Kinetic and biocompatibility investigation on the catalytic ring opening polymerization of L-lactide in bulk using cyclic Bu2Sn initiators derived from ethylene glycol, pentaerythritol and cloisite 30B

Article abstract of DOI:10.1016/j.poly.2019.114202

In the present work, considering a green approach, PLLA was synthesized by the ring-opening polymerization of L-LA using various cyclic initiators (2,2-dibutyl-2-stanna-1,3-dioxacyclo ethane (1), spirocyclic tin (3) and cloisite modified tin (5)). PLLA was characterized by size exclusion chromatography, differential scanning calorimetry, ¹H and ¹³C NMR, XRD and MALDI-TOF. The molecular weight distribution was narrow (<1.5) for both low and high molecular weights (>85,000 g/mol). An X-ray study showed the crystalline nature of PLLA, irrespective of the different types of initiators and polymerization conditions. A kinetic investigation showed that all polymerizations were first order with respect to the monomer and no termination reactions occurred during the polymerization. MALDI-TOF spectra of PLLA chains doped with Na⁺ and K⁺ cations show that the PLLA prepared using initiators 1 and 3 have hydroxyl and carboxyl end groups without incorporation of tin. The MALDI-TOF spectrum of PLLA prepared using initiator 5 also enabled the simultaneous detection of the linear and cyclic structure of PLLA without tin. Biocompatible and biodegradable PLLA was obtained, which could be used for biomedical applications.

Full text of DOI:10.1016/j.poly.2019.114202

Journal Pre-proofs
Kinetic and Biocompatibility Investigation on the Catalytic Ring Opening Pol-
ymerization of L-Lactide in Bulk Using Cyclic Bu Sn Initiators Derived from
2
Ethylene Glycol, Pentaerythritol and Cloisite 30B
Kavita K. Mohite, Baijayantimala Garnaik
PII:
S0277-5387(19)30647-3
DOI:
https://doi.org/10.1016/j.poly.2019.114202
POLY 114202
Reference:
To appear in:
Polyhedron
Received Date:
Revised Date:
Accepted Date:
10 July 2019
16 October 2019
20 October 2019
Please cite this article as: K.K. Mohite, B. Garnaik, Kinetic and Biocompatibility Investigation on the Catalytic Ring
Opening Polymerization of L-Lactide in Bulk Using Cyclic Bu Sn Initiators Derived from Ethylene Glycol,
2
Pentaerythritol and Cloisite 30B, Polyhedron (2019), doi: https://doi.org/10.1016/j.poly.2019.114202
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover
page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will
undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing
this version to give early visibility of the article. Please note that, during the production process, errors may be
discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
©
2019 Published by Elsevier Ltd.

Kinetic and Biocompatibility Investigation on the Catalytic Ring Opening
Polymerization of L-Lactide in Bulk Using Cyclic Bu Sn Initiators Derived
2
from Ethylene Glycol, Pentaerythritol and Cloisite 30B
a,b
a,b
Kavita K. Mohite , Baijayantimala Garnaik *
aPolymer Science and Engineering Division, CSIR-National Chemical Laboratory, Pune-411008,
India.
bAcademy of Scientific and Innovative Research (AcSIR), Ghaziabad- 201002, India.
Corresponding Author: Dr. Baijayantimala Garnaik; E-mail: b.garnaik@ncl.res.in
Tel.: +91-20-25902304; Fax: +91-20-25902615
ABSTRACT
In the present work, considering a green approach, PLLA was synthesized by the ring-
opening polymerization of L-LA using various cyclic initiators (2,2- dibutyl-2-stanna-1,3-
dioxacyclo ethane (1), spirocyclic tin (3) and cloisite modified tin (5)). PLLA was
1
characterized by size exclusion chromatography, differential scanning calorimetry, H and
13C NMR, XRD and MALDI-TOF. The molecular weight distribution was narrow (< 1.5)
for both low and high molecular weights ( 85,000 g/mol). An X-ray study showed the
crystalline nature of PLLA, irrespective of the different types of initiators and
polymerization conditions. A kinetic investigation showed that all polymerizations were
first order with respect to the monomer and no termination reactions occurred during the
+
+
polymerization. MALDI-TOF spectra of PLLA chains doped with Na and K cations show
that the PLLA prepared using initiators 1 and 3 have hydroxyl and carboxyl end groups
1

without incorporation of tin. The MALDI-TOF spectrum of PLLA prepared using initiator 5
also enabled the simultaneous detection of the linear and cyclic structure of PLLA without
tin. Biocompatible and biodegradable PLLA was obtained, which could be used for
biomedical applications.
Keywords: Poly (L-Lactide); cyclic tin initiator; spirocyclic; cloisite modified tin; MALDI-
TOF
2

Introduction
In recent years, environmentally friendly polymers, such as Poly (L-Lactide) (PLLA), have
attracted tremendous attention because of their biocompatibility, biodegradability and
permeability. They also aid reduction of carbon dioxide emissions which induce global
warming. PLLA is also produced from renewable resources [1,2], which is an alternative to
the exhaustion of petroleum resources. PLLA is widely used for biomedical applications,
such as in surgery, drug delivery systems, surgical sutures, implants (internal bone fixation)
and packaging [3]. PLLA is derived from biomass and is being explored as a potential
alternative to petroleum-derived polymers such as polyolefin, polyester and polystyrene [4-
7
]. Efforts have been made to synthesize PLLA through catalyst development and initiator
modification [8]. However, there are still great challenges to modify and enhance the
properties of the resulting materials.
Cyclic initiators prepared from dibutyltin oxide and various alcohols have been reported as
initiators in the ring opening polymerization (ROP) of various lactones and lactides [9-12].
It is observed that the polymerization mechanism involves Sn-O bonds as active species.
Again, the reactivity of the initiator increases with the number of attached alkoxide groups
to the tin atom and decreases with the increasing bulkiness of the alkoxide groups [13]. The
literature reports demonstrate that six-member cyclic dibutyltin alkoxides are interesting
initiators for β-D, L-butyrolactone or ε-caprolactone [14,15]. The insertion of lactones into
both Sn-O bonds produces macrocyclic polylactones in high yields, and the variation of the
monomer/initiator ratio allows easy control of the average ring size. Kricheldorf et al [16]
have shown that 10-membered rings containing tin initiate the ROP of L-Lactide (L-LA)
and racemic D, L-Lactide (D, L-LA) at 120 °C. The resulting molecular weights are
observed to be higher than expected from the monomer/initiator ratio and are not parallel to
it. The key aspect of this polymerization is the formation of macrocycles and the absence of
3

a back-biting reaction. As demonstrated in the literature reports [17], cyclic initiators allow
the synthesis of macrocyclic lactones, telechelic linear polylactones with two hydroxyl end
groups, A-B-A triblock copolymers and random copolyesters.
The ROP of L-LA has been initiated with cyclic tin alkoxide, where the microstructure and
stereo sequences of poly L-lactic acid are discussed [18]. Homo and copolymers of (D, L)-
β-butyrolactone and ϒ-butyrolactone or β-valerolactone have been synthesized using cyclic
tin alkoxide [19]. Biodegradable cyclic polyesters and tadpole-shaped polyesters have been
synthesized in the presence of cyclic tin alkoxide [20]. Ring expansion polymerization of β-
D, L-butyrolactone or Ԑ-caprolactone has been initiated with 2,2-dibutyl-2-stanna-1,3-
dioxapane and results in polylactones which are toxic for biomedical applications [21]. L-
LA macromonomers have been successfully synthesized using functionalized cyclic tin
alkoxides in order to obtain brush-like structures [22]. Dihydroxy terminated PLLA has
been obtained by controlled ROP and the control of the polymerization mechanism has also
been investigated [23]. The aggregation of active centers in metal alkoxide initiated lactones
or lactide in solution is a commonly reported phenomenon. The reactivity of the aggregated
and non-aggregated species differs, and as a result, the kinetics of the polymerization are
influenced. The propagation commonly proceeds via the non-aggregated species while the
aggregated complex is temporarily terminated and thus inactive towards polymerization.
In the literature, it has not been ascertained whether the broad distribution occurred due to
low solubility or the reactivity of these initiators. There is also a large difference between
solution studies and melt polymerization studies.
The objective of this study was to examine in depth, the ring opening polymerization of L-
LA with 1,1,6,6-tetra-n-butyl-1,6-distanna-2,5,7,10-tetraoxacyclodecane (1/2), spirocyclic
tin (3/4) and Cloisite modified tin (5) as water-tolerant initiators in bulk polymerization
4

(green route). The advantageous properties of these initiators are: inexpensive, non-
hygroscopic, non-explosive in nature, free from impurities such as water and organic acids,
thermally stable, can be stored for long periods without adopting any stringent conditions,
easy to handle and non-aggregation in bulk ring opening polymerization. Detailed
characterization of PLLA was performed using various techniques, such as MALDI-TOF
1
3
MS, quantitative C NMR analysis, size exclusion chromatography (SEC), wide-angle X-
ray diffraction (XRD) analysis and DSC. The nature of the end groups of PLLA was
1
3
determined with MALDI-TOF MS and C NMR spectroscopy. A detailed kinetic study and
the efficacy of these initiators for the ring opening polymerization of L-LA in bulk were
explored. The synthesized PLLA can be used as a biocompatible and biodegradable polymer
in biomedical applications.
STRUCTURES
H2 H2
O
O
CH2
CH2
O
O
C
C
O
O
Sn
1
Sn
Sn
C
2
C
H
H
2
2
O
Sn
O
O
O
O
O
O
Sn
O
O
O
Sn
O
Sn O
3
4
O
Sn
N
T
O
5
5

Mehrkhodavandi et al. prepared a highly active indium catalyst for living lactide
polymerization, where they showed the functional group tolerant polymerization and also
catalyst recovery [24]. The spiro polymer has been synthesized, characterized and found
show a lack of solubility, melting etc [25]. The mono and bifunctional spiro orthoesters have
been synthesized using a cationic ROP technique [26]. Organic spirocyclic initiators have
been used for the ring-expansion polymerization of -lactones [27].
Materials and methods
Materials. Dibutyltin oxide (98 %) and pentaerythritol (99 %) were purchased from
Aldrich. Ethylene glycol was purchased from Spectrochem. Cloisite 30B was purchased
from Southern Clay. All the materials were used without further purification. L-LA was
purchased from Purac and recrystallized once from ethyl acetate. The monomer (L-LA) was
dried and stored in a dry glove box.
Preparation of the initiators
1
,1,6,6-Tetra-n-butyl-1,6-distanna-2,5,7,10-tetraoxacyclodecane (1/2) was prepared
using dibutyltin oxide (4.05 g, 0.0163 mol) and ethylene glycol (0.92 mL, 1.021, 0.0163
mol, 1.0 equiv.). The reactants were taken in a three-neck flask containing dry toluene (100
mL) equipped with a Dean and Stark moisture trap. The reaction mixture was stirred under
reflux conditions (dibutyltin oxide dissolves) until a constant volume of 0.3 ml liquid had
collected in the trap.
The flask containing the product (a clear yellow toluene solution) was gradually cooled to
room temperature. The solution deposited white crystals which were isolated by filtration,
washed with toluene and dried under vacuum; yield 4.579 g (96%). These were
recrystallized from methylene chloride. The M.p. is 223-226.5 C (lit. 223-229). Exposure
of the product to air, overnight, did not affect the M.p. Found: C, 41.36; H, 7.45; Sn,
6

4
5
0.45%; mol.wt. in CHCl 650. C H O Sn Cald.: C, 40.99; H, 7.57; Sn, 40.52%; mol. wt.
3, 20 44 4 2
86). The 119Sn NMR spectrum of initiator 1 was recorded and the structure was confirmed.
3
,3,9,9-Tetrabutyl-2,4,8,10-tetraoxa-3,9-distannaspiro[5,5]undecane (3/4) was prepared
-2
according to the literature [14,23-25]. Dibutyltin oxide (2.48 g, 9.962 x 10 mol) and
-2
pentaerythritol (0.68 g, 4.981 x 10 mol dried over P O ) were suspended in dry toluene
4
10
(100 mL) and refluxed until the theoretical amount of by-product water was removed by
azeotropic distillation over a period of 24h. When the clear solution was cooled, the product
precipitated. The crude product was isolated by filtration. It was dissolved in hot toluene,
the insoluble fraction was removed by filtration over dry Celite and the product was isolated
by cooling (in the refrigerator overnight) and filtration. Yield: 1.988 g (88%). M.p. : 183-
1
85 C. Anal. Cald. for C H O Sn (597.96): C, 42.18; H, 7.42. Found: C, 41.56; H, 7.83.
2
1
44
4
2
The 119Sn NMR spectrum of initiator 3 was recorded and the structure was confirmed.
Cloisite 30B modified tin (5) was prepared by condensing (1.443 g) cloisite 30B, where T
is Tallow (65% C ; 30% C ; 5% C ) and (1 g) dibutyltin oxide in dry toluene at 110 °C
1
8
16
14
for 7h. The reaction mixture was cooled at room temperature and filtered. The crude product
was washed with n-hexane several times and dried using a vacuum to obtain a constant
weight. The yield was 2.229 g (95 %). The 119Sn NMR spectrum of the modified cloisite
3
0B initiator 5 was recorded and the structure was confirmed. No M.p. was observed. The
TGA curve of the modified Cloisite 30B catalyst shows a first decomposition temperature at
93 C.
2
Polymerization
7

L-LA (1g, .0069mol) and the requisite mole of initiators were weighed using a dry box
model No-UniLabMBroun-MB-OX-SE-1, S.L.-7120) into a reactor ampoule which had
silanized glass walls (pre-treated with Me SiCl ). The break seal technique was used for all
(
2
2
polymerization reactions. The sealed ampoule was completely immersed into a preheated
fluidized sand bath reactor. All other polymerizations catalyzed by tin modified initiators
were performed analogously. The reaction mixture was cooled to room temperature and 100
mL of chloroform were added. The resultant solution was filtered to remove the residual
initiator. The filtrate was poured into 400 mL of n-hexane for precipitating the polymer. The
resultant polymer was collected by filtration and further purified by repeated dissolution and
precipitation. A similar procedure was followed to remove the cloisite modified tin initiator.
Analysis
The molecular weights (M and M ) and dispersity were determined with respect to
n
w
polystyrene standards by size exclusion chromatography (SEC) on a Waters 150C machine
at 25 °C by eluting the PLLA solutions (10 mg/ mL in CHCl ), with toluene as an internal
3
5
4
3
standard, through a series of µ-Styragel columns (30 cm length) of pore sizes 10 , 10 , 10 ,
5
00 and 100 Å respectively. Chloroform was used as the mobile phase (flow rate 1 mL/
min) and a refractive index. detector (spectra series RI-150) was used for the detection of
the different molecular weight fractions. The correction factor is known in the literature to
be 0.58 [28].
1
Samples for H NMR spectroscopy were prepared in CDCl in 5 mm diameter NMR tubes
3
1
3
at room temperature. The sample concentration for C NMR measurements was 10 % by
1
3
weight. Proton decoupled C NMR spectra with NOE were recorded on a Bruker DRX 500
1
3
MHz NMR spectrometer working at 125.577 MHz for Carbon 13. C NMR spectroscopy
was also performed on a Bruker DRX 500 MHz NMR spectrometer in 10 mm diameter
8

1
3
NMR tubes. CDCl was used as the solvent and TMS as an internal standard for all C
3
NMR measurements. The relative peak areas were proportional to the number of carbon
atoms. The peak areas were calculated by the deconvolution method using the WIN-NMR
software. 119Sn NMR was also studied for the cloisite modified initiator.
The thermal stability was analyzed using a Perkin-Elmer TGA-7, by heating the samples
from 50 to 700 °C with a heating rate of 10 °C/ min under a nitrogen atmosphere and a flow
rate of 50 mL/ min.
Differential scanning calorimetric (DSC) measurements were performed on a Perkin-Elmer
thermal analyzer model DSC-7 in a nitrogen atmosphere. The measurements were made
from -40 to 200 °C at a heating rate of 10 °C/ min and a cooling rate of 100 °C/ min. The
samples were further heated from -40 to 200 °C at a heating rate of 10 °C/ min. The melting
temperature(T ) and glass transition temperature (T ) were calculated from the second
m
g
heating curves. Representative second heating curves of PLLA-5, PLLA-10 and PLLA-14
are shown in Figs. 11a-c.
Wide-angle X-ray scattering (WAXS) patterns of the samples were obtained in the
reflection mode using a Rigaku Dmax 2500 diffractometer and Ni-filtered copper radiation.
The samples were scanned in the range 2θ = 10 – 35° and the generator was operated at 40
kV and 150 mA. The Full Width at Half Maximum (FWHM) of the 110 peaks was
determined by the peak fitting software available with the diffractometer.
MALDI-TOF MS analysis was performed on a Kratos Kompact MALDI-IV spectrometer
equipped with 0.7 m linear and 1.4 m reflection flight tubes as well as a 337 nm nitrogen
LASER of pulse width 3 ns. All experiments were carried out at an accelerating potential of
2
0 kV. In general, mass spectra from 200 shots were accumulated to produce the final
spectrum. The samples were dissolved in tetrahydrofuran (1 mg/ mL) and mixed with the
9

matrix (15 mg/ mL of THF) before drying on the sample plate. 2,4,6-
Trihydroxyacetophenone (THAP) was used as the matrix. The sample plate was inserted
-5
into the apparatus under high vacuum (~10 Pa).
Results and discussion
The initiators 1, 3 and 5 have been synthesized by several research groups using three
different synthetic methods [12,17,29]. In the present work, all three methods were applied
to synthesize these initiators. The reaction of dibutyltin oxide with ethylene glycol was
found to be useful for the synthesis of initiator 1. The product obtained from ethylene glycol
and the dibutyltin derivatives is mainly the 10 membered ring dimer 2 and not the
monomeric form 1, as shown in Scheme 1. 1 was used as an initiator for the polymerization
1
of L-LA. The structure was confirmed by H NMR (Fig. 1). The signal appearing at δ 1.66
ppm is due to the methylene group attached to the Sn atom.
Scheme 1. Synthesis of 1,1,6,6-tetra-n-butyl-1,6-distanna-2,5,7,10-tetraoxacyclodecane
(1/2)
Sn
O
O
O
CH2
CH2
-
H O
2
Sn
HO
HO
CH2
CH2
1
H2 H2
O
O
C
C
O
Sn
Sn
O
C
H
C
H
2
2
2
1
0

The 119Sn NMR spectrum in CDCl solution exhibited two signals. The experimental results
3
showed a sharp signal at δ -176.74 ppm (relative to internal (CH ) Sn). The 119Sn NMR
3
4
measurement also suggested that the concentrated solution of initiator 1/2 in chloroform
involved an association equilibria in addition to the dimerization because of the 119Sn signal
was unusually broad, as shown in Fig. 3a. There was no significant line broadening found in
1
the H NMR spectra.
The solid spiro compound 3 was used as an initiator for polymerization of L-LA (Scheme
1
2
). The intermediate formation of the spirocycle 4 and its structure were confirmed by H
NMR spectroscopy, as shown in Fig. 1B. A signal appeared at δ 1.67 ppm due to the
methylene group attached to the Sn atom. The 119Sn NMR spectrum (Fig. 3b) depicted a
sharp signal at δ -172.47 ppm, which confirmed the structure of initiator 3/4.
Scheme 2. Synthesis of spirocyclic initiator (3/4)
O
Sn
O
O
Sn
O
Sn
O
-
H O
2
3
HO
HO
OH
OH
O
O
O
O
Sn
Sn
O
O
O
O
4
1
1

Clay plays an important role in the improvement of mechanical, thermal and barrier
properties. However, the focus here is to study its catalytic activity in the ring opening
polymerization of L-LA. Therefore, clay modified compound 5 was prepared by condensing
cloisite 30B, where T is Tallow (65% C ; 30% C ; 5% C ), and dibutyltin oxide in toluene
1
8
16
14
1
3
as shown in Scheme 3. Initiator 5 was characterized by C NMR CP/MAS spectroscopy
and is shown in Fig. 2.
Scheme 3. Clay modified tin initiator prepared from dibutyltin oxide and Cloisite 30B (5)
Sn
O
O
-
H O
2
Sn
N
T
O
HO
HO
N
T
The reaction of the cyclic tin alkoxides with 1, 3-dithion-2-one has been studied [30]. 1, 3-
Dithion-2-one allows the selective removal of the dibutyltin group from the macrocycle,
keeping the ring closure non-toxic, and forms the biodegradable macrocyclic ester. The
cyclic tin alkoxides 1, 3 and 5 have also been mentioned in the literature.
1
3
Initiator 5 was characterized by C NMR CP/MAS spectroscopy and is shown in Fig. 2.
The peaks at δ 23.65, 25.28, 27.28 ppm are attributed to the presence of dibutyl groups
attached to the tin atom. The disappearance of the peak at ~ δ 70 ppm due to a hydroxyl
group and the appearance of methylene and methyl groups at δ 23.65 and 16.04 ppm,
confirmed the structure of the initiator. The 119Sn NMR spectrum of the modified clay
initiator 5 was recorded (Fig. 3c). The high field signal at δ -176.15 ppm was attributed to
1
2

the Sn atom coordinated by two oxygen atoms and two butyl groups. The compound
showed an unusually low field signal (δ -86.82 ppm) which might be attributed to the
presence of anionic species coordinated with the tin atom. Thermal gravimetric analysis
(
TGA) was used to obtain information regarding the thermal stability and the results are
depicted in Figs. 4a-c. The initiators 1, 3 and 5 were found to be thermally stable up to ~280
C.
°
1
3

1
Fig. 1. H NMR spectra of the initiators: A) 1; B) 3
1
3
Fig. 2. C NMR spectrum of initiator 5
1
4

Fig. 3. 119Sn NMR spectra of a) 1; b) 3 and c) 5
1
5

Fig. 4. TGA graphs of initiators a) 1/2; b) 3/4) and c) 5.
Ring opening polymerization of L-Lactide
The usefulness of the synthesized three initiators was evaluated in a series of
polymerizations with L-LA (Scheme 4a). The mechanism of the L-LA polymerization has
been clearly described and is also presented in Scheme 4b. In the bulk polymerization, the
reaction temperatures have to be above the melting point of the monomers. In our case, L-
LA at just above its melting point (~98 °C), i.e. in the range 110-115 °C, has a tendency to
induce crystallization in the newly prepared PLLA. This crystallization makes the remaining
melt monomers insoluble in high molecular weight PLLA. This issue may lead to a
1
6

reduction of the conversion rate of the monomer; in this regard, we have carried out ROP at
different temperatures, ranging from 130-180°C [31].
The tin initiators 1, 3 and 5 show solubility in the monomer melt. A moderate reaction
temperature was chosen to avoid all problems connected to the phase separation of the
reagents and to limit the number of transesterification reactions. Due to the good solubility
of the initiators at the polymerization temperatures, reproducible results have been observed
for three consecutive reactions. The molecular distribution of PLLA also supports the strong
ability of these initiators [32].
The reaction pathway for the formation of PLLA was enabled by the tin initiators 1, 3 and 5.
The proposed mechanism is shown in Scheme 4b, where L-LA is coordinated with the
initiator via a nucleophilic attack of the hydroxyl group of a trace amount of water present in
the system, leading to the insertion of L-LA into the metal-oxygen bond by rearrangement
of the electrons [33, 34]. It is assumed that the trace amount of water plays the role of a co-
initiator in combination with the normal coordination insertion mechanism, allowing the
essential interpretation of all the experimental observations. The prepared PLLA retains its
stereochemistry [33]. Further, these initiators in the ROP of L-LA were used in order to gain
insight into the kinetics of the reaction and the influence of the initiator structure and
polymerization conditions on the initiator activity and polymer properties.
Scheme 4a. Polymerization of L-lactide with cyclic Bu Sn initiators derived from ethylene
2
glycol, pentaerythritol and Cloisite 30B
1
7

CH3
O
C
O
C
CH3
1
H
H
O
O
C
H
OH
OH
o
n
1
50 C
O
CH3
O
3
C
H
O
o
1
50 C
n
O
CH3
O
C
CH3
O
C
CH3
5
O
C
H
O
H
O
C
H
OH
o
2
00 C
n
n
Scheme 4b. Proposed reaction mechanism for PLLA using a tin complex
O
O
Sn
H-O-H
H
O
O
O
H
O
O
O
O
Sn
O
Tin complex
O
O
O
O
O
O
O
L-Lactide
O
O
H
O
O
O
O
HO
OH
H
O
Sn
O
n
O
O
1
8

Kinetics of the L-LA polymerization reactions
The kinetics of the L-LA polymerization reactions were investigated for initiators 1, 3 and 5
1
by monitoring the reactions using H NMR spectroscopy. The percentage of conversions
from L-LA to PLLA was determined by comparing the intensity of the PLLA peak at δ 5.16
ppm to that of the L-LA monomer at δ 5.0 ppm.
The polymerizations were conducted for various times to reach a plateau value in the
conversion ratio curve, as shown in Fig. 5. The maximum conversion was obtained after 1h
for all three initiators and therefore the reaction time of 1h was kept constant. From the
conversion curve, it was established that the initiation was instantaneous.
Fig. 5. Conversion of L-LA as a function of reaction time, for polymerizations initiated with
the initiators 1, 3 and 5.
The effects of the polymerization temperature on the ROP of L-LA using initiators 1, 3 and
5
are shown in Tables 1, 2 and 3. All polymerizations were initiated with cyclic initiators (1,
1
9

3
and 5) and conducted in bulk because all attempts to initiate the polymerization of L-LA
below 130 °C resulted in oligomers. Initiators 1 and 3 are reactive enough to catalyze
quantitative polymerizations of L-LA in bulk. With relatively short times and low
temperatures, a kinetically controlled polymerization takes place.
The effect of temperature on the polymerization kinetics of L-LA was examined for
initiators 1 and 3 by comparing the activity over the temperature range 130-150 °C. In
presence of initiator 1, the [M]/ [I] ratio was 500 and conversion ≥ 90%. Initiator 3 showed
a similar trend where the [M]/ [I] ratio was 550 and conversion ≥ 85%. In both cases, the
molecular weights (M and M ) increased up to 150 °C 31. Initiator 5 decomposes at
n
w
approximately 250 C, therefore polymerization was possible up to 220 C. The
polymerization temperatures were varied from 140 to 220 °C, where the [M]/ [I] ratio was
3
50 and conversion ≥ 90%.
Table 1. Effect of polymerization temperatures on the ROP of L-LA using initiator 1
Polymer Temp. Conversiona) Mnb)
M cald. Mnd)
c)
PDI Tg
Tme)
n
No.
(°C)
(%)
(°C) (°C)
1
2
3
4
5
130
135
140
145
150
86
89
90
92
93
61920
64080
64800
66240
66960
26332
27086
27724
28130
28420
45400 1.17 49
46700 1.20 50
161
163
167
166
172
47800 1.21 50
48500 1.20 54
49000 1.23 58
L-LA Polymerization run in bulk for 1h and [M]/ [I] = 500; a) Conversion determined from H NMR spectra of
1
the product Calculated from the molecular weight of L-LA x [M]/ [I] x conversion; c) The calibration value of
b)
n n
calculated according to formula M = 0.58M
(GPC); d) Obtained from GPC analysis and calibrated against
M
n
e)
polystyrene standard; T
m
calculated from DSC.
2
0

Table 2. Effect of polymerization temperatures on the ROP of L-LA using initiator 3
Polymer Temp. Conversiona) Mnb)
M cald. Mnd)
c)
PDI
Tg
Tme)
n
No.
(°C)
(%)
(°C) (°C)
6
7
8
9
130
135
140
145
150
84
86
86
89
90
66528 26738
68112 27956
68112 29638
70488 31494
71280 33060
46100 1.21
48200 1.21
51100 1.21
54300 1.20
57000 1.21
46
47
50
56
60
162
162
164
163
175
1
0
L-LA Polymerization run in bulk for 1h and [M]/ [I] = 550; a) Conversion determined from H NMR spectra of
1
the product; b) Calculated from the molecular weight of L-LA x [M]/ [I] x conversion; c) The calibration value
n n
calculated according to formula M = 0.58M
(GPC); d) Obtained from GPC analysis and calibrated
of M
n
e)
against polystyrene standard; T
m
calculated from DSC.
Table 3. Effect of polymerization temperatures on the ROP of L-LA using initiator 5
c)
Mnd)
Tme)
Polymer
No.
Temp. Conversiona) Mnb)
M cald.
PDI Tg
n
(°C)
(%)
(°C) (°C)
1
1
1
1
1
1
2
3
4
5
140
150
180
200
220
50
57
74
99
99
25200 12992
22400 1.13 49
34400 1.14 55
38100 1.16 54
45600 1.20 54
167
178
169
172
162
28728 19952
37296 22098
49896 26448
49896 30740
53000 1.22 60
L-LA Polymerization run in bulk for 1h and [M]/ [I] = 350; a) Conversion determined from H NMR spectra of
1
the product; b) Calculated from the molecular weight of L-LA x [M]/ [I] x conversion; c) The calibration value
2
1

n n
calculated according to formula M = 0.58M
(GPC); d) Obtained from GPC analysis and calibrated
of M
n
e)
against polystyrene standard; T
m
calculated from DSC.
The value of M and M monotonously increases with a rise in temperature. At lower
n
w
temperatures, a large number of very low-molecular-weight oligomeric species
predominate. An increase in the temperature beyond 180 °C resulted in high molecular-
weight species [35]. A similar observation has been made for PLA macrocycles containing
tin atoms in their chains. The maximum M and M were obtained at 220 °C as 53,000 and
n
w
1
00,000 Da respectively. However, the M /M value is also increased. This is presumably
w n
due to the presence of macrocycles. The reaction temperature was selected as 200 °C on the
basis of the low molecular weight distribution and a higher melting temperature.

Ring-opening polymerization of L-LA was carried out using initiators 1 and 3 at 150 C and

5
at 200 C. The reaction time was varied from 0 to 60 minutes. The first series of
polymerizations were conducted with L-LA and initiator 1. The results are compiled in
Tables 1, 4 and 6. These polymerization reactions are rapid and yield more than 90% within
an hour. There is no indication of backbiting degradation with longer reaction times. The
molecular weights, M increase with the M/I ratio, but do not parallel the [M]/ [I] ratio.
n
The kinetics of the L-LA polymerization reactions was compared to earlier literature results.
The reaction conditions and the results are summarized in Tables 1 and 2. For initiators 1, 3
and 5, the monomer-to-initiator ratio [M]/ [I] was varied from 100/1 and 800/1. The
linearity showed that no termination occurred during the polymerization, i.e. the number of
propagating chains was constant throughout the reaction, and that the initiation was fast
[36].
2
2

The polymerization proceeded easily, and high monomer conversions were achieved under
the conditions used, as illustrated in Tables 4-7. Mild reaction conditions resulted in good
control of the molecular weights and molecular weight distribution of the polymer chains.
o
a)
Table 4. Ring-opening polymerization of L-LA in bulk with initiator 1 at 140 C
Polymer [M]/[I]b) Conversionc) Mnd)
M cald. Mnf)
e)
Tg
Tm
PDI
n
o
o
no.
(%)
( C) ( C)
1
1
1
6
7
8
9
100
200
400
800
97
96
98
94
13968 8990
27648 16298
56448 28942
108288 57536
15500 1.10 47
28100 1.16 51
49900 1.16 53
99200 1.17 60
161
163
171
172
1
a)
The reaction was conducted for 1h at 140 C; b) Molar feed ratio M=L-LA, I=1; c) Conversion determined
o
from H NMR spectra of the product; d) Calculated from the molecular weight of L-LA x [M]/ [I] x
1
conversion; e) The calibration value of M
calculated according to formula M
= 0.58M
(GPC); f) Obtained
n
n
n
from GPC analysis and calibrated against polystyrene standard.
Table 5. Ring-opening polymerization of L-LA in bulk with initiator 3 at 140 C a)
o
Polymer [M]/[I]b) Conversionc) Mnd)
M cald. Mnf)
e)
PDI Tg
Tm
n
o
o
no.
(%)
( C) ( C)
2
2
2
2
0
1
2
3
100
200
400
800
97
93
97
96
13968 17864
30800 1.10 59
163
164
165
172
26784 21692
55872 29174
37400 1.11 55
50300 1.14 56
84300 1.13 56
110592 48894
a)
The reaction was conducted for 1h at 140 C; Molar feed ratio M=L-LA, I=1; c) Conversion determined
o
b)
1
d)
from H NMR spectra of the product;
Calculated from the molecular weight of L-LA x [M]/ [I] x
2
3

conversion; e) The calibration value of M
n n
calculated according to formula M = 0.58M
(GPC); f) Obtained
n
from GPC analysis and calibrated against polystyrene standard
o
a)
Table 6. Ring-opening polymerization of L-LA in bulk with initiator 1 at 150 C
Polymer [M]/[I]b) Conversionc) Mnd)
M cald.
e)
Mnf)
PDI Tg
Tm
n
o
o
no.
(%)
( C) ( C)
2
2
2
2
4
5
6
7
100
200
400
800
98
95
94
93
14112
27360
54144
6612
11400
19200
36700
69700
1.14 49
152
161
163
168
11136
21286
1.23 50
1.22 52
1.19 60
107136 40426
a)
The reaction was conducted for 1h at 150 C; Molar feed ratio M=L-LA, I=1; c) Conversion determined
o
b)
1
d)
from H NMR spectra spectrum of the product; Calculated from the molecular weight of L-LA x [M]/ [I] x
conversion; e) The calibration value of M (GPC); f) Obtained
n
n n
calculated according to formula M = 0.58M
from GPC analysis and calibrated against polystyrene standard.
o
a)
Table 7. Ring-Opening Polymerization of L-LA in bulk with initiator 3 at 150 C
Polymer [M]/[I]b) Conversionc) Mnf)
M cald. Mnd)
e)
PDI Tg
Tm
n
no.
(%)
(oC) (oC)
2
2
3
3
8
9
0
1
100
200
400
800
97
97
94
96
13968 12876
22200 1.08 51
147
156
157
158
27936 18212
54144 26622
110592 47038
31400 1.12 53
45900 1.15 54
81100 1.14 61
a)
The reaction was conducted for 1h at 150 C; Molar feed ratio M=L-LA, I=3; c) conversion from H NMR
o
b)
1
of the product; d) Calculated from the molecular weight of L-LA x [M]/ [I] x conversion; e) The calibration
2
4

n n n
calculated according to formula M = 0.58M
(GPC); f) Obtained from GPC analysis and calibrated
value of M
against polystyrene standard.
The kinetics of the L-LA polymerization reaction for all three initiators proceeded according
to simple pseudo first order kinetics, as shown in equation 1.
d[LA]/dt=k[LA]…………eq. 1
x
where k = k [I] , k = rate of chain propagation, I = initiator and x = order of reaction. The
p
p
plot shows a linear relationship with pseudo first-order kinetics with respect to L-LA for all
the initiators.
The plots of ln[LA ]/ [LA ] versus the reaction time for the polymerization reaction using
0
t
the three different initiators 1, 3 and 5 are shown in Figs. 6a-c, where [LA ] is the initial
0
monomer concentration and [LA ] is the monomer concentration at reaction time t. The plot
t
shows straight lines, at least up to 97 % monomer conversion in all three cases. However,
the polymerization temperature in the case of initiators 1 and 3 was 150 °C, whereas the
polymerization temperature of initiator 5 was 200 °C. These results show that the
polymerizations are first order reactions with respect to the monomer, when initiated at 150
(1 and 3) and 200 °C (5). The negative intercept with the x-axis may be due to initiation by
a trace amount of moisture present in L-LA (Fig. 2c). A similar negative intercept with the
x-axis has been observed in the ring opening polymerization of L-LA in the presence of N.O
Guanidine Catalysts [37,38].The rate constants for initiators 1, 3 and 5 were obtained from
Fig. 6 as 0.106, 0.126 and 0.113 min, respectively. The higher activity observed for 3 is
consistent with a literature report [35].
2
5

2
6

Fig. 6. Semi-logarithmic plots of L-LA monomer with various initiators: a) 1; b) 3; c) 5.
The PLLA molecular weight depends on the monomer-to-initiator ratio. With monomer-to-
initiator ratios from 100 to 800, the conversions were high and a low molecular weight
distribution (MWD) was observed in all polymerizations.
Nuclear magnetic resonance
1
1
The polymers were characterized using H NMR spectroscopy. Representative H NMR
spectra of PLLA-2, PLLA-7 and PLLA-14, prepared using initiators 1, 3 and 5, are shown
in Figs. 7A-C. The methine and methyl peaks of PLLA appear at δ 5.16 and 1.56 ppm in all
cases, as reported for linear PLLA [2]. The end group is assigned to the –CHOH group,
which appears at δ 4.30 ppm in all cases. The peaks at δ 1.6 (doublet) and 5.0 ppm (quartet)
are due to the methyl and methine groups of L-LA which appear because of incomplete
conversion of L-LA to PLLA (PLLA-2 and PLLA-7). However, these peaks are absent in
the case of PLLA-14, indicating the complete conversion of L-LA to PLLA.
2
7

1
Fig. 7. H NMR spectra of A) PLLA-2; B) PLLA-7; C) PLLA-14
Size exclusion chromatography
Size exclusion chromatography was used to analyze all these polymers and commercial
narrow polystyrene standards were used for calibration. Representative elugrams of PLLA
in the presence of initiators 1, 3 and 5 are shown in Figs. 8a-c. The results are presented in
Tables 1-4. The number-average molecular weight M , obtained from the SEC
n
measurements, is plotted versus the [M]/ [I] ratio for the L-LA polymerizations initiated by
1
and 3. The molar masses with initiator 3 in Fig. 4 show a strong deviation because the Sn-
2
8

O bonds are less stable and might not have undergone a clean stoichiometric insertion into
the cyclic tin initiator 3. Relatively narrow PDI values were reported and supported better
control of the ROP reactions. A possible explanation could be rapid chain transfers of the
growing polymer chains at the expense of chain propagation. Similar results have been
reported for bridged bis(amidinate) ytterbium alkoxide systems employed in the ROP of -
CL and LA [36, 37]. There is a linear relationship over the entire range of [M]/ [I] ratios and
the M value is proportional to the [M]/ [I] ratio at 140 C, as shown in Fig. 9. These results
n
confirm that the M value of PLLA can be controlled by adjusting the [M]/ [I] ratio in the
n
polymerization of L-LA. There is also a linear relationship over the entire range of [M]/ [I]
ratios and the M value is proportional to the [M]/ [I] ratio at 150 °C initiated by initiators 1
n
and 3 (Fig. 10). The M value of the PLLA obtained at 150 C was controlled by adjusting
n
the [M]/ [I] ratio in the polymerizations of L-LA initiated with 1 and 3.
Fig. 8. Representative SEC elugrams of PLLA: a) prepared using initiator 1; b) prepared
using initiator 3 and c) prepared using initiator 5.
2
9