Migration and hydrolysis of hydrophobic polylactide plasticizer

Article abstract of DOI:10.1021/bm901157h

Hydrophobic plasticizer protects polylactide (PLA) against hydrolytic degradation but still migrates to aging medium and there undergoes further hydrolysis contributing to the spectrum of degradation products. PLA plasticized with hydrophobic acetyl tributyl citrate (ATC) plasticizer showed a slower degradation rate compared with pure PLA because of the increased hydrophobicity of the material. The enhanced bulk hydrophobicity also overcame the degradation enhancing effect of hydrophilic surface grafting. In addition to plasticization with ATC, some of the samples were also surface grafted with acrylic acid. The materials were subjected to hydrolysis at 37 and 60 °C for up to 364 days to compare the effect of hydrophobic and hydrophilic bulk and surface modifications. Although considered insoluble in water, the plasticizer was detected in the water solutions immediately upon immersion of the materials, and the relative abundance of the ATC degradation products increased with hydrolysis time.

Full text of DOI:10.1021/bm901157h

Biomacromolecules 2010, 11, 277–283
277
Migration and Hydrolysis of Hydrophobic Polylactide Plasticizer
Anders Ho¨glund, Minna Hakkarainen, and Ann-Christine Albertsson*
Department of Fibre and Polymer Technology, School of Chemical Science and Engineering, Royal
Institute of Technology, S-100 44, Stockholm, Sweden
Received October 9, 2009; Revised Manuscript Received November 4, 2009
Hydrophobic plasticizer protects polylactide (PLA) against hydrolytic degradation but still migrates to aging medium
and there undergoes further hydrolysis contributing to the spectrum of degradation products. PLA plasticized
with hydrophobic acetyl tributyl citrate (ATC) plasticizer showed a slower degradation rate compared with pure
PLA because of the increased hydrophobicity of the material. The enhanced bulk hydrophobicity also overcame
the degradation enhancing effect of hydrophilic surface grafting. In addition to plasticization with ATC, some of
the samples were also surface grafted with acrylic acid. The materials were subjected to hydrolysis at 37 and 60
°C for up to 364 days to compare the effect of hydrophobic and hydrophilic bulk and surface modifications.
Although considered insoluble in water, the plasticizer was detected in the water solutions immediately upon
immersion of the materials, and the relative abundance of the ATC degradation products increased with hydrolysis
time.
the migration of the plasticizer and its further fate in the
surrounding medium. The aim of the present work was therefore
to reveal the water-soluble migrant patterns of modified PLA
during hydrolysis. Our hypothesis was that hydrophobic plas-
ticizers, despite their low water solubility, migrate from the
polymer matrix to the surrounding water, where they subse-
quently hydrolyze and thereby contribute to the spectrum of
degradation products. To confirm this, PLA was plasticized with
acetyl tributyl citrate (ATC), and some of the samples were
surface grafted with acrylic acid (AA), and the materials were
thereafter subjected to hydrolysis in deionized water for up to
364 days. The water-soluble product patterns after different time
periods were analyzed by electrospray ionization mass spec-
trometry (ESI-MS), and mass loss, molar mass changes, and
changes in thermal properties of the materials were determined
in parallel.
Introduction
Controlling the degradation rate and release of degradation
products and other migrants are key issues during design of
degradable polymers. Among the most interesting degradable
materials is polylactide (PLA), which is also derivable from
renewable resources. PLA has, mainly because of its high cost,
historically been used primarily in biomedical applications,
where it has proven good biocompatibility and favorable
mechanical properties.1 Continuous progress in the manufactur-
ing methods has, however, led to economically feasible produc-
tion of PLA for packaging applications as well.2 The homopoly-
mer of PLA has some drawbacks compared with the traditional
polyolefins, for example, somewhat inferior barrier properties
and inherent brittleness, which could be an issue in some cases.
The possibility of synthesizing copolymers of PLA with
controlled composition and advanced architectures has provided
a variety of materials with specific properties and predetermined
degradation rates.3,4 Modification of PLA with different func-
tional groups and incorporation of drugs has also been dem-
onstrated to effect the degradation rate.5,6 We have in previous
work shown the effect of copolymer composition and macro-
molecular architecture on the release of monomeric and oligo-
meric degradation products.7-10 We have also shown that
surface modification affects the degradation process and the
degradation product patterns.11
However, despite the versatility provided by copolymerization
and surface modification, these techniques are not always
sufficient to fulfill all demands. PLA suffers from the drawback
of being rigid and hard. Plasticization is a common method to
improve the flexibility of a polymer and overcome inherent
brittleness and processing limitations. In a previous work, we
showed migration and hydrolysis of polyester-based plasticizers
from PVC.12,13 The effect of citrate ester plasticizers on the
PLA material properties has been previously studied, and there
are also a few reports on the hydrolytic degradation of these
materials.14,15 However, no attention has been paid to the low-
molar-mass products formed during degradation and especially
Experimental Section
Materials. PLA pellets (95% L-LA, 5% D-LA) were used as
received. Natureworks PLA was chosen to obtain a homogeneous and
reproducible material and to make the results comparable to previous
work.11,16 PLA pellets (∼8 g) were dissolved in 200 cm3 chloroform
(Fischer Scientific, HPLC grade) and subsequently solution-casted into
films on silanized glass molds. The solvent was allowed to evaporate,
after which the films were dried for 1 week at reduced pressure. We
obtained plasticized PLA films with 10% (w/w) plasticizer in the same
way by dissolving 7.2 g of PLA pellets and 0.8 g of ATC (g99%,
Fluka). Specimens in the form of circular discs with a diameter of 10
mm and a thickness of ∼200 µm were punched from the films, washed
with ethanol (99.5%), and dried under vacuum prior to grafting or
degradation. AA (99.5%, Acros) was distilled under vacuum just before
use and stored cold. Benzophenone (99%, Alpha Aesar) was used as
received.
Vapor-Phase Grafting. The grafting procedure has been described
in detail elsewhere.17 In brief, a glass reactor with two interconnected
cylindrical compartments was used in the grafting process. The PLA
substrate specimens were placed in one of the reactor compartments,
and AA monomer and benzophenone (M/I ) 10:1) were added in the
other reactor compartment. The reactor was evacuated and thereafter
filled with nitrogen according to a freeze-pump-thawing procedure
* Corresponding author. Tel: +46-8-790 82 74. Fax: +46-8-20 84 77.
E-mail: aila@polymer.kth.se.
10.1021/bm901157h
 2010 American Chemical Society
Published on Web 11/23/2009

278 Biomacromolecules, Vol. 11, No. 1, 2010
Ho¨glund et al.
Table 1. Polymer Properties before Hydrolysis Including Number-Average Molar Mass, Polydispersity, Melting Temperature,
Glass-Transition Temperature, and Degree of Crystallinity
sample
PLA
PLA-ATC
PLA-AA
Mn
PDI
Tm (°C)
Tg (°C)
wc (%)
100 800 ( 1900
92 800 ( 2800
72 500 ( 3000
67 600 ( 3500
1.68 ( 0.03
1.76 ( 0.02
1.66 ( 0.06
1.76 ( 0.06
144.0 ( 1.1
146.5 ( 0.5
147.9 ( 1.7
144.0 ( 2.0
49.4 ( 0.3
36.2 ( 1.0
53.9 ( 1.6
46.9 ( 3.6
24 ( 4
24 ( 2
24 ( 6
26 ( 3
PLA-ATC-AA
that was repeated three times. After the last evacuation, the reactor
was sealed and immersed into a water bath at 35 or 40 °C for the
plasticized and nonplasticized PLA samples, respectively. Following
an equilibration time of 5 min, the reactor was irradiated with UV light
for 20 min. The entire procedure was repeated after the films were
turned over to obtain grafted layers on both sides of the sample
substrates. The surface-grafted PLA samples were thereafter thoroughly
washed with deionized water and ethanol (99.5%) and dried under
reduced pressure prior to hydrolysis.
200 to 0 °C at a rate of 10 °C/min before being heated again from 0 to
200 °C at a rate of 10 °C/min. The melting temperatures, Tm, were
noted as the maximum values of the melting peaks from the first heating
scan. The glass-transition temperature, Tg, was taken at the midpoint
temperature of the glass transition. The approximate degree of crystal-
linity of the PLA samples was calculated according to eq 2
∆Hf
wc )
× 100
(2)
∆H0f
Hydrolysis. Four different PLA samples, pure PLA (PLA), plasti-
cized PLA (PLA-ATC), acrylic acid-grafted PLA (PLA-AA), and
plasticized acrylic acid-grafted PLA (PLA-ATC-AA), were subjected
to hydrolytic degradation in deionized water at 37 and 60 °C. Samples
were hydrolyzed in 20 mL vials containing 10 mL of water. Deionized
water was chosen over a phosphate-buffered solution as degradation
medium because salts are known to be detrimental to the sensitive
components of the ESI instrument mass analyzer. The sample vials
were sealed with septa and placed in a thermostatically controlled
incubator at 37 °C and 60 rpm rotation or in an oven at 60 °C. After
different time periods between 1 and 364 days, duplicate samples of
each material were withdrawn from the test environment, dried under
reduced pressure, and analyzed by various techniques. In addition, the
water-soluble products in the sample solutions were analyzed after each
hydrolysis time.
where wc is the degree of crystallinity, ∆Hf is the heat of fusion of the
sample, and ∆Hf0 is the heat of fusion of 100% crystalline polymer.
The value for ∆Hf0 was 93 J/g.18
Results and Discussion
We assessed the effect of plasticization and surface grafting
on the degradation rate and the release of water-soluble products
from PLA by analyzing samples and the sample solutions during
hydrolytic degradation in deionized water for up to 364 days.
The graft yield was ∼5%, and successful grafting was verified
by Fourier transform infrared spectroscopy (FTIR), contact angle
measurement, scanning electron microscopy (SEM), and atomic
force microscopy (AFM).11
The materials used and their properties prior to degradation
are presented in Table 1. Previous studies have shown that the
molar mass of PLA films are not significantly affected during
vapor phase grafting, regardless of the grafting time and the
presence of grafting monomer.17 The AA-surface grafted PLA
chains were, however, no longer soluble in chloroform, which
could have influenced the molar mass values of PLA-AA and
PLA-ATC-AA. The polymer names are denotations of their
respective modifications, for example, PLA-ATC-AA is plas-
ticized and surface modified PLA.
Analysis of Water-Soluble Migrants. The water-soluble
migrants in the water fractions after different hydrolysis times
and at different degradation temperatures were monitored by
electrospray ionization mass spectrometry (ESI-MS) to deter-
mine the effect of plasticization, hydrolysis time, and temper-
ature on the degradation product patterns. The water-soluble
degradation products released from pure and surface grafted
PLA have been described in detail in previous work.11 In
general, water-soluble lactic acid oligomers were detected for
pure PLA after 28 days of hydrolysis at 60 °C and after 133
days of hydrolysis at 37 °C. In the case of surface-grafted PLA,
grafted oligomers were detected already after 1 day at 60 °C
and after 7 days at 37 °C, and nongrafted oligomers were
observed after 1 day at 60 °C and after 28 days at 37 °C.
Moreover, the degradation product patterns of surface-grafted
PLA showed significant variation with hydrolysis time with the
formation of short and long AA-grafted lactic acid oligomers
as well as plain lactic acid oligomers.
Electrospray Ionization Mass Spectrometry. The water-soluble
products were analyzed with a Finnigan LCQ ion trap mass spectrom-
eter (Finnigan, San Jose, CA). Methanol (Fischer Scientific, super
gradient) was added to the samples (2:1 v/v), and the solutions were
continuously infused into the ESI ion source at a rate of 5 µL/min
using the instrument syringe pump. The LCQ ion source was operating
at 5 kV, and the capillary heater was set to 175 °C. Nitrogen was used
as nebulizing gas, and helium was used as damping gas and collision
gas in the mass analyzer. Positive ion mode was used for all analyses.
Size Exclusion Chromatography (SEC). The molar mass of the
polymers was determined by SEC. Chloroform (Fischer Scientific,
HPLC grade) with 5% methanol (v/v) was used as the eluent at a flow
rate of 1.0 mL/min, and the injection volume was 50 µL. The instrument
comprised a Waters 717 Plus autosampler and a Waters model 510
solvent pump equipped with a PL-ELS 1000 light scattering evaporative
detector and three PLgel 10 µm mixed B columns (300 × 7.5 mm)
from Polymer Laboratories. Calibration was performed with narrow
molar mass polystyrene standards, and Millennium software version
3.20 was used to process the data.
Mass Loss. Samples were withdrawn from the test environment after
different hydrolysis times, washed with deionized water, and gently
wiped with a tissue. The mass loss was determined after the samples
were dried for 2 weeks under reduced pressure (0.5 × 10-3 mbar) and
the dry mass (md) was compared at a specific time with the initial mass
(m0) according to eq 1
m0 - md
∆md )
× 100
(1)
m0
Differential Scanning Calorimetry (DSC). The thermal properties
of the materials were determined with a DSC (Mettler Toledo DSC
820 module) under nitrogen atmosphere. Approximately 5 mg of the
polymer was encapsulated in a 40 µL aluminum crucible without pin.
Samples were heated from 0 to 200 °C at a rate of 10 °C/min under a
nitrogen gas flow of 50 mL/min. Then, the samples were cooled from
Migration of Water-Soluble Products from Plasticized
Polylactide. Figure 1 shows the positive ESI-MS spectra of the
compounds that had migrated from plasticized PLA after 28
days of hydrolytic degradation at 60 °C in the mass range m/z
150-1000.

Hydrophobic PLA Plasticizer
Biomacromolecules, Vol. 11, No. 1, 2010 279
Scheme 1. Illustration of Possible Hydrolysis Reactions for Acetyl
Tributyl Citratea
a Top route may occur on all three butyl ester groups.
molecules was confirmed by GC-MS and 13C NMR, which
showed two separate peaks with similar mass spectra in the GC
chromatogram and intact C-C bonds, respectively (data not
shown). Furthermore, the low energy involved in the electro-
spray ionization allows intact complexes to be directly detected
in the mass spectrometer.19,20 Parallel ESI-MS on the related
plasticizer acetyl triethyl citrate showed the exact same behavior
with the formation of molecular pairs in both MeOH and MeOH/
H2O.
Figure 1. Positive ESI-MS spectrum of the migrants from PLA
plasticized with acetyl tributyl citrate after 28 days of hydrolytic
degradation at 60 °C in the mass range m/z 150-1000.
The appearance of the peaks corresponding to the plasticizer
already after 1 day of degradation clearly shows the migration
of the plasticizer without any significant degradation of the bulk
material. Moreover, two additional peaks with lower intensity
and an m/z difference of 42 were also seen in Figure 2. These
peaks indicate that some hydrolysis of the plasticizer had
occurred. Scheme 1 illustrates possible hydrolysis pathways for
ATC.
The peaks at m/z 361 and m/z 785 in Figure 2 correspond to
the products formed after hydrolysis of the ATC acetyl group
with the formation of acetic acid according to the bottom route
in Scheme 1. It is interesting to note that the plasticizer started
to migrate from the material more or less immediately upon
immersion into water and that hydrolysis of the ester arms was
directly initiated. We confirmed the susceptibility of ATC toward
hydrolysis in water by analyzing the pure plasticizer in a parallel
control experiment. Pure ATC was treated in the same manner
as during the film preparation and was dissolved in methanol,
and this solution, with and without the addition of water (MeOH/
H2O 2:1), was subsequently analyzed by ESI-MS. Peaks
corresponding to the ATC hydrolysis products were only
detected in the case of the MeOH/H2O solution, which confirms
that the hydrolysis of the ester arms was induced neither during
contact with CHCl3 or MeOH nor during the ionization in the
ESI instrument. When the hydrolysis time was increased, the
migrated ATC was further hydrolyzed according to Scheme 1.
This is shown in Figure 3, which shows the positive ESI-MS
spectra of the migrants from plasticized PLA after 182 days of
hydrolytic degradation at 60 °C in the mass range m/z
150-1000.
In this stage, the hydrolysis products of ATC were dominating
the ESI-MS spectrum. The peaks with the highest intensity
correspond to products formed after hydrolysis of the ATC
acetyl group (∆m/z ) 42). Moreover, two series of peaks with
a mass-to-mass peak decline of 56 were also observed. These
peaks correspond to products formed after the hydrolysis of
various numbers of butyl ester groups with the formation of
butanol (top route in Scheme 1). In addition, pure lactic acid
oligomer peaks were also observed in the mass spectrum, which
also showed degradation of the polymer matrix in this stage.
The further hydrolysis of ATC with degradation time and the
Figure 2. Positive ESI-MS spectrum of the migrants from PLA
plasticized with acetyl tributyl citrate after 1 day of hydrolytic degrada-
tion at 60 °C in the mass range m/z 150-1000.
This mass spectrum is similar to the corresponding mass
spectrum of pure PLA.11 The main series of peaks, marked LAn,
correspond to water-soluble lactic acid oligomers with n
repeating units terminated by carboxyl and hydroxyl end groups.
These peaks appear at m/z ) (1 + n × 72 + 17 + 23), and
their general chemical structure is shown in Figure 1. Oligomers
with up to 12 repeating units were observed in the spectrum,
which is in correlation with previously reported results after
the hydrolysis of pure PLA.11 In addition, three other distinct
peaks at m/z 403, 425, and 827 were seen, which were not
visible in the spectra from pure PLA. These peaks were detected
already after 1 day of degradation at both 37 and 60 °C, as
illustrated in Figure 2.
In this stage, no water-soluble PLA oligomers were detected
because of the short degradation time. The observed peaks
instead originate from the added plasticizer ATC. ATC has a
molar mass of 402.5 g/mol, and the additional peaks therefore
most likely correspond to ATC (m/z 403), ATC with attached
sodium (m/z 425), and two ATC molecules with attached sodium
(m/z 827). The hydrophobic nature of the plasticizer probably
favors the formation of noncovalently bound pairs of molecules
in polar solvents. The formation of pairs of intact ATC

280 Biomacromolecules, Vol. 11, No. 1, 2010
Ho¨glund et al.
Figure 3. Positive ESI-MS spectrum of the migrants from PLA
plasticized with acetyl tributyl citrate after 182 days of hydrolytic
degradation at 60 °C in the mass range m/z 150-1000.
Figure 5. Positive ESI-MS spectrum of the migrants from acrylic acid-
grafted PLA plasticized with acetyl tributyl citrate after 182 days of
hydrolysis at 60 °C in the mass range m/z 200-2000.
In addition to the ATC plasticizer, peaks corresponding to
water-soluble lactic acid oligomers with grafted AA, marked
LAnAA, were also observed. The suggested general chemical
structure of these oligomers is shown in Figure 4. In the case
of pure and plasticized PLA, water-soluble oligomers were not
detected until after 133 days of degradation at 37 °C. Therefore,
there was a large difference in the degradation process and
products of grafted and nongrafted plasticized PLA. The AA-
grafted oligomers are more hydrophilic than the nongrafted
oligomers and therefore migrate easier and faster to the
surrounding water solution. This has also been shown in a
previous study of pure PLA and AA-grafted PLA.11 In addition,
some longer AA-grafted oligomers are also visible in the higher
mass range. These peaks probably correspond to longer AA-
grafted lactic acid oligomers that are water-soluble because of
the increased hydrophilicity from the AA units. An assignment
of the degradation products corresponding to each of these peaks
was not performed because the chemical structure of the grafted
surface layer results in the formation of multiply charged ions
and ions with several sodium atoms in the ESI-MS. This, in
turn, leads to complex mass spectra where each hydrolysis
product can induce multiple peaks corresponding to ions with
different charges and a different number of sodium atoms.11
When the hydrolysis time and hydrolysis temperature were
increased, the intensity of these peaks increased in the mass
spectra. This is illustrated in Figure 5, which shows the ESI
mass spectrum recorded from the water fraction of grafted and
plasticized PLA after 182 days of degradation at 60 °C.
Prolonged degradation of the surface-grafted and plasticized
PLA induced a shift in the water-soluble product patterns. After
182 days of hydrolysis at 60 °C, the longer AA-grafted LA
oligomers dominated the mass spectrum, although peaks cor-
responding to the ATC plasticizer and its hydrolysis products
(ATC′) were still clearly discerned.
Figure 4. Positive ESI-MS spectrum of the migrants from acrylic acid-
grafted PLA plasticized with acetyl tributyl citrate after 7 days of
hydrolytic degradation at 37 °C in the mass range m/z 150-2000.
increase in its hydrolysis products were also confirmed by
analyzing pure ATC aged in water for ∼10 months at 60 °C
with ESI-MS. Although a quantification of the amount of the
respective degradation products was not performed, it is possible
to conclude that the ATC plasticizer migrates from the PLA
matrix and subsequently undergoes hydrolysis in the surrounding
water. This, in turn, entails a large difference in the degradation
product patterns of pure and plasticized PLA. Despite the fact
that migration of the plasticizer occurred, the plasticizer also
protected the material toward degradation leading to lower total
mass loss. (See the Mass Loss section.)
Migration of Water-Soluble Products from Acrylic
Acid-Grafted and Plasticized PLA. The water-soluble product
patterns of the acrylic-acid-grafted and plasticized PLA were
quite different from the product patterns of pure and plasticized
PLA. Figure 4 shows the ESI mass spectrum recorded from
the water fraction of AA-grafted and plasticized PLA after 7
days of degradation at 37 °C.
In summary, plasticizing with ATC, surface grafting with AA,
and degradation temperature and degradation time all had a large

Hydrophobic PLA Plasticizer
Biomacromolecules, Vol. 11, No. 1, 2010 281
Figure 7. Molar mass as a function of hydrolysis time at 37 °C for
PLA, PLA plasticized with acetyl tributyl citrate (PLA-ATC), PLA
grafted with acrylic acid (PLA-AA), and AA-grafted PLA plasticized
with ATC (PLA-ATC-AA).
highest mass loss of the various PLA materials. Analogously,
this is due to the increased hydrophilicity induced by the grafted
surface layer. The AA side chains facilitate water absorption
and accelerate the cleavage of the ester bonds in the PLA main
chain. Grafted and plasticized PLA had an intermediate mass
loss profile at 37 °C, which is somewhat slower than pure PLA
but somewhat faster than plasticized PLA. Apparently, the
increased bulk hydrophobicity caused by the plasticization had
a larger influence on the degradation rate compared with the
increased surface hydrophilicity caused by surface grafting.
Molar Mass Changes. The molar mass changes during
hydrolysis were determined with SEC. Figure 7 shows the molar
mass of the various PLA materials as a function of hydrolysis
time.
In general, the molar mass decreased with hydrolysis time
for all four materials. The surface-grafted PLA chains were,
however, no longer soluble in chloroform, which influenced the
molar mass values, and the grafted and nongrafted PLA are
therefore not directly comparable. Nevertheless, in conformity
with the mass loss results, the lowest molar mass decrease was
obtained for the plasticized PLA. It is interesting to note that a
relatively large decrease in molar mass was already detected
after 7 days of degradation at 37 °C, especially for the surface-
grafted PLA. The mass loss was considerably slower, and over
90% of the original mass remained in all materials after 28 days
of degradation. This is due to the hydrophobic nature of PLA.
The degradation products formed during hydrolysis are not
water-soluble until they have a molar mass of ∼1000 g/mol
(Figure 1) and therefore remain in the polymer bulk.
Figure 6. Remaining mass as a function of hydrolysis time at (a) 37
and (b) 60 °C for PLA, PLA plasticized with acetyl tributyl citrate (PLA-
ATC), PLA grafted with acrylic acid (PLA-AA), and AA-grafted PLA
plasticized with ATC (PLA-ATC-AA).
influence on the degradation rate and the water-soluble product
patterns of PLA. The ATC plasticizer and its hydrolysis products
were detected by ESI-MS already after 1 day of degradation at
both 37 and 60 °C. With increased temperature and time, the
migrated ATC was further hydrolyzed, and water-soluble lactic
acid oligomers were released from the PLA matrix. Grafting
with AA increased the hydrophilicity of the PLA substrate, and
grafted lactic acid oligomers were detected even at shorter
degradation times. Finally, longer AA-grafted oligomers were
predominantly formed after prolonged degradation of grafted
and plasticized PLA.
Mass Loss. We also studied the degradation process by
monitoring the mass loss after different hydrolysis times. Figure
6a,b shows the remaining mass as a function of hydrolysis time
at 37 and 60 °C, respectively.
As expected, the remaining mass decreased with hydrolysis
time for all four materials, and the process was substantially
faster at 60 °C than at 37 °C. The plasticized PLA showed the
lowest mass loss of all materials. This is most likely due to the
increased hydrophobicity after the addition of ATC. The
hydrophobic nature of ATC suppresses water absorption and
thereby decreases the degradation rate. This is in line with earlier
results on hydrolytic degradation of PLA with ATC as plasti-
cizer.14 However, the water-soluble migrants were not previously
analyzed. On the contrary, the AA-grafted PLA showed the
Thermal Properties. DSC was used to evaluate the changes
in thermal properties during hydrolysis. Figure 8 shows the
melting temperature of the different PLA materials as a function
of hydrolysis time.
All four materials had approximately the same melting
temperature prior to degradation. The melting temperatures
increased slightly during the first 49 days of degradation, after
which they decreased more rapidly. Similar to mass loss and
changes in molar mass, the lowest decrease in melting temper-
ature was observed for the plasticized PLA. The initial increase
in Tm is explained by continued crystallization during hydrolysis,
where the crystal thickness increased.21 When the degradation
time was increased, the melting temperatures decreased because
of the formation of shorter polymer chains resulting in lower
Tm. In parallel to the decrease in Tm, an increase in the degree
of crystallinity, wc, was also observed (Figure 9).

282 Biomacromolecules, Vol. 11, No. 1, 2010
Ho¨glund et al.
Figure 10. Glass transition-temperature as a function of hydrolysis
time at 37 °C for PLA, PLA plasticized with acetyl tributyl citrate (PLA-
ATC), PLA grafted with acrylic acid (PLA-AA), and AA-grafted PLA
plasticized with ATC (PLA-ATC-AA).
Figure 8. Melting point as a function of hydrolysis time at 37 °C for
PLA, PLA plasticized with acetyl tributyl citrate (PLA-ATC), PLA
grafted with acrylic acid (PLA-AA), and AA-grafted PLA plasticized
with ATC (PLA-ATC-AA).
material immediately upon immersion in water. In addition,
subsequent hydrolysis of the ATC ester arms was immediately
initiated upon contact with water leading to the formation of
new degradation products from the plasticizer. The migrated
plasticizer underwent continuous hydrolysis, and the relative
abundance of ATC degradation products increased with hy-
drolysis time. The addition of ATC decreased the degradation
rate of PLA as a result of the increased overall hydrophobicity
of the material. Water-soluble lactic acid oligomers were,
however, detected in the water solutions after the same
degradation periods as those for nonplasticized PLA, although
in lower amounts because the total mass loss caused by
degradation products and plasticizer migration was lower for
plasticized PLA. This was valid for both pure PLA and AA-
grafted PLA. The increased bulk hydrophobicity caused by
the plasticizer thus overcame the degradation-enhancing effect
from the hydrophilic surface layer. The fate of the plasticizer
with respect to the migration and formation of new degrada-
tion products as well as the possible effect on the degradation
rate should be taken into consideration when designing new
materials.
Figure 9. Degree of crystallinity as a function of hydrolysis time at
37 °C for PLA, PLA plasticized with acetyl tributyl citrate (PLA-ATC),
PLA grafted with acrylic acid (PLA-AA), and AA-grafted PLA plasti-
cized with ATC (PLA-ATC-AA).
It is well known that the amorphous regions are more
susceptible to hydrolysis than the crystalline regions, which leads
to an increase in wc during hydrolysis of semicrystalline
polyesters. The higher mobility of the shorter chains formed
during hydrolysis also allows for a reorientation of the crystalline
phase with a subsequent increase in wc. As expected, larger
differences between the materials were observed when compar-
ing the glass-transition temperatures, as seen in Figure 10.
As expected, the plasticized PLA materials had a lower glass-
transition temperature, Tg, than their nonplasticized analogues.
(See also Table 1.) Grafting with AA increased the Tg of the
PLA. This is probably due to the polarity of the grafted acid
groups, which increase the secondary interactions between the
polymer chains and restrict their mobility. The Tg values of all
four materials decreased with increased degradation time. This
is also due to the formation of shorter chains with higher
mobility. The Tg values were taken from the second heating
scan because the peak from the first heating scan was too ill-
defined to be evaluated.
Acknowledgment. We gratefully acknowledge the European
Community Sixth Framework Programme Sustainable Microbial
and Biocatalytic Production of Advanced Functional Materials
(BIOPRODUCTION) under the contract number NMP2-CT-
2007-026515 for financial support of this work.
References and Notes
(1) Albertsson, A.-C.; Varma, I. K. Biomacromolecules 2003, 4, 1466–
1486.
(2) Datta, R.; Henry, M. J. Chem. Technol. Biotechnol. 2006, 81, 1119–
1129.
(3) Hans, M.; Keul, H.; Moeller, M. Biomacromolecules 2008, 9, 2954–
2962.
(4) Leemhuis, M.; Kruijtzer, J. A. W.; Van Nostrum, C. F.; Hennink, W. E.
Biomacromolecules 2007, 8, 2943–2949.
(5) Seppa¨la¨, J. V.; Helminen Antti, O.; Korhonen, H. Macromol. Biosci.
2004, 4, 208–217.
(6) Tarvainen, T.; Malin, M.; Barragan, I.; Tuominen, J; Seppa¨la¨, J.;
Ja¨rvinen, K. Int. J. Pharm. 2006, 310, 162–167.
(7) Hakkarainen, M.; Adamus, G.; Ho¨glund, A.; Kowalczuk, M.; Alberts-
son, A.-C. Macromolecules 2008, 41, 3547–3554.
(8) Hakkarainen, M.; Ho¨glund, A.; Odelius, K.; Albertsson, A.-C. J. Am.
Chem. Soc. 2007, 129, 6308–6312.
(9) Ho¨glund, A.; Hakkarainen, M.; Kowalczuk, M.; Adamus, G.; Alberts-
son, A.-C. J. Polym. Sci., Part A: Polym. Chem. 2008, 46, 4617–
4629.
Conclusions
Plasticizing PLA with ATC significantly altered the water-
soluble product patterns during hydrolysis. Despite its hydro-
phobic nature, the plasticizer started to migrate from the PLA

Hydrophobic PLA Plasticizer
Biomacromolecules, Vol. 11, No. 1, 2010 283
(10) Ho¨glund, A.; Odelius, K.; Hakkarainen, M.; Albertsson, A.-C.
Biomacromolecules 2007, 8, 2025–2032.
(11) Ho¨glund, A.; Hakkarainen, M.; Edlund, U.; Albertsson, A.-C. Lang-
muir, published online Aug 24, http://dx.doi.org/10.1021/la902166j.
(12) Lindstro¨m, A.; Hakkarainen, M. Biomacromolecules 2007, 8, 1187–
1194.
(16) Ka¨llrot, M.; Edlund, U.; Albertsson, A.-C. Biomacromolecules 2007,
8, 2492–2496.
(17) Edlund, U.; Ka¨llrot, M.; Albertsson, A.-C. J. Am. Chem. Soc. 2005,
127, 8865–8871.
(18) Fischer, E. W.; Sterzel, H. J.; Wegner, G. Kolloid Z. Z. Polym. 1973,
251, 980–990.
(13) Lindstro¨m, A.; Hakkarainen, M. J. Appl. Polym. Sci. 2007, 104, 2458–
(19) Cole, R. B. J. Mass Spectrom. 2000, 35, 763–772.
(20) Loo, J. A. Mass Spectrom. ReV. 1997, 16, 1–23.
(21) Tsuji, H.; Ikada, Y. J. Appl. Polym. Sci. 1998, 67, 405–415.
2467.
(14) Kranz, H.; Ubrich, N.; Maincent, P.; Bodmeier, R. J. Pharm. Sci. 2000,
89, 1558–1566.
(15) Labrecque, L. V.; Kumar, R. A.; Dave, V.; Gross, R. A.; McCarthy,
S. P. J. Appl. Polym. Sci. 1997, 66, 1507–1513.
BM901157H