Ring-opening polymerization of l-lactide by organotin(IV)alkoxides, R2Sn(OPri)2: Estimation of the activation parameters

Article abstract of DOI:10.1016/j.poly.2007.06.002

The ring-opening polymerization of l-lactide, l-LA, to give poly-l-lactide by R₂Sn(OPrⁱ)₂ compounds, where R = Buⁿ and p-XC₆H₄ (X = CF₃, F, H, Me and OMe) has been studied in benzene over a temperature range. There is a relatively small variation in ΔH^≠ as a function of R with all the values falling within the range 11 ± 2 kcal mol^-1. The entropy of activation, ΔS^≠, is consistently large and negative, -50 ± 5 eu, supporting the view that the ring-opening event, the enchainment step involves a highly ordered transition state. The crystal and molecular structures of the compounds Ph₂Sn(OPrⁱ)₂, (p-FC₆H₄)₂Sn(OPrⁱ)₂ and (p-Me₂NC₆H₄)₃SnOPrⁱ are also reported. While the latter compound is monomeric in the solid state the former are both dimeric with a pair of bridging OPrⁱ ligands.

Full text of DOI:10.1016/j.poly.2007.06.002

Polyhedron 26 (2007) 4436–4444
www.elsevier.com/locate/poly
Ring-opening polymerization of L-lactide
by organotin(IV)alkoxides, R2Sn(OPri)2: Estimation
of the activation parameters
*
Malcolm H. Chisholm , Judith C. Gallucci, Clemens Krempner
Department of Chemistry, The Ohio State University, 100 West 18th Avenue, Columbus, OH 43210-1185, USA
Received 1 March 2007; accepted 1 June 2007
Available online 10 June 2007
Abstract
The ring-opening polymerization of L-lactide, L-LA, to give poly-L-lactide by R2Sn(OPri)2 compounds, where R = Bun and p-XC6H4
¼
(X = CF3, F, H, Me and OMe) has been studied in benzene over a temperature range. There is a relatively small variation in DH as a
¼
function of R with all the values falling within the range 11 2 kcal molꢀ1. The entropy of activation, DS , is consistently large and
negative, ꢀ50 5 eu, supporting the view that the ring-opening event, the enchainment step involves a highly ordered transition state.
The crystal and molecular structures of the compounds Ph2Sn(OPri)2, (p-FC6H4)2Sn(OPri)2 and (p-Me2NC6H4)3SnOPri are also
reported. While the latter compound is monomeric in the solid state the former are both dimeric with a pair of bridging OPri ligands.
ꢀ 2007 Published by Elsevier Ltd.
Keywords: Tin; Alkoxides; Catalysis; Lactide polymerization
1. Introduction
metal-alkoxide bonds effect a more rapid ROP and conse-
quently lanthanides, and group 1 and 2 metal alkoxides are
especially active. These metal complexes do, however, suf-
fer some limitations in as much as deleterious side reactions
are also kinetically more rapid. These side reactions include
epimerization and trans-esterification both of which can
lead to a loss of stereoselectivity in the polymerization of
LA. Trans-esterification furthermore increases the polydis-
persity of the polyester. Consequently a great deal of
attention has been given to ROP involving well-defined sin-
gle-site metal alkoxide catalysts where the metal–oxygen
bond is more covalent. Thus numerous studies have dealt
with salen aluminum(III) alkoxides. These have attracted
great attention because of their ability to show stereoselec-
tivity [7–13]. They are, however, very slow catalysts effect-
ing the ROP of ca 50–100 equiv of LA in typically 1–14
days at 70 ꢁC. They are better considered oligomerization
catalysts. Despite the numerous studies to date the intrica-
cies of the factors leading to stereocontrol are not well
understood [14]. The various reactions involved in LA
ROP are shown in Scheme 1.
The ring-opening polymerization, ROP, of cyclic esters
derived from renewable resources provides an efficient
and sustainable route to polyesters [1]. Lactide LA, derived
from enzymatic degradation of starches, can be readily
converted to polylactide, PLA, in this manner and the
PLA finds applications from clothing, carpeting to packag-
ing, drug delivery and numerous medical devices. ROP
may be brought about by organic [2–4] or coordinate catal-
ysis [5], the latter involving a metal-alkoxide group as the
propagating species. Coordinate catalysis offers certain
advantages in terms of control of stereochemistry when
employing rac-LA and, in addition to isotactic PLA
derived from the ROP of L-LA (or D-LA), syndiotactic,
heterotactic, stereoblock and stereoplex polymers are
known [6]. A wide variety of metal complexes are known
to be effective initiators for the ROP of LAs. Polar
*
Corresponding author.
E-mail address: chisholm@chemistry.ohio-state.edu (M.H. Chisholm).
0277-5387/$ - see front matter ꢀ 2007 Published by Elsevier Ltd.
doi:10.1016/j.poly.2007.06.002

M.H. Chisholm et al. / Polyhedron 26 (2007) 4436–4444
4437
L
LM X
O
M
X
O
O
O
LM
O
O
Ring-opening
X
O
O
O
O
O
O
O
O
M = metal
L = ligand
X = initiation group
O
O
n
O
O
Propagation
O
X
X
O
LM
O
(Ring-opening Polymerization)
LM
O
O
O
O
n+1
Side reactions
O
O
O
O
O
P
P'
O
ML
O
ML
O
O
O
Interchain Transesterification
+
O
O
P'
P = propagation polymer
LM
P
LM
O
O
O
P
O
O
O
O
O
O
P
O
O
Intrachain Transesterification
LM
O
O
O
LM
O
+
O
O
O
O
O
O
n-1
n
+
+
Chain Transfer
LM' OP
LM OP
LM' OP'
LM OP'
Scheme 1. Reactions involved in ring-opening polymerization employing a coordination catalyst.
One advantage of studying a slow catalytic system is
that one may judge the relative importance of subtle
changes at the metal center on the rate of the reaction
and in this regard we were attracted to the study of organo-
tin(IV) complexes in the ring-opening and ROP of lactide.
We found previously that the rate of initial ring-opening of
LA by Ph3SnX complexes was very sensitive to the nature
of the X group and followed the order X = NMe2 >
OMe > OPri > OBut > N(SiMe3)2 > OPh which indicated
the importance of both steric and electronic factors,
namely the nucleophilicity of the X group [15]. In the
ROP of LA by Ph3SnX initiators it was also noted that
the tin(IV) complexes were subject to ligand redistribution
reactions yielding Ph4Sn and Ph2Sn(OP)2 species and
furthermore that when the initiating group X = NMe2
intrachain trans-esterification was favored yielding cyclic-
oligoesters and Ph2Sn(OCHMeC(O)NMe2)2 [16,17]. In
contrast the Ph2Sn(OPri)2 initiators produced a much
‘‘better behaved’’ catalytic system for the ROP of LA
and this prompted us to evaluate the influence of various
aryl substitutents p-XC6H4, where X = CF3, F, H, OMe
and NMe2, on the kinetics of the ROP by bisaryltin(IV)
complexes (C6H4)2Sn(OPri)2 and to compare these with
the dialkyltin(IV) complex (Bun)2Sn(OPri)2. These consid-
erations provide the basis for the work described herein.
2. Experimental
2.1. General
The manipulation of air-sensitive compounds involved
standard Schlenk line and dry box techniques. All solvents
were distilled under nitrogen from alkali metals and stored
over 4 A molecular sieves. L-Lactide was purified by subli-
mation under vacuum three times. Benzene-D6 was dried
over activated molecular sieves and stored in the glove

4438
M.H. Chisholm et al. / Polyhedron 26 (2007) 4436–4444
box. The tin dichlorides of general formula (p-
XC6H4)2SnCl2 (where X = CF3, F, H, Me, MeO, NMe2)
were synthesized via the Kochesnikov redistribution reac-
tion involving the tin tetrachloride with (p-XC6H4)4Sn
and isolated by re-crystallization from toluene as described
previously [18,19]. Samples of high purity were obtained by
a distillation under vacuum using a Kugelrohr apparatus,
except for (p-Me2NC6H4)2SnCl2 which was recrystallized
twice from toluene. Tin chlorides of formula (p-
XC6H4)3SnCl (where X = CF3, F, Me) were synthesized
via the Kochesnikov redistribution reaction involving (p-
XC6H4)2SnCl2 with (p-XC6H4)4Sn in the neat at 170 ꢁC
and purified by distillation under vacuum (Kugelrohr
apparatus) [20]. Ph3SnNMe2 was synthesized by the reac-
tion of LiNMe2 with Ph3SnCl in benzene according to
the literature procedure [4]. 1H, 13C and 118Sn NMR spectra
were obtained from either Bruker DPX-400 or DPX-250
spectrometers using benzene-D6 as solvent and elemental
analyses were performed by the Atlantic Microlab, Inc.
All kinetic data were obtained from NMR scale reactions.
Standard solutions of (p-XC6H4)2Sn(OiPr)2 and L-lactide
were made in benzene-D6 and stored in the dry box.
Appropriate aliquots of both reagents (molar ratio = 1:50)
were transferred to a J. Young NMR tube. The reaction
temperatures were regulated via a thermostatically con-
trolled oil bath. For lactide polymerizations reaction rates
152.5 (C–N(CH3)2), 136.7 (m-CH), 122.5 (ipso-C–Sn),
112.7 (o-CH), 39.1 (N(CH3)2) ppm; 118Sn NMR (C6D6,
149.2 MHz): d ꢀ3.3 ppm.
2.3. General procedure for the synthesis of (p-XC6H4)2Sn-
(OiPr)2(X = CF3, F, H, Me, MeO) and (p-Me2NAr)3-
SnOiPr
Typically, a solution of (p-XC6H4)2SnCl2 (1 equiv) in
benzene was added via a syringe to a vigorously stirred
and ice cooled suspension of LiNMe2 (2.2 equiv) in ben-
zene. After stirring at room temperature for 2 days, the sus-
pension was separated by centrifugation or filtration. All
volatiles were removed under high vacuum to quantita-
tively leave (p-XC6H4)2Sn(NMe2)2 (as shown by NMR
spectroscopy) as colorless oils or solids. The compounds
were redissolved in hexanes and a ten-fold excess of isopro-
panol was added at room temperature. After stirring for ca.
4 h all volatiles were removed under vacuum and the solid
residues were recrystallized from hexanes at ꢀ40 ꢁC to
yield analytically pure samples of the title compounds.
2.3.1. (p-CF3C6H4)2Sn(OiPr)2
(p-CF3C6H4)2SnCl2 (1105 mg, 2.30 mmol) and LiNMe2
(260 mg, 5.10 mmol). Yield 450 mg, 37%. M.p. 83–85 ꢁC;
3
1H NMR (C6D6, 400 MHz): d 7.50 (d, JH–H = 7.6 Hz,
1
3
of polymerization were determined from H NMR data
o-CH aryl, 4 H), 7.37 (d, JH–H = 8.0 Hz, m-CH aryl, 4
3
(400 MHz) where the sum of the area of monomer and
polymer peaks (CH and Me) was assumed to be 100%
and rates of disappearance of monomer were calculated
by the substraction of the polymer peaks from the mono-
mer peaks and dividing by the concentration of monomer
at t = 0 to give a ratio. Equilibrium monomer concentra-
tions were determined by polymerizing L-lactide using the
(p-XC6H4)2Sn(OiPr)2 catalyst precursors in benzene-D6
(1:25 OiPr: L-lactide; [LA]init = 0.1229 M]) at various tem-
peratures. The natural log of the equilibrium monomer
concentration was then plotted against 1/T to give a
straight line. These data were used in the various polymer-
ization reactions as indicated in Fig. 4. The value of
kprop was determined from the gradient a plot of
ꢀln[(A ꢀ Aeq)/(A0 ꢀ Aeq)] versus time (Fig. 6), where
A = [LA], Aeq = the equilibrium [LA] for a given tempera-
ture and A0 = [LA] at t = 0.
H), 4.28 (sept, JH–H = 6 Hz, OCH(CH3)2, 2 H), 1.24 (d,
3JH–H = 6 Hz, OCH(CH3)2, 12 H) ppm; 13C NMR
(C6D6, 100.6 MHz): d 142.8 (ipso-C–Sn), 136.6 (o-CH),
2
3
132.8 (d, JC–F = 40.2 Hz, C–CF3), 125.6 (d, JC–F
=
1
4.0 Hz, m-CH), 124.5 (d, JC–F = 272.6 Hz, C–CF3), 68.2
(OCH(CH3)2), 27.6 (OCH(CH3)2) ppm; 118Sn NMR
(C6D6, 149.2 MHz): d ꢀ173.7 ppm; Anal. Calc. for
C20H22F6O2Sn (527.088): C, 45.57; H, 4.21. Found: C,
45.43; H, 4.02%.
2.3.2. (p-CF3C6H4)2Sn(NMe2)2
1H NMR (C6D6, MHz): d 7.4–7.5 (m, CH aryl, 8 H),
2.81 (s, N(CH3)2, 12 H) ppm; 13C NMR (C6D6, MHz): d
2
142.4 (ipso-C–Sn), 137.2 (o-CH), 132.2 (d, JC–F
=
3
32.2 Hz, C–CF3), 125.4 (d, JC–F = 3 Hz, m-CH), 124.7
(d, JC–F = 271.6 Hz, CF3), 43.6 (N(CH3)2) ppm; 118Sn
1
NMR (C6D6, 149.2 MHz): d ꢀ83.7 ppm.
2.2. Synthesis
2.3.3. (p-FC6H4)2Sn(OiPr)2
(p-FC6H4)2SnCl2 (1270 mg, 3.38 mmol) and LiNMe2
(362 mg, 7.10 mmol). Yield 910 mg, 63%. M.p. 85–87 ꢁC;
2.2.1. (p-Me2NC6H4)2SnCl2
3
Following the procedure given in the literature [20], (p-
Me2NC6H4)4Sn (11.62 g, 19.4 mmol) and SnCl4 (2.27 ml,
19.4 mmol) were dissolved in toluene (100 ml) and heated
up to 110 ꢁC. Cooling down slowly to room temperature
gave crystals of the title compound, which were isolated
by filtration. Yield 15.04 g, 90%. 1H NMR (C6D6,
1H NMR (C6D6, 400 MHz): d 7.43 (dd, JH–H = 8.4 Hz,
3
3JH–F = 6 Hz, m-CH aryl, 4 H), 6.82 (tr, JH–H = 8.4 Hz,
3
o-CH aryl, 4 H), 4.33 (sept, JH–H = 6 Hz, OCH(CH3)2, 2
3
H), 1.28 (d, JH–H = 6 Hz, OCH(CH3)2, 12 H) ppm; 13C
1
NMR (C6D6, 100.6 MHz): d 164.9 (d, JC–F = 250.5 Hz,
3
4
C–F), 138.2 (d, JC–F = 7 Hz, m-CH), 133.1 (d, JC–F
=
3
2
400 MHz): d 7.57 (d, JH–H = 8.8 Hz, m-CH aryl, 4 H),
4 Hz, ipso-C–Sn), 116.6 (d, JC–F = 20.1 Hz, o-CH), 67.9
(OCH(CH3)2), 27.8 (OCH(CH3)2) ppm; 118Sn NMR
(C6D6, 149.2 MHz): d ꢀ152.6 ppm; Anal. Calc. for
3
6.43 (d, JH–H = 8.8 Hz, d, o-CH aryl, 4 H), 2.35 (s,
N(CH3)2, 12 H) ppm; 13C NMR (C6D6, 100.6 MHz): d

M.H. Chisholm et al. / Polyhedron 26 (2007) 4436–4444
4439
C18H22F2O2Sn (427.073): C, 50.62; H, 5.19. Found: C,
50.34; H, 5.08%.
13C NMR (C6D6, 100.6 MHz): d 161.6 (C-OCH3), 138.3
(m-CH), 128.3 (ipso-C–Sn), 115.1 (o-CH), 54.5 (C–
OCH3), 43.9 (N(CH3)2) ppm; 118Sn NMR (C6D6,
149.2 MHz): d ꢀ7.4 ppm.
2.3.4. (p-FC6H4)2Sn(NMe2)2
1H NMR (C6D6, 250 MHz): d 7.35 (dd, 3JH–H = 8.5 Hz,
3JH–F = 6.3 Hz m-CH aryl, 4 H), 6.88 (dd, 3JH–H = 8.5 Hz,
2.3.10. (p-Me2NC6H4)3SnOiPr
4JH–F = 8.6 Hz o-CH aryl, 4 H), 2.84 (s, N(CH3)2, 12 H)
(p-Me2NC6H4)2SnCl2 (858 mg, 1.99 mmol) and LiNMe2
1
ppm; 13C NMR (C6D6, 62.9 MHz): d 164.6 (d, JC–F
=
(224 mg, 4.39 mmol). Yield 170 mg, 32%. M.p. 125–126 ꢁC
3
3
245 Hz, C–F), 138.7 (d, JC–F = 7.1 Hz, m-CH), 132.7 (d,
(dec.); 1H NMR (C6D6, 250 MHz): d 7.86 (d, JH–H
=
4JC–F = 4.4 Hz, ipso-C–Sn), 116.4 (d, JC–F = 19.7 Hz,
8.4 Hz, m-CH aryl, 6 H), 6.68 (d, JH–H = 8.4 Hz, o-CH
aryl, 6 H), 4.59 (sept, JH–H = 6 Hz, OCH(CH3)2, 1 H),
2.48 (s, N(CH3)2, 18 H), 1.48 (d, JH–H = 6 Hz,
2
3
3
o-CH), 43.6 (N(CH3)2) ppm.
3
2.3.5. Ph2Sn(OiPr)2
OCH(CH3)2, 6 H) ppm; 13C NMR (C6D6, 100.6 MHz): d
151.7 (C–NMe2), 138.0 (m-CH), 125.3 (ipso-C–Sn), 113.4
(o-CH), 67.5 (OCH(CH3)2), 39.8 (N(CH3)2), 28.4
(OCH(CH3)2) ppm; 118Sn NMR (C6D6, 149.2 MHz): d
ꢀ66.6 ppm; Anal. Calc. for C27H37ON3Sn (538.312): C,
60.24; H, 6.93. Found: C, 60.12; H, 6.78%.
Ph2SnCl2 (2057 mg, 5.98 mmol) and LiNMe2 (670 mg,
13.14 mmol). Yield 1610 mg, 69%. M.p. 83–84 ꢁC (dec.).
The analytical data obtained were in agreement with those
previously reported [4].
2.3.6. (p-MeC6H4)2Sn(OiPr)2
(p-MeC6H4)2SnCl2 (974 mg, 2.62 mmol) and LiNMe2
2.3.11. (p-Me2NC6H4)2Sn(NMe2)2
(p-Me2NC6H4)2SnCl2 (2094 mg, 4.87 mmol) and
LiNMe2 (550 mg, 10.78 mmol). Yield 1900 mg, 87%. M.p.
(280 mg, 5.49 mmol). Yield 330 mg, 30%. M.p. 77-78 ꢁC
3
(dec.); 1H NMR (C6D6, 400 MHz): d 7.64 (d, JH–H
=
3
8 Hz, m-CH aryl, 4 H), 7.02 (d, JH–H = 8 Hz, o-CH aryl,
4 H), 4.47 (sept, JH–H = 6 Hz, OCH(CH3)2, 2 H), 2.04
(s, CH3, 6 H), 1.35 (d, JH–H = 6 Hz, OCH(CH3)2, 12 H)
75–76 ꢁC (dec.); 1H NMR (C6D6, 400 MHz): d 7.69 (d,
3
3
3JH–H = 8.4 Hz, m-CH aryl, 4 H), 6.66 (d, JH–H
=
3
8.4 Hz, o-CH aryl, 4 H), 3.11 (s, Sn–N(CH3)2, 12 H), 2.49
(s, p-N(CH3)2, 12 H) ppm; 13C NMR (C6D6,
100.6 MHz): d 151.7 (C–N(CH3)2), 138.0 (m-CH), 123.1
(ipso-C–Sn), 113.4 (o-CH), 44.1, (Sn–N(CH3)2), 39.7
(p-N(CH3)2) ppm; 118Sn NMR (C6D6, 149.2 MHz): d
ꢀ60.8 ppm; Anal. Calc. for C20H32N4Sn (447.205): C,
53.71; H, 7.21. Found: C, 53.44; H, 7.03%.
ppm; 13C NMR (C6D6, 100.6 MHz): d 140.6 (C–CH3),
136.5 (m-CH), 134.7 (ipso-C–Sn), 130.1 (o-CH), 67.7
(OCH(CH3)2), 28.0 (C–CH3), 21.4 (OCH(CH3)2) ppm;
118Sn NMR (C6D6, 149.2 MHz): d ꢀ 147.1 ppm; Anal.
Calc. for C20H28O2Sn (419.145): C, 57.31; H, 6.73. Found:
C, 57.19; H, 6.65%.
2.3.7. (p-MeC6H4)2Sn(NMe2)2
2.4. General procedure for the synthesis of
(p-XC6H4)3SnOiPr (X = CF3, F, Me)
3
1H NMR (C6D6, 250 MHz): d 7.60 (d, JH–H = 7.9 Hz,
3
m-CH aryl, 4 H), 7.07 (d, JH–H = 7.9 Hz, o-CH aryl,
4 H), 2.99 (s, N(CH3)2, 12 H), 2.08 (s, CH3, 6 H) ppm;
13C NMR (C6D6, 62.9 MHz): d 139.7 (C–CH3), 137.1 (m-
CH), 134.3 (ipso-C–Sn), 129.9 (o-CH), 43.9 (N(CH3)2),
21.4 (C–CH3) ppm.
Typically, a suspension of (p-XC6H4)3SnCl (1 equiv)
and LiOiPr (1.1 equiv) in benzene was vigorously stirred
for two days at room temperature and for ca 4 h at
60 ꢁC. After centrifugation and filtration of the mixture
all volatiles were removed under vacuum and the residues
were either recrystallized from hexanes at ꢀ40 ꢁC or dis-
tilled under vacuum (Kugelrohr apparatus) to yield analyt-
ically pure samples of the title compounds.
2.3.8. (p-MeOC6H4)2Sn(OiPr)2
(p-MeOC6H4)2SnCl2 (1031 mg, 2.56 mmol) and LiNMe2
(274 mg, 5.37 mmol). Yield 750 mg, 65%. 1H NMR (C6D6,
3
250 MHz): d 7.66 (d, JH–H = 8.8 Hz, m-CH aryl, 4 H),
3
6.83 (d, JH–H = 8.8 Hz, o-CH aryl, 4 H), 4.52 (sept,
2.4.1. (p-MeC6H4)3SnOiPr
3JH–H = 6 Hz, OCH(CH3)2, 2 H), 3.21 (s, CH3O, 6 H),
(p-MeC6H4)3SnCl (1735 mg, 4.06 mmol) and LiOiPr
(280 mg, 4.24 mmol). Yield 1340 mg, 73%. Bp. 135 ꢁC
3
1.40 (d, JH–H = 6 Hz, OCH(CH3)2, 12 H) ppm; 13C
NMR (C6D6, 100.6 MHz): d 162.1 (C–OMe), 138.0 (m-
CH), 128.1 (ipso-C–Sn), 115.2 (o-CH), 67.7 (OCH(CH3)2),
54.4 (CH3O), 28.3 (OCH(CH3)2) ppm; 118Sn NMR (C6D6,
149.2 MHz): d ꢀ133.7 ppm; Anal. Calc. for C20H28O4Sn
(451.144): C, 53.25; H, 6.26. Found: C, 52.98; H, 6.14%.
(4 · 10ꢀ2 mbar); 1H NMR (C6D6, 400 MHz): d 7.7 (d,
3
3JH–H = 7.6 Hz, m-CH aryl, 6 H), 7.06 (d, JH–H
=
3
7.6 Hz, o-CH aryl, 6 H), 4.42 (sept, JH–H = 6 Hz,
3
OCH(CH3)2, 1 H), 2.07 (s, CH3, 9 H), 1.36 (d, JH–H
=
6 Hz, OCH(CH3)2,
6
H) ppm; 13C NMR (C6D6,
62.9 MHz): d 139.8 (C–CH3), 137.0 (m-CH), 135.7 (ipso-
C–Sn), 129.9 (o-CH), 67.9 (OCH(CH3)2), 28.2 (C–CH3),
21.4 (OCH(CH3)2) ppm; 118Sn NMR (C6D6, 149.2 MHz):
d ꢀ85.9 ppm; Anal. Calc. for C24H28OSn (451.189): C,
63.89; H, 6.26. Found: C, 63.67; H, 6.05%.
2.3.9. (p-MeOC6H4)2Sn(NMe2)2
3
1H NMR (C6D6, 400 MHz): d 7.58 (d, JH–H = 8.4 Hz,
3
m-CH aryl, 4 H), 6.87 (d, JH–H = 8.4 Hz, o-CH aryl, 4
H), 3.29 (s, OCH3, 6 H); 3.00 (s, N(CH3)2, 12 H) ppm;

4440
M.H. Chisholm et al. / Polyhedron 26 (2007) 4436–4444
1
2.4.2. (p-FC6H4)3SnOiPr
monitored after ca. 13 h at r.t. by H NMR spectroscopy
(ratio: 1.0:0.7 for X = F:Me).
(p-FC6H4)3SnCl (1336 mg, 3.04 mmol) and LiOiPr
(220 mg, 3.33 mmol). Yield 880 mg, 63%. M.p. 53–54 ꢁC;
(2) 6.8 mg of (p-MeC6H4)3SnOiPr and 8.1 mg of (p-
Me2NC6H4)3SnOiPr and 2.5 mg of L-lactide (1:1:1 ratio)
were dissolved in benezene-D6 (0.5 mL) and the reaction
was monitored after ca. 130 min at r.t. by H NMR spec-
troscopy (ratio: 0.9:1.0 for X = Me:NMe2). In competitive
experiments the initial loss of the triaryltin complex was
1H NMR (C6D6, 400 MHz): d 7.36 (dd, JH–H = 8.4 Hz,
3
3
3
d, JH–F = 6 Hz, m-CH aryl, 4 H), 6.87 (tr, JH–H
=
3
1
8.4 Hz, o-CH aryl, 4 H), 4.18 (sept, JH–H = 6 Hz,
3
OCH(CH3)2, 1 H), 1.23 (d, JH–H = 6 Hz, OCH(CH3)2, 6
H) ppm; 13C NMR (C6D6, 62.9 MHz): d 164.7 (d, JC–F
=
1
3
1
249.7 Hz, C–F), 138.5 (d, JC–F = 6.9 Hz, m-CH), 133.5
monitored. At longer time the H NMR spectra are com-
plicated by other reactions including further LA insertion
4
2
(d, JC–F = 3.8 Hz, ipso-C–Sn), 116.5 (d, J = 20.1 Hz, o-
CH), 68.0 (OCH(CH3)2), 28.0 (OCH(CH3)2) ppm; 118Sn
NMR (C6D6, 149.2 MHz): d ꢀ89.1 ppm; Anal. Calc. for
C21H19F3OSn (463.080): C, 54.47; H, 4.14. Found: C,
54.21; H, 4.07%.
and trans esterification.
2.6. Crystallography
A summary of crystal data collection and refinement
details for compounds (p-Me2NC6H4)3SnOiPr, Ph2Sn(OiPr)2
and (p-FC6H4)2Sn(OiPr)2 is given in Table 1. The data
collection crystals for (p-Me2NC6H4)3SnOiPr were color-
less rectangular plates, Ph2Sn(OiPr)2 was a colorless plate,
2.4.3. (p-CF3C6H4)3SnOiPr
(p-CF3C6H4)3SnCl (967 mg, 1.64 mmol) and LiOiPr
(120 mg, 1.82 mmol). Yield 840 mg, 84%. M.p. 92–93 ꢁC;
1H NMR (C6D6, 400 MHz): d 7.41 (d, JH–H = 8 Hz,
3
3
o-CH aryl, 6 H), 7.37 (d, JH–H = 8 Hz, m-CH aryl, 6 H),
and (p-FC6H4)2Sn(OiPr)2 was
a colorless triangular
3
4.14 (sept, JH–H = 6 Hz, OCH(CH3)2, 2 H), 1.20
chunk. Single crystal data sets were collected on a Nonius
Kappa CCD diffractometer with Mo Ka radiation at low
temperature using an Oxford Cryosystems Cryostream
Cooler. All sets were measured out to 2h of 55ꢁ. Data
integration was done with Denzo [18]. Scaling and merg-
ing of the data was done with Scalepack [18]. All struc-
tures were solved by the direct methods procedure or
the Patterson method in SHELXS-86[19a] or SHELXS-97
[19b]. Full-matrix least-squares refinements based on F2
were done SHELXL-97 [20], as incorporated in the WINGX
package [21].
The methyl group hydrogen atoms were added at calcu-
lated positions using a riding model with U(H) = 1.5*Ueq-
(bonded C atom). For each methyl group, the torsion
angle, which defines its orientation about the C–C bond
was refined. The other hydrogen atoms were included in
3
(d, JH–H = 6 Hz, OCH(CH3)2, 12 H) ppm; 13C NMR
(C6D6, 62.9 MHz): d 142.5 (ipso-C–Sn), 137.0 (o-CH),
2
3
132.6 (d, JC–F = 32.7 Hz, C–CF3), 125.7 (d, JC–F = 3.8
1
Hz, m-CH), 124.6 (d, JC–F = 272.4 Hz, C–CF3), 68.5
(OCH(CH3)2), 27.9 (OCH(CH3)2) ppm; 118Sn NMR
(C6D6, 149.2 MHz): d ꢀ106.0 ppm; Anal. Calc. for
C24H19F9OSn (613.103): C, 47.02; H, 3.12. Found: C,
46.89; H, 2.98%.
2.5. Competitive studies on the ring-opening of lactide with
(p-X-C6H4)3SnOiPr where X = F, Me, NMe2
(1) 11.8 mg of (p-FC6H4)3SnOiPr and 11.6 mg of (p-
MeC6H4)3SnOiPr and 3.7 mg of L-lactide (1:1:1 ratio) were
dissolved in benzene-D6 (0.5 mL) and the reaction was
Table 1
Summary of crystallographic data
Compound
(p-Me2NC6H4)3SnOiPr
[Ph2Sn(OiPr)2]2
[(p-FC6H4)2Sn(OiPr)2]2
Formula
C27H37ON3Sn
538.29
monoclinic
P21/c
10.5507(1)
23.5325(3)
11.5562(1)
112.646(1)
2648.01(5)
4
C36H48O4Sn2
782.12
monoclinic
P21/c
10.538(1)
15.684(2)
11.641(1)
112.089(4)
1782.8(3)
2
C36H44F4O4Sn2
854.13
monoclinic
P21/n
10.869(1)
13.324(2)
13.220(2)
106.118(5)
1839.2(4)
2
Molar mass
Crystal system
Space group
˚
a (A)
˚
b (A)
˚
c (A)
b (ꢁ)
3
˚
V (A )
Z
q(calc) (Mg/m3)
1.35
1.457
1.542
Crystal size (mm)
Temperature (K)
Absorption coefficient (mmꢀ1
Data/restraints/parameters
R1 [I > 2r(I)]
0.04 · 0.15 · 0.38
150(2)
0.987
6046/0/289
0.030
0.15 · 0.23 · 0.27
200(2)
1.435
4070/0/194
0.020
0.31 · 0.31 · 0.35
150(2)
1.413
4213/0/212
0.021
)
R1, wR2 all data
Goodness of fit
0.064, 0.061
0.995
0.030, 0.045
1.033
0.032, 0.049
1.045
Maximum and minimum peaks
in final difference map (e/A )
0.772 and ꢀ0.359
0.748 and ꢀ0.373
0.569 and ꢀ0.435
3
˚

M.H. Chisholm et al. / Polyhedron 26 (2007) 4436–4444
4441
the model at calculated positions using a riding model with
U(H) = 1.2*Ueq(bonded C atom). Neutral atom scattering
factors were used and include terms for anomalous disper-
sion [22].
3. Results and discussion
3.1. Preparations
The new compounds described in this work were pre-
pared via the Kochesnikov redistribution reactions as
shown in Schemes 2–4.
3.2. Structures
The (p-XC6H)3Sn(OPri) compounds are monomeric in
benzene solution and in the solid state and the molecular
structure of the complex (p-Me2NC6H4)3Sn(OPri) is shown
in Fig. 1.
In contrast the bisaryltin(IV) isopropxides are loosely
associated dimeric species in the solid state. The molecular
structure of the [(Ph)2Sn(OPri)2]2 dimer is shown in Fig. 2.
Fig. 1. Molecular structure of (p-Me2NC6H4)3SnOiPr in the crystal. The
thermal ellipsoids correspond to 30% probability. Hydrogen atoms are
˚
omitted for clarity. Selected bond lengths (A) and angles (ꢁ): Sn–O1
2.0007(15), Sn–C4 2.110(2), Sn–C10 2.121(2), Sn–C16 2.120(2), C1–O1
1.423(3), C1–C2 1.501(3), C1–C3 1.514(3), O1–C1–C2 110.2(2), O1–C1–
C3 108.61(19), Sn–O1–C1 119.79(13), O1–Sn–C4 100.38(7), O1–Sn–C16
105.84(7), O1–Sn–C10 110.09(8), C4–Sn–C16 119.09(8), C4–Sn–C10
109.64(8), C16–Sn–C10 111.00(8).
SnCl4
X
Sn
X
SnCl2
4
2
X = CF3, F, H,
Me, MeO
LiNMe2
HOiPr
X
Sn(OiPr)2
X
Sn(NMe2)2
2
2
Scheme 2. Synthesis of (p-XC6H4)2Sn(OiPr)2 (X = CF3, F, H, Me, MeO).
(p-XC6H4)2SnCl2
X
Sn
X
SnCl
X= CF3,F ,M e
4
3
LiOiPr
Fig. 2. Molecular structure of [Ph2Sn(OiPr)2]2 in the crystal. The thermal
ellipsoids correspond to 30% probability. Hydrogen atoms are omitted for
clarity.
X
SnOiPr
3
The molecule has a crystallographically imposed center
of inversion. The coordination geometry about the Sn(IV)
center can be described as that of a highly distorted trigonal
bipyramid with the terminal Sn–O2 and bridging Sn–O1
bonds being mutually trans, O2–Sn–O1 = 162.71(5)ꢁ. The
bridging Sn–O bonds are very asymmetric with one being
short, Sn–O1A = 2.020(1) and one long Sn–O1 = 2.356(1)
Scheme 3. Synthesis of (p-XC6H4)3SnOiPr (X = CF3, F, Me).
HOiPr
Me2N
Sn(NMe2)2
Me2N
SnOiPr
-H NMe2
2
3
˚
A. Another way to view the geometry is to consider that
the structure is the result of two pseudo tetrahedral SnO2C2
Scheme 4. Synthesis of (p-Me2NAr)3SnOiPr.

4442
M.H. Chisholm et al. / Polyhedron 26 (2007) 4436–4444
centers weakly associating via the formation of a Sn–O
dative bond as depicted below.
Table 2
i
˚
Selected distances [A] and angles [ꢁ] of Ph2Sn(O Pr)2 and (p-FC6H4)2-
Sn(OiPr)2
Ph2Sn(OiPr)2
(p-FC6H4)2Sn(OiPr)2
Ph
Ph
R
Sn–O1
Sn–O1A
Sn–O2
C1–O1
C4–O2
2.356(1)
2.020(1)
2.013(1)
1.448(2)
1.424(2)
2.131(2)
2.131(2)
68.44(5)
111.56(5)
162.71(5)
94.31(5)
119.35(10)
120.85(11)
2.021(1)
2.292(1)
2.011(1)
1.458(2)
1.419(2)
2.133(2)
2.126(2)
68.76(5)
111.24(5)
91.95(5)
160.71(5)
129.38(10)
118.52(12)
RO
M
O
M
RO
OR
Sn–C7
Sn–C13
Ph
Ph
O1–Sn–O1A
Sn–O1–SnA
O1–Sn–O2
O2–Sn–O1A
Sn–O1–C1
Sn–O2–C4
The molecular structure of the related complex
[(p-FC6H4)2Sn(OPri)2]2 is shown in Fig. 3 and a compari-
son of structural data is given in Table 2.
3.3. Solution NMR studies
In order to interrogate the nature of the bis-aryl tin alk-
oxides in solution we undertook NMR studies that imply
that the weakly associated dimers observed in the solid
state dissociated to monomers in solution. (1) Only one
type of alkoxide group is seen by NMR spectroscopy, (2)
the 119Sn NMR signals are sharp and their chemical shifts
appear in the range of other known monomeric triarylti-
nalkoxides and (3) the DOSY NMR spectra for these
two compounds yield diffusion coefficients and estimated
radii that are comparable to the known monomeric com-
pounds Ph2SnCl2 and Ph3SnCl.
3.4. Studies of kinetics
As described in the Experimental Section the reactions
1
were monitored by H NMR spectroscopy in benzene-d6
over various temperatures with the catalyst to monomer
L-LA ratio of 1:50. The conversion of monomer LA to
polymer PLA was determined by integration as was the
equilibrium concentration of monomer as a function of
temperature. The natural log of the equilibrium monomer
concentration was then plotted against 1/T to give a
straight line. These data are shown in Fig. 4 and were used
in the various polymerization reactions.
The values of kobs were determined from the gradient of
a plot of ꢀln{[A] ꢀ [A]eq/[A]0 ꢀ [A]eq} versus time as
shown in Fig. 5.
¼
¼
The activation parameters, DH and DS , were
extracted from the Arrhenius plots of ꢀlnk/T versus 1/T
(see Fig. 6). The activation parameters are summarized in
Table 3.
We also examined the order of the reaction in [Sn] under
pseudo-first-order conditions where [LA]/[Sn] > 45. The
plot of kobs versus [Sn] for the five samples examined when
[Sn] represents the concentration of Ph2Sn(OPri)2
Fig. 3. Molecular structure of [(p-FC6H4)2Sn(OiPr)2]2 in the crystal. The
thermal ellipsoids correspond to 30% probability. Hydrogen atoms are
omitted for clarity.
Fig. 4. Plots of ꢀln[LA]eq vs. 1/T giving a straight line, which allows for
the determination of the equilibrium monomer concentration of L-lactide
(LA) at various temperatures.

M.H. Chisholm et al. / Polyhedron 26 (2007) 4436–4444
4443
employed in the polymerization at 70 ꢁC is shown in
Fig. 7a. As can be seen the plot yields a straight line with
a non-zero intercept in [Sn]. The only reasonable explana-
tion for this is that in each case studied an equivalent
amount of the catalyst is inactive and is presumably
removed from the system by hydrolysis or some other reac-
tion that leads to its inactivity. This situation is not uncom-
mon and was specifically noted by Hillmyer et al. [23] in a
related study of the polymerization of LA by a monomeric
zinc alkoxide complex.
The plot of lnk versus ꢀln[Sn] from which the order of
the reaction in tin can be deduced is shown in Fig. 7b
(order in [Sn] 1.41). In constructing this plot we have
assumed the correlation for the concentration of [Sn] by
its intercept in Fig. 7a. From this we obtain an order in
[Sn] of 0.93., which we take to imply first order reactivity.
Once again this situation is similar to that described by
Hillmyer [23].
Fig. 5. Typical plot of ꢀln[[A0] ꢀ [Aeq]/[A] ꢀ [Aeq]] vs. time in the ROP of
L-lactide by Ph2Sn(OiPr) at various temperatures.
As can be seen from Table 3 there is a very small range
¼
of DH values and for the p-XC6H4 groups it appears that
the electron withdrawing substituents lower that activation
enthalpy when compared with the electron donating OMe
group. However, the range is very small. A plausible inter-
pretation of this would be that the more electrophilic the
metal center exhibits greater Lewis acidity toward the LA
monomer thus favoring a larger Keq for the reversible bind-
ing of LA. The reactivity of (Bun)2Sn(OPri)2 which is
comparable to the (p-CF3C6H4)2Sn(OPri)2 complex is an
outrigger in as much as its steric factors are not the same.
Irrespective of the interpretation that may be placed on
Fig. 6. Plots of ꢀlnk/T vs. 1/T, giving a straight line, which allows for the
determination of the activation parameters DH* and DS*.
¼
the subtle factors influencing DH , one point that emerges
unequivocally from this study is that the entropy of activa-
Table 3
¼
tion is large in magnitude and negative, DS ꢁ ꢀ50 5 eu.
Calculated activation parameters for the polymerization of lactide with (p-
XC6H4)2Sn(OiPr)2 and Bu2nSn(OPri)2 as catalysts
This is quite a bit larger than typically seen for a bimolec-
¼
ular reaction for which DS values fall in the range of ꢀ25
Catalysts
DH*(kcal/mol)
DS* (e.u.)
to ꢀ35 eu. Values of ꢁ ꢀ50 eu imply a very highly ordered
transition state as is sometimes found for constrained cyclic
transition states. For example, 2 + 2 cycloadditions in
(p-MeOC6H4)2Sn(OiPr)2
(Ph)2Sn(OiPr)2
(p-MeC6H4)2Sn(OiPr)2
(p-FC6H4)2Sn(OiPr)2
(p-CF3C6H4)2Sn(OiPr)2
(n-Bu)2Sn(OiPr)2
12.1(5)
11.4(5)
10.9(5)
11.2(5)
10.4(5)
10.0(5)
ꢀ53(5)
ꢀ50(5)
ꢀ49(5)
ꢀ44(5)
ꢀ48(5)
ꢀ47(5)
¼
organic chemistry generally have DS values of 35 5 eu
[24]. Recent computations by Gibson et al. [25,26] on the
ROP of LA by Mg and Sn alkoxides would seem to sup-
Fig. 7. Plots of kobs vs. [cat] (a) and lnkobs vs. ꢀln[cat] (b) in the ROP of L-lactide by different concentrations of Ph2Sn(OPri)2 at 70 ꢁC.

4444
M.H. Chisholm et al. / Polyhedron 26 (2007) 4436–4444
[7] N. Spassky, M. Wisniewski, C. Pluta, A. LeBorgne, Macromol.
Chem. Phys. 197 (1996) 2627.
[8] N. Spassky, C. Pluta, V. Simic, M. Thiam, M. Wisniewski, Macro-
mol. Symp. 128 (1998) 39.
[9] C.P. Radano, G.L. Baker, M.R. Smith III, J. Am. Chem. Soc. 122
(2000) 1552.
port this view but in our present study we have not evalu-
ated the influence of solvent.
4. Supplementary material
[10] T.M. Ovitt, G.W. Coates, J. Am. Chem. Soc. 124 (2002) 1316.
[11] P. Hormnirun, E.L. Marshall, V.C. Gibson, A.J.P. White, D.
Williams, J. Am. Chem. Soc. 126 (2004) 2688.
[12] Z. Zhong, P.J. Dijkstra, J. Feijen, J. Am. Chem. Soc. 125 (2003)
11291.
CCDC 637509, 637508 and 637510 contain the supple-
mentary crystallographic data for (p-Me2NC6H4)3SnOiPr,
Ph2Sn(OiPr)2 and (p-FC6H4)2Sn(OiPr)2. These data can
be obtained free of charge via http://www.ccdc.cam.ac.
uk/conts/retrieving.html, or from the Cambridge Crystal-
lographic Data Centre, 12 Union Road, Cambridge CB2
1EZ, UK; fax: (+44) 1223-336-033; or e-mail: deposit@
ccdc.cam.ac.uk.
[13] Z. Tang, X. Chen, X. Pang, Y. Yang, X. Zhang, X. Jing,
Biomacromolecules 5 (2004) 965.
[14] M.H. Chisholm, Z. Zhou, Chem. Commun. (2005) 127.
[15] M.H. Chisholm, E.E. Delbridge, New J. Chem. 27 (2003) 1167.
[16] M.H. Chisholm, E.E. Delbridge, New J. Chem. 27 (2003) 1177.
[17] M.H. Chisholm, E.E. Delbridge, J.C. Gallucci, New J. Chem. 28
(2004) 145.
Acknowledgements
[18] DENZO: Z. Otwinowski, W. Minor, in: C.W. CarterJr., R.M. Sweet
(Eds.), Methods in Enzymology, vol. 276: Macromolecular Crystal-
lography, part A, Academic Press, 1997, pp. 307–326.
[19] (a) SHELXS-86 G.M. Sheldrick, Acta Cryst. A46 (1990) 467–473;
(b) G.M. Sheldrick, SHELXS-97, Universitat Gottingen, Germany,
1997.
We thank the Department of Energy, Office of Basic Sci-
ences, Chemistry Division for support of this work.
References
[20] G.M. Sheldrick, SHELXL-97, Universitat Gottingen, Germany, 1997.
[21] WINGX-Version 1.64.05 L.J. Farrugia, J. Appl. Cryst. 32 (1999) 837–
838.
[1] O. Dechy-Cabaret, B. Martin-Vaca, D. Bourissou, Chem. Rev. 104
(2004) 6147.
[22] International Tables for Crystallography, Volume C. Dordrecht:
Kluwer Academic Publishers, 1992.
[2] F. Neederberg, E.F. Connor, T. Glausser, J.L. Hedrick, Chem.
Commun. (2001) 2006.
[23] C.K. Williams, L.E. Breyfogle, S.K. Choi, W. Nam, V.G. Young,
M.A. Hillmyer, W.B. Tolman, J. Am. Chem. Soc. 125 (2003)
11350.
[24] R.B. Woodward, R. Hoffmann, The Conservation of Orbital Sym-
metry, Academic Press, New York, 1969.
[25] E.L. Marshall, V.C. Gibson, H.S. Rzepa, J. Am. Chem. Soc. 127
(2005) 6048.
[26] A.P. Dove, V.C. Gibson, E.L. Marshall, H.S. Rzepa, A.J.P. White,
D.J. Williams, J. Am. Chem. Soc. 128 (2006) 9834.
[3] F. Neederberg, E.F. Connor, M. Moeller, T. Glauser, J.L. Hedrick,
Angew. Chem., Int. Ed. 40 (2001) 2712.
[4] G.W. Nyce, T. Glauser, E.F. Connor, A. Mo¨ck, R.M. Waymouth,
J.L. Hedrick, J. Am. Chem. Soc. 125 (2003) 3046.
[5] B.J. O’Keefe, M.A. Hillmyer, W.B. Tolman, J. Chem. Soc., Dalton
Trans. (2001) 2215.
[6] M.H. Chisholm, Z. Zhou, in: L. Baugh, J. Canich (Eds.), Stereose-
lective Polymerization with Single Site Catalysts, Taylor and Francis
publishers, 2007.