Synthesis and characterisation of unsymmetrical Zr(iv) amine tris(phenolate) complexes and their application in ROP of rac-LA

Article abstract of DOI:10.1039/c4dt01260g

Unsymmetrical amine tris(phenolate) ligands have been prepared and complexed to Zr(iv). In the solid-state dimeric species are observed, which persist in solution. The complexes show interesting coordination chemistry and the nature of which is discussed. These complexes are analogues of a highly stereoselective C₃-symmetric Zr(iv) amine tris(phenolate) alkoxide and have been prepared, characterised and trialled for the ring-opening of rac-lactide providing insight into the effect of symmetry and ligand steric bulk on selectivity. The polymerisations have been investigated in solution and under the industrially preferred melt conditions (130 °C). The rate law for the ROP has been determined-with Zr₂(1)₂(O ⁱPr)₂, a pseudo first order dependency was seen, however for Zr₂(2)₂(OⁱPr)₂ this changed to a half order dependency in initiator. This journal is the Partner Organisations 2014.

Full text of DOI:10.1039/c4dt01260g

Dalton
Transactions
View Article Online
View Journal | View Issue
PAPER
Synthesis and characterisation of unsymmetrical
Zr(IV) amine tris(phenolate) complexes and their
application in ROP of rac-LA†
Cite this: Dalton Trans., 2014, 43,
12095
Thomas R. Forder,a,b Mary F. Mahon,a Matthew G. Davidson,a Timothy Woodmanc
and Matthew D. Jones*a
Unsymmetrical amine tris(phenolate) ligands have been prepared and complexed to Zr(IV). In the solid-
state dimeric species are observed, which persist in solution. The complexes show interesting coordi-
nation chemistry and the nature of which is discussed. These complexes are analogues of a highly stereo-
selective C3-symmetric Zr(IV) amine tris(phenolate) alkoxide and have been prepared, characterised and
trialled for the ring-opening of rac-lactide providing insight into the effect of symmetry and ligand steric
bulk on selectivity. The polymerisations have been investigated in solution and under the industrially
preferred melt conditions (130 °C). The rate law for the ROP has been determined – with Zr2(1)2(OiPr)2,
a pseudo first order dependency was seen, however for Zr2(2)2(OiPr)2 this changed to a half order depen-
dency in initiator.
Received 29th April 2014,
Accepted 24th June 2014
DOI: 10.1039/c4dt01260g
www.rsc.org/dalton
controlled ROP of cyclic esters.18 For example, Group 4 amine
tris(phenolate) complexes have received considerable interest
Introduction
The controlled ring-opening polymerisation (ROP) of rac- for their structure and reactivity for ROP of rac-LA.19–22 A C3-
lactide (rac-LA) through the use of stereo-selective metal alkox- symmetric Zr(IV) amine tris(phenolate) alkoxide has previously
ides has received continued interest due to the ability to been shown to be extremely active for the controlled ROP of
produce polymeric material with well-defined thermal and rac-LA, yielding highly heterotactic polylactide.19 This metal
mechanical properties.1 Such control provides one route to complex has been shown to have a C3-symmetric propeller-like
widening the potential applications of this environmentally arrangement similar to that for other amine tris(phenolate)
favourable renewable plastic.2–6 The polymer itself (PLA) has complexes and can undergo axial flipping between the P and
found many uses from packing material to high value bio- M enantiomers.21,23 Complexes with such ligands are prevalent
medical applications.7 There are numerous examples in the in the literature;24–26 however there are no examples of un-
literature of discrete metal complexes that are efficient (in symmetrical variants.
terms of molecular weight control and tacticity). Examples of
stereo-selective initiators vary significantly in both metal
centre and ligand architecture.8–15 Whilst recent publications
Results and discussion
Metal complex synthesis
have shown that for some metal–ligand systems it is possible
to switch from isotactically- to heterotactically-enriched poly-
lactide by altering ligand structure,16,17 it is on the whole
To further understand the influence of initiator symmetry and
difficult to predict selectivity without potentially arduous and
steric bulk of the ligand, a systematic alteration of ligand archi-
time consuming synthesis and polymerisation trials. Group
tecture was carried out to remove symmetry and provide step-
4 metals initiators are an important class of complexes for the
wise removal of the sterically-bulky substituents about the
metal centre. Unlike the one-pot synthesis of the aforemen-
aDepartment of Chemistry, University of Bath, Claverton Down, Bath BA2 7AY, UK.
E-mail: mj205@bath.ac.uk; Fax: +44 (0)1225 386231; Tel: +44 (0)1225 384908
bDoctoral Training Centre in Sustainable Chemical Technologies, University of Bath,
Bath BA2 7AY, UK
cDepartment of Pharmacy and Pharmacology, University of Bath, Bath BA2 7AY, UK
†Electronic supplementary information (ESI) available: Full characterisation and
experimental procedures. CCDC 959185 and 959186. For ESI and crystallo-
graphic data in CIF or other electronic format see DOI: 10.1039/c4dt01260g
tioned amine tris(phenolate) via a modified Mannich reac-
tion,21 one unsymmetrical analogue (H31) had to be accessed
through the utilisation of a salicylammonium salt and a 3,5-
disubstituted salicylbromide. Synthesis of amine tris(phenol-
ate) ligands with further removal of steric bulk or a fully
unsymmetrical structure (H32–3) required utilising a benzyl-
protected imine bis(phenolate) intermediate.† Zr(OiPr)4-
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 12095–12099 | 12095

View Article Online
Paper
Dalton Transactions
meric complex, the nitrogen donor and alkoxide remain in an
axial orientation.19 Contrary to the monomeric species, the
dimer motif is not fluxional and the rigid structure results in
the six protons of the methylene bridges of each ligand
1
appearing as discrete diastereotopic doublets in the H NMR
1
spectrum.† When the H NMR spectrum for Zr2(1)(OiPr)2 was
run in d8-THF, at an elevated temperature (333 K), two doub-
lets and a singlet were observed for the methylene bridges,
indicating that the complex can be broken down in a coordi-
nating solvent to form a monomeric species.† Interestingly,
from solution-state DOSY NMR spectroscopic analysis the
hydrodynamic volume of Zr2(1)2(OiPr)2 was 635 Å3 (cf. 505 Å3
for the C3 symmetric monomer complex19 and 606 Å3
for Zr2(3)2(OiPr)2 – in d8-Tol). When the DOSY was recoded in
d8-THF, which is believed to break the dimer into a monomer,
for Zr2(1)2(OiPr)2 the hydrodynamic volume is 493 Å3, ana-
logous to the known monomeric species.
Discerning dimer motifs
For Zr2(2)2(OiPr)2 and Zr2(3)2(OiPr)2 analysis of the 1H NMR
spectra indicated the presence of multiple sets of diastereo-
topic doublets, which presumably are due to three possible
isomers for this motif (Fig. 2). With a restriction that an
unsubstituted phenolate would adopt the bridging role, the
dimer can consist of an enantiomeric pair (Λ,Δ) that requires
the tBu-substituted phenolate arms to adopt a transoid
Scheme 1 Synthesis of unsymmetrical Zr(IV) amine tris(phenolate)
complexes and ligands.
t
arrangement across the dimer (Fig. 2a and b). For the Bu-sub-
stituted arms to be arranged cisoid the dimer must consist of
a Λ,Λ or Δ,Δ pairing (Fig. 2c).
Geometry can be further distinguished by the arrangement
t
of the Bu-substituted phenolate arm with respect to the brid-
ging phenolate arm of the same ligand. This leads to the
resolution of two forms of the transoid dimer (Fig. 2a and b)
whereas the cisoid dimer arrangement requires there to always
be a trans/cis pairing. Through variable temperature (208 K–
358 K) 1H NMR spectroscopy it was determined that these
isomers are not exchanging, providing further evidence of the
non-fluxional nature of the dimer motif.† We have further pre-
pared H32 and subsequent Zr2(2)2(OiPr)2 (Fig. 3) in which each
arm of the tris(phenolate) is differently substituted.
Fig. 1 Molecular structure for Zr2(1)2(OiPr)2: hydrogen atoms, lattice
solvent and ligand disorder have been removed for clarity.
(HOiPr) was coordinated with 1 eq. of H31–3 to yield dimeric
complexes Zr2(1)2(OiPr)2, Zr2(2)2(OiPr)2 and Zr2(3)2(OiPr)2
respectively (Scheme 1).
The dimer motif was observed in the solid-state with the
less-sterically demanding phenolate acting as
a bridge
between two metal centres (Fig. 1), the NMR spectra (in non-
coordinating solvents) indicates that this structure is retained
in solution.† Both Zr(IV) centres are six-coordinate and adopt a
pseudo-octahedral geometry. As with the C3-symmetric mono-
Fig. 2 Possible isomers for the dimer Zr2(3)2(OiPr)2. Methylene bridges,
phenolates and substituents have been pictorially represented for clarity.
tBu annotation indication which phenolate arm is di-tBu-substituted.
Only one enantiomer for each isomer is shown for clarity.
12096 | Dalton Trans., 2014, 43, 12095–12099
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
Fig. 4 First order plots for ln([LA]o/[LA]t) versus time using
Zr2(1–3)2(OiPr)2 and the C3-symmetric analogue (toluene-d8, 353 K,
[LA]0 = 0.50 M, [M]/[I] = 100).
pared to the C3-symmetric analogue, molecular weight control
was lost. MALDI-ToF MS of the polymer from Zr2(3)2(OiPr)2
indicated the presence of the H- and OiPr end groups, as
expected from the coordination insertion mechanism.†
Fig. 3 Molecular structure for Zr2(2)2(OiPr)2: hydrogen atoms, lattice
solvent and ligand disorder have been removed for clarity.
Solvent-based kinetic studies for the ROP of rac-LA were
carried out in toluene-d8 at 253
K
for complexes
Zr2(1–3)2(OiPr)2 and compared with the C3-symmetric ana-
logue. All polymerisations were found to be first order with
respect to monomer (Fig. 4). Under the conditions used a first
order rate constant (kapp) of 3.0 × 10−2 min−1 was found for
the C3-symmetric Zr(IV) amine tris(phenolate) complex (cf.
1.4 × 10−2 min−1 for L-LA). A larger rate constant of 13.0 × 10−2
min−1 was observed for Zr2(1)2(OiPr)2 whilst pseudo first-order
plots for the polymerisation of rac-LA initiated by Zr2(2)2(OiPr)2
and Zr2(3)2(OiPr)2 identified relatively lower rate constants of
0.86 × 10−2 and 0.41 × 10−2 min−1 respectively. Compared to
Polymerisation of rac-lactide
Complexes Zr2(1–3)2(OiPr)2 were tested for the ROP of rac-LA
under solvent free conditions and compared with that
previously reported for the C3-symmetric analogue (Table 1).18
All initiators showed high activity with comparable conversion
to the C3-symmetric analogue being achieved in under
30 minutes under industrially-relevant melt conditions.
Analysis of the polymer microstructure by homonuclear
decoupled 1H NMR spectroscopy revealed that heterotactic-
enriched PLA was obtained for Zr2(1–3)2(OiPr)2 with no signifi-
cant difference in selectivity between each initiator (Pr ≈ 0.7,
Table 1). Polymeric material produced in solution (toluene,
0.5 M) at a lower reaction temperature was found to have
similar or lower heterotactic enrichment. Analysis of polymeric
material using size-exclusion chromatography (SEC) found
that polymer weight distribution (Đ) increased with further
removal of sterically-bulky ligand substituents and that, com-
t
the C3-symmetric complex, removal of the Bu substituents
from one arm of the amine tris(phenolate) ligand to yield
Zr2(1)2(OiPr)2 increases the relative rate of polymerisation of
rac-LA. Further removal of steric bulk to Zr2(2)2(OiPr)2 and sub-
sequently Zr2(3)2(OiPr)2 leads to a lowering in rate below that
observed for the C3-symmetric analogue (Fig. 4).
It was deemed reasonable to assume that the ROP-active
species of Zr2(1–3)2(OiPr)2 would not be dimeric; due to the
lactide coordination step required in both initiation and
propagation.27 To further investigate the polymerisation
kinetics and mechanism kapp was determined for a variety of
initiator concentrations and a plot of ln kapp vs. ln[I]0 was used
to determine the order of the process with respect to the
initiator, and hence the rate law –d[LA]/dt = k[LA]1[I]x (Fig. 5).
Gradients of each linear plot indicated that, as expected, order
with respect to the monomeric C3-symmetric zirconium
initiator to be pseudo first-order (x = 1.05). The order with
respect to initiator for Zr2(1)2(OiPr)2 was also essentially first
order (x = 1.22), whilst for Zr2(3)2(OiPr)2 it was found to be of
half order dependency (x = 0.43).
Table 1 Polymerisation data for rac-LA with initiators Zr2(1–3)2(OiPr)2
d
e
Initiator
Time (h)
Conv.c (%)
Mn
Đd
Pr
C3-symm. Zr18
0.1
0.5
0.5
0.5
1
3
3
78
76
80
87
94
65
50
32 300
66 900
19 100
41 900
22 500
20 300
9750
1.22
1.40
1.54
1.76
1.54
1.08
1.08
0.96
0.72
0.71
0.70
0.64
0.65
0.69
Zr2(1)2(OiPr)2 a
Zr2(2)2(OiPr)2 a
Zr2(3)2(OiPr)2 b
Zr2(1)2(OiPr)2 b
Zr2(2)2(OiPr)2 b
Zr2(3)2(OiPr)2
a
Conditions: a [M]/[I] = 300, 130 °C, solvent free. b Solution (toluene)
[M]/[I] = 100, 80 °C ([M]0 = 0.5 mol dm−3). c As determined via 1H NMR
spectroscopy. d Determined from GPC (in THF) referenced to
polystyrene. e Pr is the probability of racemic enchainment, calculated
from the 1H homonuclear decoupled NMR spectra.
1H NMR spectroscopic studies of Zr2(1)2(OiPr)2 in a coordi-
native solvent (THF-d8) at an elevated temperature of 333 K,
found that the AX spin system for the diastereotopic protons
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 12095–12099 | 12097

View Article Online
Paper
Dalton Transactions
It is also possible that removal of the steric bulk for these
amine tris(phenolate) complexes would increase the rate of
polymerisation due to the greater ease of coordination of
lactide.15,30 Whilst this is the case for Zr2(1)2(OiPr)2 it is not
the case for Zr2(2–3)2(OiPr)2 as activity is reduced. Half order
dependency has been previously reported by Mountford et al.
for a dimeric zinc species and attributed to aggregation a cata-
lytically active monomeric species during the ROP of ε-capro-
lactone.31 Whilst for the previously published zinc species, the
dimer–monomer equilibrium is caused by the benzyl alcohol
co-initiator, herein we propose that dissociation is caused
by the coordination of the lactide monomer. For both
Zr2(1)2(OiPr)2 and Zr2(3)2(OiPr)2, the dimer is disrupted and a
propagating lactidyl alkoxide species generated; the less-steri-
cally hindered Zr2(3)2(OiPr)2 can aggregate back to a ROP-in-
active dimer whilst Zr2(1)2(OiPr)2 does not. The observed trend
in kinetics can be attributed to the relative reduction in steric
bulk in close proximity to the metal centre (C3-symm. Zr →
Zr2(1)2(OiPr)2) coupled with the potential of the propagating
species to aggregate (Zr2(1)2(OiPr)2 → Zr2(3)2(OiPr)2).
Fig. 5 ln–ln plots for determining reaction order with respect to
initiator for Zr2(1)2(OiPr)2, Zr2(3)2(OiPr)2 and the C3-symmetric analogue
(toluene-d8, 353 K, [LA]0 = 0.50 M, [M]/[I] = 50, 100, (150), 200, 400).
of the methylene bridges are no longer observed. This is attrib-
uted to dissociation of the dimer species due to coordination
of THF to the metal centre.† Furthermore, whilst heating to
333 K appeared to disrupt the AX spin system, upon cooling
the system was not re-established suggesting, for Zr2(1)2(OiPr)2
at least, a barrier to both coordination and reformation of the
dimer. The ROP of lactide via a coordination-insertion method
requires the initial coordination of the lactide to the metal
centre. We therefore suggest that the ROP-active form of these
dimer complexes is a discreet monomeric five-coordinate
complex, akin to that observed for the C3-symmetric analogue.
Conclusions
In summary we present a novel step-wise synthesis for the pro-
duction of unsymmetrical amine tris(phenolate) ligands that
when coordinated to Zr(IV) centres form dimeric complexes.
Whilst interesting in their own right, they have been shown to
be active for the ROP of lactide and have provided insight into
the highly heterotactic selectivity previously published for a
C3-symmetric Zr(IV) amine tris(phenolate).19 The investigation
of reaction order with respect to initiator provides an interest-
ing case study of the need to understand the potential nature
and behaviour of ROP-active species when considering relative
rates of polymerisation. Work is underway to utilise the step-
wise ligand synthesis methodology to access Zr(IV) amine tris
(phenolate) complexes that disrupt the C3-symmetry via the
increase in steric bulk of the phenolate substituents.
Implications for reactivity and stereochemistry
The high heteroselectivity previously reported for the C3-sym-
metric Zr(IV) amine tris(phenolate) complex has been attribu-
ted to dynamic enantiomorphic site control due to “flipping”
of the propeller geometry about the C3 axis.19,28,29
Zr2(1)2(OiPr)2 offers an active species that removes the C3-sym-
metry yet maintains considerable steric bulk about the metal
centre. A significant drop in selectivity is observed with this
removal of C3-symmetry, whilst further removal of steric bulk
has no significant further effect on the stereoselectivity
{Zr2(1)2(OiPr)2 vs. Zr2(3)2(OiPr)2}. These observations provide
further support for the origin of stereoselectivity in ROP of
lactide by C3-symmetric complexes. The lack of higher stereo-
selectivity for Zr2(1)2(OiPr)2 and Zr2(3)2(OiPr)2 at a lower temp-
erature in solution suggests the preferable enantiomeric
pairing of the P/M axial chirality and RR/SS chain-end stereo-
chemistry as previously presented in the literature29 is not
present for these unsymmetrical analogues.
Acknowledgements
We wish to thank the EPSRC (EP/G03768X/1) for funding.
Notes and references
1 R. Auras, B. Harte and S. Selke, Macromol. Biosci., 2004, 4,
835–864.
Polymeric material produced through the ROP of rac-LA
initiated by Zr2(1–3)2(OiPr)2 has a higher polydispersity than
that observed for the C3-symmetric analogue. It is proposed
that this is a combination of the reduced steric bulk about the
metal centre allowing for an increased rate in undesirable
2 M. Vert, Biomacromolecules, 2005, 6, 538–546.
3 O. Dechy-Cabaret, B. Martin-Vaca and D. Bourissou, Chem.
Rev., 2004, 104, 6147–6176.
4 C. K. Williams and M. A. Hillmyer, Polym. Rev., 2008, 48,
1–10.
5 N. Ajellal, J.-F. Carpentier, C. Guillaume, S. M. Guillaume,
M. Helou, V. Poirier, Y. Sarazin and A. Trifonov, Dalton
Trans., 2010, 39, 8363–8376.
transesterification15 reactions and
a
potentially slower
initiation step of the polymerisation due to the requirement
for the dimer to dissociate into the active monomeric species.
12098 | Dalton Trans., 2014, 43, 12095–12099
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
6 R. J. Pounder and A. P. Dove, Polym. Chem., 2010, 1, 260– 19 A. J. Chmura, M. G. Davidson, C. J. Frankis, M. D. Jones
271. and M. D. Lunn, Chem. Commun., 2008, 1293–1295.
7 A.-C. Albertsson and I. K. Varma, Biomacromolecules, 2003, 20 Y. Kim, G. K. Jnaneshwara and J. G. Verkade, Inorg. Chem.,
4, 1466–1486. 2003, 42, 1437–1447.
8 J. Börner, U. Flörke, A. Döring, D. Kuckling, M. D. Jones, 21 M. Kol, M. Shamis, I. Goldberg, Z. Goldschmidt, S. Alfi and
M. Steiner, M. Breuning and S. Herres-Pawlis, Inorg. Chem.
Commun., 2010, 13, 369–371.
E. Hayut-Salant, Inorg. Chem. Commun., 2001, 4, 177–
179.
9 B. M. Chamberlain, M. Cheng, D. R. Moore, T. M. Ovitt, 22 S. Gendler, S. Segal, I. Goldberg, Z. Goldschmidt and
E. B. Lobkovsky and G. W. Coates, J. Am. Chem. Soc., 2001,
123, 3229–3238.
10 M. H. Chisholm, J. C. Gallucci and K. Phomphrai, Inorg.
Chem., 2004, 43, 6717–6725.
11 R. A. Collins, J. Unruangsri and P. Mountford, Dalton
Trans., 2013, 42, 759–769.
12 S. L. Hancock, M. D. Jones, C. J. Langridge and
M. F. Mahon, New J. Chem., 2012, 36, 1891–1896.
M. Kol, Inorg. Chem., 2006, 45, 4783–4790.
23 M. G. Davidson, C. L. Doherty, A. L. Johnson and
M. F. Mahon, Chem. Commun., 2003, 1832–1833.
24 S. Groysman, I. Goldberg, Z. Goldschmidt and M. Kol,
Inorg. Chem., 2005, 44, 5073–5080.
25 S. Groysman, I. Goldberg, M. Kol, E. Genizi and
Z. Goldschmidt, Adv. Synth. Catal., 2005, 347, 409–
415.
13 P. Hormnirun, E. L. Marshall, V. C. Gibson, A. J. P. White 26 G. Licini, M. Mba and C. Zonta, Dalton Trans., 2009, 5265–
and D. J. Williams, J. Am. Chem. Soc., 2004, 126, 2688–2689. 5277.
14 J. Okuda and J. C. Buffet, Chem. Commun., 2011, 47, 4796– 27 C. K. Williams, Chem. Soc. Rev., 2007, 36, 1573–1580.
4798.
28 H. Ma, T. P. Spaniol and J. Okuda, Angew. Chem., Int. Ed.,
2006, 45, 7818–7821.
29 A. J. Chmura, C. J. Chuck, M. G. Davidson, M. D. Jones,
M. D. Lunn, S. D. Bull and M. F. Mahon, Angew. Chem., Int.
Ed., 2007, 46, 2280–2283.
30 E. L. Whitelaw, M. D. Jones and M. F. Mahon, Inorg. Chem.,
2010, 49, 7176–7181.
31 Y. Huang, W. Wang, C.-C. Lin, M. P. Blake, L. Clark,
A. D. Schwarz and P. Mountford, Dalton Trans., 2013, 42,
9313–9324.
15 E. L. Whitelaw, M. G. Davidson and M. D. Jones, Chem.
Commun., 2011, 47, 10004–10006.
16 C. Bakewell, T.-P.-A. Cao, N. Long, X. F. Le Goff, A. Auffrant
and C. K. Williams, J. Am. Chem. Soc., 2012, 134, 20577–
20580.
17 A. Stopper, J. Okuda and M. Kol, Macromolecules, 2012, 45,
698–704.
18 A. Sauer, A. Kapelski, C. Fliedel, S. Dagorne, M. Kol and
J. Okuda, Dalton Trans., 2013, 42, 9007–9023.
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 12095–12099 | 12099