A well defined tin(II) initiator for the living polymerisation of lactide

Article abstract of DOI:10.1039/b008770j

[HC{C(Me)NAr}₂]Sn(OPrⁱ) (Ar = 2,6-Prⁱ₂C₆H₃) is shown to be an initiator for the living polymerisation of rac-lactide to heterotactic-enriched poly(lactide).

Full text of DOI:10.1039/b008770j

A well defined tin(II) initiator for the living polymerisation of lactide
Andrew P. Dove, Vernon C. Gibson,* Edward L. Marshall, Andrew J. P. White and David J. Williams
Department of Chemistry, Imperial College of Science, Technology and Medicine, Exhibition Road, South
Kensington, London, UK SW7 2AY. E-mail: v.gibson@ic.ac.uk
Received (in Cambridge, UK) 31st October 2000, Accepted 14th December 2000
First published as an Advance Article on the web
[HC{C(Me)NAr}2]Sn(OPri) (Ar = 2,6-Pri2C6H3) is shown to
be an initiator for the living polymerisation of rac-lactide to
heterotactic-enriched poly(lactide).
There is increasing interest in the design of polymers for in vivo
applications such as sutures, artificial tissue networks and drug
delivery agents. For these applications it is desirable for the
polymer to be non-toxic, biocompatible and resorbable. One of
the most promising classes of polymer in this field is the
poly(lactide)s1 which are readily obtained in a controlled
manner via the living ring-opening polymerisation of lactide,
the cyclic diester of lactic acid.2
Initiators for lactide polymerisation are typically based on
alkoxide or alkanoate complexes of metals such as Al, Mg, Sn,
Zn and the rare earths. Tin(II) catalysts such as Sn(ethyl
hexanoate)2 are generally preferred in the commercial produc-
tion of poly(lactide)s3 for medical or pharmaceutical applica-
tions owing to the low toxicity of Sn(II) compared to other metal
ions [Sn(ethyl hexanoate)2 is a permitted food additive in many
countries]. Although a number of systems based on aluminium,4
and more recently, magnesium5 and zinc,6 have been described
that initiate the living ring-opening polymerisation of lactide,
there have to date been no reports of well defined tin initiators.
Here we describe the first example of a single-site tin catalyst
for the living polymerisation of lactide.
The initiator 2 is prepared according to Scheme 1.†
Treatment of SnCl2 with [HC{C(Me)NAr}2]Li in diethyl ether
followed by crystallisation from pentane affords [HC{C(Me)-
NAr}2]SnCl 1 as a yellow crystalline solid. This is converted to
2 by treatment with LiOPri followed by recrystallisation from
pentane. Crystals of 2 suitable for an X-ray structure determina-
tion‡ were grown from pentane; the structure is shown in
Fig. 1.
The complex has non-crystallographic Cs symmetry with the
Fig. 1 (a) The molecular structure of 2. Selected bond lengths (Å) and angles
(°); Sn–O 2.000(5), Sn–N(1) 2.206(4), Sn–N(3) 2.208(4), C(1)–N(1)
1.323(6), C(1)–C(2) 1.404(7), C(2)–C(3) 1.387(7), C(3)–N(3) 1.331(6); O–
Sn–N(1) 94.1(2), O–Sn–N(3) 92.5(2), N(1)–Sn–N(3) 83.6(2). (b) Space
filling representation of the structure of 2 showing the exposed environment
of the tin atom.
three-coordinate tin atom adopting a tripodal geometry with
inter-bond angles in the range 83.6(2)–94.1(2)°, the most acute
being associated with the bite of the chelating N,NA ligand. The
Sn–N and Sn–O distances are unexceptional, and there is the
expected pattern of delocalisation within the b-diketiminate
ligand. The six-membered chelate ring has a boat conformation
with C(2) and Sn lying 0.12 and 0.87 Å ‘above’ the N(1), C(1),
C(3), N(3) plane; the isopropoxide oxygen atom lies 0.74 Å
‘below’ this plane. As a consequence of the folded chelate ring
conformation, and retention of near trigonal planar geometries
at the two nitrogen centres, the C(12) and C(27) isopropyl
groups are drawn together, and those associated with C(15) and
C(24) are folded away exposing the non-coordinated ‘face’
containing the stereochemically active lone pair on the tin atom
[Fig. 1(b)]; the shortest intermolecular approach to the tin centre
is 3.84 Å from C(13)–H.
The polymerisation of rac-lactide (a racemic mixture of
(S,S) and (R,R) lactides) was initially investigated using
L
D
complex 2 in CH2Cl2 at ambient temperature.§ Under these
conditions it was found that 100 equivalents of monomer
1
required 96 h for complete conversion ( > 99% by H NMR).
The resultant polymer has a molecular weight close to that
calculated from the monomer+initiator ratio (observed Mn
=
17 100; calculated Mn = 14 400) and exhibits a narrow
polydispersity (Mw/Mn = 1.11), characteristics of a living
process. The polymerisation was then repeated at 60 °C in
Scheme 1 Reagents and conditions: i, SnCl2, Et2O, 18 h, recrystallisation
from pentane, 60% yield (unoptimised); ii, LiOPri, Et2O, 18 h, recrystallisa-
tion from pentane, 17% yield (unoptimised).
DOI: 10.1039/b008770j
Chem. Commun., 2001, 283–284
This journal is © The Royal Society of Chemistry 2001
283

tions are consistent with a heterotactic-biased product since the
rmr microstructure can only arise from two consecutive or
interchanges; each rmr tetrad is accompanied by two mrm
D L
–
L–D
tetrads in agreement with the NMR integration. The preference
for heterotacticity is not as strong as reported previously for the
zinc analogue of 2, but nonetheless represents the first example
of tacticity bias arising from the polymerisation of rac-lactide
using a tin catalyst.
Further studies are examining the effect of changes to the aryl
ligand substituents on the polymerisation behaviour of the tin
catalysts and their use in the ring-opening polymerisation of
other cyclic ester monomers.
The Engineering and Physical Sciences Research Council is
thanked for a studentship (to A. P. D.) and a postdoctoral
fellowship (to E. L. M.).
Fig. 2
toluene affording 85% conversion after 4 h, again resulting in a
narrow molecular weight distribution product (Mw/Mn = 1.05).
The activity of the tin catalyst is lower than that observed for the
related zinc system6 which may be in part due to the lower
electrophilicity of the tin centre, and partly a consequence of the
stereochemically active lone pair which may disfavour mono-
mer binding. The living characteristics of the polymerisation are
supported by the linear increase in Mn with conversion giving in
each case a low polydispersity product (Fig. 2).
Notes and references
† Selected spectroscopic data: for 1: dH(250 MHz, C6D6, 25 °C) 1.06 (d,
6H, 3JHH 6.8 Hz, CHMeMe), 1.18 (d, 6H, 3JHH 6.9 Hz, CHMeMe), 1.22 (d,
6H, 3JHH 6.8 Hz, CHMeAMeA), 1.45 (d, 6H, 3JHH 6.6 Hz, CHMeAMeA), 1.61
(s, 6H, HC{C(Me)NAr}2), 3.12 (sept, 2H, 3JHH 6.8 Hz, CHMe2), 3.95 (sept,
3
2H, JHH 6.8 Hz, CHMe2), 5.06 (s, 1H, HC{C(Me)NAr}2), 7.15 (m, 6H,
Haryl). MS: m/z 572 [M]+. Anal. Calc. (found) for C29H41ClN2Sn: C, 60.91
(60.77); H, 7.22 (7.32); N, 4.90 (5.07)%. For 2: dH(250 MHz, C6D6, 25 °C)
d 0.90 (d, 6H, 3JHH 6.8 Hz, OCHMe2), 1.14 (d, 3H, 3JHH 6.8 Hz, CHMeMe),
1.24 (d, 3H, 3JHH 6.9 Hz, CHMeMe), 1.27 (d, 3H, CHMeAMeA), 1.54 (d, 3H,
In order to confirm that the initiator is indeed the isoprop-
3
CHMeAMeA), 1.59 (s, 6H, HC{C(Me)NAr}2), 3.25 (sept, 2H, JHH 6.8 Hz,
1
oxide complex 2, a H NMR study was carried out in which
CHMe2), 3.86 (sept, 2H, 3JHH 6.8 Hz, CHMe2), 4.15 (sept, 1H, 3JHH 6.0 Hz,
OCHMe2), 4.73 (s, 1H, HC{C(Me)NAr}2), 7.16 (m, 6H, Haryl). Anal. Calc.
(found) for C32H48N2OSn: C, 64.55 (64.28); H, 8.13 (8.08); N, 4.70
(4.90)%.
increasing amounts of lactide were added to 2 in CDCl3. The
isopropoxide methine septet resonance at the end of the
propagating chain is shifted slightly to higher frequency
(0.005 ppm) relative to the unconsumed initiator (d 4.01). New
resonances for the b-diketiminate ligand substituents are also
observed for the propagating species. As the number of
monomer equivalents is increased, the intensities of the signals
attributable to the propagating species increase relative to those
of unconsumed 2. Owing to the overlapping nature of the
resonances, accurate integration has not been possible, but the
addition of 5 equivalents of monomer leads to an approximately
1+1 mixture of initiator+propagating species. This indicates a
favourable kp/ki ratio (the rate constant of propagation to rate
constant of initiation) which is desirable for minimising the
polydispersity.
‡ Crystal data for 2: C32H48N2OSn, M 595.4, monoclinic, space group
P21/n (no. 14), a = 13.205(2), b = 16.680(2), c = 15.527(2) Å, b =
107.42(1)°, V = 3263.1(6) Å3, Z = 4, Dc = 1.212 g cm23, m(Mo-Ka) =
8.07 cm21, T = 293 K, yellow blocks; 5742 independent measured
reflections, F2 refinement, R1 = 0.049, wR2 = 0.112, 4038 independent
observed reflections [|Fo| > 4s(|Fo|), 2q @ 50°], 326 parameters. CCDC
151597. See http://www.rsc.org/suppdata/cc/b0/b008770j/ for crystallo-
graphic files in .cif or other electronic format.
§ Typical polymerisation procedure: complex 2 (0.005 g, 0.008 mmol) and
rac-lactide (0.1181 g, 0.819 mmol) were weighed in to a 15 cm3 glass
ampoule fitted with a Teflon stopcock. The mixture was suspended in
toluene (6 cm3) and the ampoule was then sealed and transferred to an oil
bath pre-heated to 60 °C. After stirring for the allotted period of time the
volatile components were removed in vacuo. Conversion was determined
by integration of monomer vs. polymer methine resonances in the 1H NMR
spectrum of the crude product (in CDCl3). The polymer was purified by re-
dissolving in CH2Cl2 (5 cm3) and precipitating from rapidly stirring
methanol. GPC chromatograms were recorded on a Knauer differential
refractometer connected to a Gynotek HPLC pump (model 300) and two 10
mm columns (PSS) at a flow rate of 1.00 cm3 min21 using CHCl3 as the
eluent. The columns were calibrated against polystyrene standards with
molecular weights ranging from 1560 to 128 000. Samples were filtered
through a 0.45 mm filter immediately prior to injection. Analysis was
performed using Version 3.0 of the Conventional Calibration module of the
Viscotek SEC3 software package.
1
The H NMR spectrum of the poly(lactide) derived from 2
(Fig. 3) differs from the spectrum predicted from a Bernouillian
analysis of totally random poly(rac-lactide), with the rmr and
mrm tetrads much more intense than expected. These observa-
1 J. C. Middleton and A. J. Tipton, Biomaterials, 2000, 21, 2335.
2 H. R. Kricheldorf and I. Kreiser-Saunders, Macromol. Symp., 1996, 103,
85; D. K. Gilding and A. M. Reed, Polymer, 1979, 20, 1459.
3 E. E. Schmitt and R. A. Rohistina, US Pat., 3297033, 1967 (Chem. Abstr.,
1967, 66, P38656u); E. E. Schmitt and R. A. Rohistina, US Pat., 3463158,
1969 (Chem. Abstr., 1969, 71, P92382t); H. R. Kricheldorf, I. Kreiser-
Saunders and C. Boettcher, Polymer, 1995, 36, 1253 and references
therein.
4 P. Dubois, C. Jacobs, R. Jérome and P. Teyssié, Macromolecules, 1991,
24, 2266; P. A. Cameron, D. Jhurry, V. C. Gibson, A. J. P. White, D. J.
Williams and S. Williams, Macromol. Rapid. Commun., 1999, 20, 616;
N. Spassky, M. Wisniewski, C. Pluta and A. LeBorgne, Macromol.
Chem. Phys., 1996, 197, 2627.
5 M. H. Chisholm and N. W. Eilerts, Chem. Commun., 1996, 853.
6 M. Cheng, A. B. Attygalle, E. B. Lobkovsky and G. W. Coates, J. Am.
Chem. Soc., 1999, 121, 11 583.
Fig. 3
284
Chem. Commun., 2001, 283–284