.9, 29.9, 25.4, 25.3, 9.4; MS
NMR (400 MHz, CD6SO): δ 7.12 (br, 2H, ArH), 7.06 (m, 0.4H, (EI): m/z calcd for C36H56NO5Ti: 630.4 [M+H]∙+; found 630.3;
ArH), 7.00 (m, 2H, ArH), 6.87 (m. 0.4H, ArH), 5.01 (br, 0.8H, elemental analysis calcd (%) for C36H55NO5Ti: C, 68.67; H, 8.80;
CH), 4.28 (m, 0.4H, CH2), 4.04 (s, 1.5H, CH2), 3.78 (sept, 2H, N, 2.22; found: C 69.94; H 8.61; N 2.19.
CH), 3.48 (m, 3H, CH2), 3.25 (m, 0.4H, CH2), 2.88 (d, 0.4H, CH2),
Solution polymerisation procedure
2.78 (br, 1,5H, CH2), 1.41 (m, 25.2H, CH3), 1.23 (m, 25.2H, CH3),
1.05 (m, 16.2H, CH3); 13C NMR (100.6 MHz, CDCl3 / DMAP): δ Under N2, a Schlenk flask was charged with rac-lactide (1.00 g,
175.0, 174.0, 160.1, 155.3, 154.7, 150.4, 149.4, 141.5, 141.3, 6.94 mmol) and an appropriate amount of catalyst. Dry
136.1, 135.3 (br), 125.0, 124.3, 124.2, 124.0, 123.6, 106.6, toluene (6.94 mL) was added and the vessel was heated in an
106.3, 80.9, 79.7, 64.3, 64.2, 63.8, 62.5 (br), 57.6, 39.2, 39.1, oil bath at 130 °C. Once the polymerisation time was reached,
35.2, 35.1, 34.3, 31.7, 29.9, 29.8, 25.4, 25.3; MS (MALDI): m/z the vessel was removed from the oil bath and the reaction
calcd for C35H53NO5Ti: 615.3 [M+H]+; found 615.3; elemental terminated by the addition of methanol (5 mL). The solvents
analysis calcd (%) for C35H53NO5Ti: C, 68.28; H, 8.68; N, 2.28; were removed and the resulting solids were dissolved in
found: C 69.78; H 9.00; N 2.20.
dichloromethane. The polymers were precipitated with excess
methanol, washed with methanol and dried at 40 °C under
vacuum.
Complex rac-13
The orange solid was dissolved in dry toluene. An orange
precipitate was obtained at –20 °C which was filtered, washed,
Melt polymerisation procedure
dried and weighed. Yield: 0.4857 g (93.9 %). 1H NMR (400 MHz, Under N2, a vessel was charged with sublimed rac-lactide (1.00
CDCl3): δ 7.19 (d, 1H, ArH), 7.15 (dd, 1H, ArH), 7.02 (d, 1H, g, 6.94 mmol) and appropriate amounts of catalyst. The vessel
ArH), 6.97 (d, 1H, ArH), 6.54 (t, 2H, ArH), 4.67 (sept, 1H, CH), was sealed and immersed in an oil bath at 130°C. Once
4.28 (d, 1H, CH2), 4.07 (sept, 1H, CH), 3.53 (d, 1H, CH2), 3.43 (d, polymerisation time was reached, the vessel was removed
1H, CH2), 3.23 (q, 1H, CH), 3.07 (d, 1H, CH2), 1.45 (d, 3H, CH3), from the oil bath and the reaction terminated by the addition
1
1.32 (m, 30H, CH3); H NMR (400 MHz, CDCl3 / DMAP): δ 8.19 of the methanol (5 mL). The solids were dissolved in
(br, 2H, ArH DMAP), 7.17 (dd, 1H, ArH), 7.12 (br, 1H, ArH), 7.00 dichloromethane and the polymers were dissolved with excess
(d, 1H, ArH), 6.97 (br, 1H, ArH), 6.77 (d, 1H, ArH), 6.59 (d, 1H, methanol. The collected polymers were washed with methanol
ArH), 6.40 (d, 2H, ArH DMAP), 4.73 (br, 1H, CH), 3.95 (sept, 1H, and dried at 40°C under vacuum.
CH), 3.63 (d, 1H, CH2), 3.48 (m, 3H, CH2 / CH), 3.10 (m, 1H,
CH2), 2.94 (s, 6H, CH3 DMAP), 1.22 (s, 9H, CH3), 1.18 (m, 12H,
CH3), 1.13 (d, 12H, CH3); 13C NMR (100.6 MHz, CDCl3 / DMAP):
δ 176.4, 164.2 (br), 159.6 (br), 154.7 (br), 149.6 (br), 142.7,
141.7, 126.4, 124.1, 123.1, 116.6, 115.5, 106.6, 80.7 (br), 77.3,
77.0, 76.7, 64.3 (br), 61.1, 39.0, 34.1, 34.0, 31.6, 31.6, 25.4,
24.6, 24.5, 9.1; MS (EI): m/z calcd for C31H47NO6Ti: 577.3
[M+HOiPr]∙+; found 577.5; m/z calcd for C28H40NO5Ti: 518.2
[M+H]∙+; found 518.2; elemental analysis calcd (%) for
C28H39NO5Ti: C, 64.99; H, 7.60; N, 2.71; found: C 66.53; H 7.87;
N 2.79.
Conflicts of interest
There are no conflicts to declare.
Acknowledgements
We are grateful to acknowledge financial support from Heriot-
Watt University and assistance from the EPSRC UK National
Mass Spectrometry Facility at Swansea University, UK. This
research used resources of the Advanced Light Source, which
is a DOE Office of Science User Facility under contract no. DE-
AC02-05CH11231.
Complex rac-14
The orange solid was dissolved in dry THF. An orange
precipitate was obtained at –20 °C which was filtered, washed,
dried and weighed. Yield: 0.5692 g (90.4 %). 1H NMR (400 MHz,
CDCl3): δ 7.21 (dd, 2H, ArH), 6.96 (d, 1H, ArH), 6.84 (d, 1H,
References
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 9
Please do not adjust margins

Please do not adjust margins
Dalton Transactions
Page 10 of 11
View Article Online
DOI: 10.1039/C8DT01229F
ARTICLE
Journal Name
1
2
P. J. Dijkstra, H. Z. Du and J. Feijen, Polym Chem-Uk, 2011,
520-527.
M. Shavit and E. Y. Tshuva, Eur. J. Inorg. Chem., 2008, 1467- 24 24. A. J. Chmura, M. G. Davidson, M. D. Jones, M. D. Lunn, M.
2
,
23 R. Baum and G. G. Smith, J. Am. Chem. Soc., 1986, 108, 7325-
7327.
1474; J.D. Ryan, K.J. Gagnon, S.J. Teat and R.D. McIntosh,
Chem. Commun., 2016, 52, 9071-9073
F. Mahon, A. F. Johnson, P. Khunkamchoo, S. L. Roberts and
S. S. F. Wong, Macromolecules, 2006, 39, 7250-7257.
3
4
5
6
7
B. T. Ko and C. C. Lin, J. Am. Chem. Soc., 2001, 123, 7973- 25 25. E. Y. Tshuva, M. Versano, I. Goldberg, M. Kol, H. Weitman
7977.
and Z. Goldschmidt, Inorg. Chem. Commun., 1999, 2, 371-
373.
Z. H. Liang, M. Zhang, X. F. Ni, X. Li and Z. Q. Shen, Inorg.
Chem. Commun., 2013, 29, 145-147.
D. C. Aluthge, E. X. Yan, J. M. Ahn and P. Mehrkhodavandi,
Inorg. Chem., 2014, 53, 6828-6836.
26 E. Y. Tshuva, I. Goldberg, M. Kol and Z. Goldschmidt, Inorg.
Chem., 2001, 40, 4263-4270.
27 G. A. Jeffrey, Crystallography Reviews, 2003, 9, 135-176.
K. M. Osten, D. C. Aluthge and P. Mehrkhodavandi, Dalton 28 D. T. Dugah, B. W. Skelton and E. E. Delbridge, Dalton Trans.,
Trans., 2015, 44, 6126-6139. 2009, 1436-1445.
F. M. Garcia-Valle, R. Estivill, C. Gallegos, T. Cuenca, M. E. G. 29 S. Barroso, A. M. Coelho, S. Gomez-Ruiz, M. J. Calhorda, Z.
Mosquera, V. Tabernero and J. Cano, Organometallics, 2015,
34, 477-487.
D. C. Aluthge, J. M. Ahn and P. Mehrkhodavandi, Chem Sci, 30 R. Neufeld and D. Stalke, Chem Sci, 2015, 6, 3354-3364.
Zizak, G. N. Kaluderovic and A. M. Martins, Dalton Trans.,
2014, 43, 17422-17433.
8
9
2015,
A. Thevenon, C. Romain, M. S. Bennington, A. J. P. White, H.
6, 5284-5292.
31 S. Bachmann, R. Neufeld, M. Dzemski and D. Stalke, Chem-
Eur J, 2016, 22, 8462-8465.
J. Davidson, S. Brooker and C. K. Williams, Angewandte 32 B. Gupta, N. Revagade and J. Hilborn, Prog. Polym. Sci., 2007,
Chemie-International Edition, 2016, 55, 8680-8685; Y. J. Kim 32, 455-482.
and J. G. Verkade, Macromol. Rapid Commun., 2002, 23, 917- 33 C. K. Williams, Chem. Soc. Rev., 2007, 36, 1573-1580.
921.
10 For select examples of organometallic aggregates see:
34 S. Slomkowski, S. Penczek and A. Duda, Polym. Adv. Technol.,
2014, 25, 436-447.
A.Arbaoui, C. Redshaw and D.L. Hughes, Chem. Commun. 35 T. Rosen, I. Goldberg, W. Navarra, V. Venditto and M. Kol,
2008, 39
Carpentier and E. Kirillov, Chem. Commun. 2013, 49
11692−11694; S.M. Kirk, H.C. Quilter, A. Buchard, L.H.
,
4717−4719; M. Normand, T. Roisnel, J.F.
Chem Sci, 2017.
36 D. Takeuchi, T. Nakamura and T. Aida, Macromolecules,
2000, 33, 725-729.
,
Thomas, G. Kociok-Kohn and M.D. Jones, Dalton Trans. 37 A. Sauer, A. Kapelski, C. Fliedel, S. Dagorne, M. Kol and J.
2016, 45, 13846−13852; F. Isnard, M. Lamberꢀ. L. Lettieri, I. Okuda, Dalton Trans., 2013, 42, 9007-9023.
D’Auria, K. Press, R. Troiano and M. Mazzeo, Dalton Trans. 38 Y. Ikada, K. Jamshidi, H. Tsuji and S. H. Hyon,
2016, 45, 16001-16010; Y. Gao, Z. Dai, J. Zhang, X. Ma, N. Macromolecules, 1987, 20, 904-906.
Tang and J. Wu, Inorg. Chem. 2014, 53, 716-726; X.Pang , 39 M. J. Stanford and A. P. Dove, Chem. Soc. Rev., 2010, 39, 486-
R.Duan, X.Li, C. Hu, X. Wang and X. Chen, Macromolecules,
2018, 51, 906-913.
494.
40 M. Bouyahyi, N. Ajellal, E. Kirillov, C. M. Thomas and J. F.
11 P. Dubois, C. Jacobs, R. Jerome and P. Teyssie,
Macromolecules, 1991, 24, 2266-2270.
12 A. Kowalski, A. Duda and S. Penczek, Macromolecules, 1998,
31, 2114-2122.
Carpentier, Chem-Eur J, 2011, 17
alternative method of assigning Pr see: M.J. Walton, S.J.
Lancaster, C. Redshaw, ChemCatChem, 2014, , 1892-1898.
, 1872-1883. For an
6
41 T. M. Ovitt and G. W. Coates, J. Am. Chem. Soc., 2002, 124
,
13 W. M. Stevels, M. J. K. Ankone, P. J. Dijkstra and J. Feijen,
Macromolecules, 1996, 29, 6132-6138.
1316-1326.
42 B. M. Chamberlain, M. Cheng, D. R. Moore, T. M. Ovitt, E. B.
14 B. J. O'Keefe, S. M. Monnier, M. A. Hillmyer and W. B.
Tolman, J. Am. Chem. Soc., 2001, 123, 339-340.
Lobkovsky and G. W. Coates, J. Am. Chem. Soc., 2001, 123,
3229-3238.
15 B. J. O'Keefe, L. E. Breyfogle, M. A. Hillmyer and W. B. 43 Y. Q. Cui, C. J. Chen, Y. Y. Sun, J. C. Wu and X. B. Pan, Inorg
Tolman, J. Am. Chem. Soc., 2002, 124, 4384-4393. Chem Front, 2017, , 261-269.
16 For other example of catalysts incorporating amino acids 44 Z. H. Tang, X. S. Chen, Y. K. Yang, X. Pang, J. R. Sun, X. F.
4
see: D.J. Darensbourg and O. Karroonnirun, Inorg. Chem.,
2010, 49, 2360-2371; Organometallics 2010, 29, 5627-5634;
Zhang and X. B. Jing, J Polym Sci Pol Chem, 2004, 42, 5974-
5982.
D.J. Darensbourg, O. Karroonnirun and S.J. Wilson, Inorg. 45 H. B. Wang, Y. Yang and H. Y. Ma, Macromolecules, 2014, 47
Chem., 2011, 50, 6775-6787; R. L. Webster, RSC Adv., 2014, 7750-7764.
4, 5254-5260; Y. F. Al-Khafaji, T.J. Prior, L. Horsburgh, M.R.J. 46 A. Pilone, N. De Maio, K. Press, V. Venditto, D. Pappalardo,
,
Elsegood, C. Redshaw, ChemistrySelect, 2017, 2, 759-768; Y.
F. Al-Khafaji, M.R.J. Elsegood, J.W.A. Frese, C. Redshaw, RSC
Adv., 2017, 7, 4510-4517.
M. Mazzeo, C. Pellecchia, M. Kol and M. Lamberti, Dalton
Trans., 2015, 44, 2157-2165.
47 T. Biela, A. Duda and S. Penczek, Macromol. Symp., 2002,
183, 1-10.
17 A. S. Ceccato, A. Neves, M. A. De Brito, S. M. Drechsel, A. S.
Mangrich, R. Werner, W. Haase and A. J. Bortoluzzi, J. Chem. 48 E. Safaei, H. Sheykhi, A. Wojtczak, Z. Jaglicic and A.
Soc., Dalton Trans., 2000, 1573-1577; Y.-L. Wong, Y. Yan, E. S. Kozakiewicz, Polyhedron, 2011, 30, 1219-1224.
H. Chan, Q. Yang, T. C. W. Mak and D. K. P. Ng, J. Chem. Soc., 49 T. Weyhermuller, R. Wagner and P. Chaudhuri, Eur. J. Inorg.
Dalton Trans., 1998, 3057-3064.
18 Z. Taheri, B. Ghanbari and H. Hajibabaei, Chem. Pap., 2014,
68, 989-994.
Chem., 2011, 2547-2557.
19 J. M. E. P. Cols, C. E. Taylor, K. J. Gagnon, S. J. Teat and R. D.
McIntosh, Dalton Trans., 2016, 45, 17729-17738.
20 S. Barroso, P. Adao, A. M. Coelho, J. C. Pessoa and A. M.
Martins, J. Mol. Catal. A: Chem., 2016, 412, 107-116.
21 G. G. Smith and T. Sivakua, J. Org. Chem., 1983, 48, 627-634.
22 E. D. Stroud, D. J. Fife and G. G. Smith, J. Org. Chem., 1983,
48, 5368-5369.
10 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 11 of 11
Dalton Transactions