Configurationally flexible zinc complexes as catalysts for rac-lactide polymerisation

Article abstract of DOI:10.1039/C8DT02562B

Zn(N(SiMe₃)₂)₂ was reacted with pyridinemethanol and R,R-N,N′-di(methylbenzyl)-2,5-diiminopyrrole (L1H) to afford the dimeric complex (L1)₂Zn₂(μ-OR)₂. The complex showed moderate activity i

Full text of DOI:10.1039/C8DT02562B

Dalton
Transactions
View Article Online
View Journal | View Issue
PAPER
Configurationally flexible zinc complexes as
catalysts for rac-lactide polymerisation†
Cite this: Dalton Trans., 2018, 47,
16279
Pargol Daneshmand, Ina Michalsky, Pedro M. Aguiar
and Frank Schaper
*
Zn(N(SiMe3)2)2 was reacted with pyridinemethanol and R,R–N,N’-di(methylbenzyl)-2,5-diiminopyrrole
(L1H) to afford the dimeric complex (L1)2Zn2(µ-OR)2. The complex showed moderate activity in rac-
lactide polymerization to heterotactic polymer (Pr = 0.75). 2,4-Di-tert-butyl-6-aminomethyl-phenol
ligands with amino = N,N,N’,N’-tetramethyldiethylenetriamine (L2H) or di-(2-picoly)amine (L3H) were
reacted with ZnEt2 to form (L2)ZnEt and with Zn(N(SiMe3)2)2 to form the respective amide complexes. All
complexes, including (L1)2Zn2(µ-OR)2 were characterised by X-ray diffraction studies. (L2)ZnEt was
unreactive toward ethanol, but the amide complexes afforded (L2)ZnOEt and (L3)ZnOEt upon reaction
with ethanol, which were used in rac-lactide polymerization without isolation. All complexes racemise
readily at room temperature and show apparent Cs-symmetry in their NMR spectra. The ethoxide com-
plexes were highly active in lactide polymerization, with (L3)ZnOEt reaching full conversion in 15 min at
0.5 mM catalyst concentration at room temperature. In both cases, introduction of a second donor arm
on the central nitrogen introduced a slight bias for isotactic monomer enchainment (Pm = 0.55–0.60),
which for (L3)ZnOEt was dependent on catalyst concentration.
Received 22nd June 2018,
Accepted 5th October 2018
DOI: 10.1039/c8dt02562b
rsc.li/dalton
isotactic stereocontrol – which seems currently to be the
biggest challenge: not only are the traditional stereocontrol
Introduction
Using polylactic acid (PLA) as a green replacement for pet- mechanisms, chain-end and catalytic-site control, both
roleum based resources has gained lot of interest observed in lactide polymerisation (sometimes opposing each
a
recently.1–10 PLA is prepared through ring opening polymeris- other), but other mechanisms, such as selective chain transfer,
ation (ROP) of lactide which itself is obtained through fermen- can be operative.43–45 In this context, there is growing evidence
tation of corn starch.11–13 Fuelled by its industrial application, that in some instances a higher flexibility of the catalytic site
controlled lactide polymerisation has become a catalytic can be beneficial, which is untypical for “traditional” stereo-
challenge.14–42 To date, there is no catalytic system which can control mechanisms. Based on initial observations by Ma,
polymerize lactide with high stereocontrol (Pm
>
95%) Okuda and Carpentier,46,47 and follow-up work of Davidson
(Scheme 1), excellent polymer molecular weight control and and Jones,48–51 a catalytic-site mediated (or ligand mediated)
good activities under industrially relevant conditions chain-end control mechanism was proposed. While the chiral
(140–180 °C, in the presence of water and lactic acid).
information is still derived from the polymer chain end, the
The situation is rendered more interesting (and made latter does not interact directly with the incoming monomer.
thoroughly more complicated) by the number of possible Rather, the catalytic site adapts its configuration to match the
mechanistic pathways for lactide polymerisation: in addition chirality of the chiral chain end, and in turn determines
to ring-opening polymerisation by anionic or neutral organic stereochemistry by interaction with the monomer. A flexible
initiators, lactide can be polymerized by Lewis-acid activation and preferably chiral catalytic site is a prerequisite for this
of the monomer together with a suitable co-initiator (alcohol) mechanism.
or by coordination–insertion polymerisation into a metal alk-
oxide catalyst. The same mechanistic multitude is observed for
Centre in Green Chemistry and Catalysis, Department of chemistry,
Université deMontréal, C. P. 6128 Succ. Centre-Ville, Montréal, QC H3 T 3J7,
Canada. E-mail: Frank.Schaper@umontreal.ca
†Electronic supplementary information (ESI) available: Additional details on
polymerisation reactions, X-ray structures and NMR spectra. CCDC
1850465–1850469. For ESI and crystallographic data in CIF or other electronic
format see DOI: 10.1039/c8dt02562b
Scheme 1
This journal is © The Royal Society of Chemistry 2018
Dalton Trans., 2018, 47, 16279–16291 | 16279

View Article Online
Paper
Dalton Transactions
Scheme 3
Scheme 2
We have recently observed this mechanism to be active in
copper diiminopyrrolide complexes to provide moderately iso-
tactic PLA.52–55 The active species of catalyst 1 is found to be a
dinuclear species, in which the penta-coordinated copper
centres are chiral, but can readily racemise due to the presence
of the pendant imino ligand (Scheme 2). Stereocontrol seemed
to be largely invariant of the nature of the N-substituent, but a
pyridylmethoxide ligand was essential. The only pyridylmeth-
oxide complex which did not provide isotactic PLA, 1b
(Scheme 2), showed an octahedrally coordinated copper in its
crystal structure and thus an achiral catalytic site, unable to
participate in the proposed mechanism.
Due to their abundance, low price, non-toxicity and general
lack of colour, zinc-based complexes have been expansively
investigated in homogenous lactide polymerisation from the
very beginning.56,57 Zinc-based complexes typically show good
to excellent activities and good polymer molecular weight
control. But despite numerous studies,14,15,17,20,24,26 including
our own,58 stereocontrol toward isotactic PLA was difficult to
achieve. Only in recent years, zinc-catalysts with preference for
isotactic monomer insertion emerged.59–73 In the following,
we explore if the stereocontrol mechanism observed in copper
diiminopyrrolide complexes can be transferred to zinc-based
catalysts, either using the identical ligand framework as for
copper or by designing a catalyst capable of catalytic site
racemisation.
Fig. 1 X-ray structure of 2. Thermal ellipsoids are drawn at the 50%
probability level. Hydrogen atoms omitted for clarity.
Complex 2 crystallised as a dimeric, pentacoordinated
complex, with distorted bipyramidal geometry around zinc (τ =
0.7,74 Fig. 1). While τ-values can be misleading, this agrees
with the observed metal–ligand bond lengths. In the distorted
square-pyramidal coordination of 1, the ligand in the apical
position showed
a notably (0.2–0.3 Å) elongated bond
(Table 1). In distorted bipyramidal 2, all zinc–ligand bond dis-
tances fall all in the range of 2.0–2.1 Å, irrespective of position,
Results and discussion
Diiminopyrrolide complexes
Table 1 Selected geometric data for pyridylmethoxide complexes 1
and 2a
Synthesis and structure. Synthesis of L1H (Scheme 3) has
been reported previously.50 The dimeric complex (L1)2Zn2(µ-
OCH2C5H4N)2, 2, was obtained similar to the analogous
copper(II) complexes by reaction of zinc bis–bis(trimethylsilyl)
amide with one equivalent of pyridinemethanol, followed by
addition of the ligand L1H (Scheme 3).50 If dimethyl-
aminoethanol was used instead of pyridinemethanol, the
corresponding homoleptic complex (L1)2Zn was obtained. For
the sake of comparison and characterisation, (L1)2Zn was also
prepared from reaction of Zn(N(SiMe3)2)2 with two equivalents
of L1H (see Fig. S1† for its crystal structure).
1b
2
M–NPyrrole
M–Nimine
M–Oshort
M–Olong
M–NPyridine
M–M
1.945(2), 1.964(2)
2.294(3), 2.242(3)
1.915(2), 1.944(2)
1.960(2), 1.960(2)
2.025(3), 1.995(3)
3.025(5)
2.084(2), 2.0896(19)
2.1205(19), 2.1095(18)
1.9908(16), 1.9850(15)
2.0744(16), 2.0666(17)
2.1084(19), 2.112(2)
3.1091(4)
τ
0.6, 0.4
0.7, 0.7
a The second values cited refers to the second metal center of the
dimer. b Taken from ref. 52 for comparison.
16280 | Dalton Trans., 2018, 47, 16279–16291
This journal is © The Royal Society of Chemistry 2018

View Article Online
Dalton Transactions
Paper
and are comparable to what has been reported in literature.75 molecular transesterification (Fig. S6†). We thus cannot deter-
Overall though, the structure of 2 resembles very strongly that mine if only one (as in 1) or if both pyridylmethoxide ligands
of 1 (Table 1, Fig. 1). Due to the differences in preferred coordi- initiate chain growth.
nation geometry,
1 showed a better defined equatorial
Triaminophenolate complexes
complex plane with an offset of appr. 0.8 Å between the
CuON2-planes of each metal centre.49 The structure of 2 is
slightly more distorted, with a larger offset (appr. 1.5 Å)
between the ZnON2-planes.‡
Given the poor performance of the diiminopyrrolide complex,
we decided to explore if stereocontrol via site-mediated chain-
end control can be achieved with a ligand better suited for zinc.
In 2003, Williams, Hillmyer, Tolman and coworkers reported a
The 1H-NMR spectrum agrees with the unsymmetrical
coordination of the ligand observed in the crystal structure:
two sets of chemical shifts are obtained for the methyl-
benzylimino-substituent and the protons of pyrrole and of the
methylene group are diastereotopic (Fig. S15†).
tetra-coordinated, monomeric Zn(II) catalyst carrying
a
diaminophenolate ligand (3, Scheme 4).76 3 is among the most
active zinc-based catalysts reported, reaching full conversion
in only 5 min at room temperature with good polymer
molecular weight control and low polydispersities. The PLA
produced was atactic, however. The high activity was attributed
to reversible coordination of the pendant dimethylamino
group. Polyamino-phenolate based ligands have been in
the following well studied in zinc-catalysed lactide
polymerisation.59,62,69–72,77–83 Several strategies have been
employed to add isotactic stereocontrol to this catalyst system.
Mehrkhodavandi introduced chirality into the side arm and
replaced the ethylene bridge with a chiral cyclohexylene bridge
(4, Scheme 4).84 The more rigid bridge, however, drastically
reduced activity (full conversion in about 40 h), while the
resulting PLA remained atactic. Removing the methyl group on
the central nitrogen, increased activity, but did not improve on
stereocontrol.81 Ma added an aniline donor arm to the central
amine donor (5, Scheme 4).69 Stereocontrol correlated with the
coordination environment of the zinc centre, with aniline-
coordinated complexes providing heterotactic and amine-
coordinated complexes providing isotactic PLA. The same
group also successfully explored replacing the dimethylamino
substituent with chiral or non-chiral cyclic amines to provide
isotactic PLA.71,72 The results are mechanistically complex and
involve catalytic-site control and chain-end control active at
the same time. More recently, Kol added a pyridylmethyl
donor on the terminal amino group to form a tetradentate
ligand with either achiral or chiral spacers (6, Scheme 4). The
rac-Lactide polymerisation. Despite the strong structural
resemblance between 1 and 2, polymerisation results with 2
were unsatisfactory and differed strongly from those of 1. At
room temperature and in C6D6 solution, complex 1 showed
slow initiation (t0 = 11 min), followed by pseudo-first order
kinetics to produce moderately isotactic PLA after appr. 5 h
(kobs = 0.6 0.1 h−1 at 2 mM catalyst concentration). Complex
2 likewise followed a pseudo-first order regime, but did not
show an induction period. Instead, an apparent negative x-axis
intercept of the regression curve at t0 = −45 min indicated a
higher rate of monomer consumption in the very first minutes
of the polymerization. After this initial fast reaction, the reac-
tion required 48 h to reach completion (kobs = 0.15(1) h−1 at
2 mM catalyst concentration, Fig. 2) and produced moderately
heterotactic PLA (Pr = 0.75). Polymer weight control is very
poor, with appr. 5 chains produced per catalyst dimer and a
high polydispersity of 2.8. The MALDI spectrum of the
polymer shows the presence of cyclic oligomers from intra-
Fig. 2 Conversion-time profile for rac-lactide polymerisation with 2.
The inset shows the semi-logarithmic plot. Solid lines represent theore-
tical curves based on linear regression of the linear region in the semi-
logarithmic plot. Conditions: [2] = 2.0 mM, [lactide] = 0.20 M, C6D6, RT.
Final conversion : 98% after 48 h, Mn(GPC) = 3.0 kg mol−1, Mw/Mn = 2.8.
‡As an alternative description of the two differing geometries, we can define the
equatorial complex plane as containing M, O and Npyrrol. The pyridylmethoxide
ligands have an angle of 46°–48° with this plane (cf. 26°–30° in 1) and the imino-
pyrrolide ligands of 53°–62° (cf. 69°–82° in 1).
Scheme 4
This journal is © The Royal Society of Chemistry 2018
Dalton Trans., 2018, 47, 16279–16291 | 16281

View Article Online
Paper
Dalton Transactions
complexes were highly active in polymerisation and provided amino-capped side arms exchange readily on the NMR time
isotactic PLA.70,73 All these approaches relied on stabilizing a scale and only one singlet is observed for all NMe2 groups.§
specific environment of the catalytic site. We decided to inves- More importantly, the ArCH2N group appears as a singlet with
tigate whether deliberate introduction of flexibility into 3, i.e. an intensity of 2 at 3.39 ppm. The side arm exchange is thus
providing a stereochemically unstable catalytic site, would correlated with an inversion at the zinc centre and the central
allow isotactic stereocontrol via a catalytic-site mediated chain- nitrogen, i.e. an racemisation of the complex, which renders
end control mechanism. Complex 7, containing identical these two protons homotopic (Scheme 5). It should be noted
aminoethyl substituents should preserve the high activity of 3, that complex 3b shows two distinct singlets for the NMe2
while exchange of the coordinating arms would invert chirality group and a pair of doublets for ArCH2N in its NMR spec-
at the zinc and the central nitrogen atom simultaneously, and trum.76 Mehrkhodavandi showed that addition of pyridine led
would allow facile racemisation of the complex (Scheme 4).
to coalescence of the NMe2 signals, but the ArCH2N protons
Syntheses and structures. L2H was prepared by reductive remained diastereotopic.84 Racemisation thus does not occur
amination of 2-formyl-4,6-tert-butylphenol in the presence of in 3b – even in the presence of Lewis bases – since it would
NaBH3(CN) and acetic acid in methanol, adapting protocols require dissociation of both amine ligands at the same time.
for similar ligands.78 It has been previously prepared by reac- As envisioned, the presence of an additional donor arm in 8
tion of sodium dimethyl amide with the respective chloride facilitates epimerisation at the metal centre, a prerequisite for
substituted precursor.85 Reaction of zinc diethyl with one the targeted stereocontrol mechanism.
equiv. of L2H produced a yellow oil from which colourless crys-
Williams et al. reported facile transformation of ethyl
tals of (L2)ZnEt, 8, could be obtained (Scheme 5). The X-ray complex 3b into the desired alkoxide complex 3 with ethanol.
structure of 8 shows a chiral, tetrahedral Zn(II) complex in Unfortunately, the ethyl group in 8 was unreactive toward alco-
which only one of the dimethylamino arms is coordinated to holysis and, even after heating, NMR spectra confirmed the
the metal (Fig. 3). As hypothesised, the additional donor arm presence of unreacted 8 and ethanol. Similar problems were
does not influence complex geometry, and 8 is essentially iso- encountered in the reaction of 4 with ethanol,84 and we and
structural to 3b (Scheme 4, Table 2).76
others have noticed previously that alcoholysis of the second
The 1H NMR spectrum of 8 displays a higher apparent sym- zinc ethyl bond can be challenging.58,86 In fact, the major
metry than its crystal structure (Fig. S20†). The two dimethyl- product of ZnEt2 in isopropanol is EtZnOiPr.87 Zinc amide
complexes are a commonly employed alternative pathway to
prepare heteroleptic zinc alkoxides.56 Complex 9 was thus pre-
pared by addition of 1 equiv. Zn(N(SiMe3)2)2 to a toluene solu-
tion of L2H, and formed colourless crystals after recrystallisa-
tion from hexane (Scheme 6).
The X-ray structure of 9 shows the same distorted tetra-
hedral coordination with one uncoordinated amine ligand as
in 8 (Fig. 4, Table 2). The 1H NMR spectrum of 9 likewise
showed a single singlet for the aryl methylene group and one
Scheme 5
singlet for all dimethylamino groups, in agreement with fast
exchange of coordinated and uncoordinated dimethylamine
ligand, coupled with an epimerisation of the metal centre
(Fig. S22†). Only one signal is observed for the trimethylsilyl
substituents. Zn–Namide rotation is thus fast on the NMR time
scale, but this process is not connected with complex
racemisation.
The desired catalyst 7, containing an alkoxide as a suitable
initiator for rac-lactide polymerisation, was prepared by the
addition of 1 equiv. of ethanol to a C6D6 solution of 9
(Scheme 6). The reaction was followed by 1H NMR to ensure
clean conversion to the alkoxide complex. The reaction was
complete before the first NMR spectrum was taken (<10 min,
Fig. S8†). As for the amide complexes, only one singlet is
observed for the dimethylamine groups and the arylmethylene
group, respectively, indicating fast epimerisation at the metal
Fig. 3 X-ray structure of 8. Thermal ellipsoids are drawn at the 50%
probability level. Hydrogen atoms and minor fractions of disorder in
tert-butyl and dimethylaminoethylene substituents omitted for clarity.
§The same spectrum would be expected for a Cs-symmetric penta-coordinated
complex, but it is highly unlikely to observe a higher coordination in solution
than in the solid state.
16282 | Dalton Trans., 2018, 47, 16279–16291
This journal is © The Royal Society of Chemistry 2018

View Article Online
Dalton Transactions
Paper
Table 2 Geometrical details for the X-ray structures of 8–11
8
3ba
9
11
Zn–O
Zn–Nterminal
Zn–NR3
Zn–C/Namide
O–Zn–C/Namide
O–Zn–Nterminal
1.959(2)
2.159(3)
2.171(3)
1.977(4)
126.5(1)
101.6(1)
1.956(3)
2.128(3)
2.147(4)
1.998(4)
128.8(1)
99.8(1)
1.949(2), 1.941(2)
2.154(3), 2.158(3)
2.118(2), 2.124(2)
1.927(2), 1.933(3)
115.6(1), 116.0(1)
103.4(1), 103.3(1)
1.972(1)
2.210(1), 2.223(6)
2.306(1)
1.974(1)
120.45(1)
b
b
95.49(4), 99.7(2)
a Taken from ref. 76. b Nterminal: NMe2 (3b, 8, 9), pyridine (11).
be partly attributed to the difference in solvent (C6D6 here,
CH2Cl2 for 3) and partly to the presence of two diamino
groups in 7, which make dissociation of a diamino ligand,
speculated to be required for lactide coordination,76 statisti-
cally less likely. Unfortunately, 7 shows relatively poor polymer
molecular weight control with polydispersities of 2.8 and a
lower than expected polymer molecular weight. The MALDI
spectrum of the polymer confirmed the presence of intra-
molecular transesterification reactions (Fig. S7 and S8†). The
latter could be improved under immortal polymerisation con-
ditions: in presence of 4 equiv. EtOH, 7 produced PLA with the
expected molecular weight and lower polydispersities. Activity
in immortal polymerisation was only half as high (Table 3,
Fig. S2 and S3†), which is surprising since the complex should
not be sensitive towards alcohol. PLA produced with 7 showed
a very slight isotactic bias of Pm = 0.55. Using our standardised
integration protocol (see Experimental part), Pm values are
typically consistent to 1% through-out a kinetic experiment
and to 3% in repeated experiments. The small amount of iso-
tacticity observed is thus outside of typically experimental
error, but might nevertheless be influenced by transesterifica-
tion.58 Pm values did, however, not show any variation with
conversion or time and are thus not due to transesterification
reactions (Fig. 6).
Scheme 6
Dipyridylaminophenolate complexes
Syntheses and structures. To compare the influence of cata-
lytic site racemisation on stereocontrol, rac-lactide polymeris-
ation was investigated with 10, in which the dimethyl amino
donors were replaced with pyridyl (Scheme 7). Complexes
similar to 10 have been briefly investigated by Thomas and
Carpentier and produced atactic PLA.78 10 was prepared analo-
Fig. 4 X-ray structure of 9. Thermal ellipsoids are drawn at the 50%
probability level. Hydrogen atoms and the second, independent mole-
cule in the asymmetric unit omitted for clarity.
centre. After 1H NMR confirmed full conversion, C6D6 solu- gous to 7: reaction of Zn(N(SiMe3)2)2 and L3H provided the
tions of 7 were used as stock solutions in polymerisations amide complex 11. Further reaction with ethanol in C6D6
experiments.
afforded 10, which was directly used in polymerisation
rac-Lactide polymerisation. Complex 7 readily polymerized (Scheme 7). In contrast to 8 and 9, in the crystal structure of
lactide in C6D6 solution at room temperature and reached full 11 both pyridine ligands were coordinated to the metal centre
conversion after appr. 30 min at 2 mM catalyst concentration (Fig. 7). A τ-value of 0.1 would indicate square-pyramidal geo-
(Table 3). The catalyst follows clean first-order kinetics, metry, but closer inspection of the structure and the respective
without notable induction period or complex decomposition metal–ligand bond lengths propose distorted bipyramidal geo-
(Fig. 5). The pseudo-first-order rate constant at 2 mM catalyst metry as a better description (Table 2). The structure of 11
concentration is kobs = 4.1(1) h−1 (Fig. 5). Complex 7 was thus does not show mirror-symmetry, despite the coordination of
able to retain the high activity of complex 3. The slightly lower both pyridine ligands, since the geometry of the aryl methyl-
rate when compared to 3 (kobs(3) = 15 h−1 at [3] = 2 mM)76 can ene group forces a bending of the aryl group out of the ZnN2-
This journal is © The Royal Society of Chemistry 2018
Dalton Trans., 2018, 47, 16279–16291 | 16283

View Article Online
Paper
Dalton Transactions
Table 3 rac-Lactide polymerisation with 7 and 10a
b
e
Catalyst
[cat.]
2.0 mM 200 mM 1 : 100
[Lactide] Zn : lactide Final conversion (time) kobs
Mn
Mn c (calc.) Mw/Mn Chains/Znd Pm
1
2
3
4
5
6
7
8
9
7
99% (420 min)
99% (100 min)
67% (100 min)
97% (2 min)
99% (2 min)
93% (20 min)
96% (13 min)
60% (24 h)
4.1(1) h−1
2.5(1) h−1
6.0 kDa
2.7 kDa
14.3 kDa
2.9 kDa
9.6 kDa
14.0 kDa
14.3 kDa
2.8
1.2
1.1
1.9
1.1
1.9
1.7
1.2
1.2
2.4
5
1
5.8
1.5
0.4
1.7
2.7
1.2
0.55
0.55
0.55
0.49
0.50
0.55
0.55
0.55
0.60
7 + 4 EtOH 2.0 mM 200 mM 1 : 100
7
10
0.5 mM 50 mM
1 : 100
0.73(1) h−1 9.1 kDa
2.4 kDa
2.0 mM 200 mM 1 : 100
120(3) h−1 9.6 kDa
10
10
10
10
0.5 mM 50 mM
1 : 100
6.9(2) h−1
33.8 kDa 13.4 kDa
12.2(6) h−1 32.8 kDa 55.3 kDa
0.9(1) h−1
36.5 kDa 86.4 kDa
0.25(4) h−1 23.2 kDa 27.4 kDa
0.5 mM 200 mM 1 : 400
0.5 mM 500 mM 1 : 1000
0.3 mM 150 mM 1 : 500
40% (16 h)
a Conditions: C6D6, RT. b Mn and Mw determined by size exclusion chromatography vs. polystyrene standards, with a Mark–Houwink correction
factor of 0.58. c Calculated from [lactide]/([cat] + [EtOH])·conversion·Mlactide + MROH d Number of chains per zinc centre, calculated from the ratio
.
of expected and obtained polymer molecular weight. e Pm determined from decoupled 1H NMR by Pm = 1 − 2·I1/(I1 + I2), with I1 = 5.20–5.25 ppm
(rmr, mmr/rmm), I2 = 5.13–5.20 ppm (mmr/rmm, mmm, mrm).
Fig. 5 Conversion-time profiles for rac-lactide polymerisation with 7.
Conditions: C6D6, RT, 7 : lactide = 1 : 100. The inset shows the semi-log-
arithmic plot. Solid lines correspond to theoretical conversions based on
rate constants obtained from linear regression: black triangles: [7] =
2.0 mM, kobs = 4.1(1) h−1, t0 = −5 min, final conversion after >7 h: 99%;
blue diamonds: [7] = 0.5 mM, kobs = 0.73(1) h−1, t0 = −5 min, final con-
version after >7 h: 67%. The negative axis intercept might indicate partial
catalyst decomposition in the first 5 min of the reaction, inhomo-
geneous starting conditions or – although unlikely – experimental error.
Fig. 6 Variation of polymer microstructure (Pm) in dependence of con-
version or time in rac-lactide polymerisations with 7. Black triangles:
[7] = 2.0 mM, blue diamonds: [7] = 0.5 mM. With [7] = 0.5 mM, conver-
sion plateaued at 67%.
plane toward one of the pyridine ligands. The latter shows a
bending of the Zn–NPyridine bond out of the mean plane of the
pyridine to allow closer contact with the aryl group, indicating
favourable π-interactions between the two aromatic systems
(angle between planes = 34°, shortest atom-plane contact =
3.0 Å).
1
The H NMR spectrum of 11 likewise shows indications of
Scheme 7
stronger interaction of pyridine with the zinc centre. While 11
1
still displays apparent Cs-symmetry in its H spectrum, peaks
are broadened, indicative of a lower rate of exchange: the
PyCH2 groups appear as one broadened and one sharp doublet complex and undergoes slow racemisation (Scheme 8, A), or it
and the ArCH2 group is broadened to a large peak between interchanges between a tetrahedral and a five-coordinated
3.8–4.9 ppm (Fig. S26†). Only one set of pyridine signals are species by dissociation/recoordination of a pyridine ligand (B),
observed, significantly broadened and coupling is barely or the complex remains five-coordinated, with a slow “flip-
visible. 11 thus either exists in solution as a tetrahedral ping” of the phenolate ring (C). All these dynamic processes
16284 | Dalton Trans., 2018, 47, 16279–16291
This journal is © The Royal Society of Chemistry 2018

View Article Online
Dalton Transactions
Paper
Fig. 7 X-ray structure of 11. Thermal ellipsoids are drawn at the 50%
probability level. Hydrogen atoms, and the second, independent mole-
cule in the asymmetric unit omitted for clarity.
Fig. 8 Variable temperature NMR spectra of 11 showing the coalesc-
ence of the ortho hydrogen atoms of the two pyridyl units (see ESI† for
further spectra).
1.8 ppm. These observations agree best with the species at low
temperature being the five-coordinated Zn complex, observed
in the crystal structure. The kinetic process observed in the
NMR spectra is thus most likely the “flipping” of the phenolate
ring (Scheme 8, C).
1
The H NMR spectra of the respective ethoxide complex 10
shows essentially the same features as 11: only one set of
peaks for the pyridyl ligands in agreement with apparent Cs
symmetry, but with overall broadened peaks (Fig. S11†).
Regardless of the exact nature of the dynamic process, racemi-
sation – if it happens at all in 10 and 11 – is more difficult
with pyridyl donors than with dimethylamine ligands.
Scheme 8
rac-Lactide polymerisation. Complex 10 was highly active in
generate apparent Cs-symmetry and equalize the two pyridyl lactide polymerisation (Table 3). Polymerisation was com-
moieties. Derivatives of 11, previously reported by Thomas and pleted in 2 min at ambient temperature at 2 mM catalyst con-
Carpentier,78 likewise showed apparent Cs-symmetry in their centration and in appr. 15 min at 0.5 mM, placing 10 among
NMR spectra. At higher temperatures, peaks of 11 sharpen to the most active zinc-based catalysts (cf. estimated kobs at 2 mM
yield the apparently Cs-symmetric spectrum. Below 250 K, the [Zn]: 3,76 15 h−1; 5,69 10 h−1; 6,73 50 h−1; 7, 4 h−1; 10, 120 h−1).
spectrum shows one species with independent signal sets for No induction period was observed, but regression curves of
each of the pyridylmethylene arms (Fig. 8 and S32–S35†). The kinetics at 0.5 mM all showed a negative x-axis intercept,
trimethylsilyl signal likewise splits into two peaks at lower indicative of catalyst deactivation at the beginning of the reac-
temperatures. However, while activation barriers estimated tion (Fig. S4 and S5†). The same was observed for 7, but only
from the coalescence temperatures of the aromatic and to an extent still explicable by experimental error. While
methylene protons fall in the range of ΔG‡ = 53–57 kJ mol−1
,
decomposition did not visually affect polymerisations at
an activation barrier of ΔG‡ ≈ 46 kJ mol−1 is obtained for the catalyst : lactide ratios of 1 : 100 or 1 : 400, polymerisations at a
trimethylsilyl signals. The latter is thus an unrelated kinetic ratio of 1 : 1000 or at catalyst concentrations of 0.3 mM showed
process, probably involving rotation around the Zn–N bond. curved semi-logarithmic plots and did not reach completion,
Upon irradiation of the pyridyl ortho hydrogen atoms at 9.1 indicative of catalyst decomposition before the end of the reac-
and 8.8 ppm, respectively, an NOE enhancement of the tri- tion (Fig. S4 and S5†). At 0.1 mM catalyst concentration, we
methylsilyl signal was observed for both hydrogen atoms. did not observe more than 10% conversion. Polymerisation
However, only irradiation of the hydrogen atom at 8.8 ppm with 7 at reduced catalyst concentrations (0.5 mM) also failed
showed an NOE enhancement of the tert-butyl group at to reach completion (Fig. 5, Table 3). Polydispersities of PLA
This journal is © The Royal Society of Chemistry 2018
Dalton Trans., 2018, 47, 16279–16291 | 16285

View Article Online
Paper
Dalton Transactions
produced with 10 varied between 1.1 to 1.9, but in most cases slight isotactic bias in one of the most active zinc-based cata-
the polymer molecular weight was smaller than expected. lysts. While the low isotacticity and poor polymer molecular
MALDI spectra of the polymer showed the presence of cyclic weight control do not encourage further optimisation of this
oligomers, indicative again of intramolecular transesterifica- ligand system in particular, these results underline that lactide
tion (Fig. S10–S14†).
While PLA obtained with 7 showed slight isotacticity (Pm
0.55, Table 3), PLA produced with 10 was atactic (Pm
polymerisation often defies the axiom that successful control
requires a rigid environment of the catalytic site and that cata-
lytic sites with flexible conformation or even flexible configur-
=
=
0.49–0.50, Table 3). This would be in line with a higher ten- ation might offer an alternative approach to achieve
dency of pyridine to coordinate to zinc and the proposed cata- stereocontrol.
lytic-site mediated chain-end control: if the active species is a
penta-coordinated zinc complex, the catalytic site is achiral
and cannot transfer the chirality of the chain-end to zinc. In
diiminopyrrolide copper complexes of type 1, the only complex
ever observed to coordinate both iminogroups to form an
Experimental
General considerations
achiral catalytic site (1b, Scheme 2) was also the only complex All reactions were carried out using Schlenk or glove box tech-
of type 1 producing atactic PLA.54,55 Alternatively, the catalytic niques under nitrogen atmosphere. Zn(N(SiMe3)2)2,88 2,4-di-tert-
site might be slow to racemise and to adapt to the chain-end, butylsalicyladehyde,89 N,N,N,N-tetramethyldiethylenetriamine,90
also resulting in loss of the ability to transfer chain-end chiral- and L1H,53 were prepared according to literature. Solvents were
ity to the monomer. In the latter case, stereocontrol can be dried by passage through activated aluminum oxide (MBraun
influenced by reaction conditions, since insertion is depen- SPS), de-oxygenated by repeated extraction with nitrogen, and
dent on monomer concentration, while – in first approxi- stored over molecular sieves. C6D6 was dried over molecular
mation – catalyst racemisation is not. At lower monomer con- sieves. rac-Lactide (98%) was purchased from Sigma–Aldrich,
centrations, the ratio of racemisation vs. insertion rate will purified by 3× recrystallisation from dry ethyl acetate and kept
thus be higher. If lactide concentration was reduced from 200 at −30 °C. All other chemicals were purchased from common
to 50 mM, while keeping the lactide : catalyst ratio constant, commercial suppliers and used without further purification. 1H
stereocontrol indeed increased to Pm = 0.55 (Table 3). Closer and 13C NMR spectra were acquired on Bruker Advance 300 and
investigation revealed, however, that this effect was not due to 400 spectrometers. Chemical shifts were referenced to the
reduced lactide concentration, but rather due to reduced cata- residual signals of the deuterated solvents (CDCl3: 1H:
lyst concentrations: increasing the lactide concentration to 200 δ 7.26 ppm, 13C: δ 77.16; C6D6: 1H: δ 7.16 ppm, 13C:
or even to 500 mM, while keeping the catalyst concentration at δ 128.06 ppm). Elemental analyses were performed by the
0.5 mM, did not affect stereocontrol, which remained constant Laboratoire d’analyse élémentaire (Université de Montréal). All
at Pm = 0.55 (Table 3). On the other hand, lowering the catalyst UV-Vis measurements were performed in degassed and anhy-
concentration to 0.3 mM increased stereocontrol further to drous toluene at RT in a sealed quartz cell on a Cary 500i
Pm = 0.60. For 7, lowering of the catalyst concentration did not UV-Vis-NIR Spectrophotometer.
affect stereocontrol (Table 3). There are several mechanistic
2,4-Di-tert-butyl-6-((N,N,N′,N′-tetramethyldiethylenetriamine))
explications for a negative impact of catalyst concentration on phenol, L2H.85 A procedure from literature was adapted as
stereocontrol, such as chain-exchange between centres in a follows:78 To a brown mixture of 3,5-di-tert-butyl-2-hydroxy-
catalytic-site control mechanism or the formation of dinuclear benzaldehyde (0.50 g, 2.1 mmol), NaBH3(CN) (0.16 g,
species with different reactivities. Given the overall low isose- 2.5 mmol) and a few drops of acetic acid in methanol (10 ml),
lectivity of 7 and 10, in particular when compared to 5 and 6, was added
a
solution of N,N,N,N-tetramethyldiethyl-
and the mediocre polymer molecular weight control, we did enetriamine (0.67 g, 4.2 mmol) in methanol (10 ml) dropwise.
not investigate this issue further.
The reaction was stirred for 48 h. The methanol was evapor-
ated and the resulting brown residue was purified by silica gel
chromatography (2% MeOH, 1% NEt3 in CHCl3) yielding a
light yellow oil (0.62 g, 78%).
Conclusions
1H NMR (CDCl3, 300 MHz): δ 7.18 (d, J = 3 Hz, 1H, Ar), 6.82
The application of the ligand system which provided isotactic (d, J = 3 Hz, 1H, Ar), 3.74 (s, 2H, ArCH2N), 2.68–2.63 (m, 2H,
copper-based polymerisation catalysts to zinc, afforded a CH2), 2.47–2.43 (m, 2H, CH2), 2.19 (s, 12H, N(CH3)2), 1.40 (s,
complex of surprisingly similar structure, but strongly 9H, CH3), 1.27 (s, 9H, CH3); 13C{1H} NMR (CDCl3, 75 MHz):
different polymerisation reactivity. Metals with coordination δ 154.3 (Ar), 140.4 (Ar), 129.8 (Ar), 123.9 (Ar), 122.9 (Ar), 121.9
geometries closer to copper might show more similar reactiv- (Ar), 59.0 (CH2), 57.1 (CH2), 51.9 (CH2), 45.8 (N(CH3)2), 35.0
ity, but our attempts to prepare the respective iron, cobalt or (C(CH3)3), 34.2 (C(CH3)3), 31.8 (CH3), 29.7 (CH3). ESI-HRMS
manganese complexes have not been successful so far.
For aminophenolate-based complexes, the counterintuitive
(m/z): [M + H]+ (C23H44N3O) calcd 378.3478; found 378.3485.
2,4-Di-tert-butyl-6-((di-(2-picolyl)amine)phenol, L3H. A pro-
approach to provide additional flexibility to the catalytic site cedure from literature was slightly modified as follows:78
and enable fast racemisation was successful in introducing a Analogous to L2H, from 3,5-di-tert-butyl-2-hydroxybenzalde-
16286 | Dalton Trans., 2018, 47, 16279–16291
This journal is © The Royal Society of Chemistry 2018

View Article Online
Dalton Transactions
Paper
hyde (0.50 g, 2.1 mmol), NaBH3(CN) (0.16 g, 2.5 mmol), a few lysis could not be obtained. The NMR likewise indicates the
drops of acetic acid, di-(2-picoly)amine (0.84 g, 4.2 mmol) in presence of impurities.
methanol (20 ml) stirred for 4 hours to yield a brown residue
which was purified by silica gel chromatography (2% MeOH, C6D6 (0.6 ml) in a J-Young tube was added a freshly prepared
1% NEt3 in CHCl3) (0.34 g, 39%). solution of EtOH in C6D6 (0.20 M) in two portions of appr.
(L2)ZnOEt, 7. To a solution of 9 (10 mg, 16 µmol) in
1H NMR (CDCl3, 300 MHz): δ 10.62 (s, 1H, OH), 8.56 (ddd, 40 µL. The reaction was followed by NMR and the amount of
J = 6, 2, 1 Hz, 2H, Py), 7.63 (td, J = 8, 2 Hz, 2H, Py), 7.37 (d, J = the second batch of ethanol was adjusted with regard to
8 Hz, 2H, Py), 7.20 (d, J = 3 Hz, 1H, Ar), 7.15 (ddd, J = 8, 6, remaining 9. After NMR confirmed complete replacement of
1 Hz, 2H, Py), 6.87 (d, J = 3 Hz, 1H, Ar), 3.87 (s, 4H, PyCH2N), the amide by ethoxide, the solution used directly in
3.80 (s, 2H, ArCH2N), 1.45 (s, 9H, CH3), 1.26 (s, 9H, CH3); 13C polymerization.
{1H} NMR (CDCl3, 75 MHz): δ 158.3 (Py), 154.0 (Ar), 149.2 (Py),
1H NMR (C6D6, 400 MHz,): δ 7.64 (d, J = 3 Hz, 1H, Ar), 6.94
140.5 (Ar), 136.7 (Py), 135.7 (Ar), 124.7 (Ar), 123.7 (Py), (d, J = 3 Hz, 1H, Ar), 3.25 (s, 2H, ArCH2), 2.11 (s, 12 H, NMe2),
123.3 (Ar), 122.4 (Py), 121.8 (Ar), 59.7 (CH2), 58.4 (CH2), 2.01–2.08 (m, 2H, CH2), 1.96 (ddd, J = 13, 9, 4 Hz, 2H, CH2),
35.1 (C(CH3)3), 34.2 (C(CH3)3), 31.8 (CH3), 29.8 (CH3). 1.90 (s, 9H, CMe3), 1.73 (m, 2H, CH2), 1.45–1.60 (m, 7H, CH3 +
ESI-HRMS (m/z): [M + H]+ (C27H36N3O) calcd 418.2852; found 2 CH2), 1.50 (s, 9H, CMe3); 13C{1H} NMR (C6D6, 101 MHz):
418.2857.
δ 165.4 (Ar), 138.5 (Ar), 134.3 (Ar), 125.5 (Ar), 124.2 (Ar), 121.9
(L1)2Zn2(μ-O,κN-OCH2C6H2N)2, 2. Zn(N(SiMe3)2)2 (115 mg, (Ar), 59.9 (CH2Ar), 55.9 (CH2), 51.2 (CH2), 46.3 (NMe2), 45.5
0.304 mmol) was suspended in toluene (3 ml). (ZnOCH2), 35.9 (CMe3), 34.2 (CMe3), 32.4 (CMe3), 32.1
2-Pyridinemethanol (28.3 µl, 0.304 mmol) was added to the (ZnOCH2Me), 30.4 (CMe3).
red suspension, which was left to stir for 45 min. A freshly pre-
(L2)ZnEt, 8. ZnEt2 (33 mg, 0.26 mmol) was added to a
pared light orange solution of L1H (100 mg, 0.304 mmol) in freshly prepared light yellow solution of L2H (100 mg,
toluene (2 ml) was added dropwise, resulting in a light yellow 0.26 mmol) in toluene (5 ml), resulting in an orange solution.
solution. The reaction was stirred 24 hours at RT, filtered to The reaction was stirred 24 hours at RT, and filtered to remove
remove trace impurities, concentrated to 1/3 of the volume trace impurities. The solvent was removed under vacuum and
resulting in colourless crystals. The crystals were separated by the resulting orange oil was crystallised from hexane (10 ml) at
decantation and washed with ether (3 × 10 ml) (23 mg, 16%).
1H NMR (C6D6, 400 MHz): δ 8.18 (s, 1H, (NvC)H), 7.80 (s, and washed with hexane (3 × 10 ml) (45 mg, 38%).
1H, (NvC)H), 7.35 (s, 1H), 7.29 (bs, 2H, Ph), 7.04–6.95 (m,
1H NMR (C6D6, 400 MHz): δ 7.63 (d, J = 3 Hz, 1H, Ar), 6.97
−30 °C. The colourless crystals were separated by decantation
3H), 6.95–6.83 (m, 5H), 6.76 (s, 1H), 6.65 (t, J = 8 Hz, 1H), 6.15 (d, J = 3 Hz, 1H, Ar), 4.34 (bs, 2H, NCH2), 3.39 (s, 2H, ArCH2N),
(d, J = 8 Hz, 1H, pyrrole), 4.64 (d, J = 18 Hz, 1H, PyCH2), 4.56 2.50–2.40 (m, 2H, NCH2), 2.37–2.21 (m, 4H, NCH2), 1.89 (s,
(d, J = 18 Hz, 1H, PyCH2), 4.24 (q, J = 7 Hz, 1H, CH), 4.10 (q, J = 21H, C(CH3)3 + N(CH3)2), 1.68 (t, J = 8 Hz, 3H, ZnCH2CH3),
7 Hz, 1H, CH), 1.50 (d, J = 7 Hz, 3H, CH3), 1.20 (d, J = 7 Hz, 1.49 (s, 9H, C(CH3)3), 0.45 (q, J = 8 Hz, 2H, ZnCH2); 13C{1H}
3H, CH3); 13C{1H} NMR (C6D6, 101 MHz): δ 165.2 ((NvC)H), NMR (C6D6, 101 MHz): 155.2 (Ar), 133.6 (Ar), 131.5 (Ar), 124.4
157.0 (Ar), 154.8 ((NvC)H), 146.7 (Ar), 145.9 (Ar), 144.6 (Ar), (Ar), 123.0 (Ar), 58.3 (CH2), 57.2 (CH2), 51.4 (CH2), 45.5
141.6 (Ar), 140.2 (Ar), 137.4 (Ar), 129.3 (Ar), 128.5 (Ar), 127.4 (N(CH3)2), 35.5 (CMe3), 34.4 (CMe3), 32.1 (CMe3), 30.11 (CMe3),
(Ar), 127.0 (Ar), 126.9 (Ar), 126.7 (Ar), 122.0 (Ar), 120.9 (Ar), 19.2 (CH2Me), 1.4 (ZnCH2). UV-vis (toluene, 1.2 × 10−4 M)
116.3 (Ar), 115.7 (Ar), 68.4 (CH2), 67.7 (CH), 64.2 (CH), 24.7 [λmax
,
nm (ε, mol−1 cm2)]: 302 (2700). Anal. calcd for
(CH3), 24.6 (CH3). UV-vis (toluene, 2.2 × 10−6 M) [λmax, nm (ε, C25H47ZnN3O: C, 63.75; H, 10.06; N, 8.92; found: C, 63.63; H,
mol−1 cm2)]: 378 (23 500), 387 (23 800). Anal. calcd for 10.25; N, 8.84.
C56H56Zn2N8O2: C, 67.00; H, 5.62; N, 11.16; found: C, 66.34; H,
5.62; N, 10.65.
(L2)ZnN(SiMe3)2, 9. Analogous to 8, from Zn(N(SiMe3)2)2
(102 mg, 0.265 mmol), L2H (100 mg, 0.265 mmol) in toluene
(L1)2Zn. Zn(HMDS)2 (115 mg, 0.304 mmol) was suspended (5 ml). Filtration, removal of the solvent under vacuum, crystal-
in toluene (3 ml). A freshly prepared light orange solution of lisation in hexane (10 ml) at −30 °C, decantation and washing
L1H (200 mg, 0.608 mmol) in toluene (2 ml) was added drop- with hexane (3 × 10 ml) afforded 52 mg (32%) of colourless
wise, resulting in a light yellow solution. The reaction was X-ray quality crystals.
stirred 24 hours at RT, filtered to remove trace impurities, con-
1H NMR (C6D6, 400 MHz): δ 7.62 (d, J = 3 Hz, 1H, Ar), 6.87
centrated to 1/3 of the volume resulting in colourless crystals. (d, J = 3 Hz, 1H, Ar), 3.60 (s, 2H, ArCH2N), 2.45 (dt, J = 11,
The crystals were separated by decantation and washed with 6 Hz, 2H, CH2), 2.27 (td, J = 11, 6 Hz, 4H, CH2), 2.08–1.98 (br
ether (3 × 10 ml) and recrystallised two times from a mixture m, 2H, CH2), 1.93 (s, 12H, N(CH3)2), 1.78 (s, 9H, CH3), 1.47 (s,
of toluene : hexane (1 : 3) (53 mg, 24%).
9H, CH3), 0.49 (s, 18H, Si(CH3)3); 13C{1H} NMR (C6D6,
1H NMR (C6D6, 400 MHz): δ 7.80 (s, 2H, (NvC)H), 101 MHz): δ 164.4 (Ar), 138.1 (Ar), 134.9 (Ar), 126.3 (Ar), 124.9
7.04–6.98 (m, 6H, Ph), 6.93–6.86 (m, 4H, Ph), 6.76 (s, 2H, 3,4- (Ar), 120.7 (Ar), 62.1 (CH2Ar), 56.6 (CH2), 53.5 (CH2), 46.7
Pyrrole), 4.10 (q, J = 7 Hz, 2H, CH), 1.20 (d, J = 7 Hz, 6H, CH3). (N(CH3)2), 35.9 (CMe3), 34.2(CMe3), 32.4 (CMe3), 30.6 (CMe3),
UV-vis (toluene, 2.2 × 10−6 M) [λmax, nm (ε, mol−1 cm2)]: 376 7.2 (SiMe3). UV-vis (toluene, 7.6 × 10−5 M) [λmax, nm (ε, mol−1
(20 600), 389 (22 000), 433 (sh). Despite X-ray quality crystals cm2)]: 302 (4000). Anal. calcd for C29H60N4OSi2Zn·1/3C6H14 C,
and repeated re-crystallizations, a satisfactory elemental ana- 59.00; H, 10.33; N, 8.88; found: C, 59.25; H, 10.74; N, 9.16.
This journal is © The Royal Society of Chemistry 2018
Dalton Trans., 2018, 47, 16279–16291 | 16287

View Article Online
Paper
Dalton Transactions
(L3)ZnOEt, 10. Analogous to 7, from a freshly prepared 164.8 (Ar), 154.0 (Py), 149.3 (Py), 137.7 (Py), 137.6 (Ar), 134.0
solution of EtOH in C6D6 (0.20 M) and 11 (10 mg, 16 µmol). (Ar), 125.2 (Ar), 123.9 (Ar), 123.1 (Py), 121.8 (Ar), 121.6 (Py),
After NMR confirmed complete replacement of the amide by 62.3 (CH2Ar), 61.1 (CH2Py), 35.6 (CMe3), 33.9 (CMe3), 32.4
ethoxide, the solution used directly in polymerization.
1H NMR (C6D6, 400 MHz,): δ 9.42 (br s, 2H, Py), 7.60 (s, 1H, [λmax
(CMe3), 30.5 (CMe3), 6.6 (SiMe3). UV-vis (toluene, 5.7 × 10−5 M)
nm (ε, mol−1 cm2)]: 296 (4200). Anal. calcd for
,
Ar), 6.97 (s, 1H, Ar), 6.77 (t, J = 8 Hz, 2H, Py), 6.56 (t, J = 8 Hz, C33H52N4OSi2Zn: C, 61.70; H, 8.16; N, 8.72; anal. calcd for
2H, Py), 6.15 (bs, 2H, Py), 3.41 (bs, 2H, CH2), 3.2–3.7 (br, 2H, C33H52ZnN4OSi2·C6H14: C, 64.30; H, 9.13; N, 7.69; found: C,
CH2), 2.75 (bs, 1H), 2.24 (bs, 1H), 1.91 (s, 9H, CMe3), 1.6–1.8 61.99; H, 8.79; N, 8.40 (X-ray structure shows presence of 1 co-
(bm, 2H), 1.51 (s, 9H, CMe3), 1.4 (bm, 3H); 13C{1H} NMR crystallised hexane, which seems to be lost partly on drying.
(C6D6, 101 MHz): δ 165.7 (Ar), 155.6 (Ar), 151.1 (Ar), 139.4 (Ar), NMR shows the presence of 0.7 hexane, EA analysis samples
138.5 (Ar), 133.7 (Ar), 125.8 (Ar), 124.3 (Ar), 123.3 (Ar), 122.5 best agree with 0.2 hexane).
(Ar), 121.6 (Ar), 59.4 (CH2Ar), 58.6 (ZnOCH2), 57.2 (CH2Py),
36.0 (C(CH3)3), 34.2 (C(CH3)3), 32.5 (CH3), 32.0 (ZnOCH2CH3), amount of rac-lactide was placed into a J.-Young tube together
30.3 (CH3). with C6D6. If required, a stock solution of an additive
rac-Lactide polymerisation. In a glove box, the desired
(L3)ZnN(SiMe3)2, 11. Analogous to 8, from Zn(N(SiMe3)2)2 (EtOH, etc.) was added, followed by a stock solution of the cata-
(92 mg, 0.24 mmol), L3H (100 mg, 0.24 mmol) in toluene lyst (≈ 20 mM in C6D6). The reaction was followed by 1H NMR.
(5 ml). Filtration, removal of the solvent under vacuum, crystal- The reaction was quenched by addition of ≈5 equiv. of a
lisation in hexane at −30 °C, decantation and washing with CDCl3 solution of acetic acid (5 mM). The volatiles were
hexane (3 × 10 ml) afforded 54 mg (31%) of colourless X-ray immediately evaporated and solid polymer samples were
quality crystals, containing 1 equiv. co-crystallised hexane.
stored at −80 °C for further analysis. Conversion was deter-
1
1H NMR (C6D6, 400 MHz, 298 K): δ 8.78 (bs, 2H, Py), 7.12 mined from H NMR by comparison to remaining lactide. Pm
(d, J = 3 Hz, 1H, Ar), 6.79 (bs, 2H, Py), 6.62 (d, J = 3 Hz, 1H, Py), values were determined from homodecoupled 1H NMR spectra
6.50 (bs, 2H, Py), 6.27 (bs, 2H, Py), 3.72 (d, J = 14 Hz, 2H, and calculated from Pm = 1 − 2·I1/(I1 + I2), with I1
=
PyCH2N), 3.45 (d, J = 14 Hz, 2H, PyCH2N), 3.29 (bs, 2H, 5.15–5.21 ppm (rmr, mmr/rmm), I2 = 5.21–5.25 ppm (mmr/rmm,
ArCH2N), 1.68 (s, 9H, CH3), 1.32 (s, 9H CH3), 0.50 (s, 18H, Si mmm, mrm). The integration of the left multiplet and right
1
(CH3)3); H NMR (C7D8, 500 MHz, 208 K): δ 9.09 (d, J = 5 Hz, multiplet (I1 and I2) required only one, very reproducible divid-
1H, ortho Py), 8.76 (d, J = 5 Hz, 1H, ortho Py′), 7.12 (d, J = 3 Hz, ing point of the integration, which was always taken as the
1H, Ar), 6.83 (app. t, J = 7 Hz, 1H, para Py), 6.63 (d, J = 2 Hz, minimum between the two multiplets. Pm-Values obtained
1H, Ar), 6.55 (dt, J = 8, 1 Hz, 1H, para Py′), 6.50 (dd, J = 7, 6 Hz, this way were typically consistent to 1% over the course of
1 H, meta Py), 6.32 (m, 2H, meta Py, Py′), 5.98 (d, J = 8 Hz, 1H, one experiment and 3% between different experiments under
meta Py′), 3.74 (d, J = 11 Hz, 1H, ArCH2), 3.58 (d, J = 16 Hz, identical conditions. Molecular weight analyses were per-
PyCH2), 3.18 (d, J = 15 Hz, 1H, PyCH2), 3.12 (d, J = 15 Hz, 1H, formed on crude reaction products using a Waters 1525 gel per-
PyCH2), 3.09 (d, J = 16 Hz, 1H, PyCH2), 2.26 (d, J = 11 Hz, 1H, meation chromatograph equipped with three Phenomenex
ArCH2), 1.75 (s, 9H, tBu), 1.40 (s, 9H, tBu), 1.06 (bs, 9H, columns and a refractive index detector at 35 °C. THF was
SiMe3), 0.09 (bs, 9H, SiMe3); 13C{1H} NMR (C6D6, 101 MHz): used as the eluent at a flow rate of 1.0 mL min−1 and poly-
Table 4 Experimental details of X-ray diffraction studies
2
(L1)2Zn
8
9
11
Formula
C56H56Zn2N8O2
1003.82; 1.312
100; 2096
Orthorhombic
P212121
C44H44ZnN6
722.22; 1.236
100; 760
Orthorhombic
P21212
C25H47ZnN3O
471.02; 1.125
150; 1024
Monoclinic
P21/c
C29H60ZnN4Si2O
602.36; 1.115
150; 1312
C39H66ZnN4Si2O
728.50; 1.155
150; 1576
Monoclinic
P21/c
Mw (g mol−1); dcalcd. (g cm−3
T (K); F(000)
)
Crystal system
Space group
Triclinic
ˉ
P(1)
Unit cell: a (Å)
b (Å)
c (Å)
10.1814(3)
13.3856(4)
37.3018(12)
90
11.4428(7)
11.9319(8)
14.2169(9)
90
9.9872(6)
13.6954(8)
13.8027(8)
19.6030(11)
78.331(3)
18.8720(5)
18.6906(5)
12.3226(3)
90
24.3367(16)
11.7387(7)
90
α (°)
β (°)
90
90
90
90
102.843(2)
90
89.408(3)
81.562(3)
105.3650(10)
90
γ (°)
V (Å3); Z
5083.6(3); 4
0.956; multiscan
3.1–60.7; 0.97
114 196; 0.0226
11 290; 0.0487
0.0301
0.0765
1.054; 0.050(16)
0.338; −0.391
1941.1(2); 2
0.725; multiscan
2.7–60.7; 1.0
21 172; 0.0361
4452; 0.0607
0.0361
0.0909
1.120; 0.12(3)
0.336; −0.302
2781.8(3); 4
1.350; multiscan
5.3–71.4; 0.98
31 767; 0.0356
5351; 0.0454
0.0586
0.2791
1.271; —
0.709; −1.162
3589.0(4); 4
1.115; multiscan
2.8–60; 1.0
90 622; 0.0669
16 482; 0.0904
0.0644
0.1660
1.021; —
0.640; −0.380
4191.18(19); 4
1.010; multiscan
3.0–60.6; 1.0
70 912; 0.0163
9608; 0.0311
0.0509
0.1462
1.044; —
0.905; −0.751
μ (mm−1); Abs. Corr.
θ range (°); completeness
Collected reflections; Rσ
Unique reflections; Rint
R1(F) (I > 2σ(I))
wR(F2) (all data)
GoF(F2); flack-x
Residual electron density
16288 | Dalton Trans., 2018, 47, 16279–16291
This journal is © The Royal Society of Chemistry 2018

View Article Online
Dalton Transactions
Paper
styrene standards (Sigma–Aldrich, 1.5 mg mL−1, prepared and 10 R. E. Drumright, P. R. Gruber and D. E. Henton, Adv.
filtered (0.2 mm) directly prior to injection) were used for Mater., 2000, 12, 1841–1846.
calibration. Obtained molecular weights were corrected by a 11 M. Singhvi and D. Gokhale, RSC Adv., 2013, 3, 13558–
Mark–Houwink factor of 0.58.91
13568.
X-ray diffraction. Single crystals were obtained directly from 12 J. Vijayakumar, R. Aravindan and T. Viruthagiri, Chem.
isolation of the products as described above. Diffraction data Biochem. Eng. Q., 2008, 22, 245–264.
were collected on a Bruker Venture METALJET diffractometer 13 Y. Tokiwa and B. P. Calabia, Can. J. Chem., 2008, 86, 548–
(Ga Kα radiation) or a Bruker APEXII with a Cu microsource/ 555.
Quazar MX using the APEX2 software package.92 Data 14 B. J. O’Keefe, M. A. Hillmyer and W. B. Tolman, J. Chem.
reduction was performed with SAINT,93 absorption corrections
with SADABS.94 Structures were solved by dual-space refine-
Soc., Dalton Trans., 2001, 2215–2224, DOI: 10.1039/
b104197p.
ment (SHELXT).95 All non-hydrogen atoms were refined an- 15 O. Dechy-Cabaret, B. Martin-Vaca and D. Bourissou, Chem.
isotropic using full-matrix least-squares on F2 and hydrogen
Rev., 2004, 104, 6147–6176.
atoms refined with fixed isotropic U using a riding model 16 Z. Zhong, P. J. Dijkstra and J. Feijen, J. Biomater. Sci.,
(SHELXL97).96 Further experimental details can be found in
Polym. Ed., 2004, 15, 929–946.
Table 4 and in the ESI (CIF).†
17 J. Wu, T.-L. Yu, C.-T. Chen and C.-C. Lin, Coord. Chem. Rev.,
2006, 250, 602–626.
18 A. Amgoune, C. M. Thomas and J.-F. Carpentier, Pure Appl.
Chem., 2007, 79, 2013–2030.
19 R. H. Platel, L. M. Hodgson and C. K. Williams, Polym.
Rev., 2008, 48, 11–63.
20 C. A. Wheaton, P. G. Hayes and B. J. Ireland, Dalton Trans.,
2009, 4832–4846.
Conflicts of interest
There are no conflicts to declare.
21 N. Ajellal, J.-F. Carpentier, C. Guillaume, S. M. Guillaume,
M. Helou, V. Poirier, Y. Sarazin and A. Trifonov, Dalton
Trans., 2010, 39, 8363.
22 M. D. Jones, in Heterogenized Homogeneous Catalysts for
Fine Chemicals Production, ed. P. Barbaro and F. Liguori,
Springer Netherlands, 2010, vol. 33, pp. 385–412.
23 M. K. Kiesewetter, E. J. Shin, J. L. Hedrick and
R. M. Waymouth, Macromolecules, 2010, 43, 2093–2107.
24 M. J. Stanford and A. P. Dove, Chem. Soc. Rev., 2010, 39,
486–494.
Acknowledgements
Funding was supplied by the NSERC discovery program
(RGPIN-2016-04953) and the Centre for Green Chemistry and
Catalysis (FQRNT). I. M. contributed during her research
internship financed by the DAAD-RISE program. We thank
Karine Gilbert, Marie-Christine Tang and Alexandra Furtos for
support with mass spectroscopy and Francine Bélanger for
support with X-ray crystallography.
25 A. K. Sutar, T. Maharana, S. Dutta, C.-T. Chen and
C.-C. Lin, Chem. Soc. Rev., 2010, 39, 1724–1746.
26 C. M. Thomas, Chem. Soc. Rev., 2010, 39, 165.
27 J.-C. Buffet and J. Okuda, Polym. Chem., 2011, 2, 2758–
2763.
28 S. Dagorne, C. Fliedel and P. de Frémont, in Encyclopedia of
Inorganic and Bioinorganic Chemistry, John Wiley & Sons,
Ltd, 2011, DOI: 10.1002/9781119951438.eibc2416.
Notes and references
1 V. Nagarajan, A. K. Mohanty and M. Misra, ACS Sustainable
Chem. Eng., 2016, 4, 2899–2916.
2 E. Castro-Aguirre, F. Iñiguez-Franco, H. Samsudin, X. Fang
and R. Auras, Adv. Drug Delivery Rev., 2016, 107, 333–366.
3 S. Slomkowski, S. Penczek and A. Duda, Polym. Adv. 29 P. J. Dijkstra, H. Du and J. Feijen, Polym. Chem., 2011, 2,
Technol., 2014, 25, 436–447. 520–527.
4 T. A. Hottle, M. M. Bilec and A. E. Landis, Polym. Degrad. 30 S. Dutta, W.-C. Hung, B.-H. Huang and C.-C. Lin, in
Stab., 2013, 98, 1898–1907.
5 S. Inkinen, M. Hakkarainen, A.-C. Albertsson and
A. Södergård, Biomacromolecules, 2011, 12, 523–532.
Synthetic Biodegradable Polymers, ed. B. Rieger, A. Künkel,
G. W. Coates, R. Reichardt, E. Dinjus and T. A. Zevaco,
Springer-Verlag, Berlin, 2011, pp. 219–284.
6 J. Ahmed and S. K. Varshney, Int. J. Food Prop., 2011, 14, 31 C. A. Wheaton and P. G. Hayes, Comments Inorg. Chem.,
37–58. 2011, 32, 127–162.
7 C. K. Williams and M. A. Hillmyer, Polym. Rev., 2008, 48, 32 I. dos Santos Vieira and S. Herres-Pawlis, Eur. J. Inorg.
1–10. Chem., 2012, 2012, 765–774.
8 B. Gupta, N. Revagade and J. Hilborn, Prog. Polym. Sci., 33 J.-F. Carpentier, B. Liu and Y. Sarazin, in Advances in
2007, 32, 455–482.
9 E. T. H. Vink, K. R. Rábago, D. A. Glassner, B. Springs,
Organometallic Chemistry and Catalysis, John Wiley & Sons,
Inc., 2013, pp. 359–378, DOI: 10.1002/9781118742952.ch28.
R. P. O’Connor, J. Kolstad and P. R. Gruber, Macromol. 34 S. Dagorne and C. Fliedel, in Modern Organoaluminum
Biosci., 2004, 4, 551–564.
Reagents: Preparation, Structure, Reactivity and Use,
This journal is © The Royal Society of Chemistry 2018
Dalton Trans., 2018, 47, 16279–16291 | 16289

View Article Online
Paper
Dalton Transactions
ed. S. Woodward and S. Dagorne, Springer Berlin 58 F. Drouin, P. O. Oguadinma, T. J. J. Whitehorne,
Heidelberg, Berlin, Heidelberg, 2013, pp. 125–171, DOI:
10.1007/3418_2012_35.
R. E. Prud’homme and F. Schaper, Organometallics, 2010,
29, 2139–2147.
35 S. Dagorne, M. Normand, E. Kirillov and J.-F. Carpentier, 59 L. Wang and H. Ma, Dalton Trans., 2010, 39, 7897.
Coord. Chem. Rev., 2013, 257, 1869–1886.
36 B. H. Huang, S. Dutta and C. C. Lin, in Comprehensive
60 C.-Y. Sung, C.-Y. Li, J.-K. Su, T.-Y. Chen, C.-H. Lin and
B.-T. Ko, Dalton Trans., 2012, 41, 953–961.
Inorganic
Chemistry
II
(Second
Edition),
ed. 61 M. Honrado, A. Otero, J. Fernández-Baeza, L. F. Sánchez-
J. R. Poeppelmeier, Elsevier, Amsterdam, 2013,
pp. 1217–1249, DOI: 10.1016/B978-0-08-097774-4.00146-7.
37 R. Jianming, X. Anguo, W. Hongwei and Y. Hailin, Des.
Monomers Polym., 2013, 17, 345–355.
Barba, A. n. Lara-Sánchez, J. Tejeda, M. a. P. Carrión,
J. Martínez-Ferrer, A. Garcés and A. M. Rodríguez,
Organometallics, 2013, 32, 3437–3440.
62 H. Wang and H. Ma, Chem. Commun., 2013, 49, 8686–8688.
38 A. Sauer, A. Kapelski, C. Fliedel, S. Dagorne, M. Kol and 63 S. Abbina and G. Du, ACS Macro Lett., 2014, 3, 689–692.
J. Okuda, Dalton Trans., 2013, 42, 9007–9023.
39 S. M. Guillaume, E. Kirillov, Y. Sarazin and J.-F. Carpentier,
Chem. – Eur. J., 2015, 21, 7988–8003.
64 M. Honrado, A. Otero, J. Fernández-Baeza, L. F. Sánchez-
Barba, A. Garcés, A. n. Lara-Sánchez and A. M. Rodríguez,
Organometallics, 2014, 33, 1859–1866.
40 J. P. MacDonald and M. P. Shaver, in Green Polymer 65 Z. Mou, B. Liu, M. Wang, H. Xie, P. Li, L. Li, S. Li and
Chemistry: Biobased Materials and Biocatalysis, American
Chemical Society, 2015, vol. 1192, ch. 10, pp. 147–167.
41 S. Paul, Y. Zhu, C. Romain, R. Brooks, P. K. Saini and
C. K. Williams, Chem. Commun., 2015, 51, 6459–6479.
42 E. Le Roux, Coord. Chem. Rev., 2016, 306, 65–85.
43 T. M. Ovitt and G. W. Coates, J. Am. Chem. Soc., 2002, 124,
1316–1326.
D. Cui, Chem. Commun., 2014, 50, 11411–11414.
66 H. Wang, Y. Yang and H. Ma, Macromolecules, 2014, 47,
7750–7764.
67 M. Honrado, A. Otero, J. Fernández-Baeza, L. F. Sánchez-
Barba, A. Garcés, A. Lara-Sánchez, J. Martínez-Ferrer,
S. Sobrino and A. M. Rodríguez, Organometallics, 2015, 34,
3196–3208.
44 H. Du, A. H. Velders, P. J. Dijkstra, J. Sun, Z. Zhong, 68 Y. Sun, Y. Cui, J. Xiong, Z. Dai, N. Tang and J. Wu, Dalton
X. Chen and J. Feijen, Chem. – Eur. J., 2009, 15, 9836–9845. Trans., 2015, 44, 16383–16391.
45 K. Press, I. Goldberg and M. Kol, Angew. Chem., Int. Ed., 69 Y. Yang, H. Wang and H. Ma, Inorg. Chem., 2015, 54, 5839–
2015, 54, 14858–14861. 5854.
46 H. Ma, T. P. Spaniol and J. Okuda, Angew. Chem., Int. Ed., 70 T. Rosen, Y. Popowski, I. Goldberg and M. Kol, Chem. –
2006, 45, 7818–7821. Eur. J., 2016, 22, 11533–11536.
47 A. Amgoune, C. M. Thomas, T. Roisnel and J.-F. Carpentier, 71 H. Wang, Y. Yang and H. Ma, Inorg. Chem., 2016, 55, 7356–
Chem. – Eur. J., 2006, 12, 169–179. 7372.
48 A. J. Chmura, C. J. Chuck, M. G. Davidson, M. D. Jones, 72 C. Kan, J. Hu, Y. Huang, H. Wang and H. Ma,
M. D. Lunn, S. D. Bull and M. F. Mahon, Angew. Chem., Int.
Ed., 2007, 46, 2280–2283.
49 A. J. Chmura, M. G. Davidson, C. J. Frankis, M. D. Jones
and M. D. Lunn, Chem. Commun., 2008, 1293–1295.
Macromolecules, 2017, 50, 7911–7919.
73 D. E. Stasiw, A. M. Luke, T. Rosen, A. B. League,
M. Mandal, B. D. Neisen, C. J. Cramer, M. Kol and
W. B. Tolman, Inorg. Chem., 2017, 56, 14366–14372.
50 M. D. Jones, S. L. Hancock, P. McKeown, P. M. Schafer, 74 A. W. Addison, T. N. Rao, J. Reedijk, J. van Rijn and
A. Buchard, L. H. Thomas, M. F. Mahon and J. P. Lowe,
G. C. Verschoor, J. Chem. Soc., Dalton Trans., 1984, 1349–
Chem. Commun., 2014, 50, 15967–15970.
1356, DOI: 10.1039/DT9840001349.
51 M. D. Jones, L. Brady, P. McKeown, A. Buchard, 75 C. R. Groom, I. J. Bruno, M. P. Lightfoot and S. C. Ward,
P. M. Schafer, L. H. Thomas, M. F. Mahon, T. J. Woodman
and J. P. Lowe, Chem. Sci., 2015, 6, 5034–5039.
Acta Crystallogr., Sect. B: Struct. Sci., Cryst. Eng. Mater.,
2016, 72, 171–179.
52 S. Fortun, P. Daneshmand and F. Schaper, Angew. Chem., 76 C. K. Williams, L. E. Breyfogle, S. K. Choi, W. Nam,
Int. Ed., 2015, 54, 13669–13672.
53 P. Daneshmand, A. van der Est and F. Schaper, ACS Catal.,
2017, 7, 6289–6301.
54 P. Daneshmand, S. Fortun and F. Schaper, Organometallics,
2017, 36, 3860–3877.
55 P. Daneshmand, J. L. Jiménez-Santiago, M. Aragon–Alberti
and F. Schaper, Organometallics, 2018, 37, 1751–
1759.
V. G. Young, M. A. Hillmyer and W. B. Tolman, J. Am.
Chem. Soc., 2003, 125, 11350–11359.
77 P. D. Knight, A. J. P. White and C. K. Williams, Inorg.
Chem., 2008, 47, 11711–11719.
78 Z. Zheng, G. Zhao, R. Fablet, M. Bouyahyi, C. M. Thomas,
T. Roisnel, O. Casagrande Jr. and J.-F. Carpentier, New
J. Chem., 2008, 32, 2279–2291.
79 S. Yann, P. Valentin, R. Thierry and C. Jean-François,
Eur. J. Inorg. Chem., 2010, 2010, 3423–3428.
56 M. Cheng, A. B. Attygalle, E. B. Lobkovsky and
G. W. Coates, J. Am. Chem. Soc., 1999, 121, 11583– 80 V. Poirier, T. Roisnel, J.-F. Carpentier and Y. Sarazin, Dalton
11584. Trans., 2011, 40, 523–534.
57 T. M. Ovitt and G. W. Coates, J. Am. Chem. Soc., 1999, 121, 81 T. Ebrahimi, E. Mamleeva, I. Yu, S. G. Hatzikiriakos and
4072–4073.
P. Mehrkhodavandi, Inorg. Chem., 2016, 55, 9445–9453.
16290 | Dalton Trans., 2018, 47, 16279–16291
This journal is © The Royal Society of Chemistry 2018

View Article Online
Dalton Transactions
Paper
82 D. Jedrzkiewicz, D. Kantorska, J. Wojtaszak, J. Ejfler and 89 S. Mondal, S. M. Mandal, T. K. Mondal and C. Sinha,
S. Szafert, Dalton Trans., 2017, 46, 4929–4942. Spectrochim. Acta, Part A, 2015, 150, 268–279.
83 S. M. Kirk, P. McKeown, M. F. Mahon, G. Kociok-Köhn, 90 M. Savva, P. Chen, A. Aljaberi, B. Selvi and M. Spelios,
T. J. Woodman and M. D. Jones, Eur. J. Inorg. Chem., 2017,
2017, 5417–5426.
84 G. Labourdette, D. J. Lee, B. O. Patrick, M. B. Ezhova and
P. Mehrkhodavandi, Organometallics, 2009, 28, 1309–1319.
85 S. C. Marinescu, T. Agapie, M. W. Day and J. E. Bercaw,
Organometallics, 2007, 26, 1178–1190.
Bioconjugate Chem., 2005, 16, 1411–1422.
91 M. Save, M. Schappacher and A. Soum, Macromol. Chem.
Phys., 2002, 203, 889–899.
92 APEX2, Release 2.1-0, Bruker AXS Inc., Madison, USA,
2006.
93 SAINT, Release 7.34A, Bruker AXS Inc., Madison, USA, 2006.
86 M. Cheng, D. R. Moore, J. J. Reczek, B. M. Chamberlain, 94 G. M. Sheldrick, SADABS, Bruker AXS Inc., Madison, USA,
E. B. Lobkovsky and G. W. Coates, J. Am. Chem. Soc., 2001,
123, 8738–8749.
87 R. J. Herolds, L. Aggarwal and V. Neff, Can. J. Chem., 1963,
41, 1368.
1996 & 2004.
95 G. Sheldrick, Acta Crystallogr. Sect. A: Found. Adv., 2015, 71,
3–8.
96 G. M. Sheldrick, Acta Crystallogr., Sect. A: Found.
Crystallogr., 2008, 64, 112–122.
88 D. Y. Lee and J. F. Hartwig, Org. Lett., 2005, 7, 1169–1172.
This journal is © The Royal Society of Chemistry 2018
Dalton Trans., 2018, 47, 16279–16291 | 16291