Poly(lactide) Stereocomplexes: Formation, Structure ...peters/WimBras/LuigiBalzano/2011_03_11...
Transcript of Poly(lactide) Stereocomplexes: Formation, Structure ...peters/WimBras/LuigiBalzano/2011_03_11...
Poly(lactide) Stereocomplexes: Formation, Structure,
Properties, Degradation, and Applications
Hideto Tsuji
Department of Ecological Engineering, Faculty of Engineering, Toyohashi University of Technology, Tempaku-cho,Toyohashi, Aichi 441-8580, JapanE-mail: [email protected]
Received: March 23, 2005; Revised: May 18, 2005; Accepted: May 20, 2005; DOI: 10.1002/mabi.200500062
Keywords: association; biodegradable; biomaterials; blends; polyesters; stereospecific polymers
1. Introduction
When the interaction between polymers having different
tacticities or configurations prevails over that one between
polymers with the same tacticity or configuration, a stere-
oselective association of the former polymer pair takes
place. Such association is described as stereocomplexation
or stereocomplex formation. A well known and typical
example of stereocomplexation is the one between isotactic
and syndiotactic poly(methyl methacrylate) (PMMA),
which was first reported by Fox et al. in 1958.[1] The first
example of stereocomplexation (stereoselective associa-
tion) for enantiomeric polymers, i.e. between R- and
S-configured (or L- and D-configured) polymer chains, was
reported by Pauling and Corey for a polypeptide in 1953.[2]
However, it seems that the detailed structures of the
polypeptide and its assemblies are unclear. With respect to
optically active polyesters, Grenier and Prud’homme[3]
Summary: Poly(lactide)s [i.e. poly(lactic acid) (PLA)] andlactide copolymers are biodegradable, compostable, produ-cible from renewable resources, and nontoxic to the humanbody and the environment. They have been used as bio-medical materials for tissue regeneration, matrices for drugdelivery systems, and alternatives for commercial polymericmaterials to reduce the impact on the environment. Sincestereocomplexation or stereocomplex formation betweenenantiomeric PLA, poly(L-lactide) [i.e. poly(L-lactic acid)(PLLA)] andpoly(D-lactide) [i.e. poly(D-lactic acid) (PDLA)]was reported in 1987, numerous studies have been carried outwith respect to the formation, structure, properties, degrada-tion, and applications of the PLA stereocomplexes. Stereo-complexation enhances the mechanical properties, thethermal-resistance, and the hydrolysis-resistance of PLA-based materials. These improvements arise from a peculiarly
strong interaction between L-lactyl unit sequences andD-lactyl unit sequences, and stereocomplexation opens anew way for the preparation of biomaterials such ashydrogels and particles for drug delivery systems. It wasrevealed that the crucial parameters affecting stereocom-plexation are the mixing ratio and the molecular weight ofL-lactyl and D-lactyl unit sequences. On the other hand,PDLAwas found to form a stereocomplex with L-configuredpolypeptides in 2001. This kind of stereocomplexation iscalled ‘‘hetero-stereocomplexation’’ and differentiated from‘‘homo-stereocomplexation’’ between L-lactyl and D-lactylunit sequences. This paper reviews the methods for tracingPLA stereocomplexation, the methods for inducing PLAstereocompelxation, the parameters affecting PLA stereo-complexation, and the structure, properties, degradation, andapplications of a variety of stereocomplexed PLA materials.
Macromol. Biosci. 2005, 5, 569–597 DOI: 10.1002/mabi.200500062 � 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Review 569
found stereocomplexation between enantiomeric poly(a-methyl-a-ethyl-b-propiolactone) (PMEPL), R- and
S-PMEPL, in 1984. Later, Ikada et al. reported stereo-
complexation between enantiomeric polylactide [i.e. poly-
(lactic acid) (PLA)], poly(L-lactide) [i.e. poly(L-lactic acid)
(PLLA)] and poly(D-lactide) [i.e. poly(D-lactic acid)
(PDLA)] in 1987.[4] Although these polyesters, PMEPL
and PLA, are a-substituted, Voyer and Prud’homme
reported stereocomplexation between b-substituted poly-
esters, poly(b-proriolactone)s, having different side groups(1,1-dichloroethyl or 1,1-dichloropropyl).[5] The typical
polymers having stereocomplexationability are summar-
ized in Table 1.[6] Stereocomplexation can occur between
isotactic and syndiotactic polymers or R- and S-configured
polymers having different chemical structures. The
reported examples are stereocomplexation between iso-
tactic PMMA and syndiotactic poly(methacrylic acid),[7]
between isotactic PMMA and syndiotactic poly(isobutyl
methacrylate),[8] and between PDLA and L-configured pep-
tides.[9] Slager and Domb described such kinds of stereo-
complexation between polymers with different chemical
structures as ‘‘hetero-stereocomplexation’’ and discri-
minated it from stereocomplexation between polymers
having identical chemical structures (‘‘homo-stereocom-
plexation’’).[9]
Stereocomplexation between PLLA and PDLA can
occur in solution, in a solid (bulk) state from the melt,
during polymerization, or during hydrolytic degradation, as
long as L-lactide (or L-lactyl) unit sequences and D-lactide
(or D-lactyl) unit sequences coexist in a system.[6] In other
words, PLA stereocomplexation can take place both in
enantiomeric PLA-based polymer blends and in non-
blended stereoblock PLA. Here, lactyl unit means a half
lactide unit. The synthetic procedures and structures of
PLLA, PDLA, and stereoblock PLA are given in Figure 1.
Stannous octoate (stannous 2-ethylhexanoate) and lauryl
alcohol (1-dodecanol) have been used frequently as initiator
and coinitiator, respectively, for the synthesis of PLA
homopolymers. These synthetic routes in the presence of
comonomers can be utilized to prepare copolymers having
L- or D-lactyl units.
When L-chains and/or D-chains are present in a system,
various types of crystallites can be formed (Figure 2).[19]
Figure 2 parts (a) and (b) show crystallites composed solely
of L-chains and D-chains, respectively. We call such
crystallites ‘‘homo-crystallites’’ and the formation of
homo-crystallites ‘‘homo-crystallization’’. In this case,
either L- or D-chains, not both, are present in the system.
However, when both L-chains and D-chains are present in a
system, three types of crystallites, Figure 2 parts (c)–(e),
can be formed. Figure 2(c) represents stereocomplex cry-
stallites (or racemic crystallites), where L-chains and
D-chains have a peculiarly strong interaction compared
with that between identical configurations and, therefore,
L- and D-chains are packed side by side. It should be noted
that the actual shape of the unit lattice of the PLA
stereocomplex showing in Figure 5 is different from that
showing in Figure 2(c).When an interaction between chains
having identical configurations prevails against that one
between polymers with different configurations, L-chains
and D-chains assemble separately to form homo-crystal-
lites, as illustrated in Figure 2(e). The formation of
crystallites where L- and D-chains are packed randomly, as
given in Figure 2(d), has not been reported so far. Such non-
selective packing is expected to occur when interactions
between chains with the same and different configurations
are identical. In the case of PLA, when L- and D-chains are
mixed non-equimolarly or when L- and D-chains have
relatively highmolecularweights, homo-crystallites aswell
as stereocomplex crystallites are formed.[6,20–24]
Stereocomplexation of PLA is composed solely of a
stereocomplex crystallization (racemic crystallization)
process, in marked contrast to stereocomplexation between
isotactic and syndiotactic PMMA, where an association
Hideto Tsuji is an Associate Professor at the Faculty of Engineering, Toyohashi University ofTechnology (Aichi, Japan). He was born in Osaka (Japan) in 1961. He received his Ph.D. degree inpolymer chemistry in 1992 from the Kyoto University (Japan) under the supervision of ProfessorYoshito Ikada. His Ph.D. work focused on the stereocomplexation between enantiomeric poly(lacticacid)s. After he finished his Ph.D. program, he moved to the Technology Development Center,Toyohashi University of Technology as a Research Associate in 1990. In 1996 as a Visiting Researcher,he joined the group of Professor William Bonfield and Dr. Raymond Smith at the Queen Mary andWestfield College, University of London (U.K.), where he studied the preparation methods ofbiodegradable porous materials. He became an Associate Professor at the Faculty of Engineering,Toyohashi University of Technology, in 1997. His research interests include the developments ofpoly(lactide)s and other biodegradable polyesters for a variety of applications and the manipulation oftheir physical properties and degradability. He has been investigating the synthesis, treatments, blend-ing, physical properties, crystallization, hydrolytic degradation, biodegradation, thermal degradation,and recycling of poly(lactide)s and other biodegradable polyesters. He has authored and coauthoredmore than 75 original research papers, 15 review articles, 5 patent articles, and 25 book chaptersincluding Chapter 5 in ‘‘Polyesters III (Biopolymers, vol. 4, Wiley-VCH, 2002)’’.
570 H. Tsuji
Macromol. Biosci. 2005, 5, 569–597 www.mbs-journal.de � 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Table 1. Stereocomplexationable polymers.[6]
Isomeric type Polymer type Polymer Reference
Syndiotactic and isotactic Vinyl polymer Poly(methyl methacrylate) (PMMA) Fox et al.[1]
Isotactic R- and S- (or L- and D-) Polyether Poly(tert-butylethylene oxide) Sakakihara et al.[10]
(chiral center in the main chain) Poly(epichlorohydrin) Signfield and Brown[11]
Polythioether Poly(tert-butylethylne sulfide) Dumas et al.[12]
Polyketone Poly(propylene-carbon monoxide) Jiang et al.[13]
Poly(1-butene-carbon monoxide) Jiang et al.[13]
Poly(allylbenzene-carbon monoxide) Jiang et al.[13]
Polyamide (Polypeptidic) Poly(g-methyl glutamate) Yoshida et al.[14]
Poly(g-benzyl glutamate) Tsuboi et al.[15]
Polyamide (Non-polypeptidic) Poly(hexamethylene-2,3-di-O-methyl-L-tartaramide)
Iribarren et al.[16]
Polyester Poly(a-methyl-a-ethyl-b-propiolactone) (PMEPL)
Grenier and Prud’homme[3]
Poly[b-(1,1-dichloroethyl)-b-propiolactone]
Voyer and Prud’homme[5]
Poly[b-(1,1-dichloropropyl)-b-propiolactone]
Voyer and Prud’homme[5]
Polylactide, polyl(lactic acid) (PLA) Ikada et al.[4]
Isotactic R- and S- (or L- and D-) Vinyl polymer Poly(g-methylbenzyl methacrylate) Hatada et al.[17]
(chiral center in the side chain) Poly(N-methacryloyl-L-leucinemethyl ester)
Sanda et al.[18]
Figure 1. Synthesis andmolecular structures of PLLA [(a) and (b)], PDLA [(c) and (d)], and stereo-block isotactic PLA[(e) and (f)].
Poly(lactide) Stereocomplexes: Formation, Structure, Properties, Degradation, and Applications 571
Macromol. Biosci. 2005, 5, 569–597 www.mbs-journal.de � 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
process of isotactic and syndiotactic PMMA is followed by
crystallization of the associated PMMA chain pairs.[25]
Most of the reported stereocomplexation between enantio-
meric polyesters are expected to proceed via the same
process as that for PLA stereocomplexation, i.e., a stereo-
complex crystallization process. With respect to PLA-
based stereocomplexation, the ratio of stereocomplex
crystallites to homo-crystallites is affected by numerous
parameters, such as the molecular weights and optical
purities of the polymers and themixing ratio of the isomeric
chains, as elucidated by intensive studies.[6,20–24,26–29]
Those parameters expected to affect PLA-based stereo-
complexation are listed in Table 2 and detailed information
is given below.
Since PLA and lactide copolymers are biodegradable,
compostable, producible from renewable resources, and
nontoxic to the human body and the environment, they have
been used as biomedical materials for tissue regeneration,
matrices for drug delivery systems, and alternatives for
commercial polymericmaterials to reduce the impact on the
environment.[30–37] PLA stereocomplexation due to the pe-
culiarly strong interaction between L-lactyl unit sequences
and D-lactyl unit sequences is expected to improve a
variety of properties of PLA-based materials and open
novel methods to prepare such materials. Although
numerous reviews or feature articles with respect to PLA
have been published, the PLA-based stereocomplex is
described only in parts of these articles,[9,30–37] excluding a
short feature article published in 2000.[6] This paper
reviews the methods for tracing PLA stereocomplexation,
the methods for inducing PLA stereocompelxation, the
parameters affecting PLA stereocomplexation, and the
structure, properties, degradation, and applications of a
variety of stereocomplexed PLA materials.
2.Methods forTracingPLAStereocomplexation
Numerous methods have been reported for tracing PLA
stereocomplexation. The representative methods are des-
cribed in this section and such information is required to
understand the subsequent sections.
2.1. Differential Scanning Calorimetry (DSC)
DSC is one of the most effective and simple methods for
monitoring PLA stereocomplexation. Figure 3 gives typical
DSC curves for the blends of PLLA and PDLAwith diffe-
rent mixing ratio (XD) values.[21] XD is defined as follows:
XD ¼ Weight of PDLA=Weight of PLLA and PDLA
ð1Þ
The specimens were prepared by precipitation of a
methylene chloridemixed solution of PLLA andPDLA into
stirred methanol. The peak at around 180 8C for PLLA or
PDLA (XD¼ 0 or 1) is ascribed to the melting of PLLA or
PDLA homo-crystallites, while a new melting peak at
around 230 8C appears in the blend specimens. The new
peak is attributed to the melting of stereocomplex crystal-
lites. A similar increase in melting temperature (Tm)
upon stereocomplexation was reported by Grenier and
Prud’homme for blends between R- and S-PMEPL.[3] The
enthalpyofmelting(DHm)ofstereocomplexcrystallitesgives
a maximum at mixing ratio XD of 0.5 and with a deviation of
XD from 0.5 the DHm of homocrystallites increases. This
reflects the fact that equimolar mixing is favored for
stereocomplexation.
Tsuji and Ikada have estimated the equilibrium melting
temperature (Tm0 ) of PLA stereocomplex crystallites by
extrapolation of Tm0 values for different optical purities,
which were obtained by the Hoffman-Weeks procedure
using experimental Tm values, to 100% optical purity.[23]
The estimated Tm0 value of 279 8C is much higher than the
205 8C (Tsuji and Ikada[38,39]), 212 8C (Tsuji and Ikada[40]),
and 215 8C (Kalbs and Pennings[41]) reported for homo-
crystallites of PLLA. The reported DHm values for the
crystals having an infinite thickness [DHm(100%)] for PLA
Figure 2. Crystallite models for enantiomeric L- and D-poly-mers.[19] (a) a homo-crystallite composed of L-polymer chains; (b)a homo-crystallite composed of D-polymer chains; (c) a stereo-complex (racemic) crystallite where L- and D-polymers are packedside by side; (d) a crystallite where L- and D-polymer chains arepacked randomly; (e) a mixture of (a) and (b).
572 H. Tsuji
Macromol. Biosci. 2005, 5, 569–597 www.mbs-journal.de � 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
stereocomplex, 142 J g�1 (Loomis et al.[42]) and 146 J g�1
(Tsuji et al.[43]) are in the range of DHm(100%) values for
homo-crystallites, 93 J g�1 (Fischer et al.[44]), 135 J g�1
(Miyata and Masuko[45]), 142 J g�1 (Loomis et al.[42]), and
203 J g�1 (Jamshidi et al.[46]). In our case, the DHm(100%)
for the PLA stereocomplex was calculated from the
experimental DHm value (102 J g�1) and the crystallinity
(Xc) value (70%) obtained with wide-angle X-ray scatter-
ing,[43] whereas the procedure for the calculation of
DHm(100%) for the PLA stereocomplex was not given in
the literature.[42] The physical properties, including the
detailed thermal properties of stereocomplexed PLA and
nonblended PLLA in comparison with other biodegradable
polyesters, are summarized in Table 3.[36]
2.2. Wide-Angle X-Ray Scattering (WAXS)and Small-Angle X-Ray Scattering (SAXS)
In the first report on PLA stereocomplexation, the WAXS
profiles of blends having different XD values are shown
(Figure 4) and in the same report DSC thermograms are
presented. Themain peaks of PDLA (XD¼ 1) film appear at
Table 2. The parameters affecting PLA stereocomplexation.
Molecular Structures Molecular Characteristics Procedures Parameters
1. PLLA/PDLA 1. Overall molecular weight 1. Bulk 1. Mixing ratio of polymers(for polymer blending)
2. L-lactide copolymer/D-lactide copolymer
2. Averaged sequence lengths(molecular weights) of L- andD-lactide (lactic acid) units(for linear copolymers).
1.1. Crystallization at a fixedtemperature from the meltor after melt-quenching
2. Melting temperature and time3. Crystallization or polymerization
temperature and time (1.1 and 1.3)4. Cooling rate from the melt (1.2.)or heating after melt-quenching
3. PLLA-b-PDLA 3. Kinds and averagesequence lengths(molecular weights)of comonomer units(for copolymers).
1.2. Cooling from the melt orheating after melt-quenching
4. Degree of substitution ornumber of graft chain perunit length of main chain(for graft copolymers)
1.3. During polymerization
2. Solution casting 1. Mixing ratio of polymers(for polymer blending)
2. Kind of solvent3. Temperature and time for solventevaporation
3. Precipitation 1. Mixing ratio of polymers(for polymer blending)
3.1. Precipitation or gel formation ata constant polymerconcentration
3.2. Precipitation into non-solvent
2. Kinds of solvent and precipitant3. Precipitation temperature4. Share rate (Stirring rate) of solventor precipitant during precipitation
4. Stepwise assembly in solution 1. Solution concentration andtemperature
2. Immersion time (for oneimmersion) and number of times
5. Drawing or orientation(after preparation by theprocedures 1–3)
1. Drawing method (Uniaxialor Biaxial)
2. Step number of drawing(One, two, or more)
3. Drawing temperature andannealing time
6. Compression (after preparationof monolayer film)
1. Compression temperatureand pressure
Poly(lactide) Stereocomplexes: Formation, Structure, Properties, Degradation, and Applications 573
Macromol. Biosci. 2005, 5, 569–597 www.mbs-journal.de � 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
2y values of 15, 17, and 198,[4] which are comparable with
the results for the a form of PLLA crystallized in a pseudo-
orthorhombic unit cell of dimensions: a¼ 1.07 nm,
b¼ 0.595 nm, and c¼ 2.78 nm, which contains two 103helices.[36,47] The most intense peaks of equimolarly
blended film (XD¼ 0.5) are observed at 2y values of 12,
21, and 248. These peaks are for the PLA stereocomplex[4]
crystallized in a triclinic unit cell of dimensions: a¼0.916 nm, b¼ 0.916 nm, c¼ 0.870 nm, a¼ 109.28,b¼ 109.28, and g¼ 109.88, in which L-lactide and D-lactide
segments are packed parallel taking 31 helical conforma-
tion.[47] The crystal structure of the PLA stereocomplex
proposed by Okihara et al.[47] is demonstrated in Figure 5.
The lattice containing a PLLA or PDLA chain with a 31helical conformationhas the shape of an equilateral triangle,
which is expected to form equilateral-triangle-shaped
single crystals of the PLA stereocomplex.[48] On the other
hand, Okihara et al. found that in the X-ray and electron
diffraction patterns of drawn fibers of the PLA stereo-
complex, the equatorial reflections were sharp, but those on
the layer lines were largely broadened in the direction
parallel to the layer line, becomingmore diffuse as the layer
order increases. On the basis of the paracrystalline theory,
they estimated the degree of shift disorder among polymer
chains in the direction parallel to the molecular axis to be
0.1.[49] Furthermore, Brizzolara et al.[50] compared the
WAXS profiles from actual stereocomplexed specimens
and a Force-Field simulated stereocomplex. They also
proposed the growth mechanism of the stereocomplex
equilateral-triangle-shaped single crystal.
Figure 3. DSC thermograms of PLLA/PDLA blends withdifferent XD values.[21] Viscosity-average molecular weights(Mvs) of PLLA and PLLA are both 3.6� 105 g �mol�1.
Table 3. Physical properties of stereocomplexed PLA and some biodegradable polyesters.[36]
Physical properties Stereo-complexed
PLA
PLLA PDLLA SyndiotacticPLA
PCL Poly[(R)-3-hydroxybutyrate]
(R-PHB)
Poly(glycolide)(PGA)
Tm(8C) 220–230 170–190 – 151 60 180 225–230Tm0(8C) 279 205, 212, 215 – – 71, 79 188, 197 –
Tg(8C) 65–72 50–65 50–60 34 �60 5 40DHm(100%)a) (J g�1) 142, 146 93, 135, 142,
203– – 142 146 180–207
DEtdb) (kJ �mol�1) 205–297 87–104 – – – – –
Density (g � cm�3) – 1.25–1.29 1.27 – 1.06–1.13 1.177–1.260 1.50–1.69Solubility parameter (dp) (25 8C) (J
0.5 � cm�1.5) – 19–20.5,22.7
21.1 – 20.8 20.6 –
[a]58925 in chloroform(deg � dm�1 � g�1 � cm3) – �155� 1 0 – 0 þ44c) –
WVTRd) (g �m�2 � d�1) – 82–172 – – 177 13e) –Tensile strength (GPa) 0.88f) 0.12–2.3f) 0.04–0.05g) – 0.1–0.8f) 0.18–0.20f) 0.08–1f)
Young’s modulus (GPa) 8.6f) 7–10f) 1.5–1.9g) – – 5–6f) 4–14f)
Elongation at break (%) 30f) 12–26f) 5–10g) – 20–120f) 50–70f) 30–40f)
a) Enthalpy of melting of crystal having infinite size.b) Activation energy for thermal degradation estimated by thermogravimetry at a constant temperature (250–270 8C).c) 300 nm, 23 8C.d) Water vapor transmission rate at 25 8C.e) Poly[(R)-3-hydroxybutyrate-co-3-hydroxyvalerate] (94/6).f) Oriented fiber.g) Non-oriented films.
574 H. Tsuji
Macromol. Biosci. 2005, 5, 569–597 www.mbs-journal.de � 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
The result obtained from SAXS analysis was solely for
the long period of the PLA stereocomplex precipitated from
acetonitrile solutions at 80 8C and annealed at 216 8C.[43]
The estimated long period was 12 nm, which is much
smaller than the 22–35 nm reported for PLLA films
crystallized at 120–160 8C from the melt.[51] Using the
crystallinity (Xc) value obtained by a WAXS profile (70%),
the estimated thickness values of stereocomplex crystalline
and amorphous regions were 8.4 and 3.6 nm, respec-
tively.[43] The estimated thickness of stereocomplex crys-
talline regions is also much smaller than the 13–22 nm
reported for PLLAfilms crystallized at 120–160 8C.[51] The
differences in long period and crystalline thickness between
the stereocomplex and PLLA are partly ascribed to the
differences in the specimen preparation method and
conditions.
2.3. Infrared (IR) and Raman Spectroscopy
Kister et al.[52] observed IR and Raman spectral changes in
peak shapes and wavelengths upon PLA stereocomplex-
ation. Later, by the use of FT-IR, Zhang et al.[53,54] found
that a very small low-frequency shift (about 1 cm�1) of
nas(CH3) and a larger low-frequency shift (about 5 cm�1) of
n(C O) were observed during stereocomplex crystalliza-
tion from the melt (Figure 6). The low-frequency shifts of
the stretching vibration modes of the methyl and carbonyl
groups confirmed for the first time that the interaction
between the chains in the PLA stereocomplex is ascribed to
the CH3���O C hydrogen bonding. Another interesting
result is that the peak shift of the n(C O) band already
occurs in the induction period, which indicates that the
CH3���O C interaction is the driving force for the racemic
nucleation of the PLA stereocomplex.[53,54] Moreover, 2D
correlation analysis indicates that the structural adjustment
of the CH3 group occurs prior to that of the C–O–C
backbone during the stereocomplexation process. Although
van derWaals interaction between the hydrogen of CH3 and
the oxygen of O C has been suggested by Brizzolara
et al.,[50] Zhang et al. indicated for the first time that the
hydrogen bonding is the driving force for the nucleation of
PLA stereocomplex crystallites.[53,54]
Figure 4. WAXS profiles of PLLA/PDLA blends havingdifferent XD values.[4] Solid line, dashed line, and dashed/singledotted line are for XD¼ 0.5, 0.75, and 1, respectively.
Figure 5. Crystal structure of PLA stereocomplex.[47] The lines between PLLA and PDLAchains were added to original figure.
Poly(lactide) Stereocomplexes: Formation, Structure, Properties, Degradation, and Applications 575
Macromol. Biosci. 2005, 5, 569–597 www.mbs-journal.de � 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
2.4. 1H and 13C NMR Spectroscopy
Tsuji et al. estimated the degree of PLA stereocomplexation
in a concentrated solution of equimolar PLLA and PDLA
by theuseof 1HNMRspectroscopy, as shown inFigure7.[26]
Broad peaks in addition to sharp peaks appeared in the
resonance lines of the methine and methyl groups and the
areas of the broad peaks increased with time. These broad
peaks are ascribed to the chains adjacent to stereocomplex
micro-crystallites, i.e. folding chains and tie chains. The
total peak area decreased because the PLA chains in the
stereocomplex crystallites gave no peaks. The shoulder
which appeared in the resonance line of CHCl3 may be due
to the CHCl3 surrounding the folding and tie chains. The
decrease in the peak areas of themethine andmethyl groups
continued for more than fifty days and the peak areas
decreased below 50%of the initial values. Such phenomena
reflect the formation of stereocomplex crystallites in
concentrated solutions and the increase in number and/or
thickness of the stereocomplex crystalline regions.
Figure 6. Temporal changes of the IR spectra in the C–H stretching region ofmethyl group(a) and C O stretching region (b) during the melt-crystallization process of PLLA/PDLAstereocomplex at 220 8C, respectively.[54]
Figure 7. 400 MHz 1H NMR spectra of equimolarly mixed chloroform solution of PLLAand PDLA at 17.5 g � dL�1 and 25 8C.[26]
576 H. Tsuji
Macromol. Biosci. 2005, 5, 569–597 www.mbs-journal.de � 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
13C NMR spectroscopy is also an effective method for
tracing PLA stereocomplexation.[43] Figure 8 gives high-
resolution solid-state CP/MAS 13C NMR spectra of PLA
stereocomplex precipitates and nonblended PDLA pre-
cipitates, and an as-cast amorphous poly(D,L-lactide)
(PDLLA)film, togetherwith a 13CNMRspectrumofPDLA
in CDCl3. Upon stereocomplexation, a new peak appeared
at 173.3 ppm (Peak I) for carbonyl carbon, whereas explicit
new peaks were not observed for the methine and methyl
carbons. Later, a similar spectral change upon stereocom-
plexation was monitored for carbonyl carbon by Kister
et al.[52]
We performed a component analysis solely for carbonyl
carbon because of the appearance of the new peak due
to stereocomplexation. Figure 9 illustrates the component
analysis for the total 13CNMR spectrum of carbonyl carbon
in the stereocomplex precipitates, and the 13C chemical
shifts and spin-lattice relaxation times (T1C) of components
A–D in resonance lines I–III are described in the caption of
Figure 9. The assignments of components A–D in the
reported article are as follows.[43] Component A in line III is
ascribed to PLA chains in the amorphous regions because
the chemical shift of line III is very close to those of the
amorphous PDLLAfilm and the PDLA in solution, and also
because of its very short T1C value (5.4 s). Components C
and D in line I are assignable to the PLA chains in the
stereocomplex crystalline regions because line I is not
observed for the crystallized nonblended PDLA and PLLA
precipitates or for the amorphous PDLLA film. Relatively
high and low T1C values of components C and D, 128 and
17 s, suggest that the chains of these components are in a
rigid and disordered state, respectively. Component B
corresponds to line IIwhichmay be ascribed to the chains in
the homo-crystalline regions, because the T1C value is
rather high (40 s) and has a very similar chemical shift to
that of the precipitates of the nonblended PDLA or PLLA.
2.5. Light Scattering (LS) Measurements
The aggregation behavior of isotactic and syndiotactic
PMMAwas monitored by Vorenkamp and Challa using the
LS method.[55] They demonstrated that the aggregation of
isotactic and syndiotactic PMMA caused no change in the
radius of gyration (ca. 30 nm) but an increase in molecular
weight. For stereocomplexation of PLA homopolymers,
PLLA and PDLA, the LS method has not been applied, but
Portinha et al.[56,57] observed aggregation behavior in
nonblended and blended solutions of poly(L-lactide)-block-
poly(e-caprolactone) (PLLA-b-PCL) and poly(D-lactide)-
block-poly(e-caprolactone) (PDLA-b-PCL). They revealedthat the hydrodynamic radii of assemblies in the enantio-
meric blended solution, 200 nm, were higher than those of
nonblended polymer solutions and that the radius distribu-
tion in the enantiomeric blended solution was sharper than
that in nonblended polymer solutions.
2.6. Viscometry
In concentrated solutions, the formed PLA stereocomplex
crystallites acted as physical crosslinks. The crosslinks
raise the solution viscosity and finally cause gel forma-
tion.[26] Figure 10 depicts the time change of the relative
viscosity of a mixed chloroform solution of equimolar
PLLA and PDLA with different concentrations. The
increase in viscosity occurs at concentrations exceeding a
critical level. In Figure 10, the critical concentration is in the
Figure 8. 50 MHz CP/MAS 13C NMR spectra of different PLAspecimens at 40 8C:[43] (A) PDLA precipitates; (B) stereocom-plex precipitates; (C) PDLLA film; (D) 13C NMR spectrum ofPDLA in chloroform-d.
Figure 9. Component analysis for the total 13C NMR spectrumof carbonyl carbon in the stereocomplex precipitate:[43] (A)noncrystalline (amorphous) component (169.7 ppm, T1C¼ 5.4 s);(B) homo-crystalline component (172.0 ppm, T1C¼ 40 s); (C)rigid stereocomplex crystalline component (173.3 ppm, T1C¼128 s); (D) disordered sterocomplex crystalline component(173.3 ppm, T1C¼ 17 s).
Poly(lactide) Stereocomplexes: Formation, Structure, Properties, Degradation, and Applications 577
Macromol. Biosci. 2005, 5, 569–597 www.mbs-journal.de � 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
range of 10–12.5 g � dL�1 and this value depends onvarious
parameters such as the molecular weight, mixing ratio of
PLLA and PDLA, and solution temperature, as mentioned
below.
2.7. Rheological Measurements
Similar to theviscosity, the storagemodulus (G0) of the PLAsolution is increased by stereocomplexation due to the
reason given by Tsuji et al.[26] The dried specimen obtained
by solvent evaporation of a PLA stereocomplex gel has
higherG0 values for a wide temperature range.[24] Figure 11
gives the G0 of a nonblended PLLA film (XD¼ 0), and
blended films of PLLA/PDLA¼ 3/1 (XD¼ 0.25) and 1/1
(XD¼ 0.5), respectively. Because of its high brittleness
arising from a low molecular weight, measurements for
nonblended PLLA film could not be carried out for tem-
peratures exceeding 100 8C. The loss tangent (tan d) peaksat around 60 8C for all films and at around 180 8C for the 3/1
blended film are attributed to the glass transition and
melting of homo-crystallites. The G0 of all films decreased
above the glass transition temperature (Tg) and that of the 3/
1 blended film became lower above the Tm of the homo-
crystallites. Although the 1/1 blended film having solely
stereocomplex crystallites gradually decreased with tem-
perature above Tg, it retained the highestG0 among the films
from room temperature to Tm of stereocomplex (ca.
220 8C). Even 3/1 film having homo-crystalline regions as
well as the stereocomplex crystalline regions maintained
non-zero G0 values for temperatures up to 220 8C, meaning
that the presence of stereocomplex crystallites enhanced the
heat-resistance of PLA-based materials for temperatures
above Tg.
2.8. Tensile Properties
Stereocomplexation improves the tensile properties of
PLA-based materials.[24] Figure 12 shows the tensile
properties of nonblended PLLA or PDLA films and their
equimolarly (XD¼ 0.5) blended films. At the weight-
average molecular weight (Mw) of 1� 105–1� 106 g �mol�1, the blended film surpassed the nonblended films in
all the tensile properties (tensile strength, Young’smodulus,
and elongation at break). The main reason for the increased
tensile properties of equimolarly blended films at relatively
lowMw is the formation of stereocomplexmacro-gel during
solvent evaporation, as stated in Section 2.6, whereas the
reason for those at relatively high Mw is the formation of
smaller spherulites in the blended films.
2.9. Polarization Optical Microscopy (POM)
The growth rate, induction period, and morphology of the
spherulites of stereocomplex crystallites as well as those of
homo-crystallites can be monitored by POM.[22–24,58] The
PLA spherulites containing solely stereocomplex crystal-
lites can be formed in the equimolarly blended film of
PLLA and PDLA (Figure 13).[58] The spherulitic structure
of PLA stereocomplex crystallites was different from that
of homo-crystallites with polygonal shapes, which can be
seen for nonblended PLLA or PDLA having low molecular
weights at low temperatures.[59,60] On the other hand, by
the use of POMaswell as the scanning electronmicroscopy,
the porous structure of dried stereocomplex gels can be
observed (Section 3.2.1).[24]
Figure 10. Relative viscosity of mixed chloroform solutions ofequimolar PLLA [viscosity-average molecular weight(Mv)¼ 4.2� 104 g �mol�1] and PDLA (Mv ¼ 4.5� 104 g �mol�1)with different concentrations at 25 8C.[26]
Figure 11. Storage modulus (G0) and loss tangent (tan d) for as-cast nonblended and blended films from PDLA (Mw ¼ 1.2�105 g �mol�1) and PLLA (Mw ¼ 1.0� 105 g �mol�1) with diffe-rent XD values.[24]
578 H. Tsuji
Macromol. Biosci. 2005, 5, 569–597 www.mbs-journal.de � 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
2.10. Scanning Electron Microscopy (SEM),Electron Diffraction (ED), and TransmissionElectron Microscopy (TEM)
SEM observation revealed that PLA stereocomplex parti-
culate precipitates (Figure 14) are formed in acetonitrile
solution and their size and shape depends on various para-
meters, as shown inTable 2.[27] The stereocomplex particles
were disk-shaped under the conditions of 1 g � dL�1, 80 8C,and XD¼ 0.5. Upon decreasing the temperature or increas-
ing the polymer concentration, the stereocomplex particles
becamemore spherical, whereas upon deviation ofXD from
0.5, the stereocomplex particles became equilateral-triangle-
plate-shaped. The equilateral triangular shape is very
similar to that of the PLA stereocomplex single crystals
observed by TEM [Figure 15(A)].[47] Cartier et al.[61] found
that such triangular crystals as observed for the PLA
stereocomplex can be utilized as a morphological marker
for the frustrated character of the chain packing of a
polymer in the unit cell. On the other hand, Okihara et al.[47]
demonstrated that PLA stereocomplexation can be traced
by the use of obtained ED patterns.
2.11. Atomic Force Microscopy (AFM)
Similar to the SEM and TEM observation, AFM can moni-
tor the equilateral triangle-shaped single crystal and disks
of the PLA stereocomplex (Figure 16).[9]
3. Methods for Inducing PLAStereocomplexation
PLA stereocomplexation takes place in the absence of a
solvent (in bulk: crystallization from the melt or during
polymerization), or in the presence of solvent (in solution).
As tabulated inTable 2, PLA stereocomplexation is affected
by numerous parameters. Tsuji et al.[6,20–24,26–29,36] and
Murdoch and Loomis[62–66] investigated the parameters’
effects on the stereocompelxation of PLA homopolymers
utilizing different methods under a wide variety of
conditions.
3.1. In Bulk or Solid State
3.1.1. Crystallization from the Melt
There are four representative procedures for crystallization
from the melt; (1) crystallization at a fixed temperature (Tc)
directly from the melt, (2) crystallization at a fixed Tc after
quenching from the melt, (3) crystallization during cooling
from the melt, (4) crystallization during heating of melt-
quenched specimens.[67] Procedure (1) is utilized to trace
the radius growth rate of spherulites (G) and the induction
period for spherulite formation (ti). Figure 17 illustrates the
G and ti values for spheruiltes of stereocomplex crystallites
and homo-crystallites in PLLA, PDLA, and equimolar
PLLA/PDLA blended films crystallized through Procedure
(1). The polarization optical photomicrographs are given in
Figure 13.[58] The spherulites composed of stereocomplex
crystallites have an extremely high G and a short ti com-
pared with those of the homo-crystallites of either PLLA or
PDLA, as shown in Figure 17. Such rapid formation and
growth of PLA stereocomplex crystallites were observed
even when an equimolar mixture of PLLA and PDLA both
Figure 12. Tensile properties of nonblended films and equimo-larly (XD¼ 0.5) blended film from PLLA (XD¼ 0) and PDLA(XD¼ 1) as a function ofMw.
[24]
Poly(lactide) Stereocomplexes: Formation, Structure, Properties, Degradation, and Applications 579
Macromol. Biosci. 2005, 5, 569–597 www.mbs-journal.de � 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Figure 13. Spherulites of PLLA (Mw ¼ 1.0� 104 g �mol�1) (A), PDLA (Mw ¼ 2.2�104 g �mol�1) (B), and their equimolarly blended films (C, D) crystallized at 140 8C (A–C)and 190 8C (D) from the melt at 250 8C.[58]
Figure 14. SEM photographs of PLA stereocomplex particles precipitated fromacetonitrile solutions with different concentrations.[27] (A) 0.1 g � dL�1; (B) 1 g � dL�1; (C)3 g � dL�1; (D) 10 g � dL�1.
580 H. Tsuji
Macromol. Biosci. 2005, 5, 569–597 www.mbs-journal.de � 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
with low molecular weights was quenched from the
melt.[22,62] The hydrogen bonding between methyl hydro-
gen and carbonyl oxygen as revealed by Zhang et al.[53,54] is
expected to enhance such rapid spherulite formation and
growth of PLA stereocomplex crystallites. The spherulites
of stereocomplex crystallites had normal morphology very
similar to that of homo-crystallites when only stereocom-
plex crystallites were contained in the spherulites.[22,23] In
contrast, the spherulite morphology became complicated
when both stereocomplex crystallites and homo-crystallites
were formed simultaneously in the spherulites.[22,23]
Although there has been no report with respect to Pro-
cedure (2), the quenching process should enhance stereo-
complex crystallization by increasing the spherulite nuclei,
as reported for homo-crystallization of PLLA.[67] In
Procedures (1) and (2), the stereocomplex crystallites are
predominantly formed by crystallization at Tc between the
Tm values of homo-crystallites (170–180 8C) and stereo-
complex crystallites (220–230 8C).[22,62] For Procedure
(3), Brochu et al.[68] reported the epitaxial crystallization of
homo-crystallites of PLLA on the stereocomplex crystal-
lites of PLLA and poly(L-lactide-co-D-lactide) (20/80)
when the two polymers were mixed at a ratio of 80/20
and crystallized during slow cooling at 1 or 2 8Cmin�1 from
themelt.With regard to epitaxial crystallization of aliphatic
polyesters, Soldera and Prud’homme[69] observed epitaxial
crystallization of R-configured poly(a-methy-a-propyl-b-propiolactone) (PMPPL) onto S-configured PMEPL crys-
tallites. For Procedure (4), Urayama et al.[70] found that an
aluminum complex of a phosphoric ester combined with
hydrotalcite (NA) executively enhance the nucleation of
stereocomplex crystallites, whereas talc cannot suppress
the homo-crystallization of PLLA or PDLA.
3.1.2. During Polymerization
Spinu et al.[71–73] proposed a novel method for stereo-
complexation between PLLA and PDLA; polymerization
of LLA and DLA in the presence of PDLA and PLLA,
respectively, after mixing LLA and PDLA (or DLA and
PLLA). With this method, they successfully prepared well-
stereocomplexed PLA materials. In the strict sense of the
word, this method may not be ‘‘template polymeriza-
tion’’,[74] but effectively utilizes the fact that the polymer-
ized chains have strong interaction with the template
chains.
3.1.3. Upon Compression
Bourque et al.[75] and Pelletier and Pezolet[76] reported that
stereocomplexation occurs upon the compression of the
Figure 15. TEM photographs and electron diffraction patternsof PLAstereocomplex single crystal (A) and particle (B) formed indilute acetonitrile solutions.[27]
Figure 16. AFM photographs of equilateral-triangle-shapedsingle crystal (A) and disks (B) of PLA stereocomplex (Courtesyof Prof. Domb).[9] Reprinted from Adv. Drug Delivery Rev., Vol.55, Slager and Domb, ‘‘Biopolymer Stereocomplexes’’, pp. 549–583, 2003, with permission of Elsevier.
Poly(lactide) Stereocomplexes: Formation, Structure, Properties, Degradation, and Applications 581
Macromol. Biosci. 2005, 5, 569–597 www.mbs-journal.de � 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
monolayer of PLLA/PDLA mixture films. They prepared
Langmuir films of PLLA andPDLA and then observed their
structural changes by the use of surface-pressure measure-
ments and polarization modulation infrared reflecting-
absorption spectroscopy (PM-IRRAS). They indicated that
a stereocomplex bilayer in equilibrium with the monolayer
was formed at the air-water interface by compression.
3.1.4. Upon Orientation
Tsuji et al. revealed that stereocomplex cystallization of the
equimolar PLLA/PDLA fibers were enhanced by hot-
drawing to high ratios.[77] The hot-drawing process causes
expanded chains or increases the surface area per unit
molecule and, therefore, raises the probability of interaction
between PLLA and PDLA segments. This causes increased
Xc and Tm of stereocomplex crystallites and enhances
tensile properties.[62,77]
3.1.5. Upon Hydrolytic Degradaion
Li et al. found that stereocomplexation can occur in poly(L-
lactide-co-D-lactide) with a thickness of mm order after a
long-term hydrolytic degradation (25 or 30 weeks) as
mentioned in detail in Section 4.2.2.[78–82] The hydrolytic
degradation induces a decreased molecular weight, which
elevates the mobility of PLA chains in addition to the
presence of water as a plasticizer, and selective removal of
relatively atactic sequences, leaving relatively isotactic
chains. These three factors enhance PLA stereocomplexa-
tion between remaining L-lactyl unit sequences and D-
lactyl unit sequences. However, when thin specimens
(thickness< 200 mm) were used, no stereocomplexation
was observed for poly(D,L-lactide)[83] and an amorphous-
made equimolar blend of PLLA and PDLA,[84] even when
hydrolytic degradation was continued for 20–24 months.
This supports the fact that core-accelerated hydrolysis
of thick specimens should have induced PLA stereocom-
plexation.
3.2. In Solutions
Once PLA stereocomplex crystallites are formed in solu-
tion, they are insoluble in good solvents for either PLLA or
PDLA (e.g., chloroform, dichloromethane). This means
high stability of the stereocomplex crystallites compared
with that of homo-crystallites of either PLLAor PDLA. The
stereocomplex crystallites can be dissolved in extremely
good solvents at room temperature (e.g. 1,1,1,3,3,3-
hexafluoro-2-propanol) or in generally good solvents at
high temperatures near the boiling points (e.g., chloroform,
1,1,2,2-tetrachloroethane). However, the stereocomplex
crystallites become insoluble in extremely good solvents
even at elevated temperatures when the crystalline thick-
ness is high.
3.2.1. At a Fixed Polymer Concentration
The crystallites of a polymer are formed when the polymer
concentration in a solution exceeds a critical level. The
critical concentration for PLA stereocomplex crystallite
formation (stereocomplex crystallization) is much lower
than that for homo-crystallite formation of either PLLA or
PDLA (homo-crystallization). In other words, some good
solvents [e.g., chloroform and dichloromethane at room
temperature, acetonitrile around boiling temperature (ca.
80 8C)] for either PLLA or PDLA are poor solvents or non-
solvents for stereocomplex crystallites. Therefore, when an
initial polymer concentration is lower than the critical level
for homo-crystallite formation but higher than that for
stereocomplex crystallite formation, mixing two separately
prepared solutions of PLLA and PDLA results in the
formation of stereocomplex crystallites. Such stereocom-
plex crystallite formation induces particulate precipitates or
single crystals in dilute solutions (acetonitrile)[27] or gels in
Figure 17. Radius growth rate of shperulites (G) and inductionperiod of spherulite formation (ti) of PLLA, PDLA, and theirequimolarblendsasafunctionofcrystallizationtemperature(Tc).
[58]
582 H. Tsuji
Macromol. Biosci. 2005, 5, 569–597 www.mbs-journal.de � 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
concentrated solutions (chloroform, dichloromethane).[26]
The shape and size of stereocomplex precipitates in ace-
tonitrile depend on representative parameters: XD, polymer
concentration (Figure 14), molecular weights of PLLA and
PDLA, and temperature.[27] The stereocomplex particles
grow in a suspended state, which is monitored by the
turbidity of the solution (Figure 18).[27] After sedimentation
of the stereocomplex particles, the supernatant becomes
transparent again.
Probably, Murdoch and Loomis first observed gel forma-
tion upon PLA stereocomplexation.[62] Figure 19 shows a
phase diagram of a PLLA/PDLA/chloroform system.[21]
The lines between phases I and II and between phases II and
III are the critical concentrations for stereocomplex
crystallite formation (micro-gel formation) and macro-gel
formation, respectively. In stereocomplex gel formation,
the PLA stereocomplex crystalline regions act as physical
cross-links between the PLA chains (Figure 20).[24] In the
case of the PLA-based block copolymers mentioned below,
when the tie chains or sequences which do not take part in
the formation of cross-links or stereocomplex crystalline
regions are hydrophilic, the materials will be biodegradable
hydrogels in aqueous media. It is of great interest that
although L-lactyl unit sequences and D-lactyl unit sequences
become insoluble in water at a higher degree of polymer-
ization (DP), such hydrogels can be formed through
stereocomplexation in aqueous media upon mixing the
enantiomeric suspensions.
3.2.2. Casting (Increasing Polymer Concentrationby Solvent Evaporation)
In the solution castingmethod, solvent evaporation elevates
the polymer concentration of the solution from an initial
concentration to an infinite one. Therefore, during the
course of solvent evaporation, the polymer concentration
exceeds first the critical level of stereocomplex crystallite
formation, and then that of homo-crystallite formation. This
means that the stereocomplex crystallites are predomi-
nantly formed in equimolarly mixed solutions of PLLA and
PDLAwhen the solvent evaporation rate is sufficiently low.
Figure 18. Polymer concentration in supernatant of acetoni-trile solution as a function of time during stereocomplexation ofPDLA and PLLA [XD¼ 0.5, Mv (PLLA)¼ 4.2� 104 g �mol�1,Mv(PDLA)¼ 4.5� 104 g �mol�1, 1 g � dL�1, 80 8C].[27]
Figure 19. Phase diagram of PDLA/PLLA/chloroform system;homogeneous solution (I); cloudy solution containing stereo-complex microgel (II); stereocomplex macrogel (III). [XD¼ 0.5,Mv (PLLA)¼ 4.2� 104 g �mol�1, Mv(PDLA)¼ 4.5� 104 g �mol�1].[26]
Figure 20. SEM photograph and presumed structure of driedstereocomplex gel.[24]
Poly(lactide) Stereocomplexes: Formation, Structure, Properties, Degradation, and Applications 583
Macromol. Biosci. 2005, 5, 569–597 www.mbs-journal.de � 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
However, rapid solvent evaporation will not give the stereo-
complex crystallites a sufficiently long time for their
formation, and the polymer concentration will reach the
critical level of homo-crystallite formation, resulting in the
formation of homo-crystallites. Homo-crystallization pre-
dominantly occurs when the molecular weights of both
PLLA and PDLA are high, the deviation of XD from 0.5 is
large, or the solvent evaporation rate is high. Figure 21
illustrates a typical example of the effects of PLAmolecular
weights on the kind and amount of the crystallites.[20] With
increasing the molecular weights of PLLA and PDLA, a
large fluctuation of microscopic XD from 0.5 occurs even in
an equimolarly mixed solution, due to the large radii of the
PLLA and PDLA molecules. This must retard the nuclei
formation and growth of stereocomplex crystallites, result-
ing in the formation of a large amount of homo-crystallites.
de Jong et al.[85,86] synthesized polydisperse L- and
D-lactic acid oligomers using stannous octoate and 2-(2-
methoxyethoxy)ethanol as initiator and coinitiator, respec-
tively, and subsequent preparative HPLC of polydisperse
oligomers yielded monodisperse oligomers (DP¼ 1–16).
They prepared nonblended and equimolarly blended speci-
mens by solution-casting with dichloromethane. DSC[85]
and WAXS[86] revealed that L- or D-oligomers were
crystallizable forDP above 11, while an equimolar mixture
formed stereocomplex crystallites forDP above 7,meaning
high stability of the stereocomplex crystallites compared
with that of homo-crystallites. However, WAXS analysis
revealed that the mixture of L- and D-lactic acid oligomers
(DP¼ 7) contained homo-crystallites as well as stereo-
complex crystallites. This may have arisen from epitaxial
crystallization of the homo-crystallites on the stereocom-
plex crystallites.[68]
3.2.3. Precipitation into Non-solvent
Amixed solution of PLLA and PDLA into a precipitant or a
non-solvent causes removal of the solvent from the solution
and diffusion of the non-solvent into the solution, resulting
in rapid crystallization. In this method, in addition to the
aforementioned representative parameters such as XD and
the molecular weights of PLLA and PDLA (Table 2), the
shear rate of the non-solvent and polymer concentration
affect greatly the kind and amount of crystallites formed. A
high shear rate of the non-solvent and a low polymer
concentration induce the predominant formation of stereo-
complex crystallites.[21]
3.2.4. Stepwise Assembly
Serizawa et al.[87] reported that stepwise assembly between
PLLA and PDLA on a quartz crystal microbalance (QCM)
substrate gave rise to stereocomplexation. Thiswas attained
by alternate immersion of the substrate into acetonitrile
solutions of PLLA and PDLA. They also indicated that the
assembly of PLLA can grow epitaxially on the surface of
stereocomplex crystallites, as shown by Brochu et al.[68]
4. Homo-Stereocomplexation
The reported examples for the PLA-based stereocomplex
from various types of polymer blends or from copolymers
are summarized in Table 4.
4.1. Stereocomplexation in Polymer Blend
4.1.1. Homopolymers
For homopolymer blends of PLLA and PDLA, intensive
studies have been carried out by Tsuji et al.[6,20–22,26,27,36]
and Murdoch and Loomis et al.,[62–66] with numerous
varying parameters. They found that the dominant para-
meters for stereocomplexation are XD and the molecular
weights of PLLA and PDLA. As stated earlier, the ratio of
stereocomplex crystallites to homo-crystallites decreases
with the deviation of XD from 0.5 (Figure 3) and increasing
molecular weights of PLLA and PDLA (Figure 21).
However, no crystallite formation occurs when the DP
values of PLLA and PDLA are lower than 7.[85]
4.1.2. Random Stereocopolymers
Tsuji and Ikada synthesized L-lactide-rich PLA and
D-lactide-rich PLA having optical purities from 0–100%
and traced the crystallization behavior of their nonblended
and blended films at different Tc from the melt[23] as well as
during solvent evaporation.[28] The DHm and Tm of stereo-
complex crystallites in blended films decreased with
decreasing optical purity (OP), in agreement with those of
homo-crystallites in nonblended films. It was found that
Figure 21. Enthalpies of melting (DHm) of stereocomplexcrystallites and homo-crystallites of equimolar (XD¼ 0.5)PLLA/PDLAblends prepared by solvent evaporation as a functionof the averaged viscosity-average molecular weight (Mav) ofPLLA and PDLA.[20]
584 H. Tsuji
Macromol. Biosci. 2005, 5, 569–597 www.mbs-journal.de � 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Table4.
Stereocomplexation(racem
iccrystallization)oflacticacid-based
polymers.
Polymer
pairorpolymer
Initiator,catalystforsynthesis
oflacticacid
chains
Stereocomplexation(racem
iccrystallization)
Stereocomplexed
materials
Authorsandreferences
Procedure
Conditions
Param
eters
MonitoringMethods
Shape
Application
PLLA/PDLA
Stannousoctoate(Stannous
2-ethylhexanoate)/Lauryl
alcohol(1-D
odecanol)
Solution
Solvent:CHCl 3,
MolecularweightsofPLLAandPDLA,
L-and
D-Polymer
ratioand
concentration,Tem
perature,Tim
e
Viscometry,Testtube
tilting,1HNMR,DSC
Gels,Microgels
Tsujietal.[26]
Solvent:CH3CN
MolecularweightsofPLLAandPDLA,
L-and
D-Polymer
ratioand
concentration,Tem
perature,Tim
e
DSC,Polarimetry,
SEM,13CNMR
Particles,Single
crystals
Tsujietal.[27,43]
Solvent:p-X
ylene
Fixed
WAXS,ED,TEM
Singlecrystals
Okiharaetal.[47,49]
ED,TEM
Singlecrystals
Cartier
etal.[61]
Casting
Solvent:CH2Cl 2,CHCl 3,
Benzene,Dioxane
MolecularweightsofPLLAandPDLA,
L-and
D-Polymer
ratioand
concentration
DSC,POM,Tensiletesting
Films,Fibers
Tsujietal.[20,24,77]
Solvent:CH2Cl 2
L-and
D-Polymer
ratio
DSC,POM,SEM,GPC,
Tensiletesting
Films
Tsujietal.[132,134]
(HydrolyticDegradation)
Solvent:CHCl 3
Fixed
FT-IR
Films
Zhangetal.[53,54]
Precipitation
Solvent:CH2Cl 2,CHCl 3,Benzene,
Dioxane,Nonsolvent:CH3OH
MolecularweightsofPLLAandPDLA,
L-and
D-polymer
ratioand
concentration,Tem
perature,
Rotationrateofnon-solvent
WAXS,DSC,SEM,
Tensiletesting
Fibrousprecipitates,
Fibers
Ikadaetal.,[4]Tsujietal.[21,77]
Bulk
Crystallizationfrom
themelt
MolecularweightsofPLLAandPDLA,
L-and
D-Polymer
ratio,
Crystallizationtemperature
andtime
DSC,POM
Films
Tsujietal.,[22,58]
SasaiandYam
ane[144]
Takeupvelocity,Throughputrate,
Extrusiontemperature
WAXS,DSC,Tensile
testing
Fibers
Takasakietal.[141]
MolecularweightsofPLLAandPDLA,
L-and
D-Polymer
ratio
DSC,GPC,Tensiletesting
Films
Tsuji,[19]TsujiandSuzuki[134]
(Hydrolyticdegradation)
Nucleators
DSC,POM
Films
Urayam
aetal.[70]
Amorphous–
madefrom
themelt
L-and
D-Polymer
ratio
Gravim
etry,GPC
Films
Tsuji[133](H
ydrolyticdegradation),
TsujiandMiyauchi[140]
(Enzymaticdegradation)
Inthemelt
L-and
D-Polymer
ratio
TG
Melts
TsujiandFukui[123]
(ThermalDegradation)
L-and
D-Polymer
ratio,
Concentrationofcatalyst
TG
Melts
Fan
etal.[128](Thermal
Degradation)
Oncompression
Fixed
PM-IRRAS
Langmuirfilm
sBourqueetal.,[75]
Pelletier
andPezolet[76]
Bulk,Solution
Solvent:CH3CN(Solution)
Fixed
ForceField
Sim
ulation,
WAXS,AFM
Films,Particles
Brizzolalaetal.[48]
Stepwiseassembly
Alternateim
mersioninto
polymer
solutions,Solvent:CH3CN
Polymer
concentrationandTem
perature,
Immersiontime
DSC,AFM
Thin
layer
Serizaw
aetal.[87,135]
(Alkalinedegradation)
Tem
plate
polymerization
MixingratioofDLAandPLLAorLLA
andPDLA,Tem
perature,Pressure,
Tim
e
DSC
Bulks
Spinuetal.[71–73]
Precipitation
Solvent:CH2Cl 2,Nonsolvent:CH3OH
MolecularweightsofPLLAandPDLA
DSC
Murdoch
andLoomis[42,62–66]
Zincpowder
Solution
Solvent:CH3CN
Fixed
IR,Ram
an,13CNMR
Kisteretal.[52]
Stannousoctoate/2-(2-M
ethoxy-
ethoxy)ethanol
Casting
Solvent:CH2Cl 2
MolecularweightsofPLLAandPDLA
DSC,WAXS
deJongetal.[85,86,100]
Diethylaluminum
ethoxide
Bulk
Crystallizationfrom
themelt
MolecularweightofPLLA,L-and
D-Polymer
ratio,Tem
perature
program
DSC,WAXS,POM
SchmidtandHillm
yer
[143]
L-lactide-rich
PLA/D-
lactid-richPLA
Stannousoctoate/Laurylalcohol
Casting
Solvent:CH2Cl 2
L/D-Lactideratioin
PLA
DSC
Films
Tsujietal.[28]
Bulk
Crystallizationfrom
themelt
Crystallizationtemperature,L/D-lactide
ratioin
PLA
DSC,POM
Films
Tsujietal.[23]
Aluminum
isopropoxide
Bulk
Crystallizationfrom
themelt
Coolingratefrom
themelt,
L/D-Lactide
ratioin
PLA,Mixingratioof
twopolymers
DSC,POM
Films
Brochuetal.[68]
AchiralSalenAlOMe
Precipitation
Solvent:CH2Cl 2,Nonsolvent:CH3OH
L/D-Lactideratioin
PLA
DSC
Spasskyetal.[110]
P(LLA-G
A)/P(D
LA-G
A)
Stannousoctoate/Laurylalcohol
Casting
Solvent:CH2Cl 2
Lactidecontentsin
copolymers
DSC
TsujiandIkada[29]
P(LLA-CL)/P(D
LA-CL)
Precipitation
Solvent:CH2Cl 2,Nonsolvent:CH3OH
Lactidecontentsin
copolymers
DSC
Murdoch
andLoomis[62–66]
PLLA-b-PCL/PDLA
Aluminum
tris(2-propanolate)
Al[OCH(CH3) 2] 3
Fixed
DSC
Dijkstra
etal.[89]
(Continued
)
Poly(lactide) Stereocomplexes: Formation, Structure, Properties, Degradation, and Applications 585
Macromol. Biosci. 2005, 5, 569–597 www.mbs-journal.de � 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Table4.
(Continued
)
Polymer
pairorpolymer
Initiator,catalystforsynthesis
oflacticacid
chains
Stereocomplexation(racem
iccrystallization)
Stereocomplexed
materials
Authorsandreferences
Procedure
Conditions
Param
eters
MonitoringMethods
Shape
Application
PLLA-b-PCL/PDLA-b-PCL
Yttrium
isopropoxide
Precipitation
Solvent:CHCl 3,N
onsolvent:C2H5OC2H5
MolecularweightofPLAblock
DSC
Stevelsetal.[90]
Stannousoctoate
Precipitation
Solvent:CH2Cl 2,Nonsolvent:CH3OH
MolecularweightsofPLAand
PCLblocks
DSC
Pensecetal.[91]
Yttrium
alkoxide
Solution
Solvent:THF
Polymer
concentration,Tem
perature
LS,DSC,1HNMR,IR
Portinhaetal.[56,57]
PLLA-b-PCL-b-PLLA/
PDLA-b-PCL-b-PDLA
Stannousoctoate
Precipitation
Solvent:CH2Cl 2,Nonsolvent:CH3OH
Fixed
DSC
Pensecetal.[91]
PLLA-b-PEG/PDLA-b-PEG
Notspecified
Solution
Solvent:CH3CN
Fixed
WAXD,AFM
Particles
Brizzolara
etal.[50]
PLLA-b-PEG-b-PLLA/
PDLA-b-PEG-b-PDLA
Stannousoctoate/PEG
Precipitationor
Casting
Solvent:CHCl 3,Nonsolvent:Hexane
PercentageorMolecularweightofPLLA
orPDLA
DSC
Precipitates,
Films
Stevelsetal.[92]
Precipitation
Solvent:CH2Cl 2,Nonsolvent:H2O
Fixed
DSC,X-ray
Particles
DDS
Lim
andPark[93]
Casting
Solvent:THF/H
2O(v/v)¼1/2,THF
was
removed
PercentageorMolecularweightofPLLA
orPDLA,Tem
perature
Testtubetilting,Visible
lighttransm
ittance,
WAXS,Rheological
measurements
Gels
Fujiwaraetal.[94]
PLLA-b-PEG-b-PLLA/
PDLA-b-PEG-b-PDLA
andPLLA-b-PEG/PDLA-
b-PEG
Zincpowder/PEG
Solution
Solvent:H2O
MolecularweightsofPLAand
PEGblocks,Tem
perature,Tim
eDSC,WAXS,Ram
anspectroscopy,
Viscoelasticity
Gels
LiandVert[95,96]
PLLA-b-PSA-b-PLLA/
PDLA-b-PSA-b-PDLA
Polycondensation/PSA
Casting(Solution
andmelt)
Solvent:CH2Cl 2
MolecularweightsofPLAandPSA
blocks,Polymer
mixingratio
DSC,SEM
Powder
DDS
SlivniakandDomb[97]
dextran-graft-PLLA/
dextran-graft-PDLA
Stannousoctoate/
2-(2-M
ethoxy-
ethoxy)ethanol
Solution
Solvent:Acetate-buffered
solution(pH4)
Degreeofsubstitution(D
S),Molecular
weightsofPLLAandPDLAchains
Storagemodulus,
FT-IR,Swellingratio
Gels
DDS
deJongetal.[86,99–102]
Poly(H
EMA-graft-oligo
L-lactie)/Poly(H
EMA-
graft-oligo
L-lactie)
Stannousoctoate/HEMA
Casting
CHCl 3
MolecularweightsofL-orD-lactide
sidechains
DSC,WAXS
Films
Lim
etal.[98]
pHPMAm-graft-L-lacticacid
oligomers/pHPMAm-
graft- D-lacticacid
oligomers
Stannousoctoate/2-(2-M
ethoxy-
ethoxy)ethanol
Solution
Solvent:Acetatebuffer
(pH4)
Degreeofsubstitution(D
S),molecular
weightsofPLLAandPDLAchains
Rheological
measurements,Swelling
Gels
van
Nostrum
[103]
PMBLLA/PMBDLA
Stannousoctoate/Laurylalcohol
Solution
Solvent:CH3OH/CH2Cl 2/(v/v)¼2/1,
inthepresence
ofNaC
lMolecularweightsofL-orD-lactide
sidechains
DSC,WAXS
Porousfilm
sScaffold
WatanabeandIshiharaetal.[104–106]
PLLA-grafted
film
/PDLA
Stannousoctoate
Solution
Solvent:CH3CN,CHCl 3
MolecularweightsofL-lactide
sidechains,Solvent
FT-IR,WAXS
Films
Tretinnikovetal.[107]
Stereo-block
PLA
Aluminum
tris(2-propanolate)
Al[OCH(CH3) 2] 3
Precipitation
Solvent:Toluene,Nonsolvent:CH3OH
Fixed
DSC
Yuietal.,[108]Dijkstra
etal.[89]
Rac-(SalBinap)A
lOi Pr
Precipitation
Solvent:CH2Cl 2Nonsolvent:CH3OH
Fixed
DSC
Ovittetal.[114,115]
Fixed
DSC
Radanoetal.[113]
(-)-(SalBinap)A
lOMe
Initiator/Monomer
ratio,Reactiontime
DSC
Spasskyetal.[109,110]
AchiralSalenAlOMe
Aspolymerized,
Bulkfromthemelt
D/L-Lactideratio;Crystallization
temperature
andtime,Coolingrate
from
themelt
DSC,WAXS,POM
Wiesniwskietal.,[111]
Sarasuaetal.[112]
SnCl 2(Polycondensation)
As-polymerized
Molecularweightsofhomopolymersof
PLLAandPDLA,Reaction
temperature
andtime
DSC,WAXS
Fukushim
aetal.[116]
Poly(L-lactide-co-D-lactide)
(62.5/37.5),(50/50)
Zincpowder,Stannousoctoate,
orPolymersweresupplied
Bulk
during
hydrolytic
degradation
Catalystforsynthesis,Hydrolysismedia,
Tem
perature,Tim
eDSC,WAXS
Plates,Films
LiandVert[78–82]
PDLA/LHRH,
PDLA/Vapreotide
Stannousoctoate
Solution
Solvent:CH3CN
Polymer
mixingratio
DSC,SEM
Particles
DDS
Slager
etal.[117]
PDLA/Leuproide,PDLA/
Vapreotide
Stannousoctoate/(Benzyl
alcohol)
Solution,Casting
Solvent:CH3CN
Tim
e,Polymer
mixingratio,Molecular
weightofPLA
DSC,SAXS,SEM,AFM,
Cryo-TEM,Confocal
microscopy
Particles
DDS
Slager
etal.[118–121]
PDLA/Insulin,PDLA-b-
PEG/Insulin,
PDLA-b-PEG-b-PDLA/
Insulin
Stannousoctoate/Octanol
Solution
Solvent:CH3CN
Fixed
DSC,SEM,HPLC
Particles
DDS
Slager
andDomb[122]
586 H. Tsuji
Macromol. Biosci. 2005, 5, 569–597 www.mbs-journal.de � 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
stereocomplex crystallites in blendedfilms aswell as homo-
crystallites in nonblended films can be formed as long as
both L-lactide-rich PLA and D-lactide-rich PLA have OP
values above 76% (XD� 0.88,XD� 0.12) (Figure 22). Only
in this section is XD used as D-lactide unit content in a
copolymer according to the following definition:
XD¼Weight of D-lactide=Weight of L- and D-lactide
ð2Þ
The L-lactyl unit sequence length (lL) and D-lactyl unit
sequence length (lD) can be calculated fromXD values using
Equation (3) and (4):[88]
lL ¼ 2=XD ð3Þ
lD ¼ 2=ð1� XDÞ ð4Þ
These equations assume that L-lactide and D-lactide were
polymerized by their random addition and that no ester
exchange reaction occurred during polymerization and
thermal treatment. Figure 22 and Equation (2)–(4) indicate
that for the solution-casting method at least 16.4 L-lactyl
unit sequences and D-lactyl unit sequences are required for
PLA stereocomplexation (stereocomplex crystallization)
and homo-crystallization,[28] in agreement with the results
for the melt-crystallization method.[23] The critical se-
quence unit value is slightly higher than the 7 lactyl units for
pairs of the L-lactic acid oligomer and D-lactic acid
oligomer.[85] In the case of melt-crystallization, by crystal-
lization from the melt at temperatures between the Tmvalues of stereocomplex crystallites and homo-crystallites
(between the solid line and dashed/single dotted line for
blended specimens in Figure 23), well-stereocomplexed
PLA materials can be prepared.[23] On the other hand,
Brochu et al.[68] synthesized PLLA, PDLA, and poly(L-
lactide-co-D-lactide) [P(LLA-DLA)](20/80) and investi-
gated stereocomplexation between PLLA and P(LLA-
DLA)(20/80) as well as that between PLLA and PDLA
during cooling from the melt under various mixing ratios of
the two polymers.
4.1.3. Hetero Random Copolymers
Murdoch and Loomis reported the effects of incorporated
e-caprolactone units on PLA stereocomplexation by pre-
paring equimolar blends of poly(L-lactide-co-e-caprolac-tone) [P(LLA-CL)] and poly(D-lactide-co-e-caprolactone)[P(DLA-CL)][62] having e-caprolactone unit content of upto 32 wt%. The Tm of the stereocomplex decreased with
increasing the e-caprolactone unit content. However, the
critical lactyl unit sequence length for stereocomplexation
was not identified. By a procedure similar to that in Section
4.1.2., Tsuji and Ikada[29] studied the effects of incorporated
glycolyl unit (a half glycolide unit) content on PLA
stereocomplexation. The DHm and Tm of stereocomplex
crystallites in equimolarly (XD¼ 0.5) blended films
decreased with increasing glycolyl unit content, in agree-
ment with those of homo-crystallites in nonblended films.
The stereocomplex crystallites in blended films of poly(L-
lactide-co-glycolide) [P(LLA-GA)] and poly(D-lactide-co-
glycolide) [P(LDA-GA)] are formed with L- and D-lactide
unit content as low as 72.1 and 68.6 wt%, respectively,
Figure 22. Amorphous (A), stereocomplex crystalline (S), andhomo-crystalline (H) phases in as-cast binary equimolar (1:1)blends from PLAs with different D-lactide contents [XD¼D-lactide/(L-lactideþ D-lactide)].[28]
Figure 23. Amorphous (A), stereocomplex crystalline (S), andhomo-crystalline (H) phases in equimolarly blended and non-blended PLA films crystallized from the melt as a function ofoptical purity (OP); highest crystallization (annealing) temper-ature below which stereocomplexation occurs (Tc,s) in equimo-larly (1:1) blended PLAs (solid line) and highest crystallizationtemperature below which homo-crystallization occurs (Tc,h) inequimolarly blended PLAs (dashed/single dotted line) andnonblended PLAs (dashed line).[23]
Poly(lactide) Stereocomplexes: Formation, Structure, Properties, Degradation, and Applications 587
Macromol. Biosci. 2005, 5, 569–597 www.mbs-journal.de � 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
whereas the homo-crystallites in nonblended films are
formed with L- or D-lactyl unit content above 81.0 wt%
(Figure 24). The calculated critical L- or D-lactyl unit
sequence length for stereocomplex crystallization in the
blended films (5.5 lactyl units) is lower than 8.8 lactyl units
for homo-crystallites in the nonblended films, reflecting the
fact that the peculiarly strong interaction between the L-
lactyl unit sequences and D-lactyl unit sequences enhances
the stability of stereocomplex crystallites and the crystal-
lizablity of the blended films. Here, the calculation proce-
dure for the lactyl unit sequence length is similar to the
Equation (3) and (4), and is given in our recent article.[88]
This value of 5.5 lactyl units is slightly lower than the
7 lactyl units for pairs of the L-lactic acid oligomer and
D-lactic acid oligomer.[85] It should be noted that the critical
values shown here are average values and, therefore, the
copolymers contain L- or D-lactyl unit sequences longer
than the average values. The critical value for the lactide-
glycolide copolymer is lower than that for random lactide
stereocopolymers (Section 4.1.2.), suggesting that the
glycolide units in the copolymers must have enhanced the
stereocomplexation between L-lactyl unit sequences and D-
lactyl unit sequences. It seems probable that the lower steric
hindrance of the methylene group of glycolyl units
(compared with the high steric hindrance of the ethylidene
group of L-lactyl or D-lactyl units) raises the chain mobility
of glycolide copolymers, resulting in high stereocomplex-
ationability during solvent evaporation.
4.1.4. Hetero Block Copolymers
4.1.4(a). With Poly(e-caprolactone) (PCL) Blocks
Dijkstra et al.[89] prepared an A-B diblock copolymer
[number-average molecular weight (Mn)¼ 3.87� 104 g �
mol�1] of PLLA (A) and PCL (B) as well as PDLA
(Mn ¼ 9.7� 103 g �mol�1). They indicated that PLA stereo-
complexation takes place even in the presence of polymeric
impurity of a PCL block. In this blend, e-caprolactone unitsequences were phase-separated to form their crystalline
regions.
Stevels et al.[90] synthesized A-B diblock copolymers of
PLLA or PDLA (DP¼ 1–80) (A) and PCL (DP¼ 70) (B).
Upon blending PLLA-b-PCL and PDLA-b-PCL, both
having weight fractions of PLA blocks above 44%, a Tmincrease (ca. 55 8C) of the PLA crystalline regions was
observed due to stereocomplexation, while for the blends
composed of PLLA-b-PCL and PDLA-b-PCL, both having
weight fractions of PLA blocks below 22%, no melting of
the PLA crystalline regions was observed, meaning that the
PLA blocks were noncrystallizable. On the other hand,
crystallization of PCL blocks took place in the blended
specimens aswell as in the nonblended specimens. Portinha
et al.[56] prepared A-B diblock copolymers of PLLA or
PDLA (A) and PCL (B) and monitored their aggregation
behavior in nonblended and blended THF solutions. The
hydrodynamic radii of assemblies in enantiomeric blended
solutions were 200 nm, which were higher than those in
nonblended polymer solutions. The radius distribution in
enantiomeric blend solutions was sharper than that of non-
blended polymer solutions. Furthermore, the same research
group[57] continued studies on the self-assembly of PLLA-
b-PCL and PDLA-b-PCL in THF by the use of DSC,
dynamic LSmeasurements, 1HNMRspectroscopy, and FT-
IR. They revealed that, at higher concentrations such as
10 g � dL�1, stereocomplexation was in competition with a
solvophobically driven aggregation, whereas at lower con-
centrations, only the stereocomplexation process was
involved.
Figure 24. Schematic representation for the phases and Tm in Blend 11 [nonblend P(DLA-GA), Blend 31 and 32 [blended films from PDLA and P(DLA-GA)] (a), Blend 2 [blendedfilms from P(LLA-GA) and P(DLA-GA)], Blend 41 [blended films from PDLA and P(LLA-GA)], and Blend 42 [blended films from PLLA and P(DLA-GA)] (b).[29] The areas withhorizontal stripes, vertical stripes, both horizontal and vertical stripes, and without stripesmeans homo-crystallineþ amorphous, stereocomplex crystallineþ amorphous, stereocom-plex crystallineþ homo-crystallineþ amorphous, and amorphous, respectively. Tm1 andTm2 are the Tm of homo-crystallites and stereocomplex crystallites, respectively.
588 H. Tsuji
Macromol. Biosci. 2005, 5, 569–597 www.mbs-journal.de � 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Pensec et al.[91] synthesized A-B-A triblock copolymers
of PLLA or PDLA (DP¼ 58) (A) and PCL (DP¼ 34) (B)
and A-B diblock copolymers of PLLA (DP¼ 19–54) or
PDLA (DP¼ 20–57) (A) and PCL (DP¼ 24–106) (B).
They indicated that stereocomplexation occurred in all
enantiomeric polymer pairs precipitated from a dichlor-
omethane solution into methanol. The DHm and Tm of
stereocomplex crystallites were affected by the L-lactyl unit
sequence length and D-lactyl unit sequence length, but were
not affected by the e-caprolactone sequence length.
4.1.4(b). With Poly(ethylene glycol) (PEG) Blocks
Brizzolara et al.[50] prepared A-B diblock copolymers
(Mn ¼ 5.6� 104 and 5.9� 104 g �mol�1) of PLLAor PDLA
(A) and PEG (B). The molecular weight ratio of the PLA
block to the PEG block was 1.5/1. They traced stereo-
complexation between enantiomeric copolymer blends by
WAXS, and the difference in the morphology between
precipitates of nonblended and blended specimens by
AFM.
Stevels et al.[92] synthesized A-B-A triblock copolymers
(Mn ¼ 6.7� 103–2.3� 104 g �mol�1, PEG content 27–
86 wt%) of PLLA or PDLA (A) and PEG (Mn ¼ 6 000) (B)
and prepared equimolarmixtures of PLLA-b-PEG-b-PLLA
and PDLA-b-PEG-PDLA by two different procedures,
precipitation and solution-casting. Although they investi-
gated the effects of the procedures on stereocomplexation
as reported for the blends from pure PLLA and PDLA,[20,21]
stereocomplexation occurred readily in specimens prepared
by these two different procedures. Probably, the molecular
weights of the block copolymers were sufficiently low to
form stereocomplex crystallites, irrespective of the proce-
dure. Another probable explanation is that the PEG seg-
ments acted as a plasticizer for the PLA segments and
consequently enhanced stereocomplexation of the block
copolymers. On the other hand, Lim and Park[93] prepared
A-B-A triblock copolymers of PLLA (DP¼ 200, 250) or
PDLA (DP¼ 208, 262) (A) and PEG (DP¼ 23, 77) (B).
Stereocomplexation between the enantiomeric block copo-
lymers was traced by DSC and WAXS. The cumulative
release of bovine serum albumin (BSA) was lower for
stereocomplexed microspheres than for those of non-
blended blockcopolymers, when compared with the same
period. Furthermore, Fujiwara et al.[94] prepared synthe-
sized A-B-A triblock copolymers (Mn ¼ 7 200 and 6 800 g �mol�1) of PLLA or PDLA (A) and PEG (Mn ¼ 4 600 g �mol�1) (B). The enantiomeric triblock copolymers were
separately dissolved in tetrahydrofuran (THF)/water (v/
v)¼ 1/2 and then THF was removed by evaporation,
leaving 10 wt.-% aqueous dispersion. They traced stereo-
complexation between the enantiomeric triblock copoly-
mers upon heating the aqueous dispersion from room
temperature to 37 8C by rheological measurements, the test
tube tilting method, and WAXS.
Li and Vert[95,96] synthesized A-B diblock and A-B-A
triblock copolymers of PLLA or PDLA (DP¼ 12–
52, molar mass¼ 860–3 700) (A) and PEG (DP¼ 104–
454, molar mass¼ 4 600–20 000) (B). The crystallization
of PEG was dominant in these block copolymers. On
blending the enantiomeric block copolymers in an aqueous
medium, gels were formed, as seen for the homopolymer
blends of PLLA and PDLA in organic solvents.[26,62] Such
stereocomplexation was confirmed by Raman spectro-
scopy, WAXS, and rheological measurements.[95,96]
4.1.4(c). With Poly(sebacic acid) (PSA) Blocks
Slivniak and Domb[97] synthesized A-B-A triblock copo-
lymers of PLLA or PDLA (DP¼ 20–30) (A) and PSA
(DP¼ 2–40) (B) and observed stereocomplexation
between these enantiomeric block copolymers.
4.1.5. Graft Copolymers
Lim et al.[98] prepared poly[2-hydroxyethyl methacrylate-
graft-oligo(L-lactide)] and poly[2-hydroxyethyl methacry-
late-graft-oligo(D-lactide)] by radical polymerization of
macromonomers 2-hydroxyethyl methacrylate-graft-
oligo(L-lactide) and 2-hydroxyethyl methacrylate-graft-
oligo(D-lactide), respectively, in the presence of 2,20-azoisobutyronitrile (AIBN). The macromonomers were
synthesized by ring-opening polymerization of L- or
D-lactide in the presence of stannous octoate and 2-
hydroxyethyl methacrylate as initiator and coinitiator, res-
pectively. Stereocomplexation occurred during solution-
casting of mixed solutions of enantiomeric graft polymers.
The stereocomplexed films became hydrogels in aqueous
media.
de Jong et al.[86,99–102] synthesized dextran [degree of
substitution (DS, number of lactic acid side chains per 100
glucopyranose units)¼ 3–17]-graft-L-(or D-)lactic acid
oligomers [average degree of polymerization (DPav)¼ 6–
12] by forming a carbonate bond using N,N0-carbonyldii-midazole (CDI) and monitored their stereocomplexation in
an acetate-buffered solution (100� 10�3M, pH 4) by
rheological measurements. Here, lactic acid oligomers
were synthesized by ring-opening polymerization of LLA
or DLA using stannous octoate and 2-(2-methoxyethoxy)
ethanol (MEE) as initiator and coinitiator, respectively.
At 20 8C the storage modulus (G0) of dextran-graft-
enantiomeric lactic acid oligomers increased with time due
to the formation of crosslinks composed of stereocomplex
crystallites, but decreased upon heating to 80 8C due to their
melting. They confirmed stereocomplexation by FT-IR[99]
as well as WAXS.[57] The properties of stereocomplexed
hydrogel can be manipulated by alteringDP andDS. It was
found that at least 11 lactyl units are required to obtain a
stereocomplexed hydrogel.[99] These results are summar-
ized in detail in their review article.[101] Furthermore, the
same research group, van Nostrum et al.[103] prepared
Poly(lactide) Stereocomplexes: Formation, Structure, Properties, Degradation, and Applications 589
Macromol. Biosci. 2005, 5, 569–597 www.mbs-journal.de � 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
poly(2-hydroxypropylmethacrylamide) (pHPMAm)-graft-
L-(or D-)lactic acid oligomers by forming an ester linkage
using AIBN. The gels were formed from pHPMAm-graft-
enantiomeric L- and D-lactic acid oligomers in an acetate-
buffered solution (pH 4) as evidenced by rheological
measurements (G0).Watanabe and Ishihara et al.[104–106] synthesized graft–
type copolymers containing L-lactide unit sequences or D-
lactide sequences as side-chains (PMBLLA and PMBDLA,
respectively) by copolymerization of 2-methacryloylox-
yethyl phosphorylcholine (MPC), butyl methacrylate, and
PLLA or PDLA macromonomers. They prepared porous
stereocomplexed films using an extraction method with
water-soluble particles of NaCl.[104] On the other hand,
Tretinnikov et al.[107] synthesized graft layers of PLLAwith
thicknesses from 7 to 35 nm from hydroxy end-groups of a
self-assembled monolayer on gold and monitored the
stereocomplexation between free PDLA and grafted PLLA
chains on the surface by the use of WAXS and FT-IR. In
other words, PDLAmoleculeswere effectively adsorbed on
or entrapped by the PLLA-grafted surface.
4.2. Stereocomplexation in Nonblended Polymers
4.2.1. Stereo-Block Copolymers
Yui et al.[108] and Dijkstra et al.[89] synthesized A-B diblock
copolymers of PLLA (A) and PDLA (B) having L-lactyl
unit content of from 82 to 37% through a ring-opening
sequential two-step polymerization of L- and D-lactide
[Figure 1(e)] initiated by aluminum tris(2-propanolate)
Al[OCH(CH3)2]3. Although they observed stereocom-
plexation between L-lactyl unit sequences and D-lactyl unit
sequences in a block copolymer (Mn ¼ 2.01� 104 g �mol�1) even with L-lactide unit content of 55%, further
investigation on the block copolymers has not been reported
so far.
Spassky et al.,[109,110] Wisniewski et al.,[111] Sarasua
et al.,[112] Radano et al.,[113] and Ovitt and Coates[114,115]
synthesized highly isotactic stereo-block copolymers from
lactides with relatively low optical purities in one step
[Figure 1(f)] using Schiff’s base/aluminum alkoxide
initiators, [(�)-(Salbinap)AlOMe],[109,110] achiral Sale-
nAlOMe,[111,112] or racemic [(Salbinap)AlOiPr].[113–115]
They traced stereocomplexation in the stereo-block copo-
lymers by DSC, WAXS, and/or POM. Sarasua et al.[112]
showed that stereocomplexation of PLAwith [a]58925 of�66,
�73, and �95 deg � dm�1 � g�1 � cm3 in chloroform can
occur when the crystallization temperature and time are
carefully selected.
Fukushima et al.[116] proposed a novel procedure to
synthesize stereo-block PLA by solid-state polycondensa-
tion of a stereocomplexed mixture of PLLA and PDLA. In
the first step, homopolymers PLLA and PDLA having 2.0–
4.6� 104 g �mol�1 were prepared; in the second step the
enantiomeric homopolymers were melt-mixed to form
stereocomplex crystallites; in the third step further poly-
condensation of the stereocomplexed mixture was carried
out.
4.2.2. Random Stereo Copolymers
Normally, in marked contrast to stereo-block copolymers
and blends between L-lactide-rich PLA and D-lactide-rich
PLA, no stereocomplexation occurs in relatively random
stereo copolymers during materials preparation. However,
Li et al. found steteocomplexation between L-lactyl unit
sequences and D-lactyl unit sequences in poly(L-lactide-co-
D-lactide)(62.5/37.5),[78] poly(L-lactide-co-D-lactide)(50/
50) (i.e. PDLLA) prepared with zinc powder,[79] PDLLA
prepared with stannous octoate,[80] and supplied
PDLLA,[81] when thick plates of these copolymers were
hydrolyzed to a great extent in a phosphate-buffered solu-
tion at 37 8C for 30 and 17 weeks, in an aqueous solution
with a caffeine base for 27 weeks, and in a phosphate-
buffered solution at 60 8C for 4 weeks, respectively.
Stereocomplexation took place in a shorter period for plate
specimens than for film specimens,[81] reflecting the fact
that accelerated hydrolytic degradation at the core parts of
the plates enhances stereocomplexation. They suggested
that although the initial fractions of L-lactyl unit sequences
and D-lactyl unit sequences having high sequence numbers
are low in these copolymers, the fractions will be increased
by selective hydrolysis and removal of chains with
relatively random sequences, resulting in stetreocomplexa-
tion. Stereocomplexation can be ascribed to the predomi-
nantly isotactic structure of the copolymers obtained by
ring-opening polymerization of lactides with low optical
purities. Schwach et al.[82] investigated the effects of the
kinds of catalysts on stereocomplexation aswell as tacticity,
water uptake, weight loss, and release of acids. It was
demonstrated that stereocomplexation during hydrolytic
degradation was not influenced by the kind of catalyst and
that PDLLA synthesized with stannous octoate had a
predominantly isotactic structure.
5. Hetero-Stereocomplexation
In the previous sections, stereocomplexation is restricted to
that between L-lactyl unit sequences and D-lactyl unit
sequences (homo-stereocomplexation). In this section,
stereocomplexation between the PDLA and the L-enantio-
meric form of an optically active polymer or between PLLA
and the D-enantiomeric form of another kind of optically
active polymer (hetero-stereocomplexation) is described.
Slager andDomb et al. reported hetereo-stereocomplexa-
tion between PDLA and L-configured peptides such as the
luteinizing hormone-releasing hormone (LHRH),[117] leu-
proide (an LHRH nonapeptides analogue),[118–121]
and vapreotide (a cyclic octapeptide somatostatin
590 H. Tsuji
Macromol. Biosci. 2005, 5, 569–597 www.mbs-journal.de � 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
analogue)[117–119]. Hetero-stereocomplexation was
observed by the use of DSC,[117,120] SEM,[117–120]
SAXS,[118,119] cryogenic transmission electronmicroscopy
(Cryo-TEM),[118,119] and confocal microscopy.[120] In
addition to DSC results, the increase in scattering intensity
in SAXS suggested the formation of hetero-stereocomplex
particles, which was further evidenced by SEM and
Cryo-TEM and confocal microscopic photographs. The
stereocomplex particles had amean unweighed particle size
of 1.7 mm.[119]
L-Insulin and PDLA form stereocomplex porous parti-
culate precipitates and become insoluble in acetonitrile
which dissolves isotactic PLLA or PDLA at an elevated
temperature.[122] The results of DSCmeasurements and the
fact that after the formation of particles no free insulin was
detected by HPLC supported their hetero-stereocomplexa-
tion. They also prepared hetero-stereocomplexed particles
for drug delivery systems (DDS) of insulin with PDLA,
PDLA-b-PEG, PDLA-b-PEG-PDLA, PLLA/PDLA, or
PLLA-b-PEG/PDLA-b-PEG. When insulin, D-lactyl unit
sequences, and L-lactyl unit sequences were involved in a
system, two types of stereocomplex can be formed, i.e., a
homo-stereocomplex and a hetero-stereocomplex.
On the other hand, Force-Field simulation predicted that
PLLA [(�)-PLA] can form a hetero-stereocomplex with
(þ)-alternating isotactic propylene-CO-copolymers [P(P-
alt-CO)] and vice versa.[50] However, Brizzolara et al. have
not yet shown experimental evidence for hetero-stereo-
complexation.
6. Degradation
6.1. Thermal Degradation
Tsuji and Fukui performed a thermogravimetric study on an
equimolar (XD¼ 0.5) blend of PLLA and PDLA as well as
nonblended PLLA and PDLA.[123] These polymers were
synthesized with stannous octoate (0.03 wt%, Sn con-
tent< 88 ppm) and lauryl alcohol (0.5 wt% for PLLA,
0.4 wt% for PDLA) and purified by precipitation using
methylene chloride and methanol, as solvent and non-
solvent, respectively. The initial Mn values of PLLA and
PDLA before thermal degradation were 8.7� 104 and
9.5� 104 g �mol�1, respectively. In heat scanning at a cons-
tant rate in thermogravimetry (TG), a very small difference
was observed between the TG curves of the specimens,
whereas at fixed temperatures of 250 and 260 8C exceeding
the Tm of the stereocomplex, i.e. in the melt, the equimolar
blend has higher stability than the nonblended PLLA or
PDLA (Figure 25). It is expected that the 31 helical
conformation remains even in the melt of the blend at
temperatures exceeding the Tm of the stereocomplex and,
therefore, the peculiarly strong interaction between
L-lactide chains and D-lactide chains has a significant effect
in reducing the molecular mobility and, therefore, in
disturbing the thermal degradation. However, it seems that,
at temperatures far higher than the Tm of the stereocomplex,
such peculiar interaction arising from the helical conforma-
tion disappears, resulting in a small difference in thermal
stability between nonblended and blended films. The
activation energy for the thermal degradation (DEtd) values
was estimated by the method recommended byMacCallum
et al.[124–127] The obtained DEtd value of the equimolar
blend was in the range of 205–297 kJ �mol�1, which was
higher by 82–110 kJ �mol�1 than the averaged DEtd values
of nonblended PLLA and PDLA (87–104 kJ �mol�1).[123]
On the other hand, Fan et al.[128] reported DEtd values of
80–100, 100–120, and 125–180 kJ �mol�1 respectively for
as-polymerized (Sn content: 286 ppm), purified by preci-
pitation with methanol (Sn content: 266 ppm), and purified
metal-free (Sn content < 10 ppm) equimolar blend speci-
mens from PLLA (Mn ¼ 1.2–1.3� 105 g �mol�1) and
PDLA (Mn ¼ 1.2–1.3� 105 g �mol�1). The thermal degra-
dation of the three specimens proceeds through mechan-
Figure 25. The percentage of remaining weight of the filmsmeasured isothermally at the constant temperature as a function ofheating time.[123] Here, L, D, L/D represent nonblended PLLA andPDLA films, and their equimolarly blended film, respectively.
Poly(lactide) Stereocomplexes: Formation, Structure, Properties, Degradation, and Applications 591
Macromol. Biosci. 2005, 5, 569–597 www.mbs-journal.de � 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
isms of unzipping caused by the Sn-alkoxide chain end,
Sn-catalyzed selective lactide elimination, and random
degradation, respectively. In our study, constant temper-
ature analysis was made in the temperature range where
PLLAandPDLAhave a peculiarly strong interaction,while
Fan et al. analyzed the specimens under constant heating,
and the procedure for DEtd estimation is different from our
case.[129–131] The DEtd value difference between the two
articles (ours and Fan et al.) is attributable to the differences
in initial molecular weights and terminal groups of PLLA
and PDLA, the kind and concentration of the remaining
catalyst, and the method and procedure forDEtd estimation.
6.2. Hydrolytic Degradation
Tsuji found that stereocomplexed PLA specimens have a
higher hydrolysis-resistance compared with that of non-
blended PLLA and PDLA specimens when they were
hydrolyzed in a phosphate-buffered solution at pH 7.4 and
37 8C (Figure 26).[132] It is surprising that although the
decrease in tensile strength of the nonblended specimens
started even at 4 months, the stereocomplex specimens
retained their initial tensile strength for a long period of
16 months. Similarly, de Jong et al.[100] reported hydrolytic
degradation of solution-cast L-lactic acid oligomers
(DP¼ 7) and the stereocomplex of L- and D-lactic acid
oligomers (DP¼ 7) at pH7 and 37 8C.They showed that thefraction of the L-lactic acid oligomer approached nil within
4 h, whereas 50% of the stereocomplex remained even after
96 h of degradation. Here, although the L-(or D-)lactic acid
oligomer and stereocomplex are amorphous and crystalline,
respectively, the obtained results are in agreement with our
results.[132] Such stereocomplex crystallization should have
disturbed the hydrolytic degradation of stereocomplexed
specimens compared with that of the L-lactic acid oligomer.
However, hydrolytic degradation proceeds predominantly
in the amorphous regions between the stereocomplex
crystalline regions. This means that the PLLA chains and
PDLA chains should have a peculiarly strong interaction
even when they are in an amorphous state, as suggested by
the aforementioned thermal degradation results.[123] To
confirm this assumption, we prepared various types of equi-
molarly blended specimens from PLLA and PDLA, i.e.,
amorphous-made[133] and homo-crystallized,[19] and car-
ried out their hydrolytic degradation in a phosphate-
buffered solution at pH 7.4 and 37 8C, together with
nonblended PLLA and PDLA specimens. We observed
retarded hydrolytic degradation of the blended specimens
compared with the nonblended specimens, irrespective of
their state, amorphous or homo-crystallized, reinforcing the
abovementioned hypothesis. Tsuji and Suzuki[134] carried
out the hydrolytic degradation of stereocomplexed fibers
and films and revealed that the morphology of the stereo-
complexedmaterials have crucial effects on their hydrolytic
degradation rates.
There have been some reports that stereocomplexa-
tion between enantiomeric L-lactide unit sequences and
D-lactide unit sequences retarded hydrolytic degradation;
e.g., poly[2-hydroxyethyl methacrylate-graft-oligo(lac-
tide)] (Lim et al.[98]) and A-B-A triblock copolymers of
Figure 26. Weight remaining (a),Mn (b), and tensile strength (c)of nonblended films and well-stereocomplexed blended film ofPLLA and PDLA films after hydrolytic degradation in phosphate-buffered solution (pH 7.4, 37 8C).[132]
592 H. Tsuji
Macromol. Biosci. 2005, 5, 569–597 www.mbs-journal.de � 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
PLA (A) with poly(sebacic acid) (PSA) (B) (Slivniak and
Domb[97]). On the other hand, de Jong et al.[102] indicated
that hydrolytic degradation of the stereocomplex hydrogels
from dextran (DS¼ 3–12)-graft-L- and D-lactic acid oligo-
mers (DP¼ 6–12) depends on the number of lactate grafts
(DS), the length (DP) and polydispersity of the grafts, and
the initial water content, and that the degradation time
varied from one to seven days. Moreover, van Nostrum
et al.[103] showed that the hydrolytic degradation time of the
stereocomplex hydrogels from poly(2-hydroxypropyl-
methacrylamide) (pHPMAm)-graft-L- and D-lactic acid
oligomers can be readily tailored from 1 week to almost
3 weeks by changing the grafting density of the polymers
and the structure of the terminal group of the side chains.
6.3. Alkaline Degradation
Serizawa et al.[135] prepared the PLA stereocomplex assem-
bly and the assemblies composed of both a stereocomplex
crystalline layer and a PLLA homo-crystalline layer
deposited on QCM substrates by the procedure stated
earlier (Section 3.2.4), and their alkaline hydrolytic degra-
dation rates were investigated. They claimed that the
hydrolysis-resistance of the PLLA crystalline layer having
a 103 helical conformation is higher than the stereocomplex
layer having a 31 helical conformation, in marked contrast
to the abovementioned results in neutral media (Section
6.2).
6.4. Enzymatic Degradation
Proteinase K is an endo-protease having broad specificity
but with preference for the cleavage of the peptide bond
C-terminal to aliphatic and aromatic amino acids, espe-
cially alanine.[136] The similar chemical structures of lactic
acid and alanine are expected to induce the proteinase
K-catalyzed hydrolysis of the C-terminal of PLLA or
L-lactyl unit sequences. As already found, proteinase K can
catalyze the hydrolytic degradation of L-lactyl chains in
amorphous regions.[137,138] The tie chains and chains with a
long free end in amorphous regions can be enzymatically
cleaved, whereas the folding chains and the chains with a
short free end are highly resistant to enzymatic clea-
vage.[139] Since hydrolytic degradation proceeds predomi-
nantly in amorphous regions, we investigated the effects of
the presence of PDLAon proteinaseK-catalyzed enzymatic
degradation of PLLA in an amorphous state.[140] Assuming
that PLLA and PDLA have no interaction with each other,
PLLA in blends with PDLAwill be enzymatically degraded
in the presence of proteinase K as in nonblended PLLA.
However, the enzymatic hydrolysis rate (REH) of PLLAwas
largely reduced by the presence of PDLA (Figure 27). In
otherwords, the presence of PDLAdisturbed the adsorption
or cleavage process of proteinase K. The finding here
reflects the fact that PLLAandPDLAchains arewell-mixed
in the amorphous state and that they have strong interaction
with each other.
7. Applications
7.1. Biodegradable Films
Biodegrdable PLA-based stereocomplex films can be pre-
pared by the solution-casting method with organic solvents
such as chloroform and methylene chloride.[20,62] In this
method, the molecular weights of L-lactyl unit sequences
and D-lactyl unit sequences and the solvent evaporation rate
are crucial parameters. Molecular weights and solvent eva-
poration rate which are too high disturb the stereocomplex
formation, resulting in the formation of films containing a
relatively large amount of homo-crystallites or having low
crystallinities.[20] Although PLA stereocomplex films can
also be prepared by the melt-molding method, it should be
noted that the critical molecular weights for PLLA and
PDLA, below which only stereocomplex crystallites are
formed, decrease dramatically compared with those in the
solution-casting method.[22] This indicates the difficulty in
preparing well-stereocomplexed PLA materials with high
molecular weights.
7.2. Biodegradable Fibers
Murdoch and Loomis[62] prepared melt-spun PLA stereo-
complex fibers from equimolar mixture of PLLA and
PDLA, while Tsuji et al. obtained wet- and dry-spun PLA
stereocomplex fibers[77] from mixed chloroform solutions
of equimolar PLLA and PDLA. In the former study no
estimation of the fractions of stereocomplex crystallites and
Figure 27. Proteinase K-catalyzed enzymatic hydrolysisrate (REH) of blended films from PLLA and PDLA and as afunction of XD.
[140]
Poly(lactide) Stereocomplexes: Formation, Structure, Properties, Degradation, and Applications 593
Macromol. Biosci. 2005, 5, 569–597 www.mbs-journal.de � 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
homo-crystallites was carried out,[62] whereas in the latter
study fraction estimation byDSC revealed that hot-drawing
of as-spun fibers can increase the amount of stereocomplex
crystallites and reduce that of homo-crystallites, resulting in
the formation of fibers whose main crystalline kind is
stereocomplex.[77] The PLA stereocomplex fibers used by
Tsuji and Suzuki[134] for hydrolysis experiments, which
were prepared by melt-spinning and subsequent two-stage
hot-drawing, contain only stereocomplex crystallites and
no homo-crystallites.
Later, Takasaki et al.[141] revealed that stereocomplexa-
tion is favored in melt-spun fibers from equimolar mixtures
of PLLA and PDLAunder the spinning conditions of higher
take-up velocity, lower throughput rate, and lower extrusion
temperature. This result suggests that these conditions can
enhance the orientation-induced crystallization of the
stereocomplex, as described above.[77,134] They also con-
firmed that drawing at a lower temperature and annealing
between the Tm of stereocomplex crystallites and homo-
crystallites further enhanced stereocomplexation in agree-
mentwith the aforementionedfindings.[22,62]Themaximum
tensile strength and Young’s modulus were respectively
530 MPa and 7.4 GPa for melt-spun and drawn stereo-
complex fibers,[62] 920 MPa and 8.6 GPa for solution-spun
and drawn stereocomplex fibers,[77] and 400 MPa and
4.7 GPa for as-spun stereocomplex fibers.[141] These values
are so far lower than the maximum tensile strength and
Young’s modulus, 1.8 GPa and 14 GPa, of melt-spun and
drawn PLLA fibers.[142]
7.3. Biodegradable Microspheres for DDS
Loomis and Murdoch[65] prepared injectable stereocom-
plex microspheres containing a naltrexone base using the
oil-in-water (O/W) solvent-evaporationmethod (O: methy-
lene chloride, W: water with a surfactant) and the release
rate was determined in various media. The release of
naltrexonewas delayed by a lag time of 200, 230, and 240 h
in water, acid and buffer solutions, respectively. After the
lag the release occurred in zero order up to 450 h with 46,
48, and 35% of the drug released in water, acid and buffer
solutions, respectively. On the other hand, de Jong et al.[102]
prepared stereocomplex hydrogels from dextran (DS¼ 3–
12)-graft-L- and D-lactic acid oligomers (DP¼ 6–12) and
showed that the stereocomplex gels released the entrapped
model proteins (IgG and lysozyme) during six days.
Slager and Domb formulated hetero-stereocomplex
DDS particles from L-configured peptides such as insulin
with PDLA, PDLA-b-PEG, PDLA-b-PEG-PDLA, PLLA/
PDLA, or PLLA-b-PEG/PDLA-b-PEG.[122] The strong
physical entrapment of peptides by the D-lactide unit
sequence resulted in retarded release of the peptides. They
also prepared hetero-stereocomplex DDS particles from an
L-configured leuprolide and PDLA.[119–121] Various factors
affecting the release of leuprolide from the hetero-
stereocomplex particles were investigated. The release rate
of leuprolide from the particles increased with decreasing
the molecular weight of PDLA and increasing the weight
fraction of leuprolide. Continuous release of leuprolide for
over one hundred days was observed for certain stereo-
complex compositions.[121]
7.4. Biodegradable Hydrogels
Stereocomplexed PLA hydrogels can be prepared in
aqueous media by blending block or graft copolymers with
hydrophilic segments and L-lactyl unit sequences or
D-lactyl unit sequences. Stereocomplex hydrogels were
reported for enantiomeric A-B diblock and A-B-A triblock
copolymers of PLA (A) with PEG (B),[95,96] poly[2-
hydroxyethyl methacrylate-graft-oligo(lactide)],[98]
dextran-graft-lactic acid oligomers,[86,99–102] and
pHPMAm-graft-lactic acid oligomers.[103]
On the other hand, Watanabe and Ishihara et al.[104–106]
prepared porous stereocomplexed PLA films from graft-
type copolymers, PMBLLA and PMBDLA, using an ex-
traction procedure with water-soluble particles of NaCl[104]
and investigated the cell adhesion and morphology on the
porous scaffolds.[106] They revealed that the number of
adhering cells is correlated with the PLLA or PDLA con-
tent, and that cell morphology is correlated with the MPC
unit content. Fibroblast cells adhered on the surface and
intruded into the scaffolds through the connected pores after
24 h. The cell morphology became round from spreading
with the decreasing PLLA or PDLA content in the
scaffolds.
7.5. Nucleation Agents
Schmidt and Hillmyer[143] and Yamane and Sasai[144] re-
ported that stereocomplex crystallites formed by the
addition of small amounts of PDLA to PLLA and act as
heterogeneous nucleation sites for PLLA crystallization, or
that PLLA homo-crystallites are formed epitaxially on
the stereocomplex crystallites. Such nucleation agents
increase the number of PLLA spherulites per unit volume
and the total crystallization rate but does not alter spherulite
growth rate G. The nucleation effects must arise from
epitaxial crystallization of PLLA homo-crystallites on the
stereocomplex crystallites, as reported for PLLA and
poly(L-lactide-co-D-lactide) (20/80) when they were mixed
at a ratio of 80/20 and crystallized during slow cooling from
the melt.[68]
8. Conclusions and Perspectives
Stereocomplexation gives PLA-based materials higher
mechanical performance, thermal resistance, hydrolysis-
resistance, and opens a newway to produce various types of
594 H. Tsuji
Macromol. Biosci. 2005, 5, 569–597 www.mbs-journal.de � 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
biodegradable materials such as hydrogels and DDS parti-
cles. The former improvements arise from the peculiarly
strong interaction between L-lactyl unit sequences and
D-lactyl unit sequences. A variety of properties of stereo-
complexed PLAmaterials can bemanipulated bymolecular
characteristics, highly-ordered structures, and additives.
Some Lactobacilli are reported to produce exclusively
D-lactic acid (not the mixture of L- and D-lactic acids) from
numerous kinds of renewable resources,[30,145] and PDLA
can be produced from D-lactic acid by the same procedure
for PLLA production from L-lactic acid. Therefore, the
most crucial issue for stereocomplexed PLA materials,
the reduction of the production cost of PDLA, can be solved
by large-scale facilities. Reduced cost of PDLA will give
stereocomplexed PLA materials further applications, not
only as biomedical materials but also as alternatives for
commercial polymeric materials.
[1] T. G. Fox, B. S. Garrett, W. E. Goode, S. Gratch, J. F.Kincaid, A. Spell, J. D. Stroupe, J. Am. Chem. Soc. 1958,80, 1768.
[2] L. Pauling, R. B. Corey,Proc. Natl. Acad. Sci. Wash. 1953,39, 253.
[3] D. Grenier, R. E. Prud’homme, J. Polym. Sci. Polym. Phys.Ed. 1984, 22, 577.
[4] Y. Ikada, K. Jamshidi, H. Tsuji, S.-H. Hyon, Macro-molecules 1987, 20, 904.
[5] R. Voyer, R. E. Prud’homme, Eur. Polym. J. 1989, 25,365.
[6] H. Tsuji, in:ResearchAdvances inMacromolecules, R.M.Mohan, Ed., Grobal Research Network, Trivandrum,India, Vol. 1, 2000, pp. 25–48.
[7] J. H. G. M. Lohmeyer, Y. Y. Tan, P. Lako, G. Challa,Polymer 1978, 19, 1171.
[8] F. Bosscher, D. Keekstra, G. Challa, Polymer 1981, 22,124.
[9] J. Slager, A. J. Domb, Adv. Drug Delivery Rev. 2003, 55,549.
[10] H. Sakakihara, Y. Takahashi, H. Tadaokoro, N. Oguni, H.Tani,Macromolecules 1973, 6, 205.
[11] K. L. Singfield, G. R. Brown, Macromolecules 1995, 28,1290.
[12] P. Dumas, N. Spassky, P. Sigwalt,Makromol. Chem. 1972,156, 55.
[13] Z. Jiang,M.T.Boyer,A. Sen, J. Am.Chem. Soc. 1995, 117,7037.
[14] T. Yoshida, S. Sakurai, T. Okuda, Y. Takagi, J. Am. Chem.Soc. 1962, 84, 3590.
[15] M. Tsuboi, A. Wada, A. Nagashima, J. Mol. Biol. 1961, 3,705.
[16] I. Iribarren, C. Aleman, C. Regano, A. Martınez deIlarduya, J. J. Bou, S. Munoz-Guerra, Macromolecules1996, 29, 8413.
[17] K. Hatada, S. Shimizu, Y. Terawaki, K. Ohta, H. Yuki,Polym. J. 1981, 13, 811.
[18] F. Sanda, M. Nakamura, T. Endo, Macromolecules 1996,29, 8064.
[19] H. Tsuji, Biomaterials 2003, 24, 537.
[20] H. Tsuji, S.-H. Hyon, Y. Ikada,Macromolecules 1991, 24,5651.
[21] H. Tsuji, S.-H. Hyon, Y. Ikada,Macromolecules 1991, 24,5657.
[22] H. Tsuji, Y. Ikada, Macromolecules 1993, 26, 6918.[23] H. Tsuji, Y. Ikada, Macromol. Chem. Phys. 1996, 197,
3483.[24] H. Tsuji, Y. Ikada, Polymer 1999, 40, 6699.[25] E. Schomaker,G.Challa,Macromolecules 1988, 21, 3506.[26] H. Tsuji, F. Horii, S.-H. Hyon, Y. Ikada, Macromolecules
1991, 24, 2719.[27] H. Tsuji, S.-H. Hyon, Y. Ikada,Macromolecules 1992, 25,
2940.[28] H. Tsuji, Y. Ikada, Macromolecules 1992, 25, 5719.[29] H. Tsuji, Y. Ikada, J. Appl. Polym. Sci. 1994, 53, 1061.[30] G. B. Kharas, F. Sanchez-Riera, D. K. Severson, in:
Plastics from Microbes, D. P. Mobley, Ed., HanserPublishers, New York 1994, pp. 93–137.
[31] M. H. Hartmann, in: Biopolymers from RenewableResources, D. L. Kaplan, Ed., Springer, Berlin 1998, pp.367–411.
[32] H. Tsuji, Y. Ikada, in: Current Trends in Polymer Science,K. L. DeVries, P. Hodge, A. Ledwith, D. W. McCall,A. M. North, D. R. Paul, R. S. Porter, J. C. Salamone, P. L.Taylor, O. Vogl, Editorial Advisory Board, ResearchTrends, Trivandrum, India, Vol. 4, 1999, pp. 27–46.
[33] Y. Ikada, H. Tsuji, Macromol. Rapid Commun. 2000, 21,117.
[34] H. Tsuji, in: Recent Research Developments in PolymerScience, A. B. Salamone, J. Brandrup, R. M. Ottenbrite,Editorial Advisors, Transworld Research Network,Trivandrum 2000, Vol. 4, pp. 13–37.
[35] D. Garlotta, J. Polym. Environ. 2001, 9, 63.[36] H. Tsuji, in: Polyesters 3 (Biopolymers, vol. 4), Y. Doi, A.
Steinbuchel, Eds.,Wiley-VCH,Weinheim 2002, pp. 129–177.
[37] R. Auras, B. Harte, S. Selke, Macromol. Biosci. 2004, 4,835.
[38] H. Tsuji, Y. Ikada, J. Appl. Polym. Sci. 1995, 58, 1793.[39] H. Tsuji, Y. Ikada, Polymer 1996, 37, 595.[40] H. Tsuji, Y. Ikada, Polymer 1995, 36, 2709.[41] B. Kalb, A. J. Pennings, Polymer 1980, 21, 607.[42] G. L. Loomis, J. R.Murdoch, K. H. Gardner,Polym. Prepr.
1990, 31, 55.[43] H. Tsuji, F. Horii, M. Nakagawa, Y. Ikada, H. Odani, R.
Kitamaru, Macromolecules 1992, 25, 4114.[44] E. W. Fischer, H. J. Sterzel, G. Wegner, Kolloid-Z.u.Z.
Polym. 1973, 251, 980.[45] T. Miyata, T. Masuko, Polymer 1998, 39, 5515.[46] K. Jamshidi, S.-H.Hyon,Y. Ikada,Polymer1988,29, 2229.[47] T. Okihara, M. Tsuji, A. Kawaguchi, K. Katayama, H.
Tsuji, S.-H. Hyon, Y. Ikada, J. Macromol. Sci., -Phys.1991, B30, 119.
[48] D. Brizzolara, H.-J. Cantow, K. Diederichs, E. Keller, A. J.Domb, Macromolecules 1996, 29, 191.
[49] T. Okihara, A. Kawaguchi, H. Tsuji, S.-H. Hyon, Y. Ikada,K. Katayama, Bull. Inst. Chem. Res. Kyoto Univ. 1988, 66,271.
[50] D. Brizzolara, H. J. Cantow, R. Mulhaupt, A. J. Domb, J.Computer-Aided Mater. Design 1996, 3, 341.
[51] H. Tsuji, K. Ikarashi, N. Fukuda, Polym. Degrad. Stab.2004, 84, 515.
[52] G. Kister, G. Cassanas, M. Vert, Polymer 1998, 39, 267.[53] J. Zhang, H. Sato, H. Tsuji, I. Noda, Y. Ozaki, Macro-
molecules 2005, 38, 1822.
Poly(lactide) Stereocomplexes: Formation, Structure, Properties, Degradation, and Applications 595
Macromol. Biosci. 2005, 5, 569–597 www.mbs-journal.de � 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
[54] J. Zhang, H. Sato, H. Tsuji, I. Noda, Y. Ozaki, J. Mol.Struct. 2005, 735, 249.
[55] E. J. Vorenkamp, G. Challa, Polymer 1981, 22, 1705.[56] D. Portinha, J. Belleney, L. Bouteiller, S. Pensec, N.
Spassky, C. Chassenieux,Macromolecules 2002, 35, 1484.[57] D. Portinha, L. Bouteiller, S. Pensec, A. Richez, C.
Chassenieux, Macromolecules 2004, 37, 3401.[58] H. Tsuji, Y. Tezuka, Biomacromolecules 2004, 5, 1181.[59] R.Vasanthakumari, A. J. Pennings,Polymer 1983, 24, 175.[60] H. Abe, Y. Kikkawa, Y. Inoue, Y. Doi,Biomacromolecules
2001, 2, 1007.[61] L. Cartier, T. Okihara, B. Lotz,Macromolecules 1997, 30,
6313.[62] US 4719246 (1988), E. I. Du Pont Co., invs.: J. R.
Murdoch, G. L. Loomis.[63] US 4766182 (1988), E. I. Du Pont Co., invs.: J. R.
Murdoch, G. L. Loomis.[64] US 4800219 (1989), E. I. Du Pont Co., invs.: J. R.
Murdoch, G. L. Loomis.[65] US 4902515 (1990), E. I. Du PontCo., invs.: G. L. Loomis,
J. R. Murdoch.[66] US 4981696 (1991), E. I. Du PontCo., invs.: G. L. Loomis,
J. R. Murdoch.[67] H. Tsuji, Y. Ikada, Polymer 1995, 36, 2709.[68] S. Brochu, R. E. Prud’homme, I. Barakat, R. Jerome,
Macromolecules 1995, 28, 5230.[69] A. Soldera, R. E. Prud’homme, PMSE (Am. Chem. Soc.,
Div. Polym. Mater. Sci. Eng.) 1993, 68, 304.[70] H. Urayama, T. Kanamori, K. Fukushima, Y. Kimura,
Polymer 2003, 44, 5635.[71] US 5317064 (1994), E. I. Du Pont Co., inv.: M. Spinu.[72] M. Spinu, K. H. Gardner, PMSE (Am. Chem. Soc., Div.
Polym. Mater. Sci. Eng.) 1994, 71, 19.[73] M. Spinu, C. Jackson, M. Y. Keating, K. H. Gardner,
J. Macromol. Sci.-Pure Appl. Chem. 1996, A33, 1497.[74] G. Challa, Y. Y. Tan, Pure Appl. Chem. 1981, 53, 627.[75] H. Bourque, I. Laurin, M. Pezolet, Langmuir 2001, 17,
5842.[76] I. Pelletier, M. Pezolet,Macromolecules 2004, 37, 4967.[77] H. Tsuji, Y. Ikada, S.-H. Hyon, Y. Kimura, T. Kitao, J.
Appl. Polym. Sci. 1994, 51, 337.[78] S. Li, M. Vert, Macromolecules 1994, 27, 3107.[79] S. Li, M. Vert, Polym. Int. 1994, 33, 37.[80] S. Li, S. Girod-Holland, M. Vert, J. Controlled Release
1996, 40, 41.[81] S. Li, S. McCarthy, Biomaterials 1999, 20, 35.[82] G. Schwach, J. Coudane, R. Engel, M. Vert, Biomaterials
2002, 23, 993.[83] H. Tsuji, Y. Ikada, J. Appl. Polym. Sci. 1997, 63, 855.[84] H. Tsuji, C. A. Del Carpio, Biomacromolecules 2003, 4, 7.[85] S. J. de Jong, W. N. E. van Dijk-Wolthuis, J. J. Kettenes-
van den Bosch, P. J. W. Schuyl, W. E. Hennink,Macromolecules 1998, 31, 6397.
[86] S. J. de Jong,C. F. vanNostrum,L.M. J.Kroon-Batenburg,J. J. Kettenes-van den Bosch, W. E. Hennink, J. Appl.Polym. Sci. 2002, 86, 289.
[87] T. Serizawa, H. Yamashita, T. Fujiwara, Y. Kimura, M.Akashi, Macromolecules 2001, 34, 1996.
[88] H. Tsuji, Y. Tezuka, Macromol. Biosci. 2005, 5, 135.[89] P. J. Dijkstra, A. Bulte, J. Feijen, Preprints for The 17th
Annual Meeting of the Society for Biomaterials 1991, 184.[90] W. M. Stevels, M. J. K. Ankone, P. J. Dijkstra, J. Feijen,
Macromol. Symp. 1996, 102, 107.[91] S. Pensec, M. Leroy, H. Akkouche, N. Spassky, Polym.
Bull. 2000, 45, 373.
[92] W. M. Stevels, M. J. K. Ankone, P. J. Dijkstra, J. Feijen,Macromol. Chem. Phys. 1995, 196, 3687.
[93] D.W. Lim, T. G. Park, J. Appl. Polym. Sci. 2000, 75, 1615.[94] T. Fujiwara, T. Mukose, T. Yamaoka, H. Yamane, S.
Sakurai, Y. Kimura, Macromol. Biosci. 2001, 1, 204.[95] S. Li, M. Vert, Macromolecules 2003, 36, 8008.[96] S. Li,Macromol. Biosci. 2003, 3, 657.[97] R. Slivniak, A. J. Domb,Biomacromolecules 2002, 3, 754.[98] D. W. Lim, S. H. Choi, T. G. Park, Macromol. Rapid
Commun. 2000, 21, 464.[99] S. J. de Jong, S. C. DeSmedt, M. W. C. Wahls, J.
Demeester, J. J. Kettenes-van den Bosch, W. E. Hennink,Macromolecules 2000, 33, 3680.
[100] S. J. de Jong, B. van Eerdenbrugh, C. F. van Nostrum, J. J.Kettenes-van den Bosch, W. E. Hennink, J. ControlledRelease 2001, 72, 47.
[101] S. J. de Jong,C. F. vanNostrum,L.M. J.Kroon-Batenburg,J. J. Kettenes-van den Bosch,W. E. Hennink, S. J. de Jong,G. W. Bos, T. F. J. Veldhuis, C. F. van Nostrum, Int.J. Pharm. 2004, 277, 99.
[102] S. J. de Jong, S. C. Smedt, J. Demeester, C. F. vanNostrum,J. J. Kettenes-van denBosch,W. E. Hennink, J. ControlledRelease 2001, 71, 261.
[103] C. F. van Nostrum, T. F. J. Veldhuis, G. W. Bos, W. E.Hennink,Macromolecules 2004, 37, 2113.
[104] J. Watanabe, K. Ishihara, Chem. Lett. 2003, 32, 192.[105] J. Watanabe, T. Eriguchi, K. Ishihara, Biomacromoleucles
2002, 3, 1109.[106] J. Watanabe, T. Eriguchi, K. Ishihara, Biomacromoleucles
2002, 3, 1375–1383.[107] O. N. Tretinnikov, K. Kato, H. Iwata, Langmuir 2004, 20,
6748.[108] N. Yui, P. J. Dijkstra, J. Feijen, Makromol. Chem. 1990,
191, 481.[109] N. Spassky, M. Wisniewski, C. Pluta, A. Le Borgne,
Macromol. Chem. Phys. 1996, 197, 2627.[110] N. Spassky, C. Pluta, V. Simic, M. Thiam,M.Wisniewski,
Macromol. Symp. 1998, 128, 39.[111] M. Wisniewski, A. Le Borgne, N. Spassky, Macromol.
Chem. Phys. 1997, 198, 1227.[112] J.-R. Sarasua, R. E. Prud’homme, M. Wisniewski, A. Le
Borgne, N. Spassky,Macromolecules 1998, 31, 3895.[113] C. P. Radano, G. L. Baker, M. R. Smith, III, J. Am. Chem.
Soc. 2000, 122, 1552.[114] T. M. Ovitt, G. W. Coates, J. Polym. Sci.: Part A: Polym.
Chem. 2000, 38, 4686.[115] T. M. Ovitt, G. W. Coates, J. Am. Chem. Soc. 2002, 124,
1316.[116] K. Fukushima, Y. Furuhashi, K. Sogo, S. Miura, Y.
Kimura, Macromol. Biosci. 2005, 5, 21.[117] J. Slager, M. Glandnikoff, A. J. Domb, Macromol. Symp.
2001, 175, 105.[118] J. Slager, Y. Cohen, R. Khalfin, Y. Talmon, A. J. Domb,
Macromolecules 2003, 36, 2999.[119] J. Slager, A. J. Domb, Eur. J. Pharm. Biopharm. 2004, 58,
461.[120] J. Slager, A. J. Domb, Biomacromolecules 2003, 4, 1308.[121] J. Slager, A. J. Domb, Biomacromolecules 2003, 4, 1316.[122] J. Slager, A. J. Domb, Biomaterials 2002, 23, 4389.[123] H. Tsuji, I. Fukui, Polymer 2003, 44, 2891.[124] J. R. MacCallum, in: Comprehensive Polymer Science:
The Synthesis, Characterization, Reactions, and Applica-tions of Polymers, Vol. 1 (Polymer Characterization), G.Allen, J. C. Bevington, Editorial Board, Pergamon,Oxford1989, Chapter 37, pp. 903–909.
596 H. Tsuji
Macromol. Biosci. 2005, 5, 569–597 www.mbs-journal.de � 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
[125] J. R. MacCallum, J. Tanner, Eur. Polym. J. 1970, 6, 907.[126] J. R. MacCallum, J. Tanner, Eur. Polym. J. 1970, 6, 1033.[127] J. Atkinson, J. R. MacCallum, Eur. Polym. J. 1972, 8, 809.[128] Y. Fan, H. Nishida, Y. Shirai, Y. Tokiwa, T. Endo, Polym.
Degrad. Stab. 2004, 86, 197.[129] C. D. Doyle, J. Appl. Polym. Sci. 1962, 6, 639.[130] L. Reich, Polym. Lett. 1964, 2, 621.[131] T. Ozawa, Bull. Chem. Soc. Japan 1965, 38, 1881.[132] H. Tsuji, Polymer 2000, 41, 3621.[133] H. Tsuji, Polymer 2002, 43, 1789.[134] H. Tsuji, M. Suzuki, Sen’i Gakkishi 2001, 57, 198.[135] T. Serizawa, Y. Arikawa, K. Hamada, H. Yamashita, T.
Fujiwara, Y. Kimura, M. Akashi, Macromolecules 2003,36, 1762.
[136] P. J. Sweeney, J. M. Walker, in: Enzymes of MolecularBiology (Methods in Molecular Biology, vol. 16), M. M.
Burrell, Ed., Humana Press, Totowa, New Jersey 1993,Chapter 6, pp. 305–311.
[137] H. Tsuji, S.Miyauchi,Polym.Degrad. Stab. 2001, 71, 415.[138] H. Tsuji, S. Miyauchi, Polymer 2001, 42, 4465.[139] H. Tsuji, K. Ikarashi, Polym. Degrad. Stab. 2004, 85,
647.[140] H. Tsuji, S. Miyauchi, Biomacromolecules 2001, 2, 597.[141] M. Takasaki, H. Ito, T. Kikutani, J. Macromol. Sci. Part B
Phys. 2003, B42, 403.[142] J. W. Leenslag, A. J. Pennings, Polymer Commun. 1987,
28, 92.[143] S. C. Schmidt, M. A. Hillmyer, J. Polym. Sci., Part B:
Polym. Phys. 2001, 39, 300.[144] H. Yamane, K. Sasai, Polymer 2003, 44, 2569.[145] K. Fukushima, K. Sogo, S. Miura, Y. Kimura,Macromol.
Biosci. 2004, 4, 1021.
Poly(lactide) Stereocomplexes: Formation, Structure, Properties, Degradation, and Applications 597
Macromol. Biosci. 2005, 5, 569–597 www.mbs-journal.de � 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim