Bidentate salicylaldiminato tin(ii) complexes and their use as lactide polymerisation initiators

Article abstract of DOI:10.1039/b812877d

A study of the reactions of several salicylaldimines (ortho-iminophenols) with Sn(NMe₂)₂ reveals that the sterics and electronics of the N-substituent play a principal role in determining the product mixture. Thus mono(chelate) tin(i

Full text of DOI:10.1039/b812877d

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/dalton | Dalton Transactions
Bidentate salicylaldiminato tin(II) complexes and their use as lactide
polymerisation initiators†
Nonsee Nimitsiriwat,a Vernon C. Gibson,*a Edward L. Marshall*a and Mark R. J. Elsegoodb
Received 28th July 2008, Accepted 14th November 2008
First published as an Advance Article on the web 17th March 2009
DOI: 10.1039/b812877d
A study of the reactions of several salicylaldimines (ortho-iminophenols) with Sn(NMe2)2 reveals that
the sterics and electronics of the N-substituent play a principal role in determining the product mixture.
Thus mono(chelate) tin(II) amides have only been isolated with bulky N-anilido substituents, including
i
=
=
[2,4-X2–6-{CH N-2,6- Pr2C6H3}-C6H2O]Sn(m-NMe2)]2, X = Cl, 1; X = I, 2 and [2,4-Cl2–6-{CH N-2,
4,6-tBu3C6H2}-C6H2O]Sn(NMe2), 3. With smaller N-aryl and N-alkyl substituted salicylaldimines,
mixtures of mono- and bis- chelate complexes are formed, even if the phenoxide ring bears large
tert-butyl substituents. Further, when the anilido group bears electron-withdrawing substituents, the
imino carbon is activated towards nucleophilic attack (as demonstrated by the formation of
[2,4-tBu2–6-{CH(NMe2)N-2,4,6-Br3C6H2}-C6H2O]Sn, 4); no such reactivity has been observed when
the halo substituents are located on the phenolic ring. The abilities of complexes 1–4 to initiate the
ring-opening polymerisation of rac-lactide have also been studied. Complexes 1, 2 and 4 possess
comparitively similar activities, but propagation with 3 is at least one order of magnitude slower, an
observation rationalised in terms of steric congestion.
Introduction
Several single-site tin(II)-based initiators for the well-controlled
polymerisation of lactide, LA, have been reported.1–6 Initiation
via nucleophilic attack at the carbonyl carbon may be performed
by either alkoxide (LSnOR) or amido (LSnNR2) complexes7–10
and although chain termini differ (ester versus amide), subsequent
propagation steps proceed in an identical manner. We have
therefore found Sn(NMe2)2 to be a particularly useful precursor:
this compound may be readily isolated as crystalline blocks from
the reaction of SnCl2 with LiNMe2,11 and usually reacts in an
effectively irreversible and highly predictable manner (Scheme 1)
with acidic pro-ligands including b-diketimines4 and amidines.5
Scheme 2
We describe here a related study into the divalent tin coordi-
nation chemistry of bidentate salicylaldimines, with the aim of
producing less congested tin(II) coordination spheres with high
activities for LA polymerisation. Our original decision to study
tridentate analogues was based upon the known tendency of
even sterically bulky bidentate iminophenols to form bischelate
products with a range of divalent metals,12–17 but we suspected that
Sn(II) might prove an exception to this behaviour given the usual
stereochemical activity of its lone 5s2 electron pair. Here, we shall
show that: (i) if steric conditions are met, then this supposition
is largely correct; and (ii) the amine migration reaction shown
in Scheme 2 is not restricted to the tridentate salicylaldiminato
family.
Scheme 1
An exception to this behaviour is the reaction of Sn(NMe2)2 with
potentially tridentate salicylaldiminines (ortho-iminophenols with
Lewis basic N-substituents): following deprotonation of the phe-
nol, the remnant NMe2 ligand attacks the imino carbon to generate
a tetradentate, dianionic aminoamidophenoxide ligand.3,6 These
species may still be used to initiate the polymerisation of lactide
as they exist in equilibria with their terminal amide tautomers
(Scheme 2).
Results and discussion
Synthetic and crystallographic studies
As summarised in Scheme 3, ten salicylaldimines were chosen for
this study and these are abbreviated to [RONR¢]H, where R and
R¢ are the 2,4-phenoxy- and imino substituents respectively. In a
series of NMR scale experiments each of the salicylaldimines was
reacted with ca. 10 mg of Sn(NMe2)2 in C6D6. Despite our initial
hopes of circumventing bischelation, the reactions of [tBuONiPr]H,
aImperial College London, Department of Chemistry, South Kensington
Campus, London, SW7 2AZ, United Kingdom. E-mail: v.gibson@imperial.
ac.uk, e.marshall@imperial.ac.uk
bLoughborough University, Chemistry Department, Leicestershire, LE11
3TU, United Kingdom
† Electronic supplementary information (ESI) available: 1H NMR spectra
for complexes 2 and 3; crystallographic data for complexes 2 and 4. CCDC
reference numbers 696308 and 696309. For ESI and crystallographic data
in CIF or other electronic format see DOI: 10.1039/b812877d
[
tBuONtBu]H, [tBuONBn] and [MeONDIPP]H all led to mixtures
of the mono(chelate) tin(II) dimethylamides, [RONR¢]Sn(NMe2),
3710 | Dalton Trans., 2009, 3710–3715
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
1H NMR signals in the room temperature spectra of 1 and 2
is observed consistent with hindered rotation of the aryl rings
due to their proximity to the bridging NMe2 ligands. By contrast,
1
all resonances in the H NMR spectrum of 3 are sharp (see ESI
Fig. S1–S3†). An alternative argument may be advanced which is
also consistent with the spectroscopic solution data, namely that
3 also adopts a dimeric structure with bridging amides, but the
1
salicylaldiminato ligands bind in an h manner. However, lactide
polymerisation activities (vide infra) suggest this is not the case.
Further support for our postulate is provided from a single
crystal X-ray diffraction study on 2, which reveals that the amide-
bridged dimeric structure shown in Scheme 4 is adopted in the
solid state, with the two halves of the molecule related by a
centre of inversion (Fig. 1). The four-membered Sn2N2 ring is
planar, with the amide ligands bridging in an asymmetric fashion
˚
˚
(Sn(1)–N(2A) = 2.326(2) A vs. Sn(1)–N(2) = 2.192(2) A, Table 1).
One of the more remarkable features of the molecular structure is
the particularly weak interaction between the metal and the imino
Scheme 3 DIPP = 2,6-di-isopropylphenyl; DMP = 2,6-dimethylphenyl;
TBP = 2,4,6-tribromophenyl; TTBP = 2,4,6-tris(tert-butyl)phenyl.
˚
nitrogen (Sn(1) ◊ ◊ ◊ N(1) = 2.698 A), which is distinctly longer than
a typical dative covalent Sn–N bond, though still appreciably
shorter than the sum of the respective van der Waals radii
contaminated to varying degrees with the bis(chelate) species,
[RONR¢]2Sn. However, bis(chelate) species were not observed in
any other case.
Thus, the 2,4-dihalophenols, [ClONDIPP]H, [IONDIPP]H and
[ClONTTBP]H, were cleanly transformed into their mono(chelate)
tin(II) amides and when repeated on a larger scale, [ClONDIPP]-
Sn(NMe2), 1, [IONDIPP]Sn(NMe2), 2, and [ClONTTBP]Sn(NMe2),
3, were all successfully recrystallised from toluene in yields of
ca. 65, 45 and 60% respectively. The two di-tert-butylphenol sal-
icylaldimines were also converted into [tBuONDMP]Sn(NMe2) and
19
˚
(2.17 (Sn) + 1.55 (N) = 3.72 A). The long nature of this inter-
action is attributed in part to electronic repulsive forces arising
from the tin 5s2 lone pair (previously suggested as the origin of
similarly lengthy Sn–O interactions4).
[
tBuONDIPP]Sn(NMe2); however, when synthesised on a preparative
scale, the high solubility of both complexes in a range of solvents
precluded purification (via washing or recrystallisation) in our
hands.
The 1H and 13C NMR spectra of 1–3 are consistent with
mono(iminophenoxide) mono(amide) formulations with no sug-
gestion of the migration of the NMe2 unit to the imino carbon.
However, while the 119Sn NMR spectra of 1 and 2 feature
singlets consistent with four-coordinate centres at d -634 and
-639 respectively, the metal nuclei in complex 3 resonate at d
-108, indicative of a lower coordination number.18 We therefore
propose that both 1 and 2 exist as amide-bridged dimers in solution
whereas the 2,4,6-tri-tert-butyl anilino component of 3 occupies
sufficient space to confer mononuclearity (Scheme 4). Consistent
with this hypothesis, broadening of the 2,6-di-isopropylphenyl
Fig. 1 The molecular structure of complex 2. Symmetry operator A =
1 - x, 1 - y, 1 - z.
The reaction of Sn(NMe2)2 with the tenth salicylaldimine,
[
tBuONTBP]H, also proceeds cleanly and again, bischelation is not
observed. In this case, the 1H NMR spectrum of the product (4)
is noticeably different from those of complexes 1–3, and is instead
consistent with migration of the NMe2 ligand to the imino carbon
(Scheme 5), forming a tridentate dianionic aminoamidophenoxide
ligand in a fashion analogous to that reported previously with
tridentate salicylaldimines.4 The 1H NMR spectrum of 4 includes
a methine singlet at d 6.62 (cf. 7.49, 8.85 and 7.37 ppm for the
Scheme 4 Postulated solution state structures of 1–3.
This journal is
The Royal Society of Chemistry 2009
Dalton Trans., 2009, 3710–3715 | 3711
©

View Article Online
◦
◦
˚
˚
Table 2 Selected bond lengths (A) and bond angles ( ) for complex 4
Table 1 Selected bond lengths (A) and bond angles ( ) for complex 2
Sn(1)–N(2)
Sn(1)–O(1)
O(1)–C(1)
C(6)–C(7)
2.192(2)
2.103(2)
1.314(3)
1.466(4)
Sn(1)–N(2A)
Sn(1)–N(1)
C(1)–C(6)
2.326(2)
2.698(2)
1.425(4)
1.277(4)
Sn(1)–O(1)
Sn(1)–N(2)
C(7)–N(1)
2.073(2)
2.354(2)
1.481(4)
Sn(1)–N(1)
Sn(1)–Br(1)
C(7)–N(2)
2.153(2)
3.1469(6)
1.500(4)
C(7)–N(1)
O(1)–Sn(1)–N(1)
N(1)–Sn(1)–N(2)
Br(1) ◊ ◊ ◊ Sn(1)–O(1)
92.28(8)
60.91(9)
81.01
O(1)–Sn(1)–N(2)
N(1)–C(7)–N(2)
Br(1) ◊ ◊ ◊ Sn(1)–N(1)
80.63(8)
100.4(2)
69.56
Sn(1)–N(2)–Sn(1A)
N(2)–Sn(1)–O(1)
N(1)–Sn(1)–N(2)
N(1)–Sn(1)–O(1)
99.25(9)
87.77(9)
92.43(8)
74.40(7)
N(2)–Sn(1)–N(2A)
N(2A)–Sn(1)–O(1)
N(1)–Sn(1)–N(2A)
80.75(9)
90.11(8)
163.37(8)
Scheme 5
CH N resonances of 1–3 respectively), and a corresponding 13C
NMR chemical shift which is at odds with an unsaturated imino
carbon environment (d 88.44 vs. d 165.54, 157.68 and 165.43 for
1–3). In addition, two singlets at 2.34 and 1.84 ppm are attributed
to inequivalent N-methyl groups, indicating that the Me2N amino
donor is strongly bound to the tin centre.4
The molecular structure of complex 4 is shown in Fig. 2 and
Fig. 3. The N,N,O chelate adopts a tripodal geometry with the
coordination sphere of the tin completed by weak inter- and
intra- molecular interactions with the N-aryl ortho-bromine atoms
=
Fig. 3 Weak intermolecular interactions in the molecular structure of
˚
complex 4. Sn(1) ◊ ◊ ◊ Br(1A) = 3.850 A. Symmetry operator A = 1 - x,
2 - y, 1 - z.
atoms approach 90◦. As observed with several other divalent
tin complexes, this bonding picture is consistent with the metal
utilising orthogonally-aligned 4d and 5p orbitals to engage the
ligand (on the basis of previous computational studies,4 we assume
that a relativistic contraction greatly precludes involvement of the
5s orbital).
˚
˚
(Sn(1) ◊ ◊ ◊ Br(1) = 3.1469(6) A; Sn(1)–Br(1A) = 3.850 A). The
conversion of the salicylaldimine nitrogen into an anionic amide
˚
ligand results in a Sn(1)–N(1) distance (2.153(2) A) which is
notably shorter than the neutral metal–amino interaction, Sn(1)–
˚
˚
N(2) (2.354(2) A, Table 2). The tin atom lies 1.307 A out of
the N(1)–C(7)–N(2) plane, in the direction of the phenoxide
Polymerisation studies
˚
oxygen, and lies 1.439 A above the O(1)–N(1)–N(2) plane. With
The three tin(II) amides 1–3, and the amide tautomer 4, have all
been investigated as possible rac-lactide polymerisation initiators
in toluene at 60 ◦C ([LA]0 = 0.28 M; [LA]0 : [Sn] = 100; Table 3).
The most reactive initiator, 1 promotes > 90% conversion in just
1 h but deviates slightly from ideal behaviour, resulting in a rather
broad molecular weight distribution (ca 1.4). This most likely
arises from transesterification (confirmed by an rrr tetrad CHMe
13C resonance20) in line with the increased Lewis acidity of the
metal in this complex; similar levels of molecular weight control
have previously been reported with halo-substituted diketiminate-
supported counterparts.4 The polymerisations mediated by 2 and
4 are only slightly slower yet both are well-controlled chain growth
processes, with linear relationships recorded between Mn and
monomer conversion, molecular weights close to those calculated
from the monomer : initiator stoichiometries, and relatively
narrow polydispersities. A moderate heteroselectivity is observed
with all four initiators, as reported for a number of tin(II)-based
initiator families,1–6 lending further credence to a lone pair assisted
chain end mechanism.4
the exception of the N(1)–Sn(1)–N(2) chelate [60.91(9)◦] many
of the angles at the metal formed between the ligand donor
Despite the comparable activities of complexes 1, 2 and 4
(Fig. 4), the mononuclear initiator 3 is far slower. No chain growth
is observed during the first few hours and thereafter, propagation
Fig. 2 The molecular structure of complex 4.
3712 | Dalton Trans., 2009, 3710–3715
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
Table 3 rac-LA polymerisation dataa
d
d
Initiator
Time/minb
Conversionc
Induction period/min
k
app/h-1
Mn
Mw/Mn
Pre
1
2
3
4
60
100
1440
100
92
92
94
93
—
—
277
—
2.56
1.55
0.14
1.69
15 400
20 700
28 100
21 500
1.37
1.19
1.41
1.19
0.65
0.65
0.64
0.62
a Toluene, 60 ◦C, [LA]0 = 0.28, [LA]/[I] = 100. b Time required for polymerisation to attain >90% LA consumption. c Determined by 1H NMR (CDCl3)
by integration of monomer : polymer methine resonances. d Determined by GPC (CDCl3). e Pr is the probability of racemic (heterotactic) enchainment,
determined by homonuclear decoupled 1H NMR (CDCl3).
Experimental
All solvents were dried by prolonged reflux over appropriate
drying agents, except toluene, which was dried by passing through
a column of Cu2O/Al2O3 under a stream of nitrogen. All
solvents were degassed thoroughly prior to use. Sn(NMe2)2 was
prepared by an existing literature procedure.10 The salicylaldimines
were synthesised by the formic acid-catalysed condensation of
amines or anilines with commercially available salicylaldimines
in ethanol. Lactide (Aldrich) was purified by triple sublimation
under vacuum.
NMR spectra were recorded on a Bruker AM500 or DRX400
instrument. Microanalyses were carried out at the University of
North London. GPC measurements were performed in HPLC
Fig. 4 Plots of ln{[LA]0/[LA]t} versus time (min) for the polymerisation
of rac-LA with initiators 1–4 (toluene, 60 ◦C, [LA]0 = 0.28 M, [LA]0/[I]0 =
100); ꢀ = 1; ꢀ = 2; ᭜ = 3; ᭹ = 4.
grade CHCl3 at 1.0 ml min-1 using a Polymer Laboratories
LC1220 HPLC pump and a Spark Midas autosampler connected
to two 5 mm columns (300 ¥ 7.5 mm) and a Shodex RI-101
differential refractometer. The columns were calibrated with
polystyrene standards ranging in molecular weight from 1560
to 128 000 Da and chromatograms were analysed using Cirrus
software (Polymer Laboratories) in order to establish molecular
weights and polydispersities. Samples were passed through a
0.45 mm filter prior to injection.
proceeds at a rate which is at least one order of magnitude
slower than for the other three complexes. Even though one might
anticipate 3 to possess the most accessible active site, the relative
reactivity of the four initiators is readily explained by comparing
the nature of their propagating species. Examination of the solid
state structure of 2 indicates that even in the dimeric form, the
tin(II) centres are quite exposed, and we therefore propose that 1
and 2 rapidly insert one equivalent of lactide to give mononuclear
four-coordinate active sites (Scheme 6). These would be analogous
to the propagating species generated from 3, but would be rather
less bulky and hence more reactive. A similar argument justifies
the potency of the latent amide initiator 4: once initiation has
occurred the propagating species is akin to those generated from
1–3 and the rate of propagation is therefore susceptible to similar
steric and electronic effects. The large induction period witnessed
with 3 has not been probed further, but we note that similar
effects have been seen previously with sterically bulky diketiminate
tin(II) initiators and have been subject to a recent theoretical
study.4
[ClONDIPP]Sn(NMe2), 1
[ClONDIPP]H (0.811 g, 2.32 mmol) was dissolved in 20 cm3 toluene
and added to a 20 cm3 toluene solution of Sn(NMe2)2 (0.484 g,
2.34 mmol) chilled to -78 ◦C. After stirring for 2 h room
temperature, volatiles were removed and the yellow solid was
recrystallised (required several days at -30 ◦C) from warm (70 ◦C)
toluene (0.769 g yellow crystals, 1.50 mmol, 65% yield).
1
=
H NMR (400.1 MHz, C6D6): d 8.21 (br, 2H, HC N+/C6H3),
=
=
7.62 (br, 2H, HC N+/C6H3), 7.49 (s, 1H, HC N), 7.32 (d, 2H,
4JHH 2.4 Hz, C6H2), 7.14 (br s, 1H, C6H2), 7.14–6.99 (m, 3H,
4
=
C6H3+HC N), 6.78 (d, 1H, JHH 2.5 Hz, C6H2), 4.19 (br sept,
Scheme 6
This journal is
The Royal Society of Chemistry 2009
Dalton Trans., 2009, 3710–3715 | 3713
©

View Article Online
1H, CH(CH3)2), 3.05 (br sept, 2H, CH(CH3)2), 2.80 (br sept,
solution at -30 ◦C over several days afforded 3 as yellow crystals
(0.501 g, 0.84 mmol, 45% yield).
3
1H, CH(CH3)2), 2.40 (br s, 12H, N(CH3)2), 1.40 (br d, 3H, JHH
3
1
=
H NMR (400.1 MHz, C6D6): d 8.84 (s, 1H, HC N), 8.57 (d,
4 t
5.8 Hz CH(CH3)2), 1.31 (br d, 3H, JHH 5.9 Hz CH(CH3)2),
3
3
1.11 (d, 12H, JHH 6.7 Hz CH(CH3)2), 1.00 (br d, 3H, JHH
1H, JHH 2.6 Hz, 6-C6H2), 7.55 (s, 2H, 3 + 5-C6H2 Bu3), 7.21 (d,
5.9 Hz CH(CH3)2), 0.88 (br d, 3H, JHH 5.6 Hz CH(CH3)2); 13C
1H, JHH 2.7 Hz, 4-C6H2), 2.04 (s, 6H, N(CH3)2), 1.46 (s, 18H,
3
4
NMR (100.6 MHz, C6D6): d 165.54 (HC=N), 162.45 (1/2/3/5-
C6H2), 158.94 (1/2/3/5-C6H2), 148.93 (1-C6H3), 144.91 (1-C6H3),
141.17 (2/6-C6H3), 140.77 (2/6-C6H3), 138.48 (2 coincident
resonances, 2/6-C6H3), 133.63 (4/6-C6H2), 133.03 (2 coincident
resonances, 4/6-C6H2), 132.83 (4/6-C6H2), 127.23 (3/4/5-C6H3),
125.48 (3/4/5-C6H3), 123.71 (3 coincident resonances, 3/4/5-
C6H3), 122.82 (3/4/5-C6H3), 121.78 (1/2/3/5-C6H2), 120.73
(1/2/3/5-C6H2), 41.22 (N(CH3)2), 28.48 (3 coincident reso-
nances, CH(CH3)2), 27.72 (CH(CH3)2), 25.44 (CH(CH3)2), 25.28
(CH(CH3)2), 24.10 (5 coincident resonances, CH(CH3)2), 23.88
(CH(CH3)2); 119Sn NMR (186.4 MHz, C6D6): d -664. (Found: C,
49.00; H, 4.99; N, 5.60. C21H26N2OCl2Sn requires C, 49.26; H,
5.12; N, 5.47%).
2 + 6-C(CH3)3), 1.34 (s, 9H, 4-C(CH3)3); 13C NMR (100.6 MHz,
t
C6D6): d 158.47 (2-C6H2), 157.68 (HC=N), 151.44 (1-C6H2 Bu3),
t
t
144.63 (4-C6H2 Bu3), 138.86 (2 + 6-C6H2 Bu3), 132.09 (4-C6H2),
126.83 (6-C6H2), 124.13 (3/5-C6H2), 123.32 (3/5-C6H2), 121.90
t
(3/5-C6H2 Bu3), 40.40 (N(CH3)2), 36.22 (2 + 6-C(CH3)3), 34.84
(4-C(CH3)2), 32.08 (2 + 6-C(CH3)3), 31.81 (4-C(CH3)3), 1-C6H2
was not found; 119Sn NMR (186.4 MHz, C6D6):d -107. (Found:
C, 54.56; H, 6.17; N, 4.60. C27H36N2OCl2Sn requires C, 54.58; H,
6.11; N, 4.71%).
[2,4-tBu2–6-{CH(NMe2)N-2,4,6-Br3C6H2}-C6H2O]Sn, 4
A 30 cm3 diethyl ether solution containing 0.800 g [tBuONTBP]H
(1.46 mmol) was added to a diethyl ether solution (20 cm3 of
Sn(NMe2)2 (0.318 g, 1.54 mmol) at -78 ◦C. After allowing to
stir for 2 h at room temperature the reaction was concentrated
under reduced pressure to afford a yellow solid. Analytically pure
4 was then obtained by recrystallising from a minimum amount
of diethyl ether stored at -30 ◦C for 18 h (0.599 g yellow crystals,
0.85 mmol, 58% yield). Crystals suitable for X-ray crystallography
were grown by allowing a saturated diethyl ether solution to stand
at room temperature for several days.
[IONDIPP]Sn(NMe2), 2
A diethyl ether solution (30 cm3) of [IONDIPP]H (1.1015 g,
1.91 mmol) was added dropwise to a solution of Sn(NMe2)2
(0.397 g, 1.92 mmol) in diethyl ether (20 cm3) chilled to -78 ◦C. The
mixture was then stirred for 2 h at room temperature, affording a
yellow precipitate. The volatile components were removed under
reduced pressure and yellow (X-ray diffraction quality) crystals of
2 were obtained by allowing a saturated toluene solution to cool
from 70 ◦C to room temperature (0.815 g, 1.17 mmol, 61% yield).
4
1H NMR (400.1 MHz, C6D6): d 7.53 (d, 1H, JHH 2.6 Hz, 4-
4
C6H2), 7.13 (s, 2H, 3 + 5-C6H2Br3), 6.73 (d, 1H, JHH 2.7 Hz,
6-C6H2), 6.62 (s, 1H, HC-N), 2.34 (s, 3H, N(CH3)2), 1.84
(s, 3H, N(CH3)2), 1.80 (s, 9H, 3-C(CH3)3), 1.20 (s, 9H, 5-
C(CH3)3); 13C NMR (100.6 MHz, C6D6): d 154.59 (2-C6H2),
146.17 (1-C6H2Br3), 138.53 (5-C6H2), 138.17 (3-C6H2), 135.20
(2/4/6 + 3+5-C6H2Br3), 125.24 (4-C6H2), 123.88 (6-C6H2),
118.98 (2/4/6-C6H2Br3), 109.76 (2/4/6-C6H2Br3), 88.44 (HC-
N), 43.90 (N(CH3)2), 40.68 (N(CH3)2), 35.54 (3-C(CH3)3), 34.02
(5-C(CH3)2), 31.80 (5-C(CH3)3), 30.28 (3-C(CH3)3); 119Sn NMR
(186.4 MHz, C6D6): d -194. (Found: C, 38.95; H, 4.06; N, 3.90.
C23H29N2OBr3Sn requires C, 39.02; H, 4.13; N, 3.96%).
1H NMR (400.1 MHz, C6D6): d 8.14 (d, 1H, 4JHH 2.3 Hz, C6H2),
4
=
8.04 (br, 1H, C6H2/HC N), 7.95 (d, 1H, JHH 2.3 Hz, C6H2), 7.38
4
=
(s, 1H, HC N), 7.27 (d, 1H, JHH 2.3 Hz, C6H2), 7.18–7.00 (m,
=
7H, C6H3+C6H2/HC N), 4.17 (br sept, 1H, CH(CH3)2), 3.04 (br
sept, 2H, CH(CH3)2), 2.69 (br sept, 1H, CH(CH3)2), 2.61 (br s,
3
12H, N(CH3)2), 1.39 (br d, 3H, JHH 5.6 Hz CH(CH3)2), 1.31
3
3
(br d, 3H, JHH 5.7 Hz CH(CH3)2), 1.12 (d, 12H, JHH 6.8 Hz
3
CH(CH3)2), 0.93 (br d, 3H, JHH 6.4 Hz CH(CH3)2), 0.89 (br d,
3H, 3JHH 6.2 Hz CH(CH3)2); 13C{ H} NMR (100.6 MHz, C6D6):
1
=
d 165.43 (HC N), 163.42 (1/2/3/5-C6H2), 162.70 (1/2/3/5-
C6H2), 150.71 (4/6-C6H2), 150.38 (4/6-C6H2), 144.94 (4/6-C6H2),
144.47 (1-C6H3), 141.20 (2/6-C6H3), 140.86 (2/6-C6H3), 138.78
(2 coincident resonances, 2/6-C6H3), 127.13 (2 coincident res-
onances, 3/4/5-C6H3), 125.60 (2 coincident resonances, 3/4/5-
C6H3), 123.75 (2 coincident resonances, 3/4/5-C6H3), 123.42
(1-C6H3/1-C6H2), 96.52 (1/2/3/5-C6H2), 96.14 (1/2/3/5-C6H2),
78.45 (1/2/3/5-C6H2), 76.72 (1/2/3/5-C6H2), 43.42 (N(CH3)2),
28.51 (3 coincident resonances, CH(CH3)2), 28.08 (CH(CH3)2),
26.08 (CH(CH3)2), 25.71 (CH(CH3)2), 24.17 (5 coincident reso-
nances, CH(CH3)2), 23.63 (CH(CH3)2); 119Sn NMR (186.4 MHz,
C6D6): d -639. (Found: C, 36.28; H, 3.81; N, 3.90. C21H26N2OI2Sn
requires C, 36.29; H, 3.77; N, 4.03%).
Crystal data for 2. C42H52I4N4O2Sn2, M = 1389.86, triclinic,
˚
space group P¯ı, a = 8.6381(5), b = 10.0747(6), c = 14.7661(8) A,
a = 71.3596(9), b = 85.7434(10), g = 79.4997(9)◦, U = 1197.07(12)
3
-1
˚
˚
A , T = 150(2) K, Z = 1, m(Mo-Ka) = 3.659 mm , l = 0.71073 A,
10669 reflections measured, 5538 unique (Rint = 0.0170) which
were used in all calculations. The final wR2 = 0.0556 (all data) and
R1 = 0.0247 (for 4588 data with F2 > 2s(F2)). CCDC 696308.
Crystal data for 4. C23H29Br3N2OSn, M = 707.90, triclinic,
space group P¯ı, a = 9.0297(2), b = 10.2931(3), c = 14.0540(4)
◦
˚
A, a = 78.4205(12), b = 88.3895(17), g = 80.8048(18) , U =
3
-1
˚
1263.19(6) A , T = 120(2) K, Z = 2, m(Mo-Ka) = 5.773 mm ,
˚
l = 0.71073 A, 22501 reflections measured, 5749 unique (Rint
=
0.0485). wR2 = 0.0734 (all data); R1 = 0.0301 (for 4961 data with
F2 > 2s(F2)). CCDC 696309.
[ClONTTBP]Sn(NMe2), 3
A 30 cm3 diethyl ether solution of [ClONTTBP]H (0.802 g, 1.85 mmol)
was added to Sn(NMe2)2 (0.385 g, 1.86 mmol, 20 cm3 diethyl
ether) at -78 ◦C. After allowing to warm to room temperature,
the reaction was stirred for 2 h and volatiles removed in vacuo to
give a yellow powder. Recrystallisation from a saturated toluene
Polymerisation procedure
In a N2 atmosphere glove box, 10 ml toluene was added to a solid
mixture of rac-LA (0.4036 g, 0.28 mmol) and the tin-based initiator
(0.0028 mmol), and a magnetic stirrer was added. This mixture
3714 | Dalton Trans., 2009, 3710–3715
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
was then transferred immediately to a pre-warmed oil-bath and
stirred at 60 ◦C. At appropriate time intervals, aliquots were
removed and following termination with MeOH the conversion
of monomer was determined by integration of the monomer vs.
polymer methine resonances in the 1H NMR spectrum (CDCl3),
and molecular weight data was determined by GPC.
Acknowledgements
The authors express their gratitude to the Royal Thai Government
for a studentship (N.N.) and the EPSRC UK National Crystallog-
raphy Service at the University of Southampton for data collection
for compound 4.
Notes and references
Conclusions
1 A. P. Dove, V. G. Gibson, E. L. Marshall, A. J. P. White and D. J.
Williams, Chem. Commun., 2001, 283.
2 K. B. Aubrecht, M. A. Hillmyer and W. B. Tolman, Macromolecules,
2002, 35, 644.
3 N. Nimitsiriwat, E. L. Marshall, V. C. Gibson, M. R. J. Elsegood and
S. H. Dale, J. Am. Chem. Soc., 2004, 126, 13598.
4 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.
5 N. Nimitsiriwat, V. C. Gibson, E. L. Marshall, A. J. P. White, S. H.
Dale and M. R. J. Elsegood, Dalton Trans., 2007, 4474.
6 N. Nimitsiriwat, V. C. Gibson, E. L. Marshall and M. R. J. Elsegood,
Inorg. Chem., 2008, 47, 5417.
Although four of the smaller salicylaldimines are susceptible
to bischelation at divalent tin centres, this study shows that
mono(iminophenoxide) counterparts are accessible. A fine balance
between steric and electronic effects clearly operates; for example,
while bischelation is observed with [MeONDIPP]H, the sterically
similar [ClONDIPP]H yields 1 cleanly and in good yield. In this case,
the mono(amide) complex may be stabilised by the formation of
the bridged dimer, a possible repercussion of the Lewis acidic tin
when supported by the halogenated ancillary ligand.
7 J. Wu, T.-L. Yu, C.-T. Chen and C.-C. Lin, Coord. Chem. Rev., 2006,
250, 602.
One of the most intriguing results emanating from this work is
the isolation of complex 4 which confirms that the amide migration
previously reported by us is apparently not limited to tridentate
salicylaldimines. However, the weak coordination of the ortho-
bromo substituent in the solid state structure of 4 could point
towards a requirement for a third coordination site to be occupied
by the ligand. It is interesting that this rearrangement is not
observed in 1–3 with halogenated phenolic groups, indicating that
electron withdrawing N-substituents are more likely to activate the
imino carbon towards nucleophilic attack. Preliminary 1H NMR
scale experiments with the analogous N-2,4,6-trichlorophenyl
salicylaldimine with Sn(NMe2)2 indicate that amide migration also
occurs in this system too.
All four complexes isolated in this study are active for the
polymerisation of rac-lactide, though the mononuclear terminal
amide complex 3 is notably slower than 1, 2 or 4. This observation
is attributed to a common propagating species structure in all four
cases (Scheme 6), with the accessibility of the metal in 3 impeded
by the N-2,4,6-tris(tert-butyl)phenyl substituent.
8 O. Dechy-Cabaret, B. Martin-Vaca and D. Bourissou, Chem. Rev., 2004,
6147.
9 K. Nakano, N. Kosaka, T. Hiyama and K. Nazaki, Dalton Trans., 2003,
4039.
10 B. J. O’Keefe, M. A. Hillmyer and W. B. Tolman, J. Chem. Soc., Dalton
Trans., 2001, 2215.
11 P. Foley and M. Zeldin, Inorg. Chem., 1975, 14, 2264.
12 M. H. Chisholm, J. C. Gallucci, H. Zhen and J. C. Huffman, Inorg.
Chem., 2001, 40, 5051.
13 A. Recca, F. A. Bottino and P. Finocchiaro, J. Inorg. Nucl. Chem., 1980,
42, 479.
14 F. A. Bottino and P. Finocchiaro, Polyhedron, 1985, 4, 1507.
15 M. A. Torzilli, S. Colquhoun, D. Doucet and R. H. Beer, Polyhedron,
2002, 21, 697.
16 D. J. Darensbourg, P. Rainey and J. Yarbrough, Inorg. Chem., 2001, 40,
986–993.
17 B. Erk, A. Dilmac, Y. Baran and A. Balaban, Synth. React. Inorg.
Met.-Org. Chem., 2000, 30, 1929.
18 R. Hani and R. A. Geanangel, Coord. Chem. Rev., 1982, 44, 229.
19 A. Bondi, J. Phys. Chem., 1964, 68, 441.
20 M. T. Zell, B. E. Padden, A. J. Paterick, K. A. M. Thakur, R. T. Kean,
M. A. Hillmyer and E. J. Munson, Macromolecules, 2002, 35, 7700.
This journal is
The Royal Society of Chemistry 2009
Dalton Trans., 2009, 3710–3715 | 3715
©