Tert-butylamidinate tin(II) complexes: High activity, single-site initiators for the controlled production of polylactide

Article abstract of DOI:10.1039/b706663e

The tin(ii) coordination chemistry of two monoanionic N,N′-bis(2,6- diisopropylphenyl)alkylamidinate ligands is described. Complexation studies with the acetamidinate, [MeC(NAr)₂]^-, (Ar = 2,6- ⁱPr₂C₆H

Full text of DOI:10.1039/b706663e

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/dalton | Dalton Transactions
Tert-butylamidinate tin(II) complexes: high activity, single-site initiators for
the controlled production of polylactide†‡§
Nonsee Nimitsiriwat,a Vernon C. Gibson,*a Edward L. Marshall,*a Andrew J. P. White,a Sophie H. Daleb and
Mark R. J. Elsegoodb
Received 3rd May 2007, Accepted 17th July 2007
First published as an Advance Article on the web 9th August 2007
DOI: 10.1039/b706663e
The tin(II) coordination chemistry of two monoanionic N,Nꢀ-bis(2,6-diisopropylphenyl)alkylamidinate
ligands is described. Complexation studies with the acetamidinate, [MeC(NAr)2]−, (Ar = 2,6-iPr2C6H3)
are complicated by the side formation of the bis(amidinate) tin(II) compound, [MeC(NAr)2]2Sn, 1. By
contrast, the bulkier tert-butylamidinate, [tBuC(NAr)2]−, allows tin(II) mono-halide, -alkoxide and
-amide complexes to be isolated cleanly in high yields. Thus, the reaction of [tBuC(NAr)2]H with nBuLi
and subsequent treatment with SnCl2 generates [tBuC(NAr)2]SnCl, 2, in ca. 70% yield. Reactions of 2
with LiOiPr, LiNMe2 and LiNTMS2 afford [tBuC(NAr)2]Sn(OiPr), 3, [tBuC(NAr)2]Sn(NMe2), 4, and
[tBuC(NAr)2]Sn(NTMS2), 5, respectively. The molecular structures of complexes 1–4 are reported.
Complexes 3, 4 and 5 have been investigated as initiators for the ring-opening polymerisation of
rac-lactide: 3 and 4 display characteristics of well-controlled polymerisation initiators, but high
molecular weight polymer is observed with 5 due to inefficient initiation, a consequence of the steric
bulk of the NTMS2 unit. Polymerisations with 3 and 4 are faster than for the corresponding
b-diketiminate tin(II) complexes, consistent with the more open nature of the tin(II) coordination sphere.
However, we also observed an unusual stereochemical invari-
ance among the PLA samples prepared using this initiator family:
Introduction
The replacement of petrochemically derived products with
a wide range of b-diketiminate tin(II) initiators all afforded PLA
biomass based alternatives is exemplified by the commercial
with a similar heterotactic bias. Computational analysis led us
synthesis of poly(lactide), PLA, from starch.1,2 Industrially, tin(II)
to propose that this behaviour arises from a reluctance of the 5s
bis(2-ethylhexanoate), SnOct2, is employed as the polymerisation
lone pair to mix with other Sn orbitals. The orientation of bound
initiator because it is soluble and thermally stable in molten lactide,
monomer relative to the propagating polymer chain is therefore
thereby allowing the process to be conducted in the melt phase in
the absence of organic solvents.
Numerous metal alkoxides are known to initiate polymerisation
of lactide, LA3–5 and although details of the mechanism by
determined principally by the need to interact with the tin centre
through orthogonally aligned pd hybrids, relegating the steric and
electronic properties of the b-diketiminate ligand to a secondary
role.
We thus decided to target amidinate ligands containing N,Nꢀ-
which SnOct2 functions remain ambiguous, in the presence of
alcohol the propagating species is commonly believed to be
2,6-diisopropylphenyl substituents for two major reasons. Firstly,
a tin(II) alkoxide.6–10 In order to synthesise single-site, well-
defined analogues of SnOct2 we have therefore previously reported
the b-diketiminate tin(II) alkoxide complex, [HC{C(Me)N-2,6-
iPr2C6H3}2]Sn(OiPr), and shown that it polymerises rac-LA in
a well-controlled manner.11,12 Although slower than comparable
complexes of more electropositive metals (e.g. Zn,13–15 Mg,15,16
Ca17,18 and Fe19), improved activities were obtained with less
hindered and with halogenated b-diketiminates, as anticipated for
a coordinative-insertion mechanism.12
we envisaged that amidinate20,21 ligands would give more ac-
cessible metal centres than their corresponding b-diketiminates,
potentially allowing improved catalytic activities to be realised
(Scheme 1). Secondly, we also wanted to investigate whether
opening up the coordination sphere whilst retaining the same
aryl substituents would lead to a change in stereocontrol. Here
therefore, we describe the synthesis of tin(II) complexes of N,Nꢀ-
bis(2,6-diisopropylphenyl) alkylamidinate ligands and compare
aDepartment of Chemistry, Imperial College London, South Kensing-
ton Campus, London, UK SW7 2AZ. E-mail: v.gibson@imperial.ac.uk,
e.marshall@imperial.ac.uk
bChemistry Department, Loughborough University, Loughborough, Leices-
tershire, UK LE11 3TU
† The HTML version of this article has been enhanced with colour images.
‡ CCDC reference numbers 645832–645835. For crystallographic data in
CIF or other electronic format see DOI: 10.1039/b706663e
§ Electronic supplementary information (ESI) available: Crystallographic
data for complex 1 and details of kinetic studies using complex 3;
Fig. S1–S7. See DOI: 10.1039/b706663e
Scheme 1
4464 | Dalton Trans., 2007, 4464–4471
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
lactide polymerisation behaviour with their b-diketiminate ana-
logues.
97.98(8)◦]. This structure is analogous to several previously
reported tin(II) bis(amidinate)s23–28 and is therefore not discussed
in further detail.
Whilst N,N ꢀ-silyl26,27 and N,N ꢀ-cyclohexyl24 bis(amidinate) tin(II)
compounds exhibit a single set of N-substituent NMR resonances
at 298 K, axial and equatorial sites of complex 1 do not readily
exchange at room temperature. Hence, eight equally intense methyl
doublets and three isopropyl methine septets (in the ratio 2 : 1 : 1)
are observed in its1H NMR spectrum. Suchbehaviour is attributed
to the steric bulk of the N-aryl units in 1, with coalescence of the
iPr resonances only occurring at elevated temperature (348 K).
The tetra-coordinate nature of the metal in the solid state is
retained in solution as shown by a singlet 119Sn NMR resonance
at −394.8 ppm, notably upfield of signals for the three coordinate
species described below (which occur between +50 and −30 ppm).
In light of the ease with which the bis(amidinate) 1 forms,
we concluded that the {[MeC(NAr)2]Sn(II)} fragment is unlikely
to prove a suitable stabilising unit for single-site polymerisation
systems. We therefore turned our attention to the complexation
chemistry of the analogous tert-butylamidinate ligand.29 Jordan
and co-workers have previously shown that a tert-butyl group on
the amidinate carbon forces the N-substituents to project further
towards the metal30 and we rationalised that this might serve to
suppress bis(chelate) formation in our system, while still affording
markedly less steric protection than b-diketiminate ligands.
In situ lithiation of [tBuC(NAr)2]H using nBuLi followed by
addition to a toluene suspension of SnCl2 afforded the tin(II)
monochloride complex, [tBuC(NAr)2]SnCl, 2 in ca. 70% yield
following recrystallisation from heptane (Scheme 3). No evidence
for the presence of the bis(amidinate) analogue of 1 was observed
in either the crude reaction product or in the recrystallised
material.
Results and discussion
Synthetic studies
Our initial investigations were performed using the acetamidine,
[MeC(NAr)2]H, Ar = 2,6-iPr2C6H3.22 However, all attempts to syn-
thesise tin(II) complexes of this ligand resulted in the co-generation
of the bis(amidinate) complex, [MeC(NAr)2]2Sn, 1. For example,
lithiation of the acetamidine and subsequent addition to SnCl2
resulted in a mixture of the desired [MeC(NAr)2]SnCl complex
and 1 in a 2 : 3 ratio. Similarly, the reaction of MeC(NAr)2H
with Sn(NMe2)2 in Et2O afforded [MeC(NAr)2]Sn(NMe2) contam-
inated with ca. 20–25% 1. The more bulky diamide Sn(NTMS2)2
reacted very slowly with MeC(NAr)2H at room temperature, and
when the reaction was repeated at 80 ◦C overnight, an analytically
pure sample of 1 was obtained (Scheme 2).
Scheme 2
X-Ray diffraction analysis of 1 revealed the presence of two
independent C2-symmetric molecules (I and II) with essentially
identical geometries (molecule I is shown in Fig. 1, molecule II in
Fig. S2 in the ESI§). Though highly distorted, the coordination
geometry can be viewed as saw-horse like, with N(2) and N(2A)
in the axial positions, and N(1) and N(1A) residing in the
equatorial plane. Consistent with the use of 5p4d hybrid frontier
orbitals, many of the interligand angles at tin approach 90◦ [N(1)–
Sn–N(1A) = 101.33(12); N(1)–Sn–N(2A) = N(2)–Sn–N(1A) =
Scheme 3
X-Ray diffraction quality crystals of 2 were obtained by
allowing a saturated heptane solution to cool slowly from 70 ◦C to
ambient temperature. The molecular structure (Fig. 2) is similar to
that of [tBuC(NAr)2]GeCl31 with the tin centre adopting a three-
coordinate distorted pyramidal geometry. Hybridisation of the
metal 4d and 5p orbitals is supported by the near perpendicular
alignment of the Sn–Cl vector to the amidinate ligand (N(1)–Sn–
Cl = 94.93(4), N(2)–Sn–Cl = 94.89(4)◦). The anionic charge on the
ligand is clearly delocalised over the amidinate backbone as shown
by the symmetric bonding pattern within the core (N(1)–C(1) =
˚
1.3418(19), N(2)–C(1) = 1.3374(19) A; N(1)–Sn = 2.2022(13),
˚
N(2)–Sn = 2.2104(13) A). The tin atom is essentially coplanar
with the amidinate (sum of the internal angles of the CN2Sn
core = 359.91◦), whilst N(1), N(2) and C(1) all approach trigonal
planarity as anticipated for sp2 hybridisation, with the sum of the
Fig. 1 The molecular structure of one (molecule I) of the two independent
C2-symmetric molecules present in the crystals of 1. The suffix ‘A’indicates
1
the equivalent positions (1 − x, y, − z).
2
This journal is
The Royal Society of Chemistry 2007
Dalton Trans., 2007, 4464–4471 | 4465
©

View Article Online
Scheme 4
contrasts markedly with a previous study of less bulky N-silyl
benzamidinate ligands, with only exceptionally bulky alkoxide
tin(II) complexes proving isolable.27
1
Fig. 2 The molecular structure of complex 2.
The H NMR spectrum of 3 at 298 K is similar to that of
2, although some broadening of the aryl resonances is observed
(including unresolved multiplets for Hmeta and Hpara, and two broad
septets and four broad doublets for the CHMe2 groups). The
resonances of the isopropoxide ligand, and of the amidinate tert-
butyl substituent are all sharp, suggesting that in solution at room
temperature these units are free to rotate, but their steric bulk
serves to restrict rotation of the aromatic rings about the N–Cipso
vectors.
Complex 3 (Fig. 4) is essentially isostructural to 2: the tin atom
again displays a distorted pyramidal geometry and is coplanar
with the three amidinate ring atoms (sum of internal angles in the
CN2Sn core = 359.55◦). Geometries at N(1), N(2) and C(1) are
consistent with sp2-hybridisation (sum of angles = 358.27, 357.45
and 359.85◦ respectively). The separation between the isopropyl
angles around these three atoms being 357.82, 358.94 and 359.60◦
respectively. The acute N–Sn–N angle of 59.40◦ is comparable to
values previously recorded for tin(II)23–28 and tin(IV)32,33 amidinate
complexes. The aryl rings are aligned in a near-orthogonal manner
relative to the ligand chelate plane, and the N–Cipso bonds are
orientated out of the CN2Sn plane in a syn fashion, causing the
two isopropyl moieties below the amidinate chelate plane to lie
significantly closer together than the pair on the opposite face
(distances between C(15) and C(27) = 4.11, C(12) and C(24) =
˚
5.71 A). The less sterically congested face of the complex is
therefore used to accommodate both the chloride ligand and one
of the tert-butyl methyl groups (torsional angle between C(2)–C(3)
and Sn–Cl = 0.2◦).
To examine further the extent of the steric protection afforded
the metal centre by the tert-butylamidinate ligand we have com-
pared the structure of 2 to that of a b-diketiminate counterpart,
[HC{MeC(N-2,6-iPr2C6H3)}2]SnCl, which has previously been
reported by Roesky, Power and co-workers.34 As shown in Fig. 3,
the b-diketiminate ligand evidently encapsulates both the tin atom
and the chloride ligand more effectively than the amidinate unit.
˚
˚
methine carbons C(12) and C(24) is 5.59 A, 1.33 A greater than
the distance between C(15) and C(27), and the alkoxide oxygen is
located in a similar position to the chloride of complex 2, with an
average N–Sn–O bond angle of 93.8◦.
Fig. 3 Comparison of the solid states structures of (a) complex 2 with
(b) [HC{MeC(N-2,6-iPr2C6H3)}2]SnCl viewed along the Sn–Cl vector (H
atoms omitted for clarity).
The 1H NMR (C6D6) spectrum of 2 indicates that its solid state
structure is maintained in solution. The tert-butyl group on the
ligand backbone obstructs rotation of the aryl rings on the NMR
timescale, with two distinct isopropyl resonance patterns observed
(CHMe2 septets: d 4.06, 3.48; CHMe2 doublets: d 1.47, 1.31, 1.26,
1.07).
Complex 2 serves as a convenient precursor to alkoxide and
amide complexes (Scheme 4). For example, its reaction with
LiOiPr cleanly generates [tBuC(NAr)2]SnOiPr, 3. This observation
Fig. 4 The molecular structure of complex 3.
Treatment of complex 2 with LiNMe2 and with LiNTMS2
affords the mononuclear amides [tBuC(NAr)2]SnNR2 (R = Me,
4; R = SiMe3, 5), again with no bis(amidinate) contamination. A
further repercussion of the increased size of the ancillary ligand
became apparent when we tried to synthesise 5 by reaction of the
tert-butylamidine with Sn(NTMS2)2: no reaction was observed
even after heating to 80 ◦C in toluene for 70 h.
4466 | Dalton Trans., 2007, 4464–4471
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
Table 1 Selected bond parameters for 1–4
of the polymerisation and monomer conversion and molecular
weight were determined by 1H NMR and GPC, respectively.
As shown in Fig. 6, complex 3 is a well-controlled polymerisa-
tion initiator: a plot of Mn versus monomer conversion is linear,
and molecular weight distributions are narrow, although a slight
deviation from linearity is observed at high conversions (>85%)
indicative of an increased prevalence for transesterification as the
monomer concentration declines.35
1 (mol I)
2
3
4
˚
Bond lengths/A
Ca–N
Sn–N
Sn–X
1.343(3)
1.314(3)
2.444(2)
2.197(2)
—
1.3418(19)
1.3374(19)
2.2104(13)
2.2022(13)
2.4282(6)
1.345(2)
1.338(2)
2.2216(15)
2.2085(15)
2.0056(14)
1.354(2)
1.331(2)
2.2600(15)
2.2401(16)
2.0414(18)
Bond angles/◦
Ca–N–Sn
97.80(17)
87.59(16)
57.13(8)
114.1(2)
—
95.68(9)
95.44(9)
59.40(5)
109.37(13)
94.93(4)
95.93(11)
95.09(11)
59.21(5)
109.32(15)
94.37(5)
95.77(11)
95.53(11)
58.38(6)
109.66(16)
99.42(7)
N–Sn–Nꢀ
N–Ca–Nꢀ
N–Sn–X
—
94.89(4)
93.27(6)
99.36(6)
Ca–N–Cipso
Sn–N–Cipso
Cb–Ca–N
121.7(2)
119.9(2)
145.61(18)
133.96(18)
123.1(2)
122.8(2)
131.95(13)
131.93(13)
131.57(10)
130.19(10)
125.00(13)
125.23(13)
132.72(15)
132.32(15)
130.46(11)
129.20(11)
126.34(16)
124.19(15)
131.64(16)
130.57(16)
129.90(12)
127.32(12)
126.54(16)
123.51(16)
Fig. 6 A plot of Mn vs. monomer conversion for the polymerisation of
rac-LA using complex 3 as the initiator ([LA] : [3] = 100; 60 ◦C; toluene;
polydispersities given in parentheses).
A linear relationship is also observed between Mn and [LA]0/[3]
(Fig. S4, ESI§). The polymerisation of rac-LA using 3 as an
initiator is further characterised by a first-order dependence upon
monomer concentration, as demonstrated by a linear correlation
between ln{[LA]0/[LA]t} and reaction time (Fig. 7). However,
repeating this analysis over a range of initiator concentrations
(1.4–3.5 × 10−2 mol L−1), revealed a non-first-order dependence
upon [3]. Examining the variation of rate with initiator concen-
tration indicates that the polymerisation adheres to the rate law
−d[LA]/dt = k[LA][3]0.21 (i.e. a rate law order dependence on [3]
of just 0.21; see ESI§). Thus, with [LA] kept constant and the
concentration of 3 varied so as to give initial [LA]/[3] ratios of 80,
100, 133 and 200, after 60 min conversions reached 83%, 78%, 75%
and 72%, respectively. This result contrasts with our findings on
the b-diketiminate analogue, [HC{MeC(NAr)}2]Sn(OiPr), which
exhibits first-order kinetics.12
The molecular structure of 4 (Fig. 5) also features a distorted
pyramidal tin(II) centre and amidinate bond parameters similar to
those found in the structures of 2 and 3 (pertinent data for 1–4 are
summarised in Table 1). Notably, the angles N(1)–Sn–N(3) and
N(2)–Sn–N(3) (99.36 and 99.42◦, respectively) are ca. 5–6◦ greater
than those in 2 and 3, reflecting the increased steric bulk of the
NMe2 unit relative to the Cl and OiPr ligands.
Fig. 5 The molecular structure of complex 4.
Polymerisation of rac-lactide
In order to study the behaviour of the alkoxide 3 and the amides
4 and 5 as lactone ring-opening polymerisation initiators, each
complex was mixed with 100 equivalents of rac-LA in toluene at
60 ◦C, i.e. under the same conditions examined for b-diketiminate
tin(II) initiators.11,12 Aliquots were then removed over the course
Fig. 7 A plot of ln{[LA]0/[LA]t} vs. time for the polymerisation of rac-LA
using complex 3 as the initiator (60 ◦C; toluene; [LA]0 = 0.28 M; [3] =
0.0014 M; [LA]0 : [3] = 200).
This journal is
The Royal Society of Chemistry 2007
Dalton Trans., 2007, 4464–4471 | 4467
©

View Article Online
Fractional rate law orders of this type have been observed
with other lactide polymerisation initiators;27,36–39 in the most
salient study, Aubrecht, Hillmyer and Tolman found a rate
law dependence of 0.33 for an equimolar initiating mixture
◦
of [PhC(NSiMe2Ph)2]Sn(OCPh3) + PhCH2OH at 80 C.27 Low
rate orders have traditionally been explained in terms of the
structural reorganisation of the metal coordination sphere to give
multinuclear aggregates. In the amidinate tin(II) system, however,
we believe that aggregation is more likely to occur via the reversible
coordination of the propagating polyester chains to the readily
accessible active sites. Nevertheless, the polymerisation of rac-LA
using 3 is still significantly faster than with the b-diketiminate
initiators [HC{RC(NAr)}2]Sn(OiPr) which typically require ca. 4
Scheme 5
tert-butyl Hc singlet at 0.85 ppm and the emergence of a new
singlet at d 0.91 (Hh). Other resonances associated with the newly
formed compound include quartets at d 4.88 and 4.57 (He and
◦
t
(R = Me) and 8 (R = Bu) h at 60 C to consume 100 equivalents of
monomer (kapp = 0.734 and 0.384 h−1, respectively).5 By contrast,
i
Hf), and the Hd septet at 4.90 ppm, indicative of the CO2 Pr end
group. Initiation (i.e. formation of 6) is therefore rapid, whereas
the insertion of a second lactide molecule is far slower (Fig. 7
confirms that a short delay—ca. 8 min—also exists prior to
polymerisation at 60 ◦C). Similar delays have been observed with
the tin(II) diketiminate initiators and have recently been the focus
of a theoretical study.12
complex 3 attains similar levels of conversion within 90 min (kapp
=
1.78 h−1).
In order to examine the initiation process more closely, complex
3 was mixed with 5 equivalents of rac-LA in CD2Cl2 at 298 K and
the reaction was monitored by 1H NMR spectroscopy. As shown
in Fig. 8, within just 15 min the initiator and one equivalent of
monomer are cleanly converted into a new species, which in light
of our previous study12 is believed to be the first insertion product
6 (Scheme 5). Thereafter, the reactants remain unchanged for ca.
2 h, before gradual consumption of the monomer is observed.
The rapid formation of 6 is signified by the disappearance of the
The polymerisations of rac-LA initiated with 4 and 5 also
demonstrate linearly proportional increases in molecular weight
with monomer consumption (Fig. 9). However, whilst the
dimethylamide complex 4 produces molecular weights similar
to those recorded for the alkoxide initiator 3, use of the more
bulky bis(trimethylsilyl)amide 5 results in much higher molecular
weights (and appreciably broader molecular weight distributions).
For example, at 93% and 92% conversion, respectively, the Mn
values of PLA synthesised using 3 and 4 are 19 600 and 20 700;
by contrast, an Mn of 122 000 was recorded for PLA prepared
from 5 at 91% LA consumption. We ascribe this observation
to unfavourable initiation by the bulky NTMS2 unit, leading
to a smaller number of propagating chains than expected from
the initial monomer : initiator stoichiometry. The slow rate of
initiation also impacts upon the time taken for the polymerisations
to attain high conversion. Hence, under the conditions employed,
complex 3 consumes > 90% of 100 equivalents rac-LA within
90 min; similar levels of conversion with 4 and 5 require 180 and
240 min, respectively.
Fig. 9 A plot of Mn vs. monomer conversion for the polymerisation of
rac-LA using complexes 4 (ꢁ) and 5 (ꢀ) as the initiator (60 ◦C; toluene
[LA]0 : [4] = 100; [LA]0 : [5] = 100; polydispersities given in parentheses).
A moderate bias to heterotactic assembly40 is observed with
initiators 3–5, as exemplified by Fig. 10. Similar levels of hetero-
selectivity have been observed with many tin(II) initiators,11,12,27 and
our observations lend further support to computational studies12
Fig. 8 1H NMR spectra of the reaction between 3 and 5 equivalents
rac-LA (CD2Cl2, 298K, 250 MHz; * = CHDCl2). Spectrum shown at
t = 0 was recorded prior to monomer addition.
4468 | Dalton Trans., 2007, 4464–4471
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
were analysed using Cirrus Software (Polymer Laboratories) and
molecular weights are reported versus polystyrene calibrants.
Synthesis of [MeC(NAr)2]2Sn, 1
To a 30 mL toluene solution of Sn{N(TMS)2}2 (0.48 g, 1.09 ×
10−3 mol) was added a 30 mL toluene suspension of MeC(NAr)2H
(0.41 g, 1.09 × 10−3 mol) dropwise at room temperature. The
◦
reaction was stirred at 80 C overnight. Solvent was distilled off
under reduced pressure and the residue was washed with cold
pentane and then dried to give [MeC(NAr)2]2Sn, 1, as a pale yellow
solid, (Found: C, 71.52; H, 8.55; N, 6.59%; C52H74N4Sn requires
Fig. 10 The methine region of the 1H decoupled NMR spectrum (298 K,
400 MHz, CDCl3) of PLA prepared using complex 3 at 60 ◦C in toluene
(i = isotactic, s = syndiotactic dyads).
C, 71.47; H, 8.54; N, 6.41%); dH(C6D6) 7.30–7.07 (m, 12H, Hmeta
H
,
para), 3.76 (br sept, 4H, 3JHH = 6.5 Hz, CH(CH3)2), 3.33 (br sept,
2H, 3JHH = 6.7 Hz, CH(CH3)2), 3.17 (br sept, 2H, 3JHH = 6.6 Hz,
CH(CH3)2), 1.57 (br d, 6H, 3JHH = 6.6 Hz, CH(CH3)2), 1.48 (br d,
6H, 3JHH = 6.6 Hz, CH(CH3)2), 1.38 (s, 6H, (CH3)C(NAr)2), 1.36
(overlapping br d, 6H, CH(CH3)2), 1.29 (br d, 6H, 3JHH = 6.7 Hz,
which indicate that the tin 5s2 lone pair of electrons plays an
instrumental role in determining stationary point geometries along
the reaction coordinate.
3
CH(CH3)2), 1.25 (br d, 6H, JHH = 6.8 Hz, CH(CH3)2), 1.16 (br
d, 6H, 3JHH = 6.6 Hz, CH(CH3)2), 0.96 (br d, 6H, 3JHH = 6.7 Hz,
CH(CH3)2), 0.68 (br d, 6H, 3JHH = 6.7 Hz, CH(CH3)2); dC(C6D6)
168.69 (MeC(NAr)2), 145.09 (Cortho), 144.75 (Cipso), 143.88 (Cortho),
142.71 (Cipso), 141.81 (Cortho), 141.48 (Cortho), 125.71 (Cpara),
124.93 (Cpara), 124.41 (Cmeta), 123.52 (Cmeta), 123.26 (Cmeta), 29.30
(CHMe2), 28.81 (CHMe2), 28.17 (CHMe2), 27.46 (CH(CH3)2),
25.57 (CH(CH3)2), 25.26 (CH(CH3)2), 25.09 (CH(CH3)2),
24.59 (CH(CH3)2), 23.57 (CH(CH3)2), 23.40 (CH(CH3)2), 22.30
(CH(CH3)2), 16.64 ((CH3)C(NAr)2); d119Sn(C6D6) −394.77; m/z
713 (M − (CH(CH3)2)2C6H3).
Conclusion
The N,N ꢀ-bis(2,6-diisopropylphenyl)(tert-butyl) amidinate ligand
allows tin(II) coordination chemistry to be explored without the
complication of bis(chelate) side formation. X-Ray crystallogra-
phy indicates that this ligand provides the metal centre with far
less protection than its b-diketiminate counterpart, and rac-lactide
polymerisations are accordingly faster, even though the activities
are presumably tempered by aggregation. For complexes 3 and
4, the increased reactivity is coupled with good molecular weight
control, but the identity of the initiating group is a crucial factor:
the much bulkier bis(trimethylsilyl)amide group initiates the poly-
merisation of rac-LA poorly, and high molecular weights result.
The tacticity of the PLA products are similar to those obtained
using b-diketiminate tin(II) initiators (and SnOct212) giving strong
support to our hypothesis that the presence of the 5s2 lone pair
of electrons and the deployment of 4d5p hybrid orbitals exert an
influence over the monomer ring-opening event much larger than
the attendant ligand(s).
Synthesis of [tBuC(NAr)2]SnCl, 2
A suspension of [tBuC(NAr)2]Li (9.55 × 10−3 mol) in 60 mL
toluene (formed in situ from the reaction of nBuLi, 2.5 M in
hexanes, with [tBuC(NAr)2]H) was added to SnCl2 (1.813 g, 9.56 ×
10−3 mol) in toluene (30 mL) at −78 ◦C. The reaction mixture
was allowed to warm to room temperature and then stirred for
18 h. The yellow solution was filtered and volatiles removed to
give a beige solid. This was recrystallised from heptane to give
[tBuC(NAr)2]SnCl as pale yellow crystals (3.781 g, 6.59 × 10−3 mol,
69% yield). Crystals suitable for X-ray crystallography were grown
by slow cooling of a heptane solution from 70 ◦C to room
temperature, (Found: C, 60.77; H, 7.33; N, 4.85%; C29H43N2SnCl
requires C 60.70, H 7.55, N 4.88%); dH(C6D6) 7.12–6.97 (m, 6H,
Experimental
General
All solvents were distilled over standard drying agents un-
der nitrogen and were deoxygenated before use. Sn(NMe2)2,41
[MeC(NAr)2]H22 and [tBuC(NAr)2]H29 were prepared according
to literature procedures. Rac-lactide was purchased from the
Sigma Aldrich Chemical Co. and sublimed three times prior to
use. All other chemicals were purchased from the Sigma Aldrich
Chemical Co. and used as received.
NMR spectra were recorded on a Bruker DRX400 instrument.
1H (400 MHz) and 13C (100 MHz) NMR chemical shifts are
quoted in ppm relative to the residual solvent resonances; 119Sn
(186.5 Hz) spectra were referenced to the internal standards of
the machine. Microanalyses were carried out at the University
of North London. GPC chromatograms were recorded using a
Polymer Laboratories LC1220 HPLC pump and a Spark Midas
autosampler connected to two 5 lm columns (300 × 75 mm)
and a Shodex RI-101 differential refractometer. Chromatograms
H
meta, Hpara), 4.06 (sept, 2H, 3JHH = 6.7 Hz, CH(CH3)2), 3.48 (sept,
3
3
2H, JHH = 6.8 Hz, CH(CH3)2), 1.47 (d, 6H, JHH = 6.6 Hz,
3
CH(CH3)2), 1.31 (d, 6H, JHH = 6.9 Hz, CH(CH3)2), 1.26 (d,
3
3
6H, JHH = 6.9 Hz, CH(CH3)2), 1.07 (d, 6H, JHH = 6.8 Hz,
CH(CH3)2), 0.88 (s, 9H, C(CH3)3); dC(C6D6) 180.01 (tBuC(NAr)2),
145.64 (Cortho), 143.53 (Cortho), 140.23 (Cipso), 126.27 (Cpara), 124.71
(Cmeta), 123.08 (Cmeta), 44.67 (C(CH3)3), 29.10 (CHMe2), 29.06
(C(CH3)3), 28.70 (CHMe2), 28.49 (CH(CH3)2), 27.90 (CH(CH3)2),
23.02 (CH(CH3)2), 22.39 (CH(CH3)2); d119Sn(C6D6) +2.97; m/z 573
(M+), 539 (M − Cl).
Synthesis of [tBuC(NAr)2]SnOiPr, 3
A solution of LiOiPr (0.38 g, 5.79 × 10−3 mol) in 20 mL
toluene was added dropwise to a solution of complex 2 (3.02 g,
5.26 × 10−3 mol) in toluene (15 mL) at −78 ◦C. After stirring
This journal is
The Royal Society of Chemistry 2007
Dalton Trans., 2007, 4464–4471 | 4469
©

View Article Online
for 18 h at room temperature, the solvent was removed under
reduced pressure and the product was extracted into pentane
(20 mL). Recrystallisation from a saturated pentane solution at
room temperature gave colourless crystals of [tBuC(NAr)2]SnOiPr,
(1.60 g, 2.68 × 10−3 mol, 51% yield). Crystals suitable for X-ray
diffraction were grown by allowing a saturated pentane solution
to stand at room temperature for several days, (Found: C 64.40, H
8.37, N 4.79%; C32H50N2OSn requires C 64.33, H 8.44, N 4.69%);
44.44 (CMe3), 30.16 (C(CH3)3, 28.95 (CHMe2), 28.81 (CHMe2),
27.86 (CH(CH3)2), 26.47 (CH(CH3)2), 23.76 (CH(CH3)2), 23.41
(CH(CH3)2), 5.26 (Si(CH3)3); d119Sn(C6D6) +50.00; m/z 699 (M+),
539 (M − N(Si(CH3)3)2).
General polymerisation procedure
Toluene (10 mL) was added to an ampoule precharged with rac-LA
(0.4041 g, 2.80 mmol) and 3, 4 or 5 (0.0028 mmol). The reaction
was immediately transferred to an oil-bath preheated to 60 ◦C
and aliquots were sampled with time. The reaction was eventually
quenched with 1 drop of methanol and monomer conversion
was determined by 1H NMR spectroscopy. After the solvent was
removed under reduced pressure the residue was redissolved in
a small volume of chloroform and the polymer was precipitated
from excess cold acidic methanol and dried in vacuo for 18 h. The
molecular weight and PDI were determined by gel permeation
chromatography.
1
3
dH(C6D6) 7.15–7.02 (m, 6H, Hmeta, Hpara), 4.44 (sept, H, JHH
=
6.0 Hz, OCHMe2), 3.96 (br sept, 2H, CHMe2), 3.64 (br sept, 2H,
CHMe2), 1.43 (br d, 6H, CH(CH3)2), 1.37 (br d, 6H, CH(CH3)2),
1.31 (br d, 6H, CH(CH3)2), 1.26 (br d, 6H, CH(CH3)2), 1.25 (d,
6H, 3JHH = 6.0 Hz, OCH(CH3)2), 0.93 (s, 9H, C(CH3)3); dC(C6D6)
177.32 (tBuC(NAr)2), 145.46 (Cortho), 143.23 (Cortho), 141.01 (Cipso),
125.66 (Cpara), 123.98 (Cmeta), 123.17 (Cmeta), 66.09 (OCHMe2),
44.28 (C(CH3)3), 29.61 (OCH(CH3)2), 29.32 (C(CH3)3, 28.93
(CHMe2), 28.56 (CHMe2), 28.01 (CH(CH3)2), 27.66 (CH(CH3)2),
22.79 (CH(CH3)2), 22.38 (CH(CH3)2); d119Sn(C6D6) −28.58; m/z
598 (M+), 538 (M − OCH(CH3)2).
General procedure used for kinetic studies
Synthesis of [tBuC(NAr)2]SnNMe2, 4
All kinetic runs were carried out in a glovebox. Into a poly-
merisation ampoule containing rac-LA (0.4041 g, 2.80 mmol) an
appropriate amount of toluene stock solution of 3 was added
to give a 0.28 M solution of lactide with the desired [LA]0 : 3
ratio. At appropriate time intervals, 0.5 mL aliquots were removed
and quenched with 1 drop of methanol and the conversion was
determined by 1H NMR spectroscopy.
A suspension of LiNMe2 (0.047 g, 0.917 × 10−3 mol) in toluene
(30 mL) was added dropwise to a solution of complex 2 (0.501 g,
0.873 × 10−3 mol) in toluene (30 mL) at −78 ◦C. The reaction was
allowed to stir for 18 h whilst warming to room temperature. The
yellow solution was filtered and the filtrate concentrated under
reduced pressure to give the crude product as a pale yellow solid
in quantitative yield. Recrystallisation from a saturated heptane
solution at −30 ◦C gave 4 as pale yellow crystals (0.291 g, 0.499 ×
10−3 mol, 57% yield). Crystals suitable for X-ray crystallographic
analysis were grown from a pentane solution at room temperature,
(Found: C 63.74, H 8.42, N 7.09%; C31H49N3Sn requires C
Crystallography‡
Crystal data for 1. C52H74N4Sn, M = 873.84, monoclinic,
˚
P2/c (no. 13), a = 21.8742(6), b = 10.9039(3), c = 20.8390(5) A,
◦
3
˚
b = 97.598(2) , V = 4926.8(2) A , Z = 4 (two C2-symmetric
molecules), Dc = 1.178 g cm−3, l(Mo-Ka) = 0.555 mm−1, T =
173 K, colourless prisms; 16 674 independent measured reflections
(Rint = 0.032), F2 refinement, R1 = 0.076 for 15 788 independent,
observed, absorption-corrected reflections [|Fo| > 4r(|Fo|),
2hmax = 65.2◦], wR2 = 0.176 (all data), 517 parameters.
63.93, H 8.48, N 7.21%); dH(C6D6) 7. 12–7.07 (m, 6H, Hmeta
,
3
H
para), 3.84 (br sept, 2H, JHH = 6.7 Hz, CHMe2), 3.71 (br sept,
3
2H, JHH = 6.7 Hz, CHMe2), 3.26 (s, 6H, N(CH3)2), 1.33 (d,
3
3
12H, JHH = 6.9 Hz, CH(CH3)2), 1.29 (d, 6H JHH = 6.9 Hz,
3
CH(CH3)2), 1.27 (d, 6H, JHH = 6.7 Hz, CH(CH3)2), 0.94 (s,
9H, C(CH3)3); dC(C6D6) 177.83 (tBuC(NAr)2), 144.31 (Cortho),
143.99 (Cortho), 141.88 (Cipso), 125.51 (Cpara), 124.26 (Cmeta), 123.47
(Cmeta), 44.21 (CMe3), 43.31 (N(CH3)2), 29.64 (C(CH3)3), 28.90
(CHMe2), 28.56 (CHMe2), 28.21 (CH(CH3)2), 26.19 (CH(CH3)2),
22.86 (CH(CH3)2); d119Sn(C6D6) +18.34; m/z 539 (M − N(CH3)2).
Crystal data for 2. C29H43N2ClSn, M = 573.79, monoclinic,
˚
P21/n, a = 9.5201(4), b = 16.9282(6), c = 17.8114(7) A, b =
◦
3
−3
˚
90.911(2) , V = 2870.09(19) A , Z = 4, Dc = 1.328 g cm ,
l(Mo-Ka) = 1.002 mm−1, T = 150 K, colourless blocks; 6978
independent measured reflections (Rint = 0.017), F2 refinement,
R1 = 0.024 for 6069 independent, observed, absorption-corrected
reflections [|Fo| > 4r(|Fo|), 2hmax = 58.0◦], wR2 = 0.058 (all
data), 309 parameters.
Synthesis of [tBuC(NAr)2]SnN(SiMe3)2, 5
Complex 5 was prepared in an analogous manner to 4 using
0.756 g [tBuC(NAr)2]SnCl (1.31 × 10−3 mol) and 0.231 g
LiN(SiMe3)2 (1.38 × 10−3 mol). Recrystallisation from heptane
afforded yellow crystals of [tBuC(NAr)2]SnN(SiMe3)2 (0.606 g,
0.864 × 10−3 mol, 66% yield), (Found: C 59.81, H 8.60, N 5.71%;
C35H61N3Si2Sn requires C 60.16, H 8.80, N 6.01%); dH(C6D6)
Crystal data for 3. C32H50N2OSn, M = 597.43, orthorhombic,
˚
Pbca, a = 18.3898(5), b = 17.9716(4), c = 19.1257(5) A, V =
3
−3
−1
˚
6320.9(3) A , Z = 8, Dc = 1.256 g cm , l(Mo-Ka) = 0.833 mm ,
T = 150 K, colourless blocks; 7878 independent measured
reflections (Rint = 0.020), F2 refinement, R1 = 0.025 for 6329
independent, observed, absorption-corrected reflections, [|Fo| >
4r(|Fo|), 2hmax = 58.2◦], wR2 = 0.066 (all data), 338 parameters.
3
7.12–7.02 (m, 6H, Hmeta, Hpara), 3.65 (sept, 2H, JHH = 6.8 Hz,
3
CHMe2), 3.60 (sept, 2H, JHH = 6.8 Hz, CHMe2), 1.45 (d,
3
3
¯
6H, JHH = 6.8 Hz, CH(CH3)2), 1.37 (d, 6H, JHH = 6.9 Hz,
CH(CH3)2), 1.31 (d, 6H, 3JHH = 6.8 Hz, CH(CH3)2), 1.27 (d, 6H,
3JHH = 6.8 Hz, CH(CH3)2), 0.96 (s, 9H, C(CH3)3), 0.07 (s, 18H,
Si(CH3)3); dC(C6D6) 171.34 (tBuC(NAr)2), 144.91 (Cortho), 142.28
(Cipso), 142.04 (Cortho), 125.64 (Cpara), 124.32 (Cmeta), 123.79 (Cmeta),
Crystal data for 4. C31H49N3Sn, M = 582.42, triclinic, P1, a =
◦
˚
10.0495(9), b = 11.0334(10), c = 15.5989(14) A, a = 99.131(2) ,
◦
◦
3
˚
b = 107.283(2) , c = 106.213(2) , V = 1529.8(2) A , Z = 2, Dc =
1.264 g cm−3, l(Mo-Ka) = 0.857 mm−1, T = 150 K, pale yellow
needles; 6970 independent measured reflections (Rint = 0.021),
4470 | Dalton Trans., 2007, 4464–4471
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
R1 = 0.028 for 6286 independent, observed, absorption-corrected
reflections [|Fo| > 4r(|Fo|), 2hmax = 57.7◦], wR2 = 0.070 (all
data), 329 parameters.
13 M. Cheng, A. B. Attygalle, E. B. Lobkovsky and G. W. Coates, J. Am.
Chem. Soc., 1999, 121, 11583.
14 B. M. Chamberlain, M. Cheng, D. R. Moore, T. M. Ovitt, E. B.
Lobkovsky and G. W. Coates, J. Am. Chem. Soc., 2001, 123, 3229.
15 M. H. Chisholm, J. C. Huffman and K. Phomphrai, J. Chem. Soc.,
Dalton Trans., 2001, 222.
Crystal data were measured on Oxford Diffraction Xcalibur
3 (1) and Bruker AXS SMART 1000 CCD (2, 3 and 4) diffrac-
16 M. H. Chisholm, J. Gallucci and K. Phomphrai, Inorg. Chem., 2002,
41, 2785.
˚
tometers with Mo-Ka (k = 0.71073 A) radiation, using x-scans
with narrow frames. Lp and absorption corrections were applied
based on symmetry equivalent and repeated measurements. All
structures were solved by direct methods with non-H atoms
refined anisotropically and H atoms included in a riding model.
Anisotropic displacement parameter restraints were applied to iPr
and tBu C atoms in 3.
17 M. H. Chisholm, J. Gallucci and K. Phomphrai, Chem. Commun., 2003,
48.
18 M. H. Chisholm, J. Gallucci and K. Phomphrai, Inorg. Chem., 2004,
43, 6717.
19 V. C. Gibson, E. L. Marshall, D. Navarro-Llobet, A. J. P. White and
D. J. Williams, Dalton Trans., 2003, 4321.
20 J. Barker and M. Kilner, Coord. Chem. Rev., 1994, 133, 219.
21 F. T. Edelmann, Coord. Chem. Rev., 1994, 137, 403.
22 R. T. Boere´, V. Klassen and G. Wolmersha¨usen, J. Chem. Soc., Dalton
Trans., 1998, 4147.
23 U. Kilimann, M. Noltemeyer and F. T. Edelmann, J. Organomet. Chem.,
1993, 443, 35.
24 Y. Zhou and D. S. Richeson, J. Am. Chem. Soc., 1996, 118, 10850.
25 P. B. Hitchcock, M. F. Lappert and M. Layh, J. Chem. Soc., Dalton
Trans., 1998, 3113.
26 S. R. Foley, Y. Zhou, G. P. A. Yap and D. S. Richeson, Inorg. Chem.,
2000, 39, 924.
Programs: Oxford Diffraction CrysAlis CCD42 and Bruker
SMART43 (diffractometer control), Oxford Diffraction CrysAlis
RED42 and SAINT43 (data reduction), SHELXTL44 (solution and
refinement) and local programs.
CCDC reference numbers 645832–645835.
For crystallographic data in CIF or other electronic format see
DOI: 10.1039/b706663e.
27 K. B. Aubrecht, M. A. Hillmyer and W. B. Tolman, Macromolecules,
2002, 35, 644.
28 F. Antolini, P. B. Hitchcock, A. V. Khvostov and M. F. Lappert,
Can. J. Chem., 2006, 84, 269.
29 A. Xia, H. M. El-Kaderi, M. J. Heeg and C. H. Winter, J. Organomet.
Chem., 2003, 682, 224.
30 M. P. Coles, D. C. Swenson, R. F. Jordan and V. G. Young,
Organometallics, 1997, 16, 5183.
Acknowledgements
The authors express their gratitude to the Royal Thai Government
for astudentship(N. N.)and the Engineering and Physical Sciences
Research Council (UK) for a studentship (S. H. D.).
31 S. P. Green, C. Jones, P. C. Junk, K.-A. Lippert and A. Stasch, Chem.
Commun., 2006, 3978.
32 S. R. Foley, G. P. A. Yap and D. S. Richeson, J. Chem. Soc., Dalton
Trans., 2000, 1663.
33 Y. Zhou and D. S. Richeson, Inorg. Chem., 1997, 36, 501.
34 Y. Q. Ding, H. W. Roesky, M. Noltemeyer, H. G. Schmidt and P. P.
Power, Organometallics, 2001, 20, 1190.
References
1 A. H. Tullo, Chem. Eng. News, 2002, 80, 13.
2 A. J. Nijenhuis, D. W. Grijpma and A. J. Pennings, Macromolecules,
1992, 25, 6419.
3 B. J. O’Keefe, M. A. Hillmyer and W. B. Tolman, J. Chem. Soc., Dalton
Trans., 2001, 2215.
35 S. Penczek, A. Duda and R. Szymanski, Macromol. Symp., 1998, 132,
441.
4 O. Dechy-Cabaret, B. Martin-Vaca and D. Bourissou, Chem. Rev., 2004,
104, 6147.
5 V. C. Gibson, and E. L. Marshall, in Comprehensive Coordination
Chemistry II, ed. J. A. McCleverty and T. J. Meyer, Elsevier, Oxford,
UK, 2003, vol. 9, ch. 1, pp. 36–52.
6 G. Schwach, J. Coudane, R. Engel and M. Vert, J. Polym. Sci., Part A:
Polym. Chem., 1997, 35, 3431.
7 H. R. Kricheldorf, Macromol. Symp., 2000, 153, 55.
8 H. R. Kricheldorf, I. Kreiser-Saunders and C. Boettcher, Polymer,
1995, 36, 1253.
9 A. Kowalski, A. Duda and S. Penczek, Macromolecules, 2000, 33,
689.
36 T. Ouhadi, A. Hamitou, R. Je´roˆme and P. Teyssie´, Macromolecules,
1976, 9, 927.
37 P. Dubois, C. Jacobs, R. Je´roˆme and P. Teyssie´, Macromolecules, 1991,
24, 2266.
38 B. M. Chamberlain, B. A. Jazdzewski, M. Pink, M. A. Hillmyer and
W. B. Tolman, Macromolecules, 2000, 33, 3970.
39 H. Ma and J. Okuda, Macromolecules, 2005, 38, 2665.
40 H. R. Kricheldorf, C. Boettcher and K. U. To¨nnes, Polymer, 1992, 33,
2817.
41 P. Foley and M. Zeldin, Inorg. Chem., 1975, 14, 2264.
42 CrysAlis CCD and RED software, Oxford Diffraction Ltd., Abingdon,
Oxon, UK, 2004.
10 H. R. Kricheldorf, M. Berl and N. Scharnagl, Macromolecules, 1988,
21, 286.
11 A. P. Dove, V. C. Gibson, E. L. Marshall, A. J. P. White and D. J.
Williams, Chem. Commun., 2001, 283.
43 SMART and SAINT software for CCD diffractometers, Bruker AXS
Inc., Madison, WI, USA, 2001.
44 G. M. Sheldrick, SHELXTL user manual, version 6.10, Bruker AXS
Inc., Madison, WI, USA, 2000.
12 A. P. Dove, V. C. Gibson, E. L. Marshall, H. S. Rzepa, A. J. P. White
and D. J. Williams, J. Am. Chem. Soc., 2006, 128, 9834.
This journal is
The Royal Society of Chemistry 2007
Dalton Trans., 2007, 4464–4471 | 4471
©