Synthesis and structures of calcium and strontium 2,4-di-tert- butylphenolates and their reactivity towards the amine co-initiated ring-opening polymerisation of rac-lactide

[Sr₃(μ-DBP)₆(THF)₆] (3), and [Sr₂(DBP)(μ-DBP)₃(DME)₃] (4). Compound 1 is a six coordinate trans-octa-

hedral monomer, whereas in binuclear 2 two seven-coordinate Ca centres are bridged by a DME ligand.

In 3 a central Sr is connected by three bridging DBP groups to each of two terminal Sr(THF)₃moieties,

all metal atoms being six coordinate. Compound 4 has one six- and one seven-coordinate Sr, bridged by

three DBP ligands, the former Sr also having a terminal DBP and a bidentate DME ligand and the

latter two DME ligands. Complexes 2 and 4 act as ring-opening polymerisation (ROP) catalysts for the

benzyl alcohol or benzylamine co-initiated ROP rac-lactide forming atactic alcohol- or amine-terminated

polylactide H-[PLA]-XBn (X = O or NH) with reasonable control of molecular weight via an activated

Received 8th January 2013,

Accepted 13th February 2013

monomer propagation mechanism. Kinetic studies for BnNH₂found the unusual rate expression

−d[LA]/dt = k_p(Ae)[2 or 4]₀[rac-LA]²[BnNH₂]₀

(k_p(Ca)≈ 1.7 × k_p(Sr)). Preliminary studies suggest that

2.5

DOI: 10.1039/c3dt00065f

[Y(DBP)₃(THF)₂] also catalyses amine or alcohol co-initiated ROP by an activated monomer mechanism

without loss of a phenoxide ligand.

www.rsc.org/dalton

have attracted recent attention.^2fAlcohols can function as

chain transfer agents (CTAs) with metal catalysts leading to the

Introduction

Poly(lactide) (PLA) derived plastics are important renewable, “immortal” ROP (iROP) of LA in which one H-[PLA]-OR

biocompatible and biodegradable materials.¹Studies of the polymer chain is formed per added ROH and grows continu-

ring-opening polymerisation (ROP, the preferred method of ally through the ROP process.^2sFor metal complex-mediated

PLA synthesis from both an industrial and academic perspec- ROP, in the absence of any added alcohol CTA, monomer

tive) of the cyclic diester, lactide (LA) has attracted sustained enchainment almost always proceeds via the “coordination–

interest over the last 15 years in particular.²Metal alkoxides of insertion” mechanism (repeated insertion of LA into the

the type (L)M-OR (either previously isolated or generated growing M-O_PLAchain of a metal complex). In the presence of

in situ; L = supporting ligand set) are normally considered to an alcohol CTA under iROP conditions, propagation usually

be the optimum initiators since these generally give similar proceeds via coordination–insertion, but this is accompanied

rates of initiation (rate constant k_i) and propagation (rate con- by fast exchange (rate constant k_CTwhere k_CT≫ k_p) between

stant k_p). Similarly, alcohol co-initiators with organic catalysts the metal-bound polymeryl chain and a free HO-terminated

PLA chain. However, it is also possible that in the presence of

ROH, propagation can proceed via the “activated monomer”

^aChemistry Research Laboratory, Department of Chemistry, University of Oxford,

mechanism in which ROH (where ROH is either the initial

Mansfield Road, Oxford OX1 3TA, UK. E-mail: philip.mountford@chem.ox.ac.uk

^bSchool of Chemistry, Monash University, Clayton, Vic. 3800, Australia.

E-mail: glen.deacon@monash.edu

added alcohol or a HO-terminated PLA chain) externally

attacks a metal-bound LA moiety with the metal acting as a

Lewis acid catalyst.^2s,3Mechanistic studies can in principle

help distinguish between these possibilities.

Whereas there are numerous well-established examples of

the synthesis of RO-terminated PLAs (H-[PLA]-OR) (with or

without the use of CTAs), there are very few, good ROP-based

^cSchool of Pharmacy and Molecular Sciences, James Cook University, Townsville,

Qld 4811, Australia. E-mail: peter.junk@jcu.edu.au

†Electronic supplementary information (ESI) available: Further details of the

ROP experiments and NMR tube scale experiments. CCDC 914477–914480. For

ESI and crystallographic data in CIF or other electronic format see DOI:

10.1039/c3dt00065f

This journal is © The Royal Society of Chemistry 2013

Dalton Trans.

View Article Online

Paper

Dalton Transactions

routes to amine (NRR′)-terminated PLA. Such methodologies mechanistic studies were undertaken,⁸the propagation

could allow the use of functionalised amines and poly(amines) process was established as rate-limiting coordination–inser-

for the preparation of complex architectures such as block- or tion coupled with fast (non rate-limiting) exchange between

graft-copolymers, dendritic copolymers and star-shaped poly- the free H-[PLA]-NRR′ moieties and the metal-bound ones of

mers.^1hMetal amides (L)M-NR₂often give poor control of the propagating species (L)M-[PLA]-NRR′ (i.e., iROP). Specifi-

molecular weight due to a mismatch of the rates of initiation cally, while the number average molecular weights (M_n) of the

and propagation, and a propensity to form cyclic PLAs.⁴Way- PLAs formed in Scheme 1 are controlled by the initial

mouth et al. have reported a stoichiometric organocatalytic [lactide]₀: [RNH₂]₀ratio (one chain per R′RNH), the rates of

method based on the insertion of carbenes into amine N–H propagation are independent of [RNH₂]₀(i.e., the concentration

bonds at 90 °C giving 85% conversion to PLA after 71 h.⁵of polymeryl chains, [H-[PLA]-NRR′]).

Sn(octanoate)₂was employed at high temperature under melt

As part of these studies we recently described the synthesis,

conditions (up to 150 °C and 72 h) to form PLA-functionalised structures and ROP behaviour of the lanthanoid(^III) 2,4-di-tert-

amine-based dendrimers and star-polymers.⁶In this case one butylphenolate complexes [Ln(DBP)₃(THF)₃] (Ln = Nd, Gd; DBP

PLA chain forms per Sn–Oct bond and there is extensive = –OC₆H₃Bu^t₂-2,4).¹⁰These complexes on their own were fast

transesterification.⁷

but poorly-controlled initiators for the ROP of rac-LA. However,

In a recent advance in this area, as shown in Scheme 1, we use of an excess of BnNH₂aﬀorded well-controlled, amine co-

recently reported that yttrium dicationic (I), zwitterionic (II) initiated iROP (analogous results were obtained with BnOH).

and neutral amide (III) compounds can act as catalyst precur- The very fast rates of propagation and the paramagnetic nature

sors for the amine co-initiated iROP of rac-lactide using of the complexes precluded a detailed study, and a coordi-

primary or secondary amines.⁸For compounds II and III, nation–insertion mode of propagation was proposed based on

amine-terminated, highly heterotactic PLA of the type H-[PLA]- our mechanistic studies for I–III (Scheme 1) and previous lit-

NHRR′ with narrow polydispersities and well-controlled mole- erature where alcohol co-initiators were employed with Ln-

cular weights were rapidly prepared at RT. Subsequently, we (OAr)₃systems.¹¹

reported that several trivalent indium compounds were also

During their recent studies of the alcohol co-initiated ROP

eﬀective for this process, although these required longer reac- of lactide using cationic main group catalysts supported by

tion times at RT (up to 1 week) or elevated temperatures.⁹phenolate-poly(ether) ligands,³Carpentier and Sarazin briefly

Mechanistic studies^8,9established the general mechanism in reported another example of BnNH₂co-initiated ROP for a zinc

Scheme 1 whereby a number of equivalents of LA are rapidly system. Mechanistic studies were not carried out for the amine

ring-opened by a primary or secondary amine RR′NH₂(RR′NH co-initiated ROP, but we were interested to note that detailed

is typically benzylamine, BnNH₂, in most mechanistic studies analysis of the corresponding alcohol co-initiated systems

to date) via an activated monomer-type process to form an showed that in these divalent systems iROP proceeded by

amine-terminated alcohol IV which eventually forms amine- activated monomer propagation. Similarly, while this manu-

terminated PLA of the type H-[PLA]-NRR′. Where the relevant script was in preparation, Miller et al. proposed an activated

Scheme 1 Top: the ﬁrst metal complexes (I–III) for fast and well-controlled amine co-initiated iROP of rac-LA. Bottom: general mechanism of initiation and

propagation.⁸

Dalton Trans.

This journal is © The Royal Society of Chemistry 2013

View Article Online

Dalton Transactions

Paper

monomer ROP process for alcohol co-initiated ROP catalysed in complexes compared with the more commonly used sym-

by 2,6-di-tert-butyl-substituted magnesium and calcium metrical 2,6-disubstituted phenolates.

phenolates although no kinetic analysis was presented in this

case.¹²

The complexes [Ca(DBP)₂(THF)₄] (1), [Ca₂(DBP)₄(DME)₄-

(μ-DME)] (2), [Sr₃(μ-DBP)₆(THF)₆] (3), and [Sr₂(DBP)(μ-DBP)₃-

We anticipated that extension of the trivalent lanthanoid (DME)₃] (4) were prepared by RTP reactions between an excess

complexes [Ln(DBP)₃(THF)₃] to their divalent Ln^IIcounterparts of freshly filed alkaline earth metal, Hg(C₆F₅)₂and 2,4-di-

would potentially be complicated by redox reactions as found tert-butylphenol (H-DBP) in tetrahydrofuran (THF) or 1,2-

previously.¹³On the other hand, use of alkaline earth (Ae) dimethoxyethane (DME) in the presence of a drop of added

metal 2,4-di-tert-butylphenolates of the type [Ae(DBP)₂(L)_n] mercury for activation of the metal through surface amalgama-

(specifically for Ae = Ca and Sr since Ca²⁺/Yb²⁺and Sr²⁺/Eu²⁺tion (Scheme 2). After a simple workup of filtration to remove

have similar ionic radii.¹⁴) should provide access to their be- mercury and excess Ae metal and evaporation to the point of

haviour without any accompanying redox complications. In crystallization, the products were obtained in good to excellent

general terms the alkaline earth metals have attracted much yield. Monitoring progress of the reactions was possible by ¹⁹F

recent interest with regard to their stoichiometric and catalytic NMR spectroscopy since Hg(C₆F₅)₂and C₆F₅H have very

reactivity.¹⁵From the point of view of ROP, these metals (with diﬀerent spectra.

the exception of beryllium) are attractive owing to their low

An attempted reaction between Sr metal and H-DBP in THF

toxicity. In the last 10–12 years in particular, a range of sup- in the presence of Hg metal failed. Accordingly, the reactions

porting ligands have been employed to eﬀect control of are considered to proceed by redox transmetallation giving

polymer molecular weights, microstructure, end groups and highly reactive [Ae(C₆F₅)₂(L)_n] (L = THF or DME) complexes

activity.^3,12,16Most of these previous studies focussed on Mg (eqn (1)), which then undergo protolysis by the H-DBP

and Ca systems, but more recently there has been increased (eqn (2)), a step driven by the large acidity diﬀerence between

interest in the ROP of lactones and lactide using molecular Sr a phenol and C₆F₅H.^18cThere is experimental precedent for

initiators.^{3,16q,t,u,w,x}Finally, with regard to compounds of the eqn (1) from the reactions of ytterbium or europium with Hg-

type [Ae(DBP)₂(L)_n], we note that alkoxides and phenoxides of (C₆F₅)₂,^18c,21and ¹⁹F NMR evidence for formation of Ae(C₆F₅)₂

the heavy group 2 metals (Ca, Sr, Ba) show promising appli- species have been obtained.²²Attempted RTP reactions in

cations in other areas such as MOCVD and sol–gel processes.¹⁷diethyl ether failed although an analogous reaction of Nd

In this contribution we report the synthesis and solid state metal giving Nd₂(DBP)₆(OEt₂)₂has been reported.¹⁰

structures of new calcium and strontium bis(2,4-di-tert-butyl-

Ae þ HgðC₆F₅Þ₂! AeðC₆F₅Þ₂þ Hg #

ð1Þ

phenolate) complexes together with a detailed study of their

rac-lactide ROP capability, in particular using benzylamine as

a co-initiator.

AeðC₆F₅Þ₂þ 2 H-DBP ! AeðDBPÞ₂þ 2 C₆F₅H

ð2Þ

Complexes 1–4 were characterised in the solid state by X-ray

crystallography (vide infra) and their C, H analyses were con-

sistent with this composition for the bulk products (with the

exception of 3 where loss of THF in vacuo leads to a lower than

expected %C), as were their ¹H and ¹³C NMR spectra. In the

case of calcium complexes [Ca(DBP)₂(THF)₄] (1) and

Results and discussion

Synthesis of calcium and strontium bis(2,4-di-tert-

butylphenolate) complexes

Redox transmetallation/protolysis (RTP)^10,18oﬀers a simple [Ca₂(DBP)₄(DME)₄(μ-DME)] (2), the ¹H NMR spectra in C₆D₆

route to alkaline earth phenoxides and oﬀers a number of show one type of DPB ligand environment as expected from

advantages over traditional metathesis, such as the simple the solid state structures (Scheme 2 and Fig. 1 and 2 below)

one-pot synthesis, and avoiding the issue of halide or alkali with distinct signals for each of the ring protons. The para-Bu^t

metal retention.^17a,b,19The use of a donor solvent often results groups were also distinguished from the ortho-Bu^tgroups,

in the isolation of heteroleptic complexes,^10,18and the struc- separated by approximately 0.2 ppm. The NMR spectra of

tural motif of the product can vary quite substantially depend- [Sr₃(μ-DBP)₆(THF)₆] (3) in C₆D₆were more complex. At room

ing on the steric demands of the coordinated solvent, which temperature the ¹H and ¹³C NMR spectra show two diﬀerent

can be defined in terms of the steric coordination number DPB environments (1 : 1 ratio) in contrast to the solid state

(steric CN).²⁰For the coordinating solvents THF and DME the structure (Scheme 2 and Fig. 3 below) which shows only one

steric coordination numbers are 1.21 and 1.78, respectively.²⁰type of phenoxide ligand environment. However at 70 °C, the

2,4-Di-tert-butylphenol (H-DBP) was selected for these (and the ¹H NMR spectrum shows only a single set of DPB ligand reso-

previous studies of [Ln(DBP)₃(THF)₃]¹⁰) due to the steric bulk nances, indicative of fast exchange between the two environ-

in just one ortho-position, the high solubility in organic ments. In THF-d₈at 60 °C, a single set of DBP resonances is

solvent due to the tert-butyl substituents, the relative neglect of also observed, again either consistent with fast exchange of

alkaline earth complexes of asymmetrically substituted phe- DBP environments or with conversion into monomeric [Sr-

noxides^15f,17aand its relatively low cost. Unsymmetrical substi- (DBP)₂(THF-d₈)_n]. Overall these data suggest that one or more

tution of the phenolate may aﬀect both bonding and bridging coordination geometries are available to

3

in solution,

This journal is © The Royal Society of Chemistry 2013

Dalton Trans.

View Article Online

Paper

Dalton Transactions

Scheme 2 Syntheses of compounds 1–4. All reactions were performed at RT.

consistent with the well known stereochemical non-rigidity of

complexes of the larger alkaline earth metals.

The ¹H and ¹³C NMR spectra of the unsymmetric [Sr₂(DBP)-

(μ-DBP)₃(DME)₃] (4) in C₆D₆at RT shows only a single set of

DBP (and DME) ligand resonances, indicative of rapid

exchange between terminal and bridging phenolate ligands on

the NMR timescale. At −80 °C in toluene-d₈, greater complexity

is observed, consistent with the solid state structure. The ¹H

NMR spectrum in THF-d₈is analogous to that of 3 apart from

the presence of free DME. These observations are consistent

with substitution of DME by THF-d₈in an excess of the latter

although the nature of the species formed in solution (coordi-

nation number and nuclearity) is unknown.

Solid state structures of 1–4

The structures of the four complexes show remarkable variety,

ranging from a six coordinate monomer 1, through a seven

coordinate DME-bridged dinuclear species 2, to a trinuclear

complex in which all phenoxide ligands are bridging 3, and an

unsymmetrical dinuclear structure with both six and seven

coordinate metal atoms 4. We discuss each in turn and

selected distances and angles are collected in Table 1.

[Ca(DBP)₂(THF)₄] (1). Compound 1 (Fig. 1) is monomeric

with two trans-phenoxide and four equatorial THF ligands in a

close to octahedral arrangement (O–Ca–O angles 86.70(4)–

Fig. 1 Molecular structure of [Ca(DBP)₂(THF)₄] (1). Hydrogen atoms omitted

for clarity. Displacement ellipsoids drawn at 50% level.

Dalton Trans.

This journal is © The Royal Society of Chemistry 2013

View Article Online

Dalton Transactions

Paper

93.30(4)°). Similar trans-[Ca(OR)₂(THF)₄] structures are also

observed for OR = OC₆H₄Bu^t-2 or OSiPh₃and in one molecular

component of [Ca(OC₆H₃Prⁱ₂-2,6)₂(THF)₄]·[Ca(OC₆H₃Prⁱ₂-2,6)₂-

(THF)₃].²³By contrast, analogous alkoxides [Ca(OR)₂(THF)₄]

24b

(R = CPh₃,^24aC(CF₃)₃

and CHCH₂(dimeric)^24chave a cis

Fig. 2 Molecular structure of [Ca₂(DBP)₄(DME)₄(μ-DME)] (2). Hydrogen atoms

and all primary carbons of Bu^tgroups omitted for clarity. Displacement ellipsoids

drawn at the 50% level.

Fig. 3 Molecular structure of [Sr₃(μ-DBP)₆(THF)₆] (3). Hydrogen atoms, all

primary carbons of Bu^tgroups and lattice benzene molecules have been

omitted for clarity. Displacement ellipsoids drawn at the 50% level.

Table 1 Selected bond lengths (Å) and angles (°) for complexes 1–4

[Ca(DBP)₂(THF)₄] (1)

Ca(1)–O(1)

2.206(1)

89.27(4)

159.80(10)

Ca(1)–O(2)

O(2)–Ca(1)–O(3)

2.407(1)

93.30(4)

Ca(1)–O(3)

2.436(1)

93.17(4)

4.822(1)

O(1)–Ca(1)–O(2)

Ca(1)–O(1)–C(1)

O(1)–Ca(1)–O(3)

Ca(1)⋯C(7)(o-Bu^t)

[Ca₂(DBP)₄(DME)₄(μ-DME)] (2)

Ca(1)–O(1)

Ca(1)–O(4)

Ca(1)–O(7)

Ca(1)–O(2)–C(15)

2.192(3)

2.495(4)

2.462(3)

141.5(2)

Ca(1)–O(2)

Ca(1)–O(5)

Ca(1)–O(1)–C(1)

O(1)–Ca(1)–O(2)

2.236(3)

2.526(3)

159.5(3)

177.34(11)

Ca(1)–O(3)

2.518(3)

2.481(3)

4.798(4)

5.031(4)

Ca(1)–O(6)

Ca(1)⋯C(7)(o-Bu^t)

Ca(1)⋯C(21)(o-Bu^t)

[Sr₃(μ-DBP)₆(THF)₆] (3)

Sr(1)–O(1)

Sr(2)–O(1)

2.445(2)

2.582(2)

90.04(7)

Sr(1)–O(2)(THF)

Sr(1)–O(1)–C(1)

Sr(2)–O(1)–C(1)

2.650(2)

120.27(16)

119.27(16)

Sr(1)⋯C(7)(o-Bu^t)

Sr(2)⋯C(7)(o-Bu^t)

Sr(1)⋯Sr(2)

4.723(3)

5.325(3)

3.556(0)

Sr(1)–O(1)–Sr(2)

[Sr₂(DBP)(μ-DBP)₃(DME)₃] (4)

Sr(1)–O(1)

Sr(1)–O(4)

Sr(1)–O(6)

Sr(2)–O(3)

Sr(2)–O(7)

Sr(2)–O(9)

Sr(1)–O(1)–C(1)

Sr(2)–O(2)–C(15)

Sr(1)–O(3)–C(29)

Sr(1)–O(4)–C(43)

2.340(2)

2.478(2)

2.670(2)

2.496(2)

2.709(2)

2.676(2)

152.0(2)

145.4(2)

131.87(19)

125.6(2)

Sr(1)–O(2)

Sr(1)–O(5)

Sr(2)–O(2)

Sr(2)–O(4)

Sr(2)–O(8)

Sr(2)–O(10)

Sr(1)–O(2)–C(15)

O(1)–Sr(1)–O(3)

Sr(2)–O(3)–C(29)

Sr(2)–O(4)–C(43)

2.515(2)

2.769(3)

2.457(2)

2.507(2)

2.669(3)

2.632(2)

100.70(18)

179.01(8)

111.63(19)

115.6(2)

Sr(1)–O(3)

2.539(2)

4.826(4)

4.913(3)

5.037(3)

4.950(4)

5.120(4)

4.751(4)

5.256(4)

3.519(1)

Sr(1)⋯C(7)(o-Bu^t)

Sr(1)⋯C(21)(o-Bu^t)

Sr(2)⋯C(21)(o-Bu^t)

Sr(1)⋯C(35)(o-Bu^t)

Sr(2)⋯C(35)(o-Bu^t)

Sr(1)⋯C(49)(o-Bu^t)

Sr(2)⋯C(49)(o-Bu^t)

Sr(1)⋯Sr(2)

This journal is © The Royal Society of Chemistry 2013

Dalton Trans.

View Article Online

Paper

Dalton Transactions

arrangement. The BHT (BHT = OC₆H₂Bu^t₂Me-2,4,6) analogue solvate using synchrotron radiation. Each six coordinate termi-

of 1, namely [Ca(BHT)₂(THF)₃], has only three THF ligands, nal Sr atom is linked to the six coordinate central strontium by

consistent with the presence of two ortho-tert-butyl groups.¹²three bridging 2,4-di-tert-butylphenolate ions and they are

Whereas the Ca–O bond lengths (Table 1) are close to those of capped by three THF ligands. Unusually there are no terminal

the OC₆H₄Bu^t-2 and OSiPh₃analogues, the (Ar)C–O–Ca angle phenoxide ligands. The stereochemistry of the central Sr(2) is a

is substantially increased from the 2-tert-butylphenolate symmetrical triangular prism, whilst the triangle of O(THF)

complex (146.95(13)°)²³to 1 (159.8(1)°) (Table 1). Thus, the donors (O(2), O(2)#1, O(2)#2) is twisted with respect to the

addition of the 4-Bu^tsubstituent has a significant structural O(1), O(1)#1, O(1)#2 triangle by 41.81(7)°, resulting in tri-

eﬀect. On the other hand, addition of a second o-Bu^t-sub- angular antiprismatic stereochemistry for Sr(1, #1). This

stituent leads to five coordinate [Ca(OAr)₂(THF)₃] com- unusual structural motif has nevertheless been previously

plexes,^{23,24a,25,29a}and this coordination number is also observed encountered in [Eu₃(OC₆H₃Me₂-2,6)₆(THF)₆],^28band the iso-

in the tris(THF) component of the 2,6-diisopropylphenolate typic [M₃(OC₆H₂Me₂-2,4,6)₆(THF)₆] (M = Sr, Ba,^28aEu^28c), in all

above.²³

of which the phenoxide ligands are considerably bulkier (steric

[Ca₂(DBP)₄(DME)₄(μ-DME)] (2). The molecular structure of CN ≥ 1.59) than DBP (Steric CN 1.38). In 3, the sum of the

dinuclear complex 2 is shown in Fig. 2. The seven coordinate steric coordination numbers at Sr(2) is 8.28 and at Sr(1) is

calcium atoms are ligated by trans phenoxide ligands (O(1)– 7.77, by contrast with the very large 9.54 (or greater) for the

Ca(1)–O(2) 177.34(11)°) and two chelating and one bridging central metal of the reported complexes.²⁸In the present high

DME in a pentagonal bipyramidal array (O(DME)–Ca–O(DME) symmetry complex the Sr–O(DBP) bond length to the less

between adjacent oxygens 65.13(11)–80.90(12)°). Similar bond crowded Sr(1) is much shorter (0.14 Å) than to the more

lengths are observed for the two somewhat asymmetrically (Δ crowded Sr(2) (Table 1), as also observed for the reported ana-

= 0.023/0.045 Å) chelating DME ligands (Table 1). Ca(1)–O(7) of logues.²⁸In [Sr₃(OC₆H₂Me₂-2,4,6)₆(THF)₆], which is less sym-

the bridging DME is significantly shorter than Ca(1)–O(3,4,5,6) metrical, the ranges of Sr(2)–OAr (2.516(4)–2.616(4) Å), Sr(1)–

of the chelating DME ligands. In the related seven coordinate OAr (2.408(4)–2.458(4) Å, Sr(1)–O(THF) (2.597(4)–2.684(4) Å)

β-diketonato complex [Ca₂{(Bu^tCO)₂CH}₄(κ²-DME)₂(μ-DME)], encompass the corresponding values for 3, but, surprisingly

coordination of DME is far more unsymmetrical (Δ = 0.078 Å), for a bulkier phenoxide, the averages are not larger than the

and Ca–O(μ-DME) lies between the chelate values,²⁶but the values for 3.

average Ca–O(DME) length is similar to that of 3. For the

[Sr₂(DBP)(μ-DBP)₃(DME)₃] (4). Complex 4 (Fig. 4) crystallises

phenolate ligands, Ca(1)–O(2) is significantly longer than Ca(1)– in the space group P2₁/n and is dinuclear and unsymmetrical

O(1), the Ca–O–C angle is more bent for the O(2) ligand, and with three bridging phenoxides, a terminal phenoxide and a

the Ca(1)⋯C(21)(o-Bu^t) distance is longer than Ca(1)⋯C(7)- chelating DME on Sr(1), but two chelating DME ligands and

(o-Bu^t) (Table 1), all consistent with the greater crowding of the no terminal DBP on Sr(2). Consequently Sr(1) is six coordinate,

O(2) ligand. The aromatic ring of the phenolate is only

19.63(7)° out of the Ca(1)O(1,2,7) plane whereas that of the

O(1) phenolate is rotated by 78.80(12)° thereby reducing the

steric eﬀect of the o-Bu^tgroup. Despite the coordination

number diﬀerences, the Ca(1)–O(DBP) bond lengths of 2 are

comparable with those of 1, but the Ca(1)–O(DME) values are

greater than Ca(1)–O(THF) bond lengths by almost that (0.06 Å)

expected for the diﬀerence in ionic radii.¹⁴By contrast with 2,

somewhat bulkier OC₆H₃R₂-2,6 (R = Me, Prⁱ) or BHT ligands

give mononuclear six coordinate [Ca(OAr)₂(DME)₂] com-

plexes^12,27(steric coordination numbers: 1.59 (R = Me),²⁰1.70

(R = Prⁱ),²⁰1.38 (DBP)¹⁰). A bridging DME is surprising given

the propensity of phenoxide ligands to bridge,^17afor example

in

[Ca₂(OC₆H₂Me₃-2,4,6)₂(μ-OC₆H₂Me₃-2,4,6)₂(κ²-DME)₂-

(κ¹-DME)₂].^28aHowever, the present results can be rationalised

in terms of the diﬀerent steric requirements of THF and

DME.²⁰Thus, the sum of the ligand steric coordination

numbers is 7.60 for six coordinate 1 and 7.21 for seven coordi-

nate 2. Chelation of a third DME to give eight coordinate [Ca-

(DBP)₂(DME)₃] would result in a sterically more strained 8.10.

[Sr₃(μ-DBP)₆(THF)₆] (3). The unusual trinuclear complex 3

ˉ

(Fig. 3) crystallises in the space group R3c all three DBP

ligands are crystallographically equivalent. An initial determi-

nation established the connectivity, but was of a low precision.

Fig. 4 Molecular structure of [Sr₂(DBP)(μ-DBP)₃(DME)₃] (4). Hydrogen atoms

and all primary carbons of Bu^tgroups omitted for clarity. Displacement ellipsoids

A higher quality structure was then obtained for a benzene drawn at the 50% level.

Dalton Trans.

This journal is © The Royal Society of Chemistry 2013

View Article Online

Dalton Transactions

Paper

but Sr(2) seven. The former has distorted octahedral stereo- coordination).¹⁴In the case of 2, Ca(1)–O(2)–C(15) of ligand 2

chemistry, with the terminal and one bridging DBP ligands in is more bent than Ca(1)–O(2)–C(7) of ligand 1 (Table 1) and

apical positions (O(1)–Sr(1)–O(3) angle = 179.01(8)°), whilst the the quaternary carbon of the Bu^tgroup of ligand 2 is further

arrangement at Sr(2) is irregular. Analogous connectivity is from Ca(1), though ligand O(1) avoids steric problems by twist-

observed in [Eu₂(OC₆H₃Me₂-2,6)₄(DME)₃]^27awith a bulkier ing of the aryl group (above). For 3 with solely bridging phen-

phenoxide, but Eu²⁺and Sr²⁺have similar ionic radii.¹⁴Fur- oxides, the Sr(1,2)–O(1)–C(1) angles are shallow, consequent

thermore, the structural motif of three bridging and one term- on bridging, with considerable variation in the Sr(1,2)–C(7)-

inal phenoxide is being increasingly observed in Sr, Eu^IIand (o-Bu^t) distance owing to unsymmetrical bridging. With the

Ba chemistry, e.g. in a N-methylimidazole solvated euro- terminal ligand of 4, Sr(1)–O(1)–C(1) is more bent than Ca(1)–

pium(^II) 2,6-dimethylphenolate,^29aand also in the unsolvated O(1)–C(1) of 1, but the distances to the quaternary carbon of

[M₂(ODPP)₄] (ODPP = OC₆H₃Ph₂-2,6); M = Sr, Ba,^29bEu,^29co-Bu^tare comparable.

[MEu(ODPP)₄] (M = Sr, Ba), and [SrBa(ODPP)₄],³⁰where intra-

molecular π-Ph⋯M coordination occurs in place of donor

Ring-opening polymerisation of rac-lactide

solvent coordination as in 4. As expected, the terminal Sr–O One of our main objectives in this study was to determine the

bond length is the shortest (Table 1) but, unexpectedly two of eﬀectiveness of alkaline earth DPB complexes as initiators/

the three pairs of bridging Sr–O bond lengths are larger to the catalysts for the ROP of lactide, in particular using amine co-

six coordinate Sr atom, which is also less crowded (Σ steric CN initiators which have not yet been established for the alkaline

7.30 Sr(1); 7.72 Sr(2)). In the Eu analogue, all OC₆H₃Me₂-2,6 earth metals. BnNH₂was used as the representative amine for

ligands form longer bridges to the six coordinate Eu.^27aLike- comparison with previous studies (and has been shown to

wise, average Sr–O(DME) bond lengths are larger for six coordi- provide the best performance so far).^3a,8–10ROP experiments

nate Sr(1). Overall, average Sr–O bond lengths in all categories were performed in a glove-box in THF at RT using thoroughly

are similar to those of [Eu₂(OC₆H₃Me₂-2,6)₄(DME)₃] despite purified rac-LA ([rac-LA]₀= 0.69 M). Aliquots were taken at

greater steric crowding (steric CN, 8.14, 8.56 at six and seven various intervals, quenched with non-dried (as-supplied) THF

1

coordinate Eu atoms) but Eu–O bond lengths cover a wide and analysed by H NMR spectroscopy (% conversion and end

range. The asymmetry in the bridging in 4 is less marked than group and tacticity analysis). Number average and weight

in the more crowded 3. As the terminal Sr–O bond length of 4 average molecular weights (M_nand M_w) were determined using

is lengthened from the corresponding Ca–O bond length of 1 gel permeation chromatography (GPC; polystyrene standards

and 2 by significantly less than expected from ionic radii with appropriate Mark–Houwink corrections for PLA in THF at

(0.18 Å),¹⁴there must be significant steric stress in the calcium 30 °C³¹). MALDI-ToF mass spectrometry was used for

complexes.

We have earlier shown that in [Ln(DBP)₃(THF)₃] complexes, 10³g mol⁻¹. Full details are given in the Experimental section.

the Ln–O–C angles of the coordinated phenoxides become

Preliminary studies. The results of initial ROP screening

additional polymer analysis for PLAs with M_nup to ca. 3–4 ×

more bent as the Ln³⁺ion size contracts, thereby oﬀsetting experiments using [Ca₂(DBP)₄(DME)₄(μ-DME)] (2) and

increased steric stress by avoiding a decrease in the Ln–C- [Sr₂(DBP)(μ-DBP)₃(DME)₃] (4) (and also with BnOH and BnNH₂

(o-Bu^t) distance.¹⁰For complex 1, Ca(1)–O(1)–C(1) (159.80(10)° is co-initiators) are summarised in Table 2. Additional prelimi-

somewhat less bent than <La–O–C>(155.8°)¹⁷and Ca–C(o-Bu^t) nary studies found that 3 performed similarly to 4 but that for

(4.822(1) Å) is slightly shorter than <La–C(o-Bu^t)>(4.94 Å) 3 the agreement between calculated and measured M_nvalues

with Ca²⁺slightly smaller than La³⁺(1.00/1.03 for six were slightly less good and the polydispersity indices (PDIs,

Table 2 Polymerisation of rac-LA using [Ca₂(DBP)₄(DME)₄(μ-DME)] (2) and [Sr₂(DBP)(μ-DBP)₃(DME)₃] (4) as initiators/catalysts and BnOH or BnNH₂as co-initiators^a

Conv.^b(%)

M_n(calcd) (g mol⁻¹

)

M_n(GPC)^c(g mol⁻¹

)

M_w/M_n

c

Entry

Complex/co-initiator

[rac-LA]₀: [Co-initiator]₀

Time (min)

d

i

1

2

2^d

30

150

73

89

3200^e

4050

2.6

12 800^f

d

i

3

4

4^d

30

150

51

86

3100^e

12 400^f

2850^g

2870^g

2850^h

2880^h

13 200^j

1.5^j

5

6

7

8

2/BnNH₂

4/BnNH₂

2/BnOH

4/BnOH

20 : 1

10

95

96

95

96

3070

3320

3000

3010

1.8

2.0

1.8

^a[rac-LA]₀: [2 or 4]₀= 100 : 1, 2 mL THF, [rac-LA]₀= 0.69 M, RT. ^bMeasured by ¹H NMR in CDCl₃. ^cMeasured by GPC against polystyrene

standards using the appropriate Mark–Houwink corrections for PLA in THF at 30 °C.^{31 d}No co-initiator added. ^eM_n(calcd) for one PLA chain per

Ae–DBP group = (Conv.(%) × 144.1)/4. ^fM_n(calcd) for one chain per complex = (Conv. (%) × [rac-LA]₀: [Complex]₀× 144.1). ^gM_n(calcd) for one PLA

chain per added BnNH₂= (Conv. (%) × [rac-LA]₀: [BnNH₂]₀× 144.1 + 107.2). ^hM_n(calcd) for one PLA chain per added BnOH = (Conv. (%) ×

[rac-LA]₀: [BnOH]₀× 144.1 + 108.1). ⁱNot measured. ^jBimodal distribution, major peak area measured (87%).

This journal is © The Royal Society of Chemistry 2013

Dalton Trans.

View Article Online

Paper

Dalton Transactions

M_w/M_n) were slightly larger. Since work-up of 3 appeared to

Overall the data in Table 2 show that 2 and 4 alone are poor

result in some THF loss (vide supra) it was felt that the DME initiators but that use of BnNH₂or BnOH co-initiators con-

adduct 4 was more consistently well-defined on the bulk pre- siderably improve the performance. At first sight the BnNH₂

parative scale. For best comparison with 4, the DME adduct 2 co-initiator is acting in a manner analogous to that found pre-

was chosen as the calcium system.

viously (Scheme 1) and in the remainder of this paper we

In the case of 2 with [rac-LA]₀: [Complex]₀= 100 : 1, 89% address this aspect in detail. The PLA formed with 2–4 showed

conversion of monomer to PLA was achieved in 150 min (entry no heterotactic or isotactic enrichment, the polymers being

2), with intermediate aliquots at 30 and 60 min showing 73% atactic in all cases as determined from the selectively homo-

and 80% conversion, respectively. The corresponding reaction nuclear decoupled ¹H NMR spectra of the polymers.³²

with 4 proceeded over a similar timescale and 86% conversion

Lactide and co-initiator loading studies. Building on entry

of monomer to PLA was achieved in 150 min (entry 4). Inter- 5, Table 2, Further ROP experiments were performed where

mediate aliquots were taken at 30, 60 and 120 min and showed [rac-LA]₀: [2]₀was increased from 100 : 1 to 500 : 1 equiv.,

51%, 70% and 83% conversion, respectively. For the PLA holding [BnNH₂]₀: [2]₀constant at 5 : 1 and [rac-LA]₀= 0.69 M.

obtained with 2, the experimental (GPC) molecular weight, These data are plotted in Fig. 5 as M_n(GPC) vs. ([rac-LA]₀–

M_n(GPC), = 4050 g mol⁻¹, consistent with one chain forming [rac-LA]_t) : [2]₀. The line of least-squares best fit gave a gradient

for each Ca–DBP moiety (M_n(calcd) = 3200 g mol⁻¹), whereas 27(1) g mol⁻¹(R²= 0.991) in excellent agreement with the

in the case of 4 M_n(GPC) = 13 200 g mol⁻¹was more consistent expected value of 28.8 g mol⁻¹(i.e. M_Rof [rac-LA]/5 = 144.1/5 g

with one PLA per molecule of 4 (M_n(calcd) = 12 400 g mol⁻¹

)

mol⁻¹) for one chain growing per BnNH₂(5 polymer chains

than for each Sr–DBP unit (M_n(calcd) = 3100 g mol⁻¹). At first growing per equiv of 2). Fig. 6 shows plots of M_n(GPC) vs.

sight the M_n(GPC) values appear to correlate with the solid ([rac-LA]₀–[rac-LA]_t) : [BnNH₂]₀for the ROP of rac-LA for a

state structures of 2 and 4 (Scheme 2) assuming that only the related second series of runs where [rac-LA]₀: [2 or 4]₀= 200 : 1,

terminal DBP ligands can act as a initiators (i.e., in 2, all four and 3, 5, 7 or 10 equiv. of BnNH₂co-initiator were added. In

DBP ligands are terminal whereas in 4 there is one terminal these plots the expected line of least-squares fit has a gradient

DBP ligand per dinuclear complex). However, the broad PDIs of 144 g mol⁻¹(i.e. M_Rof [rac-LA]) for one H-[PLA]-NHBn chain

(2.6 and 1.5) are indicative of a poorly controlled process. In forming per equivalent of amine. In each case the experimen-

addition no –DBP end-groups could be detected in either the tal gradient of the least-squares fitted line was slightly lower

¹H NMR or MALDI-ToF mass spectra, the latter showing only than expected (133(6) g mol⁻¹(R²= 0.950) for 2 and 122(6)

cyclic PLA arising from transesterification side reactions, along (R²= 0.954) g mol⁻¹for 4) but still in good agreement and

with an unassignable minor secondary distribution. Care comparable to other systems we have studied.^8a,9

should therefore be taken in drawing conclusions between the

NMR tube scale mechanistic studies. To gain additional

solid state structures of 2 and 4 and the experimental molecu- insight into the BnNH₂or BnOH co-initiated ROP of rac-LA a

lar weights observed in Table 2. With the absence of evidence series of NMR tube scale experiments were carried out in C₆D₆

for the participation of a conventional initiating group, and THF-d₈. As expected on Brønsted acidity trends for

initiation by trace water or lactic acid cannot be ruled out, amines, aliphatic alcohols and phenols³³neither BnOH nor

despite the rigorous purification of the solvent and rac-LA.

Incorporation of either BnNH₂(entries 5 and 6) or BnOH

(entries 7 and 8) into the ROP experiments with [rac-LA]₀:

[BnNH₂or BnOH]₀: [2 or 4]₀= 100 : 5 : 1 gave a considerably

faster polymerisation system in all cases with >95% conversion

to amine (H-[PLA]-NHBn) or alcohol (H-[PLA]-OBn) terminated

PLA within 10 min. The M_n(GPC) values are consistent with

one chain forming per added co-initiator with M_n(GPC) in the

BnNH₂(4 equiv.) reacted with 2 or 4 to eliminate H-DBP,

range 3000–3320

g

mol⁻¹and M_n(calcd) in the range

1

2850–2880 g mol⁻¹. The H NMR spectra were consistent with

the expected HOC(H)Me– and –C(O)XBn (X = NH or O) end

groups. MALDI-ToF analysis (e.g. Fig. S1† of the ESI) also con-

firmed the expected chain ends but in addition revealed the

presence of cyclic PLA (presumably due to intramolecular

transesterification processes). The separation of Δ(m/z) = 72 g

mol⁻¹between the adjacent peak envelopes in both the linear

and cyclic PLAs in the mass spectra is also indicative of exten-

sive transesterification processes. The PDI values of 1.8–2.0 for

these PLAs are consistent with the concurrent formation of

cyclic PLAs and transesterification reactions. Again no evi-

dence of –DBP terminated PLAs were found by NMR or

MALDI-ToF analysis.

Fig. 5 M_n(GPC) vs. ([rac-LA]₀–[rac-LA]_t) : [2]₀, [rac-LA]₀: [2]₀= 100, 200, 300,

400 or 500, using 2 and 5 equiv. BnNH₂as co-initiator. PDIs are given in

parentheses and intercept was set to 107.2 g mol⁻¹(M_RBnNH₂).

Dalton Trans.

This journal is © The Royal Society of Chemistry 2013

View Article Online

Dalton Transactions

Paper

BnNH₂and either 2 or 4 proceeds broadly according to

Scheme 1 with initial, very rapid ring-opening of lactide by all

the amine to form H-[rac-LA]-NHBn. This is followed by iROP

type propagation catalysed by the metal complex in an acti-

vated monomer type mechanism. In this case a rate law of the

type shown in eqn (3) should apply^2s,3bwhere −d[rac-LA]/dt is

the rate of monomer consumption and k_p(Ae)is the propa-

gation rate constant for 2 (k_p(Ca)) or 4 (k_p(Sr)).

b

a

c

ꢀd½rac-LAꢁ=dt ¼ k_pðAeÞ½2 or 4ꢁ₀½rac-LAꢁ ½BnNH₂ꢁ₀

ð3Þ

Initial studies focused on the benzylamine co-initiated ROP

of rac-LA using 4 as the catalyst and a loading of [rac-LA]₀:

[BnNH₂]₀: [4]₀= 400 : 5 : n. Runs were performed where the

initial rac-LA and BnNH₂concentrations were held constant

([rac-LA]₀: [BnNH₂]₀= 400 : 5, 0.69 M and 8.6 × 10⁻³M respect-

ively) and the relative concentration of 4 was increased from

n = 0.5 to 2.5 molar equiv. with respect to this. A representative

% Conversion vs. Time plot for the run with [rac-LA]₀:

[BnNH₂]₀: [4]₀= 400 : 5 : 1.5 is shown in Fig. 7 together the

corresponding second-order plot modelling consumption of

rac-LA for this experiment. A summary of key data is given in

Table 3.

Fig. 6 M_n(GPC) vs. ([rac-LA]₀–[rac-LA]_t) : [BnNH₂]₀for the ROP of rac-LA

([rac-LA]₀: [2 or 4]₀

= 200 : 1) using [Ca₂(DBP)₄(DME)₄(μ-DME)] (2) (circles)

or [Sr₂(DBP)(μ-DBP)₃(DME)₃] (4) (triangles) and 3, 5, 7 or 10 equiv. BnNH₂as

co-initiator. The numbers in parentheses are the PDIs and the intercepts were

set to 107.2 g mol⁻¹(M_RBnNH₂). The dashed line represents M_n(calcd) based

on [rac-LA]₀: [BnNH₂]₀= 200 : n.

although they appear to compete with DME as a donor ligand

based on the changes in the chemical shifts of the respective

resonances. When rac-LA (20 equiv.) was added to the mixtures

(cf. Fig. S2† of the ESI) there was complete consumption of

both the lactide and the BnOH or BnNH₂, and concomitant

formation of H-[PLA]-XBn (X = NH or O) oligomer. This is con-

sistent with results in Table 2 (entries 5–8) and Fig. 5 and 6

that the co-initiator is consumed rapidly at the beginning of

the reaction in order to generate the multiple H-[PLA]-XBn PLA

chains (cf. Scheme 1) which are then able to propagate.^8a,9

Interestingly, while quantitative displacement of DME from

each metal complex occurred upon formation of H-[PLA]-XBn

(as judged by comparison of the chemical shifts with those for

free DME in the same solvent), there was no formation of the

phenol H-DBP which would be expected if propagation pro-

ceeded by the coordination–insertion mechanism and in situ

formed (L)Ae-[PLA]-XBn units. These results imply that ROP

catalysed by 2 and 4 in the presence of BnNH₂or BnOH may

proceed via an activated monomer mechanism, consistent

with recent proposals by Carpentier et al.³and Miller et al.¹²

This was supported by further NMR tube scale experiments (cf.

Fig. S3† of the ESI) in which up to 4 equiv. of isolated benzyl

amine-terminated PLA macromonomer (H-[rac-LA]₁₀-NHBn,

M_n= 1600 g mol⁻¹, PDI = 1.11), independently prepared using

Y(O₂N^NMe2)(HO₂N^NMe2),^8awere successively added to 2 or 4. ¹H

NMR analysis again showed no liberation of H-DBP, but DME

was displaced showing that the PLA chains can act as ligands.

Addition of 20 equiv. rac-LA showed that each system was

active for ROP (Fig. S3†) as judged by the immediate consump-

tion of rac-LA and the increase of relative intensity of the

H-[rac-LA]_n-NHBn main chain resonances.

Plots of 1/[rac-LA]_tvs. Time (e.g., Fig. 8, left) were linear over

all concentrations of 4 assessed, indicating that the polymeri-

sation reaction proceeded with a second-order dependence in

[rac-LA]. The gradients of the fitted lines in Fig. 8 aﬀord the

apparent second-order rate constants, k_app(Sr)in each case

a

c

(where k_app(Sr)= k_p(Sr)[4]₀[BnNH₂]₀, assuming that: (i) the con-

centration of growing H-PLA-NHBn chains is equivalent to

[BnNH₂]₀; (ii) the concentration of the actual catalyst is equal

to [4]₀; (iii) both remain constant throughout the reaction).

The 2nd order dependence on [rac-LA] is unusual but not

unprecedented.^8a,9,34,35A plot of k_app(Sr)vs. [4]₀in the range

molar equiv. [4]₀= 0.5–2.5 was linear (Fig. 8, right) showing a

first order dependence of k_app(Sr)on [4]₀. Thus rate of propa-

gation (eqn (3)) is proportional to [rac-LA]²[4]₀, and from Fig. 8

(right) k_p(Sr)[BnNH₂]₀^c= 182(5) M⁻²min⁻¹

.

Further runs were performed where the molar equiv. of

strontium centres (= 2 × [4]₀) were increased to exceed the

number of growing polymer chains ([rac-LA]₀: [BnNH₂]₀: [4]₀=

400 : 5 : 3.0, 400 : 5 : 4.0 and 400 : 5 : 5.0).³⁶As shown in Fig. 8

(right) and Table 3, k_app(Sr)increased only marginally after the

400 : 5 : 2.5 run in this region demonstrating kinetic saturation.

In other words, propagation became limited by the number of

propagating chains (i.e., [BnNH₂]₀) rather than the availability

of catalytic sites.

Molecular weight data were measured for selected exper-

iments (Table 3 and Fig. 9). In general, the calculated and

measured M_nof early aliquots taken from the reaction mixture

were in reasonable agreement, but as the reaction progressed

experimental molecular weights began to diverge from those

expected. Fig. 9 (left) shows M_n(GPC) data and PDI for the

run, [rac-LA]₀: [BnNH₂]₀: [4]₀= 400 : 5 : 1.5 (at completion,

M_n(calcd) = 9910 g mol⁻¹, M_n(GPC) = 7040 g mol⁻¹and PDI =

1.51, entry 9, Table 3). The lower than calculated M_n(GPC) at

Kinetic studies: metal and LA dependences. The experi-

ments described above indicate that ROP of rac-LA using

This journal is © The Royal Society of Chemistry 2013

Dalton Trans.

View Article Online

Paper

Dalton Transactions

Fig. 7 Conversion vs. time proﬁle (left) and second-order rate plot (right) for the run [rac-LA]₀: [BnNH₂]₀: [4]₀= 400 : 5 : 1.5 with [rac-LA]₀= 0.69 M. k_app(Sr)

0.501(11) (R²= 0.994) M⁻¹min⁻¹, vertical-axis intercept = 1.63(8) M⁻¹(expected value = 1/[rac-LA]₀= 1.44 M⁻¹).

=

Table 3 Apparent second-order rate constants (k_app) and molecular weight data from the metal dependence determination experiments for [Ca₂(DBP)₄-

(DME)₄(μ-DME)] (2) and [Sr₂(DBP)(μ-DBP)₃(DME)₃] (4)

[rac-LA]₀:

[BnNH₂]₀: [2 or 4]₀

Conv.^c

(%)

M_n(calcd)^c

M_n(GPC)^d

k_app(Ae)^b(M⁻¹min⁻¹

)

R^{2 b}

(g mol⁻¹

)

(g mol⁻¹

)

M_w/M_n

a

d

Entry

Complex

e

1

2

3

4

5

6

7

8

2

4

400 : 5 : 0.5

400 : 5 : 1.0

400 : 5 : 1.5

400 : 5 : 2.5

400 : 5 : 4.0

400 : 5 : 5.0

400 : 5 : 0.5

400 : 5 : 1.0

400 : 5 : 1.5

400 : 5 : 2.0

400 : 5 : 2.5

400 : 5 : 3.0

400 : 5 : 4.0

400 : 5 : 5.0

0.279(8)

0.994

0.997

0.995

0.982

0.981

0.986

0.984

0.994

0.997

0.991

0.986

0.975

0.967

0.534(12)

0.897(22)

1.323(52)

1.606(71)

1.556(53)

0.136(5)

0.281(11)

0.501(11)

0.671(11)

0.759(23)

0.780(29)

0.813(49)

0.805(45)

87

86

80

10 140

10 020

9330

8880

8200

6710

1.35

1.36

1.46

86

10 020

6350

1.37

e

80

85

9330

9910

7140

7040

1.66

1.51

9

10

11

12

13

14

76

8870

7550

1.70

e

82

74

88

9560

8640

10 250

7730

5790

5570

1.51

1.36

1.44

^a[rac-LA]₀: [BnNH₂]₀= 400 : 5, [rac-LA]₀= 0.69 M, THF, RT. ^bApparent second-order rate constants (k_app(Ae)= k_p(Ae)[2 or 4]₀[BnNH₂]₀) for

consumption of rac-LA with standard error and R²determined by linear regression analysis. ^cReactions monitored by ¹H NMR (CDCl₃) sampling;

all quoted conversions refer to the last point where M_n(GPC) was determined. M_n(calcd) = (Conv. (%) × [rac-LA]₀: [BnNH₂]₀× 144.1 + 107.2).

^dMeasured by GPC against polystyrene standards with appropriate Mark–Houwink corrections for PLA in THF at 30 °C.^{31 e}Molecular weight data

not determined.

c

higher monomer to PLA conversion and the general increase M_n(GPC) = 5570 g mol⁻¹for [BnNH₂]₀: [4]₀= 5 : 5.0, entry 14,

in PDI during polymerisation were consistent with detrimental Table 3). Thus not only does an excess of metal sites yield no

chain-shortening transesterification side reactions competing net increase in rate, but seems to increase the number of

with linear chain growth and resembled the poorer agreement chain-shortening (probably transesterification) processes.

seen in the catalyst loading experiments (Fig. 6) at lower load-

ings of amine ([rac-LA]₀: [BnNH₂]₀).

In a similar way, compound 2 was conveniently examined

under the same conditions as 4 (see Fig. S4 and S5† of the ESI

Up to the saturation limit, adding more 4 to the polymeri- and Table 3). A second-order dependence in [rac-LA] was again

sation mixture resulted in a linear increase in k_app(Sr)(Fig. 8, observed along with a first-order dependence in [2]₀, and a

right), whereas M_n(GPC) showed no systematic dependence on saturation limit was reached once the molar equiv. of calcium

[4]₀(Fig. 9, right), as expected. However, the data for when centres exceeded [BnNH₂]₀. The experimental rate of propa-

[BnNH₂]₀: [4]₀= 5 : 4.0 or 5 : 5.0, where the molar equivalents gation (eqn (3)) is therefore proportional to [rac-LA]²[2]₀as was

of Sr significantly exceeded [BnNH₂]₀(i.e. the number of also found for 4, and k_p(Ca)[BnNH₂]₀= 315(10) M⁻²min⁻¹

,

c

growing chains), showed increasingly poor agreement between suggesting that k_p(Ca)≈ 1.7 × k_p(Sr), assuming the rate depen-

measured and expected M_n(e.g., M_n(calcd) = 10 250 g mol⁻¹

,

dence on [BnNH₂]₀is the same in each case (vide infra).

Dalton Trans.

This journal is © The Royal Society of Chemistry 2013

View Article Online

Dalton Transactions

Paper

Fig. 8 Left: second-order plots of 1/[rac-LA]_tvs. time with increasing concentrations of [4]₀. Rate constants, k_app(Sr)= 0.136(5) (R²= 0.986), 0.281(11) (R²= 0.984),

0.501(11) (R²= 0.994), 0.671(11) (R²= 0.997) and 0.759(23) (R²= 0.991) M⁻¹min⁻¹. The vertical-axis intercepts of these plots are 1.68(13), 1.75(11), 1.63(8), 1.61(8)

and 1.48(14) M⁻¹(expected value = 1/[rac-LA]₀= 1.44 M⁻¹). Right: plots of k_appvs. [4]₀for the polymerisation of rac-LA. [rac-LA]₀: [BnNH₂]₀= 400 : 5. Molar equiv.

[4]₀= 0.5–2.5, 3.0, 4.0 or 5.0. k_p(Sr)[BnNH₂]₀^c= 182(5) M⁻²min⁻¹(R²= 0.983) based on k_app(Sr)data in the range molar equiv. [4]₀= 0.5–2.5.

Fig. 9 Left: M_n(GPC) vs. conversion (triangles) and M_w/M_nvs. conversion (circles) plots for the run [rac-LA]₀: [BnNH₂]₀: [4]₀= 400 : 5 : 1.5. Right: M_n(GPC) vs. con-

version plots for [rac-LA]₀: [4]₀= 400 : 1.0, 1.5, 3.0, 4.0 or 5.0, [rac-LA]₀: [BnNH₂]₀= 400 : 5, PDIs = 1.4–1.7. The dashed lines represent M_n(calcd) based on

[rac-LA]₀: [BnNH₂]₀= 400 : 5.

Kinetic studies: BnNH₂dependence. From the preliminary rate (and thus k_app(Sr)) on [BnNH₂]₀. This fast rate (>90% con-

experiments summarised in Table 2 it was clear that BnNH₂version after 8 min) is at the upper limit of the practical reac-

had a significant eﬀect on the rate of polymerisation (cf. tion-sampling window and, so runs at 2.5 and 7.5 BnNH₂

entries 1, 2 and 5, and 3, 4 and 6), consistent in general terms molar equiv. were performed (entries 9 and 11, Table 4). The

with an activated monomer mechanism as inferred from the gradient of the line of least-squares fit (R²= 0.984) from the

NMR tube scale experiments described above. To determine plot of −lnk_app(Sr)vs. −ln[BnNH₂]₀(Fig. 10, left) yielded an

the order c with respect to [BnNH₂]₀(cf. the rate expression order (c) in [BnNH₂]₀of 2.6(2).

proposed in eqn (3)), a ratio of [rac-LA]₀: [4]₀= 400 : 1 was

A similar non-linear dependence of rate and k_app(Ca)on

selected and held constant ([rac-LA]₀= 0.69 M) while [BnNH₂]₀[BnNH₂]₀was found for 2, but only for runs up to

was varied (Table 4, entries 8–11). k_app(Sr)for the run [rac-LA]₀: [BnNH₂]₀: [2]₀= 7.5 (Table 4, entries 1–4). Holding [rac-LA]₀

[BnNH₂]₀: [4]₀= 400 : 5.0 : 1 has already been determined and [2]₀constant at [rac-LA]₀: [BnNH₂]₀: [2]₀= 400 : 1 (as for

(k_app(Sr)= 0.281(11) M⁻¹min⁻¹; entry 8, Table 3; reproduced for 4), the number of amine equiv. (n) were increased in the order

convenience as entry 9, Table 4). Doubling [BnNH₂]₀to give 4.0, 5.0, 6.0, 7.5, 10.0 and 12.5 (Table 4). The use of 10 equiv.

[rac-LA]₀: [BnNH₂]₀: [4]₀= 400 : 10.0 : 1 gave k_app(Sr)= 2.033(92) of BnNH₂did not result in as great an increase in rate as

M⁻¹min⁻¹(Table 4, entry 11). This is considerably larger than expected based on the fitted line for 4.0–7.5 BnNH₂equiv.

for the 400 : 5.0 : 1 run, showing a non-linear dependence of (Fig. 10, right). The result was reproduced and a further point

This journal is © The Royal Society of Chemistry 2013

Dalton Trans.

View Article Online

Paper

Dalton Transactions

Table 4 Apparent second-order rate constants (k_app(Ae)) and molecular weight data from the [BnNH₂]₀dependence experiments for [Ca₂(DBP)₄(DME)₄(μ-DME)] (2)

and [Sr₂(DBP)(μ-DBP)₃(DME)₃] (4)

[rac-LA]₀_a: [BnNH₂]₀:

[2 or 4]₀

Conv.^c

(%)

M_n(calcd)^c

M_n(GPC)^d

k_app(Ae)^b(M⁻¹min⁻¹

)

R^{2 b}

(g mol⁻¹

)

(g mol⁻¹

)

M_w/M_n

d

Entry

Complex

e

1

2

3

4

5

6

7

8

2

4

400 : 4.0 : 1

400 : 5.0 : 1

400 : 6.0 : 1

400 : 7.5 : 1

400 : 10.0 : 1

400 : 10.0 : 1 (duplicate)

400 : 12.5 : 1

400 : 2.5 : 1

400 : 5.0 : 1

400 : 7.5 : 1

0.285(6)

0.994

0.997

0.998

0.991

0.997

0.987

0.989

0.976

0.984

0.982

0.534(13)

0.761(11)

1.264(33)

1.407(22)

1.492(49)

1.891(58)

0.048(2)

0.281(11)

0.598(20)

2.033(92)

87

10 140

8420

1.35

e

90

7030

8310

1.30

85

5010

5950

1.42

e

88

4170

4260

1.25

e

9

10

12

80

87

88

9330

6800

5180

7140

5900

5840

1.66

1.70

1.50

400 : 10.0 : 1

^a[rac-LA]₀: [2 or 4]₀= 400 : 1, [rac-LA]₀= 0.69 M, THF, RT. ^bApparent second-order rate constants (k_app(Ae)= k_p(Ae)[2 or 4]₀[BnNH₂]₀) for

consumption of rac-LA with standard error and R²determined by linear regression analysis. ^cReactions monitored by ¹H NMR (CDCl₃). All

quoted conversions refer to the last point where M_n(GPC) was determined. M_n(calcd) = (Conv. (%) × [rac-LA]₀: [BnNH₂]₀× 144.1 + 107.2).

^dDetermined by GPC against polystyrene standards with appropriate Mark–Houwink corrections for PLA in THF at 30 °C.^{31 e}Molecular weight

data not determined.

c

Fig. 10 −lnk_app(Ae)vs. −ln[BnNH₂]₀plots for [Sr₂(DBP)(μ-DBP)₃(DME)₃] (4) (left) and [Ca₂(DBP)₄(DME)₄(μ-DME)] (2) (right). The value in parentheses are number of

amine equiv. (n) for the runs [rac-LA]₀: [BnNH₂]₀: [Complex]₀= 400 : n : 1.

at 12.5 amine equiv. was likewise suggestive of a tendency to for 2 and 4 is unexpected, but arises from identical kinetic be-

saturation, representing the rate of propagation becoming haviour for the two systems in the non-saturation region where

limited, presumably due to limitations of access to the smaller –lnk_appvs. −ln[BnNH₂]₀is linear. Although the orders of 2.4(1)

metal centre. An order (c) of 2.4(1) in [BnNH₂]₀was calculated and 2.6(2) are individually close or equal to 2 (which is also

from the plot of −lnk_appvs. −ln[BnNH₂]₀, based on the line of the order in [rac-LA]) within three standard deviations, the fact

least-squares fit (R²= 0.993) through the points representing that they are both independently closer to the average value of

4.0, 5.0, 6.0 and 7.5 BnNH₂molar equiv. (Fig. 10, right).

ca. 2.5 implies that this value may more likely be the correct

Kinetic studies: concluding comments. In general terms, interpretation. In combination with the first order dependence

the observation that there is a rate dependence on both on [2 or 4]₀and second order rac-LA consumption the overall

[BnNH₂]₀(i.e. number of growing PLA chains) and [2 or 4]₀is rate law therefore becomes −d[rac-LA]/dt = k_p(Ae)[2 or 4]₀-

2.5

consistent with an activated monomer mechanism,^2s,3bas is [rac-LA]²[BnNH₂]₀

.

the observation that the polymerisation rate only significantly

This rate expression is clearly more complex than expected

increases with [2 or 4]₀while the number of metal centres is for a “simple” activated monomer mechanism.^2sFor example,

less than or equal to the number of growing chains (once each Carpentier et al. found a rate expression −d[L-LA]/dt = k_p[com-

chain has a metal centre “partner” adding more catalyst has plex][L-LA][BnOH] for the BnOH co-initiated ROP of L-LA using

little positive eﬀect and a kinetic saturation point is reached). cationic alkoxide or phenolate Ae (Ae = Mg, Ca, Sr, Ba) or Zn

However, the non-integer order dependence of k_app(Ae)on complexes.^3bHowever, in these systems the supporting ligands

[BnNH₂]₀(i.e. on the concentration of growing polymer chains) are six-coordinate mixed aza-crown ether types, providing a

Dalton Trans.

This journal is © The Royal Society of Chemistry 2013

View Article Online

Dalton Transactions

Paper

much better defined coordination environment than is possi- absence of any co-initiator and then with 5 equiv. of BnOH

ble for 2 and 4. Scheme 2 and the solid state structure in and BnNH₂respectively. In the reaction without any co-

Fig. 1–4 show the possible structural diversity available for the initiator, conversion of monomer to PLA was sluggish with

“Ae(DBP)₂” unit in presence of diﬀerent donor ligands. 34% conversion being recorded after 30 min. At this point the

Our NMR tube experiments between the macromonomer catalyst was no longer active. The M_n(GPC) = 7460 g mol⁻¹was

H-[rac-LA]₁₀-NHBn and [Sr₂(DBP)(μ-DBP)₃(DME)₃] (4) (Fig. S3† much higher than that calculated for even one chain per metal

of the ESI) found that the polymer chain could act as a ligand (M_n(calcd) = 4900 g mol⁻¹) and the PDI of 1.8 was comparable

by displacing DME. Therefore, while it is clear (cf. Fig. 8 and to values observed with 2, 4 and [Ln(DBP)₃(THF)₃] (Ln = Nd or

S4†) that each individual metal site in 2 and 4 is catalytically Gd). No –DBP end group was observed in the NMR or MALDI-

competent, the nature of the (one or more) complexes present ToF spectra, again as found for 2, 4 and [Ln(DBP)₃(THF)₃].

under polymerisation conditions, and how any equilibria

The polymerisation performance improved when BnOH

between them changes as a function of [rac-LA] or [BnNH₂]₀and BnNH₂were included with [rac-LA]₀: [BnOH and

(i.e. the concentration of PLA chains) cannot be known. BnNH₂]₀: [5]₀= 100 : 5 : 1. In these cases ∼95% conversion was

Unusual fractional orders or orders greater than unity in ROP achieved within 20 min in both cases. With BnOH the agree-

catalyst rate expressions have typically be attributed to solution ment between M_n(GPC) (2850 g mol⁻¹, BnOH) and M_n(calcd)

equilibria and/or catalyst aggregation eﬀects.^2q,35a,37

(2880 g mol⁻¹) was consistent with one H-[PLA]-OBn polymer

Based on the k_p(Ae)[BnNH₂]₀values of 182(5) M⁻²min⁻¹chain forming per equivalent of alcohol; likewise in the exper-

(Ae = Sr, Fig. 8) and 315(10) M⁻²min⁻¹(Ae = Ca, Fig. S4†) iment with BnNH₂, M_n(GPC) = 3230 g mol⁻¹compared well

found for experiments with [BnNH₂]₀= 8.6 × 10⁻³M (Table 3) with that expected (M_n(calcd) = 2890 g mol⁻¹). The PDIs of 1.2

the propagation rate constants for k_p(Ca)and k_p(Sr)are ca. 4.6 × (BnOH) and 1.3 (BnNH₂) were also consistent with well-con-

10⁷and 2.6 × 10⁷M^−4.5min⁻¹, respectively. This is also an trolled ROP processes, and the respective ¹H NMR and

unexpected result since previous studies for alkaline earth MALDI-ToF mass spectra featured –C(O)OBn and –C(O)NHBn end

catalysts operating through the activated monomer mechan- groups. However, as with [Ln(DBP)₃(THF)₃], 2 and 4 no hetero-

ism^3b,12found that k_pincreased in the order Ae = Mg < Ca < Sr tactic bias was detected and transesterification processes were

≈ Ba. Further detailed kinetic studies of other systems will be shown to operate, given separation of Δ(m/z) = 72 g mol⁻¹

needed to probe this aspect further, taking care that saturation between peak envelopes in the MALDI-ToF mass spectra.

2.5

kinetics are avoided both in terms of catalyst loading and

metal : co-initiator ratios.

Identical ¹H NMR tube scale studies to those for 2 and 4

were also performed with 5. Treatment of 5 with 2–5 equiv.

BnNH₂or BnOH did not result in liberation of H-DBP.

However, both BnNH₂and BnOH were immediately consumed

when 20 equiv. rac-LA was added to each mixture, forming

H-[PLA]-NHBn and H-[PLA]-OBn, respectively. In neither case

was free H-DBP observed. Therefore it would seem that this

Synthesis and preliminary ROP studies of [Y(DBP)₃(THF)₂] (5)

The unusual experimental rate law for 2 and 4 and the identifi-

cation of an activated monomer mechanism led us to consider

again our previously reported¹⁰ROP results for the rare earth

phenoxides [Ln(DBP)₃(THF)₃] (Ln = Nd or Gd). As mentioned

the very fact ROP for these larger lanthanoids and their para-

magnetic nature precluded detailed mechanistic studies of the

type reported here for 2 and 4. We therefore prepared the new

diamagnetic compound [Y(DBP)₃(THF)₂] (5) by a protolysis

reaction between [Y{N(SiHMe₂)₂}₃(THF)₂]³⁸and 3 equiv. of

H-DBP in THF at RT (eqn (4)). The NMR, elemental analysis

and other data are consistent with the structure shown in

class of lanthanoid and group

3 tris(phenoxide) (and

potentially others mentioned previously in the literature¹¹)

may behave in a similar manner to 2 and 4, i.e., proceeding via

an activated monomer type propagation pathway. Further

detailed studies will be required to develop this hypothesis

further.

eqn (4) which is analogous to that of the starting complex Conclusions

[Y{N(SiHMe₂)₂}₃(THF)₂].

Four new alkaline earth complexes of the asymmetrically sub-

stituted 2,4-di-tert-butylphenol, namely [Ca(DBP)₂(THF)₄] (1),

[Ca₂(DBP)₄(DME)₄(μ-DME)] (2), [Sr₃(μ-DBP)₆(THF)₆] (3), and

[Sr₂(DBP)(μ-DBP)₃(DME)₃] (4) have been synthesised by RTP

reactions and structurally characterised. The diﬀerence in

ionic radii between Ca and Sr, coupled with the diﬀering steric

coordination numbers of THF compared to DME resulted in

four distinct structural types. In addition the new yttrium com-

pound [Y(DBP)₃(THF)₃] (5) was also synthesised by an amide-

ð4Þ

A detailed investigation of the ROP capability of 5 was protolysis approach. While 2, 4 and 5 are poor initiators for

beyond the scope of the present study but some preliminary the ROP of lactide in their own right, addition of BnOH or

experiments have been undertaken. Thus 5 was treated with BnNH₂as co-initiators gave better controlled systems. This was

100 equiv. of rac-LA ([rac-LA]₀= 0.69 M) in THF at RT in the the first example for a group 2 metal of amine co-initiated

This journal is © The Royal Society of Chemistry 2013

Dalton Trans.

View Article Online

Paper

Dalton Transactions

ROP. The complex form of the rate expression, −d[rac-LA]/dt = [Ca(DBP)₂(THF)₄] (1)

k_p(Ae)[2 or 4]₀[rac-LA]²[BnNH₂]₀^2.5, diﬀers from that described

Calcium metal filings (0.20 g, 5.0 mmol), bis(pentafluorophe-

nyl)mercury (0.55 g, 1.0 mmol), 2,4-di-tert-butylphenol (0.47 g,

2.3 mmol), and Hg metal (2 drops), were added to a Schlenk

flask with dry THF (20 mL), and placed in an ultrasound bath

for ∼60 h. A pale yellow solution was observed. The solution

was filtered and concentrated under vacuum to 6 mL and left

to stand at −4 °C. Crystallisation was initiated when returned

to room temperature, and small colourless crystals grew over

24 h. Yield = 0.65 g (76%), m.p. 76–80 °C. Anal. found (calcd

by Carpentier and Sarazin and most likely reflect the poorly-

defined supporting ligand environment in the systems under

consideration here.

Experimental

All products are air-sensitive and required the use of glove-box

and Schlenk and vacuum line techniques, and were manipu-

lated under a dinitrogen or argon atmosphere. Solvents were

dried by distillation over sodium or sodium/benzophenone. All

solvents were stored in J. Young Teflon valve ampoules.

Elemental analyses (C, H) were performed by the Campbell

Microanalytical Laboratory, University of Otago, Dunedin, New

Zealand or by the Elemental Analysis Service at the London

Metropolitan University. Infrared spectra were obtained as

Nujol mulls with a Perkin Elmer 1600 FTIR or Perkin Elmer

1

for C₄₄H₇₄CaO₆): C, 71.37 (71.50); H, 9.84 (10.09) %. H NMR

(300.13 MHz, C₆D₆, 303 K): 7.44 (s, 2H, H3(Ar)), 7.18 (br s, 2H,

H5(Ar)), 6.84 (br s, 2H, H6(Ar)), 3.47 (s, 16H, OCH₂CH₂), 1.62

(s, 18H, 2-Bu^t), 1.40 (s, 18H, 4-Bu^t), 1.30 (s, 16H, OCH₂CH₂).

¹³C-{¹H} NMR (125.8 MHz, C₆D₆, 298 K): 160.8 (ArC_q-1), 137.1

and 136.5 (ArC_q-2 and ArC_q-4), 123.9 and 123.7 (ArC-5 and

ArC-3), 121.1 (ArC-6), 68.3 (OCH₂CH₂), 35.4 (2-C(CH₃)₃), 34.2

(4-C(CH₃)₃), 32.3 (4-C(CH₃)₃), 31.0 (2-C(CH₃)₃), 25.5 (OCH₂CH₂)

ppm. IR (Nujol mull): 1595 (s), 1404 (w), 1306 (m), 1256 (m),

1203 (m), 1141 (m), 1117 (m), 1074 (m), 1036 (m), 885 (m),

1

Spectrum RX1 instrument. H, ¹³C and ¹⁹F NMR spectra were

recorded on a Bruker DPX 300 MHz spectrometer, the former

being referenced to residual protio-solvent peaks. Melting

points were measured in pipettes sealed under dinitrogen, and

are uncorrected. 2,4-Di-tert-butylphenol (H-DBP) was pur-

chased from Sigma-Aldrich, and was not further purified. The

compounds [Y{N(SiHMe₂)₂}₃(THF)₂]³⁸and bis(pentafluorophe-

nyl)mercury³⁹were synthesised according to the published

procedures. Calcium and strontium metals were purchased

from Aldrich as chunks or turnings which were freshly filed

under nitrogen prior to use.

821 (m), 731 (m), 666 (m), 633 (m) cm⁻¹

.

[Ca₂(DBP)₄(DME)₄(μ-DME)] (2)

Calcium metal filings (0.25 g, 6.2 mmol), bis(pentafluorophe-

nyl)mercury (0.55 g, 1.0 mmol), 2,4-di-tert-butylphenol (0.47 g,

2.3 mmol), and Hg metal (3 drops), were added to a Schlenk

flask along with dry DME (30 mL). Upon sonication for 60 h,

the pale yellow solution was filtered and concentrated under

vacuum to 6 mL and left to stand at −4 °C. Small colourless

crystals grew over one week. Yield = 0.43 g (55%), m.p. 160 °C,

dec. temp. 220 °C. Anal. found (calcd for C₇₆H₁₃₄Ca₂O₁₄): C,

67.80 (67.51); H, 10.0 (9.99) %. ¹H NMR (300.13 MHz, C₆D₆,

MALDI-ToF MS analysis was performed on a Waters MALDI

micro equipped with a 337 nm nitrogen laser and an accelerat-

ing voltage of 25 kV was applied. The polymer samples were

dissolved in THF at a concentration of 1.0 mg mL⁻¹. The cat-

ionisation agent used was potassium trifluoroacetate (Fluka,

3

303 K): 7.52 (s, 4H, H3(Ar)), 7.23 (d, J = 6.0 Hz, 4H, H5(Ar)),

3

6.92 (d, J = 6.0 Hz, 4H, H6(Ar)), 3.11 (s, 20H, CH₃OCH₂), 2.94

(s, 30H, CH₃OCH₂), 1.70 (s, 36H, 2-Bu^t), 1.40 (s, 36H, 4-Bu^t).

¹³C NMR (125.80 MHz, C₆D₆, 298 K): 160.4, 137.8 and 136.4

(br, ArC_q-1,2,4), 124.0, 123.6 and 120.3(ArC-5,3,6), 70.7

(CH₃OCH₂), 59.7 (CH₃OCH₂), 35.3 and 34.1 (C(CH₃)₃), 32.1

and 30.5 (C(CH₃)₃). ¹H NMR (500.3 MHz, C₆D₆, 298 K): 7.50 (s,

4H, H3(Ar)), 7.20 (br s, 4H, H5(Ar)), 6.76 (br s, 4H, H6(Ar)),

2.81 (br s, 20H, CH₃OCH₂), 2.70 (br s, 30H, CH₃OCH₂), 1.66 (s,

36H, 2-Bu^t), 1.39 (s, 36H, 4-Bu^t) ppm. ¹³C-{¹H} NMR

(125.8 MHz, C₆D₆, 298 K): 160.4 (ArC_q-1), 137.8 and 136.4

(ArC_q-2 and ArC_q-4), 124.0 (ArC-5), 123.6 (ArC-3), 120.3 (ArC-6),

70.7 (CH₃OCH₂), 59.7 (CH₃OCH₂), 35.3 (2-C(CH₃)₃), 34.1 (4-C-

(CH₃)₃), 32.1 (4-C(CH₃)₃), 30.5 (2-C(CH₃)₃) ppm. IR (Nujol

mull): 1598 (m), 1525 (w), 1307 (m), 1282 (m), 1202 (m),

1143 (m), 1115 (m), 1084 (w), 1070 (m), 1029 (w), 932 (w),

>99%) dissolved in THF at a concentration of 5.0 mg mL⁻¹

.

The matrix used was trans-2-[3-(4-tert-butylphenyl)-2-methyl-2-

propenylidene]-malononitrile (DCTB) (Fluka) and was dis-

solved in THF at a concentration of 40 mg mL⁻¹. Solutions of

matrix, salt and polymer were mixed in a volume ratio of

4 : 1 : 4, respectively. The mixed solution was hand spotted on

a stainless steel MALDI target and left to dry. The spectra were

recorded in the reflectron mode.

Polymer molecular weights (M_w, M_nand PDI) were deter-

mined by GPC using a Polymer Laboratories PL-GPC50 Plus

instrument equipped with a Polymer Laboratories Plgel Mixed-

D column (300 mm length, 7.5 mm diameter) and a refractive

index (RI) detector. Samples were dissolved in THF (Fisher,

HPLC grade, stabilised with BHT, 2.5 ppm) at a concentration

of 2.0 mg mL⁻¹and filtered prior to injection. THF (Fisher,

HPLC grade, stabilised with BHT, 2.5 ppm) was used as the

905 (w), 889 (w), 863 (w), 829 (w), 665 (w), 634 (w) cm⁻¹

.

[Sr₃(μ-DBP)₆(THF)₆] (3)

eluent at 30 °C and the flow rate was set to 1.0 mL min⁻¹

.

Linear polystyrenes (Polymer Laboratories) were used as the Strontium metal filings (0.90 g, 10.3 mmol), bis(pentafluoro-

primary calibration standards and the appropriate Mark– phenyl)mercury (0.56 g, 1.0 mmol), 2,4-di-tert-butylphenol

Houwink corrections for PLA in THF at 30 °C were used to cal- (0.47 g, 2.3 mmol), and Hg metal (3 drops), were added to a

culate the experimental molecular weights.³¹

Schlenk flask containing dry THF (20 mL), and ultrasonicated

Dalton Trans.

View Article Online

Dalton Transactions

Paper

for 3 d. The resulting yellow solution was filtered and concen- THF-d₈, 333 K): 7.09 (s, 4H, H3(Ar)), 6.82 (d, 4H, H5(Ar)), 6.71

trated under vacuum. Small rectangular block crystals suitable (vbr s, 4H, H6(Ar)), 3.43 (s, 12H, CH₃OCH₂), 3.27 (s, 18H,

for X-ray crystallography grew over 2 d. Yield = 0.57g (77%), CH₃OCH₂), 1.44 (s, 36H, 2-Bu^t), 1.23 (s, 36H, 4-Bu^t). ¹³C-{¹H}

dec. temp. 150 °C. Anal. found (calcd for C₁₀₈H₁₇₄O₁₂Sr₃): C, NMR (125.8 MHz, C₆D₆, 298 K): 162.8 (ArC_q-1), 137.3 and

66.39 (67.30); H, 9.09 (9.10) %; the low %C found for 3 is 136.3 (ArC_q-2 and ArC_q-4), 124.3 (ArC-3), 124.0 (ArC-5), 120.2

1

attributed to partial loss of THF on drying in vacuo. H NMR (ArC-6), 71.0 (CH₃OCH₂), 59.1 (CH₃OCH₂) 35.5 (2-C(CH₃)₃),

(300.13 MHz, C₆D₆, 303 K): 7.42 (s, 3H, ArH), 7.25 (s, overlap- 34.2 (4-C(CH₃)₃), 32.3 (4-C(CH₃)₃), 30.7 (2-C(CH₃)₃) ppm. IR

ping, 9H, ArH), 6.75 (s, 3H, ArH), 5.79 (s, 3H, ArH), 3.43 (s, (Nujol mull,): 1598 (s), 1525 (w), 1307 (m), 1272 (m), 1201 (m),

24H, OCH₂CH₂), 1.54 (s, 27H, Bu^t), 1.41 and 1.39 (s, overlap- 1146 (w), 1111 (w), 1084 (w), 1071 (m), 1016 (w), 925 (w), 907

1

ping, 54H, Bu^t), 1.32 (s, 24H, OCH₂CH₂), 1.10 (s, 27H, Bu^t). H (w), 887 (w), 856 (m), 830 (m), 667 (m), 638 (m) cm⁻¹

.

NMR (300.13 MHz, C₆D₆, 343 K): 7.24 (s, 6H, H3(Ar)), 6.72

Attempted synthesis of [Sr₃(DBP)₆(THF)₆] in the absence of

Hg(C₆F₅)₂

(br s, 6H, H5(Ar)), 5.73 (br s, 6H, H6(Ar)), 3.51 (s, 24H,

OCH₂CH₂), 1.43 (s, 24H, OCH₂CH₂), 1.36 (s, 54H, 2-Bu^t), 1.09

1

(s, 54H, 4-Bu^t). H NMR (300.13 MHz, THF-d₈, 303 K): 7.08 (s, Strontium metal filings (0.90 g, 10.3 mmol), 2,4-di-tert-butyl-

6H, H3(Ar)), 6.84 (d, 6H, H5(Ar)), 3.61 (m, THF, integration phenol (0.47 g, 2.3 mmol), and Hg metal (3 drops), were added

could not be determined owing to solvent resonance), 1.78 (m, to a Schlenk flask containing dry THF (20 mL), and ultrasoni-

THF, integration could not be determined owing to solvent reso- cated for 3 d. The reaction mixture was filtered and concen-

nance), 1.44 (s, 54H, 2-Bu^t), 1.24 (s, 54H, 4-Bu^t), H6 evidently trated to dryness under vacuum. Analysis of the bulk material

broadened into baseline. ¹H NMR (300.13 MHz, THF-d₈, by ¹H NMR and IR spectroscopy showed only starting

333 K): 7.09 (s, 6H, H3(Ar)), 6.83 (d, 6H, H5(Ar)), 6.75 (vbr s, materials.

6H, H6(Ar)), 3.61 (s, THF, integration could not be determined

owing to solvent resonance), 1.76 (s, THF, integration could

[Y(DBP)₃(THF)₂] (5)

not be determined owing to solvent resonance), 1.44 (s, 54H, A solution of 2,4-di-tert-butylphenol (0.53 g, 2.54 mmol) in

2-Bu^t), 1.24 (s, 54H, 4-Bu^t). ¹³C-{¹H} NMR (125.8 MHz, C₆D₆, THF (10 mL) was added dropwise to a stirring solution of

298 K): 159.6 (ArC_q-1, two overlapping), 138.6 (ArC_q-2 or [Y{N(SiHMe₂)₂}₃(THF)₂] (0.53 g, 0.85 mmol) in THF (10 mL) at

ArC_q-4), 137.4 (ArC_q-2 and ArC_q-4, three overlapping), 125.3, room temperature. The reaction mixture was stirred for 16 h,

124.7 and 124.1 (ArC-3 and ArC-5, four resonances with two during which time the solution colour changed from colour-

overlapping), 119.6 (ArC-6, two overlapping), 68.0 (OCH₂CH₂), less to yellow but remained clear. The volatiles were removed

35.3 and 35.0 (2-C(CH₃)₃), 34.3 and 34.2 (4-C(CH₃)₃), 32.2 (4-C- under reduced pressure and the crude solids were dissolved in

(CH₃)₃, two overlapping), 30.7 and 30.5 (2-C(CH₃)₃) and 25.5 THF (10 mL). The solution was filtered and then concentrated

(OCH₂CH₂) ppm. IR (Nujol mull): 1599 (m), 1401 (w), 1303 (w), under reduced pressure. Hexanes were added (10 mL) and the

1269 (s), 1201 (m), 1151 (m), 1126 (w), 1072 (m), 1036 (m), title compound precipitated as a white solid, which was iso-

886 (w), 829 (m), 727 (m), 669 (w), 636 (w) cm⁻¹. Single crystals lated by filtration and dried in vacuo at 0 °C. Yield = 0.40 g

obtained from THF (unit cell similar to that of the benzene (56%). Anal. found (calcd for C₅₀H₇₉O₅Y): C, 70.56 (70.73); H,

1

4

solvate) provided only a connectivity quality structure. Small 9.29 (9.38)%. H NMR (500.3 MHz, C₆D₆, 298 K): 7.60 (d, J =

3

4

crystals for a higher quality synchrotron structure determi- 2.6 Hz), 3H, H3(Ar)), 7.35 (dd, J = 8.4 Hz, J = 2.6 Hz, 3H, H5

nation, of composition 3·C₆H₆were grown from benzene.

(Ar)), 7.13 (d, ³J = 8.4 Hz, 3H, H6(Ar)), 3.73 (br s, 8H,

OCH₂CH₂), 1.76 (s, 27H, 2-Bu^t), 1.38 (s, 27H, 4-Bu^t), 1.06 (br s,

8H, OCH₂CH₂) ppm. ¹³C-{¹H} NMR (125.8 MHz, C₆D₆, 298 K):

[Sr₂(DBP)(μ-DBP)₃(DME)₃] (4)

Strontium metal filings (0.90 g, 10.3 mmol), bis(pentafluoro- 161.3 (ArC_q-1), 138.2 (ArC_q-2), 135.9 (ArC_q-4), 123.9 (ArC-3),

phenyl)mercury (0.56 g, 1.0 mmol), 2,4-di-tert-butylphenol 123.4 (ArC-5), 121.8 (ArC-6), 68.0 (OCH₂CH₂), 35.5 (2-C(CH₃)₃),

(0.47 g, 2.3 mmol), and Hg metal (∼3 drops), were added to a 34.3 (4-C(CH₃)₃), 32.1 (4-C(CH₃)₃), 30.5 (2-C(CH₃)₃), 25.7

Schlenk flask with dry DME (30 mL). Ultrasonication for 24 h, (OCH₂CH₂) ppm. IR (Nujol mull): 1602 (w), 1485 (s), 1419 (m),

followed by filtration and concentration of the resulting pale 1361 (s), 1219 (m), 1203 (m), 1153 (w), 1086 (m), 1018 (m),

yellow solution yielded large, colourless block crystals. Yield = 827(s), 742 (m) cm⁻¹

0.62g (85%), dec. temp. 260 °C. Anal. found (calcd for

.

Representative procedure for room temperature rac-LA

C₆₈H₁₁₄O₁₀Sr₂): C, 64.54 (64.47); H, 9.22 (9.07) %. ¹H NMR

(300.13 MHz, C₆D₆, 303 K): 7.45 (s, 4H, H3(Ar)), 7.14 (br s, 4H,

polymerisations without a co-initiator

H5(Ar), 6.71 (br s, 4H, H6(Ar)), 2.98 (s, 12H, CH₃OCH₂), 2.79 (s, In a glove-box, rac-LA (0.20 g, 1.39 mmol, 100 equiv.) was

18H, CH₃OCH₂), 1.59 (s, 36H, 2-Bu^t), 1.44 (s, 36H, 4-Bu^t). ¹H weighed into a vial and dissolved in THF (1.5 mL). The

NMR (300.13 MHz, toluene-d₈, 193 K): 7.81, 7.64, 7.47(s, ArH), monomer solution was added quickly in one portion to a stir-

3.18, 3.07, 2.61(br s, DME), 1.90, 1.80, 1.60, 1.53, 1.45 (br s, ring solution of [Ca₂(DBP)₄(DME)₄(μ-DME)] (2), [Sr₃(μ-DBP)₆-

Bu^t). ¹H NMR (300.13 MHz, THF-d₈, 303 K): 7.07 (s, 4H, H3 (THF)₆] (3), and [Sr₂(DBP)(μ-DBP)₃(DME)₃] (4) or [Y(DBP)₃-

(Ar)), 6.82 (d, 4H, H5(Ar)), 3.43 (s, 12H, CH₃OCH₂), 3.27 (s, (THF)₂] (5) ([rac-LA]₀: [Complex]₀= 100 : 1) in THF (0.5 mL).

18H, CH₃OCH₂), 1.44 (s, 36H, 2-Bu^t), 1.24 (s, 36H, 4-Bu^t), H6 The vial was sealed and the reaction mixture stirred until high

evidently broadened into baseline. ¹H NMR (300.13 MHz, conversion was recorded, before immediate quenching outside

Dalton Trans.

View Article Online

Paper

Dalton Transactions

the glove-box by addition of THF (as supplied, ≈0.5 mL). Four and processed with the XDS⁴⁴software package. The structures

aliquots (≈0.2 mL) were taken and dried in vacuo. These were solved using direct methods, and observed reflections

samples were used for ¹H NMR (CDCl₃), MALDI-ToF MS and were used in least squares refinement on F², with anisotropic

GPC analysis.

displacement parameters refined for non-hydrogen atoms.

Hydrogen atoms were constrained in calculated positions and

refined with a riding model. Structure solutions and refine-

ments were performed using the programs SHELXS-97⁴⁵and

Representative procedure for room temperature rac-LA

polymerisations with co-initiators

In a glove-box, rac-LA (0.20 g, 1.39 mmol, 100 equiv.) was SHELXL-97⁴⁶through the graphical interface X-Seed,⁴⁷which

weighed into a vial and dissolved in THF (1.5 mL). The was also used to generate the Figures. Other details of the

monomer solution was added quickly in one portion to a structure solution and refinements are given in the ESI† (CIF

stirred mixture of [Ca₂(DBP)₄(DME)₄(μ-DME)] (2), [Sr₃(μ-DBP)₆- data). A full listing of atomic coordinates, bond lengths and

(THF)₆] (3), and [Sr₂(DBP)(μ-DBP)₃(DME)₃] (4) or [Y(DBP)₃- angles and displacement parameters for all the structures have

(THF)₂] (5) ([rac-LA]₀: [Complex]₀= 100 : 1) and co-initiator been deposited at the Cambridge Crystallographic Data Centre

([Co-initiator]₀: [Complex]₀= 5 : 1 in THF (0.5 mL). An appro- (CCDC 914477–914480).

priate volume of BnNH₂or BnOH was measured using a

Crystal data for 1: C₄₄H₇₄CaO₆, M = 739.11, colourless block,

Hamilton microlitre syringe. The vial was sealed and the reac- 0.08 × 0.05 × 0.05 mm³, monoclinic, space group P2₁/c (No.

tion mixture stirred for 30 min before immediate quenching 14), a = 9.778(2), b = 9.6717(19), c = 22.946(5) Å, β = 94.60(3)°,

outside the glove-box by addition of THF (as supplied, V = 2163.1(8) Å³, Z = 2, D_c= 1.135 g cm⁻³, F000 = 812, 2θ_max

=

≈0.5 mL). Four aliquots (≈0.2 mL) were taken and dried 55.0°, 19375 reflections collected, 4963 unique (R_int= 0.0750).

in vacuo. These samples were used for ¹H NMR (CDCl₃), Final GooF = 1.049, R₁= 0.0477, wR₂= 0.1230, R indices based

MALDI-ToF MS and GPC analysis.

on 3917 reflections with I > 2σ(I) (refinement on F²), 238 para-

meters, 0 restraints. Lp and absorption corrections applied,

Representative kinetics procedures for room temperature

rac-LA polymerisations using [Ca₂(DBP)₄(DME)₄(μ-DME)] (2)

or[Sr₂(DBP)(μ-DBP)₃(DME)₃] (4)

μ = 0.188 mm⁻¹

.

Crystal data for 2: C₇₆H₁₃₄Ca₂O₁₄, M = 1351.99, colourless

block, 0.2 × 0.2 × 0.1 mm³, monoclinic, space group C2/c

In a glove-box, rac-LA (0.80 g, 2.78 mmol, [rac-LA]₀: [Complex]₀= (No. 15), a = 24.136(3), b = 13.9781(14), c = 25.508(4) Å, β =

400 : 1.0–5.0) was weighed into a vial and dissolved in THF 112.206(4)°, V = 7967.5(17) Å³, Z = 4, D_c= 1.127 g cm⁻³, F000 =

(7.5 mL). The monomer solution was added quickly in one 2968, 2θ_max= 55.0°, 28504 reflections collected, 9130 unique

portion to a stirred mixture of 2 or 4 ([rac-LA]₀: [Complex]₀

=

(R_int= 0.0781). Final GooF = 1.069, R₁= 0.0719, wR₂= 0.1760,

400 : 1.0–5.0) and co-initiator ([BnNH₂]₀: [Complex]₀= 12.5 : 1, R indices based on 5273 reflections with I > 2σ(I) (refinement

10.0 : 1, 7.5 : 1, 6.0 : 1, 5.0 : 1, 4.0 : 1 or 2.5 : 1 in THF (0.5 mL). on F²), 432 parameters, 127 restraints. Lp and absorption cor-

An appropriate volume of BnNH₂was measured using a Hamil- rections applied, μ = 0.200 mm⁻¹

.

ton microlitre syringe. Aliqouts (≈0.1 mL) were taken at the

Crystal data for 3·C₆H₆: C₁₁₄H₁₈₀O₁₂Sr₃, M = 2005.44, colour-

3

ˉ

appropriate time interval and were quenched inside the glove- less block, 0.04 × 0.02 × 0.01 mm , trigonal, space group R3c

box with degassed but not dried THF (6 drops). The volatiles (No. 167), a = 21.738(3), b = 21.738(3), c = 40.897(8) Å, γ = 120°,

were removed from the aliquots under reduced pressure and V = 16 737(5) Å³, Z = 6, D_c= 1.194 g cm⁻³, F000 = 6444, 2θ_max

=

dried in vacuo. The crude aliquots were dissolved in CDCl₃for 50.0°, 68739 reflections collected, 3272 unique (R_int= 0.0561).

determination of the monomer to polymer conversion by ¹H Final GooF = 1.105, R₁= 0.0519, wR₂= 0.1237, R indices based

NMR analysis. The conversion was measured by integration of on 3272 reflections with I > 2σ(I) (refinement on F²), 201 para-

the CH(Me) resonances of PLA and unreacted rac-LA. Aliquots meters, 24 restraints. Lp and absorption corrections applied,

were then recovered for GPC and MALDI-ToF MS analysis.

μ = 1.485 mm⁻¹

.

Crystal data for 4: C₆₈H₁₁₄O₁₀Sr₂, M = 1266.83, colourless

block, 0.20 × 0.20 × 0.08 mm³, monoclinic, space group P2₁/n

X-ray structure determinations for 1–4

Intensity data were collected using an Enraf–Nonius KAPPA (No. 14), a = 16.312(3), b = 25.924(5), c = 16.982(3) Å, β =

CCD (for 1) or a Bruker X8 Apex II CCD (for 2 and 4) at 123 K 96.27(3)°, V = 7138(2) Å³, Z = 4, D_c= 1.179 g cm⁻³, F000 = 2712,

with Mo-K_αradiation (λ = 0.71073 Å) or the Australian Synchro- 2θ_max= 55.0°, 50398 reflections collected, 16384 unique (R_int

=

tron MX1 beamline (for 3) at 100 K with single wavelength (λ = 0.0935). Final GooF = 1.020, R₁= 0.0560, wR₂= 0.1197,

0.712 Å). Suitable crystals were immersed in viscous hydro- R indices based on 10598 reflections with I > 2σ(I) (refinement

carbon oil and mounted on a glass fibre on the diﬀractometer on F²), 734 parameters, 0 restraints. Lp and absorption correc-

or a cryoloop on the beamline. From KAPPA or Apex defracto- tions applied, μ = 1.545 mm⁻¹

meters, psi and omega scans, N_t(total) reflections were

.

measured and reduced to No. of unique reflections, with F_o>

2σ(F_o) being considered observed. Data were initially processed

Acknowledgements

and corrected for absorption using the programs DENZO⁴⁰

and SORTAV,⁴¹or the Bruker Apex II program suite.⁴²With the We thank the Australian Research Council and the UK EPSRC

MX1 beamline, data were collected using the Blue Ice⁴³GUI and EU COST Action CM1006 for support. Aspects of this

Dalton Trans.

View Article Online

Dalton Transactions

Paper

research were undertaken on the MX1 beamline at the Austra-

lian Synchrotron, Victoria, Australia.

2003, 27, 1167; (c) H. D. Dyer, S. Huijser, N. Susperregui,

F. Bonnet, A. D. Schwarz, R. Duchateau, L. Maron and

P. Mountford, Organometallics, 2010, 29, 3602.

5 O. Coulembier, M. K. Kiesewetter, A. Mason, P. Dubois,

J. L. Hedrick and R. M. Waymouth, Angew. Chem., Int. Ed.,

2007, 46, 4719.

References

1 (a) E. Chiellini and R. Solaro, Adv. Mater., 1996, 8, 305;

(b) K. E. Uhrich, S. M. Cannizzaro, R. S. Langer and

K. M. Shakesheﬀ, Chem. Rev., 1999, 99, 3181; (c) Y. Ikada

and H. Tsuji, Macromol. Rapid Commun., 2000, 21, 117;

(d) R. E. Drumright, P. R. Gruber and D. E. Henton, Adv.

Mater., 2000, 12, 1841; (e) A.-C. Albertsson and I. K. Varma,

Biomacromolecules, 2003, 4, 1466; (f) R. Auras, B. Harte and

S. Selke, Macromol. Biosci., 2004, 4, 835; (g) C. K. Williams

and M. A. Hillmyer, Polym. Rev., 2008, 48, 1; (h) A. P. Dove,

6 Q. Cai, Y. Zhao, J. Bei, F. Xi and S. Wang, Biomacro-

molecules, 2003, 2003, 828.

7 A. Kowalski, J. Libiszowski, T. Biela, M. Cypryk, A. Duda

and S. Penczek, Macromolecules, 2005, 38, 8170.

8 (a) L. Clark, M. G. Cushion, H. D. Dyer, A. D. Schwarz,

R. Duchateau and P. Mountford, Chem. Commun., 2010, 46,

273; (b) L. Clark, D.Phil. Thesis, University of Oxford, 2012.

9 M. P. Blake, A. D. Schwarz and P. Mountford, Organo-

metallics, 2011, 30, 1202.

Chem. Commun., 2008, 6446; (i) M. Kakuta, M. Hirata and 10 L. Clark, G. B. Deacon, C. M. Forsyth, P. C. Junk,

Y. Kimura, J. Macromol. Sci., Polym. Rev., 2009, 49, 107;

( j) E. S. Place, J. H. George, C. K. Williams and

M. M. Stevens, Chem. Soc. Rev., 2009, 38, 1139.

2 (a) H. Keul and H. Hocker, Macromol. Rapid Commun.,

2000, 21, 869; (b) M. Okada, Prog. Polym. Sci., 2002, 27, 87;

(c) X. Lou, C. Detrembleur and C. Jérôme, Macromol. Rapid

Commun., 2003, 24, 161; (d) S. Penczek, M. Cypryk,

A. Duda, P. Kubisa and S. Slomkowski, Prog. Polym. Sci.,

P. Mountford and J. P. Townley, Dalton Trans., 2010, 39,

6693.

11 (a) W. M. Stevels, M. J. K. Ankoné, P. J. Dijkastra and

J. Feijen, Macromolecules, 1996, 29, 3332; (b) W. M. Stevels,

M. J. K. Ankoné, P. J. Dijkastra and J. Feijen, Macro-

molecules, 1996, 29, 6132; (c) C. Yu, L. Zhang, X. Ni, Z. Shen

and K. Tu, J. Polym. Sci., Part A: Polym. Chem., 2004, 42,

6209.

2007, 32, 247; (e) C. K. Williams, Chem. Soc. Rev., 2007, 36, 12 H.-Y. Chen, L. Mialon, K. A. Abboud and S. A. Miller, Orga-

1573; (f) N. E. Kamber, W. Jeong, R. M. Waymouth, nometallics, 2012, 31, 5252–5261.

R. C. Pratt, B. G. G. Lohmeijer and J. L. Hedrick, Chem. 13 (a) W. J. Evans and H. Katsumata, Macromolecules, 1994,

Rev., 2007, 107, 5813; (g) D. Bourissou, S. Moebs-Sanchez

and B. Martin-Vaca, C. R. Chim., 2007, 10, 775;

(h) A.-C. Albertsson and R. K. Srivastava, Adv. Drug Delivery

Rev., 2008, 60, 1077; (i) C. Jérôme and P. Lecomte, Adv.

27, 2330; (b) C. Iftner, F. Bonnet, F. Nief, M. Visseaux and

L. Maron, Organometallics, 2011, 30, 4482; (c) A. Momin,

F. Bonnet, M. Visseaux, L. Maron, J. Takats, M. J. Ferguson,

X.-F. Le Goﬀ and F. Nief, Chem. Commun., 2011, 47, 12203.

Drug Delivery Rev., 2008, 60, 1056; ( j) M. Labet and 14 R. D. Shannon, Acta Crystallogr., Sect. A: Cryst. Phys., Diﬀr.,

W. Thielemans, Chem. Soc. Rev., 2009, 38, 3484; Theor. Gen. Cryst., 1976, 32, 751.

(k) H. Yasuda and E. Ihara, Bull. Chem. Soc. Jpn., 1997, 70, 15 (a) T. P. Hanusa, in Comprehensive Organometallic Chemistry

1745; (l) B. J. O’Keefe, M. A. Hillmyer and W. B. Tolman,

J. Chem. Soc., Dalton Trans., 2001, 2215; (m) G. W. Coates,

J. Chem. Soc., Dalton Trans., 2002, 467; (n) O. Dechy-

Cabaret, B. Martin-Vaca and D. Bourissou, Chem. Rev.,

2004, 104, 6147; (o) J. Wu, T.-L. Yu, C.-T. Chen and

C.-C. Lin, Coord. Chem. Rev., 2006, 250, 602;

(p) A. Amgoune, C. M. Thomas and J.-F. Carpentier, Pure

Appl. Chem., 2007, 79, 2013; (q) R. H. Platel, L. M. Hodgson

and C. K. Williams, Polym. Rev., 2008, 48, 11;

(r) C. A. Wheaton, P. G. Hayes and B. J. Ireland, Dalton

III, ed. R. H. Crabtree and D. M. P. Mingos, Elsevier,

Oxford, 2007, vol. 2; (b) M. Westerhausen, M. Gärtner,

R. Fischer and J. Langer, Angew. Chem., Int. Ed., 2007, 46,

1950; (c) M. Westerhausen, Coord. Chem. Rev., 2008, 252,

1516; (d) J. D. Smith, Angew. Chem., Int. Ed., 2009, 48, 6597;

(e) A. G. M. Barrett, M. R. Crimmin, M. S. Hill and

P. A. Procopiou, Proc. R. Soc. London, Ser. A, 2010, 466, 927;

(f) W. D. Buchanan, D. G. Allis and K. Ruhlandt-Senge,

Chem. Commun., 2010, 46, 4449; (g) S. Harder, Chem. Rev.,

2010, 110, 3852.

Trans., 2009, 4832; (s) N. Ajella, J.-F. Carpentier, 16 (a) M. H. Chisholm and N. W. Eilerts, Chem. Commun.,

C. Guillaume, S. M. Guillaume, M. Helou, V. Poirier,

Y. Sarazin and A. A. Trifonov, Dalton Trans., 2010, 39, 8363;

(t) M. J. Stanford and A. P. Dove, Chem. Soc. Rev., 2010, 39,

486; (u) I. dos Santos Vieira and S. Herres-Pawlis,

Eur. J. Inorg. Chem., 2012, 765.

3 (a) Y. Sarazin, V. Poirier, T. Roisnel and J.-F. Carpentier,

Eur. J. Inorg. Chem., 2010, 3423; (b) Y. Sarazin, B. Liu,

T. Roisnel, L. Maron and J.-F. Carpentier, J. Am. Chem. Soc.,

2011, 133, 9069.

1996, 853; (b) M. H. Chisholm, N. W. Eilerts, J. C. Huﬀman,

S. S. Iyer, M. Pacold and K. Phomphrai, J. Am. Chem. Soc.,

2000, 122, 11845; (c) B. M. Chamberlain, M. Cheng,

D. R. Moore, T. M. Ovitt, E. Lobkovsky and G. W. Coates,

J. Am. Chem. Soc., 2001, 123, 3229; (d) M. H. Chisholm,

J. Gallucci and K. Phomphrai, Chem. Commun., 2003, 48;

(e) M. S. Hill and P. B. Hitchcock, Chem. Commun., 2003,

1758; (f) M. Westerhausen, S. Schneiderbauer,

A. N. Kneifel, Y. Söltl, P. Mayer, H. Nöth, Z. Zhong,

P. J. Dijkastra and J. Feijen, Eur. J. Inorg. Chem., 2003, 3432;

(g) Z. Zhong, S. Schneiderbauer, P. J. Dijkastra,

4 (a) H. Y. Ma and J. Okuda, Macromolecules, 2005, 38, 2665;

(b) M. H. Chisholm and E. E. Delbridge, New J. Chem.,

Dalton Trans.

View Article Online

Paper

Dalton Transactions

M. Westerhausen and J. Feijen, Polym. Bull., 2003, 51, 175; 22 J. Hitzbleck, Ph.D. Thesis, Syracuse University, Syracuse,

(h) M. H. Chisholm, J. C. Gallucci and K. Phomphrai, Inorg.

Chem., 2004, 43, 6717; (i) A. P. Dove, V. C. Gibson,

NY, 2004; J. Hitzbleck, G. B. Deacon and K. Ruhlandt-

Senge, unpublished results.

E. L. Marshall, A. J. P. White and D. J. Williams, Dalton 23 T. J. Boyle, B. A. Hernandez-Sanchez, C. M. Baros,

Trans., 2004, 570; ( j) M. H. Chisholm, J. C. Gallucci and

K. Phomphrai, Inorg. Chem., 2005, 44, 8004; (k) T. Chivers,

L. N. Brewer and M. A. Rodriguez, Chem. Mater., 2007, 19,

2016.

C. Fedorchuk and M. Parvez, Organometallics, 2005, 24, 24 (a) J. P. Dunne, M. Tacke, C. Selinke and D. Stalke,

580; (l) J. Ejfler, M. Kobylka, L. B. Jerzykiewicz and

P. Sobota, Dalton Trans., 2005, 2047; (m) M. G. Davidson,

C. T. O’Hara, M. D. Jones, C. G. Keir, M. F. Mahon and

G. Kociok-Köhn, Inorg. Chem., 2007, 46, 7686; (n) V. Poirier,

Eur. J. Inorg. Chem., 2003, 1416; (b) W. D. Buchanan,

M. A. Guino-o and K. Ruhlandt-Senge, Inorg. Chem., 2010,

59, 7144; (c) O. Michel, C. Meermann, K. W. Törnroos and

R. Anwander, Organometallics, 2009, 28, 4783.

T. Roisnel, J.-F. Carpentier and Y. Sarazin, Dalton Trans., 25 (a) P. B. Hitchcock, M. F. Lappert, G. A. Lawless and

2009, 9820; (o) D. J. Darensbourg, W. Choi,

O. Karroonnirun and N. Bhuvanesh, Macromolecules, 2008,

41, 3493; (p) J. Ejfler, K. Krauzy-Dziedzic, S. Szafert,

B. Royo, J. Chem. Soc., Chem. Commun., 1990, 1141;

(b) K. F. Tesh, T. P. Hanusa, J. C. Huﬀman and

C. J. Huﬀman, Inorg. Chem., 1992, 31, 5572.

L. B. Jerzykiewicz and P. Sobota, Eur. J. Inorg. Chem., 2010, 26 G. Becker, M. Niemeyer, O. Mundt, W. Schwarz,

3602; (q) Y. Sarazin, D. Rosca, V. Poirier, T. Roisnel,

A. Silvestru, L. Maron and J.-F. Carpentier, Organometallics,

2010, 29, 6569; (r) M. G. Cushion and P. Mountford, Chem.

M. Westerhausen, M. W. Ossberger, P. Mayer, H. Nöth,

Z. Zhong, P. J. Dijkstra and J. Feijen, Z. Anorg. Allg. Chem.,

2004, 630, 2605.

Commun., 2011, 47, 2276; (s) V. Poirier, T. Roisnel, 27 (a) W. J. Evans, W. G. McClelland, M. A. Greci and

J.-F. Carpentier and Y. Sarazin, Dalton Trans., 2011, 40, 523;

(t) B. Liu, V. Dorcet, L. Maron, J.-F. Carpentier and

Y. Sarazin, Eur. J. Inorg. Chem., 2012, 3023; (u) B. Liu,

J. W. Ziller, Eur. J. Solid State Inorg. Chem., 1996, 33, 145;

(b) G. B. Deacon, P. C. Junk and G. J. Moxey, New J. Chem.,

2010, 34, 1731.

T. Roisnel, J. P. Guegan, J.-F. Carpentier and Y. Sarazin, 28 (a) G. B. Deacon, P. C. Junk, G. J. Moxey, M. A. Guino-o and

Chem.–Eur. J., 2012, 18, 6289; (v) B. Liu, T. Roisnel and

Y. Sarazin, Inorg. Chim. Acta, 2012, 380, 2; (w) S. Marks,

M. Kuzdrowska, P. W. Roesky, L. Annunziata,

S. M. Guillaume and L. Maron, ChemPlusChem, 2012, 77,

K. Ruhlandt-Senge,

Dalton

Trans.,

2009,

4878;

(b) W. J. Evans, M. A. Greci and J. W. Ziller, Inorg. Chem.,

2000, 39, 3213; (c) G. B. Deacon, P. C. Junk and G. J. Moxey,

Chem.–Asian J., 2009, 4, 1717.

350; (x) R. A. Collins, J. Unruangsri and P. Mountford, 29 (a) W. J. Evans, M. A. Greci and J. W. Ziller, J. Chem. Soc.,

Dalton Trans., 2013, 42, 759.

Chem. Commun., 1998, 2367; (b) G. B. Deacon,

C. M. Forsyth and P. C. Junk, J. Organomet. Chem., 2000,

607, 112; (c) G. B. Deacon, C. M. Forsyth, P. C. Junk,

B. W. Skelton and A. H. White, Chem.–Eur. J., 1999, 5, 1452.

17 (a) D. C. Bradley, R. C. Mehrotra, I. P. Rothwell and

A. Singh, Alkoxo and Aryloxo Derivatives of Metals, Academic

Press, London, 2001; (b) M. N. Bochkarev, L. N. Zakharov

and G. S. Kalinina, Organoderivatives of the Rare Earth 30 G. B. Deacon, P. C. Junk, G. J. Moxey, K. Ruhlandt-Senge,

Elements, Kluwer, Dordrecht, 1995; (c) S. A. Cotton, in Com- C. St. Prix and M. F. Zuniga, Chem.–Eur. J., 2009, 15, 5503.

prehensive Coordination Chemistry II, ed. J. A. McCleverty 31 J. R. Dorgan, J. Janzen, D. M. Knauss, S. B. Hait,

and T. J. Meyer, Elsevier, Oxford, 2004, vol. 3, ch. 3.2, p. 93.

18 (a) G. B. Deacon, G. D. Fallon, C. M. Forsyth, S. C. Harris,

B. R. Limoges and M. H. Hutchinson, J. Polym. Sci., Part B:

Polym. Phys., 2005, 43, 3100.

P. C. Junk, B. W. Skelton and A. H. White, Dalton Trans., 32 M. T. Zell, B. E. Padden, A. J. Paterick, K. A. M. Thakur,

2006, 802; (b) G. B. Deacon and C. M. Forsyth, Inorganic

Chemistry Highlights, Wiley-VCH, 2002; (c) G. B. Deacon,

R. T. Kean, M. A. Hillmyer and E. J. Munson, Macro-

molecules, 2002, 35, 7700.

C. M. Forsyth and S. Nickel, J. Organomet. Chem., 2002, 33 J. March, Advanced Organic Chemistry, 4th edn, Wiley-

647, 50. Interscience, New York, 1992.

19 (a) H. Schumann, J. A. Meese-Marktscheﬀel and L. Esser, 34 F. Bonnet, A. R. Cowley and P. Mountford, Inorg. Chem.,

Chem. Rev., 1995, 95, 865; (b) F. T. Edelmann, in Compre- 2005, 44, 9046.

hensive Organometallic Chemistry II, ed. G. Wilkinson, 35 (a) T. P. A. Cao, A. Buchard, X. F. Le Goﬀ, A. Auﬀrant and

F. G. A. Stone and E. W. Abel, Pergamon, Oxford, 1995,

vol. 4, ch. 2, p. 11; (c) R. Anwander, Top. Organmet. Chem.,

C. K. Williams, Inorg. Chem., 2012, 51, 2157; (b) P. I. Binda

and E. E. Delbridge, Dalton Trans., 2007, 4685.

1999, 2, 1; (d) R. C. Mehrotra, A. Singh and U. M. Tripathi, 36 Although 4 can polymerise rac-LA in the absence of BnNH₂,

Chem. Rev., 1991, 91, 1287; (e) W. J. Evans, New J. Chem.,

1995, 19, 525; (f) L. G. Hubert-Pfalzgraf, Coord. Chem. Rev.,

1998, 178, 967.

the reaction is very slow (30 min to convert 51 equiv. with

[rac-LA]₀: [4]₀= 400 : 4 ( = 100 : 1)) compared to when an

amine or alcohol is present. To a first approximation, poly-

merisations reactions initiated by 4 (or 2) alone may be neg-

lected over the timescale of these experiments.

20 J. Marcalo and A. P. de Matos, Polyhedron, 1989, 8, 2431.

21 C. M. Forsyth and G. B. Deacon, Organometallics, 2000, 19,

1205; G. B. Deacon and C. M. Forsyth, Organometallics, 37 R. D. Köhn, Z. Pan, J. Sun and C. Liang, Catal. Commun.,

2003, 22, 1349. 2003, 4, 33.

Dalton Trans.

View Article Online

Dalton Transactions

Paper

38 R. Anwander, O. Runte, J. Eppinger, G. Gerstberger, 42 Apex II Version 2.1, Bruker AXS Ltd., Madison, Wisconsin.

E. Herdtweck and M. Speigler, J. Chem. Soc., Dalton Trans., 43 T. M. McPhillips, S. E. McPhillips, H. J. Chiu, A. E. Cohen,

1998, 847.

A. M. Deacon, P. J. Ellis, E. Garman, A. Gonzalez,

N. K. Sauter, R. P. Phizackerley, S. M. Soltis and P. J. Kuhn,

J. Synchrotron Radiat., 2002, 9, 401.

39 G. B. Deacon, J. E. Cosgriﬀ, E. T. Lawrenz, C. M. Forsyth

and D. L. Wilkinson, in Synthetic Methods of Organometallic

and Inorganic Chemistry, ed. W. A. Herrmann, Theime, 44 W. Kabsch, J. Appl. Crystallogr., 1993, 26, 795.

Stuttgart, 1997, vol. 6, p. 48.

40 Z. Otwinowski and W. Minor, Methods Enzymol., 1997, 276,

45 G. M. Sheldrick, SHELXS-97 Program for crystal structure

solution, University of Gottingen, Germany, 1997.

307.

46 G. M. Sheldrick, SHELXL-97 Program for crystal structure

refinement, University of Gottingen, Germany, 1997.

47 L. J. Barbour, J. Supramol. Chem., 2001, 1, 189.

41 R. H. Blessing, Acta Crystallogr., Sect. A: Fundam. Crystal-

logr., 1995, 51, 33.

Dalton Trans.

Article Doi

Synthesis and structures of calcium and strontium 2,4-di-tert- butylphenolates and their reactivity towards the amine co-initiated ring-opening polymerisation of rac-lactide

DOI: 10.1039/c3dt00065f

Source and publish data:

Authors:

Article abstract of DOI:10.1039/c3dt00065f

Full text of DOI:10.1039/c3dt00065f

Products guided by the article

R&D Labs maybe for 1443003-45-9

Relevant to this article

Hot Product